Confinement effectiveness of CFRP strengthened ultra-high performance concrete cylinders exposed to elevated temperatures

Portland ASTM Type-I cement has been utilized and the type of sand used in this study is quart sand with a nominal particle size of no >2 mm. The proportions of the mixture are listed in Table 1. All the mix designs contained 900 kg/m³ cement and 222 kg/m³ micro silica. All mixtures contained 1,000 kg/m³ sand. The water-to-binder ratio (w/b) in the entire combinations was 0.146 and the concrete workability has been maintained by adding 30 kg/m³ of superplasticizer. In this study, three fiber types were evaluated, including steel, polyvinyl alcohol (PVA), and PP, as shown in Figure 1. The characteristics of these fibers are presented in Table 2. The percentage by volume of fibers, which were used in each mix design, is provided in Table 3.

Fig. 1

In this study, the testing program focused on the effect of the CFRP sheet strengthening on the performance of the heat-damaged plain and fiber-reinforced UHPC. The control (no-heat exposure), as well as the specimens, were exposed to high-temperature settings of 400°C for 3 h. The temperature range and the exposure duration of the study have simulated the temperatures of exposure in the liner of the RC chimneys, clinker silos, and the containment vessels of the power plant [34, 35]. However, six concrete mix designs were examined, as illustrated in Table 3. The first mix was a UHPC control without fibers (M0), while the subsequent five mixes contained fibers (i.e., UHPFRC). Moreover, three various types of chopped fibers were used in producing UHPFRC, including SF, PP, and PVA fibers.

After mixing, the slump of the concrete was recorded, and the results are provided in Table 4. Plastic molds were utilized to create the concrete cylinder samples, whereby the concrete was poured into one layer. For the concrete prisms, the concrete was placed in steel molds. The filled molds are shown in Figure 2. After casting, the concrete surface was complete using a steel trowel. After 24 h of moist curing at room temperature, the concrete cylinders and prisms were demolded and cured in water up until 28 days, as shown in Figure 3.

Fig. 2

Fig. 3

After 28-day water immersion, the concrete specimens were air-dried at room temperature for another 28 days to guarantee that the concrete was completely dry before being exposed to high temperatures. This has been carried out to allow drying shrinkage before the abrupt temperature increase caused by the high-temperature exposure. As shown in Figure 4, an electrical oven with an internal dimension of 1 m × 1 m × 1 m was used. The concrete specimens were heated at a rate of 8°C/min. The heating curves for each of the heating regimes was 400°C are shown in Figure 5, which includes the heating curve, as per ISO 834 [36] for comparison. A Type-K thermocouple was used to regulate the oven's temperature. When the oven temperature reached its maximum setting, the temperature was maintained for 3 h. The concrete specimens were cooled to room temperature inside the oven after heating.

Fig. 4

Fig. 5

After heating, the cylinders were strengthened by using two layers of the CFRP sheet. Before applying the strengthening schemes, the surfaces of all the cylinder specimens were roughened using sandblasting to ensure a strong bond between the applied strengthening layer and the concrete surface. Additionally, before CFRP jacketing, the specimens’ surfaces were lightly scratched to eliminate surface impurities. First, the specimen surface was properly cleaned to remove any dust. Next, the concrete's external surface was coated with an epoxy primer layer in the second step to fill the air voids and create a firm connection. Then, a thin coat of epoxy was applied to the concrete columns. After that, the CFRP layer was carefully wrapped. Special attention was paid to ensure that there were no air spaces between the fiber and the concrete surface. Before testing, the entire samples were maintained at room temperature for at least 28 days. Figure 6 shows the use of CFRP sheets to strengthen the cylinder specimens. The properties of the CFRP sheet are given in Table 5.

Fig. 6

The fresh properties of concrete were evaluated by using unit weight and slump flow. The properties of hardened concrete were also evaluated by using compressive strength. The compressive strength of concrete was measured according to ASTM C39 [37]. Before testing, sulfur capping was applied to the concrete cylinders to ensure a level testing surface. A compressometer, which comprised two LVDTs, has been used to measure the axial strain during the compressive strength test, as shown in Figure 7. Furthermore, a data acquisition system was used to record the readings of the load and deflections.

