Strength and Durability Performance of Mortars Incorporating Calcined Clay as Pozzolan

doi:10.21203/rs.3.rs-1983330/v1

The use of low-grade calcined clays is the only potential material available in large quantities to meet the requirements of eco-efficient supplementary cementitious materials by decreasing the clinker content in blended cements or decreasing the cement content in concrete.

In the present paper, six mortar mixtures with a constant w/b ratio were used to investigate mechanical and durability properties of mixtures substituted with low-grade calcined clays with/without limestone compared to PC and silica fume mixtures. The present research have shown that using calcined clays with low or intermediate purity (i.e., kaolinite content) leads to enhanced durability properties besides the comparable strength development characteristics compared to PC mortar. The results also indicate the superiority of low or intermediate calcined clay (> 40% kaolinite content) binder over silica fume binder in producing durable mortar mixture, especially in a chloride laden environment. The major reason for the better performance is attributed to the refined pore structure and dense microstructure of the cement paste with the calcined clay pozzolan compared to PC and silica fume.

Calcined clay

durability

Kaolinite

Limestone

chloride diffusion

The use of supplementary cementitious materials (SCMs) has a high potential to enhance strength characteristics, improve durability issues, reduce carbon emissions and intact resource utilization in cement and concrete production, especially for developing countries. Additionally, it was reported that the cement industry is responsible for approximately 5–8% of the world CO₂ emissions (Flatt et al., 2012; Vizcaíno et al., 2020). A forward-looking solution and strategy for decreasing CO₂ emissions during the cement manufacturing stage is reducing the clinker-to-cement ratio by substitution of a percentage of supplementary cementitious material as clinker. However, the most replacement value for pozzolans is about 35% and has been proved to be insufficient to drastically reduce the worldwide carbon emissions from cement production (Schneidera et al., 2011).

Today, in the cement manufacturing, more than 80% of the supplementary cementitious materials for replacement of cement clinker are: limestone, fly ash or slag (Fernandez et al., 2011). It was also reported that substitution of slag and fly ash, regardless of their quality, has particular problems (Fernandez et al., 2011). Regarding limestone, it has been reported that adding more than 10% limestone to cement leads to increased porosity and reduced durability (Matschei et al., 2007). In the case of natural reserves of pozzolans as alternative choice, it can be stated that reserves of natural pozzolans (i.e. volcanic ash or zeolite tuffs) are not widely distributed everywhere and they are only available in certain regions of the earth (Rodriguez-Camacho and Uribe-Afif, 2002). Therefore, their transportation for use in cement and concrete industry, especially for long distances, creates a great economic and environmental burden. Therefore the development of new cement based materials with the evaluation of pozzolanic activity is always of interest. In addition, there is also a shift and tendency to seek alternative SCM sources due to supply-and-demand concerns within the future. In this context, one of the most promising alternative sources for cement substitution is calcined clays since they have not yet reached their full potential. In fact, clay reserves with poor or medium purity in the world are abundant and widespread. Therefore transportation costs for the production of cement are often reduced considerably no matter other benefits like energy saving (Schneidera et al., 2011).

Calcined clay as a supplementary cementitious material has great potential to develop the utilization of SCMs as a partial replacement of clinker in cement (particularly in combination with limestone) or partial replacement of cement as pozzolan in concrete (Scrivener et al., 2018; Samet et al., 2007; Avet and Scrivener, 2018; Tironi et al., 2013; Hollanders et al., 2016). Based on available data, it is possible to obtain pozzolanic materials when clays are thermally heated within the range of 550–950°C (Hollanders et al., 2016; Tironi et al., 2012; Alujas et al., 2015; Avet et al., 2016). At this temperature, the crystal structure of the clay minerals is destroyed and an amorphous or disordered alumino-silicate structure is created, developing pozzolanic properties (Samet et al., 2007). It must be noticed that the need to heat the clays, as a downside to their use, has made them generally more expensive than other pozzolans, such as slag, fly ash, or natural pozzolans if such materials are locally available. To solve this problem, the combination of calcined clay and limestone, as a material with low cost and high availability, irrespective of satisfactory mechanical and durability results obtain when utilized in concrete, helps to reduce the cost of the calcination process. Recently, extensive interest has been assigned to research about calcined clay-limestone blend cement. The pozzolanic reaction of calcined clay and also the reaction of limestone with alumina phase make a synergetic influence within the hydration process (Avet et al., 2016). In this regard, it was reported that the cement, consisting of a mixture of calcined clay and limestone, provides equivalent strength to ordinary Portland cement even at a 50% clinker substitution (Antoni, 2013).

Calcined clays from different locations in the world have been largely studied (Fernandez et al., 2011; Antoni et al., 2012; Badogiannis et al., 2005; Vejmelkova et al., 2012; Castillo et al., 2010; Akindahuns et al., 2020). Clay deposits are usually composed of a mixture of different clay minerals (kaolinite, illite, montmorillonite, etc) and a large proportion of impurities of non-clay materials such as quartz, mica, feldspars, calcite, anatase, and sulfides. The pozzolanic reactivity of calcined clays depends mainly on the type and amount of clay minerals (especially kaolinite content) in the raw material (which is also directly related to the SiO₂ and Al₂O₃ content) and the degree of structural disorder resulting from thermal activation, the amount of thermal activation temperature and conditions governing the heating and cooling process (Alujas et al., 2015). The chemical reactions involved when calcined clays are used as a replacement of cement in concrete have been discussed in the previous study (Dousti et al., 2017). The preliminary reaction is between the AS₂ and the CH produced from cement hydration. As a result of this reaction, cementitious aluminum-containing CSH gel, accompanied by crystalline products, which include calcium aluminate hydrates and alumino-silicate hydrates (i.e. C₄AH₁₃, C₂ASH₈, and C₃AH₆) is formed. Currently, especially in the concrete industry, the use of calcined clays in the world is limited to the production of pure kaolinite clays for the production of metakaolin. Most previous studies on the effect of metakaolin on hydration have used relatively pure metakaolin (Antoni et al., 2012), but it should be noted that these types of materials are not economically acceptable because they typically cost around 3 times the price of Portland cement in the world. In contrast, there is little information available about calcined clays with lower calcined kaolinite. Calcined kaolinite, or metakaolin, has been well studied in recent years and has been demonstrated to have very good pozzolanic properties (Fernandez et al., 2011; Scrivener et al., 2018; Dousti et al., 2017). Its effective use in concrete has been well established; for example, it has been shown that the substitution of cement by 5– 10% of metakaolin can significantly improve the mechanical and durability properties of concrete (Singh and Garg, 2006; Zhang and Malhotra, 1995; Shekarchi et al., 2010). Recent studies have shown that, after optimization of calcination, cement substituted by 30% low-grade clay, with kaolinite content below 50%, can have the same mechanical strength as Portland cement at 28 days (Antoni, 2013).

