1-s2.0-S0950061815303093-main

Construction and Building Materials 98 (2015) 237–249

Contents lists available at ScienceDirect

Construction and Building Materials

journal homepage: www.elsevier .com/locate /conbui ldmat

Fire resistance and mechanical properties of carbon nanotubes – claybricks wastes (Homra) composites cement

http://dx.doi.org/10.1016/j.conbuildmat.2015.08.0740950-0618/� 2015 Elsevier Ltd. All rights reserved.

⇑ Corresponding author at: Chemistry Department, Faculty of Science, Ain ShamsUniversity, Cairo, Egypt.

E-mail address: [email protected] (M.S. Amin).

M.S. Amin a,b,⇑, S.M.A. El-Gamal b, F.S. Hashemb

aDepartment of Basic Sciences and Technology, Community College, Taibah University, Saudi ArabiabChemistry Department, Faculty of Science, Ain Shams University, Cairo, Egypt

h i g h l i g h t s

� Effect of carbon nanotubes (CNTs) on the fire resistance of Homra/OPC blends.� Small additions of CNTs to OPC–Homra blends improve the fire resistance.� 0.1% CNT could be considered as the optimum addition to each mix.� Presence of CNTs does not affect the hydration reaction of OPC or Homra/OPC blends.� Presence of CNTs does not affect the microstructure of the formed hydrates.

a r t i c l e i n f o

Article history:Received 4 June 2015Received in revised form 4 August 2015Accepted 9 August 2015Available online 24 August 2015

Keywords:Clay bricks wastes (Homra)Carbon nanotubesThermal treatmentCompressive strengthPozzolanic reactionComposite

a b s t r a c t

This paper aims to evaluate the effect of carbon nanotubes (CNTs) on the mechanical properties and thefire resistance of Homra/OPC blends; Homra represents the solid waste generated from clay bricks indus-try in Egypt. The nanocomposites, thus produced, were obtained by different additions of CNTs of 0.02,0.05, 0.1 and 0.2 mass% to Homra/OPC blends, prepared by the partial replacement of OPC by 10, 20and 30 mass% by Homra. Fire resistance of the nanocomposite blends was studied by firing of the variousblended cement pastes at 300, 600, and 800 �C for 3 h. The compressive strength values were determinedfor different blended cement–CNTs composites at each firing temperature, in addition, the phase compo-sition, thermal analysis and microstructure were investigated for some selected samples. The resultsobtained revealed that addition of 0.1 (mass%) of CNTs showed better improvements in the thermaland mechanical properties of the hardened Homra/OPC blended cement–CNTs composites. XRD, DSCand SEM results revealed that the presence of CNTs does not affect the rate of hydration reaction ofthe neat OPC or Homra/OPC blends; but increases their compressive strength and fire resistance throughtheir physical contribution; where it can further fill in the pores between the hydration products and actsas bridges between hydrates and across cracks.

� 2015 Elsevier Ltd. All rights reserved.

1. Introduction

Cement products are premiermaterials in construction industry.They have excellent compression properties but are weak in ten-sion. To increase the tensile strength, these products can be rein-forced with bars, rods, and fibers or prestressing. Introduction ofnanoparticles in cement based materials has gained popularity inrecent years due to their excellentmechanical properties and appli-cation potential. Addition of carbon nanoparticles in the

cementitious materials may provide extra-ordinary strengthincrease as well as controlling cracks prevention [1,2].

A great potential has been identified in the modification ofmechanical properties of cement based materials using carbonnanotubes (CNTs). Many studies demonstrated that the incorpora-tion of CNTs into cementing matrices lead to improve theirmechanical properties [3–10]. CNTs occur as single-walled carbonnanotubes (SWCNTs) and multi-walled carbon nanotubes(MWCNTs). SWCNTs are composed of a single graphite sheet rolledinto a long hollow cylinder, whereas MWCNTs are nested arrays ofSWCNTs. The average diameter of an individual SWCNT is on theorder of 1 nm whereas the average diameter of an individualMWCNT is on the order of 10 nm [2]. Carbon nanotubes havedesirable mechanical, show very high thermal conductivity of

238 M.S. Amin et al. / Construction and Building Materials 98 (2015) 237–249

1700–3000W/mK and very low electrical resistivity of 5 � 10�8–2 � 10�6 Om. In addition, upon stress, composites containingCNTs show a ‘piezoresistive response’ wherein the electrical prop-erties are changed with respect to different stress levels [11,12].

However, beside the enhancement of the mechanical and elec-trical properties of cement composites by addition of CNTs, severalchallenges must still be considered. One of these is achieving effec-tive dispersion of CNTs at the single tube level. CNTs have highaspect ratios and strong van der Waals self-attraction betweennanotubes, and tend to form CNT bundles [13,14]. Poor dispersionof CNTs leads to the formation of many defect sites in thenanocomposite and limits the efficiency of homogeneous disper-sion of CNTs in the matrix. Another major challenge is to enhancethe interfacial interaction between CNTs and hydration products ofcement; CNTs are expected to provide mechanical reinforcementbetween hydration products of cement with nano-scale dimen-sions. However, hydration products such as calcium–silicate–hydrates (C–S–H), calcium hydroxide (CH) and ettringite have similaror larger size than CNTs, and consequently only a few CNTs couldbe anchored by the hydration products in studies reported to date[15]. Carbon nanotubes were found to improve strength propertiesof Portland cement, the reason given was that the functionalizedcarbon nanotubes could provide chemical bonds between the –COOH groups of the nanotubes and the calcium silicate hydratephase (CSH) of the cement matrix, which enhanced the transferof stresses. Furthermore, previous research has demonstrated thatthe addition of multi-walled carbon nanotubes of less than 1 mass% of cement can greatly improve the mechanical properties of thecement composites [16–18].

Combination of CNT in blended cement with other materialssuch as fly ash, nanoclay, nano Fe needles, bagasse fiber and carbonnanofibers among others has also been studied [19–23]. Homra(clay bricks waste) is a solid waste material, which is constitutedmainly of silica quartz, aluminosilicate, anhydrite, and hematite.Therefore, it acts as a pozzolanic material [24]. Homra as a poz-zolanic material reacts with lime liberated from the hydration ofordinary Portland cement (OPC). This pozzolanic reaction improvesthe microstructure of cement pastes and also increases their heatresistance [25].

The aim of this study is to identify the effect of both Homra andcarbon nanotubes on the mechanical and thermal resistivity ofPortland cement pastes.

2. Experimental

2.1. Materials

The cement used in this study was ordinary Portland cement (OPC), suppliedfrom Lafarge cement company, Suez, Egypt, with Blaine surface area of 3320 cm2/g. Blender was crushed clay bricks (Homra) of particle size 60.125 mm andBlaine surface area of 3300 cm2/g. Table 1 shows the chemical oxide compositionfor OPC and Homra.

