Use of nano-silica to reduce setting time and increase

Construction and Building Materials 29 (2012) 573–580

Contents lists available at SciVerse ScienceDirect

Construction and Building Materials journal homepage: www.elsevier.com/locate/conbuildmat

Use of nano-silica to reduce setting time and increase early strength of concretes with high volumes of fly ash or slag Min-Hong Zhang ⇑, Jahidul Islam Department of Civil and Environmental Engineering, National University of Singapore, 1 Engineering Drive 2, 117576 Singapore, Singapore

a r t i c l e

i n f o

Article history: Received 8 June 2011 Received in revised form 19 October 2011 Accepted 24 November 2011 Available online 21 December 2011 Keywords: Compressive strength Fly ash Heat development Nano-silica Setting time Silica fume Slag

a b s t r a c t This paper presents an experimental study to evaluate the effects of nano-silica (NS) on rate of cement hydration, setting time and strength development of concretes with about 50% fly ash or slag. Results indicate that length of dormant period was shortened, and rate of cement and slag hydration was accelerated with the incorporation of 1% NS in the cement pastes with high volumes of fly ash or slag. The incorporation of 2% NS by mass of cementitious materials reduced initial and final setting times by 90 and 100 min, and increased 3- and 7-day compressive strengths of high-volume fly ash concrete by 30% and 25%, respectively, in comparison to the reference concrete with 50% fly ash. Similar trends were observed in high-volume slag concrete. Nano-silica with mean particle size of 12 nm appears to be more effective in increasing the rate of cement hydration compared with silica fume with mean particle size of 150 nm. Ó 2011 Elsevier Ltd. All rights reserved.

1. Introduction Fly ash and ground granulated blast-furnace slag (GGBFS) have been used in concrete for many years as mineral admixture to replace Portland cement partially either in batching plants or in the production of blended cements. Concretes with high volumes of GGBFS have been used for applications in marine environments and situations where ground water has high sulfate contents. High volumes of fly ash have also been used in environments where durability is of concern. Concretes with high volumes of fly ash or slag can develop good strengths over time, exceeding those of similar concretes without fly ash or slag. However, early strengths of such concretes are often lower than similar concretes without fly ash or slag which may affect construction progress. Recent developments in nano-technology and availability of nano-silica (NS) have made the use of such materials in improving concrete properties possible [1–12]. Several researches show that early-age and 28-day strengths of cement pastes [3,4], mortars [5,6], and concrete [7] are increased by using a small amount of NS. Higher strengths of pastes [4], mortars [5,6] and concrete [7] with NS are also reported in comparison to those with silica fume. The higher strengths are attributed to accelerated cement hydration and pozzolanic reaction [3,4,7], reduced pores [5], and improved interface bonding between hardened cement paste and ⇑ Corresponding author. Tel.: +65 6516 2273; fax: +65 6779 1635. E-mail addresses: [email protected], [email protected] (M.-H. Zhang). 0950-0618/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.conbuildmat.2011.11.013

aggregate [3]. Nano-silica has also been used to increase early strength of concrete with fly ash [8,9]. This paper presents an experimental study on the use of NS to improve setting time and early strengths of mortars and concretes with high-volumes of fly ash or slag. Nano-silica with a specific surface of 200.1 m2/g was used in the study in comparison to silica fume with a specific surface area of 21.3 m2/g. As NS particles are very fine and they tend to agglomerate due to high surface interaction, effect of dispersion methods of NS was also investigated. In addition to the compressive strength and setting time, rate of heat development of cement pastes in the first 30 h, and resistance to chloride-ion penetration of concretes at 28 days were studied and compared. The study provides information for systematic analyses on (1) effect of NS; (2) effect of particle sizes of reactive silica (NS vs. silica fume); and (3) effect of dispersion methods on the properties of fly ash and slag concretes. 2. Experimental details In this study specimens were prepared to determine effect of NS on compressive strengths of mortars (from 1 to 91 days) with about 50% ASTM C 618 class F fly ash [13] or GGBFS [14]. Silica fume was also included in the mixtures to compare with NS. A superplasticizer was used to achieve target flow from 104% to 110% of the mortars. Cement pastes with the same water-to-cementitious ratio (w/cm) and mix proportion as the mortars (except for the sand) were prepared to determine the rate of heat development and cement hydration in the first 30 h. Concrete mixtures were prepared to determine the effect of NS on setting times, compressive strength development from 3 to 91 days and resistance to chloride-ion penetration at 28 days in comparison to the reference concrete with 50% fly ash or slag and the concrete with silica fume.

