Construction and Building Materials 36 (2012) 838844

Contents lists available at SciVerse ScienceDirect

Construction and Building Materials

journal homepage: www.elsevier.com/locate/conbuildmat

Properties of concrete incorporating nano-silica

A.M. Said a,, M.S. Zeidan a, M.T. Bassuoni b, Y. Tian a
a b

Department of Civil and Environmental Engineering, University of Nevada, Las Vegas, USA Department of Civil Engineering, University of Manitoba, Winnipeg, Canada

h i g h l i g h t s
" Colloidal nano-silica was studied in concrete with and without y ash. " Two dosage levels of nano-silica in concrete mixtures were evaluated. " Adding nano-silica enhanced the mechanical and micro-structural properties of concrete. " Nano-SiO2 enhanced properties through pozzolanic reaction and ller effect.

a r t i c l e

i n f o

a b s t r a c t
This study investigated the effect of colloidal nano-silica on concrete incorporating single (ordinary cement) and binary (ordinary cement + Class F y ash) binders. In addition to the mechanical properties, the experimental program included tests for adiabatic temperature, rapid chloride ion permeability, mercury intrusion porosimetry, thermogravimetry and backscattered scanning electron microscopy in order to link macro- and micro-scale trends. Signicant improvement was observed in mixtures incorporating nano-silica in terms of reactivity, strength development, renement of pore structure and densication of interfacial transition zone. This improvement can be mainly attributed to the large surface area of nanosilica particles, which has pozzolanic and ller effects on the cementitious matrix. Micro-structural and thermal analyses indicated that the contribution of pozzolanic and ller effects to the pore structure renement depended on the dosage of nano-silica. 2012 Elsevier Ltd. All rights reserved.

Article history: Received 21 July 2011 Received in revised form 13 May 2012 Accepted 4 June 2012 Available online 15 July 2012 Keywords: Nano-silica Fly ash Durability Pozzolanic effect Filler effect Reactivity

1. Introduction Nanotechnology has been dened by Drexler et al. [1] as the control of the structure of matter based on molecule-by-molecule control of products and by-products. Nanotechnology can be considered as the most modern aspect in the elds of science and technology. Because nanotechnology has great market potential and economical impact, the need for research and exploration in this eld and of its applications has been growing signicantly during the last few decades. The aim of this research is to gain better understanding of the behavior of materials on the nano-scale level as well as determine how to improve the microstructure of cementitious materials. Concrete, which is known to be the most used manmade material, is primarily based on the cement industry; in the last few years, the total annual worldwide production of concrete exceeded
Corresponding author. Address: Civil and Environmental Engineering Department, University of Nevada, Las Vegas, 4505 S. Maryland Pkwy, Las Vegas, NV 89154-4014, USA. Tel.: +1 702 895 2722; fax: +1 702 895 3936. E-mail address: aly.said@unlv.edu (A.M. Said).
0950-0618/$ - see front matter 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.conbuildmat.2012.06.044

2600 million tons [2]. The cement industry is considered to be one of the most energy consuming industries, with a high rate of carbon dioxide (CO2) emissions. Every year, it is responsible for approximately 5% of the global manmade CO2 emissions [3]; 50% of these emissions are caused by chemical manufacturing processes and 40% are due to burning fuel. Extensive research efforts have been directed to reducing the effect of the cement industry on greenhouse gases either by improving the efciency of the cement manufacturing process [4,5] or by using supplementary cementitious materials (SCMs), which partially replace ordinary cement [6]. Various SCMs have been investigated, including y ash, ground granulated blast furnace slag, natural pozzolans, and silica fume. Recent studies [7] have indicated that the use of new technologies may lead to industrial breakthroughs for the manufacture of SCMs. It is believed that nanotechnology is one of the most promising research elds that may signicantly improve the mixture design, as well as the performance and production of cement-based materials. Fly ash, a by-product of coal consumption, is considered to be one of the most used SCMs alternatives to partially replace cement in concrete because it is cost efcient and widely available.

A.M. Said et al. / Construction and Building Materials 36 (2012) 838844

839

Furthermore, y ash is much more environmentally friendly than cement [8]. Therefore, US Green Building Council (USGBC) considers using y ash in concrete as a contributing factor to earn points in the Leadership in Energy and Environmental Design (LEED) program. In addition to its economical and environmental benets, cement blended with y ash improves durability properties of concrete [9]. For example, Alasali and Malhotra [10] showed that binary binders (ordinary cement and y ash) signicantly reduce the expansion in mortar/concrete due to alkali silica reactivity (ASR). Hence, several transportation agencies include in its specication for concrete mixtures a minimum amount of y ash to prevent/ mitigate the effect of ASR [11]. The main concern about y ash concrete, particularly that containing Class F y ash, has been its slow rate of strength development compared to normal concrete [12,13]. Numerous studies have been conducted to explore different solutions for this issue, but until this stage, concrete comprising Class F y ash is not suitable for many applications where early strength is required, such as repairs and rapid construction. Experimental results [14,15] indicated that the performance of concrete is generally improved by adding nano-silica. At rst, it was believed that the improvement in concrete performance due to the addition of nano-silica is attributed to its ller effect and its pozzolanic reaction. Recently, however, it has been reported that the small particle size of nano-silica provides a larger surface area, which speeds up the rate of cement hydration and pozzolanic reactions [16]. This may be supported by the fact that other nano-materials, such as titanium oxide (TiO2), accelerate the hydration reaction of the main clinker phases, such as tricalcium silicate (C3S) [17]. Nano-silica is available in two main forms: compacted dry grains and colloidal suspension. Dry nano-silica requires a special preparation procedure before mixing in order to insure the thorough dispersion of the nano-particles in the mixing water, or other liquid admixtures, so that it can be well distributed in the concrete mixture. On the other hand, colloidal nano-silica, which is manufactured as a suspension stabilized by a dispersive agent, is a ready-to-use form of nano-silica. Previous work [18] indicated that better behavior was achieved when colloidal nano-silica was added to mortar mixtures, compared to dry grained nano-silica. This was attributed to the better dispersion of colloidal nano-silica with highly reduced agglomeration, in contrast to the dry form. In this study, the effect of incorporating colloidal nano-silica in concrete prepared with single (OPC) and binary (OPC + Class F y ash) binders is investigated. The inuence of the nano-silica dosage on the mechanical and durability properties of concrete was evaluated in the light of micro-structural and thermal analyses.
2. Experimental program 2.1. Materials The cement used was Type II/VI Portland cement, which meets ASTM C 150 (Standard Specication for Portland Cement). The main composition and properties of the cement used are presented in Table 1. Six mixtures were prepared and Class F y ash, whose composition and properties are also shown in Table 1, was used in three mixtures. The nano-silica used was in a colloidal form of an aqueous solution with 50% SiO2 content. The mean particle size of the nano-silica was 35 nm, and the specic gravity and pH were 1.36 and 9.5, respectively. In order to achieve a constant workability level, a high-range water reducing admixture (HRWRA), based on polycarboxylic acid and complying with ASTM C 494 (Standard Specication for Chemical Admixtures for Concrete) Type F, was used at different dosages for all mixtures. Well-graded natural gravel was used as coarse aggregate. The specic gravity, absorption, and dry rodded unit weight were 2.79%, 0.60% and 1634 kg/m3, respectively. The ne aggregate was natural sand with a specic gravity, absorption, and neness modulus of 2.78%, 0.80% and 3.0%, respectively. 2.2. Procedures Constituent materials were mixed in a mechanical mixer according to ASTM C 192 (Standard Practice for Making and Curing Concrete Test Specimens in the Laboratory). In this study, three mixtures included only Type II/V cement as a binder

Table 1 Properties of cement and y ash. Type II/VI Portland Cement SiO2 (%) Al2O3 (%) Fe2O3 (%) CaO (%) MgO (%) SO3 (%) Na2O (%) Loss on ignition Specic gravity Fineness (m2/kg) 20.64 3.40 3.40 63.5 4.70 2.40 0.46 1.20 3.15 376 Class F y ash 58.25 16.60 4.63 10.23 0.84 1.52 2.35 290

