4.1. The FTIR Spectra Analysis and C/N Analysis of Tested Sorbents
The FTIR spectrum made for SW is typical for lignocellulosic biomass (
Figure 2). The wide absorption band at 3600–3000 cm
−1 indicates the presence of hydrogen-bonded hydroxyl groups [
34]. The pronounced peaks at 2920 cm
−1 and 2851 cm
−1 are ascribed to asymmetric and symmetric vibrations stretching the C-H bonds in the aliphatic lignin and saccharide chains (-CH
2- bonds) [
35]. The presence of typical lignin ester functional groups in the sorbent is evidenced by the presence of carbonyl bonds (C=O) (peak at 1730 cm
−1) [
36]. The stretching of the C=C bond corresponds to the peaks at 1630 cm
−1 and 1515 cm
−1, and the deformations of the C-H bond in the benzene rings of lignin are indicated by the peaks at 1460 cm
−1 and 1420 cm
−1 [
37]. The peaks at 1370 cm
−1 and 1032 cm
−1 indicate stretching of the C-H bond, while the peak at 1320 cm
−1 corresponds to the stretching of the C-O bond of the C5 carbon of the aromatic rings of cellulose, hemicellulose, and lignin. The presence of C-O-C bonding typical for saccharide structures is indicated by the peaks at 1160 and 898 cm
−1 [
38].
The WS-E spectrum is characterized by the presence of peaks at 807, 829, and 853 cm
−1, indicating the presence of epoxy groups in the sorbent structure. These peaks are not present in the spectra of the remaining sorbents [
39] (
Figure 2).
The spectra of WS-A and WS-EA are quite similar. The peaks at 3320 cm
−1 as well as 660 cm
−1 indicate oscillation of the N-H bond. In turn, the stretching of the C-N bond of aliphatic amines corresponds to the peaks at 1200, 1155, and 1100 cm
−1 [
40]. Additional peaks observed for N-H and C-N bonds in the WS-A and WS-EA spectra confirm the enrichment of the sorbent structure with amino groups (
Figure 2).
The analysis of the contents of the C and N showed that the nitrogen content was similar in WS and WS-E and averaged 0.60–0.61% (
Table 2). In turn, WS-A was characterized by a slightly higher nitrogen content (0.63%). The richest in nitrogen turned out to be WS-EA (0.88%). The amount of N determined in WS-EA was higher than that of the unmodified sorbent by 44%. This confirms the higher efficiency of sorbent amination in the case of its initial activation with epichlorohydrin. Carbon contents in the tested sorbents were similar and ranged from 41.0 to 43.6%.
4.2. Influence of pH on the RB5 Sorption Effectiveness
The sorption efficiency of RB5 on WS, WS-A, and WS-E was the highest at the initial pH of 2 and decreased with increasing pH, reaching the greatest decrease in the pH range of 2–4. In the pH range of 4–8, the efficiency of RB5 sorption on WS, WS-A, and WS-E remained at a similar but low level. A characteristic feature of WS-E was a slight increase in the efficiency of dye sorption at pH 9–10. For WS-EA, the RB5 binding efficiency was the highest at pH 3 and decreased with increasing pH (except for pH 9), yielding the worst result at pH 11. Contrary to WS, WS-A, and WS-E, dye sorption on WS-EA was effective in a wide range of pH values, i.e., at pH 2–10 (
Figure 3a).
The high efficiency of RB5 sorption on the tested materials at pH 2–3 resulted from the anionic nature of the dye and the acquisition of a positive charge by the surface of sorbents [
41]. At a very low pH (pH 2), the concentration of hydronium ions (H
3O
+) in the dye solution was so high that there was an intense protonation of functional groups of straw-based materials. In all sorbents, the hydroxyl groups of polysaccharides and lignins were protonated (–OH + H
3O
+ → –OH
2+ + H
2O), while WS-EA also protonated numerous primary amine groups attached to the sorbent as a result of the amination process (–NH
2 + H
3O+ → –NH
3+ + H
2O). The positively charged surface of the sorbents attracted electrostatically to the particles of the anionic dye RB5, which significantly enhanced their sorption. As the pH of the solution increased, the concentration of H
3O
+ ions decreased, which resulted in a decreasing number of protonated functional groups. The lower total positive charge on the sorbent’s surface translated into a weaker interaction with RB5 and, consequently, a worse dye sorption efficiency. At pH > 4, the hydroxyl groups practically did not undergo protonation, which explains the relatively low efficiency of RB5 binding to WS, WS-A, and WS-E in the pH range of 4–10. At pH > 7, with increasing pH, the concentration of OH- ions also increased, and this increase caused the deprotonation of hydroxyl groups (–OH + OH
− → –O
− + H
2O) at a higher pH (pH > 10). The increasing total negative charge on the surface of the sorbents additionally hindered the sorption of anionic RB5, which is particularly visible in the pH range of 9–11 (
Figure 3a).
