Abstract

The mechanical behavior of basaltic volcanic ash (VA) and fly ash (FA) as a cement replacement under elevated temperatures is mainly investigated in the current study. For this, cement content has been partially replaced with and without the presence of electric arc furnace slag (S). Four distinct ranges of temperatures (200°C, 400°C, 600°C, and 800°C) were selected, and the modified mixes were subjected to these gradually elevated temperatures. Samples were cured and cooled by using air- and water-cooling techniques. Test results were established by examining the sample weights and compressive strength before and after the exposure of each temperature level. The pozzolanic potential of volcanic ash and fly ash samples was identified using the strength activity index. After analyzing the test results, it has been found that there is a significant effect on the compressive strength of mortar mixes at the early ages of its strength gain. However, at the later ages of curing, samples modified with volcanic and fly ash with the presence of electric arc furnace slag have shown a better performance than control mix in terms of strength and weight loss.

1. Introduction

It is not rare to expose concrete to elevated temperatures especially when concrete is used in furnaces, chimneys, nuclear reactors, and subjected to fire hazards. Concrete is considered as one of the good fire resistant materials, but in the past, it has shown a severe damage and even collapses under high temperatures especially when the concrete strength is high [1]. Exposure of concrete to fire and its temperature history have been a topic of zeal interest for researchers over the years. The constituent ingredient of concrete such as cement paste undergoes a sequence of decomposition reactions which in most of the cases are irreversible. Harmathy [2] conducted thermoanalytical tests and proposed thermogravimetric techniques to gauge concrete temperatures under accidental fire. The behavior of concrete changes under fire at high temperatures, and its mechanical properties such as compressive strength, Poisson’s ratio, and stiffness are greatly affected, which may result in complete failure or collapse of structural systems [310]. A significant change in the physical and chemical composition of concrete has been witnessed by Khoury et al., when concrete was subjected to high temperatures [11]. He found that the chemically bound water releases from calcium silicate hydrates when the binding paste of concrete faces temperatures higher than 110°C. Under elevated temperatures, dehydrated calcium silicate and expansion of aggregates cause internal microcracks within the concrete structure especially when the temperature rises up to 300°C [12]. At 450°C, calcium hydroxide starts decomposing, and crystalline structures of quartz change from form a to form b. When the temperature rises beyond 530°C, Ca(OH)2 disassociates causing intense shrinkage cracking on the concrete surface and results an increase in pore pressure which affects the permeability of concrete [4, 5, 13, 14]. Decomposition of C-S-H gel has been observed when the temperature rises up to 600°C. Concrete is crumbled at 800°C and beyond this temperature concretely remains no more capable of contributing any strength and loses its integrity and durability [11]. Effect of high temperatures could be seen in the form of concrete spalling and cracking [3, 7, 9, 1517].

Structural damages occur when mortar or concrete is exposed to fire for a long time [18]. The thermal properties of concrete change at elevated temperatures due to variation in moisture content and decomposition of different hydration products. The heat resistance of concrete depends on various factors such as heating rate, cooling rate, loading rate, duration of exposure, and moisture content [19]. A well-hydrated cement paste mainly consists of calcium silicate hydrate, calcium hydroxide, and calcium sulfate aluminate hydrate. A saturated paste also contains a large amount of free water, capillary water, and gel water (chemically bonded water). The damaging effect of Ca(OH)2 can be reduced by using different pozzolans, such as fly ash, slag, silica fume, clay, and volcanic ash, as a replacement of cement in concrete. SiO2 present in these pozzolans reacts with Ca(OH)2, and a by-product of hydration reaction forms calcium silicate hydrates. As a result, the amount of Ca(OH)2 is reduced and C-S-H is increased, which improves the performance at elevated temperatures [20, 21].

Many studies have been done to evaluate the performance of pozzolan-incorporated mortar and concrete at elevated temperatures. Sarshar and Khoury [22] found that a paste containing 30% fly ash as a replacement of cement exhibited a residual compressive strength of almost double as compared to control samples at 400°C and 600°C. Yazici et al. [23] studied the compressive strength of mortar incorporating fly ash, silica fume, and pumice at elevated temperatures. They found that pumice mortar exhibited the lowest strength, while silica fume showed highest compressive strength values at all elevated temperatures. Poon et al. [24] evaluated the performance of concrete having metakaolin, which is used as a replacement of cement at elevated temperatures. The experimental results revealed that metakaolin concrete showed reduced strength as compared to unheated samples, when it was exposed to an elevated temperatures (400°C, 600°C, and 800°C) at all percentages of replacement. There was a little gain of strength at 200°C. They further demonstrated that higher percentage replacement gave much lesser strength and high durability losses. Ibrahim et al. [25] studied the fire resistance of high-strength mortar having high-volume fly ash and nanosilica. From the test results, they found that fly ash- and silica fume-based mortar exhibit almost equal or better residual strength at 400°C and 700°C as compared to control samples. The better performance of fly ash- and silica fume-based mortar at elevated temperatures was due to stable microstructure and decreased pore size distribution. Nadeem et al. [26] evaluated the compressive strength and durability performance of fly ash and metakaolin mortar at elevated temperatures. The result showed that fly ash mix (20%) showed better performance as compared to all other mixes. They further demonstrated that 400°C is a critical temperature for strength and durability performance of mortar when exposed to elevated temperatures. Khandaker and Hossain [27] evaluated the performance of volcanic ash incorporating high-strength concrete (0%–20%) exposed to elevated temperatures. They found that at 200°C, the volcanic ash concrete showed more strength result as compared to control samples. They further demonstrated that concrete containing higher percentage replacement of volcanic ash gives more residual strength as compared to control sample when the temperature was increased from 200°C to 800°C. Volcanic ash concrete also showed less spalling and cracking as compared to control samples.

There are vast reserves of basaltic volcanic ash available in the western region of the Kingdom of Saudi Arabia. Western Saudi Arabia contains a huge number of lava and cinder cones in the area known as Harrat. Volcanic basaltic ash is available inside or along the periphery of these cones [2831]. Many researchers have evaluated the pozzolanic potential of basaltic volcanic ash when it is used as a replacement of cement in mortar and concrete. They found that up to 20% of substitution of cement with volcanic ash gives better strength and durability properties at normal curing temperature [3238]. No research has yet evaluated the performance of this locally available basaltic volcanic ash in mortar and concrete under elevated temperatures. So the main focus of the present study is to evaluate the performance of basaltic volcanic ash mortar, when it is exposed to elevated temperatures (200°C, 400°C, 600°C, and 800°C), along with fly ash (FA), a well-known pozzolan, as a reference material. Also, electric arc furnace slag was used as an additive because of its very fine particle size. Different performance tests such as compressive strength and weight loss were performed on the mortar sample before and after exposure to elevated temperatures.

2. Materials and Methods

In this study, a locally available Portland type I cement manufactured by Saudi Cement Factory was used [39]. Fine aggregate, fulfilling the requirement of ASTM C109 and ISO standard, was used for the preparation of mortar samples. Fineness modulus of standard sand was 2.54. Grain size distribution of fine aggregate was calculated according to ASTM C125 [40], as shown in Table 1. Chemical and physical characteristics of cement and all other materials such as volcanic ash (VA), fly ash (FA), and electric arc furnace slag (EAFS) are mentioned in Table 2.

2.1. Volcanic Ash

Formation of a vast field of basaltic flow due to the volcanic activity occurred 25 million years ago in the western region of the Kingdom of Saudi Arabia known as Harrat [28]. These harrats are spread over an area of 900,000 km. Hundreds of scoria cone were discovered in these harrats. These cones are fully or partially formed from scoria material. Many researchers have investigated the chemical and pozzolanic potential of these scoria materials obtained from different scoria cones. The chemical analysis result showed a high percentage of silica (35–49.5%) in all the samples, and it was also found that almost 97% of the samples showed positive pozzolanic activity [32]. In the current study, the scoria material was collected from Harrat Rahat, Jabal Khada quarry in Madinah Province, Saudi Arabia. A finely ground basaltic volcanic ash (30 µm) was provided by Supper Burkanni Block Company in Jeddah. The fine material was then passed through a sieve (#635) to get the ultrafine material.

