The utilization of two kinds of iron slag in the production of pozzolanic cement is introduced. A series was created with a fixed percentage of OPC (30%) and varying amounts of imported granulated blast-furnace slag (IGBFS) and air-cooled slag (ACS) at which imported IGBFS was substituted by ACS with the mass ratios of 10, 20, and 30%. Physico-mechanical properties and hydration parameters of the hardened pastes were examined at different time intervals. Furthermore, the aggressive attack of seawater on chosen specimens was studied for up to one year of immersion. The hardened composite cement pastes were tested in terms of weight loss, compressive strength, bulk density, total porosity, and free lime at different thermally treated temperatures starting at 105 °C and ending at 800 °C for 2 h of a socking period to investigate its thermal characteristics. XRD, IR, and DTA/TGA techniques were used to examine some chosen samples. The results revealed that the incorporation of ACS reduces the water consistency and prolongs setting times. Compressive strengths are higher in samples containing 10% ACS than those specimens containing 20–30 wt% and without ACS. A 7% reduction in the compressive strength was achieved by A1 (30%OPC and 70% IGBFS), which is the lowest one over 1 year of exposure to seawater. The incorporation of ACS at the expense of IGBFS tends to lower the strength but tends to enhance the bulk density.
Notes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Introduction
Urbanization growth necessitates infrastructure expansion, which raises the output of ordinary Portland cement, the most often used binder. Approximately 8% of the world’s anthropogenic CO2 emissions are caused by the manufacturing of Portland cement [1]. Recently CO2 emissions have been reduced primarily by three methods (e.g., energy efficiency improvement, fuel switching using waste as an alternative fuel, and blended cement using industrial by-products) [2‐4]. Iron slag is perhaps one of the earliest additions that were made to cement to improve its strength and chemical durability [5]. Blast-furnace slag (BFS) is a by-product of the iron-making process that is produced when the aluminosilicate waste from the iron ores is fused with limestone and coke ash. This procedure results in the formation of a molten slag which floats on the surface of the iron. After that, it is cooled and withdrawn. Based on the cooling process, there is three different forms of iron slag are formed, namely granulated, air-cooled and pelletized blast-furnace slags which are reused in many applications, especially concrete works [6‐8]. The rapid cooling of iron slag with water is responsible for its amorphous nature which exceeds 95% amorphous calcium aluminosilicate. This kind of slag is known as granulated blast furnace slag (IGBFS) [9‐12]. Incorporating IGBFS in concrete improves various properties, including early hydration, strength, permeability, and sulfate resistance, while a higher proportion of IGBFS reduces rebar corrosion [13]. As opposed to this, ambient air is used to gradually cool crystalline material known as air-cooled slag (ACS) [14, 15]. Rapid cooling of iron slag with air is responsible for glassy or crystalline pellets known as pelletized blast-furnace slags (PBF) [6]. The World Steel Association estimates that about 90% of all by-products by weight is produced globally of steel slag because the iron and steel industry is one of the most energy and resource-intensive industries that contribute to greenhouse gas emissions [16]. The recycling of its waste, which is in line with the principle of the circular economy, has a significant environmental and economic impact. Iron slag has been extensively researched particularly for use in the production of building materials. One sink that is occasionally used to accommodate ACS is the construction industry by incorporating this by-product in concrete, building bricks, road construction etc. [17‐23]. On the other hand, due to its vitreous structure, GBFS is frequently utilized as SCM throughout the world. In the presence of a suitable activator, it exhibits both cementitious behaviour and pozzolanic properties [24]. In this respect, [25, 26] showed that replacing cement partially with GBFS in concrete mix affects positively its workability. While Lee et.al [9] found that using GBFS elongates both the initial and final setting times and [9, 25, 27] noticed slow mechanical strength development at an early age. The results obtained by [25, 27‐30] revealed that the compressive strength at later age exceeds that of free slag concrete. This improvement in mechanical properties is due to the high glass content of GBFS which increase pozzolanic activity. The partial replacement of cement by GBFS in concrete mix improves its durability against Cl− attack [31]. They also recommended further testing on Cl− and SO42− attacks for longer times. Much more recently, [32] noted that the replacement of up to 70% of the cement with IGBFS improves the heat resistance for blended cement paste. Finally, a composite cement was employed to replace cement to utilize most industrial by-products; this improved the concrete's strength and/or lifespan [33, 34]. A combination of steel slag and IGBFS up to 40% was optimal, resulting in comparable compressive strength and denser structures, while also reducing environmental impacts [35]. With the directives of the Egyptian government to benefit from industrial by-products by reusing or recycling them in the interest of the environment and the economy, we decided to prepare a composite cement by mixing two types of by-products, IGBFS and ACS, to maximize environmental and economic benefits. Up to 180 days of hydration, the physicomechanical properties of the hardened cement pastes were investigated. Seawater attack and fire resistance are also examined.
