pith. sign in

arxiv: 2606.22152 · v1 · pith:QLKLIM7Xnew · submitted 2026-06-20 · ❄️ cond-mat.mtrl-sci

Bridging Phase-Field Model and Deep Learning for Predicting 2D and 3D Microstructure Evolution in Ternary Alloys

Owais Ahmad , Aravind K , Naveen Kumar , T.A. Abinandanan , Somnath Bhowmick , Rajdip Mukherjee This is my paper

Pith reviewed 2026-06-26 11:39 UTC · model grok-4.3

classification ❄️ cond-mat.mtrl-sci
keywords phase-field modeldeep learningmicrostructure evolutionspinodal decompositionternary alloysConvLSTMnanoporous materialshybrid simulation
0
0 comments X

The pith

A hybrid phase-field and attention-based deep learning model predicts microstructure evolution in ternary alloys hundreds of timesteps ahead.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper establishes a hybrid framework that runs a phase-field model only through the initial rapid phase separation in ternary spinodal dealloying and then hands off to a trained deep learning model for the slower late-stage coarsening. The deep learning component compresses high-resolution images with an autoencoder, learns spatiotemporal patterns with an attention-augmented ConvLSTM, and extends to three dimensions via a slice-by-slice approach. This combination keeps predictive fidelity up to 400 timesteps beyond training data and works for alloy compositions outside the training set. A sympathetic reader would care because full phase-field simulations become prohibitively slow for the long times needed to design bicontinuous nanoporous materials.

Core claim

The central claim is that a hybrid strategy lets the phase-field model capture early-stage spinodal decomposition mechanisms accurately while an attention-enhanced ConvLSTM network, trained on those data, efficiently and faithfully predicts late-stage coarsening dynamics in both two and three dimensions, maintaining accuracy hundreds of timesteps ahead and generalizing to unseen compositions.

What carries the argument

An attention-augmented ConvLSTM network paired with a dimensionality-reducing autoencoder and a slice-by-slice strategy for three-dimensional data, trained directly on phase-field simulation outputs.

If this is right

  • Early rapid evolution is handled by the phase-field model while late-stage dynamics are predicted by the trained network at far lower computational cost.
  • The model generalizes to compositions outside the training distribution.
  • The slice-by-slice approach extends the same architecture from two-dimensional to three-dimensional microstructures.
  • Three distinct early-stage phase-separation mechanisms are captured in the training data.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Such a hybrid workflow could shorten the computational loop for screening alloy compositions aimed at hierarchical nanoporous structures.
  • The same hand-off strategy might apply to other microstructure evolution problems where early transients are fast and late coarsening is slow.
  • Coupling the trained model with experimental tomography data could test whether the learned patterns transfer beyond simulation.

Load-bearing premise

The phase-field model supplies accurate enough early-stage data that the learned deep learning patterns can extrapolate late-stage coarsening without explicit physical constraints.

What would settle it

Direct comparison of the deep learning predictions against independent phase-field runs continued to 400 timesteps for a held-out composition would show whether the reported fidelity holds.

Figures

Figures reproduced from arXiv: 2606.22152 by Aravind K, Naveen Kumar, Owais Ahmad, Rajdip Mukherjee, Somnath Bhowmick, T.A. Abinandanan.

Figure 1
Figure 1. Figure 1: (a) Schematic of a ternary phase diagram for a system with [PITH_FULL_IMAGE:figures/full_fig_p006_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: In particular, the semi-implicit spectral solution of the Cahn–Hilliard equations [PITH_FULL_IMAGE:figures/full_fig_p011_2.png] view at source ↗
Figure 2
Figure 2. Figure 2: Flowchart depicting the algorithm with clearly demarcated serial and parallel steps [PITH_FULL_IMAGE:figures/full_fig_p012_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Time evolution of microstructures (scalar field maps) of alloys (a) P1, (b) A1, and [PITH_FULL_IMAGE:figures/full_fig_p013_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Time evolution of microstructures (scalar field maps) of alloys P1, A1, and A3 [PITH_FULL_IMAGE:figures/full_fig_p014_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Composition and phase map comparison for both 2D and 3D. [PITH_FULL_IMAGE:figures/full_fig_p017_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Workflow of training a machine learning model with phase-field generated mi [PITH_FULL_IMAGE:figures/full_fig_p021_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Systematic analysis of prediction error accumulation across different stages of [PITH_FULL_IMAGE:figures/full_fig_p022_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Microstructure prediction for the composition A2 (0.40-0.30-0.30). (Left) Early [PITH_FULL_IMAGE:figures/full_fig_p024_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Microstructure prediction for the composition A2 ( [PITH_FULL_IMAGE:figures/full_fig_p025_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: Long-term 2D predictions for the composition A2 (0.40-0.30-0.30). Starting with [PITH_FULL_IMAGE:figures/full_fig_p027_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: Long-term 3D predictions for the composition A2 (0.40-0.30-0.30). Starting with [PITH_FULL_IMAGE:figures/full_fig_p028_11.png] view at source ↗
read the original abstract

