Materials at High Temperature Vol 20, Issue 2, 2003

Subscribers may view full papers here

The effect of deposits on waterwall corrosion in fossil fueled boilers

Water T. Bakker

EPRI, 3412 Hillview Avenue, Palo Alto, CA 94304, USA

Corrosion of water walls in fossil fueled boilers and gasifiers has traditionally been considered the result of gaseous corrodants, such as H2S and HC1, reacting with the heat exchanger tube surfaces. Under reducing conditions these corrodants prevent the formation of a protective oxide scale, leading to increased metal loss. Recent field experience in boilers, using staged combustion systems, has shown much greater corrosion rates than predicted by simple gas/solid corrosion processes. The presence of large quantities of unoxidized iron sulfide in deposits in areas where high corrosion rates were found, suggests that deposits play a role as well. Subsequent laboratory corrosion studies found that the presence of FeS can indeed lead to very high corrosion rates, but only under oxidizing conditions. Since FeS usually deposits only where reducing conditions are present, the accelerated corrosion observed requires alternating reducing and oxidizing conditions. These are usually found in load following boilers. It is also shown that chlorine corrosion may be caused or at least accelerated by chloride containing deposits in fossil fueled boilers, similar to waste incinerators, where the role of chloride rich deposits has been well established. Due to the relatively high sulfur content of fossil fuels, chloride deposits most likely form under reducing conditions only. However, once formed they can be highly corrosive under oxidizing conditions.

Keywords: corrosion of water walls, fossil fueled boilers

Materials performance in waste to energy plants

J.L.Blough and G.J.Stanko

Foster Wheeler Development Corporation, New Jersey 07039, USA

The concept of phasing out of landfills has made mass burning of refuse an attractive alternative. Even more popular is the advent of the “garbage in, energy out” philosophy. This has led to the development of boilers that convert the heat generated from burning refuse into revenue-producing process steam and electricity. Because refuse is non homogeneous in nature, the furnace zone must accommodate the combustion of a particularly aggressive and complex fuel. It must be designed to withstand the potential damaging effects of corrosion and erosion caused by the ash and flue gas generated as well as the methods used to remove the ash once it has deposited on the tube surfaces. The early mass-burning units, built with carbon and low-alloy steel waterwalls and convective passes, have experienced substantial wastage rates from corrosion and erosion. Parameters affecting the operation of the unit can be changed to minimize tube wastage, but at the expense of boiler efficiency. Upgrading the materials to more corrosion-resistant alloys, however, should not affect efficiency and should prove to be an economical solution to the problem.

Keywords: waste to energy plant, refuse, corrosion, erosion

Probes to monitor in-plant material degradation

M.P. de Jong and R.G.I. Leferink

KEMA Power Generation & Sustainables (KPS), Utrechtseweg 310, 6812 AR Arnhem, Netherlands

Fireside corrosion in coal fired boilers has been well-investigated. The main causes of water wall fireside corrosion are: (1) impurities in the fuel, such as sulphur alkali metals and chlorine; (2) the lack of control of the combustion process resulting in a reducing gaseous environment at the tube surface; (3) flame impingement; and (4) overtemperature of tube metal. Co-firing secondary fuels in coal fired boilers is becoming common practice in many power stations in Europe. Secondary fuels like wood, refuse derived fuels, meat and bone meal, straw, poultry litter or mixtures of several secondary fuels are co-fired up to 20-wt%. Most of these biomass fuels contain high concentrations of alkali chlorides. Considering the composition of these fuels, limitations on the maximum amount of secondary fuels to be co-fired in coal fired boilers are expected. In addition to the environmental benefits from biomass fired power plants, co-firing can result in “green” power labelling and governmental subsidy. Also savings on fuel costs may be a driving force for an increase of the amount of biomass or secondary fuels to be co-fired. However, without corrosion monitoring, short-term policies concerning co-firing secondary fuels in large volumes can lead to high costs in the medium or long term. These costs can be due to corrosion damage both in the furnace and superheater sections and penalties due to unplanned outages in a highly competitive electricity market. This paper summarizes practical experiences from corrosion monitoring programs with KEMA corrosion probes. The first prototype was successfully tested in 1997 at the Hemweg Unit 8 coal fired power plant of Reliant Energy in Amsterdam, the Netherlands. Other corrosion monitoring programs were carried out at coal fired power plants and at a waste incineration plant. At present a large-scale corrosion monitoring and material testing program is in progress at the Maasvlakte power station Unit 1 near Rotterdam, the Netherlands. In this 520 MWe power plant of E.on Benelux more than 10-wt% of mixtures of secondary fuels are directly co-fired. In addition to aspects such as emissions, fuel handling and fuel cost savings, co-firing secondary fuels requires corrosion monitoring to check the tolerance to different fuel types of coal fired boilers.

