Analysis of Corrosion in the Cooling Coils of Nitric Acid Absorption Towers and Countermeasures
Thread Content
Our company is one of the major nitric acid producers in the country, and it began operations in November 1990. However, from 2000 to 2003, there were 6 consecutive shutdown incidents caused by cooling coil leaks, resulting in direct and indirect economic losses of over 6 million yuan. To determine the cause, our company established a task force to analyze the common factors in these accidents and propose corresponding control measures. 1 Accident statistics (see Table 1): http://www.nmtech.com.cn/jishuwang/upload/060205844433263.jpg 2 Cause analysis: Based on the data in Table 1, the leakage points are mainly located at the junctions between the cooling coils and the tower walls; these leaks are caused by pitting corrosion and crevice corrosion. What causes these two types of corrosion? After careful research, the reasons were identified, and simulation tests were conducted to verify our conclusions. 2.1 Reason 1 – Welding: The material used for the cooling coils of our company’s absorber towers is 1Crl8Ni9 stainless steel. At room temperature and in the absence of other substances, this steel exhibits good corrosion resistance to nitric acid at concentrations below 70%; a dense film of metal oxide forms on the metal surface, protecting the underlying metal from corrosion. Although this austenitic stainless steel possesses high corrosion resistance, during welding, the areas 2–3 mm on either side are heated to 400–910°C, thereby creating what is known as a intergranular corrosion-sensitized zone. At this point, chromium and carbon at the grain boundaries combine to form Cr23C6; during natural cooling, this compound precipitates in the form of a solid solution, which prevents chromium from diffusing from within the grains to the grain boundaries. As a result, chromium-poor regions are formed at the grain boundaries, with the chromium content dropping below 11%. The chromium content in steel must be greater than 11% to achieve good corrosion resistance, as the precipitation of chromium carbide and the formation of chromium-deficient areas can cause chromium carbide to act as the cathode and the chromium-deficient areas to act as the anode, thereby creating a galvanic cell that results in the loss of corrosion resistance in stainless steel. We conducted experiments on the corrosion of the fittings both before welding and after welding, and the corrosion rate curves are shown in Figure 1. http://www.nmtech.com.cn/jishuwang/upload/060205845556438.jpg One cycle is defined as 48 hours of gentle boiling with 65% nitric acid. As can be seen from Figure 1, the corrosion rate of the unwelded pipe segments remains relatively unchanged after 5 cycles ； After welding, severe corrosion occurred at 2.0–2.8 mm on the left side and 1.9–3.0 mm on the right side of the weld seam in the pipe segments, with obvious pitting and corrosion pits appearing. 2.2 Reason 2 – The impact of Cl-: The Cl- concentration in the acid solution of our company’s absorption towers is generally high, at around (300–400)×10-6; especially in the period leading up to the accident, analysis data showed a value of (400–500)×10-6. Cl- is an active ion; when present in the environment surrounding steel, the passivation film is prone to local damage in areas with low corrosion resistance, resulting in tiny defects in the film. The metal at the hole becomes the anode, while the large area of membrane surrounding the hole acts as the cathode. The electric current is highly concentrated, causing corrosion to progress rapidly inward; that is, at the anode, the following reaction occurs: M—e→M+ (where M represents metal) ； Cathode reaction: 02 + 2H2O + 4e → 4OH- ； This causes the oxygen in the corroded area to be quickly depleted, leaving only the anodic reaction. A large amount of positively charged metal ions quickly accumulated inside the pores. To maintain electrical neutrality, the mobile Cl- ions in the environment rapidly migrate into the pores to form metal chlorides. The hydrochloric acid generated by the hydrolysis of these metal chlorides causes more metals to dissolve, which in turn leads to more Cl- ions migrating into the pores and significantly increasing the concentration of hydrochloric acid. When the molar ratio of hydrochloric acid to nitric acid reaches 3:1, the following reactions occur: 3HCl + HNO3 → Cl2 + NOCl + 2H2O; Cl2 + 2NO → NOCl; NOCl + H2O → HCl + HNO2; 2Cr + 2HNO3 → Cr2O3 + 2NO + H2O; Cr2O3 + 6HCl → 2CrCl3 + 3H2O. The occurrence of these reactions results in the presence of strong oxidizing agents such as HNO3, Cl2, and NaCl within the pores, which greatly increases their oxidation potential. This exceeds the passivation region, entering a state of over-passivation, and corrosion intensifies, causing the pores to deepen into holes that lead to pipeline leaks. Figure 2 shows the corrosion curves at different Cl- concentrations. http://www.nmtech.com.cn/jishuwang/upload/060205846455431.jpg As can be seen from Figure 2, as the Cl- concentration increases, the corrosion rate rises rapidly, reaching a maximum of 9.0 g/(m2·h); obvious corrosion pits appear, and the color of the solution changes to dark green. This further verifies the harmful effect of Cl- on 1Crl8Ni9. 