We develop a hybrid framework that integrates a phase-field model (PFM) with an attention-enhanced deep learning (DL) architecture to study ternary spinodal dealloying, a sophisticated self-organization approach used to fabricate three-dimensional bicontinuous, hierarchical nanoporous materials. The study captures three distinct phase-separation mechanisms that emerge during the early stages of spinodal decomposition in both two and three dimensions. The DL workflow consists of three key components: (i) a dimensionality-reducing autoencoder that provides compact representations of high-resolution microstructure images (256x256x3), (ii) an attention-augmented convolutional long short-term memory (ConvLSTM) network that learns complex spatiotemporal correlations governing microstructure evolution, and (iii) a novel slice-by-slice strategy that enables extension of the model to three-dimensional systems (128x128x128x3). We further demonstrate a hybrid simulation strategy in which PFM accurately captures rapid early-stage microstructure evolution, while the DL model efficiently predicts late-stage coarsening dynamics. The trained DL model achieves remarkable predictive accuracy, maintaining fidelity up to 400 timesteps ahead and generalizing to compositions outside the training distribution. By bridging the physical fidelity of PFM with the computational efficiency of DL, this framework establishes a robust platform for predictive modeling of microstructure evolution in complex multicomponent systems.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

3 major / 2 minor

Summary. The manuscript develops a hybrid phase-field model (PFM) plus deep-learning framework for ternary spinodal dealloying. An autoencoder compresses 256×256×3 (2D) or 128×128×128×3 (3D) microstructure snapshots; an attention-augmented ConvLSTM learns spatiotemporal evolution; a slice-by-slice strategy extends the model to 3D. The hybrid workflow runs PFM for early-stage decomposition and switches to the trained DL model for late-stage coarsening, claiming that the DL component maintains fidelity for 400 timesteps and generalizes to compositions outside the training set.

Significance. If the quantitative claims hold, the work would demonstrate a practical route to accelerate long-horizon microstructure simulations while retaining physical fidelity, which is valuable for designing bicontinuous nanoporous materials. The attention mechanism and hybrid PFM-DL coupling are conceptually attractive; however, the absence of reported error metrics and conservation checks limits the immediate impact.

major comments (3)
  1. [Abstract] Abstract: the central claim that the DL model 'achieves remarkable predictive accuracy, maintaining fidelity up to 400 timesteps ahead' is unsupported by any quantitative error metric (MSE, conservation error, interface-area error), baseline comparison, or error-bar analysis. Without these numbers the 400-step extrapolation cannot be evaluated.
  2. [DL workflow description] DL architecture and training (implied Methods/Results sections): the autoencoder + attention-ConvLSTM pipeline contains no physics-informed loss term, projection step, or post-processing that enforces conservation of the three component fractions. Because ternary spinodal dealloying is mass-conserving, any cumulative drift over 400 steps would invalidate both the fidelity claim and the out-of-distribution generalization result.
  3. [3D extension paragraph] 3D extension (slice-by-slice strategy): treating the 128×128×128 volume as independent 2D slices decouples the third dimension, allowing possible drift in volume fractions and interface topology across slices. No test of global conservation or topological consistency after 400 steps is described, yet this is load-bearing for the 3D generalization claim.
minor comments (2)
  1. [Abstract] The abstract states three distinct phase-separation mechanisms are captured but does not name or illustrate them; a short enumeration or figure reference would improve clarity.
  2. [Methods] Notation for the autoencoder latent dimension and ConvLSTM hidden-state size is not introduced before the results are discussed.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the constructive and detailed comments, which highlight important aspects needed to strengthen the quantitative support for our claims. We address each major comment below and will incorporate the suggested additions in a revised manuscript.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the central claim that the DL model 'achieves remarkable predictive accuracy, maintaining fidelity up to 400 timesteps ahead' is unsupported by any quantitative error metric (MSE, conservation error, interface-area error), baseline comparison, or error-bar analysis. Without these numbers the 400-step extrapolation cannot be evaluated.

