3.1. Bioadsorbent characterization
Thermal analysis by TGA-DTA for pine sawdust is presented in Fig. 1 It is very important to know the behaviour of biomass as the temperature rises, since this is closely related to the properties that the brick will reach when sintered.
The DTA diagram shows an endothermic peak at 52 °C and a small exothermic peak at 263 °C assigned to the loss of water from the sample and the combustion of volatile components, respectively. In addition, in this diagram it is also possible to recognize two exothermic peaks at 327 °C and 488 °C, attributed to the decomposition of hemicellulose (roasted in wood called active pyrolysis), and decomposition of cellulose (passive pyrolysis) and lignin (active and passive pyrolysis), respectively.
It is possible to observe three major mass losses in the TGA diagram, a first loss up 230 °C corresponding to the loss of moisture and the decomposition of volatile components, a second loss up to 311 °C assigned to hemicellulose (to give lower weight compounds molecular, mainly acetic acid), and, finally, a loss of mass up to 502 °C assigned to cellulose and lignin (to finally give CO2, H2O and ash). Hemicellulose combustion occurs at lower temperatures due to its linear structure with short side chains. Cellulose and lignin have more complex and stronger structures, with associated dispositions and aromatic compounds, so they have greater resistance to heat.
These results demonstrate that, during the sintering of the ceramic pieces, the biomass added to the clay mix the whole will gradually calcine as the temperature of the oven increases. This will allow the gases formed to diffuse slowly and the fired bricks will not crack. From the data obtained from the TGA analysis, it is possible to obtain an estimated composition for pine sawdust residues. This composition is detailed in Table 1. In particular, the value of inorganic ash was also carried out by the ASTM E 1755-01 standard method and the value obtained, 2.9%, is similar to that obtained by thermogravimetric analysis.
Figure 2 shows two SEM images of pine sawdust waste used in adsorption experiments. The images show that pine sawdust has a fibrous and irregular structure with pores smaller than 50 microns and elongated particles. These surface characteristics could make the adsorption of heavy metals possible.
Figure 3 shows the FTIR spectrum of the biomass. FTIR was used for the qualitative analysis of the biomass residues of pine sawdust and to identify the functional groups that could participate in the adsorption of contaminating metals. It is possible to observe the presence of bands corresponding to stretches O-H (3344 cm-1) of hydroxyl groups of phenol, C-H of aromatic methoxide group (2922 cm-1), and vibration of aromatic ring (1508 cm-1), assigned to lignin; stretching C=O (1600 cm-1), and C-OH and C-H (1103 cm-1), assigned to cellulose and lignin; C-H flexion of the methyl group (1327 cm-1), and stretching C-O, C=C and C-C-O (1032 cm-1), assigned to hemicellulose, cellulose and lignin; and asymmetric stretching C-O-C (1158 cm-1) and glyosidic bond (830 cm-1), corresponding cellulose and hemicellulose. Similar results have been reported in bibliography [31]. The observed peak around 2359 cm-1 of wave number corresponds to the residual signal of the asymmetric stretch O=C=O of the CO2 that the target failed to completely eliminate.
Figure 4 shows the XRD diagram of the pine sawdust residue, with the characteristic peaks at 16.2 °, 22.4 ° and 34.8 ° typical of semi-crystalline cellulose. In addition, a peak at 26.5 ° corresponding to SiO2 (quartz) and three other peaks of CaCO3 (calcite) at 24.5 °, 28.2 ° and 38.2 ° are observed in the diagram.
3.2. Qualitative study by XRF of the adsorption of Ni(II) ions as a function of time
Figure 5 shows how the amount of Ni(II) adsorbed increases as the stirring time between the Ni(II) solution and the pine sawdust biomass increases. This process was performed to estimate the optimal contact time for a certain concentration of heavy metal and a certain amount of sawdust. The analysis was performed by XRF on the solids obtained after remaining in contact with the NiCl2 solution during each of the stirring times studied. Figure 5 shows that Ni(II) adsorption on pine sawdust increases with contact time because, initially, there is a greater availability of active biomass sites. After a certain time, a plateau is reached which represents a stability in the adsorption of nickel ions. This shows a successive occupation of the active sites of pine sawdust by the pollutant, which causes, after a time, the adsorption to decrease. According to the results obtained, the stirring time for the adsorption tests was 24 h to ensure arrival at the plateau.