Starting materials and their characterization
Tap water was utilized to mix and cure all cement pastes, and Suez Gulf seawater, which had sulphate and chloride concentrations of 8786 ppm and 12,543 ppm, respectively, was used as an aggressive medium to cure all specimens and study sulphate attack. Ordinary Portland cement (OPC) from the Suez Cement Company, the Iron Steel Company provided the air-cooled slag (ACS), Helwan, Egypt, and the IGBFS is supplied by (Kokan Mining Co. LTD, Japan). A laboratory steel ball mill with (a 5 kg) capacity is used to grind the slag into a fine powder. The specific surface area was calculated with the Blaine air-permeability device following the ASTM method [36]. The XRF analysis and the Blaine surface area of utilized materials are displayed in Table 1. Figure 1 illustrates the phase composition of ACS as determined by the XRD technique. It shows that the ACS is entirely crystalline phases (quartz (Q) and akermanite (A)). The IGBFS is typically almost entirely vitreous, with an amorphous hump typical of glass structure at 20 and 38 2θo attributed to the amorphous glassy component consistent with the network-forming oxides SiO2 and Al2O3 making up the amorphous phase.
Table 1
Chemical analysis (XRF), and the specific surface areas of utilized materials (wt %)
Item
IGBFS
OPC
ACS
SiO2
37.48
20.51
40.78
Al2O3
12.86
5.07
10.95
Fe2O3
0.40
4.39
1.70
CaO
37.70
62.21
35.22
MgO
2.45
2.00
4.10
Na2O
1.84
0.23
1.91
K2O
0.93
0.29
0.58
BaO
5.31
0.00
0.0
SO3
0.01
2.25
0.16
S2−
0.75
0.0
0.0
L.O.I
0.00
2.40
0.0
Specific surface area (cm2/g)
4300
3050
4000
Fig. 1
XRD pattern of ACS and IGBFS
×
Advertisement
The OPC used in this study consists of four main phases with the following composition: Alite = 52.0%, belite = 19.20%, C3A = 8.80% and C4AF = 11.08%. XRD of IGBFS is seen in Fig. 1.
Experimental method
The IGBFS and ACS were finely processed using a laboratory steel ball mill with a 5 kg capacity after being crushed by a jaw crusher. Numerous specimens are created by mixing varying concentrations of IGBFS and ACS with a consistent quantity of OPC (30%). The mixes were denoted as A1, B1, B2, and B3 as shown in Table 2. The prepared mixes were mechanically combined for 1 h in a porcelain ball mill to achieve total homogeneity, and then tap water was used to hydrate the cement pastes. According to ASTM specifications, the water of consistency and setting times for each combination were determined [37]. Stainless steel (2.54 × 2.54 × 2.54 cm3) cubic molds were used for casting the cement pastes. The molds were then cured for the first 24 h in a humidity chamber with 100% relative humidity and a constant temperature of 25 °C. The molds were removed, and after that, they were submerged in water for the specified amount of time 3, 7, 28, 90, and/or 180 days. After any period of hydration, bulk density and total porosity were assessed as mentioned [38]. Chemically combined water content was assessed based on ignited weight loss [39, 40]. Ammonium acetate was used in a method to calculate the free lime content (FL.%) [41, 42]. As per ASTM (C-150) [43], compressive strength was assessed. The specimens underwent fire resistance testing after being cured solely in tap water for 28 days and dried at 105 °C for 24 h. They were thermally treated at a rate of 10 °C/min and held at a steady temperature for two hours at a maximum temperature of 800 °C to ensure consistency [44]. The paste specimens were subjected to temperatures of 25 °C, 250 °C, 450 °C, 600 °C, and 800 °C for a duration of 2 h, and then allowed to cool down to room temperature inside the furnace switched off. They were then covered with plastic film and placed in a desiccator to eliminate any potential moisture or carbonization effects. A set of four chips collected for each mix composition fired at appropriate temperature were immersed in kerosene for 24 h. The specimens were then the bulk density and total porosity were calculated using the density of kerosene (0.8 g/cm3). To test the aggressive media’s resistance for some selected specimens (A1, B1, B3), specimens were first cured under tap water for 28 days (zero time), then cured under seawater at 0, 1, 3, 6, and 12 months, respectively. Chemical composition of seawater is given in Table 3. The variations in compressive strength, bulk density, total porosity, combined water, and free lime contents for the immersed samples up to 1 year in Suez Gulf saltwater. The seawater from Suez Gulf as an aggressive medium was renewed every month with fresh seawater to attain the seawater nearly constant. After compressive strength test, samples were crushed and the hydration stopped using methanol/acetone/diethyl ether, followed by drying at 75 °C for 1 h [38, 39]. Some chosen hydrated samples were examined using differential thermal analysis (DTA/TGA), and differential scanning calorimetry (DSC) to confirm the mechanism anticipated by the chemical and mechanical testing. On choosing the hydrated and burnt cement pastes, X-ray diffraction method (XRD) testing was conducted. To provide further details on the hydrated products, infrared spectra were performed on a chosen paste.
