# Quenched Strip Fracture: 58 Phase-Field Simulations of Thermal-Shock Crack Propagation in Glass

## Facts

- Provider: Nidhin Ninan
- License: CC-BY-4.0 (https://huggingface.co/datasets/nidhinninan/quenched-strip-fracture)
- Size: 98 files
- Format: Parquet (in sampled files)
- Stored at: Hugging Face, https://huggingface.co/datasets/nidhinninan/quenched-strip-fracture
- Access: public, readable without an account
- Category: Physics
- Source: https://huggingface.co/datasets/nidhinninan/quenched-strip-fracture
- Page: https://scigantic.com/archive/e21ed78a-0c00-532f-aaec-e82682fe8cb5
- Keywords: Physics, Parquet, Hugging Face

## About

Finite element phase-field simulations of thermally driven brittle fracture in a hot soda-lime glass strip (30 x 16 mm, 5 mm starter notch) drawn through a quench front, the classic quenched-strip experiment in which the crack stays put, runs straight, oscillates or branches. 58 runs sweep front speed V = 0.2 to 0.8 mm/s and temperature drop of 67 to 267 C, with random notch offset and strength field, in 2D plane stress with a strength-surface phase-field model and a prescribed moving temperature front, solved with FEniCSx (dolfinx 0.9). Each HDF5 trajectory stores about 106 snapshots of displacement u and damage d on the adaptive triangular mesh (19,000 to 110,000 nodes). Rule-based regime labels give 20 straight, 20 oscillating, 15 branching and 3 no-growth runs, with train, val, test and two out-of-distribution splits. Parquet tables give runs, crack-tip tracks, crack paths and series; quenched_strip.py reads the data. About 8.4 GB. Data CC BY 4.0, reader script MIT, as stated in the card.

## Load the data

```python
from datasets import load_dataset
ds = load_dataset("nidhinninan/quenched-strip-fracture")
print(ds)
```

## Data structure

File types in sampled files: Parquet

Example file: `metadata/crack_paths.parquet`

## Open it

- In a hosted notebook with the data mounted: https://scigantic.com/archive/e21ed78a-0c00-532f-aaec-e82682fe8cb5
- Full record as JSON (schema card included): https://api.scigantic.com/api/archives/e21ed78a-0c00-532f-aaec-e82682fe8cb5
- Over MCP: `get_schema_card` with id `e21ed78a-0c00-532f-aaec-e82682fe8cb5` at https://api.scigantic.com/mcp

## Related topics

- [Parquet datasets](https://scigantic.com/topics/parquet)
- [Physics datasets](https://scigantic.com/topics/physics)
