Introduction

Concrete has an excellent fire resistance behaviour compared to other building materials. It gained good fire resistance from the aggregate and cement components [1]. Attenuation against radiation and strength against elevated temperatures are essential in constructing nuclear structures [2]. Concrete with high strength and high density fulfilled these requirements. Heavyweight concrete has been employed in constructing nuclear structures because of its good strength and attenuation ability [3]. However, external heating from the reactor itself will increase the temperature of the reactor shield due to the neutron flux and gamma radiation. Although most nuclear reactors are subjected to temperatures below 100 °C during their service life, they may rise to 350 °C in accident conditions. According to the recommendation of ACI, the temperature in concrete shield should be limited to 65 °C; other international organizations allow temperatures up to 90 °C [4]. According to Bertero and Polivka (1972), cyclic heat treatment of temperature between 20 and 150 °C is more harmful to concrete [5].

Heavyweight aggregate applied in the construction of the nuclear buildings can be a natural mineral with a high density like hematite, magnetite, limonite, and barite, or artificial like steel punching and iron shot. Other minerals with less density, like hydrous iron ore or serpentine, can be used to produce heavyweight concrete with less density [6, 7]. Due to that, mechanical, physical, and chemical properties need to be tested after exposure to high temperatures. The loss in mass and strength in heavyweight concrete (HWC) exposed to elevated temperature is a well-known phenomenon, due to the loss of physically and chemically bounded water from the hydration products, evaporation of free water from concrete pores, and formation of thermal cracks [8,9,10].

The temperature level, heating rate, thermal cycling, and temperature duration affect concrete at elevated temperatures. As a result, two different kinds of damaged have occurred in concrete, deterioration in mechanical properties, and spalling of concrete [11]. The damage mechanism in concrete could pass through four different phases, transformation phase in the cement paste, transformation phase in the aggregate, thermal incompatibility between the cement paste and the aggregate, and spalling of the concrete [12].

Deterioration of mechanical properties involves the physiochemical changes in cement paste, physiochemical changes in the aggregates, and the thermal compatibility between the aggregate and the cement paste [13]. The heat affects the properties of concrete at two levels the structural level and the microstructural levels. At the structural level, damage in concrete may occur due to thermal stress. At the microstructure level, large stress at the interface between the aggregate and the cement paste may cause microcracking due to the mismatch of thermal strains in the cement paste and aggregates [11].

Different changes occurred due to the decomposition of cement paste at various temperatures. Loss of physically bound water has happened between 20 and 200 °C, due to the movement through the capillary system and the reduction in cohesive forces with the dehydration of the C-S-H gel. The dehydration of ettringite has occurred between 80 and 150 °C, and gypsum [CaSO4.2H2O] decomposed at 150 to 170 °C. At 350 °C, some siliceous aggregates will break up, and evaporation of bound water will continue up to 374 °C. The first phase of C-S-H decomposition occurs between 460 and 540 °C [Ca (OH)2 → CaO + H2O]. At 573 °C, the quartz phase in aggregate will change from β → α. Between 600 and 800 °C, the second phase of C-S-H decomposition and the formation of β-C2S happened. At 840 °C, the dolomite decomposed, while calcite decomposed between 930 and 960 °C [CaCO3 → CaO + CO2]. At 1050 °C, basalt melts, while the total decomposition or melting of concrete will occur at 1300 °C [14].

Mechanical properties of concretes with high-density aggregates were described by Gencel and Ouda (2015). Gencel et al. have found a minor effect of hematite aggregate on the basic properties of concrete. At normal temperature, the compressive strength of high-strength concrete with hematite aggregate does not differ from the ordinary concrete [15]. Ouda (2015) investigated concrete mixes of barite, magnetite, and goethite aggregates. Therefore, he found that absorption of goethite aggregates was several times higher than barite and magnetite aggregates by 13 and 10%. Concrete made with fine magnetite aggregate showed higher physical and mechanical properties than barite and goethite concrete [16]. Sakr and Elhakim investigated the physical, mechanical, and radiation properties of the heavyweight concrete by testing the gravel, barite, and ilmenite concrete. Results showed that illuminate concrete is more resistant to elevated temperature than gravel or barite concrete. Therefore, it had the highest density, compressive, and tensile strength and the lowest absorption percentage than gravel or barite concrete [17].

