AgCl

Electrochemistry The Electrochemical Society of Japan Article

Received: August 6, 2014 Accepted: September 5, 2014 Published: December 5, 2014 http://dx.doi.org/10.5796/electrochemistry.82.1040

Electrochemistry, 82(12), 1040–1046 (2014)

Electrochemical Characterization of a Solid Embeddable Ag/AgCl Reference Electrode for Corrosion Monitoring in Reinforced Concrete Ming JIN, Jinxia XU, Linhua JIANG,* Guofu GAO, Hongqiang CHU, Chuansheng XIONG, Hailang GAO, and Peng JIANG College of Mechanics and Materials, Hohai University, Nanjing 210098, PR China * Corresponding author: [email protected] ABSTRACT In this study, a new solid embeddable Ag/AgCl reference electrode (RE) was electrochemically investigated for corrosion monitoring in reinforced concrete. The electrode was fabricated by applying the cement mortar as the bottom layer, and the polymer gel as electrolyte. Several electrochemical methods were applied. The experiments were carried out in simulated concrete solutions and concrete. Furthermore, an exposure period over 1 year was used to evaluate the stability of the electrodes embedded in concrete. It has been found that the electrode potential was not obviously affected by the chloride concentration, but affected by the pH value of solution. The Ag/AgCl RE had uniformity in concrete environments. The reversibility of Ag/AgCl RE in all solutions is only within +4 mV. Besides, potentiodynamic tests revealed that the Ag/AgCl RE had a high exchange current density, which was advantageous when using Ag/AgCl RE to conduct resistance tests. Galvanostatic application of 0.2 µA/cm2 caused little variation of potential with time (~30 mV in several days), indicating the presence of a finite polarization resistance. As a result, the new solid Ag/AgCl RE was suitable as a sensor electrode for concrete structures. © The Electrochemical Society of Japan, All rights reserved.

Keywords : Reference Electrode, Ag/AgCl, Electrochemical Tests, Steel Corrosion 1. Introduction Nowdays reinforcement concrete is the most widely used material in civil engineering on the earth, with consumption above dozens of billions of tons. The concrete industry has become the basis of the modern society. However, concrete is not free of severe degradation problems such as reinforcement corrosion, concrete physical deterioration (cracking, frost, fire, etc.) and chemical deterioration (sulphate attack, acid attack, sea weather attack, alkaliaggregate reaction, leaching, etc.).1,2 Among them, corrosion of rebars in reinforcement concrete is viewed as a major problem in the maintenance of the structural integrity of structures. Steel reinforcements embedded in reinforced or pre-stressed concrete bridges and structures need continuous monitoring because of their susceptibility to corrosion. Since the corrosion of reinforcement rebars in concrete is electrochemical in nature, therefore, various electrochemical techniques are being adopted to monitor reinforcement corrosion, such as half-cell potential, linear polarization and electrochemical impedance spectroscopy.3 In all these cases continuous monitoring of the half-cell potentials of embedded steel becomes a prerequisite. Such a situation calls for a stable and rugged reference electrode (RE), which can be permanently embedded in concrete as close as to the embedded steel so as to generate reliable data. Monitoring and control of corrosion of reinforcement steel in concrete requires reliable measurement of electrode potentials.4 Use of reference electrode in concrete presents several challenges. Conventional electrodes such as the glass membrane electrodes are not fit for in situ measurements at the steel/concrete interface due to the very high alkaline environment at the interface and their fragility. The standard hydrogen electrode (SHE) is the most accurate reference electrode and is reference to the other reference electrodes. However, the SHE is not convenient to use, and saturated calomel electrode is usually used in laboratory.5 But the calomel electrodes and other mercury-mercury electrodes are formed from mercury, therefore any leakage can result in pollution. A good embeddable reference electrode must have several 1040

