Studies in Accelerated (Predicted) Versus Observed Anticorrosive Performance for Coastal Industrialized Location.
Summary: Accelerated test for corrosion protection is an industrially accepted norm. Even though no correlations are claimed by the coating manufacturers or the users, the users specify and manufacturers comply with the specifications. Coating's performance in actual use conditions may be unrelated to its accelerated test results. In this work, visual examinations and Scanning electron microscopy (SEM)/Energy dispersive X-ray (EDX) analysis were used to study accelerated (predicted) versus observed anticorrosive performance for coastal industrialized location. Commercially available coating systems based on epoxy-polyamide/polyurethane (P1), epoxy-polyamide/epoxy-amine (P2) and alkyd/alkyd (P3) formulations were applied on mild steel test panels (4"x 6" sizes). Then accelerated testing (ASTM B-117) and natural exposure testing at marine test site in Karachi, Pakistan was performed.
It was found that the accelerated corrosion testing showed much less severity of corrosion compared to natural exposure testing for all the coatings tested. Severe blistering, filiform corrosion spread of corrosion around the scribe and loss of gloss was noticed in natural exposure testing. SEM micrographs suggested that the accelerated testing did not show surface features similar to those noticed after natural exposure testing. Principal component analysis explained a variance of 99.98 %, 99.97 % and 99.29 % for P1, P2 and P3 coating systems respectively. These results clearly indicate that no correlation exists between accelerated test method ASTM B117 and natural testing carried out at coastal industrialized location. Based on the results it is suggested to follow a more stringent protocol (inclusion of highly corrosive bromide, sulfite and nitrate ions in acidic pH (6 to 6.5) of the salt solution) for coastal industrialized locations.
Keywords: Anticorrosive performance, Mild steel, Scanning electron microscopy (SEM)/energy dispersive X- ray (EDX).
Atmospheric corrosion is an important factor responsible for deterioration of metals. This type of corrosion involves aggressive action by the atmosphere which causes deterioration in the properties of metals . Atmospheric corrosion is an electrochemical process . It involves the oxidation of a metal at the anode and reduction of an oxidant at the cathode. The dominant reaction that occurs at cathode is oxygen reduction reaction. The whole process requires the presence of an electrolyte.
The electrolyte is responsible for the completion of an electrochemical circuit between an anode and cathode [3, 4]. M. Natesan, et al. stated that worldwide the overall cost of corrosion is approximately 4-5 % of the gross national product and 20-25 % of this cost could be avoided by the use of right corrosion technology .
Use of coatings is the most effective and economical method of combating atmospheric corrosion [6-9]. Generally there are two methods for testing the anticorrosive performance of coatings i.e., natural exposure testing and accelerated testing [10-12]. Ideally, both types of testing should produce similar degradation mechanisms . Studies have been carried out all over the world for determining the anticorrosive performance and also for predicting the service life of different coatings by accelerated and natural exposure testing [7, 8, 14, 15]. Few papers reported different service life prediction approaches [16-18]. Others specified that the performance of coatings is strictly related to specific location and meteorological parameters [19-20].
Success of the service life prediction approach is related to the development of accelerated weathering techniques that can be used to identify the weathering stresses which play an important role in the degradation process . For this purpose, it is essential to study a correlation between accelerated (predicted) and observed anticorrosive performance for a specific location.
The environment of coastal industrial cities is rather harsh compared to developed cities in the main land. This is worse in the case of developing
countries. The salt content of the air due to spray drying of water droplets from sea waves, pollution due to automobiles, industries that emit SO2, HCl and NOX increase the risk of corrosion of several industrial products that are exposed to this environment. Since the manufacturer can specify only the tests that are available in the standards, the coating supplier has little say in modifying the product. In the process, it is the consumer who suffers. In view of the above, there is a need to carry out such studies in different parts of the world to have a clear vision of correlation between accelerated tests for corrosion protection and actual corrosion protection by using organic coatings. So that the user can identify and then latter rectify the causes in order to avoid losses which have an important impact on the economy of the country.