    Authors: We agree that the abstract claim requires explicit quantitative backing to be evaluable. In the revised manuscript we will add MSE, per-component conservation errors, interface-area errors, baseline comparisons (e.g., against non-attention ConvLSTM and linear extrapolation), and error bars derived from multiple independent runs to substantiate the 400-timestep fidelity. revision: yes

  2. Referee: [DL workflow description] DL architecture and training (implied Methods/Results sections): the autoencoder + attention-ConvLSTM pipeline contains no physics-informed loss term, projection step, or post-processing that enforces conservation of the three component fractions. Because ternary spinodal dealloying is mass-conserving, any cumulative drift over 400 steps would invalidate both the fidelity claim and the out-of-distribution generalization result.

    Authors: The referee is correct that no explicit conservation mechanism is present. While the training data originate from mass-conserving PFM simulations, we will add a physics-informed conservation penalty to the loss function (or a post-prediction projection onto the valid composition simplex) and report the resulting long-term conservation errors in the revision. revision: yes

  3. Referee: [3D extension paragraph] 3D extension (slice-by-slice strategy): treating the 128×128×128 volume as independent 2D slices decouples the third dimension, allowing possible drift in volume fractions and interface topology across slices. No test of global conservation or topological consistency after 400 steps is described, yet this is load-bearing for the 3D generalization claim.

    Authors: We acknowledge that the slice-by-slice approach does not inherently guarantee 3D consistency. In the revision we will add explicit global 3D conservation checks (total volume fractions across all slices) and topological consistency metrics (e.g., interface connectivity or Euler number) evaluated on the assembled 3D volumes after 400-step predictions. revision: yes

Circularity Check

0 steps flagged

No significant circularity; hybrid PFM-DL workflow is self-contained

full rationale

The paper trains an attention-augmented ConvLSTM (plus autoencoder) on microstructure snapshots generated by an independent phase-field model, then applies the trained network to forecast later timesteps or unseen compositions. This is standard supervised learning on external simulation data; the output predictions are not algebraically identical to the training inputs, nor are any load-bearing steps reduced to self-citations, fitted parameters renamed as predictions, or ansatzes imported from the authors' prior work. The 400-step extrapolation claim rests on empirical validation against held-out PFM runs rather than on any definitional equivalence. No equations or sections exhibit the enumerated circularity patterns.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

No explicit free parameters, axioms, or invented entities are identifiable from the abstract; the work relies on standard phase-field assumptions and standard DL architectures without new postulated entities.

pith-pipeline@v0.9.1-grok · 5799 in / 1134 out tokens · 41488 ms · 2026-06-26T11:39:24.906033+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

173 extracted references · 122 canonical work pages · 2 internal anchors

  1. [1]

    ACS Applied Materials & Interfaces , volume =

    Chen, Xin and Yang, Lin and Zhang, Yuan and Zhang, Donglin and Song, Jiaqi and Feng, Jiaqi and Zhao, Xiuchen and Zhang, Gang and Huo, Yongjun , title =. ACS Applied Materials & Interfaces , volume =. 2025 , doi =

    2025

  2. [2]

    ACS Applied Materials & Interfaces , volume =

    Feng, Longsheng and Huang, Sijia and Heo, Tae Wook and Biener, Juergen , title =. ACS Applied Materials & Interfaces , volume =. 2024 , doi =

    2024

  3. [3]

    Scope of generative artificial intelligence in microstructural studies: a case study , year =

    Ahmad, Owais and Panwar, Vishal and Das, Kaushik and Mukherjee, Rajdip and Bhowmick, Somnath , journal =. Scope of generative artificial intelligence in microstructural studies: a case study , year =. doi:10.1088/1402-4896/ade832 , publisher =

    work page doi:10.1088/1402-4896/ade832

  4. [4]

    Phase field simulation on the cyclic degeneration of one-way shape memory effect of NiTi shape memory alloy single crystal , year =

