Construction and Building Materials 66 (2014) 306–312

Contents lists available at ScienceDirect

Construction and Building Materials

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

Preliminary attempt of hardened mortar sealing by colloidal nanosilicamigration

http://dx.doi.org/10.1016/j.conbuildmat.2014.05.0400950-0618/� 2014 Elsevier Ltd. All rights reserved.

⇑ Corresponding author. Tel.: +34 913020440x870304.E-mail address: mercesanc@ietcc.csic.es (M. Sánchez).

M. Sánchez a,⇑, M.C. Alonso a, R. González b

a Instituto ‘‘Eduardo Torroja’’ de Ciencias de la Construcción (IETCC – CSIC), C/Serrano Galvache 4, 28033 Madrid, Spainb Universidad Autónoma de Nuevo León, Facultad de Ingeniería Civil, Pedro de Alba s/n, San Nicolás de los Garza, Nuevo León, Mexico

h i g h l i g h t s

� Efficient superficial sealing of mortar by colloidal nanosilica migration.� Significant parameters in treatment: nanosilica composition and electric field power.� Consolidation of treated mortar due to the decrease of capillary pores.� Nanosilica and cementitious matrix interaction: pozzolanic activity and C–S–H gels formation.

a r t i c l e i n f o

Article history:Received 10 December 2013Received in revised form 15 April 2014Accepted 15 May 2014Available online 18 June 2014

Keywords:Colloidal nanosilicaMigrationSealing propertiesRepair techniques

a b s t r a c t

An innovative electrochemical treatment for the sealing of hardened mortar by nanosilica migration isproposed. The composition of the colloidal suspension and the electric field power were assessed assignificant parameters of the transport efficiency. The interaction between the nano-SiO2 and the cementmatrix was characterized by different techniques: electrical resistivity measurements, DTA–TGA, BS-SEMwith EDX microanalysis coupled and MIP. Results showed that the transport of colloidal nano-SiO2 underthe action of the electric field can be achieved. An active interaction with the solid phases of the mortarsamples was confirmed from the decrease in the portlandite content of treated mortar and from thesmaller C/S ratio found in the solid hydrated phases of the treated mortars. A significant decrease of cap-illary pores was also identified at the age of 28 days after finishing the treatment. The sealing ability ofmigrated silica nanoparticles was enhanced with the ageing of the treated sample; smaller amount ofmicropores was measured with the treated mortar ageing.

� 2014 Elsevier Ltd. All rights reserved.

1. Introduction

Nowadays, nano-technology is a real aspect of the daily life andis present in almost every developed industrial field, although itsintrusion is being slower in the construction industry. The use ofnano-materials with cementitious functions coupled with theapplication of modern technologies is in continuous evolution,promoting important performance improvements in the construc-tion industry. These aspects are getting more demanded every dayin order to increase the lifetime not only of new concrete struc-tures to be built but also in the case of the existing ones. In thissense, the addition of colloidal nano-silica to fresh concrete hasbeen proposed as a promising technology to improve significantlythe mechanical and durable properties in hardened state [1–4]. Thetechnology for improving the performance of existing structures in

a non-invasive way is still in an incipient state, even though of theemerging industrial interest in developing sustainable technologyin the field of rehabilitation and retrofitting of damaged concretestructures.

Conventional electrochemical repair methods, such as electro-chemical chloride extraction (ECE) or electrochemical realkaliza-tion (ERA), are already standardized [5,6], although reticenceconcerning their long-term efficiency [7–9] and total regenerationof the passive layer still exists [10]. Then, improvements of thesetechniques to overcome those limitations arise as promisingalternatives for non-invasive repair of reinforced concrete. Thesimultaneous application of corrosion inhibitors during the electro-chemical repair treatment has been reported to significantlyimprove the repassivation of the rebar after finishing the treatment[11–14]. Applying a superficial treatment to seal the concretesurface has been also proposed to increase the efficiency of electro-chemical repair methods by limiting the further access of aggres-sive chemical species in the structure [15,16].

