Bundesanstalt für Materialforschung und -prüfung

International Symposium (NDT-CE 2003)

Non-Destructive Testing in Civil Engineering 2003
Start > Contributions >Posters > Corrosion: Print

BEM ANALYSIS ON HALF-CELL POTENTIAL MEASUREMENT FOR CORROSION ESTIMATION

Veerachai Leelalerkiet; Kumamoto University, Kumamoto, Japan
Je-Woon Kyung; Kumamoto University, Kumamoto, Japan
Masaru Yokota; Shikoku Research Institute Inc., Takamatsu, Japan
Masayasu Ohtsu; Kumamoto University, Kumamoto, Japan

ABSTRACT

Corrosion of reinforcement in reinforced concrete occurs due to an electro-chemical reaction. Accordingly, half-cell potential measurement is employed to estimate corrosion of reinforcing steel-bars (rebars). Applying the three-dimensional boundary element method (3D-BEM), potential distribution and current flow from the anodic region to the cathodic region on rebars are analyzed. The analysis simulates the half-cell potential measurement by a double-electrode probe. Thus, the influence of voids on the potential distribution and the current flow is investigated. Results of current flows reveal the presence of macro-cell mechanism and the generation of micro-cell mechanism. These mechanisms between the cathodic and the anodic regions could lead to the fact that the half-cell potential measurement is not readily applicable to estimate the corrosion near voids.

1. INTRODUCTION

In concrete structures, reinforcing steel bars (rebars) in concrete normally do not corrode because of a passive oxide film (with high pH) on the surface of the steel. However, the corrosion may occur when these structures are subjected to salt attack. Chloride ions penetrate from the surface of concrete, reach to rebars and break the passive film. As a result, the corrosion of rebar nucleates and develops cracks in reinforced concrete (RC) structures. Thus, early monitoring is necessary to detect the corrosion before cracking becomes visible and critical. To this end, nondestructive evaluation (NDE) methods are recently applied to estimate the corrosion at early stage. So far, a half-cell potential measurement is widely applied to obtain the information on the probability of corrosion [1, 2, 3]. However, the shortcoming of the measurement results from the fact that the potentials are measured not near rebars but on concrete surface [1]. Thus, compensation is unavoidable to get more reliable results. In the present paper, three-dimensional boundary element method (3D-BEM) is applied to study the results obtained from the half-cell potential measurement. Potential distribution and current flow are analyzed for identifying the corroded areas. In order to compare with analytical results, tests of RC slabs subjected to chloride penetration under the wet and dry conditions are conducted.

2. EXPERIMENT AND ANALYSIS

2.1 Specimen
Two RC slabs of dimensions 101 cm x 57 cm x 10 cm were made. One is an ordinary concrete slab and the other contains voids. To simulate voids, two styrofoam plates of dimensions 11 cm x 33 cm x 0.5 cm were embedded inside concrete at 0.75 cm cover-thickness. All rebars are 16 mm in diameter. Transverse and longitudinal rebars were embedded at 2 cm cover-thickness and 3.6 cm from the bottom surface, respectively. Locations of voids and the arrangement of rebars are illustrated in Fig.1. To accelerate the macro-cell mechanism, the left-hand side of the specimen in Fig.1 was made of concrete mixed with fresh water and the right-hand side was mixed with 3%NaCl solution. Mixture proportion of the concrete is given in Table 1. The half-cell potential measurement was conducted by using a portable corrosion-meter (SRI-CM-II) as shown in Fig.2. The bottom surface of concrete is divided into 45 meshes as illustrated in Fig.3, where the half-cell potentials were measured from No.1 through No.45 at the center of the mesh. After the measurement, the potentials were estimated as the probability of corrosion based on ASTM C876 standard [4].

Fig 1: Sketch of concrete slab and locations of rebars and voids (unit in mm).

Maximum gravel size (mm) W/C (%) s/a (%) Weight of unit volume (kg/m3) NaCl/W (%) Slump (cm) Air (%)
Water Cement Fine aggregate Coarse Aggregate
105543.1178323725114538.58
Table 1: Mixture proportion of concrete.

2.2 Boundary element method
The concrete specimen was assumed as homogeneous in BEM. Solution u(x) at boundary point x is solved by the boundary integral equation [5],

(1)

where y is also located on the boundary S surrounding the concrete. G(x,y) is the fundamental solution. In order to simulate the macro-cell phenomenon, each concrete slab is referred to electrically as two regions. The left-hand side corresponds to the cathode and the right-hand side to the anode. Thus, the potentials on rebars are prescribed as - 0.1 volts for the cathodic region and - 0.3 volts for the anodic region, respectively. By applying BEM solution, the unknown potentials and currents are calculated.


