Performance of nano titania-reinforced slag/basalt geopolymer composites

In Egypt, basalt manufacturing generates waste materials at a rate of about 33%. The disposal of basalt waste through the manufacturing of cementitious materials is an economically feasible approach. This work involved the fabrication of a one-part slag/basalt waste geopolymer cement (GP) enhanced by nano-TiO2. The physico-mechanical properties of the different GP mixes were investigated. The durability against fire at up to 750 °C, sulfate (SO4−2) attack, and irradiation with significant dosages of γ-rays at up to 1500 kGy were assessed. Phase composition and microstructure were explored through various techniques as XRD, TG, and SEM. The findings demonstrated that basalt incorporation up to 20% within the GP mix boosted its mechanical characteristics after 28 and 90 days of curing by 8.5 and 2.5%, respectively, while 60% replacement diminished the strength by 30%. Slag/basalt GP showed a comprehensible resistance to SO4−2 ions attack, and its strength was intensified by 14% after 4 months of immersion. Additionally, the blended GP mixes could preserve 40% of their strength after being exposed to 1500 kGy of γ-rays and about 50% of their strength after firing at 750 °C. Admixing GP by NT showed a synergic impact for diminishing setting times, boosting compressive strength by 5 ~ 15% compared to reference mix along with an improvement in the microstructure. Furthermore, distributing NT through GP greatly improved durability after firing at 500 and 700 °C and irradiating with high dosages of γ-rays. This feature is mostly connected to the catalytic capabilities and filling activities of NT in boosting geopolymerization processes.


Introduction
There is a significant demand for high-performance and more sustainable construction materials due to the continued economic expansion in the world.One of the most significant building materials used worldwide is Portland cement (PC).Nevertheless, owing to the burning of carbonaceous materials and fuel consumption, manufacturing of PC has run into several environmental issues [1,2].Additionally, the PC production produces thousands of tons of hazardous emissions, including nitrogen oxides, dust, greenhouse gases, and sulfur dioxide [3,4].Scientists have been developing novel ecological and eco-friendly binders to replace Portland cement partially or completely in concrete mixtures [5,6].
Alkali-activated cement or geopolymer cement (GP) is attracting an extra attention as an excellent binder that can replace PC in the building mix.Geopolymers (GPs) are cementitious composites that are free of PC [7].Their production depends on the chemical treatment of aluminosilicate raw materials by alkaline activator through chemical reactions or polycondensation process that results in solidifying products of cementitious properties.The manufacture of geopolymer binders does not require limestone firing, which minimizes CO2 emissions in the surrounding environment [8].Furthermore, industrial solid wastes can be used as primary supplies of alumina and silica in PC manufacture to reduce future accumulating problems [9][10][11][12][13][14][15][16].There are several technical benefits of employing geopolymer as construction binders, including appropriate mechanical characteristics [17] and its better endurance in both acidic and corrosive ionic environments [18,19].Furthermore, those cement binders outperformed OPC in firing actions [20,21].Recently, geopolymer binders have found numerous outstanding incorporates in restoration supplies [22], structure composites [23], and pavement constructions [24].
There are two primary forms of geopolymer, as described in the literature, contingent on how GP is prepared or handled.The first type, known as the traditional or two-part geopolymer, is created when solid aluminosilicate precursors and concentrated solution of alkaline activator are combined [25,26].Nevertheless, there are several hazardous challenges with handling the concentrated alkali activator solution, which are reflected in the expand use of GP created using this method.The second type of geopolymer, known as one part or ready mix, is made by simply adding water to a mixture of an aluminum-silicate source and a dry alkaline activator as a one mix [9,10,27].
Basalt is an igneous rock that originates when lava cools on the surface of the Earth [28].Basalt rock has a high silica content beside alumina, iron, and magnesium oxides.These minerals are present in an active state and change to a more amorphous state when the hot lava cools [29,30].Basalt stone can be used as a filter stone in drainage projects, while the crushed basalt is used to create the foundation for roads and pavements.There are two main basalt resources in South Egypt.About 240,000 m 3 of basalt aggregates and roughly 80,000 m 3 of fine-grained basalt residue are produced each year as a result of the crushing process.So about 33% of the basalt total production is produced as a waste material [31].Basalt waste (BW) can be viewed as a natural pozzolana due to its characteristics as supplemental cementitious material (SCM) to OPC that have been widely researched [32,33].Numerous studies have recently attempted to use BW in the manufacture of alkali-activated binders or geopolymer composite [34][35][36].However, a crucial factor for extending the use of basalt-based GP is its resistance to various degradation acts, which calls for more research.
A prevalent technique to promote a denser packing, lessen the tendency for cracks, and improve the compactness of the granular structure is the use of microand nanoparticles in cementitious materials.Examples of these materials with microscopic reinforcing activities are nano titania (NT), nano alumina (NA), silica (NS), carbon fibers (CF), and nano clays (NC) [37,38].NT has a wide range of uses, including the manufacture of photovoltaic solar cells [39,40] and the photocatalytic destruction of contaminants in wastewater [41].When micro-and nanoparticles are included into the geopolymer matrix, they can increase their effective volume within the first hour of reaction and densify the matrix, creating a building structure with greater strength and better durability.
Few investigations have been conducted so far on the durability performance of one-part geopolymers made from basalt waste.The objective of this study is to determine how high temperatures, sulfate attack, and gamma radiation doses affect the mechanical properties and morphological characteristics of a manufactured one-mix geopolymer made from significant quantities of basalt waste as a distinctive and environmentally friendly cementitious material.One-mix geopolymer is more likely to be utilized as a cement substitute with the suggested method.In order to improve its performance and durability against high temperatures, gamma radiation, and sulfate attack, manufactured GP was reinforced with 0.5 and 1.0 wt.% nano-TiO 2 .

