Fast Isotropic Li-Ion Diffusion in Zeolitic Imidazolate Framework Glass Electrolytes for Batteries

Kasper Tolborg; Morten M. Smedskjaer; Tao Du; Timoth\'ee Jamin; Yong Li; Yuanzheng Yue; Zhencai Li

REVIEW 3 major objections 5 minor 7 references

Fast Isotropic Li-Ion Diffusion in Zeolitic Imidazolate Framework Glass Electrolytes for Batteries

T0 review · 3 major / 5 minor · reviewed 2026-08-10 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Structural disorder from vitrification lowers the lithium migration barrier in ZIF glasses to 0.16 eV and makes diffusion isotropic.

desk verdict A plausible MLIP-MD study claiming vitrification lowers Li+ barriers and isotropizes transport in ZIF glasses, but the key numbers rest on unvalidated amorphous-phase potential transferability and Arrhenius extrapolation. read the letter →

arxiv 2608.06902 v1 pith:ZQMTAHZM submitted 2026-08-07 cond-mat.mtrl-sci

Yong Li , Tao Du , Timothée Jamin , Zhencai Li , Kasper Tolborg , Yuanzheng Yue , Morten M. Smedskjaer This is my paper · ORCID

classification cond-mat.mtrl-sci

keywords zeoliticimidazolateframeworkglasssolid-statebatteryelectrolytelithium-iondiffusionmachine-learninginteratomicpotentialactivationenergyisotropiciontransportmelt-quenchedmetal-organic

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper claims that turning a crystalline metal-organic framework into a glass is not a sacrifice but a gain for lithium-ion conduction. Simulating Li+ transport in crystalline and glassy ZIF-4 and ZIF-62 with a machine-learned interatomic potential, the authors find that vitrification lowers the migration activation energy from about 0.35 eV to 0.16 eV, raises the extrapolated room-temperature diffusion coefficient by more than an order of magnitude in ZIF-4 and nearly sevenfold in ZIF-62, and replaces rare anisotropic cage-to-cage hopping with continuous, Fickian-like isotropic diffusion. The structural cause is the randomization of imidazolate and benzimidazolate ring orientations, which removes direction-specific energy barriers. If correct, this makes ZIF glasses a concrete platform for solid-state battery electrolytes that need fast, direction-independent transport without grain boundaries.

What carries the argument

The load-bearing machinery is the structural disorder itself, produced by melt-quenching, together with a fine-tuned deep-learning interatomic potential that lets the authors run accurate nanosecond-scale molecular dynamics with only ten Li+ ions per supercell, sampled over 20 to 40 replicas for statistical convergence. The paper links disorder to transport through two quantitative tools: ring-orientation pole figures, which map the stereographic projections of imidazolate and benzimidazolate ring normals and show the loss of orientational texture on vitrification; and the diffusion tensor, whose diagonal components $D_{xx}$, $D_{yy}$, and $D_{zz}$ are drawn as projected ellipsoids and show the collapse of anisotropy in the glass. Van Hove correlation functions and the non-Gaussian parameter supply the mechanistic contrast between discrete hopping and continuous diffusion.

What would settle it

Measure the temperature-dependent Li+ conductivity of melt-quenched, Li-doped ZIF-4 glass by impedance spectroscopy or 7Li NMR; an activation energy near 0.35 eV instead of 0.16 eV would overturn the central claim. A complementary check is a DFT-based nudged-elastic-band scan of Li+ migration barriers in the glassy structure; if the barriers are not broadly distributed around 0.16 eV, the disorder mechanism is not supported.