Keywords: in-plant material degradation

Future perspective of energy system in Japan

Shigemitsu Kihara,

Ishikawajima-Harima Heavy Industries Co.,Ltd.(IHI), Japan

The Kyoto Protocol adopted at the Third Conference of the Parties to the United Nations Framework Convention on Climate Change in December 1997 commits Japan to reducing its greenhouse gas emissions to 6% below 1990 levels during the first commitment period (2012). The forecast of electricity demand and capacity to be installed at 2010 suggests that the demand and installed capacity will increase, but that the consumption for thermal power will be suppressed. People’s attitudes towards saving energy and improvement of efficiency of power generation need to change. Also because all fuels are imported into Japan, diversity of energy source, oil, coal, LNG, Nuclear, should be retained.

Keywords: energy systems in Japan

Operation experience and future perspectives with degradation at high steam parameters in WTE-plants

J. Krueger

Schwandorf, Germany

The present paper is concerned with some selected issues on the interactions between flue gas and saltmelts. Suggestions for improvements of materials are provided bearing in mind the development of new systems. The objective is to encourage the readers to join the MSB GmbH in discussions on the specific operation problems described in this paper. According to our experience, the knowledge of temperatures in the boiler walls and in the adjacent refractory is of high importance while considering the mechanisms of failures. Therefore this topic is discussed below.

Keywords: waste to energy plants, high steam parameters

Prediction and real-time monitoring techniques for corrosion characterisation in furnaces

Temi M. Linjewile1, James Valentine1, Kevin A. Davis1, N.S. Harding2 and William M. Cox3

1Reaction Engineering International, Country

2N.S. Harding & Associates, Country

3Corrosion Management, Country

Combustion modifications to minimise NOx emissions have led to the existence of reducing conditions in furnaces. As regulations demand lower NOx levels, it is possible (to a degree) to continue to address these requirements with increased levels of combustion air staging. However, in most practical situations, a number of adverse impacts prevent the application of deep combustion air staging. One of the more important limitations is the increased corrosion that can occur on wall tubes exposed to fuel rich combustion environments. Current boiler corrosion monitoring techniques rely on ultrasonic tube wall thickness measurements typically conducted over 12 to 24 month intervals during scheduled outages. Corrosion coupons are also sometimes used; typically require considerable exposure time to provide meaningful data. The major drawback of these methods is that corrosion information is obtained after the damage has been done. Management of boiler waterwall loss and system optimisation therefore requires a realtime indication of corrosion rate in susceptible regions of the furnace. This paper describes the results of a program of laboratory trials and field investigations and considers the use of an on-line technology in combination with innovative applications, also modelling and precision metrology to better manage waterwall loss in fossil fuelled boilers while minimising NOx emissions.

Keywords: computational fluid dynamics, NOx emissions, corrosion characteristics

Modelling the oxidation of FeCrAl-RE ODS alloys in simulated natural gas combustion environments

J. R. Nicholls, R. Newton, N. J. Simms and J. F. Norton

PGTC Cranfield University, Cranfield, Bedford MK43 0AL

To meet industrial needs of increased operating temperatures coupled with improved strength, many power plant designers have turned to the ODS family of FeCrAlRE alloys to provide the solution. Applications vary from combustion chambers, through heat exchangers to abraidable seals. In all of these applications the components have to work in mixed gas, but oxidative, environments for extended periods of operation; often at ultra high temperatures (> 1000C) and for some applications under cyclic operating conditions. This paper reviews the long-term cyclic oxidation of a family of ODS FeCrAl alloys, including commercial alloys MA956, ODM751 and PM2000, in air, both as foils and wrought product, and PM2000 foils in a simulated natural gas combustion environment, at temperatures between 950–1300°C. From these studies the underlying oxidation/nitridation reactions and times to breakaway have been evaluated. Oxidation data in air and the combustion environment are compared using a stochastic, cyclic oxidation life model developed as part of a European research programme LEAFA (The Life Extension of Alumina Forming Alloys). For the combustion environment the model has to be been modified to account for initial transitional alumina formation in this mixed gas environment. This modified model provides an accurate prediction of ODS FeCrAl-RE lives and also permits an assessment of the risk of early component failure due to breakaway oxidation in such combustion derived environments.

