The main aim of this work is the physical and chemical characterization of fine fraction recycled aggregate, and to study the feasibility of incorporating this material in the masonry mortar production. To that end, construction and demolition samples from two recycling plants situated in Madrid region have been used. Two of them were obtained from concrete recycling processes, and one from the ceramic recycling line.

The following materials were used in this research: cement, sand, additive and water.

Samples of fine recycled aggregates were characterized according to the technical requirements regulated by the standard UNE-EN-13139: “Aggregates for mortar”. Three types of fine recycled aggregates comply with the upper and lower size limits and the maximum percentage of fines established by the standard for masonry mortars.

Complying with the specification in the UNE-EN-13139, all particles were smaller than 8 mm in size (2D) for the three types of recycled aggregates; and particles greater than 5.6 mm in size (1, 4D) did not exceed 5% value determined as a limit. As for percentage of fines present in the samples of fine recycled aggregates passing through 0.063 mm sieve size, it did not exceed 8%. This percentage was determined as a limit value for masonry mortars. The larger retained particles of three types of recycled aggregates were saved for crushing and the finer sieved (<0.063 mm) material was discarded.

Fine recycled aggregates size distribution was determined according to the requirements established by the UNE-EN-933-2 and the UNE-EN-933-1 standards, in particular Part 1: Determination of particle size distribution “Sieving method”. Graphical representation of this test is shown in Figure 1.

As can be observed in Figure 1, the size distribution line is continuous for three types of recycled aggregates and does not exceed the limits established by the Basic Norm NBE-FL 90. Size distribution is one of the most important properties of aggregates, because it affects mortar compactness, workability and strength (2, 3). A continuous particle size distribution indicates that aggregates are uniformly graded in sizes, providing higher interaction between particles. Regarding physical characteristics of fine recycled aggregates, Table 2 summarizes the obtained values for: fine content, particle form, fineness modulus, bulk density, relative density and water absorption.

Three types of RA samples were also tested through X-Ray fluorescence method. Results of this test are shown in Table 3.

As can be observed, the fine fraction of three types of recycled aggregates is composed mainly of silicates, being the majority of them Calcium and Aluminium silicates. RA-1 presents high content of Al₂O₃, that can be explained by the higher quantities of clay and ceramics and the higher proportions of chert (11).

Obtained values of CaO and LOI fall between the range of values found in other researches (12). However, higher contents of these components were shown in other research that used aggregates originated from Italy (13).

Crystal structures and purity of the samples were tested by X-ray diffraction (XRD) analysis, where the X-ray diffraction pattern was measured using the Cu-Kα radiation (λ=1.540598 Å) with a Siemens D500 diffractometer equipped with a graphite monochromator. Data was collected at 300 K over an angular range of 10°≤2≤110° by scanning in steps of 0.05° and counting times of 10 s. The results were analyzed by the Rietveld profile refinement method, using the FullProf Suite Program 1.10 (Version Oct-2011). Figure 2 shows the diffractogram obtained for three samples used in this study. Table 4 summarizes different mineral phases. X-ray diffraction diagrams are similar for tested recycled aggregates and, with the exception of quartz, do not show a high crystalline state, although they have wide diffraction peaks with low intensities. This characteristic is typical of poorly crystalline and/or amorphous phases, due to low crystallinity of compounds formed as a result of concrete hydration and materials coming from ceramics, that have additionally undergone the grinding process. Principal phases are quartz and calcite, other phases such as Albite, Sanidine, Phlogopite and Gypsum can also be observed.

Each recycled aggregate sample was tested through TGA-DTA method; Figure 3 shows an example of this test. As can be observed in the graphic, the RA samples weight loss and derivative weight loss are presented at the same temperature regions. In order to eliminate humidity, the samples were initially maintained during 5 minutes at 100 °C. The first weight loss, -up to 110 °C-, is related to the samples humidity content, and above 160 °C–185 °C, decomposition of gypsum occurs. Above 570 °C a change of phase from αSiO₂ (alpha-quartz) to βSiO₂ (beta-quartz) takes place, characterized by the endothermic peak under this temperature without mass loss. Between 650 °C and 750 °C the decarbonation of calcite is observed. The maximum temperature value obtained from the weight loss derivative curve (720.18 °C), representing the highest speed decomposition of different compounds is marked in the Figure 3. On the basis of the data shown in the diagram, calculating quantities of gypsum and calcite becomes possible. The X-ray diagram confirms the presence of these compounds. The calculated quantities of gypsum and calcite are shown in Table 4.

Once the recycled aggregates were characterized, the study of proportions and the development of a test plan were carried out. In this study a unique proportion was used to prepare recycled mortars: 1:3:0.5 (cement, sand and water respectively), and preparation of all mixes was performed using the same technique and equipment. In this way, the goal was to clearly determine the influence of recycled aggregates on masonry mortars, using standardized sand as a reference. Recycled aggregates obtained from the treatment plants were sieved in the laboratory, eliminating material retained on the 4 mm sieve and material passed through 0.063 mm sieve. The amount of aggregate fractions reported in Table 5 refers to their dry condition. The percentages of fine recycled aggregates used were: 10%, 15%, 25%, 35% and 45% for three types of recycled aggregates.

Mix design was structured in two stages, the first one without the use of additives. The second one was performed using the Rheobuild 2100 additive, following the recommendations of the manufacturer (BASF Company) regarding proportions and usage to obtain recycled mortars with characteristics similar to the reference mortars. According to the technical data, this additive can reduce the water content of the mortar while maintaining workability and improving its strength, durability and shrinkage.

