Query Shomate Polynomial Using pymatgen


I would like to write a query to provide the Shomate polynomials for different species (e.g. the A - H values for Cu listed here). I understand that pymatgen.ext.matproj.query allows me to get information related to the DFT calculations but is there any way I can return the thermodynamic data (perhaps as a pymatgen.analysis.thermochemistry object)?


Hi Jon,

Currently, the data_type=“exp” option for MPRester.get_data, given a chemical formula as argument, will return any and all experimental (non-DFT) thermo data we have. Example:

from pymatgen import MPRester
from pymatgen.analysis.thermochemistry import ThermoData

mpr = MPRester()

data = mpr.get_data("Cu", data_type="exp")
assert(all(isinstance(d, ThermoData) for d in data)) # True


This is perfect. Thanks, Donny!

Hello @dwinston

Do you know the way to get ThermoData via query end point? I am trying to understand what property name has to be passed in the below example code to get the thermo data information.

 from pymatgen import MPRester
 from pymatgen.analysis.thermochemistry import ThermoData

  mpr = MPRester()

  data = m.query(criteria={"task_id": "mp-85"}, properties=['?']))


All of the thermo data is now available via MPContribs: https://contribs.materialsproject.org/projects/experimental_thermo/

Hello @shyamd

Thank you for the response. When we use exp mapping endpoint from the material project, does it gets data from MPContribs?


sample data:

{"@module": “pymatgen.analysis.thermochemistry”, “@class”: “ThermoData”, “type”: “B”, “formula”: “Fe2O3”, “compound_name”: “Fe2O3”, “phaseinfo”: “hemat”, “value”: 7.782e-05, “temp_range”: [298, 298], “method”: “”, “ref”: “O. Kubaschewski, C. Alcock, P. Spencer, Materials Thermochemistry, 6th ed., Oxford, Pergamom Press, 1993.”, “uncertainty”: null}


It’s still coming from our internal endpoint. When we switch to our new website + new API. All of this data will come from MPContribs, since its easier to document, and provides a means of citing the original authors and those that put together the data.

In general the MPContribs data set is easier to access and understand. You can view it directly in the portal and search there as well, or export just the systems of interest to a CSV. Those that want programmatic access still have it, with more search filters.

Thank you for the clarification @shyamd. May I know what is the timeline for new webside + API release?