Search

Realizing higher operating temperatures to increase efficiency of future applications for energy conversion and storage while minimizing cost is a challenge for development of high-temperature materials. Simultaneous optimization of mechanical properties and corrosion resistance continues to be a difficult task but is essential due to the need to significantly accelerate the transition between technology readiness levels in the future. Oxidation-induced degradation will be a critical life-limiting mechanism at increased operating temperatures. Suitable high-temperature materials cannot be solely identified by time-consuming experiments and reliable computational methods incorporating the relevant physics of processes must be considered to complement the experimental efforts. In the present work, a review of the methods employed to model oxidation-induced material degradation described in literature will be discussed. Furthermore, their capability to predict lifetime and aid in material selection will be evaluated.

Mo(AlxSi1-x)2 alloy with x in the range of 0.35–0.65 were prepared by a one-step spark plasma sintering. To study the exclusive formation of an α-Al2O3 scale, oxidation experiments were conducted in low and high oxygen partial pressure ambient at 1373 K; viz.: 10−14 and 0.21 atm. The oxidation kinetics follows a parabolic rate law after a transient period. A counter-diffusion process of O and Al along grain boundaries of Al2O3 scale is responsible for the equiaxed and columnar grain growth based on a two-layered microstructure. The formation of a dense equiaxed α-Al2O3 layer contributes to excellent oxidation resistance.

MCrAlY (M = Ni, Co) coatings are commonly used on gas-turbine components as oxidation resistant overlay coatings and bondcoats for thermal barrier systems. The present work focuses on the effect of the aluminizing process on the CoNiCrAlY coating microstructure. In the as-received condition the outer part of the coating consisted mostly of β-(Ni,Co)Al with interspersed precipitates of Cr-rich carbide and Cr-rich boride precipitates. Formation of σ-CoCr was observed at the interface between the β-layer and the inner initial CoNiCrAlY microstructure. Scanning electron microscopy (SEM) combined with energy and wavelength-dispersive X-ray spectroscopy (EDX/WDX) was employed to characterize the aluminized CoNiCrAlY coating. Phases were then identified by electron backscatter diffraction (EBSD). Detailed microstructural studies of the coating were corroborated with the help of coupled thermodynamic-kinetic calculations to model the aluminizing process. The calculations were performed with the in-house developed code employing the commercially available thermodynamic and kinetic databases (ThermoCalc). The mechanisms of the observed microstructural changes were elucidated with the help of the modelling results.

We employed correlative atom probe tomography (APT) and transmission electron microscopy (TEM) to analyze the alumina scale thermally grown on the oxide dispersion-strengthened alloy MA956. Segregation of Ti and Y and associated variation in metal/oxygen stoichiometry at the grain boundaries and triple junctions of alumina were quantified and discussed with respect to the oxidation behavior of the alloy, in particular, to the formation of cation vacancies. Correlative TEM analysis was helpful to avoid building pragmatically well-looking but substantially incorrect APT reconstructions, which can result in erroneous quantification of segregating species, and highlights the need to consider ionic volumes and detection efficiency in the reconstruction routine. We also demonstrate a cost-efficient, robust, and easy-handling setup for correlative analysis based solely on commercially available components, which can be used with all conventional TEM tools without the need to modify the specimen holder assembly.

The microstructural changes of the aluminized alloy 625 during cyclic oxidation in air + 6% H2O at 900 °C and 1000 °C were analyzed using optical metallography (OM), scanning electron microscopy (SEM) with energy and wave length dispersive X-ray analysis (EDX/WDX) as well as electron backscatter diffraction (EBSD). An in-house developed thermodynamic-kinetic procedure was employed to predict the microstructural evolution of aluminized alloy 625 during high temperature exposure by considering simultaneously occurring surface oxidation and interdiffusion processes. Due to the lack of mobility data for the relevant alloying elements in the σ-phase, assumptions for the mobilities were made based on the value of the mobilities in α-Cr. Despite these assumptions, the calculated results were found to be in good agreement with experimental observations. The complete depletion of β-NiAl in the coating observed during exposure at 1000 °C was correctly predicted by the model. The model was also able to predict dissolution of the precipitate phases α-Cr and σ in the interdiffusion zone during exposures at 900 °C and 1000 °C. The model was however unable to predict the formation of the μ-phase in the alloy after 1000 h of exposure at 1000 °C. The developed modelling approach offers the potential to predict microstructural changes of aluminized nickel base alloys thus reducing cost and time consuming experimental efforts.

The effect of alloying elements in the substrate on the interdiffusion processes between a NiCoCrAlY-coating and model Ni-base superalloys was studied. Four single-crystal substrates Ni-10Cr-8Al-X (wt%, X = 10Co, 9Ta, 6 W) were coated with a conventional NiCoCrAlY-coating and exposed to air for up to 1000 h at 1050 °C. The microstructural analyses (LOM/SEM/EDX) revealed a strong dependence of the coating degradation rate on the substrate composition. The coating applied to the ternary base Ni-10Cr-8Al alloy and to the quaternary Co-containing substrate degraded most rapidly while the Ta addition to the substrate effectively suppressed interdiffusion. The degradation phenomena such as β-NiAl depletion and propagation of the secondary reaction zone (SRZ) were interpreted in terms of a thermodynamical analysis in Thermo-Calc and by using a CALPHAD-based thermodynamic-kinetic diffusion model.

