Evaluation of Corrosion Inhibition of NO2- Intercalated LDHs on Steel Coated by Cement Paste

WU You,XU Jinxia,MEI Youjin

(College of Mechanics and Materials,Hohai University,Nanjing 210098,China)

Abstract: The purpose of this study is to apply the steel coated by cement paste to evaluate the corrosion inhibition of NO2-intercalated Mg-Al layered double hydroxides(LDHs),which was prepared by a conventional calcination-rehydration method.The chloride equilibrium isotherm,open-circuit potential (OCP)and electrochemical impedance spectroscopy (EIS) of steel in the saturated Ca(OH)2 solution contaminated by chloride ions were measured.The microstructures of as-obtained LDHs and cement paste containing the LDHs were observed by scanning electron microscope (SEM),X-ray diffraction (XRD),Fourier transform infrared spectroscopy (FT-IR) and mercury intrusion porosimetry (MIP).It is found that the technique using the steel coated by the cement paste layer containing the LDHs to represent the steel in real concrete can be applied to accurately evaluate the corrosion inhibition performance of NO2- intercalated LDHs in a short duration.The addition of NO2- intercalated Mg-Al LDHs can improve the absorption capacity of Cl- and the anti-corrosion property of cement paste layer.The combination actions of the refinement of pore structure,uptake of chloride ions and release of inhibitive NO2- contribute to the better corrosion protection property of cement paste layer.Moreover,a schematic mechanism of chloride intrusion for the steel specimen with cement paste layer is developed.

Key words: layered double hydroxides;cement paste layer;chloride equilibrium isotherm;anticorrosion property

1 Introduction

The corrosion of reinforcing steel by chloride ions is one of the main reasons for the decrease in durability and service life of concrete structures,which leads to the great damage to the national economy and people’s lives.A series of methods have been suggested and applied to protect the steel from the corrosion.Especially,the additions of corrosion inhibitors in concrete have received more and more attentions due to the merits of simplicity and relatively lower cost.However,the function of traditional corrosion inhibitors is relatively single which usually restrains the electrochemical reactions of the reinforcing steel corrosion,and their inhibition effects are often not satisfactory.

Layered double hydroxides(LDHs),also called as anionic clays and hydrotalcite-like materials can be formulated aswhere Mare divalent metal cations and can be partially replaced by trivalent metal cations M,Ais the interlayer anion and

x

is the molar ratio of M/ (M+M).The periodical stacking of positively charged sheets are separated by interlayer solvated anions which are weakly bonded with each other by hydrogen bond,and hence the anions are liable to be exchanged with other kinds of anions that are more easily intercalated into the interlayer.Owing to high anion-exchange capacity,LDHs can be used as an efficient adsorbent for removal of chloride ions from contaminated water.In addition,LDHs is considered to be added into the concrete as functional materials to increase the chloride binding capacity,and hence delay the transport of chloride ions.This ideal can be supported by the beneficial effect of hydrotalcite-like phases such as Friedel’s salt (3CaO·AlO·CaCl·10HO) on the binding of chloride ions in cement concrete.

Furthermore,LDHs materials may be applied as multifunctional corrosion inhibitor as far as the anions intercalated in the interlayer exhibit an inhibitive ability.Nitrite is one of the most common corrosion inhibitors,which is suitable for this application.When the LDHs intercalated with NOanions contacts with the chloride ions in the corrosion medium,it can not only uptake Cl,but also release the NOanions,which can arrive at the surface of the steel by diffusion and generate a passive film by redox reaction.As a result,an excellent inhibition effect of LDHs intercalated with NOanions by the dual actions can be anticipated.It is significant for us to investigate the protection of steel reinforcement from chloride-induced corrosion by the LDHs intercalated with NOanions for the application in real concrete structure exposed to chloride-laden environment.

To this date,a few experimental works have been attempted on the application of LDHs intercalated with NOanions in the protection of steel reinforcement from chloride-induced corrosion.Despite this,the investigation is not yet enough for the real application.For evaluating the inhibition effect of LDHs intercalated with NOanions,the experiments are ordinarily carried out in concrete or an alkaline solution,simulating the pore solution of concrete.For the concrete,too long duration for the onset of steel corrosion by chloride diffusion into the concrete is unacceptable for the investigator in the lab.In contrast,the experimental period can be remarkably shortened in the simulated concrete pore solution.However,the loss of the interface between the concrete and steel reinforcement makes it difficult to evaluate accurately the real condition in the reinforced concrete structure.Incorrect evaluation on the inhibition effect of corrosion inhibitor often happens.Especially for the LDHs intercalated with NOanions,the inhomogeneous dispersion and sedimentation of its solid particles further enhance the risk of incorrect evaluation.Similarly,incorrect evaluation may exist in the concrete due to the effect of complex concrete cover on the experimental result.

Based on this,a new technique using the steel coated by the cement paste layer to represent the steel reinforcement in real concrete was developed to evaluate the corrosion inhibition performance of NOintercalated Mg-Al LDHs,which was prepared by conventional calcination-rehydration method.The chloride equilibrium isotherm,open-circuit potential(OCP) and polarization resistance of steel decided by electrochemical impedance spectroscopy (EIS) in the saturated Ca(OH)solution contaminated by chloride ions were measured.Moreover,the microstructures of the as-obtained LDHs and its cement paste were observed by scanning electron microscope(SEM),X-ray diffraction (XRD),Fourier transform infrared spectroscopy (FT-IR) and mercury intrusion porosimetry (MIP).

2 Experimental

2.1 Materials and specimen preparation

The cement used was P·O 42.5 Portland cement purchased from Hai Luo Company and its chemical component was listed in Table 1.MgAl(CO)(OH)·4HO,NaNO,Ca(OH)and NaCl were analytically pure.The type of rebar used in the experiment was HPB235.Boiled deionized water was adopted to prepare aqueous solutions.

The intercalation of NOinto the interlayer of the Mg-Al-LDHs was performed by using the calcinationrehydration method.MgAl(CO)(OH)·4HO powder was heated for 5 h at 500 ℃ and then was cooled down to get the Mg-Al oxide(LDO),which was used as the precursor for the intercalation.4.14 g NaNOand 6.82 g LDO (the molar ratio was 3:1) was added into the flask with 100 mL secondary boiling deionized water (the solid-liquid ratio was around 1:10) in Nflow.The mixture was vigorously stirred by magnetic force at the temperature of 90 ℃ for 5 h in a water bath.The resultant suspension was separated by centrifuge and the collected powder was washed for several times,and then the powder was dried out at 60 ℃ for 24 h under vacuum.

2.2 Chloride equilibrium isotherm

Saturated Ca(OH)solution was applied to simulate the pore solution of concrete.Different concentrations of sodium chloride (0.005,0.01,0.05,0.1,0.3,0.6 and 1 mmol/L) were added into the solution to perform the measurement of chloride equilibrium isotherm.For the measurement,cementpastes with 0 wt%,2 wt% and 4 wt% Mg-Al-NOLDHs by cement mass (labelled by 0%-CP,2%-CP and 4%-CP,respectively) were prepared.Firstly,the Mg-Al-NOLDHs was added into the water in proportion and ultrasonic dispersion was carried out for half an hour.After this,the mixture was added into the cement in proportion and stirred with a blender for 3 min to obtain the CP with different contents of Mg-Al-NOLDHs.

The cement pastes had the cement/binder ratio of 0.5.After curing for 28 d under standard condition,the pastes were crushed to obtain particles with the diameters of 0.2-2 mm.Subsequently,the collected particles were dried out at 60 ℃for 24 h under vacuum.After this,25 g cement particles were added into the solutions with the solid/solution ratio of 1:2.The mixtures were sealed up and kept at room temperature for 10 d so that the chloride absorption equilibrium was attained.Afterwards,the free chloride concentrations in the filtrates were analyzed by means of potentiometric titration using AgNOThe concentration of the bound chloride ions (

C

(mg/g)) loaded on the CP after the absorption equilibrium was calculated by the following equation:

where

V

was the volume of solution (L),

C

was the initial chloride concentration (mmol/L),

C

was the free chloride concentration(mmol/L),and

W

was the mass of cement particle (g).

