Hydrogen permeation and diffusion in inconel 718 and incoloy 903

Hydrogen permeation and diffusion have been measured in two high strength superalloys, Inconel 718 and Incoloy 903. Measurements were made over the temperature range of 150 to 500°C, with applied hydrogen pressures of 1 to 3 atm. The permeabilityp and dif-fusivityD values were used to derive the hydrogen solubility S in these materials. For Inconel 718 the results depend to some extent on the heat treatment of the alloy. For this alloy in the solution treated condition, the results for φ andD are: D.1.07xl0^-2exp(-11,900/RT) cm²/s For Incoloy 903, the results are: φ = 9.50x10^-3 exp (-13,710/RT) cm³(NTP)/cm-s-atm¹/ RT cm-s-atm1/2D = 2.46 x 10^-2(-12,590/RT) cm²/s where the activation energies are in cal/mole. The measured diffusivities and permeabilities are very similar for the two alloys.     

Low-cycle fatigue and cyclic stress-strain behavior of incoloy 800

Strain-controlled low-cycle fatigue tests of solution-annealed Incoloy 800 were performed at temperatures of 538°, 649°, 704°, and 760°C using axial strain rates of 4 × 10^-3 and 4 × 10^-4 sec^-1. A few hold-time tests were also performed to indicate a noticeable reduction in fatigue life at hold times of 10 and 60 min. A comparison of these fatigue data with similar results for AISI 304 stainless steel indicates essentially identical behavior. An extensive study is made of the cyclic stress-strain behavior of Incoloy 800 and the relationship between the cyclic strain-hardening exponent and fatigue behavior is confirmed. Exponents onN _f in the elastic and plastic strain range terms of the total strain range equation are identified and compared with those used in the Universal Slopes equation.     

Permeation of hydrogen through alpha iron

The permeation rate of hydrogen through two alpha iron specimens was measured by means of a low pressure steady-state permeation technique under conditions such that diffusion was the rate controlling process for permeation. The observed permeation rate in these samples can be described by the equation: J=5.55 ±.52 × 10^-9 p ^1/2 exp - 8095 ± 88/RT mole s^-1 cm^-1 torr^-1/2 over a range of hydrogen pressures from 21 torr to 766 torr and over a temperature range of 342 to 619 K. Variation of permeation rate with the square root of hydrogen pressure was used as evidence that the permeation process was diffusion rather than surface reaction controlled. The permeation data were used in conjunction with the solubility expression of Gonzalez¹ to determine the diffusivity of hydrogen through alpha iron as:D = 1.01 × 10^-3 exp - 1595/RT cm²/s     

Analysis of oxide coatings on steam-oxidized incoloy 800

In-situ formed oxide coatings have been proposed as hydrogen permeation barriers for candidate materials in heat exchangers and for hydrogen containment. In this regard oxides formed by oxidation of Incoloy 800 in 243 and 532 torr (1 torr = 133.3 Pa) steam at 933 K, and in 714 torr steam at 793, 933, and 998 K for exposure times up to 5000 h, have been chemically analyzed. Analytical techniques included ion microprobe mass analysis and optical microscopy. Sample pretreatment, including electropolishing, annealing and cold working, affected the oxidation chemistry. Steam oxidation of electropolished samples formed thick iron oxide at the steam-oxide interface. Electropolished/annealed samples (1173 K for one h under vacuum) formed iron oxide at short exposure times, however longer exposure resulted in the iron oxide spalling, leaving a thin oxide rich in chromium and manganese. Subsequent cold work on annealed samples led to higher iron content in the oxide upon steam exposure. H. F. BITTNER formerly with Research Staff, Oak Ridge National Laboratory, Oak Ridge, TN.     

Carbon diffusion in Mo2C as determined from carburization of Mo

A diffusion coefficient of C in nonstoichiometric α-Mo2C has been determined from the growth kinetics of the carbide layer. The results conform to the relationship:Dc (in cm²/s) = 68.86 ± 1.51 exp [(-294.77 ± 4.98)/RT] for the temperature range of 1273 to 1673 K, with the activation energy in kJ/mole. The growth rate,Kp, of the carbide thickness can be expressed as:Kp (in cm²/s) = 32.63 ± 1.52 exp [(-319.06 ± 5.12)/RT].     