FeCrAl-base alloys are well known for their excellent oxidation resistance due to formation of a slowly growing alumina surface scale during high-temperature service. The actual scale growth mechanism and especially adherence are strongly affected by the presence of oxygen-active elements such as yttrium, titanium or hafnium. In the present study, the effect of titanium addition on the scale growth mechanisms of an yttrium oxide dispersion-strengthened FeCrAl base alloy was studied during oxidation at 1200 °C in Ar–O2. For microstructural characterization results of scanning electron microscopy and electron backscatter diffraction were combined with X-ray diffraction data. Scale growth mechanisms were investigated by two-stage oxidation using 18O tracer with subsequent scale analyses using secondary neutrals mass spectrometry. The scale on the alloy without intentionally added titanium grew virtually exclusively by oxygen diffusion along oxide grain boundaries and exhibited a columnar structure with the grain size increasing in growth direction. The addition of titanium resulted in formation of an outer oxide zone of equiaxed grains on top of the inner columnar part. The equiaxed grains increased in size with increasing exposure time. Comparison with the tracer studies revealed that the titanium-induced equiaxed zone was the result of outer scale growth. Mechanisms for the initiation of outward aluminium transport are discussed. Indications were found that the effect of titanium on the scale growth mechanisms already occurred for titanium additions as low as 0.02 wt%.

Three model alloys, Co-10Ni-20Cr-15W, Co-30Ni-20Cr-15W and Ni-20Cr-15W (all in wt.%) were investigated in Ar-50%H2O at 700 and 750 °C for up to 3000 h reaction. The results showed the formation of thin chromia scales on the sample surfaces in all cases. For the Co-base alloys, accompanied by the Cr2O3 formation, a chromium depletion zone was detected underneath the oxide scale along with Co3W with some amount of dissolved Cr and Ni. This kind of Co3W was enriched along the oxide/alloy interface. The formation of this intermetallic phase was considered to follow nucleation and subsequent growth based on morphology and composition analyses. Increasing nickel content reduced the amount of Co3W formation. For the nickel-base alloy Ni-20Cr-15W, no intermetallic phase was detected. The formation of the Co3W intermetallic was discussed based on phase transformation induced by chromium depletion. The effect of nickel on this phenomenon was also discussed according to thermodynamic phase equilibrium analysis.

The performance of MCrAlY (M = Ni, Co) bondcoats for atmospheric plasma-sprayed thermal barrier coatings (APS-TBCs) is substantially affected by the contents of Co, Ni, Cr, and Al as well as minor additions of Y, Hf, Zr, etc., but also by manufacturing-related properties such as coating thickness, porosity, surface roughness, and oxygen content. The latter properties depend in turn on the exact technology and set of parameters used for bondcoat deposition. The well-established LPPS process competes nowadays with alternative technologies such as HVOF and APS. In addition, new technologies have been developed for bondcoats manufacturing such as high-velocity APS or a combination of HVOF and APS for application of a flashcoat. Future developments of the bondcoat systems will likely include optimization of thermal spraying methods for obtaining complex bondcoat roughness profiles required for extended APS-TBC lifetimes. Introduction of the newest generation single-crystal superalloys possessing low Cr and high Al and refractory metals (Re, Ru) contents will require definition of new bondcoat compositions and/or multilayered bondcoats to minimize interdiffusion issues. The developments of new bondcoat compositions may be substantially facilitated using thermodynamic–kinetic modeling, the vast potential of which has been demonstrated in recent years.

The growth of protective α -Al 2 O 3 scales on the NiCrAl-base alloy 602 CA at 1100 °C and 1200 °C was promoted by multi-step oxidation treatments. Different microstructural changes were observed during high temperature exposures in the pre-oxidised specimens as compared to the bare specimens. A coupled thermodynamic-kinetic model was employed to describe the influence of pre-oxidation on the microstructural evolution in the alloy. The model was able to correctly predict the mechanisms of the dissolution and re-precipitation of the carbides in the pre-oxidised and bare specimens.

One of the options to manufacture MCrAlY bondcoats (M = Co, Ni) for thermal barrier coating systems is High Velocity Oxy-Fuel spraying (HVOF). In this work, particle diagnostics were applied to investigate the impact of processing conditions and feedstock characteristics on the relevant bondcoat properties. The results showed that compromises must be made on the oxygen/fuel ratio, spray distance, and particle size distribution to strike a balance between low oxidation and dense microstructures. These limitations initiated the development of the High Velocity Atmospheric Plasma Spray process (HV-APS) as a further alternative process. In this work, HV-APS process parameters were developed for a three cathode torch in combination with a 5 mm diameter high speed nozzle. A one-dimensional calculation of the expansion through this nozzle to atmospheric pressure yielded supersonic conditions with a Mach number of 1.84. The calculated plasma temperatures at the nozzle exit and in the expanded jet are 8400 K and slightly above 5200 K, respectively, which is low compared to conventional APS processes. A very fine powder with a median particle size of 18 μm was identified to be most suitable. Although the spray conditions were relatively cold, reasonable deposition efficiencies up to 61% and rather dense coatings were achieved using this feedstock. The as-sprayed porosity was ≈ 2% which was reduced by the subsequent vacuum heat treatment to Moreover, reference samples were manufactured by Low Pressure Plasma Spraying (LPPS). The oxidation behavior was compared in isothermal and cyclic oxidation tests. The oxidation rates of the HV-APS coatings were found to be significantly lower than those of LPPS coatings. The thermally grown oxide scale showed less yttrium incorporation and better adherence in case of HV-APS. The latter is suggested to be related to a unique new distribution of Y-rich nano-sized oxide precipitates. The cyclic oxidation test confirmed the better oxidation resistance of the HV-APS coatings.

