Multi- Electrode Monitoring of Transient Anodic Events in SCC Initiation Stage: Methodology Development
公開日:
カテゴリ: 第3回
1. Introduction
Recently Electro-chemical noise transients have found widespread application for monitoring different kind of corrosion and its kinetics. Most of the works involve understanding kinetics through electrochemical current transient, while others were interested in characterizing the noise signals for different kinds of corrosion or film repassivation processes!,2,4-7. But this method has difficulty at low conductivity that representing to BWR water chemistry. Present work uses electrochemical potential transients rather than current transient to monitor stress corrosion crack initiation event and its location, as it is possible to measure potential at low conductivities.locations as indicated by detection method. So it is important to locate the initiation site in 3 dimensional space. In current work 6 titanium reference electrodes were arranged around a round bar SUS 316 test piece and slow strain rate tests were conducted in 10+mol/lit Na2S203 solution at ambient temperature and pressure under atmospheric oxygen concentration. Small amount of Na2S2O3 is enough to maintain desire level of conductivity and act as an aggressive species for SCC initiation. The potential transients obtained during SSRT test was used as input for computer simulation method, which calculates the location of initiation site.
2. Experiment
SCC has been reported to occur in many plant equipments even with low carbon contents. Due to stochastic nature of SCC it is very difficult to make a maintenance plan to prevent its occurrence. So there is a necessity for detection and location of SCC at its initial stage so that preventive action can be taken only at those2.1. Procedure Stainless steel type SUS 316 with chemical composition as shown in table 2 was used as test piece. It was solution annealed at 1250 °C for 3 hours and water quenched to get bigger grain size. Grains of size as big as 350 um with average grain size of around 100 um were obtained. The purpose of bigger grain is to get SCC initiation sites separated from each other by big distance to facilitate ease of computer simulation method which at present hasPOC: Yutaka Watanabe, Management of Science & Technology, Graduate School of Engineering, Tohoku University, 01Aoba, Aramaki, Sendai, Tel 022-7956896, e-mail: yutaka@rift.mech.tohoku.ac.jp ? 144 ? resolution of around 100 um.Dc si Mn P s Ni Cr Mo Wt% 0.03 0.76 1.18 0.032 0.001 | 12.19 17.53 | 2.05Table 2 Material was then sensitized at 624 °C for 100 hours. Round bar specimen with gauge diameter 4mm and 20mm length were fabricated there after. Test piece was mirror polished up to 1 um diamond paste. A chamber made up of polycarbonate, was made to hold 6 platinum wires around the test piece at 90° to each other in order to cover maximum space around the test piece and minimize the calculation effort. Low conductive solution around 5 uS/cm solution conductivity, was prepared by adding 10 mol/lit Na2S2O3.2.2. Dummy anode test Before the actual SSRT could carried out, series of dummy anode tests were conducted in similar test condition and by applying anodic current to a 7th titanium wire placed at a predefined location, to simulate local anode on the round bar specimen. The experimental setup was shown in fig 2.3. A potentiostat with function generator was used to give anodic current of different amplitude and duration. The potential transients of 6 reference electrodes were recorded by 6 KEITHLEY nano-voltmeters.-1452.3. Actual anode test To generate natural occurring local anode, test piece was subjected to slow strain rate test of about 1.67 X 10 c/s to that passive film could break by mechanical straining and Na2S203 solution could initiate corrosion at local film breaking sites at sensitized grain boundaries. Free corrosion potentials were measured by each nano-voltmeters and straining was started after potential attained equilibrium with less fluctuation (<1 mV), after exposing the test piece to test solution for 1 day. So a constant base line potential was obtained from each reference electrode. Whenever there is film breaking event or SCC initiation, each reference electrodes sensed a transient in potential with different peak values. Test was terminated when the peak value was above certain predetermined value (50mV). SEM photographs were taken before and after the test to find out the actual position of crack initiation sites and compared against computer simulation method.divided into smaller finer grids. To calculate the correct location of the anodic site the peak ratio of each channel the initial position of local anode is decided and fed to the numerical simulation program. Potential distributions of 4 neighboring nodes were calculated using Finite element method and the error was determined using the formula (1):Error = (Poeht / Pche ? 1)2 + (Pobo / Pee ? 1)2 + (Patac / Pond ? 1)2-1Then next calculation point was shifted in the direction of minimum error. This is continued till convergence is achieved. Finally true local anode position is found out 4. Results and discussionIn order to test the validity of the computer simulation method, simulation was made to locate the dummy anode position by taking the potential peaks of each reference electrodes during dummy anode tests. Fair amount of accuracy in simulation method, was obtained in locating the dummy anode position. Simulation was carried out for different conductivities and current profiles. The results of the simulations are shown in table 4. Position of local anode was defined by polar coordinates with respect to one reference point on test piece.5uACoordi- | 30uS/cm 5uS/cm / 5uS/cm nates| 5uAlua | triangular triangular triangularO?(rad.) 3.393 Actual＝= (mm) | 13 O?