Table 2
The mix proportions for blended cement pastes, wt %
Mix No
OPC
IGBFS
ACS
A1
30
70
0.0
B1
30
60
10
B2
30
50
20
B3
30
40
30
Table 3
The ion concentration of Suez Gulf seawater
Ions
Ca2+
Mg2+
Cl−
SO42−
Na+
T.D.S*
Concentration, ppm
425
837
12,543
8786
14,522
35,939
*T.D.S Total dissolved solids
Results and interpretation
Standard water of consistency and setting times
Figure 2 shows a graphic representation of the water of consistency and initial and final setting times of composite cement pastes prepared from IGBFS and ACS as a function of mix composition. With the replacement of IGBFS with ACS, it is evident that the required water of standard consistency was slightly reduced. But when ACS concentration increases, both initial and final setting times were increased up to 10–20 wt% ACS, which gives small elongation. At 30 wt% the setting time is elongated at a higher rate than at 10–20 wt. %. As a result of the greater concentration of ACS, there is a somewhat pozzolanic reaction and a consequent decrease in the hydration products of C-S-H, C-A-H, and C-A-S-H. For cement pastes to be set, these phases are necessary.
Fig. 2
Effect of replacement of IGBFS with ACS on water of consistency, initial and final setting times of IGBFS rich cement pastes
×
Chemically combined water contents
Figure 3 shows the change in chemically combined water contents with the time of curing in. The results reveal that the combined water content rises with curing time where cement continues to hydrate and more C-S-H, C-A-H, and C-A-S-H are formed. Also, the chemically combined water content reflects the high extent of hydration reaction of OPC cement-containing phases (β-C2S and C3S) to generate C-S-H and Ca(OH)2 (CH), and CH serves as a pozzolanic activator. On the other hand, the accumulated hydration products and hence the combined water likewise drop as the IGBFS declines at the expense of ACS. Mix B3, which consists of 40% IGBFS and 30% ACS, exhibits a slower hydration rate when compared to the other mixtures. This can be primarily attributed to the slower hydration rate of ACS as compared to Portland cement. Additionally, the combined water content in pozzolanic cement pastes is lower than that in OPC paste hydration [45].
Fig. 3
Effect of replacement of IGBFS with ACS on chemical combined water of hydrated IGBFS-rich cement pastes up to 180 days
×
Advertisement
Free lime contents
Free lime contents of all specimens are varied with curing time as described in Fig. 4. The results demonstrate that the curing time affects free lime content negatively, this is due to the persistence of the pozzolanic reaction between IGBFS and the produced CH during the cement hydration [46]. Free lime content declines when ACS is substituted for IGBFS; for example, cement paste with 30 weight percent ACS gives the highest values of free lime during the entire curing time of up to 180 days. Residual Ca(OH)2 diminishes by the amount of ACS decreasing at the expense of IGBFS. This is mostly because IGBFS has a greater pozzolanic activity level. It is also obvious that free lime consumption increases with curing age, which is mainly because there is more pozzolanic activity at a later age than at an early age.
Fig. 4
Effect of replacement of IGBFS with ACS on the total free lime content of hydrated IGBFS-rich cement paste up to 180 days
×
Bulk density
Figure 5 illustrates the variation in the bulk density with a curing time of up to 180 days. The bulk density of cement pastes cured up to 180 days is reduced when a 10-weight percent of IGBFS is replaced with ACS. This is primarily because IGBFS, which has a high pozzolanic response, has better hydraulic characteristics than ACS, which has weak hydraulic qualities. The equivalent bulk density is raised to 180 days of hydration when IGBFS is substituted with ACS at 20 and 30 wt%, this is mostly caused by the fact that ACS has a larger bulk density than IGBFS. Because a portion of the accessible pores is filled with cement paste during the curing process, the overall volume of the pores is reduced, which increases the bulk density of the solid materials [47].
