Use of oxygen isotope to study the transport mechanism during high temperature oxide scale growth

Abstract

The examination of high temperature (HT) oxide scale growth mechanisms was performed using secondary ion mass spectrometry (SIMS) and secondary neutral mass spectrometry (SNMS), in conjunction with ¹⁶O₂/¹⁸O₂ HT oxidation experiments. Cr₂O₃, NiO, ZrO₂ and Al₂O₃ were studied because they constitute excellent representative thermally grown oxide scales: they grow by cationic diffusion (Cr₂O₃, NiO), anionic diffusion (ZrO₂) or mixed anionic-cationic diffusion (Al₂O₃). The oxidation tests were performed first in ¹⁶O₂ and subsequently in ¹⁸O₂ at several temperatures (600–1000°C for NiO, 600°C for ZrO₂, 1000°C for Cr₂O₃ and 1100°C for Al₂O₃). The oxygen isotope distribution observed by SIMS and SNMS profiles are discussed and related with the HT oxidation mechanisms proposed in the literature.     

Growth mechanism of chromia on hydrogen annealed chromium using imaging SIMS and 18O as tracer

Abstract

Chromium specimens were annealed in hydrogen to reduce the level of sulphur impurity and thus improve the oxide adhesion. The growth mechanism of the chromia scale in 0·1 atm oxygen at 950°C was investigated by oxidising sequentially in natural oxygen and then in gas enriched in ¹⁸O; the distribution of ¹⁸O in the scale was determined using imaging secondary ion mass spectroscopy (SIMS). From preliminary results it was found that substantial oxygen diffusion occurred. This differs from the growth mechanism of chromia on vacuum annealed material which, as far as solid state transport is concerned, mainly involves cation transport. The theory that sulphur promotes cation diffusion via grain boundaries in chromia and that removal of the sulphur favours oxygen diffusion is supported by the results. From examination of the distribution of the oxygen isotopes at the scale/gas interface, it is shown that their distribution is not uniform, from which it can be inferred that the growth of oxide at that interface was not uniform during oxidation in the ¹⁸O-rich gas.

MST/967     

Enhanced oxygen exchange near the oxide/silicon interface during silicon thermal oxidation

The effect of the SiO₂/Si interface on oxygen exchange diffusion during oxidation was investigated using oxygen isotopes. A 40-nm thick Si¹⁸O₂ layer was first grown in ¹⁸O₂ and then the sample was reoxidized in ¹⁶O₂ at 1100 °C. The ¹⁸O diffusion in Si¹⁶O₂ during the ¹⁶O₂ oxidation was investigated by secondary ion mass spectrometry measurements. A significant broadening of the ¹⁸O profile toward the newly grown Si¹⁶O₂ was observed. The average oxygen diffusivity was initially about one order of magnitude larger than the reported thermal diffusivity in the SiO₂ network. In addition, the ¹⁸O diffusion became slower with oxidation time and hence with increasing distance between ¹⁸O diffusion region and the interface. This distance-dependent ¹⁸O self-diffusion was simulated taking into account the effect of Si¹⁶O molecules generated at the interface upon oxidation and diffusing into SiO₂ to enhance the oxygen exchange. The simulation fits the SIMS profiles and shows that the SiO diffusion is greatly retarded by the oxidation with O₂ from the oxygen-containing atmosphere. Therefore, the Si¹⁶O concentration becomes smaller as the interface leaves the ¹⁸O region and the oxygen exchange becomes slower with time.     

Double perovskite cathodes for proton-conducting ceramic fuel cells: are they triple mixed ionic electronic conductors?