This study reports a detailed investigation of the influence of two intermediate available calcined clays with or without limestone as a replacement for cement on the mechanical and durability properties of mortar mixtures. The level of two grades of calcined clay and limestone in cement replacement was fixed at 30% and 15% by weight respectively based on the results of previous investigations (Scrivener et al., 2018; Avet and Scrivener, 2018; Tironi et al., 2013; Avet et al., 2016). Properties including compressive strength, water and chloride-ion permeability, as well as electrical resistivity and carbonation resistance were investigated.

Materials and characterization

Two natural clays (KZ kaolin from Zonouz and KG kaolin from Ghazvin province) from the Northwest and center of Iran were used as raw materials. The chemical and mineralogical composition was determined by X-Ray Fluorescence (XRF) and X-Ray Diffraction (XRD). XRD data were collected on Philips PW 1050 diffractometer using Cu-Ka graphite-monochromatized radiation (k = 1.5418 Å). Measurements were carried out in the 2θ range 5–40^o. In order to determine the kaolinite content of calcined clays, thermogravimetric Analysis (TGA) using the tangent method (Scrivener and Snellings, 2016) was used on raw and calcined materials according to Table 1. The cement used was Portland cement equivalent to ASTM Type II. Silica fume (SF) obtained from Azna ferro-silicon alloy manufacture was used for comparison purposes with calcined clay mixtures. Results of chemical composition for all cementitious materials are presented in Table1. It must be noticed that the two clays used in this study before the thermal treatment meet the chemical requirements stated in ASTM C618 for Class N pozzolan.

Figure 1 Shows X-ray diffraction (XRD) spectra of two clays and the mineralogical species are identified. For both clays, XRD patterns display distinct peaks of kaolinite (K) and quartz (Q), accompanied by a weak main peak of microcline (M). As can be seen, the intensity of the peaks determined to kaolinite decreases from KZ to KG-clay and agrees well according to the kaolinite content of the raw clays which was determined by Thermo-gravimetric Analysis (TGA) using the tangent method (Scrivener and Snellings, 2016) and is reported in Table 1. TG and DTG curves for the two raw clays are presented in Figure 2. Mass loss in the range of 30–180°C can be accompanied by drying of the adsorbed water. De-hydroxylation of kaolinite including the removal of the OH groups from kaolinite occurs at about 400–650^oC range.

Thermal treatment and characterization of calcined clays

The thermal treatment of clays was carried out in a laboratory programmable furnace. The two clays were heated from ambient temperature up to 800°C, a recommended temperature that transforms kaolinite into an amorphous phase (Avet and Scrivener, 2018), and then remained for 60 min at 800°C. Subsequently, samples were cooled down rapidly by removing the crucibles from the furnace and spreading the material on a steel plate at laboratory temperature (23±2)^oC. As result, two types of KZ calcined clay (CCZ) and KG calcined clay (CCG) were obtained. XRD patterns of calcined clays (Figure 3) indicate that peaks corresponding to kaolinite disappeared completely for KZ and KG patterns, confirming that heating raw materials at the temperature of 800^oC for 60 minutes is sufficient to complete the kaolin transformation to an amorphous material in these clays.

Additionally, specific surface, particle size distribution and bulk density were measured and presented in Table 1. The particle size distribution was determined by Malvern Mastersizer 2000 laser particle size analyzer and the d₉₀, d₅₀, and d₁₀ diameters were calculated and presented in Figure 4 and Table 1. Nitrogen surface area measurements are performed on two calcined clay samples by a BET sorptometer (model: Bellsorp-miniII) and reported in Table 1. As can be seen in Figure 4, CCG (type of KG calcined clay) used in this study included a larger amount of coarser particles than CCZ (type of KZ calcined clay). Subsequently, CCZ showed higher cumulative particle passing from 1μm to 100μm. Therefore CCZ presented lower d₁₀, d₅₀, and d₉₀ in comparison with CCG. Based on results obtained here it was shown that both calcined clays although classified as coarse calcined clays (Vizcaı´no Andre´s et al., 2015), but still meet the fineness requirement according to ASTM C618 which is less than 34 wt.% of particle retained on 45μm sieve for use as SCM in concrete. It should also be noted that compressive strength, especially at an early age, will considerably improve with finer calcined clay PSD in comparison with the coarser PSD. It was also revealed that finer PSD leads to higher pore refinement and a lower total porosity in comparison with coarse PSD (Vizcaı´no Andre´s et al., 2015). The BET surface area of CCZ was 4.6 m²/g, which is higher than that of CCG at 3.4 m²/g.

Pozzolanic activity tests

The pozzolanic activity of calcined clays depends on different parameters including the type and amount of clayed minerals (i.e. kaolinite, illite, or montmorillonite), the nature and amount of impurities, the thermal treatment (temperature and the time of heating used for its activation) and the specific surface. In the present study, the pozzolanic activity of calcined clay was determined with two different methods: (1) Compressive strength in mortars made with a blend of PC–pozzolan and (2) The measurement of pozzolanic activity, lime binding capacity, correspond to thermo-gravimetric analysis.

