Search

First-principles theory was developed to study the structural phase transformations in the Laves phase HfV2 alloy. We explored the energy landscape and established the role of lattice anharmonicity underlying the structural phase transitions. Our approach is based on a phenomenological Landau theory for the structural phase transition and a mean-field approximation for the free energy. First-principles calculations were utilized to obtain the distortion energy as a function of relevant deformations, and to deduce parameters for constructing the free energy. Our result for the phase transition temperature of HfV2 is in good agreement with experiment. We find that the high-temperature cubic C15 phase is stabilized by the effect of lattice anharmonicity. The theory also predicts an anomalous increase in shear modulus with increasing temperature for systems where the anharmonicity is pronounced.

Using first-principles local density functional approach, we have calculated the ground-state structural phase stabilities and enthalpies of formation of thirty-six binary transition-metal refractory TM and TM3 compounds formed by Group IV elements T(T = Ti, Zr, Hf) and platinum group elements M (M = Ru, Rh, Pd, Os, Ir, Pt). We compared our results with the available experimental data and found good agreement between theory and experiment in both the trends of structural stabilities and the magnitudes of formation enthalpies. Moreover, based on our calculated results, an empirical relationship between cohesive energies (Delta E) and melting temperatures (T-m) was derived as T-m = 0.0292 Delta E/k(B) (where k(B) is the Boltzmann constant) for both TM and TM3 compounds. (C) 2012 Elsevier Ltd. All rights reserved.

Using an ab initio density functional approach, we report on the ground-state phase stabilities, enthalpies of formation, electronic, and elastic properties of the Ti-Pd alloy system. The calculated enthalpies of formation are in excellent agreement with available calorimetric data. We found a linear dependence between the calculated enthalpies of formation of several intermetallic structures and the Pd-concentration, indicating that each of these compounds has a very limited composition range. The elastic constants for many of these Ti-Pd intermetallics were calculated and analyzed. The B2 TiPd phase is found to be mechanically unstable with respect to the transformation into the monoclinic B19' structure. A series of hydrides, Ti(2)PdH(x) (x = 1,1.5, 2, 3, 4), have been investigated in terms of electronic structure, enthalpies of hydrogen absorption, and site preference of H atoms. Our results illustrate the physical mechanism for hydrogen absorption in term of the charge transfer, and explain why TiPd(2) does not form a stable hydride. (C) 2010 Elsevier Ltd. All rights reserved.

Spin transition has attracted the interest of researchers in various fields since the early 1930s, with thousands of examples now recognized, including those in minerals and biomolecules. However, so far the metal centres in which it has been found to occur are almost always octahedral six-coordinate 3d4 to 3d7 metals, such as Fe(II). A five-coordinate centre is only rarely seen. Here we report that under pressure SrFe(II)O2, which features a four-fold square-planar coordination, exhibits a transition from high spin (S = 2) to intermediate spin (S = 1). This is accompanied by a transition from an antiferromagnetic insulating state to a ferromagnetic so-called half-metallic state: only half of the spin-down (dxz, dyz) states are filled. These results highlight the square-planar coordinated iron oxides as a new class of magnetic and electric materials., 四本の手をもつ金属原子の磁性の変換に成功 －磁気記録、スイッチ素子の革新につながるか－. 京都大学プレスリリース. 2009-07-21.

The existence of nanoclusters that are thermo- dynamically stable at elevated temperatures is truly intri- guing because of its scientific implications and potential applications. Highly stable nanoclusters have been observed by atom probe tomography in iron-based alloys at temperatures close to 1400°C (0.92Tm) that appear to defy the stability constraints of artificially created nano- structured materials. The ~4-nm-diameter Ti-, Y- and O- enriched nanoclusters are identified in the new form of a highly defective material state with vacancies as the criti- cal alloying component and with (Ti + Y):O ratio differ- ent from the stable TiO2 and Y2Ti2O7 oxides. Vacancies play an indispensable role in enhancing the oxygen solu- bility and increasing the oxygen binding energy in the presence of Ti and Y, resulting in the stabilization of coherent nanoclusters. Atom probe tomography charac- terizations and theoretical predictions indicate that vacan- cies can be exploited for the first time as a nanoscale constituent to design materials with far superior high tem- perature properties.

Neutron diffraction was used to study the structure of ternary NiAl + M (M = Ni, Fe, Co) alloys. The experiment confirmed the predictions by first-principle calculations on site preference by solute atoms. Moreover, a universal structural transformation was observed below 20 K in alloys where Al is partially replaced by M. The extra peaks at low-temperatures do not match those from known martensite phases, but are well indexed by a cubic structure. A possible mechanism for the low-temperature phase is discussed.

