Experimental Investigation of the High Temperatures Effects on Self-Compacting Concrete Properties

Abstract

Self-compacting concrete (SCC), which appeared in the 1980s in Japan, is a concrete that differs from others by its high fluidity. The constituents of SCC can be quite different from those of ordinary concretes. They can differ both in their proportions and in their choice. Given the method of installation of SCCs, particular attention is paid to the study of their physical and mechanical characteristics. In this context, experimental tests were conducted to assess the effect of high temperatures on the behavior of SCC. For this purpose, a SCC and ordinary concrete (OC) were tested at temperatures of 20, 150, 300, 450, and 600 °C. Prismatic specimens of dimensions 100 × 100 × 400 mm³, cylindrical specimens of dimensions 160 × 320 mm, and parallelepiped specimens of dimensions 270 × 270 × 40 mm³ were prepared for physical (thermal conductivity) and mechanical (compressive strength, elastic modulus, flexural strength, and ultrasonic pulse velocity) tests. The results showed an increase in the compressive strength for SCC between 150 and 300 °C following an additional hydration of the cementitious matrix. The residual flexural strength of the concretes decreases progressively with the increase in temperature. This reduction is about 90% from 450 °C to 600 °C. The results also showed that the thermal conductivity of concrete decreases as the temperature increases and can reach a value of 1.2 W/mK for the heating temperature of 600 °C.

1. Introduction

A self-compacting concrete (SCC) is a very fluid, homogeneous, and stable concrete; implemented without vibration; and givesthe structure a quality that is at least equivalent to that corresponding to conventional concrete that is implemented by vibration [1]. SCCs are distinguished from OC mainly by their properties in the fresh state. The criteria characterizing an SCC are [2]:

• Slump flow test values must be between 60 cm and 75 cm;
• The rate of passage to the L-box must be greater than 0.8;
• The concrete must be stable under the effect of gravity (no segregation) and have a limited bleeding capacity. The absence of visual segregation during the Abrams test is not sufficient.

The constituents of SCCs can be quite different from those of OC. They can differ both in their proportions and in their choice. Given the mode of its implementation, the constituents used in the manufacture of SCCs, according to their use, are grouped into three categories: base materials (cement, aggregates, and mixing water), mineral additives, and chemical additives [3]. Several studies have been carried out to analyze the behavior of concrete at high temperatures [4,5,6,7].

Ahsan et al. [8] studied the effect of adding seashell powder (0, 10, 20, and 30%) in cementitious composites to improve their fire resistance. The prepared specimens were exposed to temperatures of 200 °C, 400 °C, 600 °C, and 800 °C. They noticed a decrease in cracks at the higher temperatures compared to the conventional concrete and an improvement of some characteristics such as compressive strength, elastic modulus, and compressive toughness.

In the experimental study that was carried out byAbolhasani et al. [9], the microstructural, mechanical, and fracture features of calcium aluminate cement concrete (CACC) were assessed under the effect of exposure to high temperatures. These authors noted that the concrete became more ductile and the number of pores in the concrete structure increased. The residual fracture toughness and flexural strength decreased as the temperatures increased.

Abed et al. [10] carried out a laboratory study toinvestigate the effect of elevated temperatures on both the residual compressive and flexural strengths of self-compacting high-performance concrete (SCHPC) that was produced by incorporating recycled coarse aggregate (RCA) as a partial replacement of natural aggregate (NA) and unprocessed waste powder materials as cement-replacing material (CRMs). The unprocessed waste powder materials that were used were waste fly ash (WFA) and waste perlite powder (WPP); the fire resistance of SCHPC that is produced by incorporating unprocessed waste materials has hardly been discussed in the literature.These authors observed that using an RCA of up to 50% enhanced the residual mechanical properties of SCHPC after exposure to the elevated temperature due to the strong aggregate–mortar contact zone and the similarity of thermal expansion between them. The fire resistance of SCHPC has been enhanced by replacing the cement up to 15% of WPP; meanwhile, WFA did not affect the fire resistance of SCHPC significantly.