Watch

Extended reading notes

Core claim

Using a machine-learned interatomic potential fine-tuned on lithium-containing configurations, the authors simulate Li+ transport in crystalline and glassy ZIF-4 and ZIF-62. They find that the glassy frameworks show a Li+ migration activation energy of 0.16 eV, down from about 0.36 eV in ZIF-4 crystals and 0.34 eV in ZIF-62 crystals, and that the extrapolated room-temperature diffusion coefficient rises from 1.91e-8 to 3.21e-7 cm$^2$/s for ZIF-4 and from 1.12e-8 to 7.76e-8 cm$^2$/s for ZIF-62. The mechanistic picture is that crystalline ZIFs confine Li+ in cages and permit only rare, dynamically heterogeneous hops, whereas the glass presents a broad distribution of coordination geometries and barriers that supports continuous, Fickian-like movement. Directional analysis shows crystal diffusion is strongly anisotropic ($D_{yy} > D_{xx} > D_{zz}$, tied to ordered imidazolate and benzimidazolate ring orientations), while the glass is isotropic or near-isotropic because vitrification randomizes those ring orientations.

Load-bearing premise

The fine-tuned machine-learning potential accurately describes how Li+ interacts with the disordered glassy ZIF framework, but it was benchmarked against DFT only for Li+ in crystalline ZIF-62, not for the glassy environments where the central claims live.

Editorial extensions

If this is right

If the simulated values hold experimentally, a ZIF glass electrolyte would deliver room-temperature Li+ diffusivity roughly an order of magnitude above its crystalline parent without any grain-boundary engineering.
The 0.16 eV activation energy is close to the regime of practical solid electrolytes, meaning the glassy framework could sustain fast transport near ambient temperature.
Because diffusion in the glass is isotropic, battery cells would not need oriented or single-crystal electrolyte layers to avoid direction-dependent resistance.
The correlation between ring orientation and transport gives a design lever: linkers or processing routes that maximize orientational disorder should also lower migration barriers in other MOF glasses.
The hopping-to-continuous crossover implies that even at low lithium loadings, a substantial fraction of Li+ is mobile in the glass, which is favorable for dilute-solution electrolytes.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

Editorial inference: a direct experimental measurement of Li+ conductivity in melt-quenched, Li-doped ZIF-4 and ZIF-62 by impedance spectroscopy or 7Li NMR would test whether the predicted 0.16 eV activation energy and the order-of-magnitude speedup appear in real glasses.
Editorial inference: a natural computational check is DFT-based migration-barrier calculations for representative glassy configurations, since the machine-learning potential used here was validated against DFT only for Li+ in crystalline ZIF-62.
Editorial inference: the ring-orientation mechanism suggests a search principle: other MOF frameworks with ordered linker apertures should show the same barrier reduction and isotropization upon vitrification, so ZIF-8, SALEM-2, or mixed-linker variants are natural next targets.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Read this one if you care about MLIP predictions for amorphous electrolytes. The headline: the paper claims vitrification of ZIF-4/ZIF-62 cuts Li+ migration activation energy from ~0.35 eV to 0.16 eV, raises extrapolated room-temperature D by 7–17x, and isotropizes transport. That is genuinely new for ZIF glasses, as far as the cited record goes, and the mechanistic story is coherent: ordered ring orientations impose direction-dependent barriers and hopping dynamics, while glassy disorder broadens the barrier distribution and makes diffusion continuous and Fickian.

What is done well: the study uses a fine-tuned DeePMD potential, ships the code and data, and validates the glass structures against experimental densities and PDFs. The replica-based MSD sampling (20–40 independent runs) is a reasonable answer to the problem of having only 10 Li ions per cell. The diffusion tensor and pole-figure analysis is a nice, direct way to tie ring orientation to anisotropic transport. The authors also deserve credit for labeling the room-temperature crystal D as extrapolated rather than simulation-direct.

The soft spots are real but not disqualifying. First, the MLIP validation set is Li+ in crystalline ZIF-62 only; the glass-phase results are the headline, and there is no direct DFT or experimental check of Li in the amorphous state. With a force RMSE of ~0.11 eV/Å, a 0.16 eV barrier could shift by tens of percent. Second, at 300 K the crystalline MSDs are still in the non-Gaussian, non-diffusive regime, so the extrapolated crystal D and hence the size of the glass enhancement carry more uncertainty than the glass numbers. There are no error bars on D or Ea anywhere. I would call these caveats rather than fatal flaws: the qualitative glass-vs-crystal trend appears in two systems and is supported by several independent descriptors.