Strength activity index (SAI)

In the strength activity index method according to ASTM C618 and ASTM C311, for the assessment of pozzolanic reactivity by compressive strength in mortars cubes (50×50×50 mm³) including pozzolan, two types of mortars were used. In this regard, the strength index is specified as the ratio of compressive strength for mortar with 20% pozzolan substitution to the compressive strength of PC mortar. Reference mortar (PC mortars) cubes were prepared using a normal Portland cement (PC) with a chemical composition presented in Table 1 and w/b = 0.485. The cube specimens were demolded after 24 h and immersed in saturated lime water at 20 ± 1°C until test time. At 7, 28, and 90 days, the compressive strength was determined. The strength activity index (SAI =(A/B)×100) was calculated as the ratio of the compressive strength of PC-pozzolan mortar (A) to the strength of the reference mortar (B) at the same age, as a percentage. According to ASTM C618 and EN 450-1, the intended pozzolan is active when SAI is greater than 75% at 28-days. The results of the compressive strength index related to mortar specimens incorporating two intended calcined clays are shown in Table 2.

Based on results obtained from Table 2, it was shown that values for the strength activity index of both intended calcined clay were lower than 1.0 after 28 days of curing. It means that the compressive strengths at 7 and 28 days for both calcined clays are lower than for the PC. The authors believe that irrespective of coarse PSD for both calcined clays, it is likely the number of hydration products formed through the pozzolanic reaction did not yet compensate for the dilution effect. Results of this test confirm the higher pozzolanic reactivity of the KZ calcined clay sample in comparison with the KG calcined clay sample. After 90 days, KZ calcined clay has an SAI greater than 1.0, while the other calcined clay presented an SAI of 0.85, indicating that the contribution of pozzolanic reaction cannot compensate for the reduction of cementitious materials.

Thermo-gravimetric analysis

This method is based on the thermal decomposition of crystalline calcium hydroxide in a temperature range of 400–500°C to calcium oxide and water (Souri et al., 2015). In this study, the mixtures of calcined clay and hydrated lime were dry-mixed in a 1:1 weight ratio using a Turbula mixer for 4 h. Distilled water was added to obtain a water/solid ratio of 1:1. The mixtures were then stored in sealed containers for 7 days at the temperature of 50^oC and 90% humidity. The Ca(OH)₂ consumption over time was monitored by thermo-gravimetric analysis (TGA), using a Netzsch STA-449 C Jupiter, by calculating the mass loss in the range of 350–600°C. Figure 5 indicates DTG curves of lime–pozzolan mixtures related to both calcined clays (CCG and CCZ). As can be seen, the thermo-grams indicate thermal decomposition of Ca(OH)₂ which has not contributed to the pozzolanic reaction. The exothermic peaks located below 250^oC are possibly attributed to the release of water adsorbed on the surface and in the interlayer region of expandable clays. Second exothermic peaks in DTG curves at about 450-550^oC are attributed to the decomposition of Ca(OH)₂. In order to calculate the mass loss of Ca(OH)₂ for two samples, first the mass of Ca(OH)₂ that is converted to CO₂ can be calculated from TGA curves. In following by subtracting this amount from the initial mass of Ca(OH)₂, the final amount of Ca(OH)₂ that is consumed in the pozzolanic reaction will be calculated easily. Furthermore, by this method, the pozzolanic reactivity number can be estimated. It can be seen from Figure 5 that the partial consumption of portlandite by the pozzolanic reaction is confirmed for two calcined clay.

As can be seen in Figure 5, the higher rate of Ca(OH)₂ consumption is measured for CCZ, and the pozzolanic reactivity of clays shows a good correlation with the kaolinite content and SAI ratio. Based on the present method, the pozzolanic activity of KZ and KG calcined clay standard mixtures was determined at about 64.3% and 41.5% respectively. This result is in agreement with the results obtained for kaolinite content (Table 1) and SAI. Regarding the justification of obtained result, the authors believe that since the activity of the calcined clay, as mentioned earlier, depends primarily on the chemical and mineralogical composition of the calcined clay, as well as amorphous phase content, degree of de-hydroxylation, BET, and its fineness. Furthermore it should be noted that, greater BET and finer particle size distribution related to KZ calcined clay correspond to KG is also responsible for the higher pozzolanic activity.

Mixture proportions

The experiments were performed on six mortar mixtures containing different replacement levels of calcined clay, limestone, and silica fume for comparison. Based on the optimum percentage for replacement of calcined clay and lime available in the literature, two present calcined clays were replaced with 30% of cement by weight while replacement levels of lime were 15% by weight. The silica fume replacement level was selected at 8%. In addition, a control or reference mixture containing no pozzolanic material was also made. All six mixtures were proportioned to have the same cementitious materials content of 400 kg/m³ and the water to binder ratio (w/b) was adopted to 0.40. The total fine aggregate content was 1600-1650 kg/m³ and a high-range water reducer admixture was used in the mixtures to achieve a target flow of about 120 mm. The details of the mixture proportions of mortars and the flow values are presented in Table 3.

Test method

After the mixing process, the cast mortar specimens were consolidated by a vibrating table. Then, all the specimens were left covered with burlap in the laboratory for 24 h. The specimens were then demolded and cured in water saturated with calcium hydroxide at 23 ±1˚C for 28 days. The freshly mixed mortar was tested for air content according to ASTM C185 and flow according to ASTM C1437. The mortars were tested for compressive strength, water absorption, water penetration, electrical resistivity, rapid chloride penetration, accelerated chloride penetration and carbonation depth in the following.