JelčićRukavina et al. [11] studied the effect of mineral additives on the compressive behavior of high-strength self-compacting concrete that was exposed to temperatures up to 600 °C. A total of ten different concrete compositions were tested, in which part of the cement (by weight) was replaced by three different mineral additives (5–15% metakaolin, 20–40% fly ash, and 5–15% limestone). The damage that was caused by the high temperatures was assessed using scanning electron microscope micrographs and the stress-strain curves, compressive strength, modulus of elasticity, and the strain at peak stress were evaluated from uniaxial compression tests. After 200 °C, the authors observed a decrease in the mechanical properties and an increase in the peak strain for all the mixes that were tested. The different mineral additives that were used in this study affected the variations of the residual compressive strength by 24% and the peak strain by 38%, while the variations of the residual modulus elasticity were 14%.

In the study by Abed et al. [12], the residual density, compressive strength, flexural strength, and ultrasonic pulse velocity (UPV) of high-performance self-compacting concrete (HPSCC) were evaluated after exposure to elevated temperatures. A total of 21 HPSCC mixes were prepared by incorporating coarse recycled concrete aggregate (RA) and alternative waste materials (waste fly ash, perlite, and cellular concrete powders) as a partial replacement of coarse natural aggregate (NA) and cement, respectively. The mixes were conducted to check the correlation between the relative residual UPV and other properties of concrete after exposure to elevated temperatures ranging from 20 °C to 800 °C. The results of this study showed that the incorporation of RA as replacement of NA as well as alternative sustainable materials as cement replacing materials not only increased its sustainability but also improved its performance after exposure to elevated temperatures. It was also found that the relative residual UPV is correlated to the strength and density of HPSCC after fire exposure.

In the study of Sideris [13], the compressive strength of four SCCs and four conventional concretes (CC) of different strengths was determined. A speed of 5 °C/min was adopted for heating for temperatures at 100, 300, 500, and 700 °C. The results showed that for the same resistance class, the residual compressive strength of CC was lower than that of SCC mixtures. The two concretes SCC and CC presented the same spalling behavior and depended only on the category of resistance.

Noumowé et al. [14] examined the effect of adding of polypropylene fibers on the thermal stability of high performance SCCs during slow and fast heating. They noticed the instability of high performance SCCs during slow heating, while these same concretes to which they have incorporated polypropylene fibers did not present any disorder or instability.

Sideris [13] observed explosive bursts for water/binder (W/B) ratios of 0.45 and 0.46 corresponding to high performance SCCs (75 and 55 MPa respectively) and high performance concretes (65 and 45 MPa and W/B of 0.43 and 0.46, respectively). In this study, the SCC presented a higher residual resistance than that of the OC and spalling was observed for all the concretes that were tested.

Janson et al. [15] measured the thermal conductivity and temperature fields in three different types of concrete: OC (W/C = 0.70, Rc28 = 38.5MPa), HPC (W/C = 0.28, Rc28 = 114.2 MPa), and SCC (W/C = 0.38, Rc28 = 92.3 MPa) up to a temperature of 600 °C. They noticed that the thermal conductivity of OC was lower than that of the HPC and SCC. On the other hand, the HPC and the SCC have very close conductivities.

Persson [16] studied the evolution of porosity in SCC mixed with glass or limestone that was subjected to the following temperatures: 105, 200, 400, and 600 °C. He found that the porosity of SCC with glass fillers was greater than the other concretes that were tested. Between 400 °C and 600 °C, Persson noted a significant increase in the porosity for all the concretes due to the transformation of quartzites and the departure of water from cement hydrates during the dihydroxylation of portlandite.

In the study by Liu et al. [17], a comparison of the pore structure of SCC containing polypropylene fibers at high temperatures was performed. After heating to 500 °C, the authors observed a decrease in the porosity with the increase in fiber dosage due to the decomposition of hydrates and the increase in pores.

