Skip to content

High-harmonic generation and attosecond pulses from hydrogen - #9

Merged
rasidi3112 merged 3 commits into
mainfrom
feat/high-harmonic-generation
Oct 24, 2026
Merged

rasidi3112 merged 3 commits into
mainfrom
feat/high-harmonic-generation

Conversation

@rasidi3112

Copy link
Copy Markdown
Owner

Summary

Adds strong-field physics: a hydrogen atom in an intense laser pulse, which emits high harmonics and attosecond pulses (the three-step recollision model behind attosecond science).

  • quantumlab.strongfield:
    • laser units (wavelength → ω, intensity → field, U_p);
    • the classical three-step model;
    • dipole acceleration from Ehrenfest's theorem;
    • harmonic spectra, windowed Gabor time–frequency maps and attosecond pulse synthesis.
  • SoftCoulomb potential with an analytic gradient. In 1D with a² = 2 its ground state is −0.5 a.u., i.e. hydrogen.
  • flat_top_envelope gives long pulses with well-resolved harmonics.
  • Example 12: hydrogen at 800 nm and 10¹⁴ W/cm². It produces the spectrum, a Gabor map with classical short/long returns, the attosecond pulse train, the electron density in space-time, and an 8-intensity cutoff scan.
  • Docs: a strong-field guide page, an API page, the examples gallery and the README.

Validation

Check Result
Hydrogen ground state −0.5 a.u. to 3e-14
Three-step maximum return energy 3.1731 U_p (birth 18°, return 252°)
Odd harmonics only (inversion symmetry) even orders 2300× weaker
Emission times vs classical returns, H11–H19 median 0.02–0.08 cycle (analysis resolution 0.053 cycle)
Attosecond bursts (H11–H21) ≈ 200 as, spacing 0.499–0.503 cycle
Cutoff vs intensity linear in U_p; fitted spectral edge (3.48 ± 0.20) U_p + (1.52 ± 0.08) I_p

The cutoff row is reported with its caveat. A single 1D atom has a plateau with deep interference minima and a gradual roll-off, so the extracted edge depends on its definition: a 2-decade drop is used. The fit is therefore compared with, not claimed to equal, the classical (3.17 U_p + I_p) and quantum-corrected (3.17 U_p + 1.32 I_p) laws. 1D models also over-estimate ionisation. Both caveats are stated in the example, the docs and the README.

The version number is not changed here, to avoid conflicting with the open release PR #8.

Test plan

  • pytest: 127 passed, 12 skipped (CUDA-only), including 9 new strong-field tests
  • python benchmarks/validation_suite.py: 21/21
  • mkdocs build --strict: no warnings; examples page has 20 images, none broken
  • README: 39/39 images referenced and present
  • CI on this PR

- quantumlab.strongfield: laser units (wavelength, intensity, U_p),
  classical three-step model (max return energy 3.1731 U_p), dipole
  acceleration from Ehrenfest's theorem, harmonic spectra, windowed
  Gabor time-frequency maps and attosecond pulse synthesis
- SoftCoulomb potential with analytic gradient (1D: E0 = -0.5 a.u.)
- flat_top_envelope for long, well-resolved laser pulses
- tests for units, three-step maximum, hydrogen ground state, gradient,
  spectra, Gabor localisation, pulse trains and Ehrenfest acceleration
- 12: hydrogen in an 800 nm, 1e14 W/cm^2 flat-top pulse: harmonic
  spectrum, Gabor map against classical short/long returns, attosecond
  pulse train, electron density in space-time, and an intensity scan of
  the cutoff (reported with its definition dependence)
- validation suite: hydrogen ground state, three-step maximum, odd-only
  harmonics and half-cycle attosecond spacing (21/21 pass)
@rasidi3112
rasidi3112 merged commit 597515e into main Oct 24, 2026
8 checks passed
@rasidi3112
rasidi3112 deleted the feat/high-harmonic-generation branch October 24, 2026 08:18
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant