Activation Energy For Self-diffusion Research Articles

Phase transformation and strengthening of the gas-atomized FeCoCrNiMo0.5Al1.3 high-entropy alloy powder during annealing

Phase evolution and strengthening of the FeNiCoCrMo0.5Al1.3 powder alloy produced via inert gas atomization and annealed in the temperature interval of 300–800 °C have been studied by X-ray diffraction, scanning electron microscopy, energy dispersive X-ray spectroscopy, and microhardness testing. It was found that annealing at 300–600 °C leads to an increase of the element segregations between the several solid solutions with a rise of the lattice misfit (ε) to 1.5 % and microhardness growth to 1070 HV. It was assumed that elastic stress caused by the element partitioning is the main strengthening mechanism: microhardness rises linearly with misfit rise with dHV/dε = 43400 MPa. Sigma arises after the maximum elastic deformation (in 1.5 %) was reached. Formation of the dispersed coherent sigma phase in the annealing interval 600–800 °C results in the microhardness rise. Oxidation that began at 800 °C in 27 h is accompanied with FCC formation due to a depletion of the B2 in Al caused by Al2O3 formation. Estimation of the activation energy of the initial stage of the solid solution decomposition gives a very low value in 0.65eV, apparently caused by the high concentration of quenched vacancies. The activation energy of sigma formation approximately coincides with the activation energy of self-diffusion in BCC metals (about 2.60 eV).

Investigating evolution of microstructures and mechanical properties in metal injection molding Ti6Al4V with sintering processing by quasi in-situ strategy

Manipulation of the pore defects, adjustments of α/β phases fractions, optimization of the mechanical properties are helpful for sintered Ti6Al4V alloys fabricated Metal injection molding (MIM) powders. A pseudo quasi in-situ method with vacuum encapsulation and quenching has been proposed to control dynamical characteristics of pores, grain sizes and phase fractions of MIMed Ti6Al4V powders during sintering process. The dependency of pores annihilation and grains growth on sintering time and temperature has been established. Higher sintering temperature shortens the transition period of pore morphology from connected irregular to enclosed spherical characteristics. The annihilation of pores satisfies the exponential function as well as the power law with temperature and time, respectively. The activation energy has been calculated to be 252.2 kJ/mol comparable to self-diffusion activation energy of Ti according to Arrhenius fitting of the changes of grain size with sintering temperature. Additionally, the diffusional kinetic coefficient was 28.57 at 1200 °C, about five times larger than that at 1100 °C. The fraction of α/α′ phase decreases with increasing of sintering temperature and time. The existence of impurity elements (O, C) may result increasing the stability of α phases and morphology transforming of α phases from lamellar to equiaxed. Finally, sintered MIM-ed Ti6Al4V alloys with lowest porosity sintered at 1200 °C for 4 h have the tensile strength of 922 ± 5 MPa, the yield stress of 833 ± 4 MPa and the elongation of 12.8 ± 0.5%.

Influence of thermomechanical treatment on mechanical properties of CLAM steel

To enhance the creep resistance of CLAM steel and thus elevate its maximum service temperature, in our prior research, a superior thermomechanical treatment process was developed based on thermal simulation experiments conducted. This study employed forging to achieve the desired thermal deformation in accordance with the superior thermomechanical treatment process, followed by mechanical property testing post tempering. Comparative analysis with normalized and tempered CLAM steel (N + T CLAM steel) revealed notable improvements in yield strengths ranging from 31–55 %, ultimate tensile strengths from 30–46 %, and a reduction in total elongations by 12–21 % at both room temperature and elevated temperatures. Additionally, the ductile-to-brittle transition temperature (DBTT) increased from -65 °C to -27 °C. Creep behavior of N + T CLAM steel exhibited similarities to other RAFM steels, with stress exponents measured at 500 °C, 550 °C, 600 °C, and 650 °C recorded as 16.5, 15.0, 10.2, and 7.8, respectively. The creep activation energy was determined to be 556 kJ/mol, closely aligning with results from Eurofer 97 steel and F82H steel, and approximately twice the self-diffusion activation energy of iron. The Larson-Miller parameter method was employed to forecast stress-rupture limits under various conditions, with a predicted result of 120 MPa under typical design life conditions (550 °C / 100,000 h) for the cladding module, comparable to Eurofer 97 steel and F82H steel. TMT CLAM steel exhibited a stress exponent of 16.7 at 550 °C, with a creep activation energy of 578 kJ/mol. Under typical design life conditions (550 °C / 100,000 h) for the cladding module, the predicted stress-rupture limit increased to 176 MPa, representing a 47 % improvement over N+T CLAM steel. Moreover, under the same conditions (550 °C / 230 MPa), the rupture time of TMT CLAM steel was 54 times longer than that of N+T CLAM steel.

