Effect of Sulfur and Chlorine on Fireside Corrosion Behavior of Inconel 740 H Superalloy

Introduction

The improvement of the steam temperature and pressure of the coal fired boilers can effectively increase the power generation efficiency and reduce emissions, such as NO_x, SO_x and CO₂ [1]. Advanced, ultra-supercritical steam conditions may include temperature values between 700–760 °C and a pressure of 35–37.5 MPa [2]. The requirements imposed by these conditions, which include long-term creep strength (100 MPa for 10⁵h) and higher corrosion resistance (2 mm thickness loss for 2×10⁵h) [3] are clearly beyond the temperature, creep strength and corrosion resistance capacity of the materials employed up to date, i. e. ferritic and austenitic stainless steels, such as the T91, Super304H, TP347H FG and HR3C alloys [4, 5, 6, 7]. Even some nickel-based superalloys, such as the CCA 617, do not meet the high temperature corrosion resistance requirements for such applications [8].

Recently, a novel nickel-based superalloy, Inconel 740 H, has been designed based on the Nimonic 263 alloy. Compared to Nimonic 263, the Inconel 740H has increased Cr and decreased Mo content for increased corrosion resistance [9]. There are plenty of studies about the stability and corrosion resistance of Inconel 740H [10, 11, 12]. It has also been reported that this alloy has higher strength and improved corrosion resistance that meets the above requirements for ultra-supercritical power plants [10].

Fireside corrosion is a net result of the synergistic effect of parameters such as alloy composition, temperature, flue gas, and coal ash. Experimental results of laboratory tests employing three materials [13], four levels of SO₂ [14] and four temperature values [15], indicated that the sulfur content is an important factor for fireside corrosion, where the corrosion rate was increased with higher SO₂ levels. The study of Syed et al. [7] revealed that the corrosion damage of the alloy was increased by 20 % when the sulfur content increased from 0.13 % to 0.63 %. Hack et al. [16] applied statistical analyses on a series of alloys used for coal-fired boiler in low sulfur and high sulfur coal, pointing out the average wall thickness loss, ranging from 0.4 to 1.2 % and 0.6 to 2 % in low and high sulfur coal, respectively. Zeng et al. [17] also studied the materials corrosion behavior of boilers at 750 °C flue gas environments and there results show that the corrosion rate was increased up to 2 mm/y when the sulfur content reached 1 %. These results have substantially exceeded the corrosion resistance requirements for ultra-supercritical boilers [3].

Moreover, it was also noted that chloride-induced fireside corrosion is an issue that may dramatically increase the corrosion rate and lead to severe corrosion damage [18, 20, 20]. The acceleration of the reaction of chloride-induced corrosion has been investigated [21]. It is generally accepted that the melting point of alkali chlorides was lower and the corrosion products (such as FeCl₃, CrCl₃ and Cl₂) have higher relative vapor pressure [22]. Cl₂ has a crucial, catalytic contribution in the corrosion process by generating newly volatile metal chlorides [23]. In some cases this specific process is termed as “active oxidation”.

Although the studies on fireside corrosion behavior of various alloys are plenty, limited information is available on the influence of sulfur content on the Inconel 740H corrosion behavior, particularly in high sulfur and chlorine environments. Thus, the two main objectives of this work are, first, to compare two sulfur concentrations (0.1vol % and 1.5vol %) in order to understand the corrosive tendency of Inconel 740H and, second, to obtain more information about the role of chlorine during high temperature corrosion.

