Files

LEM 2025 energy scan — Gaussian-cosine fits at two temperatures

Worked example behind the figure docs/source/images/lem2025-energy-scan.png (used in docs/source/parameters-plots.rst). It is a real set of fits, not a mock-up.

Data

  • Instrument: LEM (PSI), proposal 20241780.
  • Sample: 85 nm Nb/Al₂O₃, applied field ≈ 100 G, spin angle −45°.
  • Data set 1: runs 1014–1027, sample temperature 3.8 K (superconducting Nb).
  • Data set 2: runs 1033–1045, sample temperature 10 K (normal state).
  • Implantation energy scanned from ≈ 0.96 keV to ≈ 17.95 keV.
  • Run files (lem25_his_01xx.root) are MUSR-ROOT files named under <data root>/lem/his/2025/. They can be fetched with muSRFlow's Download run data flow or from musruser.psi.ch.

meta.tsv records the run number, implantation energy, temperature, field, sample, and run title read from each ROOT run header (meta.C).

Fit

Each run is fitted independently to the muSRFlow Gaussian Cosine model:

asymmetry   Asy
simpleGss   Sgm
TFieldCos   Phi  Frq

with the field constrained through Frq = gamma_mu * Bfield, so Bfield (Gauss) is a free parameter and is exported by msr2data. Settings:

  • fit range 0 8 µs, packing 100;
  • LEM detectors: forward 21 25, backward 23 27, alpha 5;
  • MINIMIZE + HESSE, fittype 2 (asymmetry);
  • starting values seeded from the preceding run in energy order.

fit_all.py regenerates the 27 _asy.msr files, fits them with musrfit, and writes dataset1_3.8K.dat / dataset2_10K.dat with msr2data. example_1014_asy.msr is the generated document for the first run.

export MUSRFLOW_DATA_ROOT=/path/to/data     # contains lem/his/2025/...
python3 fit_all.py
python3 plot.py                              # rewrites the docs figure

Requires musrfit, msr2data, ROOT, numpy, and matplotlib.

Result

All 27 fits converged, χ²/NDF ≈ 0.9–2.1. The fitted field is flat at ≈ 99.6 G at 10 K, while at 3.8 K it is screened below the applied value and dips to ≈ 94.7 G around 9 keV — the Meissner response of the superconducting Nb film. The initial asymmetry rises with energy as the stopping profile moves away from the surface; σ is roughly constant away from the lowest energies.

The documentation figure plots B, σ, and A₀ versus implantation energy with data set 1 in blue and data set 2 in red, overlaid on the same three panels — the same multi-file overlay shown by the parameter-plot window.