The nature of kaolinitic clays and their impact on the performance of SCM

Nowadays, metakaolin is currently widely used as a supplementary cementitious material in the concrete industry. However, there are significant differences between kaolinitic clays, which can be used for the production of metakaolin. They can vary in the granulometry as well in the mineralogical composition (more precisely the content of varying impurities can differ widely). The main aim of this article is to assess the impact of mentioned variation on the thermal activation of such kaolinitic clays. For this purpose, four clays from the Czech Republic were chosen and examined. Primarily, their characterization composed of XRD, DSC, granulometry, and specific surface area was studied. Then these clays were thermally activated at temperatures from 500°C to 650°C. Afterwards, the changes in determined properties were investigated. Then the samples of blended cement pastes with a replacement level of 15% were produced, and basic physical and mechanical properties were determined. It was proved that for particular clays, different temperature treatments were optimal, which further lead to varying results of the mechanical performance of blended cement-based pastes. Such variation can be linked to the phase composition of studied clays.


Introduction
To produce one of the most commonly used building materials, concrete, you first need to produce Portland cement.However, due to the high firing temperature (1450 °C -1550 °C), the large amount of fossil fuels used and the large consumption of non-renewable raw materials, this is an expensive and very non-environmentally friendly process [1,2].Therefore, there is an increasing need to find a suitable supplementary cementitious material (SCM) that can at least partially replace cement in concrete without significantly degrading its properties.In addition to the commonly used SCMs such as blast furnace slag, fly ash or micro-silica, the emphasis is on finding more natural and preferably local sources of pozzolanic materials.For example, thermally activated kaolinite clays may be such a material.Kaolinite clay is defined as a material composed primarily of the mineral kaolinite (more than 80% [3]) accompanied by illite, or other clay minerals, micas, quartz or feldspar.Many researches have shown that if 10-30 % of cement is replaced by metakaolin (thermally activated kaolinite clay), the physical and mechanical properties of mortars are significantly improved [4][5][6].The amount of added metakaolin has also a significant effect on water resistance, porosity and initial setting time.In the right amount, it can increase the compressive strength of the resulting composite by up to twice [7][8][9].Nevertheless, pure kaolinite clays are frequently used applicability in the field of industrial production (such as paper production or ceramics) or medical and cosmetics applications.Therefore, their utilizable

Experimental methods
The density of studied materials was determined using helium pycnometry.ATC EVO from Thermo Fisher Scientific Inc. was used for the measurements.A laser particle size analyser Bettersizer S3 Plus was used to measure the particle size distribution and the specific surface area was determined by air permeability using the UTEST UTCM-0280 automatic Blaine apparatus.For phase analysis was used XRD, specifically Aeris from Malvern Panalytical Ltd.For data curation, two programs were combined.Firstly, HighScore from Malvern Panalytical Ltd. was used for qualitative analysis.Afterwards, quantitative analysis was performed with the help of Profex [10].The amount of amorphous and not identified phases was determined by the utilization of internal standard, specifically 20% of zincite.Thermal analysis was carried out using the Labsys Evo STA instrument.The experiments were set up as follows: the temperature range was 10 °C /min, the maximum temperature was 1000 °C and the atmosphere was argon.Compressive and bending strength was determined using the ED60 Compression Press with Compression Frame Jig Assembly (ELE International).As cement mixes were investigated in this paper, smaller samples were tested to avoid cracking due to shrinkage.The dimensions were 100 x 20 x 20 mm.Except for the sample dimensions, the measurement procedure was adopted from the standard [11].Before the actual experiments, samples were weighted and measured to obtain the bulk density of samples.

Studied clays
Four different clays were used for the need of this study, all were produced by Keramost a.s.Their physical properties are summarized in Table 1, while the phase composition is presented in Table 2

Figure 1 X-ray diffractograms of studied clays
The first reference clay KR was pure kaolinite clays, with 86% of kaolinite.The supplement components were illite and quartz.Studied kaolinitic clays KJ, KP and KW proved to have in general quite finer granulometry but they differed mainly in their mineralogical composition.They contained broadly speaking equal amount of kaolinite (about 60%) and varied between each other in the supplement component.Kaolinitic clay KJ showed only 10% of other clay minerals and had the highest amount of amorphous and not identified phases.The samples labelled KP contained the biggest amount of quartz, and the highest content of illite as well.The last sample KW had also a substantial amount of illite, but contrary to the latter clays it did not contain pyrite.
In Figure 2, the results of the simultaneous thermal analysis are delineated.The performance of all clays is in accordance with common knowledge [12,13].The first endothermic peaks indicated the evaporation of free and adsorbed water in the interlayers of clay minerals.Afterwards, the most significant exothermic peaks at temperatures in the range of 450 °C to 700 °C signified the dehydroxylation of clays minerals.The last exothermic peaks from 950 °C were associated with the crystallization of -Al-spinel.When comparing the performance of individual clays, it is visible that KJ contained the highest amount of adsorbed water (the mass change was 1.5%).Focusing on the dehydroxylation process, it is interesting to note that the highest mass change was not observed in the case of reference kaolinite clay KR, but it was clay KW which showed the biggest mass decrease (in particular 11%).Two remaining kaolinitic clays KJ and KP showed lower mass changes.This can be attributed to the higher amount of not clays minerals, especially quartz.However, based on the described performance of clay materials, three temperatures were selected for thermal activation, namely 550 °C, 600 °C and 650 °C.The thermal treatment was carried out in an electric top-cover furnace with a temperature range of 10 °C per minute, the duration of the exposure to the maximum temperature was 3 hours and then the samples were left to cool spontaneously.

