Keisuke MIWA*
So MURASUE*
Hiromitsu CHO*
Manabu NOGUCHI**
*
Ebara Environmental Plant Co., Ltd
**
Technologies, R&D and Intellectual Property Division
As one measure to improve power generation efficiency in incineration plants, we aimed at low-temperature heat recovery by lowering the temperature of feed water to the economizer. Therefore, in order to investigate the conditions under which low-temperature corrosion (acid dewpoint corrosion and salt deliquescence), which is a problem of low-temperature heat recovery, becomes apparent, a temperature-controlled test piece was exposed to the flue of an actual plant, and the relationship between temperature and corrosion rate was arranged. In addition, as a result of analyzing the gas composition in the flue and the ash deposited on the test piece and considering the factors of acid dewpoint corrosion, it was found that the deliquescence of ash had almost no effect on corrosion, and the corrosion rate rapidly increased at a temperature lower than the acid dew point estimated from the gas composition in the flue.
Keywords: Low-temperature heat recovery, Dewpoint corrosion, Economizer，Corrosion rate
The 3R initiative is being promoted from the perspective of contributing to a recycling-oriented society; efficient waste energy recovery, represented by heat recovery, such as waste power generation, is being promoted for the remaining waste. In addition, incineration plants need to be operated stably to extend their service life, so it is desirable to achieve safe plant operation and highly efficient power generation.
As one of the technological developments to further increase the efficiency of waste power generation, we are also working on increasing heat recovery capacity through the low-temperature economizer. The general flow of an incineration power plant is shown in Fig. 1. The economizer is located at the rear of the boiler and is a heat recovery device that uses the heat from the exhaust gas to heat the feed water to the boiler. Lowering the feed water temperature is an effective way of increasing heat recovery by the economizer.
Fig. 1 Flow of an incineration power plant
Acid dewpoint corrosion, however, can be caused by condensation of HCl and SOx in the exhaust gas on the heat transfer tubes, due to the low feed water temperature. It is generally known that the sulphuric acid dewpoint increases with increasing moisture content or SO3 concentration in the combustion gas, but the SO3 concentration in incineration power plants is lower than that in coal-fired power plant flue gas. The sulphuric acid dewpoint is therefore expected to be lower. In practice, however, the amount of sulphur and chlorine in the waste is not constant and it is difficult to constantly measure the acid dewpoint, so the feed water temperature is currently set at a sufficiently high temperature range where there is no risk of dewpoint corrosion occurring. In addition, corrosion may progress when the incinerator is shut down. In our case, when extracted heat transfer tubes were stored indoors for about two weeks, the tubes were generally wet and it was confirmed that a strong acid of about pH 2 was formed by the moisture absorption and hydrolysis of highly deliquescent chlorides1
）.
In achieving long-term safe plant operation while maintaining high energy recovery efficiencies, problems with low-temperature corrosion, aided by dewpoint corrosion and salts such as highly deliquescent chlorides, can be an obstacle and a factor in increasing maintenance costs. In order to clarify the details of the corrosion behavior and influencing factors in the environment in which the economizer is used in incineration power plants, we carried out exposure tests on steel in the exhaust gas of the actual equipment and investigated the corrosion behavior.
The exposure test was carried out by placing a pipe-shaped test piece, similar in shape to a heat transfer tube, in the flue duct at the rear of the economizer. A schematic diagram of the apparatus used for this exposure test is shown in Fig. 2. The test piece was an STB 340 with an outer diameter of 38.1mm, a thickness of 4.0mm, and a length of 1 200mm. A thermocouple was placed upstream of the exhaust gas to measure the temperature of the exhaust gas. In addition, thermocouples were placed at three different locations along the length of the test piece to measure the temperature on the test piece surface; by flowing cooling air through the test piece, the temperature on the outer surface of the test piece was controlled at a predetermined temperature. Two sets of test devices prepared and the temperature at the tip of the test piece was set to 120°C for the low-temperature test piece and to 140°C for the high-temperature test piece. During the exposure test, the temperature of the outer surface of the test piece was lower from the tip to the root of the test piece because the cooling air temperature was lower when the test piece was closer to the wall surface. The exposure test was carried out on two municipal waste incineration power plants (Plant A and B) for about two years each, including a period of shutdown. However, in Plant B, only the low-temperature test pieces were exposed. Each time the incinerator was stopped, the test piece was pulled out of the flue duct to remove corrosion products and the outer diameter of the test piece was measured with a caliper and the difference from the initial value was taken as the amount of thinning to calculate the corrosion rate from the exposure period.