The test program was organized in two phases: during the first phase, the fine recycled aggregate samples were characterized and the tests of consistency, flexural and compressive strength were carried out. During the second phase, and once all previously obtained results were analyzed, the batches that presented better results in the first phase were tested on bonding strength and shrinkage.

The following code was employed for the mixtures identification:

Where RA=Recycled aggregates,

N°=Waste Treatment Plant. 1 and 2 for ceramic and concrete recycling line, the plant “Palentina” and 3 for concrete recycling line, the plant “Molar”,

X%=Recycled aggregates substitution percentage,

Y%=Additive use percentage over the weight of cement.

According to the results shown in Table 2, obtained density values are lower than standardized sand, being similar to the values found in other researches (7, 8). Cement mortar attached to recycled aggregates causes poorer properties of RA, such as lower density and higher absorption. High absorption presented by fine recycled aggregates affects proportion of mortars, making the use of additives necessary in mortars manufactured with more than 10% substitution level. Fine recycled aggregate absorption is much higher compared to natural sand (7, 8, 14). This characteristic may have a negative influence on mechanical performance of recycled mortars, because of the holes that may have been produced inside the material after the water has evaporated. The smallest fractions present the highest absorption level, when obtained from the same original concrete (15). Among the samples of recycled aggregates tested in accordance with the UNE-EN 1097-6 standard, the one which presented the highest water absorption coefficient was ceramic recycled aggregate with the value of 7.48. This value is lower in comparison to the values obtained by other authors in their research works (8, 10). The water absorption coefficient for concrete recycled aggregates RA 2 and RA3 were 6.35 and 6.12 respectively.

One of the main inconveniences of such type of material is its workability. The consistency test determines the amount of water needed to manufacture mortar with an adequate workability. As it is shown in Figure 4, incorporation of recycled aggregates produces mortar consistency to decrease, being the use of additives necessary for easier mixing of diverse compounds. With increasing percentages of incorporated recycled aggregate, the addition of the additive proportions was needed to obtain consistency values between 165 mm and 185 mm according to the UNE-EN 1015-3 standard. The quantity of incorporated additives varied between 2% and 4% in cement weight; in no case the limit values established by the manufacturer were exceeded.Another possibility to improve the workability is to adjust the amount of water, reducing the additive incorporation rate. Consistency loss obtained incorporating 45% of recycled aggregate with respect to the reference batches are: 11%, 11.5% and 7% for aggregates RA-1, RA-2 and RA-3 respectively. However other studies showed better consistency incorporating up to 15% of fine concrete and 10% of fine ceramic aggregate (16, 17).

Prismatic specimens (40×40×160 mm) were manufactured, cast and wet cured at 20 °C up to 28 days. The compression and flexural strength were evaluated according to the UNE-EN 1015-11, and were tested after the curing period.

In relation to mechanical strengths, the compression strength results (see Figure 5) of three types of recycled mortars are lower compared to the reference samples. Decrease in strength for batches containing 45% of recycled aggregates are: 23.8%, 32.9% and 23.8% for aggregates RA-1, RA-2 and RA-3, respectively. These results comply with those obtained in another research (9), which proved that mortars containing recycled aggregates developed lower mechanical strength with respect to the reference mortars. However, the results obtained in recycled mortars containing RA-1 do not comply with those obtained in another studies (9, 16), which showed that replacement ratios of sand by brick waste of up to 40% increased the compressive strength of masonry mortars. Nevertheless, in accordance with existing standards, the minimum acceptable value for compression strength is 7.5 MPa. Therefore it is not necessary to modify the mortar proportions.

On the other hand, the flexural strength results (see Figure 6) show higher values for three types of recycled mortars compared to the reference samples. The mixtures prepared using RA-3 from “Molar” treatment plant stand out, reaching an average value greater than 6 MPa. Improvements obtained with 45% of recycled aggregate substitution rate are 9.03%, 9.97% and 30.34% for the aggregates RA-1, RA-2 and RA-3 respectively. There are no standards that define the minimum acceptable value for the flexural strength of mortars.

In the second testing phase, bond strength and shrinkage tests were carried out. The bond strength test was carried out on a ceramic support of 400×500×50 mm, with a specimen diameter of 50 mm and coating thicknesses of 10 mm. The obtained results are shown in Table 6.

According to the results shown in Table 6, the three types of recycled mortars containing 45% of RA showed values exceeding the limit value of 0.30 N/mm² (established by the UNE-EN 998-1). Bond strength values obtained in recycled mortar produced using RA-3 gave the best results (0.56 N/mm²) that is, higher than reference values. The mixtures elaborated with 45% of recycled ceramic aggregate presented the poorest values. This was probably because of the worse grading size distribution of this type of aggregate. Another research studies obtained values close to that of the reference using ceramic waste containing 40% of RA (9) and higher values containing 50% of RA (16).

Shrinkage tests show higher values for the three types of recycled aggregates studied. This test measures the changes in length of the mortar during the first 24 hours, taking readings every 30 minutes. Table 6 shows the obtained results:

As observed in Table 6, increase in percentage of recycled aggregates incorporated in mortars produces higher shrinkage values due to the greater porosity and water absorption presented by RA. These characteristics significantly affect water migration mechanisms. These conclusions are similar to the findings of other researches (18). The mix that presents the most unfavourable value is the RA-2-45%-4%, being this value practically double than that of the reference mortars, however, complying with the limits established by Spanish standards.