3.1 Characterization of the microparticles
Examining the scanning electron microscopy images (Fig. 2a and 2b), one can discern that both beads display a spherical shape. Nevertheless, the surface of chitosan exhibits higher roughness and porosity compared to alginate. This feature is likely to have played an important role in enabling chitosan to attain a higher adsorbed amount than alginate across all adsorption experiments.
The mean diameter measured for the wet beads was 1.89 ± 0.02 mm for chitosan and 1.98 ± 0.02 mm for alginate. After drying in a forced-air oven at 25°C for 24 hours, the diameters were 1.11 ± 0.02 mm for alginate (loss of 43.94 ± 0,08%) and 0.98 ± 0.02 mm for chitosan (loss of 48.15 ± 0,07%). Therefore, chitosan experienced a greater water loss than alginate. This difference is likely attributed to the presence of amino groups in chitosan, along with its evidently higher porosity (Fig. 2). Adriano et al. (2005) produced chitosan beads crosslinked with 5% glutaraldehyde, featuring an average diameter of 2.12 mm (wet basis). Therefore, not only do the beads of pure chitosan have their active sites unobstructed by crosslinking, but their volume is also smaller.
The TGA assesses the mass variation of the sample with respect to temperature. Figure 3a shows the thermogram of alginate beads, which shows that the evaporation of water molecules occurs within the temperature range of 25°C to 200°C. This investigation reveals a substantial mass reduction initiating at 200°C and extending to 320°C, encompassing the dehydration of saccharide rings and the rupture of C–O–C bonds. The ultimate decomposition temperatures are observed at 500°C and 600°C, signaling a virtual cessation of mass loss (Adzmi et al. 2012; Estrada-Villegas et al. 2020).
Figure 3b illustrates the thermogram of chitosan beads, in which the initial weight loss phase is evident within the temperature range of 25°C to 120°C, attributed to moisture loss (approximately 10%). Chitosan experiences non-oxidative thermal degradation under nitrogen flow, observed in the temperature span of 245°C to 600°C. This indicates the deacetylation of chitosan, involving vaporization and elimination of volatile products. The degradation process of chitosan starts with amino groups forming unsaturated structures (Dey et al. 2016).
The DSC thermogram of alginate beads (Fig. 4a) reveals two distinct endothermic peaks at 100°C and 200°C. The first is likely attributed to water evaporation, while the second indicates a melting point.
The introduction of amoxicillin into the alginate beads significantly reduces the intensity of these peaks. For the peak at 100°C, amoxicillin occupies the space previously filled by water (Patel et al., 2006; Venkateswarlu, 2017). Additionally, amoxicillin demonstrates an affinity for binding with water molecules. The second peak (200°C) is associated with the interaction between alginate and amoxicillin, as both possess various functional groups (Bankole et al., 2022). This observation implies the molecular dispersion of drugs within the beads.
The primary distinction in Fig. 4a is the peak observed around 140°C in the DSC thermogram of alginate and amoxicillin; however, this peak is not present in the thermogram of alginate alone. The endothermic peak around 140°C indicates a thermal transition in amoxicillin, potentially associated with its phase changes (Newton et al., 2014). This peak is also evident in Fig. 4b (DSC thermogram of chitosan beads containing amoxicillin), contrasting with its absence in the chitosan-only thermogram.
The DSC thermogram (Fig. 4b) of chitosan beads suggests that the initial peak corresponds to the evaporation of free water. However, chitosan exhibits a strong affinity for water, resulting in a large peak that follow the temperature increase. The presence of amoxicillin leads to the merging of the peak associated with water loss with the peak of thermal transition in amoxicillin, potentially connected to its phase changes, as observed in the thermogram of Fig. 4a (Maswadeh, 2017). This occurs due to amoxicillin's strong affinity for water (Narkar et al. 2010).
3.2 Effect of pH
In Table 1, it can be observed that the optimal pH for the adsorption of amoxicillin onto beads of chitosan and alginate was pH 4. As per Adriano et al. (2005), amoxicillin exhibits an isoelectric point of 5.7. This distinctive feature is ascribed to the functional groups inherent in amoxicillin, namely, carboxyl (with a pKa1 of 2.68), amine (with a pKa2 of 7.49), and phenolic hydroxyl (with a pKa3 of 9.63). At the pH equal to pKa3, the phenolic hydroxyl group undergoes deprotonation, resulting in amoxicillin carrying a double negative charge. When the pH falls within the range of pKa2 and pKa3 values, deprotonation of the amine group takes place. At a pH value between pKa1 and pKa2, amoxicillin remains uncharged due to the deprotonation of the carboxylic group. Lastly, amoxicillin demonstrates a cationic characteristic at pH values lower than 2.68 (Barbooti and Zahraw, 2020).