Experimental procedure

Corrosion tests

A schematic diagram of the experimental apparatus is shown in Figure 1. The samples were hold in the constant temperature region of the chamber and were covered with a synthetic mixture of deposits consisting of 39 % SiO₂, 22 % Al₂O₃, 6 % Fe₂O₃, 29 % CaSO₄, 2 % Na₂SO₄, 2 % K₂SO₄ and with either 2 % NaCl or without it (compositions in wt. %). Then, a gas flow mixture of 80 % N₂, 10 % CO₂, 3.5 % O₂, 5 % H₂O and either 0.1 % or 1.5 % SO₂ (compositions in vol. %) was fed through a Pt-based catalyst, which was maintained at above 450  °C to ensure the SO₂/SO₃ equilibrium in the gas mixture. In this study, the amount of the coated synthetic ash was approximately 40～50 mg/cm² and the gas flow through the test chamber was at a rate of approximately 100 ml/min. During the test, the specimens were exposed for a certain time, removed from the furnace, cleaned up the deposits, boiled without any soluble salts, dried and weighed. Subsequently, the exposures were continued with a fresh deposit.

Figure 1:

Setup of the experimental.

The corrosion exposure reported here was performed at 750 °C. Three samples of the tested alloy were tested under the same testing conditions. The mass gain curves were drawn by the average mass change value, which was measured using an electronic balance (sartorius balance) with a detection limit of 0.01 mg. The surfaces and cross-sections were characterized using scanning electron microscopy (SEM), coupled with X-Ray energy-dispersive spectroscopy (EDS), in a HITACHI S-4800. X-Ray diffraction (XRD) analyses were employed to determine the corrosion products, in a SHIMAZDU XRD-7000. Electroless nickel deposition was used to avoid oxide film cracking during sample preparation.

Results

Mass gain curves

The mass gain curves of the alloy under different corrosion conditions are presented in Figure 2. It can be observed that mass gain increases rapidly at the initial stages of both sulfur content atmospheres (0.1 % SO₂ and 1.5 % SO₂). After exposure for 20 h, the alloy demonstrated a substantial mass loss in the presence of 1.5 % SO₂. However, the maximum corrosion weight gain occurred at 50 h in the presence of 0.1 % SO₂. After exposure for 100 h, the mass gain curves of samples show the similar trends in both sulfur content atmospheres and the curves are approximately steady, which indicated that the alloy has entered into a stable corrosion rate stage. The mass gain of the samples in 1.5 % SO₂ atmosphere is approximately twice the value of the alloy with 0.1 % SO₂. For the samples exposed in 1.5 % SO₂ with NaCl environment, significant corrosion damage indicated by the linear decrease of mass gain is observed. The spallation of corrosion products occurred during the whole process of hot exposure.

Figure 2:

The mass gain curves of Inconel 740H at 750 °C.

Corrosion products

Figure 3 shows the XRD patterns of the tested alloy following corrosion testing under various conditions. From the diagram it is clear that the corrosion products of the corroded samples are similar for all conditions. The oxide films are primarily consist of Cr₂O₃ together with a small amount of (Co, Ni) Cr₂O₄ spinels. Some weak peaks of Al₂O₃ and TiO₂ were also identified.

Figure 3:

XRD patterns of samples corroded at 750 °C for 500 h.

Corrosion morphologies

Figure 4 presents the surface and cross-sectional morphologies of the tested samples following corrosion testing for 500 h at 750 °C. As illustrated in Figure 4(a)–(d), the morphologies of oxide films are vary. In the presence of 0.1 % SO₂, an integrated and compact oxide film was formed on the surface and the particles included in this film were relatively fine. This compact oxide film can protect the substrate from further corrosion, and therefore, a very thin and adherent oxide film was formed on the metal surface, as presented in Figure 4(b).

Figure 4:

Surface and cross-sectional morphologies of tested samples corrode at 750 °C for 500 h, (a) (b) 0.1 % SO₂, (c) (d) 1.5 % SO₂, (e) (f) 1.5 % SO₂ with NaCl addition.

On the contrary, in the presence of 1.5 % SO₂, the oxide film formed on the surface was characterized by a non-coherent texture and high porosity. The non-coherent structure is unprotected as the corrosive medium can easily penetrate through the oxide layer, leading to an increase in the thickness of the oxide layer, as presented in Figure 4(d).

