Introduction

Cationic dyes are extensively used in various industries, including papermaking, textile coloration, coatings, cosmetics, and food processing [1]. When untreated dye effluent is dumped directly into water, it can impact the ecosystem and human health via the food chain. Methylene blue (MB) is a cationic dye that can be carcinogenic and mutagenic during degradation [2]. Moreover, it can induce unpleasant symptoms, including retching, shock, jaundice, cyanosis, tissue necrosis, and even neurological harm [3]. As a result, it is critical and urgent to develop effective methods for the removal of MB from industrial effluents [4]. Therefore, many researchers are interested in removing dyes from aqueous solutions. Adsorption [5], photodegradation [6], flocculation [7], oxidation–reduction [8], and membrane filtration [9] have been employed. Adsorption has become one of the research hotspots among these methods due to its advantages of simple operation, low cost, and recyclability [10]. Therefore, it is challenging to design an adsorbent with excellent selectivity, efficiency, and stability for dye-containing wastewater. So far, adsorbents based on clay [11], activated carbon [12], zeolite [13], metal–organic framework (MOF) [14], and other materials have been reported [15].

Biopolymer hydrogel are three-dimensional network polymer materials made from renewable feedstocks such as starch [16], cellulose [17], chitin/chitosan [18], lignin [19], and agricultural wastes [20]. As the functional groups of biopolymer hydrogel can interact with metal ions or dye molecules, they have shown significant prospects as absorbents for cationic dyes. In addition, these absorbents offer additional advantages over conventional ones, such as low cost, biodegradability, high usability, and high adsorption efficiency [21].

Lignosulfonate (LS), a result of acid pulping, is abundant in the black effluent of pulp mills, has good biocompatibility and is environmental friendliness [22]. Due to its aromatic properties and functional groups (alcoholic and phenolic hydroxyl groups), LS is an attractive candidate for high-performance adsorbents among the biopolymer classes [23]. Chemical treatments, such as cross-linking and graft copolymerization, are required to make LS insoluble in water and boost their adsorption properties and improve their use for adsorbing cationic dyes and metal ions. For example, Yu et al. prepared a lignin-based hydrogel (LS-g-AA) by grafting acrylic acid (AA) on the backbone of LS in the presence of N, N'-methylene bis (acrylamide) (MBA) [24]. Shi et al. prepared a lignin-based composite (lignin-PEI) by cross-linking the lignin matrix with polyethyleneimine, which was then used to adsorb Ni2+ in polluted water [25].

Compared to biopolymer hydrogel, biopolymer/inorganic composite hydrogel exhibits outstanding thermal and mechanical properties [26]. Montmorillonite (MMT) is a layered aluminum silicate composed of octahedral aluminum and tetrahedral silica sheets [27]. It is commonly employed as an adsorbent to remove dyes and heavy metals [28]. The layer structure characteristics and the weak interaction forces between layers make MMT easily organised by the surfactants through an ion-exchange process, which ultimately results in organic montmorillonite (OMMT). As OMMT has a more pronounced layer spacing than MMT, it is easier to exfoliate as a nano-sheet during polymerization. Biopolymer/OMMT composites are highly effective in the removal of heavy metals, dyes, and pesticides from water [19, 29]. Currently, there are only a few reports exist on biopolymer/OMMT composite as an absorbent for wastewater remediation. For example, calcium alginate/OMMT composites can be potent in the rapid removal of anionic dyes from water due to their greater specific surface area and porosity [30]. Wu et al. reported a study on the synthesis of cellulose-g-poly(methyl methacrylate)/OMMT and its application in the removal of 2,4-dichlorophenol (2,4-DCP) from aqueous solution [31].

As mentioned above, LS, a natural biopolymer and OMMT, a clay mineral with a higher specific surface area and reactive –OH groups, are promising candidates for creating a cost-effective and environmentally friendly composite adsorbent. As a result, a new OMMT-containing hybrid hydrogel (LS/OMMT) was prepared, which was made by grafting acrylamide(AM) and acrylic acid (AA) onto LS. Then, LS/OMMT composite hydrogel was used to remove MB from wastewater. A study was conducted to investigate the effects of experimental factors (pH, contact time, initial MB concentration, and adsorbent dosage), adsorption kinetics, isotherms, and thermodynamics on the adsorption performance of LS/OMMT for MB and the adsorption mechanism was also postulated.

Experiment

Materials

AA, AM, MB, MBA, NaOH, and potassium persulfate (KPS) were all of analytical grade and purchased from Aladdin Reagent of China. Hydrochloric acid (HCl, 37 wt%) was purchased from Sinopharm Chemical Reagent Co., LS. OMMT was of chemically grade and purchased from Tianjin Kermel Chemical Reagent Co., Ltd. All other chemicals were purchased and used unless otherwise specified. Deionized water of high purity was utilized in the study.

