Diffusion-couple Technique Research Articles

Br diffusion in phonolitic melts: Comparison with fluorine and chlorine diffusion

Abstract Bromine diffusion was measured in two natural phonolitic melts: (1) a K2O-rich (~10 wt%) one synthesized from the white pumice phase of the 79 AD eruption of Vesuvius (Italy), and (2) a Na2O-rich (~10 wt%) one corresponding to the most differentiated melt of the 12 000 BC eruption of the Laacher See (Germany). Experiments were performed at 0.5 and 1.0 GPa, 1250 to 1450 °C, at anhydrous and hydrous (2.65 ± 0.35 wt% of dissolved water) conditions. Experiments conducted with the diffusion-couple technique in the piston cylinder were performed with only bromine diffusing and with the simultaneous diffusion of a halogen mixture (F, Cl, Br) to evaluate the interactions between the halogens during diffusion. The diffusion profiles of Br were measured by X-ray fluorescence using synchrotron radiation microprobe (SYXRF), ID18F, at the European Synchrotron Radiation Facility (ESRF, France). Bromine diffusion displays Arrhenian behavior under anhydrous conditions that is similar when it diffuses alone and when it diffuses with F and Cl. The Br diffusion coefficients range between 2 × 10–12 m2/s at 1250 °C and 1.5 × 10–11 m2/s at 1450 °C for the Na-rich melt and between 3 × 10–12 m2/s at 1250 °C and 2.5 × 10–11 m2/s at 1450 °C for the K-rich melt, at 1.0 GPa. Although Br mobility is independent of F and Cl in anhydrous phonolitic melts, its behavior may be dependent on the dominant alkali in the melt, as previously observed for Cl, but not F. For hydrous experiments, although the data are scattered, the Br diffusivity increases slightly with water and the Na/K ratio seems to influence Br diffusivity. Similarly to noble gases, halogen diffusivity at a given temperature in the phonolitic melts appears related to the ionic porosity of the silicate structure. Compared to basaltic melt, Br diffusivities are approximately one order of magnitude lower in the Na-phonolite melt, because of the difference of the pre-exponential factor. Br mobility appears to be decoupled from melt viscosity, considering the results here.

Investigation of Interface Reaction of WC/Co Using Diffusion Couple Technique

Investigation of Interface Reaction of WC/Co Using Diffusion Couple Technique

Experimental Diffusion Research on BCC Ti-Mn Binary and Ti-Al‐Mn Ternary Alloys

Interdiffusion in the BCC phase of the Ti-Mn binary and Ti-Al-Mn ternary systems was investigated between 1273 and 1473 K by applying the diffusion-couple technique. The local chemical compositions within the interdiffusion zone of the diffusion couples were acquired by electron probe microanalysis (EPMA). The raw composition profiles were then analytically represented by the error function expansion, which allow the ternary inter- and impurity diffusivities to be appropriately extracted by the Whittle–Green and generalized Hall methods, respectively. It was demonstrated that, all four diffusion coefficients $$\mathop {\tilde{D}}\nolimits_{ij}^{k}$$ (i, j = Al, Mn), both main and cross ones, increase with increasing the composition of diffusing specie at 1473 K, whereas at 1273 K $$\mathop {\tilde{D}}\nolimits_{\text{MnMn}}^{\text{Ti}}$$ and $$\mathop {\tilde{D}}\nolimits_{\text{MnAl}}^{\text{Ti}}$$ are enhanced by the addition of diffusing specie Mn but $$\mathop {\tilde{D}}\nolimits_{\text{AlAl}}^{\text{Ti}}$$ and $$\mathop {\tilde{D}}\nolimits_{\text{AlMn}}^{\text{Ti}}$$ are in weak dependence with the Al content. A complete comparison among seven Ti-Al-X (Ni, Co, Fe, Mn, Cr, V and Mo) ternary systems reveals that the order of the average main interdiffusion coefficients $$\overline{{\mathop {\tilde{D}}\nolimits_{\text{XX}}^{\text{Ti}} }}$$ (X = Ni, Co, Fe, Mn, Cr, V and Mo) exhibits DNi > DCo > DFe > DMn > DCr > DV > DMo.

Experimental Diffusion Research on BCC Ti-Al-Sn Ternary Alloys

The interdiffusion behavior of BCC alloys of Ti-Sn binary and Ti-Al-Sn ternary was investigated at 1473 K (1200 °C) by the diffusion-couple technique. The extracted interdiffusion coefficients of Ti-Sn binary was found to gradually increase with increasing the Sn content, whereas those of Ti-Al-Sn ternary, either main or cross, greatly enhance as the content of diffusing species increases. By a way of comparison, Sn diffusion in Ti-Al-Sn ternary was discerned to occur by the vacancy mechanism.

Thermodynamic Assessments of the Ni-Cr-Ti System and Atomic Mobility of Its fcc Phase

Thermodynamic assessments have been performed for the Ni-Ti binary and Ni-Cr-Ti ternary system by the CALPHAD method. Combining the experimental and ab initio calculated data on the enthalpies of formation of Ni-Ti compounds, a better description for Ni-Ti phases has been obtained. Based on the new assessments of the binary sub-systems and the recent experimental data on phase equilibria, a reassessment of the Ni-Cr-Ti system was carried out. Apart from the thermodynamic assessments, the interdiffusion coefficients for the fcc phase of the binary Ni-Ti were re-optimized and the ternary Ni-Cr-Ti system were determined experimentally over a temperature range from 1123 to 1273 K employing the diffusion-couple technique. Subsequently, atomic mobility data for the fcc phase of the Ni-Cr-Ti system were assessed and most diffusivity data were satisfactorily described.

Mapping of Diffusion and Nanohardness Properties of Fcc Co-Al-V Alloys Using Ternary Diffusion Couples

Ternary diffusion behavior in Co-Al-V ternary alloys was investigated at 1373 K and 1473 K (1100 °C and 1200 °C) by the solid-state diffusion-couple technique. The extraction and interpolation of diffusion data allows the diffusion properties of Fcc Co-Al-V alloys to be mapped in the composition arrays of Al and V. A full picture of the diffusion properties was then constructed by interpolating all accessible interdiffusivities and impurity diffusivities of Co-Al binary and Co-Al-V ternary with a Redlich–Kister polynomial, in a graphic manner depicting a rapid increase of Al diffusion with increasing Al and a weak decrease with the V addition alone. Further incorporation of a nanoindentation technique enables the nanohardness property of the Co-Al-V fcc alloys to be screened in the Al and V arrays. The hardenability in the Co-Al-V alloy system has been evidenced; specifically, the alloy arrays containing higher contents of V, being solution-and-quenching processed, exhibit more effective strengthening than those with the addition of Al. The discovery of Co-Al-V alloys with comparable nanohardness but differing alloy compositions could facilitate the strengthening design of next generation Co-based alloys.

Diffusion Research in BCC Ti-Al-Ni Ternary Alloys

Diffusion behavior in the BCC Ti-Al-Zr ternary alloys was experimentally investigated at 1273 K (1000 °C) and 1473 K (1200 °C) by means of the diffusion-couple technique. Upon the Whittle-Green and generalized Hall methods, the inter- and impurity diffusion coefficients were respectively extracted from the composition profiles acquired by the electron microprobe analysis (EPMA) and subsequently represented by the error function expansion. The extracted main interdiffusion coefficient $$\tilde{D}_{\text{AlAl}}^{\text{Ti}}$$ increases with increasing the content of either Al or Zr, and the increase is appearing more considerably at the higher temperature. However, $$\tilde{D}_{\text{ZrZr}}^{\text{Ti}}$$ was noticed to decrease with the increase of Al and Zr contents at 1273 K (1000 °C) while there is an upward trend at 1473 K (1200 °C). The impurity diffusion coefficients of Al in Ti-Zr binary alloys, $$D_{{{\text{Al}}\left( {\text{Ti - Zr}} \right)}}^{*}$$ , and of Zr in Ti-Al binary alloys, $$D_{{{\text{Zr}}\left( {\text{Ti - Al}} \right)}}^{*}$$ , increase with increasing the Zr and Al contents respectively. A comparison of average main interdiffusion coefficient $$\overline{{\tilde{D}_{\text{XX}}^{\text{Ti}} }}$$ made among ten Ti-Al-X ternary systems suggests that the Zr diffusion is most comparable to Cr and could operate via a vacancy-controlled mechanism.

