Influence of Silica Modulus and Curing Temperature on the Strength of Alkali-Activated Volcanic Ash and Limestone Powder Mortar

This present study evaluates the effect of silica modulus (Ms) and curing temperature on strengths and the microstructures of binary blended alkali-activated volcanic ash and limestone powder mortar. Mortar samples were prepared using mass ratio of combined Na2SiO3(aq)/10 M NaOH(aq) of 0.5 to 1.5 at an interval of 0.25, corresponding to Ms of 0.52, 0.72, 0.89, 1.05 and 1.18, respectively, and sole 10 M NaOH(aq). Samples were then subjected to ambient room temperature, and the oven-cured temperature was maintained from 45 to 90 °C at an interval of 15 °C for 24 h. The maximum achievable 28-day strength was 27 MPa at Ms value of 0.89 cured at 75 °C. Samples synthesised with the sole 10 M NaOH(aq) activator resulted in a binder with a low 28-day compressive strength (15 MPa) compared to combined usage of Na2SiO3(aq)/10 M NaOH(aq) activators. Results further revealed that curing at low temperatures (25 °C to 45 °C) does not favour strength development, whereas higher curing temperature positively enhanced strength development. More than 70% of the 28-day compressive strength could be achieved within 12 h of curing with the usage of combined Na2SiO3(aq)/10 M NaOH(aq). XRD, FTIR and SEM + EDX characterisations revealed that activation with combined Na2SiO3(aq)/10 M NaOH(aq) leads to the formation of anorthite (CaAl2Si2O8), gehlenite (CaO.Al2O3.SiO2) and albite (NaAlSi3O8) that improve the amorphosity, homogeneity and microstructural density of the binder compared to that of samples synthesised with sole 10 M NaOH(aq).


Introduction
Concrete is the bedrock of our built environment. The construction of critical infrastructures for social and economic development consumed approximately 35 billion tons of concrete inclusive of steel, wood and aluminium [1]. The choice of concrete as the most widely used construction material was attributed to its favourable mechanical and durability performance couple with its global availability and relatively low cost [2]. Cement is an essential binding material in concrete. Cement global production was estimated at approximately 4.1 billion tons in 2017, with an expected four-fold increase by 2050 [3]. Despite the numerous advantages of cement, it significantly contributes to 5-8% of global CO 2 resulting in climate change [4,5]. To mitigate the proliferation of CO 2 into the environment, researchers are focusing on developing viable and sustainable alternative friendly novel studied [28]. Ibrahim et al. investigated the effect of silica modulus on alkali-activated natural pozzolan concrete [29]. Curing temperature is another very important parameter that affects the strength development of alkali-activated binders. The authors of [30] investigated the effects of temperature on the strength development of alkali-activated fly ash-slag binder. It was found that curing at low temperatures hindered quick strength development of fly-ash-based alkali-activated binder due to low reactivity and dissolution of fly ash at a lower temperature of 25 • C. However, at a higher curing temperature of 60 • C and 90 • C significant strength was reported. At higher temperatures, the rate of reaction is faster, and about 70% gain in strength was reported by Kong et al. [31]. The most beneficial curing temperature of 120 • was reported for fly-ash-based geopolymer mortar [32]. Despite numerous works on the effects of alkali activators on the alkaliactivated binder, there is a limited study concerning the role of silica modulus and curing temperature on mechanical, and microstructure properties of alkaline binary blended volcanic ash powder and limestone powder mortar.
This paper reports the findings of the influence of silica modulus and curing temperature on the strength and microstructure of alkali-activated LSP/VA mortar. It is expected that the outcomes of this study will help in understanding the impact of alkaline activators and curing temperature and also contribute to a better understanding of the impact of silica modulus and curing temperature on the characteristics of the alkali-activated volcanic ash/limestone powder mortar. The utilisation of VA/LSP will be an addition to sustainable alternative binders and waste reduction.

Volcanic Ash Powder and Limestone Powder
The volcanic ash was provided by Imerys minerals Arabia, Rabigh, Kingdom of Saudi Arabia, and the limestone powder (LSP) was obtained from the quarry as a waste. The LSP was oven-dried at 105 • C ± 5 • C for 24 h. The chemical compositions of volcanic ash and the LSP determined using an X-ray fluorescence (XRF) spectrometer are provided in Table 1. It was found that LSP is composed mainly of CaO (94.1%) with a very low content of silica (2.5%) and alumina (0.8%); conversely, VA has silica (74%) as its main component with moderate content of alumina (13%). This was the reason for the binary blended alkaline activated binder using these two materials.