Relationship between magnetic properties,sensitization, and corrosion of incoloy alloy 800 and inconel alloy 600

Sensitization processes in two high Ni-Cr-Fe alloys are studied with the Huey and accelerated Strauss tests, magnetic permeability measurements and scanning electron microscopy. There is good agreement between the corrosion tests for Incoloy* Alloy 800 but not for Inconel* Alloy 600. High Huey corrosion rates are associated with large magnetic permeabilities which result from chromium depletion near the grain boundaries. The relative contribution of the chromium depleted region and of electrochemical effects to the Huey corrosion rate could not be determined because the formation and disappearance of the chromium depleted zone and of the continuous grain boundary carbides occur concurrently.     

Determination of hydrogen pęrmeation parameters in alpha titanium using the mass spectrometer

The kinetic parameters of hydrogen permeation through α-Ti were determined from mass spectrographic measurements between 400° and 800°C and 0.004 and 2.0 torr H₂. The specimens were hollow cylinders and the hydrogen source was gaseous hydrogen. Hydrogen entered the specimen wall at the inside surface and emerged from the specimen at the outside surface. The permeated hydrogen flux is given byP = 1.6 × 10⁻⁵ p exp (-14,900 ± 856/RT) mole torr/s cm² where the hydrogen pressure isp and the permeation activation energy is 14,900 cal mole⁻¹ The surface dependent nature of hydrogen permeation in α-Ti is established in these experiments by the observations that 1) permeation is not proportional to the square root of input hydrogen pressure, 2) the concentration gradient calculated from experimental permeation data is lower than it would be if equilibrium solubility had developed, and 3) the diffusional frequency factor as determined under nonsteady-state conditions is low by a factor of three compared with values determined by others.     

Steady state hydrogen transport through zone refined irons

Hydrogen permeation from the gas phase through zone refined iron was measured with the electrochemical technique over 0 to 60°C and 0.01 to 1 atm. The surface impedance problems normally encountered in this pressure-temperature range were eliminated by a thin layer of electroplated palladium on the input surface of the permeation membrane. The steady state flux was propprtional to the square root of the pressure and inversely proportional to the membrane thickness. The permeability coefficient was 2.6 × 10¹⁷ exp ((-8500 ± 600 cal/mole deg)/RT) atom H/cm s atm^1/2, in good agreement with earlier results at higher temperatures. These results can be used to calibrate electrochemical charging experiments in terms of an effective hydrogen pressure. Formerly with the Department of Materials Science and Engineering, Cornell University, Ithaca, New York     

Steady state hydrogen transport through zone refined irons

Hydrogen permeation from the gas phase through zone refined iron was measured with the electrochemical technique over 0 to 60°C and 0.01 to 1 atm. The surface impedance problems normally encountered in this pressure-temperature range were eliminated by a thin layer of electroplated palladium on the input surface of the permeation membrane. The steady state flux was propprtional to the square root of the pressure and inversely proportional to the membrane thickness. The permeability coefficient was 2.6× 10¹⁷ exp ((- 8500 ± 600 cal/mole deg)/RT) atom H/cm s atm^1/2, in good agreement with earlier results at higher temperatures. These results can be used to calibrate electrochemical charging experiments in terms of an effective hydrogen pressure. Formerly with the Department of Materials Science and Engineering, Cornell University, Ithaca, New York     

Hydrogen permeation and diffusion in iron-base superalloys

In this work, the hydrogen permeation and diffusion in three iron-base superalloys, JBK-75, Incoloy903 and GH35A, were investigated. A gaseous permeation technique was employed to measure the hydrogen permeabilities and diffusivities of specimens with different, heat treatments for a given alloy over the temperature range of 210–430°C. The effects of strengthening phase γ′ on the hydrogen permeation and diffusion in the alloys were examined. The results showed that the dependences of the hydrogen permeabilities and diffusivities in various alloys on the temperature obeyed Arrhenius relationship over the test temperature range and that the hydrogen permeation behavior in these alloys was essentially independent of their heat-treated conditions, i.e. which was not significantly affected by γt́ precipitates in the alloys. The hydrogen permeation parameters of the iron-base superalloys were approximately consistent with those of the austenitic stainless steels. The hydrogen transport in these alloys with f.c.c. structure was generally controlled only by the lattice diffusion of hydrogen.     

