Construction and Building Materials 53 (2014) 403410Contents lists available at ScienceDirect

Construction and Building Materials

journal homepage: www.elsevier .com/locate /conbui ldmatMonitoring method for the chloride ion penetration in mortar by athin-film sensor reacting to chloride ion0950-0618/$ - see front matter 2013 Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.conbuildmat.2013.11.095

Corresponding author. Tel.: +82 31 400 5181; fax: +82 31 436 8169.E-mail address: ercleehs@hanyang.ac.kr (H.-S. Lee).Won-Jun Park a, Hyun-Seok Lee b, Sung-Hyung Joh b, Han-Seung Lee b,a Sustainable Building Research Center, Hanyang University, Ansan, Gyeonggi-do, Republic of KoreabDepartment of Architecture, Hanyang University, Ansan, Gyeonggi-do, Republic of Korea

h i g h l i g h t s

A sensor reacting to chloride ion was developed using screen printing. The sensor with a weight ratio for Ag:Fe of 1:2 was proven to react to corrosion. Electrical resistance was confirmed to change with corrosion caused by salinity. The sensor could monitor the depth of salinity penetration from the mortar surface.a r t i c l e i n f o

Article history:Received 27 September 2013Received in revised form 25 November 2013Accepted 26 November 2013Available online 27 December 2013

Keywords:Monitoring methodMortarChloride ion penetrationSensorElectrical resistancea b s t r a c t

The depth of chloride ion permeation from a concrete surface can be monitored in reinforced concretestructures. The purpose of this study is to develop and apply a thin-film sensor based on the measure-ment of changes in electrical resistance of the sensing elements in order to follow the chloride ion pen-etration in mortar. The proposed thin-film sensors were placed in mortar specimens intrinsicallycontaining chlorides, and mortar specimens impregnated with chloride solutions then subjected to accel-erated corrosion in a NaCl solution. For making a film sensor, a screen printer machine was used. Silver(Ag) paste and iron (Fe) powder were coated on the thin-film sensor that reacts to chloride ion in mortar.As a result of experimental testing, the appropriate ratio of Ag to Fe was >1:2. The electrical resistance ofthe sensors increased with the degree of corrosion behavior of thin-film sensor. The time to the firstchange in electrical resistance decreased along with the degree of corrosion as the distance from the mor-tar surface decreased. The proposed thin-film sensors are thus confirmed to be capable of monitoring thedepth of chloride permeation in mortar with sufficient accuracy.

2013 Elsevier Ltd. All rights reserved.1. Introduction

The corrosion of steel reinforcements in a reinforced concretestructure starts from the surface of the steel reinforcement [1].Thus, the presence of cracks on concrete surface implies that corro-sion of steel reinforcement has made considerable progress inside[2,3]. Corrosion of steel reinforcement in concrete is the maincause for early performance degradation and the breakdown ofreinforced concrete structures. Of the deterioration phenomenathat occur in reinforced concrete structures, deterioration causedby steel reinforcement corrosion produces the most severe dam-age; the cost for maintaining and constructing these structures isvery high, and it is difficult to determine the appropriate mainte-nance period [4,5]. Therefore, the prediction and early detectionof corrosion of steel reinforcement is very important for the admin-istrator to establish efficient maintenance and enhancement plans[68].

Currently, there are several researches being conducted fordetermining the appropriate maintenance period or repair pro-cess based on the extent of damage to structures, but the mon-itoring systems that focus on the penetration of chloride ion tothe corrosion of steel reinforcement before steel reinforcementcorrosion occurs are lacking. Additionally, maintenance of mostconcrete structures usually starts after deterioration occurs. Suchmeasurements can be said to be repairs covering up the phe-nomena rather than maintenance in consideration of the future[9,10].

Thus, a sensor needs to be developed to monitor the penetrationprocess of chloride (Cl) which affects the corrosion of steel rein-forcement in concrete before it progresses, so that the propermaintenance period and appropriate repair method can be estab-lished. A system for non-destructive monitoring of building struc-tures also needs to be developed.

http://crossmark.crossref.org/dialog/?doi=10.1016/j.conbuildmat.2013.11.095&domain=pdfhttp://dx.doi.org/10.1016/j.conbuildmat.2013.11.095mailto:ercleehs@hanyang.ac.krhttp://dx.doi.org/10.1016/j.conbuildmat.2013.11.095http://www.sciencedirect.com/science/journal/09500618http://www.elsevier.com/locate/conbuildmat

