ROLE OF INHIBITORS ON HOT CORROSION OF SUPERALLOYS IN Na2SarV205 ENVIRONMENT

Abstract

Hot corrosion is an acute form of corrosion occurring at elevated temperature in the presence of an oxidizing gas and is associated with a thin electrolytic deposit (salt or ash) on alloy. The greatest corrosion is always associated with an ash composition corresponding to the sodium vanadyl vanadate (Na20.V204.5V205). Addition of Na20 to liquid V205 causes an increase in the basicity of the melt with a corresponding increase in corrosivity with respect to the acidic oxides. Addition of Na20 has also been shown to decrease the viscosity; the protective oxides become porous and non-adherent. The formation of binary and tertiary low melting eutectics increases the surface attack thereby reducing the useful life of the component. For trouble free and continuous long service of power plants running at high temperatures, suitable materials should be selected after complete hot corrosion studies of alloys. Superalloys are used to resist oil ash corrosion and success is achieved to a considerable extent. There appears to be an advantage in using high chromium alloys. Thermal barrier coatings of alloys and ceramics have been tried to counter oil ash corrosion and they were found to be effective to some extent. Inhibitors and fuel additives have also been used with varying success to combat oil ash corrosion. In the present investigation hot corrosion studies were carried out on Fe-base, Co-base and Ni-base superalloys procured from MIDHANI (Mishra Dhatu Nigham Ltd), Hyderabad (India) in Na2SO4-60% V205 with and without inhibitors. Effect of oxide additives on the hot corrosion behaviour of superalloys Superfer 800H (alloy A), Superco 605 (alloy B), Supemi 75 (alloy C), Supemi 718 (alloy D) & Superni 601 (alloy E) have been studied in Na2SO4-50%V205 under cyclic conditions. The details of these alloys are given in Table 4.1 and those of the inhibitors employed are given in Table 4.2. MgO, CaO, Mn02 and ZnSO4 were mixed with Na2SO4-60%V205 and applied on the polished preheated samples whereas in case of Ce02, SnO2, Y203 and Zr02 first a coating of oxides was iv applied; they were then heated to 900°C for 8 hours to obtain the better adhesion of the oxide film. After 8 hours heating, a coating of Na2SO4-60%V205 was applied and then they were exposed to cyclic heating at 900°C for 50 cycles (each cycle consisted of 1 hour heating and 20 minutes cooling). Weight change measurement after each cycle was noted upto 0.1 mg accuracy by electronic balance. Macrographs were taken to know the nature of corrosion product so formed. SEM, XRD and EPMA have been done for all the samples. EDAX of surface of the scale was done in few cases and in others cross-section have been analysed. Hot corrosion products observed in the scale of these corroded alloys are mostly oxides of the elements present in the alloy. The presence of continuous Cr203 layer at the substrate /scale interface is predominantly revealed in most of the alloys where weight gain and scale thickness has been very less. Spinels of composition MCr204 have formed due to reaction of oxides of Fe, Ni and Co with Cr203 as XRD indicated the formation of such phases e.g. FeCr204, NiCr204 and CoCr204 in Fe-, Ni- and Co-base alloys respectively. The weight gain values in Na2SO4-60%V205 environment have been found to be highest for Co-base alloys and lowest for Ni-base alloy Supemi 75. The general trend followed is: Superco 605 (alloy B) > Superfer 800H (alloy A) > Supemi 718 (alloy D)> Supemi 601(alloy E)> Supemi 75 (alloy C). Scale thickness values measured after experimentation follow the sequence as given below: Alloy B > Alloy A > Alloy D > Alloy E > Alloy C After mixing Mg0 in Na2SO4-60%V205 (Mg0N205 ratio 3:1), the aggressiveness of the Na2SO4-60%V205 environment has been found to decrease in case of all the alloys except alloy D where the decrease is marginal. This may be attributed to the formation of a solid phase magnesium orthovanadate (Mg3V20s, m.p. 1190°C), by reaction of V205 with Mg0 in preference to formation of liquid NaV03 and formation of above mentioned solid phase might have contributed to the decreased corrosion rate. X-ray diffraction analysis has identified the presence of Mg3V20g phase in the scale developed on the surface of the alloys A, C and E which has also been further confirmed from the EPMA micrographs as Mg, V and oxygen are co-existing in the scale. Similar effect has been observed by adding of 20wt.% CaO to Na2SO4-60(10V205 on the alloys A, B, C, D & E. Inhibiting behaviour of CaO may be owing to the formation of a protective uniform chromium enriched band present and also the formation of calcium vanadate, Ca3V2O8 that was identified by XRD and EPMA. Effect in case of alloy C is distinct as a Cr enriched continuous band type scale is formed. 