The pitting corrosion behavior of UNS S82441 duplex stainless steel annealed at six different temperatures ranging from 1,000°C to 1,130°C for 1 h has been investigated by means of the potentiostatic critical pitting temperature. The microstructure evolution and pit morphologies of the specimens were studied using optical/scanning electron microscopy. The results demonstrated that the volume fraction of the austenite phase decreased with the increasing annealing temperature. Lower critical pitting temperature values were obtained after annealing at higher temperatures. Pitting was initiated preferentially inside the ferrite domains, indicating that the ferrite phase had inferior pitting corrosion resistance as compared to the austenite phase. The pitting resistance equivalent number of the ferrite phase fell with the annealing temperature, while the values for the austenite phase rose. No equal pitting resistance equivalent number of the ferrite and austenite phases was found for this steel in the range of annealing temperatures.
Duplex stainless steels (DSS) are ferritic-austenitic steels with approximately equal volume fractions of ferrite and austenite, offering an attractive combination of mechanical properties and corrosion resistance.1–3 UNS S82441(1) is a newly introduced molybdenum and nitrogen-containing duplex stainless steel with a typical composition of 24Cr-3Mn-4Ni-0.23N-1.5Mo. This grade is designed to meet the requirements of high yield strength and high corrosion resistance in chloride-containing environments, with a lower cost compared with the conventional DSS 2205 (UNS S31803). Therefore, these characteristics make UNS S82441 well suited for various applications. One common way to describe the relative corrosion resistance is to use the pitting resistance equivalent number (PREN).4 It is well known that the higher the concentrations of the three most important alloying elements (Cr, Mo, N), the better the pitting corrosion resistance of the stainless steels. Because of the two-phase structure of DSS, the key alloying elements (Cr, Mo, Ni, and N) are partitioned between ferrite (α) and austenite (γ) with Cr and Mo being enhanced in ferrite and N, Ni being enhanced in austenite, leading to the different corrosion resistance between α phase and γ phase.5 Besides the chemical composition, the annealing treatment is another important factor influencing the corrosion resistance of DSS.6–7 At a given chemical composition, varying the annealing temperature can change the proportion of ferrite and austenite phases, redistributing the alloying elements in the two phases, causing the variation of the corrosion resistance of each phase. On the other hand, improper heat treatment temperature can result in deleterious secondary phases. It has been reported that annealing below 1,000°C will lead to the precipitation of sigma phase,8 and fast cooling from temperatures higher than 1,200°C causes the formation of chromium nitride in DSS.9–11 These secondary precipitates reduce the toughness and corrosion resistance of the DSS.
During the last few years, the corrosion resistance and mechanical properties of DSS have been studied intensively.12–21 Moura, et al., investigated the influence of microstructure on the corrosion resistance of DSS UNS S31803.22 Invernizzi, et al., studied the corrosion behavior of DSS (UNS S31803 and UNS S32750) in organic acid aqueous solutions.23 Merello, et al., discussed the influence of chemical composition on the pitting corrosion resistance of nonstandard low-Ni high-Mn-N DSS.24 Our group has investigated the critical pitting and repassivation temperatures of DSS in chloride solutions.25 Modification of the microstructures of DSS should influence greatly their overall pitting corrosion behavior. However, very little data is available regarding the effect of the annealing treatment on the microstructure and the corrosion resistance of UNS S82441. The goal of the current work was to examine the effect of the annealing temperature on the phase volume, element partitioning, and pitting corrosion behavior of UNS S82441.
In the present work, the pitting corrosion resistance of a newly developed DSS UNS S82441, annealed at different temperatures in the range from 1,000°C to 1,130°C without apparent precipitation of deleterious phases, has been investigated by a critical pitting temperature (CPT) technique, which is more sensitive and reliable than other pitting characterizing approaches such as pitting potential.25–30 The relationship between the CPT and the PREN of the single phase has been discussed for UNS S82441 DSS.
Material and Sample Preparation
A 15-mm, thick, hot-rolled plate UNS S82441 DSS (chemical composition in wt% C: 0.01, P: 0.012, Ni: 3.39, Cr: 24.01, Mn: 3.01, Si: 0.37, Mo: 1.54, Cu: 2.29, N: 0.23, Fe: bal.) was purchased. All specimens were cut into 10 by 10 by 3 mm coupons, and then solution-annealed at 1,000°C, 1,030°C, 1,050°C, 1,080°C, 1,100°C, and 1,130°C for 1 h in flowing Ar, respectively, and then water-quenched.
Microstructure examination of the as-annealed specimens before and after the CPT test was conducted using light optical microscopy and scanning electron microscopy (SEM).
Before the observation, the specimens were polished with 3.5 μm diamond paste and then electrolytically etched with 30% potassium hydroxide (KOH) electrolyte at 2.0 V vs. saturated calomel electrode (SCE) for 15 s, which made the austenite phase bright and the ferrite phase dark. The volume fraction of austenite was measured using an optical microscope equipped with a quantitative metallographical analysis system. The average of 10 measurements of the phase fraction was taken as the phase volume fraction.