This paper introduced the behaviour of heavyweight concrete at elevated temperatures by testing the properties of magnetite-based concrete and comparing it with basalt- and quartz-based concrete. The main goal was to investigate the mechanical, physical, and thermal stability properties of heavyweight concrete after exposure to elevated temperatures. The compressive strength, flexural strength, spalling observation, and porosity were measured and compared for the magnetite-, basalt-, and quartz-based concrete after being subjected to 20, 150, 300, 500, and 800 °C.

Materials and methods

Materials

Three types of concrete mixes were prepared in this research and investigated after being exposed to high temperatures. For radiation shielding structures, the cement amount is relatively high above 350 kg m−3, according to the requirements of AS3582.1 [18]. Therefore, Portland cement of the type CEM I 52,5 SR was used for the different concrete mixtures. In addition, coarse and fine aggregates were used in different mixture proportions. The size of the coarse aggregate ranged between 4 and 22 mm while the size of the fine aggregate was less than 4 mm. Figure 1 represents the grain size distribution of aggregates in each concrete series. Three different aggregates (magnetite, basalt, and quartz) were used to prepare the three concrete mixtures.

Fig. 1
figure 1

Grain size distribution for aggregates in each concrete mix

Full size image

Magnetite aggregates were used to get the heavyweight concrete type. In contrast, quartz and basalt aggregates were used to get the normal weight concrete. A 100% of magnetite aggregates with different size proportions (0–4, 4–8, and 8–16 mm) were applied to produce heavyweight concrete. About 22 and 36% coarse quartz aggregate and 42% of the fine sand aggregate mixture were used to get the normal weight concrete. The grain size of the coarse quartz aggregate was 4–8 and 8–16 mm while the grain size of the siliceous sand was less than 4 mm. Basalt-based concrete was designed by substituting 50% of quartz aggregate with basalt aggregates, so the aggregates combination is 50% basalt (22 mm) in addition to 7 and 10% of quartz (4–8 and 8–16 mm), and 33% siliceous sand.

The density of the magnetite aggregate was 4.8 [g cm−3] [19]. Quartz and basalt were used to produce the other two normal-weight concrete mixes. The specific density of gravel and basalt is 2.64 and 2.83 [g cm−3], respectively. Figure 2 illustrates the different aggregate types used in different concrete mixes and their specific mass density.

Fig. 2
figure 2

Different aggregate types and their mass density

Full size image

Mix design

Three concrete mixes were produced; cement, aggregate, water, and admixtures are the main ingredients. Magnetite, basalt with quartz and sand, and quartz with sand are the main aggregates for each mix, respectively. Table 1 illustrates the three different concrete mixes and their constituents.

Table 1 Magnetite-, basalt-, and quartz-based concrete mixes proportions
Full size table

Sample preparation and experiments

The experimental program comprised concrete mixing and casting, heat loading at 20, 150, 300, 500, and 800 °C temperature steps, spalling observation, mass loss compressive strength, flexural strength, porosity, and scanning electron microscopy (SEM) test. Figure 3 illustrates the different experimental program steps.

Fig. 3
figure 3

Experimental program

Full size image

A total of 30 samples were cast from each concrete series for different tests. Each concrete series comprises two different shapes, 150 × 150x150 mm cubes and 70 × 70x250 mm prisms. Fifteen cubes were cast for the heat loading, compressive strength, mass loss, and spalling tests. Moreover, 15 prisms were cast for the heat loading, flexural strength, and porosity tests. Three specimens were tested for each temperature step (\({\text{total }}\;{\text{no}}.\;{\text{of}}\;{\text{ samples}} = 3{ }\;{\text{samples}} \times 2{ }\;{\text{shapes}} \times 5{ }\;{\text{temperature }}\;{\text{steps}} \times 3{ }\;{\text{mixes }} = { }90\;{\text{ samples}}).{ }\)

Samples were cast in the laboratory using the standard mould. After 24 h, samples were demoulded and cured in water for 7 days. The samples were stored in the laboratory at 20 °C temperature and 35% relative humidity until 28 days and then tested.