In the pH range of 2–9, most of the primary amine functional groups were in the ionized form. The protonated amine groups are very good sorption centers for all anionic dyes [
42]. This explains the high efficiency of RB5 sorption on WS-EA over a wide pH range (
Figure 3a). The obtained result is typical for sorbents rich in amine groups, such as, e.g., chitosan. A similar result was not observed in the case of WS-A due to the low efficiency of amination of polysaccharides not modified with epichlorohydrin.
The increase in RB5 binding efficiency in the pH range of 9–11 observed in the case of WS-E was most likely due to the presence of epoxy functional groups on the sorbent surface. Under alkaline conditions, epoxy groups are able to undergo a condensation reaction with functional groups of dyes [
43]. As a consequence, at pH > 8, the physical adsorption of RB5 on WS-E was aided by chemisorption.
The slight increase in the sorption efficiency of RB5 on WS-EA, noted at pH 9, most probably resulted from the presence of the –NH
2 group in the dye. At pH 9, WS-EA already had many local negative charges on its surface, while approximately half of the amino groups of RB5 were still protonated, which theoretically could help in binding the dye. At pH > 9, both WS-EA and RB5 already had a strong total negative charge, which resulted in a significant reduction in the efficiency of dye sorption on WS-EA. A similar effect has been obtained during the sorption of RB5 on sorbents with amine groups, e.g., chitosan [
26,
31].
The tested sorbents affected a change in the pH of the dye solution in which the sorption took place (
Figure 3b). During sorption with WS at the initial pH range of 4–9, the final pH of sorption was established at pH 6.4–7.5, and in case of WS-E at pH 6.1–7.4. For the ammonium-modified sorbents, at the initial pH range of 4–9, the pH of the solution at the end of the sorption was in the range of 6.8–7.6 (WS-A) and pH 7.5–7.9 (WS-EA).
The change in the pH of the solutions during sorption was related to the system’s tending to achieve a pH close to the pH
PZC value, characteristic for the sorbent used (PZC—point of zero charge—the point of zero charge of the sorbent). The pH
PZC values determined for the tested sorbents were: pH
PZC = 7.30 (WS), pH
PZC = 7.21 (WS-E), pH
PZC = 7.49 (WS-A), and pH
PZC = 7.88 (WS-EA) (
Figure 3c). The pH
PZC values obtained for WS and WS-E indicate the relatively neutral nature of the sorbent, which results from the high content of cellulose, hemicellulose, and lignins. Higher pH
PZC values determined for the aminated sorbents (WS-A and WS-EA) are due to the attachment of amine functional groups to the sorbent, which affect its basic nature. The higher pH
PZC value of WS-EA compared to WS-A is the result of the initial modification/activation of the sorbent with epichlorohydrin prior to carrying out the amination process. The material enriched with epoxy groups showed much greater susceptibility to amination than the material with mainly hydroxyl groups [
44]. The higher content of amine functional groups in WS-EA compared to WS-A translated into a higher pH
PZC value (
Figure 3c).
For each of the tested wheat straw-based sorbents, the RB5 sorption efficiency was the highest in the pH range of 2–3. As colored industrial wastewater usually has pH > 2 [
45], we decided to conduct all other experiments (
Section 4.3 and
Section 4.4) at pH 3.
4.3. Kinetics of RB5 Sorption
In the research series with WS, WS-A, and WS-E, regardless of the initial dye concentration (10–500 mg RB5/L), the determined sorption equilibrium time was 210 min. A slightly shorter equilibrium time (180 min) was recorded in the test series with WS-EA for dye concentrations of 50 and 500 mg RB5/L (
Figure 4). A similar equilibrium time was recorded in the research on RB5 sorption on wheat straw modified with cetylpyridinium chloride (195 min) [
24], sunflower seed shell (210 min) [
46], commercial acrylic ion exchange resins (180 min) [
47], and commercial activated carbon (180 min) [
48].