574

M.-H. Zhang, J. Islam / Construction and Building Materials 29 (2012) 573–580

Table 1 Physical properties and chemical compositions of materials.

a

Portland cement

Fly ash

Slag

Silica fume

Nano-silica

Chemical composition (%)

CaO SiO2 Al2O3 Fe2O3 MgO Na2O K2O SO3 LOI

63.4 20.1 4.1 3.3 3.6 0.2 0.4 2.1 2.4

3.9 46.3 28.5 18.5 1.8 0.2 0.6 0.2 2.3

41.8 33.1 13.7 0.7 4.9 0.2 0.5 0.7 0.6

0.2 95.9 0.3 0.3 0.4 0.05 0.6 0.2 1.5

– >99.8a – – – – – – –

Mineral phases (%)

C3S C2S C3A C4AF

66.8 7.3 5.3 10.1

– – – –

– – – –

– – – –

– – – –

Physical properties (%)

Blain fineness (m2/kg) BET surface area (m2/g) Average primary particle size Specific gravity

308 – 28.2 lm 3.15

– – 27.3 lm 2.47

– – 16.7 lm 2.94

– 21.3 150 nma 2.2a

– 200.1 12 nma 2.2a

Information provided by supplier.

2.1. Materials Normal Portland cement, ASTM C 618 Class F fly ash, and ground granulated blast-furnace slag were used for cement pastes, mortars and concretes. Nano-silica1 with specific surface area of 200.1 m2/g and average particle sizes of 12 nm were used in this study in comparison to silica fume which had specific surface area of 21.3 m2/g and average particle sizes of about 150 nm. Silica fume used was undensified. Characteristics of these materials are given in Table 1. A polycarboxylate based superplastisizer2 was used in mortar, paste, and concrete mixtures for workability purposes. Natural sand with a fineness modulus of 2.97 was used for mortar and concrete mixtures. Crushed granite coarse aggregate with a maximum aggregate size of 20 mm was used for concrete mixtures. Both the coarse and fine aggregates met the requirements of ASTM C 33 [15]. Tap water was used for mortar and concrete mixing, whereas deionized water was used for cement pastes.

2.2. Mortar and concrete mixtures 2.2.1. Mix proportions Eight mortar mixtures were included in the study (Tables 2 and 3). All the mortars had a w/cm of 0.45 and a sand-to-cementitious materials ratio of 2.75. Mortars with 1% NS were compared to that with the same amount of silica fume to evaluate the effect of specific surface area and particle size of silica (Table 2). As the NS particles were extremely fine, influences of mixing and dispersing methods on properties of the mortars were also evaluated (Table 3). In addition to the mortars, effect of 2% NS on concrete properties was studied in comparison to that of silica fume. Mix proportions of the concretes are given in Table 4.

2.2.2. Specimen preparation, curing, and testing 2.2.2.1. Mortars. Mortars were mixed in a Hobart mixer at an ambient temperature of about 30 °C. For mortar mixtures with mechanical mixing, solid materials were dry mixed first. Water was added and mixed for 1 min followed by addition of superplasticizer and mixed for 1 more min at low speed and 30 s at high speed. The flow value was determined according to ASTM C 1437 [16]. For ultrasonic mixing, nano-silica and water were mixed first using ultrasonic mixer with 90 W power input for 5 min. The sonicated mixture was then mixed in a Hobart mixer with sand, cement and fly ash or slag for 1 min. After that, superplasticizer was added and mixed for 1 min at low speed and 30 s at high speed. For each mortar mixture, fifteen 50  50  50-mm specimens were cast for compressive strength test. The molded specimens were covered with wet burlap for the first 24 h to prevent moisture loss. After demold, the specimens were cured in a fog room at a temperature of about 28–30 °C until the time of testing. Compressive strengths of the mortars were determined at 1, 3, 7, 28, and 91 days according to ASTM C 109/C 109M [17].

2.2.2.2. Concretes. Concretes were mixed using a pan mixer at an ambient temperature of about 30 °C. Before concrete mixing, NS or silica fume were mixed with water using the ultrasonic mixer for dispersion of the particles. The sonicated mix1 2

AEROSIL 200, Evonik Industries. ADVA 181N, W.R. Grace (Singapore) Pte. Ltd.

