studies on early stage hydration of tricalcium silicate incorporating silica nanoparticles: part i

Construction and Building Materials xxx (2014) xxx–xxx

Contents lists available at ScienceDirect

Construction and Building Materials

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

Studies on early stage hydration of tricalcium silicate incorporating silicananoparticles: Part I

http://dx.doi.org/10.1016/j.conbuildmat.2014.08.0460950-0618/� 2014 Elsevier Ltd. All rights reserved.

⇑ Corresponding author. Tel.: +91 1332 283442; fax: +91 1332 272272.E-mail addresses: [email protected], [email protected] (L.P. Singh).

Please cite this article in press as: Singh LP et al. Studies on early stage hydration of tricalcium silicate incorporating silica nanoparticles: Part I. ConsMater (2014), http://dx.doi.org/10.1016/j.conbuildmat.2014.08.046

L.P. Singh a,⇑, S.K. Bhattacharyya a, S.P. Shah b, G. Mishra a, S. Ahalawat a, U. Sharma a

a CSIR-Central Building Research Institute, Indiab Northwestern University, Civil and Environmental Engineering, Evanston, IL, USA

h i g h l i g h t s

� Dissolution of C3S is accelerated in presence of nanosilica.� XRD results revealed the presence of additional C–S–H at early stages of hydration in nanosilica incorporated C3S.� Nanosilica causes supersaturation of calcium hydroxide 30 min later.� 50% addition C–S–H formed in the presence of nanosilica.� More polymerized C–S–H is formed using nanosilica.

a r t i c l e i n f o

Article history:Available online xxxx

Keywords:Tricalcium silicateHydrationNano silicaC–S–H

a b s t r a c t

Hydration of tricalcium silicate (C3S) in presence of nano silica at early stage has been investigated andformation of addition calcium silicate hydrate (C–S–H) has been quantitatively estimated. Prepared C3Swas hydrated in the presence of powder nanosilica (30–70 nm) using w/b ratio 0.4 for paste study and5.0 for aqueous phase study. Results of ICP and XRD showed that the stage of supersaturation of Ca2+

is delayed by 30 min and formation of secondary/additional C–S–H starts at the early stages of C3S hydra-tion in the presence of nano silica. This additional C–S–H is responsible for higher dissolution of C3S asobserved by FTIR. SEM/EDX results show the formation of denser C–S–H with low C/S ratio in the pres-ence of nano silica, whereas TG analysis revealed the formation of additional C–S–H (�50%) in nano silicaincorporated C3S system which is accountable for the higher specific surface area as obtained by N2-adsorption.

� 2014 Elsevier Ltd. All rights reserved.

1. Introduction

Hydration of cement is a complex process due to the presenceof various kinetic steps and rapidly changing mineralogy and mor-phology, especially at early stages. Presently, efforts are beingmade to improve the performance of cementitious system byincorporating nanomaterials and understanding the physical andchemical attributes from nano to mesoscale. A few theories, viz.,Protective membrane, Semi-permeable membrane, Double layertheory, Crystallographic defects, Nucleation of calcium hydroxide(CH) and calcium silicate hydrate (C–S–H), etc. have been proposedto explain the phenomena occurring during hydration. In 1997,Taylor [1] reviewed these hypotheses and concluded that a differ-ent phase of C–S–H is formed on the surface of cement grainswithin a few seconds of hydration, which prevents its continued

dissolution. In 1962 and 1964, Kantro et al. and Stevels et al.[2,3] have reported that at first the hydrates are formed within afew seconds of hydration, which then are converted into anotherhydrate that is less closely fitted to the anhydrous surface andmore permeable to water. In 1989, Gartner and Gaidis [4] also indi-cated the existence of an impermeable hydrate at the cementgrains. Thermodynamic studies consider that this impermeablemetastable C–S–H, formed during the initial stage of hydration,has greater solubility than C–S–H and lesser solubility than trical-cium silicate (C3S) and remains stable only in the induction period.In 2010, Bellmann et al. [5] have emphasized that the hydration ofC3S proceeds in two consecutive steps. In the first reaction, anintermediate phase containing hydrated silicate monomer isformed, which is subsequently transformed into C–S–H as the finalhydration product in the second step (Fig. 1).

In 1986 and 2004, Jennings et al. [6,7] has reported the exis-tence of more than one type of metastable C–S–H formed duringhydration of C3S having a unique structure and composition.

tr Build

http://dx.doi.org/10.1016/j.conbuildmat.2014.08.046

mailto:[email protected]

mailto:[email protected]


http://www.sciencedirect.com/science/journal/09500618

http://www.elsevier.com/locate/conbuildmat


Fig. 1. Schematic representation of a process consisting of two consecutivereactions, (A = Ca3SiO5, B = the intermediate phase and C = C–S–H) [5].

