Offline XRD coherently diffracting domain-size analysis. Single-file HTML with Scherrer, multi-peak and isotropic Williamson–Hall modes. v1.1.0.
Open the tool · Download HTML · XRD Toolkit
- Scherrer: D = Kλ / (β cosθ), with θ = 2θ/2 and β in radians.
- Endpoint-baseline integration or SNIP baseline + Gaussian width approximation. The SNIP mode infers a Gaussian area from peak height/FWHM; it is not a nonlinear Gaussian fit.
- FWHM or integral breadth. Instrument correction is selected independently: quadratic subtraction for a Gaussian profile, linear for a Cauchy profile. Supply instrumental width of the same width definition; mixed profiles require a more complete treatment.
- Multi-peak isotropic Williamson–Hall: β cosθ = Kλ/D + 4ε sinθ. At least three distinct clean same-phase reflections; return intercept, slope, R², D and strain. Nonpositive intercepts/negative strain suppress physical outputs rather than being converted to positive values.
- PtCo (220) empirical cross-check with a positive TEM calibration factor C. (220) is a fundamental reflection; broadening/overlap/anisotropy determine its suitability.
- CSV/JSON exports with units and settings; input edits invalidate old results.
Kα₂ stripping is not implemented. Use monochromated or externally corrected data. Scherrer domain size is not particle diameter; microstrain, faults, anisotropy and overlap can bias it. A W-H line fit and high R² do not validate those assumptions.
Open index.html, load the synthetic demo or your numeric-text/JSON scan, check peak windows, choose width and instrumental-profile settings, and calculate. For W-H, add at least three isolated reflections. Compare (111)/(220) with caution if split or overlapping fundamentals are present.
Text: .xy .xye .csv .txt .dat .asc .ras. JSON: paired x/y arrays or point pairs/objects. Descending scans are normalized; nonfinite, all-zero, duplicate and nonmonotonic input is rejected. Incomplete FWHM crossings require a wider region.
npm ci
npm testRegressions recover the analytic Gaussian Scherrer width and a synthetic W-H case (D=50 nm, ε=0.001), exercise both instrument-profile corrections, reject nonphysical cases, and check parsing, exports and result invalidation. DOM/canvas mocks do not verify browser-pixel rendering.
References: Scherrer (1918); Williamson & Hall, Acta Metallurgica 1 (1953), 22–31, doi:10.1016/0001-6160(53)90006-6; Cullity & Stock, Elements of X-Ray Diffraction; Ryan et al. (1988), SNIP baseline.