The shorter equilibrium time of RB5 sorption on WS-EA as compared to the other sorbents could be due to the sorbent having a greater amount of protonated functional groups. At pH 3, most of the amino groups are in the protonated form, while the hydroxyl groups are mainly in the non-protonated form. Therefore, under the experimental conditions, WS-EA had a greater number of protonated functional groups than WS, WS-A, and WS-E, which meant a stronger total positive charge on the sorbent’s surface. The stronger interaction of WS-EA with RB5 coupled with a higher concentration of the dye resulted in the acceleration of the process and faster saturation of the available sorption centers.
Apart from the shorter sorption equilibrium time, a large amount of amine functional groups on the WS-EA surface also resulted in significantly higher RB5 sorption capacity compared to the other tested sorbents, which is clearly visible in
Figure 4. The effect of amination of straw on its sorption capacity is described in more detail in
Section 4.4The experimental data obtained from the research on the sorption kinetics of RB5 on the tested sorbents was described using a pseudo-first and pseudo-second-order model (
Figure 4,
Table 3).
In each research series, the greatest match to the experimental data was shown by the pseudo-second-order kinetic model (
Table 3), which is typical for the sorption of dyes on biosorbents [
49,
50]. The efficiency and the rate of RB5 sorption on sorbents increased with the increase in the initial dye concentration. This can be explained by the greater probability of collisions of RB5 anions with sorption centers of sorption materials in the systems with a higher dye concentration in the solution. The sorption rate constants (k
2) determined from the model decreased with an increase in the initial concentration of the RB5 dye. A similar tendency was noted in the case of sorption of anionic pigments on chitosan sorbents [
51], apple seeds [
52], and also sorbents based on activated carbons [
53].
The experimental data was also described using the intramolecular diffusion model. The analysis of the determined model constants allowed distinguishing three main sorption phases (
Figure 5,
Table 4).
Probably in the first phase of sorption, RB5 ions were transported from the solution to the boundary phase, due to the electrostatic interaction between the positively charged surface of the sorbent and the negatively charged dye. The dye ions located at the sorbent bound to the most accessible sorption centers on the sorbent’s surface. This phase was characterized by high intensity and also relatively short duration (
Figure 5,
Table 4). The value of the intramolecular diffusion constant (k
d1) determined for each sorbent increased with the increase in the initial concentration of RB5 in the solution (
Table 4), which is typical for physical adsorption [
54].
The first phase of sorption on WS, WS-A, and WS-E occurred with similar intensity. Presumably, this was the result of having mostly functional groups with similar susceptibility to protonation (–OH groups). The WS-EA showed a much higher sorption efficiency of RB5 in the first phase. The much higher efficiency of WS-EA as compared to WS, WS-A, and WS-E was due to the presence of a high number of –NH
2 groups on the sorbent surface. Due to having easily protonated primary functional groups, WS-EA gained a large total positive charge at pH 3, which intensified the dye binding on the sorbent’s surface. A similar result could not be obtained with WS-A due to the low amination efficiency without sorbent pre-activation with epichlorohydrin, which was already explained in
Section 4.1.
After the saturation of most of the readily available sorption centers on the sorbent’s surface, the second sorption phase began. In this phase, RB5 ions competed for the last free active sites on the sorbent’s surface and bound to the sorption centers located in deeper, less accessible sorbent layers. Due to the large accumulation of RB5 at the surface, the high molar mass of the dye (992 g/mol), as well as the strong interactions between the dyes, the second phase lasted much longer and was less intense than the first. The sorption intensity of RB5 on WS and WS-E in the second phase, determined on the basis of the values of the kd2 constants, was similar. The greater efficiency of sorption on WS-A as compared to WS and WS-E was most likely due to the few amine groups attached to the sorbent’s chemical structure during direct amination of the straw. The primary amine functional groups of WS-A, although in the minority in relation to the hydroxyl groups, were easily ionized at pH 3. This generated a sufficient positive charge to aid RB5 sorption. The highest efficiency as well as the shortest duration of the second sorption phase was observed in the research series with WS-EA, which was the result of the sorbent having the highest amount of amine functional groups.
In the third phase of sorption, the last free active sites in the deeper layers of the sorbent were saturated by RB5 anions. Due to the very poor availability of sorption centers, this phase was characterized by the lowest intensity and the longest duration. The processes taking place in this phase had no major impact on the final results of RB5 sorption on the tested sorbents. In practice, shortening the process to the first two key sorption phases would be economically reasonable.