ture was then mixed with aggregate, cement, and fly ash or slag in a pan mixer for 1 min. The superplasticizer was added last and mixed for another 1–2 min to achieve a target slump of about 100 mm. Twelve 100-mm cubes and one 100  200-mm cylinders were cast for each concrete mixture for determining compressive strength and chloride-ion penetrability, respectively. Mortar sieved from the concrete was used to cast a 150-mm cube to determine setting time of concrete. Compressive strengths of concretes were determined at 3, 7, 28, and 91 days according to BS EN 12390-3 [18], setting time of concretes were determined according to ASTM C 403/C 403M [19], and resistances to chloride-ion penetration were determined at 28 days according to ASTM C 1202 [20]. Curing of the concretes was the same as that of the mortar specimens. 2.3. Cement paste mixtures 2.3.1. Mix proportions and mixing Mix proportions of cement pastes were similar to those of the mortars except for the exclusion of sand. For ultrasonic mixing, NS and water were premixed in ultrasonic machine for dispersion of fine particles. Before paste mixing, all the materials including sonicated mixtures were pre-conditioned at 30 °C for 24 h. The mixing procedures for cement pastes were similar to those of the mortars. 2.3.2. Evaluate the rate of heat generation and cement hydration Effect of NS on the rate of heat generation in cement pastes was evaluated according to ASTM C 1679 [21] by a Thermometric TAM Air 3115 isothermal calorimeter at a temperature of 30 °C. This temperature was selected to simulate weather conditions in tropical countries. The heat generation in the cement pastes reflects rate of cement hydration. The calorimeter was conditioned at 30 °C for a day before experiments, and amplifier range was set at 600 mW. After the mixing, paste sample of about 10 g was transferred into a sample ampoule with the sample mass recorded. After capping the ampoule, the sample and reference ampoules were inserted into the calorimeter. The calorimeter started to record heat at about 15 min after the cementitious materials were in contact with water, thus the heat generated initially during mixing and preparation was not captured. The heat generated from the cement hydration was monitored continuously for 30 h. The power output (in mili-watt) from the calorimeter due to the heat generated was recorded every minute. The power output was normalized based on sample mass. The normalized power output was then converted to heat generated in the sample (in joules/gram).

3. Results and discussion 3.1. Rate of heat development 3.1.1. Effect of nano-silica Figs. 1 and 2 show effect of NS or silica fume on rate of heat development of the high-volume fly ash and slag cement pastes, respectively, in comparison to corresponding reference pastes with 50% fly ash or slag. With the incorporation of 1% NS or silica fume, length of dormant period was reduced, and Peak 2 and Peak 3 of the fly ash and slag cement pastes shifted to left. The Peak 3 of

575

M.-H. Zhang, J. Islam / Construction and Building Materials 29 (2012) 573–580 Table 2 Mix proportions to compare the effect of nano-silica in fly ash and slag mortars in comparison to that of silica fume (w/cm = 0.45). Mix ID

FA0 FA11 FA1SF SL0 SL11 SL1SF a b

Mixing procedure

Mechanical a

Mechanical Mechanical a

Mechanical

Mix proportion

Flow (%) b

Water

Cement

Fly ash

Slag

Nano-silica

Silica fume

Sand

Super-plasticizer

0.45 0.45 0.45

0.5 0.5 0.5

0.5 0.49 0.49

0 0 0

0 0.01 0

0 0 0.01

2.75 2.75 2.75

0.21 0.71 0.28

107 108 106

0.45 0.45 0.45

0.5 0.5 0.5

0 0 0

0.5 0.49 0.49

0 0.01 0

0 0 0.01

2.75 2.75 2.75

0.42 0.88 0.42

110 110 107

With ultrasonic premixing of nano-silica + water. % of (cement + fly ash or slag + nano-silica or silica fume).

Table 3 Mix proportions of mortars for evaluating mixing and dispersion methods of nano-silica (w/cm = 0.45). Mix ID

a b

Mixing and dispersing procedure

FA11 FA11 (M)

a

SL11 SL11 (M)

a

Mix proportion

Mechanical Mechanical

Flow (%) b

Water

Cement

Fly ash

Slag

Nano-silica

Sand

Super-plasticizer

0.45 0.45

0.5 0.5

0.49 0.49

0 0

0.01 0.01

2.75 2.75

0.71 0.57

108 108

0.45 0.45

0.5 0.5

0 0

0.49 0.49

0.01 0.01

2.75 2.75

0.88 1.17

110 104

With ultrasonic premixing of nano-silica + water. % of (cement + fly ash or slag + nano-silica) by mass.

Table 4 Mix proportions to compare the effect of nano-silica in fly ash and slag concrete in comparison to that of silica fume (w/cm = 0.45). Mix ID

CFA0 CFA21 CFA2SF CSL0 CSL21 CSL2SF a

Mixing procedure

Mechanical a a

Mechanical a a

Mix proportion (kg/m3) Cement

Fly ash

200 200 200

200 192 192

200 200 200

– – –

Slump (mm) Slag

Nano-silica

Silica fume

CA

Sand

Super-plasticizer

– – –

0 8 0

0 0 8

1014 1014 1014

740 739 739

0.5 3.4 0.6

110 100 105

200 192 192

0 8 0

0 0 8

1014 1014 1014

774 771 771

1.4 4.0 1.7

95 85 85

With ultrasonic premixing of nano-silica or silica fume with water.