In the present paper the high temperature oxidation behavior of a free standing NiCoCrAlY coating produced by high-velocity atmospheric plasma spraying (HV-APS) is investigated and compared with that produced by conventional low pressure plasma spraying (LPPS). Isothermal thermogravimetric experiments at 1000 and 1100 °C in synthetic air revealed a lower oxidation rate of the HV-APS than the LPPS coating. Both coatings formed oxide scales based on alpha alumina, however, in the LPPS coating incorporation of coarse mixed Y/Al-oxide pegs into the scale occurred, increasing the oxidation rate by providing short circuit paths for oxygen diffusion probably due to higher diffusivities in the mixed oxide and/or along the interfaces between mixed oxide and alumina. In the HV-APS coatings most of the yttrium was tied-up in sub-μm Y-containing oxide particles and only minor amounts of mixed Y/Al oxide precipitates were found in the alumina surface scale. Cyclic air oxidation tests at 1100 °C revealed a lower oxidation rate and better scale adherence for the HV-APS coating. The results thus show that HV-APS is a promising method for the processing of MCrAlY coatings. The specific yttrium distribution in form of fine oxide precipitates in the HV-APS material prevents the formation of deleterious Y-rich oxide pegs and promotes formation of a slowly growing, protective alumina scale with excellent adherence.

Stationary applications of Solid Oxide Fuel Cell systems require operating times of 40,000 to 80,000 h for market introduction. Therefore, extended lifetime tests are essential for learning about the long-term behavior and various degradation mechanisms and to foster ideas about accelerated stack testing. The Forschungszentrum Julich has been gradually extending the testing time, resulting in successful short-stack operating times of between 20,000 and 40,000 h. This work highlights the results of these long-term tests and compares the observations for different material combinations, operating temperatures of 700 and 800 °C, including different fuel utilizations and gas compositions. An increase of temperature from 700 to 800 °C leads to an acceleration of the degradation rate by a factor of 1.5–2. Meanwhile, an increase in fuel utilization from 40 to 80% did not result in increased degradation. The same was found for higher current densities of up to 1 Acm−2.

Journal of the Electrochemical Society : JES 167(6), 061502 (2020). doi:10.1149/1945-7111/ab7d2e, Published by IOP Publishimg, Bristol

The centrifugal particle receiver is a novel concept proposed for concentrated solar power plants (CSP) to increase their operating temperature and efficiency. In this concept solar radiation is directly absorbed by a layer of ceramic particles held at the inner surface of a rotating cylindrical receiver by the centrifugal force. During operation, the hot ceramic particles (up to 1000 °C) move slowly along the receiver wall as well as other system components (e.g. tubes), which leads to their degradation through high-temperature oxidation and erosion. In the present study, a series of high temperature erosion-oxidation exposures was undertaken to experimentally evaluate performance of selected candidate metallic materials for centrifugal particle receivers. The exposures were conducted in a laboratory test facility consisting of a resistance heated furnace filled with ceramic particles, in which the specimen holder was rotated. Typical high temperature materials, such as martensitic, ferritic and austenitic stainless steels, Ni-base and Co-base alloys were investigated. The specimens were discontinuously exposed at 400–750 °C for up to 500 h and further characterized by scanning electron microscopy (SEM) and energy/wavelength dispersive x-ray spectroscopy (EDX/WDX).

Using alloy 230 as an example, a generalized oxidation lifetime model for chromia-forming Ni-base wrought alloys is proposed, which captures the most important damaging oxidation effects relevant for component design: wall thickness loss, scale spallation, and the occurrence of breakaway oxidation. For deriving input parameters and for verification of the model approach, alloy 230 specimens with different thicknesses were exposed for different times at temperatures in the range 950–1050 °C in static air. The studies focused on thin specimens (0.2–0.5 mm) to obtain data for critical subscale depletion processes resulting in breakaway oxidation within reasonably achievable test times up to 3000 h. The oxidation kinetics and oxidation-induced subscale microstructural changes were determined by combining gravimetric data with results from scanning electron microscopy with energy dispersive X-ray spectroscopy. The modeling of the scale spallation and re-formation was based on the NASA cyclic oxidation spallation program, while a new model was developed to describe accelerated oxidation occurring after longer exposure times in the thinnest specimens. The calculated oxidation data were combined with the reservoir model equation, by means of which the relation between the consumption and the remaining concentration of Cr in the alloy was established as a function of temperature and specimen thickness. Based on this approach, a generalized lifetime diagram is proposed, in which wall thickness loss is plotted as a function of time, initial specimen thickness, and temperature. The time to reach a critical Cr level at the scale/alloy interface of 10 wt% is also indicated in the diagrams.

Coupled thermodynamic and kinetic models rely on the thermodynamic and mobility databases which are compiled using critically assessed thermodynamic and diffusivity data acquired from various sources of experimental data. A continuous influx of experimental thermodynamic and kinetic data means that the respective databases have not only increased in complexity but also in size. The time and computational effort for the equilibrium calculations increases with increasing number of components and phases to be considered. In the present work, the applicability of a few methods was investigated to increase the computational efficiency of coupled thermodynamic and kinetic models. Three cases of varying complexities in terms of the number of phases, alloying elements and phenomena to be modelled were considered for demonstration. The distribution of the intensive thermodynamic calculations on multiple computing cores using MPI (Message Passing Interface) was undertaken. The interpolation scheme for dynamic storage of thermodynamic data available in the commercial software DICTRA was employed on a single computing core and the resulting performance was compared with the MPI computations. Additionally, the interpolation scheme was also parallelised to test its scaling capability in comparison to the computations performed solely with MPI. A linear scaling of computation speeds was observed with parallelisation of the thermodynamic calculations with MPI. However, the degree of scaling was dependent on the complexity of the calculation. The interpolation scheme on a single core in comparison with MPI on 48 cores was 20 times faster in one case but about 20–50 times slower in the other two cases. A parallelisation of the interpolation scheme improved its performance in the other two cases. However, the computational scaling was still poor compared to the MPI computations.