Compressive strength: Three concrete cubes of 15×15×15 cm dimension of each mixture were tested for compressive strength after 28 days of curing with a loading rate of 0.5 N/mm²/s.
Water absorption: A water absorption test was carried out on three 100-mm cube specimens of each concrete mixture. In this regard the concrete specimens initially were cured for 28 days and then oven-dried until aimed to constant weight. The increase in mass resulting from immersion in water for 30 and 60 min and 24 hours, as a percentage of the mass of dry specimen, is according to the percentage of water absorption.
Water penetration depth: Water penetration depth under water pressure for mortar mixtures was determined by EN 12390-8. In this study, a 5 bars water pressure for 72 h was applied at the bottom of the cubic specimen of 150×150×150 mm. At the end of the test, the specimen was removed from the cell and then split into two halves in order to measure the penetration depth.
Electrical resistivity: The electrical resistivity of mortar mixtures was evaluated by Wenner 4-probe resistivity meter on the lateral surface of saturated cylindrical specimens of 100mm diameter and 200mm height. The cylindrical specimens were removed from the saturated lime water and then their lateral surface, where the readings will be performed there, was wiped with a dry cloth, and then the measurements were performed immediately.
Resistance to chloride ingress: The resistance to chloride ingress was evaluated using two different tests. The rapid chloride penetration test (RCPT) according to ASTM C1202 was adopted in the first step to obtaining qualitative criteria of the mortar's resistance to chloride ingress at 28 days' age. In the second step, accelerated chloride penetration according to NT Build 443 was also adopted to estimate the diffusion coefficient of the mortar. For the RCPT test, a sample with 50 mm thickness is cut from 100 mm (diameter) × 200 mm cylindrical mortar sample. After the vacuuming process According to the instruction proposed in the standard, the specimen is placed in the test device and the total charge passed is determined for each mixture. For accelerated chloride penetration, a 100 mm thick specimen is cut from a 150mm (diameter) × 300mm cylindrical mortar sample after 28 days of curing. The slices were sealed with epoxy resin coating on all sides except the top surface to simulate one-dimensional diffusion. The slices were immersed in NaCl solution with a concentration of 165g/l. After an exposure period of 90 days, the specimens were removed from the NaCl solution and mortar powder was obtained according to the method described elsewhere (Dousti et al., 2013; Khaghanpour et al., 2016). The powder from each ground layer was collected and analyzed separately according to ASTM C1152 and ASTM C114 for measuring acid-soluble chloride content. The resulting output of the test is a plot of chloride concentration versus depth from the exposed surface. Apparent diffusion coefficient (Da) and surface chloride concentration (Cs) were determined for each mixture by fitting profiles of chloride concentration versus depths according to Fick’s second law of diffusion (Dousti et al., 2013; Khaghanpour et al., 2016; Rashetnia et al., 2012).
Carbonation resistance: In order to determine the depth of carbonation, the 100mm (diameter) × 200mm cylindrical mortar sample is used. For this purpose, five-cylinder mortar samples after 28 days of water curing were dried at a temperature of 105°C for 7 days before exposure to CO₂. After preconditioning, the samples were placed in an accelerated carbonation chamber for 3 months. The accelerated carbonation environment was set with a temperature of 23±2°C, relative humidity of 70±5%, and a CO₂ concentration of 3.5%. After the accelerated carbonation process, the cylinders are split and phenolphthalein indicator is sprayed onto the split faces of the cylinders from the plain face.

Strength property

Compressive strength

The compressive strength of different mortar mixtures is presented in Figure 6. The compressive strength of all mixtures varied between 32MPa and 65MPa after 28 days of curing. At the age of 90 days, there is a continuous improvement in the compressive strength performance of all the mixtures except L15CCG30. As can be seen in Figure 6, at 7, 28, and 90 days, the mixture containing the type of KZ calcined clay (CCZ30) and the control mixture exhibited comparable compressive strength but lower than the mixture containing 8% silica fume. This is in agreement with results obtained earlier by Antoni et al. (2012), which showed that the replacement of cement with a blend of calcined clay and limestone produced a comparable mechanical property as plain Portland cement. The 7, 28, and 90-day compressive strength of CCG30, the calcined clay with lower pozzolanic activity and greater particle size, is lower than CCZ30 about 15-29%. Based on results reported by Tironi et al. (2012), the compressive strength of blended cement with 30% of calcined clay with medium and high kaolinite content (48-94% kaolinite content) reaches or surpasses the level of compressive strength of Portland cement mortar at 28 days. For both calcined clays which are used in the present study, adding 15% limestone with specification corresponding to Table 1, was led to a reduction of compressive strength. In this regard, the mixture with 15% limestone exhibited a reduction in strength between 11% and 13% compared to those without limestone. The authors believe that the addition of both calcined clay and limestone can lead to a marked decrease in the compressive strength of the mortars due to the high substitution and dilution effect.

Durability properties

Water absorption

The results of water absorption related to mortar specimens are presented in Figure 7. It can be seen that both the SF8 and CCZ30 mortar mixtures displayed a lower amount of water absorption compared to the control mixture (for every three times of immersion processes). This event may be attributed to the fact that the hydration products formed by the pozzolanic reaction in mixtures including calcined clay are deposited in the pores and as a result, these mixtures contained fewer interconnected pores compared to the control mixture. According to Figure 7, the CCZ30 mixture exhibits the lowest amount of water absorption and the mortar containing type of KG calcined clay with and without limestone has higher water absorption values than that of all other mixtures. The results display that substitution of 30% type of KZ calcined clay slightly reduces the amount of water absorption compared to SF8. As can be seen in Figure 7, a reduction of 16% and 30% in 30 min water absorption was found for mortar with 30% of calcined clay compared to SF8 and control mixtures respectively. This is consistent with the findings of Shah et al. (2020), who reported that the water-permeable porosity and rate of water absorption were found to reduce by different replacement levels of cement with limestone calcined clay pozzolan. It is worth mentioning that for both types of calcined clay, the addition of limestone has increased the amount of water absorption.

Water permeability

Figure 8 shows the results of water penetration depth for all mortar mixtures with and without calcined clay and limestone at 28 days. As can be seen, the results clearly display that there is a reduction in the water penetration depth for all mortars including calcined clay and limestone except the L15CCG30 mixture. As can be seen, the highest penetration depth was observed in the mortar mixture of L15CCG30, while the lowest water penetration depth is observed in L15CCZ30. It is possible that adding limestone up to 15% as a substitution for cement may improve pore structure in the transition zone, and thereby decreases water permeability, but on the other hand, the same phenomenon was not observed for the type of KG calcined clay mixture. This may be attributed to the lower pozzolanic activity of KG calcined clay compared to the type of KZ. The results of this section agree with those obtained by Kenai et al. (2004) and Menadi (2009) who reported a reduction in the water penetration depth when 15% of limestone as a replacement for cement and replacement of crushed sand were used respectively. The lower water permeability of L15CCZ30 is also consistent with the results obtained from the water absorption test but is inconsistent with results obtained by compressive strength. This conflict could be explained by the fact that the amount of water permeability does not only depend on compressive strength but also tortuosity, specific surface, pore size distribution, and connectivity of pores.