The concentrations of chromium, molybdenum, nickel, and manganese in the ferrite and austenite phases were measured using energy-dispersive x-ray spectroscopy (EDX), which was equipped with the scanning electron microscope. Each value of the element concentration was an average of 10 measurements. Thermo-Calc† software was used to calculate the phase diagram of UNS S82441.
Electrochemical corrosion behavior of the specimens was carried out with a PARSTAT2273† potentiostat using three-electrode cells. A platinum foil and a SCE were used as the counter and reference electrodes, respectively. All potentials quoted in this paper referred to the reference electrode. The specimens acting as working electrodes were embedded in epoxy resin. To avoid crevices, the area of the specimen near the interface between the resin and the specimen was sealed with silica gel sealant and dried in the air. Potentiostatic measurements were performed to obtain the CPT. Prior to each CPT measurement, the working electrode was ground with successive grade silicon carbide (SiC) sandpaper up to 1000 grit, degreased with ethanol (C2H6O), rinsed with distilled water, and dried in air. The test solution, 1 M sodium chloride (NaCl), was made up of analytical grade reagent and distilled water.
The CPT measurement was performed according to ASTM G150-99.31 Before the measurement, the surface of the working electrode was polarized cathodically at −900 mVSCE for 120 s to reduce the oxide film formed in air. Then, the specimen was stabilized at open-circuit potential for 30 min in test solution at 2°C. After that, a static potential of 750 mVSCE was applied to the working electrode and the solution temperature was increased at a rate of 1°C/min controlled by a water bath. The corresponding current was recorded simultaneously with the increasing temperature. The CPT was determined from the current vs. temperature curve as the temperature where the current density exceeded 100 μA/cm2. The test solution was bubbled with pure nitrogen gas (N2) to get rid of the dissolved oxygen (O2) before and throughout the test. The CPT measurements of the same specimen were repeated three times. Any crevice corrosion observed on the specimens after testing meant that the test results were invalid and must be discarded.
Morphology of Corrosion Attacking
SEM was used to observe the morphology of the corrosion attacks produced on the specimens' surfaces after the CPT measurements.
Figure 1 presents the microstructures of specimens annealed at 1,000°C, 1,030°C, 1,050°C, 1,080°C, 1,100°C, and 1,130°C for 1 h. The typical ferrite-austenite duplex structure was observed. Austenite (γ) is light and ferrite (α) is dark. The austenite was evenly distributed as islands in the ferrite matrix and elongated along the rolling direction, showing typical ribbon-like morphology. No other deleterious phase was found in the annealed specimens. As can be seen in Figure 1, the austenitic volume fraction decreased with the increment of the annealing temperature, which was confirmed by the quantitative metallography. The volume fraction of the ferrite and austenite phases measured by the quantitative metallography was plotted against annealing temperature, as shown by Figure 2. The evolution of the two-phase microstructure was consistent with the results calculated by the software (Figure 3), which can be explained by the transformation of γ → α at elevated annealing temperatures. Figure 3 shows that only ferrite and austenite are predicted above 1,000°C, up to the initiation of melting.
The average contents of the main alloying elements analyzed by EDX in ferrite and austenite phases of all annealing treated UNS S82441 specimens are listed in Table 1. Because of the insensitivity of EDX techniques to nitrogen, an approximate calculation was used to determine the nitrogen content of each phase. The nitrogen levels in the ferrite phase were assumed to be the saturation value, 0.05 wt%; the nitrogen level in the austenite phase was thereby the remainder, which could be calculated based on the content of nitrogen in the whole alloy and the phase volume fraction.32 It shows that alloying elements such as Cr and Mo were preferentially enriched to the ferrite phase, while Ni, Mn, and N were preferentially partitioned to the austenite phase, which was in agreement with literature.25,27,33–34 With the increase of the annealing temperature, the content of Cr and Mo in ferrite decreases, which has been reported in other duplex stainless steels.27 Since the volume fraction of ferrite phase increased by raising the annealing temperature, the original ferrite-forming elements (Cr and Mo) were spread over a larger volume and diluted, resulting in a decrease of Cr and Mo concentration in the ferrite phase.
The influence of annealing treatment on the pitting corrosion resistance was estimated by the potentiostatic CPT measurement. Figure 4 illustrates the typical curves of the current density vs. temperature obtained from the UNS S82441 specimens in 1 M NaCl solution with different annealing temperatures. The current density was less than 1 μA/cm2 during the initial heating, indicating that DSS was protected by the passive film on the surface. With the increase of the solution temperature toward the CPT, some current density fluctuations were found, which were associated with a passive film breakdown in the form of metastable pits. As the temperature increased further, reaching to the CPT, the current density rose sharply as a result of the occurrence of the stable pits. The relationship between the CPT and the annealing treatment temperatures was also plotted in Figure 4. It is found that as the annealing temperature increased from 1,000°C to 1,130°C, the CPT decreased from 46°C to 36.5°C. The specimens annealed at 1,000°C had the best pitting corrosion resistance, and further discussion will be presented in the following part.