The purified multi-walled carbon nanotubes (MWCNTs) with surface area of93.81 m2/g and purity > 90%. The outside estimated diameter and length ofMWCNTs ranged from 10–40 nm to 5–10 lm respectively. The density ofMWCNTs was about 2.1 g/cm3 and their electrical conductivity was more than100 S/cm, provided by Egyptian Petroleum Research Institute (EPRI), Cairo, Egypt,were used. Morphology and microstructure of MWCNTs are shown in Fig. 1.

Polycarboxylate surfactant (Sika Viscocrete 5230 L) with specific gravity 1.08 g/ml was supplied from Sika Company, Elobour City, Egypt, to assist the dispersion ofMWCNTs.

Table 1Chemical oxide composition of OPC and Homra (mass%).

SiO2 Al2O3 Fe2O3 CaO M

OPC 18.57 4.29 3.75 62.45 1Homra 74.80 14.03 5.04 1.25 1

2.2. Methodology

Different cement blends were prepared from different OPC–Homra dry mixes.Table 2 shows the percentage composition of the different cement blends and theirdesignations. Each dry cement blend was mechanically mixed in a porcelain ballmill for 8 h to assure complete homogeneity of the dry mixture.

For the preparation of MWCNTs dispersions, the surfactant (Viscocrete), wasused; suspensions were prepared by mixing MWCNTs in an aqueous solution (usingthe whole mixing water) containing different amounts of surfactant. TheCNTs/surfactant ratio was of 1:3 and the resulting suspensions were sonicated atroom temperature for 1 h.

Different cement pastes were made using a water/solid (W/S) ratio of 0.30. Eachpaste was prepared by mixing the dry mix with the required amount of water con-taining the dispersed MWCNTs for about 3 min. After complete mixing, the resul-tant paste was molded into cubic specimens by using 1 inch cube molds. Themolds containing specimens were cured at about 100% relative humidity for 24 hto attain the final setting; then the cubic specimens were demolded and curedunder tap water at room temperature for different time intervals of 3, 7, 28 and90 days.

At each time interval, compressive strength tests were performed on the hard-ened blended cement pastes using three cubic specimens at each hydration time,and the average value was recorded. The resulting crushed specimens were ground,and the hydration reaction was stopped using the method described in an earlierpublication [26]. The samples were then dried at 90 sC for 3 h in CO2-free atmo-sphere and maintained in a desiccator containing soda lime and CaCl2 until the timeof testing was reached.

In addition, the specimens cured for 28 days were subjected to thermal treat-ment in a muffle furnace at 300, 600 and 800 sC for 3 h with a heating rate of10 sC min�1. The thermally heated specimens, after cooling in a desiccator weresubjected to compressive strength test and the residual strength values wererecorded. The percentage of residual strength was calculated as follows:

Residual strength % ¼ ½ðC:S:Þt=ðC:S:Þ0� � 100

(C.S.)t: compressive strength after firing at temperature t �C.(C.S.)0: compressive strength at room temperature.

The phase composition and microstructure of the formed hydrates were inves-tigated by X-ray diffraction (XRD), differential scanning calorimetery (DSC) andscanning electron microscopy (SEM).

3. Results and discussion

3.1. Compressive strength

The results of compressive strength of the hardened OPC–Homra composite cement pastes blends are graphically repre-sented as a function of hydration age in Fig. 2. For all of the pastesmade of the OPC and OPC–Homra blends, the compressive strengthwas found to increase continuously with increasing age of hydra-tion. This increase in compressive strength is mainly attributedto the formation and later accumulation of hydration productswhich act as binding centers between the remaining unhydratedparts of cement grains, (Fig. 2). The hydration products of compos-ite blends, mainly as calcium silicate hydrates (CSH, I & II) and cal-cium aluminate hydrates represent the main binding centersbetween the remaining unhydrated parts of OPC and Homra grains.On the other hand, the pastes made of mix B0 (90% OPC + 10%Homra) possess the highest compressive strength values especiallyat later ages of hydration (28 and 90 days) as compared to otherpastes containing 0, 20 and 30% Homra. The highest strength val-ues of the hardened pastes made of mix B0 can be attributed tothe pozzolanic reaction between the free calcium hydroxide, liber-ated from Portland cement hydration, with Homra to form exces-sive amounts of hydration products, mainly as CSH, CAH andCASH gels as well as crystalline hydrates, that serve as micro-

gO SO3 K2O Na2O Cl� LOI

.88 3.20 0.28 0.32 – 2.10

.30 0.80 – – – –

Fig. 1. TEM micrographs of multi-walled carbon nano-tubes.

Table 2The percentage composition of the different mixes and their designations.

Mix Mix Proportions (mass%) W/S ratio

OPC Homra CNTs Superplasticizer

A0 100 – – – 0.30A1 100 – 0.02 0.06 0.30A2 100 – 0.05 0.15 0.30A3 100 – 0.10 0.30 0.30A4 100 – 0.20 0.60 0.30B0 90 10 – – 0.30B1 90 10 0.02 0.06 0.30B2 90 10 0.05 0.15 0.30B3 90 10 0.10 0.30 0.30B4 90 10 0.20 0.60 0.30C0 80 20 – – 0.30C1 80 20 0.02 0.06 0.30C2 80 20 0.05 0.15 0.30C3 80 20 0.10 0.30 0.30C4 80 20 0.20 0.60 0.30D0 70 30 – – 0.30D1 70 30 0.02 0.06 0.30D2 70 30 0.05 0.15 0.30D3 70 30 0.10 0.30 0.30D4 70 30 0.20 0.60 0.30

Fig. 2. Compressive strength versus age of hydration for the hardened OPC–Homrablended cement pastes.

M.S. Amin et al. / Construction and Building Materials 98 (2015) 237–249 239

fills; this filling of gel pores resulted in densification and subse-quently enhanced the gel/space ratio and degree of hydration, con-sequently an increase in compressive strength values wereobtained. In fact, the CSH that partially fills the capillary pores alsocontains pores but these are too small to initiate cracking [27].

It is also clear from the results of Fig. 2 that the compositecement pastes containing 20 and 30% Homra (mixes C0 and D0)possess lower strength values. The decrease in strength of thehardened pastes can be attributed to a reduced amount of Ca(OH)2 to activate the waste brick Homra [28] and/or the dilutioneffect of the OPC as a results of addition of high percentages ofHomra at expenses of OPC clinker.

In conclusion, Homra (as an artificial pozzolana) can be used upto 10% as a replacement to ordinary Portland cement in the poz-zolanic OPC–Homra blends without any reduction in strength atlater ages of hydration. Further increase in the percentages ofHomra (up to 30 mass%) leads to a decrease in strength as

Fig. 3. Effect of addition of CNTs on (a) compressive strength values and (b) relativecompressive strength values for different hardened neat OPC pastes.