Figure 4(e)–(f) shows the surface and cross-sectional morphologies of the tested samples following corrosion for 500 h at 750 °C, deposited with NaCl. The oxide film was found significantly exfoliated. Thicker and multi-layered films with numerous voids were formed on the surface and a large amount of internal corrosion products appeared.

To further illustrate the structure of the corrosion film of the tested samples, EDS X-Ray elemental mapping analysis was performed, as shown in Figures 5–7. By comparing Figures 5 and 6, it can be seen that the microstructure of the corrosion films formed on the tested samples was similar. The corrosion products are rich in Cr and O, including few amounts of Ti and Al. However, the key difference of the corrosion products in different SO₂ addition is the quantity and location of sulfides within the film. There is no evidence that proves the sulfur presence in the atmosphere with 0.1 % SO₂. However, the sulfides were significantly increased in the atmosphere with 1.5 % SO₂. It is noted that sulfur not only distributes at scale/substrate interface but also diffuses into the alloy substrate concentrates in the internal penetrates. Additionally, EDS mapping also revealed a substantial Cr-depleted area in both environments.

Figure 5:

EDS elemental mapping analysis of the corroded samples at 750 °C for 500 h with 0.1 % SO₂ addition.

Figure 6:

EDS elemental mapping analysis of the corroded samples at 750 °C for 500 h with 0.1 % SO₂ addition.

Figure 7:

EDS elemental mapping analysis of the corroded samples at 750 °C for 500 h with 1.5 % SO₂+ NaCl addition.

Figure 7 shows the EDS mapping analysis of the corroded alloy at 750 °C for 500 h. The outer, non-coherent layer was rich in Cr, O, Ni, Co and with small amount of Ti. Large amount of Al-rich or Ti-rich inner oxides were concentrated beneath the Cr-rich oxide scale. Also, a large number of sulfur distributed in film/substrate interface and innermost penetrates zone.

Discussion

Superalloys are generally characterized by enhanced corrosion resistance than the ferritic or austenitic steels that are nowadays used as structural materials for coal fired boilers. The study of Natesan et al. [24], which was focused on a series of alloys with varying Cr contents between 9 % and 48 %(by mass), confirms that the materials resistance is primarily dependent on Cr content and alloy containing more than 22 % (wt%) Cr was generally exhibiting satisfactory corrosion resistance. Alloys with Cr content higher than that value have a dynamic advantage of Cr diffusion. Additionally, Cr₂O₃ nucleates and grows fast and forms a continuous, coherent layer at a very short period of time. Once the initial dense film forms, further inter-diffusion process is inhibited, resulting into a significantly reduced corrosion rate. In the present study, the Cr content of tested alloy reached up to 23.5 %. Thus, the alloy tested within the presented study should exhibit improved corrosion resistance. Based on the results, however, the alloy performs a quite different corrosion behavior in different sulfur contents, with or without NaCl deposition.

Between the two sets of conditions of sulfur content, the coal ash hot corrosion process of tested alloy can be described as follows: In the initial exposure stage, some types of oxides (Ni, Co) O and Cr₂O₃ were formed at the alloy surface and the internal oxidation reaction of some minor elements, such as Al or Ti, took place simultaneously, which led to a rapid mass gain, as presented in Figure 2. The (Ni, Co) O particles were surrounded by Cr₂O₃ and the solid-state reaction was triggered to form (Ni, Co) Cr₂O₄ spinels, which were dispersed within the outer oxide layer. The rapid formation of the Cr₂O₃ layer not only acted as an effective barrier to the inward diffusion of sulfur and oxygen, in the molten sale or flue gas, but also inhibited the outward diffusion of the metal ions to the corrosion environments.

