Nanotech News and Headlines
  • All Technology
  • AI
  • Autonomy
  • B2B Growth
  • Big Data
  • BioTech
  • ClimateTech
  • Consumer Tech
  • Crypto
  • Cybersecurity
  • DevOps
  • Digital Marketing
  • Ecommerce
  • EdTech
  • Enterprise
  • FinTech
  • GovTech
  • Hardware
  • HealthTech
  • HRTech
  • LegalTech
  • Nanotech
  • PropTech
  • Quantum
  • Robotics
  • SaaS
  • SpaceTech
AllNewsDealsSocialBlogsVideosPodcastsDigests

Nanotech Pulse

EMAIL DIGESTS

Daily

Every morning

Weekly

Sunday recap

NewsDealsSocialBlogsVideosPodcasts
NanotechNewsTwisting‐Induced Phonon Localization and Ultralow Thermal Conductivity in Penta‐PdTe2 Bilayer Revealed by a Universal Machine‐Learning Potential
Twisting‐Induced Phonon Localization and Ultralow Thermal Conductivity in Penta‐PdTe2 Bilayer Revealed by a Universal Machine‐Learning Potential
Nanotech

Twisting‐Induced Phonon Localization and Ultralow Thermal Conductivity in Penta‐PdTe2 Bilayer Revealed by a Universal Machine‐Learning Potential

•January 27, 2026
0
Small (Wiley)
Small (Wiley)•Jan 27, 2026

Why It Matters

The ability to suppress thermal transport by twist opens new pathways for high‑performance thermoelectrics and advanced heat‑management technologies in nanoscale devices.

Key Takeaways

  • •Twisting cuts bilayer κ by 76% versus untwisted
  • •ML potential predicts energies within 2.3 meV/atom
  • •Lattice κ reaches 0.30 W m⁻¹ K⁻¹
  • •Pentagonal topology enhances phonon anharmonicity
  • •Approach enables rapid design of low‑thermal‑conductivity 2D materials

Pulse Analysis

The emergence of pentagonal two‑dimensional crystals has expanded the design space for functional nanomaterials. Penta‑PdTe2 stands out due to its heavy constituent elements, which naturally lower phonon group velocities and increase scattering. However, the intrinsic thermal conductivity of a pristine monolayer remains relatively high for thermoelectric applications. By stacking two layers and introducing a controlled interlayer twist, researchers create a quasi‑periodic landscape that traps vibrational modes, effectively turning the material into a phonon glass while preserving its electronic structure.

A key enabler of this discovery is the universal machine‑learning potential NEP89, which was calibrated against high‑fidelity density‑functional data for monolayer, bilayer, and twisted configurations. The model achieves an energy prediction error of just 2.3 meV per atom, allowing large‑scale homogeneous non‑equilibrium molecular dynamics simulations that would be prohibitive with first‑principles methods alone. Coupled with Wigner transport theory, the workflow delivers a real‑space picture of phonon localization and quantifies the dramatic reduction in lattice thermal conductivity. This hybrid computational strategy demonstrates how AI‑driven potentials can accelerate the exploration of complex structural degrees of freedom in 2D systems.

The practical implications are significant. Ultrathin films with κ values near 0.3 W m⁻¹ K⁻¹ rival the performance of bulk amorphous materials, yet they retain the mechanical flexibility and integration compatibility of crystalline layers. Such properties are attractive for next‑generation thermoelectric generators, on‑chip cooling solutions, and phononic devices that manipulate heat flow at the nanoscale. Moreover, the twist‑induced phonon localization mechanism is likely transferable to other layered compounds, suggesting a broader design rule for engineering low‑thermal‑conductivity materials without sacrificing electronic conductivity. Future work will focus on experimental validation and scaling the approach to heterostructures with multiple twist angles.

Twisting‐Induced Phonon Localization and Ultralow Thermal Conductivity in Penta‐PdTe2 Bilayer Revealed by a Universal Machine‐Learning Potential

Read Original Article
0

Comments

Want to join the conversation?

Loading comments...