3. Results
Analysis of SEM: The evaluation of SEM (Universidad de los Andes, Bogotá D.C., Colombia) shows the characterization of the aquatic plant
E crassipes, observing large amounts of carbon and oxygen, and the other elements evaluated showed traces, resulting in robustness and similar characteristics to those found in [
24].
Figure 2 shows a characteristic image of the biomass of EC, where a porosity with a rough surface texture is observed.
Table 2 shows the physicochemical characterization of the EC sample, obtained through EDS.
In the microphotography of
Figure 2b, different colored dots can be seen that represent the elements in the samples: blue dots represent carbon, yellow dots represent oxygen, and green dots represent the places where chromium (VI) lodged.
In
Figure 3, different colored dots can be seen that represent the elements in the samples: green dots represent carbon, red dots represent oxygen, and yellow dots represent the places where iron lodged. Similar observations show that iron remains attached to the biomass of EC to help the chemisorption process [
22].
Additionally, in the micrograph (
Figure 3b) of ECFe with Cr (VI), it is possible to confirm this information because you can see points of different colors that represent the elements in the samples: yellow points represent iron, red points represent oxygen, green points represent carbon, and purple points represent the places where chromium (VI) is present.
The use of iron (Fe) (III) chloride in vegetable cellulose has been used for the treatment of organic and inorganic contaminants. The (Fe) (III) reacts with hydratable hydroxyl cellulose, forming iron hydroxides (FeOOH); these are responsible for the cation exchange with heavy metals. The metal ions enter the interior of
E. crassipes with FeOOH, exchanging with protons of hydroxyl groups. The ionic interaction was mainly responsible for the adsorption of As (III), As (VI) and Cr (IV).
Figure 4 shows the reaction of cellulose with iron chloride (FeCl), forming iron hydroxides (FeOH) [
22,
23], represented in
Figure 4.
Table 3 shows the characterization of the elements, and it can be observed that 8% of the weight in the sample was iron.
When reacting with cellulose, iron chloride (FeCl
3) progressively oxidizes it, creating more active sites for heavy metal adsorption [
22]. Chlorine reacts with the hydrogen in the biomass, creating (HCl) compounds; this is the reason why this biomass of ECFe tends to be acidic [
23].
The elimination of metal ion species by Fe was explained by their adsorption on the surface of the corrosion of the biomass of the original Fe particles. Therefore, when contacting a solution, Fe (0) gradually oxidizes plant cellulose, allowing the ions of different heavy metals to form internal sphere complexes with oxidized sites [
25].
The pollutant Cr (VI), in the form of dichromate (Cr
2O
7), has a complex chemical structure; when it meets the biomass of ECFe there are reactions of the (H
+) of the biomass together with the oxygens of the structure of chromium (VI), reducing it to Cr (III) and chromium oxide (Cr
2O
3) through two mechanisms. The first mechanism consists of direct reduction, in which the Cr (VI) was reduced to Cr (III) in the aqueous phase through contact with the electron donor groups present in the biomass, such as (OH) [
26] (see
Figure 5).
Figure 5 is a representation of the reduction of Cr (VI).
The second mechanism is indirect reduction, in which three stages are identified: (1) the binding of the anionic species of Cr (VI), with the functional groups present on the surface of the biomass being positively charged; (2) the reduction from Cr (VI) to Cr (III) via the adjacent electron donor groups; and, finally, (3) the release of Cr (III) ions in the aqueous phase due to the repulsion between the positively charged groups and the Cr (III) ions, or by the complexation of Cr (III) with adjacent groups capable of binding Cr. In conclusion, the mechanism of the removal of Cr (VI) by biomaterials is a combined mechanism between adsorption and reduction [
27,
28]. In
Figure 6, the process of the adsorption of Cr (III) by iron-modified cellulose can be seen.
Through
Figure 6, it is shown that when reacting with the cellulose, the iron chloride, FeCl
3, progressively oxidizes it, creating active sites for the adsorption of heavy metals; the chlorine reacts with the hydrogen of the biomass, creating compounds (HCl). The representative weight of iron in this ECf biomass is 8.3% (see
Table 3).