Materials characterization
The following materials were adopted in this study: • Ground granulated blast-furnace slag (GGBFS), a precursor for aluminumsilicates that is rich in calcium, was used.It was originated from an iron and steel factory in Helwan, Egypt.It has a 2883 cm 2 /g Blaine-specific surface area.• The crushed basalt came from the Shuq El-Thouban, Egypt.It was ground and sieved to particles with a Blaine surface area of 2643 cm 2 /g and particle size of ≤ 45-mm basalt waste (BW).• Figure 1 shows XRD patterns of BW and GGBFS and BW as determined by XRD instrument (model Xpert-2000, Philips), whereas   The adsorption/desorption isotherm of nano-TiO 2 (NT) is identified as a type 2 with a small H1 hysteresis loop according to IUPAC classifications.As shown by the monolayer capacity (V m ) of 2.0 cm 3 /g and the average pore diameter of 28.29 nm, the pores in NT are in the meso/micro range.Additionally, the total pore volume (Vp) was determined to be 0.0616 cm 3 /g, and the BET surface area (S BET ) for nano-TiO 2 was estimated to be 87.15 m 2 /g.On the other hand, Fig. 2b shows Barrett-Joyner-Halenda (BJH) curve of pore size distribution by volume for NT.The majority of the NT pores are evidenced to be mesopores according to the data of Fig. 2b, where dp max = 40 nm.SEM micrographs of nano-TiO 2 illustrated in Fig. 2c reveal that the NT particles are extremely tiny and spherical, with a consistent size distribution and a diameter that varied from 20 to 80 nm.