Role of Zr on growth kinetics and microstructural evolution of the superconductor V3Ga by the bronze technique

The influence of Zr on the growth of V3Ga is studied by a diffusion-couple technique mimicking the bronze method for superconductor production. Systematic quantitative evaluation on critical-current-density-dependent metallurgical parameters is addressed. A mixture of V3Ga and Zr-containing phases develops at the interdiffusion zone. Smaller grains on Zr addition cause higher growth kinetics owing to grain-boundary diffusion of Ga. The grains are oriented randomly irrespective of Zr addition. Refined microstructure, second-phase, and higher growth kinetics suggest a beneficial role of Zr for the better functioning of the superconductor.

Experimental Investigation and Computer Simulation of Diffusion in Fe-Mo and Fe-Mn-Mo Alloys with Different Optimization Methods

In order to simulate the diffusional phase transformations involving the fcc and bcc phases for microalloyed steels, the diffusion mobilities for fcc and bcc Fe-Mo and Fe-Mn-Mo alloys were experimentally investigated and critically assessed. The diffusion-couple technique was employed to extract the interdiffusion coefficients in Fe-Mo and Fe-Mn-Mo alloys with the Sauer–Freise and Whittle–Green methods. Based on the present experimental interdiffsivities, the mobility parameters for the fcc and bcc phases in the Fe-Mo and Fe-Mn-Mo systems were optimized using the traditional method. Simultaneously, a direct method was developed and utilized to directly fit mobilities to the diffusion profiles rather than the diffusivities in the present work. The satisfactory description of the diffusion behavior in the Fe-Mo and Fe-Mn-Mo systems has confirmed the reliability of the direct method. Particularly, the two sets of diffusion mobilities obtained with both methods could simulate the diffusion phenomenon between the fcc and bcc phases in the Fe-Mo and Fe-Mn-Mo systems successfully.

Thermodynamic assessment of the Fe–Mn–Si system and atomic mobility of its fcc phase

Thermodynamic optimizations have been performed for the Mn–Si binary and Fe–Mn–Si ternary system by the CALPHAD method. Combining the experimental and ab initio calculated data on the enthalpies of formation of manganese silicides, a better description for those silicides has been obtained. Almost all the latest data on phase equilibria in the Mn–Si system were reproduced satisfactorily. Based on the new assessments of the binary sub-systems and the recent experimental data on phase equilibria including martensitic transformations, a reassessment of the Fe–Mn–Si system has been carried out. The optimized parameters of the Fe–Mn–Si system successfully reproduced phase equilibrium data, as well as the T0 temperature between the fcc and hcp phases. Apart from the thermodynamic evaluation, the interdiffusion coefficients for the fcc phase of the binary Fe–Si and ternary Fe–Mn–Si system have been determined experimentally over the temperature range from 1273K to 1473K employing the diffusion-couple technique. Subsequently, atomic mobility data for the fcc phase of the Fe–Mn–Si system have been assessed and most diffusivity data have been satisfactorily described.

