**3. Results and Discussion**

Average compressive strength developments and NaOH concentrations of all investigated formulations are shown in Table 3, along with the relevant mix proportions of solids and liquids incorporated in the mixtures. Relationships between compressive strengths, molarity and Na2SiO3:NaOH ratios of 50:50 and 40:60 for mixtures in Case 1 are shown in Figure 4, whereas the same relationships related to Cases 2 and 3 are shown in Figures 5–7. The compressive strength values were determined as the averages of three samples for each value, and the standard deviations are included in the figures.

**Figure 4.** Effect of molarity of NaOH on the 24 h and 48 h compressive strength of formulations investigated in Case 1.

**Figure 5.** Effect of NAOH molarity and Na2SiO3: NaOH ratio on the 72 h strength of formulations at Cases 2 (**left**) and Case 3 (**right**).

**Figure 6.** Relationship between 72 h compressive strength and NaOH molarity of formulations in Cases 2 and Case 3 at Na2SiO3:NaOH ratios of 50:50 and 30:70, respectively.

The results obtained from Case 1 showed that the relationship between compressive strengths and molarities at both Na2SiO3:NaOH tested ratios (i.e., 50:50 and 40:60) and at both ages (i.e., 24 h and 48 h), was overall inversely proportional, i.e., showcasing a declining rate, although however, changing the molarities caused fluctuations on the compressive strengths, especially in the mixtures having a Na2SiO3:NaOH ratio of 50:50. In these particular mixtures, when the highest molarity NaOH was incorporated in the liquids (i.e., 12 M), a significant strength loss (66% reduction) was recorded compared to its highest strength value, which was observed at a molarity of 4 M. Another interesting point obtained was that the mixtures with Na2SiO3:NaOH ratios of 50:50 exhibited approximately 10–20% higher 48 h compressive strengths than those of the 40:60 ratio, however, this occurred only up to molarity of 8 M; beyond which, the mixtures of 50:50 ratio suffered a drastic 48 h strength loss of 66% when molarity was increased from 8 M to 10 M, and these eventually became weaker than the mixtures of 40:60 ratio, with nearly 10% lower 48 h strengths.

**Figure 7.** Effect of Na2SiO3: NaOH ratio on the 72 h strength of formulations in Cases 2 (**left**) and 3 (**right**), for MNaOH = 10.

The results in Figure 4 suggest that formulations of relatively high L/S ratio (i.e., 0.69) developed with a Na2SiO3:NaOH ratio of 50:50 and with a molarity of NaOH solution of 4 M exhibited the highest strengths from all compared formulations. However, the particular mixtures appeared to be highly sensitive to variations in the molarity, leading to significant fluctuations in strengths when molarities were progressively increased, and eventually suffering drastic strength losses beyond 10 M NaOH concentrations. In mixtures of a ratio of 40:60, the fluctuations in strengths were less significant when molarities were progressively increasing, although strengths were still declining across the tested molarities.

In Case 1 and according to Figure 4, a general trend of compressive strength decrease with the increase of the alkalinity was observed with minor differentiations in the strengths (within a range of 10–15%). This trend is attributed to the fact that the DM cannot be dissolved in the alkaline solution, according to the results obtained from the dissolution tests. Therefore, the alkali presence and the increase of molarity did not affect the materials' geopolymeric formation and the development of strength. In contrast, the increase of alkalinity seems to decrease the compressive strengths, since no geopolymeric reaction is taking place. On the other hand, the mix designs for Case 1 contain two pozzolanic materials (i.e., cement and gypsum) that favour the presence of water for developing high strengths. In mixtures of lower alkalinity, the cement and gypsum hydration reactions become dominant (due to a higher amount of unbound water, i.e., water not participating in the alkaline solution), thus resulting in higher strengths, as recorded.

Results of investigations carried on mixtures of Cases 2 and 3 are shown in Figures 5–7. It should be noted that, while observations were made and provided within the text, no correlations could be made between the two Cases (i.e., 2 and 3) due to the variation of multiple parameters governing each case, such as the curing regime, the mixture design, the ambient conditions, and also possible different activation mechanisms. As has been aforementioned, it was not the intention of this research to perform a direct comparison of the results obtained from the three Cases under study. Instead, the selection of these Cases and the variation of the parameters were aimed to reveal the factors that would positively

affect the activation of the DM, and therefore to maximise the effective utilization of the DM content in mixtures.

