3.5E+6 i
3.0E+6 ~ 3.0E~ [ (1 )
k
2.5E+6 ~ i 2
~ ( )
~" 2.0E+6 L
" I o oE+o . . . . . . , , ~ - . i
o
-r 1.fiE+6 ~ 9
5.0E+5 f / / Reinforcing Steel Bar
t * l k ~ & ~ ~ t e d Ca(OH)2 S~176
+ I ~ (1)
0.0E 0 ~ - , - ~ , , ~ 9 , . . . . . .
; !
' i
I
i I !
: i
i i
0.0E+0 5.0E+5 1.0E+6 1.5E+6 2.0E+6 2.5E+6 3.0E+6 3.5E+6
R e a l ( O h m )
Figure 6 - EIS Nyquist diagram for miM steel in saturated Ca(OH)2 solution." (1) without polarization; (2)polarization in the active and passive regions (3).
It can be seen, in all Nyquist plots, that the spectra (1) and (3) are different, even though they both correspond to passive behavior: the spectra (1) represents the natural formation of the passive layer without application of polarization, while the spectra (3) represents the formation of the passive film induced by polarization. Also the results show that all the spectra (3) display a more capacitive behavior than all spectra (1).
Therefore, there is a difference between the passive films because of the way they are formed in each of the model solutions, and it is more evident when the impedance spectra in the Bode type (Figures 9-14) are observed. Generally, three stages can be seen in the Bode diagrams: a) resistive behavior at very high frequencies; b) capacitive behavior in the majority o f the frequency range; and c) a little resistive behavior at very low frequencies.
For the reinforcing steel in cement extract solution (Figures 9,11 and 13) the spectra (3), with polarization, exhibit a behavior that goes like this: in the solution free of
VI~LEVA AND CEBADA ON REINFORCING MILD STEEL 179
chlorides (Figure 9) the steel shows a capacitive behavior that stays the same in the presence ofNaC1 5g/1 (Figure I 1), but with the addition of NaC1 10 g/1 (Figure 13) the impedance value is smaller than the other two. This kind of behavior matches with the one observed in the polarization curves (Figure 4) for the passive regions.
Figure 7 - EIS Nyquist diagram for mild steel in cement solution with 5 g/l NaCh (1) without polarization; (2)polarization in the active and passive regions (3).
180 MARINE CORROSION IN TROPICAL ENVIRONMENTS
Figure 8 - EIS Nyquist diagram for miM steel in saturated Ca(OH)2 solution with 5 gJl NaCl: (1) without polarization; (2) polarization in the active and passive regions (3).
In the saturated Ca(OH)2 solution the Bode spectra (3) (Figures 10,12 and 14) show that in the presence of 5 g/1 (Figure 12) chloride ions the capacitive behavior of steel diminishes abruptly its impedance value compare with the one obtain in the absence of chloride ions (Figure 10). With the addition of 10 g/l NaCI (Figure 14) the impedance value increases slightly compared with the one for 5 g/l. In the same way as before, this behavior matches with the one displays in the corresponding polarization curve (Figure 3).
The passive behavior of the reinforcing steel without the external polarization is shown in the Figures 9-14, spectra (l). The capacitive behavior (impedance value) of the steel in cement solution with the addition of 5g/l NaC1 (Figure 11) can be seen to be decreased compared with the one observed in the absence of chloride ions (Figure 9),
VI~LEVA AND CEBADA ON REINFORCING MILD STEEL 181
but exhibits no Change with the addition of 10 g/1 NaC1 (Figure 13). For the steel in the saturated Ca(OH)2 solution the impedance value stays the same without (Figure 10) and with 5 g/l NaC1 (Figure 12), but with 10 g/1 NaC1 (Figure 14) the value drops markedly.
These behaviors are contrary to the ones observed in spectra (3) (Figures 9-14) and in the polarization curves (Figures 3 and 4).