Keywords: FeCrAl-RE ODS alloys, natural gas combustion

Metallic interconnectors for solid oxide fuel cells – a review

W.J. Quadakkers, J. Piron-Abellan, V. Shemet and L. Singheiser

Forschungszentrum Jülich, IWV 2, 52425 Jülich, Germany

For planar solid oxide fuel cell (SOFC) designs, ceramic as well as metallic materials are being considered as construction materials for the interconnectors. Compared to the ceramics, mostly compounds on the basis of La-chromite, metallic materials have the advantage of easier fabricability, lower costs as well as higher heat and electrical conductivity. Based on the requirements in respect to oxidation resistance, low thermal expansion coefficient and electrical conductivity of surface oxide scales, Cr-based alloys and high-Cr ferritic steels seem to be the most promising metallic interconnector materials. Whereas Cr-based alloys have recently especially been developed for SOFC application, a large number of ferritic steels are commercially available in a wide range of compositions. However, it seems that the specific combination of properties required for a SOFC interconnector will necessitate the development of a new, specifically designed steel or the modification of an existing commercial steel composition.

Keywords: solid oxide fuel cells, metallic interconnectors

Measurement and compilation of materials degradation data in the COST522 programme

N. J. Simms1, S. R. J. Saunders2, S. Osgerby2 and J.E. Oakey1

1Power Generation Technology Centre, Cranfield University, Cranfield, Bedfordshire MK43 0AL, UK

2NPL Materials Centre, National Physical Laboratory, Teddington, Middlesex TW11 0LW, UK

Within the European COST522 programme, there are extensive project groupings investigating the degradation of materials, both alloys and coatings, in a wide variety of power plants. Two project groupings are addressing materials degradation in (a) gas turbines and (b) the hot gas paths/heat exchangers of boilers and gasifiers. In both of these broad topics there is a need to generate and compile quantitative information on materials degradation that is appropriate to components operating in systems using new/dirtier fuels, as well as to components operating at higher metal temperatures. The data generated from research into both of the topics are being compiled into databases. For the boiler/gasifier heat exchanger fireside, the database contains corrosion damage information generated from: (a) a range of plant and pilot plant operations (using biomass, waste and coal fuels), (b) laboratory tests (targeted at particular environments to investigate different specific degradation effects in more detail), and (c) data gathered from the open literature and previous test programmes. For the gas turbine database, data of the performance of base alloys and especially coatings are being included from burner rig and laboratory tests of isothermal oxidation, thermal cycling, hot corrosion performance (as well as data on erosion-oxidation, mechanical and thermo-physical properties). Both databases contain information that precisely specifies the material (coating and substrate, if applicable), as well as accurately describing each test environment and the data produced from each test. The databases are intended to facilitate the comparison of the performance of candidate materials under specific conditions and to enable some of their limitations to be identified, in terms of metal temperature and sensitivity to particular corrosive conditions. The most valuable corrosion data for inclusion in these databases are those reported in terms of metal loss (or metal loss distribution) and with a well-characterised exposure environment. For the data to be readily incorporated into the databases, it has been necessary to develop and apply standardised methods of data collection, computerised compilation and presentation. The preferred method of gathering corrosion damage data has been by dimensional metrology before and after exposure to obtain a distribution of damage measurements. The structured format of the data in the databases can be used to produce models of materials performance as a function of environmental exposure parameters. This is being investigated for the heat exchanger materials database using both neural network and more conventional empirical modelling.

Keywords: degradation of materials, alloys, coatings, COST522 programme

Fundamental aspects of chlorine induced corrosion in power plants

M. Spiegel, A.Zahs and H.J. Grabke

Max-Planck-Institut für Eisenforschung, 40237 Düsseldorf, Germany

Chlorine induced corrosion is the most important corrosion process in power plants, firing waste, coal and/or biomass. Depending on the process, chlorine is present as HCl gas, as solid KCl and NaCl in ashes or as eutectic melts i.e. KCl–ZnCl2 in deposits of waste fired boilers. In the presence of HCl gas and solid chlorides, ‘active oxidation’ is generally accepted as the major corrosion mechanism. Metal chlorides are formed by inward diffusion of chlorine to the metal/oxide interface, evaporating and subsequently oxidized to non-protective oxides and additional chlorine. In this mechanism, the role of alloying elements has not been fully understood until now. Hence, experiments were conducted on the influence of the alloying elements Mo, Ti, Si, Al on Fe–15Cr model alloys in an N2–5 vol.% O2 gas mixture with addition of 500 and 1500 vppm HCl with the use of thermogravimetric experiments at 600°C. The corrosive attack by ‘active oxidation’ strongly depends on the alloying elements. Fe–(15–25) wt% Cr alloys without any further addition of alloying elements show catastrophic ‘active oxidation’, characterized by the formation of metal chlorides at the metal–oxide interface. The addition of molybdenum, silicon and aluminium generally decreases the corrosive attack, whereas Ti has no beneficial effect and the corrosion is enhanced compared to Fe–15Cr. The presence of molten chlorides on heat exchanger tubes leads to catastrophic corrosion rates, even at relatively low temperatures of 250°C. Thick oxides scales of Fe2O3 are formed in contact with the gas phase. At the metal–oxide interface, a mixture of KCl–FeCl2 chloride phases are detected, probably molten at a reaction temperature of 400°C. Short term experiments on 2.25Cr–1Mo steel at 400°C in N2–5 vol.% O2–HCl gas mixtures covered with a eutectic KCl–ZnCl2 melt have shown that the metal is dissolved in the chloride melt at the melt/scale interface and precipitated as a thin layer of oxide at the melt/gas phase boundary. The kinetics of the corrosion strongly depends on gas phase composition i.e. p(HCl) and p(O2).