The solid solution effects of ternary additions of transition elements in intermetallic Ni-40% Al were investigated by both experimental studies and theoretical calculations. Co solute atoms when sitting at Ni sublattice sites do not affect the lattice parameter and hardening behavior of Ni-40Al. On the other hand, Fe, Mn, and Cr solutes, which are mainly on Al sublattice sites, substantially expand the lattice parameter and produce an unusual solid solution softening effect. First-principles calculations predict that these solute atoms with large unfilled d-band electrons develop large magnetic moments and effectively expand the lattice parameter when occupying Al sublattice sites. The theoretical predictions were verified by both electron loss-energy spectroscopy (EELS) analyses and magnetic susceptibility measurements. The observed softening behavior can be explained quantitatively by the replacement of Ni anti-site defects (potent hardeners) by Fe, Mn, and Cr anti-site defects with smaller atom size mismatch between solute and Al atoms. This study has led to the identification of magnetic interaction as an important physical parameter affecting the solid solution hardening in intermetallic alloys containing transition elements.

In Laves phase alloys with prominent size mismatch between constituent atoms and/or large negative enthalpy of formation, the existence of vacancies as the dominant point defect type is often suggested. However, there are not enough experimental data to prove or disprove these arguments. Employing first-principles calculations, we study the point defect structures of YAl 2 and ZrCo 2 C15 Laves phases, as both compounds exhibit large size mismatch between constituent atoms, and large negative enthalpy of formation. We find that one must go beyond the simple geometrical or enthalpy arguments in determining the point defect structures of these alloys. In both compounds, the point defect structure is found to be dominated by the anti-site defects on the larger atom-rich side of the stoichiometry.

First-principle calculations for the diffusion of 3d transition metal (TM) solutes in nickel demonstrate the existence of a higher diffusion energy barrier for solutes with smaller atomic sizes. The calculations reveal that smaller TM atoms are, actually, among the least compressible due to the formation of incompressible solute-host directional bonds. Magnetism is shown to have a profound effect on the solute diffusion trends across the 3d TM series: the existence of a local minimum in the diffusion energy barrier is accompanied by the occurrence of a maximum in the magnetic moment. The calculated diffusion rates disprove the traditional view that the diffusion of solutes is least rapid when the size misfit with the host is the greatest.

The coefficients of thermal expansion (CTEs) and atomic site occupation factors were measured for the pseudo-binary molybdenum–vanadium silicide Mo 5 Si 3 –V 5 Si 3 . The ratios of the CTEs in the crystallographic c and a directions range from 2.0 for Mo 5 Si 3 to 1.2 for Mo 2 V 3 Si 3 . The metal atoms can occupy chain sites (in the crystallographic c -direction), and non-chain sites. The relatively small V atoms prefer the chain sites in Mo 5 Si 3 , whereas the relatively large Mo atoms prefer the non-chain sites in V 5 Si 3 . These site-occupation characteristics reduce the thermal expansion in the c -direction and thereby the CTE anisotropy. Ab initio calculations for Mo 4 VSi 3 suggest that, energetically, V prefers the chain sites, while the measurements show that V occupies also non-chain sites. Extrapolation of the experimental results assuming the site-occupation used in the calculations agrees with the theoretical result that the thermal expansion of Mo 4 VSi 3 is isotropic.

By taking the solid solution effect of Co, Fe, Mn, and Cr solutes in Ni-rich NiAl as an example, we have determined, through a combination of alloying behavior experiments, first-principles calculations, and neutron diffraction measurements, that magnetism has a profound impact on the mechanical properties of intermetallics. Co solutes do not affect the hardening behavior of NiAl alloys. Fe solutes, on the other hand, induce the most substantial and unusual solid solution softening among these solutes. These results can only be explained by the presence of magnetic interactions induced by solutes located on the Al sublattice. The predicted magnetic moment of Fe solutes was verified by neutron diffraction experiment. Our investigation points out the inadequacy of the conventionally used Goldschmidt radii in addressing the atomic size effect in intermetallics.

The coefficients of thermal expansion (CTE) of the tetragonal silicides Mo 5 Si 3 and Ti 5 Si 3 are highly anisotropic. The CTE anisotropy values, defined as the ratio of the CTEs in the c and a directions, CTE( c )/CTE( a ), are 2.0 and 2.7 for Mo 5 Si 3 and Ti 5 Si 3 , respectively. Guided by first-principles calculations by Fu et al. (Acta Mater 51 (2003) 5033), ternary and quaternary alloying additions for reducing the anisotropy were selected. The smallest anisotropy, namely, 1.16, was found for a Mo–25Nb–12.5V–37.5Si (at.%) alloy. The experimentally determined reductions in CTE( c )/CTE( a ) were not always as large as predicted because the measured site occupations of the ternary alloying elements were not as ideal as those calculated by Fu et al. Suitable alloying also reduces the CTE anisotropy of Ti 5 Si 3 , but the reduction is not nearly as large as that claimed in previous publications.

