Search

Cerium ions were implanted into a high purity polycrystalline Ni to a dose of 2 × 1016 Ce+/cm2. The radiation damage and distribution of Ce in Ni substrate were modified by postimplantation vacuum annealing. The Ce implants significantly decreased the NiO growth rate at 973 K. Thin oxide films formed on implanted Ni were composed of three well-defined sublayers, each with an essentially different microstructure. Cerium was present in the near surface region of the oxide in the form of CeO2 particles, randomly distributed in NiO matrix, and Ce ion segregants at NiO grain boundaries. The size of CeO2 particles, formed during the initial stages of exposure to oxygen, affected the inhibition of oxide growth. Vacuum annealing following implantation decreased the beneficial effect of Ce implants by increasing the size of CeO2 particles. The role of Ce implants in inhibiting grain boundary diffusion in NiO is analyzed. As a result of this study, the growth mechanism of thin NiO films on Ce-implanted Ni is proposed.

The effect of CeO{sub 2} ceramic coatings with thicknesses in the range of 14 to 42 nm on the oxidation of high purity polycrystalline nickel at 973 K in 1 atm O{sub 2} has been studied. The ceria coatings decrease the Ni oxidation rate after 125 h by a factor up to 45, which is significantly higher than the reduction achieved during early stages. Growth features, including oxide thickness, surface morphology, and the texture for both the pure and CeO{sub 2} modified NiO, demonstrate a definite dependence on the crystallographic orientation of the Ni substrate. Oxides with thicknesses of up to 1 {micro}m developed on CeO{sub 2}-coated Ni consisted of three sublayers. The Ce-rich part, composed of small NiO grains and CeO{sub 2} particles, was located inside the scale and shifted deeper into the scale with increased oxidation time. The growth kinetics and microstructural evolutions of modified NiO are discussed in terms of the mechanism of inhibition of grain boundary diffusion in NiO by Ce{sup +4} ions.

The oxidation behavior of γ′-Ni3Al and γ′-Ni3Al(Si) (Ni75Al20Si5) intermetallic compounds was studied at 1073 K and oxygen partial pressures of 5∼9×10−6 atm, 1.2×10−14 atm, and 1.2×10−19 atm by means of a manometric apparatus, and Rhines packs of NiO/Ni and FeO/Fe, respectively. Oxidation kinetics were determined either by recording weight gains or by measuring the internal-oxidation-zone depths. The structures and morphologies of oxides were also studied. Relatively low oxidation rates occurred for both compounds when oxidized in the manometric apparatus, while fast internal oxidation was observed for both compounds in the NiO/Ni pack. The fastest oxidation occurred in the γ′-Ni3Al compound in the FeO/Fe pack. A healing Al2O3 layer was formed on the surface of γ′-Ni3Al(Si) compound in the FeO/Fe pack, indicating a synergistic effect between the solute elements in the compound at the oxygen pressures corresponding to the dissociation of FeO. The oxidation rate was found to depend on the volume expansion associated with solute-atom oxidation.

The formation and growth of oxide islands during the initial stages of oxidation of a Co-Ni alloy are studied by direct imaging of these islands using scanning Auger microscopy. These images give clear experimental evidence in support of the Holloway-Hudson model of oxide growth in which the islands nucleate at one point in time and then grow at a rate determined by the rate of gas impingement at the island perimeters, the surface crystallographic orientation, and the substrate temperature. The islands are also shown to be Co rich.