Instruments

Fourier transform infrared spectroscopy (FT-IR) was recorded using a Nicolet is50 spectrophotometer in 400–4000 cm−1. X-ray diffraction (XRD) patterns were obtained using a Smartlab SE X-ray diffractometer with a Cu-K radiation source operating at 40 kV, 30 mA in the scanning range of 2θ = 2–30° and 1° min−1 scan rate. The surface morphology and chemical composition of LS/OMMT, LS, and OMMT samples were studied using a FlexSEM1000 scanning electron microscope (SEM) with an energy-dispersive X-ray spectroscopy (EDS) detector (IXRF Model 550i). Nestal TG209F1 instrument was used to perform thermal gravimetric analysis (TGA) in N2 from 20 to 900 °C at a heating rate of 10 °C min−1. Thermo Electron's VG ESCALAB 250Xi spectrometer was applied for X-ray photoelectron spectroscopy (XPS) with an Al K X-ray source (1486 eV). LS/OMMT porosity was studied by using the N2 adsorption/desorption isotherm at 77 K, which included BET (SA3100) specific surface area, pore size, and pore volume.

Preparation of LS/OMMT Composite Hydrogel

LS/OMMT composite hydrogel was prepared according to a similar procedure described previously [19]. Graft copolymerization of AA and AM onto LS in the presence of chemical cross-linker MBA, initiator KPS, and OMMT produced the LS/OMMT hydrogel. The standard procedure was as follows: LS (1.5 g), AA (6.0 g, partially neutralized by 10 mL of 6 mol L−1 sodium hydroxide solution), AM (3.0 g), and OMMT(1.2 g) were added to a 250 mL beaker and ultrasonically dispersed for 1 h at room temperature. The mixture was added to a four-neck flask with a mechanical stirrer, a reflux condenser, and a nitrogen line. After the addition of the initiator KPS (0.25 g) and the crosslinker MBA (0.75 g), the temperature was gradually increased to 70 °C and held for 2 h. The obtained products were soaked in deionized water for two days and the deionized water was changed periodically to remove homopolymers and unreacted monomers [32]. Finally, the composite hydrogel was dried at 60 °C, and the dried samples were ground for further testing. This method of preparing LS/OMMT composite hydrogel and the mechanism of formation is shown schematically in Fig. 1.

Fig. 1
figure 1

The preparation method and mechanism for forming the LS/OMMT composite hydrogel

Full size image

Adsorption Experiments

The effects of pH, adsorbent dosage, and initial MB concentration on dye adsorption were explored using batch adsorption experiments. In conclusion, a specific amount of LS/OMMT was added to 150 mL flasks and contacted 100 mL of the desired concentration of dye solution. The adsorbent was separated by filtration after shaking the mixture at 150 rpm at a regulated temperature. The change in absorbance of the maximum absorbance (665 nm) was monitored using a UV–vis spectrophotometer to determine the concentrations of the MB solution before and after adsorption. The adsorption capacity at equilibrium (Qe, mg g−1) as well as at time t (Qt, mg g−1) and removal ratio (RR) were calculated by the following equations:

$$ Q_{e} = \frac{{C_{0} - C_{e} }}{m}V $$
(1)
$$ Q_{t} = \frac{{C_{0} - C_{t} }}{m}V $$
(2)
$$ RR = \frac{{C_{0} - C_{e} }}{{C_{0} }} \times 100\% $$
(3)

where C0 (mg L−1), Ct (mg L−1), and Ce (mg L−1) are the concentration of MB at time 0, time t and equilibrium, respectively; m (g) is the amount of the adsorbent and V (L) is the volume of MB solution.

The effect of pH was studied within the range of 1.0 to 10.0 (adjusted with 0.1 mol L−1 HCl or NaOH) with a contact time of 140 min at 293.15 K and the C0 of 50 mg L−1. Adsorption isotherms were conducted at various C0 (50–600 mg L−1) and various temperatures (283.15 K, 293.15 K, 303.15 K, 313.15 K). Adsorption kinetics studies were performed with 0.1 g LS/OMMT in 200 mL of four different initial dye concentrations, i.e., 100 mg L−1, 200 mg L−1, 300 mg L−1,400 mg L−1 at pH 7.0 under 150 rpm. Samples were collected and centrifuged at 4000 rpm for 2 min at predetermined time intervals. Qt was calculated from Eq. (2).