Mo 5 Si 3 exhibits a high anisotropy of its coefficients of thermal expansion (CTEs) in the a and c directions, i.e., CTE( c )/CTE( a )=2.0. The high CTE anisotropy is due to the existence of prominent atom chains along the c -axis. Ternary alloying additions that effectively stretch the atom chains are likely to reduce the bonding directionality/strength and CTE anisotropy. First-principles local-density-functional calculations were carried out for two alloy systems, MoNb 4 Si 3 and Mo 4 VSi 3 . Indeed, we find a significant reduction of the CTE ratio to a value of 1.5 for MoNb 4 Si 3 and a value close to 1.0 for Mo 4 VSi 3 . The decrease in the CTE anisotropy mainly comes from the reduction in CTE in the c -direction. While a conventional approach to reduce the CTE would involve an increase in lattice rigidity by increasing the bond strength, our strategy focuses on the reduction of driving force for thermal expansion by decreasing the bond directionality.

In this study, experimental work on solute effects in an intermetallic alloy, NiAl, has been complemented by first-principles quantum mechanical calculations to investigate the effect of iron and cobalt solutes on lattice parameter and hardening. Ternary additions of iron and cobalt with similar atomic sizes were added to replace nickel in NiAl containing 40% Al. Cobalt solutes did not affect the lattice parameter or the hardening behavior of NiAl alloys. Iron solutes, on the other hand, substantially expanded the lattice, resulting in unusual solid solution softening. These results could not be explained from a consideration of the size mismatch based on the Goldschmidt radii for iron and cobalt atoms. From the contrasting behavior of the iron and cobalt solutes, magnetic interactions induced by iron atoms located on the aluminum sublattice have been identified as a new factor responsible for the unusual large lattice dilation and resultant solid solution softening in these intermetallic alloys.

Dislocation structures in a nearly single phase annealed Mo5SiB2 T2 alloy have been investigated by transmission electron microscopy (TEM). The dislocations have been subjected to Burgers vector and trace analyses to determine the slip directions and planes with the aid of image simulations generated using single crystal elastic constants derived from first principles calculations. The experimental results are compared to predicted slip directions and planes from anisotropic elasticity calculations. Thermal expansion coefficients have been measured by dilatometry and are compared to both calculated and previous experimental values measured using diffraction techniques. Lastly, preliminary compression testing has been performed on the single phase material at 1200°C.

To examine the importance of thermal expansion mismatch on cracking in two-phase Cr–Cr2X (X=Hf, Nb, Ta, Zr) alloys, we have calculated the coefficients of thermal expansion (CTE) of Cr and Laves-phase Cr2X alloys by local-density-functional approach. A Debye model is used to approximate the phonon contribution through the elastic acoustic response. The sound velocities are determined by the calculated elastic constants and associated anharmonicity. The calculation shows that the CTE of Cr at high temperatures is notably larger than those of Cr2X. If the difference in CTE between Cr and Cr2X is a primary source of crack initiation, our results fail to explain the experimental observation that, among these four Cr–Cr2X alloy systems, the ingot cracking is present mainly in Cr–Cr2Nb. We suggest that for the cracking to occur,the presence of thermal mismatch stresses are retained by a hard and supersaturated Cr matrix (e.g. due to the relatively high solubility of Nb in Cr in the case of Cr–Cr2Nb). On the other hand, the softer Cr matrix can accommodate thermal misfit dislocations plastically even though the CTE difference between Cr and Cr2X is large. # 2001 Elsevier Science Ltd. All rights reserved.

The equilibrium structural parameters, enthalpies of formation and partial densities of state for Ti5Si3 and Ti5Si3Z0.5 (Z= B, C, N or O) were calculated based on first-principle techniques. Enthalpy of formation calculations suggest that of the known structures for transition metal (TM) silicide compounds containing TM5Si3 (D88, D8l and D8m) the D88 structure is the most stable form of Ti5Si3, and the stability of the structure increases as Z atoms are added. The theoretically determined structural trends as a function of interstitial element, Z, agreed well with experimentally determined values. Both indicate bonding between Ti and Z atoms based on contraction of Ti–Z separations. The calculated partial densities of state suggest that p(Si)–d(Ti) and d(Ti)-d(Ti) interactions are responsible for most of the bonding in pure Ti5Si3, which agrees with previous studies. As Z atoms are added, p(Z)–d(Ti) interactions become significant at the expense of weakening some of the d(Ti)–d(Ti) interactions.