Fig. 1. Effect of nano-silica or silica fume on the rate of heat development in fly ash cement paste (nano-silica and silica fume contents are 1% of cementitious materials by mass).

the fly ash cement paste with the NS also increased in magnitude. The incorporation of the NS, however, shortened the dormant period and increased the rate of heat development and cement

hydration of the pastes more significantly than the silica fume. Silica fume did not affect the Peak 3 of the fly ash cement paste significantly. At 24 h total cumulative heat was increased in the

576

M.-H. Zhang, J. Islam / Construction and Building Materials 29 (2012) 573–580

Fig. 2. Effect of nano-silica or silica fume on the rate of heat development in slag cement paste (nano-silica and silica fume contents are 1% of cementitious materials by mass).

pastes with NS and silica fume. However, total cumulative heat was higher in the pastes with the NS than those with silica fume. It was observed that calorimeter curves for the fly ash and slag cement pastes were different. In the fly ash pastes, Peak 2 occurred at about 10 h, whereas Peak 3 occurred before 15 h (Fig. 1). As Class F fly ash generally has limited reactivity at early stage, Peaks 2 and 3 of the fly ash cement pastes are likely corresponded to C3S hydration and C3A reaction, respectively, similar to hydration features observed in typical Portland cement paste [22]. For the slag cement pastes, Peak 2 also occurred at about 10 h but Peak 3 occurred at about 18–20 h (Fig. 2). The Peaks 2 and 3 are likely corresponded to Portland cement hydration and slag hydration, respectively, according to ACI Committee 233 Report [23]. Shift of the Peak 2 to left and shortening of the dormant period in the calorimeter curves of the fly ash pastes (Fig. 1) might be related to accelerated hydration of C3S due to increased nucleation sites provided by the fine NS. In a study on the effect of NS on the hydration of calcium silicate C3S, Stein and Stevels [24] found out that the NS accelerates the hydration rate of C3S. The acceleration is explained by increased conversion rate from a protective hydrate layer into a less protective one due to the incorporation of NS which reduced calcium and hydroxyl concentration during the first minute of the hydration in the paste. They envisage more ettringite precipitation far from C3A surface as a result of the lower calcium concentration in paste with NS at early stage. This probably enabled accelerated reaction of the C3A and earlier depletion of SO3 [25] in the system and shifted the Peak 3 to left in the paste with NS as shown in Fig. 1. For the slag pastes (Fig. 2), shift of the Peaks 2 and 3 to the left in the calorimeter curves corresponded to accelerated hydration of Portland cement and accelerated slag reaction, respectively. The accelerated cement hydration might be due to increased nucleation sites provided by the fine NS particles. In addition to the effect of fine particle size of the NS and silica fume, the accelerated slag hydration (Peak 3) might be attributed to increased concentration of calcium hydroxide in solution due to increased C3S reaction. However, the changes of the Peaks 2 and 3, and dormant period were not significant for the slag pastes with NS due to effect of retardation when higher dosage of superplasticizer was used to achieve a given workability (Table 2). It was observed that at 24 h the increase in the cumulative heat of the fly ash paste with NS was more significant (20%) than that of the slag paste (6.5%) in comparison to the corresponding reference

paste. These results are consistent with 1-day strength increases (will be discussed later) of fly ash and slag mortars shown in Table 6. 3.1.2. Effect of mixing and dispersion method Figs. 3 and 4 show effect of mixing and dispersion method of NS on the rate of heat development of fly ash and slag cement pastes, respectively, and the corresponding cumulative heat development is shown in the inset. For both fly ash and slag pastes with NS premixed with water using an ultrasonic equipment, the dormant period was reduced and both Peaks 2 and 3 shifted left and occurred before those of the corresponding reference fly ash or slag paste. For the fly ash paste prepared by mechanical mixing (Fig. 3), Peaks 2 and 3 are also shifted to left. Comparing the fly ash pastes prepared by the two different methods, the extent of the shift and increase in the magnitude of the Peaks 2 and 3 were more significant by using ultrasonicated NS than mechanical mixing method. For the slag paste prepared by mechanical mixing, however, the length of dormant period was increased, and Peaks 2 and 3 were shifted to right than those of the reference slag paste (Fig. 4). The phenomena observed for the paste by mechanical mixing might be related to less homogeneous dispersion of NS in both slag and fly ash pastes as well as retarding effect due to higher dosage of superplasticizer used in the slag cement paste (Table 3) to achieve a given workability. The NS particles are extremely small and likely agglomerate due to high surface interaction. The agglomerated NS will reduce nucleation sites for precipitation of hydration products, thus reduce the rate of reactions. Cumulative heat generated within the first 24 h in the fly ash and slag cement pastes with ultrasonicated NS was higher than that with NS mechanically mixed with other ingredients which suggests that the cement and slag hydration was increased by the sonication of NS and water. 3.2. Setting time of concretes Effects of 2% NS on the initial and final setting times of high-volume fly ash and slag concretes in comparison to corresponding reference concretes and concretes with 2% silica fume are shown in Table 5. Significant reductions of the initial and final setting times were observed for both the fly ash and slag concretes with 2% NS. The incorporation of 2% NS reduced the initial and final setting times of the fly ash concrete by 90 min and 100 min, respectively. In the slag concrete, incorporation of the same amount NS reduced