. Synthesis of Alkaline Activator
Industrial purchased aqueous sodium silicate (SS) and sodium hydroxide (NH) were used as alkaline activators. The initial silica modulus (M s = SiO 2 /Na 2 O) of SS was 3.3, and the percentage composition of the Na 2 SiO 3(aq) is as follows; H 2 O: 62.11%, SiO 2 : 29.13% and Na 2 O: 8.76%. A 10 M NaOH (aq) solution was prepared by dissolving 404.4 g of NaOH pellet (99% assays) into 1 L of distilled water and was kept at ambient temperature for at least 24 h for cooling before being used. The SS and NS were combined at a ratio of x ( SS NH ), where x varied from 0.5 to 1.5 at an interval of 0.25.

Aggregates
Dune desert sand passing the gradation size requirement of ASTM C33 was used as the fine aggregate (FA) [33]. The fineness modulus of sand was 1.82, with 2.63 as the specific gravity.

Mix Design
The mortar was prepared using a binary combination of 40% VA and 60% LSP [19]. The samples were prepared using a mass ratio of Na 2 SiO 3(aq) /10 M NaOH (aq) (SS/NH) of 0.5 to 1.5 at an interval of 0.25. The samples were designated as AANLM x (alkaliactivated VA/LSP mortar), where x = SS NH . A total of six AANLM x (where x = 0, 0.5, 0.75, 1, 1.25 and 1.5) mixtures were prepared to study the combined effect of SS and NH (silica modulus), the sole effect of 10 M NaOH (aq), and the effect of curing temperature on strengths and microstructures of the developed mortar. All the mixtures were prepared using a constant fine aggregate (FA) to the binder ratio of two based on the beneficial value from the preliminary trial mix. The alkaline activator to binder ratio of 0.5 was used for the mortar and 0.25 for the paste. The free water to precursor (pozzolanic material) ratio of 0.1 was used in all the mixtures based upon the preliminary trial mix to attain a workable mixture. Table 2 shows the proportion of the constituent materials in the alkali-activated mortar mixtures. The mixing procedure adopted in this work is the same as reported by previous researchers [19,[34][35][36][37].
The required quantities of constituent materials were measured and mixed in batches in the 5.0 L capacity Hobart planetary bench mixer. The mixing of the materials was performed in two stages. In the initial stage, the VA and LSP powder and sand were mixed in a dry condition for 3 min. In the last stage, the alkaline solution (NaOH (aq) + Na 2 SiO 3(aq) ) and water were added for the wet mixing stage, which involves low speed mixing for 2 min and another 4 min for fast or higher speed mixing until a homogeneous mixture was achieved, the total mixing time was about 9 to 10 min to ensure the homogeneity of the mix. Thereafter, the mortar was placed in the oil-smeared steel moulds of 50 × 50 × 50 mm 3 in two layers, and each layer was vibrated on the vibrating table for 30 s to remove any entrapped air from the mixture. Then, the surface was carefully smoothened with a trowel to have a smooth finish. After the placement, consolidation and finishing of the mortar, the specimens were covered with a plastic sheet to prevent moisture loss and kept in the laboratory at 20 ± 5 • C for 24 h. After 24 h of casting, the cubes were de-moulded and placed in zip plastic bags to avoid evaporation of moisture. The samples were then subjected to temperature curing in an oven maintained at room temperature (20 ± 5 • C) and various curing temperatures of 45, 60, 75 and 90 • C for 24 h. After that, the specimens were cured under a normal room temperature condition of 20 ± 5 • C until the age of testing (1, 3, 7, 14 and 28 days). The compressive strength of the mortar was determined in accordance with ASTM C 150 [38] on the 50 × 50 × 50 mm 3 cube specimens using a Matest digital compression testing machine. The compressive strength of the specimens was determined after 1, 3, 7, 14 and 28 days of curing. Three specimens were tested at each age, and the average compressive strength value was recorded.

