Effect of waste COVID-19 face masks on self-compacting high-strength mortars exposed to elevated temperature

During the pandemic, it becomes customary to wear a disposable surgical (face) mask (SM) to guard against coronavirus illness 19 (COVID-19). However, because existing disposal procedures (i.e., incineration and reclamation) emit hazardous substances, vast generations of contaminated surgical masks pose an environmental risk. Therefore, many studies are currently being carried out worldwide to dispose of SM. The easiest and cheapest of these methods is the disposal of SMs in cement-based composites. This study cut waste SMs to macro size and used them in cement-based composites such as polypropylene fiber. The elevated temperature resistance of cement-based composites decreases as their compressive strength rises. Low-melting materials like polypropylene fiber are utilized to improve the high-temperature resistance of cement-based composites. Therefore, SM with a low melting temperature was used in the design of the mixtures. SM was added to the mix at rates of 0.3, 0.5, 0.8, and 1 by weight of cement. As the SM ratio increased, the workability of the mixtures decreased. Water absorption and apparent porosity increased as SM reduced the workability of composites. The mixes' 28-day compressive strength ranges from 36.6 to 79.4 MPa. It was observed that flexural strength raised in some mixtures when SM was used. In the mixes using 0.5 % SM, about 40 MPa compressive strength was obtained after 800 °C. Additionally, SEM images showed that SM changed into microfibre during mixing. As a result, it has been determined that SM can be used at low rates to increase the elevated temperature resistance of cement-based composites.


Introduction
A new coronavirus disease 19 (COVID-19) that is extremely infectious was identified in 2019 [1,2]. COVID-19 was declared a pandemic by the World Health Organization (WHO) in March 2020. COVID-19 has started infecting millions of people on all continents [3]. Under these circumstances, most countries require non-pharmaceutical interventions to physically block COVID-19 transmission by providing personal protective equipment (PPE) (mask, glove, protective gown, etc.) [4][5][6]. As a result, the demand for PPE has grown, creating plastic waste [7]. Unfortunately, due to various factors, a viable PPE disposal platform has yet to be established [8]. Plastic pollution has increased as a result of the COVID-19 epidemic. The use of plastic is growing due to its heavy reliance on takeout and hygiene concerns, leading to the use of PPE. However, initiatives to reduce plastic use, such as a ban on single-use plastic bags, have been delayed, shelved, or outright banned [9]. Single-use face masks are the most popular personal protective equipment to stop the pandemic from spreading. Because many national and local governments have required single-use face masks, their use has grown commonplace. Face masks had a global market size of around 0.8 billion dollars in 2019, but by the end of 2020, it had grown to over 166 billion dollars [10]. Facemask waste has increased dramatically due to the increasing demand for face masks. Due to their complicated compositions and the potential for COVID-19 infection, they are difficult to recycle. Used masks are frequently disposed of inappropriately, mostly because waste management cannot keep up with the expanding amount of waste  [11]. Many incorrectly disposed face masks are in the water, polluting marine habitats. According to OceansAsia, about 1.6 billion masks will reach the ocean in 2020, resulting in 4700-6200 tons of plastic pollution [12]. Therefore, scientists and local governments are looking for alternative solutions for the disposal of waste face masks. Using waste face masks in cement-based composites such as mortar and concrete may be an alternative solution. Concrete is widespread because it is the world's most widely utilized construction material [13]. Concrete is a non-flammable substance with a slower heat transmission rate (conductivity). As a result, concrete structures function admirably in the event of a fire in the majority of circumstances. However, high-strength concrete structures are sensitive to thermal spalling due to the dense microstructure of the concrete matrix. Spalling is the breaking off of pieces of concrete from the concrete member's surface. This results in the loss of the concrete section and, in some situations, the direct exposure of the reinforcement to the fire, which affects the concrete member's loading capacity and fire resistance. As a result, fire spalling is one of the most serious risks to concrete structures, especially those constructed of high-strength concrete [14,15]. Concrete spalling is often explained by one of two mechanisms: (a) Pore pressure-induced spalling is mostly related to the thermal hydration process inside the heated concrete. As a result of heat transfer and moisture migration, pore pressure eventually builds up in the micropores. Spalling occurs when the subsequent tensile stress generated by the pore pressure exceeds the tensile strength of the concrete. (b) Delamination mechanism due to thermal stress. This is related to the generation of temperature gradients at the heated site and the resulting forced thermal expansion Pore pressure and thermal stress can act separately or in concert [15]. One of the thermo-chemo-mechanical challenges is fire damage to concrete structures. The disintegration of CASAH gels, a plate-shaped crystal of Ca(OH) 2 , CaCO 3 , Al 2 O 3 -Fe 2 O 3 -tri (AFt), and AFm are all thermos-chemo-physical phenomena that should be included in computational modeling of cementitious composites at higher temperatures, as shown in Fig. 1. The thermos-chemo-physical reactions that occur during elevated temperatures are explained by the Japanese Concrete Institute [16]. Therefore, various methods have been developed to reduce the damage in high-strength concrete during fire. The most common of these methods is the use of polymer fibers with a low melting temperature. Adding polymer fibers to prevent spalling is a well established approach. To prevent HPC from spalling, 0.3 and 0.6 vol% polymer fibers, such as nylon fibers, polypropylene (PP) fibers, and polyethylene (PE) fibers, are commonly required [17,18].
The cooling conditions of cement-based composites directly affect the mechanical properties. The firefighter usually extinguishes fires in the field with water or foam. Studies show the change in the mechanical properties of concretes exposed to elevated temperatures according to cooling conditions in the literature. Zhai et al. investigated the mechanical properties of concrete exposed to elevated temperatures after cooling with water. They observed an increase in strength in cooling with water. The content of Ca(OH) 2 in the concrete reduces while the content of CaCO 3 increases when heated to 200°C. When the water cools, it produces hydrated calcium silicate, and some of the free water fills the voids in the concrete simultaneously. The elevated temperature and water enhance the hydration reaction of the cement with the water in the concrete, and the concrete's peak stress is improved [19].
Air-entraining admixtures (AEAs) have been a major technological development in the construction sector over the last century. The development of air, gel and capillary pores is advantageous for AEA [20]. To increase workability and uniformity, stop bleeding and deterioration, and increase durability, AEA is frequently added to freshly mixed concrete [20][21][22]. Other advantage of AEA is that the increased air volume reduces thermal conductivity and raises resistance to heat flow. [23,24]. The downside of AEA is that it reduces overall density by increasing porosity, which affects compressive strength. However, water-reducing admixtures can make up for the strength loss. Additionally, research has demonstrated that using this chemical to produce more air bubbles considerably impacts frost resistance [25].
This study examined cement-based composites with compressive strength (28 days) of more than 50 MPa for their elevated temperature endurance. As stated in the literature, polypropylene fibers improve the high-temperature resistance of high-strength or high-performance composites. This study used waste surgical masks instead of polypropylene fiber to produce the mixtures. Because in the research, it has been seen that surgical masks, which are protective equipment for COVID-19, will cause significant pollution in the future. One of the easiest ways to dispose of these masks is to use them as macro fiber in cement or concrete production. This way, the wastes released during the COVID-19 process will be disposed of safely. In addition, considering that surgical masks will not melt in a short time in the mixtures, an airentraining admixture is used to reduce the internal vapor pressure that will form within the concrete. In this way, it is aimed to produce high-strength cement-based composites that are both environmentally friendly and resistant to elevated temperatures.