The heating system was applied for the 28-day specimens by heating the specimens up to the maximum temperature for two hours. The maximum temperatures were: 20, 150, 300, 500, and 800 ℃, and the heating rate was 1 ℃ min−1. In the present experiments, the heating curve was similar to the standard heating curve proposed in EN 1991-1-2 and the standard ISO 834 fire curves applicable to structural engineering. Samples were placed in an iron cage in the middle of the furnace. Figure 4a shows the placement of the specimen in the furnace, and Fig. 4b. illustrates the standard ISO 834 fire curves.

Fig. 4
figure 4

a Placement of specimen in the furnace. b the standard ISO 834 fire curves, where T: Temperature [℃] and t: Time [min]

Full size image

After the heat loading, samples were cooled in the air, left for the next day, and tested. Observation of spalling and mass loss was investigated on the cube specimens before the compressive strength test. Afterwards, investigations of the microstructure were carryout using scanning electron microscopy (SEM).

The half prisms were used to measure the porosity. Samples were submerged in water for several days to measure the porosity until full saturation or constant weight was achieved. The difference between moist and dry samples represents the pore volume. The porosity was the ratio of the pore volume to the total dry volume.

Results and discussions

Spalling observation

As a result of the heat loading and thermal stress, some visible physical changes were observed. Removal of some pieces of concrete from the surface is known as spalling [20]. These visible changes include cracks, colour changes, and the explosion of samples [21]. As shown in Table 2, changes had occurred to the concrete samples after being subjected to elevated temperatures.

Table 2 Observation of spalling
Full size table

At 150 °C, no visible changes occurred in the three concrete mix types. At 150 °C, only evaporation of the unbounded water occurred. At 300 °C, there were no visible changes on the heavyweight magnetite-based concrete; only minor spalling spots were observed on the basalt-based concrete surface. However, the size of these spots increased on the quartz-based concrete samples.

The effect of elevated temperatures is obvious at 500 °C for basalt- and quartz-based concrete. In contrast, heavyweight magnetite-based concrete still shows good thermal stability. At 500 °C, cracks were observed in basalt-based concrete, while cracks and explosive spalling were observed in the quartz-based concrete. We could see that the effect of 500 °C elevated temperature is relatively higher in quartz-based concrete than in basalt-based concrete. At 800 °C, explosive spalling occurred to the heavyweight magnetite concrete with a 20% mass loss.

Subsequently, cracks, colour changing, and minor mass losses are observed in basalt- and quartz-based concrete. Magnetite has a higher thermal conductivity than basalt and quartz aggregates. The thermal conductivity of magnetite aggregate is in the range of 6–23 W mK−1. In contrast, the normal aggregate is 1.3–2.8 W mK−1 at ambient temperature [22]. Concrete with high thermal conductivity has better thermal stability because of the better heat transfer that reduces the thermal stresses between the inner and outer surface of the concrete structure [23, 24].

The most probable factors that can increase the risk of the explosion of concrete are dense aggregate, high heating rate, thermal stresses, and thermal gradient in addition to the moisture gradient, low permeability, and closed pores containing water within the cement matrix [25]. The thermal expansion of heavyweight aggregates reduces the divergence between the aggregates and cement matrix, which is significant for the propagation of microcracks within the concrete structure [26]. High binder content can also increase the risk of explosive spalling, so high cement content would require more water; otherwise, adding supplementary cementitious material such as silica fume can decline the permeability of the cement matrix [21]

Spalling in concrete occurs during the heat loading test due to the pore pressure build-up and the thermal stresses [23]. The pressure build-up is higher in the heavyweight aggregate due to its low permeability compared to normal-weight concrete [21, 24]. Therefore, when temperatures exceed 300 °C, the internal pressure in the heavyweight concrete grows rapidly compared to the normal weight concrete [27].

Mass loss

The difference in the mass before and after heat loading is the mass loss (Fig. 4). The mass loss occurred due to evaporation, dehydration, cement paste decomposition, and concrete surface spalling [28]. These changes will develop cracks, surface spalling, or even concrete explosions with different aggregates that show a distinct trend in the mass loss [29]. The main factors affecting the mass loss in heavyweight concrete are thermal stability, composition, and water absorption.

Another factor is the high w/c ratio (above 0.4), and the high ratio of heavyweight aggregate is applied in nuclear concrete; both increase the viscosity modifying admixture (VMA). In addition, it increases water absorption and the mass loss of the heavyweight concrete [30].

As shown in Fig. 5, basalt-based concrete shows a higher mass loss trend at different elevated temperatures than magnetite- and quartz-based concrete. After exposure to a temperature of 105 °C, all the evadable water is expelled from the hardened cement paste. At 150 °C and 300 °C, the mass loss is relatively low (> 3%) in the three concrete mixes. The reason is the evaporation of bounded water and the cement paste's dehydration, which occurred up to 400 °C [31].

Fig. 5
figure 5

The mass loss of the magnetite, basalt, and quartz concrete after being subjected to elevated temperatures

Full size image

A dramatic increase in the mass loss is observed in basalt-based concrete, about 7%; it is double compared to quartz- and magnetite-based concrete. The reason is the high amount of basalt in the concrete mixture, which increments the cement paste's dehydration and decomposition.

The difference in the mass loss in magnetite-based concrete between 300 and 500 °C is relatively low. Therefore, it occurred due to the decomposition of the cement paste and the disintegration of the (CSHs). However, at 800 °C, the mass loss increases regularly in the three mixes due to the mineralogy changes in aggregates and the disintegration of calcium silicate hydrates (CSH) in concrete.

Research conducted by Horszczaruk et al. showed that concrete with magnetite aggregate has lower mass loss than concrete with quartz and barite aggregates hence identifying the good thermal stability of magnetite aggregates [32]. Also (Ling et al.) reported that the mass loss of concrete with barite aggregates is relatively high above 300 °C, due to the high water absorption of barite aggregates [33].

Compressive strength and flexural strength results

The residual compressive strength of magnetite-, basalt-, and quartz-based concrete is shown in Fig. 6. The initial compressive strength of concrete at 28 days is relatively higher in quartz-based concrete, about 84 MPa, followed by magnetite- and basalt-based concrete, about 58 and 48 MPa, respectively (Fig. 5).

Fig. 6
figure 6

Residual compressive strength values of the magnetite-, basalt-, and quartz-based concrete after subjected to elevated temperatures

Full size image

The w/c ratio and the aggregate structure directly affect the compressive and flexural strength values. Increasing the heavyweight aggregate will reduce the compressive strength because the formation of some magnetite aggregates contains weak planes that cannot withstand the compressive stress. At low w/c, the internal structure of the cement matrix for quartz aggregates was improved, which demonstrated that heavyweight magnetite aggregates have a negative impact on the compressive strength [34]. Similar findings have been reported by Mostofinejad et al. It reported that barite-based HWC has a lower compressive strength than concrete with normal limestone aggregates at a lower w/c ratio but no significant difference at a higher w/c ratio [35].

The relative compressive strength and flexural strength of magnetite-, basalt-, and quartz-based concrete after being subjected to elevated temperature are shown in Fig. 7. At 150 °C and 300 °C, the compressive strength of the heavyweight magnetite-based concrete is declined by 15 and 25%, respectively. In contrast, the reduction in basalt- and quartz-based concrete is relatively low, about 9 and 11% for basalt-based concrete and 13 and 9% for quartz-based concrete.

Fig. 7
figure 7

Relative compressive strength values of the magnetite-, basalt-, and quartz-based concrete after being subjected to elevated temperatures

Full size image

Therefore, the high absorption level of the heavyweight magnetite aggregate will elongate the evaporation phase in magnetite-based concrete compared to basalt- and quartz-based concrete. However, at 500 °C, there was no significant reduction in the compressive strength of magnetite- and basalt-based concrete. Therefore, it only declined by 15%. On the other hand, the reduction in the compressive strength of the quartz-based concrete is relatively high by 60% compared to the initial compressive strength. Finally, at 800 °C, magnetite-based concrete shows good thermal stability. Nevertheless, despite spalling from the surface, the reduction in compressive strength is still low, about 20% less than the initial value. Accordingly, the reduction in basalt- and quartz-based concrete is relatively high, about 87 and 76%, compared to the initial values.

