*2.1. Materials and Mixture Proportions of AAM Mortar*

Table 1 shows the physical properties and chemical compositions of the ground granulated blast furnace slag (GGBFS), fly ash (FA), and calcium sulfoalumiante expansive additive (CSA EA) used in AAM mortar. GGBFS satisfied the third type of KS F2563 (density: >2.8 g/cm3, fineness: 4000–6000 cm2/g), and FA met the second type of KS L5405 (density: >1950 g/cm3, fineness: >3000 cm2/g). Both GGBFS and FA were produced by the company Sampyo in South Korea [45,46]. The CSA EA used in this study was POWER CSA TYPE from Denka in Japan. It was used to compensate for the shrinkage of the AAM mortar and was added on the basis of the binder mass. The main chemical components of GGBFS were CaO (41.9%), SiO2 (13.8%), and Al2O3 (4.9%). Its basicity coefficient (Kb = (CaO + MgO)/(SiO2 + Al2O3)) and hydration modulus (HM = (CaO + MgO + Al2O3)/SiO2) were 0.99 and 1.82, respectively. Kb was close to 1.0, which is a neutral value for ideal alkali activation, and HM was higher than 1.4, which is a value for excellent hydration reaction. FA was composed of SiO2 (56.8%), Al2O3 (22.8%), Fe2O3 (6.9%), and M2O(K2O+Na2O) (1.9%). The main components of CSA EA were lime, gypsum, and bauxite, and they were composed of CaO (34.6%), SiO (30.2%), and Al2O3 (24.2%). The alkali activator was used to accelerate the reaction of the binder. The alkali activator was in the form of a white powder with a molar ratio of 0.95. In addition, alkali activators are manufactured separately by adjusting the chemical components. The SiO2/Na2O ratio of the alkali activator used in this study was 0.92. The fine aggregate used was river sand with a density of 2.53, water absorptivity of 1.08, and fineness modulus of 2.77.


**Table 1.** Physical properties and chemical composition of alkali-activated material (AAM) binder and calcium sulfoalumiante expansive additive (CSA EA).

Table 2 summarizes the mix proportions of the AAM mortar. A binary blended binder in which GGBFS and FA were mixed at a ratio of 7:3 was used, and 24% alkali activator was used compared to the unit water content. The contents of CSA EA used were 0%, 2.5%, 5.0%, and 7.5% compared to the amount of the binder. The water/binder (W/B) ratio was 45.1%, and the sand/binder (S/B) ratio was 1.2. For the mixing method, the basic binder, alkali activator, and CSA EA were placed in a test container and dry mixing was performed for 30 s. Then, water was added, and mixing was performed for ten min at 150 rpm to produce paste. In addition, fine aggregate was added, and mixing was performed for 90 s at 300 rpm.


**Table 2.** Mix proportions of alkali-activated material (AAM) mortar.

#### *2.2. Test Methods*

#### 2.2.1. Setting Time Test

To evaluate the setting characteristics of the AAM mortar, the mortar Vicat test was conducted in accordance with ASTM C191-18a [47]. The initial and final settings of the AAM mortar were determined using the penetration depth of the Vicat needle (diameter: 1.00 ± 0.05 mm, length: 50 mm or more) in a mortar for 30 s. The Vicat needle penetration tests were conducted every five min, and the time at which no trace of the Vicat needle was observed on the AAM mortar surface was determined as the final setting time.

#### 2.2.2. Compressive Strength Test

The compressive strength test was conducted in accordance with ASTM C109-16a [48]. The diameter and height of the cylindrical specimens were 100 and 200 mm, respectively, and the tests were conducted at curing ages of 1, 7, and 28 days. The average compressive strengths of the three specimens were used. The specimens were cured in a constant temperature and humidity chamber at a temperature of 23 ± 2 ◦C and a relative humidity of 90 ± 2%.

#### 2.2.3. Modulus of Elasticity Test

The concrete modulus of elasticity test was conducted in accordance with ASTM C469M-14 [49]. The diameter and height of the cylindrical specimens were 100 and 200 mm, respectively. Two strain gauges were attached to the side of each specimen to measure the modulus of elasticity. A load of up to 40% of the ultimate load was applied at a rate of 0.25 MPa/s, and the modulus of elasticity was calculated through regression based on the interpolation of strain measurements for the load. The modulus of elasticity test results was obtained by averaging the modulus of elasticity measurement results of three specimens. The specimens were cured in a constant temperature and humidity chamber at a temperature of 23 ± 2 ◦C and a relative humidity of 90 ± 2%.