The citation pattern is fine; building on the group's own prior DeePMD potential is normal, and the self-citations are load-bearing, not padding.

Who gets value: anyone working on MOF-glass electrolytes, or on MLIP transferability to amorphous phases. It deserves a serious referee. I would send it to review, with the request that the authors validate the MLIP against a DFT-computed Li insertion or migration barrier in the glass structure and report uncertainties on the Arrhenius parameters.

Referee Report

3 major / 5 minor

Summary. The paper uses a DeePMD machine-learned interatomic potential, fine-tuned to include Li interactions, to simulate Li+ transport in crystalline and glassy ZIF-4 and ZIF-62. The central claim is that vitrification lowers the Li+ migration activation energy from about 0.35 eV in the crystals to 0.16 eV in the glasses, increases the extrapolated room-temperature diffusion coefficient by more than an order of magnitude for ZIF-4 and by nearly sevenfold for ZIF-62, and makes Li+ diffusion isotropic or near-isotropic by randomizing imidazolate/benzimidazolate ring orientations. The authors support the mechanistic picture with non-Gaussian parameter analysis, van Hove correlation functions, diffusion-tensor ellipsoids, and ring-orientation pole figures. The manuscript also reports a replica-based sampling strategy (20 crystalline and 40 glassy replicas) and makes the workflow and trained potential available on GitHub.

Significance. If the findings hold, the paper offers a concrete and falsifiable design principle for glassy MOF electrolytes: structural disorder lowers migration barriers and isotropizes Li+ transport. The two-system consistency (ZIF-4 and ZIF-62) and the combination of diffusion-tensor analysis with linker-orientation descriptors are strengths, as is the reproducible MLIP workflow with documented replica sampling. However, the quantitative central claim depends on the transferability of the fine-tuned Li potential to amorphous Li+ environments, which is not directly validated, and on room-temperature crystalline diffusion coefficients whose 300 K simulations are not clearly in the diffusive regime. The result is therefore plausible but conditional in its current form.

major comments (3)

[§4.1, SI Fig. S13] The fine-tuned DeePMD potential is validated against DFT on a direct Li+ insertion model, but the parity plot and RMSEs (energy 5.51×10^-3 eV/atom, force 1.12×10^-1 eV/Å) are for Li+ in crystalline ZIF-62 only. The headline results, Ea ≈ 0.16 eV in the glasses and the room-temperature D enhancement, are computed in glassy ZIF-4 and ZIF-62, where the local Li+ coordination environments are not represented in the validation set. Because the reported force RMSE is comparable to the energy scale of the 0.16 eV barrier, a systematic error of even about 0.05 eV in Li-framework interactions in the amorphous phase could materially change the glass Ea and the crystal/glass crossover in Figure 2a. Please add DFT benchmarks, at minimum energies and forces for Li+ in the melt-quenched glass structures, and ideally short AIMD trajectories in the glass, or otherwise demonstrate transferability of the Li-related terms to disordered environments.
[§2.3, Fig. 2a] The room-temperature diffusion coefficients for the crystalline phases are not established at the same level as the glassy values. Section 2.3 states that in the crystalline phase the NGP increases continuously at 300 K within the simulated time window, which indicates that the 300 K MSD has not reached the linear diffusive regime; nevertheless, the Arrhenius fits in Figure 2a appear to include the 300 K point, and the reported room-temperature D values (1.91×10^-8 and 1.12×10^-8 cm2/s) are then used to quantify the order-of-magnitude and sevenfold enhancements. Please either exclude the non-diffusive 300 K crystalline points from the Arrhenius fit, run longer trajectories to reach the diffusive regime at 300 K, or provide a sensitivity analysis showing that the enhancement factors are robust to the treatment of this point.
[§4.5, Eq. (7)] No statistical uncertainties are reported for D, Ea, or D0. Given that the central claims are quantitative (barrier reduction from about 0.35 to 0.16 eV and an order-of-magnitude D increase), the absence of error bars leaves the significance of the crystal/glass differences unquantified. The replica-based sampling is in principle sufficient to report standard errors; please provide them for D(T), the fitted Ea, and the derived room-temperature values.