M.-H. Zhang, J. Islam / Construction and Building Materials 29 (2012) 573–580

577

Fig. 3. Effect of mixing and dispersing method on the rate of heat development in fly ash cement paste with 1% nano-silica.

Fig. 4. Effect of mixing and dispersing method on the rate of heat development in slag cement paste with 1% nano-silica.

the initial and final setting times by 95 min and 105 min, respectively. However, the incorporation of 2% silica fume did not affect the setting times of the fly ash and slag concretes significantly. 3.3. Compressive strength development 3.3.1. Effect of nano-silica on strength development of high-volume fly ash or slag mortars The effects of NS on compressive strength developments of the high-volume fly ash and slag mortars compared to those with silica fume are shown in Table 6. The compressive strengths of the fly ash and slag mortars were generally increased with the incorporation of 1% NS or silica fume in comparison to the corresponding reference mortars with 50% fly ash or slag. The NS showed more significant effect on the strength development than the silica fume, especially at early ages. For example, the strengths of the fly ash mortars with 1% NS were increased by 61% and 25% at 1 and 3 days, respectively, whereas those with 1% silica fume were increased by 13% and 15% compared to those of the reference fly ash mortars. The results are consistent with findings by Qing et al. [3]. It is noted that the incorporation of the NS had less significant effect on the 1-day strength of the slag mortars than at 3 and 7 days. This might be related to the higher dosages of the

superplasticizer used which had retarding effect in the mortars with NS to achieve a given workability. In terms of percentage of strength increase, the NS and silica fume seems to have similar effect on the fly ash and slag mortars at 3 and 7 days, but less effect on fly ash mortars at 28 and 91 days compared with the slag mortars. At the age of 1 day, 1% NS increased the strength of the fly ash mortar by 61% from 7.4 to 12.0 MPa, whereas 1% NS increased the strength of slag mortar by only 18%. The difference may be attributed to the characteristic difference between fly ash and slag. When mixed with water, slag can hydrate under the activation of cement and lime, whereas Class F fly ash is pozzolanic material and has limited reactivity at early age. Thus, early strength of slag mortar is higher than that of fly ash mortar with given volume of the slag or fly ash. Due to accelerated heat development shown in the calorimeter curve and possibility of acceleration of C3S and C3A reactions, the NS appears to have more effect on 1-day strength of the fly ash mortar than on the slag mortar. 3.3.2. Effect of mixing and dispersion method on strength development of mortars The effect of mixing and dispersion methods of NS on compressive strength developments of the fly ash and slag mortars are

578

M.-H. Zhang, J. Islam / Construction and Building Materials 29 (2012) 573–580

Table 5 Effect of nano-silica on setting time, compressive strength, and charge passed in the fly ash and slag concretes in comparison to that of silica fume (w/cm = 0.45).

a

Mix ID

Mixing and dispersing procedure

Binder type

CFA0 CFA21 CFA2SF CSL0 CSL21 CSL2SF

Mechanical

50% Fly ash, 50% cement 2% NS, 48% fly ash, 50% cement 2% Silica fume, 48% fly ash, 50% cement 50% Slag, 50% cement 2% NS, 48% slag, 50% cement 2% Silica fume, 48% slag, 50% cement

a a

Mechanical a a

Setting time (h:min)

Compressive strength (MPa)

Charge passed (C)

Initial

Final

3-day

7-day

28-day

91-day

6:05 4:35 5:45 6:05 4:30 5:50

8:15 6:35 8:05 8:10 6:25 8:05

19.5 25.3 20.4 35.4 43.1 36.7

28.1 35.0 29.9 48.1 56.8 50.5

43.0 48.7 43.5 68.0 69.8 68.4

50.2 59.3 51.9 69.3 71.1 69.7

1154 858 891 1049 786 946

With ultrasonic premixing of nano-silica or silica fume with water.