Electrical resistivity

Generally, electrical resistivity represents the interconnectivity of pores in the paste and has direct correlation with the resistance to ionic ingress. Given this assumption, Figure 9 indicates the values of surface resistivity of all the mortar mixes studied at 7, 14, 21, and 28 days. As can be seen, in all the mixtures, the results show that the CCZ30 sample indicates higher electrical resistance than other mortar mixes. As shown in this figure, all mortar mixtures increased electrical resistivity irrespective of the type of pozzolan used in mixes. Additionally, an increase in resistivity occurred with increasing age. At all ages, it was obviously shown that the electrical resistivity of the mortar mixtures with the type of KG calcined clay (with or without limestone) was lower than those recorded for the type of KZ calcined clay (CCZ30 and L15CCZ30) mortars with the same replacement level. It seems that the use of calcined clay as earlier reported by Scrivener et al (2018) even with very low original kaolinite content; has a pore structure finer than pastes made with Portland cement and increased resistivity of mortar mixtures by creating a finer pore size distribution. The authors believe that this is the most important event that occurs in the pore structure of cement paste due to the utilization of calcined clay.

Figure 10 indicates the development of electrical resistivity of mixtures with time. As can be seen, the curve of the control sample had minimal changes beyond 7 days, while CCZ30 and L15CCZ30 show a distinct increase from 7 to 28 days. These electrical resistivity trends confirm the difference in kinetics of microstructural change in the present mixtures. As can be seen, for CCZ30 and L15CCZ30 mixtures, the slope of the electrical resistivity curve is steeper correspond to other mixtures. This means that, with the development of the pozzolanic reaction, the electrical resistivity for these two mixtures has increased more than the others. This behavior is probably attributed to previously mentioned pore refinement in the calcined clay and calcined clay with limestone leads to an increase in resistivity, especially at early ages against other mixtures. In the case of the L15CCZ30 mixture, the results and pore refinement is consistent with the results obtained by other researchers (Scrivener, 2014; Fernandez Lopez, 2009). Based on data available in the literature, it was demonstrated that even calcined clay with less than 50% kaolinite content is suitable for usage in ternary systems including calcined clay with limestone (LC3). In this system, in addition to the pozzolanic reactivity of calcined clay, the aluminates phase in the calcined clay reacts with calcite from limestone to create additional hydrate phases such as Mono- (Mc) and hemi-carboaluminates (Hc) and improve pore structure of pastes (Scrivener, 2014; Fernandez Lopez, 2009). In addition, limestone influence can also be attributed to packing improvement, increasing the early hydration as effective filler and further, participating in the reaction with the highly reactive calcined clay (Berodier and Scrivener, 2014; Bentz et al., 2012). In the case of CCG30 and L15CCG30, results are consistent with previous results obtained by compressive strength, water absorption and water penetration depth tests and indicates that for both mixtures, pore refinement and improvement of microstructures are not as well as the type of KZ calcined clay.

Resistance to chloride ingress

Results of rapid chloride penetration resistance at 28 days measured in terms of electric charge passed through the specimens in coulombs are illustrated in Figure 11. Except for the mixture of L15CCG30, other mixtures indicated lower values of the charge passed correspond to the control mixture, indicating higher resistance to chloride-ion penetration. In other words, in mixtures made with the calcined clay or calcined clay with adding 15% limestone binder, the 28-day chloride resistance has improved by nearly 60–80% correspond to the control mixture. This is consistent with the excellent resistance of calcined clay binder to chloride ingress at an early age with appropriate results reported in previous sections. The improvement of the microstructure resulting from the pozzolanic reaction of calcined clay has probably led to a reduction in the rapid chloride penetration. The results of an experimental study developed by Dhandapani et al. (2018)showed that concrete made with LC3 binder has excellent resistance to chloride ingress after 28 days of age.

According to Figure 11, L15CCG30 mixture presented a higher charge value compared to all other mixtures - around 4280 coulomb. Based on ASTM C1202, this value indicates a high permeability. This is consistent with all other tests including compressive strength, water absorption, water penetration depth, and electrical resistivity. Additionally, in the present study, at 28 days, chloride-ion penetration of mixtures incorporating 15% of limestone was found to be higher than that of mixtures without limestone. These results are in agreement with those obtained by Bonavetti et al. (2000), who reported an increase in chloride ion penetration for water cured concretes containing 10% and 20% limestone filler as cement replacement.

The chloride concentration profiles as the results of the second method for the determination of resistance to chloride ingress are illustrated in Figure 12. As can be seen, generally, the chloride contents decrease higher depths. It can be also found that the use of SCMs remarkably enhances the resistance of mortar mixtures against chloride diffusion due to cement microstructure improvement. The results clearly show that at 30% calcined clay; there was a marked decrease in chloride penetration depth compared to the control mixture. It was demonstrated that from one side the number of pores with smaller size increases in mixtures including calcined clay due to the pozzolanic reaction of calcined clay helping in the refinement of the pore structure. On the other side, the refined pore structure increases the tortuosity of the system, thereby restricting the movement or transport of ions/fluids in the concrete or mortar. Apart from this pore refinement; it was demonstrated that chloride binding can also occur in mortars including calcined clay due to the higher amount of alumina hydrates. Chlorides are known to react with aluminate hydrates to form Friedel’s salt. Also, the physical binding of chloride ions on the surface of C–S–H is known to occur (Dousti et al., 2017). As a result, the above phenomena can collectively result in lower chloride ingress in mixtures containing calcined clay as compared to the control mixture. In this regard, it was observed that after 3 months of exposure to NaCl solution, chloride penetration depth was deeper than 25mm in the control mixture, while in the CCZ30 mixture; chloride ions reached a depth of about 15mm, in fact, something about 50% less. Adding 15% limestone to both mixtures incorporating calcined clay increased chloride penetration depth, but this increase is much greater for the type of KG calcined clay. Among all the mixtures, the L15CCG30 mixture as expected earlier has the highest chloride penetration depth (more than 30mm).