Fig. 5
Effect of replacement of IGBFS with ACS on the bulk density of hydrated IGBFS-rich cement paste up to 180 days
×
Total porosity
Figure 6 illustrates the change in total porosity with a time of curing of up to 180 days for ACS and IGBFS cement paste. It is evident that, up to 180 days, the total porosity of IGBFS without ACS provides practically the lowest total porosity of any composite cement paste. Conversely, composite cement pastes with 10 and 20 wt% ACS produce almost the same total porosity and are higher than those with 30% ACS and without ACS. Cement paste composed of 30% ACS with 40% IGBFS with only 30 wt% OPC shows an intermediate value between the pozzolanic cement paste with 70% IGBFS and the composite cement pastes containing 10–20% ACS. The figure also shows a sharp drop in total porosity from 90 up to 180 days. This is mainly because composite cement pastes exhibit greater pozzolanic activity at later ages [48].
Fig. 6
Effect of replacement of IGBFS with ACS on the total porosity of hydrated IGBFS-rich cement paste up to 180 days
×
Compressive strength
The results presented in Fig. 7 showed that the mechanical compressive strength was positively affected by curing time for all specimens under study. B1 achieved the highest compressive strength of all since 10% ACS service as a nucleating agent which stimulated the hydration rate which is positively reflected in the compressive strength. Cement pastes without ACS and 70 wt% IGBFS give lower compressive strength than that containing 60 wt% IGBFS and 10 wt% ACS as previously discussed which ACS acts as nucleating sites. With an increase in ACS content of more than 10%, the compressive strength is negatively affected. These cement pastes are blended since ACS has lower hydraulic material, which affects the strength negatively. All specimens behaved in the same way, as the compressive strength increased quietly in the first days, unlike what happens after three months and up to 6 months. We also notice a sharp increase in strength, this is due to the recent growth of pozzolanic activity. Also, the variation in the compressive strength with the chemical water content is presented in Fig. 8, and a mathematical linear equation has been established with a goodness coefficient greater than 0.75. These findings are in good accord with those of [49].
Fig. 7
Effect of replacement of IGBFS with ACS on the compressive strength of hydrated IGBFS rich cement paste up to 180 days
Fig. 8
Compressive strength of Imported IGBFS and ACS rich cement pastes with combined water
×
×
The relationship between the compressive strength and bound water content in hydrated cement is not necessarily linear. This relationship depends on a variety of factors, such as the composition of the cement, the curing conditions, and the testing method used. In some cases, it observed a linear relationship between compressive strength and bound water content in hydrated cement [49‐51]. However, other studies have found non-linear relationships or more complex relationships between these two factors. Scrivener et. al. discussed the complexities of the hydration process in OPC and its influence on compressive strength. It emphasizes the non-linear relationship between bound water content and strength development [52].
FT-IR spectroscopy
Figure 9 shows the FT-IR patterns of hydration products of IGBFS and ACS -cement pastes (A1 and B3) after 180 days. Absorption bands at 3448–3456 cm−1 and 1647–1653 cm−1 are related to stretching and bending vibration of O–H groups of H:OH, respectively. Absorption bands of water are due to the lattice H2O of the hydrated C-S–H, C-A-H as well as C-A-S–H. [53, 54]. The band at 1633 cm−1 decreases in B3 due to the decrease of combined water with ACS. The band at 1433 cm−1 is attributed to the presence of calcite, possibly formed due to carbonation of the pastes [55, 56]. The intensity of the band at 976 cm−1 is due to Si–O stretching vibration of C-S-H [57]. Increases in A1 (indicating the increase of C-S-H). This is observed in the higher values of compressive strength of A1 than B3 at 180 days.
