Fireside Corrosion Behavior of HVOF and Plasma-Sprayed Coatings in Advanced Coal/Biomass Co-Fired Power Plants

This article presents a systematic evaluation of coatings for advanced fossil fuel plants and addresses fireside corrosion in coal/biomass-derived flue gases. A selection of four candidate coatings: alloy 625, NiCr, FeCrAl and NiCrAlY were deposited onto superheaters/reheaters alloy (T91) using high-velocity oxy-fuel (HVOF) and plasma spraying. A series of laboratory-based fireside corrosion exposures were carried out on these coated samples in furnaces under controlled atmosphere for 1000 h at 650 °C. The tests were carried out using the “deposit-recoat” test method to simulate the environment that was anticipated from air-firing 20 wt.% cereal co-product mixed with a UK coal. The exposures were carried out using a deposit containing Na₂SO₄, K₂SO₄, and Fe₂O₃ to produce alkali-iron tri-sulfates, which had been identified as the principal cause of fireside corrosion on superheaters/reheaters in pulverized coal-fired power plants. The exposed samples were examined in an ESEM with EDX analysis to characterize the damage. Pre- and post-exposure dimensional metrologies were used to quantify the metal damage in terms of metal loss distributions. The thermally sprayed coatings suffered significant corrosion attack from a combination of aggressive combustion gases and deposit mixtures. In this study, all the four plasma-sprayed coatings studied performed better than the HVOF-sprayed coatings because of a lower level of porosity. NiCr was found to be the best performing coating material with a median metal loss of ~87 μm (HVOF sprayed) and ~13 μm (plasma sprayed). In general, the median metal damage for coatings had the following ranking (in the descending order: most to the least damage): NiCrAlY > alloy 625 > FeCrAl > NiCr.     

High-Temperature Performance of Ferritic Steels in Fireside Corrosion Regimes: Temperature and Deposits

The paper reports high temperature resistance of ferritic steels in fireside corrosion regime in terms of temperature and deposits aggressiveness. Four candidate power plant steels: 15Mo3, T22, T23 and T91 were exposed under simulated air-fired combustion environment for 1000 h. The tests were conducted at 600, 650 and 700 °C according to deposit-recoat test method. Post-exposed samples were examined via dimensional metrology (the main route to quantify metal loss), and mass change data were recorded to perform the study of kinetic behavior at elevated temperatures. Microstructural investigations using ESEM-EDX were performed in order to investigate corrosion degradation and thickness of the scales. The ranking of the steels from most to the least damage was 15Mo3 > T22 > T23 > T91 in all three temperatures. The highest rate of corrosion in all temperatures occurred under the screening deposit.     

Carburization of 9 %Cr Steels in a Simulated Oxyfuel Corrosion Environment

Laboratory corrosion tests, simulating fireside corrosion of 9 mass%Cr steels in an oxyfuel boiler of combusting bituminous 0.6 %S coal, were conducted. Thermodynamic calculation of the tube deposit chemistry indicated that sulfates, instead of carbonates, should be the stable corrosive salts. Specimens of 9 %Cr steels were coated with a synthetic salt mixture of 37.5 m/o Na₂SO₄–37.5 m/o K₂SO₄–25 m/o Fe₂O₃ and were exposed to 57.6 vol%CO₂–22.4 %H₂O–1.6 %O₂–0.16 %SO₂–18.2 %N₂ at 500–700 °C for 96 h. A two-layered scale of outer “pure” iron oxides and inner oxide/sulfide mixture of Cr and Fe formed on the specimens. The two-layered scale became thick at temperatures where the deposits were fused. The tested steals also underwent internal carburization, with the carbide-zone thickness increasing with increasing temperature for the given exposure time. Growth rate of the carburization zone was much less than the estimation derived from diffusion constants of carbon in α-Fe.     

