Utilization of steel slag in concrete: A review on durability and microstructure analysis

1.1 Steel slag (SS)

The steel industry is well known for its substantial carbon emissions and the considerable amounts of SS it produces [22]. The manufacture of SS accounts for around 10–15% of the total amount of steel produced [23]. Lim et al. [24] reported that slags are the main byproducts formed in the manufacturing of iron and crude steel, contributing over 90% of the total byproducts. SS is a substantial industrial by-product due to the high worldwide steel production volume. SS contains oxides, including calcium oxide (CaO), silicon dioxide (SiO₂), and iron oxide (Fe₂O₃), providing cementitious properties comparable to conventional binders. Furthermore, SS has a similar chemical composition to cement and can be used as a cementitious material, which further extends its applications in construction [25]. Zhu et al. [26] also highlighted that SS seems more suitable as a secondary material for producing a high-performance geopolymer composite. Janga et al. [27] also indicated that researchers throughout the globe are slowly coming toward geopolymers as viable substitutes for conventional construction materials as environmental concerns are raised. Tian et al. [28] also indicated that the utilization of SS as supplemental cementitious materials in cement and concrete production is an efficient strategy for enhancing the environmental sustainability of concrete. The total charge passed, and the penetration depth of chloride ions in SS concrete decreased due to the denser and more impermeable microstructure of the concrete, which improved its resistance to chloride ion penetration [29]. Wang et al. [30] also indicated that concrete that includes SS can exhibit durability properties such as permeability, drying shrinkage, and carbonation resistance comparable to those achieved by reference concrete. America and Europe have high utilization rates for SS, mostly used as an aggregate for asphaltic pavement or as a filler in civil engineering. The statistics show that China produces the highest slag as compared to the other countries, as presented in Figure 1. The second highest production of slag is in Canada among the other country.

Figure 1

Production of ash and slag [31].

The manufacture of iron and steel yields four distinct forms of slags, which are named after the corresponding processes from which they are produced (Figure 2). Slags may be classified into four categories: blast furnace slag, basic oxygen furnace slag, electric arc furnace slag, and ladle furnace slag. The particular technique applied in creating the crude steel, the cooling conditions applied to the slag, and the valorization technique are some of the variables that affect the slag’s characteristics [33].

Figure 2

Iron and steel refinement process [32].

The increasing SS usage can result in improved economic growth and long-term sustainability for future generations [32]. Yi et al. [34] noted that the total quantity of slag accumulated in the storage area reaches 30 million metric tons, causing substantial environmental pollution. In a recent study, Li et al. [35] reported that worldwide crude steel output reached 1868.8 million tons (Mt) in 2019, representing a significant increase of 55.2 Mt compared to the previous year. Despite the problems presented by the COVID-19 pandemic, a total of 156.4 Mt of crude steel were produced globally by September 2020. Li et al. [35] suggested that the significant increase in worldwide SS production highlights the urgent need to efficiently manage its disposal or recycle it.

In addition to conventional landfill dumping techniques, scholars have effectively investigated several approaches for recycling SS in a wide range of sectors, including construction, chemical, and agricultural [35]. Slag is produced because of impurities, such as silica, alumina, and lime, that are found in the raw materials used in the process of creating steel, including iron ore, coke, and fluxes. This slag, which was formerly regarded as waste, has found several uses owing to its unusual qualities. SS powder has hydraulic characteristics, making the hydration process comparable to that of cement [36]. Wang et al. [30] indicated that SS provides significant advantages for sustainability and enhances certain mechanical and durability characteristics. The tensile strength (TS) consistently increased when SS was used as a replacement for fine aggregate, ranging from 0 to 20%. The greatest improvement was achieved when the substitution level reached 30%. However, TS decreased when SS was replaced at a level of 40% [37]. The rough surface texture of slag aggregates provides more confinement in comparison to the smooth surface of gravel aggregates [38].

1.2 Review significance

Generally, concrete made with SS exhibits equivalent or higher strength properties as compared to natural aggregate concrete [39]. Research indicated that SS adversely affected the workability of concrete mixtures with a high replacement ratio due to its more angular shape compared to conventional rounded aggregates [40]. However, Basheer et al. [41] reported that the insufficient understanding of concrete durability and environmental impact has resulted in substantial problems. Multiple causes contribute to the degradation of reinforced concrete structures, such as the corrosion of reinforcing bars caused by carbonation or chloride ingress, the effects of freezing and thawing, sulfate assault, alkali-aggregate reaction, and other factors. Therefore, this article provides a summary of the durability characteristics and microstructural qualities associated with the use of SS in concrete. This review is essential for advancing knowledge in this field and providing insights for improving the performance and durability of concrete buildings.