4.4. Maximum Sorption Capacity
The obtained experimental data from the research on the maximum RB5 sorption capacity of the tested sorbents was described using three popular sorption isotherms: Langmuir 1, Langmuir 2, and Freundlich (
Figure 6,
Table 5). The comparison of determination coefficient (R
2) values determined from the sorption models shows that for the research series with WS, WS-A, and WS-E, the double Langmuir isotherm (Langmuir 2 model) showed the best fit to the experimental data. In the case of the research series with WS-EA, the data was equally well described by the Langmuir 1 and 2 models (the same values of R
2, K
C/K
1/K
2 and Q
max).
Better fit of the data to the Langmuir 2 model compared to the Langmuir 1 model (for WS, WS-A, and WS-E) may indicate that the sorbent has at least 2 types of sorption centers with a different degree of affinity for the dye. However, different types of sorption centers do not have to mean completely different functional groups. In the Langmuir 2 model, two different active sites may be the same functional groups (e.g., hydroxyl) but with different availability and thus interaction with and affinity to the dye.
The values of b
1 and K
1 presumably describe the sorption sites of the first type located in deeper, less accessible layers of the sorbent, or they are centers that bind to the dye via weaker ionic interactions, e.g., hydrogen bonds (
Table 5). The b
2 and K
2 values presumably describe the capacity and the degree of affinity of the RB5 dye for the second type of sorption centers located on the sorbent’s surface. In the case of WS, WS-A, and WS-E, these are mainly hydroxyl groups that protonated at pH 3. In the case of WS-A, the b2 value is probably also influenced by a few protonated amino groups and for WS-E also by epoxy groups.
The same values of Qmax, KC/K1/K2 and R2 observed in the case of WS-EA may indicate a significant dominance of one type of the sorption center in the RB5 binding process. Protonated primary amine functional groups are most likely the type of sorption center in question.
The maximum sorption capacity calculated based on the model for WS, WS-A, WS-E, and WS-EA were 16.72 mg/g, 24.12 mg/g, 18.79 mg/g, and 91.04 mg/g, respectively.
Amination had a significant effect on increasing the sorption capacity of straw-based sorbents. The sorption capacities of WS-A and WS-EA calculated from the Langmuir 2 model were on average 44.2% and 444.5% higher than WS.
As already mentioned in
Section 4.1 and
Section 4.2, during straw modification with ammonia water, the polysaccharides present in the material were aminated. Amine functional groups able to easily protonate significantly supported the binding process of RB5 on the sorbent. Presumably, in the case of WS-EA, a large number of –NH
2 groups additionally influenced the relaxation of the sorbent’s structure, which increased its specific surface area and increased access to many sorption centers located in its deeper layers. The small sorption capacity of WS-A, compared to WS-EA, resulted from a much smaller number of –NH
2 groups added during ammonization. The amination efficiency of the polysaccharides present in the sorbent was much higher for the sorbent pre-activated with epichlorohydrin.
The mere activation of the sorbent with epichlorohydrin had no significant impact on the efficiency of RB5 sorption. The sorption capacity of WS-E, compared to WS, was higher by only approximately 12%, which could be due to the more developed sorbent’s surface after activation. Theoretically, WS-E, due to its epoxy functional groups, could also permanently bind dyes by chemisorption, but the expected effectiveness of this process was low (condensation of the reactive dye requires an alkaline environment) at pH 3.
Table 6 presents the parameters of the sorption process of the RB5 dye on various sorbents.
Wheat straw subjected to amination with pre-activation by epichlorohydrin (WS-EA) had a greater sorption capacity than other unmodified plant biomass-based sorbents tested so far, such as seed husks, stems of crops, fruit peels, or compost.
Properly prepared aminated sorbents based on lignocellulose materials can show between 1760% [
26] to 1920% [
44] greater sorption capacity than their unmodified counterparts. Particularly noteworthy is the fact that wheat straw subjected to amination with initial epichlorohydrin activation as well as similarly prepared buckwheat hulls [
44] are able to achieve greater RB5 sorption capacity than the sorbents based on activated carbons [
55,
56,
57,
58,
61] (
Table 6). The method of modifying lignocellulose sorbents proposed in the research may be of great economic importance in the future, as it allows obtaining high-quality sorbents from widely available waste materials from the agri-food industry. The material based on modified straw produced during the sorption of dyes can be used as a high-energy solid fuel for direct combustion or co-combustion [
66]. It can also be a substrate to produce second and third generation fuels [
67]. An alternative method of straw management is its carbonization to activated carbon with potential use in the purification of liquids and gases.