Field test corrosion experiments in Denmark with biomass fuels. Part 1: Straw‐firing

In Denmark, straw and other types of biomass are used for generating energy in power plants. Straw has the advantage that it is a “carbon dioxide neutral fuel” and therefore environmentally acceptable. Straw combustion is associated with corrosion problems which are not encountered in coal‐fired plants. The type of corrosion attack can be directly ascribed to the composition of the deposit and the metal surface temperature. A series of field tests have been undertaken in the various straw‐fired power plants in Denmark, namely Masnedø, Rudkøbing and Ensted. Three types of exposure were undertaken to investigate corrosion: a) the exposure of metal rings on water/air cooled probes, b) the exposure of test tubes in a test superheater, and c) the exposure of test tubes in existing superheaters. Thus both austenitic steels and ferritic steels were exposed in the steam temperature range of 450–600°C. The corrosion rates were assessed by precision measurements of material loss and internal corrosion. The corrosion products and course of corrosion for the various steel types were investigated using light optical and scanning electron microscopy. Corrosion mechanisms are discussed in relation to temperature and deposit composition.     

Long‐term exposure of austenitic steels and nickel‐based alloys in lignite‐biomass cofiring

The aim of this study was to assess the long‐term impact that the addition of biomass provokes on superheater materials exposed to fireside corrosion environments. Alloys covering a broad range of commercially available materials were investigated. Their corrosion kinetics under different corrosive deposits and atmospheres was evaluated, and their corrosion products analyzed to deepen understanding of the underlying corrosion mechanisms. Therefore, three nickel‐based alloys and three austenitic steels containing 20–24 wt.% Cr were tested at 650°C for 7,000 hr. The long‐term exposure shows new mechanistic aspects of Type II hot corrosion that were revealed by accelerated material depletion. The formation of Ni–NiS eutectic and the formation of a Cr depleted zone close to the substrate corrosion product interface are indicative of the breakaway occurrence. Differences in the corrosion behavior are related to the balance of Ni, Mo, Co, and Cr and can serve as the material selection argument. The evaluation concluded with the finding that alloys presenting Mo and Ni might be preferentially used in fireside corrosion in the presence of biomass, whereas the use of austenitic steels suffer less corrosion if no biomass is present in the corrosive atmosphere.     

Fireside corrosion degradation of ferritic alloys at 600 °C in oxy-fired conditions

This paper reports the results of a study carried out to investigate the effects of simulated coal/biomass combustion conditions on the fireside corrosion. The 1000 h deposit recoat exposure (5 × 200 h cycles) was carried out at 600 °C. In these tests ferritic alloys were used 15Mo3, T22, T23 and T91. Kinetics data were generated for the alloys exposed using both traditional weight change methods and metal loss measurements. The highest rate of corrosion based on EDX results occurred under D1 deposit where provoke mainly by the formation of alkali iron tri-sulphate phase.     

Optimisation of in‐service performance of boiler steels by modelling high‐temperature corrosion

The main objective of the EU OPTICORR project is the optimisation of in‐service performance of boiler steels by modelling high‐temperature corrosion, the development of a life‐cycle approach (LCA) for the materials in energy production, particularly for the steels used in waste incinerators and co‐fired boiler plants. The expected benefits of this approach for safe and cost effective energy production are: ‐ control and optimisation of in‐service performance of boiler materials, ‐ understanding of high‐temperature corrosion and oxidation mechanisms under service conditions, ‐ improvement of reliability to prevent the failure of components and plant accidents and ‐ expanding the limits of boiler plant materials by corrosion simulations for flexible plant operation conditions (steel, fuel, temperature etc.). The technical aim of the EU OPTICORR project is the development of modelling tools for high‐temperature oxidation and corrosion specifically in boiler conditions with HCl‐ and SO₂‐containing combustion gases and Cl‐containing salts. The work necessitates thermodynamic data collection and processing. For development and modelling, knowledge about the corrosion mechanisms and exact data are needed. The kinetics of high‐temperature oxidation and corrosion are determined from laboratory thermo‐gravimetric tests (TG) and multi‐sample exposure tests. The materials studied are typical boiler tubes and fin‐steels: ferritic alloys, the austenitic steel T347 and the Ni‐based alloy Inconel 625. The exposure gases are dry air, air with 15 vol‐% H₂O, and with 2000 ppm HCl and 200 ppm SO₂. The salt deposits used are based on KCl‐ZnCl₂ and Ca, Na, K, Pb, Zn‐sulfates. The test temperatures for exposures with deposits are 320 and 420°C and, for gas exposures, 500 to 600°C. At present the tools being developed are ChemSheet based programmes with a kinetic module and easy‐to‐use interface and a more sophisticated numerical finite‐difference‐based diffusion calculation programme, InCorr, developed for prediction of inward corrosion and internal corrosion. The development of modelling tools for oxidation and high‐temperature corrosion was started with thermodynamic data collection for relevant systems and thermodynamic mappings. Further, there are needs to develop the simulation model and tool for salt‐induced hot corrosion based on the ChemSheet approach. Also, the work on modelling and simulating with the InCorr kinetic modelling tool will be continued to demonstrate the use of the tool for various steels and alloys in defined combustion environments.     