The different aggregates applied are the main factors affecting the compressive strength of concrete exposed to elevated temperatures. Concrete with low aggregate/cement ratios shows a more significant reduction in strength than concrete with high aggregate/cement ratios [36]. The initial flexural strength is quite similar for concrete with magnetite and quartz aggregate 10 MPa and relatively low for basalt-based concrete, about 6.5 MPa. Increasing the amount of the heavyweight aggregates reduces the cement matrix's ability to resist deformation by the crystallization microstructure of the heavyweight aggregate that contains weak planes (Fig. 8).

Fig. 8
figure 8

Residual flexural strength values of the magnetite, basalt, and quartz concrete after being subjected to elevated temperatures

Full size image

At 150 °C, the flexural strength declined by 3 and 10% in magnetite- and basalt-based concrete, while improved by 6% in quartz concrete. At 300 °C, the flexural strength of magnetite-based concrete improved by 6%, while it reduced in quartz- and basalt-based concrete by 5 and 15%. At 500 °C, the flexural strength of magnetite-, basalt-, and quartz-based concrete declined by 25, 22, and 55%, respectively. Therefore at 800 °C, it declined by 55, 75, and 60%, respectively (Fig. 9).

Fig. 9
figure 9

Relative flexural strength values of the magnetite, basalt, and quartz concrete after being subjected to elevated temperatures

Full size image

Porosity

Gradual heating caused a change in the pore structure of concrete. The porosity of hardened cement paste changes when exposed to elevated temperatures. The pore structure of the concrete affected the physical and mechanical properties of concrete. Samples were moisture and measured until they reached the constant mass. Therefore, the void volume ratio to the total volume represented the porosity. The volume of the voids was calculated for concrete mixtures after being subjected to elevated temperatures. Porosity was calculated using Harmathy's (1970) formula, which has calculated the true density (d), the bulk density (ρ), and the porosity (ε) at elevated temperatures, using Eq. 1 [37]

$$\varepsilon = \frac{d - \rho }{d}$$
(1)

According to the literature, the true density increased from 2.55 g cm−3 at 105 °C to 2.90 g cm−3 at 500 °C and 3.18 g cm−3 at 900 °C. On the other hand, the bulk density decreased from 1.42 g cm−3 at 105 °C to 1.32 g cm−3 at 500 °C and 1.30 g cm−3 at 900 °C. So, the corresponding values for the porosity were about 0.44, 0.54, and 0,59 at 105 °C, 500 °C, and 900 °C, respectively. The porosity increased by 34% from 105 °C to 900 °C [38]. Rostasy et al. (1980) reported that gradual heating caused changes in the pore structure of the cement mortar. The pore size distribution is also affected as the total pore volume increases. His research reported that the volume of the pores increased by 4.2, 54.9, and 84.5% after heating at 300, 600, and 900 °C, respectively [39].

Figure 10 represents changes in the porosity values with elevated temperatures. The porosity of the different concrete mixes increased gradually with elevated temperatures. According to the experiments, the porosity of basalt concrete has a higher value than magnetite and quartz concrete. Quartz concrete has the lowest porosity value compared to magnetite and basalt concrete. At 300 °C, the difference in the porosity between magnetite- and basalt-based concrete and quartz-based concrete is relatively high. There is no obvious difference in the porosity of magnetite- and basalt-based concrete at 300 °C and 500 °C, respectively, unless there was a fast increment in the porosity of the quartz concrete at 500 °C (Fig. 10).

Fig. 10
figure 10

Porosity [%] values of magnetite, basalt, and quartz concrete after subjected to elevated temperatures

Full size image

As shown in Fig. 11, the porosity of magnetite, basalt, and quartz concrete is duplicated at 150 °C due to the evaporation of the unbounded water. However, up to 300 °C, a gradual increase in the porosity was detected in the three concrete mixes, due to the evaporation of the bounded water at the earliest stage of the cement paste dehydration. The increment in the porosity is faster in magnetite and basalt concrete than the quartz concrete, due to the high absorption ratio of basalt and magnetite aggregates.

Fig. 11
figure 11

Relative porosity % values of the magnetite, basalt, and quartz concrete after subjected to elevated temperatures

Full size image

At 500 °C, the porosity of magnetite and basalt concrete remained stable. In contrast, in quartz concrete, the evaporation and dehydration of the cement paste still took place. Finally, at 800 °C, there was no significant change in the porosity of quartz and magnetite concrete, while the increment was still kept in basalt concrete (Fig. 11).