M. Sánchez et al. / Construction and Building Materials 66 (2014) 306–312 307

Electrochemical repair methods are based on the acceleratedtransport of charged particles into the pore structure by migrationmechanism under the action of an electric field. In this sense, theelectric field can be used not only to extract chloride ions fromthe porous system of the cement matrix connecting the reinforce-ment as cathode [5,6], but also to introduce positive chargedparticles to the concrete bulk. Cathodic inhibitors, such as organicbased mixed inhibitors, have been effectively migrated using con-ventional arrangement of ECE or ERA treatments by connecting thereinforcement to the cathode [13,14]. Electrokinetic transport ofpositively charged nanoparticles using similar arrangements hasbeen also performed; alumina coated silica nanoparticles havebeen penetrated into concrete specimens to promote the efficientreduction of the hardened concrete permeability [15]. The interac-tion of these particles with the cement matrix has been reported topromote the formation of a denser material, reducing the chlorideinduced corrosion of reinforced concrete [17,18].

However, this arrangement cannot be directly employed fornegatively charged compounds and an appropriate modificationof the procedure has been performed to favour the migration ofanodic inhibitors, as demonstrated in [11–12]. Furthermore, themigration of colloidal nano-SiO2 under the action of an electricfield has been also considered [19]. The practical implementationof such technology implies the use of the modified arrangementdescribed in [11,12] to promote silica nanoparticles, negativelycharged.

The interaction of colloidal nanosilica with the cement grainswhen added in fresh state has been analysed in literature [2]. Aphysical effect of nanosilica acting as filler by filling up the poresto give a cement paste with larger packing density and smallerporosity has been observed [19]. Furthermore, the chemical inter-action of nanosilica with portlandite forming additional C–S–H gelthat implies a pozzolanic reaction has been reported elsewhere[20,21]. Nanosilica has also showed ability to accelerate thecement hydration by acting as nucleating sites [22] on the anhy-drous cement grains. However, in the case of nanosilica penetra-tion through hardened concrete, scarce information can be foundconcerning the interaction between the transported silica nanopar-ticles and the cementitious matrix. Thus, in the present work amigration treatment of silica nanoparticles on hardened mortarhas been performed and the effect of nano-SiO2 on the cementmatrix after the migration process has been assessed. The effectof significant variables such as the electric field power and the sizeand concentration of nanoparticles in interaction with the hard-ened cement mortar has been analysed. The ageing of the treatedmortar has been also considered to evaluate the long-term behav-iour of this type of sealing treatments.

2. Materials and method

2.1. Cement mortar samples

Cylindrical mortar samples with 7.5 cm of diameter and 15 cm of height wereprepared with an Ordinary Portland Cement (OPC 52.5 R) sulphate resistant withlow C3A content (3.4%) and low alkali content (Na2Oeq 0.7%). The cement chemicalcomposition is detailed in Table 1. The sample preparation was carried out usingsiliceous aggregates with a chemical composition of P98% SiO2 and a moisture con-tent <0.2%. Deionised water was used in order to avoid the intrusion of other ions inthe mix that could interfere with the analysis. A water/cement ratio of 0.5 and aproportion of cement/aggregate of 1:3 were used in the preparation of the mortar

Table 1Chemical composition of the employed OPC cement (%).

Na2O K2O Al2O3 CaO Fe2O3 SiO2 MgO SO3 Loss Ign.

0.33 0.59 4.16 67.48 4.57 19.68 0.83 3.25 –

samples. An automatic mortar mixer was used to prepare the materials followingUNE-EN 196-1:2005 standard. Once prepared the mortar samples, they were curedunder controlled conditions (98 ± 2% relative humidity and 21 ± 2 �C) for 7 days.

After curing, the mortar samples were cut in slices of 1 cm, removing the topand bottom sides. The 1 cm mortar slices thickness were saturated in water undervacuum conditions and were maintained immersed in water for storage untiltesting.

2.2. Colloidal nano-SiO2

Two different commercial colloidal nano-SiO2 aqueous dispersions were used,as resumed in Table 2. As indicated by the product data-sheet [23,24], the amor-phous silica particles are discrete, spherical and monodispersed and modified witha reactive epoxy silane to have a slight superficial negative charge.

2.3. Migration cell assembly

Migration tests were carried out in a two compartments cylindrical cell. Thecompartment acting as catholyte (cathode connection) was filled with the colloidalnano-SiO2 solution and the anolyte compartment (anode connection) with distilledwater. The mortar slice was located between both compartments. The scheme ofcell is detailed in Fig. 1. The electric field was connected between two steel rebars(diameter of 6 mm) located at both compartments acting as external electrodes.Two different constant voltages (6 V and 12 V) were applied during 8 days oftreatment.