2.1 Porosity

Macro or capillary pores present in concrete are important for their mechanical characteristics since the porosity of concrete is typically inversely connected to its strength. The pores in concrete generate weaknesses in the matrix, decreasing its capacity to withstand external loads and stresses. Moreover, these holes might facilitate the entry of water, chemicals, and other harmful aspects, therefore damaging the concrete structure. Consequently, reducing porosity is essential for enhancing the strength and service life. Figure 3 shows the porosity of concrete made with steel slag aggregate (SSA) and brick aggregate (BA). The findings indicate that the porosity decreased as the percentages of SS aggregate increased. SS aggregates yield a denser concrete composition as compared to BA concrete. BA can have more porosity than SS aggregate. Therefore, concrete produced with BA would have a greater number of voids and air pockets, resulting in reduced density and strength compared to concrete generated using SSA. However, the concrete made with SSA is a denser concrete which led to enhancing the mechanical attributes, including strength, durability, and water penetration resistance. Decreased porosity further mitigates the risk of cracking and extends the durability of the concrete building.

Figure 3

Apparent porosity [42].

The denser SS aggregate (Figure 4a) has a positive effect in reducing the void spaces within the concrete, leading to dense microstructures. In contrast, BA is a more porous material (Figure 4b), which contributes to higher porosity levels when used in larger proportions. The greater porosity leads to a less dense structure, adversely affecting the strength and durability of the concrete. Increased porosity leaves concrete less resistant to water absorption, potentially resulting in problems such as freeze-thaw damage, reduced durability, and a higher likelihood of cracks. Therefore, concrete using greater quantities of BA often exhibits worse mechanical characteristics relative to that utilizing SS material.

Figure 4

(a) SS and (b) broken brick aggregate [42].

Furthermore, an increase in slag content as cementitious materials led to decreased rates of porosity and water absorption. The reduction in permeability with mineral admixture is attributed to the pozzolanic reaction, which improves the binding property of mortar. In addition, the micro-filling activity of the mineral additive plugs the voids in the aggregate, leading to compact mass and a decrease in its permeability. The pozzolanic reaction and micro-filling of voids improve the permeability of the concrete [1]. However, Qi et al. reported that permeability and void ratio depend on particle size [43]. Furthermore, although the contribution of the composite mineral additive to the enhancement of pore structure, their effectiveness is less than cement when the SS percentage is high (60%) [44]. Table 1 presents the summary of durability aspects of SS concrete.

2.2 Water absorption

Water absorption is an essential aspect of concrete durability properties because water absorption gives related information about the concrete porosity and density [50]. Water absorption of concrete (exposed to thermal variation) made with limestone aggregate (LC) and SS is presented in Figure 5. Thermal variation increased water absorption in both LC aggregate concrete and SS aggregate. The water absorption in the cement concretes, which includes SS aggregate, increased by an extent of 1.39–6.9% due to temperature fluctuations, while it was 3.3% in the concrete with LC.

Figure 5

Water absorption [45].

Shen et al. [51] revealed that as the temperature increased, the porosity increased while the pore structure experienced gradual destruction. The modifications in porosity and pore structure could be highlighted by the dehydration of water inside the concrete. Shen and Xu [52] also noted that porosity increased with increasing the temperature of concrete. The increase in porosity results in higher water absorption because the additional pores create more pathways and spaces for water to enter. However, the water absorption made with SS aggregate remained lower than that of the concrete made with aggregate, even after undergoing 60 heat cycles [45]. The decrease in the absorption properties of concrete made with SS aggregate may be ascribed to the impermeable structure of the SS aggregate in comparison to the crushed limestone (CL) aggregate. Elhadi et al. [50] found that concrete mix made with recycled aggregate (RA) shows an increase in water absorption and void volume. The increase in water absorption is attributed to the porous nature of RA. RA absorbs more water than natural aggregate, resulting in increased water absorption. Nevertheless, the use of metakaolin in concrete made using RAs resulted in a substantial decrease in voids and the ability to absorb water. However, SS aggregates possess an impermeable quality that greatly reduces their water absorption, making them an extremely durable option in situations where there are changes in temperature.

2.3 Pulse velocity

The pulse velocity of CL and SS aggregate is presented in Figure 6. The findings indicate that the pulse velocity of CL aggregate was 4.55 km·s⁻¹. In contrast, the pulse velocities of SS aggregate concrete ranged from 4.6 to 4.94 km·s⁻¹, which increased compared to the CL mix. SS aggregate produces denser concrete than CL. SS aggregates are often more dense than CL aggregates. This increased density creates a more continuous and less porous channel for ultrasonic vibrations, enabling them to travel quickly through the matrix. The ultrasonic pulse velocity depends on the maximum size of aggregate, and the ultrasonic pulse velocity of concrete improves with an increase in maximum aggregate size.

Figure 6

Pulse velocity [45].