Materials and analytical methods
Cement, fine aggregate, and water were used in this study as in conventional mortars. In addition, two chemical admixtures, superplasticizer and air-entraining (AEA), were used. Waste surgical masks (SM) were used instead of polypropylene fiber. Fig. 1. The chemical change in hydration products [16].  The ratios of Al 2 O 3 , Fe 2 O 3 , SO 3, and MgO, which are  minor cement components, are 3.7 %, 2.9 %, 3.5 %, and 1.2 %, respectively. The specific gravity of CEM I cement is 3.14. In preparing the mixtures, silica aggregate was used to conform to ASTM C 33 [26] and was well-graded. Table 1 shows the physical characteristics of silica aggregate.
A polycarboxylate ether-based superplasticizer (PCE) was used to improve the workability of the mixtures. In addition, a ligninbased air-entraining admixture (AEA) was used to create micro air voids in the samples. The characteristics of the chemical admixtures used are given in Table 2.
The waste surgical masks (SM) used in the mixtures were cut and brought to the macro fiber size. The masks used against the COVID-19 epidemic are 3-ply. The general view and layer structure of these masks are given in Fig. 2 [27]. SMs were prepared by cutting into 2 cm lengths. SMs were kept in an oven at 80°C for 2 days to ensure decontamination. In addition, it was exposed to 400°C temperature (1 h) to observe the change of SMs in the mortar. The change in SM exposed to 400°C temperature is given in Fig. 3. As seen in the figure, SM melted at 400°C. However, a fibrous structure was observed after melting. Also, the TGA curve [27] of SM is presented in Fig. 2. Yousef et al. analyzed the thermal behavior of COVID-19 masks analytically (TGA/DTG method). The results revealed that the face mask contains many volatile materials and can collapse in three phases at 360-500°C (with a mass loss of 67-96 %) [27].

Mix proportion design
In this study, different ratios of SM and AEA were used, and 15 mixtures were prepared. SM was used at ratios of 0 %, 0.3 %, 0.5 %, 0.8 % and 1 % by weight of cement. AEA was used at the ratios of 0 %, 0.35 %, and 0.70 % by weight of cement. The cement dosage was chosen as 800 kg/m 3 to reduce the strength losses that AEA and SM may cause. The w/c ratio of all mixtures is 0.30. PCE was used at a ratio of 0.5 % by the weight of cement in all mixtures. PCE ratio was determined in pre-experiments. The material amounts and ratios of the mixtures are given in Table 3. Mortars with SM and AEA are prepared based on ASTM C305 [28] standard. First, cement and silica aggregate was mixed as a dry mixture for 1 min. Afterward, 2/3 of PC and 75 % water were added to the mixture and mixed for 4 min. After the wet mixture rested for 1 min, the remaining water, PCE, and AEA were added. After the mixture was stirred for another 2 min, SM was added. Mixtures with SM were mixed for an additional 3 min. After the prepared mixtures were placed in their molds, they were kept under laboratory conditions for 24 h.

Heating procudere
The mixtures were exposed to elevated temperatures with a special electric furnace (time-adjusted). The heating rate of the electric furnace is set at 5 min/ o C. The samples were dried at 105°C for 24 h before being placed in the electric furnace. After the mixtures reached the target temperature, they were subjected to elevated temperatures for 2 h. Afterward, two different cooling conditions, air and water, were applied to the mixtures. Air cooling; The samples were taken out of the electric furnace were kept in laboratory conditions until they cooled down. Water cooling; The samples were taken out of the electric furnace, put into the water tank, and cooled. The samples, cooled in water, were then dried again at 105°C and their weight losses were determined. Additionally, the mixes exposed to elevated temperatures measured their compressive and flexural strengths. 40x40x160 mm prism samples were used to determine the elevated temperature resistance. Fig. 4 shows the samples exposed to 800°C and the heating process used in the experimental study.

Experimental methods
The workability of the mixtures was compared according to their flow diameters. The method specified by EFNARC was used to determine the workability [29]. The fresh characteristics of the mixtures and the flow diameters of some mixtures are shown in Fig. 5. The high SM ratio induced segregation in some mixes, as can be seen in the image. In addition, bleeding was observed in some mixtures, but this situation was reduced with AEA.
The compressive and flexural strengths of the mixtures were used to determine their mechanical characteristics. Standard ASTM  C349 [30] for compressive strength and ASTM C348 standard [31] for flexural strength were used. A 40 Â 40 Â 160 mm sample was utilized to determine the mechanical properties. On days 7 and 28, the mixtures' compressive and flexural strengths were also assessed. The mixtures' dry bulk density, apparent porosity, and water absorption were evaluated using the ASTM C642 standard. [32]. For the physical properties of the mixtures, cube samples of 50x50x50 mm were produced. Physical properties were measured on 28-day mixtures. All experiments were carried out on three samples (only flexural test with six samples). In addition, micro and macro studies were carried out on mixtures exposed to 800°C. A scanning electron microscope (SEM) was used for micro-examinations. An optical microscope was used for macro examinations. Macro and micro research were carried out at Gazi University, Metallurgy and Engineering Laboratory.