2.2. Electric Arc Furnace Slag

Electric arc furnace slag is a by-product of the steel industry. EAF slag is generated when the steel scrape along with pig iron is fed into the electric arc furnace for the production of steel [41]. The Kingdom of Saudi Arabia uses arc furnaces for the production of steel. So a huge amount of EAF slag is produced as a by-product. The estimated annual production of EAF slag is about 350,000 tons. In recent past, the EAF slag aggregate is successfully used in highway and concrete industry [42]. Other researchers investigated the potential use of EAF slag as a cement substitute for sustainable concrete industry [43]. But they found that EAF slag possesses a slight cementitious performance and very low pozzolanic reactivity due to its crystalline nature. So in this study, locally available EAF slag aggregate was collected form SABIC steel factory. Their aggregates were subjected to different grinding processes by altering grinding time, revolution per minutes, and grinding balls and bowl material to get fine slag using Pulverisette-5 laboratory planetary mill. After optimizing the grinding process, the EAF slag aggregates were put in a grinding bowl made of tungsten along with 5 grinding balls of the same material and subjected to grinding for 30 min in three cycles, each of 10 min duration and a pause of 10 min were provided between two cycles to keep the temperature of bowl low which was necessary for efficient grinding.

2.3. Fly Ash

Fly ash is fine spherical particle produced as a by-product, when coal is burnt as a source of energy in the thermal power plant [44]. Fly ash is a pozzolanic material and hence an important application in cement and concrete industry. The estimated annual production of fly ash is 500 million tons [45]. Past research showed that use of fly ash as a partial cement replacement improves the fresh properties, enhances the higher age strength, improves the durability properties, and adds to the sustainability of cement and concrete industry. In the current study, a commercially available fly ash (class F) is used as a reference pozzolanic material.

3. Mix Proportion, Sample Preparation, and Curing

Along with the control mix (C), two binary and two ternary mortar mixtures were prepared by replacing cement at 20% with different material and their combinations. Previous studies suggest that 20% replacement of cement by VA and FA provides the optimum performance [46, 47]. Therefore in the present study at the initial stage, binary mixes were prepared by substituting cement contents by 20% with VA and BFA. At the second stage, ternary blends were developed by replacing cement with BVA and FA at same amount (i.e., 20%), and 10% of EAFS was used as an additive. The detail of the different mix proportions is presented in Table 3.

All the test specimens were prepared as per mix design of ASTM C 109 [48] as given in (1). Mixing of all the test specimens was carefully carried out by the standard mixing procedure as described by ASTM C-305 [49]. To study the effect of BVA and FA replacement with cement on compressive strength, a set of three 50 × 50 × 50 mm cubes were prepared for each testing day (i.e., 7, 28, and 91 days of curing). In the first stage, the specimens were cast corresponding to 7, 28, and 91 days of curing at room temperature to test the compressive strength, and in the second stage, 91-day-cured samples were exposed to different elevated temperatures (200, 400, 600, and 800°C) and then cooled down by exposing the samples to air and water:

4. Fire Exposure and Cooling Methods

Prior to heat exposure, the specimens were first dried at 100 ± 5°C in an oven for 24 hours to remove the capillary water to reduce the risk of spalling and then placed inside a furnace, whose internal temperature was increased from room temperature to 200, 400, 600, or 800°C. The temperature was applied at an incremental rate of 3.3°C per minute from the room temperature of 22°C, and maximum temperature was maintained for 2 hours to attain thermal equilibrium at the center of the specimens [50]. Figure 1 shows the time-temperature curve of the furnace for heat exposure. After 2 hours, the furnace was turned off, and specimens were allowed to cool down inside the furnace for 2 hours (Figure 2), and after that, the specimens were taken out of the furnace. Half of the specimens were allowed to cool slowly for 24 hours in air at ambient temperature, and other half was cooled in water to study the effect of the cooling method on compressive strength. After the water cooling, the specimens were placed again in the oven at 100 ± 5°C for 24 hours to surface dry the specimens. Finally, all the specimens were carefully sealed with the plastic wrap to prevent rehydration until testing under compression.