Storage and Diffusion of Carbon Dioxide in the Metal Organic Framework MOF-5─A Semi-empirical Molecular Dynamics Study.

Metal-organic frameworks (MOFs) have attracted increasing attention due to their high porosity for exceptional gas storage applications. MOF-5 belongs to the family of isoreticular MOFs (IRMOFs) and consists of Zn4O6+ clusters linked by 1,4-benzenedicarboxylate. Due to the large number of atoms in the unit cell, molecular dynamics simulation based on density functional theory has proved to be too demanding, while force field models are often inadequate to model complex host-guest interactions. To overcome this limitation, an alternative semi-empirical approach using a set of approximations and extensive parametrization of interactions called density functional tight binding (DFTB) was applied in this work to study CO2 in the MOF-5 host. Calculations of pristine MOF-5 yield very good agreement with experimental data in terms of X-ray diffraction patterns as well as mechanical properties, such as the negative thermal expansion coefficient and the bulk modulus. In addition, different loadings of CO2 were introduced, and the associated self-diffusion coefficients and activation energies were investigated. The results show very good agreement with those of other experimental and theoretical investigations. This study provides detailed insights into the capability of semi-empirical DFTB-based molecular dynamics simulations of these challenging guest@host systems. Based on the comparison of the guest-guest pair distributions observed inside the MOF host and the corresponding gas-phase reference, a liquid-like structure of CO2 can be deduced upon storage in the host material.

Evidence of interstitial-mediated boron self-diffusion in boron carbide

Boron carbide is an important high-temperature material with interesting structural and functional properties. However, self-diffusion is very poorly known on this material, and also the few impurity diffusion studies, devoted in particular to helium and lithium, overlook the possible diffusion coupling with intrinsic defects. In this paper we present a study of boron self-diffusion in boron carbide based on first principles calculations and kinetic cluster expansion. Our results show that boron self-diffusion may occur with a migration activation energy on the order of 1.5–1.6 eV and with a global activation energy between 4 and 5.5 eV, depending on the doping conditions, which is significantly lower than the self-diffusion activation energies in other semiconducting carbides, like silicon carbide.We believe that our results are of interest for many applications of boron carbide and for tuning the best conditions for controlling the synthesis routes at specific stoichiometries for boron carbide.

Hot workability behaviour of two P92 creep resistant steels: Constitutive analysis

This article reports the flow stress behaviour of two P92 steels at a temperature range of 850–1000°C and a strain rate of 0.1–10 s−1 using the Gleeble® 3500 thermomechanical simulator. A physically-based constitutive model was used to analyse the effects of deformation conditions on the flow stress behaviour during deformation. This model incorporates the influence in the variation of Young’s modulus and the self-diffusion coefficient as affected by temperature. The study developed constitutive equations that predict the flow stress behaviour of the two steels investigated. From the constitutive analysis of the results, the stress exponent n was: 9.8 (steel A) and 10.3 (steel B). The model used the self-diffusion activation energy of steel. The statistical parameters: correlation coefficient of 0.99 (for steel A and B), the absolute average relative error of 2.18% (steel A) and 2.20% (steel B) quantified the applicability of the model. The quantification results show that the constitutive equations developed have high accuracy in predicting the workability of the two P92 steels. The study has shown that this method is applicable in predicting the metal flow pattern of two P92 steels in the metalworking processes.

First-Principles Study of Atomic Diffusion by Vacancy Defect of the L12-Al3M (M = Sc, Zr, Er, Y) Phase.

Atomic diffusion by the vacancy defect of L12-Al3M (M = Sc, Zr, Er, Y) was investigated based on a first-principles calculation. The point defect formation energies were firstly evaluated. Then, the migration energy for different diffusion paths was obtained by the climbing-image nudged elastic band (CI-NEB) method. The results showed that Al atomic and M atomic diffusions through nearest-neighbor jump (NNJ) mediated by Al vacancy (VAl) were, respectively, the preferred diffusion paths in Al3M phases under both Al-rich and M-rich conditions. The other mechanisms, such as six-jump cycle (6JC) and next-nearest-neighbor jump (NNNJ), were energetically inhibited. The order of activation barriers for NNJ(Al-VAl) was Al3Zr < Al3Y < Al3Er < Al3Sc. The Al3Sc phase had high stability with a high self-diffusion activation barrier, while the Al3Zr and Al3Y phases were relatively unstable with a low self-diffusion activation energy. Moreover, the atomic-diffusion behavior between the core and shell layers of L12-Al3M was also further investigated. Zr atoms were prone to diffusion into the Al3Y core layer, resulting in no stable core-shelled Al3(Y,Zr), which well agreed with experimental observation.

