Residual mechanical properties of recycled aggregate concrete at elevated temperatures

This research examines the residual mechanical properties of normal and recycled aggregate concrete when subjected to elevated temperatures. The concrete specimens containing recycled aggregate (0%, 50%, 75%, and 100%) were exposed to different temperatures (25, 200, 400, and 600°C) in a muffle furnace at a heating rate of 10°C/min. The variations in flexural strength, compressive strength, and density were then tested according to ASTM standards. Findings from this investigation indicate that the degradation in the mechanical strength of concrete does not seem to be significantly affected by the increase in the percentage of recycled aggregates. However, a significant and linear decrease in the density was observed at 400°C with an increase in the percentage of recycled aggregates. The degradation of the compressive and flexural strengths of recycled aggregate concrete with increasing temperatures obtained from the experimental analysis was compared with the analytical predictions provided by Eurocode 2. Moreover, simplified equations have been proposed to estimate the degradation of the mechanical properties of recycled aggregate concrete at higher temperatures. The incorporation of recycled aggregates into concrete resulted in satisfactory residual performance.

great potential for promoting sustainable development.The conversion of demolished concrete into recycled aggregates (RA) can aid in solving the solid disposal issue, preserving a continuous supply of natural aggregates for construction without depletion, and reducing the environmental impacts of the production and transportation of aggregates. 8,9cycled aggregate concrete (RAC) is lighter and more porous than natural aggregate concrete, owing to the presence of old mortar adhered to its surface.2][13] The rougher surface and irregular shape of RA can lead to reduced workability. 14Statistical analysis by Silva et al. identified that carbonation in RAC can be 2.5 times that of normal aggregate concrete, 15 while Guo et al. explained that the total chloride ions (charges passed) could be more than double when 100% of coarse RA were substituted for virgin aggregates. 16cording to the US Environmental Protection Agency, construction, and demolition waste (CDW) is defined as the waste that is usually generated during new construction activities, renovation, or demolition of existing construction of civil engineering structures such as streets, highways, bridges, buildings, utility plants, piers, and dams. 17Construction and demolition materials for these projects often contain bulky, heavy materials such as concrete, wood, asphalt, gypsum, metals, bricks, glass, plastics, salvaged building components, trees, stumps, earth, and rock from clearing sites. 18The largest CDW producer is China (2360 million tons), followed by the United States (600 million tons) and India (530 million tons); only approximately 35% of this waste is recycled, with the remainder being landfilled. 19e recent increase in fire incidents has generated the need to study the behavior of structural materials and components under fire conditions.1][22][23] If RA are to be used for construction purposes, extensive research must be performed to determine their mechanical properties and durability.[26][27][28][29][30] Zhao et al. 31 tested the thermal properties of RA concrete at elevated temperatures.They concluded that the thermal conductivity of concrete decreases as the percentage of RA increases.Gales et al. 26 found that the mechanical properties degradation of the RA concrete was comparable to the conventional concrete.They also stated that the reduction in the mechanical properties did not exceed the conventional Eurocode design guidelines.Sarhat and Sherwood 27 also indicated that both the partial and full replacement of natural aggregates with RA exhibited good performance at elevated temperatures.
Yang and Hou 32 pointed out that the compressive strength of concrete made with natural aggregates is higher than that of RAC upon exposure to high temperatures.Similarly, Xiao and Zhang found that RAC has a higher compressive strength at elevated temperatures when compared to natural aggregate concrete, with RA showing higher beneficial effects at higher replacement percentages. 33According to Zega et al., RAC and natural aggregate concrete exhibit similar mechanical properties at elevated temperatures. 34Baradaran-Nasiri and Nematzadeh investigated the influence of different temperatures on the strength, density, and formation of cracks in concrete containing 0%, 15%, and 30% RAC.The mechanical properties of concrete at 28 days initially decreased when the temperature was increased from 100 to 200 C.A gradual increase was noted from 200 to 450 C, whereas it suddenly dropped again from 450 to 600 C. The density did not reduce by more than 5% when exposed to elevated temperatures. 35milarly, Kou et al. found that concrete containing natural aggregates suffered greater deterioration in mechanical properties compared to RAC at elevated temperatures.With the addition of cementitious mineral admixtures, such as silica fume and GGBS, the concrete suffered less deterioration owing to the reduction in porosity.The residual tensile strength of concrete made with RAC is higher than that of concrete composed of natural aggregates. 32Based on this, it appears that the RA had either a negative or positive effect, and in some cases, had a minimal effect on the properties of concrete upon exposure to elevated temperatures.These variations are most likely a result of the different compositions of RA, age of aggregates, shape, and so on.
In this study, the residual mechanical properties of normal-weight concrete and RAC with different RA percentages of 0%, 50%, 75%, and 100% were investigated at temperatures of 25, 200, 400, and 600 C. Since the findings of the previous studies were conflicting, and the performance of the RA concrete under fire exposure is not well defined.The literature still lacks data regarding the extent of degradation of the mechanical properties of RAC with increasing temperature.Moreover, the code provisions for fire protection do not include this type of concrete.Therefore, this paper proposes simplified equations for estimating the compressive strength, flexural tensile strength, and modulus of elasticity when exposed to high temperatures.In addition, the ratios of the reduced mechanical properties at higher temperatures to the mechanical properties at the initial temperature were compared with the ratios proposed by Eurocode 2. 36