The findings indicate that ultrasonic pulse velocity in BA concrete may range from 3.10 to 3.80 km·s⁻¹, dependent upon the maximum aggregate size [53]. The maximum size of aggregates decreases the number of interfaces and transition zones within the concrete matrix, facilitating the more efficient propagation of ultrasonic waves. The ultrasonic pulse velocity also depends on the density and morphology of the concrete aggregate, the contact between the aggregate and the paste (ITZ), and the number of voids present [54]. Saxena and Tembhurkar [55] observed that the gradation curve of SS aggregate conforms to the specifications of BIS 383. This indicates that SS aggregate shows excellent particle size distribution and packing density, which led to an increase in the ultrasonic pulse velocity. The well-graded gradation of SS results in a more compact concrete matrix with fewer voids, facilitating the effective transmission of ultrasonic waves through the matrix. Thus, the use of SS aggregate leads to increased pulse velocities in comparison to CL aggregates. Roslan et al. [56] showed that SS is more difficult to break down than steel sludge. Therefore, SS has a greater density, and well-shaped aggregates provide a more continuous channel for ultrasonic wave transmission, which leads to decreased dispersion and attenuation. An effective interface between the aggregate and paste enhances the transmission of ultrasonic waves, while a reduction in voids decreases resistance and energy loss, leading to increased pulse velocities. Therefore, the concrete made with SS shows higher pulse velocity compared to concrete made with CL.

2.4 Carbonation depth

Carbonation depth measures the degree to which carbon dioxide (CO₂) infiltrates into concrete, which results in the creation of calcium carbonate and reducing the alkalinity. Therefore, the carbonation depth is important because it has the potential to induce corrosion of the reinforcement. The carbonation depth of concrete with the substitution of SS for two different water-to-binder ratios (W/B = 0.50 and W/B = 0.35) is shown in Figure 7. It can be observed that carbonation depth decreased with decreased W/B. A greater W/B (0.50) leads to a more porous microstructure because the extra water evaporated, leaving voids. Therefore, the porous microstructure allows greater penetration of carbon dioxide, resulting in increased carbonation depth. Palankar et al. [49] suggested that properly adjusting the W/B and the quantity of sodium oxide can be helpful in achieving the desired strength in alkali-activated concretes with a substitution fly ash. Furthermore, the increasing SS percentages increased the carbonation depth. A 15% substitution of SS causes a minor increase in carbonation depth at a W/B ratio of 0.50. However, the carbonation depth is greatly increased by replacing 30% and 45% of the SS. Furthermore, with a W/B ratio of 0.35, SS has less of an impact on raising the concrete’s carbonation depth than it does at a W/B ratio of 0.50.

Figure 7

Carbonation depth of concrete exposed to an accelerated carbonation chamber [30].

Liu et al. [57] highlighted that the high level of Mg(OH)₂ in reactive MgO-based cementitious materials promotes the formation of cracks in the mortar, thus decreasing its overall strength. This deterioration facilitates the more rapid diffusion of CO₂, consequently speeding carbonation and further decreasing the material’s pH. Similarly, Liu et al. found that the carbonation depth increased with the substitution of 40% GGBS and 40% SS(SS) individually as compared to concrete made with only Portland cement [44]. The GGBS and SS increased the concrete porosity at an early age, which resulted in greater penetration of carbonation. The calcium hydroxide content decreases due to a 40% decrease in cement through the substitution of GGBS and SS. Therefore, the carbonation depth of the GGBS and SS mix increased. Furthermore, it can also be observed that the carbonation of the SS mix is much greater than the GGBS mix. Therefore, the SS contributes to higher voids or associated pore networks as compared to the GGBS, which results in a greater carbonation depth. Dhivya et al. [58] reported that the decreased carbonation rate is due to the reduced alkalinity of concrete made with SS, putting concrete mix less subjected to carbon dioxide infiltration. SS consists of calcium minerals that react with calcium hydroxide, which is formed during the hydration of cement paste, producing calcium silicates and aluminates. This reaction lowers the alkalinity of the concrete, consequently decreasing the probability of carbonation as calcium hydroxide is reduced. Moreover, the use of SS in concrete could improve its density and optimize its pore structure, resulting in further decreasing the carbonation rate.

2.5 Cracking and loss of strength due to ASR

The initial stage of ASR-induced cracking for sample concrete specimens for the two slag concretes is shown in Figure 8, showing expansion stresses ranging from 1,000 to 3,000 macrostrains. It was generally found that, regardless of the cause of the cracking in plain concrete, which is a random phenomenon, the intensity of microcracking at a given expansion was comparable for both slag replacement levels. The exception was that this cracking occurred later and penetrated the specimens less deeply for the 65% level of slag. In the 50 and 65% slag concrete, extensive microcracking had been attained at around 0.3 and 0.2% expansive strain, respectively. Under increased stress conditions, the pulse velocity obtained its asymptotic values at these strain levels. Beyond this point, further expansion had no impact on the overall cracking pattern and resulted in small alterations in pulse velocity. Due to the occurrence of critical cracking, pulse velocity values exhibited little fluctuation, whereas expansions showed a consistent increase in strain values. Given the interrelationship between cracking and pulse velocity, it can be concluded from the data on expansion, cracking, and pulse velocity variations found in research that concrete with a 65% slag replacement performed better than concrete with a 50% slag replacement in terms of mitigating the effects of ASR. Yi et al. indicated that the major causes of SS concrete’s growth were free calcium and magnesium oxides [34].