HCl-Induced High Temperature Corrosion of Stainless Steels in Thermal Cycling Conditions and the Effect of Preoxidation

Gaseous HCl released during combustion is one reason for the severe materials degradation often encountered in power generation from waste and biomass. In this study, three stainless steels (the low alloyed EN 1.4982, the standard EN 1.4301 and the higher alloyed EN 1.4845) were tested by repeated thermal cycling in an environment comprising N₂?C10%O₂?C5%H₂O?C0.05%HCl at both 400 and 700 °C. The materials were exposed with ground surfaces and preoxidised at 400 or 700 °C. A positive effect of preoxidation is evident when alloys are exposed at 400 °C. Oxide layers formed during preoxidation effectively suppress chlorine ingress for all three materials, while chlorine accumulation at the metal/oxide interface is detected for surface ground specimens. The positive effect of preoxidation is lost at 700 °C and corrosion resistance is dependent on alloying level. At 700 °C metal chloride evaporation contributes significantly to the material degradation. Based on the results, high temperature corrosion in chlorinating environments is discussed in general terms.     

Effect of Zinc Chloride in Ash in Oxidation Kinetics of Ni-Based and Fe-Based Alloys

Corrosion by molten phases leads to severe corrosion of heat exchangers in waste-to-energy plants. In addition, the presence of heavy metal chlorides in ash deposit increases degradation at low temperature due to the formation of highly corrosive molten phases. In this study, two heat exchanger materials, a low alloy steel (16Mo3) and a nickel based alloy (Inconel 625) were exposed in air to three different synthetic ashes, with various chloride contents, including ZnCl₂ at isothermal temperatures of 450 and 650 °C in a muffle furnace. After the test, thickness and mass losses were evaluated on two separate samples, and metallographic cross sections of the specimens were characterized via SEM/EDX analyses. Both measurement results were in good agreement and showed that the corrosion observed on both materials was higher in the presence of zinc chloride in ash at 450 °C than in ashes without heavy metal chloride at 650 °C.     

Corrosion of austenitic steels in molten sulphate deposits

Thermal analysis has been used to determine the melting behaviour of alkali metal sulphate deposits typical of those found on superheater tubes in coal-fired power station boilers. Additions of 5–30 m/o Fe₂(SO₄)₂ to Na₂SO₄K₂SO₄ mixtures under a simulated flue gas (N₂ + 15 v/o CO₂ + 1 v/o O₂ + 0.3 v/o SO₂) are shown to reduce the melting point from 820°C to below 550°C. Alkali-iron trisulphates are formed which re-solidify on heating above 720°C by decomposition of Fe₂(SO₄)₃ at low thermodynamic activity. It is suggested that the effect of heat flux and SO₃ potential gradient on the melting behaviour of a superheater deposit could account for the observed ‘bell-shaped’ temperature dependence of corrosion rate in the range 550–750°C. Results are presented of laboratory corrosion tests on three commercial austenitic stainless steels using a melt of composition 74 m/o Na₂SO₄ + 20 m/o K₂SO₄ + 6 m/o Fe₂ (SO₄)₃. A model is proposed for corrosion by an acid fluxing mechanism involving refractory metal alloy elements in the steels. The model offers a quantitative basis for prediction of fireside corrosion resistance in coal-fired boilers, and a comparison of relative corrosion rates of a range of steels in laboratory tests, probe trials and service performance confirms the validity of the present model.     