Fresh characteristics
The effects of SM and AEA on fresh mortar properties are shown in Fig. 6. Flow diameters of mixtures without air-entraining admix-tures vary between 28 and 41 cm. As 1 % SM was added to non-AEA mixes, the flow diameter was decreased by 22.2 percent compared to the reference mix. It was found that the flow diameters showed similar properties if the air-entraining additive was 0.35. In the case of using 0.7 % AEA, the flow diameters vary between 31 and 40 cm. The use of 1 % SM in the mixtures reduces the spreading diameters. However, if air-entraining admixtures are used in these mixtures, flow diameters have been improved. In particular, if 0.35 % of AEA was used, the flow diameter increased by 32.1 %. If the air-entraining admixture ratio of 070 % is used in mixtures with SM, the flow diameters are relatively reduced. The highest flow diameters were generally observed in mixes with 0.35 % AEA. As the ratio of SM used in the mixtures increases, the flow time increases. The flow time was found to be 20.3 s when 1 % SM was used in mixes without air entrainment admixture. However, when 0.35 % AEA was used, the flow time was 17.7 s. If the AEA ratio was 0.70, the flow time decreased approximately-four times. The flow times of the mixtures with 1 % SM are relatively longer than the other mixtures. The rise in the AEA rate was more effective in the reference mixture. The use of air-entraining admixtures in self-consolidating mixtures generally increases viscosity. Therefore, the flow diameters were decreased when 0.70 % AEA was  used. Similarly, the increase in flow times occurred because of this. The workability of the mixtures decreased due to the water retention property of SM. In addition, the fact that SMs acted like fibers (Polypropylene fiber) and increased internal friction was also effective in this process.
Many studies in the literature show the effect of air-entraining additives on workability. It can be explained that air bubbles entrained in concrete only last a short period if the mixture is not constantly mixed. A portion of the ''ball bearing" effect is lost when these bubbles explode, sharply reducing the slump [33]. Similar findings were seen in the Dils et al. experiment. As the air-entraining additive ratio increased, the workability of the mixtures decreased. This effect is explained by increased yield stress and plastic viscosity of air-entraining admixtures [34]. The air bubble bridges provide a likely explanation for this phenomenon. Cement particles become positively charged when they hydrate. The cement particles, typically smaller than air bubbles, will be bonded to the anionic surfactants. The bubbles are coated with cement particles, which increases cohesiveness. These bridges cause increased flow resistance [35,36]. In this study, the waste surgical masks acted like polypropylene fiber. Hassanpouret et al. stated that the workability of concrete decreases as the polypropylene fiber content increases [37,38]. The friction between fibers and cement pastes and the usage of admixtures play a significant role in the workability loss of polypropylene fiber-reinforced concrete and mortars. Because polypropylene fiber increases granular structure and decreases maximum compactness, friction between fibers and cement particles increases, blocking granulate movement during the test. The slump value decreases, and the mixture's viscosity increases due to this behavior [39,40].
While the AEA was 0.35 %, the flow diameter was relatively increased, and the flow diameter was decreased again by 0.70 % (Fig. 7a). In this study, the optimum AEA ratio was found to be 0.35. When the optimum ratio of AEA is exceeded, the mixture's  plastic viscosity increases and its workability decreases. Increasing the AEA ratio also increases the flow times. The rise in air bubbles decreased the workability of the mixes. The mixes' flow widths get smaller as the SM ratio rises, as seen in Fig. 7b. Additionally, when SM increases, the flow time of the mixtures gets shorter. The fact that waste surgical masks have water absorption properties has reduced the workability of the mixes. In addition, the irregular shape of the SMs also increased the internal friction, and as a result, the properties of the fresh mortar were adversely affected. If the SM ratio was 0.3 %, a slight increase in the flow diameter was observed. Since the paste volume is high in the composites, the flow diameter was relatively increased with the use of low SM. However, workability can be improved as air-entraining admixtures increase the paste volume. In this way, workability has been enhanced in low AEA use. The workability losses caused by SM are reduced with air-entraining admixtures.

Physical characteristics
As shown in Fig. 8a, the apparent porosity of the mixtures increases as the AEA volume rises. The porosity increased by 10.4 % for the reference mixture when the AEA content was 0.35 % and 107.2 % when the AEA content was 0.70. The porosity values of the mixtures with an AEA content of 0.70 vary between 0.19 and 7.48 %. AEA admixture entrains air bubbles into the paste. These entrained air bubbles increase the porosity of the mixtures. If SM is used in some mixes, the porosity decreases compared to the reference mixture. For example, when the SM content of the AEA-free mixture was 1 %, the porosity was reduced by 15.8 % compared to the reference mixture. Similar behavior was observed in mixtures with an AEA content of 0.7 %. If the SM content was 0.3 %, the porosity decreased by 20.8 % compared to the reference mixture. It is thought that SM closes some of the voids between aggregates, as in polypropylene fibers. SM can act as micro aggregates like polypropylene fibers. Therefore, in some cases, SM has been observed to reduce the apparent porosity. It is seen in Fig. 8b that the water absorption values are parallel to the apparent porosity. The mixes' water absorption values increase as AEA content rises. Even though the combinations without AEA have water absorption values below 2 %, when the AEA level is 0.70, the values range from 2.63 to 4.45 %.The highest water absorption value was obtained in the mixture with an AEA content of 0.70 and an SM content of 1 %. The number of air bubbles in the paste rises with increasing AEA content. This increases the water absorption value of the mixture. In the case of SM of 0.8 % or 1 % in mixtures without AEA, the water absorption value decreased by about 12 % com-  pared to the reference mixture. Similar behavior was determined in mixtures with an AEA content of 0.70 %. SM reduced the water absorption values in some mixtures (For example, Mixture without AEA and 0.8-1 % SM) by separating the connections between the capillary voids. But because SM often makes materials less workable, mixes tend to absorb more water and become more porous. The mixes' dry bulk densities range from 2198 to 2424 kg/m 3 , as shown in Fig. 8c. The mixtures' dry bulk densities fall as AEA content rises. Since SM has a low specific gravity compared to the combinations' dry bulk densities, the change in SM concentration has little impact on those densities.
Studies have shown that AEA increases the porosity of concrete while decreasing its density [41,42]. Pundiene et al. determined that AEA reduced the density of concretes by approximately 17 %. They also observed that mixtures with AEA increased the porosity value from 17.1 % to 31 % [43]. Zeng et al. stated that as the AEA content increases, the mixture's porosity increases and the water absorption increases [44]. Wu et al. determined that the oven-dried densities of the mixtures relatively decreased as the polypropylene fiber content increased [45]. Because more air spaces are held in the fresh mix by the high polypropylene fiber content, the porosity increases and the mechanical properties are negatively impacted [46]. However, it has also been reported that fibers sometimes reduce water absorption and porosity. Previous studies discovered that low fiber content reduces water absorption and porosity because the fiber blocks the pores [47,48]. At high fiber content, the porosity of the mixtures increases due to the loss of workability. The formation of air voids causes moisture diffusion due to the high fiber content [48].