Fig. 4. Effect of addition of CNTs on (a) compressive strength values and (b) relativecompressive strength values for different hardened OPC–Homra pastes (series B).

240 M.S. Amin et al. / Construction and Building Materials 98 (2015) 237–249

compared to OPC paste, these results are in agreement with theresults obtained in previous investigations [29–31].

The effect of carbon nanotubes (CNTs) additions on the com-pressive strength values of neat OPC pastes are shown in Fig. 3a,b. All OPC mixes containing CNTs showed an enhancement in thecompressive strength compared to the neat OPC (mix A0) espe-cially at early age of hydration. The enhancement in the compres-sive strength had the order A3 > A2 > A4 > A1. These results agreewith the data obtained earlier by many researchers which can beattributed to the physical contribution of carbon nanotubes whereit can fill in the pores between the hydration products such as cal-cium silicate hydrates (CSH, I, II) and ettringite. It was proved thatthe porosity of the pastes containing CNTs decrease as a result ofthe filler effect of CNTs, leading to a denser microstructure thanthat of A0 mix [32–34].

The effect of carbon nanotube (CNT) additions, namely; 0.02,0.05, 0.1 and 0.2 mass% on compressive strength values of the dif-ferent OPC–Homra blends are shown in Figs. 4–6.

Fig. 4a shows the variation of compressive strength values ver-sus age of hydration for the composite cement made of OPC–10%Homra with different additions of CNT’s, namely; 0% (mix B0),0.02% (mix B1), 0.05 (mix B2), 0.1 (mix B3) and 0.2 (mix B4). Theresults show that the compressive strength values obtained formixes B were found to increase with increasing carbon nanotubescontent and the highest compressive strength values were found at0.1 mass% CNTs addition. In other words, the increase in thestrength values was in the order of mix B3 > mix B2 > mixB1 > mix B0 corresponding to 0.1, 0.05, 0.02, and 0 mass% additionof CNTs, respectively. On contrary, addition of 0.2 mass% leads tolower compressive strength values, i.e. the strength values of mixB4 than those of mix B3; but still has higher than those of othermixes, namely: mixes B0, B1and B2.

The increase of the compressive strength values of cementpastes containing 10% mass% Homra as a result of addition of dif-ferent ratios of CNTs agrees with the previously reported workson the use of CNTs as addition to the neat Portland cement as wellas to blended cement pastes [18,23,35–37]. The higher strengthvalues of these OPC–Homra–CNTs composites may be attributedto the physical contribution of carbon nanotubes where it can fillin the pores between the hydration products such as calcium sili-cate hydrates CSH, C–A–H and C–A–S–H. It was proved that theporosity of the pastes containing CNTs decrease as a result of thefiller effect of CNTs, leading to a denser microstructure than thatof the control mix, this effect supports the enhanced compressivestrength values of OPC–Homra mixes containing CNTs [32–34].Moreover, the reinforcing effect of CNT plays an important rolein increasing the compressive strength values of the OPC–Homrablends containing CNTs rather than the blends free from CNT. Incontrast, addition of 0.2 mass% of CNTs to mix B0 (mix B4) leadsto lower compressive strength values compared to those of mixB3 (containing 0.1 mass% CNTs), this finding could be attributedto the reagglomeration of the CNTs that tend to adhere togetherdue to strong Van der Waal’s forces leading to less dispersionamong the matrix of cement paste. It could be concluded thatthe compressive strength values of the cement blends containingCNTs is affected by not only the total porosity but also by the dis-persion of the CNTs [12].

Fig. 4b shows the relative strength of the Homra blends (90%OPC–10% Homra) containing different additions of CNTs. It wasfound that the hardened pastes made of (mixes B1–B4), all mixescontaining CNTs possess higher relative compressive strengthcompared to those made without CNTs (mix B0) indicating theenhanced and reinforcing effects of the CNT as mentioned previ-ously. At the same time mix B3, containing 0.1% CNT, has the

M.S. Amin et al. / Construction and Building Materials 98 (2015) 237–249 241

highest relative strength percentage compared to other mixes (B0,B1, B2 and B4) and its relative strength values as compared to mixB0 (0% CNTs) at 28 and 90 days are 125% and 120%, respectively.This result is believed to be due to the good dispersion of CNTs par-ticles up to 0.1% addition as well as the filler effect of the CNTsbetween the hydration products which by its role has a large effectin decreasing the total porosity of the blended cement matrix andthe subsequent increase in compressive strength. It was alsonoticed that the values of compressive strength of the hardenedpastes made of mix B3 (90% OPC–10% Homra–0.1% CNTs) at28 days and 90 days (104.02 and 108.24 MPa, respectively) arehigher than those of mix A0, (73.92 and 87.16 MPa, respectively),(Figs. 2 and 4a); this means that the values of compressive strengthof mix B3 increased with 24–40% than those of neat OPC at 28 and90 days of hydration, respectively. In conclusion, 0.1% CNTs is anoptimum addition to the blend of (90% OPC–10% Homra), whichgives the highest strength.

Fig. 5a shows the development of the mean values of the com-pressive strength obtained for the blended cement pastes (80%OPC–20% Homra) containing 0, 0.02, 0.05, 0.1 and 0.2 mass%CNTs, respectively. It was noticed that the paste made of mix C3(0.1 mass% CNTs) has the highest compressive strength values atall ages of hydration. In fact, the compressive strength increaseswith the increase of CNTs until it reaches the optimum addition(0.1%) and then it decreases by further addition of CNTs.

Obviously, the improvement of compressive strength by theaddition of CNTs up to 0.1% was attributed to the crosslinking ofCNTs fiber with the hydration products where CNTs acting asnucleating centers for the formed hydration products; this effectprevent microcracks formation and assists the decrease of the totalporosity as well as the filler effect of the CNTs. Furthermore, thehigher percentage of CNTs addition (0.2%) may permit agglomera-tion of the CNTs around the cement grains leading to partial

Fig. 5. Effect of addition of CNTs on (a) compressive strength values and (b) relativecompressive strength values for different hardened OPC–Homra pastes (series C).

hydration of cement grains and production of hydration productshaving weak binding forces [38] and/or the reagglomeration ofCNTs as a result of their bad dispersion.

Fig. 5b shows the relative strength of the cement pastes made ofOPC–Homra blends (80% OPC–20% Homra) containing differentadditions of CNTs. It was found that the hardened pastes made ofall mixes containing CNTs have higher relative strength valuescompared to those made without CNTs indicating the enhancedand reinforcing effect of the CNTs. At the same time the paste madeof mix C3, containing 0.1% CNT, has the highest relative strengthpercentages as compared to those made of other mixes (C0, C1,C2 and C3), its relative strengths to mix C0 (0% CNTs) were 127and 150% at 28 and 90 days, respectively. It was also noticed thatthe values of compressive strength values of the pastes made ofmix C3 (80% OPC–20% Homra–0.1% CNTs) at 28 and 90 days(81.57 and 102.16 MPa, respectively) are higher than those ofmix A0 (73.92 and 87.16 MPa, respectively) at the same ages ofhydration, (Figs. 2 and 5a); this means that the values of compres-sive strength of mix B3 increase by the percentages of 10 and 17%than blank mix (100% OPC) at 28 and 90 days of hydration, respec-tively. Mix C3 represents the optimum addition to the compositecement blend, which gives the highest compressive strength com-pared to both mix A0 and C0.

The same observations in compressive strength results arenoticed for the pastes containing 30% Homra without and withCNTs (Fig. 6a, b). Evidently, the results indicate that 0.1% CNTs rep-resents the optimum addition of the cement blends containing(70% OPC–30% Homra). The compressive strength values of mixD3 at 28, 90 days are 83.14 and 95.59 MPa, respectively indicatingthat the relative strength values of mix D3 to mix D0 (withoutCNTs) are 130 and 131% at 28 and 90 days, respectively. Mix D3have higher compressive strength values compared to those ofmix A0. The relative strength of mix D3 to mix A0 are 113 and

Fig. 6. Effect of addition of CNTs on (a) compressive strength values and (b) relativecompressive strength values for different hardened OPC–Homra pastes (series D).

Fig. 7. Relative residual compressive strength of hardened cement pastes fired atdifferent temperatures (series A).

242 M.S. Amin et al. / Construction and Building Materials 98 (2015) 237–249

110% at 28 and 90 days, respectively (Figs. 2 and 6). It can be con-clude that 0.1% addition of CNTs represents the optimum additionto each Homra–OPC composite blends as well as higher compres-sive strength values as compared to those of neat OPC.

Fig. 8. Relative residual compressive strength of hardened cement pastes fired atdifferent temperatures: (a) series B, (b) series C and (c) series D.

3.2. Effect of firing temperature on the compressive strength of thedifferent Homra–CNT cement pastes

Relative residual compressive strengths obtained for the hard-ened cement pastes made of OPC–Homra blends at the fired tem-perature of 300, 600 and 800 �C are shown in (Figs. 7 and 8).

Fig. 7 shows the percentages of residual compressive strengthvalues of the pastes made of mixes A0–A4 at the different firingtemperatures. All mixes showed an increase in the percentage ofthe residual strength upon heating up to 300 �C. This increase isfollowed by subsequent decrease in residual compressive strengthupon heating at 600 and 800 �C. Mix A0 showed an increase in theresidual strength about 20% (relative to the value recorded at28 days of hydration) by heating at 300 �C; while mixes A1, A2,A3 and A4 showed an increase in the residual compressive strengthby 21, 26, 41 and 28% respectively. This increase in percentage ofresidual strength can be attributed to internal autoclaving processleading to filling up of pores by additional hydration products.However, the pastes made of mixes containing different additionsof CNTs (mixes A1–A4) showed low resistance to firing at 600 and800 �C since they nearly showed the same lower values of residualstrength as mix A0 at these two temperatures.

Fig. 8a shows the relative compressive strength values of blankmix (100% OPC, A0) as well as 90% OPC–10% Homra blends con-taining different additions of CNTs (0, 0.02, 0.05, 0.1 and 0.2%). Itis observed from the results of Fig. 8a indicate that all of the pastesmade of mixes A0, B0, B1, B2, B3 and B4 show almost the sametrend upon firing at different treatment temperatures namely(300, 600 and 800 �C) in which the residual compressive strengthvalues increase with heating up to 300 �C. This increase is followedby subsequent decrease in residual compressive strength with fir-ing at 600 and 800 �C. Obviously the residual compressive strengthincreases by about 20% with thermally treated up to 300 �C. On theother hand, the mixes B0, B1, B2, B3 and B4 show an increase in therelative residual compressive strength values by about 21, 22, 29,32 and 15%, respectively upon heating at 300 �C relative to theiroriginal values recorded at 28 days of hydration. The increase inresidual compressive strength values in mix A0 by heating at300 �C may be attributed to internal autoclaving process and fillingup of pores with additional hydration products. For mix B0 (90%OPC–10% Homra) the increase in relative compressive strength

values is attributed to an additional reason, beside the internalautoclaving process, which enhances the pozzolanic reaction withfree Ca(OH)2 leading to the formation of more C–S–H, which form adenser and close-packed structure [39]. Addition of CNTs to mix B0show marked increase in the relative residual compressivestrength, especially for the mix containing 0.1% CNTs (mix B3).These higher values are attributed to the role of CNTs in densifica-tion of the hydrated matrix of the cement mix as a result of its fillerand reinforcing effects. The higher relative residual compressivestrength values of mix B3 compared to other mixes upon firingat 300 �C indicates that the percentage addition of 0.1% CNTs rep-resents the optimum addition to this blend (90% OPC–10% Homra)which is attributed to the good dispersion of 0.1 mass% CNTs.However, the residual compressive strength decreases by about23% upon firing mix A0 (100% OPC) at 600 �C; whereas, for mixesB0, B1, B2, B3 and B4, the relative residual compressive strengthvalues decreased by the values of 23, 22, 20, 18 and 29%, respec-tively, by firing at 600 �C. This decrease is attributed to the decom-position of the cementitious materials and decomposition of thefree Ca(OH)2, which loses its combined water after 500 �C [40]. Itis evident that the mix containing 0.1% CNTs (mix B3) loses theleast value of residual compressive strength values upon firing at

Fig. 9. XRD pattern of neat OPC pastes before and after firing. (a) without CNTs and(b) with 0.1 CNTs (mass%).

M.S. Amin et al. / Construction and Building Materials 98 (2015) 237–249 243

600 �C compared to the mixes containing other additions (0, 0.02,0.05, 0.2% CNTs) as well as when compared to blank OPC (mixA0). The increase of firing temperature up to 800 �C leads to thereduction of compressive strength values by 80, 66, 65, 63, 61and 68% for mixes A0, B0, B1, B2, B3 and B4, respectively; the greatloss in strength for the hardened pastes fired at 800 �C is attributedto the formation of new cracks inside the fired specimens.Evidently, the complete thermal decomposition of the bindinghydration products, such as CSH leading to an enlargement inthe microcracks. It is noticed that the partial substitution of OPCby Homra improves the fire resistance of OPC especially at 10%substitution; this is mainly due to that Homra is composed onlyof aluminosilicates; which react with the decomposed Ca(OH)2forming strong ceramic material. Moreover, addition of CNTs tothe Homra–cement blends enhanced the resistance of these pastestoward firing and 0.1% CNTs addition represents the optimumaddition which produces the highest improvement in compressivestrength when heated at 300 �C and the least loss in strength uponheating at 600 and 800 �C compared to other mixes.