The point of zero charges (pHpzc) of LS/OMMT were determined using the solid addition method [33]. The experiment was carried out in a series of 100 mL flares containing 50 mL 0.01 mol L−1 KCl. The pH was adjusted over a range of 1–11 using HCl (0.01 mol L−1) or NaOH (0.01 mol L−1). The pH of initial solutions was measured and noted as pH0. After a constant value of pH0 was reached, 0.15 g of the LS/OMMT was added to each flask. The suspensions were sealed and shaken at 150 rpm and 25 °C for 24 h, and the second pH noted as pHf was measured. A plot of ΔpH (ΔpH = pH0—pHf) vs. pH0 was made, and pHPZC was the point where the curve cuts the x-axis implying the point when pH0 = pHf.

Desorption Study and Reusability of Hydrogel

In order to recycle LS/OMMT, the method reported in the literature was adapted with few modifications [31]. LS/OMMT composite hydrogels with adsorbed MB (MB-LS/OMMT) was collected and transferred into deionized water for 1 h to remove un-desorbed MB and then dried in an oven 60 °C for 12 h for desorption experiments. The dried MB-LS/OMMT was then immersed in a solution of HCl (0.1 mol L−1) for 24 h while shaking at 300 rpm. Samples were obtained from the solution to track the amount of MB desorbed and then were treated with NaOH (0.1 mol L−1) solution to regenerate their binding sites after desorbing with HCl solution. Finally, the adsorbents were separated, washed with deionized water, and reused in the next cycle of the adsorption experiment. A total of five cycles of adsorption–desorption experiments were carried out. The amount of MB absorbed onto the LS/OMMT was determined and calculated by RR mentioned above.

Results and Discussion

Characterization

The LS/OMMT FT-IR spectra (Fig. 2) show that the absorption band at 3482 cm−1 is attributable to O–H and N–H stretching vibrations and the band at 2941 cm−1 is related to C–H stretching vibrations of –CH2 and –CH3 [34]. Aromatic skeleton vibration was responsible for the absorption peak at 1557 cm−1 and 1457 cm−1. The peak at 1217 cm−1 was assigned to the syringyl structure characteristic absorption peak from lignin [35, 36]. The sulfonic group's typical absorption peaks were 1044 cm−1 and 628 cm−1 [37]. Two new characteristic signals appear at at 1728 cm−1 and 1670 cm−1, respectively, which are attributable to the C=O stretching vibration of carboxylic acid and amide produced from AA and AM [38] compared to the FT-IR spectrum of LS. Also, the C–O (1160 cm−1) [39] stretching vibration peaks were stronger than that of LS, while the C–N stretching vibration and COO– symmetric stretching vibration exhibited absorption peaks at 1182 cm−1 and 1405 cm−1 [3, 40]. These results showed that the LS is effectively grafted with AA and AM. A peak at 1114 cm−1 for the antisymmetric stretching vibration of Si–O–Si overlaps with the asymmetric stretching vibration peak of the sulfonic group (1038 cm−1), which is more acute and shifts to a higher wavenumber than the comparable peak for OMMT. From the FT-IR spectra of LS/OMMT, the peak disappears at at 910 cm−1 (Al–OH bending vibration), suggesting that graft copolymerization happens on the surface and within OMMT. In addition, Si–O–Si symmetric stretching vibration and the Si–O–Al stretching vibration possess wavelengths of 800 cm−1 and 769 cm−1, respectively. Besides, Si–O–Al bending vibration absorption peak and Si–O–Si bending vibration are at 515 cm−1 and 466 cm−1 [3, 41], respectively. These results indicate that OMMT enters into the 3D network structure of the hydrogel.

Fig. 2
figure 2

The FT-IR spectra of LS/OMMT, LS, and OMMT

Full size image

The XRD patterns for LS/OMMT, LS, and OMMT are shown in Fig. 3. According to relevant literature sources [42, 43], the typical 001 basal reflection peak of MMT is at 2θ = 7.5°°with a d- alue of 11.77 Å. In OMMT, however, this peak was displaced to a lower angle at 2θ = 4.6° with a d value of 19.19 Å. This increase in d value supports the intercalation of surfactants. However, the 001 basal diffraction peak was not present in LS/OMMT, indicating that the LS or the polymer chains entered the OMMT layers during aqueous solution polymerization, destroying and exfoliating the original layered structure of OMMT [32].