Besides many incongruities, salt spray test is still the most widely used accelerated test. The purpose of this study is two fold. First is to compare the anticorrosive performance of different coating systems by accelerated (salt spray) and natural exposure testing at coastal industrialized location with the help of visual evaluation, gloss measurement, scribe creep measurement and scanning electron microscopy (SEM). And second is the application of the Principal component analysis on the data obtained from energy dispersive X-ray (EDX) analysis in order to study a correlation between accelerated (predicted) versus observed anticorrosive performance.
Results and Discussion
Visual examination of the anticorrosive performance of different coating systems was done according to ISO standards. The parameters assessed were blistering, rusting, cracking, flaking and filiform
Table-1: Main results of accelerated (salt spray) testing.
Coating###Coating defects after###
system codes###48 hours###72 hours###96 hours###240 hours###320 hours###480 hours###720 hours
The P2 coating system (Mild steel/Epoxy- polyamide primer/ Epoxy-amine topcoat system) with dry film thickness of 191 um showed the formation of some blisters and rust after 48 hours of salt spray testing, (Table-1). Further exposure of the coating system was stopped after 320 hours due to considerable blisters and rust along the scribe. Blistering in the coating system was revealed as the main failure mode. Natural exposure testing indicated the appearance of blisters and rust after 3 months of exposure, (Table-2). After 9 months the coating system was removed due to severe corrosion and coating delamination from the scribed region. Coating delamination from the scribed region appeared to be caused by the formation of underlying corrosion products. Both cracking and flaking were observed and rated according to ISO standards. Filiform corrosion was not observed. Fig. 1 shows the state of scribed region of unexposed, artificially and naturally exposed P2 coating system.
The P1 coating system (Mild steel/Epoxy- polyamide primer/Polyurethane topcoat system) with dry film thickness of 220 um showed the appearance of some blisters and little rust after 96 hours of salt spray testing, (Table-1). Further exposure caused increase in blister density and degree of rusting also. After 720 hours maximum degree of blistering and rusting were observed. Regular filiform corrosion was also observed. Salt spray testing revealed filiform corrosion along the scribed lines as the major mode of failure. Similar system testing indicated little blistering and rusting after 6 months of natural exposure test, (Table-2). After 18 months of exposure, maximum density of blisters and highest degree of rusting caused complete failure of coating system. Coating delamination from the scribed lines
was associated with rust filled coating blisters. The second observed failure mode was filiform corrosion. Regular filiform corrosion was observed. Filaments of filiform corrosion appeared perpendicular to the scribed lines. After natural exposure test there was no indication of cracking while the degree of flaking was rated. Fig. 1. shows the state of scribed region of unexposed, artificially and naturally exposed P1 coating system.
E. Almeida, et al. compared the performance of alkyd, epoxy and polyurethane paints on different substrates after various accelerated and natural exposure tests . They reported that the best behavior was observed for polyurethane paint in both natural atmosphere of high corrosivity and also in the accelerated test. They also found that the intermediate behavior was observed for alkyd paint. Results presented in this study also indicate best performance by polyurethane coating however in contrast alkyd coating showed worst performance in both natural and accelerated testing.
It was apparent from the results of this study that the natural exposure testing caused extremely high degradation of all the coating systems tested as compared to accelerated testing. In addition to this, the major modes of degradation in all the coating systems were completely different in accelerated and natural exposure testing. Ideally, degradation modes must be similar during accelerated and natural exposure testing . B.S. Skerry, et al. reported that coatings usually degrade by the action of light, air, temperature and water and this could be worsened by the presence of atmospheric pollutants . F. Deflorian, et al. concluded that the weathering of coatings depend on meteorological parameters .
Specular gloss measurements were made for a) control (unexposed) coating systems
b) coating systems artificially weathered using salt spray chamber
c) coating systems exposed in natural environment.
Table-2: Annual mean of different environmental parameters.
###Tmax###Tmin###RH at###RH at###Wind speed at###Wind speed at###Amount of precipitation
###0000 UTC###1200 UTC###0000 UTC###1200 UTC
Table-3: The concentration of major pollutants and particulate matter at natural exposure test site (marine).
Results are shown in Fig. 2. These results show less gloss reduction and likely durability of the epoxy-polyamide/polyurethane coating system (P1) in both accelerated and natural environments. These results also suggested that the natural exposure caused severe reduction in gloss in all coating systems as compared to accelerated test performed in salt spray chamber. F.X. Perrin, et al. found that the gloss reduction is due to the loss of organic material of the coating . Thus the severe reduction in gloss as a result of natural exposure testing could be related to high degradation of the binder in natural environment. Literature showed that the reduction in gloss of the alkyd coating was reported by F.X. Perrin, et al. . V.C. Malshe, et al. specified that the aromatic moiety and secondary hydroxyl groups play a vital role in the degradation of epoxy resin . Reduction in gloss of the polyurethane coating was reported by X.F. Yang, et al. in their study .
Degree of Corrosion Around the Scribe
Corrosion around the scribe in different coating systems and Table-5 present the results of degree of corrosion around the scribe in different coating systems. It was apparent from these results that the degree of corrosion around the scribe was many times higher in natural outdoor tested
coating systems. M. Morcillo  stated that the atmospheric pollution causes the presence of soluble salts within the corrosion products layer which in turn promotes under film metallic corrosion. Thus the high corrosion around the scribe after natural exposure testing could be related to the presence atmospheric contaminants chiefly chlorides.
Scanning Electron Microscopy (SEM) and Energy Dispersive X-ray (EDX) Analysis
EDX analysis was performed for the unexposed P3 coating system at two different points on the surface of sample. Results indicated the presence of C, O, Na, Al, Si, S, Cl, Ca, Cr, Cu, Zn and Fe in one EDX spectrum while C, O, Al, Cl, Ca,
Ti, Cu, Zn and Fe in other EDX spectrum taken for the unexposed P3 coating system [(Fig. 4) and (Table-6)]. The elemental compositions obtained for the P3 coating systems after accelerated and natural exposure testing is presented in Fig. 4 and Table-6. Chromium was conspicuously absent in the exposed coating systems probably due to its high water solubility. The atmospheric condensation due to high humidity every morning may have washed it off. Similar results are expected in accelerated salt spray testing. In order to study a correlation between accelerated and natural exposure testing, the Principal component analysis (PCA: a Statistical Method) was applied on the data obtained from EDX analysis. Fig.
5 shows loading plots of the Principal component analysis reported the results of the elaborations carried out with EDX data sets obtained for P3 coating systems. The Principal component analysis of the P3 coating systems explained a variance of 99.29 % when 2 components were considered.
The variance of the first component was explained by the negative and non-significant loadings of the unexposed i, unexposed ii, SST and NET which were not correlated each other. The variance of the second component was explained by the negative and non- significant loadings of the SST and NET, whereas the positive contributions were mainly due to unexposed i and unexposed ii. However, these contributions were also not significant.
Similar to P3 coating system, salt spray testing caused less deterioration in the surface characteristics of the P2 coating system (Mild steel/Epoxy-polyamide primer/ Epoxy-amine topcoat system) and the SEM micrograph showed that the coating surface characteristics, did not appear to be significantly different from the unexposed sample features, (Fig. 6). SEM micrographs also suggested that after natural exposure testing, the coating's surface was severely roughened. The underlying pigment matrix was seen in the micrograph. C. Ocampo, et al. compared the resistance of modified and unmodified epoxy coatings against marine corrosion with the help of SEM-EDX . They found that initial regular surface of the epoxy coating was changed to a rough superficial structure after testing. They related the roughening to the appearance of oxides in the polymeric matrix because the EDX analysis revealed an increase in oxygen contents.
