Fundamental studies on the continuous electrolytic reduction of uranyl sulfate

With the aim of developing a process for continuous electrolytic reduction of uranyl sulfate, the effects of temperature and HF concentration on the electrolytic reduction were investigated using both a rectangular cell and a bipolar cell. Some tests on the semicontinuous-type elec-trolytic reduction of U(VI) and the continuous monitoring of U(VI) in a catholyte were carried out to examine the feasibility of a process for continuous electrolytic reduction. The reduction rate and current efficiency of the electrolysis were increased with increasing temperature up to 333 K, due to an increase in the diffusion coefficient of uranyl sulfate. The presence of a small amount of HF in the catholyte improved the polarization characteristics of the electrolysis. The precipitation of uranous sulfate occurring in electrolysis above 323 K was eliminated by the addition of HF of 2.3 g dm^-3. However, the addition of the excess amount of HF causes the precipitation of UF₄ on the cathode surface during electrolysis. Increase in electric current was not effective in the semicontinuous electrolytic reduction of U(VI) for catholytes with high reduction ratios, and a catholyte with a composition of around 50 pct reduction ratio is pref-erable for continuous operation. The concentration of U(VI) in a catholyte can be monitored continuously either in the absence or in the presence of HF by using a device consisting of three small electrodes. Formerly Graduate Student in the Department of Metallurgy, Kyoto University Formerly Director, Mining and Ore Processing Division, Ningyo-toge Works, PNC Japan     

Electrolytic reduction of U(VI) to U(IV) in acidic chloride and acidic sulfate solutions

In order to examine the applicability of the electrolytic reduction process of U(VI) (originally developed for the chloride system by PNC) to the sulfate solution system, a fundamental study was made. In this study, the concentrations of various chemical species in the catholytes were calculated at 298 K at various percentages of uranium reduction, taking the chloro-complex and sulfato-complex formation reactions of uranium into consideration. The polarization characteristics of the electrolytic reduction of uranyl chloride and uranyl sulfate were determined, using titanium and platinum cathodes, respectively, at 303 ± 1 K. In conjunction with this process, the electrical conductivity of the catholyte, the electrical resistivity of the cation exchange membrane, and the diffusion coefficient of uranyl sulfate were also determined.     

Laboratory reduction rate and current efficiency studies of batch type electrolytic reduction of U(VI) in a sulfate system

The electrolytic reduction of U(VI) was investigated to improve the batch type electrolytic reduction of uranyl sulfate. For this purpose, theoretical considerations were made on the time variation of reduction rate and of current efficiency. A monitoring device consisting of two titanium electrodes and one platinum electrode was developed to determine the reduction ratio and to detect the end point of the reduction of U(VI). The monitoring device worked well for these purposes. The reduction rate in a batch type electrolytic reduction of U(VI) at constant current was larger than theoretically expected. This phenomenon was attributed to the increase in the mass transfer rate of U(VI) species toward the cathode due to the disturbance of the concentration boundary layer by hydrogen gas evolved from the cathode surface. The deterioration of titanium cathode experienced in an operating plant was investigated, and was found to be caused by platinum plating onto the titanium cathode surface during the operation. The deteriorated cathode could be restored by immersing it in a hot aqua regia or a mixture of hydrofluoric acid and phosphoric acid for a few minutes.     

Determination of the diffusion coefficients of NiCl2, ZnCl2, and CdCl2 in aqueous solution

The diffusion coefficients of NiCl₂, ZnCl₂, and CdCl₂ in the aqueous solution systems of MC1₂ and MC1₂-HC1 were measured at 298 K using a diaphragm-cell method. The data are listed as a function of molar concentrations of MC1₂ at the HC1 concentrations of 0, 0.1, 0.5, 1.0, and 2.0 mol dm^-3. It was found that the concentration dependencies of the diffusion coefficients for these metal chlorides in single-electrolyte solutions differed from each other. This could be explained in terms of changes in the mean activity coefficients of chloride and in the viscosities of those solutions. The diffusion coefficient of metal chloride in MC1₂-HC1 solution was greatly affected by the HC1 concentration; however, the behavior of the diffusion coefficients varied, depending on the kind of chloride involved. In NiCl₂-HCl solutions, an increase in HC1 con-centration caused a decrease in the diffusion coefficient value, while in ZnCl₂-HCl solutions, the addition of 2 mol dm^-3 HC1 caused an increase in the diffusion coefficient of ZnCl₂. These phenomena are quite different from those of the sulfate systems reported in our previous work. It was also demonstrated that the concentration dependency of the diffusion coefficients of MC1₂ in aqueous MC1₂-HC1 solutions could be attributed to the diffusion potential as well as the changes in the mean activity coefficient and viscosity. Masami Aoki, formerly Graduate Student, Department of Met-allurgy, Kyoto University.     