Fig. 3
figure 3

XRD patterns of LS/OMMT, LS, and OMMT

Full size image

The TGA and DTG curves of LS/OMMT, LS, and OMMT, respectively, are shown in Fig. 4a and b. Four stages of deterioration of LS/OMMT can be observed in the two figures. The first stage, which occurred between 100 and 150 °C and resulted in a 7.5% weight loss, was caused by the physically adsorbed water for LS/OMMT [41]. This is equally responsible for the weight loss for LS (30–125 °C) and OMMT (30–161 °C). In the second stage, between 215 and 300 °C, the weight loss of the sample was about 5%, possibly because the small molecules of LS structure in LS/OMMT began to decompose [25]. Similarly, the tiny molecules of LS were thermally decomposed from 160 to 337 °C, resulting in a 35.6% weight loss. The predominant weight loss of LS/OMMT occurred in the third stage, from 302 to 510 °C, with a weight reduction of 38.6%. Structural degradation of LS and decomposition of acrylic acid and acrylamide polymers from LS/OMMT led to the third stage [4, 44]. In the fourth stage, LS/OMMT lost roughly 4.5% due to OH group breakdown in the OMMT structure at temperatures ranging from 650 to 735 °C [3]. Furthermore, the maximum decomposition temperature of L/SOMMT was 374 °C, which was higher than that of LS (272 °C). These findings suggest that adding OMMT to the polymerization process can improve the thermal stability of the hydrogel.

Fig. 4
figure 4

The TGA curves of LS/OMMT, LS, and OMMT (a); the DTG curves of LS/OMMT, LS, and OMMT (b)

Full size image

SEM and EDS analysis were used to study the surface morphology and chemical content of LS/OMMT, LS, and OMMT. Figure 5a–c shows the obtained SEM pictures, while Fig. 6a–c shows the EDS spectrum. Figure 5c showed that the OMMT appears to have a heterogeneous phase and layer structure. The EDS analysis reveals that organic surfactant had been incorporated into the layered structure of MMT, with the presence of C (25.8 wt%) and N (2.19 wt%) in Fig. 6c. The surface of LS/OMMT is significantly rougher compared to LS (Fig. 5b), with many irregular folds in Fig. 5a. LS/OMMT, on the other hand, revealed an uneven macro-porous structure with thinner pore walls. The EDS spectrum of LS/OMMT (Fig. 6a) shows that the LS/OMMT contains the elements Si (2.54 wt%), Al (0.82 wt%), and Fe (0.12 wt%), which further indicates that OMMT successfully enters into the hydrogel matrix. Similar results were reported by Wang et al. [3] and Karaca et al. [45].

Fig. 5
figure 5

The SEM of LS/OMMT (a), LS (b), and OMMT (c) at ×500 magnification

Full size image
Fig. 6
figure 6

The EDS spectrums of LS/OMMT (a), LS (b), and OMMT (c)

Full size image

The specific surface area and pore size distributions were measured by the N2 adsorption–desorption experiment at 77 K, and the results are shown in Fig. 7. The pore-size distribution curves of the LS/OMMT, LS, and OMMT are shown in the insert image. According to Fig. 7c, the isotherm of OMMT is type IV [46, 47], with clear hysteresis loops in the P/Po range of 0.5–1.0 according to the IUPAC classification, which was the main feature of the type H3 isotherm and ascribed to be typical of of layered materials [48]. The BET surface area (SBET) of OMMT is 62.97 m2 g−1, as displayed in Table 1, while the average pore diameter (Dave) is 6.206 nm. This result indicates that OMMT has a higher surface area compared to natural MMT reported in the literature [3] and exhibits a mesoporous pore structure [49]. According to Fig. 7b, the isotherm of LS should be attributed to the H1 type, which is typical of polymolecular adsorption [50]. High-pressure hysteresis loops are usually associated with spherical porous structures [51]. In addition, according to the image inserted in Fig. 7b, the pore size distribution of LS is mainly around 2 nm.

Fig. 7
figure 7

N2 adsorption–desorption curves and pore-size distributions for the LS/OMMT (a), LS (b), and OMMT (c)

Full size image
Table 1 The surface parameters of the LS/OMMT (a), LS (b), and OMMT (c)
Full size table

On the other hand, as seen in Fig. 7a, the isotherm of LS/OMMT is classified with type-IV isotherms [52, 53]. The relative pressure at pressures (P/Po > 0.5) displays a type H4 hysteresis loop [54], which suggests that the LS/OMMT possesses a mesoporous structure [55, 56]. According to Fig. 7a, the pore size distribution is concentrated at 3–10 nm, confirming that the LS/OMMT is mesoporous materials [52]. The uneven pore size distribution of LS/OMMT may be due to the random recombination of polymer monomer and the OMMT during synthesis [53].

The specific surface area values and textural properties are presented in Table 1. The SBET and Dave of LS are 3.136 m2 g−1 and 1.819 nm, respectively, while the SBET and Dave of LS/OMMT are 6.712 m2 g−1 and 3.854 nm, respectively. The above results indicate a substantial increase in surface area due to the incorporation of LS into the OMMT compared to LS. Moreover, the total pore volume (Vtot) of the LS/OMMT is also higher than that of LS. However, the lower SBET of the LS/OMMT compared to the OMMT supports the compact incorporation of LS into the OMMT, and consequently causes pore-blocking [57].

