Introduction

Some natural biomolecules can help living organisms in detoxifying of heavy metals and be used in many applications such as heavy metals immobilization, detection, and prevention of pollution events (Bagbi, Sarswat, Mohan, Pandey, & Solanki, 2017; Bhattacharyya & Sen Gupta, 2008; Vandenbossche, Jimenez, Casetta, & Traisnel, 2015). Some natural chelating agents, including glutathione, phytochelatin, and metallothionein contain constant amino acids. Cysteine, tyrosine, and histidine, possessing 3 potential binding sites for heavy metals uptake, can be used for water purification (De Santana et al., 2010; Mallakpour & Dinari, 2011; Petra, Billik, & Komadel, 2015; Vandenbossche et al., 2015; Pires, Juźków, & Pinto, 2018).

l-cysteine is a physiologically active and sulfur-containing amino acid with an amino group and a sulfhydryl group. Among the proteinogenic amino acids, cysteine may be the only amino acid containing a thiol group on the side chain and has the binding site for cell detoxification (Vandenbossche et al., 2015). Cysteine could be used in the synthesis of MoS2/graphene composite with excellent electrochemical properties, an electrochemical sensor for heavy metal ions, in the nonwoven geotextile for the removal of divalent heavy metal ions, and in the prebiotic chemistry to reduce Fe3+ to Fe2+ (Chang & Chen, 2011; De Santana et al., 2010; Vandenbossche et al., 2014; Zhou et al., 2017). Copper (II) and the thiol of cysteine form a ring structure in which the sulfur on cysteine acts as a bridging ligand (Dokken, Parsons, McClure, & Gardea-Torresdey, 2009). Due to the amino, thiol and carboxyl group, cysteine can easily interact with heavy metal and is regarded as an interesting biomolecule in the field of metal capture.

Clay and clay minerals, in its natural and modified forms, are widely used for the sorption of heavy metals in aqueous environments (Uddin, 2017). Particularly the montmorillonite could be modified with organic molecules and used as a carrier for organic molecules in the environmental applications (Bergaya & Lagaly, 2013; Keng, Lee, Ha, Hung, & Ong, 2014). Compared with kaolinite and modified kaolinite, montmorillonite and its modified form have higher metal adsorption capacity (Bhattacharyya & Sen Gupta, 2008; Padilla-Ortega, Leyva-Ramos, & Flores-Cano, 2013). With the development of nanotechnology, as a low-cost adsorbent, the adaptability of montmorillonite composites and their high removal ability to various pollutants have attracted more and more attention (Pandey, 2017; Santhosh et al., 2016).

The activity of the thiol and amino group in the amino acid made l-cysteine-montmorillonite composite particularly useful in the fields of antidote of heavy metals, for example, stabilization (Fan, Li, Zhou, & Liu, 2016; Jia et al., 2016), detection (Xiao et al., 2015), removal, and adsorption (Bagbi et al., 2017; Galhoum et al., 2015) of metals. The modification of montmorillonite by amino acids or the biomolecular carried by montmorillonite is often achieved by intercalation and make those novel clays organophilic and biocompatible (Mallakpour & Dinari, 2011; Petra et al., 2015). Amine group modified clay presented greater adsorption capacity for carbon dioxide than the initial clay (Pires et al., 2018), and the adsorption of l-cysteine–modified montmorillonite–immobilized alginate was demonstrated as 100, 111, and 125 (mg/g) for Cu(II), Pb(II), and Ni(II), respectively (Mittal, Ahmad, & Hasan, 2016).

The synthesis of the montmorillonite-cysteine composites was depended on the intercalation of the cysteine into the interlayer of the montmorillonite, and the thiol and/or amino group made the stabilization of the intercalation (Ramos & Huertas, 2013). In the bonding between cysteine and montmorillonite, the amount of cysteine adsorbed by montmorillonite depends on the pH value, and cysteine replaces the water molecule in the middle layer of montmorillonite molecule (El Adraa et al., 2017). As well known, the montmorillonite could adsorb the positive charged metals through electrostatic attraction, while the amino and thiol on the cysteine could bond the covalent metals by complex adsorption. Whether the consumption of amino and sulfhydryl groups during the intercalation affect the adsorption of heavy metals on the composites? The interaction between the cysteine and the montmorillonite, and the effect of the components on the metal adsorption, was less discussed.