Water Molecules' and Lithium Cations' Mobility in Sulfonated Polystyrene Studied by Nuclear Magnetic Resonance.

The hydration of ions and charge groups controls electro mass transfer through ion exchange systems. The self-diffusion and local mobility of water molecules as well as lithium cations in poly (4-styrenesulfonic acid) and its lithium, sodium and cesium salts were investigated for the first time using pulsed-field gradient NMR (PFG NMR) and NMR relaxation techniques. The temperature dependences of the water molecule and Li+ cation self-diffusion coefficients exhibited increasing self-diffusion activation energy in temperature regions below 0 °C, which is not due to the freezing of parts of the water. The self-diffusion coefficients of water molecules and lithium cations, as measured using PFG NMR, are in good agreement with the self-diffusion coefficients calculated based on Einstein's equation using correlation times obtained from spin-lattice relaxation data. It was shown that macroscopic water molecules' and lithium cations' transfer is controlled by local particles jumping between neighboring sulfonated groups. These results are similar to the behavior of water and cations in sulfonic cation exchanger membranes and resins. It was concluded that polystyrenesulfonic acid is appropriate model of the ionogenic part of membranes based on this polymer.

Role of Mn content on processing maps, deformation kinetics, microstructure and texture of as-cast medium Mn (6–10 wt% Mn) steels

In the present work, the hot deformation behaviour of as-cast medium Mn steels with nominal Mn content of 6, 8, and 10 wt% were studied using the thermomechanical simulator in the temperature and strain rate range of 1173 to 1373 K and 10−3 - 10 s−1 respectively. The flow curves of the three steels revealed significant work hardening followed by a plateau at strain rates ≥1s−1 in the entire temperature range. At lower strain rates (<10−2s−1) noticeable softening is observed after a strain of 0.2 in all three steels. The extent of softening is lower in 10 wt% Mn steel in comparison to the 6 and 8 wt% Mn steels. Analysis of processing maps reveals that in steels with 6 and 8 wt% Mn, regions of ‘high’ strain rate sensitivity (m) (0.2 - 0.25) are found to occur in similar temperature and strain rate regimes of 1273–1323 K and 10−2s−1 respectively. These steels also showed similar apparent activation energies for hot deformation (380–390 kJmol−1), stress exponents (4.6) and dynamic recovery/recrystallization evolution with strain. The steel with 10 wt% Mn, has higher apparent activation energy for hot deformation (450 kJmol−1), higher stress exponent of 5.1, higher dynamic recovery and lower dynamic recrystallization. The observed activation energies correlate well with the activation energy for self-diffusion (QSD) of Fe in austenite containing Mn as an alloying element. In the ‘low’ m regimes, higher dynamic recovery (evaluated using kinetic analysis) was observed in the 10 % Mn steel as compared to the 6 and 8% Mn steels. In the ‘high’ m regimes, the dynamic recrystallization fraction (estimated from kinetic analysis and from parent grain reconstruction analysis) was lower in the 10 % Mn steel when compared to the 6 and 8 wt % Mn steels. This behaviour can be attributed to the higher dynamic recovery observed in the 10 % Mn steel. From the analysis of processing maps, kinetic analysis and microstructure, the optimum strain, strain rate and temperature for the thermomechanical processing of as-cast medium Mn steels with Mn in the range of 6 to 10 wt% are ≤ 0.6, ≤10−2s−1 and 1273 to 1373 K respectively.

Tellurium self-diffusion in amorphous Ge2Sb2Te5 phase change materials

Amorphous Ge2Sb2Te5 phase change material is considered as a prototype material for non-volatile memories due to the reversibility of the amorphous-to-crystalline transition on a nanosecond timescale. In this context, the kinetics of atomic self-diffusion has important bearings for the crystallization process and the switching behavior of phase change materials. It is thus important for applications as well as for the general understanding of the rapid and reversible switching behavior to understand atomic self-diffusion, especially in the amorphous phase. However, up to now, reliable data on the kinetics of atomic self-diffusion in amorphous phase change materials is almost non-existent. For this reason, Te tracer diffusion was measured using secondary ion mass spectroscopy and applying the highly enriched natural 122Te isotope. For the first time, Te self-diffusion coefficients in amorphous Ge2Sb2Te5 are experimentally measured and the activation energy of Te self-diffusion was determined as (1.43±0.08) eV.