Physical and Mineralogy Characteristics of Base Materials
The physical properties such as specific gravity, mean particle size and specific surface area of VA and LSP are depicted in Table 3. VA has a smaller mean particle size of 5.77 µm, while LSP has a larger mean particle size of 12.05 µm. The specific surface area of VA is larger than that of LSP by about five times. The particle size distribution curves of LSP and VA are provided in Figure 1. The physical properties such as specific gravity, mean particle size and specific face area of VA and LSP are depicted in Table 3. VA has a smaller mean particle s 5.77 µm, while LSP has a larger mean particle size of 12.05 µm. The specific surface of VA is larger than that of LSP by about five times. The particle size distribution c of LSP and VA are provided in Figure 1.  Figure 2 reveals a micrograph of LSP and VA examined using a JSM-5800LV ning electron microscope (SEM). LSP has a large size with a round edge polycrysta shape, while VA has a smaller particle size with an angular shape coupled with elong flakiness. Figure 3 shows the XRD result for the LSP and VA. LSP is crystalline in n and contains mainly calcite (CaCO3), while VA is amorphous in nature and contain gioclase (Ca,Na)Al2Si2O8, quartz (SiO2) and microcline (KAl2Si2O8).   Figure 2 reveals a micrograph of LSP and VA examined using a JSM-5800LV scanning electron microscope (SEM). LSP has a large size with a round edge polycrystal-like shape, while VA has a smaller particle size with an angular shape coupled with elongated flakiness.

Effect of Silica Modulus on Workability of AANL
The workability of AANLM x , where x is the mass ratio of Na 2 SiO 3(aq) /10 M NaOH (aq) (x = SS NH , x = 0, 0.5, 0.75, 1, 1.25 and 1.5), is presented in Figure 4. The mixtures were prepared to study the combined effect of SS and NH i.e., (x = 0.5, 0.75, 1, 1.25 and 1.5) and the sole effect of 10 M NaOH (aq) (x = 0). The silica modulus (SiO 2 /Na 2 O) was calculated from the mass ratios as tabulated in Table 4. The corresponding silica modulus (M s ) of x = 0, 0.5, 0.75, 1, 1.25 and 1.5 are 0, 0.52, 0.72, 0.89, 1.05 and 1.18, respectively. bility than using only NaOH(aq), as shown in Figure 5. The higher workability observed in AANLM1.0 could be as a result of higher H2O/SiO2 (5.46) H2O/Na2O (6.12) present in the mix compared to the AANLM0.    The workability of the AANLM increases as the SS NH increased from 0.5 to 1 until optimum values of silica modulus of 0.89, and thereafter, a reduction in the workability was observed. There were 18.51% and 62.96% increments in AANLM 0.75 and AANL 1 , respectively, in comparison with AANLM 0.5 , as shown in Figure 4. However, a slight reduction was observed when the silica modulus increased from 0.89 to 1.05 and 1.18. There was a 4.55% and 11.36% reduction in AANLM 1.25 and AANLM 1.5 in comparison to AANLM 1 . It is what noticing that the combination of Na 2 SiO 3 and NaOH (aq) resulted in higher workability than using only NaOH (aq) , as shown in Figure 5. The higher workability observed in AANLM 1.0 could be as a result of higher H 2 O/SiO 2 (5.46) H 2 O/Na 2 O (6.12) present in the mix compared to the AANLM 0 .

Effect of Silica Modulus on Compressive Strength Development
The initial oxide composition and the oxide composition ratio of the alkaline activators are shown in Table 4 [39]. NaOH(aq) and Na2SiO3(aq) dissolution processes are shown in Equations (1)-(4).
x NaOH(aq) + y H2O(l) → Na2O + H2O + y H2O(l) (1) Na2SiO3(aq) → Na2O + SiO2 + H2O (2) (Na2O: 8.76%, SiO2: 29.13%, H2O: 62.11%) where x, y and z represent the molar concentration of NaOH(aq), added distilled water and the mixing water, respectively. The total mass concentrations of Na2O, SiO2 and H2O and the mass ratio of silica modulus, (Ms) SiO2/Na2O, H2O/Na2O and H2O/SiO2 are as shown in Table 4 increase in Na2SiO3(aq)/10 M NaOH(aq) (x = ) from 0 to 1.5 leads to an increase in the total SiO2 and a decrease in the total Na2O and the H2O present in the mix proportion. Furthermore, the silica modulus (SiO2/Na2O) and H2O/SiO2 increases while the H2O/Na2O decrease as increased, as shown in Table 4. The impact of variation in the silica modulus on the compressive strength of the synthesised alkali-activated mortar is shown in Figure 6. Generally, the early compressive strength for all the mixes increases with the curing age till optimum values of silica modulus (0.89), and thereafter, a reduction in the compressive strength was observed. It should be noted that, after the 7-day strength, the mixes after the optimum silica modulus, i.e., silica modulus 1.25 and 1.18, showed a drop in strength.
The maximum strength of 27 MPa was obtained in with silica modulus of 0.89 after 28 days of curing; however, the compressive strength reduced sharply by 40% and 62%, respectively, in .

