pith. sign in

arxiv: 2606.03309 · v1 · pith:ZO7LZ4LSnew · submitted 2026-06-02 · ❄️ cond-mat.mtrl-sci

Unconventional crystallization pathway bypassing the intermediate cubic phase in phase-change superlattices

Bai-Qian Wang , Nian-Ke Chen , Yao-Jie Wang , Jia Sun , Yu-Ting Huang , Ming Xu , Shengbai Zhang , Xian-Bin Li This is my paper

Pith reviewed 2026-06-28 09:32 UTC · model grok-4.3

classification ❄️ cond-mat.mtrl-sci
keywords phase-change materialsGST superlatticecrystallization pathwaymolecular dynamicsdefective superlatticeanti-site defectsstacking faultsphase-change memory
0
0 comments X

The pith

Residual crystalline regions nucleate direct recrystallization of GST into a defective superlattice, bypassing the cubic phase.

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

The paper uses large-scale molecular dynamics simulations to examine how partially amorphized Ge-Sb-Te superlattices recrystallize. It shows that leftover crystalline patches act as templates that let the amorphous material reform straight into the superlattice structure rather than first forming the usual cubic phase. The resulting superlattice contains anti-site defects and stacking faults, making those regions easier to melt again. This pathway supplies a concrete atomic-scale picture for why superlattice phase-change memory can operate at lower power than conventional devices.

Core claim

Starting from a partially melted GST superlattice, residual crystalline regions serve as nuclei that drive the amorphous material to recrystallize directly into the superlattice phase without an intermediate cubic phase; the product is a structurally ordered but chemically disordered defective superlattice containing anti-site defects and stacking faults, and these defective regions melt more readily than defect-free superlattice.

What carries the argument

Residual crystalline regions acting as nuclei that template direct reformation of the superlattice order.

If this is right

  • Recrystallization avoids the cubic phase entirely when residual crystals remain.
  • The recrystallized material is a defective superlattice with anti-site defects and stacking faults rather than an ideal structure.
  • Defective superlattice regions melt at lower temperature and therefore function as the active switching volume in a device.
  • The same residual-crystal mechanism can explain repeated low-power cycling without full amorphization.

Where Pith is reading between the lines

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

  • Device models could treat the defective regions as the primary switching sites rather than assuming uniform superlattice behavior.
  • Engineering initial defect density might allow deliberate control over the active melting volume.
  • The same nucleation picture may apply to other layered phase-change compounds that retain partial crystallinity after reset.

Load-bearing premise

The machine-learning interatomic potential correctly ranks the energies, transition barriers, and defect stabilities that govern the GST superlattice trajectories.

What would settle it

Molecular-dynamics trajectories or in-situ diffraction experiments that show the cubic phase appearing as an obligatory intermediate before the superlattice re-forms.

read the original abstract

The Ge-Sb-Te (GST) superlattice phase-change material is a promising candidate for overcoming the high power-consumption of phase-change memory (PCM). However, the working mechanism of the superlattice PCM remains controversial. Partial amorphization, which is currently considered the most plausible mechanism, remains hotly debated: how does the partially amorphized GST recrystallize into its superlattice phase instead of the conventionally expected cubic phase? Here, we address this issue using large-scale molecular dynamics simulations enabled by a machine-learning interatomic potential. Starting from a partially melted GST superlattice, we demonstrate that the residual crystalline regions serve as nuclei, enabling the amorphous GST to recrystallize directly into the superlattice phase without passing through the intermediate cubic phase. Moreover, the recrystallized phase is not an ideal superlattice, but rather a structurally ordered and chemically disordered defective superlattice characterized by anti-site defects and stacking faults. The defective superlattice region is also more susceptible to melting than the defect-free superlattice, and thereby can act as the active region of the PCM device. These results help to clarify the longstanding debates concerning the mechanism of superlattice-based PCM.

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

2 major / 2 minor

Summary. The manuscript uses large-scale molecular dynamics simulations driven by a machine-learning interatomic potential to examine recrystallization in partially amorphized Ge-Sb-Te (GST) superlattices. It claims that residual crystalline regions nucleate direct recrystallization into a defective superlattice (structurally ordered but chemically disordered, with anti-site defects and stacking faults) without an intermediate cubic phase, and that the defective regions melt more readily and therefore constitute the active switching volume in superlattice PCM devices.