Fig. 7

Figure 8 displays the concrete specimens after they were exposed to elevated temperatures. The discoloration was observed in the concrete specimens, which were heated to an elevated temperature of 400°C. At a high temperature of 400°C, the examined concrete specimens without SF (M0) and the concrete specimens with SF alone (M1) exhibited a slight surface spalling at the specimens’ ends and corners. This spalling can be linked to the concrete's strong microstructure, making it difficult for water vapor to escape and, thus, leading to substantial internal pressure in certain locations. The absence of the PVA and PP fibers can also be a contributing factor because these fibers melt and evaporate at approximately 270°C and 170°C, respectively, leaving venting canals that can reduce the internal vapor pressure (Phan, 2008). The concrete specimens, which were reinforced with PVA and PP fibers, demonstrated considerably less cracking and spalling compared with the specimens, which were reinforced without PVA and PP fibers.

Fig. 8

The compressive strength test results of the UHPC specimens’ pre- and post-strengthening are provided in Table 6. Figures 9 and 10 show the obtained concrete cylinders after testing the compressive strength, and each value in Table 6 presents an average of the three cylinders. Moreover, the loss percentage of compressive strength in comparison with the primary compressive strength before being exposed to the heating regime is presented in Table 6. In general, the concrete specimens, which were exposed to high temperatures, recorded a lower compressive strength compared with those that were not exposed to high temperatures. The increase in the temperature decreased the residual compressive strength in all mixes regardless of fiber content or type.

Fig. 9

Fig. 10

Regarding the unheated concrete specimens (i.e., at room temperature), the use of SF increased the compressive strength by approximately 16% compared with the control mix (i.e., mix M0). When a hybrid PVA system and SF were used, the increase in the compressive strength ranged from 7% to 12% compared to the bare mix (i.e., mix M0). The concrete specimens, which were reinforced by using a hybrid system of steel, PVA, and PP fibers demonstrated a compressive strength improvement of approximately 10%, as shown in Table 6.

Regarding the heated concrete specimens at 400°C, the use of a hybrid fiber system has been effective in increasing the residual compressive strength capacity after the heating exposure compared with the concrete specimens without any fibers. This can be attributed to the fibers’ ability in controlling the thermal cracking and venting of the internal vapor pressure because PP and PVA fibers can melt and evaporate at high temperatures. The compressive strength tests on UHPC and UHPFRC revealed that the exposure to high temperatures of 400°C resulted in a considerable loss of strength. The compressive strength was reduced by approximately 24% in the control mix without fibers. A comparable decrease in compressive strength was observed in the concrete specimens, which were made of SF.

The use of a hybrid fiber system instead of SF has enhanced the residual compressive strength. The decrease in the compressive strength of the concrete specimens made of a hybrid system, which consists of steel and PVA fibers ranged between 18% and 24%. The concrete specimens, which were reinforced by using a hybrid system of steel, PVA, and PP fibers demonstrated a compressive strength loss of about 22%.

Figure 10 displays the failure modes of the concrete's compressive strength specimens before and after being exposed to a high temperature of 400°C. The failure of the unconfined UHPC specimens was brittle, with a quick load reduction after the peak load was attained. The UHPC specimens, which were made with or without fibers and exposed to room temperature showed more brittle and failure in an explosive nature compared with similar concrete mixes, which were exposed to 400°C. In general, the failure mode was more ductile in the concrete specimens, which were made of fibers and exposed to high temperatures at 400°C vs. the concrete specimens without fibers. The concrete specimens, which were made with a hybrid system of PP and PVA fiber showed the greatest ductile behavior in comparison with all the concrete mixes. This can be attributed to the melting and evaporation of the PP and PVA fibers, causing increased concrete porosity.

Figure 10 shows the representative samples of failure modes because of the compression tests of all types of concrete confined cylinders. All the strengthened specimens failed due to a quick and violent rupture of the CFRP sheets. The jacket has mainly failed because of the vertical and diagonal cracks. With an increasing load, the fissures become wider. Additionally, a noise was heard as the load increased, and failure occurred at the top and bottom of the specimens, while others failed in the mid-height of the specimens, as shown in Figure 10 for the unheated and heated specimens. In this region, the stress sensitivity was because of the release of high energy in the form of brittle failure of the CFRP sheets. The damaged specimens showed good contact between the concrete and the CFRP jackets, and there was no de-bonding.

Evaluating the stress-strain relation of concrete is crucial, particularly when the concrete is exposed to elevated temperatures. Figure 11(a) and 11(b) show the stress-strain curves of the concrete specimens, which were exposed to 26°C and 400°C, respectively.