Mechanical properties
As seen in Fig. 9a, the 7-day flexural strengths of the mixtures without air-entraining additives vary between 11.9 and 15.8 MPa. If the SM ratio was 0.3 % and 1 %, the flexural strengths increased compared to the reference mixture. This result shows that SM behaves like polypropylene fiber. The lowest flexural strength was obtained in the mix with 0.8 % SM, with a value of 11.9 MPa. The flexural strengths decreased when using the airentraining additive ratio of %0. 35. As the air bubbles entrained into the paste reduce the porosity, their flexural strength has reduced. The flexural strength of the mıxture wıth 1 % SM and 0.35 % AEA  Fig. 10a. In mixes lacking an air entrainment admixture, the SM ratio increases but the 7-day compressive strength drops. When 0.8 % SM was used, the compressive strength decreased by 14.6 %. If the air-entraining admixture ratio is 0.35 %, the compressive strength declines below 60 MPa. In this group, the rise in the SM rate reduced the 7-day compressive strength. 0.3 or 0.5 % SM in mixtures with 0.70 % AEA improved the compressive strength. However, compared to the reference mixture, the com-  The AEA used in the mixtures increased the porosity and decreased the compressive ve flexural strength. However, strength losses could be eliminated by SM in some cases. This effect indicates that SM behaves like a polypropylene fiber. SM reduces the strength losses by slowing down the crack development in the microstructure. In addition, it has been observed that the curing process is effective in mixtures with SM. It is also thought that SM retains some of the water during mixing, contributing to internal curing. In this way, the 28-day compressive strength of the mixtures with SM increased more. It has also been observed that very high compressive strengths, such as 70 MPa can be obtained in SM mixtures.
In the study conducted by Özcan and Koç, the compressive strength of concretes decreased as the AEA ratio increased. Increasing porosity explains the decrease in concretes' compressive strength [49]. Compressive strength and concrete porosity are inversely connected, with strength increasing as porosity decreases [50]. Because the bubbles have taken the place of the concrete, this is the case. The bubbles cannot endure the pressure, which increases in severity over time [51]. According to research, the air-void structure's properties significantly impact the compressive strength of concrete [52]. Many studies in the literature show that the compressive and flexural strengths of concretes are increased by using polypropylene fiber [53][54][55]. Polypropylene fiber has a low elastic modulus, which can delay the creation and propagation of microcracks during the early stages of hardening, reducing the number of crack sources [56]. In this study, since the elasticity modules of the SMs used were lower, they prevented crack development at an early age. Crack development, which was prevented at an early age, improved the mechanical properties at later ages. At 28 and 90 days, using 0.20 percent, polypropylene fiber resulted in a slight increase in compressive strength [57]. It has been stated that polypropylene fibers' chemical and physical properties also develop strength [58]. The great fineness and varied length of fibers in staple PP fibers, which form a network that functions as a bridge and prevents the microcrack from propagating further, may explain this behavior of fiber-reinforced concrete. However, because of decreased workability and improper mixing, when the PP fiber concentration is higher, the fiber is placed non-uniformly in concrete. As a result, these fiber masses are collected to form relatively weaker places that behave as voids, making them more vulnerable to cracking and decreasing compressive strength [59]. In this study, it was determined that the optimum SM rate was 0.3 %. The use of 0.3 % SM in mixtures can reduce crack development. However, when the SM ratio was 0.8 % and 1 %, the compressive strengths decreased because weak points increased. Fig. 11a shows that an R 2 value of 0.83 was determined between the 7-day compressive and flexural strengths. On the 28th day, a value of 0.80 was noted. The mixtures' compressive and flexural strengths were found to be highly correlated. The flexural strength of the mixtures grows along with their compressive strength. Fig. 11b illustrates the strong association between porosity and compressive strength (R 2 = 0.79). It appears that mixtures with good compressive strength have little apparent porosity. Dry bulk density and compressive strength were found to have an R 2 coefficient of 0.70. The compressive strength of the mixes grows along with their dry bulk density. Particularly, combinations with compressive strengths of 75 MPa and above have dry bulk densities of approximately 2400 kg/m 3 .
As seen in Fig. 12, a reduction in porosity results in an increase in compressive strength. If the apparent porosity is below 3.5 %, compressive strengths of 70 MPa and above were obtained in the mixtures. However, in some mixtures, it is seen that the compressive strength decreases as the flow diameters increase. The reason for this is the use of AEA admixture in the mixes. On the other hand, as AEA increases the mixes' flow diameter, their compressive strength diminishes. As a result of AEA, the matrix becomes more porous and produces more air bubbles.