The stacking fault and twin boundary energies of C15 Cr2Nb are calculated by the first-principles local-density-functional approach. It is found that the intrinsic and extrinsic stacking fault energies are 116 and 94mJm−2, respectively, and the twin boundary energy is 39mJm−2. The lower extrinsic stacking fault energy is consistent with the fact that the C36 structure has a lower energy than the C14 structure. The calculated stacking fault energies at 0K are larger than the experimental values available in the literature. The equilibrium separations between Shockley partials based on the calculated elastic constants and stacking fault energies are also calculated.

The elastic moduli, stacking fault and twin boundary energies of C15 Cr2Ta are calculated by first-principles local-density-functional approach. Similarly to the case of Cr2Nb, Cr2Ta is characterized by a low twin boundary energy (33 mJ/m 2 ). On the other hand, the intrinsic and extrinsic stacking fault energies are found to be 88 and 83 mJ/m 2 , respectively. Using the results of elastic constants and stacking fault energies, we predict the equilibrium separation between Shockley partials. The results of Cr2Ta are compared with those of Cr2Nb. # 1999 Elsevier Science Ltd. All rights reserved.

We present results of first-principles local-density-functional calculations of the structural and elastic properties of Mo5Si3. Among the three different structures (D8m, D88, and D8l), the D8m structure (referred to as the T1 phase) has the greatest binding with a high heat of formation of −3.8 eV/formula unit. The bonding in Mo5Si3 is found to have pronounced covalent components, characterized by the planar Mo–Si–Mo triangular bonding units on the (001) plane and by the unusually short Mo–Mo bonds directly along the c-axis. The calculated six elastic constants of the D8m structure are in excellent agreement with the experimental values. While the bonding in the (001) basal plane is stronger than the bonding along the [001] direction (i.e. C11+C12>C33 and C66>C44), the crystal anharmonicity is found to be higher along the [001] direction. The implication of our results on the anisotropy of thermal expansion coefficients is briefly discussed.

The phase stability and elastic moduli of Cr2Nb are investigated by first-principles calculations. Heats of formation are calculated and compared for the three Laves phases (C15, C14, and C36). It is found that the C15 phase is the ground-state structure with the lowest energy and the C36 phase is an intermediate state between C15 and C14. These three phases, however, are very close in energy, indicating low stacking fault energies in this system. For the ground-state C15 phase, we calculate three elastic constants from which the shear and Young’s moduli are obtained. It is found that these calculated moduli are smaller than the experimental values obtained from polycrystals.

We have examined a variety of structures for the {510} symmetric tilt boundary in Si and Ge, using tight-binding and first-principles calculations. These calculations show that the observed structure in Si is the lowest-energy structure, despite the fact that it is more complicated than what is necessary to preserve fourfold coordination. Contrary to calculations using a Tersoff potential, first-principles calculations show that the energy depends strongly upon the structure. A recently developed tight-binding model for Si produces results in very good agreement with the first-principles calculations. Electronic density of states calculations based upon this model show no evidence of midgap states and little evidence of electronic states localized to the grain boundary.

First-principles local-density-functional calculations have been used to investigate the defect properties of FeAl: (1) the ordering behavior and equilibrium point defects; (2) the binding interactions among vacancies; and (3) the mechanism underlying the hydrogen-induced embrittlement effect. The point defect structure is dominated by substitutional antisite defects on both sublattices and thermal vacancies on the Fe sublattice. It is predicted that the binding between vacancies is strongly attractive when vacancies are aligned along the [100] direction, and becomes weakly repulsive when vacancies are [110]a or [111]a apart (where a is the lattice constant). The binding interaction between vacancies is found to manifest the oscillations in the vacancy-induced charge density on neighboring lattice sites. Energetically, vacancies (and/or vacancy clusters) provide favorable lattice sites for hydrogen segregation. The hydrogen-induced embrittlement is explained in terms of a charge transfer mechanism and the localization behavior of induced charge at the hydrogen site.

The intrinsic values of interfacial energies based on first-principles calculations, including atomic relaxation, were obtained for the three types of γ/γ interfaces and the α2/γ lamellar boundary in two- phase TiAl alloy. The pseudo-twin boundary energy is highest, ΓP = 270 mJ/m2, and the true-twin boundary energy is lowest, ΓP = 60 mJ/m2. Planar fault energies at pseudo-twin and 120°-rotational interfaces are markedly different from those in the bulk of the γ-phase, i.e., approximately, EAPB and ECSF decreases by more than 30% and ESISF increases by threefold. Enhanced mobility of ordinary dislocations along the γ/γ and α2/α interfaces (except true twin boundaries) is predicted based on the reduced ECSF values at the interfaces.