(single binder, Group A), while the other three mixtures had 30% of the cement replaced by Class F y ash (binary binder, Group B). The cementitious materials content (390 kg/m3) and the water-to-cementitious material ratio (0.40) were kept constant for all the mixtures. Variable dosages of nano-silica were used (0%, 3%, and 6% by total mass of cementitious materials) in each of the two groups of mixtures (A and B). The amount of mixing water was adjusted for each mixture to account for the water content of the nano-silica solution. Also, different amounts of HRWRA were used for different ratios of nano-silica in order to maintain a slump between 50 and 100 mm. Higher amounts of HRWRA were needed with the larger dosages of nano-silica. Table 2 shows the different proportions for the six mixtures. 2.3. Testing methods Immediately after mixing, 100 200 mm concrete cylinders were prepared and thermocouples were inserted at mid-height of the cylinders in order to measure the temperature of the mixtures during hardening according to ASTM C 1064 (Standard Test Method for Temperature of Freshly Mixed Portland Cement Concrete). For each mixture, two cylinders were used to measure the change in temperature and their average temperature was recorded. The molds were covered and sealed to prevent moisture loss during the test, and were kept in room temperature (23 2 C). The adiabatic temperature of concrete was recorded every 2 min using a data logger for 30 h after mixing. The compressive strength at different curing ages (3, 7, 28, 90, and 365 days) and the splitting tensile strength at 28 days were evaluated for the six mixtures using 100 200 mm cylinders according to ASTM C 39 (Standard Test Method for Compressive Strength of Cylindrical Concrete Specimens) and ASTM C 496 (Standard Test Method for Splitting Tensile Strength of Cylindrical Concrete Specimens), respectively. After 24 h of mixing, the cylinders were unmolded and placed in a standard curing room (maintained at a temperature of 23 2 C and with a relative humidity of more than 95%) until they were ready for testing. At 28 days, the rapid chloride ion permeability test (RCPT) was conducted for all the mixtures to evaluate the resistance of the mixtures to the penetrability of aggressive ions. The test procedures were conducted according to ASTM C 1202 (Standard Test Method for Electrical Indication of Concretes Ability to Resist Chloride Ion Penetration). In order to measure the physical penetration depth of chloride ions, the specimens were axially split immediately after testing. Then, the inner face of each half specimen was sprayed with a silver nitrate solution, which forms a white precipitate of silver chloride in approximately 15 min. The average depth of the white precipitation was determined by measuring the depth at ve different positions along the diameter of each half specimen. This average depth is considered to be an indication of the physical ingress of the chloride ions [19]. The mercury intrusion porosimetry (MIP) technique was used to assess pore size distribution and the total porosity of the mixtures. After 28 days, small peasized chunks taken from at least two concrete cylinders were used as test samples for MIP [20]. These chunks were around 310 mm in size, and were carefully selected so that large aggregates were not included. The samples were oven dried for 72 h at a temperature of 60 2 C; then, until the time of testing, they were kept in a desiccator containing silica gel. This method of drying using a lower temperature for a longer period was adopted to avoid the formation of micro-cracks, which may occur at high temperatures. In this study, the contact angle and the surface tension of mercury were taken as 130 and 485 dynes/cm, respectively [21,22]. The effect of nano-silica on the hydration and pozzolanic reactions was assessed by determining the quantity of portlandite (calcium hydroxide) in the matrix after 28 days. Thermogravimetry (TG) at a heating rate of 10 C/min was used for this purpose. After 28 days, chunks were obtained from crushed concrete cylinders. For the TG tests, the chunks were ground to powder, passing a #200 sieve (75 lm). The content of portlandite was calculated by determining the drop in the percentage mass of TG curves at a temperature range of 420450 C (chemically bound water in portlandite) and multiplying it by 4.11 (ratio of the molecular mass of portlandite to that of water). To complement the ndings of the MIP and TG tests, backscattered scanning electron microscopy (BSEM) was used on polished thin sections, which were prepared by cutting 25 50 mm specimens from the concrete