2.3 Corrosion measurement

The steel specimens for corrosion measurements were cut with the dimensions

φ

10 mm ×5 mm.In order to have an electrical connection,an isolated copper wire was attached to one end of each specimen.Another end surface of each specimen was exposed and the remaining area was sealed by means of epoxy resin.The exposed surface was polished using a series of silicon carbide emery papers of grades 400,800 and 1 000,degreased in acetone and then washed in distilled water.Afterwards,0%-CP,2%-CP and 4%-CP were coated on the exposed surfaces of the steel specimens.The cement layer had the thickness of 3 mm.Such a thin cement layer allowed chloride ions to intrude quickly and corrode the steel bar within a short period.Then,the coated steel specimens were cured for 28 d under standard condition (20±2 ℃、95%RH).For obtaining reproducible results,all the experiments were repeated three times.

Open-circuit potential (OCP) and electrochemical impedance spectroscopy (EIS) were applied to evaluate the inhibition effect of LDHs intercalated with NOanions.The used instrument was Autolab PARSTAT 2273 potentiostat.The coated steel specimen was used as working electrode,platinum electrode was used as counter electrode,and the saturated calomel electrode was used as reference electrode,respectively.The electrolytic solution was the chloride contaminated saturated Ca(OH)solution.The aggressive chloride ions were added once a day and the concentration was increased by 0.01 mol/L at a time.The electrochemical measurements were tested before the addition of aggressive chloride ions every time.The EIS measurements were performed within the frequency range of 100 kHz to 10 mHz by using a 20 mV amplitude sinusoidal voltage at the open-circuit potential (OCP).The obtained data were analyzed using the ZsimpWin software.

2.4 Microstructure characterization

The Cambridge scanning electron microscope(model Cambridge-360) operated at 5 kV with a working distance of 4.5 mm was used.D/Max-RB X-ray powder diffractometer with Cu kα radiation(40 kV and 30 mV) at a scanning rate of 10°/ min was applied to obtain the X-ray diffraction patterns of fabricated LDHs samples and the CP with Mg-Al-NOLDHs before and after the chloride absorption equilibrium.Fourier transform infrared (FT-IR) spectra in the range 4 000-400 cmof LDHs samples before and after the chloride absorption as KBr pellets were recorded with a Nicolet IS10 spectrometer.Moreover,the pore structures of the cement pastes were measured using a Poremaster GT-60 MIP (Quantachrome) with a maximum mercury intrusion pressure of 210 MPa,mercury surface tension of 0.48 N/m and a contact angle of 140°.

3 Results and discussion

3.1 Characterization of as-obtained LDH

Fig.1 indicates the SEM images of as-obtained MgAl-COLDHs,LDO and MgAl-NOLDHs.From Fig.1(a),MgAl-COLDHs presents the characteristics of hexagonal platy morphology and the distribution of size ranges from 300 nm to 600 nm.After calcination,the lamellar structure presents obvious agglomeration into long cylinders and the width size distribution is obviously decreased,ranging from 100 to 200 nm (Fig.1(b)).After rehydration,MgAl-NOLDHs presents the similar lamellar structure and size distribution compared to MgAl-COLDHs (Fig.1(c))，which indicates the recovery of the layer structure.