The surface morphology of oxide grown in the temperature range of 873–1173K on modified high-purity nickel has been observed and quantitatively analyzed by atomic-force microscopy. The modifications included one of two finishing techniques and either CeO2 sol-gel coatings or Ce-ion implantation. There is an essential difference in the surface morphology and grain size for oxides formed on Ce/CeO2-modified nickel in comparison to pure NiO. The oxide topography also depends on the substrate-surface-finishing technique and the method of applying the cerium. Both Auger electron spectroscopy and Rutherford backscattering spectrometry were used to evaluate the depth composition of the oxide. In the modified oxide formed on chemically polished substrates the Ce generally appeared to be located close to the outer surface, even after long oxidation times. In oxide grown on mechanically polished substrates Ce was detectable by these techniques only during the very early stages of oxidation. A general correlation exists between the parameters describing the oxide-surface morphology and the Ce-depth distribution, and the reduction of the NiO growth rate achieved by applying the reactive element. Detailed analysis of the oxide surface, internal microstructure, and chemistry gives new insights into the problem of the characterization of thin oxide films and the mechanism of oxide formation.

The formation and development of oxides in Ni−4Al and Ni−4Al−xSi (at.%, x=1, 3, 5) alloys at 5–9×10−6 and 1 atm oxygen pressure at 1073 K have been studied. The oxidation rate increased with an increase of silicon content in the alloy at the early stage of oxidation, but decreased after longer time exposure due to formation of an intermediate layer composed of NiO and spinel (NiAl2O4 and Ni2SiO4) between the top NiO layer and the internal-oxidation zone. This intermediate layer became a barrier for releasing stress, generated by the volume expansion associated with oxidation of solute atoms, resulting in high dislocation density and severe distortion in the internal-oxidation zone for the Ni−Al−Si alloys. In Ni−4Al alloy where no complete intermediate-layer formation occurred, stress was easily released by an enhanced vacancy gradient, and therefore an enhanced vacancy-injection rate into the alloy, resulting in a higher oxidation rate than the situation where a sample was oxidized at an oxygen pressure associated with the dissociation of NiO.

A treatment of the kinetics of internal oxidation of ternary alloys in the absence of an external scale is presented. The analysis involves simultaneous internal oxidation of the two solutes in the ternary alloys. A comparison with measurements on Ni-4 at.% Al-1 at.% Si alloys at 800°C reveals a good agreement between theory and experiment.

The influence of 14-nm thick ceria ceramic coatings deposited by the sol-gel technique on the early-stage oxidation of polycrystalline nickel at 973 K was studied by analytical electron microscopy, Auger electron spectroscopy, Rutherford backscattering spectrometry and X-ray diffraction. The size of the ceria particles in the coating was modified prior to oxidation by vacuum annealing. It was found that ceria particle size is a crucial factor affecting the oxidation kinetics, oxide microstructure, and distribution of cerium within the oxide film. Coarse ceria particles applied to the nickel surface were ineffective in the inhibition of oxidation and were spread throughout the whole oxide. Coatings with small ceria particles markedly improved the oxidation resistance. After oxidation such particles were present in the surface region of nickel oxide, acting as the sources of cerium ions segregated at the nickeloxide grain boundaries. The stereological analysis of oxide microstructure as well as microchemical examination supported the predominant role of grain-boundary segregation of cerium ions decreasing the oxidation rate. The results are discussed in terms of reactive-element effect on the development of microstructure of nickel oxide film during initial stages of oxidation.

CeO[sub 2] sol with particle diameter of 5 nm was applied to the surface of both chemically and mechanically polished polycrystalline nickel. The oxidation kinetics during early stages of NiO formation have been measured with an ultrahigh vacuum manometric system in the temperature range of 873 to 1073 K and oxygen pressure 5 [times] 10[sup [minus]3] Torr. The ceria coatings reduced the nickel oxidation rate by up to a factor of 4, but the extent of the effect depended on coating thickness, substrate surface finish, and preoxidation prior to coating. After oxidation the Auger electron spectroscopy sputter profiles revealed the cerium concentration maximum in the vicinity of the gas-oxide interface. As was determined by scanning and transmission electron microscopy, ceria coatings resulted in formation of more uniform oxide films, with grain size much smaller than that observed on pure nickel. Ceria particle with diameter less than 10 nm were detected both inside nickel oxide grains and at oxide grain boundaries.

