*2.3. Uniaxial Compressive Loading Test*

A uniaxial compressive loading test was performed by using a mechanical testing machine. Two displacement meters were fixed on the middle part of both sides of the specimen, and the displacement in the middle part of the specimen (170–180 mm) was measured in this test as shown in Figure 2. A load sensor was placed under the steel pads. The dynamic data acquisition system was used to collect the test data.

**Figure 2.** Test setup of loading and measurement.

The peak strain (*ε*0) is the strain corresponding to peak stress and the ultimate strain (*ε*u) is the strain corresponding to the stress which is 50% of peak stress at the descending part of the stress–strain curve.

The elastic modulus (*Ec*) was calculated with the following expression [33,34]:

$$E\_{\varepsilon} = \frac{\sigma\_{0.4} - \sigma\_0}{\varepsilon\_{0.4} - \varepsilon\_0} \, ^{\prime} \tag{2}$$

where *σ*<sup>0</sup> is peak stress and *σ*0.4 is the stress corresponding to 40% of the peak stress, *ε*<sup>0</sup> is the peak strain, and *ε*0.4 is the corresponding to *σ*0.4.

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

#### *3.1. Carbonation Depth*

Figure 3 shows carbonated specimens with 0%, 50%, and 100% RCA replacement ratios. The correlation between carbonation depths and carbonation periods is shown in Figure 4. It can be seen that the carbonation depth of specimens increased as the replacement ratio of RCA increased due to the increase of the porosity of RAC. This agreed with the findings observed by C. Thomas and Miguel Bravo [35,36].

**Figure 3.** Typical specimens after different carbonation periods. (**a**) NAC; (**b**) RAC50; (**c**) RAC100.

**Figure 4.** Carbonation depth versus carbonation periods.

Moreover, the RAC100 was fully carbonated at 35.0 mm of carbonation depth, whereas RAC50 and NAC were carbonated at 31.0 and 19.2 mm, respectively. Notably, the carbonation depth of RAC50 was 12% higher than that of NAC under a similar 28-day compressive strength. However, the carbonation resistance of RAC can be improved upon by adding other materials, such as fly ash [14], etc. Consequently, it is possible to apply RAC in field applications.

#### *3.2. Failure Pattern*

Failure patterns of RAC and NAC specimens with different carbonation depths are compressive failure modes, as shown in Figure 5. For non-carbonated RAC and NAC specimens, there were a few cracks that were parallel to the loading direction, and the surfaces were slightly spalling. There is no significant difference in failure patterns of uncarbonated RAC and NAC specimens, as shown in Figure 5a–c. In contrast, for highly carbonated RAC50 and RAC100 specimens, as shown in Figure 5e,f, lots of short cracks were parallel to the loading direction, but these short cracks formed one or several long cracks. Finally, there were diagonal cracks that passed through the whole specimens, as shown in Figure 6. However, carbonated NAC specimens had the same failure pattern with uncarbonated NAC, and the spalling happened in carbonated NAC, as shown in Figure 5d. Therefore, the effect of carbonation on the failure pattern of RAC is more obvious than that of NAC. This is explained as follows: for carbonated RAC, carbonation did not change the pore and microcrack in RCA, but there were obvious micro-cracks on the interface between RCA and the new cement matrix, and the structure was looser in the interface transition zone (ITZ). Yang et al. [37] found the above phenomenon by comparing scanning electron micro-graphs of carbonated RAC with that of carbonated NAC. Consequently, carbonation mainly changes the microstructure of the paste and interface transition zone. Short cracks occurred in RAC with the increase of loading, but the more pore and microcracks in RCA prevented the development of short cracks. Therefore, lots of short cracks occurred and formed one or several long diagonal cracks, resulting in the diagonal band failure of specimens.

**Figure 5.** The failure pattern of RAC and NAC specimens. (**a**) NAC-0.0 mm; (**b**) RAC50-0.0 mm; (**c**) RAC100-0.0 mm; (**d**) NAC-19.8 mm; (**e**) RAC50-31.0 mm; and (**f**) RAC100-35.0 mm.

**Figure 6.** The development of the failure of carbonated RAC specimens.

In Figure 5f, the failure mode of RAC100-35.0 mm, which was fully carbonated, is similar to the findings in the literature [38]. Although the interior of the fully carbonated concrete has a high degree of compactness, a large number of original cracks inside RCA was responsible for the failure of carbonated RAC.

#### *3.3. Uniaxial Compressive Stress–Strain Curves*

The uniaxial compressive stress–strain curves of the specimens with different carbonation depths are shown in Figure 7. It is shown that the stress–strain curves of the specimens exhibit similar features, whereas the replacement rate of RCA and carbonation depth showed an obvious impact on the stress–strain relation of RAC.

**Figure 7.** Uniaxial stress–strain curves of the specimens with different carbonation depth. (**a**) NAC; (**b**) RAC50; and (**c**) RAC100.

The peak stress and slope of the ascending part of the curve declined with the increase of the replacement rate of RCA, which conformed to findings from other researchers [29,38]. Furthermore, as carbonation depths increased, peak stress increased and both the ascending and descending part of the stress–strain curves became steeper. During a uniaxial compressive loading experiment, the highly carbonated specimens were more brittle.

#### 3.3.1. Peak Stress

Figure 8 shows the relationship between peak stress and carbonation depth. The results show that the peak stress of NAC, RAC50, and RAC100 specimens increased with the increase of carbonation depth. This is caused by the reactions between CO2 and hydration products of cement, such as Ca(OH)2 and C-S-H, which can reduce the porosity of concrete [33,34,37–39]. The peak stress of the concrete was increased by reducing the porosity [40–42].

**Figure 8.** Peak stress versus carbonation depth.

When the carbonation depth reached about 20 mm, the peak stress of NAC, RAC50, and RAC100 specimens increased by about 30.5%, 17.6%, and 41.6%, respectively, compared with uncarbonated specimens. The result shows that the peak stress of NAC increased much more than that of RAC50 with the same carbonation depths. Moreover, the peak stress of fully carbonated RAC100 increased by about 62.8%; hence carbonation can increase the strength of RAC, which is consistent with findings by other researchers [24–26]. In Figure 8, it is found that the standard deviation values of the peak stress of RAC100 is higher than that of RAC50 and NAC, the peak stress has a clear jump from 16.8 to 21.5 mm of carbonation of RAC100. The reason may be due to the discreteness of RAC. Further experiments need to be performed in order to investigate the phenomenon.

#### 3.3.2. Strain

The peak strain and ultimate strain of the specimens with different carbonation depths are shown in Figure 9. It shows that carbonation depth has no significant effect on the peak strain of all specimens. However, the ultimate strain of NAC and RAC100 decreased with the increase of carbonation depth.

**Figure 9.** Peak strain and ultimate strain versus carbonation depth.

The ratio of the ultimate strain to peak strain (*ε*u/*ε*0) describes the trend of the descending part of stress–strain curves [43]. Figure 10 shows the relationship between the ratio of ultimate strain to peak strain and carbonation depth. The results show that the ratio of ultimate strain to peak strain of RAC50 is roughly larger than that of NAC. This means

the descending part of the stress–strain curve of RAC50 is flatter than that of NAC, and the results match with the descending part of stress–strain curves. Additionally, the ratios of ultimate strain to peak strain of NAC, RAC50, and RAC100 specimens decreased with the increase of carbonation depth. This means the descending part of the stress–strain curves of these specimens becomes steeper and steeper due to the carbonation. This is consistent with the rapid destruction of specimens after carbonation.

**Figure 10.** Ratio of ultimate strain to peak strain versus carbonation depth.