and
. due to the presence of more SiO2 than Na2O in the mix, as shown in Table 4. At lower silica modulus (0-0.72) below the optimum silica modulus (0.89), the compressive strength recorded was lower due to the presence of excess alkalis (Na2O) in the mixture, which caused negative effects such as efflorescence and brittleness of the binder product. Above the optimum silica modulus

Effect of Silica Modulus on Compressive Strength Development
The initial oxide composition and the oxide composition ratio of the alkaline activators are shown in Table 4 [39]. NaOH (aq) and Na 2 SiO 3(aq) dissolution processes are shown in Equations (1)-(4).  Table 4 increase in Na 2 SiO 3(aq) /10 M NaOH (aq) (x = SS NH ) from 0 to 1.5 leads to an increase in the total SiO 2 and a decrease in the total Na 2 O and the H 2 O present in the mix proportion. Furthermore, the silica modulus (SiO 2 /Na 2 O) and H 2 O/SiO 2 increases while the H 2 O/Na 2 O decrease as SS NH increased, as shown in Table 4. The impact of variation in the silica modulus on the compressive strength of the synthesised alkali-activated mortar is shown in Figure 6. Generally, the early compressive strength for all the mixes increases with the curing age till optimum values of silica modulus (0.89), and thereafter, a reduction in the compressive strength was observed. It should be noted that, after the 7-day strength, the mixes after the optimum silica modulus, i.e., silica modulus 1.25 and 1.18, showed a drop in strength.
The maximum strength of 27 MPa was obtained in AANLM 1 with silica modulus of 0.89 after 28 days of curing; however, the compressive strength reduced sharply by 40% and 62%, respectively, in AANLM 1.25 and AANLM 1.5 due to the presence of more SiO 2 than Na 2 O in the mix, as shown in Table 4. At lower silica modulus (0-0.72) below the optimum silica modulus (0.89), the compressive strength recorded was lower due to the presence of excess alkalis (Na 2 O) in the mixture, which caused negative effects such as efflorescence and brittleness of the binder product. Above the optimum silica modulus (0.89), the compressive strength recorded decreased drastically due to the presence of excess alkalis (SiO 2 ) in the mixture. The excess SiO 2 caused a reduction in pH, increase in viscosity and degree of polymerisation of silicate species of alkaline solution leading to a reduction in the reactivity of alkaline solution [17]. The trends observed clearly show the dependency of compressive strength development on the silica modulus of the alkaline activators. Silica modulus above and below the optimal values was found to be not suitable for achieving higher compressive strength. Similar trends were reported previously in the literature [17].
Materials 2021, 14, x FOR PEER REVIEW 9 of 20 (0.89), the compressive strength recorded decreased drastically due to the presence of excess alkalis (SiO2) in the mixture. The excess SiO2 caused a reduction in pH, increase in viscosity and degree of polymerisation of silicate species of alkaline solution leading to a reduction in the reactivity of alkaline solution [17]. The trends observed clearly show the dependency of compressive strength development on the silica modulus of the alkaline activators. Silica modulus above and below the optimal values was found to be not suitable for achieving higher compressive strength. Similar trends were reported previously in the literature [17].