There is an indirect relationship between growth in the porosity and reduction in the strength of concrete mixes (Fig. 12).

Fig. 12
figure 12

Changing the porosity Vs the compressive strength of magnetite-, basalt-, and quartz-based concrete after being subjected to elevated temperature

Full size image

The relationship proved that increasing the porosity would decrease the strength of the concrete due to changing the material's microstructure. Therefore, a comparison was necessary to define how far the growth in the porosity will affect the compressive and flexural strength of concrete after the heat treatment test.

Up to 300 °C, magnetite-based and basalt-based concretes' porosity was duplicated gradually by 3.4 and 3.7 times to the initial porosity; alternatively, their compressive strength decreased by 25 and 12%, respectively. Subsequently, the growth of the porosity of the quartz-based concrete was low compared to magnetite and basalt concrete unless there was no significant change in its compressive strength (Fig. 12).

The flexural strength showed different behaviour, slight improvement by 8% and 1% occurred in magnetite- and quartz-based concrete. In contrast, a 17% dropping rate was observed in basalt concrete (Fig. 13). Evaporation of the unbounded water is the main reason, furthermore the dissimilarity of the grain size distribution, especially in basalt-based concrete.

Fig. 13
figure 13

Changing the porosity Vs the flexural strength of magnetite, basalt, and quartz concrete after being subjected to elevated temperature

Full size image

At 500 °C, there was no significant change in the porosity and the compressive and flexural strength of the magnetite- and basalt-based concrete. Correspondingly, the compressive strength improved by 8% for magnetite-based concrete and decreased by 4% for basalt-based concrete. The apparent change in the porosity of the quartz-based concrete at 500 °C caused a fast drop in the compressive and flexural strength by more than 50%. At 800, the growth in the porosity and the compressive strength of magnetite-based concrete were constant. In contrast, the compressive strength in quartz-based concrete declined unless the porosity was stable due to C-S-H decomposition. There was no significant effect in the flexural strength of the quartz-based concrete at 800 °C, while it declined by more than 40% in the magnetite-based concrete.

Scanning electron microscopy (SEM)

5000 × observation scales are obtained and investigated for microstructural analysis. The three different minerals aggregates (magnetite, basalt, and quartz) are investigated at the origin state and after the heat treatment procedure (Fig. 14). At 20 ℃, there is no change in the microstructure of magnetite, basalt, and quartz aggregates. Nevertheless, the porous structure of the three minerals is different (medium, high, and low) in the magnetite, basalt, and quartz, respectively.

Fig. 14
figure 14

Different microstructures of different aggregates after the heat treatment procedure

Full size image

After the heat treatment at 800 ℃, changes have occurred. Therefore, thermal cracks are observed in magnetite aggregate because of the aggregate thermal expansion. Whereas there is no significant change in the microstructure of basalt aggregate, only microcracks appear between the aggregate's particles, which is essentially due to the evaporation of fixed water from the aggregate. Alternatively, a change in the microstructure was observed in quartz aggregate due to aggregate disintegration.

Table 3 represents the SEM image of magnetite-, basalt-, and quartz-based concrete before and after the heat treatment at 20, 500, and 800 ℃. At 20℃, results showed the good state of the C-S-H gel in magnetite-, basalt-, and quartz-based concrete, while the observed porosity in magnetite-based concrete is higher than in basalt- and quartz-based concrete. At 500 ℃, presence of microcracks between the cement paste and the transition zone between the paste and the aggregate (ITZ) is observed in magnetite-based concrete as well as void because of the dehydration process. Correspondingly, the microcracks and a few porosity changes are observed in basalt-based concrete. In contrast, microcrack appearance, first stage of C-S-H decomposition, and dehydration of the paste are observed in quartz-based concrete. At 800 ℃, the C-S-H gel in magnetite-based concrete becomes porous, due to the decomposition of paste and the presence of microcracks and voids. Correspondingly, the lack of cohesion in the paste and the appearance of microcracks are observed in basalt- and quartz-based concrete, due to the second phase of C-S-H decomposition. Moreover, the width of microcracks is larger, and the degree of C-S-H decomposition is higher in quartz-based compared to basalt-based concrete.