Table 4 shows the physicochemical characterization of the ECFe sample with Cr (VI).
The relationship between cellulose ECFe and Cr (VI) is shown in
Table 4, with weights of 7.4 for Cr (VI) and 7.9 for Fe. The chemical reaction is represented through the following:
where ({3(C
6H
10O
6)} *4FeOOH) represents the ECFe biomass and Cr
2O
7−2 represents Cr (VI); it can be seen from both
Figure 5 and
Figure 6 that three parts of glucose react with dichromate.
Before the process of recycling the biomass with EDTA, microphotography was performed with the goal of knowing about the biomass modification.
Figure 7 shows the biomass before the process of elution with EDTA.
Figure 7 shows how EDTA delayed the Cr (VI) in the water; this sample is in the first process of elution with EDTA.
Table 5 shows the physicochemical characterization of the EC sample after the recycling process.
Table 5 shows how the Cr (VI) is in form of 0.14%, showing the elimination with EDTA and the efficiency of this eluyent chemistry.
Mass balance in the biofilter. The biofilter is a treatment system in which biochemical diffusion processes occur inside it (see
Figure 8).
In the process of the treatment representated in
Figure 8, water contaminated with Cr (VI) is imperative for the adjusted design to be able to remove a significant amount of contamination. Important goals of the process are the retention, adsorption, and velocity of entering of system.
Adsorption is a special process in a treatment system with an adsorbent and biomass; the cationic interchange between the biomass of the EC and the Cr (VI) is represented through the following differential equation:
where
is the ablity of the biomass to adsorb to remove heavy metals, determined by the load of mailgrams of heavy metals above the grams of biomass used.
The density,
, of the bed of the adsorptions is a principle parameter of the design; due to the relationship between it and the biomass use under the volume of this, it must have a value under 0.7 gr/mL with the goal of having better space, for the contaminate is lodged through the particle and favors adsorption [
13]. The relationship between the density of the bed of treatment,
, and the density of the micro particle of this biomass,
, is represented through the following:
where
is determined by the following:
The mass of the microparticle is its weight, and the volume of the microparticle is obtained through the following:
where r is the radius of the biomass microparticles used in the treatment process [
29], which is related to the densities and established that the relationship between
pb/pp should be less than 0.5; in other terms, ε values greater than 0.6.
To comply with this design parameter of treatment processes through chemisorption, it is essential to use particle diameters less than 0.212 mm; with this diameter there would be a direct relationship with the pollutant particle.
The retention or accumulation in the treatment process is dependent on the relationship between the densities and the concentration of the pollutant that enters the treatment system:
where
is the entry of the pollutant into the treatment system in terms of its initial concentration. The input to the treatment system is represented by the design speed and the amount of the pollutant that the treatment system can treat:
With these equations the general balance of matter will be completed, which is summarized in Equation (7):
Through Equation (8), the most important design parameters when treating water in adsorbent-based treatment processes could be adjusted, leaving Equation (9):
where
V = system volume (mL),
ε = ratio between densities, Co = initial concentration of Cr (VI) (mg/mL), Q = design flow (mL/min), Tb = break time (min),
M = biomass used (g), and
q = adsorption capacity of the biomass used (mg/g).
Depending on the most important parameters when building a treatment system, Equation (9) could be used in order to model and validate the best form of treatment; for example, the necessary amount of biomass to be used to treat a certain amount of initial pollutant was used in the present investigation to establish the adsorption capacity in these initial conditions of treatment.
Equation (10) can be used to determine the adsorption capacity:
To solve this equation, preliminary experiments will have to be carried out in order to establish the relationship between the densities (ε); the capacity to retain chromium (VI) via the biomass of EC will be established later.
In the development of the fixed-bed column Equation (3) was used, with the goal of determining the relationships between densities.