5. Testing Procedures and Methods

5.1. Compressive Strength Test

All the mortar cubes were tested under compression using a universal testing machine. During compression test, a loading rate of 1 mm/min was maintained as specified by ASTM C 109. For each parameter of the study (i.e., type of specimen, heat exposure, and cooling method), the average value of three cube tests’ results was reported.

5.2. Measurement of Weight Loss

The weight of each sample before and after the exposure of an elevated temperature was measured. The weight loss was then calculated as the ratio of original weight before exposure to heating to the residual weight after getting exposed to specified elevated temperature.

5.3. Particle Size Analysis

Particle size analysis of all the samples (Figure 3) in powder form was carried out using Microtrac S3500. The particle size analysis curves of cement, volcanic ash, fly ash, and EAF slag are given in Figure 4. The curves show that all the substitute materials are finer than cement. The EAFS is the finest material, and because of its high fineness, it is used as an additive to increase the internal packing of mortar mix. The d10, d50, and d90 sizes were also calculated for all the material used in this study and are given in Table 4.

6. Result and Discussion

6.1. Chemical Analysis and Strength Activity Index

The pozzolanic potential of basaltic volcanic ash was evaluated according to ASTM C618. The chemical analysis result shows that the summation of SiO2, Al2O3, and Fe2O3 is 84.65%, which satisfies the minimum requirement (70%) set by the standard. Additionally, the amount of SO3 is 0.10 and LOI is 2.71; both these values fall within the limits set by ASTM C618. The chemical analysis result shows that the strength activity index is used to evaluate the reactivity of mineral admixture with cement. According to ASTM C618 [50], the strength activity index value for each mix should be at least 75%, which means that the mortar containing SCMs must have compressive strength equal to 75% of control at 7 and 28 days. The strength activity index was calculated according to ASTM C311 [51]. Table 5 shows the compressive strength results of all mixes subjected to standard curing (water cured at 20°C) along with their strength activity index values. The results show that the all binary and ternary mixes containing SCMs (FA, VA, and EAFS as an additive with FA and VA) attained compressive strength more than 75% of the reference specimen (C) at all ages.

6.2. Weight Loss

Figures 5(a) and 5(b) show the effect of elevated temperatures on the weight loss of C, FA20, VA20, F20S10, and V20S10 under air and water cooling, respectively. The residual weight (%) along the vertical axis in Figure 5 is a measure of percentage weight retained compared to the weight of the specimen at room temperature. Residual weights of the specimens were evaluated at four distinct temperature ranges, that is, 200°C, 400°C, 600°C, and 800°C. For all of the tested specimens, with the increase in temperature, the weight loss is increased for air-cooled specimens as shown in Figure 5. In case of air-cooled specimens, from 20 to 200°C, the FA20 specimen has shown the highest reduction among all, whereas the control sample and V20S10 have shown the minimum value. F20S10 and VA20 exhibited a higher weight loss than FA20 and C but lesser than FA20. The weight loss in this temperature range is mainly attributed due to moisture evaporation from the sample surface to the atmosphere. The supporting evidence for this argument is lesser or almost no weight loss of tested specimens when cooled under water as shown in Figure 5(b). When the temperature is raised from 200 to 400°C, the reduction in weight increases significantly, especially for the water-cooled specimens. In both of the cooling conditions, the control specimens have shown the maximum weight retained followed by V20S10, VA20, F20S10, and FA20. At this temperature range, a similar amount of reduction in weight of specimens was observed under both air-cooling and water-cooling conditions: for example, the percentage retained in case of air-cooling conditions (at 400°C) was found to be 97.7, 97, 96, 95.7, and 95.4 compared to water-cooling values of 98%, 97.1%, 96.6%, 96%, and 95.7% for C, V20S10, VA20, F20S10, and FA20 samples, respectively. The highest lost in the weight was observed in the case of the F20S10 specimen which is 3.1% (from 98.7% to 95.6%) for air-cooled conditions and 3.4% (from 99.4 to 96%) for water-cooling conditions. This temperature range has shown the maximum weight loss which could be due to the further evaporation of residual moisture content retained at a temperature level of 200°C. For both of the cooling cases, the percentage reduction in weight loss was less at a temperature range of 400°C to 600°C compared to 200°C to 400°C as shown in Figures 5(a) and 5(b). Almost no reduction or very slight reduction in specimen’s weight was observed when the temperature was further raised from 600°C to 800°C. The weight loss in the aforementioned temperature ranges is mainly attributed due to evaporation of free water and binding water through the structure of C-S-H and subsequent decomposition of Ca(OH)2. It changed the stiffness and mechanical properties of the substance resulting in lower compressive strength values. FA20 has shown the maximum loss of water which could be due to higher retention of water content in the presence of fly ash.