Determination of the diffusion coefficients of CuSO4, ZnSO4, and NiSO4 in aqueous solution

The diffusion coefficients of CuSO₄, ZnSO₄, and NiSO₄ in the aqueous solution systems of MSO₄ and MSO₄-H₂SO₄ were measured at 298 K using a diaphragm-cell method, and are listed as a function of molar concentrations of MSO₄ and H₂SO₄. It was found that the concentration dependencies of the diffusion coefficients for CuSO₄, ZnSO₄, and NiSO₄ in each single metal sulfate solution are very similar. The presence of H₂SO₄ generally causes a less significant concentration dependency of the diffusion coefficients of MSO₄. The concentration dependencies of the diffusion coefficients of CuSO₄ in aqueous solutions of CuSO₄ and CuSO₄-H₂SO₄ are attributed to the changes in the mean activity coefficient of CuSO₄ and the viscosity of the solutions. formerly Graduate Student, Department of Metallurgy, Kyoto University, Kyoto, Japan     

Calculation of equilibrium species for the aqueous solution systems of UO2SO4−U(SO4)2−H2SO4−HF and UO2Cl2−UCl4−HCl−HF at 298 K

Speciation models for aqueous solutions of UO₂SO₄−U(SO₄)₂−H₂SO₄−HF and UO₂Cl₂−UCl₄−HCl−HF were proposed based on chemical reaction equilibria, mass balances, charge balance, and stoichiometry of UF₄(s). The equilibrium concentrations of uranium and fluoride species in these solutions were calculated at 298 K, and are of relevance to the electrolytic reduction of U(VI), followed by the precipitation of UF₄(s). In these calculations, the reduction ratios of U(VI) were set at 25, 50, 75, and 100 pct. In the sulfate system the stable domains of U⁴⁺, U(SO₄) _n ⁴⁻²ⁿ , UF _n ⁴⁻ⁿ , and UF₄(s) as U(IV) species and UO ₂ ²⁺ , UO₂(SO₄) _n ²⁻²ⁿ , and UO₂F _n ²⁻ⁿ , as U(VI) species are strongly dependent on theC _T(F)/C _T(U(IV)) value. On the other hand, the stable domains of U⁴⁺ UCl³⁺, UF _n ^/4−n , and UF₄(s) as U(IV) species and UO ₂ ²⁺ , UO₂Cl⁺, and UO₂F _n ²⁻ⁿ as U(VI) species are also strongly affected by theC _T(F)/C _T(U(IV)) ratio in the chloride system. The initiation and precipitation of UF₄(s) in both the sulfate and chloride systems are a function of the reduction ratio of U(VI). The higher the reduction ratio, the lower theC _T(F)/C _T(U(IV)) values required. Compared to the chloride system, UF₄(s) precipitation in the sulfate system starts at a lower value ofC _T(F)/C _T(U(IV)). The addition of an excess amount of HF does not cause the dissolution of UF₄(s) precipitates because HF is a weak acid. KOJI SATO, formerly Graduate Student, Department of Metallurgy, Kyoto University, Kyoto, Japan     

Effect the tungsten(VI) and molybdenum(VI) ions on the electrical conductivity of ammonium hydroxide solutions

The effect of the W(VI) and Mo(VI) ions on the electrical conductivity of ammonium hydroxide solutions (7 mol/L) is studied in the tungsten or molybdenum concentration range 0.1–0.5 mol/L and the temperature range 20–60°C. The electrical conductivities of the solutions are shown to depend linearly on the solution temperature and the metal concentration, and the coefficients of linear regression equations are calculated. The temperature coefficients of the electrical conductivities of the solutions are determined.     