In memory of John Stringer (1934–2014), one of the leaders in studying the reactive element (RE) effects, this paper reviews the current status of understanding of the effect of RE dopants on high-temperature oxidation behavior, with an emphasis on recent research related to deploying alumina-forming alloys and coatings with optimal performance in commercial systems. In addition to the well-known interaction between indigenous sulfur and RE additions, effects have been observed with C, N, and O found in commercial alloys and coatings. While there are many similarities between alumina-forming alloys and coatings, the latter bring additional complicating factors such as the effects of O incorporation during thermal spraying MCrAlY coatings, coating roughness, and heat treatments that must be considered in optimizing the beneficial dopant addition. Analogies can be seen between RE effects in alloys and in the substrates beneath diffusion M–Al coatings. Recently, there has been more interest in the influence of mixed oxidant environments, since these may modify the manifestation of the RE effect. Finally, some thoughts are provided on optimizing the RE benefit and modeling oxidation of RE-doped alloys.

Numerical modelling of the degradation of P92 and VM12 martensitic stainless steels in gaseous (i) 70% CO 2 -30%H 2 O and (ii) Ar-50%CO 2 at 550 ° C is reported, emphasising the formation of duplex oxide scale and internal carburisation. A 1D numerical model is presented for the evolution of multiphase regions, subject to a Robin-type boundary condition representing a Boudouard surface reaction. The surface reaction kinetics at the spinel/substrate interface and the carbon transport kinetics in the substrate result in unusual carbon concentration profiles, measured here by glow-discharge optical emission spectroscopy (GDOES). Analysis of experimental data indicates that moisture decreases the rate of the surface reaction, despite higher carbon activity of the gas.

Ni-14Cr-9Co (wt.%) based alloys with additions of Al, Ti and Ta were exposed in dry and wet (+20 vol.%H2O) air at 950 °C for up to 100 h and characterized with GD-OES and SEM/EDX. Ni-14Cr-9Co-alloy exhibited initially Cr2O3-scale formation followed by breakaway Ni-oxidation. 3%Al addition to Ni-14Cr-9Co stabilized the Cr2O3-scale and largely suppressed Ni-oxide formation. Ni-14Cr-9Co-3Al-5Ti formed a fast growing Cr2O3-scale attributed to p-type Ti-doping, but Ni-oxide formation was completely inhibited. 5%Ta addition to Ni-14Cr-9Co-3Al promoted formation of a thin alumina scale. Addition of H2O to the test gas increased growth of Ni-rich oxides and shortened time to breakaway for the Al-free alloy.

The oxidation behaviour and microstructural evolution of aluminide coatings on the Ni base superalloys Rene 80 and alloy 625 were investigated during 1050 °C oxidation in air. Alumina based oxide scales formed on the coating surfaces of both alloys and only minor differences in scale growth and spallation kinetics were found. A coupled thermodynamic-kinetic model was employed to describe the effect of the base material on the depletion processes and microstructural changes in the aluminide coating thus providing a tool for coating life time prediction without very extensive test programs.

Single crystal Ni-base superalloys contain marginal amounts of carbon and as a consequence, diffusion coatings on these superalloys are generally expected to result in carbide free interdiffusion zones (IDZ) containing intermetallic precipitates. The experimentally observed presence of carbides in the IDZ of the aluminised (CVD) single crystal SC2000 led to a thorough investigation of the source of carbon as well as the precipitate phases occurring in the coating and in the IDZ. Scanning electron microscopy (SEM) combined with energy- and wavelength-dispersive X-ray spectroscopy (EDX/WDX) was employed to characterise the aluminised superalloy specimens. Average concentration profiles were determined by glow discharge optical emission spectroscopy (GDOES). Phases were then identified by X-Ray diffraction (XRD) and electron backscatter diffraction (EBSD). The carbides in the interdiffusion zone were found to be of the type M 6 C/M 12 C. By aluminising specimens of SC2000 with different thicknesses, the available carbon reservoir for the precipitation of carbides was varied. The amount of carbides in the IDZ after coating decreased with decreasing thickness of the substrate specimen. Hence it could be confirmed that carbon indeed diffused from the substrate to the coating/substrate interface and did not originate from the service environment.

A diffusion couple between directionally solidified nickel and the single crystal Ni-base superalloy CMSX-10 was produced by hot pressing in vacuum. The diffusion couples were heat treated at temperatures between 1050 and 1250 °C. The exposed samples were characterized by SEM/EBSD/EPMA. The interdiffusion results in dissolution of the γ′-Ni3Al in the superalloy and in growth of nickel grains towards CMSX-10. Rapid diffusion of aluminum from the superalloy into pure nickel leads to a significant formation of pores in the superalloy. The interdiffusion processes were modelled using the finite-element simulation software DICTRA with the databases TCNi5 and MobNi2, tailored specially for Ni-base superalloys. The effect of alloying elements on the interdiffusion profiles is discussed in terms of alloy thermodynamics. The calculated element concentration profiles are in good agreement with the EPMA measurements. The interdiffusion modeling correctly predicts the shapes of the concentration profiles, e.g. kinks on the Al and Ti profiles in the vicinity of the original interface in the joint. The calculation predicts with reasonable accuracy the extent and the location of the Kirkendall porosity.

The investigation aims at obtaining a life time extension of FeCrAlY heating elements by increasing the emissivity of the alumina surface scales. This approach will allow the use of a lower heating element temperature without decreasing the usable heat output. For this purpose, the oxidation mechanisms of Zr-doped FeCrAlY model alloys were investigated. For thick specimens a high Zr addition is accompanied by an undesired rapid increase of the oxidation rate adversely affecting the life time of the component. However, for thin foil heating elements (typically 50–200 μm thickness) an optimized high Zr content might be a suitable concept for obtaining a high emissivity, because apparently, a dark appearing, high emissivity oxide may be obtained without occurrence of dramatic internal oxidation.