840 Table 2 Proportions of mixtures per cubic meter of concrete. Mixture A-0 A-1 A-2 B-0 B-1 B-2
*

A.M. Said et al. / Construction and Building Materials 36 (2012) 838844

Cement (kg) 390 390 390 273 273 273

Fly ash (kg) 117 117 117

Colloidal nano-SiO2 (kg) 23.4 46.8 23.4 46.8

Water* (kg) 156 144.3 132.6 156 144.3 132.6

HRWRA (mL/100 kg of binder) 326 457 914 261 326 653

Coarse aggregate (kg) 1184 1175 1162 1163 1154 1143

Fine aggregate (kg) 789 783 774 775 769 762

The content of mixing water was adjusted by the amount of water in the nano-silica solution.

29
A-0 (0% N.S.)

140 120

28 27

Compressive Strength (MPa)

A-1 (3% N.S.) A-2 (6% N.S.)

100 80 60
A-0 (0% N.S.)

Temperature (C)

26 25 24 23 22 21 0 5 10 15 20 25 30 35
* Note N.S.: Nano Silica

40 20 0

A-1 (3% N.S.) A-2 (6% N.S.) B-0 (0% N.S.) B-1 (3% N.S.) B-2 (6% N.S.)

50

100

150

200

250

300

350

Time (hrs)

Age (Days)
Fig. 3. Compressive strength versus curing time.

Fig. 1. Adiabatic temperature test results for Group A mixtures.

29 B-0 (0% N.S.) 28 B-1 (3% N.S.) 27

Temperature (C)

B-2 (6% N.S.) 26 25 24 23 22 21

10

15

20

25

30

35

Time (hrs)
Fig. 2. Adiabatic temperature test results for Group B mixtures.

either for Group A (without y ash) or for Group B (with 30% y ash) with the exception of mixture A-2. For example, comparing mixtures B-1 and B-2 to mixture B-0 (Fig. 2), the peak temperatures were about 20% higher and were reached in a shorter time period (4 h earlier). This increase in the peak temperature, within 15 h after mixing, indicates acceleration in the rate of hydration due the presence of nano-silica. This increase cannot be ascribed to the pozzolanic effect of nano-silica, as the pozzolanic reaction usually takes place later after adequate formation of portlandite [23]. Hence, it may be attributed to the very high surface area of nano-silica particles (average particle size of 35 nm and a specic surface of 80 m2/g), which act as nucleation sites for the hydration reactions. These results indicate that the role of nano-silica is not limited to pozzolanic or ller effects, but it also speeds up the kinetics of hydration due to its ultrane nature. 3.2. Mechanical properties The average compressive strength of the specimens at different curing ages and the average splitting tensile strength at 28 days are

cylinders after 28 days. Subsequently, they were impregnated by a low-viscosity epoxy resin under pressure and polished by using diamond surface-grinding equipment.

3. Results and discussion 3.1. Adiabatic temperature The temperature recorded over 30 h was plotted against time, as shown in Figs. 1 and 2. For all mixtures, the temperature increased just after mixing within the acceleration period and the setting period until a peak was reached. The increase in temperature was followed by a decrease during the deceleration period until a relatively constant temperature was recorded. Generally, the peak temperature was higher for mixtures containing nano-silica

Tensile Strength (MPa)

7 6 5 4 3 2 1 0 A-0 A-1 A-2 B-0 B-1 B-2

Mixture
Fig. 4. Splitting tensile strength of all the mixtures at 28 days.