Keywords: chlorine induced corrosion, power plants

Future perspectives including fuel cells, gas turbines, USC and HGCU the US perspective

John Stringer1 and Lawrence A. Ruth2

1EPRI, Palo Alto, California 94304, USA

2National Energy Technology Laboratory, Pittsburgh, PA 15326-0940, USA

Both the US Department of Energy (DOE) and EPRI are developing models for the evolution of a secure energy future for the USA. Our general views are very similar. However, there are some differences in approach. DOE is concerned with all energy issues in the US future, including electricity, transportation fuels, and the industrial, commercial, and residential energy sectors; EPRI is concerned specifically with the electricity component, principally in the USA, but, as does DOE, also takes a global view. Both organizations take what is now known as a ‘Roadmapping’ approach. Roadmapping is an example of a ‘Top Down’ planning method: it involves the specification of a “destination” which the research and development program is aimed towards. In the DOE case, the destination refers to a secure energy future. Typically, Roadmapping is concerned with relatively long time scales. Time scales for different technologies are, of course, very different; in a fast-moving technology such as semiconductors, five to ten years may be a long time. For energy, the equipment is large; planning and construction times are long, and the expected lifetimes of the major components are not less than twenty years, and more typically up to forty years. The time scale that both of our organizations talk about is in the range 20–50 years in the future. The DOE model is called ‘Vision 21.’ The specific destination for Vision 21 is the technical design bases for near-zero emission fossil fueled energy plants. The EPRI model is called the ‘Electricity Technology Roadmap’, and more recently we have ‘A Vision of the Electricity System of 2020.’ An important aspect of the method common to both DOE and EPRI is that the destination is developed by what is called a ‘Stakeholder’ group: this involves not only the researchers and developers, but also the eventual customers for the technology, and the users of the products. This will include members with environmental and societal concerns. In this paper, we will highlight some of the scenarios that emerge from these models. The first part will concentrate on the Department of Energy program; the latter part on the EPRI view, remembering that we are in close agreement on most aspects.

Keywords: full cells, gas turbines, USC, HGCU, US perspective

High temperature carbon corrosion in solid oxide fuel cells

C.H. Toh1, P.R. Munroe1, D.J. Young1 and K Foger2

1School of Materials Science & Engineering, University of New South Wales, Sydney NSW 2056, Australia

2Ceramic Fuel Cells Limited, 170 Browns Road, Noble Park, VIC 3174

Operating conditions in a current design for a planar geometry oxide fuel cell plant are briefly reviewed and the danger of encountering “metal dusting” conditions identified. Laboratory tests were designed to produce accelerated metal dusting by exposing heat resisting alloys to a CO–26 H2–6 H2O (vol. pct) gas mixture at 680oC under thermal cycling conditions. The hot gas composition corresponded to ac = 2.9 and an oxygen potential high enough to oxidise chromium and aluminium, but not iron or nickel. The alloys tested included ferritic and austenitic chromia formers and two ferritic alumina formers, all with electropolished surfaces. Thermal cycling of the chromia formers led to oxide scale damage followed by internal carburisation, metal dusting and coking. This failure occurred very rapidly on most austenitic materials (Alloy 800, Inconel 601, 690, 693, Alloy 602CA), but did not commence until after approximately 50 one-hour cycles for the ferritic steel Fe–27Cr–0.001Y (wt %). The alloy with the best performance was Inconel 625, which was still protected by its Cr2O3 scale after 500 cycles. The alumina forming alloys showed superior performance, with no damage apparent after 1200 cycles. Additional tests using ground metal surfaces showed that they were more resistant to dusting in the case of chromia formers, but more susceptible in the case of alumina formers, metal dusting.