The impact of natural aggregates on concrete residual mechanical properties varies with their mineral composition. Tufail et al. [18] noted that natural granitic aggregate concrete retained the highest residual compressive strength at all tests elevated temperatures up to 800 °C compared to the natural limestone and quartzite. Likewise, Khaliq [19] compared the residual compressive strength of concrete that was made from natural crushed limestone and river gravel coarse aggregate during exposure to temperatures that rangedfrom 200–1200 °C. He observed a significant difference between the relative compressive strength in concrete with limestone (90%) and river gravel (50%) of unheated concrete strength after being exposed to temperatures up to 600 C.

In this study, the two types of concrete SCC and OC were tested at temperatures of 20, 150, 300, 450, and 600 °C. These values have been chosen according to the main transformations of the cementitious matrix that was observed by most researchers during the rise in temperature and which are [20]:

-

20–120 °C: departure from the open air and first signs of ettringite decomposition

-

130–170 °C: double endothermic reaction during the decomposition of gypsum

-

450–550 °C: decomposition of calcium hydroxide (CH) into free lime and water

-

600–750 °C: decomposition of calcium carbonate C-S-H.

The novelty of the study is the comparison between the physical and mechanical properties of the two types of concrete, SCCs and OC, at the same value of compressive strength. Several tests were carried out on the concrete specimens and the results are presented as a function of the temperature for each of the formulations that was studied.

Given the large volume of paste that is present in self-compacting concretes due to the large amounts of fines, we seek to understand the behavior of SCC under the action of high temperatures. From a durability point of view and following a heat treatment, are SCC as efficient as OC for identical resistance. In addition, we want to know if the regulations that are established for OC are applicable for SCC regarding the risk of bursting during heating.

2. Materials and Methods

2.1. Materials

2.1.1. Cement

A Portland cement (CEMI 42.5N) was selected for the preparation of the mixtures. The properties of the cement are presented in

Table 1. Chemical composition of cement.

Chemical Compound	Value
Insoluble Residue (IR)	≤5
Sulphur Trioxide (SO3)	≤3.5
Magnesium Oxide (MgO)	≤5
Chloride (CI)	≤0.1
Soundness (mm)	≤10
Initial Setting Time (min)	≥60

and

Table 2. Physical and mechanical properties of cement.

Property	Value
ρ (g/cm3)	3.13
BSS (cm2/g)	3100
True class (MPa)	44

, respectively.

2.1.2. The Limestone Filler

The limestone filer was used in our study to improve the stability and viscosity of SCC as well as its compressive strength.

Table 3. Physical and chemical properties of limestone filler.

Chemical Properties
CaCO3 (%)	SiO2 (%)	Na2O (%)	MgO	SO3 (%)
98.4	0.20	0.01	0.69	0.06
Physical properties
Curvature coefficient	Uniformity coefficient	Absolute density (g/cm3)	Bulk density (g/cm3)	Blaine specific surface (cm2/g)
1.87	8.60	2.71	0.81	4405

gives the physical and chemical properties of limestone filler.

2.1.3. Admixture

A superplasticizer (SP) Sika ViscoCrete Tempo 12, based on a modified polycarboxylate, was used to ensure good fluidity for the various formulations. It belongs to the family of highly water-reducing superplasticizers and is certified according to the requirement of NF EN 934-1 [21]. As its pH ranges between 4.5 and 6.5, the SP is particularly recommended for cements with a low or medium alkaline sulphate content. The dosage of the superplasticizer is about 1.1% of the weight of cement. The superplasticizer has a density of 1.06 ± 0.01 and CI⁻ content less than 0.1%. The SP must be stored in a closed room away from direct sunlight and frost between 5 and 35 °C.

