3.1 Adsorbents characterization
XRD patterns for SBA-15 P/C and SBA-15 APTES are shown in Fig. 2. The diffractogram peaks corresponding to the (100), (110) and (200) planes are indicated at 2𝜃=0.85, 1.43 and 1.65, respectively for SBA-15 P/C, and 0.83, 1.43 and 1.63 for the modified material. The results showed that the mesoporous materials are characterized by 2-D hexagonal symmetry (P6mm). (Laaz et al., 2016; Zhao D., 1998)
After APTES modification, there was a minor pattern change due to the decrease in pore size, but the symmetrical structure was preserved. For SBA-15 APTES, the (100) and (200) planes shifted to a lower diffractogram angle compared to the unmodified parent material. This change is certainly due to the filling of pores with amino functional groups. (Li et al., 2008)
Table 1 recapitulates the structural characteristics of the mesoporous material before and after amino groups incorporation.
Table 1 : Structural characteristics of SBA-15 before and after modification, based on XRD data.
Sample
|
2𝜃
|
d100 a (nm)
|
a0b (nm)
|
SBA-15 P/C
|
0.85
|
10.39
|
12
|
SBA-15 APTES
|
0.83
|
10.64
|
12.3
|
a interplanar distance; b mesoporous structure parameter.
As demonstrated by the result of nitrogen adsorption-desorption studies in Fig. 3. All samples showed type IV isotherms according to the IUPAC classification (K. S. W. SING, 1985), and are thus characteristic of mesoporous materials.
For both SBA-15 P/C and SBA-15 APTES materials, the part of the isotherm at low relative pressures (P/P0) (between 0 and 0.6) corresponds to monolayer surface filling, while the significant increase in adsorbed volume at pressures between 0.6 and 0.8 is associated with capillary condensation of nitrogen inside the channels and consequently with pore volume filling.(Dubinin, 1975) Interestingly, the width of the hysteresis loop did not change significantly after calcination, indicating preservation of the pore arrangement.
The apparent decrease in pores size, surface area and micropore volume is primarily attributed, as indicated in the XRD analysis, to pore blocking after modification with silanols groups. (See Fig. 4)
The physicochemical properties of pure SBA-15 and modified material are detailed in Table 2.
Table 2
The physicochemical properties of SBA-15 and the modified material.
Sample
|
SBET c (cm3.g − 1)
|
Dp d (nm)
|
Micropore volume (cm3.g − 1)
|
SBA-15 P/C
|
894
|
1.03
|
8.86
|
SBA-15 APTES
|
396
|
0.62
|
8.24
|
c The BET surface area estimated in the relative pressure range P/P0; d Pore diameter calculated using the BJH method.
Zeta potential of the mesoporous materials as a function of pH solution is shown in Fig. 5.
Each SBA-15 sample’s zeta potential was conducted at numerous pH spectrum to identify the specific pH at which the surface charge becomes neutral (isoelectric point IEP).(Tang et al., 2011) In a pH increasing matter, zeta potentials indicated low results even at the highest values (> 15mV). For SBA-15 APTES, zeta potential continuously decreases and IEP is around 6. (Tanev et al., 1997)
According to this results APTES can be used as an adsorbent for both negative and positive charged substances, however SBA-15 P/C can be used with only positive charged substances.
Thermogravimetric analysis in air atmosphere was also conducted for the samples before and after modification. Figure 6 present the evolution of the mass of the materials during the course of the stoichiometric reaction of the thermogrammes. Figure 6.a, compares the thermogrammes of the different materials as function of temperature, while Fig. 6.b compares the curves derived from these thermogrammes.
Samples heated between 25 and 160°C lost 13.58% of their mass. Due to water desorption, functionalized samples lost more mass (20.61%) in this range. Between 160 and 600°C, organic compounds decomposed and evaporated, resulting in further mass loss. The non-functionalized samples lost 10.87%, while functionalized samples lost 32.50%. At the highest temperature (+ 800°C), the silicate network underwent dihydroxylation, leading to minimal mass loss: 1.01% for non-functionalized and 3.82% for functionalized samples.(Bérubé & Kaliaguine, 2008; Boukoussa et al., 2018)
FT-IR spectroscopy was employed to characterize the synthesized and functionalized materials. (see Fig. 7)
According to the spectral results, the materials analyzed exhibit very similar spectral features. The observation of a characteristic band between 3000–3700 cm− 1 indicated the presence of the O-H band for silanols groups.(Chong & Zhao, 2003) Two bands were also observed in the region of 800 cm− 1 and between 950 and 1300 cm− 1, which are attributed to the symmetric and asymmetric vibration of Si-O-Si group, respectively. In addition, the functionalized material shows typical bands for silanols groups, demonstrating the incorporation of amino groups into the parent material.(Yuan et al., 2020) The new bands marked at approximately 2850cm− 1 and 1500cm− 1 are characteristic of the formation and deformation of NH2 respectively.
