Polymer-silicate composite based on alternative raw materials under conditions of increased temperature

Procedia Engineering 151 ( 2016 ) 375 – 382 Available online at www.sciencedirect.com 1877-7058 © 2016 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of the organizing committee of ICEBMP 2016 doi: 10.1016/j.proeng.2016.07.398 ScienceDirect International Conference on Ecology and new Building materials and products, ICEBMP 2016 Polymer-silicate composite based on alternative raw materials under conditions of increased temperature Tomáš Melichar*, JiĜí Bydžovský Brno University of Technology, Faculty of Civil Engineering, VeveĜí 331/95, 602 00 Brno, Czech Republic Abstract The paper presents research into studying the properties of composite materials with polymer-silicate matrix and porous (lightweight) filler. Modification of matrix even filler was carried out by raw materials from alternative sources. Specifically, the matrix based on cement substituted by 35% of alternative raw material was considered. In case of filler aggregate based sintered ash was used. One of the requirements for the parameters of the examined materials is resistance to extreme temperatures. Investigated materials were therefore exposed at high temperatures. Thermal loading took place gradually in temperature increments of 22 °C (reference conditions), 400, 600, 800 and 1000 °C with subsequent controlled cooling. Attention was also focused on various ways of cooling – slow in the furnace (1 °C/min) and immersion in water (bath temperature c. 18 °C). Suitability of matrix modification and use of porous fillers based on alternative raw materials was proved by set of analyses. After exposure in an environment with a temperature of 1000 °C followed by quench cooling was set residual compressive strength of about 40%, which is relatively high for this type of the matrix. The curves of residual strength with different types of cooling are intriguing too. Selected samples were also investigated by using micro/structural techniques, where the emphasis was placed on the formation of any defects related. © 2016 The Authors. Published by Elsevier Ltd. Peer-review under responsibility of the organizing committee of ICEBMP 2016. Keywords: Composite; extreme; temperature; regime; cooling * Corresponding author. Tel.: +420-541-147-463; fax: +420-541-147-502. E-mail address: melichar.[email protected].cz © 2016 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of the organizing committee of ICEBMP 2016 376 Tomáš Melichar and Jiří Bydžovský / Procedia Engineering 151 ( 2016 ) 375 – 382 1. Introduction Composite materials on the basis of a cement matrix are non-flammable, however, the negative effects of elevated temperatures is generally known and has been demonstrated in a number of studies [1,2]. Several principal aspects play a significant role in the degradation of the parameters of cement mortars and concretes. The first and foremost is the matrix itself. The matrix is a complex of compounds created as the product of the hydration reactions of the binder - CSH in various forms, portlandite, calcite, ettringite, etc. or potentially of the non-reacted components of the original binder or admixtures used. The matrix structure decomposes as a result of elevated temperatures. Water is eliminated during the first stage. Potential polymers, CSH phase, portlandite and calcite therein also decompose. Also of significance is the modification transformation at 573 °C which is accompanied by considerable changes in volume. Ceramic bonds which substitute original hydraulic bonds start to be created in the last stage i.e. at temperatures above 800 °C. Later, at temperatures above 1100 °C, matrix clinkering is apparent. The processes mentioned above are accompanied by considerable matrix shrinkage. A positive effect of modification of the cement binder under extreme thermal stress has been demonstrated in many research studies [3,4,5]. Substitution on a larger scale with, for example, blast-furnace slag or high-temperature fly ash or with a minor admixture of Al2O3, TiO2, micro-silica, etc. is worthy of note. Aggregate represents another component of cement composites. The changes that take place are somehow different in character. The modification transformations are dependent on the origin of a specific aggregate. The decomposition or formation of new phases is also not an exception. The antagonistic behaviour (with respect to the matrix) is substantial because of the aggregate expansion and the subsequent changes in volume that occur. A characteristic that is of particular importance is the modulus of elasticity which participates in the characterization of the relationship between the aggregate used and the matrix. The use of porous aggregate proved to be very advantageous in this respect. Furthermore, if this aggregate is made by firing clayey soils or fly ash at high temperatures, it gains considerable thermal resistance and stability. From the viewpoint of the behaviour of the given composite system, the isothermal endurance and cooling conditions are as important as the rising temperatures. Under real conditions various situations may occur during a fire. When a fire is extinguished by means of water, severe thermal shock occurs. Depending on the maximum temperature of a fire, shock cooling may occur (over a period of just a few seconds) e.g. from 1200 °C (surface temperature) to approximately 15 °C. This can also take place when there is no water contact with the hot surface. Rapid cooling may occur by ambient air (according to local climatic conditions) within the range (most frequently) of 0 °C to 35 °C. Various authors have dealt with the different conditions of cooling of composite materials in their studies [6,7]. The differing dilatation/contraction properties of a matrix and aggregate are more distinctive as a result of rapid cooling. Furthermore, some components may rehydrate due to volume expansion as a result of coming into contact with moisture (from both air and water). A typical example of this is CaO (created, for example, as a result of the decomposition of portlandite, calcite or CSH phases), which hydrates relatively quickly back to Ca(OH)2. It is for this reason that this paper focuses on the various conditions of cooling polymer-silicate composites. 2. Materials and methods 2.1. Raw materials and mix proportions The positive effect of other types of light aggregate on the thermal resistance of cement composites has been proved. The results of such research are presented in [8,10]. In order to improve the thermal resistance of the resulting composite and to eliminate the different dilatation contraction parameters of the matrix and aggregate, only porous aggregate was used. Agloporite was specifically used on the basis of high-temperature fly ash sintered by self-firing. The thermal resistance of this type of filler is the result of the conditions under which it is produced and the character of the input raw material (fly ash) which is analyzed in detail in [9]. The effect of this aggregate on the thermal resistance of polymer-silicate composites has not been investigated or explained very much. However, it can be assumed that agloporite represents a suitable filler for thermally resistant composite materials. Aggregate of 0– 2 mm fraction with bulk weight of 780 kg·m-3 was used. The porosity of this aggregate is approximately 37.1%, with an associated absorption capacity of 34.5%. From the viewpoint of its chemical and mineralogical composition 377 Tomáš Melichar and Jiří Bydžovský / Procedia Engineering 151 ( 2016 ) 375 – 382 it is apparent and can therefore be stated to be similar to the original input raw material – blast-furnace fly ash. With regards to minerals, it predominantly contains mullite, with lesser quantities of ȕ-silica, hematite and anorthite, as well as trace amounts of calcite. An observation furnace was used for the verification of the thermal resistance of agloporite. Several grains of agloporite were stressed thermally at the rate of 35 °C·min-1 up to the start of melting in the furnace. The first minute changes were noticed at a temperature of 1,249 °C in the observation furnace. The start of grain surface melting was observed at a temperature of 1,288 °C. With respect to other findings in professional literature (see the previous references to research studies in the introduction), Portland cement of the class 42.5 (CEM I 42.5 R) was selected as the binding component. The cement was substituted by blast-furnace slag (Kotouþ Štramberk; hereinafter BFS) and high-temperature fly ash (power plant Poþerady; hereinafter FA), namely in the proportion of 35%. Micro-silica (hereinafter SF) and a polymer modifier for the matrix (on the basis of vinyl acetate and ethylene; hereinafter PM) were used in minor amounts. Dispersed polymer reinforcement – polypropylene fibres with a length of 3 mm (Fibrin 315; hereinafter PPF) - also formed an integral part of the composite material. The precise composition of both recipes is specified in the table below (see Table 1). Table 1. Mixture proportions of developed composite material. Mixture Composition (kg·m-3) Cement FA BFS SF Aggregate PM PPF Water PSFA 452 215 - 60 1,261 20 1.2 85 PSFS 439 - 241 61 1,262 20 1.2 74 It was necessary to saturate the agloporite with water before mixing the components. The exact quantity of water freed from the agloporite during mixing was very difficult to determine. It is for this reason that no value for the water coefficient is given in the table above (see Table 1). Test specimens were created for testing in the shape of a prism with the dimensions 40×40×160 mm. A layer of 20 mm thick of each mixture was then applied to the concrete substrate. 2.2. Exposition procedure The composite materials on the basis of the polymer-silicate matrix were stored in a humid environment for 72 hours and then at a temperature of 22 °C and relative air humidity of 55% for 28 days before testing. Thermal exposure was tested after 28 days. The basic requirements for testing fire resistance are specified in the standard ýSN EN 1363-1. The test specimens were exposed to temperatures of 400, 600, 800 and 1,000 °C respectively. The thermal exposure of the test specimens (prism shape - dimensions 40×40×160 mm) took place in an electric furnace with temperature rises at a rate of 15 °C·min-1. The controlled cooling process started with 90 minutes isothermal dwell. The last stage of this cooling process was carried out in either one of two ways. The first half of the test specimens were cooled gradually i.e. at rate of approximately 1 °C·min-1. The second half of the thermally exposed specimens were submerged into a water bath with a temperature of approximately 18 °C immediately after termination of the isothermal holding period. The mixtures applied to the concrete substrate were then exposed to a flame, whereby the temperature on the surface and at the composite/substrate interface was scanned. At a temperature of 1,000 °C the exposure was stopped and the specimen allowed to undergo 30 minutes isothermal dwell. The specimens were then cooled with a stream of water with a temperature of approximately 18 °C (simulation of fire-fighting). The thermal exposure (with exception to the cooling stage) of the mixtures applied to the substrate corresponded approximately to the curve contained in the ýSN EN 1363-1 standard. 2.3. Testing details The dimensions, weight, volume weight, compressive strength and flexural tensile strength of the test specimens, and changes therein, were measured before and after thermal exposure. The parameters were determined in accordance with the ýSN EN 12190 and ýSN EN 196-1 standards. In the case of the mixtures applied to the 378 Tomáš Melichar and Jiří Bydžovský / Procedia Engineering 151 ( 2016 ) 375 – 382 concrete substrate, changes in adhesion to the substrate at a temperature of 1,000 °C were also monitored. The determination of adhesion was carried out in accordance with the ýSN EN 1542 standard. Three test specimens were tested for each determination; only the average values are presented below. In the following stage the micro/structural properties were also monitored for the selected substances. The assessment of the potential failures was significant with regards to the effect of shock cooling. It should be noted that these failures may also occur as a result of the rapid cooling of a given substance. The difference in thermal dilatation/contraction of the matrix and aggregate used may be problematic. As a result, cracks inside the given substance or intensive peeling of the surface layer may occur. It is for this reason that SEM and CT methods were used. 3. Results and discussion 3.1. Physical and mechanical parameters In the diagrams below (Fig. 1, 2) a comparison of the average values of the monitored parameters is shown. The volume changes were monitored before and after thermal exposure; each dimension of the specimen in the shape of a prism was monitored separately (Fig. 1a). It can be stated that all the composite specimens, with the exception of one, underwent dimensional changes as a result of contraction. The most significant change was noted in the width of the mixture containing BFS i.e. approximately -0.9%. The smallest change occurred in the longitudinal dimension i.e. not more than -0.4%. It is clear that the use of FA as a substitute component has a more positive effect (than in case of BFS) – dimension change not more than -0.4%. Unambiguous dependencies cannot be deduced when comparing the effects of the cooling conditions on the dimensional changes. The volume weight and change of weight were also monitored parameters. The substitution of cement by FA and BFS both had a similar effect on volume weight. However, this applied only to the thermally non-exposed substances which had a volume weight of 1,420 kg·m-3 and 1,410 kg·m-3 (PSFS and PSFA respectively). The composite on the basis of the matrix containing FA showed a higher level of stability when exposed to increasing temperatures. The one exception to this was the PSFS recipe. When the PSFS recipe was exposed to a temperature of 1,000 °C, there was significantly less weight loss after shock cooling. The recorded loss of weight was in the range of -11.9 to -15.8 %. It is possible to state that a smaller drop in weight resulted from shock cooling. This can only be assumed to a certain extent because gradual cooling may still result in the degradation of individual components of the composite. Shock cooling could prove negative where severe volume changes occur which are diametrically different for the matrix and the filler. In the worst case scenario the composite may peel off the surface or be completely destroyed thereby also resulting in a considerable loss of weight. This did not occur. It can therefore be assumed that the proposed composite is very resistant to sudden temperature changes. This means that elevated temperatures have a dominant effect, rather than rapid drops in temperature. - 0.9 - 0.8 - 0.7 - 0.6 - 0.5 - 0.4 - 0.3 - 0.2 - 0.1 0.0 0.1 22 400 600 800 1000 Dimensional changes [%] Exposition temperature [°C] ǻw - PSFS ǻw - PSFS-W ǻw - PSFA ǻw - PSFA-W ǻh - PSFS ǻh - PSFS-W ǻh - PSFA ǻh - PSFA-W ǻl - PSFS ǻl - PSFS-W ǻl - PSFA ǻl - PSFA-W -25 -20 -15 -10 -5 0 0 200 400 600 800 1000 1200 1400 1600 22 400 600 800 1000 Change of weight [%] Bulk density [kg.m -3 ] Exposition temperature [°C] BD - PSFS BD - PSFS-W BD - PSFA BD - PSFA-W ǻm - PSFS ǻm - PSFS-W ǻm - PSFA ǻm - PSFA-W Fig. 1. (a) dimensional changes; (b) bulk density, weight and their changes. a b 379 Tomáš Melichar and Jiří Bydžovský / Procedia Engineering 151 ( 2016 ) 375 – 382 In the majority of cases, strength parameters are of principal importance for building materials. It is for this reason that the focus of attention was also on the behaviour of the strength characteristics (Fig. 2). As expected, higher strength values for the thermally non-exposed substances were registered for the PSFS recipe, namely 33 N·mm-2 (PSFA – 28 N·mm-2). The residual compressive strength after exposure to a temperature of 1,000 °C ranged between 33 and 41 % (11 and 12 N·mm-2). This can be considered to be a very favourable finding. It is clear from the course of the curves that characterize the changes in compressive strength that the drops in strength are not as considerable as in the case of gradual cooling. Of particular interest is the thermal resistance of PSFA up to a temperature of 600 °C, where a residual strength of approximately 90% (30 N·mm-2) was recorded. The differences between the monitored methods of cooling of the individual substances were negligible after reaching a temperature of 1,000 °C. High-temperature fly ash seems to be a very advantageous component for matrix stabilization from the viewpoint of compressive strength. The developed substances behaved differently when tested for their bending tensile strength (see Fig. 2a). Of particular interest is the effect of the different cooling conditions. The differences between the values of strength achieved after quick and slow cooling were relatively marked. The fact that the PSFS composite is more resistant under gradual cooling, whereas the PSFA substance is more resistant under sudden cooling, is an important finding. The residual bending tensile strength reached approximately 37 to 39 % (1.9 to 2.3 N·mm-2) after exposure of the substances to a temperature of 1,000 °C and gradual cooling. Shock cooling then resulted in a drop in flexural tensile strength up to approximately 18 % of the original value (1.0 N·mm-2). It is clear from the diagrams below (see Fig. 2) that shock cooling had a more noticeable and negative effect on flexural tensile strength. This fact closely emanates the findings in [7]. -70 -60 -50 -40 -30 -20 -10 0 0 5 10 15 20 25 30 35 22 400 600 800 1000 Change of strength [%] Compressive strength [N.mm-2] Exposition temperature [°C] fc - PSFS fc - PSFS-W fc - PSFA fc - PSFA-W ǻfc - PSFS ǻfc - PSFS-W ǻfc - PSFA ǻfc - PSFA-W -90 -80 -70 -60 -50 -40 -30 -20 -10 0 0 1 2 3 4 5 6 7 22 400 600 800 1000 Change of strength [%] Flexural strength [N.mm -2 ] Exposition temperature [°C] ff - PSFS ff - PSFS-W ff - PSFA ff - PSFA-W ǻff - PSFS ǻff - PSFS-W ǻff - PSFA ǻff - PSFA-W Fig. 2. (a) compressive strength and its changes; (b) flexural strength and its changes. Adhesion to the substrate (bond strength) was another significant characteristic. The final average values are given in the table below (see Table 2). Table 2. Bond strength by pull-off of tested mixtures and change by thermal shock. Mixture Bond strength (N·mm-2) Change of bond strength (%) 22 °C 1000 °C 22 °C 1000 °C PSFS 1.78 0.66 0.0 -62.9 PSFA 1.59 0.64 0.0 -59.7 As is apparent from the values mentioned above for the simulation of fire with a temperature of 1,000 °C with consequent shock cooling, the tested composites keep approximately 40% adhesion to the substrate. This is a favourable finding. The PSFA recipe showed slightly better resistance. With regards to the