20wt% Mn02 additions to Na2SO4-60% V205, has also been found to be beneficial in almost all the alloys. The extent of reduction in weight gain is more in alloy D & follows the trend as: Alloy D > Alloy C> Alloy A> Alloy B> Alloy E. A thick scale is indicated in case of alloys B & E and a thin irregular scale in alloy A & C whereas a continuous band type regular uniform scale is indicated in alloy D. Internal oxidation is indicated in alloys A, C, D & E. In case of alloy C & D, Cr is present throughout the scale where as in alloy D, it is present as a rich continuous band at the interface which is probably helping in reducing the rate of corrosion. Mn & V are coexisting in the scales of alloys B,C,D&E. With addition of 10% ZnSO4 in Na2SO4-60% V205, the scale formed is relatively very thick in case of alloys A and E. Due to severe sputtering experimentation had to be terminated after 8 cycles in case of alloy B and after 30 cycles in alloys A, C, D and E. ZnSO4 is only marginally effective in all the alloys. In case of alloys A, C and D, very thin Cr2O3 layer has formed just above the substrate but internal oxidation is indicated. In case of CeO2 coated samples, a thick scale is indicated in alloy A where as it is of medium size in alloys B, C, D & E. But there is indication of overall reduction in the rate of corrosion for all the alloys. Formation of CeVO4 is proposed on the basis of EPMA and EDAX analysis in case of alloys C & E. EDAX analysis of the corroded product of alloy A scanned along cross-section mainly indicated Cr, Ni and Fe and showed the presence of Ceria in small amounts near the substrate. EDAX of surface of the same scale confirmed the presence of Ce and vi V in the scale. In case of alloy C, ceria is present in high concentration in the scale and is coexisting with V and also there is growth of Ni-rich crystals in Ce-enriched layer. Overall wt. change is very less as compared to that without Ce02. But the scale of alloy D shows presence of Ce in pockets at few spots. Presence of Ti pockets near scale/substrate interface is indicated. Just above the interface, there is a thin layer rich in Cr where both Ni and Fe are absent. Presence of unreacted salt is also indicated. Very little change in weight was observed in case of alloy E. Superficially applied Y203 seems to be most effective in combating aggressive environment of Na2SO4-60% V205 in case of alloy A as scale thickness was only 20)t. This scale is rich in Cr and is continuous. Alloy B has shown the formation of medium size scale rich in Cr and Co. Scale thickness is nearly 50% of that without Y203 barrier coating. In case of alloy C again a very thin scale is observed about 16.51.un in size and contains mainly Cr and Ni. Y is present as a thin layer along with V & S. A medium size scale has been formed in alloy D that is rich in Ni, Cr and Fe. The scale thickness values in case of superficially applied Y203 follow the trend as: Alloy C< Alloy D < Alloy A < Alloy B and Alloy E In case of Sn02 coated samples a thick scale is formed in alloys A, B & E whereas medium size scale in alloys C & D. There is a thick continuous Sn rich layer present at the top of the scale formed in alloys A, C & E indicating presence of unreacted Sn02. V is indicated on the top of the scale and even above the thick tin layer which gives the possibility of the presence of some unreacted salt. Macrographs of the same samples indicate presence of unreacted salt on the surface, which has spalled off at many places. Percentage decrease in weight gain in alloys A, B, C, D and E with superficially applied Sn02 is 17.2, 66, 50, 17 and 32 percent respectively after 30 cycles as compared to without inhibitor. The effectiveness of the Sn02 is perhaps due to its non-reactive nature with the corroding species and high melting point (1630°C). With Zr02 superficial coating also, overall weight gain is less in case of alloys B, C, D & E. Zr02 is marginally effective in alloy A & B. A thick scale is observed in alloy A, rich in Cr, Ni, vii Fe & V. Absence of protective continuous chromia layer and presence of less protective Ni0 is the main reason for more corrosion rate in alloy A. Similar can be the reason for thicker multilayered scale formed in alloy B where innermost layer is again Ni-rich. Presence of Cr in very high concentration just above the interface indicated by EPMA and EDAX and presence of Zr in the scale in the top and middle area detected by EDAX might have helped in providing some inhibiting effect. Scale of alloy C is very thin mainly consisting of Cr and Fe along with V, S & Zr in small percentages and weight gain is also reduced considerably. Scale of alloy D is slightly thicker and consists of mainly Cr with small amounts of nickel. Zr is present in the top most layer indicating presence of unreacted Zr02 on the surface which is even visible from the macrographs. ZrV202 might have formed and reduced the corrosion rate as Zr, V and oxygen are present in the scale at same areas as observed from the EDAX/EPMA. Chromium rich layer is present in both the alloys, which may perhaps be contributing to reduced reaction rates. But a single layer scale has been observed in case of alloy E which is rich in Cr, Fe and V. Various spinals and iron vanadate have been identified. Internal oxidation of Cr and Al is indicated by EPMA. The role of all the inhibitors used in this study superficially applied or mixed with the aggressive environment is beneficial in decreasing the extent of corrosion attack under the given aggressive environment. The beneficial effect in most of the cases is found to be mainly due to formation of vanadates which could be solid at the given reaction temperature. The observance of a continuous band of Cr203 mostly along the interface between the scale and the substrate may be the other reason for providing protection to the alloy. In case of superalloys where the relatively higher extent of corrosion have been indicated, formation of vanadates and presence of exclusive continuous Cr203 band is not generally observed especially in case of alloy D in MgO, alloy B in Mn02, alloy A in ZnSO4, alloy E in Sn02 and Alloy A in Zr02 whereas extent of reaction is still lower than that with inhibitor. From this it can be inferred that the formation of Cr2O3 rich band as well as solid vanadate gives the maximum protection to the alloys. In most of the cases this Cr203 layer has formed by diffusion of Cr from the substrate that is clearly marked by the depletion of Cr from the substrate.