This paper reports the cysteine adsorption on/in montmorillonite and the Cd(II) adsorption on the composites synthesized with cysteine and montmorillonite at various pH. Due to the pH dependency of the protonation of amine groups, the adsorption of cysteine and Cd(II) at various pH were discussed. X-ray diffraction patterns were recorded to discuss the basal spacing (d001) changes of the composites. The characteristic bands of cysteine after the intercalation were analyzed with the Fourier transform infrared spectroscopy method. The study of the interaction between amino acid and montmorillonite is helpful to further understanding on its mechanism of action and provide an important theoretical basis for the application of the bonding of metals and metalloids. In addition, the interaction mechanisms of the composite are also helpful for the medicine application.

Materials and methods

Materials

The montmorillonite supplied by Sanding Tech. Co. (Zhejiang, China) was saturated with sodium chloride solution and washed with deionized water until no Cl− was detected. The sediment after the centrifuge was freeze dried and stored in the refrigerator for use as sodium montmorillonite. The cation exchange capacity of the montmorillonite was 76 cmolc/kg as measured by the Ca-Mg exchange method. The l-cysteine (MW = 121) and other reagents were of analytical grade and purchased from Sinopharm Chemical Co. (Shanghai, China). The deionized water used in the experiments had a resistivity of 18.25 MΩ.

Methods

Adsorption of cysteine on the montmorillonite

One gram sodium montmorillonite was sonicated for 1 h in 200 mL DI water and was continuously magnetic stirred for 12 h to obtain a 5-mg/mL montmorillonite dispersion. The pH of the mixtures of 5-mL dispersion and 50 mL 50 mg/l-cysteine solution were adjusted to 2.5–8, respectively, with 0.01–1 mol/L HCl and NaOH. The mixtures were magnetically stirred at 600 rpm for 30 min and then centrifuged at 5000 rpm for 30 min. The 5-ml supernatants were added with 1 ml 2% ninhydrin solution (1 g ninhydrin powder was blended with 50 ml DI water and mixed with 40 mg SnCl2 before the supernatant was used for the measurement) and 4 ml KH2PO4-K2HPO4 buffer solution (pH 8) into tubes, respectively, and then in boiling water bath for 15 min (Elgubbi & Mlitan, 2015; Gaitonde, 1967). The samples were cooled to room temperature and settled for 15 min; the cysteine concentration of the samples was measured at 570 nm on the UV spectrophotometer (UV1800, Jinghua Technology Instrument Co., Ltd., Shanghai, China). The adsorption amount was calculated according to the difference between the concentration of the initial and equilibrium in the supernatant. In the following experiments, the solutions of cysteine were stored in brown bottles; the treatments were conducted in the triple at room temperature with unless otherwise stated.

Five milliliters of the 5 mg/mL montmorillonite dispersion was added with 50-ml cysteine solutions of various concentrations into the beakers, and the pH of the samples was adjusted to 4.5, respectively. The initial concentration of cysteine was of 9.04–178.59 mg/L, respectively. The equilibrium concentration of the cysteine in the supernatants was measured same as above. The adsorption amounts were fitted with Langmuir and Freundlich isotherm models (Freundlich, 1906; Langmuir, 1916).

The linear form of Langmuir is:

$$ \frac{1}{q_e}=\frac{1}{q_{\mathrm{ml}}}+\frac{1}{b{q}_{\mathrm{ml}}{C}_e} $$
(1)

where qe is the amount of adsorption at equilibrium, mg/g; qml is the theoretical maximum adsorption, mg/g; b is the binding site between the metal and the adsorbent. The affinity of the point, which can be used to calculate the dimensionless separation factor; Ce is the equilibrium concentration of the metal in the supernatant, mg/L.

The linear form of Freundlich is:

$$ \log {q}_e=\log {K}_F+\frac{1}{n}\log {C}_e $$
(2)

where qe and Ce are consistent with the meaning of Formula 1; KF and 1/n are empirical constants, respectively representing relative adsorption capacity and strength associated with metal affinity.