Abstract

¹⁸O and ²H diffusion has been investigated at a temperature of 300 °C in the double perovskite material PrBaCo₂O_5+δ (PBCO) in flowing air containing 200 mbar of ²H₂¹⁶O. Secondary ion mass spectrometry (SIMS) depth profiling of exchanged ceramics has shown PBCO still retains significant oxygen diffusivity (~1.3 × 10^?11 cm²s^?1) at this temperature and that the presence of water (²H₂¹⁶O), gives rise to an enhancement of the surface exchange rate over that in pure oxygen by a factor of ~3. The ²H distribution, as inferred from the ²H₂¹⁶O^? SIMS signal, shows an apparent depth profile which could be interpreted as ²H diffusion. However, examination of the 3-D distribution of the signal shows it to be nonhomogeneous and probably related to the presence of hydrated layers in the interior walls of pores and is not due to proton diffusion. This suggests that PBCO acts mainly as an oxygen ion mixed conductor when used in PCFC devices, although the presence of a small amount of protonic conductivity cannot be discounted in these materials.     

Effects of a Reactive Element on isothermal and cyclic oxidation of chromia-forming alloys: SEM/EDX,TEM and SIMS investigations.

Abstract

Scanning electron microscopy (SEM), transmission electron microscopy (TEM), secondary ion mass spectrometry (SIMS), secondary neutral mass spectrometry (SNMS) and X-ray diffraction (XRD) were used to study the reactive element effect on chromia-forming alloys. The reactive element, neodymium, was introduced as an oxide film at the surface of the alloys. The analyses were performed during the early stages of oxidation at 1,273 K. Uncoated and Nd₂O₃-coated alloys have been oxidised for 1, 5, 30, 60, 120 minutes and 50 hours in air at atmospheric pressure. Chromia growth mechanisms were studied by two-stage ¹⁶O₂/¹⁸O₂ oxidation exposures followed by SIMS and SNMS analyses. Chromia grains quickly grew on uncoated samples, whereas they slowly developed on Nd₂O₃-coated specimens. A neodymium-containing phase rapidly evolved from Nd₂O₃, to NdCrO₃ and then NdTi₂₁O₃₈. Indeed, the main phase evolution appeared during the first 60 minutes of the oxidation process. Chromia growth mechanism was not changed after 1 hour of oxidation because Nd was not yet incorporated into Cr₂O₃ scales. During the early stages of oxidation, Nd was mainly concentrated in the outer part of the scale composed of a spinel phase, Mn_1.5Cr_1.5O₄. After two hours of oxidation, Nd was incorporated inside the chromia scale, leading to inward diffusion of oxygen. These results clearly demonstrated that the incorporation of the reactive element in the chromia scale as grain boundary segregant is the main explanation of the reactive element effect in the case of chromia-forming alloys.     

Relation between the oxidation growth rate of chromia scales and self-diffusion in Cr2O3

In most cases, chromia scales are assumed to grow by predominant chromium diffusion. However, results of Atkinson and Taylor indicated that chromium bulk diffusion could not account for the growth rate of chromia scales. Moreover, recent results of Parket al. showed that oxygen diffusion in chromia was faster than chromium diffusion. So, at this date, the controlling process of the growth of chromia scales is not elucidated.To interpret such a phenomenon, oxygen and chromium self-diffusion coefficients in Cr₂O₃ single crystals and polycrystals were determined in the same materials and in the same experimental conditions, thus allowing a direct comparison. Tracers were introduced by ion implantation, thick film methods, and isotopic exchange, using the⁵⁴Cr,⁵⁰Cr and¹⁸O isotopes. Depth profiling was made by secondary ion mass spectroscopy (SIMS). The bulk diffusion coefficients were computed by using a general solution of the Fick's law taking into account evaporation and exchange at the surface. Grain-boundary diffusion coefficients were computed by using the Whipple-Le Claire equation for type B intergranular diffusion. Lattice and grain boundary self-diffusien coefficients were determined as a function of temperature and oxygen pressure.The diffusion coefficients are lower than results given in the literature and do not depend on the oxygen pressure. Moreover, it is found that oxygen diffusion is faster than chromium diffusion. These results are compared to the oxidation constants of chromia-forming alloys and it is shown that neither lattice self-diffusion, nor grain-boundary self-diffusion can justify the growth rate of chromia scales. Such a situation is compared to NiO case, for which authors found important differences in grain-boundary diffusivity, according to the elaboration mode of NiO (thermal oxidation or growth from the melt).     