Determination of phase equilibria in the Co-rich Co–Al–W ternary system with a diffusion-couple technique

Phase equilibria in the Co-rich Co–Al–W ternary system were determined with a unique diffusion-couple technique in which Co–27Al and Co–15W binary alloys (at. %) were first coupled for interdiffusion and then heat-treated for precipitation. After a diffusion process at 1300 °C for 20 h, concentration gradients of Al and W were formed in the γ-Co(A1) matrix in the vicinity of the coupled interface. After a heat treatment at 900 °C for 500 h the γ′-Co 3(Al,W)(L1 2) phase was formed with a coarsened shape in contact with the γ, CoAl(B2) and Co 3W(D0 19) phases. Additionally, it appeared with a submicron cuboidal shape within the γ matrix. After 2000 h, however, the coarsened γ′ phase became infrequent and the three phases of γ, CoAl and Co 3W came into frequent contact with each other. These results clearly demonstrate that the γ′ phase is metastable and the three phases of γ, CoAl and Co 3W are thermodynamically in equilibrium at 900 °C in the Co–Al–W ternary system.

F and Cl diffusion in phonolitic melts: Influence of the Na/K ratio

Fluorine and chlorine diffusion were measured in two natural phonolitic melts, from Vesuvius (Italy) and from Laacher See (Germany), at 0.5 and 1.0 GPa, between 1250 and 1450 °C at anhydrous conditions and with about 2 and 5 wt.% of dissolved water. The two different starting materials allow us to investigate the alkali effect, Na vs. K, on halogen diffusion. One composition was a K-rich (~ 10 wt.%) phonolitic melt corresponding to the white pumice phase of the 79 AD eruption of Vesuvius, and the other a Na-rich (~ 10 wt.%) phonolitic melt corresponding to most differentiated melt of the 12,000 BC eruption of Laacher See. The diffusion-couple technique in a piston cylinder was used for the experiments. Experiments were performed with only one halogen diffusing and with the simultaneous diffusion of a halogen mixture (F, Cl, Br) in order to evaluate the interactions between the halogens during diffusion. Diffusion coefficients for F range between 2 × 10 − 11 m 2/s at 1250 °C and 7 × 10 − 11 m 2/s at 1450 °C for the Na-rich melt and between 1 × 10 − 11 m 2/s at 1250 °C and 8 × 10 − 11 m 2/s at 1450 °C for the K-rich melt at anhydrous conditions. Diffusion coefficients for Cl range between 2 × 10 − 12 m 2/s at 1250 °C and 1 × 10 − 11 m 2/s at 1450 °C for the Na-rich melt and between 7 × 10 − 12 m 2/s at 1250 °C and 2 × 10 − 11 m 2/s at 1450 °C for the K-rich melt at anhydrous conditions. Fluorine diffusivity is higher than Cl in the Na-rich-phonolitic melt by one order of magnitude, whereas in the K-rich-phonolitic melt F and Cl diffusivities are similar. The effect of water is significant for Cl in both Na-rich and K-rich melts: the addition of water enhances Cl diffusivity by up to one order of magnitude, but water does not significantly affect F diffusion. F and Cl diffusivities always differ from one another in the same phonolitic melt composition. F diffusivities are similar in both compositions. Conversely, Cl diffusion depends upon the dominant alkali. These results evidence that halogen diffusivity may represent a limiting factor for their degassing during rapid syneruptive decompression and vesiculation of H 2O-rich-phonolitic melts. The contrasting volatile diffusivities of F and Cl in silicate melts during magma vesiculation may be a key, controlling factor of the composition of the vapour phase (bubbles) produced. Such diffusion controlled degassing model may explain the absence of F and Cl degassing observed during the 79AD eruption of Vesuvius.