Bulk diffusion and solubility of silver and nickel in lead,lead-silver and lead-nickel solid solutions

The results of a study of solubility and bulk diffusion of ¹¹⁰Ag and ⁶³Ni in lead, lead-silver and lead-nickel solid solutions in the temperature range 220 to 88°C are reported. Owing to the low solubility of silver and nickel in lead, Pick's solution corresponding to the boundary condition of a constant concentration of solute at the surface has been used. Depth-profile concentration analysis of ¹¹⁰Ag and ⁶³Ni suggests a fundamental difference between the diffusion mechanisms of silver and nickel. Since silver penetration profiles in pure lead give diffusion coefficients independent of the penetration depth and silver concentration, it is suggested that slight decrease of silver diffusivity in lead-silver solid solutions (which remain within experimental error) have no significance. This implies that the interstitial silver atoms do not associate significantly with each other to form Ag-Ag dimers. In contrast, different behaviors of ⁶³Ni depth profile concentration in pure lead and saturated PbNi solid solutions agree with a strong Ni-Ni interaction leading to the formation of less mobile dimers near the surface in pure lead. The model proposed by Decker to describe the diminution of gold diffusivity in PbAu solid solutions yields H = 9500 cal/mol and S = 0.9 k for the binding enthalpy and entropy of Ni-Ni substitutional dimers.     

Identification of a novel actin binding site within the Dp71 dystrophin isoform

A method is described by which the diffusion coefficients of electrically charged micelles can be determined using micellar electrokinetic chromatography (MEKC). The determination is based on a theory for the dispersion, at low electric field strengths, of analytes that are solubilized by only the micellar phase. The dispersion is represented by contributions from instrumental sources and from longitudinal diffusion. The latter depends on the micellar diffusion coefficient. The theory is used to determine the micellar diffusion coefficients of the surfactant sodium dodecyl sulfate in three solutions having different surfactant concentrations. These diffusion coefficients compare very favorably with those determined by diffusion-ordered two-dimensional NMR spectroscopy. An argument is presented justifying that the diffusion coefficients determined by MEKC are self-diffusion coefficients.     

An increase in purity of ammonium perrhenate solutions with respect to molybdenum(IV) with the sorption recovery of rhenium(VII) from Mo-containing solutions

The possibility of the additional purification of ammonia rhenium desorbates with respect to molybdenum in the course of the sorption recovery of rhenium from Mo-containing solutions with the help of Purolite A170 and Purolite A172 weak base anion-exchange resins is considered. The pH-dependence of sorption of Re(VII) and Mo(VI) on these anion-exchange resins is investigated in static conditions with the 1 M (NH₄)₂SO₄ background in the solution. It is shown that the range of pH, in which anion-exchange resins retain the ability to sorb Re(VII), is also spread to a weakly basic region. A substantial decrease in the adsorption of Re(VII) starts already with an increase in pH above 7.5. The capacity of anion-exchange resins with respect to Mo(VI) starts to decrease noticeably with an increase in pH of solutions above 5.0, and molybdenum almost ceases to sorb by both anion-exchange resins upon reaching pH ~ 7.0. In order to decrease the Mo(VI) content in rhenium desorbates with the sorption recovery of Re(VII) from Mo-containing solutions on weak base anion-exchange resins, the following flowsheet is suggested. Initially, the main amount of sorbed Mo(VI) is desorbed by contacting the saturated anion-exchange resin with the ammonium sulfate solution upon mixture stirring and holding constant pH of the solution in limits of 7.0–7.5 due to the addition of dosed amounts of ammonia solution. Then anion-exchange resin is separated from the ammonium sulfate solution containing Mo(VI), washed with water, and Re(VII) is desorbed by ammonium solution in dynamic conditions. The verification of the proposed method for the resins saturated by sorption from the model solution of the composition, g/L, 98 H₂SO₄, 4 Mo(VI), and 0.5 Re(VII) showed the occurrence of desorption of no less than 90% sorbed molybdenum during the treatment of anion-exchange resins with ammonium sulfate solution. Herewith, concentration ratio Re(VII) : Mo(VI) in ammoniacal rhenium desorbates when using A170 anion-exchange resin increases 11-fold and when using A172 anion-exchange resin, it increases 20-fold compared with that attained without the additional washing of Mo(VI). Losses of Re(VII) with the Mo-containing desorbate (reversible) do not exceed 5.2% of the amount of sorbed Re(VII).     

