Explore effective charge in electromigration
The driving force of electromigration is typically considered as a combination of the electron wind force, which is due to electron-ion scattering, and the direct force, which originates from the external electric field. The species’ effective charge (z*) sets the scale and diffusion direction of the EM driving force. The larger the effective charge is, the greater the driving force at a given current density level. The negative and positive sign of effective charge represent the diffusion direction toward anode and cathode side, respectively.
In this work, machine learning approaches were employed to model the effective charge as a linear function of physically meaningful elemental properties. Average fivefold (leave-out-alloy-group) cross-validation yielded root-mean-square-error divided by whole data set standard deviation (RMSE/σ) values of 0.37 ± 0.01 (0.22 ± 0.18), respectively, and R2 values of 0.86. Extrapolation to z* of totally new alloys showed limited but potentially useful predictive ability. The model was used in predicting z* for technologically relevant host–impurity pairs.
Full fit and 5-fold cross validation test.
Leave-out alloy group cross validation test.
Leave-out element group cross validation test.
Used the ML model to decipher the complex physics of effective charge.
Explore effective charge in different host element over the whole periodic table.
For more info, please visit:
Y.-C. Liu, B. Afflerbach, R. Jacobs, S.-K. Lin and D. Morgan, “Exploring effective charge in electromigration using machine learning”, MRS Communications, 1-9 (2019) (Special Issue Research Letter: Artificial Intelligence [by invitation only]) (IF = 1.997, Rank: 197/314 = 63% in MATERIALS SCIENCE, MULTIDISCIPLINARY -- SCIE). <Journal Pub>
Youtube video: