Manabu NOGUCHI*
Hiroshi YAKUWA*
*
Technologies, R&D Division
For incineration plants that use chlorine-containing fuel, chlorine-induced corrosion is an unavoidable issue. The most important key for high temperature corrosion protection is to make the formation of protective corrosion product layers; however, in the case of chlorination-induced corrosion, formed chloride is low in density, and serves poorly as a protective layer. Furthermore, chlorides generally have characteristics, such as low melting points and high vapor pressure levels, which cause various types of corrosion with different characteristics. This part discusses high temperature corrosion caused by chlorination in various sections other than boilers, including volatilization of chlorides at ultrahigh temperatures, grain boundary corrosion of cast metal, and erosion-related corrosion, as well as protective measures against such corrosion. Additionally, corrosion due to dew condensation, which is an unavoidable corrosion problem in actual incinerators, is briefly described.
Keywords: High temperature corrosion, Chlorination, Volatilization, Grain boundary corrosion, High temperature erosion-corrosion, Condensation, Void formation, Stoker, Inbed tube, Casting alloy
Chlorine (Cl) is a substance that typically causes high temperature corrosion in waste and biomass incineration plants. Taking SUS304 as an example, a critical temperature for high temperature corrosion is reportedly 800 ℃ or higher when it is caused by oxidation, 500 ℃ or higher by sulfidation, and 350 ℃or higher by chlorination 1) . Key factor for high temperature corrosion protection is to form a protective corrosion product layer. If corrosion is caused by chlorination, however, the formed chloride is too low in density to act as a protective layer against corrosion. Furthermore, compared with oxides and sulfides, chlorides generally have characteristics, such as a low melting point and a high vapor pressure, which cause various types of corrosion with different characteristics.
Part 3 of this lecture dealt with boilers in waste-toenergy plants with a focus on the influence of ash adhesion on corrosion. Part 4 discusses high temperature corrosion in various components other than boilers in waste and biomass incineration plants, as well as protective measures against such corrosion.
It has long been reported that incinerators undergo severe corrosion caused by molten salt contained in adhered ash at 300 ℃ or higher and in those operating with a temperature range between 500 and 700 ℃ particularly severe corrosion occurs 2). With few findings on corrosion behavior in high temperature ranges of 700 ℃ or higher, it has been believed that, at such high temperatures, corrosion is likely to decrease to a level similar to the level of corrosion in the gas phase. A high temperature air heater has been developed as a method of recovering heat in a super high temperature range around 1000 ℃ to avoid temperatures at which corrosion is caused by adhered ash 3). Figure 4-1 shows a conceptual scheme of a plant using a high temperature air heater 4). High-temperature heated air obtained by heat exchange in a high temperature flue gas atmosphere around 1000 ℃ is used as a heat source for steam superheating or as combustion air for an ash melting furnace. A top priority issue for this system is selecting an appropriate material for a heat exchanger. Since heat transfer tubes themselves are heated to around 1000 ℃, tube materials must be strong enough at high temperatures, and among other characteristics, resistant to corrosion at such high temperatures. The metal surface temperature of a heat exchanger currently used for an incinerator does not exceed 500 ℃. This method, however, allows the temperature to increase to nearly 1000 ℃, which is an extremely challenging approach, providing an opportunity to study corrosion at super high temperatures in an incinerator.
Fig. 4-1 Conceptual scheme of gasification and ash melting plant using high temperature air heater<sup>4）</sup>
For an actual high temperature air heater, heat transfer tubes made of various commercially available materials were fabricated by centrifugal casting, and exposure tests using a gasification and ash melting furnace (demonstration furnace) were conducted to evaluate the tubes. Figure 4-2 shows typical results of the tests. The surface temperature of the heat transfer tubes was raised to approximately 700 to 1100 ℃, which is above the temperature range that has been believed to accelerate corrosion caused by ash adhesion. Although only mild corrosion was expected, extremely severe metal loss occurred. Based on the premise that the linear law is applicable, a metal loss after 1000 hours of approximately 3 mm for a heat-resistant cast steel was obtained and approximately 2 mm for Ni-Cr-Fe cast alloys. Even in the Ni-Cr-W cast alloy that showed the best results in the tests, a metal loss of approximately 0.5 mm was observed. All of the cast alloys underwent grain boundary corrosion along a network of precipitated carbide phases; grain boundary corrosion of 2 mm or more was observed in the Ni-Cr-Fe cast alloys.
When the heat transfer tubes after the exposure tests were exposed to the atmosphere for a period of time, droplets frequently formed on their surfaces, as shown in Figure 4-3. After similarly exposing test pieces cut from the heat transfer tubes for crosssectional observation for a period of time, droplets were also observed on the corroded parts.