#### 2.2.4. Ultrasonic Pulse Value (UPV) Test

The system used for monitoring the UPV in the AAM mortar specimens is shown in Figure 1. A hole was made in the Styrofoam container, and an oscillator and a receiver were placed so that they could face each other at a distance of 30 mm. After filling the container with mortar, the transit time of ultrasonic waves from the oscillator to the receiver was measured every 30 s. The Pundit-2 model from PROCEQ was used to measure the UPV. UPV (Vc) was obtained by dividing the length of the specimen (L) by the transit time (T), and it has the following relationship with the dynamic modulus of elasticity (Ed) and density (ρ) [50–52].

$$V\_c = \frac{\mathcal{L}}{\mathcal{T}} = \sqrt{\frac{E\_d}{\rho}}\tag{1}$$

Figure 2 shows the typical evolution curve of the UPV of the AAM mortar, which was altered with three sections. According to the curve, points A and B can be determined in terms of altered time points. The UPV begins to increase at point A, and it begins to slow down and converges to a certain value at point B [53–55].

**Figure 1.** Schematic of ultrasonic pulse velocity (UPV) monitoring.

**Figure 2.** Typical evolution curve of ultrasonic pulse velocity.

#### **3. Results and Discussion**

#### *3.1. Setting Time*

The setting of AAM mortar is affected by the conditions of the alkali activator, and it is generally faster than that of ordinary Portland cement (OPC) [56,57]. Table 3 shows the setting time of the AAM mortar according to the content of EA. For 0 EA without EA, the initial setting time (IST) was 101 min and the final setting time (FST) was 292 min. As for paste (w/c = 0.5) using ordinary Portland cement (OPC), the initial setting is within 6 to 7 h according to the literature [58]. Under the same mixing conditions, the AAM mortar exhibited a faster setting compared to mortar that used cement as a binder. This is because the alkali activator stimulated the binder quite early [59].


**Table 3.** Setting time and strength characteristics of the AAM mortar mixed with EA.

The setting time was accelerated as the CSA EA content increased. For 7.5 EA with the highest CSA EA content, the initial setting time and final setting time were 55 and 171 min, respectively, which were 45.4% and 41.4% shorter compared to those of 0 EA. When a CaO-based CSA EA is used in mortar, Ca2<sup>+</sup> ions diffuse and generate hydrates (Ca(OH)2). The setting characteristic of OPC is highly associated with the hydration of C3S (Ca3SiO5, alite). An OPC paste reaches its initial setting state once C3S has been sufficiently hydrated, and the final setting occurs when a sufficient amount of C-S-H has precipitated. On the other hand, the main constituent of CSA expansive additives is C4A3s (Ye'elimite), which dissolves in contact with water and forms ettringite (Ca6Al2(SO4)3(OH)12·26H2O) at a much faster rate than C3S does.

This action, which is similar to the alkali activation reaction of AAM, accelerates the reaction of AAM by increasing the dissolution of Ca2<sup>+</sup> ions in the mixing water, thereby further accelerating setting [59,60].

#### *3.2. Compressive Strength and Modulus of Elasticity*

When a CSA EA is used in mortar, the mortar temporarily expands at the initial stage owing to the hydrates of EA, and its internal structure becomes dense. This increases the strength and modulus of elasticity of the mortar [1]. Table 3 shows the compressive strength and modulus of elasticity of AAM mortar mixed with CSA EA.

The average compressive strength of the three AAM mortars exceeded 40 MPa, which was the target strength set during the mix design. The AAM mortar without EA (0 EA) exhibited high initial strength, with a compressive strength of 3.2 MPa and modulus of elasticity of 1.2 GPa at 1 day of age. However, for the AAM mortar mixed with CSA EA, the compressive strength was higher than 5 MPa, and the modulus of elasticity was higher than 2 GPa at 1 day of age, which were 63–87% and 76.3–103.5% higher, respectively, compared to those of 0 EA. AAM mortar containing CSA EA exhibited higher compressive strength values at 28 days of age compared to 0 EA. At one day of age, the strength and modulus of elasticity increased as the EA content increased. However, at 28 days of age, the compressive strength of 5.0 EA was the highest (49.6 MPa), and the compressive strength of 7.5 EA was 47.3 MPa, which was lower than that of 2.5 EA.