Table 1
Effect of pH variation on the amoxicillin adsorption capacity.
| Alginate | Chitosan |
pH | mg g− 1 | mg g− 1 |
3 | 58.4 \(\pm\) 0.2 | 55.4 \(\pm\) 0.3 |
4 | 74.2 \(\pm\) 0.3 | 80.4 \(\pm\) 0.2 |
5 | 55.4 \(\pm\) 0.2 | 62.4 \(\pm\) 0.2 |
6 | 30.6 \(\pm\) 0.1 | 42.3 \(\pm\) 0.1 |
7 | - | 33.2 \(\pm\) 0.2 |
8 | - | 20.4 \(\pm\) 0.3 |
Mean \(\pm\) standard deviation (n = 3) |
Concerning chitosan, the positive surface charge of chitosan tends to increase with a decrease in pH, below pH 6 (Camara et al. 2020). This phenomenon arises from the protonation of the amino group. In contrast, the surface charge of amoxicillin tends to remain neutral between pH 2.7 and 7.5. Therefore, at pH below 4, there is an increase in the repulsion between amoxicillin and chitosan, because amoxicillin contains an amino group that tends to be protonated with a decrease in pH. With an increase in pH, deprotonation of chitosan occurs, resulting in a diminished attraction between chitosan and amoxicillin.
Alginate exhibits a negative charge with an increase in pH above 3, due to the presence of the carboxylic group. The presence of the carboxylic group in alginate and the amino group in amoxicillin are likely responsible for the interaction and, consequently, adsorption, with a maximum interaction at pH 4. Adsorption above pH 6 was not feasible because alginate beads dissolved in water (Gao et al. 2020).
In the literature, amoxicillin adsorption has been investigated using various adsorbents, and the maximum adsorption capacity observed by fitting the Langmuir model varies widely: it ranges from 2.28 mg g⁻¹ for Organobentonite (Xing Zha et al. 2013) to 909 mg g⁻¹ for magnetic Bionanocomposite (Mosavi et al. 2023). Other reported values include 7.3 mg g⁻¹ using methylene blue (Barbooti and Zahraw, 2020), 135 mg g⁻¹ at 30°C for activated carbon prepared from durian shell (Yazidi et al. 2020), 345.4 mg g⁻¹ for activated carbon from Arundodonax biomass (Chayid and Ahmed, 2015), 357 mg g⁻¹ at pH 9 for tecomachip wood waste into microwave-irradiated adsorbent (Khan et al. 2023), and 526 mg g⁻¹ for Graphene/copper oxide nanocomposites (Moradi et al. 2022).
However, the fabrication of all these adsorbents involves numerous intricate steps and a variety of chemical reagents, making the process considerably more complex when compared to the relatively simpler production of chitosan and alginate beads. Additionally, these more complex fabrication methods may have potential ecological impacts, such as increased chemical waste generation and energy consumption.
3.3 Effect of contact time
Figure 5 shows the adsorption kinetics of amoxicillin in alginate and chitosan beads, as well as the fitting of data to the Boyd and Weber and Morris models. Chitosan took 200 minutes to reach an adsorption capacity of 78.6 mg g⁻¹, while alginate took 250 minutes to reach 71.3 mg g⁻¹.
When adsorption is physical, the influence of mass action may be neglected. As will be demonstrated, the enthalpy obtained in this study is characteristic of physical adsorption. Therefore, adsorption is governed by either liquid film diffusion or intraparticle diffusion. Both mechanisms are often present simultaneously in many heterogeneous systems, and the overall rate of mass transfer is usually determined by the slowest step, which may vary depending on the specific conditions of the system. Experimental techniques, such as fitting experimental data to mathematical models like the Weber-Morris equation, can help elucidate the relative contributions of liquid film diffusion and intraparticle diffusion to the overall mass transfer process. Liquid film diffusion refers to the process of mass transfer that occurs as a solute in a fluid phase moves from the bulk of the fluid to the surface of a solid particle, where adsorption or reaction takes place. Intraparticle diffusion, on the other hand, occurs within the solid particles themselves. After reaching the surface of the solid, the solute must then diffuse into the pores or internal structure of the solid particle.