The 5-ml dispersion of montmorillonite and the 50 ml 50 mg/L solution of cysteine were mixed at pH 4.5 and stirred at 600 rpm for 1–50 min, respectively, and centrifuged at 5000 rpm for 30 min. The concentration of cysteine in supernatants was measured same as above, and the data were fitted with pseudo first-order, pseudo second-order, and intraparticle diffusion model.

According to the Lagergren model (Ho, 2004; Lagergren & Bergström, 1999; Sen Gupta and Bhattacharyya 2011), the pseudo first-order dynamic equation has a linearized form:

$$ \log \left({q}_e-{q}_t\right)=\log {q}_e-\frac{k_1t}{2.303} $$
(3)

where qe and qt are the amount of adsorption at equilibrium and time t, mg/g; t is the reaction time, min; k1 is the kinetic constant, min−1.

The pseudo second-order dynamics are given by:

$$ \frac{t}{q_t}=\frac{1}{k_2{q}_e^2}+\frac{t}{q_e} $$
(4)

where t, qe and qt are consistent with the meaning of Formula 3; k2 is the constant for pseudo second-order, g/mg min.

The equation for the intraparticle diffusion model is:

$$ {q}_t={k}_i{t}^{0.5}+{C}_i $$
(5)

where t and qt are consistent with the meaning of Formula 3; ki is the rate constant for controlled adsorption, mg/g min1.5; Ci is the relative thickness of the boundary layer.

Characterization of the cysteine-montmorillonite composites

The powder samples of cysteine-montmorillonite composites synthesized at pH 2.4–6.5 were freeze-dried and grounded to pass through 140 mesh sieve. The Bruker D8 advance diffractometer was operated using Cu-Kα radiation at a tube voltage of 40 kV and a current of 40 mA with 5 s counting per 0.05°. The X-ray diffraction (XRD) patterns were recorded with the powder in 5–35° (2θ). Fourier transform infrared spectroscopy (FTIR) (Spectrum 100, Perkin-Elmer) was used in transmission mode with the mixtures of the samples and KBr. Each spectrum was average of the 64 times scans with a resolution of 2 cm−1.

The adsorption of Cd(II) on the composites

The composites were synthesized with 0.1 g montmorillonite and 50 ml 20 mg/L cysteine at pH 2.3 and 7.8, respectively, and centrifuged at 5000 rpm for 30 min. The sediments were freeze-dried and grounded for the adsorption experiments of Cd(II).

0.1-g composites were dispersed in 100 ml DI water and mixed with 50 ml 6 mg/L Cd(NO3)2 solutions. The pH of the mixtures was adjusted to 3.0, 3.5, 4.0, 4.5, and 5.0, respectively, and the mixtures were magnetic stirred at 400 rpm for 5 h. The supernatants after 10 min of 10,000 rpm centrifuge were diluted various times for the measurement on the atom adsorption spectrometer (TAS, Beijing Puxi General Apparatus Ltd., Beijing, China). The adsorption of Cd(II) was calculated by the difference between the initial and final concentration of the supernatant.

Results and discussion

The effect of pH on the cysteine adsorption by the montmorillonite

As shown in Fig. 1, the adsorption amount of cysteine on the montmorillonite changed significantly with pH change (Duncan, P < 0.01). The adsorption amounts at pH 2.4–4.6 (53.86–57.87 mg/g) were greater than that at pH 5.4–7.8 (46.17–49.16 mg/g). In the range of pH 2.4–4.6, the adsorption amount of cysteine on montmorillonite decreased with the increase of pH, and reached a maximum of 57.87 mg/g at pH 2.4; in the range of pH 5.4–7.8, the adsorption amount increased with the increasing pH, and the lowest amount of the adsorption was 46.17 mg/g when pH was 7.8.