Fig. 9
FT-IR spectroscopy of A1 and B3 mixes cured up to 180 days
×
Thermogravimetric analysis (DTA/TGA)
Figure 10 illustrates the DTA thermograms of mixes A1 and B3 at 180 days. The dehydration of water and combined free from C-S-H within the hydrated cement matrix causes endotherms to occur at 70–120 °C [58]. The endothermic peaks associated with C-A-H and C-A-S-H occur at 269, and 345 °C [59]. The endothermic peaks around 664–758 °C, are related to the breakdown of carbonate, hydrated aluminates, and the final phase of C-S–H decomposition [60]. The exothermic peak at 845–900 °C is caused by wollastonite recrystallizing [61]. Additionally, the endothermic peaks at the low-temperature range due to the decomposition of hydration products increase with IGBFS content. This means that mix A1 with 70 wt% IGBFS gives better hydration product values as seen from the peak area of the hydrated pastes. As the ACS content increases at the expense of IGBFS the hydration products decrease. This is mainly due to the higher degree of hydration of IGBFS in comparison to ACS. The breakdown of well-crystallized CaCO3 is what causes the endothermic peak at 878 °C. The ACS contains a small endothermic peak in the range of 345 °C due to the decomposition of brucite Mg(OH)2. The dehydroxylation of portlandite is responsible for the endothermic peak in the 395–450 °C range. Due to the strong pozzolanic activity of IGBFS in comparison to ACS, it is obvious that the peak of portlandite in the case of A1 is more intense than that of B3. As a result, B3 paste has more portlandite than A1 paste.
Fig. 10
The DTA thermograms of hydrated Imported IGBFS and ACS rich cement paste of mixes A1 and B3 at 180-days
×
Figure 11 illustrates the thermogravimetric analysis (TGA) of hydrated A1 and B3 cement pastes for up to 180 days. TGA losses of A1, and B3 because hydration products are decomposed up to 250 °C are 8.489 and 7.795 wt% for A1 and B3, respectively. On the other hand, the TGA losses due to the dehydroxylation of portlandite are 3.318 and 4.016 wt% for A1 and B3, respectively. The last decomposition of the CO32− is 2.622 and 3.105 wt% for A1 and B3, respectively. The cement paste A1 is more hydraulic than B3. This is mainly due to the incorporation of ACS (B3) instead of IGBFS (A1). The IGBFS has a pozzolanic reaction whereas ACS is less hydraulic material. Cement paste's combined water content is 11.113 and 12.505 wt% of B3 and A1.
Fig. 11
The TGA thermograms of hydrated Imported IGBFS and ACS rich cement pastes of A1 and B3 at 180-days
×
It is obvious that IGBFS results in higher combined water content. The combined water as determined by the chemical method includes the loss of both Ca(OH)2 and CaCO3. The IGBFS is more hydraulic than ACS and these values are in according with the determined values of combined water. Cement paste containing 70 wt% IGBFS gives weight loss of 8.489, 12.505, and 15.51 wt% whereas cement paste with 40% IGBFS and 30% ACS shows 7.795, 11.113, and 13.735 wt. %. This means the cement with higher content of IGBFS is more hydraulic.
Effect of firing temperature
Weight loss
A graphical plot of the weight loss of hardened specimens (A1, B1, B2 and B3) varied with the thermal treatment temperatures up to 800 °C, is presented in Fig. 12. Because of the breakdown of some hydration products, weight loss typically increases with treatment temperature for all cement pastes [10]. Below 105 °C, the free water was eliminated, and below 200 °C, calcium silicate, aluminate, sulphoaluminate, and aluminum–silicate hydrate partially decomposed. At 200–250 °C, gehlenite hydrate (C2ASH8) phase starts to break down. Between 420 and 520 °C, Ca(OH)2 underwent dehydroxylation [62]. The decomposition of C-S-H, C-A-H, calcium sulphoaluminate hydrates, gehlenite hydrate, and Ca(OH)2 causes the weight loss to increase up to 800 °C [10, 63]. It has been demonstrated that up to 800 °C, A1 loses more weight than B1, B2, and B3. The reduction of mixing water, which results in a drop in the quantities of free and chemically mixed water, may be responsible for the decreased weight loss of B1, B2, and B3. Additionally, the rate of hydration and subsequent weight loss of cement pastes are both reduced when ACS is added in place of IGBFS.