Fig. 2 Schematic diagram of the exposure corrosion test piece
To investigate the corrosive environment during the exposure test, we analyzed the exhaust gas composition at the inlet side of the bag filter, which is in the same environment as the economizer. Concentrations of acid gases in the exhaust gas were measured in accordance with JIS K 0301:2012 (Methods for determination of oxygen in flue gas) for HCl and JIS K 0103:2011 (Methods for determination of sulfur oxides in flue gas) for SOx. In actual operation, ash is deposited on the heat transfer tubes with a certain thickness and it is believed that the ash affects the corrosion reaction on the test piece surface in the exposure test. Therefore, out of the ash deposited on the test piece, the ash near the surface of the test piece was collected and characterized for chemical composition, pH, and other properties. The analysis samples were homogenized and powdered by grinding in a mill (TI-100N by Rigaku Corporation) for approximately five minutes. The samples were vacuum dried and analyzed for chemical composition using an X-ray fluorescence spectrometer (ZSX Primus IV by Rigaku Corporation). The pH was measured using a pH meter (HM-25R by DKK-TOA Corporation) for aqueous solutions with mass ratios of ash:water = 1:10 and ash:water = 1:100.
Fig. 3 shows an example of temperature measurements taken by thermocouples placed at various locations on the test piece during a certain week of the exposure test. Although the exhaust gas temperature of the actual equipment varied between 170°C and 190°C, the tip temperatures of both the low-temperature and high-temperature test pieces were fixed at 120°C and 140°C, respectively, as set, and temperature controllability was realized. The temperature of the low-temperature test piece was approximately 110°C to 120°C from the root to the tip, and the temperature of the high-temperature test piece was approximately 130°C to 140°C from the root to the tip.
Fig. 3 Example of temperature measurement at each part of the test piece
Appearance photos taken immediately after the test piece was pulled out of the flue duct are shown in Fig. 4. White ash was uniformly deposited on the test piece surface with a thickness of approximately 1 cm. When the ash was removed, a thin red rust layer was formed covering the test piece surface, as shown in Fig. 4 (a). The same appearance was observed in the area where the outer surface temperature of the test piece was above 110°C. Red rust was also observed on the side of the test piece where the ash naturally peeled off. On the other hand, as shown in Fig. 4 (b), the deep root and its surrounding areas exposed to low temperatures below 110°C near the wall surface were covered with hard reddish brown and yellow corrosion products, indicating severe corrosion.
Fig. 4 Test piece appearance immediately after pulling out from the duct (surface temperature of the relevant part during operation is indicated in [ ])
Appearance photos of the test piece after removal of the ash and corrosion products deposited on the test piece are shown in Fig. 5. The area where the surface temperature of the test piece was controlled above 110°C was covered with thin red rust, but when the red rust was removed from the surface of the test piece, it was smooth with almost no corrosion marks, such as microscopic pores, as shown in Fig. 5 (a). On the other hand, as shown in Fig. 5 (b), corrosion marks were scattered and the surface was rough in the areas where thermocouples were not placed and the temperature was below 110°C during the exposure test.