Results obtained in this study were also consistent with their findings. Smooth surface of the unexposed coating was converted in to rough surface after accelerated and natural exposure testing.
EDX analysis showed the presence of C, O, Al, Si, Ti, and Fe in the unexposed P2 coating [(Fig. 6) and (Table-7)]. The elemental compositions obtained for the P2 coating systems after accelerated and natural exposure testing are presented in Fig. 6 and Table-7. Fig. 7. shows loading plots of the Principal component analysis reported the results of the elaborations carried out with EDX data sets obtained for P2 coating systems. The Principal component analysis of the P2 coating systems explained a variance of 99.97 % when 2 components were considered. The variance of the first component was explained by the negative and non-significant loadings of the unexposed, SST and NET which were not correlated each other. The variance of the second component was explained by the negative and non- significant loadings of the NET, whereas the positive contributions were mainly due to unexposed and SST. However, these contributions were also not significant.
SEM micrograph for the P1 coating system (Mild steel/Epoxy-polyamide primer/Polyurethane topcoat system) obtained as a result of salt spray testing showed little degradation of the surface when compared with the unexposed coating surface features, (Fig. 8). Similar system testing by natural exposure revealed drastic deterioration of the coating surface characteristics. Surface roughening because of some depositions or structure was observed in the micrograph. Effects of natural exposure testing and different accelerated testing on the surface of polyurethane coating by SEM were reported by C. Merlatti, et al. . They found that the surfaces of samples exposed at natural exposure sites were rougher than the surfaces of samples tested by accelerated testing. They also found good correlation between SEM studies and gloss measurements. Results presented in this study were completely consistent with their findings.
EDX analysis indicated the presence of C, O, Al, Pb, Ti, Fe, Cu and Zn in the unexposed P1 elemental compositions obtained for the P1 coating systems after accelerated and natural exposure testing is presented in Fig. 8. and Table-8. Fig. 9. shows loading plots of the Principal component analysis reported the results of the elaborations carried out with EDX data sets obtained for P1 coating systems. The Principal component analysis of the P1 coating systems explained a variance of 99.98 % when 2 components were considered. The variance of the first component was explained by the negative and non-significant loadings of the unexposed, SST and NET which were not correlated each other. The variance of the second component was explained by the negative and non-significant loadings of the NET, whereas the positive contributions were mainly due to unexposed and SST. However, these contributions were also not significant.
SEM-EDX analysis is widely employed as a powerful tool for the study of coatings [27-30]. Scanning electron microscopy (SEM) revealed high surface degradation of all the coating systems due to natural exposure testing. EDX analysis as well as the results obtained from PCA indicated lack of correlation between accelerated and natural exposure testing.
Reason for the smooth and featureless surfaces of the unexposed coating systems was well elucidated by B.S. Skerry, et al. . They explained that this could be attributed to the formulation of topcoats as a gloss finish systems with a relatively low pigment volume concentration.
X.F. Yang, et al.  specified the causes for the formation of cracks and increase surface roughening as a result of weathering. According to them, the action of UV radiation, oxygen and humidity on coating material breaks down macromolecules and produce gaseous products, such as carbon monoxide and carbon dioxide, ultimately. Loss of coating material takes place which results in the formation of cracks and pigments erode on the surface.
Principal component analysis (PCA) was applied on the data obtained from EDX analysis in order to study correlation between accelerated and natural exposure testing. The data was processed by using the software Statistica (Version 10). The results were presented on a bi-dimensional plot and the significant loadings were marked when [greater than or equal to]|0.7|. PCA is a method of statistics used to trim down the dimensionality of a data set which contains a number of interrelated variables by maintaining the variation present in the data set.