| MATERIALS AND METHODS
This section describes the composition of the concrete used in the test along with the percentage of each component.The experimental program, including the test matrix, test standards, and parameters tested, is summarized in this section.

| Materials
Ordinary Portland cement (OPC) Type-1 conforming to ASTM C150 was used in all concrete mixes. 37The OPC was comprised of 79% CaO, 13.8% Al 2 O 3 , and 5.8% SiO 2 .Natural river sand (with a water absorption of 0.8, specific gravity of 2.65, and fineness modulus of 1.83) was used as the fine aggregate.Two different coarse aggregates were used: natural coarse aggregates from a local quarry and recycled coarse aggregates (obtained from an old building demolition site).The concrete mixtures used both fine and coarse aggregates in the saturated surface dry condition.

| Experimental program
The mix designs were created according to ACI 211, with the test matrix provided in Table 1. 38The physical properties of the aggregates are given in Table 2.A total of 32 concrete cylinders and 24 concrete prisms were cast with RA percentages of 0%, 50%, 75%, and 100%.The compressive and flexural strengths were determined using standard cylinders (200 mm Â 100 mm) and prisms (500 mm Â 100 mm Â 100 mm), as per ASTM C-39 39 and ASTM C-109, 40 respectively.The specimens were demolded after 24 h and water cured for 28 days, at an average temperature of 23 C, according to ASTM C192, 41 then left to dry for 8 days in the lab environment.For the compression tests, the specimens were ground to smoothen the cylinder's surface and to meet the end tolerance as per ASTM C-617. 42The parameters varied in this study were the temperature ( C) and quantity of RA (0%, 50%, 75%, and 100%).The ensuing degradation of the compressive strength (cylinder), flexural strength (prism), density, mass loss, elastic modulus, and development of stressstrain curves were analyzed, and the results are presented.
The air-dried specimens were subjected to elevated temperatures in an electric oven (600 mm Â 600 mm Â 700 mm) to examine their properties at temperatures of 25, 200, 400, and 600 C. The rate of heating and hold time followed RILEM TC 129-MHT guidelines. 43The target temperatures were reached by heating the specimens at a rate of 10 C/min, and heating the specimens for 60 min to fully expose them to the targeted temperature.This ensured that a steady-state condition was reached at the center of the specimens.After the steady exposure period of 60 min, the oven was switched off, and the specimens were allowed to cool down naturally for 24 h. 44e flexural strength (modulus of rupture) was determined according to ASTM C78 guidelines on a Matent Flexural Testing Machine using the center-point loading method. 45The tested prism specimens were standard in size (500 mm Â 100 mm Â 100 mm), and the load was incrementally applied perpendicular to the face of the prism until failure of the tested specimens.The compressive strength test was performed according to the ASTM C39 standards using cylindrical specimens (200 mm Â 100 mm) on a SANS-MTS Compression Machine, 39 as shown in Figure 1.The concrete specimens were placed inside the compression testing machine, and a compressive load was gradually applied until the tested specimens failed.A color change from gray to yellowish-brown was observed, corresponding to an increase in temperature.During the heating process, there was no explosive spalling on any of the concrete specimens, even though it has been widely reported by other sources, particularly for high-strength concrete.The absence of spalling in this study may be due to the slow heating rate (at 10 C/min).According to the available literature, spalling occurs at any heating rate above 7 C/min; however, in this study, spalling was absent even at a heating rate of 10 C/min.The thermal expansion between RA and the surrounding cement paste may also be one of the reasons for the resistance to spalling.With the increase in RA percentage, the number of cracks also increased, raising concerns regarding rebar corrosion.Therefore, proper compaction and curing are recommended for concrete containing a higher percentage of RA.

