Electronic structure and elasticity of the Ta-W solid solution
Kareem Abdelmaqsoud, John R. Kitchin, and M. Widom, “Electronic structure and elasticity of the Ta-W solid solution”, Physical Review Materials 10, 053605 (2026). https://doi.org/10.1103/ph3y-8lnq
This summary was written by a language model from the paper and has not been checked line by line. Rely on the published paper for details.
Question
Ta(1-x)Wx alloys pass from ductile behavior on the Ta-rich side to brittleness at high W. Ductility is commonly judged from elastic constants, through a high Pugh ratio P = K/G, a high Poisson ratio, or a positive Cauchy pressure, but these rules are empirical. The authors’ earlier study of Ta-W with a little Nb found that the shear moduli dip slightly up to x = 1/2 and then turn upward, while the bulk modulus rises almost linearly, so the Pugh ratio shows a broad maximum followed by a steep fall (Abdelmaqsoud et al. 2026). This paper asks why: can the composition dependence of the elastic constants be traced to specific features of the electronic structure (Abdelmaqsoud et al. 2026)?
Approach
Elastic constants of the binary were computed with VASP and the PBE functional (Tapv and Wsv PAW potentials, 330 eV cutoff), using fully relaxed 16-atom cubic structures spanning x = 0 to 1 in steps of 1/8 and finite-difference strains. The electronic analysis rests on a rigid band model: the band structure and density of states (DOS) of equiatomic TaW in the ordered B2 (cP2) structure are taken as fixed, and composition only moves the Fermi level according to the valence electron count (VEC), with 5 electrons per Ta and 6 per W. An appendix tests this by comparing the DOS of pure Ta, pure W, B2 TaW, a 16-atom special quasirandom structure and the ground-state TaW structure. The authors then examine orbital-projected DOS and “fat band” plots, crystal orbital Hamilton populations (COHP, computed with LOBSTER) for nearest-neighbor (NN) and next-nearest-neighbor (NNN) bonds, and real-space plots of individual wave functions (WaveTrans). Ground-state orderings from the ATAT toolkit were used to gauge the effect of chemical order, and five other group V-VI binaries were checked for the same behavior.
Key findings
The cubic elastic constants reproduce the ternary trends: the tetragonal shear modulus C’ = (C11 - C12)/2 is nearly flat for x below 1/2 and climbs steeply above it, C44 follows the same pattern with smaller variation, and K = (C11 + 2C12)/3 rises steadily. The DOS of TaW has a high shoulder just below the Fermi level and a deep pseudogap above it. Between VEC 5.0 and 5.5 (0 to 1/2 W) the Fermi level DOS is high and dominated by T2g (dxy, dxz, dyz) states. Between VEC 5.5 and 6.0 it drops abruptly. The bands explain the drop: the three T2g bands meet with threefold degeneracy and zero slope at the Gamma point only 0.12 eV above the TaW Fermi level, and above that energy two of them are nearly flat.
COHP analysis shows that NN bonding is strong at low VEC and turns weakly antibonding at high VEC, mostly because T2g-T2g interactions shift from strongly bonding to nearly neutral. NNN interactions are antibonding throughout, but less so at high VEC. All of these changes occur near the 0.12 eV degeneracy. The wave functions supply the mechanism. Electrons added up to x = 1/2 fill T2g orbitals that bond along the close-packed <111> directions, raising both C11 and C12, but the same orbitals are antibonding along the <001> NNN bonds, which contribute only to C11, so C11 grows slowly. Once two of the three T2g orbitals are full, the NNN antibonding weakens and C11 rises faster while C12 levels off, producing the upturn in C’. Shear strain of type uxy affects NN bonds only weakly, which the authors suggest is why C44 varies less. The Pugh ratio peaks broadly for x below about 1/2, and the Cauchy pressure is positive at every composition but peaks over the same range.
The rigid band test passes for Ta-W: all structures, including the random and ground-state ones, share the shoulder from VEC 5.0 to 5.5 and the pseudogap from 5.5 to 6.0. The energy-minimizing orderings (for example an oC12 structure at TaW) are noncubic, and their Voigt-Reuss-Hill Pugh ratios fall below those of the cubic cells, most strongly for x below 1/2, so chemical short-range order is expected to reduce ductility. VCr, NbMo, NbW and TaMo (with TaW) all show Pugh ratios peaking below x = 1/2 and falling at higher x, though peak positions and heights differ and the rigid band picture holds less well across systems than within Ta-W.
Significance
The paper gives an orbital-level account of the “valley of brittleness” in group V-VI refractory alloys: the change in slope of the shear moduli near equiatomic Ta-W coincides with the Fermi level passing a T2g degeneracy and entering a pseudogap, which switches off NN T2g bonding and NNN antibonding. It supports the view that valence electron count, rather than volume, sets the trend of the Pugh ratio across these alloys, and provides the electronic explanation for the C’ jump used in the companion alloy design and printing study (Abdelmaqsoud et al. 2026).