Categories

New Problems in the Oxidation and High Temperature Corrosion of Metals 1998-2002

Some radical changes have occurred in both the energy sector and automotive field over the past few years, which have impinged on studies in the high temperature corrosion.

In the USA and Europe, where natural gas is cheap, most of the new power generating plant is of the CCGT (Combined Cycle Gas Turbine) type with the newest units having turbine inlet temperatures above 1400°C. Where coal plant is favoured, this is still of the steam plant type, but peak steam pressures and temperatures are now over 250 bar and up to 600°C. Hence in both CCGT and steam plant oxidation is a key concern but it must be admitted that the main R&D focus is that of thermal fatigue. The new kid on the block, in terms of power generation, is the circulating fluidised bed boiler. This eliminates fireside corrosion, but increases the prospect of high temperature erosion-corrosion. In contrast, coal gasification and novel designs of high temperature nuclear plants, the latter using helium as a heat exchange medium, are very much at the concept stage, although corrosion work is still being funded.

Turning to the automotive field, environmental factors are pushing technology. To reduce catalyst warm up times, so reducing emissions at start up, peak temperatures will need to be around 1100°C. This implies the need for materials with very low thermal mass and as such there is great interest in thin foil metallic catalyst supports. The author has also pointed out that the elimination of lead compounds from motor fuels, and the reduction of sulphur levels should permit the formulation of new exhaust valve alloys, but so far there are no reported developments in this area.

The need for low lead fuels has had, however, an indirect affect on high temperature corrosion studies, as MBTE (C5H12O) is needed to increase gasoline octane rating to acceptable levels. MBTE is produced with the help of synthesis gas, a mixture of CO and H2. This combination of gases gives rise to metal dusting, a very aggressive form of attack.

Brief descriptions will now be given of current work in these areas. It is noteworthy that although much of the work is carried out with sophisticated techniques and is backed up with much computer simulation, so that corrosion mechanisms can be explained in much detail, the overall aim is to produce data for design purposes. Hence the work has a strong engineering science underpinning.

For a more comprehensive update on current concerns and related high temperature studies, a short set of references is given at the end of this update

CCGT Plant

As was pointed out by the author elsewhere, there has been little change in gas turbine blade temperatures over the past decade, whereas combustion path temperatures in gas turbines for CCGT plant have increased by about 400°C. Most of the improvement has come about through advanced internal cooling techniques. Latterly TBCs (Thermal Barrier Coatings) have contributed about a 70°C increase. These require the formation of a slow growing alumina scale at the interface between the TBC and the blade material or substrate as it is called in oxidation parlance.

The ability of turbine blade and coatings to withstand thermal cycling is becoming critical as many plants are now shut down overnight. As the oxide grows thicker the risks of oxide spalling increase, as do the risks of coating loss. Hence much of the theoretical work focuses on the stresses which lead to oxide spallation. The other problem with coatings is the loss of protective elements due to selective oxidation of elements such as alumina, or back diffusion of elements into the substrate. Here again computational techniques are available.

Efficient Steam Plant

One of the biggest developments over the last twenty years has been the introduction of the so-called P91/P92 steels for superheater and reheater construction. This class of materials is helping to make USC (Ultra Supercritical Steam ) plant possible.

These alloys are based on 9 Cr martensitic steels containing percentage levels of Mo and W to slow diffusion rates, but also containing Nb, V, Cu, C and N to induce precipitation of strengthening carbides, nitrides and copper rich phases.

Unfortunately the levels of chromium and silicon are on the borderline for good steam side corrosion resistance at temperatures much above 550°C. Clearly there will be loss of material with time, leading to premature failure due to creep. If excessive quantities of oxide were to spall off it could lead to blockage of superheater tubing and erosion of steam turbine blading.

More serious is the effect on superheater and reheater metal temperatures. The formation of oxide on the steam side surface has an insulating effect, preventing the heat getting through the tubing. Metal temperatures will rise, gradually increasing with time, as the oxide thickens, so that creep life will be reduced. However at high rates of heat transfer, the increase in temperature will have a noticeable affect on oxide growth so that creep life under these conditions will be seriously undermined.

Here the main thrust of the work is to assess the effect of minor changes to composition on oxidation rate. So far it appears that Si, Mn, Y and Ti may beneficial.

Automotive Catalytic Converters

The major concern is to reduce the rate of oxide growth of protective alumina scales. The growth of these will lead to the gradual loss of aluminium from the foil material, which when the aluminium content falls below a critical level, will result in breakaway oxidation. Much of the work in this area originally stemmed, from the efforts, by the author, to build very high temperature heat exchangers for advanced energy conversion systems. Increasingly sophisticated algorithms have since been produced to model breakaway corrosion, taking into account the drop in aluminium levels due to spalling of the scale, and regrowth, or from over rapid scale growth without spalling.

It is a moot point, however, whether this modelling work is truly applicable to thin foils. Of more fundamental importance is the prospect of changes in size and shape of the foils due to the stresses produced during oxide growth, and the differential stresses between the oxide and foil induced by heating and cooling. Here it should be noted that at temperatures above 800°C, the foil material, being a simple Fe-Cr-Al ferritic alloy, is extremely weak. Studies are in progress to model the stress effects.

Metal Dusting

As noted above much of the present work on metal dusting is directed to forming an understanding of the phenomena during the production of synthesis gas. Earlier work, within and sponsored by the UK Gas Industry focused on attack in hydrocarbon rich atmospheres. Metal dusting results in the formation of large volumes of material consisting of carbon, metallic particles and metal carbides. Attack in extreme cases is at a rate of millimetres per day.

The mechanism of metal dusting is not well understood. One view is that it is due to the initial formation of iron carbide, which later becomes unstable, degenerating into a mixture of carbon and metal rich particles. The latter may induce the growth of filamentous carbon, which can assist in the break up of the metal surface, as well as causing processing difficulties. Hence a number of different, interacting, mechanisms are at work.

Many of the studies are aimed at predicting the onset of metal dusting and how it can be mitigated by gaseous additions to the synthesis gas environment, through increases in steam or carbon dioxide. These will have an effect on the carbon activity, since it is known that in synthesis gas, an activity of between 3 and 10 will result in serious metal dusting. The addition of ppm levels of H2S will also reduce the amount of wastage by metal dusting. There are efforts to predict the levels required, as the presence of sulphur can interfere with subsequent chemical processing. It seems clear that the sulphur is acting in an anti-catalytic mode, since the calculations involve surface rather than bulk thermodynamics.

References 

Special Issue of Oxidation of Metals in Honor of Prof David.L. Douglas on the Occasion of His Retirement Vol 44 (1and 2) August 1995

High Temperature Corrosion and Protection of Materials 5 Ed R. Streiff, IG. Wright, RC. Krutenat, M. Caillet and A. Gallerie, Trans Tech Publications 2001 (Generally referred to as Les Embiez 5 from the Conference held there)

Lifetime Modelling of High Temperature Corrosion Processes, ed M. Schuetze, WJ. Quadakkers and JR. Nicholls, Maney Publishing 2001

Materials for High Temperature Power Generation and Process Plant Applications, ed A. Strang IOM Communications 2001

Parsons 2000: Advanced Materials for the 21st Century Turbines and Power Plant, ed A. Strang, WM. Banks, RD. Conroy, GM. McColvin, JC. Neal and S. Simpson IOM Communications 2000