Fig. 12
Effect of replacement of IGBFS with ACS on the weight loss of hydrated IGBFS-rich cement paste at different temperatures
×
Bulk density
Figure 13 illustrates the relationship between the bulk density of treated pozzolanic cement pastes with varying IGBFS and ACS dosages and the thermal treatment temperature up to 800 °C. It is evident that treatment temperatures up to 250 °C, result in a decrease in bulk density. At all treatment temperatures, cement pastes without ACS had the lowest bulk densities. The breakdown of some hydration products causes the bulk density to decrease, but the ACS has a higher specific gravity than IGBFS, which tends to increase the bulk density. The elimination of free or adsorbed as well as bound water of C-S-H or C-A-H has taken place. There are three phases involved in changing the bulk density of variously treated cement pastes. The apparent density typically declines with treatment temperature up to 250 °C (first step). The density is virtually the same in the second step, between 250 and 400 °C. For all pozzolanic cement pastes, a rise in density is observed during the third step (from 400 to 800 °C). In comparison to other pozzolanic cement pastes, B3 rich in ACS cement paste has a higher apparent density. This is mostly because of ACS's denser structure and higher specific gravity compared to GBFS.
Fig. 13
Effect of replacement of IGBFS with ACS on the bulk density of hydrated IGBFS-rich cement paste at different temperatures
×
Total porosity
A graphical plot of the porosity of hardened specimens (A1, B1, B2 and B3) varied with the treatment temperature up to 800 °C, is presented in Fig. 14. The breakdown of hydration products Ca(OH)2 and CaCO3 causes the total porosity typically increases with treatment temperature from 400 to 800 °C for all cement pastes [64]. The internal cracks growth in B2 and B3 leads to a total porosity rise up to 800 °C. In addition, increasing the percentage of ACS negatively affects the hydration rate, which in turn increases the porosity. It can be concluded that mix A1 (30% OPC with 70% IGBFS) gives lower porosity in comparison to others because of product accumulation growth in its open pores.
Fig. 14
Effect of replacement of IGBFS with ACS on the total porosity of hydrated IGBFS rich cement paste at different temperatures
×
Compressive strength
Figure 15 illustrates the compressive strength as a function of treated temperature from 105 °C up to 800 °C. Compressive strength can be found to rise with temperature up to 250 °C before decreasing up to 800 °C. The increase in cement paste hydration products and the pozzolanic reaction of IGBFS with CH to produce additional amounts of C-S-H with a low Ca/Si ratio and high strength are both responsible for the increase in compressive strength up to 250 °C [65]. This is also a result of the internal and self-autoclaving action, in which steam accumulates due to the removal of capillary, adsorbed, and bound water at high temperatures. This increases the pressure of steam, which speeds up the hydration of anhydrous cement paste grains to form dense-microstructure matrixes [66]. At 800 °C, the dissociation of cementing ingredients and swelling of the water layer in cement pastes cause the compressive strength to diminish [67]. Additionally, the coarsening of the pore size distribution causes an increase in porosity [65]. The A1 (70 wt% IGBFS) is more heat resistant than the other specimens at all temperatures. This is because increased amounts of calcium silicate, aluminate, and aluminum silicate hydrates are produced as a result of the reaction between released lime and IGBFS. The compressive strength is increased by these hydrates that have been deposited within the pore system. The compressive strength of thermally treated cement pastes tends to decrease when equal amounts of IGBFS are substituted with ACS. The compressive strength of B1 cement paste is almost identical to that of A1, although at lower values. Additionally, the compressive strength rises with ACS content i.e., 30% up to 250 °C before declining. The thermal resistance of ACS up to 300 °C may be to blame for this rise. The slow rate of hydration of B3 at 105 °C explains the lower strength.
Fig. 15
Effect of replacement of IGBFS with ACS on the compressive strength of hydrated IGBFS-rich cement paste at different temperatures
×
The visual inspection of pozzolanic cement thermally treated at 600–800 °C is shown in Fig. 16. The surface cracks appeared at 600 °C, which propagated at 800 °C. These cracks diminished with A1 mix. Hence, the pozzolanic cement pastes, A1 mix, improve the thermally treatment resistance.
Fig. 16
The visual inspection of pozzolanic cement thermally treated at 600 up to 800℃
×
Effect of seawater attack
Chemically combined water content
Figure 17 illustrates the combined water content of IGBFS and ACS cement pastes submerged in seawater. The combined water content increases with an immersing period for A1 cement paste, and increases up to 6 months for cement pastes B1 and B3 containing ACS with IGBFS then decreases gradually up to 12 months. The increase of combined water of A1 mix containing 70% IGBFS up to one year of seawater immersion primarily as a result of the high pozzolanic nature of IGBFS. On the other side, the pastes containing 10 and 30 wt% ACS show lower water contents than A1 up to 1 year. The combined water content also drops as ACS increases because of ACS inert nature. The decrease in combined water contents after 6 months in seawater may be due to silicate polymerization with low water content.