GP mix design
Along with the TiO 2 nanoparticles, two precursors were used in the current investigation.Various dry blends were developed by substituting GGBFS by 0, 20, 40, and 60 wt.%BW.The solid alkaline activators were added to the dry blend at a rate of 15 wt.% (5% NaOH, 10% Na 2 Si 2 O 3 ) of the total mass.To guarantee the homogeneity of the mix, each dry mix (containing the dry raw precursors and the solid activators) was stirred for 6 h in a porcelain ball mill.Designation and composition of the examined dry mixes are disclosed in Table 2.
Each dry mix was mixed with water to create geopolymer pastes.All of the created pastes had the same consistency, suggesting that substituting BW for GGBFS did not affect the workability of GP paste as well as the water/solid (W/S) ratio necessary for standard consistency.To assure the uniform distribution of the nanoparticles in the mixing water within the pastes containing nano-TiO 2 , they were initially suspended in a mixture of water along with 3 wt.%superplasticizer, and the suspension was sonicated by ultrasonic homogenizer (LUHS0A12, 650W, 220 V/50HZ) for 1 h [43].To begin the geopolymerization processes, the dry mixes were combined with water, and the resulting pastes were molded into cubic specimens utilizing one-inch cubic molds and kept overnight at RH≈98 ± 2%.
The hardened cubic specimens were taken out of the molds after 24 h and were cured under humid conditions (≥ 99%) to retain the geopolymerization processes [44].

Testing routine
According to ASTM C191, the Vicat apparatus was used to measure the initial and final setting times for all the freshly developed geopolymer pastes [41].This apparatus comprises a frame that contains a mobile rod, equipped with a disc at one side and a needle that can be affixed at the opposite end.
Compressive strength tests were done using a Ton Industrial Machine (West Germany) for maximum load of 60 tons.The test was performed on three cubic specimens, for each geopolymer mix, after 3, 14, 28, and 90 days to monitor the strength that had evolved, according to ASTM C109 M16-a [13,45].
After 3 and 28 days of curing, total porosity (P, %) and bulk density (BD, g/cm 3 ) were performed by measuring the following weights of the hardened cube that represent each GP mix: the weight of the cube suspended in air (saturated surface dry) (W1), its weight immersed in water (W2), and then the weight of the cube after drying at 100 °C for about 24 h (W3).Bulk density and total porosity were calculated using formulae 1 and 2:

Fire resistance
The thermal resistance test was performed on the 28-day cured samples.Three cubes which represent each GP mix were dried in an oven at 90 °C overnight.They were then fired for 3 h at temperatures of 250, 500, and 750 °C in a muffle furnace at a heating rate

Sulfate attack test
Three cubes for each GP mix, after hydration for 28 days, were submerging in MgSO 4 solution (7%) for up to 4 months, to evaluate the detrimental effects of sulfate ions penetration.Samples were removed after 1, 2, and 4 months of curing in sulfate solution, and their compressive strength was evaluated, and the average value was adopted.

Irradiation by γ-rays
Resistance of GP mixes to irradiation of γ-rays was tested by exposing the dried 28-day samples to a gamma-ray source (Co 60 , dose rate 985 kGy/h) applying 500, 1000, and 1500 kGy with a dosing rate 1.4 kGy/h.After irradiating the samples, the compressive strength test was assigned after the sample had been exposed to radiation, and the residual strengths (RS %) rad were determined using Eq.4: Where CS rad is the compressive strength after radiation given dosage of γ-ray, CS 0 is the compressive strength prior to radiation (28-day value).

Phase composition and microstructure
GP phases that developed during the hydration of the various geopolymer specimens were identified using XRD and thermogravimetric techniques.XRD was performed using cobalt target (λ = 0.17889 nm) and filter made of nickel under 40 kV and 40 mA.The scanning range spanned from 5 to 60 (2θ°), with a scanning speed of 1 s/step and a resolution of 0.02/step.TGA test was done using TGA:TA instrument, model SDT Q600.A nitrogen environment was maintained while heating 15 mg of powdered sample (≤ 25 µm) in the temperature range of 50 to 1000 °C, with a heating rate of 10 °C/min.The microstructure of various GP samples was explored using SEM-Quanta 250 FEG equipment.

