3.3. Particle Size, ζ Potential, and SEM Images of the Clay and Hydrocolloid
The particle size of natural HMB clay showed three groups (
Table 2), 53.0% corresponded to a mean size of 839.5 nm, 46.4% reported 239.8 nm, and 0.6% 37.6 nm. After activation, it was observed that the size decreases considerably, and 97.4% presented a mean size of 487.5 nm.
This decrease in size is due to the treatment of the clays with phosphoric acid, which dissolves the carbonates, eliminates oxides present in the octahedral structure of the native clay, and destroys the mineral part, generating amorphous silica and active sites that will improve the adsorption centers. On the other hand, the acid action during activation produces dehydroxylation and the elimination of metal cations from octahedral sites produces new pores suitable for adsorption [
12,
43,
44].
Regarding the CH hydrocolloid, it was observed that 95.8% reported a mean size of 421.7 nm, this size would allow the improvement of the electrostatic and chemical interactions during the adsorption process of heavy metals and the formation of colloids and suspension due to its ζ potential (−27.14 mV) (
Table 3) [
18,
45,
46,
47,
48].
The ζ potential allows knowing the solution stability of powdered materials, the absolute value between 21 to 40 mV indicates medium stability, and <20 mV allows easy agglomeration and sedimentation [
49,
50,
51]. In this study, it was observed that activated clay presents a medium stability (39.91 mV, measured at neutral pH) (
Table 2).
The reported values of ζ potential for HMB-Act and CH allow the establishment of a good electrostatic attraction due to the greater number of carboxyl, hydroxyl, and carbonyl groups, as evidenced in the IR analysis. These results show that these materials can easily hydrate and interact with metal cations [
9,
52,
53,
54]. As for HMB-Act, the activation with NaCl allows the potential to improve, because Na
1+ ions are extending in a diffuse electric double layer, according to the diffuse double layer theory [
9,
55,
56,
57].
The ζ potential is associated with particle size, degree of hydration, chemical nature, surface topography, and charge density on the surface of a material [
52,
57,
58]. High absolute values are an indicator of smaller particle sizes being able to reach nanometric levels, and this was observed in activated clay (
Figure 2).
3.4. Metal Adsorption
It was observed that the adsorption percentage followed the order HMB-Act/CH > Natural HMB/CH > HMB-Act > Natural HMB (
p-value < 0.05), that is, the activation of the clay improves the multimetal adsorption and the addition of the hydrocolloid improves it even more (
Table 3). On the other hand, it was observed that Pb has a better adsorption affinity, removing 78.35% in natural clay, and up to 99.51% for the HMB-Act/CH composite, followed by As (32.32%), Cd (14.16%), and Zn (10.31%), adsorbing up to 108.14 mg/g of Pb in equilibrium at 120 min.
The Pb affinity is due to the competition by the metal ions in the multimetal system for the active sites of the composite materials. This fact gives rise to an antagonistic effect, which largely depends on the ionic radius, hydration radius, hydration enthalpy, and cation solubility [
22,
23,
59]. Thus, Pb
2+ (1.20 Å) presents a higher ionic radius, which justifies its greater adsorption [
40,
60,
61,
62], followed by Cd (0.97 Å) > Zn (0.74 Å) > As (0.47 Å), substituting the Na
1+ ions (0.95 Å) of the three-dimensional network of activated clays [
63,
64]. However, the opposite happens with the hydration radius [
59,
65], although this also depends on the type of adsorbent material, and the initial multimetal concentration decreases when this is higher, due to the overlapping of active sites [
19,
60].
This demonstrates the viability of activated natural clay as an effective adsorbent, due to increased contact area, pore-volume, ion exchange capacity, and decreased particle size [
16,
40,
41], presenting specific adsorption at the edges of the nanocomposite structure, caused by the formation of complexes with the hydroxyl and oxygen groups of Si-O [
40,
66,
67].
Materials formulated with activated clays show adsorption levels of around 95% for Pb [
40,
41,
66]. While for activated materials of plant origin, values of around 90% removal are reported [
7,
17,
25,
30]; thus, the values found are encouraging for the use of this formulated compound.
