*2.2. Experimental Procedures*

Compaction tests were performed for the soil and the mixture with PET according to [37,38]. They related the moisture content to be used with the specific weight of the soil samples and the mixtures with PET. To perform the compaction test, a cylindrical mold (Normal Proctor) was used, with a base attached to a metal socket which has a drop control apparatus [40,41].

The molding process of the bricks was performed according to [35] at the company Arte Cerâmica Sardinha, located in Campos dos Goytacazes. A hydraulic press, Model 7000 Turbo II, of the manufacturer Ecomáquinas, Navegantes, Santa Catarina, Brazil, was used, which has a molding capacity of up to 2 bricks at a time and a compression pressure of approximately 15 tons. The bricks were made in the size of 30 × 15 cm (length × width) and variable height of 7 cm ± 1 cm, with 2 holes and a useful area of 80%.

After making the bricks, the curing process was carried out by water sprinkling for 28 days. The process occurred in a humid chamber, where the specimens were placed on pallet racks covered with plastic sheeting, and during the 28 days constant humidity was added to the bricks through the water pump sprayer to help in the hydration process of the cement. The compressive strength was analyzed after 7 and 28 days of curing. For the statistical analysis, 7 samples of each treatment were used. The dimensions of the samples were measured with a calliper, according to the ABNT standards [40,41]. Then the bricks were cut in half (Figure 2a) in the transverse direction, having the upper part of the socket removed, as recommended by the ABNT standard [41]. The cuts were made with the aid of an electric saw (Figure 2b). Next, the cut halves were overlapped and joined by cement paste and then the capping was performed.

**Figure 2.** Brick capping step: (**a**) Bricks cut in half; (**b**) upper part of socket removed.

The cement used to make the capping paste was the same CPV-ARI used in the manufacture of the bricks. After capping, the bricks were submerged in water for 24 h to ensure their saturation before rupture, which was performed in the press Model 100 T Manual Hydraulic Press DIG.110/220, Solotest, São Paulo, Brazil.

The water absorption test of the bricks was performed according to the ABNT standard [42,43] at 28 days after manufacturing. Three bricks of each mixture were used, which were dried in a SOLAB oven, SL-100 model, with temperatures ranging between 105 ◦C and 110 ◦C until reaching constant mass, for 48 h and then stored. After this period, the weight of each of the bricks was measured with the aid of a digital balance brand Marte, Model AD5002, São Paulo, Brazil to obtain the dry mass (g), as recommended by the ABNT standard [43]. Then, the bricks were immersed in water for 24 h, and after this saturation step, they were dried superficially with the help of a clean and dry cloth and then weighed again to obtain the wet mass (g). The water absorption of the bricks was obtained, according to the ABNT standard [43]:

$$\mathbf{A}(\%) = \frac{(\mathbf{m1} - \mathbf{m2})}{\mathbf{m2}} \times 100 \tag{1}$$

where: A—is the water absorption (%); m1—is the dry mass of the specimen (g); m2—is the mass of the saturated specimen (g).

Analysis of Variance (ANOVA) was used to verify the existence of significant differences between the results obtained. Statistical differences were confirmed by means of the comparison of means test, using Tukey's method (*p* < 0.05). The experimental design used was the Completely Randomized Design for the two variables analyzed: water absorption and compressive strength. For water absorption, three specimens were used for each treatment (28 days of curing): 0%, 10%, 20% and 30%. For compressive strength, seven specimens were used for each of the four treatments (7 days and 28 days of curing).

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