2 L.P. Singh et al. / Construction and Building Materials xxx (2014) xxx–xxx

Further, in Jennings et al., 2000 [8] proposed a ‘J–T model’ andaccordingly, hydration process is completed in early, middle andlate periods. During hydration, two types of C–S–H are formed:low density (LD) and high density (HD) C–S–H (Fig. 2). First stageis a period of possibly diffusion-controlled slow reaction, calledthe ‘‘early period,’’ that ends at the time of initial set. Secondly,there is a nucleation and growth stage enduring until 12–14 h ormay be up to 24 h, called the ‘‘middle period.’’ Finally, there isthe ‘‘late period’’ or diffusion-controlled stages, for the remainingperiod of the reaction. They reported that under normal conditionsgenerally LD C–S–H is formed during the middle period, whereasduring the later stage, formation of HD C–S–H is predominant.

In contrast, Scrivener et al., 2011 [10] has reported that theslowdown in the initial reaction of C3S is caused by the changesin ion concentration in the solution, not by the presence of inhib-iting layers formed on the surface of C3S grains. Further, Scriveneret al., 2012 [11] have reported that hydration of alite is well simu-lated by two mechanisms: Solution Controlled Dissolution (SCD)up to the end of the induction period and Nucleation with Densify-ing Growth (NDG) to capture the main heat evolution peak. Theyproposed a scheme for early stage hydration and according to this,as C3S grains come to the contact of water, some primary hydratesare precipitated on the surface of C3S grains, but they do not form acontinuous layer [12]. Change in dissolution rate of alite is mainlyresponsible for the onset of the acceleration period (Fig. 3).

Other theories postulate that the length of the induction periodis governed by the nucleation and growth of hydrates, i.e. C–S–H or

Fig. 2. 2-D schematic representation of HD C–S–H (A

Please cite this article in press as: Singh LP et al. Studies on early stage hydrationMater (2014), http://dx.doi.org/10.1016/j.conbuildmat.2014.08.046

(CH) [13,14]. According to nucleation theory of CH, the nucleationof CH is rate determining parameter and induction period occursbecause CH is not precipitated, even after the saturation of thesolution due to the poisoning of the nuclei by silicate ions. Theinduction period ends when the level of super-saturation is suffi-cient to overcome this effect [15–17]. In nucleation theory of C–S–H, the rate of reaction during the induction period is controlledby nucleation and growth of the initially formed C–S–H and theinduction period ends when growth of C–S–H begins [19].

An important unresolved issue in the cement hydration is tounderstand the hydration chemistry during the dormant periodwith additives. Several additives are being used to improve theperformance of cementitious system, but understanding theirchemistry during hydration is a challenging task for cement andconcrete technologists. Applications of nano materials in construc-tion have brought a new revolution by improving the properties oftraditional building materials [20–23]. Aim of the nano materialsin cementitious system is to improve the characteristics of materi-als. Shah et al., 2010–2013 [24–29] have studied various engineer-ing properties of cementitious systems using different kinds ofnanomaterials such as colloidal nano silica, nanolime, carbonnanotubes, etc. Nano silica addition significantly alters the propor-tion of low and high stiffness C–S–H [30]. Several studies show thatthe application of nano silica in cementitious system improves thecompressive strength at early age of hydration [31–33]. Land andStephan, 2011 [34] have reported that the acceleration occurs dur-ing hydration of C3S, due to the formation of C–S–H seed on thesurface of nano silica. Singh et al., 2012 [35] have reported thataddition of nano silica into the cement paste refined the micro-structure of the paste and calcium leaching is significantly reducedas nano silica reacts with CH and forms additional C–S–H gel. Thechemistry of hydration in the presence of nano silica is an impor-tant issue from the point of research, especially during the induc-tion period. In the present studies, influence of nano silica onhydration during pre-induction and induction period of C3S wasexamined using ICP, XRD, TGA, FTIR and SEM/EDX techniques.

2. Material and methods

2.1. Preparation and characterization of tricalcium silicate (C3S)

C3S was prepared by high-temperature solid-state reaction between calciumcarbonate and silicic acid. Starting materials were mixed in the stoichiometric ratio(3:1) and then heated up to 1000 �C with a rate of 10 �C/min and a curing of 5 h atthis temperature. The mixture was further heated to 1500 �C with a rate of 5 �C/minand a curing of 12 h at the final temperature. The prolong heating at 1500 �C is

) and semidispersed LD C–S–H (B), respectively.

of tricalcium silicate incorporating silica nanoparticles: Part I. Constr Build


Fig. 3. Hydration scheme at early age of reaction, (a) an alite grain (b) when the alite grain is put in contact with water, hydrates precipitate (C–S–H in dark gray) but not as acontinuous membrane, (c) finally stable nuclei of CH and C–S–H start to grow [11].

Table 1Characterization of C3S.

CaO SiO2 1–10 l 10–100 l Specific surface area (m2/g)

74.2% 24.9% 41% 59% 3.01

L.P. Singh et al. / Construction and Building Materials xxx (2014) xxx–xxx 3

required to obtain a free lime content of 60.4%. Prepared C3S was grounded in amilling device using isopropanol as a dispersing agent to reduce the agglomerationand finally, the sample was dried at 40 �C after grinding.