3.5. IR Analysis of Composites Subjected to Adsorption
Regarding the hydrocolloid CH, a peak was observed around 3400 cm
−1, which corresponds to a vibration of the -OH and -NH bond (
Figure 3a), characteristic of amides and carboxylic acids, which would allow the establishment of hydrogen bridge bonds; at 2927 cm
−1 asymmetric stretching vibrations of the C-H bond are presented, corresponding to the hydrocarbon chains of carbohydrates; another zone with high intensity is found around 1646 cm
−1, it corresponds to a vibration of the stretching of the carbonyl group -C=O, and -OH stretching of the water present, which suggests high hygroscopicity [
68,
69].
Around 1530 cm
−1, low intensity spectra are observed, corresponding to the stretching vibrations of the COO- and -C=O bonds of the carboxylate anions [
70]; at 1416 cm
−1 stretching of the -C-O, -C-H, and -OH single bonds were observed; at 1060 cm
−1 a high intensity peak is presented, which would be due to the manifestation of bond stretching -C-O, C-O-C, C-OH, a peak at 815 cm
−1 of low intensity that indicates deformation of the -CH
2 bond corresponding to methylene groups, while between 800 and 580 cm
−1 different low intensity spectra are presented, this area is known as the “fingerprint” of the materials, these spectra are attributed to stretching of the -C-H and -C-O bonds, belonging to starches and glucose, which is characteristic of hydrocolloids from algae [
13,
27,
70,
71,
72,
73].
Regarding the natural clay (HMB Natural) and activated (HMB-Act) (
Figure 3b), a band at 3694 cm
−1 corresponds to the stretching -OH of the hydroxyl groups of the internal surface (Si-OH) of the tetrahedral layer, while the band around 3621 cm
−1 corresponds to -OH stretching the internal hydroxyl groups (Al-OH) of the octahedral layer, confirming the 2:1 arrangement of the smectite. On the other hand, the band around 3426 cm
−1 and 1629 cm
−1 is due to -OH stretching vibrations of adsorbed water molecules [
74,
75].
The shoulder-type band at 1090 cm
−1 corresponds to Si-O stretch (out of plane), while at 1019 cm
−1 Si-O stretching (in the plane) is present; at 917 cm
−1 the natural clay HMB presents a peak and is attributed to the bending vibrations of AlAlOH, AlFeOH, and AlMgOH, causing deformation of hydroxyl groups on the internal surface [
76,
77]; however, HMB-Act does not present this band, which would confirm the substitution of trivalent and bivalent cations by Na
1+ [
12,
77]. While the peaks 749 and 686 cm
−1 correspond to quartz vibrations present in the clay matrix [
26,
78].
The band at 530 cm
−1 and 466 cm
−1 corresponds to an Al-O-Si deformation and Si-O-Mg and Si-O-Fe vibration bending, respectively [
26], presenting higher intensity for HMB-Act, which shows the substitution of trivalent ions (Al
3+ and Fe
3+) to monovalent Na
1+ in the octahedral sheet [
12,
75].
Regarding the HMB-Act/CH composite (
Figure 3b), a peak with high intensity is observed around 2900, 1500, and 1400 cm
−1, which are attributed to the hydrocolloid CH, and is mainly due to the presence of -OH, -CO, -COO groups, which allows the increase of the active metal adsorption centers [
17,
20,
23].
In materials and composites subjected to adsorption, a considerable decrease in peak intensity was observed around 3400, 2926, 1630, 1430 cm
−1, and in the fingerprint field, mainly for HMB-Act/CH (
Figure 3c). This would be due to the functional groups of the clay and hydrocolloid that would be responsible for the complexation of Pb
2+ and Cd
2+ and Zn
2+, due to the O and N binding atoms [
15,
79,
80]. Likewise, this would be due to the synergistic effect of the functional groups -OH of Al-OH and Si-OH of the octahedral sheet in the activated clay and -NH and -OH of the hydrocolloid [
40,
41,
80,
81,
82].
On the other hand, the appearance of a 1380 cm
−1 peak with high intensity was observed (
Figure 3c), which clearly indicates the complexation of heavy metals, mainly Pb [
17,
80].
In this sense, it can be considered that the formulated composite (HMB-Act/CH) has a high capacity for heavy metal removal, evidencing the synergy of the functional groups of each material, so inorganic materials at the nanoparticulate level and biological materials, such as the
Nostoc Sphearicum algae, present good ability to complex monovalent, bivalent, and trivalent heavy metals [
17,
24,
30,
41,
42,
80,
83,
84,
85,
86].
