This section entails six experimental investigations including (1) ion concentration of hydroxyl, aluminate, and calcium, (2) flowability of fresh mixtures, (3) initial setting time, (4) compressive strengths at different curing times, (5) 28-d porosity, and (6) Da
3.1 Ion Concentration
The effect of hydration time on the hydroxyl ion concentrations for the control, 20%, and 50% fly ash systems mixed at 0 and 400 rpm is shown Figs. 2a and 2b, respectively. Results indicate that the hydroxyl ion concentrations of all systems increase when hydration time increases. The slope of the fitted curve is referred to as the dissolution rate of the hydroxyl ions in solution. Larger slope indicates higher dissolution rate of hydroxyl ions in solution.
In Fig. 2a, the slopes of the control, 20%, and 50% fly ash systems are 9.0, 7.7, and 5.2 mmol/l/min, respectively. The larger slope of the control system corresponds with higher rate of dissolution of hydroxyl ions than the 20% and 50% fly ash systems. Increased percent replacement level of fly ash in the cementitious systems has been reported to result in decreased hydroxyl ion concentration in solution [26]. This is believed to occur because fly ash acts as “inert.” Hence, the fly ash mixtures contain less cement. Results in Fig. 2b are similar with the results in Fig. 2a, of which the slope of the control system mixed at 400 rpm (17 mmol/l/min) is larger than the slopes of the 20% (15 mmol/l/min), 50% fly ash (13 mmol/l/min) systems mixed at 400 rpm. Increased replacement level of fly ash results in decreased dissolution rates of hydroxyl ions, irrespective of mixing speed.
Comparing the results in Fig. 2b with the results in Fig. 2a indicates that the slopes of the control, 20%, and 50% fly ash systems mixed at 400 rpm are approximately 89%, 95%, and 150% higher than the slopes of the control, 20%, and 50% fly ash systems mixed at 0 rpm, respectively. The steeper slope is attributed to faster rate of dissolution of hydroxyl ions. This indicates that mixing speed strongly affects the dissolution rate of hydroxyl ions of all systems and could affect early-age characteristics of mixtures (as discussed later). Mixing speed increasingly affects the dissolution rate of hydroxyl ions in the systems containing higher replacement levels of fly ash.
Figure 3a shows the time-variant concentrations of aluminate ions of the control, 20%, and 50% fly ash systems mixed at 0 rpm. Results indicate that the aluminate ion concentrations of the control system are lower than the aluminate ion concentrations of the 20% and 50% fly ash systems. The plot of aluminate ion concentrations of the control, 20%, and 50% fly ash systems mixed at 400 rpm as a function of the hydration time is shown in Fig. 3b. Like the results of all systems mixed at 0 rpm in Fig. 3a, the aluminate ion concentrations of the control system are lower than the 20% and 50% fly ash systems. More importantly, the aluminate ion concentrations mixed at 400 rpm for all systems are not stable, when compared with the aluminate ion concentrations in solutions mixed at 0 rpm. This “unstable” state likely occurs due to mixing speed. Because the aluminate ion concentrations are not stable in the systems mixed at 400 rpm, it is expected that there would be the differences of early-age characteristics (e.g., setting) between the systems that are continuously mixed and the systems that are mixed and then stop mixing.
Figures 4a and 4b show the plots of the calcium ion concentrations as a function of hydration time for the control, 20%, and 50% fly ash systems mixed at 0 and 400 rpm, respectively. Results in both figures indicate that the control systems mixed at 0 and 400 rpm have higher calcium ion concentrations than the 20% and 50% fly ash systems. The presence of fly ash results in reduced calcium ion concentrations at early ages due to less cement contents. In addition, these lower calcium ion concentrations could be a result of fly ash particles acting as nuclei for lime precipitation. Lawrence [27] and Fraay et al. [28] reported that the lime precipitation can occur when replacing cement with SCMs. Like the results of the aluminate ion concentrations in solutions mixed at different speeds (shown in Figs. 3a and 3b), the calcium ion concentrations in solutions mixed at 400 rpm are not stable compared to the solutions mixed at 0 rpm. Apparently, results from the aluminate and calcium ion concentrations indicate that when mixing continuously progresses, ions are likely unstable, thereby affecting other early-age characteristics.
Table 3 summarises the effects of the replacement of fly ash in cementitious systems on the concentrations of hydroxyl, aluminate, silicate, and calcium ions in solutions at early ages. Regarding the ion concentration studies for extended mixing processes, replacing cement with fly ash can lead to (1) decreased hydroxyl ion concentration due to less cement contents, (2) increased aluminate ion concentration due to a formation of aluminate-rich gel layers, (3) increased silicate ion concentration due to more silicon from fly ash in cementitious systems, and (4) decreased calcium ion concentration due to lime precipitation and less cement content [29, 30].