Setting times
Setting times are thought to be crucial factors that can impact how the tested materials are handled and what applications they can be used for [9].The initial and final setting times of geopolymer mixes without and with basalt are shown in Fig. 3. Initial and final setting times are found to be 25 and 44 min, respectively, for the geopolymer mixture consisting entirely of GGBFS (mix M).One-part geopolymer sets up more quickly than OPC owing to the heat released during the solid NaOH activator's dissolution [46].Furthermore, the interaction between GGBFS and alkaline activators is an exothermic reaction in which the produced heat promotes the geopolymerization process [47].
Replacement of GGBFS by BW causes an enlargement in the setting time.Additionally, the enlargement of setting times becomes greater as replacement percentage of BW increases, reaching about 280% for the initial and 190% for the final setting times for mix M60 (40% GGBFS + 60% BW compared to control mix (M, 100% GGBFS).This discovery is consistent with earlier research and may be explained by the fact that the geopolymer mixture's Si/Al ratio decreased as the basalt percentage increased which is a crucial factor in governing both setting times and compressive strength measurements [48].Table 2 shows that the Si/Al ratios for M, M20, M40, and M60 are 5.45, 5.31, 4.91, and 4.17, respectively.The quantity of soluble Si that stimulates the geopolymer reaction is decreased by lowering the Si/Al ratio [34].However, these blended mixes exhibit setting times that are satisfactory for the ASTM-C150M standards, demonstrating the proper application of these geopolymer blends as an alternative to conventional OPC [49].
Furthermore, setting times significantly decrease by adding 0.5 or 1% of nano-titanium oxide (NT) to mix M20 (80% GGBFS + 20% BW).Evidently, seeded nucleation occurs due to NT particles' dispersion in the one-part geopolymer matrix [50].According to Hajimohammadi et al. [51], this nucleation inhibits Al from adhering to silica particles, which regulates silica release and speeds up the development of products with high silica phase content in the early geopolymerization processes.

Compressive strength
The strength of any cementitious component determines its appropriateness as a construction material.Figure 4a shows the compressive strength variation for mixes M, M20, M40, and M60 with geopolymerization time.As the curing period increases, the compressive strength of the control mix (100% GGBFS, mix M) increases continuously, − .These species operate as monomers, initiating geopolymerization reactions that result in strength-giving phases like CSH and CASH gels [13,52].The gained strength often diminishes when GGBFS is replaced by BW.As the BW percentage is increased from 20 to 60% in the mix, the strength decreases.This occurs because of the replacement of slag (a material with a high calcium content) with BW (a substance with a high silica content), which results in the creation of hydration products like sodium aluminosilicate hydrate (NASH) gel [53].NASH has weaker binding characteristics than calcium silicate hydrate (CSH) gel [54].Nevertheless, mix M20 exhibits higher strength at 28 and 90 days than the control mix, whereas mix M40 and M60 provide CS at 28 and 90 days that are around 70% of control values.This demonstrates that the M20 mix represents the optimal ratio of soluble Ca +2 ions provided by GGBFS which interacts with silicate ions from BW leading to the formation of an extra gel that provides strength [55].
Figure 4b shows the compressive strength of geopolymer blends comprised of 80% GGBFS + 20% BW and admixed with 0.5 or 1% nano-TiO 2 (M20-T1 and M20-T2, respectively).Evidently, the strength of M20-T1 and M20-T2 composites is improved when compared to M and M20, especially at early curing ages.Such finding is consistent with the outcomes of setting times and may be relevant to the following: (i) the effective dispersion of NT in the matrix of geopolymer and (ii) the substantial surface area (87.15 m 2 /g) and superior fineness (~ 20-80 nm) for NT.The combination of the above factors causes the geopolymerization of the precursor materials to start rapidly.Additionally, due to their filling activity, NT tiny particles reduce porosities and serve as catalysts for geopolymerization reactions.Combining these traits causes the hydration products with strong binding characteristics to accumulate, increasing the strength [42,45,56].

