4. Airborne salinity and atmospheric corrosion of steel at Cuba and the Yucatan Peninsula
Airborne salinity can be determined using different methods. In corrosion research the standard method (wet candle method) is established on ISO 9225: 1992 [21]; however, it is not the only method traditionally used. In the case of Cuba it has been widely used the method named as dry plate method, consisting in the employment of a dry cotton fabric of known area exposed under a shed. The amount of chloride deposition on the gauze is determined analytically at the end of the exposure period (2 months) and the deposition rate is calculated.
A report [22] about the simultaneous comparison between values obtained using wet candle and dry plate methods at different corrosion stations in Cuba showed that there is not a good correlation for the rural station having lower values of salinity; however, a good correlation is obtained for stations having higher values of salinity. The following regression equation was obtained: ½ClW:C: ¼ 54:5 þ 1:6½ClD:P:; r ¼ 0:98; P < 0:005 where [Cl]W.C. is the chloride deposition rate determined
using wet candle method and [Cl 195 ]D.P. is the chloride deposition rate determined using dry plate method.
Chloride deposition rate was determined using wet candle and dry plate methods in the Cuban corrosion stations Santiago de las Vegas (rural–urban), Casa Blanca (industrial–marine–urban), Via Blanca (industrial–urban–marine) and Cojimar (marine).
On Fig. 2 it can be noted a significant difference between steel corrosion rate and chloride deposition between the north and the south shores. Weight lost of steel samples was determined according to the methodology reported in [1]. This behaviour is explained because, in general, trade winds in the north shore come from the Ocean and in the south shore from the earth. In addition, cold fronts always come from the north. The territory is flat, presenting only small hills, so Chloride deposition reaches almost all the territory. It can be seen that even in places located at 15–30 km from the north seashore a significant chloride deposition rate is determined (3.4–8.5 mg/m2 d-classified as S1 according to ISO 9223). Sulphur compounds deposition has also some relation with chloride deposition. The higher values correspond to coastal and industrial stations. It means that, taking into account that the determination using alkaline surfaces is sensible to different sulphur compounds; it includes, in addition to sulphur oxides, sulphatescoming in airborne salinity. Another possible sulphur compound determined could be H2S (see Figs. 3 and 4).
A similar behaviour is observed for the east side of the Cuban Isle. A very high annual steel corrosion rate (in this case steel helix specimens were used) and chloride deposition in the north shore and a significantly lower corrosion rate of steel and chloride deposition in the south shore. The methodology of exposure and evaluation was very similar to [1]: weight lost evaluation and exposure in front to the south of the cylinder in which helix specimen was supported. Unfortunately, there is no data corresponding to distances to the south shore lower than 1 km, but the tendency is toward a lower value in this region. It is a wider territory including mountains, that is why inside the territory chloride deposition rate is lower than in the west side of Cuba (classified as so according to ISO 9223). The inner
region of this part of the Cuban territory is the place where the influence of chloride deposition is lower.
On Fig. 5, the annual average chloride deposition rate determined at three zones of the Yucatan Peninsula is presented: Puerto Progreso and Puerto Morelos in the mouth of the Gulf of Mexico and Campeche and Veracruz inside the Gulf of Mexico as function of the distance to the shoreline. As can be seen, chloride deposition is very similar at Puerto Progreso, Puerto Morelos and Veracruz; however, the values of Chloride deposition are significantly lower for Campeche. It could be explained based on the fact that predominant winds in Campeche most of the year are coming from the earth. Another factor is that Campeche Sea usually has no significant movement, that is, very few sea waves. It is a similar situation to the south shore of the Cuban Isle where a lower chloride deposition is determined.
Changes of corrosion rate of steel as function of distance to the shoreline are lower in Campeche regarding Veracruz. Higher values of corrosion of steel are determined at Puerto Progreso and Puerto Morelos at almost the same distance to the shore than in Campeche. It is in perfect agreement with the influence of chloride deposition.
As can be observed on Table 1, corrosivity >C5 is reported always at distances to the shoreline lower than 150 m, excepting data reported for Veracruz (0.8 and 1.0 km).
In the case of samples exposed inside a bay, chloride deposition rate usually diminishes, but very frequently sulphur compounds deposition rate increases, because in general these are industrial sites.
Corrosivity C3 in coastal regions is reported for Campeche, the south shore of Cuban Republic and two places in India. The distance to the shoreline of the sites is always higher than 200 m, excepting the Indian sites.
5. Long term corrosion rate in coastal atmospheres
The behaviour of carbon steel and copper at Cojimar coastal atmospheric test station (Cuba) in the following exposure periods: copper outdoor – 4 years, steel outdoor – 3 years, copper indoor – 3 years, steel indoor – 2 years is presented on Fig. 6. As it can be seen, for this station a very high corrosion rate is determined for steel and copper. In the case of steel, samples sizing 100 mm 150 mm 10 mm were used because the normal size of samples (100 mm 150 mm 1 mm) was not possible to use because before 1 year of exposure samples are completely corroded and disappear.
Data for outdoor exposure were processed and fit to the very well known equation: K ¼ atb where K is the weight loss (g/m2), a the constant and b is
the exponent coefficient (considered as an indication of the protective nature of corrosion products).
The results are presented on Table 2: it is very interesting to note the value of b coefficient for steel. It is over the normal range of 0–1, indicating a marked acceleration of corrosion rate with time. For example, weight loss of steel corresponding to the first year of exposure is 2019.1 and 7000.0 g/m2 for the second year, indicating a notable acceleration with time. That is not the case for copper, because weight loss determined for the first year was 43.1 and 54.4 g/m2 for the second year. A remarkable difference in long term corrosion behaviour of steel and copper is observed. Outdoor corrosion of steel accelerates with time, but indoor corrosion not. In the case of copper, there is no acceleration of outdoor corrosion with time, but the difference in corrosion rate in outdoor and indoor conditions is much smaller than for steel.
The same acceleration is reported for Campeche CRIP station during 1 year’s exposure in the period 1996–1997. In the semester March 1996 to September 1996 weight loss of steel was of 431.4 g/m2 and in the period March 1996 to March 1997 it was of 2257.3 g/m2. This station is very near to the seashore (4 m). Data accumulated for other years at this station did not reach such acceleration rate. Acceleration rate may change depending on the extent of rain according to [26].
6. Corrosion aggressivity in coastal sheltered and indoor atmospheres
The influence of chloride deposition in corrosion rate could be higher in case of sheltered conditions, because rain and other precipitations do not clean the surface and chloride and other contaminants remain in the metallic surface. In outdoor conditions, chloride and other pollutants deposit on the surface, but rain and other precipitations periodically clean the surface. When the metallic surface is sheltered, the same quantity of chloride and other pollutants deposits on the surface, but it is not cleaned by precipitations, so an accumulation of pollutants takes place and if the relative humidity is high, even a higher corrosion rate than outdoors could be determined.
It is not always the case, as can be seen on Table 3 [3]. It can be observed that corrosion of copper in sheltered conditions is higher than outdoor, it means that corrosion aggressivity increases when copper does not receive the cleansing effect of rain. Even in a ventilated shed, copper corrosion increases faster in time respecting outdoor conditions. Aluminium is another metal in which corrosion rate is higher under a shed. This behaviour is probably due to the accumulation of contaminants in the surface that makes possible a higher water adsorption and produces conditions that does not stabilizes the formation of a protective layer as it usually occurs in outdoor conditions.