Fig 2:
Portable corrosion meter (SRI-CM-II).

Fig 3:
Measuring points in dry condition.

Fig 4:
Saturated in a tub of 3 % NaCl solution.

2.3 Test procedure
After 28-day cured in fresh water, the half-cell potentials were firstly measured in the specimens. Subsequently, the specimen was weekly dried in ambient temperature and put into a tub of 3%NaCl solution as shown in Fig.3 and Fig.4, respectively. At the end of every week, the half-cell potentials were measured. The test was completed when potentials in dry condition reach to - 350 mV (CSE). This is because, in wet condition, the depletion of oxygen inside concrete may cause high negative potential than the actual values [2, 3].

2.4 Visual inspection and levels of corrosion
Fig 5: Degree of corrosion.
To inspect the corrosion in the specimens, concrete cover was removed after the test. Then, corrosion of rebars was rated as three corrosion levels as illustrated in Fig.5. Level 1 represents no visible rust on the surface. Level 2 shows a little rust at the surface. Level 3 indicates heavy corrosion. It is noted that this corrosion rating was based on visual inspection on the top surface of rebar in which the pitting corrosion was not taken into account, because it is not easy to observed by the naked eye [3].

3. RESULTS AND DISCUSSION

3.1 Potential distribution by BEM
Potential distributions where the electrode is located at position No. 26 in both intact and void models are shown in Fig.6. It is found that the electrode influences the potential distribution only in the vicinity of the electrode. This confirms that potentials inside concrete or flows of current are measured around the rebars near the electrode.

In the intact model, the potentials in either the cathodic region or the anodic region show almost common values for entire area of each part, except the position of the electrode. In contrast, in the void model, the potentials around voids become higher than the areas right over the voids in both the cathode and the anode regions. This implies that the micro-cell mechanism is generated around a void. Locally, the area of the rebar under the void becomes the cathode, while the rebar around the void becomes the anode. Thus, it is roughly concluded that only the macro-cell phenomenon is found in the intact case, while the micro-cell phenomenon is observed due to the presence of voids in addition to the macro-cell.

Fig. 6 Potential distribution on concrete surface in the case of electrode on mesh No.26.
Fig.7 Potential distributions at rebars 1and 6 in the case of electrode on mesh No.26.

The potential distributions inside concrete at elevation views of rebars 1 and 6 in the case of the electrode on mesh No.26 are shown in Fig.7. The effect of the electrode is apparent on rebar 6, because the electrode is placed right over. In the intact model, the affected areas by the electrode in the depth are slightly larger than the void model, while those are almost the same in the reinforcing direction. Similarly to the potentials on concrete surface, the potential distribution is almost uniform in the left-hand and the right-hand sides. In the void model, the potentials are varied at void areas in which potentials are negatively high around the voids as the same as the potentials on concrete surface. Again, it is observed that the macro-cell phenomenon is found in intact model, while both macro-cell and micro-cell mechanisms are present in the void model.

3.2 Current flow by BEM
The flow of current along the surface of rebar is analyzed. A cross-section of the rebar is divided into 4 segments as illustrate in Fig.8. The current ratios out of total are obtained as 4 ratios as shown in Table 2. An upper current ratio is calculated by the summation of currents at segments S1 and S2 and divided by the total current on rebar.

Current ratio Zone of segments
Upper ratio(S1 + S2)/ST
Lower ratio(S3 + S4)/ST
Left- side ratio(S1 + S4)/ST
Right-side ratio(S2 + S3)/ST
Table 2: Current ratio.

*Total current, ST = S1 + S2 + S3 + S4


Fig 8: The segment of rebar.

Fig 9: Cross-over point of rebar on mesh No.11.

As mentioned before, the potentials both on the surface of concrete and inside the concrete are influenced by the electrode (see Figs.6 and 7). Hence, the effective length should be taken into account as illustrated in Figs.9 and 10. The elective lengths of the electrode on meshes No.11 and No.23 are referred to as the shaded lengths.

Fig 10: Center of specimen on mesh No.23.

The electrode places on mesh No.11 exactly over the cross-over point of two rebars as illustrated in Fig.9. The currents on the effective areas of rebars 3 and 1 are illustrated in Fig. 11, where the size and direction of arrows represent the amount and direction of current. It is clearly seen that all current values on rebars are negative, and the current on segments of the side near the electrode is more negative than that of the opposite side. In other words, the electrode has great influence on the current at the near-side of the rebar. Especially in rebar 3 (upper rebar), the upper segments of the rebar have more negative current ratios than those of the lower segments. In the case of the void model, it is observed that the presence of voids causes the decrease in the upper current ratio (UR) of both rebars 3 and 1.