Effect of Heat Treatment on Microstructure and Mechanical Behavior of Ultrafine-Grained Ti-2Fe-0.1B.

In the present study, a novel Ti-2Fe-0.1B alloy was processed using equal channel angular pressing (ECAP) via route Bc for four passes. The isochronal annealing of the ultrafine-grained (UFG) Ti-2Fe-0.1B alloy was conducted at various temperatures between 150 and 750 °C with holding times of 60 min. The isothermal annealing was performed at 350-750 °C with different holding times (15 min-150 min). The results indicated that no obvious changes in the microhardness of the UFG Ti-2Fe-0.1B alloy are observed when the annealing temperature (AT) is up to 450 °C. Compared to the UFG state, it was found that excellent strength (~768 MPa) and ductility (~16%) matching can be achieved for the UFG Ti-2Fe-0.1B alloy when annealed at 450 °C. The microstructure of the UFG Ti-2Fe-0.1B alloy before and after the various annealing treatments was characterized using electron backscatter diffraction (EBSD) and transmission electron microscopy (TEM). It was found that the average grain size remained at an ultrafine level (0.91-1.03 μm) when the annealing temperature was below 450 °C. The good thermal stability of the UFG Ti-2Fe-0.1B alloy could be ascribed to the pinning of the TiB needles and the segregation of the Fe solute atoms at the grain boundaries, which is of benefit for decreasing grain boundary energy and inhibiting the mobility of grain boundaries. For the UFG Ti-2Fe-0.1B alloy, a recrystallization activation energy with an average value of ~259.44 KJ/mol was analyzed using a differential scanning calorimeter (DSC). This is much higher than the lattice self-diffusion activation energy of pure titanium.

Through-thickness heterogeneity in creep properties of friction stir welding 7B50-T7451 aluminum alloy thick plate joint: Experiments and modeling

Through-thickness heterogeneity in creep properties of 7B50-T7451 aluminum alloy Friction Stir Welding (FSW) joints was investigated. Creep tests for three slices of the FSW joint were conducted at the temperature of 150–200 °C and applied stress of 60–225 MPa. The theta projection method was used to predict creep curves and minimum creep rate. The results show that the minimum creep rate increases and creep rupture life decreases with the increase of creep temperature and applied stress. Creep properties of the FSW joint deteriorate along the thickness direction from the top to the bottom. The threshold stress of all three slices of the FSW joint decreases with the increase of creep temperature and even disappears at 200 °C for the bottom slice. Creep activation energy approaches the activation energy of the lattice self-diffusion of aluminum. The value of true stress exponent for different slices is approximately equal to three. The predominant creep mechanism of the FSW joint is dislocation viscous glide by lattice self-diffusion. What is more, a constitutive model is established based on the theta method to accurately describe creep behavior of different slices of the FSW joint.

Features of Self-Diffusion of Tridecane Molecules in a Porous Medium of Kaolinite Used as a Model of a Chemically Inert Membrane

The present work focused on the experimental study of the specific features of self-diffusion of tridecane molecules in macroporous kaolinite, which is used as a raw material for the production of chemically inert membranes. The measurements of self-diffusion coefficients by pulsed magnetic field gradient nuclear magnetic resonance (PMFG NMR) revealed an increased translational mobility of tridecane molecules in kaolinite with incomplete filling of the pore space. This effect was accompanied by a sharp change in the slope of the Arrhenius plot of the self-diffusion coefficients of tridecane molecules in kaolinite. An analysis of the diffusion spin echo decay in the tridecane-kaolinite system revealed a discrepancy between the experimental data and the theoretical predictions, considering the effect of the geometry of porous space on molecular mobility. It was shown that the experimental results could be interpreted in terms of a model of two phases of tridecane molecules in the pores of kaolinite, in the gaseous and adsorbed state, coexisting under the fast-exchange conditions. Within the framework of the model, the activation energies of self-diffusion were calculated, which agreed satisfactorily with the experimental data. Additionally, the effects of the internal magnetic field gradients arising in a porous medium loaded with a gas or liquid on the data of the PFG NMR measurements were calculated. It was shown that the effect of magnetic field inhomogeneities on the measured self-diffusion coefficients of tridecane in kaolinite is small and could be neglected.