Fig. 8b shows the relative compressive strength values of blankmix (100% OPC, A0) as well as 80% OPC–20% Homra blends con-taining different additions of CNTs (0, 0.02, 0.05, 0.1 and 0.2%).Evidently, the results of Fig. 8b indicate the same trend observedas obtained in Fig. 8a, in which the residual compressive strengthincrease with firing up to 300 �C. This increase is followed by sub-sequent decrease in residual compressive strength with heating at600 and 800 �C. The gain in strength by firing mixes C0, C1, C2, C3and C4 at 300 �C was found to be 26, 32, 42, 45 and 15%, respec-tively. It is seen, that the values of the relative residual obtainedfor the different pastes made of mixes Co, C1, C2, C3 and C4 arehigher than those of mixes B0, B1, B2, B3 and B4, respectively.The higher the relative compressive strength values of the mixescontaining 20% Homra are due to the higher pozzolanic reactionof Homra up to 300 �C, leading to formation of additional hydrationproducts (C–S–H, C–A–H, and C–A–S–H), deposit in the open poreswith a consequent densification of the structure of the hardenedpaste [25]. On the other hand, the gain of strength of the mixesC1, C2, C3 and C4 made of 0.02, 0.05, 0.1 and 0.2% CNTs additionsto mix C0 (80% OPC–20% Homra) are higher than the strength val-ues of the same blend without CNTs additions by heating at 300 �C,obtained by Heikal [25]. This observation reveals the role of CNTsin enhancing the fire resistance of the blended cement pastes. Onthe other hand, the pastes made of mixes C1, C2, C3 and C4 losetheir strength upon firing at 600 and 800 �C, but with lesser extentscompared to those of mixes B1, B2, B3 and B4. In addition replace-ment of OPC by 30% Homra (series D) enhanced the fire resistanceof all mixes (without and with CNTs) compared to those of othermixes (containing 10 and 20% Homra), (Fig. 8c). It was found thatfor mix D3 (70% OPC–30% Homra–0.1% CNT) the compressivestrength increases to the extent of 50% upon firing at 300 �C, whichis the highest gain in strength compared to all other specimensunder investigation, and the compressive strength decrease bythe values of 5 and 50% upon firing at 600 and 800 �C, respectively.These results prove that the blend (70% OPC–30% Homra) could beconsidered as a successful blend in fire resistance if small additionsof CNTs, especially 0.1%, are used; this conclusion contradicts theresults obtained by Heikal [25]. Obviously the positive effect ofaddition of small percentages of CNTs to the cement–Homrablends (OPC replaced by 30% Homra) greatly exceeds the negativeeffect as a result of decreasing the clinker content in this blend.

The main conclusions derived from the thermal resistance stud-ies are: (1) replacement of OPC by 10, 20 and 30% Homra enhancethe residual compressive strength of OPC–Homra blends upon fir-ing at 300, 600 and 800 �C compared to neat OPC (mix A0); (2)small additions of CNTs to the blends containing Homra (up to30% Homra) improve the fire resistance of the mixes compared

to the Homra blends without CNTs; (3) 0.1% CNTs could be consid-ered an optimum addition to each mix; (4) mix D3 (70% OPC–30%Homra–0.1% CNTs) has the highest fire resistance compared to allother mixes as a result of the extra pozzolanic reaction of Homraas well as to the effect of the well-dispersion of CNTs.

3.3. X-ray diffraction

XRD patterns of hardened OPC paste (mix A0) before and afterthermal treatment up to 800 �C, are shown in Fig. 9a. After 28 daysof hydration the XRD pattern of the neat OPC paste indicated theformation of portlandite (CH), calcium silicate hydrates (CSH) asmain hydration products and the peaks characteristic to CaCO3

are also appeared in the XRD pattern. The peaks characteristic forunhydrated parts of calcium silicate phases (C3S and b-C2S) as wellas minor amounts of quartz were also appeared in this pattern.Firing the specimens at 300 �C caused a notable increase in theintensities of the peaks characterizing to the main hydration prod-ucts (CH and CSH), which accompanied by a decrease in the inten-sity of peaks characteristics for anhydrous phases (C3S and b-C2S).The increase in the intensities of CH and CSH peaks is attributed tothe internal autoclaving (hydrothermal reaction) occurring for theremaining unhydrated part of cement grains by firing at 300 �C[39,41,42]; such increase in the intensity of the peaks of the hydra-tion products confirms the increase in the compressive strengthvalues noticed up to 300 �C (Fig. 9a). The XRD patterns of the spec-imens after thermal treatment at 800 �C reveals the presence of CH,as a result of recrystallization of the formed amorphous part ofportlandite and/or the rehydration of lime (CaO) from the decom-position of portlandite as well as minor amounts of CaCO3 [42–44];the intensities of the peaks characterized to anhydrous phases (b-C2S and C3S) were increased with complete disappearance of the

Fig. 10. XRD pattern of OPC cement Pastes blended with 10% Homra before andafter firing. (a) without CNTs and (b) with 0.1 CNTs (mass%).

Fig. 11. XRD pattern of OPC cement Pastes blended with 30% Homra before andafter firing. (a) without CNTs and (b) with 0.1 CNTs (mass%).

Fig. 12. DSC curves for hardened cement pastes made from mixes A0, B0 and D0after 28 days of hydration.

244 M.S. Amin et al. / Construction and Building Materials 98 (2015) 237–249

peak characteristics for CSH. The previous results are attributed totheir recrystallization and/or to the transformation of CSH gel tolarnite (b-C2S) and hartutite (C3S) [40,42,44,45]. Addition of(0.1 mass%) CNTs to OPC does not affect the hydration of OPC asshown from the XRD patterns of the specimens made of mix A3after 28 days of hydration followed by firing at 300 and 800 �C;Fig. 9b; the same hydration products are formed with nearly sim-ilar peaks intensities as A0 specimens. Such results lead to animportant conclusion that, the increase in thermal stabilityobserved for the specimens containing CNTs is believed to be dueto their physical contribution; where it can further fill in the poresbetween the hydration products giving as a result a more efficientcrack bridging at the very preliminary stage of crack propagationwithin composites and also due to their insulation character (18,38).

XRD patterns of OPC specimens blended with 10 and 30%Homra (mix B0 and D0) after 28 days hydration followed by firingat 300 and 800 �C, with gradual cooling in air are shown inFigs. 10a and 11a) respectively. The XRD patterns indicate nearlythe same hydrated phases as in case of OPC specimens (A0) after28 days hydration, with a notable increase in the intensity of thepeaks characteristic for quartz due to the presence of Homra anda slight decrease in the intensity of the peak characteristic for CHdue to its consumption in the pozzolanic reaction with Homra.Upon thermal treatment of the specimens for 3 h at 300 and800 �C, the same behavior of change of hydration products withtemperature was noticed in both B0 and D0 specimens as in OPCspecimens (A0). After firing the specimens at 800 �C small peakscharacteristic for CH were observed in case of B0 while this peakis completely diminished in case of D0 after heating. Addition of(0.1 mass%) CNTs, on OPC –Homra blends containing 10 and 30%Homra (mixes B3 and D3) give nearly the same hydrated phasesand also the same trend as in case of B0 and D0 specimensFigs. 10b and 11b.