minor comments (5)

[Abstract] The phrase 'an grainboundary-free' should read 'a grain-boundary-free'.
[§4.1] The reference to 'Supporting Figure S12' for the DeePMD parity plot should be 'Supporting Figure S13'.
[§4.5] In the discussion of Eq. (17), the abbreviation 'NPG' should be 'NGP' to match the notation used throughout the paper.
[Supporting Information, Fig. S10 caption] The caption lists 'MSDxy, MSDxy and MSDyz'; the second term should be 'MSDxz'.
[§4.1] The force RMSE unit appears as 'ev/Å' and should be 'eV/Å'.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the Li+ diffusion results are produced by a DFT-trained MLIP and standard MSD/Arrhenius analysis, not by fitting or renaming the target quantities.

full rationale

The paper's central claims—lower activation energy, enhanced room-temperature diffusivity, and isotropization in ZIF glasses—are derived from molecular dynamics simulations using a DeePMD potential that was fine-tuned with Li-specific DFT data and validated on a separate DFT test set (Section 4.1, SI Figure S13). The room-temperature diffusion coefficients are explicitly labeled as extrapolated from Arrhenius fits to simulated high-temperature data; this is standard analysis of simulation output, not a parameter fitted to the target experimental result. The main self-reference is the group's prior ZIF DeePMD potential (ref 23), but that potential was DFT-trained and experimentally benchmarked in its original work, and the present paper further benchmarks the Li-containing model against DFT. Per the review rules, such externally validated citations are independent support and do not constitute load-bearing circularity. The limited validation of Li+ behavior in the amorphous phase is a legitimate transferability/correctness concern, not a circularity: the glassy-state predictions are not defined in terms of the quantities they claim to predict, and no equation in the paper reduces an output to an input by construction. No self-definitional, fitted-as-prediction, uniqueness-imported, ansatz-smuggled, or renaming step was found.

Assumptions & free parameters 6 free parameters · 5 assumptions · 0 invented entities

The MLIP is the main instrument; its weights are trained on DFT data, and the paper's diffusion results are not fitted to experiment. The Arrhenius extrapolation supplies the crystalline room-temperature D used in the headline enhancement factor, so it functions as a fitted input for that specific claim.

free parameters (6)

Activation energy Ea for crystalline ZIF-4 = 0.36 eV
Obtained by Arrhenius fit to simulated D(T) data; used to extrapolate the room-temperature crystalline D.
Activation energy Ea for crystalline ZIF-62 = 0.34 eV
Obtained by Arrhenius fit to simulated D(T) data; used to extrapolate the room-temperature crystalline D.
Activation energy Ea for glassy ZIF-4 and ZIF-62 = 0.16 eV
Obtained by Arrhenius fit to simulated D(T) data for both glassy systems.
Pre-exponential factor D0 per system = Not reported numerically
Arrhenius prefactors fitted to simulated D(T); not tabulated in the paper.
Lithium loading = 10 Li per 3x3x3 supercell
Dilute-regime choice for tracer diffusion; does not represent a practical battery electrolyte carrier concentration.
Melt-quench cooling rate and pressure = 5 K/ps, 1 GPa in first stage
Glass preparation protocol chosen to reproduce experimental densities; not fitted to the diffusion target.