First-principles atomic cluster calculations have been carried out in the local density approximation to understand the segregation behavior and strengthening effects of boron in Ni3Al. The binding energy of boron is calculated in lattice fragment clusters representing the perfect crystal, as well as various defect sites. The agreement between trends in energetics determined for small clusters and periodic supercells indicates the dominant role of boron’s interaction with nearest-neighbors of the host. The stereochemical factor underlying boron’s preferential bonding to nickel atoms in four-fold planar coordination (i.e., sp3 hybridization) suggests a mechanism for the boron-effect in Ni3Al: increased cohesion provides a driving force for B segregation to open sites, such as at Ni-enriched grain boundary sites, and the strengthening is a result of strong localized Ni–B covalent bond formation.

First-principles local-density-functional calculations have been used to investigate the equilibrium point defect structure and boron-defect interactions in Ni 3 Al. The dominant point defect types in off-stoichiometric Ni 3 Al are substitutional antisite defects on both sublattices. The boron binding energy is dependent on lattice coordination; it is strongest near vacancy sites with a nearst-neighbour nickel coordination number of about four (instead of six as in the defect-free interstitial site) and with no aluminium atom nearest-neighbours. This suggests that boron tends to segregate to “open” defect sites and to enhance cohesion through the formation of localized Ni-B covalent bonds. Comparison of the binding energies of boron and carbon in Ni 3 Al shows that boron has a stronger tendency to segregate to “open” sites than carbon.

We have examined a number of symmetric tilt grain boundary structures in diamond, using a transferable tight-binding (TB) carbon potential. Utilizing O(N) tight-binding molecular dynamics, we are able to simulate several thousands of atoms, allowing us to examine low-angle boundaries with large periodicities. The lowest energy structure is the {111} twin boundary, corresponding to the \ensuremath{\Sigma}=3 70.53\ifmmode^\circ\else\textdegree\fi{} grain boundary. For angles less than 70.53\ifmmode^\circ\else\textdegree\fi{}, the boundaries are composed of a series of b=〈011\ifmmode\bar\else\textasciimacron\fi{}〉 edge dislocations. The core structures of these dislocations are composed of five- and seven-atom rings. For these low-angle structures, the tight-binding energies have trends similar to those found using a Tersoff potential; both models give results that are consistent with linear elasticity theory. For the high-angle {211} and {311} boundaries, the tight-binding model predicts a reconstruction along the 〈011〉 direction, in order to eliminate dangling bonds. \textcopyright{} 1996 The American Physical Society.

First-principles calculations have been performed to investigate the site preference of ternary alloying additions in FeAl and NiAl. In FeAl, Cr and Ti are found to occupy the Al sublattice whereas Ni has a distinct preference for the Fe sublattice. The site substitutional behavior of 3d ternary elements in FeAl can be explained in terms of the trends in the heat-of-formation. In Al-rich NiAl, Fe atoms occupy exclusively the Ni sublattice. In Ni-rich NiAl, because of the small enthalpy difference between Fe occupying Al and Ni sublattices (i.e. less than 0.1 eV with a preference of Fe for Al sites at 0 K), the site distributions of Fe in these alloys are found to vary with alloy composition and temperature. Due to the large difference in the local magnetic moments between Fe atoms occupying Al and Ni sublattices (with values of 2.4 μ B and less than 0.1 μ B , respectively) in NiAl, magnetic susceptibility measurement should be the most effective way to measure the site distributions of Fe in NiAl.

We have presented results of first-principles calculations of the elastic constants of Ti3Al, and the planar fault energies in Ti3Al and at the TiAl Ti 3 Al interfaces. Similar to those found in Ti, the bonding in Ti3Al is dominated by the multi-centered bonding interactions, but is enhanced by the presence of Al atoms. The bonding enhancement is manifested in stiffer elastic moduli and considerably high APB energies (even in the case where the nearest-neighbor coordination of atoms is not altered by the presence of APB, i.e., APB-I on the prism plane). The APB and CSF energies are reduced at the TiAl Ti 3 Al interface from those of either constituent phase, since the long-range bonding interactions are disrupted at the interface.