Characterization of the Corrosion Attack
Surface morphologies of the specimens were studied after the CPT test. The metastable pits of specimens annealed at different temperatures were located at the ferrite phase. Figures 5(a) and (b) show metastable pits with diameters of 1 μm to 2 μm formed on the surface of the specimen annealed at 1,000°C and 1,080°C, respectively, below the CPT. Metastable pits always preferentially nucleated inside the ferrite domains. Figures 5(c) and (d) illustrate the stable pits formed on the surface of the specimen annealed at 1,000°C and 1,080°C above the CPT. Pit morphologies for all the annealed specimens showed the same tendency. It was found that the characteristic mechanism of pitting corrosion for the specimen was the selective dissolution of the ferrite phase. Pits first initiated inside the ferrite grain and then propagated into the ferrite phase until they reached the austenite phase, indicating that pit growth was restrained by the austenite phase. The locations marked in Figures 5(c) and (d) clearly reflect these findings.
The microstructural investigation of UNS S82441 in the temperature range from 1,000°C to 1,130°C exhibited the expected results for this steel; i.e., the volume fraction of ferrite phase increased with annealing temperature, as shown by Figure 1. Meanwhile, the pitting corrosion resistance of specimens decreased. The observed occurrence of pits in the ferrite phase was contrary to the findings of some previous studies, in which both austenite and ferrite phases were attacked.35–36 This can be explained by the PREN of the two phases. As mentioned previously, a higher PREN value has widely been considered to result in better pitting corrosion resistance.5 PREN was defined by the following equation:34 PREN = wt% Cr + 3.3 wt% Mo + 30 wt% N – wt% Mn. Based on the compositions of alloy elements listed in Table 1, the relationship between the calculated PREN for a single phase vs. annealing temperatures was illustrated in Figure 6. Since the PREN values of the ferrite phase were always lower than that of the austenite phase, the evolution of the pitting corrosion behavior of this kind of alloy depended on the ferrite phase. Figure 6 shows that the PREN of ferrite decreased with annealing temperature, while the PREN of austenite increased, and no equal PREN value in ferrite and austenite phases was found within the annealing temperature range of 1,000°C to 1,130°C. This behavior is similar to that of another type of high nitrogen-containing DSS, UNS S32101.34 The varying of PREN of austenite and ferrite at annealing treatment condition was primarily because of the diffusion of key alloying elements and the evolution of the two-phase volume fraction. Higher annealing temperatures increased the ferrite content and diluted the key alloying elements in the ferrite phase, lowering the corrosion resistance of the ferrite phase. The concentration of the Cr and Mo in the ferrite phase decreased as the annealing temperature increased from 1,000°C to 1,130°C. Meanwhile, as seen in Table 1, the calculated levels of N in the austenite increased drastically (the N levels in the ferrite were assumed to be 0.05 wt% and that the austenite phase was thereby the rest). This is consistent with other findings.34,37 Therefore, the PREN value of ferrite decreased and that of austenite increased. In fact, pitting corrosion was the selective dissolution of ferrite phases, as shown in Figures 5(c) and (d). Experimental results have shown that the CPT of UNS S82441 DSS decreased as the annealing temperature increased, which was consistent with the decreasing PREN for the ferrite phase. In summary, the evolution of pitting corrosion resistance of UNS S82441 was determined by the weaker phase (ferrite phase).
❖ The volume fraction of austenite in UNS S82441 decreased continuously as the annealing temperature increased from 1,000°C to 1,130°C. Raising the annealing temperature within this range lowered the CPT of UNS S82441.
❖ The pitting corrosion of UNS S82441 in chloride solutions took place preferentially inside the ferrite domains, which can be explained reasonably by the change of PREN of the two single phases.
❖ The PREN of the ferrite phase decreased with increasing temperature from 1,000°C to 1,130°C, while the PREN of austenite increased. There was no equal PREN value between the ferrite and austenite phases for UNS S82441.
❖ The pitting corrosion resistance of UNS S82441 was determined by the weaker phase (ferrite), which has been proved by the pit locations and morphologies in specimens annealed at different temperatures.
The authors would like to acknowledge the helpful collaboration of Baosteel Co., Ltd. This work was supported by the National Science Foundation of China (Grant nos. 50871031, 51131008, 51134010, and 51071049) and Industrialization Project of Shanghai New and High Technology.
(1) UNS numbers are listed in Metals and Alloys in the Unified Numbering System, published by the Society of Automotive Engineers (SAE International) and cosponsored by ASTM International.
*Department of Materials Science, Fudan University, Shanghai 200433, P.R. China.
**Research and Development Center, Baosteel Co., Ltd, Shanghai 201900, P.R. China.