When the modulus of elasticity of the AAM mortar at 28 days of age was compared, the modulus of elasticity of the AAM mortar containing CSA EA was higher than that of 0 EA, but there was no clear improvement in the modulus of elasticity due to EA. This is because the modulus of elasticity was not significantly affected by the binder, as it was also affected by factors other than the cement matrix, such as fine aggregate.

#### *3.3. Ultrasonic Pulse Velocity*

UPV can measure changes in the internal structure of mortar caused by a hydration reaction without physical damage using the speed of sound waves, and it is possible to identify sudden changes in the internal structure of mortar through the monitoring of UPV. According to Lee et al., the UPV of mortar and concrete has three different sections [61]. The initial fluid state section (step 1) represents the unhardened state of the mortar. In this instance, the UPV exhibits a level similar to that of the fluid state and ranges from 300 to 500 m/s. In step 2, pores are filled with the hydration products of cement, and moisture unsaturated porous solid structures are connected, thereby increasing the UPV. In the step 3 section, high UPV is maintained because the hydrates are connected to each other, and mortar begins to develop strength owing to the curing process. Micropores are filled with hydration products, and UPV continuously increases, even though it does not increase as rapidly as in step 2.

Figure 3 shows the UPV curves of the AAM mortar according to the EA content. The UPV of the AAM mortar incorporating CSA EA was 348–412 m/s, corresponding to the values typically shown by a mixture in the fluid state. An increase in the UPV value was observed after 0.5 h by 7.5 EA, which incorporated the highest dosage of CSA EA and was the first to enter step 2, followed by 2.5 EA and 0 EA after 1.5 h. It was observed that the UPV rapidly increased in step 2, while its increasing rate decreased after 3–8 h, reaching step 3. Similarly, the time taken to reach step 3 was the shortest in 7.5 EA. The rate at which the UPV increased tended to gradually decrease 48 h after the specimens reached step 3.

**Figure 3.** AAM mortar of ultrasonic pulse velocity curve: (**a**) Time: 3 h, (**b**) Time: 48 h.

The curves show shapes similar to those of the UPV curves of other researchers, but there are some differences. In a study by Lee et al. [61], the duration of step 1 for OPC mortar was 6–10 h, whereas that for the AAM mortar was 0.5–1.5 h, suggesting that the duration of step 1 was much shorter for AAM. In addition, while step 2 was reached after a sudden increase from step 1 for the UPV curves of the AAM mortar, the UPV curves of the OPC mortar continuously increased without clear boundary points to reach step 2. This is because the rapid hydration reaction occurred as AAM mortar-stimulated GGBFS using the alkali activator. In addition, CSA EA further reduced step 1 as it accelerated the hydration reaction. Lee et al. [61] suggested that OPC mortar reaches step 3 after 16–20 h, while it was shown that AAM took 3–8 h to reach step 3, showing a much faster rate than OPC mortar. This is because the rapid reaction of AAM mortar occurred as described above. The AAM mortar containing EA exhibited a higher UPV at the same age. This is due to the generation of ettringite, which is the main hydration product of EA.

#### *3.4. Setting Time in UPV Test*

The UPV curve can reflect changes in the internal structure of mortar caused by the hydration reaction. Pessiki and Carino et al. analyzed the relationship between UPV and setting or strength using the UPV curve [62]. According to a report by Lee et al., the analysis of the relationship between the UPV and the setting of concrete revealed that UPV increased when setting began. It was confirmed that initial and final settings occurred when UPV reached certain values [61,63,64].

The UPV measurement for AAM according to each step defined in Figure 2 is analyzed in Figure 4. As can be seen from the figure, the UPV curves are divided into three sections. According to a report by Lee, the intersections of the straight lines created by connecting UPV values in each section are related to setting. As the UPV curves of the AAM mortar have three different sections, straight lines were created in each section by connecting the data, and the intersections of the lines were compared with the setting time by the ASTM C191-18a test method.

Step 1 lasted for approximately 60 min (30 min for EA7.5), and the UPV results were used to create a straight line (black curve in Figure 4). The UPV sharply increased afterwards, and step 2 began approximately 90–120 min after mixing. In this instance, the use of EA advanced the time at which step 2 began. During step 2, the UPV results for an hour were used to create a straight line (blue curve in Figure 4). Step 3 began approximately eight hours after mixing. Mortar mixtures tend to reach step 3 within 3–8 h after setting. The point at which the UPV converges has been determined by regression analysis (blue curve in Figure 4).