Transport properties
The water penetration depths of the mixtures without AEA vary between 2.8 and 3.6 mm (Fig. 13a). If the SM is 1 % in the mixtures without AEA, the depth of water penetration decreased by about 1.3 times compared to the reference mix. Since the SM seals the connections between the capillary gaps, the depth of water pene-tration is decreased. However, in the combinations where the SM ratio was 0.8 %, the water penetration depth was found to be 3.6 mm. The water penetration depths decreased by 3.5 % compared to the reference combination when using a low SM fraction (0.3 % or 0.5 %). Fig. 13b shows that the water penetration depths increase when the AEA ratio is 0.35 %. The reference mixture with 3.5 mm had the lowest water penetration depth. Compared to the reference mixture, the mixture incorporating 0.8 % SM had a 32.5 % deeper water penetration depth. The water penetration depth closest to the reference mix was observed in the mixes with 0.3 % SM content. In the mixtures using 0.35 % AEA, the depth of water penetration generally increases as the SM volume increases. The converting of SM to microfibre structure during mixing and nonhomogeneous mixing are crucial aspects of this process.
As seen in Fig. 13c, when the AEA content is 0.70 %, the depth of water penetration varies between 3.9 and 6.6 mm. In the case of using 0.3 % SM, the water penetration depth decreased negligibly. However, if the SM was 1 %, the water penetration depth increased 1.6 times than the reference mixture. This ratio was obtained 1.2 times in mixtures with 0.5 % SM. The SM ratio is 0.8 % or 1 % in the mixtures using AEA increases the water penetration depths considerably. Since AEA affects the viscosity of the mixes, the self-compacting property may decrease. As a result, the depth of water penetration is increased by combinations with high SM concentration and AEA. Creating a blocking feature when SM material is used sparingly can lower the depth of water penetration.
The relationship between water penetration depth and porosity was found to have an R2 value of 0.79, as shown in Fig. 14. The  expected decrease in water penetration depths occurs when the mixes' porosity diminishes. The combinations with an apparent porosity value of less than 4 %, in particular, have a water penetration depth of 3.5 mm or less. The correlation between compressive strength and water penetration depth was found to have a substantially lower R2 value (0.66). However, water penetration depths decrease as the mixes' compressive strengths rise. Mixtures with a 28-day compressive strength of more than 75 MPa have a water penetration depth of less than 3.5 mm. The findings showed an inverse link between strength and sorptivity, as Kanellopoulos et al. noted [60]. Capillarity is accepted as a chemical resistance index in concrete technology. Because it is directly related to the resistance of materials to the penetration of aggressive chemicals [61]. The porosity and absorbency coefficient of cement-based composite materials raise with the increase in the pore volume [62]. In the investigation done by Elango et al., it was found that sorptivity rose as the mixtures' porosity rose [63]. Foam concretes with various foam contents were created by Bayraktar et al. The depths of the combinations' water penetration rise as foam concentration in foam concrete increases [64]. Usman Rashid showed that concretes' sorptivity decreased using polypropylene fiber and steel fiber [65]. Similar findings were also determined in the study by Akid et al. [66]. Studies in the literature show that fibers reduce sorptivity to a certain ratio [67]. The inclusion of fibers up to a specific volume percentage decreased the sorptivity, according to Ramezanianpour et al., but above that range, the sorptivity coefficient value raised again [67]. In this study, SM generally reduces the depth of water penetration at low ratios (0.3 % or 0.5 %). The water penetration depth rises when used at a 1 % ratio, though.

Compressive strength
In Fig. 15a, the compressive strengths of the mixtures are exposed to temperatures up to 800°C after cooling in the air. The compressive strength of the mixes diminishes as the airentraining additive ratio rises, as seen in the figure. After 600°C, the strength loss in mixtures utilizing air-entraining admixtures primarily accelerated. Particularly, after 800°C, the compressive strength of the 0.70 % AEA mixture declined around five times less than that of the reference mixture. This mixture was more susceptible to the effects of elevated temperatures since its 28-day com-pressive strength was likewise low. The mixes chilled in the air at 800 oC have compressive strengths ranging from 17.8 to 45.2 MPa. Since the air-entraining admixtures reduce the 28-day compressive strength, relatively lower strength values were determined after elevated temperatures (600 and 800°C).
In Fig. 15b, the compressive strengths of mixtures exposed to different temperatures are given after water cooling. The compressive strengths of the mixtures cooled with water vary between 13.6 and 88.3 MPa. Water cooling at low temperatures such as 100-150°C did not damage the mixtures much. In the mixtures cooled in water, strength losses occurred from 400°C. After 400°C in air-entraining mixtures, water cooling reduced their compressive strength below 50 MPa. After 800°C, the mixture containing 0.70 % AEA had an 85 % lower compressive strength than the reference mixture. The rise in the air-entraining admixture ratio in the water-cooled mixtures reduced the compressive strength. However, using AEA in the cooling water application at 150-200°C reduced the strength loss. Water cooling applied to the mixtures decreased their compressive strength more than air cooling-thermal shocks resulting from water cooling cause more damage to the ITZ between paste and paste-aggregate. As a result, the compressive strength decreases more in water-cooling.
The compressive strength of the mixes reduces after elevated temperatures, as seen in Fig. 16a, as the SM ratio rises. The mixtures containing 1 % SM produced compressive strengths of 50 MPa and higher up to 200°C and 30 MPa and up to 600°C. When the mixture with 1 % SM was cooled in the air after 800°C, the compressive strength decreased approximately-three times compared to the reference mix. This ratio was found to be 1.5 times in mixtures without SM. Compressive strength increased after the 150°C effect in mixtures without SM. This effect can be explained by the elevated temperature activating the nonhydrated cement particles. In this process, although 100°C was ineffective, 150°C helped to increase the strength. Since the slow cooling process did not damage the microstructure, the strength increased. In Fig. 16b, strength increases up to 150°C were observed in water-cooled mixtures. However, the compressive strength of 400°C and later decreased considerably with the effect of water cooling. The rise in the SM rate reduced the compressive strength of the water-cooled mixtures. This effect shows that SM effectively reduces thermal shocks after 400°C. When compared to the reference combination, the compressive strength of the mix with 1 % SM after 800 oC fell roughly-four times. Due to its low elastic modulus, the SM is unable to withstand the stresses caused by the heat effect.