In most cases, it is assumed that a 100 hours of exposure in salt spray by ASTM B-117 corresponds to about 1 year in the outdoors (Though no one says it with any degree of certainty, neither the coating manufacturer, nor the equipment manufacturers or the coating buyers). Many automotive industries specify 1400 hours of salt spray test for the ED primers by ASTM B-117 in a fond hope that the coating would last about 12-14
years, about the useful life of an automobile. The accelerated corrosion testing completely failed to correlate these norms with the results of natural exposure in marine environment of Karachi, Pakistan where several other factors appear to be causing much more accelerated corrosion. Following observations are note worthy.
1. The accelerated corrosion testing showed much less severity of corrosion compared to natural exposure testing for all the three coatings tested. This is most likely due to the presence of several corrosive components of environment emerging from the mist of salt spray in the atmosphere and pollution due to industry and automobiles. Air, as seen from the environmental data, contains large concentration of particulates; nitrous oxide sulfur compounds and sea salt are the most detrimental.
2. The loss of gloss was much higher in natural exposure testing for all the coatings tested.
3. Severe filiform corrosion was noticed in natural exposure testing compared to accelerated testing. This may be due to diffusion of corrosive salt /salts solution from the edges of the damage that were not present in the mist of accelerated salt spray test.
4. Spread of corrosion was far more severe innatural exposure testing.
5. Blistering was severe in natural exposure testing.
6. Surfaces of all the unexposed coating systems were smooth and featureless.
7. SEM micrographs suggested that the accelerated(salt spray) testing did not show surface features similar to those noticed after natural exposure testing. Accelerated (salt spray) testing revealed less degradation of the coating surface characteristics. In contrast, natural exposure testing caused drastic degradation of the coating surface characteristics.
8. Application of the Principal component analysis(PCA) on EDX data sets explained a variance of99.98 %, 99.97 % and 99.29 % for P1, P2 and P3 coating systems respectively. These results clearly indicate that no correlation exists between accelerated test method ASTM B117 and natural exposure testing carried out at coastal industrialized location.
Based on the above observations, it is felt there is a need to devise a more stringent test protocol for coastal, industrialized locations than ASTM B 117 and it should be used for specifying the coatings for automotive paints. Some of the suggested changes are inclusion of highly corrosive bromide, sulfite and nitrate ions in acidic pH (6 to 6.5 or lower caused by acid rains) of the salt solution. In absence of such test methods, the unsuspecting manufacturers of coating products and buyers of coating products are likely to suffer unexpected damage to their products. We propose to take up this work in collaboration with other researchers located in coastal industrialized locations of the world.
The authors very specially thank Prof. V. C. Malshe [(Retd) Prof. of Paint Technology, Head Surface Coatings Technology, University of Mumbai (India)] for his guidance.
The authors express their gratitude to the following persons, companies, and institutions in Karachi, Pakistan: Anis-Ur-Rehman Siddiqui, Fakhrul Arfin and Dr. M. Nasir uddin Khan, Yousuf Khan, M. Farooq Wahab, Furqan uddin and Dr. Sheikh Mohiuddin, Hino Pak motors limited (Body operation plant) and Berger paints Pakistan limited for providing mild steel and application facilities, ICI paints Pakistan for the gloss measurement facility, Centralized science laboratories for providing SEM facility, Dean, Faculty of Science, University of Karachi, and Dr. Fatima Bi of PCSIR laboratories complex for the financial support and Pakistan Meteorological department. Special thanks to Dr. Badar Munir Khan Ghauri, Director, Pakistan space and Upper Atmosphere Research Commission (SUPARCO), Karachi, Pakistan for his support.