According to Table 2, the best-fitted model for chitosan was Boyd's external diffusion, indicating that the rate-limiting step is the external diffusion model. However, for alginate, a better fit was obtained with the Weber and Morris model, thus indicating that the rate-limiting step is the internal diffusion model. Intraparticle diffusion can be influenced by pore size, and as observed in Fig. 2, the surface of chitosan exhibits higher roughness and porosity compared to alginate, which may facilitate the internal mass transfer in chitosan beads. This difference in beads morphology may explain the results regarding the rate-limiting step for adsorption observed in chitosan and alginate beads.
Table 2
Parameters for the study of amoxicillin kinetics.
| Chitosan | Alginate | Chitosan | Alginate |
| Boyd's external diffusion | Weber and Morris |
\({\chi }^{2}\) | 74 \(\pm\) 1.2 | 82 \(\pm\) 1.1 | 142 \(\pm\) 1.8 | 0.961 \(\pm\) 0.001 |
R2 | 0.973 \(\pm\) 0.001 | 0.961 \(\pm\) 0.001 | 0.991 \(\pm\) 0.001 | 0.933 \(\pm\) 0.001 |
\({k}_{\text{W}\&\text{M}}\) | - | - | 4.13 + 0.11 | 5.51 + 0.057 |
q∞ | 82.72 \(\pm\) 3.12 | 136.75 \(\pm\) 8.2 | - | - |
R | 0.0125 \(\pm\) 0.0014 | 0.00251 \(\pm\)0.0012 | - | - |
Mean \(\pm\) standard deviation (n = 3) |
3.4 Effect of temperature
Observing Fig. 6, it is evident that the increase in temperature resulted in an enhancement of the adsorption capacity. This is because amoxicillin may experience solvent solvation effects, and thus, at higher temperatures, amoxicillin has more energy to overcome solvation.
Analyzing the results presented in Table 3, it is noticeable that the best-fitted model for chitosan beads was the Freundlich model. Furthermore, the Freundlich constant n was greater than 1 in all cases, indicating that the surface of chitosan is non-homogeneous, contrary to what the Langmuir model suggests (Perwitasari et al. 2021). This suggests that one adsorption site might interfere with adsorption at another site. Regarding alginate beads, generally, the Freundlich model also provided a better fit. The behavior of n for alginate beads was similar to that observed for chitosan beads. That is, the adsorption of one molecule favors the adsorption of another molecule, indicating positive cooperativity.
Table 3
Equilibrium isotherm parameters for amoxicillin adsorption.
| Chitosan | Alginate |
Langmuir |
T | qm | Kd | R2 | qm | Kd | R2 |
278 | 98.4 \(\pm\)0.8 | 0.0348 \(\pm\)0.0002 | 0.857 \(\pm\)0.001 | 60.8 \(\pm\)1.2 | 0.0191 \(\pm\)0.0002 | 0.859 \(\pm\)0.001 |
293 | 102.2 \(\pm 0.9\) | 0.057 \(\pm\)0.0002 | 0.981 \(\pm\)0.001 | 104.3 \(\pm\)0.8 | 0.0209 \(\pm\)0.0002 | 0.981 \(\pm\)0.001 |
308 | 94.4 \(\pm 0.7\) | 0.083 \(\pm\)0.0002 | 0.984 \(\pm\)0.001 | 99.0 \(\pm\)0.7 | 0.0344 \(\pm\)0.0002 | 0.994 \(\pm\)0.001 |
Freundlich |
278 | 2.7 \(\pm 0.4\) | 1.7 \(\pm\)0.1 | 0.992 \(\pm\)0.001 | 0.7 \(\pm\)0.4 | 1.6 \(\pm\)0.2 | 0.996 \(\pm\)0.001 |
293 | 15.1 \(\pm 0.5\) | 2.9 \(\pm 0.2\) | 0.991 \(\pm\)0.001 | 10.8 \(\pm\)0.4 | 2.5 \(\pm\)0.2 | 0.991 \(\pm\)0.001 |
308 | 11.3 \(\pm 0.7\) | 2.7 \(\pm\)0.1 | 0.996 \(\pm\)0.001 | 7.1 \(\pm\)0.5 | 2.2 \(\pm\)0.2 | 0.990 \(\pm\)0.001 |
Mean \(\pm\) standard deviation (n = 3) |
Adriano et al. (2005) investigated the adsorption of amoxicillin on chitosan beads crosslinked with 5% glutaraldehyde at pH 6.5. They achieved a maximum adsorption capacity of only 8.71 mg g− 1, according to the Langmuir model. This diminished capacity is likely attributable to the glutaraldehyde, which reduces the number of active sites on the chitosan available for adsorption. Furthermore, in this study, a significantly higher maximum adsorption capacity of 102 mg g− 1 was observed, according to the Langmuir model. This disparity underscores how glutaraldehyde not only hampers drug adsorption but also complicates the synthesis of chitosan beads by introducing additional steps. Mirizadeh et al. (2024) investigated the removal of amoxicillin by magnetic chitosan/microalgae biocomposites, achieving optimal results between pH 5 and 6, obtaining a maximum adsorption capacity of 125.6 mg g− 1. Yeo et al. (2023) investigated the adsorption of amoxicillin using a bentonite-chitosan composite and observed a maximum capacity of 66 mg g− 1 at 50°C. Danalioğlu et al. (2017) adsorbed amoxicillin using magnetic activated carbon/chitosan and achieved an adsorption capacity of 98 mg g− 1 for 15 mg of adsorbent; however, the pH of the medium was not controlled. Our adsorption capacity was similar to those reported in the literature; however, it is noteworthy that the adsorption sites on chitosan beads are more accessible compared to the mentioned adsorbents in the literature. This is because the beads did not undergo any crosslinking process, nor did the amino groups form chemical bonds with other compounds.
