3.1. Sintering Temperature of Clinkers
Comprehensive thermal analysis can characterize the mass and heat changes in the clinker formation process to approximate preliminarily the temperature of the reaction among the raw materials. As shown in
Figure 2, the DSC-TG curves of the mixtures (i.e.,
samples I-1,
II-1 and
III-1) show a similar change rule. Combined with the DSC-TG curves of the raw materials in
Figure 3, it is not difficult to obtain some meaningful results: when the temperature is below 200 °C, the endothermic peaks and mass loss in the DSC-TG curves should originate from the evaporation of the physical water in all the raw materials; when temperature reaches 400–500 °C, an endothermic peak should be caused by the decomposition of the Ca(OH)
2 in the petroleum coke desulfurization slag (formed by the CaO absorbing water from the air) and in the carbide slag; at approximately 710 °C, 910 °C and 1050 °C, there are different sized endothermic peaks with mass loss, which are still caused by the petroleum coke desulfurization slag and carbide slag. It appears that there was no new mineral formation before 1050 °C, and in this process, the mass loss and heat change of the fly ash and bauxite were also not significant. The reason may be that these two raw materials have excellent resistance to elevated temperatures. After 1050 °C, all the raw materials were involved in the reaction, including a series of chemical changes such as the decomposition of the C
S, the formation and decomposition of the C
2AS (gehlenite, 2CaO·Al
2O
3·SiO
2) [
21], C
4A
3S, C
5S
2S (ternesite, 4CaO·2SiO
2·CaSO
4) [
21,
25,
26] and β-C
2S. The comprehensive effect was to form exothermic peaks in the 1050–1350 °C range, which was also used preliminarily as the sintering temperature.
To determine the clinker sintering temperature and assess the formation of products more accurately, an X-ray diffraction analysis was carried out. The X-ray diffraction (XRD) patterns of the sample
II-1 cement clinker at different temperatures
are given in Figure 4. When the sintering temperature is in the range of 950–1100 °C, the main products are C
S, C
2AS, CaO, Al
2O
3 and SiO
2, and C
4A
3S begins to form in a small amount at 1050 °C, which is consistent with the results of the comprehensive thermal analysis. The temperature rises to 1150–1225 °C, C
4A
3S gradually forms in large quantities; additionally, the C
2AS disappears and C
5S
2S appears, and the sintering products are mainly C
4A
3S and C
5S
2S. As the temperature continues to rise to 1250–1300 °C, the C
5S
2S decomposes, and β-C
2S gradually forms; the main products are C
4A
3S, β-C
2S and C
S. At the same time, the intensity of the C
4A
3S and β-C
2S diffraction peaks increases with the increase in temperature, while that of the C
S decreases, indicating that the increase in temperature promotes high temperature reactions. As the temperature increases again, the intensity of the diffraction peaks of each product begins to decrease, and the C
S diffraction peaks are hardly observed, especially after 1350 °C. This finding indicates that the increase in temperature has caused the decomposition of the mineral phases. It can be seen that the sintering temperature of the sample II-1 cement clinker ranges from 1250 °C to 1350 °C, and the optimum sintering temperature is 1300 °C. It is worth pointing out that the iron element did not form the anticipated C
4AF but mainly formed C
4A
2.85Fe
1.5S(3CaO·2.85Al
2O
3·1.5Fe
2O
3·CaSO
4) by solid solution in the C
4A
3S [
25], regardless of the sintering temperature.
In accordance with the same principle of analysis, all the other mix proportions besides sample II-1 were tested to study the influence of the mix proportion on the sintering temperature. Some of the results are shown in
Figure 5,
Figure 6 and
Figure 7. As revealed in these figures, the sintering temperature ranges of the Series
I, II, and III cement clinkers are approximately 1225–
1325 °C,
1250–
1350 °C, and
1275–
1350 °C, respectively. Thus, with the increase of the C
S content in the mix design, the lower limit of the sintering temperature increases gradually, which is controlled by the decomposition of C
5S
2S, and the range of the sintering temperature is basically maintained at 100 °C. In addition, the optimal sintering temperature of all the mix proportions is approximately 1300 °C.