6.3. Residual Compressive Strength at Elevated Temperature under Air and Water Cooling

Figures 6(a) and 6(b) show the compressive strength of the mortar specimens with different pozzolanic materials under elevated temperatures for the specimens cooled under air and water, respectively. To have a better tracing of pozzolanic effect under fire, results are provided in the form of compressive strength and residual compressive strength ratio. The vertical axis on the left shows the residual compressive strength values, while on the right side, the residual strength ratio is given which is calculated by dividing the retained compressive strength of the specimens with its 28-day compressive strength prior to heat exposure. All the specimens were subjected to four distinct ranges of temperatures starting from 200°C to 400°C, 600°C, and 800°C, respectively.

By increasing the temperature to 200°C, a significant increase in compressive strength of all the specimens was observed. This increase in strength was highest in the case of the control sample and was lowest for the V20S10 specimen. This high-strength increment in the case of control samples is attributed mainly due to the loss of free water which has increased the friction between the failure planes causing higher strength values. The other reason could be the hydration and chemical bonding process of nonreactive cementitious particles at elevated temperature level. Figure 6(a) also shows a considerable increase in the residual compressive strength ratio of VA20. It could be due to its higher finesse values that have triggered a pozzolanic reaction of the fine ash particles. At the same level of heating, 200°C, no significant increase in the compressive strength of the mortar specimens was observed when the samples were cooled under water as shown in Figure 6(b). In the case of the control sample, a small reduction in strength was noticed. This could be due to slower catalyzed hydration and chemical binding that has occurred due to the rapid cooling provided by the water. Additionally, a sudden temperature change between surface and core of the specimen by water cooling leads to the microcracks, which is also one of the reasons for the reduction in compressive strength [52]. Figure 7 shows the microcracks on the surface of the mortar cube after water cooling. A very small increase in the residual strength ratio of F20V10 has been observed, and this increase was lesser than the increase in strength when samples were cooled in the air after heating at 200°C.

When the samples were heated up to 400°C, a reduction in strength of the mortar specimens was observed as shown in Figures 6(a) and 6(b). In Figure 6(a), it can be observed that this reduction in strength was more when the samples were cooled in air. The main reason is that the hydrated cement contains a large amount of free Ca(OH)2, which loses its water at a temperature level above 400°C. If the cement is subjected to moist condition (e.g., water cooling) after exposing to fire, it rehydrated to calcium hydroxide, which increases its strength as compared to the air-cooled specimen after exposing to fire [53].

In the case of control samples, the compressive strength became equal to the value corresponding to the no heating which shows that gain in strength up to 200°C is almost equal to the loss in strength at 400°C. The similar phenomenon was observed in the cases of FA20, VA20, and VA20S10. When the samples were cooled under water after heating at 400°C, a slight decrease in compressive strength was observed (Figure 6(b)). However at 400°C, both VA20S10 and FA20S10 have shown better strength than the control sample irrespective of the method of cooling as the addition of slag reduces the dehydration of cement and forms the new hydrated product with slag. Figure 8 shows the failure modes of air-cooled VA20 and VA20S10 specimens tested under compression after exposure to 400°C. In Figure 8(a) for the specimen VA20 without slag, spalling is quite evident due to the dehydration of Ca(OH)2, whereas for the specimen VA20S10, only crushing (Figure 8(b)) was observed without spalling as addition of the slag resists the dehydration of mortar and forms new hydrated product.