3.6. Metal Adsorption Kinetics in the Composite
The study of adsorption kinetics is important in the design of adsorption systems, allowing residence times, reaction rates, and reactor sizing being established [
24,
87].
The pseudo first-order model describes the adsorption of liquid–solid phase systems as a function of adsorption capacity, considering that the binding site occupancy rate is proportional to the number of unoccupied sites on the sorbent [
14,
88].
The pseudo second-order model considers the adsorption capacity of solid phases, due to the chemical bonds (chemisorption with strong interactions) in the adsorbent monolayer. It also describes the occupancy rate of the adsorption sites, proportional to the square of the number of unoccupied sites in the sorbent [
24,
88]. The intraparticle diffusion model considers the probability that the adsorbate is transported from a concentrated zone to the adsorbent through diffusion, this being the stage that limits the speed in many adsorption processes, generally for discontinuous agitation processes, where the adsorption varies almost proportionally to t
1/2 at the point of contact time t [
88].
The pseudo first-order model showed values of
R2 > 0.944, and the pseudo second-order model reported
R2 > 0.980 (
Table 4). The fact that the fit of the pseudo first-order model is slightly lower could be attributed to the limitations of the boundary layers that control the physical adsorption processes [
21,
22]. Therefore, the adsorption of metals would be subject to chemisorption processes, which allows a better description of the pseudo second-order model [
16,
23].
It was observed that adsorption takes place rapidly during the first 30 min (
Figure 4), due to the availability of the active sites on the surface of the composite, establishing the complexation of the -NH, -C=O, -OH, Al-O-Si, and Si-O groups of the clay and hydrocolloid (chemisorption process); for longer times, adsorption is very slow, with intraporous adsorption occurring mainly from 90 min onwards. The increase of adsorption is not significant, this behavior is usual for clay materials subjected to Pb, Cd, and Zn adsorption [
16,
80]. The adsorption equilibrium is influenced by the nature of the adsorbent and adsorbate, mainly by the functional groups of the active sites, particle size, ion exchange capacity, and ζ potential [
16,
21].
The kinetic parameters of the pseudo second-order model (
Table 4) reported that q
e (equilibrium adsorption capacity) is higher for Pb, followed by As, Zn, and Cd; demonstrating greater specificity of the composite for Pb, which is characteristic for clay adsorbents [
19,
21,
23,
62,
80,
85,
89,
90].
The adsorption rate,
k2, ranged from 0.002 to 0.025 g/mg·min. These values depend on the nature of the adsorbent and adsorbate and the medium conditions.
qe does not show any behavior, although low values suggest lower adsorption capacity in multimetal systems when treated with inorganic and organic adsorbents [
14,
21,
80,
84,
89,
91].
Adsorption processes can be explained through four processes: (
i) surface migration, (
ii) film diffusion, (
iii) intraparticle or pore diffusion, and (
iv) sorption at interior sites. Stages
i and
iii occur spontaneously due to the availability of metals near the surface, whereas stages
ii and
iv generally control the rate of adsorption, and the intraparticle model makes it possible to explain this phenomenon [
19,
89].
The intraparticular model (
R2 > 0.70) showed that the
kid intraparticulate rate constant is higher for Pb, followed by As > Zn > Cd (
Table 4), suggesting that surface migration and intraparticulate diffusion occur rapidly, especially for Pb and As. While low
kid values would be subject to diffusion in the film and sorption in interior sites of the adsorbent, and the higher hydration radius of Zn and Cd [
65], would be the conditioning factors of the adsorption rate [
89].
On the other hand, the constant
C, related to the thickness of the boundary layer, reported higher values for Pb and As (
Table 4). If
C is equal to zero, the only control step is intraparticle diffusion; if
Ci ≠ 0, it indicates that the adsorption process is quite complex and involves more than one diffusive resistance [
41,
90,
92], confirming the information reported by
kid, which is that the metals diffuse slowly in the pores of the HMB-Act/CH composite, which constitutes a limiting step [
40,
89,
90].
3.7. Metal Adsorption Isotherms in the Composite
The interaction behavior of adsorbate and adsorbent at equilibrium is described through adsorption isotherms [
10,
24].