XRD Characterisation of Varied Silica Modulus on the Mortar Binder
The XRD spectra of the binder product that gave the best and worst results for compressive strength are shown in Figure 7. Sole usage of sodium hydroxide solution (10 M NaOH(aq)) as alkaline activator resulted in the formation of calcite (CaCO3), quartz (SiO2), anorthite (Na48Ca52(Si2.5Al1.5)O8 and kaolinite (Al2Si2O9H4). However, when a combined sodium hydroxide solution and sodium silicate were used, the alkaline activated products formed were anorthite (CaAl2Si2O8), gehlenite (CaO.Al2O3.SiO2) and albite (NaAlSi3O8). Anorthite (CaAl2Si2O8) and gehlenite (CaO.Al2O3.SiO2) compounds are similar to the calcium-aluminosilicate-hydrate (C-A-S-H) compound. The use of combined sodium hydroxide solution and sodium silicate enhanced the strength development than using only sodium hydroxide solution.

XRD Characterisation of Varied Silica Modulus on the Mortar Binder
The XRD spectra of the binder product that gave the best and worst results for compressive strength are shown in Figure 7. Sole usage of sodium hydroxide solution (10 M NaOH (aq) ) as alkaline activator resulted in the formation of calcite (CaCO 3 ), quartz (SiO 2 ), anorthite (Na 48 Ca 52 (Si 2.5 Al 1.   Figure 8 shows the FTIR spectra of AANLM binder, activated using only 10M NaOH(aq) (Ms = 0) and combined NaOH(aq) and Na2SiO3(aq) (Ms = 0.89). From Figure 8a Figure 8 shows the FTIR spectra of AANLM binder, activated using only 10M NaOH (aq) (M s = 0) and combined NaOH (aq) and Na2SiO3 (aq) (M s = 0.89). From Figure 8a

SEM + EDX Characterisation of Silica Modulus Effect
Figures 9 and 10 show the micrograph and EDX results of the AANLM binder, activated using only 10 M NaOH(aq) (Ms = 0) and combined NaOH(aq) and Na2SiO3(aq) (Ms = 0.89) paste for the best and worst compressive strength results, respectively. Figure 9 shows the SEM and EDX pattern result of the binder using only 10M NaOH(aq) (Ms = 0) The image reveals non-homogeneous and non-compacted structures with connected micropores. This indicates that the degree of polymerisation at low silica modulus (Ms = 0) is slow, resulting in the low compressive strength recorded in AANLM0. However, as the silica modulus increases, the SEM image for AANLM1 revealed a homogenous and denser microstructure due to more formation of alkaline activated products. The EDX results for AANLM0 indicates a higher ratio of Si/Na (7.89) at spectrum 1 compared to the values of Si/Na (0.96) at spectrum 3 and Si/Na (1.4) at spectrum 4 in AANLM1. This indicates the presence of excess unreacted silica gel in AANLM0. Furthermore, the Si/Ca (14.2) in AANLM0 (spectrum 1) is higher than the Si/Ca (2.33) present in AANLM1 (spectrum 4).  Figures 9 and 10 show the micrograph and EDX results of the AANLM binder, activated using only 10 M NaOH (aq) (M s = 0) and combined NaOH (aq) and Na 2 SiO 3(aq) (M s = 0.89) paste for the best and worst compressive strength results, respectively. Figure 9 shows the SEM and EDX pattern result of the binder using only 10M NaOH (aq) (M s = 0). The image reveals non-homogeneous and non-compacted structures with connected micropores. This indicates that the degree of polymerisation at low silica modulus (M s = 0) is slow, resulting in the low compressive strength recorded in AANLM 0 . However, as the silica modulus increases, the SEM image for AANLM 1 revealed a homogenous and denser microstructure due to more formation of alkaline activated products. The EDX results for AANLM 0 indicates a higher ratio of Si/Na (7.89) at spectrum 1 compared to the values of Si/Na (0.96) at spectrum 3 and Si/Na (1.4) at spectrum 4 in AANLM 1 . This indicates the presence of excess unreacted silica gel in AANLM 0 . Furthermore, the Si/Ca (14.2) in AANLM 0 (spectrum 1) is higher than the Si/Ca (2.33) present in AANLM 1 (spectrum 4).