Field test of superheater corrosion in a CFB waste boiler: Part I – Metal loss characteristics

The corrosion was investigated on a superheater test coil in a CFB waste boiler. The alloys ranged from ferritic steel T22 to nickel‐based Alloy 65 and the metal temperatures were between 460 and 540°C. The thickness of the deposit was alloy and temperature dependent. The low‐alloyed steels developed thick deposits at all temperatures while the deposit thickness increased with the temperature on the high‐alloyed steels and the nickel‐based alloy. The corrosion attack was alloy dependent and related to the deposit crest. The nickel‐based Alloy 65 was preferentially attacked directly under the crest of the deposit while the other alloys were preferentially attacked at the edge. The corrosion rate increased with temperature for X20, Alloy 304L, Alloy 310 and Alloy 825; decreased on Alloy 65; and was bell shaped on T22 and Alloy 28. Alloy 310 suffered from severe pitting corrosion in a line following the edge of the deposit crest. The best overall corrosion resistant alloy was Alloy 28.     

The Effect of Temperature on the Formation of Oxide Scales Regarding Commercial Superheater Steels

This study addresses the surface changes of three commercial steels (a low alloy ferritic 10CrMo9-10 steel, a Nb-stabilized austenitic AISI347 steel, and a high alloy austenitic Sanicro 28 steel) by comparing the oxide scale thicknesses, chemical compositions, and surface morphologies of samples after pre-oxidation at 200, 500 and 700 °C with different exposure times (5 and 24 h) under humid or dry conditions. With all three steels, the oxide scale thickness increased as functions of temperature and exposure time, the effect of temperature being more prominent than the effect of exposure time. The presence of water resulted in thicker oxide scales at the studied low alloy ferritic steel, whereas in the two austenitic steels, the presence of water increased chromium diffusion to the oxide scale rather than the scale thickness. The oxide layers characterized and analyzed in this paper will be further studied in terms of their abilities to resist corrosion by exposing them under corrosive conditions. The results regarding the corrosion resistance of the steels will be published in a sequel paper.     

The effect of an impinging liquid‐solid jet on the electrochemical corrosion of stainless steels

Stainless steels rely on their passive film for corrosion protection in saline environments and localised disruption of the passive film can lead to high local rates of material loss. Materials which exhibit passivity in static conditions are often susceptible to erosion‐corrosion under severe hydrodynamic regimes. In this paper the corrosion behaviour of two superaustenitic (UNS S31254 and UNS S32654) and one superduplex (UNS S32750) stainless steels was examined in static conditions at a range of temperatures from 18°C to 70°C and during exposure to an impinging saline jet containing varying concentrations of solid particles. In static conditions the materials exhibit decreasing resistance to passivity breakdown as the temperature is increased and a critical breakdown temperature for UNS S31254 and UNS S32750 was identified. Under liquid‐solid impingement the materials exhibit an active regime near to the free corrosion potential followed by a stabilised current regime as more positive potentials. The complex electrochemical response is dependent on the material grade and the solid particle concentration. Also in the paper assessment of the „recovery”︁ of the stainless steels after exposure to slurry impingement is addressed.     

Hot corrosion of coated and uncoated single crystal gas turbine materials

Single crystal superalloys were originally developed for use in the hot gas paths of aero‐engines with fuels and air that could both contain low levels of elements that had the potential to produce significant corrosive damage to these materials. However, in the development of power systems with increased efficiencies, single‐crystal materials are now being used in industrial gas turbines, which operate with higher levels of potentially corrosive elements. Therefore, the hot corrosion of single crystal materials, both bare and with coatings, is of interest in assessing the viability of their use in combusted gas environments that contain progressively higher levels of potentially damaging elements. The paper summarises work that has been carried out to investigate the hot corrosion of two single crystal alloys (CMSX‐4 and SC²‐B), both bare and with Pt‐Al coatings. A series of laboratory tests has been carried out using the ‘deposit replenishment’ technique to investigate the sensitivity of the corrosion damage to single crystal materials and coatings to a range of exposure conditions anticipated for gas turbines using combustion gases that contain high levels of potentially damaging elements. The exposure variables studied during the series of 500 h tests included:

• Gas composition: the effects of different levels of SO_x, HCl and steam in air;
• deposition flux: the effects of 4/1 (Na/K)₂SO₄ deposits applied with deposition fluxes of 1.5, 5 and 15 µg/cm²/h.
• temperature: 700 and 900 °C.