Figure 8

(a) First visible cracking at 50% slag, (b) 1000 micro-strain at 50% slag, (c) first visible cracking at 65% slag, (d) 1000 micro-strain at 65% slag, (e) 2000 micro-strain at 50% slag, (f) 2000 micro-strain at 65% slag, (g) 3000 micro-strain at 50% slag, and (h) 3000 micro-strain at 65% slag [46].

The results of the SS concrete in terms of loss in CS and FS at different expansion levels are shown in Figure 9. These results demonstrate that both concrete experienced significant reductions in FS, but the decrease in CS for the concrete mix with 50% slag was only about 10%, while the concrete containing 65% slag showed an improvement in CS of almost 25%, probably because of the slag’s increased reactivity in the existence of moisture and the resulting pore-filling. As previously stated, these findings demonstrate that a decrease in compressive capacity is not a reliable indicator of the harm that ASR has caused to concrete [59]. According to Sheen et al., the primary causes of the growth of SS concrete were free calcium and magnesium oxides [60]. As a result, the collapse of ASR-affected buildings owing to a loss in CS is improbable, but a loss in FS is still significant, which would substantially impair the concrete’s capacity to withstand the increased deflection and deformation caused by these losses. Thomas et al. [61] also indicated that the expansion of blocks incorporating fly ash was significantly less than that of comparable blocks with identical Portland cement content but without fly ash. The findings conclude that fly ash, at 25–40%, does not substantially contribute alkalis to the ASR.

Figure 9

Strength loss and expansion of (a) 50% SS and (b) 60% SS; data source [46].

2.6 Chloride ion permeability

The grade of chloride penetrability and charge passing for concretes with a water-to-binder ratio (w/b) of 0.45 is displayed in Figure 10(a). It is evident that at ages 28, 90, and 360 days, all the mix display the same penetrability grade to chloride ions. These results suggest that when the substitution rate of fine SS is not more than 30%, it does not affect the chloride ion permeability of standard concrete. The grade of chloride permeability and charge passed for concretes with w/b of 0.30 are shown in Figure 10(b). Regardless of age, the concrete from sample 30% SS displays greater chloride ion permeability than that from sample control, 10% SS, and 20% SS, in contrast to the findings for normal concrete. According to Pan et al. [62], adding 10% SS powder to self-compacting concrete instead of cement boosted the material’s mechanical qualities and strengthened its durability features, including carbonation and chloride penetration. As stated in ASTM C1202 [63], the 30% SS sample has a moderate chloride ion permeability grade at the age of 28 days compared to reference concrete, 10% SS, and 20% SS, which has low penetrability grades. The chloride ion penetrability grade of 30% SS sample is low at the ages of 90 and 360 days, in contrast to extremely low grades for sample control, 10% SS, and 20% SS. Therefore, fine SS does not affect the chloride ion penetrability of high-strength concrete when its substitution rate is less than 20%. However, it is impossible to overlook the detrimental impact of fine SS on the chloride ion penetrability of high-strength concrete when its alternative ratio approaches 30%. According to Wang and Yan [64], the detrimental impact of SS on concrete permeability is less pronounced at lower W/B.

Figure 10

Chloride ions penetration: (a) W/C 0.30 and (b) W/C 0.30; data source [47].

However, the outcome is not fully constant for fine SS. The detrimental impact of fine SS replacement at a 30% level on concrete permeability is more pronounced at lower w/b. The reason for this is that fine SS has particles that are less than 20 μm in size, giving the unhydrated particles a somewhat effective physical filling effect. The microstructures of the hardened fine SS paste and the ITZ in concrete are comparatively permeable at a W/C of 0.45. As a result, unhydrated fine SS’s physical filling action might somewhat increase the chloride ion penetrability of conventional concrete.

2.7 Acid attacks

Concrete durability is the capacity of concrete to withstand the effects of strong environmental media that endanger the regular operation of concrete components [65]. These hostile environmental media acts are sometimes described as acid attacks [66]. Early cracking and softening of OPC-based concrete, when exposed to acid, has been seen owing to the breakdown of calcium hydroxide and the creation of significant amounts of gypsum. As a consequence, prolonged contact with an aggressive acid-rich environment, such as a sewage system, causes concrete to significantly lose strength [67]. The mass loss resulting from exposure to sulfuric acid increases with increasing OPC binder concentration [68]. Figure 11 shows the concrete sample before and after the acid was immersed. It can be noted that the outer surface of the cubes submerged in H₂SO₄ has degraded more than the sample immersed in HCl.