Particle size distribution studies show that 41% particles have a size of 1–10 lmwhile 59% particles exhibit size in the range of 10–100 lm. The specific surface areaof prepared C3S as measured by BET is 3.01 m2/g, SEM micrograph revealed that theparticles of prepared C3S have polygonal shape (Fig. 4a) and its C/S ratio was 3:1 asanalyses by XRF (Table 1). XRD profile of prepared C3S shows its triclinic poly-morphs (Fig. 4b).

2.2. Preparation and characterization of nano silica

Sol–gel method was used for the preparation of nano silica using water glass asa silica precursor. In the preparation, cetrimonium bromide (CTAB), hydrochloricacid (HCl) and deionized water were mixed and stirred for 45 min. CTAB was usedas a dispersing agent and 1N HCl was used as a catalyst. Further, 1M solution ofsodium silicate was added drop wise with stirring at room temperature until thepH of the reaction system reached up to �9. The sol–gel mixture was washed withdeionized water to remove the sodium chloride. The synthesized powder was driedat 50–60 �C and then muffled at 700 �C for 4 h. SEM results of synthesized nano sil-ica show that the particle size varies from 30 to 70 nm (Fig. 5a). XRD profile as pre-sented in Fig. 5b shows that synthesized nano silica is amorphous in nature, and itsspecific surface area is around 116 m2/g as measured by BET (Table 2).

Fig. 4a. SEM micrograph of prepared unhydrated C3S.

Fig. 4b. XRD pattern of prepared unhydrated C3S.


2.3. Sample preparation of pure and nano silica incorporated C3S for hydration studies

For the determination of ionic concentration in the aqueous phase, pure andnano silica incorporated C3S samples were hydrated with a w/b ratio 5.0. Thehydrated mixtures were stirred continuously in a magnetic stirrer. These hydratedmixtures were filtrated at different time intervals starting from 30 s to 3:00 h.

For the study of hydration at early stage in the presence of nano silica, experi-ment was carried out by adding a fixed quantity (10% by wt. of C3S) of powder nanosilica in C3S. The w/b ratio was fixed 0.4 for all the samples. Hydration was stoppedat different time intervals starting from 2 min to 4:00 h, which completely coversthe pre-induction and induction period of hydration at early stage. To observe theeffect of nano silica on cement, same quantity (10% by wt. of cement) of powdernano silica was added to the cement keeping alike w/b ratio. Ordinary Portlandcement (OPC), 43 grade, conforming to IS: 8112 was used as such. The chemicaland mineralogical characteristics of the OPC are given in Table 3.



Fig. 5a. SEM micrograph of synthesized nanosilica.

Fig. 5b. XRD pattern of synthesized nanosilica.

Table 2Characterization of Synthesized n-SiO2.

Nanosilica SSA (m2 g�1) Particle size (nm)

n-SiO2 116 30–70

Table 3Chemical and mineralogical properties of OPC.

Items Content (%)

SiO2 22.3Al2O3 6.1Fe2O3 2.2CaO 61.1MgO 3.1SO3 2.1LOI 2.3C3S 49.4C2S 27.2C3A 10.9C4AF 11.3

840

850

860

870

880

0 30 60 90 120 150 180

Ca2+

Con

cent

ratio

n (p

pm)

Time (min)

Pure C3S

C3S+NS

Fig. 6. Ca2+ concentration in pure and nanosilica incorporated C3S.


2.4. Characteristics techniques

In the present studies, influence of nano silica on hydration during the pre-induction and induction period of C3S have been studied. The techniques, InductiveCouple Plasma (ICP) (model Prodigy XP, Teledyne Leeman lab, USA) for the determi-nation of ionic concentration, Powder X-ray diffraction (model DMax-2200, Rigaku,Japan) was used for the study of different phases formed during the hydration atearly stage. All the experiments were carried out at room temperature using copperradiation (Cu-Ka) at 40 kV/40 mA. Scanning was performed with a step width of0.02� 2h over an angular range from 5� to 80� with a scanning rate of 0.5�/minthroughout the experiments. FTIR (model NEXUS (1100), Thermo Nicolet, FTIR,USA) used for the study of chemical bonding in hydrated phases and specific surfacearea of hydrated samples was analyzed with BET (model Adsotrac DN-04, MicrotracSSA, USA). TGA (make: Perkin Elmer; model: Diamond) studies were performed at aheating rate of 5 �C/min under nitrogen flow. Morphological changes were observedusing Scanning Electron Microscope (model LEO 438 VP, Carl Zeiss AG, Oberkochen,Germany) at an accelerating voltage of 15–20 kV and the samples were analyzedunder variable pressure mode. For elemental analysis, Energy Dispersive X-ray(model X-Flash Detector 5010, Bruker, Nano GmbH, Germany) attached with SEMwas used.


