Announcement: changes to elastic tensor data


Hello all,

We’ve recently decide to make a few small changes to the presentation of elastic tensors on our website. Due to numerical errors, our elastic tensor workflow does not necessarily result in an elastic tensor that is precisely consistent with the crystal symmetry of its corresponding structure. In addition, the default view of the elastic tensor was in “POSCAR”-format, i. e. relevant to the basis implicit in the conventional standard structure. To be more consistent with the published reference on elastic constants, which chooses the IEEE standard convention as the default, we’re changing a few things in the next few days with respect to how the elastic tensors are presented on the web and in the API.

  • Elastic tensors presented on are in IEEE format, are symmetrized such that they are consistent with the crystal structure, and are rounded to the nearest GPa.
  • “elastic_tensor_original” is now included in the API elasticity document, referring to the unsymmetrized, “POSCAR”-format tensor (i. e. corresponding to the conventional standard unit cell as generated by pymatgen).
  • Derived values (compliance tensor, K, G, poisson ratio, and anisotropy) are shown such that they are derived from the symmetrized IEEE tensor, rather than the raw tensor.
  • The calculations wiki will be updated to reflect the new schema

If you have any questions or feedback regarding this, please feel free to post a reply in this thread or to e-mail me at montoyjh [at] lbl [dot] gov. Thanks for using the Materials Project!