2.3 Reason 3 – Erosion: The nitric acid in the absorption tower is a flowing liquid that causes scouring and wear in the corroded areas. Combined with the reasons in 2.1 and 2.2, this further exacerbates the corrosion of stainless steel. Strategy 3.1: Improving welding quality. To prevent the formation of Cr23C6 solid solution precipitates during the cooling process after welding, which could lead to chromium-deficient areas, water quenching is employed immediately after welding, thereby preventing Cr23C6 from having time to precipitate and ensuring a uniform crystal structure. 3.2 Reducing carbon content in stainless steel: Some stainless steels are replaced with 0Crl8Nil0. The self-healing and self-repairing ability of this material is several times greater than that of 1Crl8Ni9, and its carbon content is lower than that of 1Crl8Ni9. 3.3 Control of Cl- concentration (1) Strictly control the Cl- content in process water to keep it at ≤2×10-6. (2) Periodically analyze the Cl- in the absorption tower and remove it promptly to keep the Cl- concentration in the acid below 200×10-6. 4. Results: Since the repair of our company’s cooling coil last year, there have been no further leaks. A hydrostatic test was conducted during the maintenance on April 28, 2004, and the cooling coils were in perfect condition.
The former dilute nitric acid production unit of Lanzhou Petrochemical Fertilizer Plant consisted of 3 sets of dilute nitric acid generation equipment (No. 6, No. 7, No. 8). The absorption towers were of the foam sieve-plate type, with specifications of φ3000mm×8mm and a height of 46150mm; they were made of 1Crl8Ni9Ti material. These units were capable of producing 4.6 tons of dilute nitric acid with a concentration of 47%–49% per hour. They have been in use for nearly 40 years (since 1964), with an average effective utilization rate of 60%. In July 2003, corrosion and leakage were detected in multiple areas of the lower tower wall and the welds at the base of Tower No. 7. When grinding and welding repairs were carried out from the outside of the tower, it was found that most of the fusion zone and heat-affected zone of those welds had suffered corrosion penetration, making it impossible to carry out welding. As a result, several severely corroded welds had to be repaired from inside the tower (using a lap welding method) before the tower could be put back into use. At the beginning of 2004, about three months after the tower had been in operation, leakage occurred again at the original leak site. Moreover, during a pressure test to check for air leaks at the welds on both sides of the tower’s waist section, perforation and corrosion were found in the base material of the tower wall, and the situation was quite serious. The other two absorption towers were also inspected promptly, and the same situation was found there as well. Since the areas most severely affected by corrosion are located in the lower part of the absorption tower (below 4 meters), the upper part of the tower is in relatively good condition and can still be used, so a proper plan has been developed for its repair. 1 Analysis of corrosion causes 1.1 Knife-like corrosion on the base metal on both sides of the welds 1.1.1 Cause analysis Observations inside the tower revealed that groove-shaped cracks (with a width of 1–2 mm) appeared to varying degrees at the fusion lines on both sides of the welds on the tower walls and bottom. As shown in Figure 1. Based on its morphology and the material composition of the tower, this is likely to be a typical knife-edge corrosion (a special type of intergranular corrosion). It occurs primarily in Ti-containing 18-8 austenitic stainless steels (such as 1Crl8Ni9Ti) that have been stabilized through heat treatment at temperatures between 850°C and 950°C for several hours. This type of corrosion differs significantly from the intergranular corrosion that takes place in the heat-affected zone of austenitic stainless steels, where such corrosion occurs over a wider area at greater distances from the weld seam; it is also more dangerous. http://www.nmtech.com.cn/jishuwang/upload/060912826116901.jpg Austenitic steel