404 W.-J. Park et al. / Construction and Building Materials 53 (2014) 403410This study carried out a performance assessment for a devel-oped thin-film sensor reacting to chloride ion. The reactivity ofthe sensor to changes in the salinity of the mortar was examined,and the monitoring of chloride ion penetration factors for a build-ing by the sensor was considered at different depths. The sensorwas built into the structure as shown in Fig. 1 and was monitoredin real time. The penetration of chloride ions was detected by theelectrical resistance of the corrosion of sensor.2. Consideration of existing literature

Sensors are defined in the IEEE Standard Dictionary of Electricaland Electronics Terms (1996) 6th Edition and classified accordingto temperature instrumentation type and by devices for testing,production, instrumentation, diagnosis, etc. In other words, a sen-sor refers to devices having the functions of metering, detecting,determining, or measuring various types of quantities such as tem-perature, sound, pressure, and light and delivering them as signalsor in the applicable unit of measurement [5]. For monitoring corro-sion or chloride ion diffusion, the actual methods are divided intostatic measurements and polarization measurements. This studyfocused on the static methods as half-cell potential, macro-cell cur-rent and electrochemical noise measurements for corrosion moni-toring which various sensors are applied [11].

During the corrosion process, corrosion macro-cells are formedwith a distribution of anodic and cathodic areas. Different measur-ing configurations for in situ testing are developed. A step-typeprobe (Schiebl probe or ladder system) sensor system was de-signed to monitor the risk of corrosion from airborne chlorides pe-netrating concrete [4,12]. The system consists of the steelelectrodes and insulting supports. The sensor can be built in anew construction or during the repair. Steel electrodes are placedat different depths which makes depassivation front monitoringpossible [11,13]. Another configuration is a multi-probe system.The test unit is exposed to chloride ions diffusing from one side.Initiation of corrosion can be detected by a sudden rise in the ano-dic current. This test method has an advantage in providing directindication of electrochemical activity in the system [11,1318].

Upon the pumping of concrete, this probe is placed close to thesteel reinforcement, providing a monitoring signal hourly. If iron isinitially put under strong alkaline conditions, a passive film formson the iron surface, and hence, no corrosion occurs [4,11,12]. Atthis point of time, even if the iron is electrically connected to metalwith an ionization tendency less than that of closely placed steelCl-CO2

Cl-CO2

Cl-CO2

Fig. 1. Sensor reacting to chloride ion in the structure.reinforcements (e.g., stainless steel), almost no flow of current oc-curs between the two types of metal. However, if the passive filmon the iron surface is damaged by the penetrating corrosion factorsand if corrosion is progress further, the amount of current from theiron to the stainless steel increases according to the degree of cor-rosion. Sequential measure of the electrical reaction at the iron sur-face at different depths allows the penetration depth and speed ofthe corrosion factors to be monitored [1,19].

On the other hand, wire sensor system was developed to mon-itor the factors that cause corrosion along the depth. Whereas theSchiebl probe sensor system measures the electrical flow with ironby inserting a type of metal (noble metal), this sensor places a thiniron wire at a depth, and monitors the change in wire resistancebetween both ends; the wire disconnects in reaction to corrosionfactors [20,21]. Another wire sensor covered by PVC in the formof a wire is wrapped around the steel to be monitored. A potentialbetween the steel and electrode can be measured by using the half-cell. The advantage of the method is great sensitivity, which makesthe method suitable for measurements of pitting corrosion at thelarge concrete structures [11,22].

Finally, corrosion sensor system by sputtering was developed.Sputtering is a method of applying a metal or oxide (target) usingplasma on the material surface [23,24]. However, there are weak-nesses to deposition technology using sputtering: it is complex; ittakes a long time to mass-produce sensors; the sensor is attachedto a specific substrate; and the attachment stress with the sub-strate is weak [2326]. Therefore, a screen printer machine wasused for making a film sensor in this study. To improve the previ-ous disadvantage, Ag paste and iron Fe powder were coated on thethin-film sensor that reacts to chloride ion in mortar environment:it is not complex; it take short time to mass-produce sensor; thesensor is small size; and the sensor has high response speed withelectrical resistance.

The comparison of electrochemical methods for chloride iondiffusion and corrosion characterization in reinforced concrete issummarized in Table 1.3. Sensor development

3.1. Sensor system using a screen printer

Screen-printer technology adds a basic solution (Ag paste) thatcan be printed. If this solution is not put in, screen printer equip-ment cannot be used. However, this study attempted to print asensor reacting to chloride ion using screen-printer technologyby mixing a suitable amount of 99.9% pure iron powder (powdersize 10 lm) with Ag paste to develop a sensor that reacts in thesame manner as iron. Screen-printer technology is a simple silkscreen method for producing a sensor by masking and etchingusing a photolithographic method [25,27]. In addition, becausethe sensor drawing can be replaced, various sensor forms can beproduced. Thus, the method of printing the sensor directly ontothe substrate with screen-printer technology was employed.