To study this corrosion behavior, laboratory tests were conducted using the Ni-Cr-W cast alloy (Ni-33Cr15W-0.3C, hereinafter the “base material”) that showed the best results in the demonstration furnace tests 5) . Corrosion tests were conducted on test pieces of the alloy embedded in adhered ash collected from the demonstration furnace in N2-10 % O2-1000 ppm HCl gas for 200 hours (embedding tests). The test temperature range was set between 700 and 1200 ℃, exceeding the temperature range of corrosion caused by molten salt. In comparison tests conducted in the same gas without using adhered ash, the level of corrosion was less than that in the tests using adhered ash. In other words, while the corrosiveness of HCl gas was limited, the corrosive environment became severer as Cl2 was formed by the catalytic effect of adhered ash. Maximum thickness loss (= metal loss + grain boundary corrosion depth) values obtained from the results of the embedding tests are shown in Figure 4-4 6). While the thickness loss tended to increase in temperatures between 800 and 900 ℃ in the exposure tests using the demonstration furnace, a temperature dependence curve with a peak between 800 and 900 ℃ was also obtained from the laboratory tests. Furthermore, by observing the cross-sections of the test pieces at 900 ℃ and 1100 ℃ (Figure 4-5), we found that the test pieces underwent grain boundary corrosion along Cr carbide at both 900 ℃ and 1100 ℃ and that droplets similar to those found in the exposure tests formed at 900 ℃.Consequently, the results of the laboratory tests were determined to reflect the results of the tests using the demonstration furnace. According to the results of the analysis, the droplets were identified as having been formed by deliquescent Cr and Ni chlorides. On the other hand, at 1100 ℃, no such droplets formed on the test pieces with less thickness loss; instead, a number of voids were formed within the alloy. According to the results of the elemental analysis of the cross-sections, almost no Cl was found. That is to say, the existence of Cl within the alloy had a significant influence on the corrosion rate.
Based on the above results, the corrosion mechanism of a Ni base cast alloy in this environment is shown in Figure 4-6 5). First, Cl2 with strong corrosiveness was formed from HCl by the catalytic effect of adhered ash (Reaction Equation (1)). Cl2 and O2 penetrated into the alloy through boundaries between the Cr carbide phase and the substrate to cause grain boundary corrosion. In an atmosphere with 10 % O2 in which oxides are stable, chlorides were formed within the alloy (at corrosion tips) in which the partial pressure of O2 was sufficiently reduced. As the amount of penetrated O2 increased over time, the partial pressure of O2 within the alloy increased.Consequently, as the chloride-stable atmosphere changed to an oxide-stable atmosphere, the chlorides converted into oxides (Reaction Equations (2) and (3)). Simultaneously, Cl2 was formed from the chlorides; formed Cl2 was involved in corrosion within the alloy again, as shown in Reaction Equations (4) and (5). This recycling reaction of Cl2 within the alloy accelerated corrosion. The corrosion rate increased dramatically as the temperature rose from 700 ℃ to 800 ℃.Although the melting point of CrCl2 is 815 ℃, mixed salt with NiCl2 had a lower melting point; the chlorides melted to promote the recycling reaction of Cl2, resulting in a sharp increase in the corrosion rate. By increasing the test temperature to 800 ℃ or higher, the corrosion rate began to decrease. The chlorides were volatilized at the increased temperature, the amount of the chlorides within the alloy decreased, and then, the recycling reaction of Cl2 was suppressed, resulting in a decrease in the corrosion rate. It is likely that a number of voids, such as those observed in the test piece at 1100 ℃, were formed as the traces of volatilization.
As described above, we found that, because of a low melting point and high volatility, chlorides exhibited a complex behavior with temperature, resulting in significant effects on corrosion. A similar chlorine behavior was also found in a study conducted by Sato, et al 7) .