Fig.1 SEM images of as-obtained (a)MgAl-CO3 LDHs,(b) LDO,(c)MgAl-NO2 LDHs

Fig.2 XRD patterns of Mg-Al-CO3 LDHs,LDO and Mg-Al-NO2 LDHs

Fig.2 indicates XRD patterns of Mg-Al-COLDHs,LDO and Mg-Al-NOLDHs.The peaks in the XRD pattern of Mg-Al-NOLDHs are similar to those in the XRD pattern of Mg-Al-COLDHs.In contrast,the primary peaks disappear for the calcinated product LDO.Compared to that of Mg-Al-COLDHs,the(003) diffraction peak of Mg-Al-NOLDHs moves to the smaller 2

θ

angle of 11.22 from 11.58.The corresponding basal spacing is increased from 0.76 to 0.78 nm,which indicates that the NOhas been successfully intercalated into the interlayer.In addition,the reconstructed layered structure happens.

Fig.3 FT-IR spectra of Mg-Al-CO3 LDHs,LDO and Mg-Al-NO2 LDHs

Fig.3 indicates the FT-IR spectra of Mg-Al-COLDHs,LDO and Mg-Al-NOLDHs.After calcination,the intense broad band from 4 000 to 2 700 cmassociated with the stretching vibrations of the hydrogen-bonded hydroxyl group of both the hydroxide and interlayer water weakens obviously compared to Mg-Al-COLDHs.Meanwhile,the band for COanions at 1 117 cmdisappears,indicating the collapse of the layered structure.In case of Mg-Al-NOLDHs,there is the band for NOanions at 1 263 cm,indicating the appearance of the intercalation of NO-.However,the stretching vibration of nitrate group at 1 372 cmis also indicated,which may attribute to the oxidizing of the small portion of nitrite by mixed air during the process of intercalation.

3.2 Chloride absorption equilibrium isotherms

Fig.4 indicates equilibrium isotherms of chloride absorption on 0%-CP,2%-CP and 4%-CP.There is an obvious nonlinear relationship between the chloride absorption loading and chloride concentration at equilibrium.Two nonlinear equations are applied to fit the experimental data as follows:

where

α

and

β

are the constants of Freundlich,γ and ε are the constants of Langmuir.All the values of the constants can be obtained by the isotherm fitting.The values of the constants and determination coefficients(

R

) are listed in Table 2.Compared to Langmuir isotherm,the fitting results by Freundlich isotherm have higher coefficients of determination for all the CPs.Besides,with the increase of Mg-Al-NOLDHs content,the

R

increases obviously.Hence,Freundlich isotherm provides a better fit to the experimental data than Langmuir isotherm.

Table 2 Fitting parameters for the chloride absorption isotherms of CP with Mg-Al-NO LDHs

Fig.4 Freundich (a) and Langmuir (b) equilibrium isotherms of chloride absorption on the 0%-CP,2%-CP and 4%-CP

Furthermore,the chloride absorption amount of 0%-CP,2%-CP and 4%-CP are respectively 6.487,6.743 and 7.871 mg/g,respectively,when the added chloride concentration is 1 mol/L.The amount of chloride absorption is increased by 3.9% and 21.3%with the additions of 2% and 4% Mg-Al-NOLDHs compared to pure CP,respectively.It can be found that the chloride absorption capacity is increased nonproportionally with the addition of Mg-Al-NOLDHs.The reason may be attributed to the complicated chloride binding of cement paste with Mg-Al-NOLDHs,which can be affected by many factors,including the NOreleased from the Mg-Al-NOLDHs.This non-proportional growth indicates that the addition of Mg-Al-NOLDHs in CP can effectively increase the chloride absorption capacity.

3.3 Anti-corrosion performance

3.3.1 Open-circuit potential (OCP) measurement

Fig.5 The evolutions of corrosion potentials of steel specimens with the 0%-CP,2%-CP and 4%-CP layers