During the course of the tests, the performance of the materials was monitored using conventional mass‐change methods. But the main aim of the work was to develop predictive models for the hot corrosion damage to the materials in terms of metal loss. Such quantitative data on the damage to the materials in terms of metal losses have to be obtained using dimensional metrology: for this study this was carried out using pre‐exposure contact measurements and post‐exposure measurements of features on polished cross‐sections. These measurement methods have allowed distributions of damage data to be determined and permitted different approaches to corrosion modelling to be investigated and their viability to be assessed.     

Effect of Pressures on the Corrosion Behaviours of Materials at 625°C

The corrosion behaviors of austenitic stainless steels (SS) 310, 304 and Ni- and Fe-based A-286 exposed to 0.1 MPa, 8 MPa and 29 MPa at 625°C for 1000 h were investigated. These represent exposure to superheated steam, subcritical and supercritical water (SCW) at 625°C, respectively. As SS 310 showed the smallest weight change, the oxide cross-sections made from 310 samples were examined by transmission electron microscopy. The results revealed a single-layer oxide at 0.1 MPa and dual-layer oxides at 8 MPa and 29 MPa, followed by a Cr-depleted region into the austenite substrate. The compositions of the inner oxides at 8 MPa and 29 MPa are Cr-rich and largely similar to those of the single-layer oxides at 0.1 MPa exposure. These results suggest that corrosion testing in superheated steam may be a suitable surrogate for scoping tests of materials under SCW conditions at >650°C.     

Corrosion behavior of various steels in the simulated fluidized bed boiler environment

The corrosion behavior of various austenitic stainless steels and high-alloy steels has been studied in simulated fluidized bed boiler environment to develop a new corrosion resistant austenitic stainless steel for the superheater tube. The superheater is usually not installed within the bed position, which is different from the evaporator installed within the bed position. Therefore, the superheater tubes are exposed to an oxidizing environment; but it is also necessary to estimate the corrosion resistance of the steels in a reducing environment. It is already known that the high temperature corrosion behavior in conditions where CaSO₄ is coated on the steels is more important than the erosion of the superheater tubes. The main results in this present study are as follows: The Nb bearing steels and low C steels showed good resistance to high-temperature corrosion in CaSO₄/CaO, e.g. 347, 304L and HR3C. The corrosion rate of all steels used increased with increase in temperature, particularly at temperatures higher than 650°C. Internal penetration was not detected at temperatures lower than 550°C, but it was detected at temperatures higher than 600°C, in particular, higher than 650°C. The corrosion thickness loss was almost the same as the internal penetration depth at 700 and 750°C in the 300 series steels placed in CaSO₄/CaO, including the fine grained 347 steel, while the internal penetration depth was larger than the corrosion thickness loss in high-alloyed materials such as Alloy 800 and 310 steels. At temperatures higher than 800°C, the same result was also obtained for the fine grained 347 steel. The corrosion during exposure to oxidizing or reducing gases without CaSO₄/CaO or CaS was slight, but when the test specimens were placed in CaSO₄/CaO or CaS, the corrosion rate sharply increased, regardless of the atmospheric gas composition. Cr, Si, Mn (less than 5 %), Mo and Nb are beneficial elements while C, Cu and Al are harmful elements. From the above results, the following steel was developed for high temperature corrosion resistance in CaSO₄/CaO: low C-22/25Cr-17/25Ni-3/5Mn-(2Mo)-Nb-0.08/0.2N-Al-(B).     