During testing, periodical electrical measurements, such as the current passedthrough the system and the real voltage between both sides of the mortar sample,were carried out with an Autoranging Microvolt DMM multimeter from Keithley.The effective voltage was measured between two stainless steel meshes locatedclose to the mortar surfaces, as observed in Fig. 1, in order to avoid the polarizationcontributions at the electrodes and to minimize the electrolyte contribution. Cur-rent passing through the system was also periodically measured. The evolution ofthe mortar electrical resistance was estimated by Ohm’s Law.

2.4. Characterization of the nanosilica/mortar interaction

2.4.1. Electrical resistivity measurementsThe resistivity of the mortar sample was measured before and after the treat-

ment by a direct measurement using two electrodes attached to both sides of thetreated sample according to UNE 83988-1:2008. This parameter was utilized asan indirect indicator of the mortar consolidation due to the migration nanosilicatreatment.

2.4.2. Mercury Intrusion Porosimetry (MIP)After nano-SiO2 treatment, mortar samples taken from the surface directly

exposed to the nanosilica compartment during the migration treatment (4–5 mmdepth) were analysed in an AutoPore IV 9500 V1.05 equipment from MicromeriticsInstrument Corporation. The total porosity values and the pore size distributionwere determined. Before the analysis, the samples were dried at 40 �C during48 h. Two different ages after the treatment were considered: 28, and 90 days.

2.4.3. Differential Thermal Analysis (DTA) and Thermogravimetric Analysis (TGA)The DTA/TGA analysis was conducted on a TA Instruments SATQ600 thermal

analyser in a dynamic nitrogen atmosphere. The samples were heated at a rate of4 �C/min from 25 to 1000 �C. Small pieces of mortar were taken from the surfacethat had been exposed to nanosilica compartment during the migration treatment(negative side). Mortar samples were powdered for the analysis by grinding at theambient air. The powdered samples were not previously dried before the thermalanalysis. Platinum and aluminium crucibles were used as references. The weightloss associated to the dehydration of cement products such as C–S–H gel and port-landite were estimated for mortar samples at the age of 28 days after finishing thetreatment. A non-treated mortar with the same age was taken as reference.

2.4.4. Back-Scattering Microscopy with X-ray Microanalysis (BSE/EDX)The nanosilica interaction and reactivity with the cement matrix were also eval-

uated by characterizing the microstructure and chemical composition of mortars atthe age of 28 days after finishing the treatment by BSE/EDX with a JEOL JSM-6360LVwith an EDX detector coupled for elemental chemical analysis. The C/S ratio of C–S–H gel phases was estimated as a qualitative indicator of the chemical interactionbetween the silica nanoparticle and the cement solid matrix.

Table 2Properties of the studied colloidal nanosilica suspensions.

ID Average particle size (nm) Silica (wt%) Density (g/cm3) pH

B-30 7 29 1.2 8B-40 12 37 1.3 8

Fig. 1. Scheme of the migration cell set-up.

308 M. Sánchez et al. / Construction and Building Materials 66 (2014) 306–312

3. Results and discussion

3.1. Nanosilica transport characterization

The mortar electrical resistance, estimated during the connec-tion of the electric field, is represented in Fig. 2. A clearly differentbehaviour was observed for both colloidal suspensions studied, B-30 and B-40. An increase of the mortar electrical resistance wasregistered from the beginning of the treatment migrating the lessconcentrated suspension with smaller nanoparticles size, B-30. Inthe case of B-40 suspension, the electrical resistance of mortar onlyincreased after a certain induction t time of the electrical fieldsrunning, that varies with the level of the voltage used. Thus, theincrease of the resistance against the ions movement through mor-tar sample was progressively detected during the nanosilica migra-tion treatment. The ionic transport seemed to be more hindered forthe migration of the smaller nanoparticles (B-30 suspension)through the mortar pores.

3.2. Nanosilica/mortar interaction

The sealing ability of the colloidal nanosilica was characterizedon mortar samples at the age of 28 days after finishing the treat-ment by the relative resistivity of the treated mortar, taken as a

Fig. 2. Evolution of the electrical resistance through mortar sample during theelectric field connection.

qualitative indirect indicator. The relative resistivity values,defined as the ratio between the final and the initial resistivity,are included in Fig. 3 for the different studied situations.