    Bo Xu and Guozheng Kang and Qianhua Kan and Chao Yu and Xi Xie , journal =. Phase field simulation on the cyclic degeneration of one-way shape memory effect of NiTi shape memory alloy single crystal , year =. doi:10.1016/j.ijmecsci.2019.105303 , keywords =

    work page doi:10.1016/j.ijmecsci.2019.105303 2019

  5. [5]

    Phase field modeling of fracture in fiber reinforced composite laminate , year =

    Peng Zhang and Xiaofei Hu and Tinh Quoc Bui and Weian Yao , journal =. Phase field modeling of fracture in fiber reinforced composite laminate , year =. doi:10.1016/j.ijmecsci.2019.07.007 , keywords =

    work page doi:10.1016/j.ijmecsci.2019.07.007 2019

  6. [6]

    An extension of the phase field method to model interactions between interfacial damage and brittle fracture in elastoplastic composites , year =

    Pengfei Li and Julien Yvonnet and Christelle Combescure , journal =. An extension of the phase field method to model interactions between interfacial damage and brittle fracture in elastoplastic composites , year =. doi:10.1016/j.ijmecsci.2020.105633 , keywords =

    work page doi:10.1016/j.ijmecsci.2020.105633 2020

  7. [7]

    Numerical simulation of three-dimensional multicomponent Cahn–Hilliard systems , year =

    Shiwei Zhou and Yi Min Xie , journal =. Numerical simulation of three-dimensional multicomponent Cahn–Hilliard systems , year =. doi:10.1016/j.ijmecsci.2021.106349 , keywords =

    work page doi:10.1016/j.ijmecsci.2021.106349 2021

  8. [8]

    Deep learning-driven prediction of microstructure evolution via latent space interpolation , year =

    Gaikwad, Sachin and Kasilingam, Thejas and Ahmad, Owais and Mukherjee, Rajdip and Bhowmick, Somnath , journal =. Deep learning-driven prediction of microstructure evolution via latent space interpolation , year =. doi:10.1103/5ngk-4v9j , issue =

    work page doi:10.1103/5ngk-4v9j

  9. [9]

    2024 , volume =

    Ahmad, Owais and Maurya, Rakesh and Mukherjee, Rajdip and Bhowmick, Somnath , title =. 2024 , volume =. doi:10.4028/p-6W4IXl , keywords =

    work page doi:10.4028/p-6w4ixl 2024

  10. [10]

    1994 , issn =

    E.Bruce Nauman and David Qiwei He , journal =. 1994 , issn =. doi:10.1016/0032-3861(94)90757-9 , keywords =

    work page doi:10.1016/0032-3861(94)90757-9 1994

  11. [11]

    2025 , issn =

    Saurabh Tiwari and Prathamesh Satpute and Supriyo Ghosh , journal =. 2025 , issn =. doi:10.1016/j.commatsci.2024.113518 , keywords =

    work page doi:10.1016/j.commatsci.2024.113518 2025

  12. [12]

    2017 , issn =

    Cong Yang and Qingyan Xu and Baicheng Liu , journal =. 2017 , issn =. doi:10.1016/j.commatsci.2017.04.031 , keywords =

    work page doi:10.1016/j.commatsci.2017.04.031 2017

  13. [13]

    2023 , issn =

    Pengzhi Li and Yan Pei and Jianqiang Li , journal =. 2023 , issn =. doi:10.1016/j.asoc.2023.110176 , keywords =

    work page doi:10.1016/j.asoc.2023.110176 2023

  14. [14]

    2002 , doi=

    Yoshihiro Suwa and Yoshiyuki Saito and Kazumi Ochi and Takahiro Aoki and Kanako Goto and Kotaro Abe , journal =. 2002 , doi=

    2002

  15. [15]

    Chen and Jie Shen , journal =

    L.Q. Chen and Jie Shen , journal =. 1998 , issn =. doi:10.1016/S0010-4655(97)00115-X , keywords =

    work page doi:10.1016/s0010-4655(97)00115-x 1998

  16. [16]

    2012 , issn =

    Park, Jin Man and Han, Jun Hee and Mattern, Norbert and Kim, Do Hyang and Eckert, Jurgen , journal =. 2012 , issn =

    2012

  17. [17]