Oxidation and diffusion study on AlCrVN hard coatings using oxygen isotopes O and O

The present work aims to investigate the oxidation behaviour of AlCrVN hard coatings of equal composition but of different crystal structure. In order to gain more information about the mechanisms that are active during oxidation, a two-stage oxidation procedure has been applied where different isotopes, ¹⁶O and ¹⁸O, were introduced in each step. The analysis by means of secondary ion mass spectrometry depth profiling with its inherent isotope selectivity provided information on the general oxidation behaviour as well as the oxygen diffusion during the oxidation process. The single-phase coating with its face-centred cubic (fcc) structure presents a higher oxidation resistance as compared to the dual-phase coating containing a wurtzite and an fcc phase. After the annealing treatment the surface of the latter is entirely covered by VO₂ and V₂O₅ as evidenced by Raman spectroscopy. The single-phase coating, on the other hand, reveals unoxidised coating material and AlVO₄ crystals. However, even though exhibiting a significantly different oxidation resistance, the oxygen diffusion is similar. In both cases the peak values of ¹⁸O, which was introduced in the second stage, were found near the oxide-nitride interface indicating that O atoms diffused through the already formed oxides. Additional experiments using a gas mixture comprising natural water vapour H₂¹⁶O and ¹⁸O₂ revealed that mainly the presence of molecular oxygen causes oxidation as with increasing water vapour partial pressure the oxide layer thickness was significantly reduced.     

Breakdown of the protective oxide on 11 % Cr steel at high temperature in the presence of water vapor and oxygen,the influence of chromium vaporization

Abstract

We report on the effect of water vapor and oxygen on the oxidation of a ferritic/martensitic 11 % Cr steel (CrMoV11 1). The influence of pH₂O, exposure time, gas velocity and temperature was investigated. The samples were exposed to dry O₂, O₂+10 or 40 % H₂O for up to 336 hours. Total pressure was 1 atm (1.02 × 10⁵ Pa). The gas velocity was between 0.05 and 10 cm/s while temperature was in the range 450–700°C. The samples are investigated by thermogravimetry, GI-XRD, SEM/EDX, GDOES, FIB and TEM/EDX. Oxidation is strongly affected by the vaporization of CrO₂(OH)₂ in H₂O/O₂ environment. The mechanism of vaporization of CrO₂(OH)₂ from a Cr₂O₃ surface is modelled by DFT calculations. In the absence of chromium vaporization the alloy forms a protective oxide consisting of a corundum-type solid solution (Fe_1–xCr_x)₂O₃. The vaporization of chromium tends to deplete the oxide in chromium. In some cases the oxide remains protective in spite of chromium depletion while in other cases there is a transition to breakaway oxidation. In the latter case a thick layered scale forms, consisting of an outer hematite part and an inner iron-chromium spinel. Oxidation behavior in an O₂+H₂O environment is to a large extent determined by the ability of the metallic substrate to supply the oxide with chromium by diffusion in order to compensate for the losses by vaporization. The corrosivity of the environment increases with the concentration of water vapor and oxygen, with the gas velocity and with temperature.     

Growth mechanisms of oxides formed on Fe–9Cr steel in high pressure carbon dioxide studied using imaging SIMS

Abstract

Specimens of Fe–9Cr steel were oxidised sequentially, at 4·1 MPa pressure, in natural carbon dioxide and then in carbon dioxide enriched in ¹⁸O. The specimens were cross-sectioned and the distribution of the ¹⁸O was determined using imaging secondary ion mass spectroscopy (SIMS). Scales developed at the oxide/gas interface, at the oxide/metal interface, and within the oxide layer. When internal oxidation occurred, growth took place at the interface between the internally oxidised zone and the metal. It is shown that imaging SIMS, when used in conjunction with ¹⁸O as a tracer, is a powerful technique for studying oxide growth mechanisms.