assumptions (5)

domain assumption The fine-tuned DeePMD potential reproduces DFT accuracy for Li+ in ZIF frameworks, including amorphous environments.
Validated only against a crystalline Li-ZIF test set (SI Figure S13); glassy Li+ environments are not directly benchmarked. Entered in Section 4.1.
domain assumption DFT training and test data are accurate for Li-ZIF interactions.
The DFT functional is not specified in the text; all MLIP predictions inherit this assumption.
domain assumption Arrhenius behavior holds from 400-700 K down to 300 K in both phases.
The crystalline 300 K D is extrapolated; NGP at 300 K shows no diffusive regime within 1 ns, so the extrapolation is unverified. Sections 2.3 and 4.5.
domain assumption 10 Li ions per supercell represent the dilute limit, so the tracer diffusion coefficient is the relevant ionic mobility.
Replica averaging mitigates statistics, but assumes no Li-Li interactions and no finite-size effects. Section 4.5.
domain assumption 1 ns production runs are sufficient to reach the diffusive regime for the reported D values.
Holds for glasses at 300 K and crystals above 400 K; for crystals at 300 K the diffusive regime is not reached, making the reported crystalline RT D an extrapolation. Section 2.3.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Fast Isotropic Li-Ion Diffusion in Zeolitic Imidazolate Framework Glass Electrolytes for Batteries." pith.science (2026). https://pith.science/paper/ZQMTAHZM

@misc{pith2026260806902,
  author       = {Pith},
  title        = {Pith review of: Fast Isotropic Li-Ion Diffusion in Zeolitic Imidazolate Framework Glass Electrolytes for Batteries},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ZQMTAHZM}},
  note         = {Machine review of arXiv:2608.06902}
}

read the original abstract

All-solid-state lithium batteries require solid electrolytes that combine rapid room-temperature ion transport with mechanical robustness and interfacial compatibility. Zeolitic imidazolate framework (ZIF) glasses, with ZIFs being a sub-set of metal-organic frameworks, offer an attractive yet relatively underexplored platform because they combine an grainboundary-free and amorphous topology with chemically tunable frameworks. Here, we reveal that structural disorder unlocks fast and isotropic lithium diffusion in ZIF glasses. This is realized by using a machine learning interatomic potential to simulate Li+ transport in crystalline and glassy ZIF-4 and ZIF-62. Structural disorder reduces the activation energy for Li+ migration from ~0.35 eV to 0.16 eV and increases the extrapolated room-temperature diffusion coefficient by more than one order of magnitude for ZIF-4 and nearly sevenfold for ZIF-62. Analyses of non-Gaussian dynamics and van Hove correlation functions reveal that Li+ diffusion in crystalline ZIFs occurs via rare, dynamically heterogeneous hopping events among well-defined cages, whereas Li+ diffusion in glassy ZIFs is more homogeneous, continuous, and Fickian-like, benefiting from a wide distribution of coordination geometries and migration barriers. Li+ diffusion in crystalline ZIFs is strongly anisotropic, reflecting that ordered orientations of imidazolate and benzimidazolate rings impose distinct energy barriers along different crystallographic directions. Upon vitrification, these ring orientations become randomized, and hence, the diffusion of Li+ becomes isotropic or near-isotropic. These findings imply that well-designed metal-organic framework glasses are a promising candidate as high-performance solid-state electrolytes.

Figures

Figures reproduced from arXiv: 2608.06902 by the authors.

**Figure 2.** Disorder-enabled lithium diffusion in crystalline and glassy ZIF electrolytes. (a) Arrhenius analysis of Li+ diffusion coefficients in crystalline and glassy ZIF-4 and ZIF-62. Diffusion coefficients were calculated at 300, 400, 500, 600, and 700 K, with additional simulations at 350, 450, and 550 K for ZIF-62 to improve the reliability of the Arrhenius fits. (b,c) Temperature-dependent nonGaussian parameter (NGP) f… view at source ↗