Bulk density and porosity (%)
The variation of the geopolymer system's bulk density (BD, g/cm 3 ) and porosity (P, %) with curing time is an indication of the progress of the polymeric reactions and hence the strength development.Figure 5a and b shows the BD and P% of M, M20, and M20-T2 mixes with compressive strength after 3 and 28 days of curing.Deposition of the various geopolymerization products leads to a subsequent clogging of the accessible pores (reduction in P%) and an increase in the bulk density from 3 to 28 days in all investigated mixes.After 3 days, the blended mix (M20) possessed higher porosity and lower BD than the control mix (M), and after 28 days, the behavior changed, showing that the geopolymer system's produced hydrates have changed in both quantity and type as a result of the partial replacement of GGBFS by BW [57].Such behavior is in line with the obtained compressive strength of these two combinations.Among the examined mixes, M20-T2 composites had the lowest P% and the greatest BD and CS after 3 and 28 days, indicating that this composite has the best-consolidated structure.This demonstrates how the nanoparticles serve to speed up the geopolymerization process, especially in the early stages [58].

Fire test
Figure 6 shows the residual strength (RS) t for specimens M, M20, M40, and M20-T2 upon fire at 250, 500, and 750 °C for 3 h and progressive cooling to ambient temperature in desiccators, and Fig. 7b shows the visual appearance of GP samples after firing at 500 °C.Evidently, firing at 250 °C boosts compressive strength for M, M20, M40, and M20-T2 by 10, 12, 5, and 22%, respectively, compared to their strengths at 28 days.The unreacted GGBFS grains and the alkaline hydroxyl anions are activated by heating the hardened geopolymer mix at 250 °C, which increases the steam pressure inside the geopolymer matrix and continues the hydration and geopolymerization processes [30,43,58].As a result, additional binding products such as CSH gel, CASH, NSAH, and NCASH will be generated.The aggregation of these compounds improves the geopolymer mix's mechanical characteristics.These outcomes agree with earlier publications [59,60].Compressive strength for the M, M20, and M40 mixes decreased slightly after firing at 500 °C by 8.3, 4.2, and 14.8%, respectively, yet the composite M20-T2 still gained strength by around 5% exceeding its 28-day value.When GP is heated to 500 °C, the geopolymer matrix suffers thermal degradation, and the unstable phases dissolve, resulting in low (RS) t %.This decrease in (RS) t is mostly dependent on the type and/or quantity of geopolymer hydrates, which is connected to the constitution of the starting precursors.This explains the variations in (RS) t across the studied mixes.However, the M20-T2 composite's RS t at 500 °C is still greater than its original value after 28 days of hydration, indicating the presence of additional quantities of hydrates with reduced porosities and the contribution of NT particles to the improvement of this composite's thermal stability [61,62].Upon firing at 750 °C, all GP mixes show a significantly decrease in compressive strength.RS t for M, M20, M40, and M20-T2 mixes were 46.2, 41, 53.2, and 22%, respectively.Evidently, at 750 °C, dehydration/dehydroxylation of all the produced hydrates took place, causing the pore matrix's high pressure to create microcracks, which greatly diminished the mixes' mechanical characteristics as well as their residual strengths (RS) t .Again, at this high temperature, mix M20-T2 experiences the least strength loss.Such findings will be validated by XRD and SEM analyses, which will be presented in the next sections.SO 4 −2 ions penetration M, M20, and M20-T2 mixtures were subjected to the attack by SO 4 2− ions test by submerging the samples in MgSO 4 solution (7%) for up to 4 months.Figure 7 illustrates the results of the compressive strength measurements at 1, 2, and 4 months of immersing in MgSO 4 solution.For the three examined mixes, the behavior of GP mixes in MgSO 4 solution exhibits different tendencies.In the first month of curing, mix M showed an increase in CS, followed by a progressive failure in strength.M20 mix showed a similar trend, but the boosting in CS extended to 2 months of immersion.However, the CS of M20-T2 remained almost unaltered for the first 2 months before increasing after 4 months of curing.Attack of any hardened cementitious material (CSH or CASH) by SO 4 −2 ions is resulting in the development of expansive hydration products like Ettringite [63,64].The accumulation and precipitation of these hydrates inside the pores can improve the strength after short time of attack by sulfate ions.Meanwhile, as we observe in mix M, the accumulation of these hydrates causes internal stress and the creation of fractures owing to continual immersion in sulfate ions [65].The nature of hydration products produced in the hardened mix and its pore system primarily determine the degree to which CS is deteriorated by SO 4 −2 attack.The resistance of this mixture to strength depletion driven on by attack by SO 4 −2 ions is explained by the improved pore system in M20-T2.Similar observations were made with fly ash-based geopolymer concrete, which indicated that there was less of a loss in CS following immersion in solutions of sulfate and chloride [47,66].