Figure 9 - EIS Bode diagram for mild steel in cement solution: (1) without polarization;
(2) polarization in the active and passive regions (3).
Figure 10 - E I S Bode diagram for miM steel in saturated Ca(OH)e solution: (1) without polarization; (2) polarization in the active and passive regtons (3).
182 MARINE CORROSION IN TROPICAL ENVIRONMENTS
F i g u r e 11 - EIS Bode diagram for mild steel in cement solution with 5 g/l NaCI: (1) without polarization; (2)polarization in the active and passive regions (3).
F i g u r e 12 - EIS Bode diagram for mild steel in saturated Ca(OH)2 solution with 5 g/l NaCI: (1) without polarization; (2)polarization in the active and passive regions (3).
VI~LEVA AND CEBADA ON REINFORCING MILD STEEL 183
F i g u r e 13 - EIS Bode diagram for mild steel in cement solution with 10 g/l NaCh (1) without polarization; (2)polarization in the active and passive regions (3).
F i g u r e 1 4 - EIS Bode diagram for mild steel in saturated Ca(OH)2 solution with 10 g/l NaCh (1) without polarization; (2)polarization in the active and passive regions (3).
184 MARINE CORROSION IN TROPICAL ENVIRONMENTS
Figures 15-17 show the SEM topography images of the layer formed on the reinforcing steel surface in both model solutions with the addition of 10 g/l NaC1 (Figures 16, 17), and SEM of the control specimen surface immersed in no solution at all (Figure 15). The film formed in the cement solution (Figure 16) is apparently more homogeneous than the one form in saturated Ca(OH)2 (Figure 17) and its appearance is different. Also can be seen from the EDAX film surface analysis that the peaks for Ca, Si, C, Cr and oxygen are higher in the cement solution (Figure 18) than the ones for the steel in saturated Ca(OH)2 (Figure 19). Therefore, the passive film formed in each one of the model solutions is different in composition, homogeneity, anticorrosion protectiveness and perhaps thickness. All those observations explain why the two model solutions do not produce similar behavior of steel.
Figure 15 - SEM image of the reinforcing steel surface without being immersed in any solution.
V#LEVA AND CEBADA ON REINFORCING MILD STEEL 185
F i g u r e 16 - SEM image of the passive layer formed on the reinforcing steel surface in the cement extract solution with NaCI 10 g/l, after 21 days of immersion.
F i g u r e 17 - SEM image of the passive layer formed on the reinforcing steel surface in the saturated Ca(OH)2 solution with NaCl lOg~l, after 21 days of immersion,
186 MARINE CORROSION IN TROPICAL ENVIRONMENTS,
6 0 0
550 500- 450 4OO 35O 300 +
2OO ~
IO0 5O :
o I
Ca
o ,, J
i 2
Fe
4 S 6 7 8 9 10
E n e r g y (keV)
F i g u r e 18 - E D A X analysis of the passive layer formed on the reinforcing steel surface in the cement extract solution with NaCI 10 g/l, after 21 days of immersion.
1000' 900 800 700 C 600' O U S00 N T 400, S 300.
100 i Ca
C o ~ r .~i _ . ~ ~
1 2 3 4 s 6
E n e r g y (keV) F e
F i g u r e l 9 - EDAX analysis of the passive layer formed on the reinforcing steel surface in the saturated Ca(OH)2 solution with NaC110 g/l, after 21 days of immersion.
VI~LEVA AND CEBADA ON REINFORCING MILD STEEL 187
Therefore, our future investigation is directed towards further study of the topography and composition of the layers formed on the metal surface immersed in both solutions, at different values of polarization corresponding to the passive state of the metal. On the other hand, EIS circuit equivalent simulation could be done in the cases where it is possible, because of the very complex system metal-interface-model solutions.