Figure 11

Concrete sample (a) before acid, (b) H₂SO₄, and (c) HCl immersed [48].

Palankar et al. [49] noted that a concrete sample containing SS aggregates presents expansion cracks and pop-outs on the surface as a result of the expansion of the SS aggregates (Figure 12). The expansion of SS particles induces tension inside the concrete matrix and exerts stress on the surrounding concrete. This stress leads to the development of cracks once it exceeds the TS of concrete. The internal pressure from expanding aggregates can cause the concrete surface to detach, resulting in pop-outs. These are little, conical particles that separate from the surface. Furthermore, a white paste-like substance is also noted on the surface of the aggregates. Liu et al. [69] also indicated that the sulfate solution penetrates the concrete surface by capillary action, which results in the crystallization of salt on the concrete surface.

Figure 12

Deterioration concrete: (a) 0% SS, (b) 50% SS, and (c) 100% SS [49].

Scanning electron microscopy (SEM) findings (Figure 13) indicate the presence of intact coarse aggregates (EAF slag, stabilized AOD slag, and silico-calcareous). However, it exhibits cracking inside the cement paste and, in some regions, at the interfaces between the paste and EAF slag or stabilized AOD aggregates. Intergranular cracking is absent in the aggregates, which is important since such cracking could suggest internal expansive activities, such as the ASR or delayed ettringite formation. However, the absence of this form of cracking indicates that the aggregates do not contribute to expanding reactions that could put the integrity of the concrete at risk. Therefore, the findings suggest that, although cement paste exhibits cracking, the aggregates remain stable and do not increase the probability of long-term degradation through expanding actions.

Figure 13

(a) Interface between slag aggregates and paste, and (b) interface between stabilized slag aggregates and paste [70].

Figure 14(a) and (b) shows the weight loss of the cube while submerged in H₂SO₄ and HCl for various percentages of replacement. According to the test results, a substitution of 40% natural sand with SS results in less weight loss than a replacement of 30% coarse material and conventional concrete. Better acid resistance than our standard concrete is achieved by using SS, either in fine or coarse aggregate. Because sulfuric acid causes more rapid dehydration than hydrochloric acid, weight loss while submerged in it is greater than when submerged in the latter. When submerged in sulfuric acid, the amount of CSH gel produced by the hydration process will be somewhat less than when submerged in HCl.

Figure 14

Weight loss of samples: (a) H₂SO₄ immersed and (b) HCl immersed; data source [48].

The dehydrated gypsum layer produced by the interaction of sulfuric acid with blast furnace slag sand is denser and may stop sulfuric acid from penetrating the interior of the cement specimen when blast furnace slag sand is employed. Additionally, a lesser w/b results in greater strength and resistance to sulfuric acid [71]. The SS concrete sample with very harsh exposure conditions was found to withstand 7.5% better than the other exposure-conditioned samples and normal concrete specimens, according to Kumar et al. [72]. In contrast, Hewayde et al. found that although pozzolanic materials significantly enhanced CS and decreased porosity, their impact on the resistance to sulfuric acid was quite minimal. The strength and physical properties of concrete (compressive capacity and porosity) and its resistance to sulfuric acid attack could not be correlated. Additionally, the linear connection between the fall in compressive strength of concrete specimens exposed to H₂SO₄ attack and their mass loss was discovered [73].

Figure 15 illustrates the alterations in the grayscale curves of SS concrete exposed to erosion by sulfate solutions with mass fractions of 5 and 15%, respectively. The results indicate that the grayscale curves of each sample group modify to different levels after various erosion durations. The alteration is due to the degradation of the concrete’s surface morphology resulting from sulfate attack. Liu et al. [69] reported that a sulfate attack could effectively influence the performance of high-strength concrete. It can be noted that the increased concentration of the sulfate solution results in the concentration gradient between the concrete surface and the solution growing larger, promoting the penetration of sulfate ions into the concrete and the formation of expansive compounds. This could clarify why the compressive strength of concrete subjected to a saturated sulfate solution is lower than that of concrete exposed to a 10% sulfate solution.

Figure 15

Grayscale curve: (a) 5% and (b) 10% sulfate solution [74]. (a) 0% substitution rate of SS, (b) 30% substitution rate of SS, (c) 60% substitution rate of SS, and (d) 90% substitution rate of SS.