**Figure 4.** UPV curves of AAM mortar: (**a**) 0 EA, (**b**) 2.5 EA, (**c**) 5.0 EA, (**d**) 7.5 EA.

To determine the setting time of the AAM mortar using the UPV curves, the intersections of the straight lines created in each section of the curves were examined. The intersection of the straight lines of the step 1 and step 2 sections was defined as the initial setting, and that of the straight lines of the step 2 and step 3 sections was defined as the final setting. The analysis results showed that the initial setting time of 0 EA was 102 min, and its final setting time was 294 min. In the same manner, the initial setting and final setting of 2.5, 5.0, and 7.5 EA are shown in each graph. The initial and final setting times determined using UPV became shorter as the EA content increased. Table 4 shows the differences between the initial and final setting times determined by the UPV curves and those determined by ASTM C 191-18a. The difference in initial setting time ranged from 1 to 14 min, and that in the final setting ranged from 1 to 10 min. In particular, when the EA content was 2.5% or less, the difference in setting time was 3 min or less, resulting in no significant difference in the results of the two test methods.


IST \*: Initial setting time; FST \*\*: Final setting time.

#### *3.5. Elasticity in UPV Test*

The dynamic modulus of elasticity of mortar can be obtained through an empirical equation if dynamic characteristics, such as UPV, are used.

$$E\_d = \rho V\_c^2 \tag{2}$$

where *Ed* is the dynamic modulus of elasticity (GPa), ρ is the density of the specimen (ton/m3), and *Vc* is UPV (m/s). Table 5 shows the UPV results of the AAM mortar at one day of age, the Ed values obtained using the results, and the static modulus of elasticity (Es) values at one day of age obtained through an experiment. The density of the AAM mortar used was 2.4 ton/m3.


**Table 5.** UPV and dynamic and static moduli of elasticity of the AAM mortar at one day of age.

As the EA content increased at 1 day of age, both Ed and Es increased. As the content of CSA EA increased to 2.5, 5.0, and 7.5%, Es increased by 23.5, 32.1, and 45.9% compared to that of 0 EA and Ed increased by 29.1, 37.6, and 49.1%. Here, the difference between the increase rates of Ed and Es of the AAM mortar that used CSA EA based on 0 EA ranged from 3.2% to 5.6%, indicating that Ed and Es showed relatively similar tendencies. This means that Es can be indirectly calculated through UPV because Ed is obtained using the equation of UPV. This approach has been mainly used to analyze the characteristics of rocks, and the correlation between the static and dynamic characteristics of rocks has been verified by several researchers [65–69].

In general, the dynamic modulus of elasticity tends to be higher than the static modulus of elasticity [65,70–72]. This can be because an increase in the crack and pore volume in concrete and mortar decreases their static modulus of elasticity to a larger extent [65,73]. In the test results, Ed by UPV exhibited higher results than Es. It was found that Ed was 62.4–55.4% higher than Es.

Table 6 shows the static modulus of elasticity of the AAM mortar at the final setting time obtained using UPV. The UPV at the accurate final setting time was unknown because it was measured every 30 min. Therefore, the dynamic modulus of elasticity was calculated using UPV at a time point similar to the final setting time. At the final setting time of the AAM mortar, UPV exhibited similar values, ranging from 2039.9 to 2205.0 m/s. Based on this, the dynamic modulus of elasticity was measured to be between 0.98 and 1.14 GPa. The AAM mortar developed stiffness as it was cured, and UPV and the dynamic modulus of elasticity were found to be constant at the final setting time. They were similar regardless of EA content.


**Table 6.** UPV and dynamic modulus of elasticity at the final setting time of the AAM mortar.

#### **4. Conclusions**

In this study, the initial hardening characteristics of alkali-activated material (AAM) mortar were analyzed using the four levels of CSA expansive additive (CSA EA) content. For the analysis of the initial hardening characteristics using ultrasonic pulse velocity (UPV), the results of the Vicat setting test, a conventional test method, were compared with those of the modulus of elasticity test. The main observations and findings of this study can be summarized as follows:


**Author Contributions:** Conceptualization, G.-S.R., S.C.; setting time test, S.C., G.-H.A.; data analysis, S.C.; compressive strength & modulus of elasticity test, G.-H.A., K.-T.K.; data analysis, G.-S.R., H.-Y.K. Y.-J.Y.; UPV test & data analysis, S.C., G.-S.R.; writing, G.-S.R., S.C. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Acknowledgments:** This work was supported by the Korea Institute of Civil Engineering and Building Technology (Project No. 2020-60).

**Conflicts of Interest:** The authors declare no conflict of interest
