*1.2. Sources of Recycled Aggregate*

There have been studies on recycled aggregate in concrete for some time. Reliable studies mentioned the properties of concrete and the behavior of the construction from recycled aggregate concrete. They found that recycled aggregate concrete plays a significant role in water absorption in aggregate, leading to lowered permeability resistance [17]. The water absorption capacity of recycled aggregate concrete is considered high (>7 percent) according to the standardized measure provided by [18] JIS A5002. Moreover, recycled aggregate concrete has a lower compressive strength compared to natural aggregate concrete by 21 percent [19]. However, if the amount of recycled aggregate in the concrete is restricted to not more than 20–30 percent of the total concrete, there is a negligible difference between the properties of recycled aggregate concrete and those of natural aggregate concrete. [20] In 2014, [21] Duan and Poon conducted research on the properties of concrete with mortar on the stone surface and from different sources. The result was that recycled aggregate with a little amount of mortar on the stone surface had low water absorption capacity and therefore can be a substitute for natural aggregate. Likewise, the research by [22] Pedro et al., 2014 also found that the water absorption capacity of recycled aggregate was medium as represented by 3.9–7.6 percent. As a result, a recycled aggregate could be used in concrete with high compressive strength.

Still, the aforementioned research did not analyze the cause of different behaviors of three different types of aggregate concrete. Also, it is worth mentioning that the research did not take the recycled aggregate which came from different sources into account. More importantly, the area of ITZ which significantly affects the properties of the concrete was not considered.

#### **2. Materials and Methods**

The recycled aggregate made from 3 sources was investigated (Figure 1). The first was the leftover cube concrete specimens from laboratory testing, with an existing strength ranging from 40–45 MPa. The second was the waste from the rejected product from the precast hollow core slab industry with an average strength of 35 Mpa. The last one was the building demolition waste, especially the building demolition concrete from the building of which the location interposes the new underground MRT Orange line in the capital city of Thailand. This example emphasizes the increasing demand for infrastructure renovation from the fast-growing city. The coring sample of concrete structure from the building was tested to verify its existing compressive strength which turned out to be 16 Mpa on average. B-RA, P-RA, and L-RA represent recycled aggregate from building demolition, aggregate from precast concrete waste, and aggregate from laboratory waste respectively. The concrete waste was crushed and graded according to ACI 555 [23]. Then, the natural aggregate concrete was replaced by L-RCA, P-RCA, B-RCA with the proportion of 30 percent, 60 percent, and 100 percent respectively as indicated in Table 1. Short-Term properties of concrete from recycled aggregate were verified in terms of strength and elastic modulus in the period of 7, 14, and 28 days. The durability of the concrete was represented by a rapid chloride penetration test (ASTM C 1202) [24] at 28 day-age concrete.

**Figure 1.** Recycled aggregate from various sources.

**Table 1.** Mixed proportions of concrete mixture.


The standard controlled concrete mix ratio for comparison with recycled aggregate concrete was Natural Concrete Aggregate (NCA). The mixing ratio was calculated to obtain the required tensile strength of approximately 25 Mpa. Then in the experiment, different proportions of the aggregates from natural stone were substituted. By using aggregate from the recycling as mentioned above. Table 1 shows the proportion of concrete mix used in the sample for this test.

#### *2.1. Aggregate*

The aggregate used in the concrete mixture was from three sources. From Figure 2, the particles of L-RCA and B-RCA were larger than that of P-RCA due to the smaller size of natural crushed stone used in precast concrete. Another point was that L-RCA possessed less cement paste on the aggregate's surface because of its higher parent strength compared to that of B-RA. The shape of L-RA aggregates is angular, similar to those obtained from natural stone. The B-RA is also angular, but from a comparative physical appearance, it appears to have more porosity from old-adhered mortar. The final aggregates that make up the fraction of Precast Hollow Core Slab are round but small and very porous as well.

According to the test, the distribution of different particle sizes of the coarse aggregate corresponded to ASTM C135 [25] Standard Test Method for Sieve Analysis of Fine and Coarse Aggregates as shown in Figure 3. However, it could be observed that the distribution of precast concrete waste of different sizes with mostly fine aggregate was higher than that of other types of aggregate.

**Figure 2.** Recycled aggregate from laboratory waste (L-RA), building demolition waste (B-RA) and precast concrete waste (P-RA).


**Figure 3.** Aggregate particle size distribution.

#### *2.2. Unit Weight and Absorption Rate of Aggregate*

The unit weight of natural aggregate is a little bit higher than that of recycled aggregate, as represented by 14–19%. From Figure 4, the unit weight of natural aggregate was higher

than that of every type of recycled aggregate. This was because there was mortar waste on the surface and the crushing process might leave the shape of the surface distorted. Moreover, the particle size of recycled aggregate was mostly at the same size range, resulting in more rooms in concrete and eventually in lower unit weight. Considering the water absorption rate, it was obvious that the unit weight of natural aggregate was lower than that of recycled aggregate. From Figure 5, the water absorption percentage of recycled aggregate was 6.65, 8.72, and 6.78 times higher than that of natural aggregate. This was because there was mortar present on the outer surface of recycled aggregate, resulting in higher porosity, lower specific gravity, and higher water absorption when compared to natural aggregate.

**Figure 4.** The unit weight of aggregate.

**Figure 5.** Percentage of Water Absorption for each type of Aggregates.

### *2.3. Rapid Chloride Penetration*

The chloride penetration concrete test according to ASTM C 1202 was conducted using a rapid chloride permeability test after a 28-day period of curing. From Figure 6, the test consisted of a clamp tab, a concrete cube containing a 3 percent concentration of sodium chloride (NaCl) solution on the cathode side and a 0.3M concentration of sodium

hydroxide (NaOH) on the anode side. Then, it is connected to a 60-volt DC power supply and data logger.

**Figure 6.** Test setup of the Rapid Chloride Penetration Test.

According to the experiment, although the chloride ion permeability, which is an indicator of concrete durability, could not be measured, the total amount of chloride ion permeating through the concrete was measured. Still, the result might not be easily applicable in real-life situations. However, ASTM C1202 presented a table to determine the concrete quality with different chloride ion permeabilities as shown in Table 2.


**Table 2.** Chloride permeability based on the total charge passed.

#### *2.4. Image Processing*

Figure 7 shown vertical cross section of concrete sample so that the properties of the concrete could be carefully examined. Then, the concrete was captured with a digital camera with at least 5 MB resolution. The image was then analyzed by the MATLAB program, in which the process involved converting the image to greyscale so that the color intensity of each element in the concrete could be clearly seen. After this process, the image was denoised. Subsequently, the area of ITZ of each element was processed. The result was the differences in color intensity, with the void being the highest, followed by mortar and aggregate respectively. In order to analyze the area of ITZ, the perimeter needed to be identified. Then, the area of ITZ could be determined by multiplying the perimeter by an average ITZ thickness.

**Figure 7.** Color pixel population from image processing.