Phase stability and bulk properties of TiAl and Ti 3 Al are investigated based on the first-principles local-density-functional approach, and deformation and fracture behavior of two-phase lamellar γ-TiAl alloys are analyzed using the linear elasticity theory applied to interfaces, dislocations, and cracks. In terms of the calculated elastic constants, Young's and shear moduli are higher in Ti 3 Al than in TiAl. The coupling effect due to elastic incompatibility reduces the misfit strain at an interface when a tensile stress is applied normal to it. The mode mixity of stress concentration by a dislocation pile-up plays an important role in determining whether slip transfer or microcracking occurs at the interface. The fracture morphology reported in notched polysynthetically twinned crystals is explained in terms of the elastic interaction of a (11 2 )[1 1 0] crack with an α 2 -Ti 3 Al plate. Additional factors that are involved in the brittle-to-ductile transition behavior of TiAl are discussed.

First-principles calculations have been performed to study the chemical bonding trends and the phase stability of 3[ital d] transition-metal (TM) aluminides with equiatomic composition. The physical mechanisms for ordering in both the [ital L]1[sub 0] and [ital B]2-type aluminides are presented. It is shown that the dominant factor for early TM aluminides is the directional bonding between the [ital d] orbitals of TM atoms, whereas for late TM aluminides, charge transfer and hybridization between Al [ital s] and [ital p] states, and TM [ital d] states play more important roles in the bonding mechanism and the properties of lattice defects. Our calculations also show the existence of a ferromagnetic phase for MnAl in the [ital L]1[sub 0] structure, which is stabilized by the formation of a magnetic moment on the Mn sites (with a value of 2.0 [mu][sub [ital B]]). The chemical trends in the structural stability and heats of formation of 3[ital d] TM aluminides are found to correlate well with the densities of states at the Fermi level.

First-principles theory has been developed to understand the unique material state underlying the ultra stability of O-enriched (with Ti and Y solute additions) nanoclusters (NCs) observed in nanostructured Fe. We show that an essential condition for the formation of these 2--4 nm diameter NCs is the existence of their exceptionally stable interface, as manifested in a higher (i.e., more negative) O-binding energy on the interface (${E}_{S}$) than in the interior (${E}_{B}$) of the NCs. The difference between ${E}_{S}$ and ${E}_{B}$ is due to the effect of strain, originating from the solute-solute repulsion and the presence of O in the form of O-vacancy (O:V) pair. A Friedel-type oscillation pattern is found for the distance dependence of the solute-(O:V) and (O:V)-(O:V) interaction energies. The strain is also important in controlling the size of NCs, which are predicted to exist only within a narrow O concentration range.

Point-defect energetics and diffusion mechanisms in graphite are investigated using a semi-empirical tight-binding-force model. Possible diffusion processes associated with point-defect (i.e., vacancies and interstitials) and nondefect (i.e., atomic exchange) mechanisms are analyzed. It is found that self-diffusion in graphite in the direction parallel to the basal plane can be mediated by vacancies. However, since the calculated vacancy- and interstital-formation energies are nearly equal, it is argued that Frenkel pairs could exist as equilibrium defects. In this case, at high enough temperatures, self-diffusion parallel to the basal plane should occur by an interstitial mechanism because the migration energy of the interstitial is much lower.

Equilibrium point defects and their relation to the contrasting mechanical behavior of NiAl and FeAl are investigated. For NiAl, the defect structure is dominated by two types of defects---monovacancies on the Ni sites and substitutional antisite defects on the Al sites. The defect structure of FeAl differs from that of NiAl in the occurrence of antisite defects at the transition-metal sites for Al-rich alloys and the tendency for vacancy clustering. The strong ordering (and brittleness) of NiAl is attributed mainly to the difference in atomic size between constituent atoms.

The surface and interface magnetism of metastable bcc Ni as an overlayer on Fe(001)---originally synthesized by Heinrich et al.$mdash---is investigated using the highly precise total-energy full-potential linearized augmented plane-wave method. Two systems---a monolayer (1Ni/Fe) and a bilayer (2Ni/Fe) coverage on Fe(001)---are investigated and compared. The Ni-Fe interlayer spacings are found to be relaxed downward by 2.5% and 4% for 1Ni/Fe and 2Ni/Fe, respectively. The effect of the Ni-Fe hybridization on the electronic and magnetic properties is examined. In both overlayer systems the magnetic moments at the interface Ni and Fe sites are enhanced relative to their corresponding bulk values. The magnetic moment (0.69${\mathrm{\ensuremath{\mu}}}_{\mathit{B}}$) of the surface Ni atom of 2Ni/Fe is found to be reduced from that (0.86${\mathrm{\ensuremath{\mu}}}_{\mathit{B}}$) of the surface Ni atom on 1Ni/Fe. A manifestation of the moment enhancement at the interface Fe site is a positive contribution to the hyperfine field from the valence electrons.