Figure 16 illustrates the grayscale standard deviation (Gstd) of each concrete sample subjected to erosion in sulfate solutions with differing mass percentages. A higher Gstd value indicates that the concrete has had more significant degradation, reflecting a stronger influence of sulfate solution erosion on its surface appearance. The Gstd values range from a minimum of 60% to a maximum of 90% for SS. The findings indicate that the observable degradation of SS coarse aggregate concrete first decreases and then worsens when the SS replacement rate increases. It is apparent that at an SS replacement rate of 60%, the concrete surface remains mostly unaltered, exhibiting negligible morphological changes pre- and post-erosion. This indicates that, at this replacement rate, the concrete exhibits better integrity when subjected to sulfate attack. In contrast, the remaining three groups exhibit cracks, cement paste delamination, and several different signs of degradation. This indicates that using an SS replacement rate below 60% in the composition of SS coarse aggregate concrete mitigates the observable degradation resulting from sulfate erosion. The underlying process could be ascribed to the ideal equilibrium of the physical and chemical characteristics of the concrete at the 60% substitution rate, which enhances resistance to sulfate-induced degradation. When used in suitable quantities, SS can enhance concrete’s durability under adverse conditions owing to its inherent qualities. Nonetheless, variation from this ideal replacement rate causes an imbalance in material composition, which raises susceptibility to sulfate attack.

Figure 16

Gstd of concrete eroded in sulfate solution: (a) 5 and (b) 10% sulfate solution [74].

2.8 Shrinkage

The findings indicate that SS tends to accelerate the drying shrinkage development in this period (Figure 17). However, the later age of dry shrinkage rate is decreased, and, at 90 days, becomes comparable to the reference concrete. Furthermore, the GGBS shows lower dry shrinkage as compared to the SS aggregates at all days and even lower than reference concrete after 28 days. Choi et al. noted that the drying shrinkage of all specimens first increases significantly and continues to increase for the first 50 days of the curing period. During the last phases of the curing process, drying shrinkage significantly diminishes and occurs at a moderate rate [75]. The concrete experiences initial drying when extra mixing water evaporates from the surface, particularly in the early days, which results in a reduction in volume (shrinkage). The fast loss of water results in considerable shrinkage in the early days. However, dry shrinkage decreases with time when the concrete’s pore structure becomes densifies over time. The reduced drying shrinkage in concrete with GGBS, compared to SS aggregates and the reference concrete after 28 days, can be ascribed to factors including particle size, morphology, chemical composition, pozzolanic activity, and hydration properties of GGBS.

Figure 17

Dry shrinkage [44].

Mostly, GGBS has smaller particles than SS aggregates, resulting in denser packing and less porosity within the concrete matrix. This mitigates moisture loss, which decreases shrinking. The less rough surface texture of GGBS particles, in contrast to the rougher texture of SS, enhances compaction and reduces water consumption, consequently minimizing shrinkage. Also, the chemical composition of GGBS improves its capacity to participate in additional hydration processes (creating secondary cementitious compounds), which consequently refine the pore structure and minimize shrinkage. Bayraktar et al. [76] also indicated that marble waste as fine aggregate decreased the drying shrinkage. Wang et al. [30] noted that the concrete using 15% SS has a drying shrinkage pattern similar to that of conventional concrete, with the only variation being a slightly accelerated progression of drying shrinkage after 70 days. Furthermore, concrete made with 30% SS demonstrates accelerated drying shrinkage compared to pure cement concrete over the first 30 days; however, this rate subsequently decreases. At 90 days of age, the final drying shrinkages of all three concrete mixtures (0, 15, and 30% SS) are almost identical. Therefore, SS influences the initial drying shrinkage characteristics of concrete. However, the long-term shrinkage results are similar to those of normal concrete, irrespective of the SS percentages. Additionally, incorporating fibers can increase TS [77] and enhance resistance to shrinkage crack formation in SS concrete. Furthermore, Zhang et al. [78] recommended ultra-high-performance concrete instead of traditional concrete to improve crack resistance. However, further detailed research is necessary.

3.1 Pozzolanic reactivity

The pozzolanic reactivity can be judged by the amount of Ca(OH)₂ formed during the hydration of cement. The decrease in Ca(OH)₂ is a positive sign of pozzolanic reaction, indicating that Ca(OH)₂ is reacting with silica and forming additional cementitious compounds. Pozzolanic materials in the presence of water and Ca(OH)₂, a chemical reaction occurs in which the silica in the pozzolan reacts with the calcium hydroxide generated during the hydration of cement and produce additional cementitious compounds. The following equation illustrates the pozzolanic reaction:

The concentration of Ca(OH)₂ in samples made with SS and steel sludge appears to be equal at 3 days in all mixes (control, SS, and steel sludge) but lower than that in the control paste at a later age (7, 28, 60, and 90 days), as shown in Figure 18. The justification for this is that the Ca(OH)₂ concentration may be inadequate for interaction with the pozzolanic materials at an early stage. Nonetheless, chemical interactions inside the pozzolanic materials at later ages may result in a decreased quantity of Ca(OH)₂. Therefore, the resulting chemical reactions in later stages promote the development of CSH gel, which enhances the concrete qualities and explains the reduced amounts of Ca(OH)₂. Therefore, the decrease in Ca(OH)₂ content shows that the materials act as pozzolanic materials and produce secondary CSH. He et al. [79] also noted that active Si from amorphized clays reacts with Ca(OH)₂ to form calcium silicate hydrate (C–S–H), which is the main strength contributor in hardened mortars.