conditions: stable, invariant to chemical and thermal changes, capable of passing small currents with a minimum of polarization and hysteresis effects and so on. Reference electrodes used in concrete structures are described by several authors.6,7 The ASTM C876 specifies the use of Cu/CuSO4 as reference electrode for measurement of rebar potential in concrete.8 However, the leakage of copper sulphate solution can cause contamination of concrete, and the IR drop within the concrete cover can also lead to erroneous consequence. Duffo and Farina9,10 have used the metal-metal oxide (MMO) activated titanium rod (ATR) as the reference electrode in concrete, nevertheless, the potential of this reference electrode depended on the pH and became less reproducible after almost 1 year in mortar. Castro10 and Duffo11 both reported that the graphite electrodes are sensitive to oxygen partial pressure. Muralidharan3 chose MnO2 as a sensor material due to the stability of MnO2 in alkaline environment and reported that MnO2 reference electrode shows perfect uniformity, good reliability as well as good reversibility in simulated concrete pore solutions. But the MnO2 sensor performance in concrete need further testing. For many years, the study of silver/silver chloride (Ag/AgCl) electrode has been a hot topic. Gurusamy12 investigated the longterm performance of Ag/AgCl electrodes in concrete over a period of 4 years and reported a mean potential difference of 30–35 mV with respect to saturated calomel electrode (SCE) in 3% CaCl2 contaminated concrete. Climent-Llorca13 reported the possibility of using Ag/AgCl wire electrodes as in situ sensors of chloride concentration in concrete, which was studied by embedding them in a series of mortar specimens with different admixed sodium chloride (NaCl) contents. They showed a sensitive potentiometric response to Cl¹ concentrations. Besides, the Ag/AgCl reference electrode (RE) had internal potassium chloride (KCl) solution, which the Ag/AgCl electrode was embedded in to get a stable potential. However, traditional Ag/AgCl RE used the porous ceramic as the permeating layer, which had a drawback of continuous exudation of the KCl internal solution. The exudation of KCl is one of the most important parameters for reliable performance of traditional Ag/AgCl

Electrochemistry, 82(12), 1040–1046 (2014) reference electrode.14 What’s more, Brezinski15 pointed out that AgCl clogged the liquid junction in a Ag/AgCl reference electrode. Satoshi16 found that chelating resins could eliminate Ag ions existing in the KCl inner solution. Huang17 introduced a novel agarose-stabilized KCl-gel membrane to serve as a polymersupported solid reference electrolyte. Glanc18 studied the performance of fabrication parameter variations of thick-film screen printed Ag/AgCl RE. But these processes are too much complex. It can be clearly indicated from the foregoing that there is an utmost need to develop the reliable, easily manufactured and low-cost Ag/AgCl RE for steel in concrete. Such embeddable electrodes should have the following features:6,19 it should be rugged and maintenance-free, and have desirable micro-reversibility, negligible liquid junction potential, adequate stability and reliability when compared to a standard reference electrode such as SCE. Besides, it should be capable of being embedded under any orientation, and should not lead to undersirable contamination of surrounding concrete. The aim of the present study was to fabricate a new Ag/AgCl RE which possessed KCl inner solution mixed with water retention polymer. The water retention polymer could effectively prevent the leakage of inner solution and the dissolution of AgCl coating. In addition, we used a large water-cement ratio mortar instead of the porous ceramic as the bottom layer. Such a bottom layer has a pH value of 13.0, which corresponds to that of pore solution in concrete. So, the electrode can keep a good balance with the surrounding environment of concrete. This investigation addressed performance of the material for short- and long-term absolute potential determination, with emphasis on the open-circuit behavior of the electrodes. The polarizability of the Ag/AgCl RE, the effect of sodium chloride (NaCl) additions, and the temperature also were studied. 2. Materials and Methods 2.1 Fabrication of Ag/AgCl reference electrode Silver rod with 99.98% purity was used. All the silver segments, which were cut from one silver rod, had the same length of 8.5 cm. The anodic polarization with constant current density was applied to produce silver chloride on the surface of silver segment. The configuration of Ag/AgCl RE is shown in Fig. 1. Ag/AgCl RE was fabricated in the laboratory as follows: Step-1: First, a hole with the depth of 1 mm was drilled at one end of each silver segment. After this, a copper wire was wrapped inside the hole. Subsequently, the hole to the end of segment was sealed with silica gel, leaving 8 cm long silver segment exposed. Exposed area was about 7.536 cm2.