Fig. 4-2 Results of exposure tests of commercially available materials using gasification and ash melting furnace (demonstration furnace)
Fig. 4-3 Photo showing external view of heat transfer tube after exposure test
Fig. 4-4 Results of embedding tests of Ni-Cr-W cast alloys<sup>6）</sup>
Fig. 4-5 Cross-sectional optical micrographs of Ni-33Cr-15W-0.1C alloy after embedding tests
Fig. 4-6 Corrosion mechanism of Ni base cast alloy<sup>5）</sup>
According to the results of the exposure tests using the demonstration furnace, existing alloys do not have service life long enough for high temperature air heaters. Thus, we undertook to improve the corrosion resistance of the base material (Ni-33Cr-15W-0.3C) that showed the best results in the exposure tests 6). Because the accumulation of Cl within an alloy promotes corrosion, as described in Section 2-2, we examined the possibility of improving corrosion resistance by adding Al to the base material and obtaining AlCl3 with a high volatility so as to massively remove Cl from the alloy. We aimed to suppress the recycling reaction of Cl by the reaction of Cl penetrated within the alloy with Al and the volatilization of Cl because AlCl3 sublimes at 200 ℃ or lower. Figure 4-4 also shows the results of corrosion resistance evaluation from the embedding tests described above.For comparison purposes, a Si-added alloy was experimentally produced for evaluation by adding Si, a typical element that improves high temperature corrosion resistance, to the base material. While the Si-added alloy exhibited a corrosion behavior similar to that of the base material, in the Al-added alloy, corrosion, similar to the corrosion that peaked at 800 ℃ or higher in the base material, was substantially suppressed, resulting in significant improvement in corrosion resistance. By adding Si and Al simultaneously to the base material, the corrosion resistance of the alloy improved further. This is attributable to the removal of Cl by Al, which leads to the effective action of Si. Figure 4-7 shows a cross-sectional image of a test piece of the Si/Al-added alloy at 900 ℃. No droplets such as those found in the base alloy at 900 ℃ were observed; instead, voids were observed. The effect of Al on the prevention of Cl accumulation was also confirmed by cross-sectional observation.
The corrosion of centrifugally cast alloys is characterized as grain boundary corrosion along Cr carbide in a network structure, as described above. Therefore, aiming at further improvement in corrosion resistance, we studied the possibility of breaking up the Cr carbide network by adding Nb, which has a greater affinity for carbon than Cr, to suppress such grain boundary corrosion. Figure 4-8 shows the dependence of maximum thickness loss on Nb content. By adding Nb, the thickness loss decreased; the minimum value was obtained by adding 1.5 mass% of Nb.
The advantages of these developed alloys were verified in exposure tests using test pieces in an actual gasification and ash melting furnace. The test pieces were placed at three points, upstream, midstream, and downstream of flue gas at the tips of the high temperature air heater shown in Figure 4-1. The gas temperature values at these points were assumed to be approximately 1000 ℃, 900 ℃, and 800 ℃, respectively. However, the temperature values of the test pieces were likely to be less than the gas temperature values because they were cooled by the high temperature air heater. After the 1-year exposure tests, the maximum thickness loss of each test piece was evaluated. Figure 4-9 shows the results of the exposure tests. Because the lower halves of 7 mm-thick test pieces made of the base material were lost upstream and midstream of flue gas, the evaluation was performed using the remaining halves. There was no change in the appearance of the developed materials; the corrosion resistance of the developed materials was significantly improved compared with that of the base material.
As described above, in corrosion caused by chlorination, a protective corrosion product layer hardly formed, compared with corrosion caused by oxidation or sulfidation; under certain conditions, severely accelerated corrosion is caused by problems associated with corrosion product. Therefore, by massively volatilizing chlorides by using Al, we minimized the effect of Cl and suppressed an increase in the corrosion rate. The developed alloys have been put into practical application in materials for high temperature air heaters in gasification and ash melting furnaces.
Fig. 4-7 Cross-sectional optical micrograph of Si/Al-added alloy at 900 ℃
Fig. 4-8 Dependence of maximum thickness loss in Si/Al-added alloy on Nb content
Fig. 4-9 Results of exposure tests of Ni base cast alloys in actual furnace
Figure 4-10 shows a schematic diagram of grates. The fixed and movable grates are placed in a stoker furnace in a stepped manner. After input into the stoker furnace, waste moves over the grates as it burns, using combustion air supplied through the grates from below, and then, becomes incineration ash to be carried out of the furnace. Meanwhile, the high temperature corrosion environment is mitigated by cooling the back side of the grates with air for combustion to suppress the overheating of the grates by heat from combustion. Waste is moved by the forward and backward movement of the moving grates; the grate temperature fluctuates when the grates move forward to push burning waste and then backward, when contact with waste decreases. For grates, heat-resistant cast steels are commonly used. The following section describes a case of metal loss in a heat-resistant cast steel caused by corrosion and thermal fluctuations.
Fig. 4-10 Schematic diagram of grates
Grates are subject to high temperature corrosion caused by exposure in a high temperature environment created by heat from combustion and to fluctuations in the surface temperature during forward and backward movement. Figure 4-11 shows images of how a grate made of SCH2 (Fe-26Cr-0.3C) in an actual furnace was damaged. Severe metal loss was observed on the front surface of the grate, in the area that pushes waste, particularly, from the center to the lower part of the front surface. Figure 4-11 also shows the cross-sectional observation results of the upper part of the front surface that underwent relatively mild metal loss and the center of the front surface that underwent severe metal loss. Both parts underwent grain boundary corrosion; particularly severe grain boundary corrosion was observed in the upper part. Several cracks were also found in the corroded grain boundaries.The cause of grain boundary corrosion of heat-resistant cast steels in waste incineration environments has been identified as the preferential corrosion of Cr carbide precipitated adjacent to grain boundaries by Takahashi, et al. 8)-10). As described in Chapter 2 of this paper, cast alloys in incineration environments typically undergo the preferential corrosion of Cr carbide.