The interaction between deuterium and irradiation defects in nickel is studied using the gas-phase permeation technique. Irradiation defects were introduced using 90 keV Ni + ions with fluences of 10 14 −10 17 ions cm −2 , or 30 keV 4 He + ions with fluences of 10 15 −5 × 10 16 ions cm −2 . Permeation results show that crystalline defects such as dislocation loops produce no measurable effect on deuterium permeation in the temperature range 370–570 K. By comparison, helium defects decrease the apparent diffusivity by four to six orders of magnitude, indicating a strong trapping effect for deuterium. These results are discussed using a simple trapping model.

Sulfidation of an Fe-23.4Cr-18.6Al (at.%) alloy was investigated in H2S-H2 atmospheres,\(10^{ - 6} \leqslant {\text{P}}_{{\text{S}}_{\text{2}} } \leqslant 10^2 \) Pa, at 973 K. It was found over this pressure range that sulfidation after an early transient period followed the parabolic rate law, being diffusion controlled. An investigation was carried out of the scales formed during early transient sulfidation over the sulfur pressure range\({\text{p}}_{{\text{S}}_{\text{2}} } = 10^{ - 6} - 10^{ - 2} \) Pa. Fully developed scales were multilayered consisting of an inner compact layer of equiaxed grains, an intermediate layer of equiaxed and columnar grains exhibiting a small degree of porosity, and an outer porous layer of distinct plates and needles. The grains of the inner and intermediate layers contained quarternary sulfide phases. The following phases were identified: spinels (CrFe)Al2S4 and (FeAl)Cr2S4, hexagonal (FeCr)Al2S4, (CrAlFe)2S3, and (CrAlFe)5S6. The plates and needles were composed of hexagonal (FeCr)Al2S4 and (CrAlFe)2S3 at\({\text{p}}_{{\text{S}}_{\text{2}} } \geqslant 10^{ - 6} \) and 10−5 Pa from which pyrrhotite, FeS, grew at\({\text{p}}_{{\text{S}}_{\text{2}} } \geqslant 10^{ - 4} \).

The sulfidation behavior of an Fe-26.4Cr-8Al-6Mn (at.%) alloy in H2S-H2 atmospheres,\(10^{ - 4} \leqslant {\text{P}}_{{\text{S}}_{\text{2}} } \leqslant 10^2 {\text{Pa}}\).

La reaction du degagement d'hydrogene dans une solution aqueuse d'hydroxyde de potassium est etudiee pour l'alliage de surface nickel-palladium, produit par melange ionique

Oxidation properties of Ni‐2 and 6 w/o Al alloys were investigated in 1 atm oxygen at temperatures in the range 1273°–1573°K. Ni‐2 w/o Al oxidized parabolically, one order of magnitude more rapid than pure nickel, to a duplex scale by metal diffusion consisting of outer and inner layers of Al‐doped and , respectively, and an internal zone of rodlike precipitates by oxygen diffusion through the Al‐depleted alloy. Oxidation kinetics of Ni‐6 w/o Al alloy were irreproducible due to early‐stage formation of an imperfect scale containing nodules. Impingement of favorably oriented rodlike precipitates growing beneath the nodules accompanied with lateral diffusion of aluminum to react with oxygen at the precipitation front eventually led to transition from internal oxidation to development of a multilayered scale.