2.1.4. Aggregates

The fine aggregates that were used in this study were natural sand 0/2 and crushed sand 2/4 with a fineness modulus of 1.79 and 4.8, respectively. For coarse aggregate, gravel 4/8 and gravel 8/12 were used. The aggregates properties are given in

Table 4. Physical parameters of aggregates.

Aggregates	Bulk Density (g/cm3)	Absolute Density (g/cm3)	Equivalent of Sand (%)	Absorption (%)	MDE (%)	Los Angeles (%)
Sand 0/2	1.48	2.46	76	-	-	-
Crushed sand 2/4	1.32	2.43	100	0.8	-	-
Gravel 4/8	1.42	2.50	-	0.8	18	20
Gravel 8/12	1.40	2.55	-	0.8	18	20

.

The classic sieving method was used to determine the particle size distribution of the aggregates in conformity with standard NF EN 933-1 [22]. Figure 1 gives the particle-size distribution curve of the sand and gravel.

2.2. Mixtures

A self-compacting concrete, noted SCC35, and an ordinary concrete, noted OC35, with a compressive strength of 35 MPa were prepared. The composition of the two types of concrete is shown in

Table 5. Composition of 1m³ of concrete.

Type of Concrete	Cement CEMI 42.5 N	Sand 0/2	Crushed Sand 2/4	Gravel 4/8	Gravel 8/12	Filer	Water
SCC35	350	460	400	440	300	130	200
OC35	370	460	450	400	390	-	200

.

For the various mixtures, the quantity of cement was calculated to obtain the values of the suitable strength. The dosage of superplasticizer was determined to ensure adequate slump (≥65 cm) for SCC35.

The water/binder (W/B) ratio of the different mixtures was calculated using the following relationship: (Equation (1))

$$\frac{\mathrm{W}}{\mathrm{B}}=\frac{\mathrm{W}}{\mathrm{C}+\mathrm{K}.\mathrm{F}}$$

(1)

where: C: cement dosage; K = 0.25; F: limestone filler dosage.

A concrete mixer with vertical axis was used for mixing the concrete according to the following steps:

-

Wet the walls of the mixer to prevent part of the mixing water from being absorbed,

-

Introduction of constituents (gravel, sand, limestone fillers, and cement) and mixing for 30 s,

-

Introduction of water + superplasticizer and mixing for 60 s,

For the vibrated concrete, the filling of the molds (cylindrical and prismatic) was done in three layers as indicated in the ASTM standard. After each layer, the molds were vibrated for 10 s using a vibrating table. The finishing operation was carried out using a straightedge.

Cylindrical specimens (160 × 320 mm), Parallelepipedic specimens (270 × 270 × 40 mm), as well prismatic specimens (100 × 100 × 400 mm) were made for each mixture (Figure 2). After pouring the concrete, the specimens must be stored at an ambient temperature of 20 ± 2 °C for the first seven days. Then, they are removed from the mold and put in airtight bags to ensure 100% relative humidity.

The specimens must be at least 90 days old at the time of the mechanical tests and must not remain outside for more than 2 h before the tests are carried out. The flow chart of the experimental study is given in Figure 3.

The compressive strength test, the ultrasonic pulse velocity test, and the elastic modulus of concrete were carried out on the cylindrical specimens, while the prismatic specimens were used to measure the tensile strength by four-point bending. The thermal conductivity values were determined using parallelepipedal specimens.

Table 6. Shape and dimensions of prepared specimens.

Test	SpecimenShape	Specimen Dimensions	Standard	Specimen Number
Compressive strength	Cylindrical	160 × 320 mm	NF EN 12390-3	45
Four-point bending	Prismatic	100 × 100 × 400 mm	NF EN12390-5	15
Elastic modulus	Cylindrical	160 × 320 mm	NF EN 12390-13	45
Thermal conductivity	Parallelepiped	270 × 270 × 40 mm	EN ISO 8990	15
Ultrasonic pulse velocity	Cylindrical	160 × 320 mm	NF EN 12504-4	15

presents the different tests that were carried out on different concrete specimens.