The Leaching of Hematite in Acid Solutions

The reactions of hematite in aqueous hydrochloric acid, perchloric acid, and sulfuric acid solutions with or without the addition of common or uncommon salts were studied using monosized particulates in a well-stirred reactor and dilute solid concentration to obtain fundamental details of the reaction kinetics. The experimental rate data suggest that the entire leaching reaction is controlled by a chemical process. The leaching rate of hematite was seen to be first order with respect to hydrogen ion activity, a(H⁺), in hydrochloric acid or perchloric acid solutions, with or without the addition of common salts, while the rate was of a half order in sulfuric acid solutions with or without the addition of sodium sulfate. A theoretical analysis showed that the anions next to the surface in the double layer were chloride ion and perchlorate ion in hydrochloric acid and perchloric acid solutions, respectively, and sulfate ion in sulfuric acid solutions, with or without the addition of sodium sulfate. The fact that the leaching rates of hematite were quite different in various acids having identical α(H⁺ values indicates the importance of anion adsorption. The dependency of the leaching rate upon α(H⁺) appeared to be controlled by adsorbed anions next to the surface in the double layer. TAKUMI MISHIMA, formerly Graduate Student, Kyoto University     

生物聚合硫酸铁去除U(VI)的絮凝试验

采用生物聚合硫酸铁絮凝剂对低浓度含铀废水进行絮凝试验研究,考察U(VI)溶液pH、絮凝剂投加量以及U(VI)初始浓度对絮凝效果的影响。结果表明,反应最佳pH范围在5~7,反应平衡时间为5 min,其絮凝过程符合Lagergren准二级反应动力学模型。含铀废水经生物聚合硫酸铁絮凝处理过后,残余铀浓度低于《铀加工与燃料制造设施辐射防护规定》（EJ 1056-2005）中的排放限值（0.05 mg/L）。     

Solubility of uranous sulfate in aqueous sulfuric acid solution

To provide important thermodynamic data for use in uranium hydrometallurgy, solubilities of uranous sulfate were determined as a function of free acid concentration and temperature. Two sets of experiments were performed in this study. One set was the precipitation experiments of uranous sulfate crystals, in which concentrated uranous sulfate solution was mixed with sulfuric acid solution of suitable concentration. The other set was the dissolution experiments of uranous sulfate crystals in aqueous sulfuric acid solutions. It is noteworthy that good agreement exists between the solubilities determined by the two methods. At elevated temperatures, say, 363 K, the presence of free sulfuric acid is required to avoid precipitation of uranous hydroxide resulting from the hydrolysis of uranous sulfate. Generally speaking, however, an increase in free sulfuric acid concentration results in a slight decrease in uranous sulfate solubility. The elevation of solution temperature causes a decrease in solubility of uranous sulfate. It should be noted that the solid uranous sulfates equilibrated with saturated solutions at 298 K were U(SO₄)₂ 2H₂O in dilute sulfuric acid solution and U(SO₄)₂ 4H₂O in concentrated sulfuric acid solution, while those at 333 K and 363 K were mainly U(SO₄)₂ 4H₂O. Formerly Director, Mining and Ore Processing Division, Ningyo-toge Works, Power Reactor and Nuclear Fuel Development Corporation, is Director, Ningyo-toge Nuclear Service Company, Tomata-gun, Okayama Prefecture 708-06, Japan.     

New agents for in vivo chelation of uranium(VI): efficacy and toxicity in mice of multidentate catecholate and hydroxypyridinonate ligands

Soluble uranyl ion [UO2(2+), U(VI)] is a kidney poison. Uranyl ion accumulates in bone, and the high specific activity uranium isotopes induce bone cancer. Although sought since the 1940's, no multidentate ligand was identified, until now, that efficiently and stably binds U(VI) at physiological pH, promotes its excretion, and reduces deposits in kidneys and bone. Ten multidentate ligands patterned after natural siderophores and composed of sulfocatechol [CAM(S)], carboxy-catechol [CAM(C)], or hydroxypyridinone [Me-3,2-HOPO] metal-binding units have been tested for in vivo chelation of U(VI). Ligands were injected intraperitoneally (i.p.) into mice 3 min after intravenous (i.v.) injection of 233U or (232+235)U as UO2Cl2 [ligand-to-metal molar ratio 75 to 92]. Regardless of backbone structure, denticity, or binding unit, all 10 ligands significantly reduced kidney U(VI) compared with controls or with mice given CaNa3-DTPA, and four CAM(S) or CAM(C) ligands also significantly reduced skeleton U(VI). Several ligands removed U(VI) from kidneys, when injected at 1 or 24 h. Injected at molar ratios > or = 300, 5-LIO(Me-3,2-HOPO) and TREN-(Me-3,2-HOPO) reduced kidney U(VI) to about 10% of control. Given orally to fasted mice at molar ratios > or = 300, those ligands significantly reduced kidney U(VI). In mice injected i.v. with 0.42 micromol kg(-1) of 235U and given 100 micromol kg(-1) of one of those Me-3,2-HOPO ligands i.p. daily for 10 d starting at 1 h after the U(VI)) loss of kidney U(VI) was greatly accelerated, and the kidneys of treated mice showed no microscopic evidence of renal injury. Crystals of uranyl chelates with linear tetradentate ligands containing bidentate Me-3,2-HOPO groups demonstrate a 1:1 structure. Considering low toxicity, effectiveness, and reasonable cost, the structurally simple linear tetradentate ligands based on the 5-LI backbone (diaminopentane) offer the most promising approach to a clinically acceptable therapeutic agent for U(VI). Work is in progress to identify the most suitable CAM or HOPO binding unit(s).     