Significance. If the underlying potential is shown to reproduce the relevant phase stabilities, barriers, and defect energetics, the work would supply a concrete atomistic mechanism that resolves the long-standing debate on whether partial amorphization in GST superlattices recrystallizes via the cubic phase or directly into the superlattice, while also identifying a structural origin for the low-power operation of these devices. The scale of the simulations is a computational asset, but the absence of any reported validation of the potential against DFT or experiment for the quantities that control the reported pathways substantially limits the strength of the conclusions.

major comments (2)
  1. [Methods] Methods section (description of the machine-learning interatomic potential): No benchmarking is reported against DFT or experimental values for (i) the relative energy of the cubic versus superlattice phases, (ii) the energy barriers separating the direct recrystallization path from the two-step (cubic-intermediate) path, or (iii) the formation energies of the anti-site defects and stacking faults invoked in the central claim. Because every trajectory and mechanistic conclusion rests exclusively on dynamics generated by this potential, the lack of such validation makes it impossible to determine whether the observed nucleation mechanism and defect susceptibility are physical or artifacts of the potential.
  2. [Results] Results (recrystallization trajectories): The manuscript provides no information on system size, number of independent trajectories, or statistical sampling of the nucleation events. Without these details it is impossible to assess whether the reported preference for direct superlattice recrystallization is robust or a consequence of limited sampling or finite-size effects.
minor comments (2)
  1. [Abstract] The abstract states that the recrystallized phase is "structurally ordered and chemically disordered," but the precise order parameters or metrics used to quantify this distinction are not defined until later in the text; a brief definition in the abstract would improve readability.
  2. [Figures] Figure captions should explicitly state the simulation temperature, timestep, and potential cutoff used for each panel so that the trajectories can be reproduced without cross-referencing the Methods text.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive and detailed report. The comments identify important omissions that limit the strength of the conclusions. We address each major comment below and will revise the manuscript to incorporate the requested information and validations.

read point-by-point responses

  1. Referee: [Methods] Methods section (description of the machine-learning interatomic potential): No benchmarking is reported against DFT or experimental values for (i) the relative energy of the cubic versus superlattice phases, (ii) the energy barriers separating the direct recrystallization path from the two-step (cubic-intermediate) path, or (iii) the formation energies of the anti-site defects and stacking faults invoked in the central claim. Because every trajectory and mechanistic conclusion rests exclusively on dynamics generated by this potential, the lack of such validation makes it impossible to determine whether the observed nucleation mechanism and defect susceptibility are physical or artifacts of the potential.

    Authors: We agree that the manuscript does not report these specific benchmarks and that they are necessary to substantiate the mechanistic claims. The potential was trained on extensive DFT data for GST, but targeted validation for phase energy differences, transition barriers, and defect formation energies was not included. We will add these benchmarks to the revised Methods section, performing additional DFT calculations (including relative phase energies, nudged-elastic-band estimates of barriers between direct and two-step paths, and defect formation energies) and reporting the comparisons explicitly. revision: yes

  2. Referee: [Results] Results (recrystallization trajectories): The manuscript provides no information on system size, number of independent trajectories, or statistical sampling of the nucleation events. Without these details it is impossible to assess whether the reported preference for direct superlattice recrystallization is robust or a consequence of limited sampling or finite-size effects.

    Authors: We acknowledge the omission. The original submission does not specify system sizes or the number of independent runs. We will revise the manuscript to include a clear description of the supercell sizes employed, the number of independent trajectories initiated from different partially amorphized configurations, and a brief discussion of how the observed nucleation preference holds across these runs, thereby addressing concerns about sampling and finite-size effects. revision: yes

Circularity Check

0 steps flagged

No circularity: results emerge from independent MD trajectories

full rationale

The paper's central claims (residual crystals nucleate direct superlattice recrystallization without cubic intermediate; defective superlattice with anti-site defects and stacking faults) are obtained from large-scale molecular dynamics trajectories. These trajectories are generated by an ML interatomic potential whose outputs are not algebraically forced to match any target result inside the paper; the simulation dynamics are independent of the specific nucleation or defect observations reported. No self-definitional equations, fitted inputs renamed as predictions, or load-bearing self-citations appear in the derivation chain. The result is a computational experiment whose validity rests on the potential's fidelity (a separate benchmarking question) rather than any internal reduction to its own inputs.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