    Sohn and W

    S.W. Sohn and W. Yook and W.T. Kim and D.H. Kim , journal =. 2012 , issn =. doi:10.1016/j.intermet.2011.12.021 , keywords =

    work page doi:10.1016/j.intermet.2011.12.021 2012

  18. [18]

    Chang and W

    H.J. Chang and W. Yook and E.S. Park and J.S. Kyeong and D.H. Kim , journal =. 2010 , issn =. doi:10.1016/j.actamat.2009.12.034 , keywords =

    work page doi:10.1016/j.actamat.2009.12.034 2010

  19. [19]

    , journal =

    Pattanayak, Pulok and Asokan, S. , journal =. 2004 , issn =

    2004

  20. [20]

    2010 , issn =

    Norbert Mattern and Thomas Gemming and J. 2010 , issn =. doi:10.1016/j.jallcom.2009.10.013 , keywords =

    work page doi:10.1016/j.jallcom.2009.10.013 2010

  21. [21]

    Mattern and G

    N. Mattern and G. Goerigk and U. Vainio and M.K. Miller and T. Gemming and J. Eckert , journal =. 2009 , issn =. doi:10.1016/j.actamat.2008.10.028 , keywords =

    work page doi:10.1016/j.actamat.2008.10.028 2009

  22. [22]

    Goerigk and N

    G. Goerigk and N. Mattern , journal =. 2009 , issn =. doi:10.1016/j.actamat.2009.04.028 , keywords =

    work page doi:10.1016/j.actamat.2009.04.028 2009

  23. [23]

    and Thomas, G

    Okada, M. and Thomas, G. and Homma, M. and Kaneko, H. , journal =. 1978 , volume=

    1978

  24. [24]

    Chu, S. N. G. and Nakahara, S. and Strege, K. E. and Johnston, W. D., Jr. , journal =. 1985 , issn =. doi:10.1063/1.335368 , eprint =

    work page doi:10.1063/1.335368 1985

  25. [25]

    and Ralph, Brian , journal =

    Louise Makin, P. and Ralph, Brian , journal =. 1984 , issn =

    1984

  26. [26]

    doi:10.1007/BF02668132 , adsurl =

    The effect of prior deformation on spinodal age hardening in Cu-15 Ni-8 Sn alloy. doi:10.1007/BF02668132 , adsurl =

    work page doi:10.1007/bf02668132

  27. [27]

    and Lele, S

    Singh, J. and Lele, S. and Ranganathan, S. , journal =. 1980 , issn =

    1980

  28. [28]

    A thermodynamic-consistent phase-field model for fracture in temperature-dependent materials , year =

    Yaode Yin and Hongjun Yu , journal =. A thermodynamic-consistent phase-field model for fracture in temperature-dependent materials , year =. doi:10.1016/j.ijmecsci.2025.110382 , keywords =

    work page doi:10.1016/j.ijmecsci.2025.110382 2025

  29. [29]

    and Yang, W

    Chen, L.-Q. and Yang, W. , journal =. Computer simulation of the domain dynamics of a quenched system with a large number of nonconserved order parameters: The grain-growth kinetics , year =. doi:10.1103/physrevb.50.15752 , source =

    work page doi:10.1103/physrevb.50.15752

  30. [31]

    Cahn , journal =

    John W. Cahn , journal =. On spinodal decomposition in cubic crystals , year =. doi:10.1016/0001-6160(62)90114-1 , issue =

    work page doi:10.1016/0001-6160(62)90114-1

  31. [32]

    W. A. Tiller and G. M. Pound and J. P. Hirth , journal =. A critique on the mathematical theory of spinodal decomposition , year =. doi:10.1016/0001-6160(70)90027-1 , issue =

    work page doi:10.1016/0001-6160(70)90027-1

  32. [33]

    J. W. Cahn and J. E. Hilliard , journal =. Spinodal decomposition: A reprise , year =. doi:10.1016/0001-6160(71)90127-1 , issue =

    work page doi:10.1016/0001-6160(71)90127-1

  33. [34]

    Phase-field simulations of ferroelectric/ferroelastic polarization switching , year =