2.3. Experimental Tests

2.3.1. Thermal Treatment

The behavior of the concrete was studied by applying heat treatments at a low heating rate of 1 °C/min [23]. The heat treatment cycles included heating to 150, 300, 450, and 600 °C, followed by a one-hour dwell to ensure homogeneity temperature within the sample, then cooling to room temperature. The heating/cooling cycles were carried out in a programmable oven which allows the heating up to 1200 °C.

When heated to temperatures above 300 °C, the SCC35 specimens burst violently into several pieces (Figure 4a). These explosions were produced around 320 °C and only concerned the cylindrical specimens of dimensions 160 × 320 mm. On the other hand, the OC35 specimens showed good thermal stability and retained overall cohesion; no detachment of pieces of concrete on the specimens or bursting was observed (Figure 4b). This same behavior has been observed by other authors [24,25] who attributed these bursting to the low permeability and the drop in the tensile strength of the concrete.

2.3.2. Concrete Properties at Ambient Temperature

The characterization of the SCC in the fresh state was carried out according to the recommended European Guidelines. A slump test, V-funnel test, and an L-box test were used to determine, respectively, the flow ability, the viscosity, and the passing ability of concrete. To characterize the risk of segregation, the sieve stability test was also carried out to characterize the risk of segregation of concrete. The specifications for the different tests are also taken from EFNARC [2]. The equipment that was used for the different tests is presented in Figure 5.

2.3.3. Four-Point Bending Test

The four-point bending test was carried out on prismatic specimens of dimensions 100 × 100 × 400 mm in conformity with standard EN 12390-5 [26]. The flexural strength was calculated with the following formula (Equation (2)):

$${\mathrm{R}}_{\mathrm{f}}=\frac{{\mathrm{F}}_{\max }.\mathrm{L}}{{\mathrm{b}}^2}$$

(2)

where: F_max: Maximum bending force (N); L: Distance between supports (30 cm); b: Width of the specimen (10 cm).

2.3.4. Thermal Conductivity Measurement

The measurement of the thermal conductivity was carried out on the parallelepipedal test specimens of dimensions 270 × 270 × 40 mm in accordance with the requirements of EN ISO 8990 [27]. The principle of this method is to send a unidirectional heat flow through a sample of material to be tested, then taking the measurements after obtaining the steady state (Figure 6). The sample to be tested is placed between a cold isothermal capacitor and a constant flux heat source; this flux is assumed to be unidirectional. The temperature gradient that is established between the two faces is then measured. Once the steady state is established, the thermal conductivity can be determined by the following formula (Equation (3)) [28]:

$${\mathsf{\lambda }}_{\exp }=\frac{\mathrm{e}}{\mathrm{S}.({\mathrm{T}}_1-{\mathrm{T}}_2)}\left[\frac{{\mathrm{V}}^2}{\mathrm{R}}-\mathrm{C}.({\mathrm{T}}_{\mathrm{B}}-{\mathrm{T}}_{\mathrm{a}})\right]$$

(3)

where: T₁ and T₂: Sample surface temperatures (K); T_B − T_a: Temperature difference between the inside and the outside of the box (K); S: Sample section (m²); e: Sample thickness (m); V: potential difference (V); R: Resistance (Ω); C: Coefficient of box losses (W K⁻¹).

2.3.5. Compressive Strength Test

The compressive strength test was carried out on the cylindrical specimens in accordance with the requirements of EN 12390-3 [29]. The test consists of placing the test specimen between the steel plates of the hydraulic press and applying a continuous load until the specimen breaks (Figure 7). A compressive strength value R_C was calculated using the following formula (Equation (4))

$${\mathrm{R}}_{\mathrm{C}}=\frac{{\mathrm{F}}_{\mathrm{C}}}{\mathrm{S}}$$

(4)

where, F_C: Maximum breaking load (KN); S: Specimen section (cm²).