The transition metals when exposed to oxygen at low and intermediate temperatures form thin, protective oxide films of up to some thousands of Angstroms in thickness. At higher temperatures, thick (> 1/z) oxide scales are generally formed. The kinetics of the high-temperature process are well established, whilst those of the low-temperature reaction are a matter of contention in the literature. Our discussion reflects this difference by incorporating a detailed consideration of lowtemperature kinetics. Oxidation kinetics are, of course, determined by the reaction mechanism. The mechanism is in turn determined by the properties of the solid oxide reaction product. It should therefore be possible to predict the kinetics of an oxidation reaction from the appropriate properties of the solid oxide or, conversely, to determine those properties from kinetic measurements. This discussion therefore commences with a review of the relevant structural, thermodynamic and transport properties of transition metal oxides. A confrontation of theory and experiment for both the lowand hightemperature r6gimes then follows, and both the successes and the inadequacies of oxidation theories are pointed out. Finally, the model employed for the oxide product is extended to include line defects and internal surfaces, and the importance of these parameters on transition metal oxidation is assessed.

Equilibrium oxygen pressures of the univariant systems , and were determined between 1123 and 1423 K using galvanic cells with calcia‐stabilized zirconia as the solid electrolyte. Slipcast tubes of a two‐phase mixture of were used as electrolyte in cells to determine the equilibrium oxygen pressures of the univariant between 923 and 1173 K and the dissociation pressure of equilibrated with (Ni‐Al) alloys containing 0.17, 1.19, and 2.36 atomic percent Al at 1213 K. The free energy of formation of was determined between 923 and 1423 K. An electron probe microanalyzer was used to determine the composition of oxides and alloy phases of the ternary Ni‐Al‐O system that coexist with each other at 1273 K. The Ni‐Al‐O isotherm at 1273 K and Ni‐Al‐O equilibrium oxygen pressure diagrams at 1213 and 1273 K are presented.

The sulfidation behavior of Mn in H2S-H2 atmospheres,\(10^{ - 8} \leqslant P_{S_2 } \leqslant 10^{ - 4} \) atm, maintained at a total pressure of 1 atm was investigated at temperatures between 1073 and 1273 K. The reaction kinetics obeyed the parabolic rate law; the parabolic rate constant was proportional to\(P_{S_{_2 } }^{{1 \mathord{\left/ {\vphantom {1 n}} \right. \kern-\nulldelimiterspace} n}} \) where n=6.1±0.3. A value of 21,000±2000 cal/mole was estimated for the activation energy of the growth of the sulfide scale under constant sulfur pressure. Polycrystalline columnar α-MnS scales were formed which exhibited texture with preferred (111) orientation. The parabolic rate constants at different sulfur pressures from this and previous investigations were used to evaluate the manganese self-diffusion coefficient in α-MnS at\(P_{S_{_2 } } \).

Growth of layers on the (100) surface of single crystal was investigated in air at 700° and 800° C. These layers were of duplex structure due to simultaneous inside and outside growth. Measurements on the position of inert gold markers demonstrated that cobalt migration outward in was the dominant diffusion mechanism. The oxidation kinetics were parabolic; the parabolic rate constant at 800°C was used to obtain an estimate of for the self‐diffusion coefficient of cobalt in .

Auger electron spectroscopy in combination with depth profiling by Ar ion sputtering was employed as a surface analytical technique to determine the positions of thin (3–5 nm) inert Pd markers during growth of oxide films on Al, Ni, and β-NiAl. Alumina films 35 and 120 nm thick formed on Al and β-NiAl, respectively, grew by inward diffusion of oxygen. Metal migration accounted predominantly for the growth of a NiO film 150 nm thick on the (111) crystal face of Ni. However, an assessment could not be made with respect to oxygen diffusion, if any, in this film due to the limitations imposed on precise depth profiling of the marker position caused by uneven sputtering of the oxide and metal phases.

Equilibrium oxygen pressures of the univariant ternary coexistences hematite + spinel + alumina, alloy + wustite + spinel, and alloy + spinel + alumina were determined at temperatures in the range 1123 to 1423 K using solid state electrochemical cells. The compositions of alloys and oxide phases of this Fe-Al-0 system that equilibrated with one another at 1273, 1423, and 1573 K were determined. These results and those from the literature are used to determine stabilities of the ternary oxide phases as a function of oxygen pressure and composition by constructing equilibrium oxygen pressure diagrams at 1073, 1173, and 1273 K.