The calculated apparent diffusion coefficients (D_a) and surface chloride concentrations (C_s) for all mixtures after 3 months of exposure to the saline solution are presented in Table 4. As can be seen, the results agree well with those obtained earlier by the RCPT test. As shown in Table 4, regardless of the addition of limestone, the use of SF and calcined clay decreases D_a dramatically compared to the control mixture. The use of 30% type of KZ and KG calcined clay reduced apparent chloride diffusion of mortar by about 80 and 40% respectively and the use of 15% limestone and 30% type of KZ calcined clay decreased D_a by 73%. As can be seen, for the L15CCG30 mixture the D_a increased by about 100%. This increase in D_a was expected earlier due to poor results obtained by other durability factors. In a similar investigation but on an LC3 binder system, Sui et al. (2019) revealed that LC3 mortar obtained the lowest apparent chloride diffusion coefficient among different binary and ternary cementitious materials.

Resistance to carbonation

To determine the carbonation resistance of mortar mixtures, the carbonation depth was measured by spraying 1% phenolphthalein indicator solution on the freshly broken surface of cylinder samples after 3 months of exposure. It can be seen from Figure 13 that on the un-carbonated surface of the sample where the mortar is still highly alkaline, purple-red color has appeared. On the carbonated surface, no coloration occurs. The carbonation depth values of the samples are shown in Figure 14. The minimum carbonation depth of 5mm was observed for the control mixture and maximum carbonation depth including the whole surface of samples (> 50 mm) was obtained for both CCG30 and L15CCZ30 mixtures. For CCZ30 and SF8, the carbonation depth was less than 20 mm. The results clearly indicate that the carbonation resistance of mixtures containing SCMs is generally lower as compared to PC due to lower total alkalinity and agree with data available in the literature (Sui et al., 2019; Branch et al., 2016; Khunthongkeaw et al., 2006). It should be noted that the lower carbonation depth obtained for the control sample compared to SF8 and CCZ30 mortars should not be taken into account as an advantage to reinforced concrete structures. The results obtained in other studies indicate that the reinforced concrete structures with no pozzolan are more vulnerable to the combined attack of carbonation and chloride ingress due to the higher chloride ion diffusion coefficient (Branch et al., 2016; Khunthongkeaw et al., 2006; Shah and Bishnoi, 2018; Šavija and Lukovic, 2016).

In this study, the feasibility of using a new type of supplementary cementitious materials as cement replacement was investigated. The influence of cement replacement levels with two medium-grade kaolinite clay and limestone on strength development, transport properties, and carbonation was investigated and the following conclusion can be drawn:

It was observed that mortars made with 30% KZ calcined clay (> 40% kaolinite content), with a similar mixture proportion, showed comparable strength development characteristics with PC mortar and lower strength development compared to SF8. The addition of 15% limestone in mortar caused a reduction of 11–13% in compressive strength regardless of the calcined clay type.
The water absorption and water permeability were found to reduce by replacement of cement with calcined clay. Based on present results, a reduction of 16% and 30% in 30 min water absorption was found for mortar with 30% of calcined clay compared to SF8 and control mixtures respectively. The reduction in value of both parameters could be primarily associated with the reduction in total volume of capillary pores and increased tortuosity.
Based on the results obtained, mortar made with calcined clay presents significantly higher electrical resistivity than the control mortar mixture. This also results in better resistance against ingress of chloride ions which were evaluated by RCPT and accelerated chloride penetration.
Except for the mixture of L15CCG30, other mixtures, showed lower values of the charge passed compared to the control mixture. Furthermore, the chloride resistance has improved by nearly 60–80% with respect to control mixture by 28 days, indicating higher resistance to chloride-ion penetration.
The results of the accelerated chloride penetration test and calculated diffusion coefficient are in agreement with the results obtained by the RCPT test. The results show that at 30% calcined clay; there was a marked decrease (about 80 and 40%) in D_a and an increase in C_s compared to the control mixture. The use of 15% limestone and 30% KZ calcined clay also decreased D_a by about 73%. However, the mixture including the KZ calcined clay had a better influence on improving diffusion characteristics of mortar compared to the KG calcined clay mortar at the same level of substitution.
It was observed that the utilization of 15% limestone with 30% KZ calcined clay, unlike KG, improved all durability parameters (except carbonation resistance) and reduced compressive strength in comparison with PC mortar.
The results indicate that the carbonation resistance of mixtures containing calcined clays and SF is lower compared to the control mixture due to lower total alkalinity.

Acknowledgments

The authors would like to acknowledge the support for this Project provided by the Construction Materials Institute (CMI), School of Civil Engineering, College of Engineering, University of Tehran and the Road, Housing and Urban Development Research Center (BHRC).