**Figure 3.** Transition from discrete hopping to continuous Li-ion migration in ZIF-62. (a) Self-part van Hove correlation function Gs(r,t) for Li ions in crystalline and glassy ZIF-62 at 700 K. (b) Schematic illustration of the contrasting Li+ migration mechanisms in crystalline (left) and glassy (right) ZIF-62 at 700 K. In the crystalline framework, Li ions migrate through discrete hopping between well-defined sites, whereas i… view at source ↗

Figures from the paper (1 more)

**Figure 4.** Figure 4: Disorder-induced transition from anisotropic to isotropic Li-ion diffusion. (a) Workflow for visualizing Li-ion diffusion anisotropy from the diffusion tensor. The tensor is computed from ionic trajectories, represented as a diffusivity ellipsoid, transformed using log…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

7 extracted references · 7 canonical work pages

[1]

Building better batteries

(1) Armand, M.; Tarascon, J.-M. Building better batteries. Nature 2008, 451 (7179), 652–657. (2) Schmuch, R.; Wagner, R.; Hörpel, G.; Placke, T.; Winter, M. Performance and cost of materials for lithium-based rechargeable automotive batteries. Nature energy 2018, 3 (4), 267–

work page 2008
[4]

D.; Yue, Y.; Li, P.; Qiao, A.; Tao, H.; Greaves, N

(27) Bennett, T. D.; Yue, Y.; Li, P.; Qiao, A.; Tao, H.; Greaves, N. G.; Richards, T.; Lampronti, G. I.; Redfern, S. A.; Blanc, F. Melt -quenched glasses of metal–organic frameworks. Journal of the American Chemical Society 2016, 138 (10), 3484–3492. (28) Stepniewska, M.; Østergaard, M. B.; Zhou, C.; Yue, Y. Towards large -size bulk ZIF-62 glasses via opt...

work page 2016
[125]

(26) Kraß, H.; Huang, J.; Moosavi, S. M. MOFSimBench: evaluating universal machine learning interatomic potentials in metal- organic framework molecular modeling. NPJ Computational Materials 2025, 12 (1),

work page 2025
[278]

28 (3) Whittingham, M. S. History, evolution, and future status of energy storage. Proceedings of the IEEE 2012, 100 (Special Centennial Issue), 1518–1534. (4) Porz, L.; Swamy, T.; Sheldon, B. W.; Rettenwander, D.; Frömling, T.; Thaman, H. L.; Berendts, S.; Uecker, R.; Carter, W. C.; Chiang, Y. M. Mechanism of lithium metal penetration through inorganic s...

work page 2012
[2025]

R.; Zhou, D.; Yang, Y.; Smedskjaer, M

29 (13) Li, Z.; Hu, J.; Du, Z.; Ge, X.; Jensen, L. R.; Zhou, D.; Yang, Y.; Smedskjaer, M. M.; Qiu, J.; Yue, Y. Tailoring Optical Properties of Rare‐Earth Doped Oxyfluoride Glass‐Ceramics by Tuning Crystal Structure. Advanced Optical Materials 2024, 12 (34), 2401648. (14) Li, Z.; Wang, Z.; Zhang, H.; Ge, X.; Hung, I.; Yin, B.; Cao, F.; Banerjee, P.; Xu, T....

work page 2024
[6296]

Metal–organic frameworks (MOFs)

(9) Kitagawa, S. Metal–organic frameworks (MOFs). Chemical Society Reviews 2014, 43 (16), 5415–5418. (10) Furukawa, H.; Cordova, K. E.; O’Keeffe, M.; Yaghi, O. M. The chemistry and applications of metal-organic frameworks. Science 2013, 341 (6149), 1230444. (11) Smirnova, O.; Hwang, S.; Sajzew, R.; Ge, L.; Reupert, A.; Nozari, V.; Savani, S.; Chmelik, C.;...

work page 2014
[8079]

M.; Yue, Y

(18) Stepniewska, M.; Januchta, K.; Zhou, C.; Qiao, A.; Smedskjaer, M. M.; Yue, Y. Observation of indentation- induced shear bands in a metal− organic framework glass. Proceedings of the National Academy of Sciences 2020, 117 (19), 10149–10154. (19) Ding, J.; Du, T.; Jensen, L. R.; Sørensen, S. S.; Wang, D.; Wang, S.; Zhang, L.; Yue, Y.; Smedskjaer, M. M....

work page 2020

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.