| Compressive strength
The average compressive strength of the concrete cylinders with four different RA percentages (0%, 50%, 75%, and 100%) exposed to 25, 200, 400, and 600 C temperatures is given in Figure 4.The failure patterns of the concrete cylinders can be seen in Figure 5.In addition, Figure 6 illustrates the normalized compressive strength with respect to the different RA percentages used in the mixtures.From the graph, it is evident that the compressive strength decreases with both increasing temperatures and the percentage of RA.When the concrete specimen was exposed to 200 C temperature, the compressive strength was reduced by 18.3%, 11.8%, 17.2%, and 18.9% for RA percentages of 0%, 50%, 75%, and 100%, respectively.Where these percentage differences are between the heated specimens and the ones tested at initial temperatures.It can be noted that the specimens having a 50% RA experienced less degradation in compressive strength compared to the 0% RA.Also, a negligible difference is observed between specimens having 75% RA compared to the ones having 100% replacement.At 400 C, the compressive strength decreased by 35.8%, 37.3%, 45.0%, and 34.5% for RA percentages of 0%, 50%, 75%, and 100%, respectively.This also indicates a deviation in the results, which means that increasing the percentage of RA does not necessarily cause a more severe deterioration in the compressive strength.Similarly, for the specimens exposed to 600 C temperature, the compressive strength reduced by 62.7%, 66.0%, 79.2%, and 72.5% for RA percentages of 0%, 50%, 75%, and 100%, respectively, with variations of 62.7%-79.2%.Based on this information, the reduction in strength was much more significant when the specimens were exposed to 600 C when compared to that of 200 and 400 C temperatures.
An increase in temperature was found to cause a reduction in the compressive strength of the concrete, especially above 600 C.
This abrupt loss can be attributed to the decomposition of calcium hydroxide, which is known to occur above 500 C. 46 At high temperatures, there is a loss of moisture, leading to contractions within the mixture and weakening the bond between the aggregate and the cement paste.This may be due to the micro-cracking caused by the differential thermal strain during heating and cooling.The loss of moisture also leads to an increase in the porosity, thus leading to more reduction in the compressive strength of the concrete.Furthermore, for a constant temperature, the compressive strength decreases as the percentage of RA increases.From Figure 3, a gradual reduction in the compressive strength of the concrete can be observed, correlating to an increase in the temperature.For the 100% RA specimen, there was a sudden reduction in the compressive strength when exposed to 600 C, breaking the previous moderation of the trend.
F I G U R E 1 Compressive test apparatus.

| Flexural tensile strength
The results for the flexural tensile strength of concrete cylinders incorporating different RA percentages (0%, 50%, 75%, and 100%) exposed to 25, 200, 400, and 600 C temperatures are given in  F I G U R E 4 Compressive strength of concrete.
in flexural strength was 39.40%, significantly higher than the results found from the previous three.
Analyzing with respect to the RA percentages, with the inclusion of 0%-50% RA, there was a reduction of 33.33%, 36.09%, and 37.27% at 25, 200, and 400 C, respectively.It was evident that there was a significant decrease in flexural strength when the RA was substituted from 0% to 50%.For 0% to 75% RA, there was a 48.48%, 50.36%, and 48.76% reduction at 25, 200, and 400 C, respectively, and for 0%-100% RA, the reduction was 57.33%, 67.62%, and 66.41% for the same temperatures as above.
The highest variation was therefore noted between 0% and 100%.
As the temperature was increased from ambient to 400 C, there was a significant reduction in the flexural strength for the 100% recycled specimens when compared with other RA percentages, as shown in Figure 8.It can be seen that the strength reduction ratio for the 100% RA replacement is substantially less than other percentages.However, this deviation did not occur in the compressive strength results nor in the modulus of elasticity or density.Therefore, the higher reduction in strength could be attributed to the casting or test conditions.(A) Flexural strength vs percentage of recycled aggregates (B) Flexural strength vs temperature F I G U R E 8 Reduction in flexural strength versus temperature.

