**3. Materials and Methods**

#### *3.1. Materials*

Portland cement CEM I 42.5 R, meeting the requirements of the EN 197-1:2011 [28] standard, was used for manufacturing cement composites. The natural aggregate sand with a maximum size of 2 mm was used. The procedure of obtaining RCM was described in Section 2.2 and in [23]. For further studies, only the fraction <250 μm of RCM was used as 25% cement replacement. RCM sieve analysis was performed in accordance with EN 933-1:2012 [29], and the grading curve of RCM is presented in Figure 3. As shown in Figure 3, after the recycling process, approximately 20% of the fraction classified as fine aggregate (<4000 μm) according to EN 206:2016 [30] is the dust fraction (<63 μm). For further studies, the fraction <250 μm was used, which was about 50% of the whole material. In future applications, an additional grinding of RCM is recommended immediately after the recycling process to increase the proportion of the finest particles.

**Figure 3.** Grading curve of RCM after recycling process.

#### *3.2. The Cement Composites with RCM*

The mix compositions of composites with RCM used as a partial cement substitute are shown in Table 2. The compositions were designed as standard cement mortar according to EN 196-1:2016 [31]. In the series 1–10, 25% of the mass of Portland cement was replaced by RCM after thermal and mechanical treatment (RCM calcination temperature and time of treatment were based on the assumptions of the experimental plan, which is described in detail in Section 4). Series 12 had an analogous composition, but the RCM used was not subjected to calcination. Series 11 was made as a standard cement mortar: it did not contain RCM.

**Table 2.** The composition of cement composites.


#### *3.3. Methods*

#### 3.3.1. Physical and Mechanical Properties Test Methods

Specimens of composites (40 × 40 × 160 mm) were prepared in accordance with EN 196-1:2016 [31]. After 28 days of curing, the compressive strength and flexural strength tests were carried out [31]. The water absorbability test was performed by determining the percentage increase in the weight of the specimens saturated with water in relation to the weight of the specimen in the dry state. The consistency measurement of fresh cement composite mixtures was made by the flow table method according to EN 1015-3:1999 [32]. The strength activity index (SAI) of the RCM was determined according to EN 450-1:2012 [33]. The heat of hydration of RCM was tested using a semi-adiabatic method based on the standard EN 196-9:2010 [34].

#### 3.3.2. Analysis of RCM Properties and Microstructure

In order to determine the effect of the thermal and mechanical treatment on the phase composition of the RCM, X-ray diffraction analysis was conducted using a D8 Discover A25 instrument (Bruker) with CuKα radiation. All diffraction patterns were obtained by scanning the goniometer from 10 to 70 (2θ) at the rate of 0.05 min<sup>−</sup>1.

The differential thermal analysis and thermogravimetric analysis were carried out using a model STA 409 PG analyzer (Netzsch, Selb, Germany) under a nitrogen atmosphere. The specimens were heated at rate of 10 ◦C/min to the temperature 1100 ◦C. The content of RCM components was calculated using DTA/TG DTG curves based on the instructions [35,36].

The morphology of RCM and cement composites with RCM was investigated using a Tescan high-resolution scanning microscope (Aztek Automated, Oxford Instruments, UK) equipped with an X-ray microanalysis system based on the method of X-ray spectrometry with energy dissipation (EDS) and a high-resolution microscope (Quanta 250 FEG, FEI, ThermoFisher Scientific, USA), digitally controlled and equipped with an electron gun with thermal field emission (the Schottky emitter). The shapes of fine particles were classified according to EN ISO 3252:2002 [37].

### **4. Design of the Experiment**

*Selection of Variables and Development of the Experimental Plan*

For better understanding the relations among the factors determining the characteristics of RCM as a partial substitute for cement, an experiment was performed based on two variables: *X*1—temperature of concrete rubble calcination, *X*2—time of thermal treatment. The range of variation and the levels of analyzed factors are shown in detail in Table 3.

**Table 3.** Variables in the plan of experiment.


The calcination temperatures at which the effects of phase changes can be expected were selected. At temperatures up to 350 ◦C, dehydration of C–S–H silicates, hydrated aluminates and aluminum calcium sulphates occurs along with gypsum decomposition. However, up to 650 ◦C, portlandite breaks down into CaO and H2O. The temperature 500 ◦C is the center of the plan, and the other extreme like 288 ◦C and 712 ◦C are the star points and result from the construction of the adopted rotational plan.

Statistical analysis was carried out in accordance with the rotatable central composite design with a double repetition of the experiment at a central point. The design of the experiment (DoE) enables to check repeatability of results, to find which input factors and their interactions can influence the output properties significantly, to calculate regression equation and to check its adequacy with the test results. The following output properties were selected for analysis: compressive strength, flexural strength and water absorbability of composites with RCM.

On the basis of the above-mentioned variables, the experimental plan including 10 test series and 2 additional control series was established. Table 4 shows the detailed experimental plan with the real and normalized values of the variables.


**Table 4.** The rotatable central composite design of experiment.

Apart from the series described in Table 4, additional series were also tested for comparison 11 and 12. For the first control series 11, the cement composite included only cement as a binder, while the other control series 12 was made with RCM without thermal treatment.

Combinations of real values of the examined factors *X*<sup>1</sup> and *X*<sup>2</sup> were established on the basis of the assumptions of the design of experiment [38]. The dimensionless normalized values—*x*<sup>1</sup> and *x*2—related to them were used to develop the functions describing the influence of the analyzed factors on the resulting quantities.

The test results were statistically analyzed in order to determine an approximating function describing the influence of the tested variables on the selected properties of the composites with RCM. The analyses included analysis of variance, calculation of regression coefficients and assessment of the regression coefficients' significance. The function describing the changes in the physical and mechanical properties of cement composites adopted the form of a second-degree polynomial (1):

$$y = b\_0 + b\_1 \mathbf{x}\_1 + b\_2 \mathbf{x}\_2 + b\_3 \mathbf{x}\_1 \mathbf{x}\_2 + b\_4 \mathbf{x}\_1^2 + b\_5 \mathbf{x}\_2^2 \tag{1}$$

where: *y*—dependent variable, explained; *x*1, *x*2—independent variables; *bi*—coefficients; *b*0—free term in expression. Calculations were performed according to [38] using software Statistica Version 13 (StatSoft, Poland).

### **5. Test Results and Discussion**

#### *5.1. Characteristics of Recycled Cement Mortar (RCM)*

#### 5.1.1. Heat of Hydration

In order to assess the rehydration reactivity of RCM, calorimetric tests were performed, and the results are presented in Figure 4. Two types of RCM were analyzed: calcined at the temperature 650 ◦C for 60 min and non-calcined material (without thermal treatment). CEM I 42.5 R was used as a standard for comparison. The specific density of RCM equal to 2.66 g/cm<sup>3</sup> was slightly lower than cement CEM I 42.5 R density, which was 3.05 g/cm3.

**Figure 4.** Changes in the amount of heat accumulated of tested materials.

Based on the analysis of data in Figure 4, it can be concluded that the thermal treatment of RCM has a significant impact on its rehydration reactivity properties and applicability as an active supplementary material. Non-calcined RCM showed no change in the value of heat released over a 48-h measurement period, and the maximum recorded value of heat accumulated was 40 J/g. The RCM calcined at the temperature 650 ◦C showed the best rehydration reactivity, the amount of accumulated heat increased successively during the test, and after a 48-h measurement period it reached the level of 125 J/g. In relation to the value of accumulated heat obtained for CEM I 42.5 R, it was a decrease of 35%, while compared to non-calcined RCM, an increase of almost 70% was observed.