Fig. 2 shows a mask drawing. For printing (Fig. 3), a maskwas attached to the silk screen equipment, and the substrate wasbonded to the silk screen equipment. Printing startedafter the Ag paste and iron powder were mixed and applied onthe mask.

3.2. Sensor development methods

3.2.1. Design of sensorDrawing of the sensor is shown in Fig. 4. This drawing was pre-

viously developed by sputtering technology [24], but its variousweaknesses were mitigated using screen-printer technology. The

Table 1Comparison of electrochemical methods [11].

Classification Measurement principle Material composition Measurementtarget

Sensor location Remarks (Sensitivity) Ref.

Cla Dcb Coc

Macro-cell current Galvanic current betweensteel and stainless steel

Steel and stainless steel s s Near the rebar Qualitative approach, measurementerror (Low sensitivity)

[4,12] s s Location free [13,14]

Steel and stainless mesh s s Location free Quantitative approach, problems ofsize and price, Additional equipment(Low response speed)

[15,16]Steel and stainless steel s s s Location free [18]

Half-cell potential Electrical resistance at bothends of wire or circuit

Steel wire ( 0.1 mm) s s s Location free Quantitative approach, change ofwire resistance (Middle responsespeed)

[2123]

Thin-film (Proposed method) s s s Location free Qualitative approach, small size,low price (High response speed)

[2427]

a Cl: Measurement of chloride ion.b Dc: Monitoring of chloride ion diffusion.c Co: Measurement corrosion of rebar and sensor.

Fig. 2. Mask drawing.

Fig. 3. Silk screen equipment.

Fig. 4. Drawing of sensor.

W.-J. Park et al. / Construction and Building Materials 53 (2014) 403410 405source technologies and materials to produce the corrosion sensordiffer. Each sensor was arranged in 10 thin lines with 0.158 mmline widths. The sensors have dimensions of 12.7 mm (horizon-tal) 15.88 mm (vertical). If chloride ions penetrate the mortar,each line of the sensor rusts and corrodes. This means that an elec-tric current does not flow along each line. Therefore, the total elec-trical resistance value of the sensor increases; if all 10 lines becomeshort-circuited, electric current does not flow through the sensor.Based on this principle, the penetration of chloride ions into mortarcan be monitored. Therefore, if this sensor is arranged and built atregular intervals from the concrete cover, the penetration depth ofchloride ions that affect steel reinforcements corrosion can bedetermined.3.2.2. Sensor solution mixingAg paste is a solution used for silk screens. Yet, owing to Ags

chemical properties, it is difficult for corrosion caused by chlo-ride ions to take place. Thus, to produce a sensor with the samecorrosive reaction as a steel reinforcement, iron powder wasmixed with the Ag paste. The particle size of the iron powderwas 10 lm. The mixes of Ag paste and iron powder accordingto the weight ratio are presented in Table 2. Printing was doneusing silk screen equipment up to a weight ratio of 1:1 for theAg paste and iron powder. However, as the rate of iron increases,the viscosity of the mixed solution increases, which became aproblem. If viscosity increases, printing is not possible. Thus,from the rate of 1:2, the issue of viscosity was solved by mixinga diluted Ag paste solution.

As shown in Table 2, as the rate of iron powder is increased,the sensors initial electrical resistance value tends to increase.This can mean that Ags electrical conductivity is better thanthat of iron, so the electric current flows better. If the flow ofelectric current is good, electrical resistance decreases. Therefore,the initial electrical resistance value can be adjusted by control-ling the rate of iron powder. The initial electrical resistance of aFe/Ag = 1 sensor had an average value of 1.4 O and the initialelectrical resistance of a Fe/Ag = 2 sensor had an average valueof 21 O. The initial electrical resistance of 17 randomly selectedsensors printed using a screen printer was measured. Each sen-sor had a small error in its initial electrical resistance value;however, this error was due to the mixing of the iron powderand Ag paste and was judged to have little effect on the exper-iment results.

Table 2Mixture solution ratio.