| Density of concrete
The density of the specimens was measured before and after being subjected to elevated temperatures in order to scrutinize the variations for each group of concrete specimens.It was generally observed that the percentage of loss increased along with the increase in exposure temperature.The structural integrity also degenerated, as verified by the emergence of cracking.The results for the relative reduction in density are summarized in Figure 9.When the cylinders were exposed to 200 C temperature, density was reduced by 0.81%, 1.08%, 1.87%, and 2.68% for 0%, 50%, 75%, and 100% RA, respectively.The reduction was 3.67%, 5.40%, 7.18%, and 9.38% at 400 C, and 4.77%, 7.12%, 9.38%, and 11.30% at 600 C. The obtained results show that there is a gradual reduction in the density of concrete as the temperature increases.However, a significant reduction was noted when the concrete specimens were exposed to 400 C temperature.In control normal aggregate concrete (0% RA), there was a 0.81%, 3.67%, and 4.77% reduction in density when the specimens were exposed to 200, 400, and 600 C, respectively.
A linear reduction in concrete density can be observed as the percentage of RA increases, unlike the pattern with temperature variation.
The conclusion was formed that as the recycled percentage and temperature exposure increase, there is a reduction in the density of the concrete.This may be attributed to the mass loss in the concrete as a result of water evaporation.Concrete contains capillary and chemically bound water; most of the capillary and physically absorbed water evaporates when the temperature reaches 200 C or above.In this investigation, it was observed that most of the water loss occurred when the specimens were exposed to 400 C temperature.The loss was more significant at 600 C, which can be attributed to the withdrawal of all chemical bonds and the transposition of aggregate structure.

| Stress-strain curves
The Stress-strain response curves obtained from compression tests on cylinders are presented in Figure 10.The figure shows the stressstrain curves for 0%, 50%, 75%, and 100% RA when exposed to 25, 200, 400, and 600 C temperatures.Ductile behavior is clearly visible in Figure 9 as the temperature increases towards the maximum (A) Density vs percentage of recycled aggregates (B) Density vs temperature 0.81% 1.08% values tested.In addition, as the percentage of RA increases at a constant temperature, the ductile failure mode appears more clearly.The strain values corresponding to the maximum stress at temperatures 25, 200, and 400 C are very close.In contrast, there is an escalation in the strain value that corresponds to the maximum stress at 600 C when compared to the other temperatures.This can also be observed in the different aggregate percentages.For example, for 100% RAC, the strain value corresponding to the maximum stress at 25, 200, and 400 C is approximately 0.005, whereas it is 0.008 at 600 C.

| Elastic modulus
Elastic modulus is one of the most important properties of concrete in terms of design, as well as structural behavior; it is utilized to evaluate the structural deflection during the service period and calculate drift as well as deformation in the seismic analysis. 47mpressive stress-strain data has been employed to calculate the elastic modulus of the tested specimens.The influence of elevated temperatures on the average modulus of elastic modulus at 25, 200, 400, and 600 C is depicted in Figure 11.The normalized values for elastic modulus with respect to temperature are given in Figure 12.
As observed in Figures 10 and 11, the trend in the loss of elastic modulus (E) with increasing temperature is quite similar in both normal aggregate and RA concrete.The elastic modulus of the concrete decreases with the increase in temperature and with the increase in the percentage of RA.At the ambient (reference) temperature, the highest modulus of elasticity occurred with the 100% virgin aggregate specimens.The modulus of elasticity of the 0%, 50%, and 75% RAC decreased significantly more as the temperature increased when compared to the 100% RA specimen, which showed a much smaller reduction in the elastic modulus, as shown in Figures 10 and 11.The elastic modulus of the natural aggregate specimens was found to be slightly higher at room temperature when compared to the RA specimens.
However, the variance shrinks when the temperature is increased to 200 C.The pattern continues, and at the temperature range of 200-400 C, the variations in elastic modulus were negligible when the specimens were compared.At 400 to 600 C, the loss of modulus was found to be less in the 100% RAC specimens, which may be attributed to the lower deterioration in the microstructure.The higher loss in elastic modulus may be accredited to the chemical changes, dehydration, micro-cracking, physical deterioration, enhanced thermal cracking, and microstructure deterioration that arise from exposure to higher temperatures.

| ANALYTICAL PREDICTIONS OF THE MECHANICAL PROPERTIES
This section presents a comparison between the analytical predictions of the mechanical properties of concrete under elevated temperatures with the experimentally measured values.These predictions include flexural tensile strength and compressive strength.Simplified equations were proposed to describe the extent of degradation of these properties with rising temperatures.

| Compressive strength
The predictions of the compressive strength of concrete under elevated temperatures were computed according to EC 2. 36 Figure 13 illustrates that all the experimental values of the normalized com-

Safe zone
Unsafe zone strength of RAC exposed to high temperatures.Figure 14 shows the actual reduction ratios versus the predicted ones, and it is obvious that all the values lie below the unity line, which indicates that they are less than the analytical predictions.Using the experimental data in this study, Equation 1 was proposed to predict the reduction ratio in compressive strength of RAC concrete (R f'c ) under elevated temperatures.
where T is the ambient temperature in C.

