Materials and Corrosion 2011, 62, No. 10 Acidic/caustic alternating corrosion on carbon steel pipes 975 wPP个一7解产 Figure 11. Changing situation of the ter pH values from four inspection sites in the year 2009 Table 3. Statistical data of the pH values from the four inspection sites H value Lowest Highest <7. 5( times) >9.5(times) Site A 1041 12 Site B 5.26 11.50 Site C 11.55 Site D 11 generator(site B, the top-right in Fig. 11), and after the QO/Ds concentrations were, respectively, 3.89, 0.71, 0.33, and heat exchanger(site C, the bottom-left in Fig. 11) had already (mg/L). Although their contents were all not pretty high, but converged within the expected alkalescent range. However, ph corrosion would still be engendered when these ions accumu values of these two sites also exceeded over 9.5 for many times, some specific pre-existing defects on the pipes surfaces even reached as high as around 11.5, seen in Table 3. But in fact, the most distinct phenomenon in Fig. 11 is that all the three 2.5.3 /CP-AES particularly in some specific points the pH value jumped from determine the cations in the proce ofIC, ICP-AES was used to ater. The results showed 5 to 9, or dropped from 9 to 6 in just 1 day. Such a violent and that the predominant cation was sodium ion Nat*,whose frequent acidic/caustic alternating environment would exert concentration reached 28.54 ppm. This result verified that a more serious corrosion effect on the equipments here, especially relatively high content of naoH was contained in the process the QO/DS heat exchanger, than either the acidic or the caustic water, which may be the major factor of the failure. However, one did solely alue of the iron ions Fe"f(including Fe and Fe't) was just 0.16 ppm, which manifested that corrosion in this event only took 2.5.2 lon chromatograph place in some specific small areas. In other words, it was the In order to testify the EDS results that formation of the pits in the localized corrosion rather than the uniform corrosion occurred peanut-like concave may have been caused by Cl elements, IC was on the pipes in the Qo/DS heat exchanger pplied. It can be learned from Fig. 12 that the major anions in the process water were F, CI, PO4, and soz-, and thei 3 Discussion Based on the above analysis results, chemical compositions, and metallographic structures of the failure pipe were both qualified what is more, no inappropriate operations were investigated during service of the Qo/DS heat exchanger according to the designed requirements as well. Thus, it can be now concluded that among the four possible factors of the failure including matrix materials, process media, service conditions, and maintenance 120140 management, the first and the third one could be ruled out. On the other hand, it is quite clear that the frequent sharp fluctuations of Figure 12. lon chromatograph results of the process wate the process water pH values, the concentrated content of NaoH www.matcorr.com o 2011 WILEY-VCH Verlag GmbH& Co KGaA, Weinheimgenerator (site B, the top-right in Fig. 11), and after the QO/DS heat exchanger (site C, the bottom-left in Fig. 11) had already converged within the expected alkalescent range. However, pH values of these two sites also exceeded over 9.5 for many times, even reached as high as around 11.5, seen in Table 3. But in fact, the most distinct phenomenon in Fig. 11 is that all the three curves behaved sharp fluctuations during the whole year, particularly in some specific points the pH value jumped from 5 to 9, or dropped from 9 to 6 in just 1 day. Such a violent and frequent acidic/caustic alternating environment would exert more serious corrosion effect on the equipments here, especially the QO/DS heat exchanger, than either the acidic or the caustic one did solely. 2.5.2 Ion chromatograph In order to testify the EDS results that formation of the pits in the peanut-like concave may have been caused by Cl elements, IC was applied. It can be learned from Fig. 12 that the major anions in the process water were F, Cl, PO3 4 , and SO2 4 , and their concentrations were, respectively, 3.89, 0.71, 0.33, and 0.28 ppm (mg/L). Although their contents were all not pretty high, but localized corrosion would still be engendered when these ions accumulated in some specific pre-existing defects on the pipes surfaces. 2.5.3 ICP–AES After examining the anions by means of IC, ICP–AES was used to determine the cations in the process water. The results showed that the predominant cation was sodium ion NaR, whose concentration reached 28.54 ppm. This result verified that a relatively high content of NaOH was contained in the process water, which may be the major factor of the failure. However, value of the iron ions FenR (including Fe2R and Fe3R) was just 0.16 ppm, which manifested that corrosion in this event only took place in some specific small areas. In other words, it was the localized corrosion rather than the uniform corrosion occurred on the pipes in the QO/DS heat exchanger. 3 Discussion Based on the above analysis results, chemical compositions, and metallographic structures of the failure pipe were both qualified, what is more, no inappropriate operations were investigated during service of the QO/DS heat exchanger according to the designed requirements as well. Thus, it can be now concluded that among the four possible factors of the failure including matrix materials, process media, service conditions, and maintenance management, the first and the third one could be ruled out. On the other hand, it is quite clear that the frequent sharp fluctuations of the process water pH values, the concentrated content of NaOH Materials and Corrosion 2011, 62, No. 10 Acidic/caustic alternating corrosion on carbon steel pipes 975 Figure 11. Changing situation of the process water pH values from four inspection sites in the year 2009 Table 3. Statistical data of the pH values from the four inspection sites pH value Lowest Highest <7.5 (times) >9.5 (times) Site A 5.07 10.41 103 12 Site B 5.26 11.50 6 247 Site C 6.68 11.55 5 141 Site D 7.21 11.94 1 7 Figure 12. Ion chromatograph results of the process water www.matcorr.com
2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim