Skip to content

Prediction of the Glass Transition Temperatures of Zeolitic Imidazolate Glasses through Topological Constraint Theory

Evidence and attribution

Authority of statements

Prose sections below (Summary, Methods, Findings, etc.) are curated summaries of the publication identified by doi, title, and pdf_path in the front matter above. They are not new primary claims by this wiki.

For definitive numerical values, reaction schemes, and interpretations, use the peer-reviewed article (and optional records under normalized/papers/ when present)—not this page alone.

Summary

For agZIF-62 glasses of composition Zn(Im₂₋ₓbImₓ), the work builds a topological constraint model for glass transition temperature (Tg) that combines experimental data with ReaxFF molecular dynamics to define a hierarchy of bond constraints. The model reproduces Tg versus benzimidazolate content with reported ~3.5 K error and is extended to 5-methylbenzimidazolate, yielding a ternary Tg diagram for a hypothetical three-ligand framework. Yang, Shin, and van Duin anchor the reactive MD contribution; Mauro and collaborators provide the constraint-theory framework for MOF glasses.

Methods

Topological constraint theory (TCT) backbone

The letter develops a constraint hierarchy for agZIF-62 compositions Zn(Im\(_{2-x}\)bIm\(_x\)), treating ZnN\(_4\) tetrahedra and imidazolate / benzimidazolate linkers as network formers. Hydrogen is excluded from connectivity counting because it is monovalent and does not change the Im/bIm backbone graph. Bond-stretching (BS) and bond-bending (BB) constraints are assigned per Figure 1, with mean coordination ⟨r⟩ feeding the usual TCT counting formulas (eq. (1) in the paper). Temperature-dependent constraint survival probabilities \(q_n(T)\) follow the Mauro onset-temperature construction (their eq. (2)), yielding ordered onsets \(T_\beta < T_\gamma < T_\alpha\) so the network is stressed-rigid at low \(T\) and progressively unlocks floppy modes as \(T\) rises (Figure 2).

ReaxFF MD used as a rigidity probe (SI)

The main text does not reproduce full LAMMPS decks on the printed pages; instead it cites Supporting Information ReaxFF MD of ZIF-62 to measure relative standard deviations (RSD) of C–C and C–N bonds and of C–N–C / N–C–N angles (RSD < 0.04–0.05) versus temperature from 300 K to 1200 K, while N–Zn–N remains comparatively floppy (RSD ~0.12–0.19). The authors caution that Zn–N(bIm) stiffening inferred from MD appears near ~500 K but the true onset may be higher because the ZIF ReaxFF parameterization was trained on small Zn/Im clusters rather than full frameworks (Gaillac et al. barrier discussion).

  • Engine / code: ReaxFF molecular dynamics; LAMMPS is the standard engine for this parameter line in the broader corpus, but operators should copy the exact package callouts from the SI PDF bundled with papers/Yang_ZIF_JPCL_2018.pdf.
  • System / PBC: ZIF-62 supercell for the RSD study (atom totals and periodic lattice vectors N/A — SI-only in this workspace curation).
  • Ensemble, timestep, barostat, production length: N/A — not restated on the letter pages; the article directs readers to the Supporting Information for equilibration/production segment lengths in ps, NVT/NPT labels, and fs timesteps.
  • Thermostat: N/A — thermostat brand/damping is not named in the letter excerpt used here—confirm in the SI trajectory description.
  • Pressure: N/A — MD barostat targets for the RSD validation runs are not stated in the letter body (see SI if pressure coupling was used).
  • Electric field / replica / metadynamics: N/A — not used for the RSD survey described above.

Findings

  • Outcomes / mechanisms: The TCT + ReaxFF pipeline reproduces \(T_g\) vs benzimidazolate fraction for Zn(Im\(_{2-x}\)bIm\(_x\)) glasses with mean absolute error ~3.5 K (letter headline; abstract quotes <4 K vs DSC data). Constraint onsets (\(T_\beta<T_\gamma<T_\alpha\)) explain how the network moves from stressed-rigid to floppy as temperature increases (Figure 2 narrative).
  • Comparisons: Predicted \(T_g\) tracks experimental DSC traces for literature agZIF compositions; RSD metrics from ReaxFF MD justify treating Im/bIm internal coordinates as rigid relative to the floppy N–Zn–N hinge at sub-\(T_g\) temperatures.
  • Sensitivity / design levers: Varying \(x\) shifts the balance of Zn–Im vs Zn–bIm constraints, which feeds directly into the \(T_g\) curve; extension to 5-methylbenzimidazolate produces a ternary \(T_g\) diagram for a hypothetical three-ligand glass.
  • Limitations / outlook (as authored): The authors flag that MD-predicted Zn–N(bIm) stiffening near ~500 K may underestimate the true onset because ReaxFF was trained on small clusters; TCT itself remains a coarse rigidity factorization that may need full cooling MD near borderline chemistries.
  • Corpus honesty: Verbatim abstract paste and publisher boilerplate were removed from this page; operators needing DSC scan rates (10 K/min), constraint tables (Table 1), or SI MD settings should read papers/Yang_ZIF_JPCL_2018.pdf directly.

Limitations

  • TCT assumes a coarse decomposition of rigidity; local chemical detail may require full MD cooling curves for borderline compositions.
  • ReaxFF must be trusted for each ligand chemistry before constraint parameters are exported to theory.

Relevance to group

Links Penn State ReaxFF on ZIF glasses to Mauro-group topological constraint methods for Tg design.

Citations and evidence anchors

  • DOI: https://doi.org/10.1021/acs.jpclett.8b03348 (papers/Yang_ZIF_JPCL_2018.pdf).