Diffusion in the titanium-nickel system: II. calculations of chemical and intrinsic diffusion coefficients

Diffusion coefficients in the Ti-Ni system have been calculated by the aid of equations given by Sauer and Freise, and Wagner. Values for the TiNi (50 at. pct Ni) phase were found to be:D ^u (cm²/s) = 0.0020 exp - 142,000/R for the temperature range between 650 and 940°C. The heat of activation, expressed in J/mol, has an accuracy of ±6000. For the β-Ti(Ni) phase containing 6 at. pct Ni the temperature dependence of the diffusion coefficient is expressed by:D ^u (cm²/s) = 0.0688 exp - 141,000/RT. The uncertainty in the energy of activation is ±12000 J/mol. No clear variation of the diffusion coefficient with concentration could be detected. It was found that Ni is by far the fastest moving component in β-Ti(Ni), Ti₂Ni and TiNi (at least in the composition range between 50 and 53 at. pct Ni). Values ofD _Ni/D _Ti have been calculated with an equation derived by van Loo. The significance of the calculated values is critically examined. By means of a practical example it is shown that the calculated ratio of the intrinsic diffusion coefficients can be extremely sensitive to slight variations in the position of the marker interface.Diffusion coefficients in the Ti-Ni system have been calculated by the aid of equations given by Sauer and Freise, and Wagner. Values for the TiNi (50 at. pct Ni) phase were found to be:D ^u (cm²/s) = 0.0020 exp - 142,000/R for the temperature range between 650 and 940°C. The heat of activation, expressed in J/mol, has an accuracy of ±6000. For the β-Ti(Ni) phase containing 6 at. pct Ni the temperature dependence of the diffusion coefficient is expressed by:D ^u (cm²/s) = 0.0688 exp - 141,000/RT. The uncertainty in the energy of activation is ±12000 J/mol. No clear variation of the diffusion coefficient with concentration could be detected. It was found that Ni is by far the fastest moving component in β-Ti(Ni), Ti₂Ni and TiNi (at least in the composition range between 50 and 53 at. pct Ni). Values ofD _Ni/D _Ti have been calculated with an equation derived by van Loo. The significance of the calculated values is critically examined. By means of a practical example it is shown that the calculated ratio of the intrinsic diffusion coefficients can be extremely sensitive to slight variations in the position of the marker interface. This paper is based on a Thesis submitted by G. F. BASTIN in fulfillment of requirements for the degree of Doctor in Technological Sciences.     

Diffusivity and solubility of oxygen in solid copper using potentiostatic and potentiometric techniques

The diffusivity and solubility of oxygen in solid copper have been determined in the temperature range 700 to 1030 °C using potentiostatic and potentiometric techniques. The results are summarized by the following equations: _Do ^Cu = _1.16-0.31 ^+0.42 × 10⁻² exp(−67300 ± 3000/RT) cm² per second; _No ^s = 154 exp(−149600/RT)atom fraction of oxygen where R is in joules/degree/mole. The experimentally determined value of the pre-exponential factor in the diffusivity equation is found to be consistent with Zener’s model for an interstitial diffusion mechanism. on leave of absence from the Banaras Hindu University, India     

The diffusion of antimony in alpha iron

Diffusion coefficients of antimony in α-iron were determined in the temperature range 700 to 900°C using the residual activity method. Specimens were large-grained polycrystals for the higher temperature measurements and single crystals for the low temperature measurements. Above 800°C the data may be represented by the equationD _sb(cm²/s) = (440 ± 200) exp [- (270,000 ± 7000)/RT]. The activation energy (reported in J/mole) is approximately equal to that measured for iron self-diffusion in this same temperature range, although the antimony diffusion coefficients are a factor of ten larger than the iron self diffusion coefficients. The potential for strongly coupled vacancy-antimony motions is demonstrated, based on the observed enhancement of iron self diffusion in dilute iron-antimony alloys. Finally molybdenum is shown to have a negligible effect on the diffusion of antimony in α-iron. These results are discussed in relation to the phenomenon of temper brittleness in steels. Embrittlement kinetics in iron-antimony alloys are shown to be consistent with an antimony diffusion controlled segregation mechanism.     