MST/969     

The effect of gas flow rate on the evolution of the surface oxide on a molten low carbon Al killed steel

The oxide phase formation on a molten Al killed low carbon steel surface under a flowing Ar atmosphere with an oxygen partial pressure of Po ₂ = 1–5 × 10^?5 atm has been visualized with a Confocal Scanning Laser Microscope (CSLM) equipped with a gold image furnace. In this study, the effect of gas flow rate variation (170–300 cm³/min) on the oxide evolution under isothermal conditions of 1600^°C was investigated. Al₂O₃, rather than the thermodynamically stable phase FeAl₂O₄, was found to precipitate under all the experimental conditions studied and the apparent rate of evolution was found to increase with increasing gas flow rate. The oxide evolved as a network that started from the container wall and grew towards the crucible center. At low flow rates the growth was a result of primarily crystal growth resulting in distinctly dendritic crystals. As the flow rate was increased, growth due to the attachment of discrete inclusions to the advancing front was observed which resulted in a final oxide network that constituted of smaller facetted particles. In the latter case, the transport of the individual inclusions to the advancing front could be caused by surface Marangoni flow due to gradients in both temperature and dissolved oxygen concentration.     

Composition profiles of silicon-silicon oxide and silicon-rare earth oxide structures

Elemental composition profiles of the structures comprising the layers of dysprosium oxide (Dy₂O₃), lutetium oxide (Lu₂O₃), or silicon oxide (SiO₂ with Dy and Lu impurities) on silicon substrates have been studied by means of Rutherford backscattering spectrometry and nuclear reaction analysis. The structures were prepared by high-temperature diffusion in oxygen or air. The results of depth profiling are compared to the concentrations of electrically active impurities in the diffusion layers on silicon, which were determined from measurements of the capacitance-voltage characteristics and the surface resistance and from the data of transmission electron microscopy and X-ray diffraction.     

SIMS studies of scale growth processes

Abstract

The penetration of ¹⁸O tracer into single layered and duplex NiO scales formed on Ni has been examined by sputter depth profiling and imaging modes of secondary ion mass spectrometry (SIMS). Preoxidation annealing in hydrogen and superficial contamination during substrate surface preparation were found to promote inward scale growth. The principles for selecting SIMS techniques for studying oxide scales and the influence of substrate condition on mechanisms for gaseous oxygen transport in NiO are discussed.

MST/950     

Oxygen gas-sensing behaviour of V2O5–SnO–TeO2 glass

The d.c. conductivity, σ, and the oxygen gas-sensing behaviour of V₂O₅–SnO–TeO₂ glass prepared by press-quenching were studied in argon and oxygen gas atmospheres at temperatures ranging from 303–473 K. The glass of 50V₂O₅·20SnO·30TeO₂ (mol %) was n-type semiconducting. The high-temperature conductivity was lower in oxygen and higher in argon than that in air. This was explained by the V⁴⁺ ions in the glass being oxidized by oxygen which had diffused into the glass, resulting in an increase in V⁵⁺ with time. The experimental relationship between σ and oxygen partial pressure, P _O2, agreed quantitatively with the theoretical relation σ ∝ P _O2 ^-1/4 . Changes in conductivity by switching the atmospheres between oxygen and argon gases were found to be reproducible. From the data of these dynamic changes, the oxygen gas sensitivity, S, at 473 K was obtained to be 1.3 in oxygen atmosphere. The dynamic changes could be quantitatively explained by an oxygen diffusion model. Throughout these discussions, the present tellurite glass was found to possess a potential applicability as an oxygen gas sensor. This revised version was published online in November 2006 with corrections to the Cover Date.     