Summary
Reinforcing steel bars (mild steel) were studying using electrochemical techniques, in two model alkaline solutions (12.7+0.2, at 24•176 saturated calcium hydroxide (Ca(OH)2) and cement extract (ordinary portland cement, Type I), as such and with the addition of 5 g/1 and 10 g/1 NaC1. The use of electrochemical techniques, such as polarization curves and impedance spectroscopy (EIS), provides a very powerful information related to real corrosion state of the metal immersed in model solutions.
Also some additional SEM and EDAX information is presented about the topography and composition of the films formed on the metal surface in both model solutions.
Conclusions
According to the polarization curves, in the absence of chloride ions, both model solutions (cement extract and saturated Ca(OHh) show similar behavior of reinforcing steel. However, the open circuit potential in the cement solution shows a slightly shift to more positive values with the increasing of the immersion time, that indicates distinct film characteristics in both model solutions. This fact is more pronounce in the presence of chloride ions (5 and 10 g/l).
The impedance spectra obtained with the application of external polarization (corresponding to the passive region of the polarization curves), are in agreement with the behavior of the steel displayed by the polarization curves in the same region,.
although these measurements do not show the real behavior exhibited by the metal in natural conditions (without external polarization). Therefore the EIS spectra registered at open circuit (without external polarization) provide more confident information.
The SEM and EDX surface analysis confirm that there are topography and composition differences in the passive layers formed in each of the model solutions. It seems that the cement extract solution allows the metal to develop a more protective film than the saturated Ca(OH)2.
All the results presented above show that the two model solutions do not produce similar behavior of steel.
188 MARINE CORROSION IN TROPICAL ENVIRONMENTS
Acknowledgments
The authors wish to recognize the financial support provided by Mexican CONACyT grant project 29649U, and also for the scholarship to M. C. Cebada. We appreciate also CINVESTAV-IPN, Unidad M6rida, for allowing us to carry out this research work.
Finally, the authors are very grateful to CINVESTAV-IPN, Unidad Saltillo, especially to Dr. Cecilia Montero Ocampo for performing the SEM and EDAX analysis.
References
[1] Berke, N.S. and Hicks, M.C., "Electrochemical Methods of Determining the Corrosivity of Steel in Concrete," Corrosion Testing and Evaluation: Silver Anniversary Volume, ASTM STP 1000, R. Baboian and S.W. Dean, Eds., American Society for Testing and Materials, Philadelphia, 1990, pp.425-440.
[2] Callaghan, B.G., "The Performance of a 12% Chromium Steel in Concrete in Severe Marine Environments," Corrosion Science, Vol. 35, 1993, p. 1535.
[3] Nagayama, M. and Cohen, M., "The Anodic Oxidation of Iron in a Neutral Solution," Journal of Electrochemical Society, Vol. 110, 1963, p. 670.
[4] Kruger, J. and Calvert, J.P., "Ellipsometric-Potentiostatic Studies of Iron Passivity,"
Journal of Electrochemical Society, Vol. 114, 1967, p.43.
[5] Stainless Steel Rebar International Bulleting, Tullmin Coonsulting, Ontario, Canad~i, 1999.
[6] Castro, P., Veleva, L. and Balanc~in, M., "Corrosion of reinforced concrete in a tropical marine environment and accelerated tests," Construction and Building Materials, Vol. 11, 1997, pp. 75-81.
[7] Veleva, L, Castro, P., Hemandez-Duque, G. and Schorr, M., "The Corrosion Performance of Steel and Reinforced Concrete in a Tropical Humid Climate,"
Corrosion Reviews, 1SSN 0048-7538, M. Schorr, Ed., Freund Publishing House, London, England, 1998, pp. 235-284.
[8] Andrade, C., et al., "Cement Paste Hardening Process Studied by Impedance Spectroscopy," Electrochimica Acta, Vol. 44, 1999, pp. 4313-4318.
[9] Dawson, J.L. and Langford, P.E., "The Electrochemistry of Steel Corrosion in Concrete Compared to Its Response in Pore Solution," The Use of Synthetic Environments for Corrosion Testing, ASTM STP 970, P.E. Francis and T.S.