Deep learning of surface elastic chemical potential in strained films: from statics to dynamics

JSON Export

{
  "revision": 3, 
  "id": "2053", 
  "created": "2024-01-16T16:21:22.507151+00:00", 
  "metadata": {
    "doi": "10.24435/materialscloud:ta-fz", 
    "status": "published", 
    "title": "Deep learning of surface elastic chemical potential in strained films: from statics to dynamics", 
    "mcid": "2024.10", 
    "license_addendum": null, 
    "_files": [
      {
        "description": "Description of the record content", 
        "key": "README.txt", 
        "size": 413, 
        "checksum": "md5:bd5486347916e4cc267ee6fa4edd1665"
      }, 
      {
        "description": "Dataset and test cases (statics and dynamics) with elastic chemical potential computed by Green's approximation", 
        "key": "GA.zip", 
        "size": 3086889897, 
        "checksum": "md5:378976c315b6b10f3e1d0af0c6b54228"
      }, 
      {
        "description": "Dataset and test cases (statics and dynamics) with elastic chemical potential computed by Finite Element Method", 
        "key": "FEM.zip", 
        "size": 1065631808, 
        "checksum": "md5:d87ce41cc5f4728e372d92428b4fae23"
      }
    ], 
    "owner": 1221, 
    "_oai": {
      "id": "oai:materialscloud.org:2053"
    }, 
    "keywords": [
      "deep learning", 
      "elastic chemical potential", 
      "surface evolution", 
      "thin films"
    ], 
    "conceptrecid": "2039", 
    "is_last": true, 
    "references": [
      {
        "type": "Preprint", 
        "comment": "Preprint where the method is described and selected cases from the datasets in the record are illustrated", 
        "citation": "L. Mart\u00edn-Encinar, D. Lanzoni, A. Fantasia, F. Rovaris, R. Bergamaschini, F. Montalenti \"Deep learning of surface elastic chemical potential in strained films: from statics to dynamics\" (in preparation)"
      }
    ], 
    "publication_date": "Jan 23, 2024, 11:05:40", 
    "license": "Creative Commons Attribution 4.0 International", 
    "id": "2053", 
    "description": "We develop a convolutional neural network (NN) approach able to predict the elastic contribution to chemical potential \u03bc\u2091 at the surface of a 2D strained film given its profile h(x). Arbitrary h(x) profiles are obtained by using a Perlin Noise generator and the corresponding \u03bc\u2091 profiles are calculated either by a Green's function approximation (GA) or by Finite Element Method (FEM). First, a large dataset is produced by exploiting the GA method and it is then used for the training of the NN model. The performance of the trained NN is extensively examined, demonstrating its ability to predict \u03bc\u2091 looking both to the training/validation set and to an additional testing set containing different profiles, including sinusoids, gaussians and sharp peaks never considered in the NN training. The NN is then applied to simulate the morphological evolution of strained Ge films, where the predicted \u03bc\u2091 at each integration timestep plays the role of driving force for material redistribution in competition with a surface energy term (proportional to local curvature) and eventually a wetting energy contribution. Both surface diffusion and evaporation/condensation dynamics are considered and the proposed NN approach is shown to well match the evolution expected by using GA. On this basis, a smaller dataset is built with \u03bc\u2091 profiles calculated by FEM and a new NN model is trained on it. Once again the trained NN-model returns reliable prediction of the FEM \u03bc\u2091. The findings suggest that the proposed NN-based strategy can be used in replacement of the computationally intensive FEM calculations, enabling the simulation of larger scales and longer time scales untreatable by direct FEM calculation.", 
    "version": 2, 
    "contributors": [
      {
        "email": "luis.martin.encinar@uva.es", 
        "affiliations": [
          "Dpto. de Electricidad y Electr\u00f3nica, E.T.S.I. de Telecomunicaci\u00f3n, Universidad de Valladolid, Paseo Bel\u00e9n 15, 47011, Valladolid, Spain"
        ], 
        "familyname": "Mart\u00edn-Encinar", 
        "givennames": "Luis"
      }, 
      {
        "email": "d.lanzoni@campus.unimib.it", 
        "affiliations": [
          "L-NESS and Dept. of Materials Science, University of Milano-Bicocca, Via R. Cozzi 55, 20125,  Milano, Italy"
        ], 
        "familyname": "Lanzoni", 
        "givennames": "Daniele"
      }, 
      {
        "affiliations": [
          "L-NESS and Dept. of Materials Science, University of Milano-Bicocca, Via R. Cozzi 55, 20125,  Milano, Italy"
        ], 
        "familyname": "Fantasia", 
        "givennames": "Andrea"
      }, 
      {
        "affiliations": [
          "L-NESS and Dept. of Materials Science, University of Milano-Bicocca, Via R. Cozzi 55, 20125,  Milano, Italy"
        ], 
        "familyname": "Rovaris", 
        "givennames": "Fabrizio"
      }, 
      {
        "email": "roberto.bergamaschini@unimib.it", 
        "affiliations": [
          "L-NESS and Dept. of Materials Science, University of Milano-Bicocca, Via R. Cozzi 55, 20125,  Milano, Italy"
        ], 
        "familyname": "Bergamaschini", 
        "givennames": "Roberto"
      }, 
      {
        "email": "francesco.montalenti@unimib.it", 
        "affiliations": [
          "L-NESS and Dept. of Materials Science, University of Milano-Bicocca, Via R. Cozzi 55, 20125,  Milano, Italy"
        ], 
        "familyname": "Montalenti", 
        "givennames": "Francesco"
      }
    ], 
    "edited_by": 576
  }, 
  "updated": "2024-01-23T10:05:40.167327+00:00"
}