Organic Semiconductor Molecules with Predicted Reorganization Energies

Yamato Nakanishi published a dataset on figshare in June 2026 detailing a data-driven framework for designing organic semiconductor molecules. The dataset likely contains molecular structures and their predicted hole and electron reorganization energies, generated via hierarchical variational autoencoders and Gaussian mixture regression. The file is 79.8 KB in size and is provided in XLSX format.

Use Cases

• Predicting hole and electron reorganization energies for organic molecules based on their latent representations.
• Generating novel molecular candidates for organic semiconductors under structural constraints like ring number and size.
• Identifying key structural descriptors linked to high carrier mobility using Bayesian optimization.

Strengths

• The dataset is derived from a published computational framework combining hierarchical variational autoencoders, Gaussian mixture regression, and Bayesian optimization.
• It includes a molecule identified via sulfur-to-nitrogen substitution with a new minimum hole reorganization energy.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• The dataset's small size (79.8 KB) suggests a limited scope of molecular candidates.

Provenance

Source

figshare

Collection Method

Generated via a computational framework combining hierarchical variational autoencoders (HVAE), Gaussian mixture regression (GMR), and Bayesian optimization.

Freshness

Last updated 2026-06-04 05:03:40; freshness should be verified.