Flexural strength
The flexural strengths of the combinations cooled by air after elevated temperatures range between 1.1 and 9.1 MPa, as shown in Fig. 17a. The flexural strengths typically decrease as the AEA ratio rises. The flexural strength of the mixture with 0.7 % AEA decreased approximately-nine times after 800°C. If the AEA ratio was 0.35 %, the flexural strength values increased relatively after 150 and 200°C temperatures. The vapor pressure formed in the microstructure at low temperatures is thought to contribute to internal curing. Since the rise in the AEA rate increased the porosity, the flexural strengths decreased. In addition, the flexural strength decreases as more micro-cracks occur with the thermal effect at temperatures 400°C and beyond. As shown in Fig. 17b, the flexural strengths of the water-cooled mixtures decreased below 9 MPa. Relatively higher flexural strength was obtained in water-cooled mixtures at 150 and 200 oC temperatures. Since the internal vapor pressure formed at 150 and 200°C temperatures is not very high, it increases the compressive strength. However, increasing steam pressure from 400°C increased the microcracks in the microstructure and considerably reduced the flexural strengths. Since the water cooling process creates more thermal shock after 400°C, the flexural strengths are decreased more than air cooling. Flexural strengths can rise by up to 200°C with an increase in the SM ratio, as shown in Fig. 18a. When cooled in the air after 100°C, it has been found that the mixture containing 1 % SM has characteristics similar to those of the reference mixture. After 800°C, the flexural strength of the mix containing 1 % SM declined almost 4.5 times less than that of the reference mixture. The flexural strength of the mixtures exposed to 800°C fell below 3 MPa. Although the mixtures cooled slowly, most damage occurred at 600 and 800°C. Fig. 18b shows the effect of SM on water-cooled mixtures. The flexural strength is raised by using SM in watercooled mixtures at temperatures of 100, 150, and 200°C. In particular, the flexural strength of more than 5 MPa was obtained up to 200°C in mixtures with 1 % SM. In mixtures, the use of SM was not effective in compressive strength but more effective in flexural strength. However, this positive effect also disappears at temperatures of 400°C and beyond. Since SM behaves like a polypropylene fiber, it increases flexural strength rather than compressive strength. SM melted relatively at elevated temperatures and increased its flexural strength by filling the air bubbles caused by AEA. In addition, the transformation of SM into a fibrous structure when melted could increase the flexural strengths after elevated temperatures. SM has determined that thermal shocks after elevated temperature are more effective in flexural strength. Although SM filled the gaps caused by AEA, it could not positively affect the compressive strength as it was soft and had low strength. However, after melting, the fibrous structure of SM improves its flexural strength.
Not many studies in the literature show the effect of AEA on mortars and concrete exposed to elevated temperatures. Akca and Özyurt Zihnioglu investigated the effect of AEA in highperformance concrete. The addition of AEA to the specimens reduced the loss in residual strength, although the results were unstable after 300°C for thick specimens [68]. In a study conducted by Seçer, AEA was utilized to entrain air into concrete in volume ratios of 4 %, 6 %, and 8 %, with the concretes being exposed to temperatures of 300°C, 500°C, and 700°C. According to the findings, the loss in the strengths of concretes exposed to elevated  temperatures decreased as the air content of the concrete increased [68]. Pliya et al. investigated the behavior of highstrength concrete at elevated temperatures (150, 300, 450, and 600°C) using polypropylene and steel fibers. The amount of polypropylene fibers with a 6 mm length was estimated to be between 1 and 2 kg/m 3 . The concrete samples were subjected to four heating and cooling cycles. The results showed that samples containing 1 kg/m 3 polypropylene fiber had higher residual compressive and splitting tensile strength results when heated to 300°C. However, polypropylene fibers did not affect the remaining compressive strength after reaching 600°C [69]. Using a 15 mm long polypropylene fiber, Ding et al. examined the ultimate load, residual compressive strength, flexural toughness, fracture mechanism, and energy of self-compacting high-performance concrete maintained at 300, 600, and 900°C for 3 h. The results demonstrated that while PP fibers had no significant effect on the material's mechanical properties, they significantly reduced the number of surface cracks [70]. Yermak et al. investigated the behavior of steel and polypropylene fiber-reinforced concrete at elevated temperatures. Polypropylene fibers can now shield steel fiber from visual fire damage by shield [71]. The thermal behavior of polypropylene fiber-reinforced concrete at elevated temperatures was the subject of Rudnik and Drzymaa's research. Temperatures of 100, 200, 300, 400, 500, and 600°C were applied to the samples. The researchers found that polypropylene fibers had no impact on the thermal stability of concrete samples. After being analyzed, the compressive strength values showed behavior similar to that of the samples before elevated temperature. This circumstance is consistent with the conclusions of the prior investigations. Compressive strength is unaffected by polypropylene fibers [72]. When polypropylene fibers melt at 170°C, they generate channels, which allow gases to escape; the pore pressure falls, and the degree of concrete damage decreases [73]. The melting of fibers allows water to be drained from the first centimeters of concrete, resulting in a significant rise in the permeability of the exposed concrete [74]. The cement matrix absorbs melted polypropylene as it approaches the melting temperature of fibers [74]. On the other hand, Fibers expand by 10 % when they melt, resulting in the formation of cracks and an increase in concrete permeability [75]. In this study, SM realized these behaviours  exhibited by polypropylene fiber. However, the fact that SM has a lower modulus of elasticity compared to polypropylene fiber has increased the strength losses. In addition, although SM closes the gaps belonging to the AEA when it melts, the strength losses increased because the gaps formed when it melted were larger. In the literature, increases in compressive strength have also been observed in concretes exposed to elevated temperatures. Abid et al. studied the high-temperature resistance of RPCs up to 900°C. By 120°C, compressive strength began to decline, but at 300°C, all types of RPC showed a partial recovery [76]. The initial decline in compressive strength is caused by the interaction of stress and internal vapor pressure build-up [77]. The expansion of water between the CASAH gel layers also reduced the adhesive forces [78]. The rise in Van der Walls forces (surface forces) due to the removal of free water was primarily responsible for the strength recovery at 300°C [79]. The porosity of something like concrete directly influences strength when free water is decreased [80]. When subjected to elevated temperatures, cementitious materials lose substantial mechanical strength [81]. It has been observed that concrete loses compressive strength mostly between the temperatures of 400°C and 800°C, with significant strength loss occurring between 600°C and 800°C [82]. When hydration products like Ca(OH) 2 and CASAH decay at elevated temperatures, the average pore size of cement paste grows, which causes a loss in compressive strength [83,84]. According to Qi Zhang's research, the structure of CASAH remains steady as the temperature rises from 105 to 400°C. When the temperature rises above 500°C, certain gels change into crystalline particles, and the content of capillary voids expands dramatically [85,86]. In this study, some increases were observed in the strength of cement-based composites exposed to temperatures of 100, 150 and 200°C. Since the permeability of high-strength cementitious composites is quite low, the efficiency of water curing application decreases. Therefore, internal curing method is recommended for high-strength cementitious composites. Hydration cannot be completed due to low permeability. However, the particles that cannot be hydrated by the effect of elevated temperature participate in hydration and increase strength. Fig. 19a shows that as the temperature applied to the mixtures and the AEA content rise, so make the mass losses. As a result of the applied temperature up to 200°C, the weight losses in the mixtures are negligible. However, after 800°C temperature applica-tion, the mass losses of the mixtures vary between 8.01 and 16.31 %. At elevated temperatures, H 2 O in hydration products such as CSH and CH decomposes and evaporates. As shown in Fig. 19b, the mass losses increased comparatively after the mixes were heated to an elevated temperature. Particularly, the range of mass losses for mixes heated to 800°C is 13.01-21.53 %. It was observed that the mass losses mainly increased at 600 and 800°C. The mass losses increase with the increasing AEA rate in the water-cooling mix. The reason for the high mass loss in water-cooled mixtures is the breakage of parts due to thermal shock. Mass losses were significant since the compressive strength of the mixes frequently decreased as the AEA rate raised.