Keywords: high temperature carbon corrosion, solid oxide fuel cells, metal dusting

Formation and disappearance of an internal oxidation zone in the initial stage of the steam oxidation of Fe–9Cr–0.26Si ferritic steel

Mitsutoshi Ueda1, Makoto Nanko2, Kenichi Kawamura1 and Toshio Maruyama1

1Department of Metallurgy and Ceramics Science, Graduate School of Science and Engineering, Tokyo Institute of Technology, 2-12-1, Ookayama, Meguro-ku, Tokyo, 152-8552, Japan

2Department of Mechanical Engineering, Nagaoka University of Technology, 1603-1, Kamitomioka-machi, Nagaoka, Niigata, 940-2188, Japan

High temperature steam oxidation of the commercial Fe–9mass%Cr–0.26mass%Si steel (ASME T91) was carried out at 973 K, and the formation and disappearance of an IOZ (Internal Oxidation Zone) is discussed based on the microstructure observation by OM, and TEM with EDS. The IOZ formed at the initial stage and disappeared within about 100 ks. Sample oxidized for 61.2 ks showed a continuous IOZ, and TEM observation at the alloy/IOZ interface clarified that a sheet-like amorphous SiO2 layer formed intermittently. On the sample oxidized for 626.4 ks, the IOZ had disappeared and a sheet of amorphous SiO2 was located at the alloy/inner scale interface. The inner scale progressively grew next to the SiO2 layer due to an increase of oxygen potential at the interface because of the extremely low oxygen permeability in the amorphous SiO2.

Keywords: Fe–9Cr–0.26Si ferritic steel, steam oxidation, internal oxidation zone, amorphous SiO2 layer

Erosion of thermal barrier coatings

J. R. Nicholls1, R. G. Wellman1 and M. J. Deakin2

1PGTC, Cranfield University, Cranfield, Bedford MK43 0AL

2Rolls Royce Plc, PO Box 31, Derby DE24 8BJ

Thermal barrier coatings have been used within gas turbines for over 30 years to extend the life of hot section components. Thermally sprayed ceramics were the first to be introduced and are widely used to coat combustor cans, ductwork, platforms and more recently turbine aerofoils of large industrial engines. The alternative technology, electron beam physical vapour deposition,(EB-PVD) has a more strain-tolerant columnar microstructure and is the only process that can offer satisfactory levels of spall resistance, erosion resistance and surface finish retention for aero-derivative engines. Whatever technology is used, the thermal barrier must remain intact throughout the turbine life. Erosion may lead to progressive loss of TBC thickness during operation, raising the metal surface temperatures and thus shortening component life. Ballistic damage can lead to total TBC removal. This paper reviews the erosion behaviour of both thermally sprayed and EB-PVD TBCs relating the observed behaviour to the coating microstructure. A model for the erosion of EB-PVD ceramics is presented that permits the prediction of erosion rates. The model has been validated using a high velocity erosion gas gun rig, both on test coupons and samples removed from coated components. The implications of erosion on component life are discussed in the light of experimental results and the model predictions.

Keywords: thermal barrier coatings, gas turbines

Materials issues in bubbling PFBC systems

I. G. Wright1, J. Stringer2, and J. M. Wheeldon3

1Oak Ridge National Laboratory, Oak Ridge, Tennessee, USA, 2EPRI, Palo Alto, California, USA and 3EPRI, Wilsonville, Alabama, USA

In recent years, considerable experience and insight regarding the behavior of materials and components in the bubbling pressurized fluidized-bed combustion (PFBC) of coal has been developed, largely as a result of the operation of a fleet of 80-MW(e) PFBC-based power plants installed by ABB Carbon. The first plant went into operation over 10 years ago and it appeared timely to review this practical experience and to document areas where improvements have been made, or can be suggested. In keeping with general bubbling-bed experience, the in-bed heat exchanger and water-wall tubes experienced metal loss. Other plant areas that experienced difficulties were the hot-gas cyclone system, the gas-turbine expander, and some balance-of-plant items including solids-handling equipment, valving, and expansion joints. The captured dust removal lines from the cyclones sometimes plugged sending high dust loadings over to the turbine expander. Consequently the turbine blades experienced material deposition and significant erosion damage. Concerns about turbine longevity in this application have led to attempts to develop high-temperature filter systems that protect the turbine by removing all the dust from the flue gas prior to expansion. These filters have themselves suffered from a range of materials problems, which is not unexpected for a relatively new technology. The current experience of these and other materials issues is reviewed.

Keywords: bubbling pressurized fluidized-bed combustion of coal