Abstract

This study examines the phase stability of perovskite SrTiO3 in Na2SO4 + 50 wt.% V2O5 and Na2SO4 + 10 wt.% NaCl environments at 900°C. Hot corrosion results show the formation of Sr2VO4, SrV2O6, and SrTiV5O11 phases in Na2SO4 + 50 wt.% V2O5 environment and Sr3Ti2O7, Na4TiO4, and TiO2 phases in Na2SO4 + 10 wt.% NaCl environment. Morphological observations revealed the austerity of hot corrosion attack on SrTiO3. The Sr2+ ions leached out from SrTiO3 and reacted with corrosive environments. These observations clearly indicate the destabilization of SrTiO3 in both environments.

1. Introduction

Thermal barrier coatings (TBCs) are generally applied as thermal insulating layers on the hot metallic surfaces such as combustion engine parts, gas turbine blades, and aeroengine parts. TBCs can enhance the operating temperature and reduce the cooling requirements ensuing higher engine efficiency and part life of an engine [15]. In order to gratify the requirements of TBC materials, the properties such as low thermal conductivity, higher melting point, and matching thermal expansion coefficient with bond coat are needed. Another predominant factor for determining the durability of TBCs is the stability against hot corrosion. Hot corrosion occurs when the engine runs with low quality fuel which contains sodium, vanadium, and sulfur as impurities in it and also in chloride (marine) environments. The sodium and vanadium salts are having lower melting point (700–880°C) than gas turbine operating temperature (>1300°C). Hence, their salts can easily condensate in the form of Na2SO4, V2O5, and NaCl on the top coat of TBCs and undergo hot corrosion in two types (Types I and II). Type I generally occurs above the melting point of sodium sulphate (Tm 884°C) with Type II vice versa. In both the types, these salts react with ceramic top coat and show the way for destabilization and degradation of TBCs.

In recent years, 6–8 wt.% yttria stabilized zirconia (YSZ) has been widely used as a top coat material for TBC and much research has been involved in understanding the hot corrosion behavior of YSZ in corrosive environments. The authors reported that these molten salts could cause several microcracks and spallation of the coating [69]. Particularly, in vanadium containing environments, YSZ is found to be unstable and forms YVO4 and m’-ZrO2 (destabilization). Destabilization causes the volume change and leads to spallation of the coating.

Therefore, there is a need to develop alternate materials to enhance the hot corrosion stability of the coating. The candidate materials, such as garnets (Y3Al5O12), La, Sm, Nd, Yb, and Gd doped ZrO2, YSZ-Ta2O5, monazite (LaPO4), and lanthanum aluminates (LaMgAl11O19) were developed in recent years and their stability was examined [1014]. Perovskite (ABO3) based ceramics (SrZrO3, SrCeO3, and BaZrO3) have also drawn much attention towards TBCs due to their low thermal conductivity, higher melting point, and matching thermal expansion coefficient with bond coat material. Particularly, the SrTiO3 has higher melting point of 2080°C, thermal expansion coefficient 10−5 K−1 (1727°C), and thermal conductivity in the range of 11 to 2.5 W·m−1 K−1 (25°C to 1100°C). But the review of literature showed that the published work on SrTiO3 is mostly devoted to understanding the thermophysical properties [1520]. However, the literature on hot corrosion behavior of SrTiO3 is sparse and reactions between SrTiO3 and molten melts remain unclear. Therefore, further studies are needed to understand the hot corrosion behavior of SrTiO3.

Hence, the present work is focused on studying the phase stability of perovskite type SrTiO3 in Na2SO4 + 50 wt.% V2O5 and Na2SO4 + 10 wt.% NaCl environments at 900°C.

2. Materials and Methods

2.1. Sample Preparation

Strontium titanate powders were synthesized via hydrothermal route [21] using strontium hydroxide (99%, Alfa Aesar) and titanium tetrachloride (99.9%, Fisher Scientific) as precursor materials. To make the pellets, obtained strontium titanate powders were pressed with 370 MPa pressure using a uniaxial die (1.3 cm internal diameter). Polyvinyl alcohol (PVA) was used as a binder (1 : 10 ratio) in the preparation of strontium titanate pellets. Compacted pellets were kept in a furnace at 500°C for 48 h to remove the binder. Then these pellets were heat treated at 900°C for 48 h.

2.2. Hot Corrosion Tests

A mixture of Na2SO4 and V2O5 salts in 1 : 1 weight ratio and another mixture of Na2SO4 and NaCl salts in 9 : 1 weight ratio were prepared and dissolved separately in distilled water for getting 50 wt.% of each salt solution. Samples were preheated at 200°C for 30 min to achieve good adhesion of salts. Salt solution was applied on all sides of surface using camel hair brush. Salt coated samples were kept in a furnace and heated at 120°C for 2 h to remove the moisture. The samples were weighed to determine the salt coverage and found in the range of 20–30 mg/cm2. Then these salt coated samples were kept in an alumina boat and heated at 900°C for 100 h in air. After hot corrosion, samples were taken out and washed with running water and dried.