(rad.) 3.4 3.583.31 Simulation＝= (mm) | 13.74 12.99 12.94 Table 4 Simulation result and actual location of dummylocal anode. For finding location of natural occurring local anode during SSRT, one set of potential transients from 6 reference electrodes were used as input into the simulation program. One such set is shown in fig 4 with base line potential shifted to zero. Test was stopped after few transients occurred and SEM photographs were taken all over the exposed gauge section to find the actual location of crack initiation site.As can be seen from the figure, there is differences inpotential transient peaks of each electrodes owing to their distance from the crack initiation site. These peak values were used as input to simulation program to reduce the calculation space to a smaller region where there is possibility of presence of local anodic site. But there is a discrepancy in the location of anodic site as predicted by simulation method and that of actual anodic site, as revealed by SEM photographs. There are few reasons for such discriPeak 10.01-0.02-0.03-- Ch1-0.04Potential, vCh2-0.07SSRT with chamber SUS 316 1250_34 SA, 624 ?_fool Setsitized conductivity 5uScmCh3 Ch4Ch5- Ch6 -0.08 -0.09 -0.10+ 30800 31000 31200 31400 31600Time, sec Fig 4 Potential transients from SCC initiation site sensed by6 electrodes pancy. One was being, variation of local conductivity around the anodic site from the bulk of solution. The simulation program doesn't take into consideration such variation. Another source of error may be alignment of chamber while fitting it manually. Also in computation method there were some assumptions like double layer capacitance and cathodic site reaction to have fixed properties which may not hold true during actual crack initiation site. So there is a need for more detailed calculation method to include variation in local chemistry and geometry. 5. ConclusionIn this paper, electrochemical potential transients from multiple electrodes, has been used as a tool for locating SCC initiation site. The 6 electrode arrangement around the electrodes reduces the computation time and effort. Anode position could be successfully calculated with fair amount of accuracy in dummy anode test conducted in varying test condition. But same kind of accuracy could not be obtained for location of natural occurring anode during SSRT. The main reason of this discrepancy was attributed to change in146local chemistry during natural occurring anode, while there was no such change during dummy anode test. However this technique has a great potential in locating anodic sites. Further improvement of simulation code would help in achieving desire level of accuracy in locating SCC initiation site. Thus present methodology has a great role to play in risk based maintenance of critical plant equipments, as one can be able to visualize the location of the crack initiation event well before its propagation. AcknowledgementsThis study was conducted under the sponsorship of JNES open application research project for enhancing the basis of nuclear safety. (JNES: Japan Nuclear Energy Safety Organization)ReferencesVol.49, 2004, pp.2795-2801. [3] Manahan, M., Mcdonald, D., “Determination of fate ofCurrent in Stress Corrosion Cracking of Sensitized type 304SS in high temperature aqueous system”, CorrosionScience, Vol.37, No.1, 1995, pp.189-208 [4] Hickling, J., “Use of Electrochemical Noise to detectStress Corrosion Crack initiation in simulated BWR environments”, Material and Corrosion, Vol. 49, 1998,pp.651-658. [5] Watanabe, Y., Kondo, T., “Current and PotentialFluctuation Characteristics in Intergranular Stress Corrosion cracking processes of Stainless Steel”,Corrosion, Vol. 56, No. 12, 2000, pp1250-1255. [6] Song, F.M., Raja, K. S., “A Film Repassivation Kineticmodel for Potential-controlled Slower ElectrodeStraining”, Corrosion Science, Vol.48,2006, pp285-307. [7] Zhou, X. Y., Lvov, S.N., “Quantitative Evaluation ofGeneral Type 304 Stainless Steel in Sub-critical and Super critical Aqueous Solution via Electrochemical Noise Analysis”, Corrosion Science, Vol. 44, 2002, pp 841-860.[1] Watanabe, Y., Shoji, T., “Electrochemical NoiseCharacteristics of IGSCCC in Stainless Steels in Pressurized High-temperature Water”, Corrosion '98,paper no. 129, 1998. [2] Leban, M., Bajt, Z., “Detection and Differentiationbetween Cracking Processes based on Electrochemical and Mechanical Measurements“, Electrochemica Acta.147“ “ Multi- Electrode Monitoring of Transient Anodic Events in SCC Initiation Stage: Methodology Development“ “Sujit BIDHAR,Yutaka WATANABE,T. UCHIMOTO
Recently Electro-chemical noise transients have found widespread application for monitoring different kind of corrosion and its kinetics. Most of the works involve understanding kinetics through electrochemical current transient, while others were interested in characterizing the noise signals for different kinds of corrosion or film repassivation processes!,2,4-7. But this method has difficulty at low conductivity that representing to BWR water chemistry. Present work uses electrochemical potential transients rather than current transient to monitor stress corrosion crack initiation event and its location, as it is possible to measure potential at low conductivities.locations as indicated by detection method. So it is important to locate the initiation site in 3 dimensional space. In current work 6 titanium reference electrodes were arranged around a round bar SUS 316 test piece and slow strain rate tests were conducted in 10+mol/lit Na2S203 solution at ambient temperature and pressure under atmospheric oxygen concentration. Small amount of Na2S2O3 is enough to maintain desire level of conductivity and act as an aggressive species for SCC initiation. The potential transients obtained during SSRT test was used as input for computer simulation method, which calculates the location of initiation site.