Table 3
Summary of ion concentration of cementitious containing fly ash.
Cementitious System | Ion concentration |
Hydroxyl | Aluminate | Silicate | Calcium |
With fly ash | ↓ | ↑ | ↑ | ↓ |
Without fly ash | ↑ | ↓ | ↓ | ↑ |
Possible cause | Less cement content | Gel formation | More silicon from fly ash | Lime precipitation/ Less cement content |
3.2 Flowability
The effect of mixing time on normalised flowability of the control, 20%, and 50% fly ash systems mixed at 285 rpm is shown in Fig. 5a. The flow values are normalised with the maximum flow values (183 mm) of the 50% fly ash systems. Results indicate that a flow reduction of all systems is a result of increased mixing time. The flow of the fresh mortars of the 20% and 50% fly ash systems mixed for 2 minutes exhibits approximately 12% and 31% larger flow than the control system, respectively. The flow of the fresh mortars of the 20% and 50% fly ash systems mixed for 15 minutes exhibits approximately 30% and 48% larger flow than the control system, respectively. Lastly, the flow of the fresh mortars of the 20% and 50% fly ash systems mixed for 60 minutes exhibits approximately 50% and 43% larger flow than the control system, respectively. Bentz and Ferraris [31] reported that as hydrating products form, the early stiffening behavior is controlled by the gradual loss of free water from hydration reactions.
Paya et al. [32] reported that flow of cementitious systems containing fly ash is influenced by several factors such as size distribution, morphology, surface condition, fineness, and loss on ignition of the fly ash particles. Replacing cement with fly ash is believed to make more water available for flow. Gopalan [33] reported the water absorption characteristics of the cementitious systems are reduced when fly ash is present. In addition, the spherical particles and better particle size distribution of fly ash (which is shown in Fig. 1) lead to the reduction in friction between constituent particles in the systems (this commonly known as the “ball-ability” or “ballbearing effect”) [34, 35]. Thus, although increased mixing time results in reduced flow of the fly ash systems, the fresh mixtures can be mixed longer and still be castable due to improved flow.
The plot of the normalised flow of the control, 20%, and 50% fly ash systems as a function of mixer revolution counts is shown in Fig. 5b. Results indicate the flow of all systems significantly reduces with increasing mixer revolution counts. The control system has less flow than the 20% and 50% fly ash systems. Replacing cement with fly ash results in a significant increase in flow of the fresh mixture and that allows proper consolidation process of mixtures mixed at higher mixer revolution counts can be performed. Based on test data, specification’s limits on mixing time and mixer revolution counts may not be appropriated when fly ash is present in the cementitious systems.
3.3 Setting Time
Figure 6a shows the effect of mixing time on the initial setting time of the control, 20%, and 50% fly ash systems. Results indicate that increased percent replacement level of fly ash results in delayed initial setting times. Possible reasons of delaying effect is likely due to the adsorption of calcium ions on the surface of fly ash [36] and lowered hydroxyl ion concentrations. These lead to delayed nucleation and precipitation process of the Ca(OH)2, C-S-H, and ettringite. Results also indicate that increased mixing time leads to increased initial setting time of the control and 20% fly ash systems (ANOVA test with p-value < 0.05). This is likely because the ions in the systems are in the unstable state (as discussed) and the constituent’s particles are still deformed by a movement of mixing tools during continued mixing. This deformation of the particles likely disturbs the adhesion bonding of hydrating particles to form a larger structure. Hence, during mixing, the hydrating particles lose their load-supporting ability, consequently resulting in slower setting. After mixing stops, the adhesion bonding begins to form a larger structure and this structure, with time, begins to support external loads. However, it is assumed that part of water is consumed by hydration reactions during extended mixing since the hydration reactions can generate substantial amount of heat, and this heat of hydration leads to increased water evolution rate. Hence, less amount of water available for flow before casting and consolidating the specimens. This less amount of water of mixtures finally results in increased energy to cast and consolidate the specimens [22].
Figure 6b show the effect of the initial setting time as a function of mixer revolution counts of the control, 20%, and 50% fly ash systems. According to the findings, it appears that the initial setting time tends to prolong as a function of the number of mixer revolutions. The initial setting times of cementitious systems containing fly ash are delayed as a result of increased mixing times as well as increased mixer revolution counts.
3.4 Compressive Strength
Figures 7a and 7b show the 1-day fc of the control, 20%, and 50% fly ash systems as a function of mixing time and mixer revolution counts, respectively. Results indicate that increased percent replacement level of fly ash leads to lower 1-day compressive strength. Statistical analyses indicate that the 1-day fc of all systems is not influenced by mixing time and mixer revolution counts (ANOVA test with p-value > 0.05).