The oxidation behavior of the nickel base Alloy 602 CA in atmospheres, relevant to gas separation units in fossil fired power plants, was investigated in the temperature range 800 to 1100 °C. Isothermal oxidation kinetics were determined in Ar–7 %H2O, Ar–4 %H2–7 %H2O, Ar–20 %O2–7 %H2O and, for comparative purposes, in Ar–20 %O2. The alloy formed an external alumina scale during oxidation at 800 °C, regardless of the atmosphere. Increasing the temperature results in gradual replacement of the alumina scale by an external chromia layer and internal alumina precipitates. The chromia growth rate is affected by the gas composition, due to titanium incorporation in the scale and outwardly protruding metallic nodules. The external alumina scale formation was strongly affected by surface cold work during specimen grinding. This effect, which promoted external alumina scale formation, was retained during exposure at 800 °C but was rapidly lost at higher temperatures resulting in external chromia scale formation and internal oxidation of aluminum.

An accurate procedure for predicting oxidation-induced damage and lifetime limits is crucial for the reliable operation of high-temperature metallic components in practical applications. In order to develop a predictive oxidation lifetime model for Ni–Cr alloys, specimens of wrought NiCrW alloy 230 with different thicknesses were cyclically oxidized in air at 950–1050°C for up to 3000 h. After prolonged exposure, two types of carbides as well as a Cr-rich nitride (π-phase) precipitated in the γ-Ni matrix. The oxidation-induced loss of Cr from the alloy resulted in the formation of subscale zones, which were free of the Cr-rich carbide and nitride but also of the Ni-W rich M6C. The width of the M6C-free zone was smaller than that free of the Cr-rich precipitates. Thermodynamic and diffusion calculations of the observed time- and temperature-dependent Cr depletion processes identified that back diffusion of C occurred which resulted in an increased volume fraction of M23C6 in the specimen core. With increasing time and temperature, the amount of π-phase in the specimen core increased. The subscale depletion of the initially present Cr-nitrides and the formation of Cr-nitrides in the specimen center is believed to be related to a mechanism which is qualitatively similar to that described for the Cr carbide enrichment. However, with increasing time and decreasing specimen thickness, N uptake from the atmosphere becomes apparent. As a result, the precipitates present in the specimen center eventually consisted almost exclusively of nitrides.

The isothermal and cyclic oxidation of a thin alloy 214 foil (130μm thick) has been studied in air at 1200°C for up to 720 h. Haynes 214 forms a duplex oxide scale consisting of pure, columnar alumina overlaid by mixed Ni(Al,Cr) 2 O 4 spinel with minor inclusions of (Al,Cr) 2 O 3 . A transition from external alumina to chromia scaling (abbreviated as TACS) occurs at the oxide-alloy interface after the Al reservoir of the foil is exhausted due to oxidation. The tensile cracking of the oxide scale upon thermal cycling is shown to promote the early transition from protective to accelerated breakaway oxidation. © 2015 Elsevier Ltd.

The high-temperature oxidation resistance of the carbide strengthened nickel-base alloy 602 CA is provided by the formation of a protective outer Cr 2 O 3 scale accompanied by internal precipitation of alumina and dissolution of Cr-rich carbides. The carbide free zone formation has been discussed in the literature to be a consequence of either chromium depletion or alloy decarburization. A new dissolution mechanism is proposed in this paper. Carbide dissolution in alloy 602 CA exposed to oxidizing atmospheres at 1100 °C was studied. The alloy microstructure was characterized by means of SEM/EDX/EPMA and GDOES. The experimental observations were substantiated by CALPHAD based computations.

A binary Fe–30 wt.%Cr alloy and corresponding ternary alloys containing manganese, molybdenum or tungsten were studied with respect to σ-phase formation at 650 °C. Although even after 3000 h exposure complete equilibration was not attained, the presence of tungsten and especially molybdenum was found to promote σ-phase formation. More extensive σ-phase formation was observed in the tungsten and especially in the molybdenum-containing alloys than in the binary and manganese-containing alloy. Apparently the bulk free energy decrease driving the nucleation of σ-phase is substantially larger when tungsten or molybdenum are present in the alloy. The presence of a nickel layer, to simulate the contact between ferritic steel interconnects and nickel mesh in a Solid Oxide Fuel Cell (SOFC) results in the formation of an austenitic zone and in accelerated formation of a σ-phase rich layer at the ferrite/austenite interface, due to interdiffusion processes. This interface acts as a highly efficient heterogeneity for the nucleation of σ-phase. The nucleation is enhanced by an increased Cr/Fe-ratio at that interface. Several possible modes for the growth of the σ layer were identified but the available experimental data were not sufficient to distinguish among these. The σ-rich layer, which appears to act as an interdiffusion barrier, is thicker in the case of the binary Fe–Cr and the Fe–Cr–Mn alloy than for the molybdenum- or tungsten-rich alloys. The results show that the stability range of σ-phase is larger than indicated by the presently used thermodynamic data bases.