Improvement of Kryukov and Zhukhovitskii method for measurement of diffusion. Application to self-diffusion in alpha iron

Abstract

The method developed by Kryukov and Zhukhovitskii for the measurement of diffusion coefficients has been improved by deriving the exact relation between the measured activities and the diffusion time. The new equation is similar to that of Kryukov and Zhukhovitskii except for an additional term and is linear for all values of time. The method has been used to measure the self-diffusion coefficient of alpha iron between 809 and 889°C. The self-diffusion coefficient (in cm²/sec) is given by

D _a = 5.4 exp (?60.3/RT)

The extrapolation of this result to the delta phase is in good agreement with the value determined experimentally in that phase.

Résumé

Les auteurs ont amélioré la méthode de Kryukov et Zhukhovitskii pour la mesure des coefficients de diffusion en dérivant la relation exacte entre les activités mesurées et le temps de diffusion. La nouvelle relation est semblable à celie de Kryukov et Zhukhovitskii sauf pour un terme additionnel; elie est lineaire pour toute valeur du temps. La methode a ete utilisee pour mesurer le coefficient d'autodiffusion du fer en phase alpha entre 809 et 889°C. Le coefficient d'autodiffusion, exprime en cm²/sec, est donne par

D _a = 5.4 exp (?60.3/RT)

L'extrapolation de ce resultat à la phase delta est en bon accord avec les résultats expérimentaux obtenus dans cette phase.     

The diffusivity and solubility of oxygen in liquid tin and solid silver and the diffusivity

The diffusivity and solubility of oxygen in liquid tin and solid silver in the temperature range of about 750° to 950°C (1023 to 1223 K) and the diffusivity of oxygen in solid nickel at 1393°C (1666 K) were determined using the electrochemical cell arrangement of cylindrical geometry: Liquid or Solid Metal + O (dissolved) | ZrO₂ + (3 to 4%)CaO | Pt, air The diffusivity and solubility of oxygen in liquid tin are given by:D _O(Sn) = 9.9 × 10⁻⁴ exp(−6300/RT) cm²/s (9.9 × 10⁻⁸ exp − 6300/RT m²/s) andN _O ^S (Sn) = 1.3 × 10⁵ exp(−30,000/RT) at. pct The diffusivity and solubility of oxygen in solid silver follow the relations:D _O(Ag) = 4.9 × 10⁻³ exp (−11,600/RT) cm²/s ( 4.9 × 10⁻⁷ exp − 11,600/RT m²/s) andN _O ^S (Ag) = 7.2 exp (−11,500/RT) at. pct The experimental value for the preexponential in the expression forD _O(Ag) is lower than the value calculated according to Zener’s theory of interstitial diffusion by a factor of 11. The diffusivity of oxygen in solid nickel at 1393°C (1666 K) was found to be 1.3 × 10⁻⁶ cm²/s (1.3 × 10⁻¹⁰ m²/s). Formerly Graduate Student, Department Formerly Graduate Student, Department Formerly Graduate Student, Department This paper is based upon a This paper is based upon a This paper is based upon a This paper is based upon a     

The nitridation of chromium and Cr-Ti alloys

Reaction zones and growth kinetics were studied after exposing high-purity chromium and alloys containing 0.5, 3.0, and 5.0 wt pct Ti to 1 atm of nitrogen between 1000° and 1400°C. Outer layers of Cr₂N and regions of internal nitridation, containing dispersed TiN particles, grew in a parabolic manner. An exact solution of Maak’s simplified analysis for internal oxidation provided calculations of nitrogen diffusion in the internal-nitride zone of each alloy. Extrapolation gave the relationshipD = 9.6 × 10^-3 exp (-28,500/RT) cm² sec^-1 for nitrogen diffusion in high-purity chromium. Increasing titanium to 5.0 wt. pct gaveD = 2.5 × 10^-3 exp (-24,000/RT) cm² sec^-1.     