Ag paste iron powder ratio Dilute solution (g) Initial resistance (O)

1(50 g):0 0.61(50 g):0.5(25 g) 0.81(50 g):0.75(37.5 g) 1.01(50 g):1(50 g) 1.41(50 g):1.5(75 g) 7 81(50 g):2(100 g) 10 211(50 g):3(150 g) 18 35

Fig. 5. Change in sensors electrical resistance in 1% NaCl solution.

Fig. 6. R0/R value change with time by the NaCl density of the sensor with Fe/Ag = 2.

Fig. 7. SEM analysis of sensor with Fe/Ag = 1 (1000).

Fig. 8. SEM analysis of sensor with Fe/Ag = 2 (1000).

Fig. 9. Shape of the test body of inner salinity mortar corrosion sensor response.

Table 3Mortar composition.

W/C [%] Weight

Water Cement Sand NaCl

50 1(kg) 2(kg) 6(kg) 0(g)(0%)6(g)(0.6%)10(g)(1%)30(g)(3%)

Fig. 10. Corrosion stimulation condition by thermo moisture machine.

406 W.-J. Park et al. / Construction and Building Materials 53 (2014) 4034103.3. Sensor response test in NaCl solution

To examine the degree of corrosion of the sensor reacting tochloride ion and the change in its electrical resistance accordingto salinity, sensors were deposited in a NaCl solution at salinitydensities of 0%, 0.6%, and 1% for 12 h and dried in air for 12 h asone cycle. This stimulated sensor corrosion, and the electrical resis-tance in the aqueous solution was measured. Fig. 5 shows the

Fig. 11. Shape of test body with corrosion sensor laid at the position.

W.-J. Park et al. / Construction and Building Materials 53 (2014) 403410 407change in sensor electrical resistance for the NaCl 1% solution.There was almost no change in electric electrical resistance forthe sensor with Fe/Ag = 1. However, for the sensor with Fe/Ag = 2, the electrical resistance changed over time. At 147 h, allthe lines became red-cyan, so electrical resistance became 0. Theelectrical resistance first changed at 28 h; half of the sensor be-came red-cyan after 75 h. Later, the electrical resistance of the sen-sor increased rapidly.

Fig. 6 shows the change in R0/R with the NaCl density for thesensor with Fe/Ag = 2 over time. R0 is the sensors initial electricalresistance of 21 O. R is the electrical resistance measured with timeand increases upon the occurrence of corrosion. Therefore, R0/Rtends to decrease as corrosion progresses. However, the sensorreacting to chloride ion deposited in a 0% NaCl solution did not rustover the same period of time, and the change in electrical resis-tance of the sensor was very small. Thus, because the density ofthe NaCl solution was high, the sensors change in electrical resis-tance occurred quickly, and the electrical resistance increased.Thus, the developed sensor responds to salinity in the aqueoussolution, and the corrosion reaction or electrical resistance alsochanges with the salt concentration.

The sensor with Fe/Ag = 1 was investigated as to why no corro-sion occurred with time, using SEM analysis. Fig. 7 shows an SEMimage of the sensor with Fe/Ag = 1, whereas Fig. 8 shows the(a) Developed corrosion sensor (b) Connect w

(d) Place mortar (e) Accelera

Fig. 12. Experiment pSEM image for the sensor with Fe/Ag = 2. For the sensor with Fe/Ag = 2, corrosion progressed and electrical resistance increased be-cause the Ag powder particles were not connected to each other;instead, the Fe particles were arranged between different Ag parti-cles. Therefore, if Fe corrodes, no electric current flows.4. Sensor response tests in mortar

4.1. Sensor response test for inner salinity

The sensors reactivity was examined as the inner mortar salin-ity content (Cl ion weight) was changed. Fig. 9 presents the shapeof the test body for the inner salinity mortar sensor reacting tochloride ion response test. The sensor is placed in the mortar testbody (40 mm 40 mm 160 mm) so that the distance betweenthe surface and sensor is 10 mm. Table 3 shows the mortar combi-nations used. The ratio of water to cement was set to 50%, whereasthe ratio of cement to sand was 1:3. Because the KS F 4009 stan-dard (Korean standard for ready-mixed concrete) states that theCl amount should be less than 0.4% of the cement weight, if trans-lated into molecular weight, the amount of NaCl should be lessthan 0.6%. Thus, the NaCl amount was set to 0%, 0.6%, 1%, and 3%of the cement weight. Fig. 10 shows the corrosion stimulation con-ditions set for a thermo moisture machine. The test body waswater-cured for 1 day after insertion of the sensor and then driedwith air for 1 day. The test body was then placed in the thermomoisture machine and underwent 1 cycle, consisting of 12 h at atemperature of 60 C and RH of 95% and 12 h at a temperature of60 C and RH of 35%. The electrical resistance of the sensor reactingto chloride ion then was measured.