Abed and Faisal (2023) explored the adsorption of amoxicillin using a calcium/iron-layered double hydroxides-sodium alginate nanoadsorbent at pH 7, achieving a maximum capacity of 6.7 mg g− 1 according to the Langmuir model. Kaur and Maity (2020) successfully removed amoxicillin using a graphene oxide/calcium alginate biocomposite, achieving a maximum adsorption capacity of 50 mg g− 1 at pH 6. In their study, Karimi and Namazi (2022) effectively adsorbed amoxicillin utilizing alginate/glycodendrimer beads, achieving a maximum adsorption capacity of 48.8 mg g− 1 at pH 7. Alhattab et al. (2023), using alginate/bentonite-impregnated TiO2 beads for amoxicillin adsorption, achieved an adsorption capacity of 85 mg g− 1 at pH 6. The adsorption capacity achieved in this study for the alginate beads was 74.2 ± 0.3 mg g− 1 at pH 4, higher than several previously cited adsorption capacities. Additionally, the manufacture of these aforementioned adsorbents involves numerous intricate steps and the use of various chemical reagents, significantly complicating the process compared to the relatively straightforward production of alginate beads. Furthermore, these more complex fabrication methods could potentially result in ecological ramifications, such as increased generation of chemical waste and energy consumption.
3.5 Thermodynamic parameters
Table 4 reveals that the change in Gibbs free energy (ΔG) was consistently negative for both beads at all temperatures, indicating a spontaneous process. The rise in temperature further enhanced adsorption, leading to a more negative ΔG. The notably more negative values for chitosan beads may be linked to their higher adsorption capacity.
Table 4
Thermodynamic parameters of amoxicillin adsorption by chitosan and alginate beads.
T | \(\varDelta G\) (kJ mol− 1) | \(\varDelta H\) (kJ mol− 1) | \(\varDelta S\) (kJ mol− 1 K− 1) |
| Chitosan | Alginate | Chitosan | Alginate | Chitosan | Alginate |
5°C | -1.5\(\pm 0.2\) | -0.10\(\pm\)0.01 | | | | |
20°C | -2.8\(\pm\)0.2 | -0.30\(\pm\)0.01 | 20.6\(\pm 0.8\) | 13.8\(\pm\)0.4 | 0.080\(\pm\)0.001 | 0.049\(\pm\)0.001 |
35°C | -3.9\(\pm\)0.1 | -1.60\(\pm\)0.01 | | | | |
Mean \(\pm\) standard deviation (n = 3) |
The positive ΔS value indicates an augmentation in system disorder at the solid-liquid interface following adsorption. At the same time, the positive ΔH value is ascribed to the solvation of amoxicillin molecules by the solvent. As a result, the energy needed to overcome solvation exceeded the energy released during the bond formation between the adsorbent and the adsorbate. Similar results are found in the literature regarding the adsorption of amoxicillin by Mirizadeh et al. (2024) (magnetic chitosan/microalgae biocomposites) and Yeo et al. (2023) (bentonite-chitosan composite). Both studies reported that the adsorption process was spontaneous and endothermic.
Enthalpy values around 20 kJ mol⁻¹ and 40 kJ mol⁻¹ are indicative of physical adsorption. Typically, this holds true for weak Van der Waals forces. However, it is possible that hydrogen bonding occurs between the beads and amoxicillin, given the presence of nitrogen, oxygen, and hydrogen atoms.