3.3. Effect of the initial concentration metal and dose of bioadsorbent
Ni(II) adsorption tests were carried out by modifying the initial concentration of the metallic solutions, but maintaining the volume of the solution, and studying adsorption on 3 different amounts of biomass, 10, 20 and 40 g L-1. After a 24 h stirring time, the mixtures were filtered and the remaining solutions were analysed to determine their residual nickel concentration. Figure 6a shows the relationship between adsorbent dose, initial adsorbate concentration, and Ni(II) adsorption efficiency. R values are higher for low NiCl2 concentrations because there are many binding sites available in pine sawdust biomass but then decrease for all residue doses studied. In Fig. 6b it is possible to observe how at higher initial concentrations of NiCl2 greater adsorption capacity. As the concentration of metal ions increased, the collisions between these and the adsorbent increased. Finally, it seems that at high metal concentrations there is a tendency to reach a plateau. According to Fig. 6b, in the concentration range studied, at the concentration of 1 mol L-1 NiCl2 (5.87 x 104 mg L-1 Ni+ 2) the adsorption capacity is greater. This value increases when the adsorbent dose of 20 g L-1 is used. Furthermore, it is observed in Fig. 6a that when the amount of biomass doubles from 10 g L-1 to 20 g L-1, the R increases 2.5 times, but when the sawdust dose is doubled again, the R increases only 1.5 times. Therefore, the adsorbate/adsorbent ratio of 1 mol L-1 NiCl2 and 20 g L-1 sawdust was considered the most efficient and was then used in the construction of clay bricks.
Figure 7 shows qualitative XRF analysis of the amount of Ni(II) absorbed in the solids obtained after contact with NiCl2 solutions for a concentration range of 0.125 to 1 mol L-1 of NiCl2 using a dose of 20 g L-1 of adsorbent. As the initial concentration of NiCl2 increases, the nickel adsorbed on the pine sawdust increases and there is a tendency to reach a plateau that marks the saturation of the biomass. This NiCl2 concentration is 1 mol L-1.
In Fig. 8, the FTIR spectrum of pine sawdust is compared with that measured after 24 h of stirring with a solution of 1 mol L-1 of NiCl2. A slight change in the position and intensity of the bands in the biomass residue can be observed after the metal adsorption, due to the probable complexation or electrostatic interaction reactions and the Van der Waals forces. It is important to highlight the evidence of new bands at 3487, 1606, 717 cm-1 (indicated with * in Fig. 8), visible from the 1 mol L-1 concentration of NiCl2 and belonging to this compound, giving an indication of its adsorption as such in the biomass.
The Fig. 9 shows the XRD of sawdust after the sorption process with 1 mol L-1 NiCl2. It is from this concentration of NiCl2 that it can be seen, in addition to the peaks corresponding to sawdust, peaks belonging to NiCl2(H2O)2 in concordance with what was observed in Fig. 8.
The biomass electron micrographs after adsorption are shown in Fig. 10. Agglomerates of pine sawdust particles exceed 25 microns. From EDS, the homogeneous presence of adsorbed Ni can be identified. Semi-quantitative chemical analysis by EDS for the pine sawdust residues after the adsorption process shows that the percentage composition by mass, without considering the carbon content in the sample, is 22.88% oxygen, 0.18% silicon, 36.72% chlorine and 40.22% nickel.
3.4. Characterization and evaluation of the bricks
The composition of the clay used as raw material in the manufacture of the bricks is the same as that used by a local brick maker and is shown in Fig. 11.
Figure 12 shows the physical appearance of the manufactured sintered ceramic pieces, (a) without added residue (ARC), and (b) with added 20% by volume of sawdust containing nickel adsorbed (AN20). Each AN20 brick weighs approximately 90 g and an estimate of its nickel content from Fig. 6 is 0.8 g. Both pieces are compact, but ARC has a reddish color due to the Fe present in the clay, while AN20 has a darker color that could be contributed by high levels Ni(II) retained on the biomass and then immobilized in the cooked brick.
Figure 13 shows the DRX obtained for AN20 after cooking. It is possible to observe the presence of SiO2 together with peaks corresponding to numerous secondary phases because the clay used was obtained from natural quarries. Furthermore, it is possible to observe peaks corresponding to NiO.