An increase on the resistivity was measured after the treatmentof sealing with the colloidal nanosilica, independently on theexperimental conditions. While a non-significant influence of theelectric field power was deduced, the effect of the product compo-sition seemed to be more relevant. The larger increase on the resis-tivity values obtained after the treatment with B-30 suspensionshould be related to an improved sealing ability of this productwith smaller nanoparticle size. A more significant mortar densifi-cation and consolidation could be expected in this case.

MIP measurements were carried out to obtain additional infor-mation about the sealing ability of the penetrated nanosilica. TheMIP characterization was carried out on mortar samples at theage of 28 days after finishing the treatment, taken from the surfacedirectly exposed to the nanosilica during the treatment. A non-treated mortar with the same age was taken as reference. InFig. 4(a) the cumulative pore volume curve is represented for eachcase.

Smaller cumulative porosity was found for the treated mortarsconfirming the sealing ability of nanosilica after the migrationtreatment. However, not significant effect neither of the electricfield nor of the nature of the nanosilica dispersion could bededuced, Fig. 4(a). In Table 3 the total porosity values are resumedand the slight porosity decrease in the treated mortar at the age of28 days after finishing the treatment can be perceived.

For a more detailed analysis, the differential pore size distribu-tion is represented in Fig. 4(b). Treated mortars showed similarporosity distribution, with two marked peaks at around 0.5 and0.07 lm, independently of the electric power applied and the sus-pension composition. However, for the reference case, the pore sizedistribution appeared slighted displaced to larger pore diameters.The sealing effect of nanosilica treatment seems to affect mainlythe capillary pores (<0.1 lm). Similar behaviour was observed incement pastes with nanosilica addition [19] where the filler effectof nanosilica by filling up the voids between the cement grains wasreported to promote a reduced capillary porosity of the cementpaste. The estimated capillary pore content for the different sam-ples is resumed in Table 3, where the clear decrease is registeredfor the treated mortars in comparison to the reference mortar. Nei-ther the applied voltage nor the nature of the nanosilica suspen-sion used for the treatment showed a crucial influence inporosity results.

MIP results do not inform the type of sealing obtained afterthe nanosilica migration treatment; both a physical effect due tothe filler action of nanosilica and a chemical effect associated to

ΔΔρ

ρρ

Fig. 3. Relative resistivity increase of treated mortar at the age of 28 days afterfinishing the treatment.

Fig. 4. MIP at 28 days after the treatment finished. (a) cumulative pore volumecurve and (b) logarithmic differential pore volume curve.

Table 3Total porosity and capillary pores content at the age of 28 days after finishing thetreatment.

Sample Total porosity (%) Capillary pores (<1 lm) (%)

R0_28d 16.76 88.05B30_6V_28d 15.41 78.81B40_6V_28d 14.35 82.10B30_12V_28d 14.34 81.57B40_12V_28d 15.06 83.56

Fig. 5. TGA of reference and treated mortars at the age of 28 days after finishingtreatment.

Table 4Composition of the main solid phases of treated mortars at the age of 28 days afterfinishing the treatment compared the reference mortar estimated from TGA results.

Sample % Mass loss

C–S–H gel (100–350 �C) Portlandite (350–450 �C)

R0_28d 1.95 1.43B30_12V_28d 1.89 1.09B40_12V_28d 1.66 1.00

M. Sánchez et al. / Construction and Building Materials 66 (2014) 306–312 309

the nanosilica reactivity can be expected [20–22]. To confirm if thenanosilica is able to react even when applied on a hardenedcementitious matrix, DTA/TGA analyses were carried out. InFig. 5, the TGA curve is included for the mortars at the age of28 days after finishing the treatment with 12 V electrical connec-tion and also for a reference mortar, without treatment. The sameregion than the studied with MIP was analysed.

No significant differences between treated and non-treatedmortars were detected from the DTA curve, although larger port-landite content in the reference sample could be deduced fromthe peak at 450 �C. For a more quantitative analysis, the contentof the main solid phases of the cement matrix was estimated fromthe weight losses data, as shown in Table 4. The different charac-teristic intervals of temperature considered were the related towater from C–S–H gel (100–350 �C) and water from portlandite(350–450 �C).