When a desorption-elution process is involved for the reuse of biomass, Equation (10) would be as follows:
where Q = design flow (mL/min), Tbj = break time of use number j (min), Co = initial concentration of Cr (VI) (mg/mL), C = final concentration Cr (VI) in the treated solution (mg/mL), V = system volume (mL), ε = porosity, M = amount of biomass used (g), and q_T = total adsorption capacity of the biomass used (mg/g).
This model, Equation (11), is designed to determine the adsorption capacity when different elution processes have been carried out; it will be used to determine the new adsorption capacity and is one of the contributions of the present investigation. Result of process adsorption: Using Equation (3), we proceeded to develop the initial conditions of the treatment bed of fixed columns with dry and crushed material of EC and ECFe with the three diameter settings.
Figure 9 shows the biomass yields of ECFe, with initial concentrations of 400 mg/L and 100 mg/L, respectively. These results are given by the average of the three pieces of data obtained in the experiment and have the bars of mistake product of standard deviation.
In the particle diameters via the treatment of water contaminated with Cr (VI), the best results were those with a diameter of 0.212 mm, treating around of 2 L of water; results with this particle diameter were reported by the authors of [
12].
In the treatment processes with diameters of 0.3 mm and 0.425 mm, their removals were below that of a diameter of 0.25 mm; it has been evidenced that the smaller the particle, the better the contact with the contaminant, processes that were seen to be of equal magnitude in the treatment of EC (
Figure 10).
In this process, remotions lower than those of ECFe were given, due to the fact that iron oxide, the organic material of cellulose, created more hydroxyl groups. The most significant diameter was 0.25 mm, though the diameters almost gave the same results in the processing of initial amounts of 100 mg/L.
Using these parameters, we proceeded to consider the relationship of the density of the bed of contact and the density of particles.
With these parameters, the density of particles to the same volume of particles with diameters of 0.25 mm is given:
The bed of treatment has the following relationship between these two densities [
13]:
The densities were related and it was established that the relationship between them should be less than 0.5; in other terms, ε values greater than 0.6.
Table 6 shows all of the parameters used in the experiment.
According to
Table 5, the relationships between these densities are some parameter fundamentals in the design of a system with an adsorbent bed. It is necessary for the density of the bed to have free space, with the objective of assimilating the contaminate, and the density of particles must have a direct relationship with the contaminate. Both types of biomass had the same results in terms of density.
Equation (10) was utilized to establish the capacity of adsorption in the process of treatment in bed adsorptions:
where q: capacity of adsorption, Co: 0.4 mg/mL, M: (ECFe) 50 g, Tb: time rupture (min): 150 min, and Q: caudal (mL/min).
In the treatment with ECFe, it was established that there was a treatment of 2.0 L with a rupture time of 150 min. A new adsorption capacity of 17.6 mg/g was obtained. In the treatment with EC, the graph can establish that there was a treatment of 1.5 L with a rupture time of 70 min, its adsorption capacity being 8.0 mg/g. Through this equation, acquiring a new interpretation of biomass being involved in the retention of heavy metals is possible; with the aim of establishing the viability of this equation, it was validated through the Thomas model.
Mathematical models of adsorption. Through the Thomas model, Equation (10) was validated and the comportment of the process of treatment was established. The Thomas model is used to estimate the maximum adsorption capacity and predict the breakdown curves; assuming the kinetics of second-order reversible reactions and the Langmuir isotherm [
29,
30,
31,
32,
33,
34], the representative equation is Equation (11).
The graphical representation is shown in
Figure 10 of the fit to this mathematical model, where the Thomas constants of all the experimental processes are shown, of the initial 400 and 100 mg/L of Cr (VI), both for the EC and ECFe biomass. Two graphs were obtained for each biomass, and from each of these the adsorption constant was obtained.
Considering the results of
Figure 9 and
Figure 10 of the adsorption process, for both the EC and ECFe biomass, there were interesting adjustments to this mathematical model of Thomas, with all of the R
2 being above 95%. Evidencing an adjustment to Langmuir’s isotherm and second-order kinetics, it could also be argued, in all processes, that there was a diffusivity in a monolayer of the biomass of EC and ECFe. This validated Equation (10), and, with the graph, it can determine that the adsorption capacity (q) and capacity of adsorption of the equation of the design were significant; therefore, both capacities were almost the same.