is generally supplied in a solution-treated state. Taking 18-8 steel with 0.08% carbon as an example, it is generally solution-treated by water quenching at 1050–1150°C. At this point, a small amount of carbon and trace amounts of Ti dissolve into the solid solution, while the majority of the remaining carbon and Ti combine to form free TiC. This is because the solubility of TiC in steel is low at temperatures below 1150°C; on the other hand, the chromium carbide Cr23C4 can dissolve entirely in the solid solution. However, during welding, the first change that occurs in the overheated region where the temperature exceeds 1200°C is that TiC can continuously dissolve into austenite to form a solid solution. The higher the peak temperature, the greater the solubility of TiC. At this point, only a small amount of large TiN particles in the superheated zone cannot undergo solid solution. When TiC dissolves, the separated carbon atoms insert into the interstices of the austenite lattice, while Ti occupies the vacant positions at the lattice nodes of austenite. During subsequent cooling, due to the high reactivity of carbon atoms at high temperatures, which gives them a stronger diffusion capacity than Ti, these carbon atoms tend to diffuse toward the peripheries of the austenite grains, while Ti does not have enough time to diffuse and remains at the lattice nodes of austenite. Therefore, carbon will accumulate near the grain boundaries, reaching a supersaturated state. If heated again via meso-temperature sensitization, carbon atoms can preferentially diffuse toward the grain boundaries at a rapid rate, thereby enriching those boundaries with carbon. At this time, although the diffusion of Cr is not as fast as that of carbon, it is faster than that of Ti; therefore, it is easy to form precipitates of the chromium carbide Cr23C6 near the grain boundaries. The greater the amount of TiC dissolved in a region (i.e., the area immediately adjacent to the weld line), the more Cr23C6 precipitates, and thus the greater the tendency for intergranular corrosion in that area. That is, the distribution of tool erosion areas and Cr23C6 deposits is consistent, presenting as tool-cutting corrosion in the vicinity of the seam. It can be seen that the sequential action of \"high-temperature superheating\" and \"moderate-temperature sensitization\" is a necessary condition for the occurrence of knife erosion. 1.1.2 Preventive measures: To prevent knife erosion, it is best to use ultra-low carbon stainless steel. For steels containing strong carbide-forming elements such as Ti or Nb, the carbon content is preferably less than 0.06%. In terms of welding techniques, it is first necessary to reduce overheating in the area near the weld seam, and in particular, avoid any heating effects that could lead to medium-temperature sensitization during the welding process. For joints or components that inevitably have to operate at the sensitization temperature, stabilization treatment should be applied to improve them. 1.2 Pitted corrosion in the base metal of the tower wall 1.2.1 Cause analysis During the airtightness test on the welds after repair work was carried out on the bottom of Tower No. 7, pinhole defects were found in certain areas of the tower wall’s base metal, as shown in Figure 2. This indicates that deep perforations (resembling ant nests) have formed in the tower wall; this type of corrosion is also known as pitting or pit corrosion, and it is a common form of localized corrosion in stainless steel. Nitric acid absorption towers often contain some Cl‑ ions, which can penetrate the passivation film on certain areas of the stainless steel surface – mainly those areas with higher concentrations of impurities and defects – thereby creating active (metal-exposed) and passive (passivation film-present) corrosion cells. Since the anode area is much smaller than the cathode area, the current density at the anode is high; as a result, corrosion progresses deeper, rapidly forming anthill-like holes. Although pitting results in a small loss of weight, the corrosion rate is high, and in many cases it exacerbates intergranular corrosion, stress corrosion, and fatigue corrosion, making it more dangerous. http://www.nmtech.com.cn/jishuwang/upload/060912827349014.jpg Pitting corrosion of stainless steel occurs in specific corrosive environments; it typically takes place in solutions containing halide anions, among which chlorides and bromides are the most corrosive. Pitting can occur only when erosive halide anions reach a certain concentration; oxidizing metal ions (such as Hg2+, Fe3+, Cu2+) can promote the formation of pitting. O2, H2O2, and other oxidants in the solution are necessary for the formation of pitting, as