Smaller weight losses were registered for the treated mortarsamples than for the reference one. A significant decrease of theportlandite content was registered in both treated mortars

confirming that the nanosilica particles are able to promote thepozzolanic reaction even when applied to hardened mortar.However, certain effect of the electric field should not be neglectedneither; in fact, the enhancement of the portlandite dissolution dueto the electric field connection was already reported elsewhere[25].

Even though similar pozzolanic activity is expected for bothcolloidal nanosilica suspensions, the weight loss associated toC–S–H gels was different in each case; when using the more con-centrated solution with larger nanoparticle size (B-40), a more sig-nificant decrease in this parameter was detected, while thisdiminution was not so evident for the solution with less concentra-tion of smaller silica nanoparticles (B-30). The decrease of theweight loss associated to the C–S–H gel content could be explainedby a different composition of these calcium silicate hydrates in thetreated mortars. However, a certain accelerated leaching of C–S–Hdue to the application of electrical field could also being contribut-ing [26,27], mainly during the treatment with B-40 where smallerefficiency of the nanosilica penetration is expected. Nevertheless,further studies are in way to identify the real contribution of theelectric field.

It should be remarked that the mortar densification wasdetected by MIP measurements at the age of 28 days after finishingthe treatment (Fig. 4), and thus, an effective reactivity of nanosilicaparticles can be deduced. A different reactivity of both solutionscan be expected probably due to the different size of the nanopar-ticles that may have different nucleating capacities and even pro-mote the formation of C–S–H gel with different compositions(that is, lower C/S ratio, larger polymerization degree, etc.).

The nanosilica penetration through the mortar due to the treat-ment was confirmed by BS-SEM coupled with EDX microanalysis,as represented in Fig. 6 for the case of treatment with B-40 at 12 V.

Spherical particles with high silica content were clearlyobserved in the first mm beyond of the mortar surface connectedduring the treatment to the negative side of the electrochemicalcell. The nanoparticle flocculation was expected to occur since

Fig. 6. Micrograph of treated sample at the age of 28 days after finishing thetreatment (B-40 at 12 V). Mortar exposed to the cathode (0.1 mm from surface).

310 M. Sánchez et al. / Construction and Building Materials 66 (2014) 306–312

the diameter was significantly larger (5 lm) than their initial meandiameter (7 nm). The chemical EDX microanalysis of these spheri-cal particles is included in Fig. 7 and compared with the analysis ofa region with C–S–H gel in the centre of the treated sample. Thepresence of such flocculated nanosilica particles found beyondthe treated mortar surface was the evidence confirming that nano-particles had certain penetrability into the treated sample and notonly the superficial deposition occurred. Besides it has to point outthat the use of silica aggregate could masked the identificationnanosilica particle.

The interaction of the nanosilica particles with the cementmatrix was qualitatively evaluated with the C/S ratio of the C–S–H gel zone at different regions of the transversal section of the trea-ted sample at the age of 28 days after finishing the treatment. TheC/S ratio of the C–S–H gel was also analysed in a 28 days reference

Fig. 7. EDX microanalysis in treated mortar at the age of 28 days after finishing the treatmin the central region of the treated mortar.

mortar (R0). In Fig. 8 the C/S ratio profile through the mortar trans-versal section is represented for treated samples as well as themean value obtained for the reference case.

The C/S profile in the transversal section of the treated samplesshowed an increasing tendency from the negative side (in contactwith the colloidal nanosilica) to the positive side (in contact withwater during the treatment); similar C/S ratios were found inde-pendently of the applied suspension (B-30 or B-40). The smallerC/S measured in the negative mortar surface should be related tothe formation of a Si enriched C–S–H gel probably due to the reac-tion of the additional silica particles with the C–S–H gel structureor even with the present portlandite forming C–S–H gels withlower C/S ratio by following a pozzolanic reaction. This could bea similar phenomenon than those detected in cement pastes withcolloidal nanosilica additions [21] or even, when using mineraladditions with high silica content as silica fume or fly ash [28]. Thiseffect was only detected until 1.5–2 mm from the mortar surfaceexposed to the colloidal nanosilica and must be indicative of thenanoparticle migration capacity. From the middle of the sample(around 5 mm depth) the typical ratio of a conventional mortar(1.5–1.8) was measured.