Diffusion coupling of an Fe6Al alloy with FeS in the form of compacts results in periodic precipitation of FeS + FeAl 2 S 4 bands in the alloy. The thickness of sulfide bands increases with depth satisfying the Jablczynski spacing relationship for Liesegang-type of precipitation. The initial stage of sulfide band formation is explained by a model involving precipitation of Al 2 S 3 which is subsequently converted to FeAl 2 S 4 and FeS. Growth of each sufide band is supported by rapid diffusion of iron and aluminum in the FeS phase. It is suggested that aluminum depletion from the alloy in front of the innermost sulfide band ultimately leads to its cessation of growth, a new sulfide band beginning to form in the alloy at a distance where the concentration product for the precipitation of Al 2 S 3 is satisfied.

Nickel spinel, Ni1−xAl2(1+x/3)O4, is the only intermediate compound in the quasibinary NiO-α-Al2O3 system at temperatures between 1000 and 1920°C. The spinel equilibrated with NiO occurs at its stoichiometric composition, NiAl2O4, independent of temperature. An alumina rich spinel, 0.17 ⩽x⩽ 0.62, equilibrated with α-Al2O3 increases in alumina content with increasing temperature. Aluminum oxide solubility in NiO increases from 1 mole % at 1000°Cto 3 mol % at 1800°C. Nickel oxide solubility in α-Al2O3 was found to increase from 2 mole % at 1000°C to 3 mole % at 1920°C.

Croissance de trois couches distinctes de sulfures en presence de H 2 S-H 2 : NiS (ou Ni 3 S 2 ) Al 2 S 3 , Ni 3 S 2

An investigation is reported on the sulfidation properties of an Fe-23.4Cr-18.6Al(at.%) alloy at 1073 and 1173 K in H2S-H2 atmospheres, 104 > PS2 ≥ 105Pa. The sulfidation kinetics exhibited an early transient period before onset of parabolic kinetics. Values of the parabolic sulfidation rate constants increased by as much as 105 from their smallest values at low sulfur pressures, PS2 ≦ 10−4 Pa, to maximum values at sulfur pressures PS2 ∼ 102 Pa. Multilayered scales were formed, the number and types of layers dependent on sulfur pressure. A fully developed scale at sulfur pressures PS2 > 10−3 Pa.

The sulfidation kinetics and morphological development of reaction products are reported for Fe-9 and 18 at.% Al alloys exposed at 1173 K to H2S-H2 atmospheres at sulfur pressures in the range 10−1−103 Pa. The Fe-9 Al alloy sulfidized parabolically at\({\text{P}}_{{\text{S}}_{\text{2}} } = 10^{ - 1} \) Pa giving rise to a duplex scale composed of an outer Al-doped FeS layer and an inner FeS + FeAl2S4 lamellar layer and to an internal sulfidation zone containing Al2S3 precipitates. The Fe-18 Al alloy which was sulfidized at\(10^2 {\text{Pa}} \geqslant {\text{P}}_{S_2 } \geqslant 10\).