A.M. Said et al. / Construction and Building Materials 36 (2012) 838844 Table 3 Early age compressive strength of concrete mixtures. Mixture Compressive strength (MPa) 3-days A-0 A-1 A-2 B-0 B-1 B-2 33 36 39 27 28 31 7-days 49 51 58 39 40 44

841

shown in Figs. 3 and 4, respectively (three replicates were used for each test). For concrete with or without y ash, the results generally indicate that the compressive and tensile strengths markedly increased with the addition of nano-silica. This was statistically supported by analysis of variance (ANOVA), at a signicance level a = 0.05. For example, ANOVA for the compressive strength results at 7 and 28 days showed that the increase in the dosage of nanosilica from 0% to 6% had F values of 32.38 and 18.54, respectively which are larger than the corresponding critical F value of 3.89. According to Montgomery [24], exceeding the critical value of an F-distribution density function reects that the tested variable signicantly affects the mean of the results. For specimens from Groups A and B, the average early age (3 and 7 days) strength increased by about 18% and 14%, respectively with nano-silica additions (Table 3). At 28 days, the increase in strength was further improved up to 36%. It was also observed that the addition of 3% and 6% nano-silica to mixtures containing 30% y ash (Mixtures B-1 and B-2, respectively) led to compressive strength that matched or exceeded the strength of the control mixture without y ash (A-0) at or before 28 days (Fig. 3). Comparatively, the mixture containing y ash without nano-silica (B-0) matched the compressive strength of the control mixture (A-0) at approximately 90 days. This suggests that the low rate of strength development for concrete incorporating Class F y ash can be controlled by the addition of nano-silica. For long-term strength, the mixtures containing nano-silica continued to gain strength with a relatively high rate after 28 days. In general, the improvement in the mechanical properties for the mixtures incorporating nano-silica can be ascribed to the pozzolanic and ller effects of nano-silica, as shown by the thermal and micro-structural analyses, which are discussed later in the text. 3.3. Rapid chloride permeability test (RCPT) In the rapid chloride ion permeability test (ASTM C 1202), charges passing through concrete disc specimens within 6 h were recorded, and the depth of chloride penetration was physically measured after testing by the colorimetric method [19]. The average passing charges and penetration depths for two replicate specimens

Fig. 5. Physical penetration (white line) of chloride ions for Specimens B-0 and B-2.

0.05 0.045

Cumulative Intrusion (ml/g)

A-0 (0% N.S.) A-1 (3% N.S.) A-2 (6% N.S.)

0.04 0.035 0.03 0.025 0.02 0.015 0.01 0.005 0 0.001 0.01 0.1 1 10 100 1000

Pore Size (m)

Fig. 6. Pore size distribution for Group A mixtures.

Table 4 Rapid chloride ion permeability test (RCPT) results. Mixture A-0 A-1 A-2 B-0 B-1 B-2 Passing charge (Coulombs) 1837 939 294 958 491 357 Penetrability class (ASTM C 1202) Low Very Very Very Very Very low low low low low Average penetration depth (mm) 10.2 3.1 4.6 8.1 4.1 3.3

are listed in Table 4, which also shows qualitative evaluation for the penetrability of the chloride ions, as specied by ASTM C 1202. Considerable improvement was noted with the addition of nano-silica. ANOVA, at a signicance level a = 0.05, for the results of passing charges and penetration depths showed that the variation of the nano-silica dosage from 0% to 6% had F values of 189.50 and 108.23, respectively, which are more than the critical F value of 3.89. This means that the tested variable signicantly affects the mean of the results. For Group A mixtures, the penetrability class changed from low to very low with the addition of nano-silica (Table 4). Fig. 5 shows an example of the physical penetration of chloride ions in two specimens from Group B, with the penetration depth indicated by the white line. In this gure, marked reduction of the penetration depth can be visually observed for the specimen with 6% nano-silica (B-2), compared to the specimen without nano-silica (B-0). In most cases, as the dosage of nano-silica increased, the passing charge and the physical penetration depth decreased. This indicates that adding small dosages of nano-silica had

842
0.065 0.06

A.M. Said et al. / Construction and Building Materials 36 (2012) 838844

B-0 (0% N.S.) B-1 (3% N.S.) B-2 (6% N.S.)

Cumulative Intrusion (ml/g)

0.055 0.05 0.045 0.04 0.035 0.03 0.025 0.02 0.015 0.01 0.005 0 0.001

0.01

0.1

10

100

1000

Pore Size (m)

Fig. 7. Pore size distribution for Group B mixtures.