Fig.5 shows the evolutions of corrosion potentials for the steel specimens with 0%-CP,2%-CP and 4%-CP layers.In general,the corrosion potentials are gradually decreased with the increase of immersion time.The shift of potential to further negative values indicates the increased risk of corrosion by the diffusion of aggressive chloride ions into the CP layer.Notably,the corrosion potentials of steel specimens with three kinds of CP layers reveal a similar value in initial few days.Subsequent to this,sharp decreases for all the evolutions of the corrosion potentials of steel specimens with CP layers appear.The sharp decrease should be attributed to the rupture of passive film on the steel surface and the initiation of active corrosion.Furthermore,the sharp decrease for the evolution of the corrosion potential of steel specimen with 0%-CP layer happens at the earliest time.With the addition of Mg-Al-NOLDHs,the time for the sharp decrease of the potential value clearly moves backwards.Such a result indicates that a longer duration is required to initiate the corrosion of steel specimens with 2%-CP and 4%-CP layers than that with 0%-CP layer.The corrosion of steel specimen with 4%-CP layer happens at the latest.As a result,that the addition of Mg-Al-NOLDHs can improve the anti-corrosion property of CP layer and the ultimate effect is tightly associated with the content of LDHs.

3.3.2 Electrochemical impedance spectroscopy (EIS)measurement

Fig.6 shows the Nyquist and bode impedance plots for the steel specimens covered by 0%-CP,2%-CP,4%-CP layers with different immersion times.At the beginning of immersion,the Nyquist plots of all specimens exhibit a pure capacitive loop and similar radius value.With the increase of immersion time,all the radius value of capacitive loop gradually decreases,which may be due to the intrusion of aggressive chloride ions.This chloride intrusion induces instability of the passive film on the steel surface.Furthermore,a straight line on behalf of Warburg component begins to emerge within the low frequencies of the Nyquist plots when the immersion time is increased to a certain value.In addition,the moments for the appearances of Warburg component for all the steel specimens have a sequence of 0%-CP＜2%-CP＜4%-CP.

Fig.6 Nyquist and bode impedance plots for the steel specimens with the 0%-CP,2%-CP,4%-CP layer.The added concentration of NaCl in solution and immersion time are gradually increased,1 d (a-1,2),6 d (b-1,2),8 d (c-1,2),10 d (d-1,2),12 d(e-1,2),15 d (f-1,2)

It is well known that the Warburg component represents the constant corrosion controlled by the mass transfer.Obviously,the aggressive chloride ions can penetrate through the CP layer due to the porous structure.As far as the invaded chloride ion attains a critical level,the corrosion of steel will be initiated.When the corrosion of steel happens,the corrosion product will be released into the pore solution by diffusion.In case the rate of production is faster than the diffusion,the corrosion product could be deposited on the surface of the steel and make it hard to strike which could prevent the further development of corrosion.Hence,the diffusion rate of corrosion product decides the whole corrosion rate of the system in this period.As a result,the appearance of Warburg component within the low frequencies of the Nyquist plots can be anticipated.The sequential order of moments for the appearances of Warburg component for all the steel specimens indicates that the addition of Mg-Al-NOLDHs can improve the anti-corrosion property of CP layer.

The initial shapes of the phase angle plots for various steel specimens with the CP layers suggest the presence of two time constants partially superimposed.Considering the main characteristic features of the CP layer,the equivalent-circuit models (a and b) in Fig.7 are applied to analyze the EIS results of steel specimens with different immersion times.In the model a,

R

is the solution resistance,

Q

the constant phase element representing layer capacitance and

R

the layer resistance.

Q

and

R

are the double-layer capacitance of reinforcement and charge-transfer resistance,respectively.Their values are obtained by using a constant phase element (CPE),which represents a nonideal frequency dependent capacitance.The CPE is given by the equation:

where parameter

n

generally ranges between 1 and 0.5.It describes the distribution of the dielectric relaxation time in the frequency domain.For

n

=0,the CPE is a resistor;for

n

=1,the CPE is a capacitor.The CPE behavior can be the consequence of the fractal nature of the electrode interface or heterogeneity of the steelsurface.The model a is to fit the EIS results in the initial immersion time.However,as soon as the straight line on behalf of Warburg component appears in the Nyquist plots,the model b with Warburg impedance(

Z

) is used.