In the presence of the atmosphere with 0.1 % SO₂, sulfur induced hot corrosion could not occur, due to the low SO₂ content. As shown in Figure 2, the corrosion rate is lower. Figure 4(a) and 4 (b) indicates that the oxide film formed was compact, dense and adherent to the substrate. The XRD patterns (Figure 3) and EDS analyses in Figure 5(a) confirmed that a Cr₂O₃ oxide film is predominant at the surface of the alloy, without any sulfides detected in the film or the film/substrate interface.

In the present of the atmosphere with 1.5 %SO₂, substantial corrosion took place. Due to the higher sulfur content, various sulfurs, including S or SO₂, diffused inward through the cracks to the film/substrate interface and a significant amount of sulfides was formed. Although, through XRD analysis the sulfides were not identified, EDS analyses (Figure 5(b)) indicted an amount of sulfides distributed at the film/substrate interface and in the innermost area. With an increase in time, the previously formed sulfides were oxidized and generated S, an amount of which can then migrate inward to form new sulfides. The sulfides were self-catalytic and re-sulfidation occurred in the substrate by the following reactions [25]:

In this way, the internal sulfidation was able to continue and then both the Cr₂O₃ film and internal sulfides were kept growing. Together with that, Cr was also migrated inward to form Cr₂O₃, leading to Cr-depletion. With the increase of Cr-depleted areas, the relative content in Co and Ni in the internal precipitation area were increased. Due to the high diffusion rate of Co in the oxide film, CoO was formed on surface of film and reacted with SO₃ to form CoSO₄, which then led to the formation of the low melting point Na₂SO₄-CoSO₄ (melting point is 565 °C) [12]. The dissolution of the cobalt oxide induces severe low temperature hot corrosion. High SO₃ content is essential to sustain hot corrosion, thus the corrosion in the atmosphere with 1.5 % SO₂ is more severe than the atmosphere with 0.1 % SO₂.

During the NaCl additions to coal ash, the alloy suffered substantial corrosion damage (Figure 6). At the presence of NaCl, the corrosion of the tested alloy was significantly accelerated. The corrosion mechanism could be described as follows [21]: In the initial exposure stage, the molten NaCl salt destroyed the protective Cr₂O₃ film very fast, following its formation (Figure 2). Together with that, NaCl reacted directly with elemental Cr through the micro-cracks that existed in the surface film. The possible reactions that took place are listed as follows [21, 26]:

An amount of Cl₂ was generated from the above reactions and was diffused inward through the cracks and voids of the oxide film and reacted with the various alloy elements, such as Cr. Then, unstable chlorides were formed:

The volatile chlorides may diffuse outward to the outer surface through the defects. At the outer surface, where the oxygen potential is high, the chlorides may be re-oxidized, according to the equation:

This mechanism is a self-sustaining process. In addition, the volatile chlorides produced from the reactions (3)-(6), will accelerate the cracking and spalling of the protective oxide films and then the corrosive medium (O, S and Cl) would rapidly diffuse inward and react with the alloying elements. Compared to the high volatility of chlorides, sulfides would remain in the oxide film and the substrate interior. As is known, sulfides have more crystal defects and higher PBR values than metal oxides. During the corrosion procedure, sulfides would segregate in the oxide film and substrate, making the corrosion products spalling off easily, and therefore, accelerating the corrosion.

Conclusions

Fireside corrosion behavior of Inconel 740H superalloy was studied at 750 °C, and the influence of SO₂ content and chlorine on fireside corrosion behavior was presented. The corrosion rate increased with the increment of SO₂ content from 0.1 % to 1.5 % by volume. In the atmosphere with 0.1 % SO₂, the alloy exhibited enhanced corrosion resistance. A higher content of SO₂ accelerated the formation of Al₂O₃, TiO₂ and sulfides beneath the Cr-rich oxide scale. NaCl additions to coal ash made the Cr₂O₃ film more porous and easier to spall off, leading to a significantly increased formation of sulfides in the alloy substrate, beneath the oxide film, accelerating the hot corrosion.