Fig. 1
figure 1

The adsorbed cysteine versus the change of pH

Full size image

Given the pI 5.01 of the l-cysteine (Dashman & Stotzky, 1982), the reason for the obvious adsorption decreasing from pH 4.6 to 5.4 might be the change of the surface charge on the cysteine: the ZPC of the montmorillonite was about pH 2–3 (Langmuir, 1992), and montmorillonite had negative charges on the surface, so the cysteine with positive charge was more steadily to be adsorbed on the interlayer. While the isoelectric points of –NH2, –COOH, and –SH on the cysteine were 10.3, 2.0, and 8.4 (Sillero & Ribeiro, 1989), respectively, indicated that the adsorption on montmorillonite greatly dominated by bonding with –SH and/or –NH2. The electrostatic adsorption enhanced with decreasing pH was significantly stronger than the competitive adsorption caused by the increasing concentration hydrogen ion.

The isotherm adsorption of cysteine on the montmorillonite

Within the equilibrium concentration range of 0–106.34 mg/L, the amount of cysteine adsorption increased with the increasing concentration and reached the maximum of 161.81 mg/g at the equilibrium concentration 106.34 mg/L (Fig. 2). The adsorption data fitted with the Langmuir equation (R2 = 0.948) better than that with Freundlich (R2 = 0.901) (Ahmad, Bhat, & Buang, 2018; Ngah & Fatinathan, 2010).

Fig. 2
figure 2

The equilibrium adsorption of cysteine on the montmorillonite and the fitting with Freundlich and Langmuir modes

Full size image

The molecular weight of cysteine was 121 g/mol and the montmorillonite had about 0.76 mmol/g negative charge on the surface (from the cation exchange capacity). Based on the balance of the charge, the montmorillonite could adsorb cysteine no more than 92 mg/g through the electrostatic adsorption. The extensive amount of the adsorption might be due to the physical sorption in the interlayer of the montmorillonite and/or complexation reaction between groups of cysteine and montmorillonite.

The kinetics adsorption of cysteine on the montmorillonite

The adsorption of cysteine at pH 4.5 reached the equilibrium amount at 55.24 mg/g in 1 min (Fig. 3a), and then the amount of the adsorption was stable at about 55.21–57.78 mg/g. The adsorption data fitted pseudo first-order (Fig. 3a), pseudo second-order model (Fig. 3b), and intraparticle diffusion model (Fig. 3a) showed R2 = 0.9977, 0.9999, and 0.7903, respectively. Those indicated not only the physical but also the chemical interaction during the adsorption. Besides, the intraparticle diffusion was not a limiting phase in the adsorption (Jabli, Gamha, Sebeia, & Hamdaoui, 2017; Sebeia, Jabli, & Ghith, 2019; Swayampakula, Boddu, Nadavala, & Abburi, 2009).

Fig. 3
figure 3

The kinetics adsorption and the fitting with models. a The kinetics adsorption and the fitting with pseudo-first and intraparticle diffusion models; b the fitting with pseudo-second model)

Full size image

The characterization of the composites synthesized at different pH

With the intercalation of cysteine at various pH (Fig. 4a), the d001 of the montmorillonite increased from 1.2267 to 1.4967 and 1.4718 nm, respectively (Fig. 4b). The d001 increased with 0.27 nm of composites synthesized at pH 2.4, 3.5, and 4.3, while 0.2451 nm of the composites synthesized at pH 4.5 and 6.5. It appeared that the lower pH led to more cysteine intercalation, which was consistent with the former discussion that the lower pH causes more adsorption of cysteine.

Fig. 4
figure 4

The XRD patterns of the composites of cysteine and montmorillonite (synthesized at pH 6.5, 4.5, 4.3, 3.5, and 2.4) a 2 theta from 5 to 35; b 2 theta from 5 to 9

Full size image

El Adraa et al. (2017) demonstrated that the two possible orientations were due to the pillar bonding conducted by the –NH3+ and –SH/–COO−, respectively, and the difference was 0.28 nm. Figure 4b illustrated the difference was 0.0249 nm, which far smaller than 0.28 nm, indicated there was no enough evidence for the two orientations of the cysteine in the interlayer of montmorillonite. The previous discussion showed that cysteine had about 68 mg/g (the difference of 160 and 92 mg/g) retained in the interlayer of montmorillonite, the 0.0249 nm might due to the excessive cysteine in the interlayer of montmorillonite through physical adsorption at lower pH conditions (< 4.5).

