3.1. Alkali-Activated Binder Based on Metakaolin
The material densities and compressive strength of the monolithic AABs are shown in
Table 4. According to the obtained results, the material densities of the AABs ranged from 1.052 to 1.198 g/cm
3 (
Figure 4), depending on the liquid-to-solid ratio (L/S) and curing time. As seen in
Table 4, the samples of series AAB-1.2 exhibited lower material densities (1.052–1.062 g/cm
3) compared to samples of series AAB-0.8 (material density: 1.193–1.198 g/cm
3) depending on curing day. The samples of series AAB-1.0 exhibited material densities of 1.127–1.178 g/cm
3 depending on time.
As seen in
Table 4 and
Figure 5, the studied AABs had a compressive strength ranging from 3.38 to 12.55 MPa, depending on the L/S ratio and curing time. Mechanical properties are very important for zeolite granules due to abrupt gas and packing movements and other mechanical influences in the adsorption bed during the long operational period (at least 3–5 years) [
22,
23]. Typical zeolite granules should have compressive strengths of at least 10 MPa for the gas adsorption-pressure vessels [
24]. The compressive strength of most of the obtained binders reported in this work was not above this compressive-strength threshold.
All samples were cured at ambient pressure and temperature (20–23 °C). Sample series AAB-0.8 presented the highest compressive strength: 12.55 MPa on day 7 and 9.01 MPa on day 28. The AAB-1.2 samples had the greatest L/S ratio and exhibited the lowest compressive strength: 3.38 MPa on day 7 and 4.56 MPa on day 28. AAB-1.0 samples exhibited 6.64 MPa on day 7 and 7.00 MPa on day 28.
Among the three obtained series, series AAB-0.8 exhibited the highest compressive strength, although the strength dropped by 3 MPa from day 7 to day 28. AAB-1.0 exhibited the most stable compressive strength in the curing timeframe, increasing its value from 6.64 MPa on day 7 to 7.00 MPa on day 28, but the overall compressive strength was lower than needed for adsorbent granules.
When looking at
Figure 5, we can see that when increasing the L/S ratio in the activation solution, compressive strength decreased.
The results shown in
Table 4 can be compared with the results from a study by Liang Chen [
25], who did a similar study of preparation and property assessment of alkali-activated metakaolin. In that research, a compressive strength of 26.42 to 37.40 MPa was achieved, using elevated temperature for curing and sodium silicate solution as part of the mixture design.
Figure 6 shows the X-ray diffractograms for the produced AABs taken on day 28 of production. According to the data, an alkali activation reaction took place because the X-ray diffractograms show a halo shift of the amorphous phase (curvature at 15–30°) to the right (curvature at 20–35°), and the main product of the alkali activation reaction—N-A-S-H gel—was obtained. The effect of the L/S ratio on the mineralogical composition was observed.
As seen in
Figure 6, crystalline phases of quartz (SiO
2) and muscovite (KAl
3Si
3O
10(OH)
2) could observed after an alkali-activation reaction. In addition to the L/S ratios affecting observation, alkali-activation byproducts—zeolites—were obtained. Following the analysis of the data, zeolites contained in the AABs corresponded to zeolite 4A (Na
12Al
12Si
12O
48). Increasing the L/S ratio increased the crystalline phase intensity of zeolite 4A. Zeolite 4A crystalline phase intensities for AAB-0.8 and AAB-1.0 were quite similar, which could be due to the lack of water in the system to initiate the crystallization process [
26], but the crystalline phase of 4A in AAB-1.2 was significantly more intensive. The effects of L/S on the formation of zeolites in the AABs consisted of both increased Na
2O levels and increased quantities of H
2O, i.e., by adding more activation solution, more water entered the system as a whole, which helped the raw materials and formed more zeolite 4A [
27].
The specific surface area was determined for the AABs (
Table 5.) to see how the L/S ratio influenced surface area. Samples were degassed in order to free their surfaces so it would be possible to obtain more accurate results for their surface areas. The largest surface area was for the AAB-1.0 sample, which was 5.138 m
2/g, while AAB-0.8 was second, with 5.088 m
2/g. AAB-1.2 exhibited the lowest surface area of all samples, with 2.487 m
2/g.
Samples of AAB were subjected to scanning electron microscopy to study their microstructure. For AAB samples, images were taken at magnifications of 5000× and 15,000×, as shown in
Figure 7. According to the XRD results, crystals of plate-like formations corresponded to N-A-S-H gel, and crystals of cubic structure corresponded to zeolite 4A.
When looking at
Figure 7a,c,e, we can see that there were not many pronounced cubic crystal structures that corresponded to zeolite 4A, but when looking at
Figure 7b,d,f, showing images that were captured at greater magnification, we can see the crystal formation of a cubic structure much clearer. As shown in
Figure 7, without monocrystalline cubic crystals, polycrystalline formations formed from zeolite 4A monocrystals.
When comparing
Figure 7b,
Figure 7d,
Figure 7f and data obtained from the XRD, the previously mentioned relationship between the L/S ratio and crystalline phase intensity was confirmed, because as we can see in
Figure 7b, there were fewer cubic crystals than those shown in
Figure 7f.