3.2. Mineral Composition of Clinkers
The micromorphology of the minerals in the clinker was observed by scanning electron microscopy (SEM), and the mineral composition was qualitatively analyzed. Taking the sample
II-1 cement clinker prepared at 1300 °C as an example, the results are shown in
Figure 8. The cement clinker system was mainly composed of tabular C
4A
3S, blocky granular β-C
2S, and radical and needle-bar C
S. The mineral composition is basically consistent with the X-ray diffraction analysis.
In addition to the qualitative analysis of the mineral composition, a quantitative analysis of the clinker minerals was also carried out. The chemical composition of the cement clinkers prepared at 1300 °C was measured as shown in
Table 3. The clinker mineral content is calculated according to Bogue’s equation [
27]. However, because of the existence of residual C
S and C
4A
2.85Fe
1.5S and the absence of C
4AF in the cement clinker, Bogue’s equation needs to be modified [
28,
29,
30,
31], as shown in Equations (1)–(4), while the alkalinity coefficient is also been modified, as shown in Equation (5).
As revealed in
Table 4, the actual mineral content of the cement clinker is not significantly different from the design content in
Table 2. More specifically, the content error of the C
4A
3S and β-C
2S is approximately 5%, while that of the C
S is within 2%, even the content error between the C
4A
2.85Fe
1.5S and C
4AF is less than 5%. In addition, the alkalinity coefficients of all the clinkers are slightly greater than 1.0, indicating that the CaO in the raw materials can meet the requirements for the formation of various useful minerals.
By the qualitative and quantitative analyses mentioned above, the mineral composition of the clinkers basically conforms to the expected product, and it is indirectly proven that the mix design and sintering conditions are reasonable.
3.3. Physical and Mechanical Properties of the Clinkers
The physical and mechanical properties are the most basic elements of cement performance. In this study, the water requirement of normal consistency, setting time and mechanical strength of the clinkers with different proportions prepared at 1300 °C were tested. As shown in
Table 5, the water requirement of normal consistency, which was prepared by using various solid wastes, is in the range of 36%–40%, which is slightly larger than that of typical clinkers (approximately 30%). The initial setting time and the final setting time are 17–25 min and 23–40 min, respectively. The setting time is relatively short, which is more suitable for projects involving emergency rescue and repair. Further analysis shows that the water requirement of normal consistency decreases and the setting time prolongs as the C
4A
3S content decreases in each series. In view of the fact that the early hydration of HBSAC is dominated by C
4A
3S [
32,
33,
34], it can be considered that the C
4A
3S content plays a direct and decisive role in the water requirement of normal consistency and in the setting time. Comparing sample I-1 with II-1, or sample I-2, II-2 with III-1, or sample I-3 with III-2, it is easy to find that the water requirement of normal consistency decreases and the setting time shortens as the C
S content increases under the same C
4A
3S content, which indicates that an increase in the C
S content in the range of 10%–20% helps to accelerate the C
4A
3S hydration.