Halogen diffusion in a basaltic melt

The diffusion of the halogens fluorine, chlorine and bromine was measured in a hawaiitic melt from Mt. Etna at 500 MPa and 1.0 GPa, 1250 to 1450 °C at anhydrous conditions; the diffusion of F and Cl in the melt was also studied with about 3 wt% of dissolved water. Experiments were performed using the diffusion-couple technique in a piston cylinder. Most experiments were performed with only one halogen diffusing between the halogen-enriched and halogen-poor halves of the diffusion couple, but a few experiments with a mixture of halogens (F, Cl and Br) were also performed in order to investigate the possibility of interactions between the halogens during diffusion. Fluorine and chlorine diffusivity show a very similar behavior, slightly diverging at low temperature. Bromine diffusion is a factor of about 2–5 lower than the other halogens in this study. Diffusion coefficients for fluorine range between 2.3 × 10 −11 and 1.4 × 10 −10 m 2 s −1, for chlorine between 1.1 × 10 −11 and 1.3 × 10 −10 and for bromine between 9.4 × 10 −12 and 6.8 × 10 −11 m 2 s −1. No pressure effect was detected at the conditions investigated. In experiments involving mixed halogens, the diffusivities appear to decrease slightly (by a factor of ∼3), and are more uniform among the three elements. However, activation energies for diffusion do not appear to differ between experiments with individual halogens or when they are all mixed together. The effect of water increases the diffusion coefficients of F and Cl by no more than a factor of 3 compared to the anhydrous melt ( D F = 4.0 × 10 −11 to 1.6 × 10 −10 m 2 s −1; D Cl = 3.0 × 10 −11 to 1.9 × 10 −10 m 2 s −1). Comparing our results to the diffusion coefficients of other volatiles in nominally dry basaltic melts, halogen diffusivities are about one order of magnitude lower than H 2O, similar to CO 2, and a factor of ∼5 higher than S. The contrasting volatile diffusivities may affect the variable extent of volatile degassing upon melt depressurization and vesiculation, and can help our understanding of the compositions of rapidly grown magmatic bubbles.

Water diffusion in natural potassic melts

Abstract Water diffusion experiments were performed on a trachytic melt from the Agnano-Monte Spina explosive eruption (Phlegrean Fields, South Italy). Experiments were run in a piston cylinder apparatus at 1 GPa pressure, at temperatures from 1373 to 1673 K and for durations of 0 to 255 s, using the diffusion-couple technique. Water concentration profiles were measured by Fourier transform infrared spectrometry. Water diffusion coefficients at different temperatures and water concentrations were calculated from the total water profiles, using the Boltzmann-Matano technique. Over the investigated range of temperatures and water concentrations, the diffusivity of water in potassic melts ( D water ), m 2 /s can be described by Arrhenius equations that can be generalized for water concentrations between 0.25 and 2 wt% as follows: D water = exp ( − 11.924 − 1.003 ln C H 2 O ) exp ( − ( exp ( 11.836 − 0.139 ln C H 2 O ) ) RT ) where C H 2 O is the water concentration in wt%, R is 8.3145 (J K −1 mol. −1 ) and T is the temperature in Kelvin. Water diffusivities in trachytic melts were compared with water diffusivities in rhyolitic and basaltic melts. The activation energies for water diffusivity in trachyte and basalt are comparable, and higher than the haplogranitic melt. This results in a convergence of water diffusion coefficients in all melts at lower (magmatic) temperatures.

Co-Fe-Gd (Cobalt-Iron-Gadolinium)