Effect of Curing Temperatures on Compressive Strength Development
The compressive strength of alkali-activated limestone and volcanic ash powder mortar cured at various curing temperatures is revealed in Figure 11. Generally, the compressive strength for all the mixes increases with curing age as well as the curing temperature. It was observed that the early strength (one and three days) of the mortar increases with the corresponding increased in the curing temperature. The sample cured at room temperature (25 • C) exhibited the lowest compressive strength of 6.85 MPa and 7.73 MPa after 1 day and 3 days of curing. The 1-day compressive strength increased by 55.9%, 149.05%, 199.27% and 240.88% after curing at 45 • C, 60 • C, 75 • C and 90 • C, respectively. A similar trend was observed for the 3-days compressive strength. However, at 7, 14 and 28 days, the compressive strength increases at the temperature increased up to 75 • C, after which a dropped in the compressive strength was observed.     after 1 day and 3 days of curing. The 1-day compressive strength increased by 55.9%, 149.05%, 199.27% and 240.88% after curing at 45 ℃, 60 ℃, 75 ℃ and 90 ℃, respectively. A similar trend was observed for the 3-days compressive strength. However, at 7, 14 and 28 days, the compressive strength increases at the temperature increased up to 75 ℃, after which a dropped in the compressive strength was observed. Figure 11. Effect of temperature on compressive strength of AANL mixes.
The maximum 28-days compressive strength of 27 MPa was achieved at the optimum curing temperature of 75 ℃. Further increase in the curing temperature to 90 ℃ caused a reduction in the compressive strength by 4% due to drying effects. The curing temperature has a significant effect on the compressive strength development of alkali-activated mortar. An increase in temperature from room temperature of 25 ℃ to a higher curing temperature of 75 ℃ enhanced the geopolymerisation process of the alkaline activated binder.

XRD Characterisation of Varied Silica Modulus on the Mortar Binder
The XRD spectra of the binder cured at 25 ℃, 45 ℃, 75℃ and 90℃ are shown in Figure 12. The binder cured at room 25 ℃ lead to the formation of albite (NaAlSi3O8), calcite (CaCO3), dolomite (MgCa(CO3)2) and unreacted quartz (SiO2), as revealed in Figure  12a (25 ℃). However, when the curing heat increased to 45 ℃, anorthite (CaAl2Si2O8), calcite (CaCO3) and quartz (SiO2) were formed as revealed in Figure 12b. The anorthite present in this binder, which is similar to the C-A-S-H product, enhanced the slight strength development of the binder. Further increased in the curing temperature to 75 ℃ favoured the geopolymerisation process of the activated binder, which resulted in the formation of anorthite (CaAl2Si2O8), gehlenite (CaO.Al2O3.SiO2) and albite (NaAlSi3O8) as revealed in Figure 12c The maximum 28-days compressive strength of 27 MPa was achieved at the optimum curing temperature of 75 • C. Further increase in the curing temperature to 90 • C caused a reduction in the compressive strength by 4% due to drying effects. The curing temperature has a significant effect on the compressive strength development of alkali-activated mortar. An increase in temperature from room temperature of 25 • C to a higher curing temperature of 75 • C enhanced the geopolymerisation process of the alkaline activated binder.

XRD Characterisation of Varied Silica Modulus on the Mortar Binder
The XRD spectra of the binder cured at 25 • C, 45 • C, 75 • C and 90 • C are shown in Figure 12 Figure 12d, that the peak of quartz compound at 2-theta angle of 27.24 • is lesser than the peak present in the sample cured at 75 • C. Furthermore, the anorthite (CaAl 2 Si 2 O 8 ) and gehlenite (CaO.Al 2 O 3 .SiO 2 ) compounds present in the binder 75 • C curing temperature disappeared at 90 • C curing temperature. This implies that at an optimum curing temperature of 75 • C, geopolymerisation and condensation of silica compounds present in the mixing matrix are better enhanced. This aforementioned process contributes positively to the high compressive recorded for the binder-activated binder cured at 75 • C. Therefore the optimum temperature for the synthesised binder is 75 • C.
(CaO.Al2O3.SiO2) compounds at the 2-theta angle of 27.35° and 32.34°, respectively, enhanced the strength development. However, upon an increase in the curing temperature to 90 °C, it was observed, as shown in Figure 12d, that the peak of quartz compound at 2-theta angle of 27.24° is lesser than the peak present in the sample cured at 75 °C . Furthermore, the anorthite (CaAl2Si2O8) and gehlenite (CaO.Al2O3.SiO2) compounds present in the binder 75 °C curing temperature disappeared at 90 °C curing temperature. This implies that at an optimum curing temperature of 75 °C , geopolymerisation and condensation of silica compounds present in the mixing matrix are better enhanced. This aforementioned process contributes positively to the high compressive recorded for the binder-activated binder cured at 75 °C . Therefore the optimum temperature for the synthesised binder is 75 °C.