Akindahunsi, A.A. Avet, F. Scrivener, K. (2020), The Influence of some calcined clays from Nigeria as clinker substitute in cementitious systems, Case Studies in Construction Materials. 13, e00443.
Alujas, A. Fernández, R. Quintana, R. Scrivener, K.L. Martirena, F. (2015), Pozzolanic reactivity of low grade kaolinitic clays: Influence of calcinations temperature and impact of calcination products on OPC hydration, Applied Clay Science. 108, 94–101.
Antoni, M. (2013), Investigation of cement substitution by blends of calcined clays and limestone, (PhD Thesis), ÉCOLE POLYTECHNIQUE FÉDÉRALE DE LAUSANNE, Switzerland.
Antoni, M. Rossen, J. Martirena, F. Scrivener, K. (2012), Cement substitution by a combination of metakaolin and limestone, Cement and Concrete Research. 42, 1579–1589.
Avet, A. Scrivener, K. (2018), Investigation of the calcined kaolinite content on the hydration of Limestone Calcined Clay Cement (LC³), Cement and Concrete Research. 107, 124–135.
Avet, F. Snellings, R. Alujas Diaz, A. Ben Haha, A, M. Scrivener, K. (2016), Development of a new rapid, relevant and reliable (R3) test method to evaluate the pozzolanic reactivity of calcined kaolinitic clays, Cement and Concrete Research. 85, 1–11.
Badogiannis, E. Kakali, G. Dimopoulou, G. Chaniotakis, E. Tsivilis, S. (2005), Metakaolin as a main cement constituent Exploitation of poor Greek kaolins, Cem Concr Compos. 27 (2), 197–203.
Berodier, B and Scrivener, K. (2014), Understanding the Filler Effect on the Nucleation and Growth of c-s-h, 3773, 3764–3773.
Bentz, D.P. Sato, T. de la Varga, I. Weiss, W.J. (2012), Fine limestone additions to regulate setting in high volume fly ash mixtures, Cem. Concr. Compos. 34, 11–17.
Bonavetti, V. Donza, H. Rahhal, V. Irassar, E. (2000), Influence of initial curing on properties of concrete containing limestone blended cement, Cem Concrete Res. 30, 703–8.
Branch, J.L. Kossona, D.S. Garrabrants, A.C. He, P.J. (2016), The impact of carbonation on the microstructure and solubility of major constituents in micro concrete materials with varying alkalinities due to fly ash replacement of ordinary Portland cement, Cement and Concrete Research. 89, 297-309.
Castillo, R. Fernandez, R. Antoni, M. Scrivener, K. Alujas, A. Martirena, J.F. (2010), Activation of low grade clays at high temperatures, Rev Ing Construccion. 25 (3), 329–52.
Dhandapani, Y. Sakthivel, T. Santhanam, M. Gettu, R. Pillai, R. (2018), Mechanical properties and durability performance of concretes with Limestone Calcined Clay Cement (LC3), Cement and Concrete Research. 107, 136–151.
Dousti, . Beaudoin, J.J. Shekarchi, M. (2017), Chloride binding in hydrated MK, SF and natural zeolite-lime mixtures, Construction and Building Materials. 154, 1035–1047.
Dousti, A. Rashetnia, R. Ahmadi, B. Shekarchi, M. (2013), Influence of exposure temperature on chloride diffusion in concretes incorporating silica fume or natural zeolite, Construction and Building Materials. 49, 393–399.
Fernandez, R. Martirena, F. Scrivener, K. (2011), The origin of the pozzolanic activity of calcined clay minerals: A comparison between kaolinite, illite and montmorillonite, Cement and Concrete Research. 41, 113–122.
Flatt, R.J. Roussel, N. Cheeseman, C.R. (2012), Concrete: an eco material that needs to be improved, J. Eur. Ceram. Soc. 32, p.2787–2798.
Hollanders, S. Adriaens, R. Skibsted, J. Cizer, Ö. Elsen, J. (2016), Pozzolanic reactivity of pure calcined clays, Applied Clay Science. 132-133, 552-560.
Khaghanpour, R. Dousti, A. Shekarchi, M. (2016), Prediction of Cover Thickness Based on Long-Term Chloride Penetration in a Marine Environment, Journal of Performance of Constructed Facilities (ASCE). 04016070.
Kenai, S. Soboyejo, W. Soboyejo, A. (2004), Some engineering properties of limestone concrete, Mater Manuf process. 19 (5), 949–61.
Khunthongkeaw, J. Tangtermsirikul, S. Leelawat, T. (2006), A study on carbonation depth prediction for fly ash concrete, Construction and Building Materials. 20 (9), 744–753.
Matschei, T. Lothenbach, B. Glasser, F.P. (2007), The role of calcium carbonate in cement hydration, Cem. Concr. Res. 37 (4), 551–558.
Menadi, B. Kenai, S. Khatib, J. Mokhtar, A. (2009), Strength and durability of concrete incorporating crushed limestone sand, Construction and Building Materials. 23, 625–633.
Rashetnia, R. Dousti, A. Ahmadi, B. Khatami, D. Shekarchi M. (2012), Effect of natural zeolite and silica fume as a pozzolanic materials on the chloride diffusion of concrete, International congress on durability of concrete (ICDC), Norways.
Rodriguez-Camacho, R.E. and Uribe-Afif, R. (2002), Importance of using the natural pozzolans on concrete durability, Cem. Concr. Res. 32(12), 1851–1858.
Samet, B. Mnif, T. Chaabouni, M. (2007), Use of a kaolinitic clay as a pozzolanic material for cements: Formulation of blended cement, Cement & Concrete Composites. 29, 741–749.
Šavija B, Lukovic M. (2016), Carbonation of cement paste: Understanding, challenges, and Opportunities, Construction and Building Materials. 117:285–301.
Schneidera, M. Romer, M. Tschudin, M. Bolio, H. (2011), Sustainable cement production—present and future, Cem. Concr. Res. 41 (7), 642–650.
Scrivener, K. Martirena, F. Bishnoi, S. Maity, S. (2018), Calcined clay limestone cements (LC3). Cement and Concrete Research. 114, 49-56.
Scrivener, K. Snellings, R. Lothenbach, B. (2016), A Practical Guide to Micro structural Analysis of Cementitious Materials, Taylor & Francis.
Scrivener, K.L. (2014), Options for the future of cements, Indian Concr. J. 88, 11–21.
Fernandez Lopez, R. (2009), Calcined Clayey Soils as a Potential Replacement for Cement in Developing Countries, Ph.D Thesis EPFL.
Shah, V. Bishnoi, S. (2018), Carbonation resistance of cements containing supplementary cementitious materials and its relation to various parameters of concrete, Construction and Building Materials. 178, 219-232.
Shekarchi, M. Bonakdar, A. Bakhshi, M. Mirdamadi, A. Mobasher, B. (2010) Transport properties in metakaolin blended concrete, Construction and Building Materials. 24, 2217–2223.
Shah, V. Parashar, A. Mishra, G. Medepalli, S. Krishnan, S. Bishnoi, S. (2020), Influence of cement replacement by limestone calcined clay pozzolan on the engineering properties of mortar and concrete, Advances in Cement Research. 32 (3), 101-111.
Singh, M. Garg, M. (2006), Reactive pozzolana from Indian clays–Their use in cement mortars, Cement and Concrete Research. 36, 1903.
Souri, A. Golestani-Fard, F. Naghizadeh, R. Veiseh, S. (2015), An investigation on pozzolanic activity of Iranian kaolins obtained by thermal treatment, Applied Clay Science. 103, 34–39.
Sui, S. Georget, F. Maraghechi, H. Sun, W. Scrivener, K. (2019), Towards a generic approach to durability: Factors affecting chloride transport in binary and ternary cementitious materials, Cem. Concr. Res. 124, 105783.
Tironi, A. Trezza, M.A. Alberto, N. Scian, A.N. Irassar, E.F. (2013), Assessment of pozzolanic activity of different calcined clays. Cement & Concrete Composites. 37, 319–327.
Tironi, A. Trezza, M.A. Alberto, N. Scian, A.N. Irassar, E.F. (2012), Kaolinitic calcined clays: Factors affecting its performance as pozzolans, Construction and Building Materials. 28, 276–281.
Vizcaíno, L. Antoni, M. Alujas, A. Martirena, F and Scrivener, K. (2020), Industrial Manufacture of a Low-Clinker Blended Cement Using Low-Grade Calcined Clays and Limestone as SCM:The Cuban Experience, Calcined Clays for Sustainable Concrete, RILEM Bookseries 10, DOI 10.1007/978-94-017-9939-3-43.
Vejmelkova, E. Keppert, M. Rovnanikova, P. Keršner, Z. Cˇerny´, R. (2012), Application of burnt clay shale as pozzolan addition to lime mortar, Cem Concr Compos. 34 (4), 486–92.
Vizcaı´no Andre´s, L.M. Antoni, M.G. Alujas Diaz, A. Fernando Martirena Herna´ndez, J. Scrivener, K. (2015), Effect of fineness in clinker-calcined clays-limestone cements, Advances in Cement Research. 27 (9), 546–556.
Zhang, M.H and Malhotra, V.M. (1995), Characteristics of a thermally activated aluminosilicate pozzolanic material and its use in concrete, Cement and Concrete Research. 25, 1713.