Cu wire Silica gel

Step-2: Prior to the anodizing process, the electrode was immersed in acetone for 5 minutes to remove surface oil. Then it was immersed in 5% mass fraction nitric acid (HNO3) for 1 minute and immersed in absolute ethyl alcohol with ultrasonic oscillation for 10 minutes; at last, the electrode was rinsed with distilled water. Then the electrode was anodized galvanostatically with a current density of 0.1 mA/cm2 in 0.1 M hydrochloric acid (HCl) for 1 hour. After anodizing, the electrode was kept in 0.1 M KCl solution out of direct sunlight with a small amount of AgCl powder. Step-3: 3.725 g KCl was added in 500 ml double-distilled water to get 0.1 M KCl solution, then the solution was heated to 80 degree. After that, the solution was changed into gel-solution with addition of 50 g carboxymethyl hydroxyethyl cellulose in that degree. Step-4: A rugged 10 cm long PVC pipe was used as the shell of the reference electrode, a 1 cm thick cement paste layer was placed in the bottom of the pipe. Then the KCl gel-solution was poured into the pipe in which the Ag/AgCl electrode was inserted. Last but not least, the top of PVC pipe was sealed with silica gel. 2.2 Tests in aqueous solutions All reagents used in our work were analytical grade. For the electrochemical measurements, the fabricated Ag/AgCl RE, cylindrical platinum foil and SCE were connected to serve as working electrode, counter electrode and as reference electrode, respectively. All the electrochemical tests were performed with an Princeton Applied Research (PAR) STAT 2273 Potentiostat. If not specified, all the potentials were monitored using a SCE as reference and all tests were carried out at room temperature (25°C). Commercial Ag/AgCl reference electrode (INESA Scientific InstrumentCo., Ltd., China). 2.2.1 Open-circuit potential measurements 2.2.1.1 Stability test The pH value of a fresh concrete is 13.5–12.5, approximately. Along with the carbonation process, the pH value of the concrete pore solution drops, however, the pH value of concrete pore solution can be not lower than 7. In our work, aqueous buffer solutions whose pH values vary from 13.5 to 6.86 were used. The solutions abbreviations, compositions and pH values are given in Table 1. The solution E, the so-called simulated concrete pore solution (SPS), and solution D, the saturated Ca(OH)2 solution (SCS) accounted for the pH of a fresh concrete. Solution C had a lower pH value and was chosen to account for a carbonated concrete; while solution B and A were chosen to investigate the effect of lower pH values. What’s more, the effect of pH values on the potential of Ag/AgCl RE was investigated. When the potential of Ag/AgCl RE was stable in SCS, we did the test in other solutions. The potential of the Ag/AgCl RE was investigated for about five weeks. Moreover, we test the potential evolution of commercial reference electrodes in the five solutions for five weeks too. 2.2.1.2 Reliability test A newly fabricated Ag/AgCl RE was placed in the saturated Ca(OH)2 solution. Inserted another Ag/AgCl RE in the same test solution. Besides, the potential of commercial Ag/AgCl reference electrodes was also investigated for about five weeks in SCS.

PVC pipe Table 1. Composition and pH of the solution used.

KCl gel Silver electrode Epoxy resin

Cement layer Figure 1. A structure of Ag/AgCl reference electrode.

Solution A B C D E

Composition

pH

0.025 M mixed phosphate 0.01 M Na2B4O7 0.025 M NaHCO3 + 0.025 M Na2CO3 saturated Ca(OH)2 solution 8 g NaOH + 35.6 g KOH per liter of saturated Ca(OH)2 solution

6.86 9.18 10 12.5 13.5

1041

Electrochemistry, 82(12), 1040–1046 (2014) 2.2.1.3 Effect of chloride concentration test The chloride ion content of fresh concrete is very little, but the content will increase rapidly such as using de-icing salt in winter. In the present work, different amounts of chloride ions were added into the saturated Ca(OH)2 solutions. The concentration of chloride ion is 0.001 M, 0.01 M, 0.1 M, 0.5 M, 1.0 M, respectively. In addition, we also test the influence of chloride concentration on the potential of commercial electrodes. 2.2.1.4 Effect of temperature test The heat released by cement hydration is considerable, so the temperature of new mixing concrete is high, but the temperature will continue to drop when the hydration is finished until equaling to outside environment. In our work, we chose several temperatures that were 0, 5, 15, 25, 35, 45°C, respectively. The testing solution was still saturated Ca(OH)2 solution. 2.2.2 Polarization measurements Potentiodynamic polarization measurements were performed in all solutions A-E. The potentiodynamic condition corresponded to a potential sweep rate of 1 mV/s and potential ranges of ¹80 to +80 mV from the open circuit potential (OCP). It should be noted that time intervals of 15–20 minutes were given for the system to get a stationary state and the OCP was observed. Cyclic polarization measurements adopted a similar process as above. The potentiodynamic condition corresponded to the potential sweep rate of 0.1 mV/s and potential ranges of ¹20 to 20 mV from the OCP. The experiment was run for two cycles. The difference in the shifting of potential from the first cycle and the second cycle was recorded. Galvanostatic polarization tests were performed with Ag/AgCl RE in saturated Ca(OH)2 solutions. A cathodic direct current (DC) of 0.2 µA/cm2 was applied to the Ag/AgCl RE for 6 days. After this, current was interrupted, and the Ag/AgCl RE was allowed to recover to their initial OCP. 2.3 Tests in concrete The concrete specimen with the size of 10 © 10 © 10 cm was fabricated. Its w/c ratio was 0.6. The chloride ions were introduced in concrete by adding sodium chloride into the mixing water. The added contents of chloride ions were 1.0%, 2.0% and 3.0% by mass of cement. Besides, The control specimen without the addition of chloride addition was also prepared. Each concrete specimen was embedded with a carbon steel, a Ag/AgCl RE and an auxiliary electrode of titanium. The diameter of the carbon steel was 1 cm and the length of the steel was 5 cm. It should be noted that the reference electrode and the titanium electrode were all placed near to the steel.20 These specimens were cast and cured for two weeks before the test.

Figure 2. (Color online) Potential evolution with time of different Ag/AgCl reference electrodes following immersion in SCS. Fabricated Ag/AgCl reference electrodes

(a)

Commercial Ag/AgCl reference electrodes

(b)

3. Results and Disscusion 3.1 Uniformity test of Ag/AgCl reference electrodes The Ag/AgCl RE potential evolution with time in saturated Ca(OH)2 solution is indicated in Fig. 2. In this experiment, 8 numbers of Ag/AgCl REs manufactured in different batches were made and tested in SCS. It can be seen that, the potentials of electrodes become stable after about three weeks. Among them, three electrodes show ¹65 mV, two electrodes show ¹60 mV and others show ¹69 mV. The maximum variation observed is only «5 mV vs. SCE and this difference in potential is almost negligible, indicating that Ag/AgCl reference electrodes show uniformity in concrete environments under laboratory condition.

Figure 3. Stability of fabricated Ag/AgCl reference electrodes (a), commercial Ag/AgCl reference electrodes (b) in solutions of different pHs.

3.2 Stability test of Ag/AgCl reference electrode Figure 3 showed the potential evolution with time of the fabricated and commercial Ag/AgCl REs immersed in solutions A-E. The electrochemical stability of the Ag/AgCl RE was

investigated with respect to SCE for the exposure period of five weeks. Interestingly, fabricated Ag/AgCl reference electrodes (Fig. 3a) in all five solutions showed stable behavior after two weeks. The stability of the Ag/AgCl RE was lasted after two weeks

1042

Electrochemistry, 82(12), 1040–1046 (2014)

Figure 4. Potential evolution of fabricated Ag/AgCl reference electrodes and commercial Ag/AgCl reference electrodes in SCS.