    Jie Wang and San Qiang Shi and Long Qing Chen and Yulan Li and Tong Yi Zhang , journal =. Phase-field simulations of ferroelectric/ferroelastic polarization switching , year =. doi:10.1016/j.actamat.2003.10.011 , issue =

    work page doi:10.1016/j.actamat.2003.10.011 2003

  34. [35]

    and Warren, J.A

    Boettinger, W.J. and Warren, J.A. and Beckermann, C. and Karma, A. , journal =. Phase-field simulation of solidification , year =. doi:10.1146/annurev.matsci.32.101901.155803 , source =

    work page doi:10.1146/annurev.matsci.32.101901.155803

  35. [36]

    Quantitative phase-field modeling of dendritic growth in two and three dimensions , year =

    Karma, Alain and Rappel, Wouter-Jan , journal =. Quantitative phase-field modeling of dendritic growth in two and three dimensions , year =. doi:10.1103/physreve.57.4323 , issue =

    work page doi:10.1103/physreve.57.4323

  36. [37]

    Phase-field formulation for quantitative modeling of alloy solidification , year =

    Alain Karma , journal =. Phase-field formulation for quantitative modeling of alloy solidification , year =. doi:10.1103/physrevlett.87.115701 , issue =

    work page doi:10.1103/physrevlett.87.115701

  37. [38]

    Cogswell and Guanglong Xu and Srdjan Milenkovic and Yuwen Cui , journal =

    Bin Tang and Daniel A. Cogswell and Guanglong Xu and Srdjan Milenkovic and Yuwen Cui , journal =. The formation mechanism of eutectic microstructures in NiAl-Cr composites , year =. doi:10.1039/c6cp01921h , issue =

    work page doi:10.1039/c6cp01921h

  38. [39]

    Chen and J

    L. Chen and J. Chen and R. A. Lebensohn and Y. Z. Ji and T. W. Heo and S. Bhattacharyya and K. Chang and S. Mathaudhu and Z. K. Liu and L. Q. Chen , journal =. An integrated fast Fourier transform-based phase-field and crystal plasticity approach to model recrystallization of three dimensional polycrystals , year =. doi:10.1016/j.cma.2014.12.007 , keywords =

    work page doi:10.1016/j.cma.2014.12.007 2014

  39. [40]

    Danan Fan and S. P. Chen and Long Qing Chen and P. W. Voorhees , journal =. Phase-field simulation of 2-D Ostwald ripening in the high volume fraction regime , year =. doi:10.1016/s1359-6454(01)00393-7 , issue =

    work page doi:10.1016/s1359-6454(01)00393-7

  40. [41]

    Rodney and Y

    D. Rodney and Y. Le Bouar and Alphonse Finel , journal =. Phase field methods and dislocations , year =. doi:10.1016/s1359-6454(01)00379-2 , issue =

    work page doi:10.1016/s1359-6454(01)00379-2

  41. [42]

    Y. M. Jin and A. G. Khachaturyan , journal =. Phase field microelasticity theory of dislocation dynamics in a polycrystal: Model and three-dimensional simulations , year =. doi:10.1080/09500830110062825 , issue =

    work page doi:10.1080/09500830110062825

  42. [43]

    Dynamic instabilities of fracture under biaxial strain using a phase field model , year =

    Hervé Henry and Herbert Levine , journal =. Dynamic instabilities of fracture under biaxial strain using a phase field model , year =. doi:10.1103/physrevlett.93.105504 , issue =

    work page doi:10.1103/physrevlett.93.105504

  43. [44]

    Phase field modeling of crack propagation , year =

    Robert Spatschek and Efim Brener and Alain Karma , journal =. Phase field modeling of crack propagation , year =. doi:10.1080/14786431003773015 , issue =

    work page doi:10.1080/14786431003773015

  44. [45]

    L. O. Eastgate and J. P. Sethna and M. Rauscher and T. Cretegny and C. S. Chen and C. R. Myers , journal =. Fracture in mode I using a conserved phase-field model , year =. doi:10.1103/physreve.65.036117 , issue =

    work page doi:10.1103/physreve.65.036117

  45. [46]

    A phase field model for rate-independent crack propagation: Robust algorithmic implementation based on operator splits , year =

    Christian Miehe and Martina Hofacker and Fabian Welschinger , journal =. A phase field model for rate-independent crack propagation: Robust algorithmic implementation based on operator splits , year =. doi:10.1016/j.cma.2010.04.011 , issue =

    work page doi:10.1016/j.cma.2010.04.011 2010

  46. [47]