3.2 Adsorption experiments
3.2.1 Effect of adsorbent selectivity
The nature of the adsorbent was investigated by adding similar amount of different adsorbents based on the SBA-15 nanomaterial in NGB and copper aqueous solutions; the results are shown in Fig. 8
The quantitative study of the materials previously synthesized shows that the APTES-functionalized material has a higher adsorption capacity than the parent material as regards both of the heavy metal and the dye selected.
This study reveals distinct adsorption behaviors among the synthesized materials. Grafting generally leads to increased pollutant uptake within the material’s pores due to the introduction of silanols groups (Si-OH), (Makhlouf Mourad, 2018; Sengupta et al., 2017) this explains the enhanced adsorption observed in grafted materials compared to non-grafted ones.
The maximum adsorption capacity of functionalized materials towards NGB and copper molecules is significant, which suggest it as an efficient adsorbent for multiple substances regardless of their chemical structure.
Therefore, the rate of pollutant adsorption is strongly influenced by the properties of the synthesized materials through an increase in pore size. (Boukoussa et al., 2017)
3.2.2 Adsorption kinetics
Adsorption kinetics of the selected pollutant were carried out on the selected material previously according to its binding capacity (see Fig. 9); indicating complete adsorption of the organic and inorganic pollutants within approximately 90 min per contact time with the adsorbent. This study was based on the result of pseudo-first-order, pseudo-second-order and intraparticle diffusion models formulated according to Eqs. (2), (3) and (4) respectively. (Gunasundari et al., 2020; Rizzi et al., 2014)
Where k1 and k2 represent rate constants of first and second-order adsorption reactions, respectively. kd is the rate constant of the intraparticle diffusion model (mg/g/min1/2) qe and qt (g.mg− 1) are the amounts of the adsorbed pollutant respectively at t time and equilibrium; parameters of the adsorption kinetics of the selected pollutants are shown in Table 3.
Table 3
Kinetic parameters of the adsorption of NGB dye on SBA-15-C and copper by SBA-15 APTES.
Pollutant
|
pseudo-first-order model
|
pseudo-second-order model
|
qe(mgg− 1)
|
k1(L.min− 1)
|
R2
|
qe(mg.g− 1)
|
k1(g.mg− 1.min− 1)
|
R2
|
NGB
|
6.15
|
0.05
|
0.931
|
111.11
|
0.0008
|
0.994
|
Copper
|
93.70
|
0.043
|
0.962
|
92.60
|
0.0009
|
0.996
|
According to the correlation coefficient (R2), and the linear study of both of the models; kinetic adsorption is best fitted with the pseudo-second-order model with a maximum adsorption of 111.11 mg/g of NGB and 92.60 mg/g of copper.
These results suggest that there is an interaction between the contaminants and the functional groups of SBA-15 APTES, confirming a chemisorption-based adsorption. (Hamzaoui et al., 2016)
In the case of intraparticle diffusion, the molecules are presumed to diffuse into the liquid and penetrate the pores, this model proposes the retention of an adsorbate (Qt) varying linearly with t1/2 (see Fig. 10). The parameters of this model are listed in Table 4.
Table 4
Intraparticle diffusion parameters for adsorption of NGB and copper onto SBA-15 APTES.
Pollutant
|
Intraparticle diffusion model
|
l
|
kid(mg.g− 1.min− 1/2)
|
R2
|
NGB
|
34.64
|
7.85
|
0.934
|
Copper
|
22.87
|
8.52
|
0.970
|
The initial, rapid phase of adsorption corresponds to the diffusion of pollutants molecules onto the external surface of the adsorbent. The subsequent, slower phase is attributed to intraparticle diffusion, which indicated the rate-limiting step of the adsorption process. Finally, the last stage shows a plateau observed at equilibrium which signifies the achievement of a steady state, where molecules diffusion is restricted to the inner pores of the adsorbent particles. (Kumar et al., 2010)
3.2.3 Effect of initial concentration and isotherm study
Adsorption isotherms are often used to determine the interaction between adsorbates (pollutants) and adsorbent surfaces (synthesized materials). These experimental techniques enable the determination of the maximum binding capacity of a material for a specific pollutant, as well as to identify the type of adsorption. (Foo & Hameed, 2010)
For a comprehensive understanding of the adsorption isotherm study Langmuir, Freundlich and Sips models were used. (see Fig. 11) Formulated according to Eqs. (5), (6) and (7), respectively.(Al-Ghouti & Da’ana, 2020; Tamer et al., 2022)
$${Q}_{e}={Q}_{m}\times \frac{ {K}_{L}\times {C}_{e}}{(1+{K}_{L}\times {C}_{e})}$$
5
$$Qe={K}_{F} \times {{C}_{e}}^{1/n}$$
6
$${Q}_{e}= {Q}_{m}\frac{{K}_{S}\times {{C}_{e}}^{\beta } }{\begin{array}{c} \left(1+ {K}_{S} \times {C}_{e}^{\beta }\right) \\ \end{array}}$$
7
Where Ce (g.mg− 1) is the equilibrium concentration of the pollutant in solution, Qe (mg.g− 1) is the adsorbed amount of the pollutant at equilibrium concentration, Qmax is the maximum adsorption capacity of adsorbent and KL, KF and KS are the constants of Langmuir, Freundlich and Sips respectively; the obtained isotherms parameters are reported in Table 5.