2.3.6. Ultrasonic Pulse Velocity Test

The ultrasonic pulse velocity (UPV) was determined in accordance with the requirements of EN 12504-4. The general principle of the UPV is to measure the speed of the mechanical wave propagating in concrete in transmission, in reflection, or on the surface (Figure 8). A pair of transducers is used for this, one serving as a source and the other as the receiver.

Knowing the distance L from the transmitter to the receiver, it is possible to know the ultrasonic pulse velocity UPV of the wave by applying the following formula (Equation (5)):

$$\mathrm{U}\mathrm{P}\mathrm{V}=\frac{\mathrm{L}}{\mathrm{T}}$$

(5)

where, L: Distance from the transmitter to the receiver(m); T: Time (s).

3. Results and Discussion

3.1. Concrete Properties at Ambient Temperature

Table 7. Properties of concrete before thermal treatment.

Test	Normative Value	Concrete		Specimen Number
		SCC35	OC35	SCC35	OC35
Slump (mm)		-	180	-	-
Slump flow (mm)	66 à 75	700	-	-	-
L-box	H2/H1 ≥ 0.8	0.82	-	-	-
V-Funnel	8 ≤ Time ≤ 14 s	9.25	-	-	-
Sieve stability	<15%	1.13	-	-	-
Compressive strength Rc (MPa)		41.7	37.1	3	3
Flexural resistance (MPa)		4.7	3.5	3	3
Elastic modulus (GPa)		36.05	38.25	3	3

presents the mechanical properties of the concrete specimens as well as the properties of the concrete mixtures before heat treatment. It is noted that the compressive strength and the elastic modulus of SCC35 presents much lower dispersions than those of the OC35. This difference in behavior is related to the high homogeneity and reproducibility of SCC compared to that of OC.

3.2. Properties of Concrete under the Effect of Heat Treatment

3.2.1. Compressive Strength

Figure 9 presents the results of the compressive strength test on the different specimens at the age of 90 days. We note that the behavior of OC is different from that of SCC. Indeed, for ordinary concrete, we observe a monotonous decrease in the relative compressive strength which accelerates from 300 °C. On the other hand, for the SCC, we note a moderate decrease in the resistance between 20˚ and 150 °C.

According to Figure 9, we also note for the SCC, an increase of 6% in the relative compressive strength between 150 and 300 °C as a result of the additional hydration of the cementitious matrix [30]. According to Khoury [31], this increase in strength can be explained by a loss of bonds with water of the silanol groups. This increase in resistance was also confirmed by Xu [32] using micro-hardness tests which highlight an additional fragility of the transition ring.

At 600 °C, the relative compressive strength values were very low and reached values that were lower than 10 MPa. This significant decrease is due to a coupled degradation of the cementitious matrix and the beginning of disintegration of the aggregates causing significant cracking [33].

Some researchers observed an increased residual compressive strength at initial elevated temperature range of 100–300 °C, followed by a sharp decrease in strength. An increased compressive strength was observed by Ahmad et al. [34] at 100 °C. In the study that was conducted by Tai et al. [35], the increase in the compressive strength was noticed at the temperature range of 200–300 °C.

According to Vieira et al. [36], the increase in strength at the initial high temperature range is mainly due to dry hardening of the cement paste that is caused by further hydration of cement and internal steam curing due to hot vapor through internal autoclave conditions at this temperature range.

Concrete mechanical properties generally decrease with increasing temperature; concrete loss approximately 25% of its original compressive strength when heated to 300 °C and about 75% when exposed to temperature above 600 °C [37]. These results are very close to the values that were found in our study since the compressive strength has been reduced by 20% for heating at 300 °C and about 78% for heating at 600 °C.