FTIR Analysis of Silica Modulus Effect
The FTIR spectra of the binder cured at 25 • C, 45 • C, 75 • C and 90 • C are shown in Figure 13. The spectra revealed some structural changes in the bond characteristics due to changes in curing temperature. From Figure 13a-c, the FTIR broad bands located at 3700-2900 cm −1 signify the presence of O-H stretching. The band is broader in both room cured samples and samples cured at 45 • C. The depth of the peak reduces as the curing temperature increases, and this peak disappeared at a curing temperature of 90 • C Figure 13d. Similar trends were observed for H-O-H bending that occurred between wavenumbers of 2380 and 2287 cm −1 and 1700 cm −1 to 1600 cm −1 ; this peak also disappeared at a curing temperature of 90 • C. The binder cured at room temperature and 45 • C had a broader trough compared to the one cured at 75 • C and 90 • C. This shows that there were more weak O-H and H-O-H bonds present in the former than the latter. This caused the low strength observed at lower curing temperatures. Figure 13. The spectra revealed some structural changes in the bond characteristics due to changes in curing temperature. From Figure 13a to Figure 13c, the FTIR broad bands located at 3700-2900 cm −1 signify the presence of O-H stretching. The band is broader in both room cured samples and samples cured at 45 °C. The depth of the peak reduces as the curing temperature increases, and this peak disappeared at a curing temperature of 90 °C Figure 13d. Similar trends were observed for H-O-H bending that occurred between wavenumbers of 2380 and 2287 cm −1 and 1700 cm −1 to 1600 cm −1 ; this peak also disappeared at a curing temperature of 90 °C. The binder cured at room temperature and 45 °C had a broader trough compared to the one cured at 75 °C and 90 °C. This shows that there were more weak O-H and H-O-H bonds present in the former than the latter. This caused the low strength observed at lower curing temperatures. Stretching vibration of C-O-O was observed between wavenumbers of 1420 and 1313 cm −1 in the samples cure at 25 ℃, 45 ℃, 75 ℃ and 90 ℃. The depth of the peak reduces as the curing temperature increases. The FTIR pattern between 900 cm −1 and 1300 cm −1 associated with the Si-O-T (T = Si or Al) asymmetric vibration were also observed in the binder at all the curing temperature. In-plane bending vibration of C-O was also observed at a band of the weak and broad absorption peak of 657 cm −1 . Figures 14-16 show the SEM and EDX analysis of AANL paste for the room cured sample, the optimum temperature (75 ℃) and the sample cured above the optimum temperature (90 ℃), respectively. Figure 14 shows the SEM and EDX result of the activated binder cured at 25 ℃. the micrograph reveals a big size particle of non-homogeneous and non-compacted microstructure. This implies that the level of polymerization at low curing temperature such as room temperature does not favour the geopolymerisation process of alkaline activated LSP and VA, resulting in the low compressive strength recorded in the sample cured at room temperature. However, as the curing temperature increased to the optimum curing temperature of (75 ℃), the SEM image ( Figure 15) showed a homogenous Stretching vibration of C-O-O was observed between wavenumbers of 1420 and 1313 cm −1 in the samples cure at 25 • C, 45 • C, 75 • C and 90 • C. The depth of the peak reduces as the curing temperature increases. The FTIR pattern between 900 cm −1 and 1300 cm −1 associated with the Si-O-T (T = Si or Al) asymmetric vibration were also observed in the binder at all the curing temperature. In-plane bending vibration of C-O was also observed at a band of the weak and broad absorption peak of 657 cm −1 . Figures 14-16 show the SEM and EDX analysis of AANL paste for the room cured sample, the optimum temperature (75 • C) and the sample cured above the optimum temperature (90 • C) respectively. Figure 14 shows the SEM and EDX result of the activated binder cured at 25 • C. The micrograph reveals a big size particle of non-homogeneous and non-compacted microstructure. This implies that the level of polymerization at low curing temperature such as room temperature does not favour the geopolymerisation process of alkaline activated LSP and VA, resulting in the low compressive strength recorded in the sample cured at room temperature. However, as the curing temperature increased to the optimum curing temperature of (75 • C), the SEM image ( Figure 15) showed a homogenous and denser microstructure due to more formation of alkaline activated product. The EDX results for the sample cured at room temperature (25 • C) indicates a higher ratio of Si/Na (4.96) at spectrum 46 and Si/Na (5.67) at spectrum 47 compared to the values of Si/Na (0.96) at spectrum 5 and Si/Na (1.4) at spectrum 6 of sample cured at (75 • C). This indicates the presence of excess unreacted silica gel in AALN mortar cured at room temperature. Furthermore, the Si/Ca (12.54) in the room cured samples (spectrum 46) is higher than the Si/Ca (0.83) and 2.33 present in samples cured at the optimum temperature as revealed in spectrum 5 and spectrum 6, respectively. However, upon increasing the curing temperature to 90 • C, three distinct layers were observed in Figure 16, spectrum 64 revealed Si/Ca (0.89) slightly above the Si/Ca (0.83) obtained in sample cured at the optimum temperature, while spectrum 65 showed Si/Ca (5.4) higher than the one observed at the optimum curing temperature and lower than the one obtained at room cured temperature. This also implies that the formation of alkaline activated products was also formed at a 90 • C curing temperature; however, the formation of CaO at spectrum 66 showed the decomposition of the alkaline product could have occurred at 90 • C, which could have been responsible for the slight decline in the compressive strength observed at this particular curing temperature.