| Flexural tensile strength
The flexural tensile strength was determined in accordance with Eurocode (EC) 2. 36 The predictions of the code were conservative for all strength reduction values, as shown in Figure 15, except for the strength of the 100% RAC, which showed a deviation from its counterparts at the same temperature.Moreover, a comparison of the ratios of the residual tensile strength to the initial strength obtained from the experimental results (R exp ) and from EC 2 (R predicted ) is plotted in where T is the ambient temperature in C.  • It is evident that there is a reduction in the density of concrete correlating to an increase in temperature.A significant reduction in the density and moisture content may be observed at 400 C, exhibiting a linear reduction in the density.
• The stress-strain relationship graphs show a direct relationship between ductility, RA percentage, and temperature.The modulus of elasticity of the concrete containing 0%, 50%, and 75% RA significantly reduced with temperature.
• The number of cracks increases at higher temperatures and increasing quantities of RA in the concrete.This may be of concern considering the corrosion of rebars.Therefore, proper compaction and curing are recommended for concrete containing higher percentages of RA.
• Eurocode predictions for the compressive strength of concrete under elevated temperatures were unconservative.Simplified equations were proposed to predict the reduction in RA concrete's compressive strength and tensile strength when exposed to high temperatures.
• Concrete with RA exhibited satisfactory residual performance.
However, this can be further confirmed with durability studies as a future research scope.

3 |
RESULTS AND DISCUSSION 3.1 | Visual observations of specimens after exposure to elevated temperatures Visual observations of both natural aggregate concrete and RAC after exposure to elevated temperatures helped identify concrete damage T A B L E 1 Test matrix of the recycled concrete specimens.
after exposure.The specimens were inspected immediately after their removal from the electric furnace.The evaluation of the specimens through visual inspection included checking for surface spalling, cracking, and color changes.Figure2shows the physical conditions of the concrete specimens after exposure to the different target temperatures.As shown in Figure1, temperatures of up to 200 C did not cause any noticeable effects on the concrete surface, except for changes in coloring.Cracks appeared once the temperatures were raised above 400 C and were significantly visible at 600 C (as shown in Figure3).

Figure 7 ,
Figure 7, and the normalized values are provided in Figure 8.The graphs show that the flexural strength decreases with the increase in both temperature and the percentage of RA.When the concrete specimen was exposed to 200 C temperature, the flexural strength was reduced by 11.49%, 15.16%, 14.72%, and 32.82% for RA percentages

1 0
Stress-strain curves of the test samples.
pressive strength (f'c residual /f'c initial ) are less than the expected values obtained from EC 2, including the natural aggregate concrete.It must be noted that Eurocode's provisions are applicable to traditional concrete; this comparison indicates that such provisions can be unconservative in predicting the degradation of the compressive

Figure 16 .
Figure 16.It can be seen from the figure that all the values except for one fell above the unity line; this means that the ratios of the reduction suggested by EC 2 are less than the actual experimental values.Therefore, the code provisions are conservative in predicting the reduction of flexural tensile strength under elevated temperatures.In addition, a simplified equation that predicts the ratio of reduction in the tensile strength (R ft ) was proposed.According to the data collected in this experimental investigation, R ft can be estimated as in Equation 2:

5 |
flexural strength.The highest percentage difference in the strength reduction factor between Equation 2 and the test results was 10.75%.Both EC 2 and Equation 2 slightly underpredicted the reduction in flexural strength of RAC when subjected to elevated temperatures.SUMMARY AND CONCLUSIONS This research examines the residual mechanical properties of normal and RA concrete with the goal of incorporating the effect of fire during design.The concrete specimens containing RA (0%, 50%, 75%, and 100%) were exposed to different temperatures (25, 200, 400, and 600 C) in a muffle furnace at a heating rate of 10 C/min.The variations in flexural strength, compressive strength, and density were then tested according to ASTM standards.Based on this experimental investigation, some conclusions can be drawn regarding the behavior of concrete with RA at elevated temperatures.• The mechanical properties of concrete decrease with increased temperatures.The percentage of RA did not highly affect the extent of degradation since the strength reduction percentages were close for specimens having different RA replacement percentages.The reduction in strength is much more significant when the specimens are exposed to 600 C compared to that of 200 C and 400 C, which may be attributed to the decomposition of calcium hydroxide and reduction in moisture content.

Table 3
Comparison between test results, EC2, and suggested equation for compressive strength.Comparison between test results, EC2, and suggested equation for flexural strength.