Effect of Adsorbent Dosage

The adsorbent dosage determines the adsorbent's ability to remove adsorbate. By changing the adsorbent dose from 0.2 to 0.9 g L−1, the effect of LS/OMMT dosage on MB removal was investigated. Figure 8 shows that when the adsorbent dosage was increased from 0.2 to 0.5 g L−1, the RR increased rapidly from 48.4% to 96.7%. Continue to increase the adsorbent dosage from 0.6 to 0.9 g L−1. RR increases slightly to 98.4% and then remains constant. However, the Qe drops from 121.1 to 54.7 mg g−1 when the dosage of adsorbent increased from 0.2 to 0.9 g L−1. The increase of the amount of adsorbent resulted in the increase of adsorption sites for dye adsorption and the the RR increased. On the other hand, the Qe per unit mass of adsorbent will decrease because the initial dye concentration was constant [58]. As a result, from an economic and efficiency point of view, the optimal adsorbent dosage for further research was determined to be 0.5 g L−1.

Fig. 8
figure 8

The effect of LS/OMMT dosage on adsorption of MB (initial MB concentration, 50 mg L−1; pH 7.0; temperature,293.15 K; contact time, 140 min)

Full size image

Effect of pH

The pH of the adsorption medium is generally considered to be a critical factor in general as it affects the properties of the adsorbent and adsorbate and the adsorption processes [59]. Figure 9 shows how the Qe of MB changes with pH. (1.0–10.0). The Qe increased as the pH went from 1.0 to 7.0 and then tended to level off as the pH went further up. The Qe of LS/OMMT is greatest when the pH was at 7.0. The protonation-deprotonation process of groups can describe this adsorption performance for RR of MB at variable pH [60]. Excess protons caused functional groups (hydroxyl, phenolic hydroxyl, methoxyl, sulfonic acid, amino) on LS/OMMT to be protonated at pH < pHpzc (6.56, as shown in insert image in Fig. 9) [38]. More positive potential of the LS/OMMT surface and the repulsion of the positively-charged MB prevented the adsorption of MB, leading to a decrease in Qe. In addition, more H+ compete with dye for binding sites, decreasing the number of active groups available and diminishing the Qe [48, 49]. At a pH > pHpzc, the surface potential of the LS/OMMT became negative due to deionization and electrostatic attraction occurred between the LS/OMMT and MB molecules, thereby enhancing the Qe. Finally, there is no change in Qe as the adsorption sites are saturated at pH 7–10.

Fig. 9
figure 9

The effect of pH on adsorption of MB by LS/OMMT (initial MB concentration, 50 mg L−1; absorbent dosage, 0.5 g L−1; temperature, 293.15 K; contact time 140 min)

Full size image

Effect of Initial Dye Concentration and Adsorption Isotherms

The relationship between the initial MB concentration and the Qe is crucial for optimizing the adsorption process and comprehending the adsorption behavior. MB solution was introduced into 50 mg of LS/OMMT at four different temperatures (283.15 K, 293.15 K, 303.15 K, and 313.15 K), with C0 of MB ranging from 50 to 600 mg L−1 at pH 7.0. As shown in Fig. 10a, the Qe improved dramatically when the C0 increased from 50 to 600 mg L−1 because a more significant initial concentration of dye accelerates the adsorption process. Subsequently, as the MB starting concentration was further increased, Qe slowly increased until it approached equilibrium. This phenomenon is a result of the active sites of the adsorbent being filled and balanced [60]. Furthermore, the Qe increased with rising temperature. LS/OMMT has a maximum Qe of 492.7 mg g−1 at 313.15 K. This demonstrates that the adsorption of MB on LS/OMMT is an endothermic process.

Fig. 10
figure 10

The effect of initial concentration on adsorption of MB by LS/OMMT (a) (pH 7.0; absorbent dosage, 0.5 g L−1; temperature, 283.15 K, 293.15 K, 303.15 K, and 313.15 K; contact time, 140 min)), adsorption isotherms: Langmuir (b), Freundlich (c) and Sips (d)

Full size image

Adsorption isotherms illustrate how adsorbents interact with the adsorbate and the link between Qe and Ce. Three adsorption isotherms, including the Langmuir [3], Freundlich [61], and Sips [62] isotherm models, were employed to analyze the experimental data and identify the most suitable isotherm mode.