these oxidants have a depolarizing effect ； However, certain oxygen-containing anions in the solution (such as hydroxides, chromates, nitrates, and sulfates) can prevent pitting, as they displace the chloride ions on the metal surface. The state of the solution also affects corrosion resistance; pitting is more likely to occur when the solution is at rest compared to when it is in motion ； At the same time, the pH value of the solution also has a significant impact on pitting; as the pH value increases, the tendency for pitting decreases markedly. Raising the temperature accelerates the adsorption of aggressive chloride ions onto the surface of stainless steel, as well as the movement of the substances involved in the reaction; this leads to an increase in activation sites, thereby exacerbating pitting corrosion in stainless steel. 1.2.2 Preventive measures Based on the causes of pitting corrosion, preventing it can be approached from two aspects: selecting materials resistant to pitting corrosion ； Improve the environment in which materials are used. Specifically, the following prevention methods are adopted: a) In austenitic stainless steel, appropriate amounts of molybdenum are added and the chromium content is increased, along with the use of a suitable heat treatment regime. b. Reduce the concentration of halide ions in the solution, especially chloride ions, and avoid local concentration of the solution. c. Stir the solution to homogenize the concentrations of oxygen and oxidizer in it, keeping the solution from remaining static. d. Increase the flow rate of the solution to prevent impurities from adhering to the steel surface. Increase the pH value of the solution. e. Add a corrosion inhibitor. f. Lower the temperature of the medium. g. Use cathodic protection to keep the potential of the material below the critical pitting potential. 1.3 Intergranular corrosion of tower wall welds and their heat-affected zones 1.3.1 Cause analysis There are many reasons for intergranular corrosion to occur. It is generally believed to be due to the depletion of intergranular alloying elements. Stainless steel is corrosion-resistant mainly because it contains certain amounts of elements such as chromium and molybdenum that can act as passivators. If chromium- and molybdenum-rich phases precipitate at the grain boundaries, these phases are primarily composed of M23C6 and M7C3 (where M represents chromium, molybdenum, and iron; copper is also present in copper-containing alloys). The chromium content in these precipitated phases can be as high as 95%, resulting in a region lacking chromium and molybdenum along those grain boundaries. When the chromium and molybdenum contents in the depleted zone drop below the limits required for passivation, the grain boundaries in that zone act as anodes, and intergranular corrosion occurs under the action of the corrosive medium. Another view is that intergranular corrosion occurs due to the formation of certain precipitates at the grain boundaries; these precipitates are corroded first in the corrosive medium, thereby causing intergranular corrosion. As for intergranular corrosion in some non-sensitized steels, it is believed to be primarily caused by the adsorption of impurities such as phosphorus and silicon at the grain boundaries. In short, the theory more widely accepted as the cause of intergranular corrosion is the theory of alloy element depletion at grain boundaries. Intergranular corrosion can occur in austenitic stainless steels, ferritic stainless steels, or austenitic-ferritic duplex stainless steels. The sensitization heating temperature range for intergranular corrosion in austenitic and duplex stainless steels is between 450 and 850°C ； Ferritic stainless steels, on the other hand, are above 850°C. The original nitric acid absorption tower was made of 1Crl8Ni9Ti; due to the presence of the strong carbide element Ti, intergranular corrosion should generally not occur. However, inspections of the tower revealed varying degrees of network-like or honeycomb corrosion in the welds and heat-affected zones at the lower part of the tower. Based on the morphology pattern of intergranular corrosion, it can be determined that this belongs to intergranular corrosion. Why does this happen? We believe the main reasons are as follows: a) Due to the limitations of the smelting technology at that time, the levels of various harmful elements in the material of the absorption tower were kept too high, especially the carbon content ； Moreover, the ferrite content in the tissue is too low (it should be ≥5%). b. The temperature at the bottom of the tower is relatively high (60°C–70°C); 1Crl8Ni9Ti is generally used in dilute nitric acid at low temperatures