The bonding mechanism and point defect structure in TiAl were investigated using a first-principles quantum mechanical calculation. The most remarkable feature found in the calculated binding charge density is the polarization of the p-electron at the aluminum sites, which gives rise to a large bond bending force between the Ti and Al layers. For the point defects, the absence of structural vacancies is predicted in TiAl, and the deviations from stoichiometry are accommodated by the substitutional antisite defects on both sublattices. High vacancy formation energies are obtained, which are closely related to the strong TiAl bonding and the similar atomic radii of Ti and Al.

The structural and electronic properties of the high-pressure molecular compound ${\text{SiH}}_{4}{({\text{H}}_{2})}_{2}$ have been calculated using density-functional theory. We identify the molecular hydrogen positions within the face-centered cubic unit cell and further find that pressure-induced intermolecular interaction between ${\text{SiH}}_{4}$ and ${\text{H}}_{2}$ units plays an important role in stabilizing this new compound. The electronic structure is characterized by a wide band gap of 6.1 eV at 6.8 GPa, which closes with pressure and finally becomes metallic at 200 GPa due to electronic band overlap accompanied by a structure change. These findings have potential implications for understanding metallization and superconductivity in ${\text{H}}_{2}$.

The immense interest in carbon nanomaterials continues to stimulate intense research activities aimed to realize carbon nanowires, since linear chains of carbon atoms are expected to display novel and technologically relevant optical, electrical and mechanical properties. Although various allotropes of carbon (e.g., diamond, nanotubes, graphene, etc.) are among the best known materials, it remains challenging to stabilize carbon in the one-dimensional form because of the difficulty to suitably saturate the dangling bonds of carbon. Here, we show through first-principles calculations that ordered polymeric carbon chains can be stabilized in solid Li$_2$C$_2$ under moderate pressure. This pressure-induced phase (above 5 GPa) consists of parallel arrays of twofold zigzag carbon chains embedded in lithium cages, which display a metallic character due to the formation of partially occupied carbon lone-pair states in \emph{sp}$^2$-like hybrids. It is found that this phase remains the most favorable one in a wide range of pressure. At extreme pressure (larger the 215 GPa) a structural and electronic phase transition towards an insulating single-bonded threefold-coordinated carbon network is predicted., Comment: 10 pages, 6 figures

The electronic and magnetic properties of recently synthesized tetragonal CuO with c/a>1 is calculated by means of hybrid density-functional theory. We predict that this tetragonal phase orders antiferromagnetically and has an exceptionally high Neeacutel temperature T(N)approximate to 800 K, which makes it an ideal candidate for doping experiments and a potential parent of superconductors. The electronic structure is characterized by a charge-transfer gap of 2.7 eV whereas the magnetic properties are dominated by the antiferromagnetic Cu-O-Cu interactions along the nearest-neighbor [100] direction. In addition, we predict the second tetragonal CuO phase with a c/a ratio < 1 with a different antiferromagnetic ordering and a similar high T(N). We suggest that this phase could be synthesized by epitaxial growth.

The role of deformation twinning in the strength and ductility of metals and ordered interrnetallic alloys is examined on the basis of crystallography, energetics and kinetics of deformation twinning. A systematic analysis is made by taking Ti, T13Al, TiAl, and A13Ti as four model systems. In comparison with profuse twinning in Ti, the intrinsic difficulty of twinning in T13Al is rationalized in terms of the interchange shuffling mechanism. A fault (SISF) dragging mechanism based on the interaction torque explains the physical source for the low mobility of screw superdislocations in TiAI, which may lead to (1Ii) [1Ii] twin nucleation. In TiAI and A13Ti alloys, the twin-slip (ordinary) conjugate relationship makes an important contribution to the strain compatibility for high,temperature plasticity. Potentially beneficial alloying additions to promote twinning are discussed. (I) Research sponsored by the Division of Materials Sciences, Office of Basic Energy Sciences, U-SDepartment of Energy, under contract DE-ACo5-840R21400 with Martin Marietta Energy Systems, Inc. (2) Permanent address.Department of Metallurgical and Materials Engineering, Michigan Technological University, Houghton, MI 49931, U-S-Aloss JOURNAL DE PHYSIQUE III M 6

Results of a spin polarized full potential linearized augmented plane wave (FLAPW) study of the hcp Co(0001) surface is reported. The calculated surface magnetic moment (≈1.76μ B ) is enhanced by ≈7% from the calculated bulk value (≈1.64μ B ) and by ≈9% from the experimental bulk value (≈1.57μ B ), a result of the narrowing of the 3d electron bands at the surface. In this system, the surface effect is seen to extend into the second atomic layer below the surface, in contrast with the one layer range found for the non-close-packed cobalt surfaces (e.g., bcc or fcc (001)). As estimated from the total energy calculations, the surface energy of hcp Co(0001) is ≈2 J/m 2 .