Fig. 11

As shown in these figures, the concrete specimens, which were exposed to room temperature, had a higher elastic modulus compared with their corresponding concrete mixtures, which were exposed to a high temperature of 400°C. The concrete mixture's modulus of elasticity, which was subject to 400°C, was the lowest in comparison with corresponding concrete mixes, which were exposed to room temperature. The concrete mixes, which were heated to 400°C, demonstrated an increased ductility and energy dissipation. It has also been observed that when the UHPFRC specimens were exposed to an elevated temperature of 400°C, the stiffness loss was greater than the compressive strength reduction, probably because of the deterioration of the concrete matrix and the formation of thermal cracks. UHPFRC reinforced with SF exhibited a higher elastic modulus and energy absorption (i.e., area under the stress-strain curve) compared with the concrete specimens without fibers or hybrid UHPFRC. This has been expected, given that SF have a higher tensile strength and modulus of elasticity than the PVA or PP fibers. Additionally, SF are more resistant to high temperatures than PVA and PP fibers. The stress-strain relationships of the unheated and heated concrete specimens showed that the modulus of elasticity decreased as the temperature exposure increased from an ambient temperature to an elevated temperature of 400°C, as shown in Figures 11(b) and 12(b). The stress-strain curves for all strengthened specimens with CFRP showed that the ultimate axial and lateral strains increased substantially, thereby indicating enhanced ductility. Such a significant enhancement in ductility is owing to significant confinement to micro-cracked heat-damaged specimens.

Fig. 12

The exposure temperatures were plotted vs. the normalized UHPC and UHPFRC strength after strengthened (i.e. the ratio of the residual strength of UHPC after high-temperature exposure to the UHPC strength at room temperature) for different fiber types to investigate the effect of fiber type and combination on the performance of test specimens after high-temperature exposure, as shown in Figure 13. The addition of fibers in the plain of UHPC has a considerable impact on the residual compressive strength of concrete, as seen in these data. The inclusion of fibers in the concrete mixture significantly reduced the loss of compressive strength caused by high temperatures. As a result, post-heated UHPFRC cylinders lose significantly less residual compressive strength than plain UHPC specimens. When compared to SF alone and the hybrid system consisting of steel and PVA fibers, the hybrid fiber system consisting of steel, PVA, and PP fibers was the most successful in preserving compressive strength at 400°C. The superior behavior of PP as well as PVA fibers can be attributed to their melting at high temperatures, which produced channels inside the concrete matrix [38].

Fig. 13

The effect of CFRP wrapping on the behavior of stress-strain response of heated-damaged UHPC has been clearly observed. After exposure to high temperatures, the strength and ductility of concrete were significantly increased. The CFRP wrapping can restore the original compressive strength at room temperature even after 3 h of exposure at 400°C, but the modulus of elasticity cannot be restored due to the confinement effect on the modulus of elasticity of heated-damaged concrete, which was minimal, particularly with a limited number of layers. Moreover, for all the specimens with and without fibers, the results of the ultimate strain of CFRP confined concrete tended to increase with the fire damage level.

Figure 12 shows the effect of the wrapping scheme on the stress-strain curves of all the UHPC specimens before and after strengthening at room temperature and varying elevated temperatures. Obviously, this demonstrates that the strength and ductility of all the mix types improved considerably; it has been more obvious for the specimens, which were exposed to elevated temperatures.

Figure 13 shows the influence of temperature exposure on the residual compressive strength of heated-damaged concrete. The ratio of increased with the temperature exposure. The effect of exposure was more obvious on the grade of normal strength, and the greatest effect exposure was for the high strength concrete.

It is, therefore, obvious that the strength improvement level by CFRP confinement for heated-damaged UHPC was greater in comparison with the unheated concrete. The confinement effectiveness increased with the level of heated damage, especially for the UHPFRC specimen.

The CFRP wrapped cylinders showed higher compressive strengths compared to the unwrapped specimens. This increase ranged between 25% and 32% for UHPC and UHFRP wrapped specimens. This indicates that the confinement effectiveness of UHPC using CFRP sheets was more effective in increasing the capacity of the concrete columns. Regarding the post-heated specimens, they showed a significant increase in compressive strength after being strengthened with two layers of the CFRP sheets. The increase ranged between 33% and 59% for all concrete types. It was observed that enhancing the strength and ductility of heat-damaged cylinders, which were wrapped in two layers of CFRP sheets was more effective in retrofitting the ultimate load capacity of UHPC and UHPFRC cylinders. The axial and lateral ultimate strains increased considerably, exhibiting ductility improvement – this improvement in ductility can be attributed to significant confinement to micro-cracked heated-damaged concrete cylinders.