Investigation of H2S Diffusion in Transcritical and Supercritical Water: A Molecular Dynamics Simulation Study

In supercritical water (SCW) gasification technologies, H2S is one of the important intermediate products. Studying the self-diffusion coefficient of H2S in subcritical water and SCW can help us better understand the mass transfer characteristics. In this work, we used the molecular dynamics method to simulate the diffusion behavior of H2S in the H2S–H2O system and calculated the self-diffusion coefficient of H2S under different conditions (pressure: 235–280 atm, H2S molar concentration: 1–5%, temperature: 600–1173 K). The result shows that the molar concentration and pressure of H2S have little effect on its self-diffusion coefficient, but temperature and water density have a greater effect on it. In addition, the self-diffusion activation energy of the supercritical region of H2S is only 32.58% of the subcritical region. Finally, we fitted new parameters of the empirical equation to predict the self-diffusion coefficient of H2S in the H2S–H2O system, whose mean relative error in the supercritical region is 3.05%.

Ethylene-Vinyl Acetate Copolymers as Potential Thermoplastic Modifiers of Photopolymer Compositions.

The possibility of using thermoplastic polymers in photopolymer compositions for SLA and DLP is discussed in this article. The diffusion and mutual solubility of uncured systems based on tert-butyl acrylate (tBA) and ethylene-vinyl acetate copolymers (EVA) or low-density polyethylene (LDPE) were studied. The solubility and diffusion of tBA with EVA containing 7, 20, and 40 wt.% vinyl acetate (VA) and with LDPE in the temperature range 20-75 °C were studied by optical micro-interferometry method. Phase diagrams of LDPE-tBA, EVA-7-tBA, and EVA-20-tBA systems were obtained. It is shown that the compositions are characterized by the phase-state diagrams of amorphous separation with the upper critical solution temperature (UCST). The concentration dependences of the interdiffusion coefficients as well as dependences of the self-diffusion coefficients on VA content and on temperature were plotted. The activation energy of self-diffusion of EVA and LDPE was calculated. It was shown that the most promising tBA modifier is EVA-40, which is completely soluble at all studied temperature ranges. The obtained data on the mixing of the initial components is valuable for further studies of the processes of structure formation during photocuring of compositions, regulation of the phase structure and, as a consequence, the performance characteristics of the 3D printable materials.

Creep properties and deformation mechanisms of a Ni2Co1Fe1V0.5Mo0.2 medium-entropy alloy

The creep properties of a Ni2Co1Fe1V0.5Mo0.2 medium-entropy alloy (MEA) in the temperature range 973–1073 K, and under the stresses 150–200 MPa, have been investigated in detail. The Ni2Co1Fe1V0.5Mo0.2 MEA possesses excellent creep properties at 973 K/150 MPa and 973 K/175 MPa. The stress exponent n was calculated to be 3.1 at 973 K, suggesting the dominance of viscous dislocation glide mechanism in creep deformation. The activation energy for creep deformation was measured to be 456.3kJ·mol−1, which is comparable to the activation energy for self-diffusion of Mo, indicating a strong effect of Mo diffusion on viscous dislocation glide. Detailed transmission electron microscopy analysis reveals: (1) increased stress and strain can alter the dislocation configurations and promote dislocation climb, and thus to increase the creep rate; (2) increased temperature and strain can lead to active recovery and recrystallisation, and thus to cause grain refinement.

Hot deformation behavior and microstructure evolution of non-equimolar Ti2ZrHfV0.5Ta0.2 refractory high-entropy alloy