mal sections that agree reasonably well with the limited experimental results available. The isothermal section at 1050 C determined by [1990Ati] (reviewed in [1992Rag]) shows that the pairs Fe17Gd2-Co17Gd2 ,F e 3Gd-Co3Gd, and Fe2Gd-Co2Gd form Binary Systems continuous solid solutions. These solutions are denoted here by M17Gd2 ,M 3Gd, and M2Gd, respectively. [1995Liu] evaluated the thermodynamic properties of this ternary system, The Co-Fe phase diagram was reviewed by [1993Nis]. The fcc phase forms through a peritectic reaction at 1499 using the experimental results of [1990Ati]. [1995She] studied the phase equilibria in this system by Cbetweenbccandliquid.TheCoandFeformacontinuous fcc solid solution () over a wide range of temperature, the diffusion-couple technique and presented five isothermal sections at 1200, 1050, 1000, 950, and 900 C. Tie-line comwhich transforms to bcc () below 985 C. An order-disorder transformation occurs in the equiatomic phase at 730 C positions were determined for the pairs -M17Gd2 ,M 17Gd2M3Gd, M3Gd-M2Gd, M17Gd2-Co5Gd, Co5Gd-Co7Gd2, and and the CsCl-type ordered phase exists over a wide composition range below this temperature. The Co-Gd phase diagram Co7Gd2-M3Gd. Using the experimental data of both [1990Ati] and [1995She], [1998Su] optimized the thermodywas calculated by [1995Liu]. There are seven intermetallic phases in this system: Co17Gd2 ,C o 5Gd, Co7Gd2 ,C o 3Gd, namic properties of the ternary system and recalculated the ternary equilibria at 1200, 1050, 1000, 950, 900, and 427 Co2Gd, Co3Gd4, and CoGd3. With the exception of Co17Gd2, theyformthroughperitecticreactions.Thefinalsolidification C. The calculated and experimental tie-line compositions agree reasonably well. The computed isothermal sections at

The synthesis of (Bi,Pb)2Sr2Ca2Cu3Ox (2223) by the diffusion-couple technique

The formation of (Bi,Pb)2Sr2Ca2Cu3Ox (2223) by various reaction paths was studied by diffusion experiments on selected composition couples in the Bi(Pb)-Sr-Ca-Cu-O system at 810-850 degrees C and analysed by electron microscopy and EDX. An important grain growth of 2223 at the interface between solid pellets was observed in the following diffusion-couple configurations: (i) (Ca0.5Sr0.5)CuO2+(Bi,Pb)2CuO4 at 845 degrees C and (ii) (Ca0.85Sr0.15)CuO2 ("n= infinity " phase)+Pb-doped 2212 at 835-852 degrees C. The presence of a liquid phase is noted in both cases. 2223 crystals up to 100 mu m long growing perpendicular to the diffusion front were obtained in 4 d for both composition assemblages. The critical temperature of the diffusion zone product was 108 K. In case (ii), it is shown that the (Ca0.85Sr0.15)CuO2 phase is unstable with respect to (Ca,Sr)2CuO3 and CuO in the temperature range 827-852 degrees C.

Experimental and theoretical study of the phase equilibria in the fe-ni-w system

Experimental data on the Fe-Ni-W phase diagram have been obtained at 1273, 1373, 1473, 1573, and 1673 K by means of the diffusion-couple technique and electron microprobe analysis. These data were combined with selected literature information on the liquidus surface and the complete Fe-Ni-W phase diagram was evaluated. The evaluation was performed by analyzing the experimental data in terms of thermodynamic models for the Gibbs energy of the various phases and by computer-optimizing the model parameters involved. The optimized phase diagram satisfactorily describes the available data.

拡散対法によって得られた高温融体の拡散データーの解析法

A method of analysis has been proposed in this paper to determine the diffusion coefficient of a component in a high-temperature melt from the data on concentration profiles measured using the so-called diffusion-couple technique. The procedure, which is believed to be simpler and more reliable than the previous methods, is summarized as follows:(1) Plot the dimensionless concentration (C−C0)⁄(C1−C0) against the corresponding distance X on a probability paper. A straight line will be produced if the experiment is successfully done.(2) Obtain the statistical variance σ2 of the distance travelled by diffusing particles from the slope of the straight line; σ2=1/(slope)2.(3) The variance is expected to be measured for several different diffusion times. If the variance is plotted against the experimental time, a straight line will be yielded. The diffusion coefficient D to be obtained is then a half of the slope of the line.The analysis is illustrated using the data by the present authors on the diffusion of manganese in molten iron saturated with carbon. Comparison of the proposed method with those by previous authors is also given.