Mass loss
As seen in Fig. 20a, similar properties were obtained up to 200°C in air-cooled samples. Similar results were observed in mixtures exposed to 400°C when the SM ratio was 0.8 %. However, using 1 % SM increased mass losses at 400°C and above temperatures. While the mass losses of the mixtures exposed to 100°C vary between 0.59 and 2.09 %, the weight losses of the mixtures exposed to 800°C vary between 11.09 and 14.35 %. More weight losses occurred in the mixtures with 1 % SM. As seen in Fig. 20b, water cooling applied to the mixtures increased the mass losses. The rise in SM rate up to 200°C did not affect the mass loss much. However, if the SM ratio was 1 % in the mixtures exposed to 800°C, the mass loss increased compared to the reference mix. In water cooling in mixtures exposed to 600°C, the mass loss exceeds 10 %. In mixtures with SM, the higher temperature than the usage rate affects the weight loss. As the temperature increases, the SM suffers more damage. In addition, SM could not greatly reduce the thermal shock effect of water cooling. Because in mixtures with SM, more mass loss occurred with the impact of water cooling. The reason for the mass loss is the breaking of parts due to the effect of thermal shock. SM polypropylene fiber has a low modulus of elasticity, so it has not prevented part breakage.

The variance of compressive strength
Changes in compressive strength were determined by comparison with 28-day compressive strength. The compressive strengths of the mixtures without air-entraining admixture increased up to 200°C (Fig. 21a). This effect was also observed in mixtures with 0.35 % and 0.70 % AEA. The compressive strength of the mixtures with 0.35 % AEA increased by 16.8 %-22.8 %, depending on the temperature. Losses occurred in the compressive strength of the mixtures at 400°C and above temperatures. As the AEA rate rises, the losses in the compressive strength of the mixes increase. As Fig. 19. Effect of AEA on mass loss after elevated temperature. seen in Fig. 21b, it was observed that the compressive strength raised (up to 200°C)    ature, unhydrated cement particles can react. However, at 400°C and above temperatures, the crack network formed in the microstructure reduces the compressive strength.
As seen in Fig. 22, the compressive strength reduces as the SM rate rises in both air and water cooling. The compressive strength of the mixture with 1 % SM cooled in the air decreased by 54.4 % Fig. 23. General view of mixtures exposed to 600°C temperature.  ( Fig. 22a). This ratio was observed as 63.2 % in the mixtures cooled with water (Fig. 22b). When SM melts, it can fill some of the voids caused by AEA, but this is not enough. Therefore, as the SM rate rises, the compressive strength reduces. When SM melts, it creates larger voids and irregular voids than AEA, reducing its compressive strength.