Table 5 Mercury intrusion porosimetry (MIP) test results. Mixture A-0 A-1 A-2 B-0 B-1 B-2 Apparent total porosity (%) 10.13 6.91 6.44 12.56 9.30 8.21 Threshold pore diameter (lm) 0.10 0.075 0.060 0.144 0.092 0.075 Percentage of small pores (<0.1 lm) (%) 69.31 75.39 72.16 73.41 79.69 77.34

6% 5% 4%

Mass (%)

3% 2% 1% 0% A-0 A-1 A-2 B-0 B-1 B-2

Fig. 9. Image from back scattered electron microscopy (BSEM) for: (a) Specimen A-0 and (b) Specimen A-2.

Mixture
Fig. 8. Thermogravimetry (TG) results for portlandite (CH) peaks at about 450 C.

a pronounced effect on decreasing the conductivity of concrete and rening the pore structure, which discounted the physical penetration of chloride ions into the cementitious matrix. 3.4. Micro-structural and thermal analyses The trends from the mercury intrusion porosimetry (MIP) for Group A and B mixtures are shown in Figs. 6 and 7, respectively. Also, the apparent total porosities, threshold pore diameters and proportion of micro pores (less than 0.1 lm) for the mixtures are shown in Table 5. It should be noted that the MIP tests were done on at least four small chunks extracted from two replicate cylinders for each mixture, which were put in the same test compartment (porosimeter). Thus, the results shown in Figs. 6 and 7 and Table 5 can be reasonably considered the averages from representative populations. In compliance to the mechanical and durability properties, it can be noted that the total porosity markedly decreased with the

addition of nano-silica (Table 5). Furthermore, the threshold pore diameters of mixtures containing nano-silica were less than those of the control mixtures without nano-silica (A-0 and B-0). More pore structure renement was achieved by increasing the dosage of nano-silica (Figs. 6 and 7 and Table 5). For example, the decrease in the threshold pore diameters for Mixtures B-1 (3% nano-silica) and B-2 (6% nano-silica) was 36% and 48%, respectively relative to that of Mixture B-0 (0% nano-silica). Correspondingly, the increase in the percentage of micro pores for those mixtures was 8.5% and 5.4%, respectively relative to that of Mixture B-0. Again, this trend is ascribed to the pozzolanic and ller effects of nano-silica; however, the contribution of each effect to rening the pore structure could not be readily differentiated by the MIP results. The thermogravimetry (TG) results are shown in Fig. 8. Compared to the control mixtures without nano-silica, the addition of nano-silica in Mixtures A-1, A-2, B-1 and B-2 led to considerable consumption of portlandite (CH) in the cementitious matrix. This can be attributed to the effect of the pozzolanic reaction, which partially led to the pore structure renement observed in the MIP results (Table 5). However, increasing the nano-silica dosage from 3% to 6% did not seem to have a pronounced effect on the depletion of CH (Fig. 8). For example, Mixtures B-1 (3% nano-silica) and B-2 (6% nano-silica) had comparable CH contents of 2.69% and 2.66%, respectively. These mixtures had threshold diameters of 0.092 and 0.075 lm, respectively (Table 5). This may suggest that the general improvement in the mechanical and durability proper-

