Machine Learning Prediction of Heat Capacity for Solid Inorganics

  • Steven K. Kauwe
  • Jake Graser
  • Antonio Vazquez
  • Taylor D. Sparks
Technical Article


Many thermodynamic calculations and engineering applications require the temperature-dependent heat capacity (Cp) of a material to be known a priori. First-principle calculations of heat capacities can stand in place of experimental information, but these calculations are costly and expensive. Here, we report on our creation of a high-throughput supervised machine learning-based tool to predict temperature-dependent heat capacity. We demonstrate that material heat capacity can be correlated to a number of elemental and atomic properties. The machine learning method predicts heat capacity for thousands of compounds in seconds, suggesting facile implementation into integrated computational materials engineering (ICME) processes. In this context, we consider its use to replace Neumann-Kopp predictions as a high-throughput screening tool to help identify new materials as candidates for engineering processes. Also promising is the enhanced speed and performance compared to cation/anion contribution methods at elevated temperatures as well as the ability to improve future predictions as more data are made available. This machine learning method only requires formula inputs when calculating heat capacity and can be completely automated. This is an improvement to common best-practice methods such as cation/anion contributions or mixed-oxide approaches which are limited in application to specific materials and require case-by-case considerations.


Neumann-Kopp Heat capacity Inorganic solids Machine learning Material design High-throughput 



The authors gratefully acknowledge support from the NSF CAREER Award DMR 1651668.

Supplementary material (783 kb)
(ZIP 782 KB)


  1. 1.
    Altmann A, Toloşi L, Sander O, Lengauer T (2010) Permutation importance: a corrected feature importance measure. Bioinformatics 26(10):1340–1347. CrossRefGoogle Scholar
  2. 2.
    Barin I, Platzki G (1989) Thermochemical data of pure substances. vol 304, Wiley Online LibraryGoogle Scholar
  3. 3.
    Gaultois MW, Oliynyk A, Mar A, Sparks TD, Mulholland GJ, Meredig B (2016) Perspective: Web-based machine learning models for real-time screening of thermoelectric materials properties. APL Mater 4 (5):053213. CrossRefGoogle Scholar
  4. 4.
    Gaultois MW, Sparks TD, Borg CK, Seshadri R, Bonificio WD, Clarke DR (2013) Data-driven review of thermoelectric materials: performance and resource considerations. Chem Mater 25(15):2911–2920. CrossRefGoogle Scholar
  5. 5.
    Ghadbeigi L, Harada JK, Lettiere BR, Sparks TD (2015) Performance and resource considerations of li-ion battery electrode materials. Energy Environ Sci 8(6):1640–1650. CrossRefGoogle Scholar
  6. 6.
    Graser J, Kauwe SK, Sparks TD (2017) Machine learning and energy minimization approaches for crystal structure predictions: A review and new horizons, Chemistry of Materials, in pressGoogle Scholar
  7. 7.
    III LMC, Taylor RE (1975) Radiation loss in the flash method for thermal diffusivity. J Appl Phys 46 (2):714–719. CrossRefGoogle Scholar
  8. 8.
    Kittel C (2005) Introduction to solid state physics. Wiley, HobokenGoogle Scholar
  9. 9.
    Leitner J, Voka P, Sedmidubský D, Svoboda P (2010) Application of Neumann–Kopp rule for the estimation of heat capacity of mixed oxides. Thermochimica Acta 497(1):7–13. CrossRefGoogle Scholar
  10. 10.
    Mostafa ATMG, Eakman JM, Montoya MM, Yarbro SL (1996) Prediction of heat capacities of solid inorganic salts from group contributions. Ind Eng Chem Res 35(1):343–348. CrossRefGoogle Scholar
  11. 11.
    Narasimhan S, de Gironcoli S (2002) Ab initio. Phys Rev B 65:064302. CrossRefGoogle Scholar
  12. 12.
    Oliynyk A, Antono E, Sparks TD, Ghadbeigi L, Gaultois MW, Meredig B, Mar A (2016) High-throughput machine-learning-driven synthesis of full-Heusler compounds. Chem Mater 28(20):7324–7331. CrossRefGoogle Scholar
  13. 13.
    Oliynyk A, Mar A Discovery of intermetallic compounds from traditional to machine-learning approaches. Accounts of chemical research.
  14. 14.
    Parker WJ, Jenkins RJ, Butler CP, Abbott GL (1961) Flash method of determining thermal diffusivity, heat capacity, and thermal conductivity. J Appl Phys 32(9):1679–1684. CrossRefGoogle Scholar
  15. 15.
    Pedregosa F, Varoquaux G, Gramfort A, Michel V, Thirion B, Grisel O, Blondel M, Prettenhofer P, Weiss R, Dubourg V, Vanderplas J, Passos A, Cournapeau D, Brucher M, Perrot M, Duchesnay E (2011) Scikit-learn: Machine learning in Python. J Mach Learn Res 12:2825–2830Google Scholar
  16. 16.
    Seshadri R, Sparks TD (2016) Perspective: Interactive material property databases through aggregation of literature data. APL Mater 4(5):053206. CrossRefGoogle Scholar
  17. 17.
    Sparks TD, Fuierer PA, Clarke DR (2010) Anisotropic thermal diffusivity and conductivity of la-doped strontium niobate Sr2Nb2O7. J Amer Ceram Soc 93(4):1136–1141. CrossRefGoogle Scholar
  18. 18.
    Sparks TD, Gaultois MW, Oliynyk A, Brgoch J, Meredig B (2016) Data mining our way to the next generation of thermoelectrics. Scr Mater 111:10–15. CrossRefGoogle Scholar
  19. 19.
    Tehrani AM, Ghadbeigi L, Brgoch J, Sparks TD (2017) Balancing mechanical properties and sustainability in the search for superhard materials. Integr Mater Manuf Innov 6(1):1–8. CrossRefGoogle Scholar
  20. 20.
    Thermart: Freed-thermodynamic database (2017).

Copyright information

© The Minerals, Metals & Materials Society 2018

Authors and Affiliations

  1. 1.Department of Materials Science and EngineeringUniversity of UtahSalt Lake CityUSA

Personalised recommendations