A high-temperature rechargeable oxide battery (ROB) comprises a regenerative solid oxide cell (SOC) and oxygen-ion storage that consists of a porous Fe 2 O 3 base redox material. This material possesses good redox kinetics, a high oxygen-ion storage capacity, and an acceptable long-term stability. Yet, observations demonstrate degradation effects such as particle coarsening and an outward diffusion of iron leading to layer formation during operation of the ROB at 800 °C. To clarify the influence of the material composition on degradation, various oxides were added as a stabilizing scaffold for the Fe 2 O 3 base material. Pressed samples of the binary mixtures were sintered in air at 900 °C and subsequently redox-treated up to 20 times under conditions that simulate those present in an actual ROB (800 °C, Ar–2%H 2 or Ar–7%H 2 O–2%H 2 ). Afterwards, the degradation properties were analyzed by laser microscopy and the phase composition was measured using X-ray diffraction. Results indicate that the addition of yttria-stabilized zirconia (8YSZ) or pure zirconia (ZrO 2 ) can suppress structural degradation thus maintaining reaction kinetics. In contrast, the use of yttria (Y 2 O 3 ) does not significantly mitigate degradation phenomena. Consequently, storage components consisting of 8YSZ and Fe 2 O 3 were employed in an ROB test, resulting in more than 200 cycles with a current density of 150 mA/cm 2 and cycle durations of up to 70 min.

Ausgewahlte austenitische Stahle und Nickelbasislegierungen wurden im Hinblick auf ihr Korrosionsverhalten bei 700 °C in einem H2-H2O-H2S-Modellgas untersucht. Werkstoffe, die unter diesen Bedingungen die geringsten Korrosionsraten aufwiesen, wurden detailliert bezuglich der Temperaturabhangigkeit der Korrosionsraten bei 500–700 °C erprobt. Dabei zeigten Alloy 693, 625 und 740 eine anomale Temperaturabhangigkeit der Korrosionsraten, wahrend alloy HR-160 bei allen Temperaturen sehr geringe Korrosionsraten aufwies. Bei Grundlagenuntersuchungen an Modelllegierungen zeigte sich, dass hohe Co-Zusatze zu binaren NiCr-Legierungen die Korrosionsraten deutlich reduzieren und bei 650 sowie 700 °C zur Bildung von Cr-Oxidschichten fuhrten. Zusatze von Ti oder Si in einer NiCrCo-Modelllegierung forderten die Cr-Oxidbildung bei niedrigen Temperaturen (500–600 °C). Auch Zusatze von Nb zu einer binaren NiCr-Modelllegierung bewirkten im Gegensatz zu Zusatzen von Mo eine Reduzierung der Korrosionsraten durch Bildung Nb-reicher intermetallischer Phasen an der Schicht/Legierungs-Grenzflache. Allerdings trat dieser positive Effekt lediglich bei den hochsten Versuchstemperaturen (650 und 700 °C) auf.

MCrAlY (M = Ni, Co) coatings are commonly used to protect the underlying superalloy component in industrial gas turbines and aeroengines from oxidation attack. They also function as bond coats for thermal barrier coatings (TBC). The chemical life time of the coated component is mainly governed by the depletion of the β-phase in the bond coat as a result of simultaneously occurring surface oxidation and interdiffusion between the coating and the substrate. A new computational approach to model the microstructural evolution in MCrAlY (M = Ni, Co) coatings on Ni base superalloys was undertaken in the present study. Scanning electron microscopy (SEM), energy dispersive X-ray spectroscopy (EDX) and wavelength dispersive X-ray spectroscopy (WDX) was employed to characterise MCrAlY coated superalloy samples after exposure at 1000 and 1100°C. Phases were identified by electron backscatter diffraction (EBSD) and correlated with SEM/EDX/WDX analyses.Modelling of the microstructural evolution was carried out...

The high chromium nickel base alloy 602 CA is used as construction material for high temperature components in aggressive environments such as gas burners or heat exchangers. The high temperature oxidation behaviour of this alloy is characterised by the formation of an external Cr2O3 scale and internal precipitation of alumina. The main goal of the present investigation was to elucidate the effect of aluminum on the subscale chromium depletion phenomena. The chromium concentration profiles were found to be flat and sometimes even increasing towards the oxide-alloy interface. The experimental observations were interpreted in terms of alloy thermodynamics, i.e. effect of oxidation induced compositional changes on chemical potentials of the alloy constituents. The concentration profiles in alloy 602 CA observed during high temperature exposure were modelled using the CALPHAD approach for thermodynamic and kinetic calculations. Thermodynamic calculations revealed that the Al depletion in the alloy subsurface ...

In the present work it is shown that the lifetime of air plasma sprayed thermal barrier coatings systems with high velocity oxyfuel bondcoats can be substantially extended by application of a thin air plasma sprayed “flashcoat” layer onto the base, high velocity oxyfuel bondcoat. This approach allows improvement of the bondcoat roughness profile to the extent typically obtained by optimized vacuum plasma spraying. The oxidation resistance of the flashcoat strongly depends on the spray parameters used. Deviation from the set of the optimized spray parameters was found to result in early catastrophic oxidation of the flashcoat and corresponding very rapid thermal barrier coating-failure. In order to explain the thermal barrier coating lifetime variations between the various bondcoat morphologies, the roughness profiles of high velocity oxyfuel, vacuum plasma sprayed and air plasma sprayed bondcoats were evaluated using fractal analysis. It is suggested that such an approach provides better correlation between the bondcoat morphology and thermal barrier coating-lifetime than the calculation of the arithmetic mean roughness ( R a ) of the bondcoat frequently used in industrial practice.

In the present study the long-term behaviour of two ferritic steels, Crofer 22 APU and Crofer 22H, in contact with a Ni-mesh during exposure in simulated anode gas, Ar–4%H 2 –2%H 2 O, at 700 and 800 °C for exposure times up to 3000 h was investigated. Ni diffusion from the Ni-mesh into the steel resulted in the formation of an austenitic zone whereas diffusion of iron and chromium from the steel into the Ni-mesh resulted in the formation of chromia base oxides in the Ni-mesh. Depending on the chemical composition of the steel, the temperature and the exposure time, interdiffusion processes between ferritic steel and Ni-mesh also resulted in σ -phase formation at the austenite–ferrite interface and in Laves-phase dissolution in the austenitic zone. The extent and morphology of the σ -phase formation are discussed on the basis of thermodynamic considerations, including reaction paths in the ternary alloy system Fe–Ni–Cr.

