Loughborough University
Browse
1/1
2 files

The corrosion behavior of Fe-Cr-Ni alloys in complex high temperature gaseous atmospheres containing the reactants oxygen, sulphur and carbon

thesis
posted on 2014-05-16, 11:35 authored by Jonathan. A. Kneeshaw
A systematic in-depth study has been undertaken to establish the corrosion mechanism of a Model 25Cr-35Ni-Fe alloy and four commercial alloys HP40Nb, AISI314, HP40Al and Alloy 800H in low oxygen, high sulphur and carbon containing environments typically found in coal gasification and fluidised bed combustion processes. A review of present knowledge of corrosion processes in purely oxidizing, sulphidizing and carburizing environments and multiple reactant carburizing/ oxidizing, carburizing/sulphizing and oxidizing/sulphidizing environments is given. The experimental programme was designed to establish the role of sulphur on the corrosion process by studying corrosion mechanisms in a sulphurfree H2-7À-1.5%H2o gas, a low sulphur H2-7À-1.5%H20-0.2%H 2 S gas (pS2_8= 10 bar), and a high sulphur H 2 -7À-1.5%H 2 0-0.6%H 2 S gas (pS = lO bar) at 800'C. All_21j_hree environments had a constant partiaf pressure of oxygen (po2 = 10 bar) and carbon activity (ac = 0.3). In the sulphur-free gas the Model alloy formed a thin uniform cr 2 o 3 layer which grew at a constant parabolic rate throughout the exposure period of 0 - 5000 hours. Surface working increased the growth rate and thickness of the Cr 2 o 3 layer but created a large number of cracks and pores which allowed carbon containing gaseous species to diffuse through the oxide to form carbide precipitates in the alloy substrata. Alloying additions of Si promoted the formation of an inner SiO layer which reduced the corrosion rate by cutting off the outward diffusion of Cr, Mn and Fe. Alloying additions of Mn promoted the formation of an additional outer (Mn, Fe )Cr 2o 4 layer. The 3. 5% Al content of the HP40Al was insufficient to form a complete Al 2 o3 layer. Alloy 800H was susceptible to localised internal oxidation. Adding a low level of sulphur (0.2% H 2 S) to the gas increased the corrosion rate of the Model alloy in the 1nitial stages. This rate gradually slowed down before becoming parabolic after 1000 - 2000 hours. This was due to the nucleation of sulphides in addition to oxides. The oxides and sulphides grew side by side until the oxides overgrew the sulphides to form a complete Cr 2o3 layer which cut off further ingress of sulphur from the gas. The entrapped sulphides promoted localized thickening of the oxide layer. Eventually the sulphur redistributed from the sulphides in the scale to internal sulphide precipitates in the alloy with the corrosion rate returning to that of the sulphur-fre,e gas for the rest of the exposure period (5000 hours total). In the commercial alloys the internal sulphide precipitates prevented the inner Si02 layer becoming complete. Sulphur doped the (Mn, Fe) Cr 2 0 4 outer layer ana the intermediate Cr 2o3 layer formed from the spinal layer, increasing the number of cation . vacancies and the growth rate of the scale. These factors caused a massive Cr depletion of the alloy substrata after several thousand hours. The internal carbides became unstable which led to a massive amount of internal attack and a dramatic increase (breakaway) in the corrosion rate. Due to its thickness and the presence of Si02 inner layer the external scale became susceptible to spallation. If this occurred the oxides and sulphides nucleated on the alloy surface again but sulphides. protective alloy. insufficient Cr was available for the oxides to overgrow the The sulphides therefore grew to form a fast growing nonsulphide scale which soon led to catastrophic failure of the Increasing the level of sulphur in the gas to 0.6% H2S caused oxides and sulphides to nucleate on the surface, but in this case the sulphides overgrew the oxides to form thick fast growing non-protective sulphide scales on all the alloys.

History

School

  • Aeronautical, Automotive, Chemical and Materials Engineering

Department

  • Materials

Publisher

© J. A. Kneeshaw

Publication date

1987

Notes

A Doctoral Thesis. Submitted in partial fulfilment of the requirements for the award of Doctor of Philosophy of Loughborough University.

EThOS Persistent ID

uk.bl.ethos.329608

Language

  • en