    P. H. Leo and J. S. Lowengrub and H. J. Jou , journal =. A diffuse interface model for microstructural evolution in elastically stressed solids , year =. doi:10.1016/s1359-6454(97)00377-7 , issue =

    work page doi:10.1016/s1359-6454(97)00377-7

  47. [48]

    M. P. Gururajan and T. A. Abinandanan , journal =. Phase inversion in two-phase solid systems driven by an elastic modulus mismatch , year =. doi:10.1080/14786430701647984 , issue =

    work page doi:10.1080/14786430701647984

  48. [50]

    Phase-field simulation of phase decomposition in Fe-Cr-Co alloy under an external magnetic field , year =

    Toshiyuki Koyama and Hidehiro Onodera , journal =. Phase-field simulation of phase decomposition in Fe-Cr-Co alloy under an external magnetic field , year =. doi:10.1007/bf03185980 , issue =

    work page doi:10.1007/bf03185980

  49. [51]

    Phase-field modeling of microstructure evolutions in magnetic materials , year =

    Toshiyuki Koyama , journal =. Phase-field modeling of microstructure evolutions in magnetic materials , year =. doi:10.1088/1468-6996/9/1/013006 , issue =

    work page doi:10.1088/1468-6996/9/1/013006

  50. [52]

    and Mukherjee, R

    Mukherjee, A. and Mukherjee, R. and Ankit, K. and Bhattacharya, A. and Nestler, B. , journal =. Influence of substrate interaction and confinement on electric-field-induced transition in symmetric block-copolymer thin films , year =

  51. [53]

    Electric-field-induced lamellar to hexagonally perforated lamellar transition in diblock copolymer thin films: Kinetic pathways , year =

    Arnab Mukherjee and Kumar Ankit and Andreas Reiter and Michael Selzer and Britta Nestler , journal =. Electric-field-induced lamellar to hexagonally perforated lamellar transition in diblock copolymer thin films: Kinetic pathways , year =. doi:10.1039/c6cp04903f , issue =

    work page doi:10.1039/c6cp04903f

  52. [54]

    L. X. Lv and L. Zhen and C. Y. Xu and X. Y. Sun , journal =. Phase field simulation of spinodal decomposition under external magnetic field , year =. doi:10.1016/j.jmmm.2009.12.001 , issue =

    work page doi:10.1016/j.jmmm.2009.12.001 2009

  53. [55]

    An electro-thermo-mechanical coupling phase-field model of defect evolution induced by electromigration in interconnects , year =

    Xin-Wei Wu and Mingyang Chen and Liao-Liang Ke , journal =. An electro-thermo-mechanical coupling phase-field model of defect evolution induced by electromigration in interconnects , year =. doi:10.1016/j.ijmecsci.2024.109792 , keywords =

    work page doi:10.1016/j.ijmecsci.2024.109792 2024

  54. [56]

    1974 , issn =

    R.J Livak and G Thomas , journal =. 1974 , issn =. doi:10.1016/0001-6160(74)90156-4 , url =

    work page doi:10.1016/0001-6160(74)90156-4 1974

  55. [57]

    Butler and G

    E.P. Butler and G. Thomas , journal =. 1970 , issn =. doi:10.1016/0001-6160(70)90150-1 , url =

    work page doi:10.1016/0001-6160(70)90150-1 1970

  56. [58]

    Findik and H

    F. Findik and H. M. Flower , journal =. 1992 , number =. doi:10.1179/mst.1992.8.3.197 , publisher =

    work page doi:10.1179/mst.1992.8.3.197 1992

  57. [59]

    Copetti , journal =

    M.I.M. Copetti , journal =. 2000 , issn =. doi:10.1016/S0378-4754(99)00153-6 , keywords =

    work page doi:10.1016/s0378-4754(99)00153-6 2000

  58. [60]

    2001 , issn =

    Ma, Yu-qiang , journal =. 2001 , issn =. doi:10.1063/1.1343838 , eprint =

    work page doi:10.1063/1.1343838 2001

  59. [61]

    Travasso, Rui D. M. and Buxton, Gavin A. and Kuksenok, Olga and Good, Kevin and Balazs, Anna C. , journal =. 2005 , issn =