4.2. Sensor response test for airborne chlorides

To examine the response of the sensor reacting to chloride ionto salinity penetration into mortar, as shown in Fig. 11, mortarire on sensor (c) Embedded sensor

te corrosion (f) Measurement

rocess sequence.

408 W.-J. Park et al. / Construction and Building Materials 53 (2014) 403410was poured to the test body (40 mm 40 mm 160 mm) with thesensor placed at positions of 10 mm, 20 mm, and 30 mm from themortar surface. After 1 d of water curing, surfaces other than thesalinity penetration surface were coated with epoxy. After 12 hof deposition in a NaCl 3% solution at room temperature, the testbody was placed in a thermo moisture machine with a tempera-ture of 60 C at a relative humidity of 35% for 1 cycle, and the sen-sors corrosion was stimulated. The sensors electrical resistancewas then measured in the period immediately after that. Fig. 12shows the sequential experimental process.Fig. 13. The sensors electrical resistance change with the lapse of time in mortar bysalinity density.

Fig. 14. Relationship of the electrical resistance of the sensor with the time changedinitially.5. Results

5.1. Result of sensor experiment in inner salinity

Table 4 shows the electrical resistance of the sensor by thesalinity density of the mortar.

Fig. 13 shows the change in electrical resistance of the sensorwith time, by the salinity density of the mortar. The sensor placedin a 3% salinity mixed test body experienced its first electricalresistance change after 4 cycles of stimulated corrosion; as corro-sion progressed, electrical resistance increased. The test body with-out mixed salinity experienced its first electrical resistance changeafter 16 cycles; however, later on there was no increase in electri-cal resistance. The 1% and 0.6% salinity mixed test bodies experi-enced their first changes in electrical resistance after 6 cycles and10 cycles, respectively; with corrosion stimulation, electrical resis-tance was gradually increased. In particular, the electrical resis-tance of the 1% salinity mixed test body rapidly increased after10 cycles, whereas that of the 0.6% salinity mixed test body in-creased after 14 cycles. This was due to the fact that parts of thelines of the sensor become short-circuited as corrosion progressed.In addition, in the 1% salinity mixed test body, even after thinningwas disconnected, electrical resistance tended to keep increasing.Overall, electrical resistance changed in the form of discrete steps.Thus, electrical resistance of the sensor increased smoothly if lightcorrosion occurred on the surface; however, it increased sharplywith disconnection caused by thinning of the sensor with increas-ing side defects. Fig. 14 shows the relationship of the electricalresistance of the sensor to the salinity density with the time of ini-tial change. The greater the salinity density in mortar, the faster theinitial electrical resistance cycle of the sensor occurs. This isTable 4Electrical resistance value of sensor by salinity density in mortar.

Salinity density in mortar

NaCl 3% (O) NaCl 1% (O) NaCl 0.6% (O) NaCl 0% (O)

1cycle 21 21 21 212cycle 21 21 21 213cycle 21 21 21 214cycle 25.5 21 21 215cycle 25.5 21 21 216cycle 30.2 25.5 21 217cycle 30.2 25.5 21 218cycle 61.1 29 21 219cycle 61.1 30.2 21 2110cycle 61.1 45.4 25.5 2111cycle 88.3 48 25.5 2112cycle 93 49 25.5 2113cycle 94 50 30.2 2114cycle 97 61.1 35 2115cycle 105 69.4 45.4 2116cycle 116.6 70.5 45.4 23.517cycle 116.6 78 52.6 23.518cycle 120 83 61.1 23.519cycle 126 88.3 61.1 23.520cycle 148.2 89.1 73.4 23.5because the sensor with thinning made from iron had a high salin-ity density, and therefore, corrosion occurred quickly.

Thus, the electrical resistance change and time can be quanti-fied because the sensor developed in this study not only has a cor-rosion response to salinity in mortar but also differs in responsetime based on the salinity density in the mortar.

5.2. Result of sensor experiment on airborne chlorides

Table 5 shows the electrical resistance values of the sensor withrespect to the depth of the sensor in the mortar. Fig. 15 shows theTable 5Electrical resistance value of the sensor by depth of the sensor built in mortar.