2.3. X-Ray Diffraction Analysis

Before and after hot corrosion, the nature of phases was investigated by X-ray diffractometer (Bruker, D8 advance) with Cu Kα radiation with a scan rate of 6°/min. All the phases were identified using the JCPDS-PCPDFWIN software package.

2.4. Morphological Analysis

The surface morphologies and elemental analysis of samples were examined by Field Emission Scanning Electron Microscope (FESEM-ZEISS-SUPRA 55VP) attached with EDX (OXFORD instruments).

3. Results

3.1. X-Ray Diffraction Analysis

Figure 1 shows the X-ray diffraction patterns of the SrTiO3 sample (A) before hot corrosion and after subjecting to hot corrosion at 900°C in (B) Na2SO4 + 50 wt.% V2O5 environment and (C) Na2SO4 + 10 wt.% NaCl environment. Before hot corrosion, it is observed that SrTiO3 is in cubic phase (JCPDS-79-0176). After exposure to hot corrosion in Na2SO4 + 50 wt.% V2O5 environment (Figure 1(B)), additional phases such as Sr2VO4, SrV2O6, and SrTiV5O11 (JCPDS-81-0854, 28-1267, and 48-0540, resp.) were identified. After exposure to hot corrosion in Na2SO4 + 10 wt.% NaCl environment (Figure 1(C)), additional phases such as Sr3Ti2O7 (JCPDS-78-2479), Na4TiO4 (JCPDS-42-0513), and TiO2 were identified. It is worthwhile to mention that SrTiO3 phase is found to be unstable in both Na2SO4 + 50 wt.% V2O5 and Na2SO4 + 10 wt.% NaCl environments at 900°C. Succinctly, this indicates destabilization of SrTiO3.

3.2. Morphological and Microchemical Analysis

Surface morphology of SrTiO3 sample before and after exposure to hot corrosion in both Na2SO4 + 50 wt.% V2O5 and Na2SO4 + 10 wt.% NaCl environments at 900°C was examined to identify the severity of hot corrosion attack. Figure 2 shows the macroscopic view of the SrTiO3 before and after exposure to hot corrosion. Figure 2(a) (before hot corrosion) reveals that surface is free from cracks and scales. Nonuniform and severe hot corrosion attack was noticed on the surfaces which were subjected to hot corrosion in both Na2SO4 + 50 wt.% V2O5 (Figure 2(b)) and Na2SO4 + 10 wt.% NaCl environments (Figure 2(c)) at 900°C.

Figure 3 shows the surface morphology of SrTiO3 before subjecting to hot corrosion. Figures 3(a) and 3(b) show the lower and higher magnification FESEM images of SrTiO3. Figure 3(b) shows the spherical shape morphology. EDX analysis was performed to identify the chemical compositions at region 1 (Figure 3(b)) and the data is presented in Table 1. Region 1 is composed of Sr, Ti, and O (corresponding to SrTiO3). These results are in good agreement with the XRD results (Figure 1(A)). The surface morphology of SrTiO3 after subjecting to hot corrosion in Na2SO4 + 50 wt.% V2O5 environment at 900°C reveals the octahedron shape morphology (Figure 4(a)). Higher magnification image of Figure 4(a) is shown in Figure 4(b) and EDX analysis was done at region 1 (data is presented in Table 1). Region 1 is composed of Sr, Ti, and O (corresponding to SrTiO3). Spherical type of morphology of strontium titanate changes onto octahedral shape when treated in sulphate solution of venadate at 900°C. Figure 4(c) shows irregular shape morphology. EDX point analysis was carried out at region 1 (data is presented in Table 1) and this location is composed of Sr, V, and O (corresponding to strontium vanadates). Figure 5 shows the severity of hot corrosion on SrTiO3 after exposure to Na2SO4 + 50 wt.% V2O5 environment and it shows the porous layer and corrosion products on the surface of SrTiO3.

The surface morphology of SrTiO3 after exposure to Na2SO4 + 10 wt.% NaCl environments at 900°C is shown in Figure 6. Lower magnification image (Figure 6(a)) reveals that nonuniform, severe hot corrosion and groove attack were observed in Na2SO4 + 10 wt.% NaCl environment. Higher magnification image (Figure 6(b)) reveals the plate type of morphology. EDX analysis was carried out at region 1 (data is presented in Table 1) in Figure 6(b) and this location is composed of Sr, Ti, and O (corresponding to Sr3Ti2O7). Similar type of morphology for Sr3Ti2O7 has also been observed by previous author [22].

4. Discussion

The phase stability of SrTiO3 against hot corrosion is being studied in Na2SO4 + 50 wt.% V2O5 and Na2SO4 + 10 wt.% NaCl environments at 900°C. Though SrTiO3 is an oxide, it underwent destabilization in both environments. Massive hot corrosion attack is being observed on surface of SrTiO3 after being subjected to both environments at 900°C. Significant literature is available for SrO-V2O5 binary systems but not for SrO-NaVO3. It is reasonable to predict the chemical reaction between SrTiO3 and NaVO3 using SrO-V2O5 [23, 24] and SrO-VO2-V2O5 phase diagrams [25]. First, SrTiO3 decomposed into SrO and TiO2 phases by leaching of Sr2+ ions from SrTiO3 due to the presence of corrosive environments (see (1)). When the temperature reaches 630°C in furnace, NaVO3 forms and acts as good solvent medium for oxides. At this stage, the reactions between sodium vanadate and oxides are faster.

