1. Introduction
As the main residue of the sugar industry, sugarcane bagasse ash (SCBA) has the disadvantage of an accumulation derived from its low demand. This residue is currently used as fertilizer in sugarcane plantations, however, only in Mexico, approximately 500 kt of this waste is produced annually. SCBA is not merely an inorganic residue, since it retains a carbon content in its composition, so its accumulation leads to bad environmental practices such as burning the ash in the open. Considering the high concentration of SiO
2 in SCBA, its use has been proposed in the synthesis of ceramic materials, mainly light blocks, tiles, and glass-ceramics, however, the problem of increased porosity in the materials arises, which is explained by the remaining carbon which generates CO
2 during the synthesis of the materials. This porosity decreases the mechanical properties and in turn increases the absorption of water, which, in environments where the temperature reaches ranges lower than the freezing point, it can cause the material to collapse [
1,
2].
In the case of glasses and glass ceramics, performance problems are caused due to replacing the SCBA with alkaline carbonates that act as fluxing agents, the decomposition of said carbonates and the burning of the remaining carbon in the ash promote the generation of CO
2, and therefore pores in the vitreous material. The pores or bubbles in the glass material can decrease if the temperature is increased enough so that the viscosity of the glass allows the flow of gas, but this implies a high energy consumption [
3,
4,
5].
Instead to synthetize glass with SCBA, there are other products with diverse applications such like sodium silicate which can be used as binder in concrete, synthesis of aerogels for water remediation, raw material in cosmetic and paper industries, in synthesis of high-performance magnets and the synthesis of anticorrosive coatings for metals and hydrophobic coatings for ceramics [
6,
7,
8,
9,
10]. With the drawback that SCBA does have a diverse and complex composition, its amount of SiO
2 is not always the same and its other components can interfere with the synthesis. However, unlike the synthesis of glasses, Na
2SiO
3 requires lower temperatures for its synthesis regardless of whether Na
2CO
3 or Na
2O are used as reagents.
The common process in the synthesis of sodium silicate is the thermochemical method, which consists of the reaction of silica with sodium hydroxide (NaOH) at temperatures above 300 °C[
11]. However, to isolate SiO
2 from ashes, the use of HNO
3 has been reported [
12], suggesting an impact on both principles 4 and 6 of green chemistry[
13]. Besides since the synthesis temperature is lower than the temperature used in other methods, there are some inconveniences with the use of NaOH. The first one is that it is a highly corrosive agent which can damage steel alloys [
14], the second is that not all sodium hydroxide reacts during synthesis, so it must be neutralized using chlorine gas and finally the use of NaOH represents significant damage to the environment and human health as reported by Ingrao [
15].
In this work, the use of the ceramic method is proposed which only impact principle 6 of green chemistry. For the synthesis, pellets were prepared with SCBA together with sodium carbonate in a 1: 1 molar ratio between silicon oxide and sodium oxide. The use of a carbonate instead an oxide has some implications such as stoichiometric calculations, the increase of reaction time because the carbonate must be decomposed previous the reaction and the subsequent release of CO
2 to the atmosphere because of decomposition (Equation 1). Since the applications of sodium silicate are focused on the solution[
9,
10,
16], the increase in porosity and the mechanical properties of the material obtained are not relevant, so in this case, the disadvantages presented using SCBA, have no relevant impact on the result. Since the composition of the ash varies according to geographical location, a characterization is necessary prior to its use. In addition to this, as it was mentioned before, traces of carbon and other elements were found in the ash, so it was necessary a purification process.
3. Results and discussions
As it was reported in our previous work[
19], according to the elemental analysis before and after the calcination, a decrease of a %wt/wt. from 15.86 to 0.12% of carbon content was observed, meaning an almost total elimination. From the XRF analysis (
Table 1) it is possible to identify the main inorganic components in SCBA, in which, silicon oxide is the main component and calcium stands out as the second major component. The high concentration of calcium becomes relevant considering that the ionic radii of Ca
2+ and Na
+ are very similar (1.14 and 1.16 Å respectively), which may imply competition in the selectivity on crystallization reactions of silicates [
20]. Citric acid leaching process was suggested to reduce calcium concentration from ash.
In the pressure test of the pellets of SCBA and sodium carbonate, with higher time and pressure, the pellets become fragile. This behavior can be explained due to the increase in pressure, as the pressure between the powder and the dice is greater, the friction between the pellet and the walls of the dice becomes greater, culminating in the destruction of the pellet during the procedure to release it from the dice [
21]. Thus, the best result corresponds to a pressure of two tons which is the lowest pressure tested and a time of 5 minutes. At these conditions the stress over the pellet was lowest and the sample was easily removed from the dice.
Figure 1 shows the diffractogram corresponding to the SCBA, where quartz is observed as the main crystalline phase, this result is different from that reported by Sultana[
17], where at a higher temperature the predominant crystalline phase in the SCBA was reported as cristobalite after a heat treatment at 800 °C for 2 hours. Even so, it could be inferred that the cristobalite formed during calcination reacted with CaO present in the SCBA forming the calcium silicate as it’s observed in the diffractogram. Quartz has an interstitial space in its cell of approximately 3 Å similar to interstitial space of cristobalite [
22]. 3 Å is considered as sufficient distance for the sodium and calcium ions to be able to diffuse through the silicon dioxide crystalline cell, since calcium silicate appears as a secondary phase, it is possible to suggest that other elements react with the silicon dioxide of the ash besides sodium oxide.