Two martensitic steels were reacted with Ar–50%CO 2 at 550 °C for exposure times up to 150 h. Microstructural analyses and glow discharge optical emission spectroscopy revealed that internal carburisation of both steels beneath external oxide scales occurred, in spite of the very low equilibrium carbon activity of the atmosphere. The carbon concentration at the metal–scale interface increased throughout the reaction and the carbon uptake varied approximately linear with time. A model is proposed to describe the non steady-state carburisation kinetics whereby the usual diffusion equation describing carbon movement into the alloy is modified to account for loss of carbon as precipitated carbides.

The aim of the present study is the investigation of the microstructural features and the oxidation resistance of the austenitic steel Sanicro 25 during long-term steam exposure up to 10 000 h in the temperature range 600–750 °C. Steel microstructures of specimens after exposure times of 1000, 3000, and 10 000 h were analyzed by scanning and transmission electron microscopy as well as X-ray diffraction. The steam oxidation behavior was estimated by gravimetry in combination with microstructural investigations using the above mentioned analysis tools as well as depth profiling by secondary neutrals mass spectrometry. In the as-received condition, the only detectable precipitate was Z-(Nb,Cr)N phase. After exposures in the temperature range 600–750 °C various amounts of the following additional phases were identified: chromium nitride, µ-phase, and carbides with various compositions. Chromia base surface scales formed during exposure at 600–700 °C, whereas more rapidly growing Fe/Cr-rich oxides were found at 750 °C. The oxidation rates were only slightly higher than those of a typical nickel base alloy such as alloy 617. In the subsurface zones, depletion of scale forming elements was accompanied by enrichment of tungsten and copper-rich phases.

Metallische Werkstoffe fur hochbeanspruchte Sauerstoffdusen und Abhitzekessel wurden auf ihre Korrosionsbestandigkeit in einem H2/H2O/H2S-Laborgas untersucht. In Screening-Versuchen an ferritischen und austenitischen Stahlen und Nickelbasislegierungen wurden vielversprechende Legierungen ermittelt. Fur alle untersuchten Werkstoffe, mit einer Ausnahme, ergab sich eine Bell-Shape-Abhangigkeit der Oxidationskinetik. Durch Untersuchungen an Modelllegierungen wird die Wirkung einzelner Legierungselemente auf die Korrosionsmechanismen ermittelt, um hieraus optimierte Legierungszusammensetzungen herzuleiten. Metallic materials for highly stressed oxygen nozzles and recovery boilers were studied in respect to their corrosion resistance in H2/H2O/H2S laboratory test gas. Screening tests using ferritic and austenitic steels and nickel-base alloys were carried out and from the results promising alloys selected and tested at various temperatures. With one exception, all alloys show bell-shape temperature dependence of the oxidation kinetics. Results from studies on model alloys will be used to derive optimised alloy compositions which exhibit protective oxide scale formation in the whole temperature range considered in the present studies.

Nickel-based alloys are being considered as construction materials for various components in high-efficiency steam turbines with envisaged operating temperatures around 700 °C. In the present study, the steam oxidation behaviour of the nickel-based alloy 625 in the temperature range of 700–800 °C was investigated whereby exposures up to 10000 h were carried out. Gravimetric data in combination with results from a variety of post exposure analysis techniques showed in all cases the formation of protective oxide scales mainly consisting of chromia with minor amounts of outer Cr/Mn spinel and internal silica. The phases found in the bulk alloy after long-term exposure were mainly needle-shaped δ-Ni3(Nb,Mo) phase, μ-phase and Si-rich η-M6C carbide. Microstructural features and phase formation were found to be related to minor variations in the alloy composition, especially iron and silicon content. The oxidation-induced chromium depletion caused a number of microstructural changes in the subsurface depletion layer. Most important was an enrichment of the intermetallic δ-phase at the scale–alloy interface. DICTRA modelling revealed this effect to be caused by uphill diffusion as a result of a negative niobium activity gradient in the subscale chromium depletion zone. Although the available kinetic and thermodynamic data allowed qualitative explanation of the δ-phase enrichment, the databases do not correctly describe the high molybdenum solubility in the δ-phase.

Microstructural features, mechanical and steam oxidation properties of ferritic 18–23 wt% chromium model steels, strengthened by the precipitation of intermetallic phases, were investigated. The behavior of the commercially available ferritic steel Crofer® 22 H [1] , [2] , [3] , [4] , originally developed for the application in automotive solid oxide fuel cell (SOFC) powered auxiliary power units (APUs) [5] , [6] , [7] , was compared with that of optimized model alloys with systematic variations in Cr, Nb, W and Ti content. The main aim of the studies was the estimation of the potential suitability of these steels as construction materials in power plants with live steam temperatures around 600 °C. Some of the optimized trial alloys demonstrated remarkable characteristics at 600 °C. Results of constant load creep experiments and steam oxidation tests indicate the potential of these alloys as candidate materials for application in highly efficient steam power plants with steam temperatures above 620 °C. At high stress levels some of the presented alloys, especially those containing enhanced levels of W and Nb, even exhibit higher creep strength than standard ferritic–martensitic steel P92. The amount of strengthening (Fe,Si,Cr)2(W,Nb) – Laves phase particles is found to correlate well with the level of W- and Nb-alloying and in contrast to ferritic–martensitic steels, the formed Laves phase particles are highly stable at elevated temperature. Owing to the high contents of chromium, the model alloys have excellent steam oxidation resistance. A way to preserve favorable steam oxidation properties and diminish undesirable (Fe,Cr)-σ-phase formation is outlined.