3.2.2. Elastic Modulus

The measurement of the elastic modulus was carried out on cylindrical specimens 160 × 320 mm in conformity with the standard of EN 12390-13 [38]. The test consists of applying three loading-unloading cycles at a speed of 0.5 MPa/s between 0.5 MPa and one third of the value of the compressive strength of the concrete. At the end, the specimen is subjected to a load until rupture and the modulus of elasticity is then deduced by taking the average of the secant modulus that is obtained on the loading curves of the last two cycles [39].

Figure 10 shows the variation in the elastic modulus of the two concretes as a function of temperature. It is noted that as soon as the temperature reaches the value 150 °C, and contrary to the compressive strength, the elastic modulus is entirely deteriorated by the heating. Between temperatures of 20 and 150 °C, the variation in the elastic modulus is identical for the different types of concrete.

According to the results, we also note a gradual decrease in the elastic modulus up to 300 °C. In this case, SCC35 has a modulus that is greater than 50% of the initial modulus, on the other hand, the elastic modulus of OC35 is lower. This decrease is due to the appearance of micro-cracks on the specimens that were tested as well as the increase in porosity. At the heating temperature of 600 °C, the concrete specimens have very low rigidities and a modulus of elasticity less than 0.5GPa.

According to the results that are presented in Figure 10, it is noted that the modulus of elasticity is the most affected mechanical propertyof concrete during exposure to elevated temperatures. Indeed, the degradation of modulus of elasticity is much faster than that of the compressive strength and tensile strength due to the loss of free water and the effect of crack development on concrete specimens that are exposed to high temperature [40].

Several authors have studied the effect of the heating-cooling cycleon the elastic modulus. According to Tolentino [41], the decrease in the elastic modulus is due to the growth in pore volume in the concrete and the cracking of the paste-aggregate interface. Liu [42] showed that the W/C ratio is not directly related to the elastic modulus and does not influence its evolution. According to Phan [24], a loss of 70% of the elastic modulus was observed during the heating process of the concrete to 300 °C and for a W/C ratio varying from 0.22 to 0.57. According to a study that was carried out by Kanema [43], a loss of 95% of the modulus of elasticity was found for high performance concretes as well as a loss of 85% of the modulus of elasticity for ordinary concrete.

3.2.3. Flexural Strength

The evolution of flexural strength that was obtained from four-point flexural test depending on the temperature is presented in Figure 11. According to the results, it is noted that the residual flexural strength of the concretes decreases progressively with the increase in temperature. From 450 °C to 600 °C, the loss of residual tensile strength is remarkable. Moreover, the intersection of the curves takes place at 600 °C and this is due to the loss of mechanical properties of the concretes.

During the tensile test, the rupture mechanisms of the specimens are different at 300 °C and 600 °C. Indeed, at 300 °C, concrete exhibits ductile behavior, while at 600 °C the concrete is rather brittle.

The degree of cracking has a greater influence on the resistance in tension than in compression. The appearance of cracks following heating at 300 °C does not promote an immediate loss of compressive strength [30] but lowers the tensile strength, and therefore the bending resistance.

The decrease in the flexural strengths of the tested SCC was confirmed in the results of de Xu et al. [32] who showed by micro-hardness tests on OC that the ring of transition is affected by heating: it is weakened by the appearance of cracks between 150 and 300 °C. This behavior generates micro-cracking at the interface and within the transition ring [44].

According to Khaliq et al. [19], as the temperature rises, the physical and chemical changes that take place due to hydrothermal processes lead to a reduction of the flexural strength. The addition of fibers and supplementary cementitious materials in concrete mix has been observed to significantly reduce the residual flexural strength loss, thereby improving its residual tensile strength after being exposed to elevated temperatures.