Figure 5. Effect of chloride activity on the potential of fabricated and conmercial Ag/AgCl reference electrodes.

up to five weeks of exposure. The commercial Ag/AgCl reference electrodes (Fig. 3b) stabilized shortly (7/8 days) after immersion in the test solutions. The fabricated electrodes were only several days later to get stable potential than commercial ones. Here it was concluded that the fabricated electrodes maintained the good stability in concrete environments. Effect of pH values—According to,10,11 the effect of pH values on potential of MMO electrode is great and the Ag/AgCl electrodes shows the lower standard deviation than the MMO and Graphite electrodes. Figure 3a indicated that the potential of fabricated electrodes was affected by the pH values of solutions. The potential at pH = 6.86 was highest. The potential at pH = 13.5 was 15 mV lower than pH = 6.86 and the potential at pH = 9.18 was 20 mV lower than pH = 10. The potential at pH 12.5 is about 40 mV lower than potential at pH = 6.86, which is the lowest of all 5 levels. The reason was attributed to the fact that pH 12.5 is similar to the pH of the bottom layer of the reference electrode. Figure 3b showed the potential of commercial electrodes was also influenced by pH values. The potential was highest at pH 6.86 and the potential was lowest at pH 12.5. However, the potential of commercial electrodes at pH 12.5 was also 35 mV lower than pH = 6.86. In a word, the influence of pH value on the potential of commercial electrodes was similar to the fabricated electrodes. Nevertheless, there was no clear trend between the potential of electrodes and the pH values of solutions, which need further experiments and investigation.

3.4 Effect of Cl− concentration The potential of fabricated Ag probe coated with AgCl on its surface exhibits a linear relationship with the logarithm of chloride activity,13,20,22–25 which is widely used as a selective chloride electrode. The equilibrium at the surface is

3.3 Reliability test of Ag/AgCl reference electrodes The reliability of Ag/AgCl RE was testing in saturated Ca(OH)2 solution (SCS) too. Here it was observed that as expected the potential difference between the two Ag/AgCl electrodes was found to be vary small. The maximum potential difference was «3 mV. This indicates that the reliability of all the Ag/AgCl reference electrodes was found to be excellent. Figure 4 shows the potential evolution of fabricated electrodes and commercial Ag/AgCl reference electrodes in SCS for about five weeks. The potential of commercial electrodes is earlier than fabricated electrodes to become stable. The potential of fabricated electrodes is about 20 mV lower than the commercial electrodes, which can be attributed to the inherent membrane potential of cement mortar layer.21 The potential of fabricated electrodes get stable condition soon and last a long time like the commercial electrodes, which indicates that the fabricated electrodes have good reliability.

AgCl þ e $ Ag þ Cl

ð1Þ

and, the equilibrium potential is given by Nernst equation E ¼ Eo 

RT  lnðCl Þ nF

ð2Þ

which, at room temperature (T = 298 K) and for n = 1 becomes E ¼ Eo  0:059 logðCl Þ

ð3Þ

According to Eq. (3), the potential of Ag probe coated with AgCl should be lager with the chloride concentration becomes less and the potential was sensitive to chloride concentration. Figure 5 shows the potential varies of fabricated and commercial reference electrodes with the chloride activity. On the contrary, the potential of Ag/AgCl RE became smaller with the decrease of the activity of chloride. And when the chloride concentration is lower than 0.1 M, the potential changed very little. The potential of commercial electrodes at 0.001 M was 15 mV lower than the potential of 1 M, but the potential of fabricated electrodes at 0.001 M was only 22 mV lower than the potential of 1 M. It can be clearly found that the Ag/AgCl RE is not sensitive to chloride concentration. In other words, the chloride activity has little effect on the potential of the reference electrodes. The explanation is that the Ag probe with a coat of AgCl is immersed in gel electrolyte and the bottom is layer of cement paste, which isolates the Ag probe with the external environment. 3.5 Effect of temperature Figure 6 shows the change of Ag/AgCl reference electrode potential with temperature in SCS. The curve revealed a linear response, with a coefficient of 0.9980, the slope was 0.266 which was in good agreement with the value 0.22 mV/°C according to Yang.5 3.6 Polarization behavior of Ag/AgCl reference electrodes 3.6.1 Potentiodynamic polarization test Results of the potentiodynamic polarization tests are shown in Fig. 7. The experiments were done twice in order to get the same results. According to,26,27 exchange current density can be calculated from Eq. (4): 1043

Electrochemistry, 82(12), 1040–1046 (2014)

Figure 6. (Color online) Effect of temperature on the potential of Ag/AgCl reference electrodes.

Figure 7. Potentiodynamic polarization curves of Ag/AgCl reference elctrodes in solutions of different pHs.

Table 2. Exchange current density for Ag/AgCl reference electrode in concrete environments. pH = 6.86 pH = 9.18 pH = 10 pH = 12.5 pH = 13.5 Io (mA/cm2)

130

Io ¼

RT ¤I F ¤U

65

88

200

95

ð4Þ

The exchange current density characterize the capacity of polarization resistance. Brossia and Kelly25 used the micro-polarization tests to determine the exchange current density of Ag/ AgCl(Methanol) electrode which was found to be 30 µA/cm2, an acceptably high exchange current density for a reference electrode. The exchange current density of Ag/AgCl RE in concrete environments derived from the potentiodynamic polarization curves is given in Table 2. The trend of exchange current density in all five solutions is sequenced in the order: pH = 12.5 > pH = 6.86 > pH = 13.5 > pH = 10 > pH = 9.18. Here it was observed that, the exchange current density was largest at pH 12.5, the reason may be attributed to that the pH of bottom layer of electrodes was near to 12.5. The exchange current density was second largest at the lowest pH and the third largest exchange current density was at the highest pH. However, the solutions whose pH values were close to neutral had the minimum of exchange current density. The minimum of exchange current density was 65 µA/cm2 that was two times as large as 30 µA/cm2, which proved the sensor has good ability of resisting polarization. Besides, Duffo11 reported the potentiodynamic polarization tests of Ag/AgCl electrodes showed that an anodic peak was observed only at pH 13.5, this anodic peak was probably due to the formation of some solver oxides.28–30 However, our Ag/AgCl reference electrodes show little hysteresis in potentiodynamic polarization testes and no an anodic peak is observed. 3.6.2 Cyclic sweep behaiour of Ag/AgCl reference electrodes The reversibility characteristics of the Ag/AgCl reference electrodes in concrete environments were investigated by cyclic sweep method. The typical cyclic sweep curves of Ag/AgCl electrode in all solutions with different pH values are given in Fig. 8. Moreover, the differences of potentials between the first and second polarization at pH = 13.5, 12.5, 10, 9.18 and 6.86 are 1.6, 3, 3.8, 2.1, 1.0, respectively. The maximum difference between the first and second cycle is 3.8 mV and the minimum difference is 1.0 mV, 1044

Figure 8. Cyclic polarization curve for Ag/AgCl reference electrode in solutions with different pHs.

these differences in potential are tolerable for reference electrodes used in concrete. Cyclic polarization behaviour of Ag/AgCl electrodes in concrete environments is good and the trend in the existence of the reversibility of the Ag/AgCl reference electrode in five solutions is as follow: pH = 6.86 > pH = 13.5 > pH = 9.18 > pH = 12.5 > pH = 10. 3.6.3 Galvanostatic polarization Results of the galvanostatic test are shown in Fig. 9. The potential of the Ag/AgCl reference electrode shifted about 30 mV and showed little hysteresis when the current was applied, besides, it recovered immediately after the interruption of the current. 3.7 Tests in concrete Figure 10 shows the Ag/AgCl reference electrode potentials as a function of time about a year, when exposed to four different conditions (chloride concentrations in concrete were 0, 1%, 2%, 3%, respectively). At the first month of test, all the potentials were increasing about 30 mV which could be attributed to the pH values increase in concrete. The potential then reduced to a stable value, it’s not difficult to note that the potentials increase with the higher concentration of chloride ions, but the potential of 1% chloride concentration is near to 2% and 3% chloride. The potential of 1% chloride concentration is only 20 mV lager than the potential without chloride. The potential of graphite electrodes embedded in mortar immersed in NaCl was 100 mV lower than the ones embedded in mortar in laboratory environment.11 So we conclude that the