Figure 14 shows the XRF diagrams of the AN20 and ARC bricks together with a commercial brick (COM). The presence of lines corresponding to nickel corroborates that the metal remains in the clay matrix after the sintering process.
The EDS images of powder AN20 are shown in Fig. 15. These images allow to corroborate the presence and immobilization of Ni after the sintering process. Semi-quantitative chemical analysis by EDS for AN20 shows that the percentage composition by mass, without considering the carbon content in the sample, is 42.04% oxygen, 0.79% sodium, 1.07% magnesium, 11.48% aluminum, 27.88% silicon, 2.25% potassium, 1.77% calcium, 0.61% titanium, 8.32% iron and 3.79% nickel.
Table 2 shows the average values obtained from LOI and the parameters determined from the apparent porosity test of the manufactured bricks. The LOI was greater for AN20 than for ARC due to the combustion, during sintering, of the aggregate biomass. This combustion of sawdust waste generates a greater apP observed in ceramic pieces made from clay and the addition of 20% biomass. Higher values of apP and H2OAbs determine lower values of apD and apsW for AN20. In addition, the mechanical properties of the AN20 and ARC ceramic matrices were evaluated and the average values obtained are presented in Table 2. The average values of σstr obtained are within the commercially required standards according to IRAM 12566-1:2005, "the bricks must have a characteristic compressive strength equal to or greater than 4 MPa". In addition, in Argentina, CIRSOC 501R - 2007 regulates the use in construction, and sets a lower limit of 5 MPa. These σstr values were lower for AN20 than for ARC due to aggregate biomass. This same behavior is observed in MOR. According to the literature [32], the MOR varies between 10 and 30% of the σstr, so AN20 still has a MOR value within the appropriate parameters of the market (MOR = 23 % σrot).
Very similar values of these properties were obtained for the brick manufactured from sawdust with Zn(II) adsorbed (AZ20) [33]. Also, in Table 2, the values for a COM are presented. AN20 has superior physical and mechanical characteristics compared to COM.
A common concern when adding contaminated waste to fired brick is the possible migration of these contaminants into the environment. In this sense, it is important to evaluate the leachability of the manufactured AN20 ceramic matrices. National Law 24051 and its regulatory decree establish that leachates in dangerous products must be analyzed in accordance with TCLP. The measurements were obtained in triplicate following the protocol (using fragments of AN20 and a solution of pH 4.93 ± 0.05 as extraction solvent, determined from the sample's alkalinity according to TCLP). Table 3 shows the average Ni(II) result obtained in the leached liquid. This value was less than 0.05 mg L-1 nickel and was compared with the maximum concentrations allowed according to Decree No. 2020 of 2007 of Law No. 2214 of the City of Buenos Aires (CABA), US EPA Code of Federal Regulations 2012 and the regulations in force in Spain and China. Therefore, the metal concentrations obtained by TCLP of AN20 are below the established maximum. A previous investigation has allowed obtaining similar behavior in AZ20 bricks [33]. These tests demonstrate that during the brick firing process, biomass residues are destroyed and heavy metals could encapsulate in the clay matrix, form oxides, and interact with silicon, iron, or aluminum oxides in the clay to form minerals stable, according to the literature [34, 35]. The phase transformations and the evolution of the heavy metals during the firing of the ceramic pieces determine the fixation of the contaminants. However, the mechanism of immobilization of heavy metals during the sintering process has not yet been understood [36].
The calculation of the efficiency of the retention is made according to the Eq. 3 [37]
See formula 3 in the supplementary files.
where, ER = efficiency retention of metal in the brick (%), MMIB = mass of metal incorporated in AN20 and MML = mass of metal extracted in the leachate. MMIB is estimated from the nickel mass adsorbed by pine sawdust in the ratio of highest efficiency selected and corrected for the mass of sawdust added as pore-forming agent in AN20. MML is estimated from the nickel concentration in the leachate corrected for the mass of the entire brick. ER for AN20 bricks is 99.99%.
Therefore, it is possible to immobilize Ni(II) in fired clay bricks and thus reuse the pine sawdust previously used as an adsorbent of these contaminants. In addition, a common paint such as latex, could act as an additional "barrier" that guarantees the non-leaching of Ni from inside the brick [28]