### Abstract

Original language | English |
---|---|

Pages (from-to) | 777-805 |

Journal | Rendiconti Lincei / Matematica e Applicazioni |

Volume | 28 |

Issue number | 4 |

DOIs | |

Publication status | Published - 2017 |

### Fingerprint

### Keywords

### Fields of Expertise

- Information, Communication & Computing

### Cite this

**Partial Differential Equations — Towards a gradient flow for microstructure.** / Eggeling, Eva.

Research output: Contribution to journal › Article › Research › peer-review

*Rendiconti Lincei / Matematica e Applicazioni*, vol. 28, no. 4, pp. 777-805. https://doi.org/10.4171/RLM/785

}

TY - JOUR

T1 - Partial Differential Equations — Towards a gradient flow for microstructure

AU - Eggeling, Eva

PY - 2017

Y1 - 2017

N2 - A central problem of microstructure is to develop technologies capable of producing an arrangement, or ordering, of a polycrystalline material, in terms of mesoscopic parameters, like geometry and crystallography, appropriate for a given application. Is there such an order in the first place? Our goal is to describe the emergence of the grain boundary character distribution (GBCD), a statistic that details texture evolution discovered recently, and to illustrate why it should be considered a material property. For the GBCD statistic, we have developed a theory that relies on mass transport and entropy. The focus of this paper is its identification as a gradient flow in the sense of De Giorgi, as illustrated by Ambrosio, Gigli, and Savaré. In this way, the empirical texture statistic is revealed as a solution of a Fokker–Planck type equation whose evolution is determined by weak topology kinetics and whose limit behavior is a Boltzmann distribution. The identification as a gradient flow by our method is tantamount to exhibiting the harvested statistic as the iterates in a JKO implicit scheme. This requires several new ideas. The development exposes the question of how to understand the circumstances under which a harvested empirical statistic is a property of the underlying process.

AB - A central problem of microstructure is to develop technologies capable of producing an arrangement, or ordering, of a polycrystalline material, in terms of mesoscopic parameters, like geometry and crystallography, appropriate for a given application. Is there such an order in the first place? Our goal is to describe the emergence of the grain boundary character distribution (GBCD), a statistic that details texture evolution discovered recently, and to illustrate why it should be considered a material property. For the GBCD statistic, we have developed a theory that relies on mass transport and entropy. The focus of this paper is its identification as a gradient flow in the sense of De Giorgi, as illustrated by Ambrosio, Gigli, and Savaré. In this way, the empirical texture statistic is revealed as a solution of a Fokker–Planck type equation whose evolution is determined by weak topology kinetics and whose limit behavior is a Boltzmann distribution. The identification as a gradient flow by our method is tantamount to exhibiting the harvested statistic as the iterates in a JKO implicit scheme. This requires several new ideas. The development exposes the question of how to understand the circumstances under which a harvested empirical statistic is a property of the underlying process.

KW - Coarsening

KW - texture development

KW - arge metastable networks,

KW - critical event model

KW - entropy-based theory

KW - free energy

KW - okker–Planck equation

KW - Kantorovich–Rubinstein–Wasserstein metric

U2 - 10.4171/RLM/785

DO - 10.4171/RLM/785

M3 - Article

VL - 28

SP - 777

EP - 805

JO - Rendiconti Lincei / Matematica e Applicazioni

JF - Rendiconti Lincei / Matematica e Applicazioni

SN - 1120-6330

IS - 4

ER -