Electrochemistry, 82(12), 1040–1046 (2014) most advantage of the electrode is chloride-free. The electrolytic contact of electrode to the concrete environments is through the bottom layer which is made up of diffusion barrier of cement paste gives a good protection to the electrode unlike other electrodes made up of glass or ceramic. This will give a good bond to the concrete. The liquid junction potential across this interface is very minimum because of the pH is nearly the same in the bottom layer and in the concrete. This will expect not to develop any junction potential at the layer/concrete interface, if this electrode is used in the field. Last but not least, through one year, almost no internal gel electrolyte leaked from this reference and the potential lasted stable state. The internal electrolyte gel can effectively prohibit the electrolyte from leaking. The superiority of this reference electrode is very compact, low leakage rate of internal elecrolyte, easy to use in the field, chloride-free and mercury-free sensor for concrete structures. 4. Conclusion Figure 9. Potential of Ag/AgCl reference electrode in SCS. Vertical dotted lines incatet the start and end of the galvanostatic pulse.

Figure 10. Potential evolition with time of Ag/AgCl reference electrodes in concrete with different chloride concentration.

In conclusion, a new solid embeddable Ag/AgCl reference electrode (RE) was electrochemically investigated for corrosion monitoring in reinforced concrete. The electrode was fabricated by applying the cement mortar as the bottom layer, the polymer gel as electrolyte. This fabricated electrode had the uniformity and good reliability in saturated Ca(OH)2 solution. The electrode proved to be stable and reliable sensor electrode in concrete environments. Moreover, the chloride concentration had little influence on the potential and the potential showed a linear relation with temperature. But the potential was affected by the pH value of solution. The potentiodynamic polarization («80 mV) test for the Ag/AgCl reference electrode in solutions indicated high exchange current density confirmed the better performance of resisting polarization. Cyclic polarization («20 mV) test for the electrode in the same solutions showed a maximum difference of 3.8 mV between the forward and reverse scans which certified the good reversibility of the reference electrode in concrete environments. Galvanostatic application of 0.2 µA/cm2 in electrode for several days caused little variation of potential (³30 mV), which indicated the presence of a finite polarization resistance and good reversibility of the reference electrode. Testing in concrete indicated the reference electrode was stable over a year and can be used to identify the real-time potential of steel which was near the reference electrode. Acknowledgment

chloride concentration has little influence on the potential of the Ag/ AgCl reference electrodes. The potentials of electrodes were kept stable over a year which proved our reference electrodes can be used in concrete as reliable reference electrodes. The real-time potentials of steels in concrete can be tested easily and quickly by the reference electrodes embedded near the steels which can eliminate the effect of concrete on the half-cell potential testing method. 3.8 Mechanistic aspects of reference electrodes in concrete environments The electrochemical stability of reference electrodes in simulated concrete environments was measured with respect to saturated calomel electrode. The stability of the any reference electrode depends on the measure of the activity of the reactive species. So all the reference electrodes depend on how far they maintain the stability and the stability depends on the activity of the electrode element. In the case of Ag/AgCl the galvanic-coupling between the reference electrode and the embedded rebar in concrete is a measure of the AgCl/Cl¹ equilibrium potential and Fe/Fe2+. The half-cell potential of Ag/AgCl is a complex function of the reduction state of AgCl. But the potential being determined by AgCl/Cl¹ equilibrium potential. In Ag/AgCl reference electrodes, the AgCl has gotten a equilibrium potential with the internal electrolyte, thus, one of the