A.M. Said et al. / Construction and Building Materials 36 (2012) 838844

843

effects induced by nano-silica particles. This conforms to the results from the strength, durability, MIP, and TG tests. For Group B specimens containing 30% Class F y ash, a more porous structure was observed in the control Specimen (B-0) at 28 days. For example, Figs. 10ab shows the BSEM images for Specimen B-0 and B-2, respectively. For Specimen B-0, without nano-silica, the micrograph in Fig. 10a shows a coarse microstructure of the cementitious matrix incorporating 30% y ash. Considering that Class F y ash was used, the microstructure reects an insufcient degree of hydration of the matrix at this age (28 days), which links to the lower strength results shown in Fig. 3 and Table 3. Also, this observation conforms to the MIP pore size distribution curve of this mixture (Fig. 7), which had the highest total porosity and threshold pore diameter at this age (Table 5). Conversely, for Specimen B-2, which incorporated 6% nano-silica with 30% y ash, the BSEM micrographs, shown in Fig. 10b, showed higher degree of hydration, as depicted by the rened and denser microstructure and ITZ. Hence, it appears that the addition of a small dosage of well-dispersed colloidal nano-silica efciently catalyzed the kinetics of Class F y ash reaction in the cementitious matrix. This suggests that the delay in strength gain, microstructure development and durability improvement of concrete comprising Class F y ash can be mitigated by the addition of small dosages of nano-silica. 4. Conclusions With the advent of nano-size materials to the concrete industry, rigorous research data should be provided to gain better understanding on the effects of such materials on the macro- and micro-scale properties of concrete. Considering the materials, mixture designs, and testing methods implemented in the present study, the following conclusions can be drawn:  The overall performance of concrete, with or without y ash, was signicantly improved with the addition of variable dosages of nano-silica.  For mixtures incorporating nano-silica, the increase in the peak temperature recorded within 15 h after mixing indicated that the ultrane nature of nano-silica was responsible for speeding up the kinetics of hydration reactions.  At all curing ages, the strength generally increased with the addition of nano-silica up to 6%. In particular, at 28 days, the compressive strength was considerably improved for mixtures incorporating 30% Class F y ash and nano-silica, which indicates that the inherently slower rate of strength development of concrete containing Class F y ash can be controlled by the addition of small dosages of nano-silica.  The RCPT results showed that the passing charges and physical penetration depths signicantly decreased with the addition of nano-silica. This suggested that the incorporation of small dosages of nano-silica has a pronounced effect on reducing the conductivity and rening the pore structure of the cementitious matrix.  MIP results showed that the total porosity and the threshold pore diameter were signicantly lower for mixtures containing nano-silica. More renement of the pore structure was achieved with increasing the nano-silica dosage up to 6%.  TG results indicated that the addition of nano-silica led to signicant consumption of portlandite (CH) in the pozzolanic reaction. However, increasing the dosage of nano-silica from 3% to 6% did not increase the consumption of CH, which suggests that the general improvement in performance associated with the increase of the nano-silica addition from 3% to 6% may be mainly attributed to the physical ller effect in the cementitious matrix.

Fig. 10. BSEM Image for: (a) Specimen B-0 and (b) Specimen B-2.

ties and also in pore structure renement associated with the increase in the nano-silica addition from 3% to 6% may be mainly attributed to the physical ller effect of incorporating a larger volume of ultrane particles in the cementitious matrix. This statement may be substantiated by further testing in future research by combining nano-silica with other types of inert nanoparticles. Back-scattered electron microscopy (BSEM) was conducted on thin sections to complement the trends observed in the MIP and TG tests. In general, BSEM observations conformed to the results for strength, durability, MIP, and TG tests. It should be mentioned that some of the observed cracks in the micrographs are artifacts due to the sample preparation procedures. However, the BSEM analysis can still be considered as a reasonable indication for the microstructure in the light of the trends observed from the other tests. For Group A mixtures (without y ash), signicant densication was observed in the matrix at the interfacial transition zone (ITZ) for specimens containing nano-silica. For instance, Figs. 9a-b show the ITZ as indicated by the arrows for Specimens A-0 (0% nano-silica) and A-2 (6% nano-silica), respectively. It can be observed that the ITZ for Specimen A-0 (Fig. 9a) is full of voids and micro-cracks due to the wall effect between aggregates and the cementitious matrix. On the other hand, the ITZ in the section prepared from Mixture A-2 (Fig. 9b) appears to be densied and rened, owing to the chemical (pozzolanic reaction) and physical (ller action)