Shielding test
The shielding characteristics of any construction material are critical in determining its potential use in a wide range of applications.Figure 8a shows the residual strength (RS) rad % of mixes M, M20, and M20-T2 after exposure to a γ-ray source ( 60 Co-γcell-220, Atomic Energy Commission, Canada) with varying intensities (500, 1000, and 1500 kGy), while Fig. 8b depicts the visual appearance of GP samples after 1500-kGy irradiation.Obviously, increasing the power of an irradiated gamma ray reduces (RS) rad , for all the studied GP mixes, suggesting the destructive characteristics of irradiated rays to target hydration products and a subsequent negative influence on compressive strength [67,68].When compared to the control (mix M), mix M20 exhibits a resistant behavior, particularly at high irradiating doses of 1000 and 1500 kGy.This finding can be related to the hydration/geopolymerization products created in mix M20 (low Ca ion and high Si/Al ratio) as opposed to those formed in mix M (control).According to Al-Hamarneh [69] and Lambertin et al. [70], irradiation can readily create additional binding phases by cross-linking the aluminosilicate's linear chains.Incorporating 1% nano-TiO 2 into M20-T2 composite had a beneficial effect on the mix's shielding characteristics at all the measured γ-ray dosages, as evidenced by the high residual strength (RS) rad .The shielding resistance in this case is due to the greater hydration gel generated in this composite compared to M or M20.Actually, NT particles operate as an activator, promoting geopolymerization and generating more gel-binding products.

XRD
XRD patterns for mixes M and M20 after 3 and 28 days of geopolymerization processes are shown in Fig. 9. XRD diffractograms of mix M after 3 days of geopolymerization displayed peaks at 2θ = 29.54°(PDF nos.00-033-0306, 00-034-0002) and 32.21° (PDF no.00-020-0452), which could be attributed to CSH gel and hydrogarnet (CASH), respectively, as the principal hydration products [13,68].The occurrence of such phases in the geopolymer network of M mix can be attributed to an excessive amount of Ca +2 ions derived from GGBFS.It has been terminated that the presence of calcium ions induces the precipitation of CASH, which serves as nucleation sites for geopolymerization process [71].Additionally, CSHs and CASH established an essential framework of percolating solids that triggered the hardening process [72].Furthermore, peaks of the illite phase and unreacted quartz (2θ = 29.64°)appeared.Nevertheless, the XRD pattern of mix M20 (80% GGBFS + 20% BW) after 3 days of geopolymerization shows peaks for new hydration products, specifically garronite (NCASH) at 2θ = 32.2°(PDF no.01079-2424) and tetragonal Na-alumino-silicate hydrates (NASH) at 2θ = 27.8° and 32.1° (NASHs, PDF nos.00-039-0217 and  [73].The production of these phases is influenced by the Si/Al ratio in the polymeric system as well as the cations obtained from the alkaline activator.As previously mentioned, NCASH may arise in a polymeric system as a result of the transition of gismondine, a highly calcium-rich phase, into NCASH via the exchange of Ca 2+ ions for Na + ions in a highly concentrated alkaline medium [59].After 28 days, the peak intensities of the hydrated phases for mixes M and M20 grew, while those of the unreacted dropped indicating the continuity of polymerization processes.XRD patterns of mixes M and M20 upon firing at 250 °C and 750 °C, are shown in Figs. 10 and 11, respectively.The sharpness of the peaks associated with CSH gel and (CASH) increases in the XRD pattern of mix M at 250 °C, demonstrating the rise in quantities of these products during the self-autoclaving process [66].However, burning at 750 °C causes nearly disappearance of hydrated phases, as evidenced from the XRD pattern, and dehydrated phases like gehlenite (Ca 2 Al 2 SiO 7 , PDF no.01-079-2423) appeared, which have poor mechanical characteristics (Fig. 11).Similar findings to those of mix M can be seen in the XRD patterns of M20 at 250 °C, where there is a reduction in the intensities of NASH-related peaks.At 750 °C, however, additional peaks in the XRD pattern can be observed that may be connected to sodium-rich phases which are thermally stable, such as nepheline (NaAlSiO 4 , PDF no.01-070-1582) [74].
Figure 12 shows XRD diffractograms of composite M20-T2 after 3 and 28 days of geopolymerization as well as after firing at 250 °C and 750 °C.Evidently, mix M20-T2 showed the identical peaks of mix M20 with relatively high intensities and almost no unhydrated phases, demonstrating that the NT dispersion in the geopolymer matrix enhances and speeds up the hydration/geopolymerization processes [75,76].This also implies that several hydrated products, such as CSH, CASH, and NASH, continued to