The central claim depends on the machine-learning interatomic potential being an accurate surrogate for the true atomic interactions; this potential is a fitted model whose parameters are not described in the abstract.

free parameters (1)
  • machine-learning interatomic potential parameters
    The potential is trained on reference data; its parameters are therefore fitted quantities that control all simulated energies and forces.
axioms (1)
  • domain assumption The machine-learning interatomic potential accurately models the relative stability and kinetics of ideal and defective GST superlattices
    Invoked as the basis for interpreting all molecular-dynamics trajectories as physically meaningful.

pith-pipeline@v0.9.1-grok · 5762 in / 1343 out tokens · 36742 ms · 2026-06-28T09:32:44.005053+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

59 extracted references

  1. [1]

    Figure 1a shows the structure of the constructed model

    Demonstration of the recrystallization of superlattice To simulate the recrystallization of the partially amorphized superlattice phase-change materials, a partially amorphized model containing Ge2Sb2Te5, Ge 1Sb2Te4 and Sb 2Te3 sublayers, is firstly constructed by the melt-quench MD simulations, where part of the crystal is fixed to guarantee the coexiste...

  2. [2]

    Figure 3 illustrates atomic rearrangements at the vdW interface during crystallization, with highlighted atoms helping to visualize the process

    Atomic mechanism of the recrystallization During the crystallization process, the migration of the crystal -amorphous interface and the formation of vdW interfaces are crucial for the formation of the superlattice structure. Figure 3 illustrates atomic rearrangements at the vdW interface during crystallization, with highlighted atoms helping to visualize ...

  3. [3]

    The identification and impact of the defective superlattice. To further analyze the structural characteristics of the re crystallized phase , we use the smooth overlap of atomic positions (SOAP) based kernel similarity function to evaluate the similarity between the recrystallized phase and the perfect superlattice phase (see method for more details ). Fi...

    2023

  4. [4]

    Wang et al., Amorphous phase-change memory alloy with no resistance drift

    X. Wang et al., Amorphous phase-change memory alloy with no resistance drift. Nature Materials 25, 456-462 (2026)

    2026

  5. [5]

    M. J. Rasch et al. , Hardware -aware training for large -scale and diverse deep learning inference workloads using in-memory computing-based accelerators. Nature Communications 14, 5282 (2023)

    2023

  6. [6]

    Ambrogio et al., An analog-AI chip for energy-efficient speech recognition and transcription

    S. Ambrogio et al., An analog-AI chip for energy-efficient speech recognition and transcription. Nature 620, 768-775 (2023)

    2023

  7. [7]

    Ortner et al

    T. Ortner et al. , Rapid learning with phase -change memory -based in -memory computing through learning-to-learn. Nature Communications 16, 1243 (2025)

    2025

  8. [8]

    Le Gallo et al., A 64-core mixed-signal in-memory compute chip based on phase-change memory for deep neural network inference

    M. Le Gallo et al., A 64-core mixed-signal in-memory compute chip based on phase-change memory for deep neural network inference. Nature Electronics 6, 680-693 (2023)

    2023

  9. [9]

    Meng et al., Electrical programmable multilevel nonvolatile photonic random-access memory

    J. Meng et al., Electrical programmable multilevel nonvolatile photonic random-access memory. Light: Science & Applications 12, 189 (2023)

    2023

  10. [10]

    Zhang et al., Metavalent Bonding in Layered Phase-Change Memory Materials

    W. Zhang et al., Metavalent Bonding in Layered Phase-Change Memory Materials. Advanced Science 10, 2300901 (2023)

    2023

  11. [11]

    Zheng et al., Extended switching endurance of phase change memory through nano -confined cell structure

    J. Zheng et al., Extended switching endurance of phase change memory through nano -confined cell structure. Nature Communications 16, 5788 (2025)

    2025

  12. [12]

    Park et al., Phase-change memory via a phase -changeable self-confined nano-filament

    S.-O. Park et al., Phase-change memory via a phase -changeable self-confined nano-filament. Nature 628, 293-298 (2024)