[1] [1]

Building better batteries

(1) Armand, M.; Tarascon, J.-M. Building better batteries. Nature 2008, 451 (7179), 652–657. (2) Schmuch, R.; Wagner, R.; Hörpel, G.; Placke, T.; Winter, M. Performance and cost of materials for lithium-based rechargeable automotive batteries. Nature energy 2018, 3 (4), 267–

work page 2008

[2] [4]

D.; Yue, Y.; Li, P.; Qiao, A.; Tao, H.; Greaves, N

(27) Bennett, T. D.; Yue, Y.; Li, P.; Qiao, A.; Tao, H.; Greaves, N. G.; Richards, T.; Lampronti, G. I.; Redfern, S. A.; Blanc, F. Melt -quenched glasses of metal–organic frameworks. Journal of the American Chemical Society 2016, 138 (10), 3484–3492. (28) Stepniewska, M.; Østergaard, M. B.; Zhou, C.; Yue, Y. Towards large -size bulk ZIF-62 glasses via opt...

work page 2016

[3] [125]

(26) Kraß, H.; Huang, J.; Moosavi, S. M. MOFSimBench: evaluating universal machine learning interatomic potentials in metal- organic framework molecular modeling. NPJ Computational Materials 2025, 12 (1),

work page 2025

[4] [278]

28 (3) Whittingham, M. S. History, evolution, and future status of energy storage. Proceedings of the IEEE 2012, 100 (Special Centennial Issue), 1518–1534. (4) Porz, L.; Swamy, T.; Sheldon, B. W.; Rettenwander, D.; Frömling, T.; Thaman, H. L.; Berendts, S.; Uecker, R.; Carter, W. C.; Chiang, Y. M. Mechanism of lithium metal penetration through inorganic s...

work page 2012

[5] [2025]

R.; Zhou, D.; Yang, Y.; Smedskjaer, M

29 (13) Li, Z.; Hu, J.; Du, Z.; Ge, X.; Jensen, L. R.; Zhou, D.; Yang, Y.; Smedskjaer, M. M.; Qiu, J.; Yue, Y. Tailoring Optical Properties of Rare‐Earth Doped Oxyfluoride Glass‐Ceramics by Tuning Crystal Structure. Advanced Optical Materials 2024, 12 (34), 2401648. (14) Li, Z.; Wang, Z.; Zhang, H.; Ge, X.; Hung, I.; Yin, B.; Cao, F.; Banerjee, P.; Xu, T....

work page 2024

[6] [6296]

Metal–organic frameworks (MOFs)

(9) Kitagawa, S. Metal–organic frameworks (MOFs). Chemical Society Reviews 2014, 43 (16), 5415–5418. (10) Furukawa, H.; Cordova, K. E.; O’Keeffe, M.; Yaghi, O. M. The chemistry and applications of metal-organic frameworks. Science 2013, 341 (6149), 1230444. (11) Smirnova, O.; Hwang, S.; Sajzew, R.; Ge, L.; Reupert, A.; Nozari, V.; Savani, S.; Chmelik, C.;...

work page 2014

[7] [8079]

M.; Yue, Y

(18) Stepniewska, M.; Januchta, K.; Zhou, C.; Qiao, A.; Smedskjaer, M. M.; Yue, Y. Observation of indentation- induced shear bands in a metal− organic framework glass. Proceedings of the National Academy of Sciences 2020, 117 (19), 10149–10154. (19) Ding, J.; Du, T.; Jensen, L. R.; Sørensen, S. S.; Wang, D.; Wang, S.; Zhang, L.; Yue, Y.; Smedskjaer, M. M....

work page 2020