The mechanism of metal loss in a grate described above is summarized in Figure 4-12. Cr carbide precipitated at grain boundaries was preferentially corroded. The surface temperature fluctuated as the grate moved forward and backward. However, since the center to the lower part of the grate was thinner than the upper part, it was susceptible to the effect of air-cooling from the back side, and subject to severe thermal fluctuations. Cracks were caused by thermal stress due to thermal fluctuations in the corroded parts, and grains fell out, causing metal loss as a result. The upper part, in which thermal fluctuations were gradual and less grains fell out, suffered less metal loss, but underwent grain boundary corrosion instead. Meanwhile, the central to the lower part, in which thermal fluctuations were severe and grains fell out frequently, underwent severe metal loss.
Fig. 4-11 External views of grate in actual furnace and cross-sectional observation results
Fig. 4-12 Schematic diagram of metal loss in grate
To suppress preferential corrosion that is a primary cause of metal loss, we examined a possibility of improving the Cr carbide phase by adding Nb to SHC2. Figure 4-13 shows the microstructural observation of a SCH2+1Nb alloy. While Cr carbide was precipitated in a network structure in SCH2 without Nb, we found that, by adding Nb to SCH2, a part of Cr carbide was converted to Nb carbide and that the Cr carbide phase network was cut. Figure 4-14 shows the results of exposure tests of side seals made of SCH2+0.5Nb and SCH2+1Nb cast alloys conducted for about six months in an actual furnace 11). Side seals are metalwork placed onto the side walls of a stoker furnace to avoid the transverse thermal expansion of grates.Therefore, although side seals are in almost the same corrosion environment as grates, they are subject to higher temperatures than grates because they are not cooled down. Round-shaped metal loss was observed at a corner of the side seal made of conventional SCH2. The diagonal length of the cross section was measured to obtain the metal loss. While a metal loss of approximately 5 mm was observed in SCH2, metal loss was significantly suppressed in the SCH2+0.5Nb and SCH2+1Nb alloys; the metal loss of the SCH2+0.5Nb alloy was less than 1 mm. As described above, cutting the Cr carbide network by adding Nb to SCH2 was proven to be effective in suppressing grain boundary corrosion.
However, unlike side seals made of the SCH2+0.5Nb and SCH2+1Nb alloys, these alloys did not demonstrate a sufficiently significant superiority over conventional SCH2 when they were used for grates in the actual furnace. The grate temperature was lower than the side seal temperature because the grates were cooled with air from the bottom and underwent corrosion caused by molten salt contained in adhered ash as described in Section 2-1 of this paper; thus, improvement in general corrosion resistance as well as grain boundary corrosion resistance was considered to be required. We undertook the development of an alloy by adding not only Nb, but also Mo, which is an element that effectively improves corrosion resistance in municipal waste environments 12). In the actual furnace, demonstration tests were conducted on grates made of the alloy developed based on the optimum composition we explored in laboratory tests. Figure 4-15 shows a photo representing the results of a demonstration test of a grate made of the developed alloy in the furnace after approximately 4 000 hours of operation. For the purpose of comparison, a grate made of SCH2 with Ni base self-fluxing alloy sprayed on the surface, which is costly but has good corrosion resistance, was also prepared for evaluation. SCH2 with Ni base selffluxing alloy sprayed and the developed alloy (1 mass% Nb + 5 mass% Mo added to SCH2) clearly demonstrated a similar superiority to SCH2 in the metal loss on the front surface. The mass loss of the grate made of the developed alloy was also measured for evaluation; it was about half (0.63 %) of that of SCH2 (1.2 %), demonstrating a superiority over the conventional SCH2 in terms of mass loss.
Currently, the service life of grates are being extended by taking further measures to lower the grate temperature, such as employing forced air-cooling and water-cooled grates, in addition to using these developed alloys.