Surface states, electronic structure and surface energy of a clean Ag(001) surface are determined using the local density full potential linearized augmented plane wave calculations for an 11 layer slab.

First-principles theory was used to investigate the roles of bond topology and covalency in the phase stability and elastic strength of 5d transition-metal diborides, focusing on elements (M=W, Re, Os) that have among the lowest compressibilities of all metals. Among the phases studied, the ReB(2)-type structure exhibits the largest incompressibility (c axis), comparable to that of diamond. This ReB(2) structure is predicted to be the ground-state phase for WB(2) and a pressure-induced phase (above 2.5 GPa) for OsB(2). Both strong covalency and a zigzag topology of interconnected bonds underlie these ultraincompressibilities. Interestingly, the Vickers hardness of WB(2) is estimated to be similar to that of superhard ReB(2).

Recently, the unusual high-pressure cubic gauche structure of solid nitrogen (cg-N) was experimentally synthesized. Our first-principles calculations reveal that cg-N represents a new class of covalent solids in which the atoms are threefold coordinated, yet the structure is stabilized by the presence of near-tetrahedral $s{p}^{3}$-hybridized electronic states. This bonding scheme results in a strong covalency, exceptional mechanical properties, and the stabilization of cg-N as a high-energy-capacity material. The cg-N is an insulator with a wide gap that is nearly constant over a wide pressure range. The mechanical failure mode in cg-N is dominated by the shear type, and the calculated (110) $⟨1\overline{1}0⟩$ shear strength of $41.3\phantom{\rule{0.3em}{0ex}}\mathrm{GPa}$ (at which the lattice becomes unstable) sets an upper bound for its ideal strength. This shear-induced structural instability is associated with the structural transformation from the cg-N structure to the monoclinic structure.

A second-variation full-potential linear augmented-plane-wave total-energy method for thin-film ferromagnetic systems is used to study the spin-orbit-interaction contribution to the magnetic anisotropy. For a free-standing Fe monolayer, the spin magnetization is determined to lie in the plane. Results for Fe monolayers on Au(001), Ag(001), and Pd(001) substrates indicate a preference for the spin direction to be perpendicular to the plane of the film. Computational details for this magnetic anisotropy are also discussed.

Magnetism in low dimensional systems continues as an exciting area of research both experimentally and theoretically. Recent developments of electronic structure theory include realistic theoretical local spin-density predictions of the electronic and magnetic properties of bulk materials, their surfaces, thin films and multi-layers. These results are essential ingredients in guiding the choice of systems to be studied experimentally, and in the interpretation of experimental results. This paper reports some recent predictions of theory including first principles determinations of magnetic anisotropy.

First-principles studies identify a vacancy mechanism underlying the unusually high O solubility and nucleation of stable O-enriched nanoclusters in defect-containing Fe. Oxygen, confined as an interstitial, shows an exceptionally high affinity for vacancies, an effect enhanced by spin polarization. If vacancies preexist, the O-vacancy pair formation energy essentially vanishes, allowing the O concentration to approach that of the vacancies. This O-vacancy mechanism enables the nucleation of O-enriched nanoclusters, that attract solutes with high O affinities (Ti and Y) and strengthen Fe-based alloys.

This study investigates the absorption of hydrogen molecules between graphite layers using both first principles calculations and classical grand canonical Monte Carlo simulations. While a recent theoretical study showed that graphite layers have high storage capacity at room temperature, previous simulation results on hydrogen-graphite systems showed otherwise. Our first-principles calculations suggest that it is possible to store hydrogen molecules between the graphene layers if the energetically unfavorable initial absorption stage could be overcome. The barrier to the initial absorption originates from the large lattice strain required for H2 absorption: small amounts of initial absorption cause an interlayer expansion of more than 60%. To determine if significant storage is indeed possible at finite temperature (and pressure), we performed grand canonical Monte Carlo H2-absorption simulations with variable graphite interlayer spacing. Using two different potentials for the H2-C interaction, we found low H2 mass uptake at room temperature and moderate pressures (e.g., close to 2 wt-% at 298 K and 5 2 MPa.). Our results suggest that a graphite pore width or interlayer spacing around 6 has the optimum absorption capacity. PACS numbers: 68.43.-h, 81.05.Uw, 82.20.Wt, 83.10.-y