3.5. Possible mechanism of Cu(II) and Pb(II) ions adsorption onto Thymus serpyllum L.
The investigation of adsorption mechanism is a very important issue with the potential to predict and control the adsorption process. Therefore, the available advanced methods like SEM/EDX, TGA, FT-IR, XPS and ζ-potential measurements were utilized to estimate the removal mechanism during the adsorption of Cu(II) and Pb(II) ions on the SER surface. Previous studies have revealed that the elimination of these contaminants can be realised by different processes, among which the complexation with surface carboxylic groups, the chelation with nitrogen moieties with formation of a coordinate bond between the lone pair of nitrogen and d-orbitals of heavy metals and the ion exchange are the most common (Mahaninia et al. 2015; Forgionny et al. 2021; Xu et al. 2021).
In order to prove the adsorption process as well as to estimate the distribution of the adsorbed ions on the biosorbent surface, the SEM/EDX observations were done. As illustrated in Fig. 3, the adsorbed Cu(II) and Pb(II) ions are evenly distributed on the SER surface in the form of precipitates, as small spheres and particles with indefinite shape for Cu and Pb adsorption, respectively (Fig. 3c and f). The EDS mapping has shown the inhomogeneous distribution of Cu and Pb elements (Fig. 3b and e). The amount of lead in the selected region in Fig. 3e is much higher than that of Cu in Fig. 3b.
To determine the nature of the precipitates, the XRD measurements of the metal-laden adsorbents were performed. However, no crystalline phases could be identified on the diffractograms (see Figure S3 in Supplementary information). Hence, it can be suggested that the available amount of crystalline phase cannot be detected because of the interfering effect of the plant matrix (this might be the reason in the case of Cu(II) ions adsorption) or because the precipitates most probably have an amorphous nature (this might be the case of Pb(II) ions adsorption).
To better understand the processes present during the adsorption of Cu(II) and Pb(II) ions on SER surface, the thermogravimetric analysis of the samples was performed. Figure 4 represents the thermogravimetric curves of TG-DTG/DTA analysis of the samples. As can be seen from the Fig. 4a, there are three exothermic peaks at 90°C, 324°C and 480°C on the thermogram of native SER plant. The mass loss at 90°C includes the loss of moisture present in the sample (~ 10.6 wt.%). Next, the fast weight losses at 324°C (~ 53.1 wt.%). and 480°C (~ 33.5 wt.%). were detected. These two processes can be considered as the decomposition of the chemical groups on the SER surface. According to (Shafeeyan et al. 2010), the first exothermic peak can be related to the decomposition of the carboxylic groups and the second peak can be assigned to the decomposition of the N-containing chemical moieties. The residual mass after the complete burning of the SER plant was near 0.8 wt.%.
In the case of SER sample loaded with Cu(II) ions (Cu-SER sample) in Fig. 4b, the peaks that are related to the carboxylic and N-containing surface groups are shifted on thermogravimetric diagram. The DTA peak related to carboxylic surface groups is shifted to 310°C and the peak associated with N-containing groups is moved to 413°C. This fact indicates the formation of chemical bonds between the Cu(II) ions and these moieties. Moreover, the additional small endothermic peak at 1009°C was detected. It can be suggested that this peak corresponds to the Cu2+ → Cu + reduction process or melting of metallic copper (Luiz and Nunes 2020). The residual mass after burning was near 5.9 wt.%. The similar situation is observed for Pb-SER sample (SER sample loaded with Pb(II) ions) in Fig. 4c. The peaks that are assigned to the carboxylic and N-containing surface groups are shifted on thermogravimetric diagram: for COOH groups from 324°C to 350°C and for N-containing groups from 480°C to 439°C. The mass of unburned residue was ~ 11 wt.%, that was much higher than obtained after complete burning of the raw SER plant adsorbent and twice as high as after the adsorption of Cu.
To shed more light on the functional groups potentially involved in the adsorption process, the FT-IR spectra of native SER adsorbent and the Cu-SER and Pb-SER samples were recorded (see Fig. 5). As can be seen, the stretching vibrations of hydroxyl surface groups in native SER plant (ν(-OH)bonded at 3322 cm− 1) are shifted to ν(-OH)bonded at 3290 cm− 1 and 3274 cm− 1 in Pb-SER and Cu-SER samples, respectively. The surface carboxylic groups possess their stretching vibrations at ν(-C = O) = 1730 cm− 1 and ν(-C-O) = 1510 cm− 1 in bare SER plant adsorbent. A shift to ν(-C = O) = 1735 cm− 1 and no shift for ν(-C-O) = 1508 cm− 1 in the Pb-SER sample confirms the participation of surface carboxylic groups in the elimination of Pb(II) ions. There is no shift in the stretching vibrations of ν(-C = O) and ν(-C-O) for Cu-SER sample which demonstrates the absence of interactions between these moieties and the Cu(II) ions. In the case of N-containing surface groups, the large shift in bending vibrations of aminogroups (δ(NH2) = 1600cm− 1, 1629 cm− 1 and 1614 cm− 1 for SER adsorbent and Pb-SER and Cu-SER samples, respectively) can be observed. The bending vibrations of hydroxyl surface groups in the native SER adsorbent δ(C-OH) at 1024 cm− 1 can be related to plant polysaccharide components and are shifted during the adsorption: for Pb-SER sample δ(C-OH) = 1014 cm− 1 and for Cu-SER sample δ(C-OH) = 1016 cm− 1. Therefore, this verifies the interactions between the hydroxyl moieties of plant components and the Pb(II) and Cu(II) ions. To sum up the observations from FT-IR, the interactions of carboxylic, hydroxyl and amino surface groups play an important role in the adsorptive removal of targeted heavy metals (Lu et al. 2012).