physical-mechanical a b 380 Tomáš Melichar and Jiří Bydžovský / Procedia Engineering 151 ( 2016 ) 375 – 382 parameters, it is evident that high-temperature fly ash has a better ability to stabilize a composite where it concerns extreme thermal resistance. It can be stated, from the viewpoint of changes to the physical-mechanical parameters as a result of elevated temperatures, that the modification of the FA binding component is more successful. The positive effect of substituting 40% cement with fly ash was also demonstrated e.g. in the study [11] where the resistance of modified mortars was investigated, however, only to 800 °C. It can be stated, on the basis of the exploration of facts, that the issues surrounding the effect of fly ash agloporite on thermal resistance under different cooling conditions i.e. up to temperatures of 1,000 °C, have not been investigated in detail so far. 3.2. Micro/structure Pieces of knowledge from other professional studies dealing with the effect of elevated temperatures on composites on the basis of the mixed cement matrix were utilized for the proposed recipes. It is for this reason that substances were developed with a relatively high thermal resistance for which the degradation of the structure (cracks, peeling, etc.) would not be visually (on the surface) apparent after thermal exposure and consequent cooling. This is why a combination of SEM and CT was utilized for the identification of potential defects in the structure. Attention was particularly focused on the occurrence of defects inside the structure of the individual components and also at their interface. Almost no cracks were caused in the structure of the analyzed substances under gradual cooling even after exposure to 1,000 °C. De facto no cracks were noticed in the structure of the substances exposed to 400 and 600 °C. Only an insignificant percentage of cracks was noticed after rapid cooling of the specimens exposed to 800 °C. The pictures below (see Fig. 3 and 4) of the specimens (dimensions 40×40×160 mm) exposed to 1,000 °C and shock cooled (with water) illustrate this. As is evident, cracks can be seen inside the structures of the specimens after rapid cooling from 1,000 °C. The course of the cracks is irregular. Fig. 3. CT – structure of sample cooled from 1000 °C by water shock (a) PSFS; (b) PSFA. Both the matrix and the aggregate were damaged by cracks. However, the cracks were predominantly found in the matrix. Identifying a contact zone between the matrix and the aggregate was relatively difficult; however, it can be stated that no cracks were noticed in this zone. The pictures from CT also correspond to the ascertained strength values which assess FA as having greater thermal resistance. PSFS shows more cracks with a larger width. However, the maximum identified crack width did not exceed 0.12 mm. This finding is most probably associated, amongst other things, with the slower involvement of FA in hydration reactions. FA proved to be considerably more useful as an inert filler, the composition of which ensured the stability of the resulting composite with respect to the shock effect of exposure to elevated temperatures. 4.5 mm a b 381 Tomáš Melichar and Jiří Bydžovský / Procedia Engineering 151 ( 2016 ) 375 – 382 Fig. 4. CT – detail of structure of sample cooled from 1000 °C by water shock (a) PSFS; (b) PSFA. The findings from CT were confirmed using SEM. More distinct cracks were only identified in the shock cooled specimens from 1,000 °C. The selected SEM pictures below illustrate this (see Fig. 5). This particularly applies to those substances exposed to 1,000 °C and cooled gradually in the furnace. The contact zone, matrix and aggregate were observed to be intact. Micro-structural analysis revealed that the use of porous aggregate on the basis of sintered fly ash – agloporite results in the almost perfect bonding of the matrix to the surface of the aggregate i.e. even at elevated temperatures. Fig. 5. SEM of sample cooled in furnace from 1000 °C (a) PSFS; (b) PSFA. 4. Conclusions The aim of the research presented in this paper was to assess newly developed recipes for polymer-silicate composites for exposure to extreme thermal stress (up to 1,000 °C). Attention was focused on the different conditions for cooling the substances. Based on the aforementioned results the following conclusions can be drawn: (1) Porous aggregates on the basis of sintered fly ash represents a suitable filler for the production of composite materials that are required to be resistant to the effects of extreme thermal conditions. (2) High-temperature fly ash seems to be a more advantageous stabilization component for the matrix from the viewpoint of thermal resistance, specifically residual strength. The residual compressive strength reached approximately 41% after exposure to a temperature of 1,000 °C and shock cooling with water. The residual compressive strength of the substance modified with blast-furnace slag was approximately 33%. It means that fly ash will find its application rather as an inert filler when compared with 28 days old slug. It is necessary to consider the aforementioned properties over a longer time horizon e.g. 90 and 180 days, with regards to continuing related research. 