3. Results

3.1. ICP

Concentration of Ca2+ was measured in the aqueous phase ofpure and nano silica incorporated C3S samples at early stage ofhydration. Results of these studies revealed that in the presenceof nano silica, a sudden fall in the Ca2+ concentration was observedwithin first few minutes of hydration (Fig. 6). This may be due tothe consumption of Ca2+ by nanosilica and thus, forming additionalC–S–H gel. The formation of C–S–H takes place during hydration ofC3S by dissolution and precipitation process [9]. In Fig. 6, in case ofpure C3S samples, Ca2+ concentration reduces due to precipitationof C–S–H & CH and again recovers (nearly 100%) due to solubility ofsparingly soluble CH. In case of C3S + nano silica, the concentrationof Ca2+ falls down in the very first few minutes of hydration (Fig. 6)due to consumption of Ca2+ by nano silica and form C–S–H seeds.These C–S–H seeds accelerate the hydration process through seed-ing. Although, the possibility of pozzolanic reaction of nano silica isalso be there because as sparingly soluble CH releases Ca2+, it willconsumed by nano silica. It will be too tough to state that pozzola-nic reaction is not there at this stage, however we can say that thenucleation effect of nano silica is predominant. Further, a suddenfall in Ca2+ concentration was observed at 2:30 h in pure C3S sys-tem showing the super-saturation stage of Ca2+ in pore solution,resulting in the crystallization of CH. However, this stage wasdelayed by 30 min (i.e. at 3:00 h) in nano silica incorporated C3Ssamples (Fig. 6). The most probable reaction chemistry duringthe hydration in presence of nano silica can be represented as:

Nucleation Reaction

O2�ðlatticeÞ þHþðaqÞ ! OH�ðaqÞ

Ca2þðlatticeÞ þ OH� ! CaðOHÞ2ðKsp ¼� 5:4 � 10�6Þ



Fig. 7a. XRD patterns of hydrated pure C3S.

Fig. 7b. XRD patterns of nanosilica incorporated hydrated C3S.


SiO4�4 ðlatticeÞ þ nHþðaqÞ ! HnSiOð4�nÞ�

4 ðaqÞ

SiO4�4 ðnSiO2Þ þ nHþðaqÞ ! HnSiOð4�nÞ�

4 ðaqÞ

Complete Reaction

Pozzolanic Reaction

CaðOHÞ2 þ nSiO2 ! C—S—H

Solubility product of CH is well known (�5.4 ⁄ 10�6) and theconcentration of lime in pore solution mainly control the hydrationprocess i.e. dissolution of ions from C3S grains and precipitation of


C–S–H. The solubility product of C3S is �9.6 ⁄ 10�6 as reported inliterature [18]. The stability of C–S–H is depends upon the C/S ofC–S–H and it’s reported that the C–S–H having C/S ratio in therange 1–2 have stable structure and lower concentration of Ca2+

accelerates the growth of C–S–H [19].

3.2. XRD

The complexity of the mineralogical analysis through XRD forcement based material is well known. Detailed XRD studies werecarried out for pure and nano silica incorporated hydrated C3S sys-tem. Each peak shift and intensity was monitored and carefullyanalyzed. Characteristic peak of the unhydrated C3S (2 h = �32)gets overlapped with the C–S–H peak in hydrated samples [36].In hydrated C3S samples, the intensity of this peak decreases as



0

1000

2000

3000

5 8 11 14 17 20

Inte

nsity

(Cou

nts)

2-Theta (0)

C3S at 4:00hC3S at 3:00hC3S at 2:30hC3S at 1:00h

CH

0

1000

2000

3000

5 8 11 14 17 20

Inte

nsity

(Cou

nts)

2 Theta (0)

CH

C3S + NS at 4:00h C3S + NS at 3:00h C3S + NS at 2:00h

C3S + NS at 1:00h

b

a

Fig. 8. Appearance of CH peak in pure hydrated C3S (a) and nanosilica incorporatedhydrated C3S (b).

0

1000

2000

3000

4000

5000

5 8 11 14 17 20

Hyd. Cement at 1:00hHyd. Cement at 2:00hHyd. Cement at 2:30hHyd. Cement at 3:00hHyd. Cement at 4:00h

Inte

nsity

(Cou

nts)

2-Theta (0)

CH

0

2000

4000

6000

8000

5 8 11 14 17 20

Hyd. Cement+NS at 1:00hHyd. Cement+NS at 2:00hHyd. Cement+NS at 2:30hHyd. Cement+NS at 3:00hHyd. Cement+NS at 4:00h

Inte

nsity

(Cou

nts)

2-Theta (0)

CH

a

b

Fig. 9. XRD patterns of hydrated Portland cement (a) and nanosilica incorporatedhydrated Portland cement (b).

15

25

35

45

55

65

75

85

95

105

700 750 800 850 900 950 1000 1050 1100 1150 1200

III

III

IV

I-Unhyd.C3SII-Hyd. C3S at 15minIII-Hyd. C3S at 60minIV-Hyd. C3S at 180min

Wavenumber (cm-1)

Tran

mis

sion

(%)

Fig. 10a. FTIR of hydrated C3S at different time intervals.