TEM investigation of the oxide scales formed on a FeCrAlRE alloy (Kanthal AF) at 900°C in dry O2 and O2 with 40% H2O

Abstract

The base oxide scales on a commercial FeCrAl alloy oxidized isothermally at 900°C in dry O₂ or O₂ with 40% H₂O were studied in detail using analytical electron microscopy. Electron transparent cross-section foils prepared with a FIB/SEM in-situ lift-out technique were investigated using STEM/EDX and CBED. The oxide scales on the samples exposed to dry O₂ are slightly thinner than the scales formed in O₂+H₂O. The oxide scales exhibit a multilayered structure, with a Cr-rich layer in the middle, indicating the original metal/gas interface. An almost pure inner α-Al₂O₃ layer, containing columnar grains, was formed by inward oxygen diffusion, after exposures in both the dry and wet atmospheres. The outer oxide layer consisted of γ-Al₂O₃ in the wet case and of α-Al₂O₃/MgAl₂O₄ in the dry case. It is suggested that the α-Al₂O₃/MgAl₂O₄ phases resulted from a phase transformation of initially grown γ-Al₂O₃. The observations indicate that water vapour may stabilize the γ-Al₂O₃ phase.     

Correlation of electrical properties and internal friction in mixed conduction glasses containing ionic and electronic conduction (1) Fe2O3-Na2O-P2O5 glasses

The electrical properties and internal friction in (40–x)Fe₂O₃·xNa₂0.60P₂O₅ glasses were measured. Two or three peak on internal friction were observed in the temperature range of –100 to 300° C at a frequency of about 1 Hz. The peak area of internal friction could be explained quantitatively by the additivity law of diffusion of Na⁺ ion and hopping of electrons which are carriers similar to those of dielectric loss. Activation energy, peak temperature of dielectric loss and internal friction showed almost the same value. Both relaxation phenomena have the same mechanism which is due to the diffusion of Na⁺ion and the hopping of electrons between Fe²⁺ Fe³⁺. The high-temperature peak is assumed to result from the interaction between protons or alkali ions and non-bridging oxygen.     

Praseodymium compound formation in silicon by ion beam synthesis

The compound formation in the ternary system Pr-Si-O initiated by ion beam synthesis inside bulk-Si was studied by transmission electron microscopy and X-ray diffraction. The oxygen content was varied by additional O⁺ ion implantation and by oxidation of the bulk-Si. For annealing temperatures of 1100 °C, Pr silicate nanoclusters were observed consisting of Pr_9.33Si₆O₂₆ or Pr₂Si₂O₇. These silicates were the dominating and most stable Pr compounds. The interfaces between Pr silicate and the crystalline Si were atomically abrupt after high-temperature annealing. Pr silicide (PrSi₂) was detected for lower annealing temperatures such as 900 °C and for higher annealing temperatures in minor fraction also in samples with enhanced O content. Pr oxide (Pr₂O₃), the promising high-k material, was not definitely verified. In ion beam synthesis, the energy related to structural reordering during solid-state compound formation is a parameter that controls the proceeding processes in addition to other parameters like chemical reactivity and the compound interface matching.     

18O-tracer diffusion along nanoscaled Sc2O3/yttria stabilized zirconia (YSZ) multilayers: on the influence of strain

Abstract

The oxygen tracer diffusion coefficient describing transport along nano-/microscaled YSZ/Sc₂O₃ multilayers as a function of the thick­ness of the ion-conducting YSZ layers has been measured by isotope exchange depth profiling (IEDP), using secondary ion mass spec­trometry (SIMS). The multilayer samples were prepared by pulsed laser deposition (PLD) on (0001) Al₂O₃ single crystalline substrates. The values for the oxygen tracer diffusion coefficient were analyzed as a combination of contributions from bulk and interface contributions and compared with results from YSZ/Y₂O₃-multilayers with similar microstructure. Using the Nernst–Einstein equation as the relation between diffusivity and electrical conductivity we find very good agreement between conductivity and diffusion data, and we exclude substantial electronic conductivity in the multilayers. The effect of hetero-interface transport can be well explained by a simple interface strain model. As the multilayer samples consist of columnar film crystallites with a defined inter­face structure and texture, we also discuss the influence of this particular microstructure on the interfacial strain.     