3.2. Alkali-Activated Zeolite Granules
Table 6 contains material, bulk, and tapped densities that were measured for the produced zeolite granules. The Hausner ratio was measured by dividing tapped density with bulk density; this was important because it measured granule flowability. If the ratio is lower than 1.25, granules have a great flowability, meaning granules will flow freely and will not block the feeder; if the ratio is greater than 1.50, the flowability of granules is poor.
To obtain information about granule strength, an attrition test was conducted for two levels of pressure—4 and 5 atm.
Tapped density was measured by tapping the granules with the shock table used in the ASTM C143 standard. Thirty handle revolutions were used for compacting the granules. The highest material density (1.500 g/cm3) was measured for granules ZG-0.8, with bulk densities in free and compacted states of 0.703 and 0.781 g/cm3, respectively. The material density of ZG-1.2 was 1.472 g/cm3, with bulk densities in free and compacted states of 0.685 and 0.753 g/cm3, respectively. Bulk densities in the compacted states were higher than in the free states because the granules filled more voids between granules when compacted, thereby occupying a smaller volume. Material densities were greater than both bulk densities because the values for the material densities were for the materials without pores and voids.
Data that we obtained from the attrition test showed how much fines were produced under each pressure, which should be as low as possible. If fewer fines are produced from adsorbents, then the equipment will work properly for a longer period. If a lot of fines are produced, it will clog the compressors and the reactor will have to shut down for maintenance, and that can be expensive. Attrition was measured for a 2–4 mm fraction of produced zeolite granules at atmospheric pressures of 4 and 5. ZG-0.8 attrition was 2.42% at 4 atmospheric pressures and 4.55% at 5 atmospheric pressures. ZG-1.2 also showed similar attrition results: 2.73% and 4.85% at 4 and 5 atmospheric pressures, respectively. ZG-1.0 exhibited 3.64% and 5.76% attrition at 4 and 5 atmospheric pressures, respectively. Attrition was affected by several factors, such as particle size and shape, chemical composition, and mechanical compressive strength. Higher attrition resulted in lower mechanical compressive strength. From data obtained in this study, these granules were not suitable for adsorption above 4 atmospheric pressures; however, they could be used in some reactors that operate at a lower pressure.
Figure 8 shows an X-ray diffractogram for the ZG-0.8 samples, in which we can assess the crystalline phase’s development over time. When comparing the 14th- and 28th-day diffractograms, it could be assessed that zeolites kept developing through time, because the 28th-day diffractograms show a more intense zeolite 4A crystalline phase.
The ZG-1.0 sample X-ray diffractograms are shown in
Figure 9. When comparing these diffractograms, we could assess similarities with the ZG-0.8 samples, i.e., on the 28th day, the crystalline phase of zeolite 4A was more intense than on the 14th day.
Figure 10 shows the ZG-1.2 X-ray diffractograms on the 14th and 28th days. Using an L/S ratio of 1.2, it was possible to create a material that, compared to AAB-0.8 and AAB-1.0, did not show a sharp change in crystalline phase intensities during the curing process, which means that zeolite granules made with an L/S ratio of 1.2 could develop a more intensive zeolite 4A crystalline phase in a shorter period.
X-ray diffractograms of the studied zeolites are shown in
Figure 11. According to the data, an alkali-activation reaction occurred, because the amorphous phase (curvature at 15–30°) was offset to the right (curvature at 20–35°) in the X-ray diffractogram, and the main product of the alkali-activation reaction—N-A-S-H gel—was obtained. The L/S ratio affected the mineralogical composition.
In
Figure 11, we can observe that in the ZG-1.2 sample, the crystalline phase of zeolite 4A was slightly more intense than those of the ZG-0.8 and ZG-1.0 crystalline phases. There were no significant changes between ZG-0.8 and ZG-1.0, but the ZG-1.0 crystalline phases were slightly more intense than the ZG-0.8 crystalline phases. From this data, we can assess a regularity that increasing the L/S ratio led to a more intensive crystalline phase of zeolite 4A in the ZG samples.
Table 7 shows the surface area of the ZG samples. Zeolite granule samples exhibited similar values for the 0.8 and 1.2 L/S ratios (3.841 and 3.635 m
2/g, respectively), but the ZG-1.0 sample showed the highest surface area, 4.314 m
2/g.
Figure 12,
Figure 13 and
Figure 14 show the microstructures of the zeolite granules at 1000×, 5000×, and 15,000×.
Figure 12 illustrates the microstructures of the zeolite granule samples, in which their appearance on the surface may be observed. In this image, suitable particle distribution and homogenous pore placement can be observed. The SEM micrographs were taken with broken zeolite granules to assess their cross-section’s inner structure.
The produced zeolite granule microstructure, at a magnification of 5000×, can be observed in
Figure 13. The figure shows a large number of cubic crystalline formations, corresponding to the results of the XRD for zeolite 4A, as well as plate-like formations corresponding to the N-A-S-H gel.
Figure 14 shows formations of cube-shaped crystals corresponding to zeolite 4A and plate-like crystals formations corresponding to N-A-S-H gel. There is also a pronounced polycrystallinity on the surface of zeolite 4A, expressed as growth on new zeolite 4A crystals on the surface of existing zeolite 4A crystals.