TGA
TGA diagrams for mixes M, M20, and M20-T2 after 3 and 28 days of geopolymerization are shown in Fig. 13a and b, respectively.A progressive mass loss by heating to 1000 °C is for the three mixes after 3 days.The mass loss percentages (%) for M, M20, and M20-T2 were 6.19, 7.73, and 8.53%, respectively, reflecting the quantities of hydration/geopolymerization products produced in these mixtures.M20-T2 offers the highest amounts of hydrates among all mixes which confirm the catalytic role of NT to promote the geopolymerization reactions.The mass loss in the three samples is detected at three temperature stages: 50-300°, 300-600°, and 600-950 °C.Table 3 shows the mass loss % for all mixes at various stages.The first stage of mass loss occurs between 50 and 300 °C and is due to the release of unbinding water as well as the partial disintegration of CSH and CASH gels.The second stage, which accounts for the main of mass loss, occurs between 300 and 600 °C and is triggered by the dehydroxation of hydrogrant (C 3 ASH 4 ) and hydrotalcite (Mg 6 Al 2 (CO 3 ) (OH) 16 ⋅4H 2 O) [73].Nevertheless, the third stage, which experiences the least mass loss %, is associated with decarbonation of CaCO 3 , which occurs as a consequence of carbonation of the samples during handling [77].All samples possessed an increase in mass loss % after 28 days of curing, suggesting that the geopolymerization processes were progressing.For mixes M, M20, and M20-T2, the respective mass loss percentages were 8.31, 10.27, and 10.0%.

Microstructure
SEM micrographs of mixes M, M20, and M20-T2 after 3 and 28 days of curing are shown in Fig. 14((a1, a2), (b1, b2), and (c1, c2)), respectively.After 3 days, the SEM picture of the mix M shows a semi-compact structure with unreacted spherical slag particles and solid alkaline activator (as hexagon plates) as well as newly developed phases such as short CSH gel and cubic plates of CASH (Fig. 15a1) [78].The mixture got more solidified after 28 days due to the deposition of hydrates, which formed a cylinder of geopolymer chains that joined the unreacted slag particles (Fig. 15a2).Such photos demonstrate the geopolymerization process's development and explain the improvement in mechanical characteristics.For mix M2, analogous observations are seen after 3 and 28 days (Fig. 14b1, b2, respectively).When comparing the geopolymer matrix for mix M20-T2 composite to those of mixes M and M20, SEM  photos clearly demonstrate that the geopolymer matrix is more compact, and that the quantity of CSH gels is increased, especially after 3 days of curing, demonstrating the beneficial effects of NT on GP structure (Fig. 15c1, c2) [66].On the other hand, SEM micrographs of mixes M, M20, and M20-T2 after firing at 250 and 750 °C are shown in Fig. 15((a1, a2), (b1, b2), (c1, c2), respectively).After firing at 250 °C for 3 h, SEM pictures of mixes M, M2, and M20-T2 samples (Fig. 16a1, b1, and c1, respectively) reveal the production of larger quantities of CSH and CASH gel which surrounded the uncreated grains.These phases facilitate the improvement of the mechanical characteristics for these mixes.This observation agrees well with the mechanical features that have already been mentioned.Additionally, firing GP samples at 250 °C seemed to have an impact on the morphology of the produced hydrates as evidenced by the emergence of semi-crystals and short fibers of CSHs linking together inside the pores in mixes M and M20 or chain formation in mix M20-T2 micrographs.In micrographs of mixes M and M20 after firing at 750 °C (Fig. 16a2, b2, respectively), significant macropores and fractures are seen that have a deleterious impact on the mechanical characteristics [79].On the other hand, mix M20-T2 exhibits minimal microcracks in SEM images at 750 °C, demonstrating the heat resistance of such composite (Fig. 15c2).The impact of SO 4 −2 ions on GP mixes due to their immersion in MgSO 4 solution is clearly noticeable in their SEM micrographs.SEM images of the M20-T2 composites reveal hexagonal plates of gypsum, whereas images of the M and M20 mixtures show ettringite (E) as crystalline needles [80] (Fig. 16).