Depth of the corrosion sensor

10 mm (O) 20 mm (O) 30 mm (O)

1cycle 21 21 212cycle 21 21 213cycle 23.5 21 214cycle 25.5 21 215cycle 28 21 216cycle 30.2 23.5 217cycle 36 23.5 218cycle 45.4 25.5 219cycle 57.2 25.5 2110cycle 61.1 36 2111cycle 74.8 36 2112cycle 81.5 48.6 2113cycle 88.3 49 23.514cycle 95.7 52.1 23.515cycle 103.8 52.6 25.516cycle 116.6 57.2 25.517cycle 129.5 61.1 30.218cycle 145.6 63.5 3619cycle 168.2 65.6 45.420cycle 177.2 74.8 52.6

0

20

40

60

80

100

120

140

160

180

Fig. 15. Electrical resistance change of the sensor as time passes by building it in bydepth.

Fig. 16. Initial electrical resistance cycle by the location of the sensor.

W.-J. Park et al. / Construction and Building Materials 53 (2014) 403410 409change in electrical resistance of the sensor with time, with respectto the depth. In other words, it shows the change in electrical resis-tance of the sensor as it corrodes with salinity penetration from themortar surface over time, during the corrosion stimulation experi-ment after sensors were placed 10 mm, 20 mm, and 30 mm fromthe surface of a mortar test body that was deposited in 3% NaClsolution for 12 h. The electrical resistance of the sensor located10 mm from mortar surface, which was closest to the surface,experienced its initial change in electrical resistance after 3 cycles;the electrical resistance then increased rapidly to 177 O after 20cycles. This was because part of the lines of the sensor became dis-connected. Meanwhile, there was almost no electrical resistancechange in the other inner sensors up to 10 cycles. This was becausethe salinity density differed at the position of each sensor based onsalinity penetration from the mortar surface, which affected thedegree of each sensors corrosion. The sensors placed 20 and30 mm from the mortar surface experienced their first changes inelectrical resistance after 6 cycles and 13 cycles, respectively, ofstimulation corrosion. However, there was no sharp increase inelectrical resistance after the first change in electrical resistance.The sensor placed at a 20 mm depth experienced a rapid increasein electrical resistance after 10 cycles. Part of the lines of the sensormay have become short-circuited at this time. The sensor placed ata 30 mm depth showed a gradual increase in electrical resistanceafter 16 cycles. The sensor is believed to have incurred corrosionwith the increase in salinity density, and electrical resistance in-creased as the corrosion progressed; however, there was no sharpincrease in electrical resistance, and hence, the lines of the sensordid not become short-circuited up to 20 cycles.

Fig. 16 shows the initial electrical resistance cycle with respectto the location of the sensor from the mortar surface. The closer thesensor was to the mortar surface, the faster the initial electricalresistance cycle of the sensor occurred. As salinity penetrated themortar surface, the salinity density of the sensor changed. Thus,the developed sensor was sensitive to salinity penetration of themortar surface. The deeper the sensor was placed, the greater thenumber of cycles before the first change in electrical resistance.Therefore, salinity penetration by depth can be quantified. Thus,nondestructive monitoring of chloride ion penetration caused byenvironmental factors can be realized.6. Conclusions

A sensor reacting to chloride ion was developed using screenprinting. The weakness of the existing sputtering technique wasmitigated by this new technique. To use screen printing, a sensorthat reacts in a similar manner to a steel reinforcement can be pro-duced by mixing Ag paste and iron powder. The sensor with aweight ratio for Ag:Fe of 1:2 was proven to react to corrosion,through SEM analysis and a NaCl solution experiment. When theFe/Ag ratio was more than 2, the electrical resistance of the sensorincreased in the 3% NaCl solution. To examine the reactions of thesensor in mortar to NaCl density, test bodies were produced;changes in the electrical resistance of the sensor according to itsdepth in the test body were examined. The following conclusionscan be drawn:

(1) The electrical resistance of the sensor changed with corro-sion in the NaCl solution and mortar with salinity; highersalinity densities caused the initial electrical resistance cycletime to decrease; the degree of corrosion and electrical resis-tance over the same corrosion stimulation time to increase.

(2) Measurement of the change in electrical resistance of thesensor owing to salinity penetration of the mortar surfaceaccording to the depth from the surface showed that the clo-ser the sensor was to the surface, the faster the first electri-cal resistance cycle time and the greater the degree ofcorrosion and electrical resistance over the same corrosionstimulation time.