Table 6 Effect of nano-silica on the compressive strength of fly ash and slag mortars in comparison to that of silica fume (w/cm = 0.45). Mix ID

Binder type

Compressive strength (MPa) 1-day

3-day

7-day

28-day

91-day

FA0 FA11 FA1SF

50% Fly ash, 50% cement 1% NS, 49% fly ash, 50% cement 1% Silica fume, 49% slag, 50% cement

7.4 12.0 8.4

18.1 22.5 20.8

26.9 31.5 29.7

40.8 43.6 42.3

48.3 50.4 48.9

SL0 SL11 SL1SF

50% Slag, 50% cement 1% NS, 49% slag, 50% cement 1% Silica fume, 49% slag, 50% cement

13.8 16.3 15.2

27.3 35.8 32.1

40.4 47.0 46.1

53.8 60.2 59.8

55.9 64.3 63.0

Table 7 Effect of mixing and dispersion methods of nano-silica on the compressive strength of mortars (w/cm = 0.45). Mix ID

Mixing and dispersing procedure

Binder type

1-day

3-day

7-day

28-day

91-day

FA0 FA11 FA11 (M)

Mechanical

50% Fly ash, 50% cement 1% NS, 49% fly ash, 50% cement

7.4 12.0 9.8

18.1 22.5 20.3

26.9 31.5 30.0

40.8 43.6 41.1

48.3 50.4 49.3

50% Slag, 50% cement 1% NS, 49% slag, 50% cement

13.8 16.3 15.9

27.3 35.8 35.7

40.4 47.0 47.3

53.8 60.2 59.3

55.9 64.3 63.5

SL0 SL11 SL11 (M) a

a

Compressive strength (MPa)

Mechanical Mechanical a

Mechanical

With ultrasonic premixing of nano-silica + water.

shown in Table 7. The compressive strengths were higher for fly ash mortars prepared by the sonication of NS than those of mechanical mixing method. However, no significant difference was observed in strength development of the slag mortars prepared by ultrasonicated NS with water and mechanical mixing method. This difference might be attributed to the lower density of the slag mortar SL11 than that of the mortar SL11 (M). Careful examination of the data shows that the density of the mortar prepared by ultra-sonication of nano-silica (2231 kg/m3) was lower than that prepared by mechanical mixing (2275 kg/m3). The difference on the density of the mortars might have affected the strength. The lower density of the mortar prepared by the ultrasonication of nano-silica (SL11) might have resulted in lower compressive strength. This suggests that if the density of the two mixtures SL11 and SL11 (M) was the same, the former might have higher strength than the latter, consistent with the heat and rate of cement and slag hydration as discussed in Section 3.1.2. The lower density of the mortar SL11 might be due to inadequate consolidation. The results indicate that ultrasonication of NS with water is probably a better method to disperse the NS particles properly in comparison to the mechanical mixing. However, further research is needed to confirm this. 3.3.3. Strength development of high-volume slag or fly ash concrete with nano-silica Compressive strength development of fly ash and slag concretes with 2% NS or silica fume in comparison to that of the reference fly

ash and slag concretes is presented in Table 5. The compressive strengths of the fly ash and slag concretes were increased with the incorporation of the NS in comparison to the corresponding reference concretes, especially at early ages. However, silica fume did not show significant effect on the strengths of the fly ash and slag concretes. For example, with the incorporation of NS, the compressive strengths of the fly ash concrete were increased by 30% and 25% at 3 and 7 days, respectively, whereas those with silica fume were increased by only 5% and 6%, respectively, compared to the reference fly ash concrete. Similar trends of strength increase due to the NS and silica fume were observed for the slag concrete at early ages as well. Comparing the fly ash and slag concretes, the NS and silica fume did not show significant effect on the slag concrete at 28 and 91 days, whereas the NS increased fly ash concrete by 13% and 18% at 28 and 91 days, respectively, in comparison with the reference fly ash concrete. This might be attributed to the coarse aggregate used which appears to have reached its strength limit at about 70 MPa in the slag concretes. In ordinary concrete, interface transition zone (ITZ) between the coarse aggregate and mortar matrix is usually the weakest link. As a result, cracks generally go through the ITZ around coarse aggregate particles. With reduction of w/cm and use of fine mineral admixture such as NS or silica fume, the ITZ and mortar matrix can be improved substantially. Thus, the strength of the concrete can be improved. However, if the strengths of the ITZ and mortar

M.-H. Zhang, J. Islam / Construction and Building Materials 29 (2012) 573–580

579

In addition to nucleation effect, NS may have acted as reactive filler which reduces bleeding and increases packing density of solid materials by occupying space between cement, fly ash, and slag particles. These physical effects of the NS may have contributed to the reduced setting time and increase early strength of the fly ash and slag concretes observed. However, as both the NS and silica fume are finer than Portland cement, fly ash, and slag, the filler effect of the NS and silica fume may be similar.