3.3. Treated mortar ageing

The evolution of the interaction between the nanoparticle andthe cement matrix due to the colloidal nanosilica reactivity wasassessed by MIP and the changes registered in the capillary poresare represented in Fig. 9, where different sealing ability of bothtreatments can be deduced. Using B-30 treatment the main reduc-tion in capillary pores was detected after 28 days of treatment; inthe case of the most concentrated solution (B-40), a gradual

ent (B-40 at 12 V). (a) Flocculated nanoparticles observed in Fig. 6 and (b) gel C–S–H

Fig. 8. C/S ratio of reference mortar and profile at different regions of the treatedsamples (28 days after finishing the treatment at 12 V).

Fig. 9. Evolution of capillary pore content at the ages of 28 and 90 after finishingthe treatment.

M. Sánchez et al. / Construction and Building Materials 66 (2014) 306–312 311

decrease of this capillary pores was observed, and an increasedefficiency could be deduced after 90 days of finishing the treat-ment. Then, a faster reactivity was expected for the most dilutedsolution containing smaller particles (B-30). However, the activityof the most concentrated solution could neither be neglected, as itsevolution with time was more significant even reaching higherlong-term efficiencies.

Although the paper is a preliminary attempt on migration ofcolloidal nanosilica, the results are promising and show clear evi-dences about the viability of drive the nanoparticles through thepores of a hardened cement matrix under an electrical field. Fur-thermore, the chemical interaction with the substrate is confirmedthat allow a sealing phenomenon to occur in the matrix of themortar.

4. Conclusions

� Colloidal nanosilica particles were successfully infiltrated inhardened mortar forced by migration under the action of anelectric field.� The nanoparticle size and the concentration were found to be

significant variables affecting the transport and sealing abilityof the treatment. However, no significant effect of the electricfield can be deduced.� The decrease of the capillary pore volume confirmed the sealing

ability of the treatment which evolves to larger efficiencies with

ageing. A chemical interaction of the nanosilica with the cementmatrix was found; nanosilica maintained its pozzolanic activityeven when applied to hardened concrete. The supplementarysilica incorporated by the treatment seems to react with theC–S–H gel of the matrix reducing the C/S silica ratio of thesegels.

Acknowledgements

Authors acknowledge ZEUS QUÍMICA S.A. the material supply.M. Sánchez also acknowledges the Spanish Ministry of Economyand Competitiveness (MINECO) her ‘‘Juan de la Cierva’’ postdoc-toral position (JCI-2011-09572).

References

[1] Said AM, Zeidan MS, Bassuoni MT, Tian Y. Properties of concrete incorporatingnano-silica. Constr Build Mater 2010;36:838–44.

[2] Singh LP, Karade SR, Bhattacharyya SK, Yousuf MM, Ahalawat S. Beneficial roleof nanosilica in cement based materials – a review. Constr Build Mater2013;47:1069–77.

[3] Shakhmenko G, Juhnevica I, Korjakins A. Influence of sol–gel nanosilica onhardening processes and physically–mechanical properties of cement paste.Proc Eng 2013;57:1013–21.

[4] Senff L, Labrincha JA, Ferreira VM, Hotza D, Repette WL. Effect of nano-silica onrheology and fresh properties of cement pastes and mortars. Constr BuildMater 2009;23:287–91.

[5] Technical Specification CEN/TS 14038-1:2004. Electrochemical realkalizationand chloride extraction treatments for reinforced concrete. Part 1:Realkalization, 2011.

[6] Technical Specification CEN/TS 14038-2:2011. Electrochemical realkalizationand chloride extraction treatments for reinforced concrete. Part 2: Chlorideextraction, 2011.

[7] Garcés P, Sánchez de Rojas MJ, Climent MA. Effect of the reinforcement bararrangement on the efficiency of electrochemical chloride removal techniqueapplied to reinforced concrete structures. Corros Sci 2006;48:531–45.

[8] Miranda JM, Cobo A, Otero E, González JA. Limitations and advantages ofelectrochemical chloride removal in corroded reinforced concrete structures.Cem Concr Res 2007;37:596–603.

[9] Tong YY, Bouteiller V, Marie-Victoire E, Joiret S. Efficiency investigations ofelectrochemical realkalisation treatment applied to carbonated reinforcedconcrete – part 1: SACRIFICIAL anode process. Cem Concr Res 2012;42:84–94.

[10] Ribeiro PHLC, Meira GR, Ferreira PRR, Perazzo N. Electrochemical realkalisationof carbonated concretes – influence of material characteristics and thickness ofconcrete reinforcement cover. Constr Build Mater 2013;40:280–90.