844

A.M. Said et al. / Construction and Building Materials 36 (2012) 838844 [11] Malvar L, Cline G, Burke D, Rollings R, Sherman T, Greene J. Alkalisilica reaction mitigation: state of the art and recommendations. ACI Mater J 2002;99(5):4809. [12] Carette G, Bilodeau A, Chevrier RL, Malhotra VM. Mechanical properties of concrete incorporating high volumes of y ash from sources in the US. ACI Mater J 1993;90(6):53544. [13] Naik TR, Singh SS, Ramme BW. Mechanical properties and durability of concrete made with blended y ash. ACI Mater J 1998;95(4): 45462. [14] Ghasemi AMR, Parhizkar T, Ramezanianpour AA. Inuence of colloidal nanoSiO2 addition as silica fume replacement material in properties of concrete. In: Proceeding of the second international conference on sustainable construction materials and technologies. Ancona, Italy; 2010. [15] Li H, Xiao H, Yuan J, Ou J. Microstructure of cement mortar with nano-particles. Composites Part B 2004;35(2):1859. [16] Belkowitz J, Armentrout DL. The investigation of nano-silica in the cement hydration process. ACI Special Publication; 2009 [SP267-08:87-100]. [17] Lee BY, Thomas JJ, Treager M, Kurtis KE. Inuence of TiO2 nanoparticles on early C3 S hydration. ACI Special Publication; 2009 [SP267-08:35-44]. [18] Campillo I, Dolado JS, Porro A. High-performance nanostructured materials for construction. In: Proceeding of the rst international symposium on nanotechnology in construction (NICOM1). Paisley, Scotland, UK; 2003. p. 21525. [19] Bassuoni MT, Nehdi M, Greenough T. Enhancing the reliability of evaluating chloride ingress in concrete using the ASTM C 1202 rapid chloride penetrability test. J ASTM Int 2006;3(3):113. [20] Kumar R, Bhattacharjee B. Study on some factors affecting the results in the use of MIP method in concrete research. Cem Concr Res 2003;33(3): 41724. [21] Shi D, Winslow DN. Contact angle and damage during mercury intrusion into cement paste. Cem Concr Res 1985;15(4):64554. [22] Auskern A, Horn W. Capillary porosity in hardened cement paste. J Test Eval 1973;1(1):749. [23] Mehta PK, Monteiro PJM. Concrete: microstructure, properties, and materials. New York: McGraw-Hill; 2006. [24] Montgomery D. Design and analysis of experiments. New York: John Wiley & Sons; 2001.

 BSEM analysis showed notable densication in the ITZ for specimens containing nano-silica. Also, for specimens containing Class F y ash and nano-silica, BSEM showed higher degree of hydration at 28 days relative the control mixture containing only Class F y ash. This suggests that the delay in microstructure development and durability improvement of concrete comprising Class F y ash can be mitigated by the addition of small dosages of nano-silica.

References
[1] Drexler KE, Peterson C, Pergamit G. Unbounding the future: the nanotechnology revolution. New York: William Morrow; 1991. [2] USGS.org [Internet]. Reston, VA: United States Geological Survey; c2010 [cited 2011 May 9]. <http://minerals.usgs.gov/minerals/pubs/commodity/cement/ mcs-2010-cemen.pdf>. [3] World business council for sustainable development WBCS. The cement sustainability initiative: our agenda for action. Geneva, Switzerland, July 2002. [4] Deja J, Uliasz-Bochenczyk A, Mokrzycki E. CO2 emissions from polish cement industry. Int J Greenhouse Gas Control 2010;4(4):5838. [5] Barker D, Turner S, Napier-Moore P, Clark M, Davison J. CO2 capture in the cement industry. Energy Procedia 2009;1(1):8794. [6] Gartner E. Industrially interesting approaches to low-CO2 contents. Cem Concr Res 2004;34(9):148998. [7] Damtoft J, Lukasik J, Herfort D, Sorrentino D, Gartner E. Sustainable development and climate change initiatives. Cem Concr Res 2008;38(2): 11527. [8] Haque M, Langan B, Ward M. High y ash concretes. ACI J Proc 1984;81(1): 5460. [9] USGBC, LEED rating system, version 3.0, US Green Building Council, Washington, DC; 2007. [10] Alasali MM, Malhotra VM. Role of concrete incorporating high volumes of y ash in controlling expansion due to alkali-aggregate reaction. ACI Mater J 1991;88(2):15963.