As shown in Fig. 5a and b, 3626 cm−1 (structural O–H stretching), 1643 cm−1 (O–H bending), and 1045 cm−1 (Si–O stretching) were ascribed to that of montmorillonite. 1424 (–CH2 bending), 1386 (–CO2 stretching), 1348 (–NH3 bending), 1298 (–CH2 wag vibration), 1196 (–CH2 twist vibration) cm−1 were ascribed to that of cysteine (Pawlukojć, Leciejewicz, Ramirez-Cuesta, & Nowicka-Scheibe, 2005). The spectra of composites showed part of characteristic bands of both montmorillonite and cysteine. The bands at 1424, 1348, and 1298 cm−1 of cysteine disappeared in the spectra of composites, and the band at 1386 shifted to 1398 cm−1, indicated the consumption of –NH3 and the effect of carboxyl reaction on CH2 at C2. Besides, there was no evidence to prove the existence of complexation.

Fig. 5
figure 5

The spectra of the montmorillonite, cysteine, and composites synthesized at various pH (the wavenumber from a 4000–700 cm−1; b 2200–1100 cm−1)

Full size image

The adsorption of Cd(II) on the composites at various pH

The adsorption amounts of Cd(II) on the montmorillonite(Mt), composites synthesized with montmorillonite and cysteine at pH 2.3 and 7.8 (Com-2.3 and Com-7.8, respectively) all increased with the pH from 3.0 to 5.0(Fig. 6). The adsorbed masses of Cd(II) at pH 3.0 were 2.54, 2.03, and 2.31 mg/g on the Com-2.3, Com-7.8, and montmorillonite, respectively. When the pH were 3.5, 4.0, and 4.5, the differences of the amounts were obvious: the adsorption amounts on Com-2.3 (4.89, 11.78, 17.58 mg/g, respectively) were greater than that on the Com-7.8 (3.18, 7.24, 13.15 mg/g, respectively) and montmorillonite(4.45, 8.53, 15.12 mg/g, respectively), and the adsorption on Com-7.8 were smaller than those on montmorillonite, which indicated the interlayer sites of montmorillonite were occupied by cysteine. At pH 5.0, the adsorption on Com-2.3, Com-7.8, and montmorillonite were 22.96, 17.15, and 17.24 mg/g, respectively.

Fig. 6
figure 6

The adsorption amount of Cd(II) on the composites versus the change of pH

Full size image

Those demonstrated the composite synthesized at pH 7.8 had not improved the adsorption amount of Cd(II) compared with that on montmorillonite, i.e., the composite synthesizes at pH 7.8 had the intercalation of the cysteine in the montmorillonite, but the Cd(II) adsorption sites on both montmorillonite and cysteine were partly consumed in the interaction between the cysteine and montmorillonite. Besides, the composite synthesized at pH 2.3, as discussed above, had more cysteine adsorbed on/in montmorillonite and led more active sites of cysteine for the bonding of metals.

Meanwhile, the cysteine-based composites showed great differences in the adsorption of heavy metal cations (Table 1). Such as the adsorption of Cd(II), the capacities were 17.24 (present work), 25.29 (El Adraa et al., 2017), 59.01 (Faghihian & Nejati-Yazdinejad, 2009), and 64.35 mg/g (Fan et al., 2016). The lowest adsorption amount of Pb(II) was 18.8 mg/g (Bagbi et al., 2017), and the highest can reach 459.33 mg/g (Zou, Yin, Zhao, Chen, & Dong, 2015). This made it necessary to further studying on the bonding of cysteine and the carrier.

Table 1 The comparison of the metal adsorption by the cysteine based adsorbents
Full size table

Conclusions

From the adsorption experiments of cysteine on the montmorillonite and the characteristics and metal adsorption properties of the composites, the composites showed that the characters were dominated by the pH during the synthesis. Lower pH (< 4.5) was beneficial to the loading of cysteine on the montmorillonite and the extensive adsorption of cysteine might depend on the physical adsorption. The extensive adsorption led a slight increase of the interlayer of composite and greater adsorption amount for the Cd(II).