2. Experiment
SCC has been reported to occur in many plant equipments even with low carbon contents. Due to stochastic nature of SCC it is very difficult to make a maintenance plan to prevent its occurrence. So there is a necessity for detection and location of SCC at its initial stage so that preventive action can be taken only at those2.1. Procedure Stainless steel type SUS 316 with chemical composition as shown in table 2 was used as test piece. It was solution annealed at 1250 °C for 3 hours and water quenched to get bigger grain size. Grains of size as big as 350 um with average grain size of around 100 um were obtained. The purpose of bigger grain is to get SCC initiation sites separated from each other by big distance to facilitate ease of computer simulation method which at present hasPOC: Yutaka Watanabe, Management of Science & Technology, Graduate School of Engineering, Tohoku University, 01Aoba, Aramaki, Sendai, Tel 022-7956896, e-mail: yutaka@rift.mech.tohoku.ac.jp ? 144 ? resolution of around 100 um.Dc si Mn P s Ni Cr Mo Wt% 0.03 0.76 1.18 0.032 0.001 | 12.19 17.53 | 2.05Table 2 Material was then sensitized at 624 °C for 100 hours. Round bar specimen with gauge diameter 4mm and 20mm length were fabricated there after. Test piece was mirror polished up to 1 um diamond paste. A chamber made up of polycarbonate, was made to hold 6 platinum wires around the test piece at 90° to each other in order to cover maximum space around the test piece and minimize the calculation effort. Low conductive solution around 5 uS/cm solution conductivity, was prepared by adding 10 mol/lit Na2S2O3.2.2. Dummy anode test Before the actual SSRT could carried out, series of dummy anode tests were conducted in similar test condition and by applying anodic current to a 7th titanium wire placed at a predefined location, to simulate local anode on the round bar specimen. The experimental setup was shown in fig 2.3. A potentiostat with function generator was used to give anodic current of different amplitude and duration. The potential transients of 6 reference electrodes were recorded by 6 KEITHLEY nano-voltmeters.-1452.3. Actual anode test To generate natural occurring local anode, test piece was subjected to slow strain rate test of about 1.67 X 10 c/s to that passive film could break by mechanical straining and Na2S203 solution could initiate corrosion at local film breaking sites at sensitized grain boundaries. Free corrosion potentials were measured by each nano-voltmeters and straining was started after potential attained equilibrium with less fluctuation (<1 mV), after exposing the test piece to test solution for 1 day. So a constant base line potential was obtained from each reference electrode. Whenever there is film breaking event or SCC initiation, each reference electrodes sensed a transient in potential with different peak values. Test was terminated when the peak value was above certain predetermined value (50mV). SEM photographs were taken before and after the test to find out the actual position of crack initiation sites and compared against computer simulation method.divided into smaller finer grids. To calculate the correct location of the anodic site the peak ratio of each channel the initial position of local anode is decided and fed to the numerical simulation program. Potential distributions of 4 neighboring nodes were calculated using Finite element method and the error was determined using the formula (1):Error = (Poeht / Pche ? 1)2 + (Pobo / Pee ? 1)2 + (Patac / Pond ? 1)2-1Then next calculation point was shifted in the direction of minimum error. This is continued till convergence is achieved. Finally true local anode position is found out 4. Results and discussionIn order to test the validity of the computer simulation method, simulation was made to locate the dummy anode position by taking the potential peaks of each reference electrodes during dummy anode tests. Fair amount of accuracy in simulation method, was obtained in locating the dummy anode position. Simulation was carried out for different conductivities and current profiles. The results of the simulations are shown in table 4. Position of local anode was defined by polar coordinates with respect to one reference point on test piece.5uACoordi- | 30uS/cm 5uS/cm / 5uS/cm nates| 5uAlua | triangular triangular triangularO?(rad.) 3.393 Actual＝= (mm) | 13 O?