Figure 18

Calcium hydrate percentages [56].

3.2 Heat of hydration

The five phases of cement’s hydration evolution – the quick exothermic period, the dormant period, the acceleration period, the deceleration period, and the stable period can be applied to the hydration development of fine slag (particle size smaller than 20 μm). The exothermic rate curves of cement and fine slag (material with particles smaller than 20 μm in diameter) are very similar, especially during the first 10 h of hydration (Figure 19). Even before that of cement, the second exothermic peak of fine slag first arises. Some claim that SS has substantially lower hydration activity than cement and that this causes the inactive phase to last longer [80]. As a result, fine slag has a substantially greater hydration activity than SS prior to particle size categorization.

Figure 19

Heat of hydration [47].

Similar studies have shown that the pozzolanic reaction, which eventually results in a decrease in heat, particularly during the early stages of hydration, continues progressively in conjunction with the hydration of cement [81]. Similar to cement, Wang et al. [82] found that SS hydrates at a rate that is significantly slower than cement at first but increases to cement levels by 90 days. After 10 h, the exothermic rate of fine slag is still lower than that of cement. Additionally, as shown in Figure 19, its second exothermic peak is similarly less than that of cement. Note that fine slag has a lower fineness than cement. Thus, it may be inferred that fine slag contains fewer active components than cement. The exothermic rate of coarse slag (particle size more than 20 μm) is much lower than that of fine slag during the whole hydration history. The exothermic rate curve of coarse slag also lacks a clear second exothermic peak. The rate of coarse slag exothermic is insignificant after about 24 h. These findings show that coarse slag activity is lower than fine slag activity. The primary cause of the low reactivity of coarse slag as compared to fine slag is the much larger particle sizes of coarse slag. Another aspect is that coarse slag has a higher proportion of inert components compared to fine slag and exhibits a higher proportion of inert components compared to cementitious compositions [47]. Wang et al. suggested that increasing the fineness of SS might increase the speed of hydration during both the early and late stages. The rate of hydration decreases at 90 days, and by 180 days, the level of hydration is similar to that of SS, with different levels of fineness [82].

3.3 SEM

SEM may be used to analyze the surfaces of concrete samples. These techniques are used in procedures such as the investigation of product failures, bonds, detection of damage angle, and several other applications. Concrete is a permeable material, and when water and cement are mixed, they form a solidifying mixture that may bind sand and stone. The hydration products will not occupy the original internal space that was filled with water, leading to the formation of pores. The frost resistance of concrete is significantly influenced by these holes, which are often referred to as fine pores. The size of these pores is typically a few nanometers to several microns [83]. SEM images may be used to examine how the microstructure changes as a result of drying or dehydration. Figures 20 and 21 depict the SEM findings for a concrete sample including and excluding SS.

Figure 20

ITZ: (a) control and (b) SS concrete [84].

Figure 21

Hydration product: (a) control and (b) SS concrete [84].

Figure 20(a) demonstrates how the unfavorable bonding between the coarse aggregate and cement paste caused evident microcracks of around 2 μm wide in the ITZ of conventional concrete. Even as freezing–thawing cycles increased in frequency, the performance of the bonding interface between binder and aggregate in ordinary concrete would be substantially affected. The aqueous solution constantly froze and thawed once it reached the concrete. The force required to make concrete sustains an increase in ice crystal expansion pressure [85]. The first damage region of concrete is in the interfacial transition zone (ITZ), where the linking between crystals may be thought of as a micro-spring system from the perspective of random damage in damage mechanics. Due to the tight connections between the SS aggregate and the cement, Figure 20(b) demonstrates that the ITZ of SS was denser. In other words, there were more micro-springs. As a result, when the two materials were exposed to identical freezing–thawing damage conditions, the SS concrete suffered from less cracking damage and had superior macroscopic freezing–thawing resistance than regular concrete specimens.

The chemical makeup of the SS and cement was comparable. Tricalcium silicate and dicalcium silicate underwent a hydration reaction afterward. Figure 21 shows that when hydration took place later in the interval, more gels were formed (Figure 21(b)). The SS surfaces had pores that made it possible for the aggregate and gels to be more deeply implanted. This has simultaneously decreased the number of holes and successfully stopped the flow of the aqueous solution [86]. Concrete performed better when pozzolanic reaction and pozzolanic material filling were combined.