Fig. 17
Effect of replacement of IGBFS with ACS on the chemically combined water of hydrated IGBFS-rich cement paste immersed in seawater up to a year
×
Free lime content
Figure 18 shows a graphic description of the values of free lime of various cement pastes immersed in seawater for a year. According to the results, the free lime contents of all cement pastes decrease gradually up to one year of immersion in seawater; this results from IGBFS- pozzolanic reaction as activated by lime leading to the extra-formation of C-S–H and C-A-S-H [46]. As the amount of IGBFS slag increases the free lime content decreases due to the high pozzolanic properties of GBFS. Because ACS has a low pozzolanic ability to react with residual free lime, the amount of free lime in cement pastes containing ACS increases.
Fig. 18
Effect of replacement of IGBFS with ACS on the free lime of hydrated IGBFS-rich cement pastes immersed in seawater up to a year
×
Bulk density
The variation in bulk density of the hardened specimens immersed in seawater against curing age is plotted in Fig. 19. According to the results, the cement is stimulated by Cl− and SO42− ions in seawater to produce wet hydrated products that fill some open pores, increasing the bulk density over a year of immersion in seawater. The bulk density of the specimens, on the other hand, improves when the ACS content rises. This suggests that ACS anhydrous grains have higher specific gravities than IGBFS grains.
Fig. 19
Effect of replacement of IGBFS with ACS on the bulk density of hydrated IGBFS-rich cement pastes immersed in seawater up to a year
×
Total porosity
The total porosity of the hardened pastes immersed in seawater for 1, 3, 6, 9 and 12 months are shown in Fig. 20. Owing to the precipitation of hydrated products into the accessible pores of cement pastes [48], the total porosity of all specimens steadily decreases with immersion time. The total porosity of the immersed specimens is reduced when IGBFS is substituted with ACS due to the ACS nucleating effect, in addition to the role of (Cl− and SO42−) which accelerated the cement hydration.
Fig. 20
Effect of replacement of IGBFS with ACS on the total porosity of hydrated IGBFS-rich cement pastes immersed in seawater up to a year
×
Compressive strength
The compressive strength of the specimens immersed in seawater for a year is plotted in Fig. 21. The results indicated that cement pastes strength improved for up to six months before decreasing for up to a year. The fundamental reason for this is because the pastes activation with Cl− and SO42− in seawater produced additional hydration products, particularly hydrated calcium silicate hydrates which affect the strength positively. Since pozzolanic material and a nucleating agent are presented at the early 6 months of immersion, both activated the hydration of IGBFS-rich cement pastes containing ACS, which results in an increase in compressive strength of B1 at early and later ages of immersion. Due to the larger concentration of ACS with poor hydraulic characteristics, B3 has a lower compressive strength. In general, the acceleration action of both sodium and magnesium salts i.e. Cl− and SO42− on cement pastes is the major cause of the strength improvement up to six months. A portion of Mg2+ is diffused and adsorbed into the C-S–H particles which stimulate its crystallization and lead to strength improvement. As a result of this activation, calcium silicate, chloroaluminate, and sulphoaluminate hydrates (ettringite minerals) are produced. After six months up to 12 months in seawater, the compressive strength begins to diminish as more calcium chloroaluminate hydrates and ettringite forms as represented in Fig. 21. The visual photos of pozzolanic cement pastes of A1, B1 and B3 cured at 6 and 12 months are given in Fig. 22.
Fig. 21
Effect of replacement of IGBFS with ACS on the compressive strength of hydrated IGBFS-rich cement pastes immersed in seawater up to a year
Fig. 22
The visual photos of pozzolanic cement pastes of A1, B1 and B3 cured at 6 and 12 months
×
×
FT-IR spectroscopy
The reduced intensity of the portlandite band at 3454 cm−1 at 12 months is shown in Fig. 23 which displays the FT-IR spectra of A1 at 0, 6 and 12 months of immersion in seawater. This is explained by the IGBFS's pozzolanic reaction with portlandite involved in cement hydration. CH is converted into CaSO4.2H2O and Mg(OH)2 by the acidic action of MgSO4 in seawater. A further portion of Ca(OH)2 is also broken down by MgCl2 to produce Mg(OH)2.