Fig 11: Currents on rebars 3 and 1 in the case of electrode on mesh No.11 (mA).

Fig 12: Currents on rebars 4 and 5 in the case of electrode on mesh No.23 (mA).

The case of the electrode on mesh No.23 is illustrated in Fig.10 and the results are given in Fig.12. It is observed that the currents on segments S1 and S4 of rebar 4 are positive. Due to the influence of electrode, the currents on segments S2 and S3 of rebar 4 become negative. In rebar 5, all current values are negative. The currents on rebars 1 and 2 are quite the same. The positive currents are noticed in the cathodic region, while the negative currents are observed in the anodic region. Accordingly, the current flow are suggested from rebar 4 and rebars 1 and 2 in cathodic region to rebar 5 and rebars 1 and 2 in anode region.

Current flows of the entire rebars are represented by the signs of current values as shown in Fig.13. Compared to the results of visual inspection in Fig.14, it is found that the current flows by BEM analysis identifies the corroded areas. In the intact model, the negative currents on rebars in the anode region correspond to the corroded portions in the right-hand side of the specimen in Fig.14 (a). Although the negative currents are observed in the rebar 3 in the cathodic region, the current values are actually low compared with the others in the anode region. In the void model, the current flows elucidate the micro-cell mechanism around the void areas. High negative currents are found around the voids, while in the rebar areas under the voids the positive currents are observed. Results are also conformable to the actual corroded areas in the void specimens as illustrated in Fig.14 (b). Thus, it is concluded that current flow analyzed show good agreement with corroded areas by visual inspection.

Fig 13: Current flows of the entire rebar in the case of the electrode on mesh No.23.

Fig 14: Corrosion levels by visual inspection.

3.3 Half-cell potentials measured
Fig 15: Measured potentials on concrete surface (V, CSE).
After 63 and 77 days (including the curing time), half-cell potentials of the intact and the void specimens in dry condition reached to - 350 mV, respectively. The eqipotential contours are shown in Fig.15. It is observed that all potentials are lower than - 300 mV and - 350 mV in the intact and the void specimens, respectively. Referring to the ASTM standard, potentials more negative than - 350mV(CSE) indicate very high probability (more than 90%) of active corrosion. The potentials between - 200mV and - 350 mV imply uncertainty on corrosion. Accordingly, in the intact specimen, it implies that the corrosion probability area is only the right-hand side of slab, while corrosions at the other places are uncertain. In the void specimen, the all rebars are possibly corroded. Compared to the visual inspection and the corrosion levels in Fig.14, it is confirmed that the half-cell potential measurement might lead to erroneous results on the actual corrosion without compensation [1]. In particular, corrosion estimation near the voids is not easy.

CONCLUSION

The potentials and current flows of rebars inside concrete are quantitatively studied by BEM. Results obtained are concluded as follows:

  1. By BEM analysis, it is found that only the macro-cell is found in the intact model, while both the macro-cell and the micro-cell mechanisms are observed in the void model.
  2. The analysis of current flows elucidates the micro-cell mechanism around the voids. The presence of the void causes the micro-cell current around the void position in which the rebar areas around the void become the anode and the rebar areas under the void become the cathode in the micro-cell circuit. This implies that rebars around the void could corrode faster than those under the void.
  3. The results of half-cell potential measurement show the higher negative values than the actual potentials, especially in void specimen. It is found that BEM analysis of potential distributions and current flows identify the corroded areas.

REFERENCES

  1. J.W.Kyung and Ohtsu.M., "Study on Half-Cell Potential Measurement for NDE of Rebar Corrosion", Structural Faults and Repair, London, UK., 2001.
  2. Naish C., Harker A. and Carney R.F.A., Concrete inspection: "Interpretation of potential and resistivity measurements", Proc. Corrosion of Reinforcement in Concrete, Elsevier Applied Science, pp.314-332, 1990.
  3. Qian, S.Y, Chagnon,N., "Evaluation of Corrosion of Reinforcement in Repaired Concrete", Structural Faults and Repair, London, UK.,2001.
  4. ASTM C876 (1991), "Standards Test Method for Half-cell Potentials of Reinforcing Steel in Concrete".
  5. Brebbia.C.A., "Topics in Boundary Element Research", Springer-Verlag, Heidelberg, Vol.3, 1987.
STARTPublisher: DGfZPPrograming: NDT.net