The Langmuir equation is applicable to homogeneous sorption, each molecule has the same sorption activation energy and is based on the assumptions that (
i) adsorption can only occur at a fixed number of defined localized sites, (
ii) each site may contain only one adsorbate molecule (monolayer) at all sites, and (
iii) there is no interaction between adsorbed molecules even at adjacent sites [
10,
31,
93].
Regarding the Langmuir isotherm (
R2 > 0.92 and chi-sq < 51.99), many multimetal sorption systems have been represented by this model. The
qmax value, which represents the maximum adsorption capacity at the monolayer level [
31] for the nanocomposite, showed selectivity in the order Pb > As > Cd > Zn (
Figure 5), being usual behavior for clays, although the values obtained are higher than those reported elsewhere [
63,
85,
89,
94]. This would be due to the activation of the clay, nanoparticulate size, and the ζ potential that the clay and the hydrocolloid present, giving them stability in suspension, which allows them to be in greater contact with the multimetal solution.
The
KL parameter showed a higher value for Pb (
Table 5), which indicates that the nanocomposite shows a high affinity for this metal due to the assumption of a finite number of identical active sites in the nanocomposite. While As, Cd, and Zn show similar values, this is related to the similar ionic radius that they present. This behavior is characteristic for clays [
23,
63,
64,
85,
89,
94].
The separation constant
RL indicates whether the adsorption system is favorable or unfavorable, when
RL > 1, unfavorable;
RL = 1, linear; 0 <
RL < 1, favorable;
RL = 0, irreversible [
10,
31,
95]. The results found showed favorable adsorption for metals (
Table 6); however, the behavior for Pb at the initial concentrations under study was more favorable, although at the initial concentration of 10 mg/L for the metals As, Cd, and Zn, they tended to be unfavorable due to antagonism with Pb, which has a greater preference for the active sites of the monolayer, caused by a greater ionic radius, which allows it to substitute the Na
1+ ions of tetrahedral structure of the nanoclay and the functional groups of the hydrocolloid in the nanocomposite [
10,
31,
96].
The Freundlich isotherm, which proposes multilayer sorption with a heterogeneous energetic and surface distribution on active sites and/or interactions between sorbed species can be used to describe heterogeneous and multicomponent systems [
63,
97,
98].
The Freundlich model (
R2 > 0. 95 and Chi-sq < 41.71) (
Table 5) reported values of the relative adsorption capacity K
F in the order Pb > Zn > Cd > As. The nanocomposite shows a relative preference for Pb, which should not be confused with the percentage of removal since adsorption is relative to agitation conditions, pH, temperature, particle size, and hydration radius [
22,
23,
59,
63,
94,
97,
99].
The 1
/n parameter describes the nature of the process, with 1
/n < 1 indicating higher saturation, measured as the potential availability of different sorption sites on the nanocomposite surface for the adsorbed metals [
33,
63,
100]. If 0 < 1
/n < 1, it indicates a heterogeneous surface structure with an exponential distribution of the active sites [
41]. According to the reported values (
Table 5), it was observed that the nanocomposite presented a heterogeneous surface, which is due to the surface characteristics of the nanoclay and the atomized nostoc, giving it high saturation for the adsorption of metals. Although a scale of the values of 1
/n has not been reported, it could indicate a rapid saturation of Zn in the available active sites, followed by Pb; however, Zn presented a lower percentage of removal, and this could be attributed to its atomic radius [
60,
61].
On the other hand, values of
n between 1 and 10 represent favorable absorption [
10,
23,
31,
33,
40,
80,
89]. In addition, it was observed that
n reported values that are in the range of 1.64 to 3.35, which suggests favorable adsorption in the nanocomposite.
The Redlich–Peterson empirical model, called the “three-parameter equation”, which makes it possible to represent adsorption equilibria over a wide concentration range [
31,
63,
97], reported
R2 values between 0.83 and 0.99, suggesting a good fit, and would allow criteria to be taken in reactor designs based on the initial concentrations of multimetal mixtures. Although the parameters of this model lack a physicochemical and thermodynamic explanation; a correlation of the parameter
KR and g with the parameters of the Langmuir and Freundlich model, percentage of removal, ionic radius, and functional groups (IR analysis) has not been observed, although the
aR parameter shows a correlation with the percentage of metal removal.