Conclusions
The following conclusions may be drawn from our investigation's findings: 1. Blast furnace slag and basalt waste powders can be used to create one-part geopolymer cement with recognized mechanical properties and durability against distinct degradation processes.2. The one-part geopolymer that was created had shorter setting times than OPC, but it nevertheless matched to ASTMC 150 M specifications.3. Greater blending ratios (up to 60%) diminished the strength by around 30% after 28 and 90 days of curing, whereas substituting up to 20% of the slag with basalt powder improved the mechanical properties.4. A geopolymer composed of 80% slag and 20% basalt exhibited resistance to gamma radiation, particularly at high doses (1000 and 1500 kGy).A suitable level of durability in MgSO4 solution and fire action up to 750 °C are also featured.5. One-part slag/basalt-based geopolymer composite with 1% NT retains 100% and 65% of its strength when fired at 500 and 750 °C, respectively.In addition, this mixture is resistant to the effects of SO 4 −2 ions and large doses of gamma radiation.This is related to the catalytic action and filling properties of nano-TiO 2 .

Fig. 1
Fig. 1 XRD patterns of a basalt and b GGBFS

Fig. 2 a
Fig. 2 a Adsorption-desorption isotherm, b BJH curve, and c & d SEM images of NT

Fig. 3
Fig. 3 Setting times of the various GP composites

Fig. 4
Fig. 4 Compressive strength (MPa) of the various GP composites.a Without and b admixed with NT

Fig. 5 Fig. 6 aFig. 7
Fig. 5 Correlation between a total porosity (P, %) and b bulk density (BD, g/cm 3 ) with compressive strength of GP composites after 3 and 28 days of curing

Fig. 8 a
Fig. 8 a strength (RS) rad of the GP composites after irradiating by various doses of rays.b Visual appearance of various samples after irradiating by 1500 KGy

Fig. 9
Fig. 9 XRD patterns of mixes M and M20 after 3 and 28 days of curing

Fig. 10
Fig. 10 XRD patterns of mix M after firing at 250 and 750 °C

Fig. 12
Fig. 12 XRD patterns of mix M20-T2 after 3 and 28 days of curing and after firing at 250 and 750 °C

Fig. 13
Fig. 13 TGA curves for the various mixes after a 3 and b 28 days of curing

Fig. 16
Fig. 16 SEM micrographs of mixes.a M, b M20, and c M20-T2 after immersion in MgSO 4 solution for 4 months

Table 1
Chemical oxide composition of the raw materials

Table 2
Mixes notation and their composition of 10 °C per minute.The burned specimens were tested for compressive strength after being cooled progressively to room temperature in desiccators.Applying Eq. 3, residual strength (RS) t was evaluated.

Table 3
Mass loss % of various GP mixes