3.2.4. Thermal Conductivity

In general, when the temperature increases, the thermal conductivity of concrete decreases. This behavior highlights the deterioration of the microstructure and the appearance of micro-cracks that limit heat transfer. The water content, the type of aggregate, the type of cement, and the formulation of the concrete are the main parameters of this variation [45]. Figure 12 shows the thermal conductivity measurement results for the different types of concrete during heating and return to room temperature. It is noted that during the rise in temperature, the conductivity gradually drops from 20 °C to 150 °C, then it drops to 300 °C and slows down between 300 °C and 600 °C. On cooling, the conductivity value is stable (around 1.2 W/mK). Such a stagnation of the conductivity between 20 °C and 600 °C makes it possible to conclude that the physico-chemical reactions that are taking place within the concrete during heating are irreversible.

Diederichs et al. [25] studied the evolution of the thermal conductivity as a function of temperature for different types of concrete that had not previously been dried. The results show that the thermal conductivity initially increases with temperature and then decreases above 90 °C. For limestone concrete, for example, this increase is 15% and is mainly due to the presence of water whose thermal conductivity varies from 0.6 W/m.K to 0.68 W/m.K for temperatures from 25 °C to 130 °C. The thermal conductivity begins to decrease when the water begins to evaporate.

3.2.5. Ultrasonic Pulse Velocity Test

The measurement of the speed of the sonic waves (UPV) was carried out at the temperatures of 150, 300, 450, and 600 °C. According to the results that are presented in Figure 13, it is noticed that UPV decreases in a continuous way with the increase of the temperature because of the changes in the structure of the concrete.

Table 8. Concrete quality according to UPV.

Ultrasonic Pulse Velocity (m/s)	Quality of Concrete
>4500	Excellent
3500–4500	Good
3000–3500	Medium
<3000	Doubtful

gives the concrete quality according to the pulse velocity. We notice that all the specimens show good concrete quality at 20 °C and after heating at 150 °C since their UPVs are between 3500 m/s and 4500 m/s. For temperatures above 300 °C, an acceleration in the reduction of the speed of the sonic waves is observed. This acceleration is due to an increase in the porosity and the deterioration of the microstructure, thus causing new cracks in the concrete. At 600 °C, the UPV values of all the specimens are less than 3500 m/s and the reduction in the speed of the sonic waves is very high and reaches 48% for the SC35 and 51% for the OC35.

4. Conclusions

This paper presented the behavior of self-compacting concrete under the effect of high temperatures and their effects on the physical and mechanical properties. The following results can be deduced:

For temperatures between 150 and 300 °C, the relative compressive strength of SCC35 significantly increases.

At 600 °C, the relative compressive strength values are very low (lower than 5 MPa) due to a coupled degradation of the cementitious matrix and the beginning of disintegration of the aggregates.

The variation in the elastic modulus is identical for the different types of concrete between temperatures of 20 and 150 °C.

Up to a temperature of 300 °C, the SCC35 has an elastic modulus that is greater than 50% of the initial modulus.On the other hand, the elastic modulus of the OC35 is lower due to the appearance of micro-cracks on the specimens that were tested.

The residual flexural strength of the mixtures decreases progressively with the increase in temperature. During the tensile test, the rupture mechanisms of the specimens are different at 300 °C and 600 °C. Indeed, at 300 °C, concrete exhibits ductile behavior, while at 600 °C the concrete is rather brittle.

The thermal conductivity of concrete decreases with increasing temperature. This behavior is due to the degradation of the microstructure of the concrete and the appearance of micro-cracks which limit heat transfer.

UPV continuously decreases as the temperature increases due to changes in the structure of the concrete.

During heating, cylindrical specimens of dimensions 160 × 320 mm made of SCC35 present a greater risk of bursting than those that are made of OC35.

This study has made it possible to understand the behavior of SCC under the effect of high temperatures. Physico-chemical transformations have created significant repercussions on the physical properties, especially the appearance of cracks that are visible to the naked eye due to the decrease in the elastic modulus of the heated concrete.

This experimental study dealt with only a few mechanical and physical properties of mixtures of SCC under the effect of high temperatures. Further research can be done to study other properties such as water absorption and fire resistance. It is also important to determine the acoustic characteristics of the material to see the possibility of its use as a sound insulator.