Cases 2 and 3 are summarized in Figures 5–7 together due to the common presence of metakolin in the two different formulations, their identical age of testing (i.e., 72 h), and also some similar trends observed in the obtained experimental results. It was observed that the progressive increase in NaOH molarity caused an increase in the 72 h strengths of both sets of mixtures with an almost identical effect on strength values for 2 M and 4 M, and then, at higher molarities, reaching up to 9.43 MPa (Case 2) and 7.53 MPa (Case 3), respectively. When the molarity value increased from 6 M to 8 M, both sets of formulations experienced their highest increase rate in strengths (121% and 93% increase for Cases 2 and 3, respectively). Beyond 8 M and towards 10 M, Case 3 mixtures still exhibited an appreciable increase in strengths, which was even higher than that observed from 4 M to 6 M. In the same concentrations, however, the mixtures of Case 2 were not associated with any significant increase, indicating an optimum concentration between 8–10 M for their highest achievable strengths. The results in Figures 5 and 7 suggest that the addition n of small quantities (4%) of CEM I in low L/S ratio DM-MK mixtures (i.e., Case 2), when conditioned in air ambient temperatures, led to a more significant increase in strengths at NaOH concentrations between 6 M and 8 M within the 50:50 Na2SiO3:NaOH ratio, although this effect was ceased at concentrations beyond 8 M and towards 10 M. This difference can be also observed in Figure 6 when comparing strength values between the 6 M–8 M range. At the same figure, when incorporating NaOH of molarity 2 M and 4 M, both air-cured and oven-cured mixtures yielded almost identical strengths regardless of their different Na2SiO3 ratios and regardless of the presence of CEM I in the formulation. Beyond a concentration of 4 M, and at least until the 8 M, there is a significant increase in strengths, which appears to be considerably sharper in the Case 2 mixtures. However, such an increase was less significant beyond 8 M in Case 2, in contrast to Case 3 mixtures. An additional observation for Case 3 made in Figure 5 (right hand part of the graph) was that a low L/S ratio DM-MK mixture with a 20:80 alkaline solution and its NaOH molarity at 10 M yields approximately the same strength as that of a mixture at 30:70 alkaline solution with a NaOH molarity of 6 M.

When both Cases were compared at equivalent molarity of 10 M (Figure 7), the maximum achievable strengths were found at 30:70 Na2SiO3:NaOH ratio, regardless of the presence of CEM I and regardless of the ambient conditions. It can be also seen, on the same figure, that the absence of CEM I in Case 3 appeared to have enhanced the strengths more significantly when the ratio was tending from 10:90 towards 30:70 when compared to Case 2 (i.e., when containing CEM I).

Fundamentally, the increase in compressive strengths in both Cases may be attributed to the increase of the alkalinity, which is mainly attributed to the presence of metakaolin in the mixtures, which predominantly reacts with the alkaline activator. Moreover, in Cases 2 and 3, the addition of the sodium silicate solution seems to have a positive effect on the evolution of the compressive strength. The continuous increase of Si content enhanced the content of dissolved elemental silicon, and the increase of Si(OH)4 monomer promotes the formation of more -Si-O-Si- bonds, thus forming a more stable bond structure.

Generally, the higher the Si content, the more stable it is, since the chemical bond strength of -Si-O-Si- is higher than the corresponding of -Si-O-Al- and -Al-O-Al-, and therefore higher energy is required to break the particular bond.

With Si/Al ratio increasing up to a certain extent (i.e., Case 3), the content of elemental silicon dissolved in the system is much higher than that of aluminum. Meanwhile, a part of Si(OH)4 will form a dimer after condensation reaction and then react with Al(OH)4 to form a stable long chain (-Si-O-Al-O-Si-O-) PSS polymer (Si/Al = 2) or more stable long chain (-Si-O-Al-O-Si-O-Si-O-) PSDS polymer (Si/Al = 3). This phenomenon is not presented in Case 2, since the presence of cement creates C-S-H phases in parallel that are not enhanced with the presence of additional Si, while it in contrast intercepts the geopolymerisation process, and thus the compressive strength decreases (Figure 7).