Application of fracture mechanics tofailure analysis and to residual life assessment of components operating at high temperatures

Abstract

MCrAlY overlay coatings have been successfully used as a means of improving the oxidation performance of gas turbine blades operating at elevated temperatures. However, depletion of aluminium can limit the ability of such coatings to form a protective oxide layer should spallation of the original α-Al₂O₃ oxide layer occur under thermal cycling conditions. It is the objective of the current research to evaluate the potential of NiAl₃ as a reservoir phase for a NiCrAlY overlay coating on a IN738LC superalloy substrate at 1,100°C in air. The morphologies and microstructures of the conventional NiCrAlY and NiAl3-modified NiCrAlY overlay coatings in the as-sprayed and oxidised conditions were characterised using SEM, EDX and XRD techniques.     

On the early stages of scale development on Ni–22Al–30Pt with and without Hf additions at 1150°C

Abstract

Platinum modified Ni –Al-based alloys play an important role in the aeronautical industry as materials applied as Bond Coats (BCs) in thermal barrier coating systems (TBC), which provide protection against high temperature oxidation. Therefore, it is crucial for the assessment of the performance of the TBC to understand the oxidation behaviour of the Bond Coat material and the properties of the thermally grown oxide developing on it.

This paper reports on the scale growth mechanism of Pt-modified γ′-Ni₃Al-based alloys with and without Hf additions at the early stages of oxidation. Samples were oxidized at 1150°C for up to 50 min. The consecutive stages of the scale evolution were followed by a systematic approach using a two-stage oxidation treatment in atmospheres containing different amounts of ¹⁸O₂ oxygen isotope. The scale growth mechanism was followed with the aid of the in-depth analysis of elemental distribution, in particular of the oxygen isotopes, across the scale using secondary ion mass spectrometry (SIMS). The surface morphology of the scales was observed using scanning electron microscopy, while its local phase composition was determined using photoluminescence spectroscopy. In addition, the imaging SIMS was used to generate distribution maps of the oxygen isotopes. Because of the limited lateral resolution of the SIMS technique (down to 0.5 μm) they were conclusive only for the most prolonged exposures (50 min).

Typical stages of scale evolution were observed for the growing scales: initially flat layers were formed, followed by more or less developed blade-like surface grains, phase-transformation-related cracks and round patches and, finally, ridges. The latter were only found on the Hf-free material. The scale growth mechanism evolved from an initially predominant outward growth mechanism towards an increasing contribution of the inward mechanism. The relative extents of the outward and inward growth mechanisms depended strongly on the surface fraction of patches and through-scale cracks. The phase composition analyses showed that cracks and patches are regions where the transient aluminas were preferentially transformed into the α-Al₂O₃ phase which is the required protective scale. The ridges formed in cracks essentially consisted of α-Al₂O₃. For the most of studied exposure periods both types of phases: transient aluminas and α-Al₂O₃ co-existed in the scales.

The results obtained indicate that: (i) phase transformations occur locally and not simultaneously in the entire scale; (ii) Hf additions retard the phase transformation; (iii) the phase transformation is completed only on Hf-free material and not after oxidation for 50 min; (iv) the two-stage oxidation approach should be carefully applied to study the growth mechanism of evolving scales in which regions of different phase compositions develop.     

Determination of diffusion coefficient of oxygen in α-iron from internal oxidation measurements in Fe-Si alloys

The diffusion coefficient of oxygen in -iron was determined by internal oxidation measurements on iron alloys with various contents of silicon in a temperature range from 1073 to 11 73 K with particular attention to the effect of oxide particles in the oxidation layer. The oxide in the oxidation layer and the concentration of silicon present as an oxide, as well as the rate constant for penetration of the oxidation front were determined. The diffusion coefficient of oxygen in the layer, D _O ¹⁰ , calculated using the rate equation for internal oxidation, increases with the increase in volume fraction of the oxide, f ¹⁰. This result indicates that the existence of oxide particles accelerates oxygen diffusion. Therefore, the diffusion coefficient of oxygen in -iron is determined by extrapolating D _O ¹⁰ to f ¹⁰ = 0, giving good agreement with results obtained in our recent investigations.