Fig. 5 Test piece appearance after removing corrosion products (surface temperature of the relevant part during operation is indicated in [ ])
Fig. 6 shows the relationship between the median temperature measured by each thermocouple during the exposure period when the incinerator was operating and the amount of thinning at the thermocouple location. For the area below 110°C, where no thermocouples were placed, temperatures were approximated. The corrosion rate was less than 0.1mm/year in the area where the surface temperature of the test piece was approximately 110°C to 140°C, and almost no corrosion was observed. The area exposed to low temperatures below 100°C was observed to be heavily corroded, but the corrosion rate increased with decreasing temperature. In addition, the trend of the corrosion rate with respect to the surface temperature of such test pieces was similar between Plant A and B. From the above, it was found that the temperature at which corrosion becomes apparent and the corrosion rate increases is approximately 100°C in this exposure test environment. In addition, the relationship between the surface temperature of the test piece and the corrosion rate corresponds to the results of the appearance observation, which confirmed that the corrosion rate was greater in the heavily corroded area at the root of the test piece.
Fig. 6 Relationship between test piece surface temperature and corrosion rate
Table 1 shows the results of the exhaust gas composition analysis during the exposure test period. The water content was about 20% and varied little, while the SO3 concentration varied widely from 0.1ppm-dry to 0.5ppm-dry. Various formulas and correlations have been proposed to estimate the sulfuric acid dewpoint2), 3), and there is some variation in the estimated values. In general, the dewpoint of hydrochloric acid is almost the same value as that of water, and the dewpoint of hydrochloric acid is below 100°C. On the other hand, the dewpoint of sulfuric acid depends on the SO3 concentration in the gas and the moisture content, and, when the SO3 content is about 0.1 ppm, the dewpoint of sulfuric acid exceeds 100°C. In this exposure test environment, the sulfuric acid dewpoint is higher than the hydrochloric acid dewpoint, indicating that sulfuric acid dewpoint corrosion is likely to occur. The dewpoint of sulfuric acid in this exposure test environment was estimated to be approximately 107°C to 112°C in Plant A and 104°C to 118°C in Plant B. In other words, from the analysis results of the exhaust gas composition in this exposure test environment, the sulfuric acid dewpoint was estimated to be below 120°C, even considering the variation of the sulfuric acid dewpoint due to the variation of SO3 concentration and moisture content by season and time of day.
Table 1 Exhaust gas analysis results in exposure test environment
The chemical composition and pH measurement results of the ash deposited on the test piece are shown in Table 2. The ash was found to be rich in Ca, Na, O, and Cl, and was composed of oxides and chlorides. In addition, when analyzing the ash, it was observed to be highly deliquescent because it absorbed moisture in a short period of time. On the other hand, the pH of the solution of ash dissolved in pure water was about 10, indicating that the deliquescent ash was less corrosive, and the pH did not change around 10 when the concentration of the solution was changed, confirming the buffering effect.
Table 2 Chemical composition and pH of ash taken from the test piece
From this exposure test, it was found that the temperature at which corrosion becomes apparent in this environment is about 100°C. The corrosion factors were investigated based on the composition analysis results of the exhaust gas, ash deposited on the test piece, and corrosion products. Based on the results of exhaust gas composition analysis, the sulfuric acid dewpoint was estimated to be below 120°C. However, the exposure test results showed that corrosion was mild even at temperatures near 100°C, which should be below the estimated sulfuric acid dewpoint. In other words, the temperature at which corrosion becomes apparent in the exposure test results does not coincide with the acid dewpoint estimated from the composition of the exhaust gas, and the corrosion rate increases rapidly at a temperature about 30°C below the estimated sulfuric acid dewpoint. According to the previous study4), the amount of sulfuric acid condensed is small near the sulfuric acid dewpoint, and iron sulfate is formed due to condensation of high-concentration sulfuric acid, resulting in a small amount of corrosion; but, as the temperature decreases from the dewpoint, the amount of low-concentration sulfuric acid condensed increases. Therefore, based on the relationship between the amount of corrosion and metal surface temperature in dewpoint corrosion, it has been reported that the amount of corrosion is small near the dewpoint and reaches a maximum at surface temperatures between 20°C and 60°C below the dewpoint, which is consistent with the trend obtained in this exposure corrosion test.