2.0 mm a b a b 382 Tomáš Melichar and Jiří Bydžovský / Procedia Engineering 151 ( 2016 ) 375 – 382 (3) The behaviour of the tested substances was very different from the viewpoint of compressive and tensile stress. Bending tensile strength is much more sensitive than compressive strength with respect to shock cooling. However, the fact that cohesion (which represents also tensile stress), or its drop as a result of thermal exposure, corresponds more to compressive strength, is an important finding. This is critical, for example, for repair compounds where, in accordance with the applicable standards, emphasis is not laid on bending tensile strength, but only on compressive strength and cohesion. (4) Assessment of the formation of structural failures represented a substantial part of the research. Unusually for building materials, CT apparatus was utilized for this purpose. Potential cracks were identified by means of CT. It was unambiguously demonstrated that the formation of noticeable cracks occurred only after stressing under temperatures of 1,000 °C with subsequent shock cooling (in a water bath with a temperature of approx. 18 °C). However, no cracks were identified on the surfaces of the test specimens. A slightly higher number of cracks were identified in those substances containing blast-furnace slag. This fact is associated with the more intensive involvement of slag in the formation of hydration products in the matrix. In contrast, fly ash acted rather as an inert filler which most probably resulted in a less compact structure which was more resistant to shock stressing under elevated temperatures. The reason for this is the fact that the re-hydration of CaO to Ca(OH)2 takes place during shock cooling to which a more flexible (less compact) structure is more resistant. (5) A temperature of 1,000 °C seems to be critical for the compressive strength of PSFS and PSFA substances after shock cooling, whereas in the case of bending tensile strength, the most considerable drop already occurs at 800 °C. Acknowledgements This paper has been worked out with the financial support of The Czech Science Foundation, project GA1507657S, “Study of kinetics of processes occurring in the composite system at extreme temperatures and exposed to an aggressive environment”, as well as from the budget of the project FAST-S-16-3274 “Research and development of advanced materials for industrial floors”. References [1] O. Arioz, Effects of elevated temperatures on properties of concrete, Fire Safety J. 42 (2007) 516–522. [2] S.K. Handoo, S. Agarwal, S.K. Agarwal, Physicochemical, mineralogical, and morphological characteristics of concrete exposed to elevated temperatures, Cement Concrete Res. 32 (2002) 1009–1018. [3] R. Kh. Ibrahim, R. Hamid, M.R. Taha, Fire resistance of high-volume fly ash mortars with nanosilica addition, Constr. Build. Mater. 36 (2012) 779–786. [4] N. Farzadnia, A.A. Abang Ali, R. Demirboga, Characterization of high strength mortars with nano alumina at elevated temperatures, Cement Concrete Res. 54 (2013) 43–54. [5] N. Farzadnia, A.A. Abang Ali, R. Demirboga, M.P. Anwar, Characterization of high strength mortars with nano Titania at elevated temperatures, Constr. Build. Mater. 43 (2013) 469–479. [6] M.M Shoaib, S.A Ahmed, M.M Balaha, Effect of fire and cooling mode on the properties of slag mortars, Cement Concrete Res. 31 (2001) 1533–1538. [7] S. AydÕn, Development of a high-temperature-resistant mortar by using slag and pumice, Fire Safety J. 43 (2008) 610–617. [8] F. Koksal, O. Gencel, M. Kaya, Combined effect of silica fume and expanded vermiculite on properties of lightweight mortars at ambient and elevated temperatures, Constr. Build. Mater. 88 (2015) 175–187. [9] V. ýerný, Quality of the structure of ash bodies based on different types of ash, Materiali in tehnologije 49 (2015) 601–605. [10] J.-P. Won, H.-B. Kang, S.-J. Lee, S.-W. Lee, J.-W. Kang, Thermal characteristics of high-strength polymer–cement composites with lightweight aggregates and polypropylene fiber, Constr. Build. Mater. 25 (2011) 3810–3819. [11] A. Nadeem, S.A. Memon, T.Y. Lo, Mechanical performance, durability, qualitative and quantitative analysis of microstructure of fly ash and Metakaolin mortar at elevated temperatures, Constr. Build. Mater. 38 (2013) 338–347.