A method of long-term Sulfidation that involves generating sulfur vapor from dissociation of Ag2S or from transformation of NiS to Ni3S2+x is described. Sulfidation tests of iron samples showed the approach to be essentially valid. At 1023 K, Ag2S dissociation of Ag2S gives rise to\(p_{s_2 } = 7.9 \times 10^{ - 6} \) atm and, in the presence of an inert carrier gas, efficient transport of sulfur to the metal was effected. Liberated silver grew in the form of discrete whiskers, hence did not interfere with continued dissociation of Ag2S. Equilibration of NiS with Ni3S2+x occurs at\(p_{s_2 } = 1.4 \times 10^{ - 4} \) atm at 1023 K but, with continued loss of sulfur to the metal, the nonstoichiometric Ni3S2±x predominates and the ambient\(p_{s_2 } \) is no longer constant. Thus for long-term studies, Sulfidation with NiS/Ni3S2+x mixtures is suitable only for metals that sulfidize slowly or at rates independent of the ambient\(p_{s_2 } \). Using mainly the NiS-Ni3S2+x mixture as sulfur source, sulfidation tests up to 72 h at 1023 K were conducted on an Fe-20 at.% Al alloy. Sulfidation was apparently insensitive to the ambient\(p_{s_2 } \)and formation of a three-layer FeS/(Fe, Al)3S4/Al2S3 scale with internal precipitation zone proceeded at similar rates, and a relationship involving coupled diffusion is proposed. At longer exposure times, transformation of FeS to (Fe, Al)3S4 became extensive. Mechanical failure of the scale, followed by rapid healing beneath the original precipitation, also occured after extended reaction periods (>50 hr).

A ternary diffusion model is invoked to interpret growth of the scale on alloys containing greater than 40 atomic percent (a/o) Fe in atmospheres, , at temperatures 533°– 980°C. Concentration profiles in the solid phases were determined from electron probe microanalyses; the iron concentration across a scale continuously decreased from its highest value at the scale/alloy interface. Computations demonstrated that the relative self‐diffusivity of cobalt to iron in was dependent on concentration increasing asymptotically from a value of 1.3 in essentially pure to a value of . A numerical computing method capable of determining sulfidation rates of a binary alloy from diffusional and thermodynamic data was used to demonstrate that calculated metal composition profiles in the scales and values of the parabolic sulfidation rate constants were in good agreement with the experimental determinations.

Sulfidation properties of Fe‐6, 9, 18, and 28 atomic percent (a/o) Al alloys were investigated at 1173 K in sulfur vapor at the dissociation pressure of . The Fe‐6 Al alloy exhibited internal sulfidation; the Fe‐9 and 18 Al alloys exhibited internal sulfidation along with formation of an external scale. Compositions of the internal sulfide precipitates corresponded to in the Fe‐6 Al alloy and to in the alloys containing 9 and 18 a/o Al. The Fe‐28 a/o Al alloy upon sulfidation only formed an external scale. The apparent permeability of sulfur in the alloy of the internal sulfidation zone as defined by product increased linearly with alloy composition. Growth of the sulfide precipitates which extended as plates across the internal sulfidation zones is satisfactorily interpreted by a model involving enhanced sulfur diffusion along the interfaces between the internal sulfides and alloy matrix. Transition from internal sulfidation to internal sulfidation plus external scale formation and finally to only external scale formation based upon the alloy aluminum composition is discussed.

SIMS was used to locate O18 in oxide scales grown on niobium specimens oxidized at 600, 700, 780, and 825°C. Analyses were performed in depth profile from the oxide-gas interface and in step-scan across polished sections. The measured distributions confirm that oxygen is the most mobile species in T -Nb2O5 and gave no evidence that pore diffusion was important in the growth of compact scales. In these scales, which grow essentially at a parabolic rate, O18 had a sigmoidal profile across the scale, plateauing near the oxide-gas interface. The O18 content of the plateau was typically less than 90% of the O18 content of the oxidizing atmosphere. At temperatures below 720°C the measured O18 distributions are compatible with a paralinear growth mode of the layered scales. In this mode the reaction rate is controlled by the diffusional properties of the oxide layer in contact with the specimen.

Oxidation of a Fe-7.4 at.% Al alloy at 1173 K in oxygen at 1 atm pressure resulted in the formation of a rust-colored film, shown by Auger spectroscopy to be a duplex layer consisting of an outer Fe-rich oxide over an Al-rich oxide. Mossbauer spectroscopy revealed that the outer layer was hematite containing ∼10 at.% Al3+ and that the inner alumina layer contained some Fe3+.