Figures 7c and 7d show the effect of mixing time and mixer revolution counts on the 7-day fc for the control system and systems containing 20% and 50% fly ash, respectively. Results indicate that the 7-day fc of all systems is not significant affected by mixing time (ANOVA p-value > 0.05). In addition, the 7-day fc of the control systems does not have a significant effect by mixing time (ANOVA p-value > 0.05); however, the 7-day fc of the 20% and 50% fly ash does (ANOVA p-value < 0.05).
The effects of mixing time and mixer revolution counts on the 28-day fc of the control, 20%, and 50% fly ash systems are shown in Figs. 7e and 7f, respectively. Similarly with the results of the 1- and 7-day fc, results indicate that mixing time does not a significant effect on 28-day fc of all systems. However, here, the mixer revolution counts do influence 28-day fc. Increased mixer revolution counts result in a reduction of 28-day fc of the control system but result in an increase of 28-day fc of the systems containing fly ash. It has been reported that an increase in the consolidation energy consumption is responsible for the 28-day fc drop that occurs as a function of the number of revolutions of the mixer [22]. The fc will begin to decrease as soon as the energy available is insufficient to consolidate the specimens. However, the systems that contain fly ash show an increase in 28-day fc, and it is assumed that this is because there is less water in their mixtures (which results in a lower w/b value). Increased rates of water evaporation and hydration reactions were claimed to be the cause of less water being present in the mixtures by Dewar and Anderson [37].
Results herein indicate that mixing time does not have a significant influence on 1-, 7-, and 28-day fc. The summary of the influence of mixer revolution counts on the fc of the control, 20%, and 50% fly ash systems is shown in Table 4. The early-age fc values (1- and 7-day) do not have a significant impact by increasing the mixer revolution counts. On the other hand, the 28-day fc will decrease for the control system as the number of mixer revolutions increases, but it will rise for the systems that include fly ash. Under the extended mixing conditions, the presence of fly ash in cementitious systems can improve the long-term compressive strength. Hence, the ready-mix concrete products should partially contain fly ash especially when a long haul is needed.
Table 4
Summary of the effect of mixer revolution count on the fc of mixtures at different ages (↑ = increase; ↓ = decrease; and - = similar).
Age (days) | Effect of mixer revolution counts on the fc |
Control | 20% fly ash | 50% fly ash |
1 | - | - | - |
7 | - | - | - |
28 | ↓ | ↑ | ↑ |
3.5 Porosity
The effects of mixing time and mixer revolution counts on the 28-day porosity of the control, 20%, and 50% fly ash systems are shown in Figs. 8a and 8b, respectively. Results indicate that increased mixing time does not significantly affect the 28-day porosity for all systems (ANOVA p-value > 0.05). Increased percent replacement level of fly ash results in higher porosity. Figure 8b shows that increasing the mixer revolution counts at lower revolutions (less than approximately 3000) results in a significant increase in porosity (ANOVA test with p-value = 0.013). However, increasing the mixer revolution counts at higher revolution counts (more than approximately 3000) has less influence on the porosity of mixtures (ANOVA test pvalue > 0.05). The porosity of the systems containing fly ash does not exhibit this effect like the control system.
Figure 9 shows the correlation between the normalised 28-day compressive strength and the 28-day hardened porosity of the control, 20%, and 50% fly ash systems for different mixer revolution counts. The 28-day fc values are normalised with the mean 28-day fc for all mixtures. The linear fitted curve was illustrated in this study. The logarithmic fitted curve shown is reported by Neveille [38] and the exponential and linear fitted curves are reported by Brandt [39]. Although different type of the fitted curve is not similar with the curves from previous studies, the linear fitted curve in this study is not distinguished from others. Results indicate that the 28-day fc is reduced with increasing porosity and this increased porosity is likely due to reduced flow caused by higher mixer revolution counts. It is assumed that the primary criteria reflecting the overall performance of macrostructures are parameters relating to the adequate flow of fresh cement mixtures resulting from extended mixing.
3.6 Apparent Chloride Diffusivity
Figures 10a and 10b show the effects of mixing time and mixer revolution counts on the Da of the control, 20%, and 50% fly ash systems, respectively. Results indicate that neither mixing time nor mixer revolution counts has a significant effect on the Da for all systems (ANOVA p-value > 0.05). The control system exhibits higher Da than the systems containing fly ash (ANOVA p-value = 0.026). Consequently, only the influence of material components influences corrosion resistance, not mixing activities. It is common knowledge that adding fly ash to cement systems can result in a denser microstructure and reduced porosity. Sabet et al. [38] reported that the presence of fly ash in the cementitious systems can react with Ca(OH)2 to produce CS-H products and also bind chloride ions by the aluminate phases during chloride exposed period. These lead to reduced transport rates, and finally the service life of concrete structure can be extended.