Considering that the reaction was carried out in solid state, the way it occurs is through a diffusion mechanism between the crystalline cells of silicon dioxide and the other reactant oxides. If the diameter is similar for both ions (sodium and calcium), a kind of competition can be generated to occupy the interstices in the silicon oxide network. Thus, allowing the generation of both sodium silicate and sodium-calcium silicate, as shown in
Figure 2 [
23].
It is worth mentioning that even though the presence of disodium calcium silicate is an inconvenience for this project, this material is identified as a bioceramic with important applications in bone regeneration and localized drug delivery [
24]. To increase the ratio of sodium silicate in the synthesized material, samples were synthesized in a molar ratio of 1:2 SiO
2-Na
2O, but a clear lack of reaction was observed, in the FTIR spectrum of
Figure 3. The signal at 1500 cm
-1 is characteristic of the bond C=O present in carbonate, the lack of reaction can be explained due to the thermal stability of the internal carbonate, requiring a higher temperature or a longer synthesis time considering the heat transfer phenomena. Even so, the presence of the signal at 950 cm
-1 corresponds to the Na-O-Si bond, suggesting the presence of a chemical reaction[
25]. About the signals between the range of 500 to 880 cm
-1 correspond to the bending of the C=O bond outside (600 to 700 cm
-1) and inside the plane (840 to 880 cm
-1), while the signal around 1050 to 1100 cm
-1 it is characteristic of the stretching of the Si-O bond [
25,
26]. The most important signals are those corresponding to 1500 and 950 cm
-1, because although the signal at 1050-1100 cm
-1 also corresponds to the stretching of the Si-O bond, it can refer to both the presence of SiO
2 like some silicate. The signal at 1500 cm
-1 is typical of the C=O bond while the signals at a lower wavenumber may correspond to other bonds such as the bending of the Si-O bond [
27].
As per the results of the synthesis developed with the untreated ash (
Figure 2), the use of a purification technique for the ash was proposed, carrying out a leaching process with citric acid at a concentration of 2% wt/v at a temperature of 60 °C for 2h[
19]. Citric acid, unlike the nitric acid used to extract SiO
2 from the ashes, is a weak acid with a low environmental impact. In this way, the calcium concentration was reduced, achieving a reduction of approximately 5% of the calcium in the ash (
Table 1), and obtaining a purer sodium silicate as observed in the diffractogram of
Figure 4.
With sodium silicate as the main crystalline phase, it is possible to separate both silicates considering that disodium calcium silicate is a water-insoluble compound, so a decantation and filtering process would allow obtaining a solution of sodium silicate. Unlike Vinai’s work [
11], which used the thermochemical method with NaOH, the results obtained in this research present a single crystalline phase of sodium silicate Na
2SiO
3, and in Vinai’s report, there are signs corresponding to three polymorphs of sodium silicate. The importance of the homogeneity of the crystalline phases lies in the fact that, in the case of sodium silicate, each crystalline phase has different applications and properties.
Before presenting the results of the Rietveld refinement, it should be considered, in addition to the results presented, that the XRD technique has a detection limit of 5%, so the presence of other crystalline phases after heat treatment is possible, such as a possible potassium silicate, since observing what is presented in
Table 1 as well as the melting temperature of potassium oxide (740 °C) it is possible that a synthesis process occurred in parallel.
The Rietveld refinement was carried out starting from the results obtained from the XRD analysis by means of the software “Match3!” and making use of the Maud software. For this process, in addition to analyzing the results of the synthesis after 3 hours of reaction, the behavior of the reaction was examined from 0.5 to 3 hours.
Figure 5 shows the diffractograms corresponding to this analysis.
From
Figure 5, letter B represents the serial experiments while 800 is the synthesis temperature and the last numbers the reaction represented time from 0.5 to 3 hours. It is possible to observe how starting from a diffractogram with a considerable noise content corresponding to amorphous signals coming from the ash, the intensity of the signals becomes more and more intense until there are no significant differences between 2.5 and 3 hours of elapsed time, for The Rietveld refinement identified the crystalline phases presented in
Table 2. Unlike the other diffractograms presented, the CIF identifications were used instead of JCPDS because the Maud software only admitted the crystallographic sheets corresponding to the CIF files.
Unlike the other diffractograms presented, the CIF identifications were used instead of JCPDS because the Maud software only admitted the crystallographic sheets corresponding to the CIF files. However, the identifications correspond to the same phases identified with their respective JCPDS number.
Table 3 shows the compositions corresponding to the crystalline phases identified at each reaction time. For the construction of
Table 3, the different SiO
2 polymorphs were integrated into one because when analyzing them individually, confusing information is generated due to the constant change between polymorphs due to temperature, the same behavior studied by Sultana [
17].
The presence of sodium superoxide in the first half hour of reaction, despite being an unstable crystalline phase under reaction conditions, has been reported as a common intermediate in the formation of Na
2O from Na
2CO
3, according to what was reported by Maheshwari [
28], even thus, this intermediate disappears immediately to produce Na
2O. After 2.5 hours of reaction, the presence of SiO
2 is still observed, which disappears after half an hour, since the amount of unreacted calcium present is unknown, it is hypothesized that part of the Na
2CaSiO
4 decomposes to give rise to Na
2SiO
3 while that the rest of the SiO
2 continues to react with the calcium present in the ash. This hypothesis is unfounded considering that if Na
2CaSiO
4 decomposed to produce Na
2SiO
3, the amount of Na
2SiO
4 would be less than the 19% reported at the end of the experiment.