SEM + EDX Characterisation of Curing Temperature Effect
(0.96) at spectrum 5 and Si/Na (1.4) at spectrum 6 of sample cured at (75 ℃). This indicates the presence of excess unreacted silica gel in AALN mortar cured at room temperature. Furthermore, the Si/Ca (12.54) in the room cured samples (spectrum 46) is higher than the Si/Ca (0.83) and 2.33 present in samples cured at the optimum temperature as revealed in spectrum 5 and spectrum 6, respectively. However, upon increasing the curing temperature to 90 ℃, three distinct layers were observed in Figure 16, spectrum 64 revealed Si/Ca (0.89) slightly above the Si/Ca (0.83) obtained in sample cured at the optimum temperature, while spectrum 65 showed Si/Ca (5.4) higher than the one observed at the optimum curing temperature and lower than the one obtained at room cured temperature. This also implies that the formation of alkaline activated products was also formed at a 90 ℃ curing temperature; however, the formation of CaO at spectrum 66 showed the decomposition of the alkaline product could have occurred at 90 ℃, which could have been responsible for the slight decline in the compressive strength observed at this particular curing temperature.

Conclusions
The impact of silica modulus (M s ) and curing temperature on workability, compressive strength, bond properties, reaction products and the microstructures of alkali-activated binder synthesised from the binary blending of volcanic ash powder (VA) and limestone powder (LSP) were investigated. This study has contributed to understanding the impact of silica modulus and curing temperature on the strength and microstructural characteristics of the alkali-activated volcanic ash/limestone powder mortar. The utilisation of VA/LSP will be an addition to sustainable alternative binders and waste reduction.
The following conclusions were deduced: 1.
The flow of the developed alkali-activated binder increases as silica modulus increased from zero to the optimum values of 0.89, and thereafter, a reduction in the workability was observed. All the flow values for all the developed binder were in an acceptable range of 135 mm to 220 mm; 2.
The maximum strength of 27 MPa was obtained with silica modulus of 0.89 after 28 days of curing; however, the compressive strength reduced sharply above the optimum silica modulus due to the presence of more SiO 2 than Na 2 O in the mix; 3.
More than 70% of the 28-day compressive strength could be achieved within 12 h of curing with the usage of combined Na 2 SiO 3(aq) /10 M NaOH at optimum M s ; 4.
Samples synthesised with sole 10 M NaOH (aq) activator resulted in a binder with a low 28-day compressive strength (15 MPa) compared to combined usage of Na 2 SiO 3(aq)/ 10 M NaOH (aq) activators;