The Langmuir adsorption isotherm model can be seen as below:

$$ Q_{e} = \frac{{Q_{{{\text{max}}}} K_{L} C_{{{e} }} }}{{1 + K_{L} C_{e} }} $$
(4)
$$ R_{{{L} }} = \frac{1}{{1 + K_{L} C_{m} }} $$
(5)

The Freundlich adsorption isotherm model:

$$ Q_{{{e} }} = K_{F} C_{e}^{1/n} $$
(6)

Sips isotherm adsorption isotherm model:

$$ Q_{e} = \frac{{Q_{{{S} }} K_{{{S} }} C_{e}^{N} }}{{1 + K_{{{S} }} C_{e}^{N} }} $$
(7)

where KL (L mg−1) is the Langmuir constant, Qmax (mg g−1) is the maximum adsorption capacity. Cm is the highest dye initial concentration, RL indicates the affinity of adsorbent to adsorbate. In case of RL = 0, the adsorption is irreversible; 0 < RL < 1, which is beneficial to adsorption; RL = 1, which indicates linear adsorption; RL > 1, which suggests unfavourable adsorption.n and KF (mg g−1) are the Freundlich constant. KS (L mg−1) is the Sips isotherm constant, N indicates the system heterogeneity when the value of N is 1. The above equation is reduced to the Langmuir adsorption isotherm. On the other hand, when KS or Ce approaches zero, the isotherm equation is reduced to Freundlich isotherm.

The relevant results and parameters are shown in Fig. 10b–d and Table 2. From Fig. 10b–d and Table 2, the equilibrium data fit better with the Langmuir model than others because the correlation constants (R2 = 0.9924, 0.9847, 0.9868, and 0.9982) are highest at four different temperatures. The calculated Qmax (409.1 mg g−1, 415.5 mg g−1, 451.5 mg g−1, and 489.7 mg g−1) approximates the experimental Qmax (374.4 mg g−1, 417.5 mg g−1, 463.4 mg g−1, and 492.7 mg g−1), implying the monolayer adsorption exhibits control over the adsorption process [4]. Moreover, the values of RL are between 0 and 1, demonstrating that the adsorption on the LS/OMMT is a favorable process. Additionally, the calculated value of the Freundlich constant n is greater than 1, which further illustrates the favorable condition for the adsorption and implys the monolayer adsorption exerts a controlling effect on the adsorption process.

Table 2 The isothermal adsorption model parameters
Full size table

Adsorption Thermodynamic Analysis

Temperature is a critical factor in the adsorption process as it affects the diffusion rate of adsorbate molecules and the adsorption capacity [2]. Furthermore, the corresponding thermodynamic parameters include the entropic change of adsorption (ΔH0), the Gibbs free energy (ΔG0), and the enthalpy change of adsorption (ΔS0), which can judge the absorption and heat release of the adsorption process, the spontaneity of the adsorption reaction and the degree of chaos in the adsorption process, respectively. As a result, experimental data collected at various temperatures were utilized to compute thermodynamic parameters such as ΔH0 (kJ mol−1), ΔG0 (kJ mol−1), and ΔS0 (J mol−1 K−1) by the following equation [62]:

$$ Lnk_{d} = \frac{{\Delta S^{{0}} }}{R} - \frac{{\Delta H^{0} }}{RT} $$
(8)
$$ \Delta G^{{0}} = - RTLnk_{d} $$
(9)
$$ k_{d} = \frac{{Q_{e} }}{{C_{e} }} $$
(10)

where Kd (L g−1) is an equilibrium constant, R represents the gas constant 8.314 (J mol−1 K−1) and T (K) is the absolute temperature. ΔH0 and ΔS0 were calculated from the slope and the intercept of the plot of lnKd versus 1/T, respectively (see Fig. 11).

Fig. 11
figure 11

The plot of lnKd versus 1/T

Full size image

As demonstrated in Table 3, ΔG0 are negative at all temperatures tested, showing that the adsorption is spontaneous [64]. Furthermore, ΔG0 decreases as the temperature increases, indicating that higher temperatures are more favourable for adsorption. A positive ΔS0 value confirmed the enhanced unpredictability at the solid/liquid interface during adsorption. The positive value of ΔH0 demonstrates the endothermic nature of the adsorption process, which is supported by the higher temperature adsorption capability. It is believed that LS/OMMT hydrogel is a polymer material with a three-dimensional network structure. This material shrinks at low temperature and expands at high temperature when it gets energy, that is to say, it can swell and form pores to boost dye adsorption [63]. In addition, high temperatures can increase the fluidity of dye molecules, enabling them to penetrate LS/OMMT gel layer [64]. In general, the ΔG° for physisorption ranges from − 20 to 0 kJ mol−1, whereas chemisorption ranges from − 80 to − 400 kJ mol−1 [45]. Therefore, it can be determined that the adsorption of MB on LS/OMMT is a physical process. Furthermore, similar results have been found for MB adsorption on lignin cross-linked composite [65].