    2005

  60. [62]

    and Nauman, E

    Alfarraj, Abdulrahman A. and Nauman, E. Bruce , journal =. 2007 , number =. doi:10.1002/mats.200700025 , keywords =

    work page doi:10.1002/mats.200700025 2007

  61. [63]

    1993 , issn =

    Long-Qing Chen , journal =. 1993 , issn =. doi:10.1016/0956-716X(93)90419-S , url =

    work page doi:10.1016/0956-716x(93)90419-s 1993

  62. [64]

    1994 , issn =

    Long-Qing Chen , journal =. 1994 , issn =. doi:10.1016/0956-7151(94)90482-0 , url =

    work page doi:10.1016/0956-7151(94)90482-0 1994

  63. [65]

    , journal =

    Eyre, David J. , journal =. 1993 , number =

    1993

  64. [66]

    2013 , doi=

    Atsuo Yasumori , journal =. 2013 , doi=

    2013

  65. [67]

    and Nishio, K

    Masuda, H. and Nishio, K. and Baba, N. , journal =. 1994 , issn =

    1994

  66. [68]

    McMillan, P. W. and Matthews, C. E. , journal =. 1976 , issn =

    1976

  67. [69]

    James, P. F. , journal =. 1975 , issn =

    1975

  68. [70]

    2023 , pages =

    Ahmad, Owais and Kumar, Naveen and Mukherjee, Rajdip and Bhowmick, Somnath , journal =. 2023 , pages =. doi:10.1103/PhysRevMaterials.7.083802 , issue =

    work page doi:10.1103/physrevmaterials.7.083802 2023

  69. [71]

    Computation , VOLUME =

    Taye, Mohammad Mustafa , TITLE =. Computation , VOLUME =. 2023 , NUMBER =

    2023

  70. [72]

    , journal =

    Abdi, Herve and Williams, Lynne J. , journal =. 2010 , number =. doi:10.1002/wics.101 , keywords =

    work page doi:10.1002/wics.101 2010

  71. [73]

    1958 , number =

    Cahn, John W and Hilliard, John E , journal =. 1958 , number =

    1958

  72. [74]

    2003 , pages =

    Bhattacharyya, Saswata and Abinandanan, TA , journal =. 2003 , pages =. doi:10.1007/bf02712812 , publisher =

    work page doi:10.1007/bf02712812 2003

  73. [75]

    1984 , number =

    Kramer, Edward J and Green, Peter and Palmstr. 1984 , number =. doi:10.1016/0032-3861(84)90205-2 , publisher =

    work page doi:10.1016/0032-3861(84)90205-2 1984

  74. [76]

    2017 , number =

    Ghosh, Supriyo and Mukherjee, Arnab and Abinandanan, TA and Bose, Suryasarathi , journal =. 2017 , number =. doi:10.1039/c7cp01816a , publisher =

    work page doi:10.1039/c7cp01816a 2017

  75. [77]

    2012 , publisher=

    Lee, Hyun Geun and Choi, Jeong-Whan and Kim, Junseok , journal =. 2012 , publisher=

    2012

  76. [78]

    1961 , number =

    Cahn, John W , journal =. 1961 , number =. doi:10.1016/0001-6160(61)90182-1 , publisher =

    work page doi:10.1016/0001-6160(61)90182-1 1961

  77. [79]

    1993 , number =

    Pfleiderer, S and Ball, DGA and Bailey, RC , journal =. 1993 , number =. doi:10.1016/0098-3004(93)90053-8 , publisher =

    work page doi:10.1016/0098-3004(93)90053-8 1993

  78. [80]

    Catherine and Apley, Daniel W

    Bostanabad, Ramin and Zhang, Yichi and Li, Xiaolin and Kearney, Tucker and Brinson, L. Catherine and Apley, Daniel W. and Liu, Wing Kam and Chen, Wei , journal =. 2018 , issn =. doi:10.1016/J.PMATSCI.2018.01.005 , keywords =

    work page doi:10.1016/j.pmatsci.2018.01.005 2018

  79. [81]

    SHI, Xingjian and Chen, Zhourong and Wang, Hao and Yeung, Dit-Yan and Wong, Wai-kin and WOO, Wang-chun , booktitle =

  80. [82]

    2022 , doi =

    De, Subhayan and Reynolds, Matthew and Hassanaly, Malik and King, Ryan N and Doostan, Alireza , journal =. 2022 , doi =

    2022

Showing first 80 references.