    2024

  13. [13]

    A. I. Khan et al., Ultralow-switching current density multilevel phase -change memory on a flexible substrate. Science 373, 1243-1247 (2021)

    2021

  14. [14]

    Ding et al., Phase-change heterostructure enables ultralow noise and drift for memory operation

    K. Ding et al., Phase-change heterostructure enables ultralow noise and drift for memory operation. Science 366, 210-215 (2019)

    2019

  15. [15]

    Ahn et al

    C. Ahn et al. , Energy- efficient phase- change memory with graphene as a thermal barrier. Nano Letters 15, 6809-6814 (2015)

    2015

  16. [16]

    Acharya, O

    D. Acharya, O. Abou El Kheir, D. Campi, M. Bernasconi, Crystallization kinetics of nanoconfined GeTe slabs in GeTe/TiTe 2-like superlattices for phase change memories. Scientific Reports 14, 3224 (2024)

    2024

  17. [17]

    R. E. Simpson et al., Interfacial phase-change memory. Nature Nanotechnology 6, 501-505 (2011)

    2011

  18. [18]

    Kwon et al., Uncovering thermal and electrical properties of Sb2Te3/GeTe superlattice films

    H. Kwon et al., Uncovering thermal and electrical properties of Sb2Te3/GeTe superlattice films. Nano Letters 21, 5984-5990 (2021)

    2021

  19. [19]

    Térébénec, F

    D. Térébénec, F. Hippert, N. Bernier, N. Castellani, P. Noé, GeTe/Sb 2Te3 Super-Lattices: Impact of Atomic Structure on the RESET Current of Phase -Change Memory Devices. Advanced Electronic 12 Materials 11, 2400290 (2025)

    2025

  20. [20]

    R. -J. Zhu et al. , High -Stability van Der Waals Structures of GeTe/Sb 2Te3 Superlattices for 100× Increased Durability Phase -Change Memory (PCM) by Low -Temperature Atomic Layer Deposition. Advanced Functional Materials 34, 2408897 (2024)

    2024

  21. [21]

    Zhao et al

    J. Zhao et al. , Probing the melting transitions in phase -change superlattices via thin film nanocalorimetry. Nano Letters 23, 4587-4594 (2023)

    2023

  22. [22]

    Li, N.-K

    X.-B. Li, N.-K. Chen, X.-P . Wang, H.-B. Sun, Phase-change superlattice materials toward low power consumption and high density data storage: Microscopic picture, working principles, and optimization. Advanced Functional Materials 28, 1803380 (2018)

    2018

  23. [23]

    Woo Lee et al., Phase change in GeTe/Sb 2Te3 superlattices: Formation of the vacancy -ordered metastable cubic structure via Ge migration

    C. Woo Lee et al., Phase change in GeTe/Sb 2Te3 superlattices: Formation of the vacancy -ordered metastable cubic structure via Ge migration. Applied Surface Science 602, 154274 (2022)

    2022

  24. [24]

    Yoo et al., Atomic Layer Deposition of Sb 2Te3/GeTe Superlattice Film and Its Melt- Quenching- Free Phase-Transition Mechanism for Phase-Change Memory

    C. Yoo et al., Atomic Layer Deposition of Sb 2Te3/GeTe Superlattice Film and Its Melt- Quenching- Free Phase-Transition Mechanism for Phase-Change Memory. Advanced Materials 34, 2207143 (2022)

    2022

  25. [25]

    Kim et al

    D. Kim et al. , Conversion between Metavalent and Covalent Bond in Metastable Superlattices Composed of 2D and 3D Sublayers. ACS Nano 16, 20758-20769 (2022)

    2022

  26. [26]

    Lim et al., Advanced interfacial phase change material: Structurally confined and interfacially extended superlattice

    H. Lim et al., Advanced interfacial phase change material: Structurally confined and interfacially extended superlattice. Materials Today 68, 62-73 (2023)

    2023

  27. [27]

    Asanuma et al

    S. Asanuma et al. , Room temperature -produced chalcogenide superlattices for interfacial phase- change memory. Applied Physics Letters 125, 043504 (2024)

    2024

  28. [28]