In addition, as mentioned earlier, ash contains chlorides and is highly deliquescent, so there is concern that deliquescence of ash will result in overall thinning and a higher corrosion rate. However, the overall corrosion rate did not increase from the results of the exposure test carried out for about two years, including the shutdown period, and there were areas where the corrosion rate was below 0.1mm/year. Therefore, although the deliquescence of ash may have contributed to the formation of corrosion products such as red rust, its effect on the amount of thinning and the corrosion rate was not confirmed. This was because the aqueous solution of the ash was weakly basic, indicating that the corrosiveness of the deliquescent ash to the steel is low. On the other hand, the buffering effect was confirmed from the pH measurement results of aqueous solutions with different mass ratios of ash and water, suggesting the possibility of reducing sulfuric acid dewpoint corrosion by adjusting the pH against sulfuric acid condensation. In addition, the deposition of ash on the test piece is considered to be an obstacle to the diffusion of oxygen in the atmosphere to the test piece surface, thereby working to inhibit corrosion.
The above results show that the relationship between temperature and corrosion rate obtained from the corrosion test indicates that sulfuric acid did not condense in areas where the tube surface temperature exceeded the sulfuric acid dewpoint and thus did not cause thinning, and that the amount of thinning on the test piece was slight even in areas where the temperature was 100°C to 110°C below the dewpoint, suggesting that there was little condensation near the sulfuric acid dewpoint, and thus no thinning occurred. In addition, it is believed that the thinning was severe in the areas significantly below the sulfuric acid dewpoint, since it was an environment where sulfuric acid was continuously condensed. Through the exposure test, the relationship between temperature and corrosion rate was arranged and an important indicator was obtained for setting the economizer feed water temperature to improve power generation efficiency.
In order to clarify the details of the corrosion behavior and influencing factors in the environment where the economizer is used, the exposure test on steel was carried out using actual equipment to investigate low-temperature corrosion, which is an issue in achieving long-term safe plant operation while maintaining high-efficiency waste power generation.
The corrosion conditions of the test piece and the relationship between the surface temperature of the test piece and the corrosion rate were investigated. The results showed that the temperature at which corrosion becomes apparent does not match the acid dewpoint estimated from the composition of the exhaust gas, and that the corrosion rate increases rapidly at a temperature about 30°C below the estimated sulfuric acid dewpoint. No influence on the amount of thinning due to the deliquescence of ash was observed in this exposure test environment.
On the other hand, when considering stable and long-term plant operation, in addition to economizers, air heaters and bag filters are devices of concern for low-temperature corrosion, used in corrosive gas environments, including HCl and SOx at low exhaust gas temperatures. When austenitic stainless steel, such as SUS 304 or SUS 316, is used for devices operated at low temperatures, the devices are constantly exposed to chloride ions contained in ash and exhaust gas, resulting in a high risk of stress corrosion cracking (SCC) occurring when wet. Therefore, design and maintenance must be considered for the entire process where low-temperature corrosion is a concern.
We are currently developing a corrosion prevention technique for low-temperature corrosion using sensors, aiming to construct a method to directly monitor the degree of corrosion caused by condensation and wetting, in addition to the indirect investigation of the corrosion environment, such as estimating the dewpoint and arranging the relationship between temperature and corrosion rate. Details on this method will be reported in Report No.2.
1）M. Noguchi and H. Yakuwa: Lecture on Fundamental Aspects of High Temperature Corrosion and Corrosion Protection Part 4: High Temperature Corrosion and Corrosion Protection in Incineration Plants., Ebara Engineering Review, No. 254, p. 29-40, (2017).
2）P. Mueller: Chemie-Ing-Techn., 31- 5, 345-350 (1959).
3）F. H. Verhoff and J. T. Banchero: Chem. Eng. Prog., 70 (8), 71-72 (1974).
4）H.Nagano: Corrosion Engineering, 26，731-740 (1977).