Interdiffusion in copper(rich)-chromium solid solutions

Abstract

Interdiffusion was studied in copper (rich)-chromium solid solutions in the composition range approximately 0.2 to 0.8 at% Cr and in the temperature range 852°C – 1050°C. Diffusion couples consisted of cylindrical sections of OFHC copper electroplated with pure chromium. The concentration profiles of the diffusion couples were determined using an electron probe micro analyzer, and the diffusion coefficients were determined by fitting the error function solution of the diffusion equation to the experimental curves. The Arrhenius plot of the diffusion coefficients obtained for five different temperatures gives 48.4 ± 1.2 kcal/mole and 1.11_?0.44^+0.78 cm²/sec for the activation energy (E_D) and frequency factor (D_o) respectively (where the limits given are standard deviations for a 99 per cent confidence level), and approximate published chromium tracer diffusion and copper self diffusion results.

Résumé

Les auteurs ont étudié l'interdiffusion dans des solutions solides, riches en cuivre et con tenant environ 0.2 à 0.8% atomique de chrome, pour une gamme de temp;amp;#x00E9;ratures de 852 agrave; 1050°c. Les couples de diffusion étaient des sections cylindriques de cuivre OFHC électroplaqués avec du chrome pur. Les profils de concentration des couples ont été mesurés par microsonde électronique et les coefficients de diffusion ont été déterminés en lissant les courbes expérimentales grâce à la solution de la fonction erreur pour l'équation de diffusion. La courbe déArrhenius des coefficients de diffusion obtenus à cinq températures différentes donne une énergie d'activation (ED) de 48.4 ± 1.2 kcal/mole et un facteur de fréquence (Do) de 1.11 _?0.44^+0.78 cm²/sec. (oú les limites indiquées sont les déviations standards pour un niveau de con fiance de 99%). Ces valeurs sont approximativement égales à celles publiées pour la diffusion de traceur de chrome et pour l'autodiffusion du cuivre.     

Copper extraction by LIX 64N: Comparison of chloride and sulfate solutions according to pH and acid concentration

Thermodynamic data for distribution coefficients and maximum loading are reported for cupric ion extraction by LIX 64N solution in xylene. The influence of ionic strength and chloride concentration in aqueous phase on distribution coefficients and measured pH is given and interpreted. The comparison of chloride and sulfate aqueous solutions shows that if extraction seems higher in chloride solution at the same measured pH, it is of the same order of magnitude at the same true concentration of H⁺ and it is higher in sulfate solution at the same apparent concentration of acid.     

Electrical conductivity of acidic sulfate solution

The electrical conductivities of the aqueous solution system of H₂SO₄-MSO₄ (involving ZnSO₄, MgSO₄, Na₂SO₄, and (NH₄)₂SO₄), reported by Tozawaet al., were examined in terms of a (H₂O) and H⁺ ion concentration. The equations to compute the concentrations of various species in aqueous sulfuric acid solutions containing metal sulfates were derived for a typical example of the H₂SO₄−ZnSO₄−MgSO₄−(Na₂SO₄)−H₂O system. It was found that the H⁺ ion concentrations in concentrated sulfuric acid solutions corresponding to practical zinc electrowinning solutions are very high and remain almost constant with or without the addition of metal sulfates. The addition of metal sulfates to aqueous sulfuric acid solution causes a decrease in electrical conductivity, and this phenomenon is attributed to a decrease in water activity, which reflects a decrease in the amount of free water. The relationship between conductivity and water activity at a constant H⁺ ion concentration is independent of the kind of sulfates added. On the other hand, any increase in H⁺ ion concentration results in an increase in electrical conductivity. A novel method for the prediction of electrical conductivity of acidic sulfate solution is proposed that uses the calculated data of water activity and the calculated H⁺ ion concentration. Also, the authors examined an extension of the Robinson-Bower equation to calculate water activity in quarternary solutions based on molarity instead of molality, and found that such calculated values are in satisfactory agreement with those determined experimentally by a transpiration method.     