1. K. Barton, Protection against Atmospheric Corrosion: Theories and Methods, New York, NY: Wiley-Interscience, (1976).
2. H. H. Lawson, Atmospheric Corrosion Test Methods, Houston, TX: NACE International (1995).
3. J. Dong, E. Han and W. Ke, Science and Technology of Advanced Materials, 8, 559 (2007).
4. R. Balasubramaniam, A. V. Ramesh Kumar and P. Dillmann, Current Science, 85, 1546 (2003).
5. M. Natesan, G. Venkatachari and N. Palaniswamy, Corrosion Prevention and Control, June, 43 (2005).
6. V. S. Moynot, S. Gonzalez and J. Kittel, Progress in Organic Coatings, 63, 307 (2008).
7. D. Santos, C. Brites, M. R. Costa and M. T. Santos, Progress in Organic Coatings, 54, 344 (2005).
8. E. Almeida, D. Santos, F. Fragata, D. de la Fuente and M. Morcillo, Progress in OrganicCoatings, 57, 11 (2006).
9. P. A. Sorensen, S. Kill, D. Johansen and C. E. Weinell, Journal of Coatings Technology and Research, 6, 135 (2009).
10. M. J. Crewdson and P. Brennan, Journal of Protective Coatings and Linings, September, 17 (1995).
11. T. Bos, Journal of Protective Coatings and Linings, June, 73 (2008).
12. B. S. Skerry and C. H. Simpson, Corrosion, 49,663 (1993).
13. F. X. Perrin, M. Irigoyen, E. Aragon and J. L.Vernet, Polymer Degradation and Stability, 72, 115 (2001).
14. X. F. Yang, J. Li, S. G. Croll, D. E. Tallman and G. P. Bierwagen, Polymer Degradation and Stability, 80, 51 (2003).
15. C. Merlatti, F. X. Perrin, E. Aragon and A.Margaillan, Polymer Degradation and Stability, 93, 896 (2008).
16. S. G. Croll, B. R. Hinderliter and S. Liu, Progress in Organic Coatings, 55, 75 (2006).
17. S. G. Croll and B. R. Hinderliter, Journal of Coatings Technology and Research, 4, 217 (2007).
18. S. G. Croll and B. R. Hinderliter, Journal of Material Science, 43, 6630 (2008).
19. F. Fragata, D. de la Fuente, E. Almeida, D. Santos and M. Morcillo, Journal of Coatings Technology and Research, 4, 75 (2007).
20. F. Deflorian, S. Rossi and M. Fedel, Corrosion Science, 50, 2360 (2008).
21. L. F. E. Jacques, Progress in Polymer Science, 25, 1337 (2000).
22. F. X. Perrin, C. Merlatti, E. Aragon and A.Margaillan, Progress in Organic Coatings, 64, 466 (2009).
23. V. C. Malshe and G. Waghoo, Progress in Organic Coatings, 51, 267 (2004).
24. X. F. Yang, C. Vang, D. E. Tallman, G. P. Bierwagen, S. G. Croll and S. Rohlik, Polymer Degradation and Stability, 74, 341 (2001).
25. M. Morcillo, Progress in Organic Coatings, 36, 137 (1999).
26. C. Ocampo, E. Armelin, F. Liesa, C. Aleman, X.Ramis and J. I. Iribarren, Progress in OrganicCoatings, 53, 217 (2005).
27. F. Fay, I. Linossier, V. Langlois, D. Haras andK. V. Rehel, Progress in Organic Coatings, 54,216 (2005).
28. A. M. P. Simoes, R. O. Carbonari, A. R. Di Sarli, B. del Amo and R. Romagnoli, Corrosion Science, 53, 464 (2011).
29. A. Gulec, O. Cevher, F. Ustel, A. Turk, A.Akinic and F. Yilmaz, Materials Performance,50, 30 (2011).
|Printer friendly Cite/link Email Feedback|
|Author:||Bano, Humaira; Khan, Mazher Islam; Kazmi, Syed Arif|
|Publication:||Journal of the Chemical Society of Pakistan|
|Date:||Jun 30, 2013|
|Previous Article:||Synthesis of Acyl Analogues of Coniferyl Alcohol and their Antimycobacterial Activity.|
|Next Article:||Simple Growth and Characterization of a-Sb2O4:Evaluation of their Photo-catalytic and Chemical Sensing Applications.|