Habibi et al. [14] have also reported that the liquid NaVO3 will increase the atom mobility of oxide elements. SrO reacts with NaVO3 liquid melt and forms SrV2O6 (see (2)). Again Na2O reacts separately with V2O5 and SO3 and forms NaVO3 and Na2SO4, respectively. This reaction is cyclic. SrO and TiO2 can also react with NaVO3 and form SrTiV5O11. At 900°C, SrO reacts with NaVO3 and forms the Sr2VO4 phase (see (3)). This phase is identified as predominant corrosion product phase in XRD (Figure 1(B)). The above sequence of reactions describes the formation mechanism of corrosion products. Higher magnification image of SrTiO3 (Na2SO4 + 50 wt.% V2O5 at 900°C) (Figure 5) shows the severe hot corrosion and corrosion products on it. Due to damage caused by Na2SO4 + 50 wt.% V2O5, a porous layer and corrosion products were formed and observed in the Figure 5. This reveals the severity of hot corrosion on SrTiO3 in Na2SO4 + 50 wt.% V2O5 environment.

The Na2SO4 + 10 wt.% NaCl environment also caused the leaching Sr2+ ions and subsequently formation of SrO and reacts with chloride to form SrCl2 (see (4)). In general thermodynamic stability of oxides is higher than the chlorides. Hence formed chlorides are further transformed into oxides (see (5)). Further they reacted with themselves and formed Sr3Ti2O7 at 900°C (see (6)). Tilley [26] has studied the Sr-Ti-O system at 1100–1473°C and reported the formation of three phases such as Sr2TiO4, Sr3Ti2O7, and SrTiO3. Jacob and Rajitha [27] have studied the pseudo-binary system (SrO + TiO2) at 775 to 977°C and reported the thermodynamics properties of Sr2TiO4, Sr3Ti2O7, Sr4Ti3O10, and SrTiO3 using solid state galvanic cells (see (6)). Based on the previous works [22, 27] the reaction between SrO and TiO2 can be expressed as follows:

The TiO2 phase can react with Na2O to form Na4TiO4 (see (7)). There has been no previous work on hot corrosion behavior of SrTiO3 in NaCl medium. The possible role of Na2O against SrO and TiO2 phases is discussed here. According to Coulomb’s law (lattice energy depends on the ionic radii and charges of the ions) the compounds with +2 ions have more lattice energy than the compounds with +1 ion. In addition to this, Ti atoms are having smaller ionic radius (94 pm) than that of Sr atoms (112 pm) [28]. Hence, the Ti-O structured atoms would be expected to have larger lattice energy which gives higher attractive force towards Na2O. Based on our above discussions the order of lattice energy can be expressed as

The lattice energy of TiO2 (12150 kJ/mole) is more than that of SrO (3223 kJ/mole) [29]. As per theory, the lattice energy is directly proportional to the phase stability of the compound, which indicates the less stability for SrO at 900°C. But, in the case of TiO2, it needs higher energy to separate the Ti2+ and O2− ions. As we discussed earlier, the strong electrical attraction between the Na+ (Na2O) and TiO2 ions might cause the formation of Na4TiO4 phase (see (7)). These discussions are well in agreement with XRD (Figure 1(C)) results. This is the possible reason why the reaction between TiO2 and NaCl was highly favored compared with that with SrO. According to the binary phase diagrams of Na2O-TiO2 [30, 31], Na4TiO4 melts at 850°C and this was the reason for Na4TiO4 being identified as minor corrosion product compared to that of Sr3Ti2O7 in XRD (Figure 1(C)).

5. Conclusions

The hot corrosion behavior of perovskite type SrTiO3 was studied in Na2SO4 + 50 wt.% V2O5 and Na2SO4 + 10 wt.% NaCl environments at 900°C for 100 h.(1)SrTiO3 has been found to be unstable in both Na2SO4 + 50 wt.% V2O5 and Na2SO4 + 10 wt.% NaCl environments at 900°C.(2)The Sr2+ ion leaches from its own phase and reacts with NaVO3 to form Sr2VO4, SrV2O6 and SrTiV5O11 in Na2SO4 + 50 wt.% V2O5 environment.(3)The Na2SO4 + 10 wt.% NaCl environment also caused the leaching of Sr2+ ions. The SrO reacted with TiO2 and formed Sr3Ti2O7. The reaction between Na2O and TiO2 caused the formation of Na4TiO4.(4)This study clearly indicates the destabilization of SrTiO3 in both mediums.

Conflicts of Interest

The authors declare that they have no conflicts of interest.

Acknowledgments

The author Dr. G. Sreedhar would like to acknowledge the DST, India, for sponsoring Project SB/EMEQ-036/2014 (GAP 09/15).