Table 3 Fitting parameters with equivalent-circuit models (a and b) immersion times

Fig.7 Equivalent-circuit models applied to analyze the EIS results of steel specimens in different immersion times

The obtained fit parameters for the EIS results in Fig.6 are listed in Tables 3.The result indicates that the resistance of CP layer is gradually decreased with the increase of immersion time.The decrease of the resistance should be attributed to the intrusion of chloride ions into the CP layer.Besides,with the same immersion time,the resistance of 4%-CP is always higher than those of 0%-CP and 2%-CP,indicating that the addition of 4 wt% LDHs could obviously slow down the intrusion of chloride ion into the CP layer.

Fig.8 Changes of charge-transfer resistance with immersion time for the steel specimens with the 0%-CP,2%-CP and 4%-CP layers

The changes of charge-transfer resistance with the increase of immersion time for the steel specimens with 0%-CP,2%-CP and 4%-CP layers are pictured in Fig.8.From Fig.8,the charge-transfer resistances for all the steel specimens do not display an obvious change in the initial few days.Subsequent to this,sharp drops of the charge-transfer resistances appear,which should be attributed to the initiation of steel active corrosion.With the increase of the LDHs content,the initiation of active corrosion is delayed.Besides,the corresponding moment for the corrosion initiation is well identical to that determined by the OCP measurement.All the results further prove that the addition of Mg-Al-NOLDHs can improve the anti-corrosion property of CP layer.Also,the best effect for the 4% CP layer has been obtained.

3.4 Anti-corrosion mechanism

3.4.1 Mercury intrusion porosimetry (MIP)

The mercury pressure curves (see Fig.9) show the pore size distributions of pure CP and 4%-CP.With the addition of 4% Mg-Al-NOLDHs,the number of the harmful large size pores(＞100 nm) is obviously decreased and the number of the tiny pores (＜100 nm)is obviously increased.The proportions of the large size pores and tiny pores are 40.75% and 59.25% for 4%-CP,respectively.In contrast,their values are 44.25%and 54.75% for pure CP,respectively.In addition,the porosities of pure CP and 4%-CP are 11.5% and 9.23%,respectively.As a result,a refined pore structure of sample has been obtained due to the addition of 4%LHDs.That is,LDHs material can effectively improve the pore structure.Obviously,the refined pore structure due to the addition of LDHs is beneficial to prevent the penetration of chloride ions.

Fig.9 Mercury pressure curves for the pure CP and 4%-CP layers

3.4.2 XRD analysis

Fig.10 XRD patterns of pure CP and 4%-CP before and after the chloride absorption

Fig.10 exhibits the XRD patterns of pure CP and 4%-CP before and after the chloride absorption.Compared to pure CP,the XRD pattern of 4%-CP before the chloride absorption appears a little sharp peak at the 2

θ

angle of 11.2,which is the characteristic diffraction peak (003) of the Mg-Al-NOLDHs.After the chloride absorption,there is a diffraction peak at the 2

θ

angle of 11.5,which belongs to the LDHs intercalated with Cl.This result indicates that the NOanions have been exchanged by Clanions in the interlayer of LDHs.Besides,the peak intensity becomes stronger.The reason may be attributed to the additional contribution of the diffraction of Friedel’s salt.

3.4.3 FT-IR analysis

Fig.11 FT-IR spectroscopes of pure CP and 4%-CP before and after the chloride absorption

Fig.11 exhibits the FT-IR spectra of pure CP and 4%-CP before and after the chloride absorption.Compared to pure CP,there is an additional weak band at 1 270 cmin the spectra of the 4%-CP before the chloride absorption.This band is related to NOanions in the interlayer of LDHs.After the chloride absorption,the band for NOis still survived,but its intensity becomes smaller.This result indicates the occurrence of chloride ions uptake by the exchange of anion in the interlayer of NOintercalated LDHs.In addition,a series of bands between 1 550 cmand 1 350 cmwhich are related to Friedel’s salt appear in the spectra of 4%-CP after the chloride absorption.