Table 7 shows the parameters of the equations of Thomas and the equation of the design.
The table appreciated that the validations of the equation of the design were representative and adjusted.
Table 6 and
Table 7 show the summary of the Bohart, Yoon, and Thomas model counts, where they were the best representative adjustments of the different removals of Cr (VI) by the EC and ECFe biomass. In
Table 8, the summary of the EC experiment is shown, where the counts found through the equations of the graphs in
Figure 11 obtained averages around 0.048 (mL/mg∙min), this being their speed of Cr (VI) removal in the biofilter; in [
30], the kinetics of
E. crassipes were the same.
It is seen in
Table 9 that the Thomas constant obtained results in its adsorption rate of 0.068, demonstrating that this biomass adsorbs at a higher rate than the biomass of EC [
31,
32,
33,
34,
35,
36].
Recycling process. The results concerning the elutions with EDTA are represented in
Figure 11.
The original ECFe biomass capacity is 17.6 mg/g, and after the elution in the second process of treatment the capacity was 14.2 mg/g. Though being less, the process was significant; the third process was also important, with a capacity of 9.2 mg/g, finishing the process of the elutions with a capacity of 3 mg/g. It has one summation with Equation (10), validating all of the variables, this equation remains Equation (11):
The EC biomass capacity was originally 8 mg/g, and after the elution in the second process of treatment the capacity was 6.5 mg/g; after the third and quarter process it was 5.6 for both, finishing the process of elutions with 4.2 mg/g. With Equation (11):
The process of recycling with EDTA was very important due to it increasing the remotions of the Cr (VI); in the ECFe biomass it has achieved reuse in more than four processes, increasing the process of treatment 2.7 times. In the case of EC the process increases 3.6 times.
Through Equation (11) and with different bibliographic references, representative data were obtained to feed this equation, determining the capacities of each of these biomasses together with the new capacities determining the desorption power of the different eluents shown and summarized in
Table 10.
For the EDTA eluent and with Equation (10), satisfactory results were evidenced by removing Al (II), reaching almost 150% of its adsorption capacity, corroborating what was presented in the present investigation; additionally, the EDTA reagent obtained interesting yields in recycling cassava biomass, increasing up to 40 mg/g. The authors of [
37] used the biomass of
Phanera vahlii to remove Cr (VI), obtaining results of 30 mg/g, and with NaOH they reached capacities in the reuse process of this biomass up to 62 mg/g, reaching almost double of its total capacity. The authors of [
40] also used NaOH for desorption processes with green-synthesized nanocrystalline chlorapatite biomass, achieving results of 75% more. The eluent HCl is also a good chemical agent to use in desorption processes, since it reached more than 100% in the reuse of biochar alginate for Cr (VI), but it is not so significant with A. barbadensis Miller biomass to remove Ni (II), and in [
39] significant results were also obtained to remove Pb (II) with
pine cone shell biomass. With the chemical agent HNO
3, interesting contaminant recycling processes were obtained, since more than 100% of the adsorption capacity of the biomass used in this process was used [
41].
Mathematical models in recycling processes. The continuous desorption process with its fit to the Thomas model for ECFe biomass always shows the fit of this model with significance, because this type of model fits representatively to desorption processes with good performance [
6]. It can also verify that with values of qt it is close to the experimental values of Equation (10) designed and presented in this research, indicating the validity of this equation again, where it reflects the maximum capacity obtained.
Table 11 shows a summary of the experiments obtained with ECFe material in the process of desorption.
In
Table 12 the EC biomass had a different behavior, and in its second and third cycle it adjusted to the Yoon model and later to the Bohart model.
This behavior is due to the alkalinization of the biomass, and this process makes the biomass a little more unstable. The values of qt, although evidencing a resemblance, were not so representative due to the little adjustment that there was with respect to the Thomas model.