Fig. 5. Crack pattern in slag concrete with 2% nano-silica under compressive load. Crack goes through coarse aggregate particles.

matrix are improved to such extents that coarse aggregate becomes the weakest link in concrete, crack will go through the coarse aggregate particles. In such conditions, reducing w/cm and using fine mineral mixtures will no longer increase the concrete strength because coarse aggregate is a limiting factor for the strength of concrete. Fig. 5 shows that cracks went through coarse aggregate particles in the slag concrete with 2% NS after a compression test. This might explain why the strength of the slag concrete with 2% NS was not increased further at 28 and 91 days. 3.4. Effect of nano-silica on resistance to chloride-ion penetration Information on the chloride-ion penetrability of the high-volume fly ash and slag concretes with 2% NS or silica fume in comparison to that of the reference concretes is presented in Table 5. The results show that the total charge passed through the fly ash and slag concretes with the NS or silica fume was lower than that of the corresponding reference concretes. However, differences on the charge passed through the concretes with NS or silica fume were not significant. The concretes with NS or silica fume had charges passed below 1000 C, which is considered ‘‘very low’’ according to ASTM C 1202 [20]. 3.5. Discussion Both the NS and silica fume are nano-sized highly reactive silica, but the average primary particle size of the former is about 10 times smaller than that of the latter. The mechanisms by which silica fume modifies cement paste, mortar, and concrete are summarized in ACI Committee 234 Report [26]. These mechanisms are also applicable to NS. As the particle sizes of the NS are much smaller than those of the silica fume, the physical and chemical effect of the former is likely more substantial than the latter. Their effects on setting time, early strength development and resistance to chloride penetration of the high-volume fly ash and slag concretes will be discussed below. 3.5.1. Physical effect From physical perspective, extremely fine particle size of the NS may have accelerated cement and slag hydration by providing high amount of nucleation sites for precipitation of cement hydration products in the high-volume fly ash and slag concretes. In the slag concrete, the accelerated cement hydration also results in increased amount of calcium hydroxide in solution which activates and speeds up slag hydration.

3.5.2. Chemical effect From chemical point of view, NS is highly reactive pozzolanic material and reacts with calcium hydroxide (CH) from cement hydration to form calcium silicate hydrates (CSH). Zhang and Gjørv [27] reported pozzolanic reaction of silica fume as early as 1-day of cement hydration. Since the NS had specific surface area about 10 times higher than that of silica fume, the pozzolanic reaction of the NS might have started before 24 h. The pozzolanic reaction of the NS at very early age might have also contributed to the reduced setting times and increase early strength of the fly ash and slag concretes observed. 3.5.3. Microstructure modification In terms of microstructure modifications, the NS might reduce porosity in cement paste and in ITZ between the cement paste and aggregate due to the physical and chemical effects discussed above. It is also reported that NS can reduce CH crystal size in the ITZ more effectively than silica fume [28]. The above effects increased the density of the cement paste and improved bonding between the cement paste and aggregate which might have contributed to the strength development and reduced chloride penetrability of the concretes. 4. Conclusions Based on the experimental results using nano-silica in pastes, mortars, and concretes with about 50% of fly ash or slag at w/cm of 0.45, following conclusions can be drawn: 1. Length of dormant period was shortened, and rate of cement and slag hydration was accelerated with the incorporation of 1% NS in the cement pastes with high volumes of fly ash or slag. 2. The incorporation of 2% NS by mass of cementitious materials reduced initial and final setting times by 90 and 100 min, and increased 3- and 7-day compressive strengths of high-volume fly ash concrete by 30% and 25%, respectively, in comparison to the reference concrete with 50% fly ash. Similar trends were observed in high-volume slag concrete. 3. At 28 and 91 days, the NS increased strength of fly ash concrete compared to the reference fly ash concrete. However, the NS did not increase the strengths of the slag concrete at these ages, which might be related to the coarse aggregate used which appears to have reached its strength limit. 4. Nano-silica with mean particle size of 12 nm appears to be more effective in increasing the rate of cement hydration compared with silica fume with mean particle size of 150 nm. The NS reduced the setting times and increased early strengths of the high-volume fly ash or slag concrete. However, the setting times and early strength of these concretes were not affected by the silica fume significantly. 5. The 28-day charges passed through the fly ash or slag concretes with 2% NS or silica fume were similar, and were lower than those of the corresponding reference concretes.