15

25

35

45

55

65

75

85

95

700 750 800 850 900 950 1000 1050 1100 1150 1200

Tran

mis

sion

(%)

III

III

IV

I-Unhyd.C3SII-Hyd. C3S+NS at 15minIII-Hyd. C3S+NS at 60minIV-Hyd. C3S+NS at180min

Wavenumber (cm-1)

Fig. 10b. FTIR of nanosilica incorporated hydrated C3S at different time intervals.


hydration proceeds up to 1:00 h, showing the dissolution of C3Swith the time (Fig. 7a). However, in the case of nano silica incorpo-rated hydrated C3S samples, the intensity of this peak increaseswith the time (up to 1:00 h) (Fig. 7b), indicating the formation ofadditional C–S–H, which is not observed in the hydrated C3S sam-ples. Therefore, in the presence of nano silica, a secondary C–S–H isformed at the early stage of hydration, as evident by new peak.

To study the effect of nano silica on the dormant and accelera-tion period, characteristic peak of CH at 2h = 18.2, was monitoredcarefully. XRD patterns of pure and nano silica incorporated C3Sin this theta range are presented in Fig. 8(a and b), respectively.In case of hydrated C3S samples, CH peak appeared at 2:30 h ofhydration (Fig. 8a) indicating that at this point the concentrationof ions reaches to its supersaturation level. Thus, crystallizationof CH takes place, which can be correlated with the ending of dor-mant period and start of the acceleration period [36,14,16]. On theother hand, in the case of nano silica incorporated hydrated C3S;this peak appeared at 3:00 h of hydration showing the shift inthe Ca2+ supersaturation state, i.e. 30 min later (Fig. 8b). SimilarXRD patterns were obtained in pure and nano silica incorporatedhydrated OPC samples as presented in Fig. 9a and b, respectively.These studies revealed that in OPC system, crystalline CH peakappeared at 2:30 h (Fig. 9a) of hydration. Nevertheless, in the caseof nano silica incorporated hydrated OPC samples, this peakappeared at 3:00 h of hydration, showing delay in CH crystalliza-tion (i.e. 30 min later) in nano silica incorporated OPC system(Fig. 9b). Therefore, effect of nano silica addition on C3S and OPCis comparable and leads to similar results.

3.3. FTIR

In the present study, hydration process was monitored for first3:00 h in pure and nano silica incorporated C3S samples with FTIR.The spectra of hydrated C3S and nano silica incorporated C3S at dif-


ferent time intervals starting from 15 to 180 min are presented inFigs. 10a and 10b). A broad peak in the region 850–950 cm�1



40

50

60

70

80

90

800 900 1000 1100 1200Tr

anm

issi

on (%

)Wavenumber (cm-1)

I

II

I- C3S at 1:00hII- C3S+NS at1:00h

50

60

70

80

90

800 900 1000 1100 1200

I

II

I- C3S at 3:00hII- C3S+NS at 3:00h

a b

Wavenumber (cm-1)

Tran

mis

sion

(%)

Fig. 11. Comparative FTIR spectra of hydrated C3S and nanosilica incorporated C3S at 1 h (a) and 3 h (b), respectively.

Table 4Quantification of C–S–H in pure and silica modified C3S system.

Time(min)

Quantification of C–S–H (%) inpure hydrated C3S

Quantification of C–S–H (%) inhydrated C3S + NS

30 0.56 0.9360 0.60 1.14

150 0.65 1.25180 0.76 1.9240 1.42 3.0

0

1.5

3

4.5

15 30 60 180 240

Pure C3S C3S + NS

Surf

ace

Are

a (m

2 /g)

Time (min)

Fig. 12. SSA of hydrated C3S and nanosilica incorporated C3S samples.


shows the characteristics peak of unhydrated C3S [37,38]. A grad-ual decrease in the transmittance peak at this region indicatesthe consumption of C3S during hydration [39]. Depletion rate ofthis peak is higher in nano silica incorporated C3S samples at earlystages of hydration. Depletion rate of this transmittance peakdecreases from 15 min onwards in pure hydrated C3S samples(Fig. 10a) showing the dormant period, when the dissolution rateslows down. However, in nano silica incorporated system, at thistime interval consumption of C3S was higher showing the acceler-ation in the dissolution process (Fig. 10b). Slower dissolution of C3Sin nano silica incorporated C3S samples occurs due to delay in theacceleration period. In nano silica incorporated C3S, samples a newpeak (�1010 cm�1) appeared at 1:00 h, represents Q2 peak of C–S–H [40], which become more intense at 3:00 h, indicating the poly-merization of the silicate chain in C–S–H gel, while this peak wastotally absent in hydrated C3S samples up to 3:00 h (Fig. 11a andb).

3.4. BET and TGA

Quantification of C–S–H in the pure and nano silica incorpo-rated C3S systems at different time intervals of hydration was car-ried out using following Eq. (1) by TG analysis [41].