Fig. 4-13 Microstructural observation of SCH2 and SCH2+1Nb alloy
Fig. 4-14 Results of demonstration tests of side seals made of SCH2+0.5Nb and SCH2+1Nb cast alloys<sup>11）</sup>
Fig. 4-15 Results of demonstration test of grate made of Nb/Mo-added alloy
This Section introduces an example of high temperature corrosion in an internally circulating fluidized bed boiler (hereinafter “ICFB”) that is applicable to various types of fuel, including biomass fuel. Figure 4-16 shows a conceptual scheme of ICFB. ICFB can be characterized by a fluidized bed divided into a combustion chamber and a heat recovery chamber and by in-bed tubes arranged in the heat recovery chamber. By providing a partition wall between the combustion chamber and the heat recovery chamber, in the heat recovery chamber, sand flows more slowly than in the combustion chamber, and the concentration of corrosive gas decreases to about one tenth of that in the combustion chamber.Therefore, both erosion and corrosion conditions in the heat recovery chamber are substantially improved compared with those in the combustion chamber, contributing to improvement in the durability of in-bed tubes. However, when fuel containing Cl, such as biomass fuel, is used, Cl is fed to the heat recovery chamber as it adheres to the fluid medium circulating between the combustion chamber and the heat recovery chamber. Figure 4-17 shows the results of cross-sectional EDS (energy dispersive x-ray spectroscopy) analysis of the fluidized medium used in the actual boiler. In the fluidized medium, Ca and Cl layers were formed around silica sand particles containing Si and O. In other words, silica sand particles were coated by chlorides. A similar example of adhesion of chlorides to fluidized medium has also been observed in an externally circulating fluidized bed boiler 13); thus, this phenomenon is supposed to be specific to fluidized bed boilers. The following section describes an example of material damage caused by high temperature erosion-corrosion, a combination of corrosion caused by chlorination and erosion.
Fig. 4-16 Conceptual scheme of ICFB
Fig. 4-17 Results of cross-sectional EDS analysis of fluidized medium
Figure 4-18 is a photo showing an external view of an inbed tube after use. The surface of the tube was thermal sprayed with Ni base self-fluxing alloy. Red rust occurred in portions of the bend, where the sprayed layer disappeared to expose the carbon steel substrate (circled in white in the figure). From the appearance, metal loss as shown in the photo may be regarded as caused by erosion because no attachment or corrosion product is observed. If fuel containing almost no Cl had been used, metal loss of the in-bed tube would have been less. On the other hand, the metal loss of the tube increases as the Cl content in fluidized medium increases; therefore, it is obvious that a corrosion phenomenon due to Cl plays a certain role in damage.
Fig. 4-18 Photo showing external view of in-bed tube in actual boiler
To study high temperature erosion-corrosion phenomena, we conducted laboratory tests using rotary erosioncorrosion test equipment in which test pieces rotate in silica sand and examined metal loss behavior under different corrosion and erosion conditions 14). The effects of both corrosion and erosion conditions were evaluated by changing the atmosphere and the rotational speed of the test pieces, respectively. 24-hour tests were conducted at 550 ℃ by changing the rotational speed of the test pieces under two corrosion conditions: 1) in a corrosive environment where 1 mass% salt (NaCl-KCl-Na2SO4-K2SO4 system eutectic salt with a melting point of 512 ℃ ) was added to silica sand and N2-10 %O2-1000 ppm HCl gas flew; 2) in an atmospheric environment containing no salt or HCl gas.
Figure 4-19 shows the dependence of the mass loss of carbon steel and corrosion resistant SUS347 on the rotational speed. Metal loss hardly occurred in the erosion-dominant atmospheric environment, where corrosion has little effect. Meanwhile, in the corrosive atmosphere, significant increases in metal loss were observed as the rotational speed increased. Extreme increases in metal loss were not observed up to 250 rpm, in the case of SUS347, and up to 750 rpm, in the case of carbon steel. Once these rotational speed thresholds were exceeded, however, metal loss was found to increase significantly. In other words, it is experimentally confirmed that extremely severe metal loss was caused by the synergetic effect of corrosion and erosion, as compared to an erosion- or corrosion-dominant environment.In the corrosive environment, carbon steel was compared with SUS347; there was less metal loss in SUS347 with a good corrosion resistance than in carbon steel under a condition with no erosion (0 rpm). Less metal loss was observed in SUS347 than in carbon steel under mild erosion conditions up to 250 rpm. Less metal loss occurred, however, in carbon steel with an inferior corrosion resistance than in SUS347, when the rotational speed increased to between 500 and 750 rpm. Photos showing the external views of the test pieces (Figure 4-20) revealed that entire test pieces of carbon steel were covered with scales at both 250 rpm and 750 rpm. Meanwhile, the entire test piece of SUS347 was covered with scales at 250 rpm.However, metal loss at an angle of 45° in the moving direction was observed in the test piece of SUS347 at 750 rpm; a very smooth metal surface with no scales appeared in the metal loss part and had external appearance similar to that of the in-bed tube in the actual boiler shown in Figure 4-18. We conclude that metal loss was reduced by scales covering the surface. With fewer scales, however, metal loss became greater, resulting in the acceleration of corrosion caused by the continuous removal of scales by erosion.