3.4.4 Schematic mechanism of chloride intrusion

According to all the above results,the addition of Mg-Al-NOLDHs can not only make up for the porous imperfection of the CP layer to some extent (called as the refinement of pore structure),but also can release NOions while absorbing chloride ions.The released NOions will migrate towards the steel surface by the diffusion.As soon as the NOions arrive at the steel,it is anticipated that they will react with the corrosion products of steel to repair the passive film and finally prevent further corrosion.The working mechanism is listed as the following chemical formula according to the literature:

Fig.12 The schematic representation of CP layer applied to interpret its protection from chloride-induced corrosion

Accordingly,it is concluded that the combination actions of the refinement of pore structure,uptake of chloride ions and release of inhibitive NOare contributed to the better corrosion protection property of CP layer.

Fig.12 represents CP layer with different contents of Mg-Al-NOLDHs.The addition of LHDs can improve significantly the pore distribution.Besides,a large part of the LDHs material can contact with corrosion medium with the increase of immersion time.It can absorb Clmeanwhile release NOwhich can partly arrive at the surface of the reinforcement and protect it.The ultimate inhibition effect also depends on the content of its addition into the cement paste.

It can be anticipated that the MgAl-NOLDHs has a better inhibition effect than the NOsince it can release the inhibitive NOto act as inhibitor besides the uptake of chloride ions and refinement of pore structure.In addition,MgAl-NOLDHs will also exhibit a better inhibition effect than MgAl-NOLDHs due to the better inhibition effect of NOthan NOin the interlayer.Such a result has been demonstrated by the investigation in the simulated concrete pore solution.

It should be pointed out that the results in this study are in well accordance with those in the literatures,which further demonstrates the better corrosion inhibition of MgAl-NOLDHs.In addition,this accordance indicates the feasibility of the developed technique using the steel coated by the cement paste layer containing the LDHs to represent the steel in real concrete.Despite this,the advantages of the developed technique can be clearly determined.The experimental period (only 20 d is required to initiate the steel corrosion in this study) can be remarkably shortened than that in the real concrete.Also,more correct evaluation can be obtained than that in the simulated concrete pore solution since the interface between the concrete and steel reinforcement is kept and the effect of MgAl-NOLDHs on the pore distribution of interface is considered.

4 Conclusions

a) The technique using the steel coated by the cement paste layer containing the LDHs to represent the steel in real concrete can be applied to accurately evaluate the corrosion inhibition performance of NOintercalated LDHs in a short duration.

b) Mg-Al-LDHs intercalated with inhibitive NOanionshas been successfully synthesized by usingthcalcination-rehydrationmethod.The fabricated NOintercalated Mg-Al-LDHshasthe(003)basalspacing of0.78 nm,whichiscloseto thatofCOintercalated Mg-Al-LDHs.There is the band for NOanions at 1 263 cmin the FT-IR spectra of Mg-Al-NOLDHs.

c) The addition of Mg-Al-NOLDHs in cement paste can significantly improve the absorption capacity to Cl.The adsorption capacities of 2%-CP and 4%-CP are respectively 6.743 and 7.871 mg/g when the added chloride concentration is 1 mol/L,which are increased by 3.9% and 21.3% relative to the 0%-CP,respectively.Freundlich isotherm provides a better fit to the experimental data than Langmuir isotherm.

d) The corrosion potentials and charge-transfer resistances for all the steel specimens with the 0%-CP,2%-CP and 4%-CP layers do not display an obvious change in the initial few days.Subsequent to this,they display a sudden change.The moments for the sudden change have a sequence of 0%-CP＜2%-CP＜4%-CP.Accordingly,the addition of Mg-Al-NOLDHs can improve the anti-corrosion property of CP layer.Also,the best effect for the 4%-CP has been obtained.

e) The addition of Mg-Al-NOLDHs can significantly improve the pore distribution of cement paste.The large size pores decrease by 4.5% and the tiny pores increase by 3.5% after the addition of 4%LDHs into cement paste.Besides,the total porosity decreases by 2.27%.