3.4. Differential scanning calorimetry

The DSC thermograms obtained for the specimens made fromneat OPC (mix A0) and OPC–Homra blends containing 10% and30% Homra (mixes B0 and D0) after 28 days of hydration areshown in Fig. 12. The DSC curves indicate two main endothermicpeaks located at 90–105 �C and 420 �C. The first endotherm locatedat 90–105 �C is mainly due to the removal of residual free water aswell as dehydration of the sulphoaluminate hydrates and theamorphous part of calcium silicate hydrates (CSH) [46]. The enthal-pies of this endotherm are 14.6, 23.82 and 16.2 J/g for specimensmade from mixes A0, B0 and D0 respectively; the marked increasein the enthalpy value in case of specimens made of mix B0 is attrib-uted to the formation of additional amount of CSH, as a result of

Fig. 13. DSC curves for hardened cement pastes made from mixes A3, B3 and D3after 28 days of hydration.

Fig. 14. SEM micrographs of OPC cement containing 10% Homra (Bo series). (a1, a2

M.S. Amin et al. / Construction and Building Materials 98 (2015) 237–249 245

the pozzolanic reaction of Homra with free calcium hydroxide lib-erated from OPC hydration. The notable lower enthalpy valueobtained in case of specimens made of mix D0 (containing 30%Homra), as compared to that of mix B0 (containing 10% Homra)is attributed to the dilution effect of OPC (the main source ofCSH) by large amount of Homra. The second peak located at420 �C, is mainly related to the decomposition of portlandite(CH) [24]. The enthalpy of this endotherm decreases from34.18 J/g to 29.22 J/g and to 23.08 for mixes A0, B0 and D0, respec-tively. These values reveal a marked decrease in the amount ofportlandite with increasing the amount of Homra due to its con-sumption in the pozzolanic reaction with Homra. The lowerenthalpy value of this endotherm in case of D0 specimens as com-pared to that of B0 specimens is due to the substitution of OPC (themain source of liberated lime) by larger amount of Homra. Also, athird endotherm peak appeared in the range of at 670–700 �C incase of mixes A0 and B0, this endotherm is due to the

C-S-H

C-S-H

CH

C-S-H

C-S-H

CH

) before firing, (b1, b2) after firing at 300 �C and (c1, c2) after firing at 800 �C.

CNTs

CNTs

CNTs

CNTs

CH

C-S-H

C-S-H

C-S-H

C-S-H

C-S-H

CNTs

CH

Fig. 15. SEMmicrographs of OPC cement containing 10% Homra with addition of 0.1 wt.% CNTs (B3 series). (a1, a2) before firing, (b1, b2) after firing at 300 �C and (c1, c2) afterfiring at 800 �C.

246 M.S. Amin et al. / Construction and Building Materials 98 (2015) 237–249

decomposition of calcium carbonate. The enthalpies of thisendotherm are 6.04 and 5.4 J/g in case of mixes A0 and B0, respec-tively. The small enthalpy values of this endotherm reveal the pres-ence of minor amount as well as the nearly amorphous nature ofcalcium carbonate [47].

The thermograms obtained for hardened pastes made from OPC,OPC–Homra blends containing 10, 30% Homra with addition of 0.1(mass%) CNTs (mixes A3, B3 and D3), respectively are shown inFig. 13. As shown in the figure, nearly the same endothermic peaksare obtained as in case of mixes A0, B0 and D0. The values ofenthalpy change for the endothermic peak characteristic fordecomposition of CSH are 15.28, 10.72 and 17.27 J/g for mixesA3, B3 and D3, respectively. The observed decrease in the valueof enthalpy in case of mix B3 can be explained in terms of decreasein the degree of crystallinity of the formed CSH, not due to decreasein its amount, since the results of compressive strength values

indicate a marked increase as compared to that of B0 specimens.On the other hand, the values of enthalpy of this endotherm in caseof hardened pastes made from mixes A3 and D3 are nearly thesame as those of specimens made from mixes A0 and D0, whichconfirm that the presence of CNTs did not affect the hydrationreaction of OPC and its role in improving the compressive strengthand fire resistance is mainly via its physical contribution (as previ-ously mentioned in XRD section). The values of enthalpy for theendothermic peak characteristic for decomposition of CH are34.68, 31.39 and 31.35 for mixes A3, B3 and D3, respectively. Thedecrease in the value of the enthalpy from A3 to B3 and D3 isexplained in term of consumption of the free CH in the pozzolanicreaction with Homra; moreover, the notable higher enthalpy val-ues obtained in case of specimens made of mixes B3 and D3 ascompared to those obtained in case of specimens made of mixesB0 and D0 are explained in term of increase in the degree of

C-S-H

CH

C-S-H

Fig. 16. SEM micrographs of OPC cement containing 30% Homra (Do series). (a1, a2) before firing, (b1, b2) after firing at 300 �C and (c1, c2) after firing at 800 �C.

M.S. Amin et al. / Construction and Building Materials 98 (2015) 237–249 247

crystallization of formed CH in presence of CNTs; the presence ofCNTs enhance the formation of crystalline products. A thirdendothermic peak observed at 670–700 �C in case of specimensmade from mix A3, which account for the decomposition of cal-cium carbonate, the enthalpy of this endotherm is 6.64 J/g (nearlyas that of specimens made from mix A0) and this peak is nearlydiminished in case of specimens made from mixes B3 and D3 .

3.5. Morphology and microstructure

High resolution SEM provides a good information on the mor-phology and the microstructure of the formed hydrates. Fig. 14a1and a2 shows the SEMmicrographs of hardened OPC paste blendedwith 10 mass% Homra (mix B0) after 28 days of hydration. Themicrograph revealed a well-developed hydrated phase such as cal-cium hydroxide (CH) crystals intermixedwith small wrinkled fibersof calcium silicate hydrate (C–S–H) and calcium sulfoaluminate

hydrate (ettringite), also the pore spaces are still available for thedeposition of hydration products. The micrograph obtained afterfiring at 300 �C shows the formation of excess amount of nearlyamorphous and micro-crystalline C–S–H as the main hydrationproducts obtained as a results of both the internal autoclaving ofthe remaining parts of unhydrated OPC particles as well as the poz-zolanic interaction between free calcium hydroxide and Homra.The C–S–H phases are deposited within the pore system, leadingtomore a dense structurewith relatively high compressive strengthvalues, Fig. 14b1 and b2. The microstructure obtained at highertemperature (800 �C) showed the formation of several microcracksleading to a destruction of binding forces as represented by the lowcompressive strength and high porosity of OPC blended paste [25],as shown in Fig. 14c1 and c2. The micrographs of 90% OPC–10%Homra in presence of 0.1 mass% CNTs (mix B3) after 28 days hydra-tion and after thermal treatment at 300 and 800 �C are given inFig. 15a1, a2, b1, b2 and c1, c2, It can be observed that the CNTs

C-S-H + CNTs

CNTs

C-S-H

CNTs

C-S-H

CNTs

C-S-H

CH

Fig. 17. SEM micrographs of OPC cement containing 30% Homra with addition of 0.1 wt.% CNTs (D3 series). (a1, a2) before firing, (b1, b2) after firing at 300 �C and (c1, c2)after firing at 800 �C.