Figure 4-21 shows a schematic diagram of this phenomenon. The horizontal axis in Figure 4-21 (a) indicates a corrosion scale growth rate that corresponds to the severity of a corrosion environment in the case an identical material is used, or the corrosion resistance of each material in an identical corrosion environment. The vertical axis indicates the severity of erosion conditions. Figure 4-21(b) shows a schematic diagram of the metal loss rate of an identical material in the corrosion only region, the corrosion-dominant region, and the erosion-corrosion region, respectively.In the corrosion only region, the metal loss follows the parabolic rule, and the metal loss rate decreases over time (the black line in Figure 4-21 (b)). When the corrosion scale growth rate is sufficiently larger than the erosion rate (in the corrosion-dominant region), the surface of the alloy is covered with scales, the metal loss of the alloy is caused by corrosion, and corrosion scales are removed by erosion. The corrosion rate is mitigated by scales; as is the case with corrosion caused by volatilization described in Section 5-1, Part 1 of this series 15), corrosion is considered to progress steadily at a constant rate because the thickness of scales was determined by a good balance between the scale growth rate caused by corrosion and the damage rate caused by erosion (the red line in Figure 4-21 (b)). On the other hand, when the erosion rate increases up to a level where the continuous removal of corrosion scales starts (erosion-corrosion region), metal loss progresses at the same rate as that of initial corrosion where the almost entire surface of the alloy was exposed without scales (the yellow line in Figure 4-21 (b)).As the erosion conditions become severer, the alloy undergoes erosion-dominant damage (erosion-dominant region). According to the results of the erosion-corrosion tests described above, the surfaces of both SUS347 and carbon steel test pieces were covered with scales and in the corrosion-dominant region up to 250 rpm. Between 500 and 750 rpm, carbon steel with an inferior corrosion resistance and a high scale growth rate was in the corrosion-dominant region while SUS347 with a low scale growth rate was in the erosion-corrosion region with scales removed and underwent severer metal loss than carbon steel did.
Fig. 4-19 Results of erosion-corrosion tests using rotary laboratory testing equipment
Fig. 4-20 Photos showing external view of test pieces after tests
Fig. 4-21 Relationship between corrosion and erosion; schematic diagram of metal loss rate
There are two types of preventive measures against metal loss: environmental measures and proper material selection. For easing erosion conditions and improving corrosion environments, environmental measures include those that decrease the concentration of corrosive substances and change the temperature. Among them, a case has been reported in which the erosion-corrosion depth was decreased by a temperature rise 16). However, in actual plants, it can be difficult to take environmental measures because they often require changes in operating conditions.
Proper material selection typically includes the use of protectors. It is possible to reduce metal loss in the system by attaching protectors to parts with metal loss and performing regular maintenance. For a heat exchanger, the attachment of protectors decreases the heat transfer efficiency; therefore, although this is a very effective countermeasure against local metal loss, the attachment of protectors in a wide area may have a non-negligible effect on equipment performance. As an example of proper material selection, surface treatment, such as thermal spraying or cladding, is often selected. If erosion is dominant, use of hard materials with hard facings or Cr carbide dispersion spraying can be taken as measures against metal loss. In erosion-corrosion environments, surface treatment using a Ni base self-fluxing alloy is widely used in practical implementations because of its superiority in corrosion resistance, erosion resistance, and workability.
As renewable energy is promoted, various types of biomass fuel will be used more in the future, and corrosion environments are expected to vary. The more corrosion environments vary, the more complex they become; as shown in Figure 4-21, optimum materials change depending on the magnitude of corrosion and erosion effects. However, the mechanism of metal loss and optimum materials in erosion-corrosion environments have not been fully understood yet; we are currently working on research and development to enhance measures against metal loss 17).
Lastly, apart from high temperature corrosion, we present examples of corrosion in components exposed to low temperature flue gas, such as an economizer, air preheater, and bag filter. Dew condensation must be avoided in these components; if carbon steel is used, it is naturally corroded by hydrochloric acid formed by HCl contained in flue gas. Therefore, these components are designed free from condensation. However, design operating conditions are not always maintained in actual plants; in fact, the author is often asked for advice on condensation-related corrosion problems. Among them, cases in which austenitic stainless steels, such as SUS304 and SUS316, are used, are more problematic. These materials are very easy to handle because of their superior properties including versatility, workability, and weldability. When used at 50 ℃ or higher in a solution with a high chloride ion concentration, however, we should be aware that they carry a high risk of stress corrosion cracking. Figure 4-22 shows the effect of Ni content on stress corrosion cracking in Fe-Cr-Ni alloys 18).Each circle in the figure indicates the occurrence of stress corrosion cracking. It is known that alloys containing about 10% Ni are most susceptible to cracking and that SUS304 containing 8 % Ni is a material prone to stress corrosion cracking. The more Ni an alloy contains, the less prone to cracking. The material cost increases substantially, however, as the Ni content increases. On the other hand, alloys containing less than 8 % are also less prone to cracking; thus, ferritic stainless steels without Ni are unlikely to undergo stress corrosion cracking.