    Sklénard et al

    B. Sklénard et al. , Electronic and thermal properties of GeTe/Sb 2Te3 superlattices by ab initio approach: Impact of Van der Waals gaps on vertical lattice thermal conductivity. Applied Physics Letters 119, 201911 (2021)

    2021

  29. [29]

    N. K. Chen et al., Metal–insulator transition of Ge–Sb–Te superlattice: An electron counting model study. IEEE Transactions on Nanotechnology 17, 140-146 (2018)

    2018

  30. [30]

    Kalikka et al., Strain-engineered diffusive atomic switching in two- dimensional crystals

    J. Kalikka et al., Strain-engineered diffusive atomic switching in two- dimensional crystals. Nature Communications 7, 11983 (2016)

    2016

  31. [31]

    Wu et al., Novel nanocomposite-superlattices for low energy and high stability nanoscale phase - change memory

    X. Wu et al., Novel nanocomposite-superlattices for low energy and high stability nanoscale phase - change memory. Nature Communications 15, 13 (2024)

    2024

  32. [32]

    Wang et al., Partial melting nature of phase-change memory Ge-Sb-Te superlattice uncovered by large-scale machine learning interatomic potential molecular dynamics

    B.-Q. Wang et al., Partial melting nature of phase-change memory Ge-Sb-Te superlattice uncovered by large-scale machine learning interatomic potential molecular dynamics. Acta Materialia 276, 120123 (2024)

    2024

  33. [33]

    A. I. Khan et al. , Electro -Thermal Confinement Enables Improved Superlattice Phase Change Memory. IEEE Electron Device Letters 43, 204-207 (2022)

    2022

  34. [34]

    Boniardi et al., Evidence for Thermal-Based Transition in Super-Lattice Phase Change Memory

    M. Boniardi et al., Evidence for Thermal-Based Transition in Super-Lattice Phase Change Memory. Physica Status Solidi-Rapid Research Letters 13, 1800634 (2019)

    2019

  35. [35]

    Térébénec et al., Improvement of Phase-Change Memory Performance by Means of GeTe/Sb2Te3 Superlattices

    D. Térébénec et al., Improvement of Phase-Change Memory Performance by Means of GeTe/Sb2Te3 Superlattices. physica status solidi (RRL) – Rapid Research Letters 15, 2000538 (2021)

    2021

  36. [36]

    Abou El Kheir, L

    O. Abou El Kheir, L. Bonati, M. Parrinello, M. Bernasconi, Unraveling the crystallization kinetics of the Ge2Sb2Te5 phase change compound with a machine-learned interatomic potential. npj Computational Materials 10, 33 (2024)

    2024

  37. [37]

    Y.-J. Choi, M. Ghim, S.- H. Jhi, Study of vacancy ordering and the boson peak in metastable cubic Ge-Sb-Te using machine learning potentials. Physical Review Materials 8, 013606 (2024)

    2024

  38. [38]

    Wang et al., Machine learning interatomic potential: Bridge the gap between small -scale models and realistic device-scale simulations

    G. Wang et al., Machine learning interatomic potential: Bridge the gap between small -scale models and realistic device-scale simulations. iScience 27, 109673 (2024). 13

    2024

  39. [39]

    Zhang, K

    X. Zhang, K. Li, J. Zhou, Z. Sun, Chalcogenide Phase -Change Materials: Unveiling New Horizons with Big Data and Machine Learning. Journal of Materials Chemistry C 13, 14168-14186 (2025)

    2025

  40. [40]

    Xu et al., Machine learning for discrimination of phase‐change chalcogenide glasses

    Q. Xu et al., Machine learning for discrimination of phase‐change chalcogenide glasses. InfoMat 7, e70006 (2025)

    2025

  41. [41]

    Abou El Kheir, M

    O. Abou El Kheir, M. Bernasconi, Million -Atom Simulation of the Set Process in Phase Change Memories at the Real Device Scale. Advanced Electronic Materials 11, e2500110 (2025)

    2025

  42. [42]

    Acharya, O

    D. Acharya, O. Abou El Kheir, S. Marcorini, M. Bernasconi, Simulation of the crystallization process of Ge2Sb2Te5 nanoconfined in superlattice geometries for phase change memories. Nanoscale 17, 13828- 13841 (2025)

    2025

  43. [43]

    K. Wan, J. He, X. Shi, Construction of High Accuracy Machine Learning Interatomic Potential for Surface/Interface of Nanomaterials—A Review. Advanced Materials 36, 2305758 (2024)

    2024

  44. [44]

    Yu et al., High-Accuracy Machine-Learned Interatomic Potentials for the Phase Change Material Ge3Sb6Te5

    W. Yu et al., High-Accuracy Machine-Learned Interatomic Potentials for the Phase Change Material Ge3Sb6Te5. Chemistry of Materials 35, 6651-6658 (2023)

    2023

  45. [45]

    Y . Zhou, D. F. Thomas du Toit, S. R. Elliott, W. Zhang, V . L. Deringer, Full -cycle device- scale simulations of memory materials with a tailored atomic -cluster-expansion potential. Nature Communications 16, 8688 (2025)

    2025

  46. [46]

    Xu et al., Enhancing elemental phase -change chalcogenide glass through tailored alloying

    Q. Xu et al., Enhancing elemental phase -change chalcogenide glass through tailored alloying. Acta Materialia 289, 120896 (2025)

    2025

  47. [47]

    Zhang et al., Atomistic Understanding of 2D Monatomic Phase-Change Material for Non-V olatile Optical Applications

    H. Zhang et al., Atomistic Understanding of 2D Monatomic Phase-Change Material for Non-V olatile Optical Applications. Advanced Science 13, e13157 (2026)

    2026

  48. [48]

    Y . T. Liu et al., High‐throughput screening for phase‐change memory materials. Advanced Functional Materials 31, 2009803 (2021)

    2021

  49. [49]

    Z h a o et al

    T.-Y. Z h a o et al. , Heterogeneous Coherent Interface Enabling Nonstochastic Crystallization for Phase-Change Memory. Advanced Functional Materials 36, e23223 (2026)

    2026

  50. [50]

    J. -J. Wang et al. , In -Plane Twinning Defects in Hexagonal GeSb 2Te4. Advanced Materials Technologies 7, 2200214 (2022)

    2022

  51. [51]

    Wang et al., Genesis and Effects of Swapping Bilayers in Hexagonal GeSb 2Te4

    J.-J. Wang et al., Genesis and Effects of Swapping Bilayers in Hexagonal GeSb 2Te4. Chemistry of Materials 30, 4770-4777 (2018)

    2018

  52. [52]

    Cojocaru -Mirédin et al., Thermally Assisted Atomic-Scale Intermixing and Ordering in GeTe – Sb2Te3 Superlattices

    O. Cojocaru -Mirédin et al., Thermally Assisted Atomic-Scale Intermixing and Ordering in GeTe – Sb2Te3 Superlattices. ACS Nano 19, 6130-6141 (2025)

    2025

  53. [53]

    Lotnyk, I

    A. Lotnyk, I. Hilmi, U. Ross, B. Rauschenbach, Van der Waals interfacial bonding and intermixing in GeTe-Sb2Te3-based superlattices. Nano Res 11, 1676-1686 (2018)

    2018

  54. [54]

    A. P. Thompson et al., LAMMPS - A flexible simulation tool for particle -based materials modeling at the atomic, meso, and continuum scales. Computer Physics Communications 271, 108171 (2022)

    2022

  55. [55]

    A. P. Bartok, M. C. Payne, R. Kondor, G. Csanyi, Gaussian approximation potentials: The accuracy of quantum mechanics, without the electrons. Physical Review Letters 104, 136403 (2010)

    2010

  56. [56]

    Y . Zhou, W. Zhang, E. Ma, V . L. Deringer, Device -scale atomistic modelling of phase -change memory materials. Nature Electronics 6, 746-754 (2023)

    2023

  57. [57]

    A. P. Bartók, R. Kondor, G. Csányi, On representing chemical environments. Physical Review B 87, 184115 (2013)

    2013

  58. [58]

    Xu et al

    Y . Xu et al. , Unraveling crystallization mechanisms and electronic structure of phase- change materials by large-scale ab initio simulations. Advanced Materials 34, e2109139 (2022)

    2022

  59. [59]

    S. P. Ong et al., Python Materials Genomics (pymatgen): A robust, open- source python library for materials analysis. Computational Materials Science 68, 314-319 (2013)

    2013