To investigate the surface chemistry of the pure SER and the corresponding metal-laden samples after the adsorption, XPS method was applied. Namely, the XPS spectra of C 1s,O 1s, Pb 4f and Cu 2p have been recorded (Fig. 6) and the changes of binding energies of target elements are shown in Table S2 in the Supplementary information.
The surface chemical composition of the SER biosorbent powder before and after the ionic adsorption was revealed by the XPS spectrum in Fig. 6a, where the carbon-based compounds of the organic SER plant and the adsorbed Cu and Pb peaks are present in the corresponding loaded samples. The bare SER plant showed the presence of impurities such as fluorine and silicon, which were washed out after the adsorption experiments. Despite the presence of the mentioned impurities, the SER biosorbent plant revealed its organic nature by the characteristic C 1s and O 1s peaks in Fig. 6b and Fig. 6c, respectively.
The core-level spectra decomposition of the C 1s (Fig. 6b) for pure SER plant was fitted in 4 main peaks positioned at 284.7 eV, 286.4 eV, 287.9 eV, and 288.9 eV, which are generally assigned to C-C, organic -C-O-, -C = O, and -C-F/COO species, respectively (Smith et al. 2016). Similarly, C 1s spectra of the SER biosorbent after the Cu and Pb adsorption was decomposed into 4 peaks, which binding energies were close to the ones detected for the pure plant. Interestingly, the peak attributed to the -C-O- group increased its intensity at the binding energy of 286.2 eV.
The decomposition of the O 1s spectrum (Fig. 6c) suggested that for pure SER biosorbent the presence of oxygen is only attributed to organic C-O or C = O compounds (binding energy, BE = 532.5 eV) (Smith et al. 2016; Wang et al. 2017). After the adsorption, the O 1s core-level spectra of both Cu-SER and Pb-SER samples were broadened by lateral bumps with deconvoluted peaks around 531.2 eV and 533.7 eV, which are assigned to metal carbonates or oxides and alcohol groups, respectively (Wang et al. 2017).
The analysis of the Pb 4f (Fig. 6d) spectrum revealed the presence of the characteristic doublet with the 4f7/2 and 4f5/2 spin-orbital levels distanced by 4.8 eV, suggesting the presence of Pb(II) species, which, after analysing the O 1s peak, can be connected to oxidized compounds such as Pb(OH)2 or PbO (Taylor and Perry 1984; Rondon and Sherwood 1998). However, it is difficult to determine the nature of the adsorbed lead substances, as almost all oxygen-containing lead compounds have a peak in this BE range (Laajalehto et al. 1993).
In the case of Cu adsorption, the characteristic doublet of the Cu 2p core-level spectrum (Fig. 6d) presented the spin-orbital levels positioned at 933.3 eV and 953.2 eV, attributed to the Cu 2p3/2 and Cu 2p1/2 spin orbitals with a ΔBE of 19.9 eV. The presence of weak intensity satellite features between 938 and 946 eV suggests the presence of Cu(I) species, such as Cu2O, rather than Cu(II) compounds (Biesinger 2017; Li et al. 2018; Liu et al. 2019b).
The changes of pH that occurred at each point of Cu(II) and Pb(II) adsorption isotherms are illustrated in Fig. 7. The calculation was done by subtracting the final pH after adsorption from the pH of the starting solution at the beginning of the adsorption (pHs-pHf). As demonstrated, both curves have negative and positive ranges. At the beginning of the adsorption isotherms (namely, at low concentrations of target ions), the changes of pH are negative, meaning that pHs is lower than pHf. This can be explained by the release of alkali and alkaline earth metals (see Table S3 in Supplementary information) that increased the pH. This release is related to the ion exchange mechanism of total adsorption process (Cui et al. 2016). Other parts of the curves showed the positive changes of pH, implying that pHs is higher than pHf. This can be interpreted as the release of additional hydrogen cations from the reactive carboxylic, amino and hydroxyl groups on the SER surface. It can be concluded that at higher Cu(II) and Pb(II) concentrations, the complexation and chelation are involved in the adsorption. Therefore, the adsorption of target metals consists of the ion exchange at the lower Cu(II) and Pb(II) concentrations and the complexation and chelation at the higher contaminants’ concentrations.
The measurements of ζ-potential for the bare SER plant and the samples Cu-SER and Pb-SER after adsorption showed the decrease of the negative charge after the adsorption processes (Table 3). This can be explained by the binding of negatively charged surface groups (mainly carboxylic) with metal ions.
Table 3
The ζ-potential and pH values changes of the point of zero charge of the native SER plant and samples after adsorption
Sample
|
ζ-potential
|
pH
|
SER
|
-25.4
|
5.9
|
Pb-SER
|
-18.1
|
5.2
|
Cu-SER
|
-18.1
|
5.2
|