Incineration plants use chlorine-containing fuel. Dew condensation water, therefore, is very likely to contain highconcentration chloride ions; a number of components are subject to temperatures above 50 ℃ due to heat produced by incineration. Such incineration plant environments have a high risk of stress corrosion cracking; in fact, a percentage of stress corrosion cracking among all corrosion problems is not so low. A risk of stress corrosion cracking can be overlooked when austenitic stainless steel, such as SUS304, is used in a component in which high temperature corrosion occurred in the past. In such a component, cooling may also be provided as a measure against expected high temperature corrosion. Too much cooling, therefore, may cause dew condensation, resulting in stress corrosion cracking.
As preventive measures, condensation-free design is most important. For proper material selection, carbon steel can be used, if timely replacement is possible or ferritic stainless steels should be used to avoid stress corrosion cracking. However, taking the risk of pitting corrosion into consideration, some environments may require the use of a material with a high pitting resistance equivalent (PRE) obtained from Formula (4-1), such as ferritic stainless steel with high Cr content or duplex stainless steel.
PRE＝Cr＋3.3（Mo＋0.5 W）＋16 N …………………（4-1）
Or，PRE＝Cr＋3.3 Mo＋16 N …………………………（4-2）
Note: N is not applicable to ferritic stainless steels.
Although austenitic stainless steel is very easy to handle, it should not be used for components in incineration plants in which the temperature can be 200 ℃ or lower. If it is absolutely necessary to use austenitic stainless steel for such components, one must recognize that special attention should be paid.
Fig. 4-22 Dependence of stress corrosion cracking in Fe-Cr-Ni alloys on Ni content
Part 3 and Part 4 of this series discuss high temperature corrosion in waste and biomass incineration plants. As mentioned in the beginning of Part 4, chlorine is a key element of corrosion in actual environments. Compared with oxidation, chlorination exerts a strong effect on corrosiveness, and its behavior is far more complex. For high temperature air heaters, the recycling reaction and volatilization of chlorides was a key issue. Corrosion resistant materials have been developed by using this volatilization behavior. A top priority issue for grates was how to suppress grain boundary corrosion of Cr carbide precipitated in a network structure; structural improvement has been implemented by additive alloys. For in-bed tubes, understanding the synergetic effect of corrosion and erosion is a key.Although all of these issues relate to chlorination, they involve various phenomena requiring different measures. There is no doubt that the most expedient solution for problems in actual environments consists of a series of steps: first, finding out what is happening in actual equipment; second, verifying those findings in laboratory tests; and third, demonstrating the verified results in actual equipment. Corrosion under dew condensation conditions occurs far more frequently than high temperature corrosion. Most corrosion phenomena caused by dew condensation, however, have already been understood and can be handled by consulting textbooks. In any case, it is obvious that corrosion problems cannot be ignored in processes in which chlorine must be handled, no matter whether the processes involve high or ordinary temperatures.
It is our hope that this series can be help in solving similar corrosion problems and we would be deeply gratified if it did.
1) Rikio Nemoto: Basics of High Temperature Corrosion, Nippon Yakin technical report, No. 5, p. 62 (1996).
2) V. K. Fassler, H. Leib and H. Spahn: Korrosionen an Müllverbrennungskesseln, Mitteilungen der VGB, Vol. 48, pp. 126 - 139 (1968).
3) Takahiro Oshita: The Prospects of Gasification and Slagging Combustion Technology in the Context of Recycling, Journal of the Japan Institute of Energy, Vol. 78, No. 9, p. 712 (1999).
4) Manabu Noguchi, Kei Matsuoka, Hiroyuki Fujimura: The Corrosion Behavior of Cast Alloys in High Temperature Combustion Gas Environment of a Waste Gasification and Ash Melting System, Zairyo-to-Kankyo, Vol. 51, No. 2, p. 67 (2002).
5) Manabu Noguchi, Kei Matsuoka, Hiroyuki Fujimura, Yoshiyuki Sawada, Ueta Shigeki: The Corrosion Mechanism of Ni Base Cast Alloy in High Temperature Combustion Gas Environment of a Gasification and Ash Melting System, Zairyo-to-Kankyo, Vol. 51, No. 2, p. 75 (2002).
6) Manabu Noguchi, Kei Matsuoka, Hideyuki Sakamoto, Ueta Shigeki, Yoshiyuki Sawada: Effect of Alloying Elements for Corrosion Resistance of Ni-base Cast Alloys in Chlorinecontaining High Temperature Environments, Zairyo-toKankyo, Vol. 54, No. 5, p. 218 (2005).
