*2.2. Performed Tests and Specimens*

In this study, compressive strength, flexural strength, and UPV tests were performed on NCa-ECC specimens. Then, 3D fractal analysis was conducted. Finally, the microstructure of the specimens was investigated using SEM analysis. First, the ultrasonic pulse velocity (UPV) was tested according to ASTM C 597 to determine the UPV values of three 50 × 50 × 50 mm cube samples for each series after 28 days [40]. Then, 50 × 50 × 50 mm cube specimens were placed in a compressive strength testing machine with a capacity of 2000 kN and subjected to compression at a rate of 0.602 MPa/s. Factors such as loading speed, size, and age of the samples were entered into the pressure machine before loading to obtain compressive strength values automatically. Thus, the compressive strength test was performed on three cubes of each series according to ASTM C109 [41]. The flowability of the fresh mixtures was tested according to ASTM C230 [42]. The spreading diameter of the mixtures obtained from this experiment was equal and measured approximately 210 mm. Thus, the fresh mixtures given in Table 3 showed good fluidity without segregation.

In this study, the flexural strength test was performed on 15 × 50 × 350 mm samples to obtain flexural strength, mid-span displacement, strain, and stress-deflection curves according to ASTM C348 [43]. For this proposal, a closed-loop controlled universal testing machine (Figure 3) with a loading rate of 0.003 mm/s was used. Three samples were used for each series, and the average results of the samples were obtained. The flexural stressdeflection curves were obtained using flexural strength values and deflections, recorded by the computer data recording system on the testing machine. Erdem and Gurbuz [44] performed a similar test on hybrid fiber reinforced engineered cementitious specimens.

**Figure 3.** The flexural test set-up.

Figure 4 shows the multiple cracks formed in the samples subjected to the flexural strength test. The crack of the prismatic sample was obtained using a 40× magnification microscope to observe the PVA inside the crack. Figure 5. shows the crack of the sample subjected to the flexural test at 40× magnification. After the strength tests, small pieces of the specimen were taken and subjected to SEM. The microstructure of the samples was studied using SEM analysis.

(**a**)

**Figure 4.** *Cont*.

(**b**)

**Figure 4.** The specimens were subjected to the flexural strength test. (**a**) 0 NCa; (**b**) 1 NCa.

**Figure 5.** The appearance of a crack in a specimen subjected to the flexural test at 40× magnification.

In addition, following the flexural tensile tests, the fractal dimensions of cracks were determined. Images of the samples after the flexural testing were firstly captured using a high-resolution camera. Then, these images were converted from RGB mode to an 8 bytes greyscale and scaled up to reflect the actual dimensions. The main flexural bending moment-induced cracks at the same point for all the samples were digitized for thresholding using open-access digital image analysis software called Image J. Then, these were covered by imaginary meshes with rectangular box sizes containing the number of pixels of the crack image (Figure 6). Next, the number of grid squares to cover the cracks was counted for the plot of In (box count) versus In (box size), which were used to compute the average value of fractal dimension that is the slope of the line joining the logarithm of the number of grid squares encountered by the crack and the logarithm of the square grid dimension.

**Figure 6.** Example of crack analysis with different box sizes using Image-J software. (**a**) Boxing size of 1 unit (**b**) Boxing size of 4 unit.

Then, using the formula established by Guo et al. [45], the composites' dissipated fracture energy (Ws/Gf) was approximated at the macro scale as a function of the surface macro-cracks. The ratio of energy (Ws) produced by crack propagation to fracture energy (Gf) is shown by the value of Ws/Gf.

$$\text{Ws/Gf} = \text{a} \times \text{ (\\$/a) } \text{D}^{1-\text{d}} \tag{1}$$

where a denotes the Euclidean length (equal to the diameter of the tested composite), and D1−<sup>d</sup> denotes the fractal dimension of the crack.

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

#### *3.1. Compressive Strength and Ultrasonic Pulse Velocity (UPV) Results*

The compressive strength and UPV tests after 28 days of the fabricated series with three different ratios of nanocalcite are shown in Figures 7 and 8. Compressive strength results were obtained by averaging three 50 × 50 × 50 mm cubic samples for each series. The maximum increase in compressive strength and UPV values were obtained for 1 NCa (1% by mass of binder) specimens. When the NCa content was increased from 0.5 to 1 percent, the compressive strength of NCa-ECC increased steadily by 2.17 to 6.92 percent compared to the 0 NCa series. When the NCa content was increased to 1.5 percent, the enhancement of compressive strength decreased to 4.64 percent. The increased strength could be attributed to both the filling effect and the chemical effect associated with NCa. NCa can react with C3A to form mono-carbonate, which has a unique structure with strong hydrogen bonds between oxygen atoms and interlayer waters in carbonate groups [46]. In addition, CaCO3 can increase the stability and nature of ettringite [47]. The difficulty of uniform distribution may be the reason for the less apparent positive effect of NCa at higher dosage on compressive strength [39]. As a result, increasing the nanocalcite content in the mixes increased the compressive strength values. These increases were 50.17, 51.26, 53.64, and 52.50 MPa for 0, 0.5, 1, and 1.5 NCa series. When the content of Nano-CaCO3 increased from 1% to 1.5%, the improvement of strengths was reduced. Because excessive NCa addition led to poor dispersion of the matrix, insufficient hydration, and limited the improvement of the strength of the samples [48]. In addition, The matrix had an agglomeration effect due to the increase of nano-materials. In addition, free water cannot reach the cement particles, which reduces hydration and reduces the strength of the concrete [49].

**Figure 7.** Compressive strength results of NCa-ECC.

**Figure 8.** UPV results of NCa-ECC.

Figure 8 shows the ultrasonic pulse velocity (UPV) testing of NCa-ECC samples with different mixtures. Looking at the UPV results, it is observed that the result has a parallel relationship with the compressive strength results. The UPV increased with increasing nanocalcite content in the mixes by 3630.06, 3700, 3755, and 3630 m/sec for 0, 0.5, 1, and 1.5 NCa series. Adesina and Das [50] investigated the UPV values of 50 × 50 × 50 mm cubic ECC samples by replacing crumb rubber with silica sand. They found that the UPV value of ECC samples without adding crumb rubber was 3689 m/s, while the UPV value decreased to 2976 m/s when they used 100% crumb rubber. The UPV result of ECC without the addition of crumb rubber obtained in their study was consistent with the UPV result of the ECC without the addition of NCa obtained in this study. In addition, the correlation between compressive strength and UPV for all the mixes is shown in Figure 9. Figure 9 shows that the R coefficient is 0.9264. This R coefficient indicates a strong relationship between UPV and the compressive strength of the specimens.

**Figure 9.** Correlation between compressive strength and UPV.