Figure 9 shows the mechanical strength of the cement clinkers measured by using the mortar pieces at a water-cement ratio of 0.52. As shown in
Figure 9a, the bending strength of each series increases as the curing age increases, and the bending strength increases rapidly within 3 days and slows down after 3 days. For example, as the lowest early bending strengths of all the mixtures, the bending strengths of sample I-3 at 1, 3, 7 and 28 days are 2.4 MPa, 3.8 MPa, 5.2 MPa and 6.2 MPa, respectively, while the strength increases at 3, 7 and 28 days are 58%, 37% and 20%, respectively. Sample II-2, as the middle early bending strength sample of all the mixtures, has bending strengths of 5.5 MPa, 6.1 MPa, 6.2 MPa and 6.4 MPa, respectively at 1, 3, 7 and 28 days, and strength increases of 11%, 2% and 3%, respectively at 3, 7 and 28 days. Of all the mixtures, sample III-1 has the highest early bending strength; its bending strengths at 1, 3, 7 and 28 days are 6.0 MPa, 6.3 MPa, 6.6 MPa and 6.8 MPa, respectively, while the strength increases at 3, 7 and 28 days are 5%, 5%, and 3%, respectively. In addition, there are several other findings: (1) The increase in the bending strength of the samples in Series II and III at 7 days has decreased to less than 5%, while that of the samples in Series I is approximately 30%, which indicates that the increased C
S content in the clinker has an obvious positive effect on the development of early bending strength [
35]. (2) Comparing the bending strength of samples I-1 and II-1 (or samples I-2, II-2 and III-1) at different curing ages, it can be seen that under the same C
4A
3S content, the bending strength at 1, 3, and 7 days increases as the C
S content increases. However, there is no similar trend for the bending strength at 28 days, which shows from another perspective that the increased C
S content in the clinker has an obvious positive effect on the development of early bending strength. The bending strength at 28 days decreases with the β-C
2S content decreases, which means that the hydration of β-C
2S is the main factor for the strength development of the cement clinker in the later stage [
36]. (3) During the entire 28-day curing process, the bending strength of each series decreases as the C
4A
3S content decreases. The β-C
2S content does not change the development trend of the later strength, indicating that the hydration of β-C
2S in the clinker is not so strong that the bending strength of the cement clinkers at 28 days is still dominated by the C
4A
3S content.
In terms of compressive strength, as shown in
Figure 9b, there are obvious differences among the three series with the increase in the curing age. The compressive strength of Series I and III samples decreases at 28 days, while that of Series II samples maintains good growth during the whole curing age. It can be seen that the C
S content in the clinker is not as high as possible, a lower or higher C
S content in the clinker may have a negative impact on the compressive strength. The reasons are as follows: when the C
S content is lower, the AFt phase formed in the early hydration stage is partly converted to the AFm (ettringite, C
3A·CaSO
4·12H
2O) phase in the later hydration stage [
37,
38,
39], as shown in
Figure 10, which results in a decrease in the compressive strength; when the C
S content is higher, the rate of the AFt phase formation in the early hydration stage is too fast, and the crystal structure continues to grow in the later hydration stage, which causes expansion and microcrack damage [
26], and also results in a decrease in the compressive strength. According to the results of this experiment, 15% is the optimal C
S content in this kind of cement clinker; under this content, the lowest compressive strengths are 42.1 MPa, 46.3 MPa, and the highest are 52.8 MPa, 64.3 MPa at 3 days and 28 days, respectively. The change rule of the compressive strength before 28 days is the same as that of the bending strength: the strength increases rapidly in 3 days but slows down after 3 days, and the increase of the C
S content in the clinker has an obvious positive effect on the development of the early compressive strength. The specific strength value can be referred to in
Figure 9b, and will not be repeated here.
The formation of the hydration products can reasonably reflect the development of the cement strength. As shown in
Figure 11, the main hydration product of the cement is the AFt phase, and the C
4A
3S hydration is the main hydration reaction during the 7 days, while C
S is gradually consumed. The C
2S hydration is slow, and C
2S diffraction peaks in the XRD patterns still change slightly at 28 days. In
Figure 12, the needle bar phase is the AFt phase, and at the initial hydration stage, the AFt phase are fine needle rods, which gradually develop into thick needle rods as the curing age increases. At the later hydration stage, the flocculent C-S-H gel phase appears and intercalates with the needle AFt phase, which makes the structure of the cement paste denser and further increases the strength. Thus, the formation process of the hydration products coincides with the strength development law as the curing age increases. In addition, the XRD patterns and SEM micromorphology of the hydration products of the Series I cement clinker before 28 days and the Series II cement clinker during the whole curing age, are basically similar to those of the Series II cement clinker and are not listed here.