(rad.) 3.4 3.583.31 Simulation＝= (mm) | 13.74 12.99 12.94 Table 4 Simulation result and actual location of dummylocal anode. For finding location of natural occurring local anode during SSRT, one set of potential transients from 6 reference electrodes were used as input into the simulation program. One such set is shown in fig 4 with base line potential shifted to zero. Test was stopped after few transients occurred and SEM photographs were taken all over the exposed gauge section to find the actual location of crack initiation site.As can be seen from the figure, there is differences inpotential transient peaks of each electrodes owing to their distance from the crack initiation site. These peak values were used as input to simulation program to reduce the calculation space to a smaller region where there is possibility of presence of local anodic site. But there is a discrepancy in the location of anodic site as predicted by simulation method and that of actual anodic site, as revealed by SEM photographs. There are few reasons for such discriPeak 10.01-0.02-0.03-- Ch1-0.04Potential, vCh2-0.07SSRT with chamber SUS 316 1250_34 SA, 624 ?_fool Setsitized conductivity 5uScmCh3 Ch4Ch5- Ch6 -0.08 -0.09 -0.10+ 30800 31000 31200 31400 31600Time, sec Fig 4 Potential transients from SCC initiation site sensed by6 electrodes pancy. One was being, variation of local conductivity around the anodic site from the bulk of solution. The simulation program doesn't take into consideration such variation. Another source of error may be alignment of chamber while fitting it manually. Also in computation method there were some assumptions like double layer capacitance and cathodic site reaction to have fixed properties which may not hold true during actual crack initiation site. So there is a need for more detailed calculation method to include variation in local chemistry and geometry. 5. ConclusionIn this paper, electrochemical potential transients from multiple electrodes, has been used as a tool for locating SCC initiation site. The 6 electrode arrangement around the electrodes reduces the computation time and effort. Anode position could be successfully calculated with fair amount of accuracy in dummy anode test conducted in varying test condition. But same kind of accuracy could not be obtained for location of natural occurring anode during SSRT. The main reason of this discrepancy was attributed to change in146local chemistry during natural occurring anode, while there was no such change during dummy anode test. However this technique has a great potential in locating anodic sites. Further improvement of simulation code would help in achieving desire level of accuracy in locating SCC initiation site. Thus present methodology has a great role to play in risk based maintenance of critical plant equipments, as one can be able to visualize the location of the crack initiation event well before its propagation. AcknowledgementsThis study was conducted under the sponsorship of JNES open application research project for enhancing the basis of nuclear safety. (JNES: Japan Nuclear Energy Safety Organization)ReferencesVol.49, 2004, pp.2795-2801. [3] Manahan, M., Mcdonald, D., “Determination of fate ofCurrent in Stress Corrosion Cracking of Sensitized type 304SS in high temperature aqueous system”, CorrosionScience, Vol.37, No.1, 1995, pp.189-208 [4] Hickling, J., “Use of Electrochemical Noise to detectStress Corrosion Crack initiation in simulated BWR environments”, Material and Corrosion, Vol. 49, 1998,pp.651-658. [5] Watanabe, Y., Kondo, T., “Current and PotentialFluctuation Characteristics in Intergranular Stress Corrosion cracking processes of Stainless Steel”,Corrosion, Vol. 56, No. 12, 2000, pp1250-1255. [6] Song, F.M., Raja, K. S., “A Film Repassivation Kineticmodel for Potential-controlled Slower ElectrodeStraining”, Corrosion Science, Vol.48,2006, pp285-307. [7] Zhou, X. Y., Lvov, S.N., “Quantitative Evaluation ofGeneral Type 304 Stainless Steel in Sub-critical and Super critical Aqueous Solution via Electrochemical Noise Analysis”, Corrosion Science, Vol. 44, 2002, pp 841-860.[1] Watanabe, Y., Shoji, T., “Electrochemical NoiseCharacteristics of IGSCCC in Stainless Steels in Pressurized High-temperature Water”, Corrosion '98,paper no. 129, 1998. [2] Leban, M., Bajt, Z., “Detection and Differentiationbetween Cracking Processes based on Electrochemical and Mechanical Measurements“, Electrochemica Acta.147“ “ Multi- Electrode Monitoring of Transient Anodic Events in SCC Initiation Stage: Methodology Development“ “Sujit BIDHAR,Yutaka WATANABE,T. UCHIMOTO