The authors gratefully acknowledge the support provided by the Natural Science Foundation of China (Nos. 51278167, 51278168 and 51478164) and the Natural Science Foundation of Jiangsu province under Project (No. BK20131374). References 1. P. K. Mehta and P. J. Monteiro, Concrete: microstructure, properties, and materials, Vol. 3, McGraw-Hill New York (2006). 2. M. Gani, Cement and Concrete, Chapman and Hall, London (1997). 3. S. Muralidharan, V. Saraswathy, K. Thangavel, and N. Palaniswamy, Sens. Actuators, B, 130, 864 (2008). 4. S. Muralidharan, T. H. Ha, J. H. Bae, Y. C. Ha, H. G. Lee, K. W. Park, and D. K. Kim, Mater. Lett., 60, 651 (2006). 5. L. Yang, Techniques for corrosion monitoring, Elsevier (2008). 6. W. J. McCarter and Ø. Vennesland, Constr. Build. Mater., 18, 351 (2004). 7. H. Arup and B. Srensen, Corrosion’92, Paper No. 208, NACE, Houston, TX, USA (1992). 8. C, A. Standard test method for half-cell potentials of uncoated reinforcing steel in concrete. 2009: American Society for Testing and Materials West Conshohocken. 9. G. S. Duffo and S. B. Farina, Constr. Build. Mater., 23, 2746 (2009). 10. P. Castro, A. A. Sagüés, E. I. Moreno, L. Maldonado, and J. Genescá, Corrosion, 52, 609 (1996). 11. G. S. Duffó, S. B. Farina, and C. M. Giordano, Electrochim. Acta, 54, 1010 (2009).

1045

Electrochemistry, 82(12), 1040–1046 (2014) 12. K. N. Gurusamy and M. P. Geoghegan, Corrosion of reinforcement in concrete, p. 333 (1990). 13. M. A. Climent-Llorca, E. Viqueira-Pérez, and M. M. López-Atalaya, Cem. Concr. Res., 26, 1157 (1996). 14. A. K. Covington, P. D. Whalley, and W. Davison, Anal. Chim. Acta, 169, 221 (1985). 15. D. P. Brezinski, Anal. Chim. Acta, 134, 247 (1982). 16. S. Ito, H. Hachiya, K. Baba, Y. Asano, and H. Wada, Talanta, 42, (1995). 17. I. Huang, R. Huang, and L. Lo, Sens. Actuators, B, 94, 53 (2003). 18. M. Glanc, M. Sophocleous, J. K. Atkinson, and E. Garcia-Breijo, Sens. Actuators, A, 197, 1 (2013). 19. M. Raupach and P. Schießl, NDT and E Int., 34, 435 (2001). 20. M. F. Montemor, J. H. Alves, A. M. Simões, J. C. S. Fernandes, Z. Lourenço, A. J. S. Costa, A. J. Appleton, and M. G. S. Ferreira, Cem. Concr. Compos., 28, 233 (2006).

1046

21. J. Zhang and N. R. Buenfeld, Cem. Concr. Res., 27, 853 (1997). 22. C. P. Atkins, J. D. Scantlebury, P. J. Nedwell, and S. P. Blatch, Cem. Concr. Res., 26, 319 (1996). 23. M. A. ClimentLlorca, E. ViqueiraPerez, and M. M. LopezAtalaya, Cem. Concr. Res., 26, 1157 (1996). 24. B. Elsener, L. Zimmermann, and H. B. Hni, Mate. Corros., 54, 440 (2003). 25. C. S. Brossia and R. G. Kelly, Electrochim. Acta, 41, 2579 (1996). 26. R. G. Kelly and P. J. Moran, Corros. Sci., 30, 495 (1990). 27. A. W. Hassel, K. Fushimi, and M. Seo, Electrochem. Commun., 1, 180 (1999). 28. R. D. Giles and J. A. Harrison, J. Electroanal. Chem. Int. Electrochem., 27, 161 (1970). 29. N. Sato and Y. Shimizu, Electrochim. Acta, 18, 567 (1973). 30. M. L. Teijelo, J. R. Vilche, and A. J. Arvia, J. Electroanal. Chem. Int. Electrochem., 162, 207 (1984).