Data for “Fragmentation Dynamics Encode Molecular Architecture”

Each material was measured at 80 nominal gate delays, with 10 stored repeats per delay, yielding 6,400 averaged single-delay spectra.Binary_mixture_spectra.zip contains the emission spectra of PA–PS, PP–PA, PP–PS, and HDPE–LDPE binary-mixture pellets. PA–PS, PP–PA, and PP–PS were prepared at first-component mass percentages of 10%, 25%, 50%, 75%, and 90%. HDPE–LDPE mixtures were prepared at HDPE mass percentages of 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, and 90%.

Each composition was measured at 80 nominal gate delays, with five stored repeats per delay.Source_data.zip contains numerical source data for the figures and tables in the manuscript and Supplementary Materials.Acquisition and interpretationExcitation was provided by three synchronized, non-collinear femtosecond pulses with a central wavelength of 800 nm, a pulse duration of 50 fs, and a total pulse energy of 2.1 mJ.

Emission spectra were collected at nominal gate delays from 10 to 800 ns in 10-ns increments using a 200-ns electronic gate width. The spectral analyses used 22,104 wavelength variables between 220 and 850 nm.Each stored spectrum is an average of signals accumulated over 10 laser shots delivered to fresh sampling positions. Spectra sharing a stored-repeat index across the nominal delays were assembled into ensemble delay trajectories.

These trajectories do not track a single plasma event or ablation crater continuously. The 10-ns delay increment is a sampling increment, not an independently established temporal resolution; adjacent gate-integration intervals overlap.Analysis protocolsPolymer-sheet trajectories support analyses of temporal order, continuous visible-window position and length, training-data requirements, and relative emission-channel evolution.

Pure-pellet spectra support construction and repeat-block evaluation of material-specific dynamic bases. Binary-pair identification and concentration inversion were evaluated separately.For concentration inversion with known material pairs, PA–PS, PP–PA, and PP–PS used 10%, 50%, and 90% compositions for calibration and fixed 25% and 75% compositions for evaluation. HDPE–LDPE calibration used 10/50/90%, 10/30/50/70/90%, or 10/30/40/50/60/70/90%, with evaluation fixed at 20% and 80%.

These fixed-composition evaluations are retrospective rather than newly conducted blind tests.ScopeThis release contains data only. Source code and analysis scripts are not included. The data support inspection of the reported spectral trajectories, model-evaluation results, and mixture-concentration analyses.

The samples are laboratory polymer sheets and pellets, not field-collected environmental microplastics.