Modern gas turbine materials are subjected to increasing operating temperatures in service environments, which contain high concentrations of gaseous species originating from fuel impurities, e.g. sulfur. In the present study, the oxidation behavior of two commercially available nickel-base alloys, Rene 80 and PWA 1483, was investigated for up to 500 h duration in synthetic air and synthetic air containing 2% SO 2 at 1050 °C. In order to investigate the mechanisms of SO 2 attack, specimens of the two alloys after different oxidation times were characterized using a number of surface analytical techniques such as scanning electron microscopy, XRD, and glow discharge optical emission spectrometry. The corrosion reactions in the presence of SO 2 were found to be strongly alloy dependent and could not be explained simply in terms of the contents of the main scale-forming alloying elements, Cr and Al. Severe internal sulfidation was observed underneath the chromium-rich oxide scale on Rene 80, leading to breakaway oxidation after longer exposure times. For PWA 1483 sub-scale formation of a near-continuous alumina occurred, which resulted in substantial suppression of sulfidation. Consequently, no breakaway oxidation of PWA 1483 was observed up to 500 h exposure at 1050 °C in synthetic air with 2% SO 2 . The formation of a dense alumina sub-scale on PWA 1483 is believed to be correlated with the Ta-addition in this material. It is proposed thatTa can tie up Ti by forming a mixed oxide of the type TiTaO 4 . In this way, it prevents the incorporation of Ti into the chromia scale, which can lead to enhanced scale growth.

This report is a description of research performed by the Julich Research Centre. Their conclusions outline the effects of SO2 additions on oxide formations and the interaction of other gases in oxvyfuel environments.

The SOFC group at JÜLICH works on the development of materials, cell and stack since the early 90s, strongly supported by post-operational analysis. Important achievements are e.g. the development of ferritic chromium steels which are now commercially available and anode substrate cells with power densities exceeding 2 W/cm² at 0.7 V and 700 °C. Special highlights are long term tests with short stacks of which one has reached 50,000 h of continuous operation. In the year 2000 also systems and system components development has been started. In 2012 a 20 kW system, based on four 5 kW stacks, was successfully set into operation. The envisaged 20 kW were achieved and the system has been operated for about 4,000 h up to now.

In new and future generation power plants, creep-resistant steels selected for high-temperature components are subjected to several service environments. In many of these applications, water vapor has been found to substantially alter the technologically relevant properties, mainly growth rate and adherence, of the surface oxides. Various data sets exist in the literature for steam oxidation for this type of steel; however, these data show a substantial variability. These differences can be related to, e.g., “batch-to-batch” variations, oxidation conditions, surface treatment, and differences between laboratory/power plants data. Therefore, these literature data need careful and critical verification when being used to estimate the long-term materials behavior in real power generation plants. The paper discusses the existing steam oxidation data for the martensitic steels P91 (9%Cr-V-Mo), P92 (9%Cr-W-V-Mo), and VM12 (12%Cr-Co-W-V-Mo), with the main emphasis on discussing the parameters that are responsible for the observed variations in oxidation rates for the various types of materials. Differences in minor alloying additions and surface treatment have as a result that for the 9%Cr steels the scale thickness after a given oxidation time may easily differ by a factor of two. Because of the “batch-to-batch variation” in oxidation behavior, the common procedure used to determine service temperatures of tubes made of low-Cr steels on the basis of measured oxide layer thicknesses, has to be used with some care in case of the 9% to 12%Cr martensitic steels. This is especially the case for the 12%Cr steels because their oxidation rates between 500°C and 650°C may not steadily increase with increasing temperature.

The effect of oxygen content in NiCoCrAlY bondcoat on the cyclic oxidation lifetimes of EB-PVD and APS thermal barrier coatings (TBC) has been studied. The EB-PVD TBC system with an oxygen content of 0.05 wt. % in NiCoCrAlY bondcoat shows five times longer lifetimes compared to a TBC system with 0.2 wt% oxygen in the bondcoat. In the bondcoat with the high oxygen content the minor (0.3 wt.%) yttrium addition was found to be tied up by oxygen into fine precipitates of yttrium aluminates. Thereby the beneficial effect of yttrium onto the adherence of the alumina scale was significantly reduced. The critical scale thickness at failure was by about a factor of two lower for the high oxygen bondcoat than for the low oxygen bondcoat. In contrast to EB-PVD TBC systems, no detrimental effect of increasing oxygen content on the lifetime of APS-TBC systems was observed. This can be explained by a different failure mechanism of APS-TBC systems, whereby the lifetime is mainly determined by the rate of crack propagation through the ceramic topcoat.

Previous studies revealed that initial sample thickness affects the growth rate of oxide scales formed during 800 or 900 °C air exposure. The effect is partially related to differences in depletion of minor alloying additions such as Mn, Ti, La in thick and thin specimens. However, it has previously been proposed that the specimen thickness dependence is partially governed by differences in creep strength of thick and thin substrates. To investigate this hypothesis, discontinuous air oxidation experiments were carried out with the Laves phase strengthened ferritic steel Crofer 22 H at 800 °C. Comparing the data for solution annealed and pre-aged (500 h, 900 °C) materials it could be shown that intrinsic creep strength of the alloy substantially affects oxidation rates. The observations can qualitatively be explained by assuming the relaxation of oxide growth stresses by plastic deformation of the metallic substrate to be an important parameter affecting the kinetics of oxide scale growth.