Table 3 Thermodynamic parameter for MB adsorption on LS/OMMT
Full size table

Effect of Contact Time and Adsorption Kinetics

The effect of contact time on the adsorption of MB onto the adsorbent LS/OMMT is shown in Fig. 12a. The influence of contact time and starting dye concentrations on Qe was investigated by adding 100 mg of LS/OMMT to 200 mL MB solution (pH 7.0) at 293.15 K. For four different initial dye concentrations (100 mg L−1, 200 mg L−1, 300 mg L−1, and 400 mg L−1), the adsorption process was rapid during the first 40 min because of an abundance of adsorption sites on LS/OMMT. As the adsorption process progressed, the number of vacant adsorption sites on the outer surface reduced, slowing down the adsorption rate until equilibrium [61]. The adsorption equilibrium was established at about 80 min to 140 min, and the maximum Qe was 180.4 mg g−1, 314.2 mg g−1, 375.1 mg g−1, and 417.1 mg g−1 with initial MB concentrations of 100 mg L−1, 200 mg L−1, 300 mg L−1, and 400 mg L−1, respectively. Moreover, it was observed that there was a a faster adsorption equilibrium at lower C0 and a slower adsorption equilibrium at higher C0. This finding arises from the fact that there are many available binding sites in lower C0 than in higher C0 solutions [38].

Fig. 12
figure 12

The effect of contact time on adsorption of MB by LS/OMMT (a) (pH 7.0; absorbent dosage, 0.5 g L−1; temperature, 293.15 K), adsorption kinetics: Pseudo-first order (b), Pseudo-second order (c), Intraparticle diffusion (d)

Full size image

Adsorption kinetics is essential because the adsorption rate and mechanism can be derived from kinetic studies. In this study, the adsorption was treated using the Pseudo-first order [4], Pseudo-second order [4], Intraparticle diffusion [66]. Their equations are expressed as follows:

$$ Q_{t} = Q_{e} (1 - e^{{ - k_{{{{1}} }} t}} ) $$
(11)
$$ Q_{t} = \frac{{k_{{{{2}} }} Q_{e}^{2} t}}{{1 + k_{{{{2}} }} Q_{e} t}} $$
(12)
$$ Q_{t} = k_{p} t^{0.5} + C $$
(13)

where C is the Intraparticle diffusion constant. k1 (min−1), k2 (g mg−1 min−1/2), kp (mg g−1) are the Pseudo-first-order, Pseudo-second-order, and Intraparticle diffusion adsorption rate constant, respectively. The fitted data and parameter values are shown in Fig. 12b–d and Table 4, respectively. The correlation coefficient (R2) for the Pseudo-second order kinetic model is higher than than that of the Pseudo-first order kinetic model, and the Qe of the model is near the experimental data. Therefore, it can be concluded that the Pseudo-second-order adsorption model is more suitable for describing the adsorption kinetics of MB on LS/OMMT. The data in Fig. 12d show multi-linear plots, and all curves have three parts: (i) the film diffusion stage 67] (ii) the intraparticle diffusion [68] and (iii) the equilibrium adsorption process. The result indicate that intraparticle diffusion is involved in the adsorption process [17]. Furthermore, the straight lines do not pass through the origin (C ≠ 0), indicating that intraparticle diffusion is not the only rate-limiting step, and the film diffusion can simultaneously be interpreted in the adsorption kinetics [20, 21].

Table 4 Kinetic parameters for the adsorption
Full size table

Adsorbent Comparison

In order to evaluate the adsorption performance of LS/OMMT toward MB, its maximum Qe was compared with some other reported ones in the literature, as summarized in Table 5. It is clear that LS/OMMT is a promising candidate for the removal of MB removal from wastewater, in view of the maximum Qe and pH values.

Table 5 Comparison of the MB maximum adsorption capacities on LS/OMMT
Full size table

Reusability Study

One of the essential properties of adsorbents used to remove dyes from wastewater is their ability to maintain their adsorption capacity over multiple adsorption–desorption cycles [25]. Figure 13 shows that the RR can retain more than 84% after five cycles. This desorption behavior can be detected because H+ ions rapidly replace the cationic dye molecules in acidic conditions, showing that the MB is attached to the hydrogel by electrostatic contact. Since the H+ ions linked to the binding point of the hydrogel are replaced by OH− present in the NaOH solution, the desorbed materials are highly effective in reabsorbing the MB after being regenerated. As can be seen from the above results, LS/OMMT exhibits good reusability.