Visual analysis
The damage in the mixtures was more visible at 600 and 800°C. Fig. 23a shows no significant damage occurred when the mixtures exposed to 600°C were cooled in air. It was observed that only small parts were broken off from the corners. In Fig. 23b, it was observed that the mixtures were damaged in water cooling. Espe-   cially in mixtures with low 28-day compressive strength, more damage occurred. As seen in Fig. 24a, spills occurred only at the corners of the mixtures cooled in air. In the samples cooled in water, both spills at the corners and map cracks appeared on the surface-the thermal shock created by water cooling after elevated temperature increases visual damage (Fig. 24b).
As seen in Fig. 24c, map cracks were observed in some mixtures. It was determined that the map cracks were mostly formed in the casting direction. In addition, it is seen that map cracks are more apparent in mixtures with AEA. Map cracks are relatively less in mixtures with high compressive strength.
3.5.6. Macrostructure after elevated temperature Fig. 25 shows the macroscopic view of mixtures with AEA and SM exposed to 150°C. The figure shows the regional filamentous structure of the SM. It has been determined that SM is mostly in the form of fiber clusters in the paste. After melting, the remaining threads of SM made bridging behavior like fibers. However, the bridging behavior was very weak due to the SM's very low modulus of elasticity. Fig. 26 shows the macro structures of the mixtures exposed to 800 oC. It is seen that SM forms square and elliptical structures on the paste surface after melting. Numerous cracks were observed on the surface where the SM came into contact with the paste. These figures show that the SM fills the ventilation slots in the mask after it melts.
3.5.7. Microstructure after elevated temperature Fig. 27 shows that the reference mixture (AEA:0%-SM:0%) has a dense microstructure at 25°C. Needle CSH gels and hexagonal CH plates were detected in SEM images. Since no pozzolan was used in the mixtures, CH plates were frequently observed in the hydration products.
SEM images of the mixture with an AEA content of 0.7 % and SM ratio of 1 % are given in Fig. 28. In SEM images, fiber-like threads are seen in some regions. These threads were formed on the cutting surfaces during the sizing of the SM. Furthermore, some washable masks release many microfibers during the washing process, which are classed as polluting elements [87,88]. In this study, contact of SMs with mixing water and exposure to a basic environment transformed them into microfiber. In addition, the rapid  mixing process in the mixer has transformed most of the SMs into microfiber. SM threads seen in SEM images are defined as microfibers. If SM was used at a low ratio (%0.3), these threads acted like fibers, increasing the flexural strength of the blends relatively. In addition, spherical air bubbles were observed with the effect of the air-entraining additive. SEM images show dense needle CSH gels. Hexagonal CH plates are also seen between the CSH gels. Fig. 29 shows 1500 and 2500 magnification SEM images of the reference mixture exposed to 150°C temperature. As seen in the figure, it was found that a dense tobermorite structure (CSH) was formed within the paste. In addition, hexagonal CH plates were also seen. It was observed that the CSH gels preserved their existence despite the temperature of 150°C. A denser CSH structure was formed as the non-hydrated cement particles hydrated, thanks to the 150°C temperature. This effect resulted in increased strength in mixtures exposed to 150°C temperature.
SEM images of mixtures with SM and AEA at 150°C are given in Fig. 30. The filamentous form of SM was determined in SEM pictures. It was also observed that SM had good adherence to paste. It has been determined that CSH gels and CH plates are frequently formed from hydration products. Since the water/cement rate of the mixtures was low, capillary voids were not observed much. It is known that the voids formed are mostly spherical, and they are formed due to the air-entraining admixture.
SEM image of the reference mixture cooled with water after 800°C temperature is given in Fig. 31. It is observed that the water cooling effect damages the hydration products in the microstructure. It has been determined that the microstructure generally resembles a spongy structure with the effect of elevated temperature and shock cooling. In addition, many crack formations in the microstructure were detected.  The width of the formed air ducts is about 10 lm. In addition, many air bubbles were observed with the effect of the airentraining additive. The air ducts formed by the SM balance the internal hydrostatic pressure caused by the elevated temperature effect. However, if SM is used at a very high rate, high strength losses occur because the porosity increases too much. Crack development in air-cooled mixtures is relatively less than in watercooled mixtures. Square-shaped structures were observed in SEM images as in stereo microscope images. These symmetrical shapes show the molten form of SM. The size of the square-shaped molten SMs ranges from 600 to 700 lm. It indicates that SM behaves like micro aggregate after melting. However, the low strength and modulus of elasticity of SM negatively affected the mechanical properties. Such patterns were obtained on the paste surface because the ventilation slots on the mask are square or elliptical.

Conclusion
If the SM content is more than 0.5 %, the workability of the mixtures decreases. However, the air-entraining additive improved the workability in the mixtures using 1 % SM. As the air-entraining additive content increased, the workability of the reference mixture increased. The flow times of the mixtures generally increase as the SM content increases. As SM increases internal friction, like polypropylene fiber, workability decreases. This is especially observed at high SM content (0.8 % or 1 %). For workability, optimum AEA content of 0.35 % and SM content of 0.3 % is appropriate.
The combinations' apparent porosity and water absorption both rise when AEA level does. The porosity values increased due to AEA increasing the quantity of air bubbles in the mixture. Water absorption and porosity ratings typically rise as SM content does. The mix's air content rises as SMs often make it less workable. Low SM (0.3 % or 0.5 %) and high AEA (0.7 %) content both reduced porosity and water absorption. By limiting the connection between the voids in mixtures with adequate workability, SMs also decreased porosity. The mixtures' dry bulk densities dropped as AEA and SM concentrations rose.
The increase in SM content in the mixes without AEA raised the 28-day flexural and compressive strengths. At high AEA content (0.7 %), the use of SM generally increases the 28-day flexural strengths. The microfiber behavior can explain the increase in flexural strength of SM. Similar results were observed for compressive strength. Although the modulus of elasticity of SM is low, it can still slow the propagation of cracks. This effect was more pronounced in mixtures with high porosity.
Increasing the SM ratio in mixtures without AEA reduces the depth of water penetration; however, as the SM content rises in the mixtures using AEA, the water penetration depth of the mixes increases. Since AEA affects the viscosity of the mixtures, the selfcompacting property decreases. As a result, SMs did not distribute homogeneously, increasing the water penetration depth of the mixtures. SM should be used in low proportions in mixtures with AEA or low workability.
After 400 oC, the mixtures' mechanical qualities primarily started to deteriorate. After being exposed to elevated temperatures, the compressive strength of the combinations reduced as the AEA level rose. The compressive strength of the mixes exposed to elevated temperatures also reduced as the SM content rose. Water cooling after the elevated temperature in SM mixtures negatively affected the flexural strengths. Such an effect was observed as the shock cooling reduced the adherence between the SM and the paste. While the increase in SM content is not very effective in mass loss, the mass loss rises as the AEA content rises. It was observed that SM melted and filled some voids in the mixture. However, as the SM and AEA volume rises, the compressive strength reduces considerably. More than 50 % strength loss has occurred in mixtures with 1 % SM depending on water and air cooling conditions. SEM and stereomicroscope examinations revealed that the SM was partially converted to microfiber. It has also been determined that microfibers have bridging properties. It was seen in the SEM images that SM melted and filled some voids in the paste. In addition, CSH and CH structures were observed in the mixtures due to the high cement content. CSH gels (tobermorite) were also determined in the mixtures exposed to 150°C temperature. A spongy microstructure and a network of cracks were observed in the mixtures exposed to 800°C.
As a result, it can be substituted for SM polypropylene fiber at low rates to create high-strength cement-based composites. It can be used in the AEA at modest rates to further lessen the detrimental effects of elevated temperatures.