Hydrogen permeation through oxide and passive films on iron

Hydrogen permeation through thin films of Fe_I–Y O on iron and through chemically polished iron were investigated by the sensitive electrochemical technique. The oxide was formed on the exit side of the sample membrane. The hydrogen arriving at the iron/oxide interface is in an atomic or protonic state which renders the hydrogen uptake by the oxide possible. The wustite films were formed by oxidation in a H₂O-H₂-atmosphere. The dependence of the hydrogen permeation current on temperature and film thickness and different degrees of nonstoichiometry in Fe_I–Y O was studied. Hydrogen permeation through these oxides is possible, but very low permeation coefficients have been found, of the magnitude of at 25°C. The diffusion coefficient of diffusible hydrogen was determined to be about 4 · 10^?10 cm²/s. Measurements of the potential dependence of permeation across the film indicate that hydrogen migrates in the oxide as a charged particle (proton). In the case of the passive surface film on iron formed by chemical polishing, the dependence of the permeation current on temperature and anodic potential was measured. The electrochemical behaviour of the film was studied by cyclic voltametry. Electron transfer reactions were investigated by means of the hexacyanoferrate (II/III) redox system. Further information on the film composition were obtained by Auger electron spectroscopy. On the one hand, electron transfer across the film can occur, but on the other hand, the film is nearly impermeable for hydrogen, even if the hydrogen is in the atomic or protonic state. Cyclic voltamograms show the formation of an oxygen adsorption layer on the film in a range of anodic potential.     

The interfacial kinetics of the reaction of CO2 with nickel: Part II. the decarburization of liquid nickel

The kinetics of decarburization of liquid nickel in CO₂-CO mixtures have been studied at 1400 and 1500°C, using the experimental arrangement of the impinging jet. At carbon concentrations above about 1 wt pct, pressures of CO₂ ⪯ 0.1 atm, and for total gas flow-rates above about 40 l/min (STP) impinging on a metal surface of 2.08 cm², it is concluded that the interfacial reaction step controls the rate. Comparison with isotope exchange studies indicates that dissociative chemisorption of CO₂ is the rate determining step. Rate constants, based on the nominal surface area, are 1.2 ×10⁻³ and 1.4 × 10⁻³ mol/cm² · s · atm at 1400 and 1500°C, respectively. on leave of absence from the Homer Research Laboratories of the Bethlehem Steel Co.     

Specific permeability of partially solidified dendritic networks of Al-Si alloys

Porous dendritic networks of Al-4 pet Si and Al-4 pct Si-0.25 pet Ti alloys with volume fraction solid,g _sA > 0.628 were prepared by removing the segregated interdendritic liquid from partially solidified samples of the alloys. In the equiaxed samples, available channels for flow were predominately between the grains. Specific permeabilities of the porous dendritic networks were measured with a triaxial cell permeameter. Measured values of specific permeability were 1 × 10^-9 to 3.5 × 10^-11 cm² in the Al-4 pet Si alloy for volume fractions solid of 0.655 to 0.94, respectively. Specific permeabilities in Al-4 pct Si-0.25 pct Ti alloy were 4.82 × 10^-10 to 7.6 × 10^-11 cm² for volume fractions solid of 0.628 to 0.837, respectively. For equivalent volume fractions solid, the measured specific permeabilities were consistently lower for the grain refined samples. Flow through the porous dendritic networks obeys D’Arcy’s law and equations derived from the capillaric flow model for volume fraction liquid less than ∼0.35. Formerly a graduate student in the Department of Metallurgy and Materials Science, M.I.T.     

Nitrogen diffusion and solubility in tungsten

The diffusion and solubility of nitrogen in tungsten were determed using an ultrahigh vacuum-and mass-spectrometric technique capable of measuring concentrations of 10^?2 ppm and degassing rates of 10^?3 ppm N per hr. The technique is based on measuring the degassing rate of nitrogen as a function of time from a resistivity heated tungsten wire previously engassed with nitrogen between 1 and 25 torr. The diffusion and solubility constants between 1000° and 1800°C may be summarized by $$D = (2.37 \pm 0.43) \times 10^{ - 3} \exp [( - 35,800 \pm 3900)/RT] cm^2 /\sec ,$$ , and $$S = (0.21 \pm 0.06) \exp [( - 17,600 \pm 5900)/RT] torr \cdot liter cm^{ - 3} torr^{ - 1/2} .$$ . The concentration of nitrogen in tungsten at 760 torr according to these results are 0.4 and 9.2 ppm at 1000° and 2000°C, respectively. The expression for the permeation constants calculated fromD andS is $$K = 5 \times 10^{ - 4} \exp ( - 53,400/RT) torr \cdot liter cm^{ - 1} sec^{ - 1} torr^{ - 1/2} .$$ .