High temperatures associated with arcing cause Cu–Ag alloys to oxidize and acquire increased resistivities when utilized in electrical switches. This oxidation involves formation of copper oxides at the surface and as precipitates randomly distributed in the arced layer and bulk of the alloy. As such, the change of resistivity with the extent of oxidation resembles a percolation process: specimens with an initial good electrical conductivity exhibit an abrupt transition to a poor conductivity when the buildup of oxide precipitates blocks paths in the conductive metal matrix.

An investigation is reported on the sulfidation properties of Fe-Cr alloys over their complete compositional range in H2S -H2atmospheres at 1073 K. Sulfidation of alloys containing less than 60 at. % Cr obeyed parabolic kinetics and alloys of larger chromium content and chromium followed linear kinetics. Scales were composed of two distinct layers: an outer columnar grained sulfide and an inner equiaxed grained sulfide layer of increased relative thickness and porosity at larger alloy chromium contents. The scales on quenched specimens were composed of either (CrFe)5S6or (FeCr)S plus small amounts of (FeCr)3S4at\(P_{S_2 }\) =10−2Pa and of (CrFe)5S6at\(P_{S_2 }\) =10−5Pa. Internal sulfidation was most pronounced in the high chromium content alloys and at the low sulfur pressure. The parabolic sulfidation rate at\(P_{S_2 }\) =10−2Pa was large and of practically constant value irrespective of alloy composition. These kinetics at\(P_{S_2 }\) =10−5Pa, however, increased from a very low value by three orders of magnitude as the alloy chromium content was increased from 5–60 at.%.

Comportement de l'alliage lors d'expositions a des atmospheres statiques Ar-S 2 (plus de 72 heures) et a des flux de melanges H 2 -H 2 S (6 heures). Suivi des cinetiques de croissance des couches formees et observation de la morphologie des produits de corrosion. Influence de la pression partielle de soufre presente dans l'atmosphere

The structures of films of 300A‐15μ thick formed on (100), (110), and (111)Ni crystal faces at temperatures in the range 500°–800°C were investigated by electron microscopy and x‐ray diffraction. The oxide was initially polycrystalline but rearranged as the oxide thickened to major epitaxial relationships with the metal substrate. These relationships were characterized by coincidence between the close‐packed directions of the metal and oxide; the number and type of oxide orientations were correlated with the number of close‐packed directions in the metal faces. Oxide thicker than 1μ exhibited preferred {100} and {111} orientations. The oxide on (100) and (111)Ni exhibited the highest preferred orientation and a duplex layer structure consisting of equiaxed crystallites in the inner layer and columnar ones in the outer layer. These findings are rationalized in terms of a diffusional model for the growth of these oxide layers in which the reaction rate is determined by boundary and lattice diffusion of nickel through nickel oxide.

EDAX spectrometry was employed as a semiquantitative analytical technique to determine the penetration depth of NiAl 2 O 4 into the internal oxidation zone of Ni-5 wt. % Al alloy oxidized in air at 1273 K. Depth profiles obtained by spot counting Al to Ni K α peak intensities demonstrated that a sharp NiAl 2 O 4 -Al 2 O 3 phase boundary exists in this internal oxidation zone at a normalized fractional depth which can be determined to an accuracy of 0.70±0.05. X-ray reflection diffraction analyses and color examination of chemical deep-etched tapered cross sections substantiated these results.

Surface segregation measurements were made on three Co-Rh alloys (19.3, 47.8 and 84.7 at% Rh) in the temperature range 623–1123 K using AES. Cobalt segregation occurred on the Co-84.7Rh alloy with a heat of segregation of −3.7 ± 7 kJ mol −1 while rhodium enrichment was observed on the Co-19.3Rh alloy with a heat of segregation of 1 ± 1 kJ mol −1 . No variation from the bulk composition could be conclusively demonstrated on the surface of the Co-47.8Rh alloy. There is good quantitative agreement between these results and the combined bond breaking and lattice strain theories of segregation.