(3) Based on the results of the above sensor response test inmortar, the electrical resistance of the developed sensorwas confirmed to change with corrosion caused by salinity,and the proposed sensor was proven to be able to suffi-ciently monitor the depth of salinity penetration.Acknowledgement

This work was supported by a Grant (12, Advanced-City, D02)from Architecture & Urban Development Research Program fundedby Ministry of Land, Infrastructure and Transport of Korean gov-ernment (MLIT).References

[1] ACI Committee 222. Corrosion of Metals in Concrete. ACI J 1985;82:332.[2] Arya C, Ofori-Darko FK. Influence of crack frequency on reinforcement

corrosion in concrete. Cem Concr Res 1996;26:34553.[3] Senecal MS, Darwin D, Lock C. Evaluation of corosion resistant steel reinforcing

bar. SM Report 40: University of Kansas; 1995. p. 142.[4] Schiebl P, Raupach M. Monitoring system for the corrosion risk for steel in

concrete. Concr Int 1992;14:525.[5] Roberge PR. Corrosion inspection and monitoring. John Wiley & Sons; 2007.[6] Isecke B. Collapse of the Berlin Congress Hall Prestressed Concrete. Roof,

Materials Performance 21: Berlin concress; 1982. p. 3639.[7] Engineering News Record. 212: McGraw-Hill; 1984. p. 11.[8] Borgard B, Warren C, Somayaji S, Heidersbach R. Proc Symp. on the corrosion

rate of steel in concrete. Baltimore: ASTM Commitee; 1990. p. 174.[9] Bentur A, Diamond S, Berke NS. Steel Corrosion in Concrete. UK,

London: Baltimore; 1997.[10] Broomfield JP, Davies K, Hladky K. The use of permanent corrosion monitoring

in new and existing reiforced concrete structures. Cem Concr Compos2002;24:2734.

http://refhub.elsevier.com/S0950-0618(13)01128-8/h0005http://refhub.elsevier.com/S0950-0618(13)01128-8/h0010http://refhub.elsevier.com/S0950-0618(13)01128-8/h0010http://refhub.elsevier.com/S0950-0618(13)01128-8/h0020http://refhub.elsevier.com/S0950-0618(13)01128-8/h0020http://refhub.elsevier.com/S0950-0618(13)01128-8/h0025http://refhub.elsevier.com/S0950-0618(13)01128-8/h0040http://refhub.elsevier.com/S0950-0618(13)01128-8/h0040http://refhub.elsevier.com/S0950-0618(13)01128-8/h0045http://refhub.elsevier.com/S0950-0618(13)01128-8/h0045http://refhub.elsevier.com/S0950-0618(13)01128-8/h0050http://refhub.elsevier.com/S0950-0618(13)01128-8/h0050http://refhub.elsevier.com/S0950-0618(13)01128-8/h0050

410 W.-J. Park et al. / Construction and Building Materials 53 (2014) 403410[11] Dubravka Bjegovic, Dunja Mikulic, Dalibor Sekulic. Non-destructive corrosionrate monitoring for reinforced concrete structures: 15th World Conf. on Non-Destructive Testing, 2000. No. 750. .

[12] Raupach M, Schiebl P. Macrocell sensor systems for monitoring of thecorrosion risk of the reinforcement in concrete structures. NDT&E Int2001;34:43542.

[13] [14] Germann, CorroWatch Multisensor & ERE 20 Reference Electrode: Germann

Instruments A/S, Copenhagen; 2008. .[15] Bjegovic D, Meyer JJ, Mikulic D, Sekulic D. Corrosion measurement in concrete

utilizing different sensor technologies. NACE International; 2003: No. 03435.[16] Climent-Llorca Miguel A, Viqueira-Prez Estanislao, Lpez-Atalaya Ma Mar.

Embeddable Ag/AgCl sensors for in-situ monitoring chloride contents inconcrete. Cem Concr Res 1996;26(8):115761.

[17] Montemor MF, Alves JH, Simes AM, Fernandes JCS, Loureno Z, Costa AJS,et al. Multiprobe chloride sensor for in situ monitoring of reinforced concretestructures. Cem Concr Compos 2006;28(3):2336.

[18] Bjegovic D, Miksic B, Stehly R. Test Protocols for Migrating Corrosion Inhibitors(MCI) in Reinforced Concrete, Emerging Trends in Corrosion Control -Evaluation, Monitoring, Solutions. Akademia Books International and NACEInternational, New Delhi, India 1999;1:318.

[19] Materials Science of Concrete I, Ed. J.P. Skalny, American Ceramic Society:Westerville; 1989.[20] Fontana MG. 3rd Int. Congr. on the Corrosion Engineering by McGraw-Hill;1986. p. 1419.