[11] Sánchez M, Alonso MC. Electrochemical chloride removal in reinforcedconcrete structures: improvement of effectiveness by simultaneousmigration of calcium nitrite. Constr Build Mater 2011;25:873–8.

[12] CSIC Patent 200803545. Dispositivo y procedimiento de uso para laeliminación de agentes agresivos y repasivación de la armadura de unhormigón armado con introducción de inhibidores aniónicos. Spain, 2008.

[13] Sawada S, Page CL, Page MM. Electrochemical injection of organic inhibitorsinto concrete. Corros Sci 2005;47:2063–78.

[14] Liu YJ, Shi XM. Electrochemical chloride extraction and electrochemicalinjection of corrosion inhibitors on concrete: state of the knowledge. CorrosRev 2009;27:53–81.

[15] Cárdenas HE, Struble LJ. Electrokinetic nanoparticle treatment of hardenedcement paste for reduction of permeability. J Mater Civil Eng 2006;18:554–60.

[16] Ryou JS, Otsuki N. Experimental study on repair of concrete structuralmembers by electrochemical method. Scr Mater 2005;52:1123–7.

[17] Kupwade-Patil K, Gordon K, Xu K, Moral O, Cárdenas H, Lee L. Corrosionmitigation in concrete beams using electrokinetic nanoparticle treatment. In:Limbachiya, Kew, editors. Excellence in concrete construction throughinnovation. London: Taylor & Francis Group; 2009. p. 365–71.

[18] Cárdenas H, Kupwade-Patil K, Eklund S. Corrosion mitigation in maturereinforced concrete using nanoscale pozzolan deposition. In: Zachar J, ClaisseP, Nailk TR, Ganjian E, editors. 2nd International conference on sustainableconstruction materials and technologies. Ancona: Università Politecnica delleMarche; 2010.

[19] Sánchez-Moreno M, Alonso MC, Díaz I, González R. Improvement of hardenedconcrete durability by nanosilica electromigration. In: de Belie N, van derZwaag S, Gruyaert T, van Tittelboom K, Debbaut B. (editors). Proceedings ofICSHM 2013 fourth international conference on self-healing material. Ghent:Magnel Laboratory for Concrete Research; 2013.

[20] Kontoleontos F, Tsakiridis PE, Marinos A, Kaloidas V, Katsioti M. Influence ofcolloidal nanosilica on ultrafine cement hydration: physicochemical andmicrostructural characterization. Constr Build Mater 2012;35:347–60.

[21] Madani H, Bagheri AL, Parhizkar T. The pozzolanic reactivity of monodispersednanosilica hydrosols and their influence on the hydration characteristics ofPortland cement. Cem Concr Res 2012;42:1563–70.

312 M. Sánchez et al. / Construction and Building Materials 66 (2014) 306–312

[22] Gaitero JJ, Campillo I, Guerrero A. Reduction of the calcium leaching rate of cementpaste by addition of silica nanoparticles. Cem Concr Res 2008;38:1112–8.

[23] Björnstroöm J, Martinelli A, Matic A, Börjesson L, Panas I. Accelerating effectsof colloidal nanosilica for beneficial calcium-silicate-hydrate formation incement. Chem Phys Lett 2004;392:242–8.

[24] http://www.dbbecker.com/images/stories/MarketingFolder_BrianMorris/DBB_Technical_DataSheets/EKA_NOBEL/PDS_BCC401.pdf.

[25] http://www.dbbecker.com/images/stories/MarketingFolder_BrianMorris/DBB_Technical_DataSheets/EKA_NOBEL/PDS_BCC301.pdf.

[26] Castellote M, Andrade C, Alonso MC. Changes in concrete pore size distributiondue to electrochemical chloride migration trials. ACI Mater J 1999;96:314–9.

[27] Castellote M, Andrade C, Alonso C, Turrillas X, Kvick A, Terry A, et al.Synchrotron radiation diffraction study of the microstructure changes incement paste due to accelerated leaching by application of electrical fields. JAm Ceram Soc 2002;85:631–5.

[28] García Calvo JL, Hidalgo A, Alonso C, Fernández Luco L. Development of low-pHcementitious materials for HLRW repositories. Cem Concr Res 2010;40:1290–7.