248 M.S. Amin et al. / Construction and Building Materials 98 (2015) 237–249

were dispersed uniformly in the cement paste and there was noobvious aggregation of CNTs. The microscopic observation revealedalso that the surface of CNTswas engulfedwith CSH, also themicro-graphs indicated nearly the same microstructure as B0 specimenswith the presence of CNTs embedded as individual fibers in thepaste and acting as bridges between hydrates and across crackswhich is the main reason for improving the thermal stability ofthese specimens at both 300 and 800 �C as compared to both neatOPC and B0 specimens Fig. 15a1, a2, b1, b2 and c1, c2. The micro-graphs of the hardened specimens made of 70% OPC–30% Homrablend (mix D0) after 28 days hydration and after thermal treatmentat 300 and 800 �C indicates the same hydration products but withless denser and more porous structure as compared to the B0 spec-imens Fig. 16a1, a2, b1, b2 and c1, c2. The addition of 0.1 mass%CNTs in these specimens (mix D3) improve their microstructure

as compared to D0 specimens; via filling the pores between thehydration products giving, as a result, a more efficient crack bridg-ing at the very preliminary stage of crack propagation within thecomposites, Fig. 17a1, a2, b1, b2 and c1, c2.

4. Conclusion

On the basis on the results obtained in this study the followingconclusions can be derived:

(1) Replacement of OPC by 10, 20 and 30% Homra enhance theresidual compressive strength of the hardened pastes madeof these blends upon firing at 300, 600 and 800 �C as com-pared to those made of the blank mix (100% OPC).

M.S. Amin et al. / Construction and Building Materials 98 (2015) 237–249 249

(2) Small additions of CNTs to OPC–Homra composite blendscontaining up to 30% Homra improve the fire resistance ofresulted OPC- Homra-CNTs blends as compared to theOPC–Homra blends without CNTs, and also 0.1% CNT couldbe considered as the optimum addition to each mix.

(3) Mix D3 (70% OPC–30% Homra–0.1% CNT) can be chosen to bea suitable mix, which shows the higher compressivestrength values as well as fire resistance as compared tothose of neat OPC.

(4) The presence of CNTs does not affect the hydration reactionof OPC or Homra/OPC composite blends, but increases theircompressive strength and fire resistance up to 300 �C

(5) From SEM micrograph it can be concluded that CNTs act asbridges between hydrates and across cracks which improv-ing the thermal stability of the composite blends up to800 �C.

(6) Flexures test program will be concluded in our future inves-tigation about the effect of CNTs on the mechanical proper-ties of composite blends.

References

[1] N. Yazadani, V. Mohanam, Carbon nano-tube and nano-fiber in cement mortar:effect of dosage rate and water–cement ratio, Int. J. Mater. Sci. 4 (2014) 45–52.

[2] K.R. Abu Al-Rub, A.I. Ashour, B.M. Tyson, On the aspect ratio effect of multi-walled carbon nanotube reinforcements on the mechanical properties ofcementitious nanocomposites, Constr. Build. Mater. 35 (2012) 647–655.

[3] N. Nochaiya, A. Chaipanich, Behavior of multi-walled carbon nanotubes on theporosity and microstructure of cement-based materials, Appl. Surf. Sci. 257(2011) 1941–1945.

[4] G.M.S. Konsta, Z.S. Metaxa, S.P. Shah, Multi-scale mechanical and fracturecharacteristics and early-age strain capacity of high performance carbonnanotube/cement nanocomposites, Cem. Concr. Compos. 32 (2010) 110–115.

[5] G.M.S. Konsta, Z.S. Metaxa, S.P. Shah, Highly dispersed carbon nanotubereinforced cement based materials, Cem. Concr. Res. 40 (2010) 1052–1059.

[6] P. Balaguru, K. Chong, Nanotechnology and concrete: research opportunities,in: Proceedings of ACI Session on Nanotechnology of Concrete: RecentDevelopments and Future Perspectives, Denver, USA, 2006, pp. 15–28.

[7] B. Han, X. Yu, E. Kwon, A self-sensing carbon nanotube/cement composite fortraffic monitoring, Nanotechnol 20 (2009) 1–5.

[8] A. Cwirzen, K. Habermehl-Cwirzen, V. Penttala, Surface decoration of carbonnanotubes and mechanical properties of cement/carbon nanotube composites,Adv. Cem. Res. 20 (2008) 65–73.

[9] S. Polizu, S. Maugey, S. Poulin, P. Poulin, L.H. Yahia, Nanoscale surface of carbonnanotube fibers for medical applications: structure and chemistry revealed byTOF–SIMS analysis, Appl. Surf. Sci. 252 (2006) 6750–6753.

[10] Q. Tan, K. Zhang, S. Gu, J. Ren, Mineralization of surfactant functionalizedmulti-walled carbon nanotubes (MWNTs) to prepare hydroxyapatite/MWNTsnanohybrid, Appl. Surf. Sci. 255 (2009) 7036–7039.

[11] X. Yu, E. Kwon, A carbon nanotube/cement composite with piezo resistiveproperties, Smart Mater. Struct. 18 (2009) 1–5.

[12] H.K. Kim, I.W. Namb, H.K. Lee, Enhanced effect of carbon nanotube onmechanical and electrical properties of cement composites by incorporation ofsilica fume, Compos. Struct. 107 (2014) 60–69.

[13] F. Sanchez, C. Ince, Microstructure and macroscopic properties of hybridcarbon nanofiber/silica fume cement composites, Compos. Sci. Technol. 69(2009) 1310–1318.

[14] C. Gay, F. Sanchez, Performance of carbon nanofiber-cement composites with ahigh-range water reducer, Transp. Res. Rec.: J. Transp. Res. Board 2142 (2010)109–113.

[15] H.K. Kim, I.W. Nam, H.K. Lee, Microstructure and mechanical/EMI shieldingcharacteristics of CNT/cement composites with various silica fume contents.In: UKC 2012 on Science Technology, and Entrepreneurship, California, USA,August 2012, pp. 8–11.