[21] Jones DA. Principle and Prevention of Corrosion. Prentice Hall; 1997. p. 44041.

[22] .[23] Duff GS, Farina SB. Development of an embeddable sensor to monitor the

corrosion process of new and existing reinforced concrete structures.Construct Build Mater 2009;23:274651.

[24] Quandt E, Halene C, Holleck H, Feit K, Kohl M, Schlobmacher P, et al. Sputterdeposition of TiNi, TiNiPd and TiPd films displaying the two-way shape-memory effect. Sens Actuator A-Phys 1996;53(13):4349.

[25] Kim Young-Geun, Li Seon Yeob, Jung Sungwon, Lee Seong-Min, Kim Jiyoung,Kho Young-Tai. Corrosion behaviors of sputter-deposited steel thin film forelectrical resistance sensor material. Surf Coat Technol 2006;201(34):17318.

[26] Iwamori Satoru, Yoshino Kiyoshi, Matsumoto Hiroyuki, Noda Kazutoshi,Nishiyama Itsuo. Active oxygen sensors used a quartz crystal microbalance(QCM) with sputter-coated and spin-coated poly(tetrafluoroethylene) thinfilms. Sens Actuator B-Chem 2012;171172:76976.

[27] Jung Sungwon, Li Seon Yeob, Kim Young-Geun. Corrosion properties ofsputter-deposited steel thin film for electrical resistance sensor material.Electrochem Commun 2006;8:65864.

http://refhub.elsevier.com/S0950-0618(13)01128-8/h0060http://refhub.elsevier.com/S0950-0618(13)01128-8/h0060http://refhub.elsevier.com/S0950-0618(13)01128-8/h0060http://refhub.elsevier.com/S0950-0618(13)01128-8/h0080http://refhub.elsevier.com/S0950-0618(13)01128-8/h0080http://refhub.elsevier.com/S0950-0618(13)01128-8/h0080http://refhub.elsevier.com/S0950-0618(13)01128-8/h0085http://refhub.elsevier.com/S0950-0618(13)01128-8/h0085http://refhub.elsevier.com/S0950-0618(13)01128-8/h0085http://refhub.elsevier.com/S0950-0618(13)01128-8/h0090http://refhub.elsevier.com/S0950-0618(13)01128-8/h0090http://refhub.elsevier.com/S0950-0618(13)01128-8/h0090http://refhub.elsevier.com/S0950-0618(13)01128-8/h0090http://refhub.elsevier.com/S0950-0618(13)01128-8/h0105http://refhub.elsevier.com/S0950-0618(13)01128-8/h0105http://refhub.elsevier.com/S0950-0618(13)01128-8/h0115http://refhub.elsevier.com/S0950-0618(13)01128-8/h0115http://refhub.elsevier.com/S0950-0618(13)01128-8/h0115http://refhub.elsevier.com/S0950-0618(13)01128-8/h0120http://refhub.elsevier.com/S0950-0618(13)01128-8/h0120http://refhub.elsevier.com/S0950-0618(13)01128-8/h0120http://refhub.elsevier.com/S0950-0618(13)01128-8/h0125http://refhub.elsevier.com/S0950-0618(13)01128-8/h0125http://refhub.elsevier.com/S0950-0618(13)01128-8/h0125http://refhub.elsevier.com/S0950-0618(13)01128-8/h0125http://refhub.elsevier.com/S0950-0618(13)01128-8/h0130http://refhub.elsevier.com/S0950-0618(13)01128-8/h0130http://refhub.elsevier.com/S0950-0618(13)01128-8/h0130http://refhub.elsevier.com/S0950-0618(13)01128-8/h0130http://refhub.elsevier.com/S0950-0618(13)01128-8/h0135http://refhub.elsevier.com/S0950-0618(13)01128-8/h0135http://refhub.elsevier.com/S0950-0618(13)01128-8/h0135

Monitoring method for the chloride ion penetration in mortar by a thin-film sensor reacting to chloride ion1 Introduction2 Consideration of existing literature3 Sensor development3.1 Sensor system using a screen printer3.2 Sensor development methods3.2.1 Design of sensor3.2.2 Sensor solution mixing

3.3 Sensor response test in NaCl solution

4 Sensor response tests in mortar4.1 Sensor response test for inner salinity4.2 Sensor response test for airborne chlorides

5 Results5.1 Result of sensor experiment in inner salinity5.2 Result of sensor experiment on airborne chlorides

6 ConclusionsAcknowledgementReferences