At 600°C, all of the mortar specimens have lost significant compressive strength as shown in Figures 6(a) and 6(b). The retained strength or residual strength ratio was reduced from 100% to 74%, 73%, 68%, 65%, and 62% for C, FA20, VA20, FA20S10, and VA20S10, respectively, under air cooling as shown in Figure 6(a). All these samples have shown a higher reduction in compressive strength when cooled using water as shown in Figure 6(b). For instance a residual strength of 52%, 56%, 54%, 47%, and 49% was noticed for C, FA20, VA20, FA20S10, and VA20S10, respectively. At this temperature level, it is difficult to differentiate between the effectiveness of different pozzolanic materials on the mortar compressive strength due to the poor level of stability attained after heating the specimens at the elevated temperature of 600°C. This reduction in strength is mainly attributed due to decomposition of C-S-H and dehydration of calcium hydroxide to free lime. Due to these changes, the volume of these cementitious product increases and cohesion within the mortar matrix decreases resulting in hairline cracks within the mortar. This phenomenon significantly reduces the overall strength of the mortar. Although the residual strength ratio of FA20S10 and VA20S10 was the lowest among all, still these two mortar mixes have shown the highest strength for both air- and water-cooling methods. This higher strength of aforementioned pozzolanic mixes is due to their ability to replace C-H partly and giving a higher resistance to degradation at elevated temperatures.

Exposure of heating temperature from 600 to 800°C has caused the further strength degradation. Although a rapid decrease in strength was observed when the samples were cooled under air, air-cooled samples have shown comparatively high strength than water-cooled samples as shown in Figures 6(a) and 6(b). However, in both of the cooling methods, pozzolana mortar specimens have shown higher strength than the control samples, and VA20S10 has given the maximum compressive strength. In case of air cooling, the residual compressive strength varies from 28 to 21 MPa for VA20S10 and control samples, respectively, whereas these corresponding values for water cooling are 24 to 10 MPa. The highest residual strength reduction was observed in the case of control samples which have reduced from 74 to 34% for air cooling and from 52% to 16% for water cooling as shown in Figures 6(a) and 6(b). For example, a newly formed compound at 600°C undergoes recrystallization, and a rapid rate of expansion and shrinkage within the mortar mass take place. From the test results, it could be seen that VA20S10 has shown a good behavior under heating compared to others. The main reason is the development of C-S-H-like gel when unhydrated slag reacts with calcium hydroxide at higher temperature (400–800°C), and this effect is also proven in a similar study incorporating the slag [54].

7. Conclusions

The result presented in this paper outlined the performance of basaltic volcanic ash when it is exposed to elevated temperatures and then cooled under two different conditions, that is, air and water. Compressive strength and weight loss were calculated based on the experimental test. The experimental result and observation made in this study led to the following conclusions:(1)The chemical analysis verified the pozzolanic potential of locally available basaltic volcanic ash. The natural pozzolans have the ability to replace cement up to 20% with no significant reduction in loss of strength at later ages.(2)Experimental results showed that the residual properties of all mortar mixes are significantly affected by cooling methodology. At identical temperature, the mortar mixes showed better residual compressive strength when cooled in air as compared to water cooling. On the other hand, the water cooling results in reduced weight loss values except for samples having FA (20%). Weight and strength loss under elevated temperatures is mainly due to evaporation of free water followed by the removal of binding water through C-S-H and subsequent decomposition of Ca(OH)2. (3)The presence of slag could not contribute well at the early ages of curing. However, by the increase of curing time, it has shown that the strength values are even higher than the control sample. Especially, the strength increment in the cases of FA20S10 and VA20S10 was mainly attributed due to the packing effect provided by the finely ground electric arc furnace slag particles and slightly due to the cementitious nature of electric arc furnace slag. Moreover, the blend of these pozzolanic materials has resulted in the formation of refractory compounds of minerals which can hold a significant strength even at elevated temperature ranges.(4)Among various blended samples, VA20S10 has shown a better behavior at elevated temperatures. It is due to the presence of ultrafine slag as an additive which has delayed the dehydration of cementations compounds by holding free water for a considerable duration under high temperatures.

Conflicts of Interest

The authors declare that they have no conflicts of interest.

Acknowledgments

This research was supported by the Deanship of Scientific Research (DSR) at King Faisal University (KFU) through its “Seventeenth Annual Research Project no. 170085.” The authors wish to express their gratitude for the financial support that has made this study possible.