C—S—Hð%Þ ¼ Total LOI� LOIðCHÞ � LOIðCCÞ ð1Þ


where LOI (CH) is the dehydration of calcium hydroxide around400–500 �C and LOI (CC) is the carbon dioxide release around600–800 �C. Table 4 shows the amount of C–S–H in pure and nanosilica incorporated hydrated C3S systems at different time intervals(30–240 min). Results of these studies revealed that in the nano sil-ica incorporated C3S system, the amount of C–S–H was significantlyhigher than pure C3S system due to formation of secondary C–S–Hgel. This addition C–S–H is formed by the both nucleation and poz-zolanic reaction of nano silica. BET results of pure and nano silicaincorporated C3S systems revealed that the specific surface area ofthe nano silica incorporated C3S samples is higher than the pureC3S samples (Fig. 12), which indicates that in the presence of nanosilica more C–S–H was formed, because in hydrated samples C–S–His responsible for the high specific surface area due to its nano sizestructure.

3.5. SEM/EDX

SEM micrographs of pure and nano silica incorporated hydratedC3S samples at different time intervals are presented in Fig. 13a–f.Hydrated C3S samples at 30 min of hydration show few hydrationproducts on the surface of C3S grains (Fig. 13a), which increases at60 min (Fig. 13c). A thin hydrated layer is formed around the C3Sgrains at 180 min (Fig. 13e). In nano silica incorporated C3S sam-ples, whole C3S grain is covered by thin hydrated layer, within0.50 h of hydration (Fig. 13b). This fine hydrated layer convertedinto foil like structure at 1:00 h (Fig. 13d), which become denserat 3:00 h (Fig. 13f). These results revealed that in the presence ofnano silica, hydration process was accelerated and more hydratedproducts were formed. EDX analysis of these hydrated productsshows that at the initial stage, very high C/S ratio of C–S–H isformed, which later converted into low C/S ratio. Nano silica incor-porated hydrated samples shows that the C–S–H formed duringhydration has low C/S ratio in comparison to pure hydrated C3Ssamples (Fig. 14).

4. Discussion

Hydration of the cementitious system is a complicated processand its complexity increases with the use of additives. Though theaddition of nano silica improves the early age strength, the precisechemistry behind the phenomenon is still not very clear. The con-centration of Ca2+ in pore solution measured by ICP, show a suddenfall within first few minutes of hydration in the presence of nanosilica (Fig. 6). This attributes to the higher kinetics of nano silica(due to its high specific surface area), which reacts with the Ca2+

released from C3S grains and forms secondary C–S–H. XRD resultsalso show the existence of additional C–S–H at early stages ofhydration. In hydrated samples, characteristic peak of C3S over-lapped with the C–S–H peak at 2h = �32 [36]. The intensity of thispeak gradually decreases up to 1:00 h in pure C3S samples,showing the dissolution of C3S (Fig. 7a). However, in the case of



Fig. 13. SEM micrographs of pure C3S (a, c and e) and nanosilica incorporated C3S (b, d and f) at 30, 60 and 180 min of hydration, respectively.

1

2

3

4

15 30 60 80 150 180

C3SC3S+NS

C/S

Rat

io

Time (min)

Fig. 14. EDX results of hydrated C3S and nanosilica incorporated C3S samples.


nano silica incorporated C3S, intensity of this peak increases up to1:00 h (Fig. 7b) showing the formation of C–S–H. This new peakmay be associated to the formation of secondary C–S–H, becausethis trend was absent in hydrated C3S samples. Crystalline CH peakat 2h = 18.2 in hydrated samples represents the stage of supersat-uration of ions in the pore solution. In hydrated C3S samples thispeak appeared at 2:30 h of hydration (Fig. 8a), while this stagecomes 30 min later i.e. at 3:00 h in the presence of nano silica(Fig. 8b). These findings are also in consistent with the results


obtained by ICP. The concentration of Ca2+ falls at 2:30 h of hydra-tion in case of pure C3S samples showing the stage of supersatura-tion when crystallization of CH takes place. This stage appeared at3:00 h of hydration, i.e. 30 min later in the presence of nano silica.This shift in the stage of super saturation occurs because, in thepresence of nano silica, Ca2+ are regularly consumed by the nanosilica and thus, forming secondary C–S–H. Therefore, limitedamounts of Ca2+ were available for the crystallization of CHthereby, delaying in appearance of peak. However the effect crys-tallization of CH on induction period will further studied by thecalorimetry results.

Further, in FTIR studies, a broad band at 850–930 cm�1(charac-teristic peak of crystalline C3S) [37,39] gradually decreases withthe hydration (Fig. 10a and b) showing the consumption of C3Sduring hydration. Depletion rate of this transmittance peak washigher in nano silica incorporated C3S than pure C3S samples atearly stages of hydration. This indicates that the dissolution pro-cess is accelerated in the presence of nano silica [39]. This acceler-ation occurs due to the formation of additional C–S–H at earlystage. In nano silica incorporated C3S samples, a new peak(�1010 cm�1) appeared at 1:00 h represents the Q2 peak of C–S–H [40] showing the polymerization of the silicate chain in C–S–Hgel. This peak was absent in hydrated pure C3S samples up to3:00 h (Fig. 11a and b), indicating that the polymerization of thesilicate chain in C–S–H gel speeds up with nano silica. Higher sur-face reactivity has been observed by SEM in the presence of nano