580

M.-H. Zhang, J. Islam / Construction and Building Materials 29 (2012) 573–580

Acknowledgments Grateful acknowledgement is made to National University of Singapore for providing partial scholarship to the second author. Appreciation is made to past undergraduate student Huynh Thanh Binh, Structure Laboratory Technologist Ang Beng Oon, and Research Fellow Li Wei who contributed to some of the experimental work. References [1] Sobolev K, Gutierrez MF. How nanotechnology can change the concrete world. Am Ceram Soc Bull 2005;84(10):14–7. [2] Sanchez F, Sobolev K. Nanotechnology in concrete – a review. Constr Build Mater 2010;24(11):2060–71. [3] Qing Y, Zenan Z, Deyu K, Rongshen C. Influence of nano-SiO2 addition on properties of hardened cement paste as compared with silica fume. Constr Build Mater 2007;21(3):539–45. [4] Porro A, Dolado JS, Campillo I, Erkizia E, Miguel Yd, Ibarra YSd. Effects of nanosilica additions on cement pastes. In: Dhir RK, Newlands MD, Csetenyi LJ, editors. Applications of nanotechnology in concrete design. London: Thomas; 2005. p. 87–96. [5] Li H, Xiao H-G, Yuan J, Ou J. Microstructure of cement mortar with nanoparticles. Composites Part B 2004;35(2):185–9. [6] Jo B-W, Kim C-H, Tae G-h, Park J-B. Characteristics of cement mortar with nano-SiO2 particles. Constr Build Mater 2007;21(6):1351–5. [7] Schoepfer J, Maji A. An investigation into the effect of silicon dioxide particle size on the strength of concrete. ACI Spec Publ 2009;267(5):45–58. [8] Li G. Properties of high-volume fly ash concrete incorporating nano-SiO2. Cem Concr Res 2004;34(6):1043–9. [9] Said AM, Zeidan MS. Enhancing the reactivity of normal and fly ash concrete using colloidal nano-silica. ACI Spec Publ 2009;267(7):75–86. [10] Collepardi M, Ogoumah-Olagot JJ, Skarp U, Troli R. Influence of amorphous colloidal silica on the properties of self-compacting concretes. In: Proceedings of the international conference, challenges in concrete construction – innovations and developments in concrete materials and construction, Dundee, UK; 2002. p. 473–83. [11] Björnström J, Martinelli A, Matic A, Börjesson L, Panas I. Accelerating effects of colloidal nano-silica for beneficial calcium–silicate–hydrate formation in cement. Chem Phys Lett 2004;392(1–3):242–8.

[12] Flores I, Sobolev K, Torres-Martinez L, Cuellar E, Valdez P, Zarazua E. Performance of cement systems with nano-SiO2 particles produced by using the sol–gel method. Transport Res Rec: J Transport Res Board 2010;2141:10–4. [13] ASTM C 618. Standard specification for coal fly ash and raw or dalcined natural pozzolan for use in concrete. Annual Book of ASTM Standards, vol. 04.02; 2008. [14] ASTM C 989. Standard specification for slag cement for use in concrete and mortars. Annual Book of ASTM Standards, vol. 04.02; 2010. [15] ASTM C 33. Standard specification for concrete aggregates. Annual Book of ASTM Standards, vol. 04.02; 2008. [16] ASTM C 1437. Standard test method for flow of hydraulic cement mortar. Annual Book of ASTM Standards, vol. 04.01; 2005. [17] ASTM C 109/C 109M. Standard test method for compressive strength of hydraulic cement mortars (using 2-in or [50-mm] cube specimens). Annual Book of ASTM Standards, vol. 04.01; 2005. [18] BS EN 12390-3. Testing hardened concrete – Part 3: compressive strength of test specimens. British Standards Institution; 2002. [19] ASTM C 403/C 403M. Standard test method for time of setting of concrete mixtures by penetration resistance. Annual Book of ASTM Standards, vol. 04.02; 2008. [20] ASTM C 1202. Standard test method for electrical indication of concrete’s ability to resist chloride ion penetration. Annual Book of ASTM Standards, vol. 04.02; 2008. [21] ASTM C 1679. Practice for measuring hydration kinetics of hydraulic cementitious mixtures using isothermal calorimetry. Annual Book of ASTM Standards, vol. 04.02; 2008. [22] Mindess S, Young JF, Darwin D. Concrete. 2nd ed. Upper Saddle River, NJ: Prentice Hall; 2003. [23] ACI Committee 233. Slag cement in concrete and mortar, ACI 233R-03, American Concrete Institute; 2003. [24] Stein HN, Stevels JM. Influence of silica on the hydration of 3CaO, SiO2. J Appl Chem 1964;14(8):338–46. [25] Buil M, Paillère AM, Roussel B. High strength mortars containing condensed silica fume. Cem Concr Res 1984;14(5):693–704. [26] ACI Committee 234. Guide for the use of silica fume in concrete, ACI 234R-06, American Concrete Institute; 2006. [27] Zhang M-H, Gjørv OE. Effect of silica fume on cement hydration in low porosity cement pastes. Cem Concr Res 1991;21(5):800–8. [28] Qing Y, Zenan Z, Li S, Rongshen C. A comparative study on the pozzolanic activity between nano-SiO2 and silica fume. J Wuhan Univ Technol – Mater Sci Ed 2006;21(3):153–7.