Table 5
Kinetic parameters of the adsorption of NGB dye and copper onto SBA-15 APTES.
Pollutant
|
Langmuir
|
Freundlich
|
Sips
|
Qmax(mg.g− 1)
|
KL(L.mg− 1)
|
R2
|
KF(L.mg− 1)
|
1/n
|
R2
|
Qmax(mg.g− 1)
|
Ks(g.L− 1)
|
ß
|
R2
|
NGB
|
227.25
|
0.07
|
0.996
|
48.90
|
0.26
|
0.896
|
227.22
|
0.007
|
1.0007
|
0.995
|
Copper
|
219.05
|
0.066
|
0.995
|
43.46
|
0.3
|
0.911
|
221.006
|
0.07
|
0.97
|
0.995
|
In summary of the results collected in Table 5, and, under the experimental conditions of this tests, the correlation value (R2) calculated using the Langmuir isotherm is equivalent to 0.996 for the adsorption of NGB dye with a maximum removal of 227.25 mg/g, which is higher compared to those of Sips and Freundlich, indicating the best adsorption capacity of the selected dye contaminant by this model.
For copper removal, the value of R2 is 0.995 with a maximum adsorption of 221.006 mg/g indicating that this model is better fitted with the removal of the selected heavy metal. The Sips model offers a unified approach combining the Langmuir and Freundlich isotherms. At low adsorbate concentrations, the model approximates the Freundlich isotherm, implying a dominance of energetically heterogeneous site interactions, characteristic of chemisorption. Conversely, at high concentrations, the model approaches the Langmuir isotherm, predicting the formation of a saturated monolayer, characteristic of physisorption.(Cárdenas et al., 2022; Krishna Prasad & Srivastava, 2009)
The fundamental characteristic of the Langmuir isotherm can be expressed in terms of a separation factor with a constant given by the following Eq. 5:
$${\text{R}}_{\text{L}}=\frac{1}{(1+{K}_{L}\times {\text{C}}_{0})}$$
5
According to the results collected in Table 5 and further calculation, the separation factor is 0 < RL< 1; for copper is equal to 0.222 and for NGB is 0.233, meaning that adsorption of both of the substances is a favorable process, and those of (1/n) are less than 1, referring to even more favorable adsorption at low concentrations.
3.2.4 pH’s effect on pollutants removal
Variation in pH levels can result in alteration of the equilibrium characteristics of the reaction, which can influence the adsorption process. To evaluate the impact of pH on copper and NGB dye adsorption from aqueous solutions onto the APTES functionalized SBA-15 adsorbent, a series of tests were carried out over a pH interval ranging from 2 to 9 (see Fig. 12).
Experimental results for copper and NGB adsorption onto functionalized SBA-15 suggest a maximum removal of approximately 76% at pH = 5.2. With increasing pH values up to 7 and 9 for copper and NGB, respectively, adsorption capacity decreased as a consequence of pore blockage by precipitation resulting from deprotonation of NH3+ ions to NH2 (basic environment) (Y. Wu et al., 2014) The highest adsorption capacity was reported increasing till pH values of approximately 5, 6 and 7 for copper and pH level 7 for NGB.
The result collected from the pH study aligns with those of the zeta potential analysis (see Fig. 5); according to the reported results, the IEP was between 5.9 and 6, indicating that the adsorbent surface was positive at pH values < 6, furthermore, within this pH region strong electrostatic interaction governs the interaction between the negatively charged anionic dye and the positively charged adsorbent surface. (Attallah et al., 2013; Da’na & Sayari, 2011)