This work aims to elucidate the hot deformation behavior and microstructure evolution of a non-equimolar TiZrHf-rich refractory high-entropy alloy (RHEA) with a composition of Ti2ZrHfV0.5Ta0.2 (at. %). The as-cast Ti2ZrHfV0.5Ta0.2 alloy composed solely of a body-centered cubic phase showed superior tensile mechanical properties with a yield strength of 865 MPa and a total elongation of 22.5%. The uniaxial compression tests were performed at temperatures ranging from 900 to 1100 °C and strain rates varying from 10−3 to 10−1 s−1. The relation among the flow stress, deformation temperature, and strain rate was represented by the Arrhenius-type constitutive equations. The calculations show a comparatively low apparent activation energy of 144–119 kJ/mol over the entire range of strains, revealing a noticeable contradiction between the concentration of group IV elements (Ti, Zr, Hf) and the high-temperature strength. The excessive addition of Group IV elements tends to deteriorate the high-temperature strength due to the relatively low melting points and self-diffusion activation energies of Group IV elements. The misorientation analysis based on electron back-scattered diffraction indicated that the dynamic recrystallization mechanism depends strongly on the thermo-mechanical processing conditions. A low deformation temperature and high strain rate results in an abnormal discontinuous dynamic recrystallization mechanism. Additionally, the dominant role played by the edge dislocation in RHEAs during deformation is supposed to be the reason for the abnormal discontinuous dynamic recrystallization mechanism. A series of stress drops were observed under low deformation temperatures and high strain rates. Based on an interrupted hot-compression experiment and a strain aging experiment, the stress drop can be attributed to the unlocking of dislocations pinned by the Cottrell atmosphere or short-range ordering.

Atomic-scale interfacial diffusion of diamond into titanium: Phase transition and layer dependence

The diffusion wear of diamond is a commonly recognized wear mode at high temperatures, while the atomic-scale diffusion behavior and the initiation mechanism are scarcely reported. In this work, we conduct a reactive molecular dynamics study to unveil the diffusion mechanism of diamond into titanium from an atomistic perspective. It is found that substantial diffusion of carbon atoms into titanium at a nanoscale temperature occurs at the temperature over 3000 °C, and the diffusion intensifies rapidly with the further increase of temperature. By studying the evolution of the phase transition, we find that the dangling atoms with a coordination less than two play a critical role in initiating the diffusion of diamond, and subsequently more sp-bonded carbon phases are involved in promoting the diffusion process during the non-equilibrium period. Moreover, the diffusivity of carbon atoms in diamond is found to have a strong layer dependence, which can be attributed to the unique phase composition and activation energy for self-diffusion. Eventually, an effective activation energy for the interfacial diffusion from diamond to titanium of 543.9 kJ/mol is obtained. This study provides a new insight into the initiation mechanism of the interfacial diffusion of diamond materials.

Prediction of flow stress characteristics of P92 steel using a simple physically-based constitutive modelling

This study reports the flow stress behavior of three P92 steels with different compositions. Uniaxial compression tests were conducted in the deformation temperature range of 575 °C to 650 °C and strain rate range of 0.001-0.5 s-1 using a Gleeble® 3500 thermo-mechanical simulator. A simple physically-based constitutive model was used to analyse the effects of deformation conditions (temperature and strain rate) on the metal flow stress behavior during the deformation process. The method accounts for the temperature dependence of Young’s modulus and the lattice self-diffusion coefficient of Fe in the ferrite. Constitutive equations describing the flow stress behavior of the three P92 steels were developed. From the results, the stress exponent n of 26.13(steel A), 21.61(steel B) and 27.55 (steel C) were obtained using the self-diffusion activation energy in the physically-based constitutive equation. From the results, the three steels had variation in the stress exponent values, which was attributed to differences in elemental content, such as chromium and tungsten. The developed constitutive equations were verified using statistical parameters: Pearson’s correlation coefficient (R) and average absolute relative errors (AARE). Statistical analysis showed that the three steels had the same R of 0.98, while AARE was: 1.68 (steel A), 1.72 (steel B), and 1.82 (steel C). The constitutive equations developed showed a good correlation between the experimental and predicted flow stress data. Hence, the method is applicable in describing flow stress behavior in the metalworking process in the industry.

Improvement of the Electromigration Resistance of Al Thin-Film Interconnection by Controlling the Crystallinity of Grain Boundaries

The degradation process of the crystallinity of grain boundaries in Al thin-film interconnections was visualized by applying EBSD analysis under high current density. The grain boundary diffusion was accelerated by electromigration only along random grain boundaries with low crystallinity. The crystallinity (IQ value obtained from EBSD analysis) of grain boundaries varied drastically as strong functions of manufacturing process such as deposition method, temperature of thin-film deposition, gas pressure during the deposition, substrate material, and so on. The effective activation energy of self-diffusion of component element along random grain boundaries was found to be dominated by the crystallinity of random grain boundaries. The measured activation energy obtained from the low-crystallinity interconnections was much lower than that obtained from the thermodynamically stable bulk material, and therefore, the diffusion constant due to electromigration was drastically increased, resulting in the drastic decrease of the lifetime of the interconnections. The lifetime of the interconnections was, therefore, improved by increasing the crystallinity of grain boundaries.