7) Yoshiyuki Sato, Motoi Hara, Yutaka Shinata, Toshio Narita: Effect of a Small Amount of Hydrogen Chloride on High Temperature Oxidation of Nickel-Chromium Alloys, Journal of the Japan Institute of Metals and Materials, Vol. 61, No. 1, p. 56 (1997).
8) Hidenori Takahashi, Yasuki Miyakoshi, Syuichi Kamota, Shigenari Hayashi, Toshio Narita: High Temperature Corrosion Behavior of a Heat Resistant Steel SCH13 in Waste Incinerator Environments, Zairyo-to-Kankyo, Vol. 47, No. 12, p. 777 (1998).
9) Hidenori Takahashi, Yasuki Miyakoshi, Syuichi Kamota, Shigenari Hayashi, Toshio Narita, Kazuhiro Kuroda, Toshio Saito, Akio Kaji: Intergranular Corrosion Mechanism of SCH 13 Heat Resistant Steel in Waste Incineration Environment, Zairyo-to-Kankyo, Vol. 48, No. 9, p. 583 (1999).
10) Hidenori Takahashi, Yasuki Miyakoshi, Syuichi Kamota, Shigenari Hayashi, Toshio Narita: High Temperature Corrosion Mechanism of SCH2 Heat Resistant Steel in Waste Incineration Furnace, Zairyo-to-Kankyo, Vol. 49, No. 7, p. 426 (2000).
11) Hidenori Takahashi, Takehiro Oka, Yoshinobu Urakami, Hajime Jimbo, Hiroshi Yakuwa, Manabu Noguchi, Toshio Narita: The Effect of Nb for the Intergranular Corrosion of SCH2 Heat Resistant Cast Steel in the Waste Incineration Environment, Zairyo-to-Kankyo, Vol. 54, No. 5, p. 215 (2005).
12) Hidenori Takahashi, Yasuki Miyakoshi, Syuichi Kamota, Takehiro Oka, Yoshinobu Urakami, Manabu Noguchi, Hiroshi Yakuwa: The Effect of Nb and Mo for High Temperature Corrosion Behavior of SCH2 Heat Resistant Cast Steel in the Waste Incineration Environment, Proceedings of the 53rd Japan Conference on Materials and Environments, A-303, 105 (2006).
13) High temperature Corrosion and Prevention Methods by Combustion Gas in Boiler, Masaharu Nakamori, Ed.; Technosystem, p. 294 (2012).
14) Manabu Noguchi, Hiroshi Yakuwa, Matsuho Miyasaka, Hideyuki Sakamoto, Shigeru Kosugi, Toshio Narita: High Temperature Erosion-Corrosion Behavior of Boiler Tube Materials in Fluidized-Bed Waste Incinerator Conditions, Proceedings of HTCP2000, 573 (2000).
15) Manabu Noguchi, Hiroshi Yakuwa: Lecture on Fundamental Aspects of High Temperature Corrosion and Corrosion Protection, Part 1: Basic Theory, Ebara Engineering Review, No. 252, p. 31, (2016-10).
16) Keiji Sonoya, Ichiro Kajigaya, Muneharu Shimazaki: Erosion Properties of the Stainless Steel and Cr-Mo Steel in Fluidized Bed Condition, Tetsu-to-Hagane, Vol. 84, No. 12, p. 45 (1998).
17) Mohammad Emami, Shigenari Hayashi, Takashi Kogin, Manabu Noguchi: Erosion-Corrosion Behavior of Metals in Chlorine Containing Oxidative Atmospheres, Proceedings of the 63rd Japan Conference on Materials and Environments, A-301, 65 (2016).
18) Boshoku Gijutsu Binran (Corrosion Protection Handbook), Japan Society of Corrosion Engineering, Ed.; Nikkan Kogyo Shimbun, p. 103 (1986).
Under the Scenes of our Lives High-pressure pump - Role and Application -
50% capacity boiler feed pump(BFP）playing an active role in a 1 000 MW thermal power plant
Large-capacity, Ultrahigh-efficiency, High-pressure Pumps for Seawater RO Desalination Delivered to Carlsbad Desalination Plant in the U.S.
Streamlines in crossover passage and velocity distributions at inlet of the second-stage impeller (Left:original,Right:optimized)
Discussion Meeting Symposium Ebara research system - Cooperation between research and business to create a new future -
Discussion Meeting (Mr. HIYAMA, Mr. SOBUKAWA, Mr. GOTO）
Under the Scenes of our Lives Standard Pumps - Essential Part of our Everyday Lives -
Examples of standard pumps
Inquiry about Ebara Engineering Review
Inquiry Form