Fig. 13
figure 13

Adsorption desorption cycles for LS/OMMT(adsorption conditions: initial MB concentration, 50 mg L−1; pH 7.0; absorbent dosage, 0.5 g L−1; temperature, 293.15 K; contact time, 140 min; desorption conditions: temperature, 293.15 K; contact time, 24 h)

Full size image

Adsorption Mechanism

FT-IR and XPS were carried out to explore the mechanism of MB adsorption on LS/OMMT. FT-IR spectra of LS/OMMT before and after adsorbing MB (MB-LS/OMMT) were recorded to understand the adsorption mechanism. The results are shown in Fig. 14a. After MB adsorption, three additional peaks of MB are detected around 1593 cm−1, 1332 cm−1, and 1355 cm−1, which are connected to the bending vibrations of –CH3, the stretching vibrations of (CH3)2N–, and the stretching vibration of C=N of the MB heterocycle, respectively [69]. Additionally, the peaks at 1729 cm−1 (–COOH) and 1114 cm−1 (–SO3) shift to 1716 cm−1 and 1119 cm−1 after adsorption, implying the involvement of carboxyl and sulfonic acid groups in the adsorption process. At 3410 cm−1, peaks were overlapping –NH/–OH stretching vibration grew more substantial and broader. It can be explained by hydrogen-bonding interactions between the surface groups of LS/OMMT and MB [4].

Fig. 14
figure 14

FT-IR of LS/OMMT before and after adsorption (MB-LS/OMMT) (a), wide scan XPS spectrum of LS/OMMT before and after adsorption MB (b), high-resolution XPS spectra of N1s, O1s and C1s (c, d, e) for LS/OMMT before and after MB adsorption

Full size image

To gain further insight into the interactions between MB and LS/OMMT, XPS studies of LS/OMMT and MB-LS/OMMT were undertaken. As shown in Fig. 14b, photoelectron peaks exist at 73.50 (Al 2p) and 120.58 (Si 2p) eV, indicating that the OMMT was successfully embedded into the hydrogel matrix.

The high-resolution XPS spectra of N1s spectra of LS/OMMT can be fit by three peaks with binding energies of 398.67 eV (N–H), 399.37 eV (H–N–H), and 399.97 eV (O=C–N), respectively. According to the literature [70], the positive charge in MB molecules is uniformly distributed as a result of the delocalized π bond and the nitrogen atoms in the MB molecule are in the same chemical state. Therefore, the high-resolution N1s scan spectrum of the MB molecule is symmetrical without splitting, and the binding energy (BE) peak is visible at 399.1 eV. Following MB adsorption, a new peak with binding energies of 402.03 eV (HN+) appears in Fig. 14c, which is more robust than the BE of N1s of NH4Cl (401.50 eV). It can be inferred that the nitrogen atom in the MB molecule was involved in MB adsorption via hydrogen bonding and electrostatic attraction. When MB is adsorbed on the surface of LS/OMMT, the positive charge of MB is concentrated on this nitrogen atom, and this positive charge must be on the nitrogen atom of dimethylamino groups to retain the most stable MB molecule [71].

The high-resolution XPS spectra of C1s and O1s for LS/OMMT are shown in Fig. 14d and e. The fitted O1s XPS spectra are shown in Fig. 14d with peaks at 529.98 eV, 530.52 eV, 531.12 eV, 531.55 eV, 532.40 eV, and 533.56 eV, respectively, corresponding to O–H, O–Al, O–Si, C–O–C, O=C, and COOH [19]. The peaks of C1s are at 283.63 eV, 284.14 eV, 285.20 eV, 286.21 eV, 287.64 eV, and 288.41 eV, corresponding to C=C, C–C/C-H, C-O/C-N, C–O–C, C=O, and COOH. After MB adsorption, the peaks of O–H, O-C-O, and COOH shift to 530.21 eV, 531.87 eV, and 533.81 eV in Fig. 14d and shift to 285.39 eV, 286.37 eV, and 288.74 eV [25, 72] in Fig. 14e, implying that the functional groups mentioned above are involved in the adsorption process. In conclusion, the proposed uptake mechanism is shown in Fig. 15.

Fig. 15
figure 15

Schematic illustration of the interaction mechanism between MB and LS/OMMT

Full size image

Conclusions

A new lignin-based hybrid hydrogel (LS/OMMT) was prepared by free radical polymerization and served as an adsorbent to remove MB from aqueous solutions. Under the optimized adsorption conditions, the adsorbent exhibited an adsorption capacity of 492.7 mg g−1 at 313.15 K and pH 7.0. Adsorption of MB from the LS/OMMT could be easily achieved by electrostatic attraction and hydrogen bonding. In addition, desorption studies predicted that the hydrogel could preserve relatively high reusability. This study demonstrates that LS/OMMT has emerged as a promising adsorbent for the removal of dyes from aqueous solutions.