silica within 0.50 h of hydration (Fig. 13a and b). At the later stage(i.e. at 1:00 and 3:00 h) denser C–S–H is formed in the presence ofnano silica (Fig. 13c–f) indicating the existence of additional C–S–H. EDX analyses show that at early stage of hydration, in pure C3Ssamples, C–S–H has high C/S ratio �3.0, which reduces (�2.5) asthe hydration proceeds up to 4:00 h. In nano silica incorporatedC3S samples, C–S–H has low C/S ratio (�2.5) than pure C3S samplesat the early stage which converted to the C–S–H having C/S ratio�1.8 at 4:00 h (Fig. 14) and this C–S–H with low C/S ratio isresponsible for the formation compact or denser C–S–H duringhydration. Quantification of C–S–H with TG analysis showed thatin nano silica incorporated C3S samples, the quantity of C–S–Hformed is approximately double than the pure one (Table 4) at4:00 h of hydration. This can be correlated with the formation ofadditional C–S–H (�50%) at early stage of hydration in the pres-ence of nano silica which is responsible for high specific surfacearea as displayed in BET results (Fig. 12).

5. Conclusions

Nano-silica is being incorporated into the cement to giveenhanced mechanical property, lower porosity and permeability,which are crucial factors for increased durability. To understandthe chemistry of hydration in the presence of nano silica, C3S washydrated with a w/b ratio 0.4 for the paste study and 5.0 for aque-ous phase study. Hydration process was monitored for the first fourhour of hydration, which completely covered the preinduction andinduction period of hydration and findings are as follows:

� XRD results show the presence of additional C–S–H at earlystages of hydration in the presence of nano silica as evidentby new peak appearance.� CH peak appeared 30 min later in nano silica incorporated C3S

system, showing the shift in the stage of supersaturation, whichmay be correlated with the shift in the induction period.� ICP results also show the existence of additional C–S–H and

shift in the induction period in the presence of nano silica.� FTIR results revealed that in the presence of nano silica, dissolu-

tion of C3S grain accelerates and polymerization in the silicatechain speeds up.� SEM/EDX results show a denser C–S–H with low C/S ratio was

formed in the presence of nano silica showing the higher reac-tivity of nano silica and existence of additional C–S–H.� TG analysis shows that in the presence of nano silica�50% addi-

tional C–S–H is formed.

Acknowledgement

Author (Usha Sharma) is grateful to University Grant Commis-sion, New Delhi for grant of Junior Research Fellowship.

References

[1] Taylor HFW. Cement chemistry. London: Thomas-Telford; 1997.[2] Kantro DL, Brunauer S, Weise CH. Development of surface in the hydration of

calcium silicates. II. Extension of investigations to earlier and later stages ofhydration. J Chem Chem 1962;66(10):1804–9.

[3] Stein HN, Stevels JM. Influence of silica on hydration of C3S. J Appl Chem1964;14:338–46.

[4] Gartner EM, Gaidis JM. Hydration mechanisms I. In: Skalny JP, editor. Mater SciConcr I Am Ceram Soc; 1989. p. 95–125.

[5] Bellmann F, Damidot D, Moser B, Skibste J. Improved evidence for the existenceof an intermediate phase during hydration of tricalcium silicate. Cem ConcrRes 2010;40:875–84.

[6] Jenning HM. Aqueous solubility relationship for two types of calcium silicatehydrate. J Am Ceram Soc 1986;69(8):614–8.

[7] Chena JJ, Thomas JJ, Taylor HFW, Jenning HM. Solubility and structure ofcalcium silicate hydrate. Cem Concr Res 2004;34:1499–519.


[8] Jennings HM. A model for the microstructure of calcium silicate hydrate incement paste. Cem Concr Res 2000;30:101–16.

[9] Bullard et al. Mechanisms of cement hydration. Cem Concr Res2011;41:1208–23.

[10] Scrivener KL, Nonat A. Hydration of cementitious materials, present andfuture. Cem Concr Res 2011;41:651–65.

[11] Kumar A, Bishnoi S, Scrivener KL. Modelling early age hydration kinetics ofalite. Cem Concr Res 2012;42:903–18.

[12] Juilland P, Gallucci E, Flatt R, Scrivener K. Dissolution theory applied to theinduction period in alite hydration. Cem Concr Res 2010;40(6):831–44.

[13] Wu ZQ, Young JF. Formation of calcium hydroxide from aqueous suspensionsof tricalcium silicate. J Am Ceram Soc 1984;67:48–51.

[14] Barret P, Ménétrier D. Filter dissolution of C3S as a function of lime concentrationin a limited amount of lime water. Cem Concr Res 1980;10:521–34.

[15] Fierens P, Verhaegen JP. Hydration of tricalcium silicate in paste – kinetics ofcalcium ions dissolution in the aqueous phase. Cem Concr Res 1976;6:337–42.

[16] Young JF, Tong HS, Berger RL. Compositions of solutions in contact withhydrating tricalcium silicate pastes. J Am Ceram Soc 1977;60(5–6):193–8.