Table 3 The SEM images of magnetite-, basalt-, and quartz-based concrete at 20, 500, and 800 ℃
Full size table

Conclusions

The thermal spalling, mass loss, changes in porosity, compressive, and flexural strength properties were investigated for the magnetite-, basalt-, and quartz-based concrete. The three mixes contain different types and amounts of aggregate and different grain size proportions; the water–cement ratio is the same for the three mixes. The results can be summarized on the following points.

  • Magnetite-based concrete has better heat/fire resistance than basalt- and quartz-based concrete, thus, due to the good thermal conductivity and the high water absorption ratio of the magnetite aggregate. Accordingly, there was no significant change up to 500 °C on the spalling observation. However, explosion spalling occurred at 800 °C, and about 20% of the mass was removed from the surface. Nevertheless, basalt-based concrete has a better fire/heat resistance up to 500 °C compared to quartz-based concrete. Therefore, due to its high water absorption ratio and low thermal conductivity, basalt concrete can elongate the evaporation phase up to 500 ℃. As a result, at 800 °C extra cracks, colour changing, and few mass losses were observed on the surface.

    The fire/heat resistance of the quartz-based concrete has lower fire/heat resistance compared to basalt- and magnetite-based concrete. Therefore, there was no significant change up to 300 °C, cracks and explosive spalling were observed at 500 ℃, and extra cracks, colour changes, and explosive spalling were observed at 800 °C.

  • The maximum change in the porosity is achieved at 300 °C, 800 ℃, and 500 ℃ in magnetite-, basalt-, and quartz-based concrete, respectively. Therefore, it reflects the thermal stability of magnetite-based concrete above 300 ℃ and the continuous evaporation of unbounded water in basalt aggregate up to 800℃ in basalt-based concrete.

  • The mass loss of basalt-based concrete is relatively higher than the magnetite and quartz aggregates. However, the effect of the mass loss on strength properties of basalt-based concrete is low compared to quartz-based concrete because of the long evaporation phase in cement paste and basalt particles. Nevertheless, magnetite-based concrete has a lower mass-loss percentage than basalt and quartz-based concrete due to its high thermal stability.

  • There is no significant change in the compressive strength of magnetite-based concrete up to 800 ℃; however, it only lost 25% of its initial value at 300 ℃. On the other hand, it is relatively stable up to 500℃ in basalt-based concrete by a 15% reduction rate; then, it rapidly declined by 76% at 800 ℃. Correspondingly, the reduction in compressive strength of the quartz-based concrete is relatively low, up to 300 ℃, with an 11% reduction rate. Afterwards, the declination rate increased rapidly by 60 and 87%, at 500 and 800 ℃, respectively.

  • After the heat treatment procedure, results showed a gradual reduction in magnetite-, basalt-, and quartz-based concrete flexural strength. Up to 300 ℃, the reduction in the flexural strength of magnetite-based concrete is low compared to basalt- and quartz-based concrete. At 500 ℃, the flexural strength of magnetite-, basalt-, and quartz-based concrete declined by 24, 21, and 55%, respectively, while the declination rate was 54, 75, and 58 at 800 ℃, respectively.

  • There is a direct correlation between changes in porosity and the reduction in compressive and flexural strength values. As well as the porosity increased, the compressive strength decreased, unless in magnetite- and basalt-based concrete. Therefore, the effect of changes in the porosity on the compressive strength of magnetite-based concrete is relatively low, above 300 ℃ and between 300 and 500℃ in basalt-based concrete.

  • The SEM observations and analysis obtained the appearance of microcracks, voids, and degradation of C-S-H at 500 and 800℃. At 500 ℃, presences of microcracks and voids are observed in magnetite-based concrete, while the decomposition of C-S-H, the appearance of microcracks, and voids are observed at 800℃. Correspondingly, at 500 ℃, the presence of the microcracks and a few porosity changes are observed in basalt-based concrete. In contrast, the appearance of microcracks, dehydrated and first stage of C-S-H, is observed in quartz-based concrete. At 800 ℃, the lack of cohesion in the paste and the appearance of microcracks are observed in basalt- and quartz-based concrete. Moreover, the width of microcracks is relatively large, and the degree of C-S-H decomposition is high in quartz-based concrete.