Research also found that, at the age of 360 days, fine slag had formed a certain number of hydration products. The majority of the hydration byproducts of fine slag are amorphous and colloidal. The amorphous hydration products of fine slag are mostly calcium silicate hydrate (C–S–H) gels, according to the findings of the EDS study. The morphologies of the hardened fine slag paste and the hardened cement paste are extremely similar [47]. Particles of pozzolanic materials have a filler action that results in the development of an incredibly dense structure as a result of their presence. Concrete’s cohesiveness is increased while segregation and bleeding are reduced with the inclusion of pozzolanic elements. To limit the porosity of the cured concrete and counteract the negative effects of trapped air, a small quantity of nano-silica is added to the cementitious matrix. It should be mentioned that when calcium hydroxide and nano-silica (as pozzolanic material [87]) react, a secondary C–S–H gel is produced. The secondary C–S–H gel makes the cured concrete denser and increases its strength and durability by filling all of the pores and capillary channels in the concrete.

Furthermore, Figure 22 illustrates the microcracks of different samples after 180 days. The findings show that the surface microcrack widths increased with SS powder and stone powder as cement replacement. The crack width for the control, 30% SS powder, 30% stone powder, and a mixed mix of 15% SS powder and 15% stone powder are 1.5, 5, 4, and 2.5 µm, respectively. Therefore, both SS powder and stone powder, either individually or in combination, result in an increase in crack width. The individual use of SS powder results in a more brittle mortar mixture, resulting in wider surface cracks. Therefore, the SS powder could reduce the mortar’s overall strength, leading to more significant cracking and splitting. Additionally, the incorporation of 30% stone powder raises the surface microcrack width to 4 µm, higher than that of the control sample but lower than 30% SS powder (5 µm). Like SS powder, stone powder alone makes the mortar more prone to cracking, although its effect is slightly less than that of SS powder. When SS powder and stone powder are combined at 15% each, the microcrack width is reduced to 2.5 µm. Although this is still larger than the control group, it is smaller than the individual additions of 30% SS or 30% stone powder. This suggests that a balanced combination of these two powders moderates the detrimental effects of microcrack formation. The mixture seems to provide some improvement in reducing the overall crack width, likely due to a complementary interaction between the two additives. Similar to SS powder, stone powder individually also increases the mortar’s susceptibility to cracking and splitting. However, stone powder shows a lesser crack width than SS powder. Furthermore, the combined use of 15% SS powder and 15% stone powder reduces the microcrack width to 2.5 µm, which remains higher than the reference mix but is individually used. Therefore, the findings show that the combined mix mitigates the adverse impact on microcrack development. The reduction of crack width with the combined use of SS powder and stone is due to pozzolanic action. The pozzolanic action increased the binding ability of the paste, which led to resistance to the crack development.

Figure 22

(a) Control, (b) 30% SS powder, (c) 30% stone powder, and (d) 15% SS powder and 15% stone powder [28].

3.4 Thermal analysis

Figure 23 shows a similar pattern in weight loss; however, the weight loss between the reference and the SS sample is different. As a result of the SS’s small promotion influence on the cement hydration, the weight loss in the sample (5% SS, particle size 45–80 μm) was discovered to be 1.7% greater than that in the control sample. The derivative weight curves clearly show three significant peaks that are caused by the dehydration of the C–S–H, AFt, and AFm phases (40–235°C) [89], the decarbonization of CaCO₃ (529–929°C), and the dehydroxylation of CH (368–499°C) [90]. The computed CH content is also shown in Figure 23. The control sample’s CH concentration was 11.7%, whereas the sample with 5% SS had a significantly lower amount (11.6%). A previous study suggests three potential causes for the drop in CH content: The SS partially replaces the cement, lowering the cement’s relative concentration and resulting in less CH being released during cement hydration. When SS is added, the effective water-to-binder ratio increases, boosting some cement’s degree of hydration and producing more CH. In addition to pozzolanic action, SS consumes calcium hydrates at a later curing age. The CH content difference between SS and the reference concrete, as well as the overall weight reduction, were assumed to be the results of the pozzolanic action of the SS. In addition, SS consumed CH and created additional C–S–H, which might be advantageous for compacting the microstructure of the hardened paste, resulting in a higher CS for the concrete.

Figure 23

Thermogravimetry curve [88].

Qiang et al. [47] came to a similar conclusion that the amounts of Ca(OH)₂ in cement and SS paste were determined to be 17.2 and 7.9%, respectively. According to these findings, at 360 days old, the Ca(OH)₂ concentration of the sample SS only makes up around 46% of the cement. Note that cement only has a 0.79% f-CaO concentration compared to FSS (1.32%). As a result, f-CaO in FSS produces more Ca(OH)₂ than cement does. The quantity of Ca(OH)₂ formed by the cementitious elements in FSS is much less than that in cement after 360 days. Even though Ca(OH)₂ should be present with the hydration of C₃S and C₂S, it was difficult to detect it at an early age owing to the slow rate of hydration. After 28 days, however, Ca(OH)₂ was found. After 90 days, it was still possible to differentiate the C₃S and C₂S peaks, indicating that some C₃S and C₂S with low activation remained in the system [64]. A similarity between the hydration byproducts of cement and SS was observed. However, compared to cement, SS still produces fewer hydration products.