Sodium vanadate induced corrosion of MCrAlY coatings – Burner rig studies

The corrosion of several MCrAlY-coatings (M = Ni, Fe and Ni + Co) has been studied in a high velocity burner rig at 650, 800 and 950°C. The fuel used was diesel oil with additions of 3% sulphur, 200 ppm vanadium and 100 ppm sodium. The deposits formed on the specimens mainly consisted of sodium vanadates which were molten at the test temperature. Sodium sulphate was only found at and below 650°C. The corrosion mechanism involved was vanadate-induced hot corrosion. This corrosion mechanism is characterized by the formation of an oxide layer adjacent to the metal, the dissolution of oxide in the molten deposit, and precipitation of vanadates or oxides near the outer surface of the deposit. The continuous deposition of fresh vanadate on the surface served to maintain high corrosion rates for extended exposures.     

Na<Subscript>2</Subscript>SO<Subscript>4</Subscript>-Deposit-Induced Corrosion of Mo-Containing Alloys

Disk alloys used in advanced gas turbine engines often contain significant amounts of Mo (2 wt% or greater), which is known to cause corrosion under Type I hot corrosion conditions (at temperatures around 900 °C) due to alloy-induced acidic fluxing. The corrosion resistance of several model and commercial Ni-based disk alloys with different amounts of Mo with and without Na₂SO₄ deposit was examined at 700 °C in air and in SO₂-containing atmospheres. When coated with Na₂SO₄ those alloys with 2 wt% or more Mo showed degradation products similar to those observed previously in Mo-containing alloys, which undergo alloy-induced acidic fluxing Type I hot corrosion even though the temperatures used in the present study were in the Type II hot corrosion range. Extensive degradation was observed even after exposure in air. The reason for the observed degradation is the formation of sodium molybdate. Transient molybdenum oxide reacts with the sodium sulfate deposit to form sodium molybdate which is molten at the temperature of study, i.e., 700 °C, and results in a highly acidic melt at the salt alloy interface. This provides a negative solubility gradient for the oxides of the alloying elements, which results in continuous fluxing of otherwise protective oxides.     

Long-term Hot Corrosion Behavior of Boiler Tube Alloys in Waste-to-Energy Plants

Accelerated corrosion of candidate alloys was induced by metal chlorides/sulfates at 500 °C. Results suggest that the corrosivity of the studied metal chlorides increases in the order CaCl₂ < NaCl < KCl < ZnCl₂ < PbCl₂ < FeCl₂. Mechanisms to explain the different impacts of chlorides were proposed. It is believed that materials exposed to chloride salts corrode through vicious cycles, in which a shorter path of the cycle leads to a higher corrosion rate. Experimental results confirmed that FeCl₂ with the shortest path of the corresponding vicious cycle has the highest corrosion rate. It is also confirmed that the sulfates of Zn and Pb are less corrosive than their chlorides for the alloys tested. A kinetic study on the hot corrosion of T22, Esshete 1250 and Sanicro 28 was carried out under simulated waste-to-energy (WTE) ashes at 500 °C for 1000 h. Results from the kinetic study show that T22, Esshete 1250, and Sanicro 28 exhibited comparable performance for short-term exposure; however, the degradation thickness presented a clear trend after the 1000-h exposures in terms of decreasing resistance to corrosion: T22 > Esshete 1250 > Sanicro 28. EDX maps confirmed the role of Ni/Cr for slowing the corrosion kinetics of these three alloys.     

Contribution to the discussion on Localized Corrosion of Stainless Steels in natural sea water

Electrochemical hysteresis methods are employed to develop experimental potential-pH diagrams for some commercial austenitic and ferritic stainless steels in 3% sodium chloride solutions at room temperature. The results were compared with the electrochemical and the gravimetric behaviour of the single stainless steels in natural sea water. The comparison suggested some considerations about the mechanism of initiation and growth of pitting and crevice corrosion prevailing in natural environment. The higher probability of pits observed in field exposure was correlated to the deposit of a microbiological slime on metal surface that ennobles the free corrosion potentials of passive surfaces up to +400 ÷ +450 m V SCE. The crevice corrosion initiation was in turn attributed to a moderate local acidification originating in conditions of reduced diffusion under the action of passivity current that makes possible pit nucleation which is otherwise unlikely to occur. The pitting growth speeds up the acidification process in the interstice until it causes general corrosion of the walls.