Recovery of uranium from phosphoric acid by means of supported liquid membranes

U(VI) was transported at 23 ± 1°C from 5–6 M phosphoric acid solutions through liquid membranes of kerosene solutions of di(2-ethylhexyl) phosphoric acid and trioctyl phosphine oxide (D2EHPA/TOPO) supported on porous polytetrafluoroethylene to a solution of phosphoric acid of equal or greater molarity containing ferrous ion as a reducing agent. The ferrous ion could be omitted when the higher molarity acid was used. The uranium flux was proportional to the U(VI) concentration. The overall resistivity of the membranes to uranium flux had a diffusional component that was proportional to the membrane thickness and an interfacial component that resulted from rate-limiting uranium complexation/decomplexation kinetics. The interfacial component accounted for over 80% of the resistivity of a membrane 75 μm thick. Increasing the temperature to 60°C only slightly diminished the interfacial resistivity. A theoretical model was constructed that accommodated data obtained from uranium transport through the membranes and through quiescent layers of phosphoric acid and D2EHPA/TOPO in kerosene. The average uranium flux from simulated solutions of wet-process phosphoric acid at 90% uranium transfer was estimated to be 1.3 × 10^?11 mol cm^?2 sec^?1, or 0.09 lb ft^?2 yr^?1. The flux was judged to be too low for supported liquid membranes to be competitive with liquid/liquid extraction for recovery of uranium from wet-process phosphoric acid.     

Dissolution of malachite in aqueous ethylenediaminetetraacetate solution

The kinetics of malachite dissolution in aqueous ethylenediaminetetraacetate (EDTA) solution has been investigated in the temperature range of 298 to 318 K. The dissolution rate of malachite determined under the present set of experimental conditions was found to be independent of agitation speed. The dissolution rate increased with increasing EDTA concentration, but leveled off at higher concentrations. At constant EDTA concentration, an increase in dissolution rate was detected at higher temperatures. A dissolution mechanism involving Langmuir-type EDTA adsorption was proposed, in which the dissolution rate of malachite is controlled by the removal of Cu(II)-EDTA complex from the malachite lattice. The proposed mechanism can explain the dependency of the dissolution rate on EDTA concentration. The activation energies determined at pHs 5, 7.5, and 10 were found to be 51.4, 50.2, and 57.5 kJ mol⁻¹, respectively. The calculated enthalpy changes of EDTA adsorption equilibrium were −43.2, −35.2, and −45.0 kJ mol⁻¹ for pHs 5, 7.5, and 10, respectively. These values are in agreement with the proposed mechanism. Formerly Graduate Student, Department of Metallurgy, Kyoto University, is Formerly Professor, is Professor Emeritus, Department of Metallurgy, Kyoto University.     

Diffusion of silicon in aluminum

Interdiffusion coefficients in Al-Si alloys were determined by Matano’s method in the tem-perature range from 753 to 893 K with the couple consisting of pure aluminum and an Al-Si alloy. Temperature dependence of the impurity diffusion coefficients of Si in Al, obtained by extrapolation of the concentration dependence of the interdiffusion coeffi-cient to zero mole fraction of Si, is given by the following equation: DSi/Al = (2.02^+0.97 _-0.66 × 10^-4 exp [-(136 ±3) kJ mol^-1/RT] m²/s. p ] The Kirkendall marker was found to move toward the Si-rich side, indicating that the Si atom diffuses faster than the Al atom in Al-Si alloys. From the interdiffusion coeffi-cient and the marker shift, the intrinsic diffusion coefficients were calculated. The difference in the activation energies (ΔQ) between the impurity diffusion of Si in Al and the self-diffusion of Al was estimated by means of the asymptotic oscillating po-tential and the Le Claire theory. The calculated value of ΔQ is in fair agreement with the experimental value. The vacancy-solute binding energy for Si in Al was also dis-cussed based on the diffusion data. formerly Undergraduate Student, Tohoku University