[16] S. Wansom, N.J. Kidner, L.Y. Woo, T.O. Mason, AC-impedance response of multiwalled carbon nanotube/cement composites, Cem. Concr. Compos. 28 (2006)509–519.

[17] S.P. Shah, G.M.S. Konsta, Z.S. Metaxa, P. Mondal, Nanoscale modification ofcementitious materials, Nanotechnol. Constr. 3 (2009) 125–130.

[18] A. Chaipanich, T. Nochaiya, W. Wongkeo, P. Torkittikul, Compressive strengthand microstructure of carbon nanotubes–fly ash cement composites, Mater.Sci. Eng., A 527 (2010) 1063–1067.

[19] Y.F. Zhu, L. Shi, J. Liang, D. Hui, K.T. Lau, Synthesis of zirconia nanoparticles oncarbon nanotubes and their potential for enhancing the fracture toughness ofalumina ceramics, Composites 39 (2008) 1136–1141.

[20] H. Lee, S. Mall, P. He, D. Shi, S. Narasimhadevara, Y.H. Yun, V. Shanov, M.J.Schulz, Characterization of carbon nanotube/nanofiber-reinforced polymercomposites using an instrumented indentation technique, Composites 38(2007) 58–65.

[21] A.M.K. Esawia, K. Morsi, A. Sayed, A. Gawad, P. Borah, Fabrication andproperties of dispersed carbon nanotube–aluminum composites, Mater. Sci.Eng., A 508 (2009) 167–173.

[22] W.X. Chen, J.P. Tu, L.Y. Wang, H.Y. Gan, Z.D. Xu, X.B. Zhang, Tribologicalapplication of carbon nanotubes in a metal-based composite coating andcomposites, Carbon 41 (2003) 215–222.

[23] G.Y. Li, P.M. Wang, X. Zhao, Mechanical behavior and microstructure of cementcomposites incorporating surface-treated multi-walled carbon nanotubes,Carbon 43 (2005) 1239–1245.

[24] M. Heikal, H. El-Didamony, Pozzolanic activity of Homra with lime, MansouraSci. Bull. 26 (1999) 79–95.

[25] M. Heikal, Effect of temperature on the physico-mechanical and mineralogicalproperties of Homra pozzolanic cement pastes, Cem. Concr. Res. 30 (2000)1835–1839.

[26] S.A. Abo-El-Enein, M. Daimon, S. Ohsawa, R. Kondo, Hydration of low porosityslag–lime pastes, Cem. Concr. Res. 4 (1972) 299–312.

[27] K. Kendall, A.J. Howard, J.D. Birchall, The relation between porosity,microstructure and strength, and the approach to advanced cement-basedmaterials, Philos. Trans R. Soc. Lond. A 310 (1983) 139–154.

[28] K.L. Lin, C.S. Chiou, B.Y. Chen, A. Cheng, Waste brick’s potential for use as apozzolan in blended Portland cement, Waste Manage. Res. 28 (2010) 647–652.

[29] M. O’Farrell, S. Wild, B.B. Sabir, Pore size distribution and compressive strengthof waste clay brick mortar, Cem. Concr. Compos. 23 (2001) 81–91.

[30] M. O’Farrell, B.B. Sabir, S. Wild, Strength and chemical resistance of mortarscontaining brick manufacturing clays subjected to different treatments, Cem.Concr. Compos. 28 (2006) 790–799.

[31] H. El-Didamony, M. Heikal, M.M. Shoaib, Homra pozzolanic cement, Sil. Ind. 65(2000) 39–43.

[32] Thanongsak Nochaiya, Amon Chaipanich, Behavior of multi-walled carbonnanotubes on the porosity and microstructure of cement-based materials,Appl. Surf. Sci. 257 (2010) 1941–1945.

[33] J.L. Fraga, J. María del Campo, J.A. García, Carbon nanotube–cement compositesin the construction industry, in: Second Int Conf on Emerging Trends in EngTech, May 30–31, London (UK), 2014.

[34] T. Nochaiya, A. Chaipanich, Behaviour of multi-walled carbon nanotubes onthe porosity and microstructure of cement-based materials, Appl. Surf. Sci. 257(5) (2011).

[35] T. Nochaiya, P. Tolkidtikul, P. Singiai, A. Chaipanich, Microstructure andcharacterizations of Portland-carbon nanotubes pastes, Adv. Mater. Res. 55(2008) 549–552.

[36] G.Y. Li, P.M. Wang, X. Zhao, Pressure-sensitive properties and microstructureof carbon nanotube reinforced cement composites, Cem. Concr. Compos. 2(2007) 377–382.

[37] S. Musso, J.-M. Tulliani, G. Ferro, A. Tagliaferro, Influence of carbon nanotubesstructure on the mechanical behavior of cement composites, Compos. Sci.Technol. 69 (2009) 1985–1990.

[38] M.S. Morsy, S.H. Alsayed, M. Aqel, Hybrid effect of carbon nanotube and nano-clay on physico-mechanical properties of cement mortar, Constr. Build. Mater.25 (2011) 145–149.

[39] M. Heikal, Effect of elevated temperature on the physico-mechanical andmicrostructural properties of blended pastes, Build. Res. J. 56 (2008) 156–172.

[40] H. El-Didamony, E. Abd El-Rahman, R.M. Osman, Fire resistance of fired claybricks–fly ash composite cement pastes, Ceram. Int. 38 (2012) 201–209.

[41] M.S. Morsy, S.S. Shebl, Effect of silica fume and metakaolin pozzolana on theperformance of blended cement pastes against fire, Ceram. Silik. 51 (2007) 40–44.

[42] E.E. Hekal, M.A. Abdel-Khalik, F.S. Hashem, Influence of heating andrehydration on compressive strength of some hardened blended cementpastes, L’indust Ital del Cemen J 6 (2000) 518–525.

[43] S.M.A. El-Gamal, F.S. Hashem, M.S. Amin, Thermal resistance of hardenedcement pastes containing vermiculite and expanded vermiculite, J. Therm.Anal. Calorim. 109 (2012) 217–226.

[44] M. Heikal, Effect of temperature on the structure and strength properties ofcement pastes containing fly ash alone or in combination with limestone,Ceram. Silik. 50 (2006) 167.

[45] C. Alonso, L. Fernandez, Dehydration and rehydration processes of cementpastes exposed to high temperature environments, Mater. Sci. 29 (2004) 3015.

[46] F.S. Hashem, M.S. Amin, S.M.A. El-Gamal, Improvement of acid resistance ofPortland cement pastes using rice husk ash and cement kiln dust as additives,J. Therm. Anal. Calorim. 111 (2013) 1391–1398.

[47] S.M.A. El-Gamal, M.S. Amin, M. Ramadan, Hydration characteristics andcompressive strength of hardened cement pastes containing nano-metakaolin, HBRC (2015), http://dx.doi.org/10.1016/j.hbrcj.2014.11.0082015.