Blended cement pastes
In order to assess the performance of thermally activated clays used as a supplementary cementitious material, they were used as a cement substitute in cement-based pastes.The reference cement paste was CEM I 42.5R from Českomoravský cement, a.s., Mokrá factory, with a w/c ratio of 0.33.Afterwards, 12 mixtures with 10% of thermally activated clays were designed and produced.Their designation is derived from the used clay and the temperature treatment from 550 °C to 650 °C.To complement the study, a further set of mixes was designed with 10% untreated clays.Blended cements were produced with equal water dosage as reference paste.Three prismatic specimens were prepared from the fresh mixes.These were cured in a climatic chamber at a controlled temperature of 20 °C and relative humidity of 80%.After 28 days of curing, the physical properties were examined.

Properties of cement pastes
The results of the physical properties of blended cement pastes are shown in Table 3.The results of the physical properties of designed blended cement pastes are shown in table 1. Regarding the values of bulk density, although the varies were within the range of 4%, some visible trends can be observed.Notactivated kaolin clay KR showed slightly lower values, while the untreated kaolinitic clays KJ, KP and KW showed slightly higher values of bulk densities.When focused on the thermal treatment, the exposure to 550 °C caused a small decrease, and with the increasing temperatures, the values went up.The results of compressive strength visibly proved the potential of kaolinitic clays to be used as SCM.However, when first focused on the performance of not-activated clays, surprisingly only a worsening of compressive strength was observed in the case of pure kaolinite KR.The value was 22% lower compared to pure Portland cement.In the case of kaolinitic clays, the compressive strengths went up (even by 15% for KW).The explanation of this performance can be found in the varying granulometry (see Table 1); the finer material is, the better compressive strength reached.Regarding the thermal activation, it is obvious, that for the studied clays, the impact of temperatures was immense.The biggest growth of compressive strength was observed in the case of kaolinite clay KR treated at 650 °C; the growth was by 35% compared to untreated clay KR.However, the difference compared to the Portland cement pastes was 19%.Despite that kaolinitic clays showed a lower increase compared to not treated samples (namely by 7%, 13% and 7% for KJ, KP and KW), compared to the Portland cement pastes they performed even better than reference KR-650.They showed 20%, 22% and 22% better values for KJ, KP and KW.Moreover, for these results, a lower temperature was needed.The optimal temperature was in this case only 550 °C.Bending strengths showed similar tendencies as the latter described compressive strengths.Except that the not treated clays reached better results in all cases of studied materials.
In order to explain such performance, the specific surface areas of treated clays were measured as well.
It is included also in Table 3. Reached results of compressive strengths are in accordance with the behaviour of specific surface areas.In addition, when focused on the composition of studied kaolinitic clays, the impact of the clay minerals content can be also observed.The lowest value of compressive strength reached sample KJ, which had the lowest amount of clay minerals (70%, see Table 2), while the best performance showed sample KW (with 78% of clays).Moreover, the higher amount of quartz and other crystal impurities in the case of clay KP seemed to have no negative effect, as the blended cement paste KP-550 reached almost equal value as the best KP-550.

Conclusion
The presented paper is focused on the performance of kaolinitic clays used as supplementary cementitious materials.Three varying clays were chosen, and their performance was compared to the behaviour of pure kaolin clay.Raw materials were primarily characterized, and their physical properties and composition were determined.Afterwards, with the help of thermal analysis, three temperatures for their thermal activation were selected, namely 550 °C, 600 °C and 650 °C.Finally, blended cement pastes containing 10% of thermally activated kaolinitic clays were produced.It was proved that kaolinitic clays can propose suitable behaviour, as they reached even slightly higher values of mechanical strengths compared to pure kaolin.Moreover, for proper activation, they required lower temperatures.The optimal treatment was found to be 550 °C for all three kaolinitic clays.

Figure 2 .
Figure 2. Results of simultaneous TG and DSC

Table 1 .
. Characterization of studied clays

Table 2 .
Mineralogical composition of studied clays

Table 3 .
Properties of blended cement pastes