Fig. 23
FTIR spectroscopy of A1 (30% OPC + 70% IGBFS) immersed in seawater for 0, 6 and 12 months
×
XRD analysis
The XRD patterns of A1 at 0, 6, and 12 months are presented in Fig. 24. The hydration process causes the concentrations of Ca(OH)2, CaCO3, C-S–H, -C2S, and C3S to drop with time. 28-day immersion in tap water does not result in the formation of calcium chloroaluminate hydrate. On prolonged hydration, the intensity of the peak corresponding to the calcium chloroaluminate hydrate increases. Additionally, due to the pozzolanic reaction of IGBFS with portlandite, CH is entirely consumed after 6 months of immersion. Until a year, anhydrous phases such β-C2S and C3S are still present.
Fig. 24
XRD patterns of A1 at 0, 6 and 12 months under seawater
×
Thermal analysis of cement pastes immersed in seawater
The effect of seawater attack on the hydration products of IGBFS cement pastes immersed in seawater for up to one year can be studied with the aid of DTA Fig. 25. The thermograms indicate different endothermic peaks in addition to one exothermic peak at about 955 °C. The lower endotherms up to 300 °C are due to the decomposition of C-S-H, ettringite and calcium chloroaluminate hydrate. There are other three endothermic peaks at 316, 428 and 459 °C; which appear with the decomposition of C2ASH8, brucite and portlandite, respectively. Two endothermic peaks located in the range of 700–900 °C appear with calcination of the amorphous and crystalline calcium carbonate [68, 69]. The exothermic peak at 956 °C is characteristic of the C-S-H involved with the crystallization of monocalcium silicate CaO.SiO2 (wollastonite).
Fig. 25
DTA thermograms of A1 immersed in seawater for 0, 6 and 12 months
×
The thermogram of mix A1 at zero time shows the C-S-H, AFt, C2ASH8 and hydro garnet series at 315 °C as represented in Fig. 25. There is a characteristic endothermic peak of Ca(OH)2 and endothermic peaks of CaCO3 in addition to an exothermic peak for the crystallization of wollastonite [70]. A1 at 6 months immersion in seawater gained the highest hydration products yield which reflects the ability of seawater to activate the hydration reaction to some extent at an earlier age. The amounts of chloroaluminate, sulphoaluminate. C-S-H and hydrogarnet hydrates series increase with the decrease of the endotherm of C-S-H as well as Ca(OH)2 due to the reaction of MgCl2 and NaCl2 with Ca(OH)2 to form brucite Mg(OH)2, as small endotherm at 438 °C. As the immersing time increases the amounts of hydration products as well as the hydrogarnet and Mg(OH)2 increase with the consumption of Ca(OH)2. This is also a result of Ca(OH)2 and chloride's reaction. The endothermic peaks due to the carbonation are also detected. The exothermic peak at 955 °C is also present, which is sharp at zero month (0 M) (in tap water). Then the intensity and the sharpness of the exothermic peak decrease. This is also a result of chloride or sulphate ions attacking C-S-H, as seen in the equation below.
Therefore, the amount of C-S-H decreases with the CaCl2 content because of the chloroaluminate hydrate growth. Hence, C-S-H decreases at later ages (12 months) as seen from the DTA thermograms.
Conclusion
1.
When ACS was used in place of GBFS, the water consistency of cement pastes marginally decreased.
2.
The initial, as well as final setting time, are prolonged with ACS content, 10–20 wt% ACS gained small elongation.
3.
As the IGBFS decreases on the expense of ACS the accumulated hydration products and hence the combined water decreases and the free lime increases.
4.
It is clear that the IGBFS is more hydraulic than ACS and these values are in according with the determined values of combined water.
5.
All specimens gain compressive strength with curing age up to 180 days and that of 10% ACS show higher compressive strength than those containing 20–30 wt% and without ACS.
6.
The incorporation of ACS at expense of IGBFS tends to increase the bulk density of thermally treated specimens. On the other side, the substitution of IGBFS with ACS decreases the compressive strength up to 800 °C.
7.
The compressive strength improved for all specimens submerged in seawater at an earlier age and then decreases up to a year, the ACS-rich cement paste has the lower compressive strength. In contrast, a 7% reduction in the compressive strength is less than achieved at zero time by IGBFS-rich cement paste.
While this study examined cement pastes with a specific W/C ratio, further work is needed to determine the effects of W/C on the properties of these composite cements. Based on the promising results of this initial characterization, future studies will investigate composite slag cements over a range of W/C ratios. Optimizing both strength and workability properties through adjustments in W/C will be the focus of continued research in this area.
Declarations
Institutional review board statement
The study was conducted and approved according to the guidelines of the declaration of the ethical committee of the Faculty of Science, Benha University (BUFS-REC-2023-22 chm).
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