[17] Odler I, Dörr H. Early hydration of tricalcium silicate II. The induction period.Cem Concr Res 1979;9:277–84.

[18] Nicoleau L et al. The di- and tricalcium silicate dissolutions. Cem Concr Res2013;47:14–30.

[19] Garrault S, Nonat A. Experimental investigation of calcium silicate hydrates(C–S–H) nucleation. J Cryst Growth 1999;200:565–74.

[20] Sobolev K, Gutierrez MF. How nanotechnology can change the concrete world:Part 1. Am Ceram Soc Bull 2005;84(10):14–7.

[21] Sobolev K, Gutierrez MF. How nanotechnology can change the concrete world:Part 2. Am Ceram Soc Bull 2005;84(11):16–9.

[22] Poole CP, Owens FJ. Introduction to nanotechnology. John Wiley and Sons;2003.

[23] Korpa A, Trettin R, Bottger KG, Thieme J, Schmidt C. Pozzolanic reactivity ofnanoscale pyrogene oxides and their strength contribution in cement-basedsystems. Adv Cem Res 2008;20:35–46.

[24] Hou P, Kawashima S, Kong D, David JC, Qian J, Shah SP. Modification effects ofcolloidal nano SiO2 on cement hydration and its gel property. Composites: PartB 2013;45:440–8.

[25] Konsta-Gdoutos MS, Metaxa ZS, Shah SP. Multi-scale mechanical and fracturecharacteristics and early-age strain capacity of high performance carbonnanotube/cement nanocomposites. Cem Concr Res 2010;32(2):110–5.

[26] Kong D, Du X, Wei S, Zhang H, Yang Y, Shah SP. Influence of nano-silicaagglomeration on microstructure and properties of the hardened cementbased materials. Constr Build Mater 2012;37:707–15.

[27] Kawashima S, Hou P, David JC, Shah SP. Modification of cement-basedmaterials with nanoparticles. Cem Concr Comp 2013;36:8–15.

[28] Konsta-Gdoutos MS, Metaxa ZS, Shah SP. Highly dispersed carbon nanotubereinforced cement based materials. Cem Concr Res 2010;40:1052–9.

[29] Metaxa ZS, Konsta-Gdoutos MS, Shah SP. Carbon nanofiber cementitiouscomposites: effect of debulking procedure on dispersion and reinforcingefficiency. Cem Concr Compos 2013;36:25–32.

[30] Mondal P, Shah SP, Marks LD, Gaitero JJ. Comparative study of the effects ofmicrosilica and nano silica in concrete. Transp Res Rec 2010:6–9.

[31] Singh LP, Karade SR, Bhattacharyya SK, Yousuf M, Ahalawat S. Beneficial role ofnano silica in cement based materials – a review. Constr Build Mater2013;47:1069–77.

[32] Singh LP, Bhattacharyya SK, Ahalawat S. Preparation of size controlled silicanano particles and its functional role in cementitious system. J Adv ConcrTechnol 2012;10:345–52.

[33] Singh LP, Bhattacharyya SK, Singh P, Ahalawat S. Granulometric synthesis andcharacterisation of dispersed nano silica powder and its application incementitious system. Adv Appl Ceram 2012;111:220–7.

[34] Land G, Stephan D. The influence of nano-silica on hydration of ordinaryPortland cement. J Mater Sci 2011;47(2):1011–7.

[35] Singh LP, Bhattacharyya SK, Mishra G, Ahalawat S. Reduction of calciumleaching in cement hydration process using nanomaterials. Mater Technol2012;27:233–8.

[36] Jadhav R, Debnath NC. Computation of X-ray powder diffractograms of cementcomponents and its application to phase analysis and hydration performanceof OPC cement. Bull Mater Sci 2011;34(5):1137–50.

[37] Ylmén R, Jäglid U, Steenari BM, Panas I. Early hydration and setting of Portlandcement monitored by IR, SEM and Vicat techniques. Cem Concr Res2009;39:433–9.

[38] Fernandez L, Alonso C, Hidalgo A, Andrade C. The role of magnesium during thehydration of C3S and C–S–H formation. Scanning electron microscopy andmid-infrared studies. Adv Cem Res 2005;17(1):9–21.

[39] Bjornstrom J. Accelerating effects of colloidal nano-silica for beneficialcalcium-silicate-hydrate formation in cement. Chem Phys Lett2004;392:242–8.

[40] Yu P, Kirkpatrick RJ, Poe B, McMillan PF, Cong X. Structure of calcium silicatehydrate (C–S–H): near-, mid-, and far infrared spectroscopy. J Am Ceram Soc1999;82(3):742–8.

[41] Aono Y, Matsushita F, Shabita S, Hama Y. Nano structural changes of C–S–H inharden cement paste during drying at 50 �C. J Adv Concr Technol2007;5:313–23.


http://refhub.elsevier.com/S0950-0618(14)00958-1/h0005






































































































studies on early stage hydration of tricalcium silicate incorporating silica nanoparticles: part i

Documents