Table 1

Chemical and mineralogical composition of the materials.

*Oxides (wt-%)*	KZ	KG	Portland cement (PC)	Silica fume (SF)	Limestone (LS)
SiO₂	74.0	73.0	20.7	92.1	12.4
Al₂O₃	15.5	14.5	4.0	0.3	1.8
Fe₂O₃	0.3	0.7	4.1	0.2	2.1
CaO	0.8	4.0	63.2	0.9	40.5
MgO	0.3	0.4	1.5	1.2	5.3
SO₃	-	-	2.2	0.3	-
Na₂O	0.3	0.3	0.5	-	-
K₂O	0.4	0.4	0.4	-	-
TiO₂	0.04	0.04		-	-
L.O.I	6.5	8.0	1.5	1.6	37.9
SiO₂ + Al₂O₃ + Fe₂O₃	89.8	88.2	-	-	-
*Mineralogical composition of raw clays (%)*
Kaolinite	44.7	40	-
Quartz	34.2	28.5	-
Anatase	0.7	0	-
Calcite	1.2	0	-
Muscovite	8.4	20.7	-
Microcline	0	3.7	-
Amorphous	10.8	7.1	-
C₃S	-	-	61
C₂S	-	-	13
C₃A	-	-	4
*Physical properties of calcined clays*
Specific gravity	2.7	2.8	3.1	2.2	-
BET specific surface (m²/g)	4.6	3.4	0.9	-	1.2
*Particle size distribution*
d₁₀ (µm)	1.3	3.1
d₅₀ (µm)	4.3	47
d₉₀ (µm)	97.5	233

Table 2

Strength Activity Index of standard mortar with KZ and KG calcined clays according to the ASTM C618.

	Strength Activity Index (%), 7days	Strength Activity Index (%), 28days	Strength Activity Index (%), 90days
Mortar with KZ calcined clay	80	84	1.06
Mortar with KG calcined clay	76	78	0.85

Table 3

Mixture proportions of fresh mortars

Mixtures	Percent of Cementitious materials substitution					w/cm	Sand (kg/m³)	Superplacticizer (%)	Flow of mixtures (mm)	Air content (%)
Mixtures	Cement (%)	KZ (%)	KG (%)	SF (%)	LS (%)	w/cm	Sand (kg/m³)	Superplacticizer (%)	Flow of mixtures (mm)	Air content (%)
PC	100	0	0	0	0	0.4	1631	0.83	167	4.8
SF8	92	0	0	8	0	0.4	1623	0.75	151	4.8
CCZ30	70	30	0	0	0	0.4	1603	0.75	128	4.7
CCG30	70	0	30	0	0	0.4	1608	0.85	120	5
L15CCZ30	55	30	0	0	15	0.4	1617	0.80	122	5
L15CCG30	55	0	30	0	15	0.4	1622	0.80	129	5.1

Table 4

Apparent chloride diffusion coefficients and surface chloride concentrations in the mortar mixtures.

Mixture	Apparent diffusion coefficient-D_a (× 10^-12 m²/s)	Chloride surface concentration-C_s (% by wt of concrete)	Correlation of determination, r²
C	17.5	0.526	0.996
SF8	6.47	0.558	0.986
CCZ30	3.07	0.889	0.992
CCG30	10.4	0.785	0.997
L15CCZ30	4.61	0.808	0.991
L15CCG30	35.7	0.913	0.996

No competing interests reported.

Strength and Durability Performance of Mortars Incorporating Calcined Clay as Pozzolan

Status:

Version 1

Abstract

Figures

Introduction

Experimental Program

Results And Disscussion

Conclusions

Declarations

References

Tables

Additional Declarations

Status:

Version 1