What you need

• Standard: Prefer LaB₆ (NIST SRM 660x) for line shape; Si (SRM 640x) also works (very low size/strain). Avoid plastically deformed powders.
• Same conditions: Use the exact optics you’ll use for samples (Bragg–Brentano, slit sizes, divergence/receiving slits, monochromator or not, detector type, scan range, step, time/step).
• Cu Kα wavelengths (if no monochromator): λ₁ = 1.540593 Å, λ₂ = 1.544427 Å, I(Kα₂)/I(Kα₁) ≈ 0.5.

Measurement (high statistics)

• Zero-background holder; gently spread a thin layer; spin the stage if available.
• Range: e.g., 10–140° 2θ, step 0.01–0.02°, counting long enough that mid-angle peaks have very good S/N (thousands of counts).
• Do not apply Kα₂ stripping if you will model the Kα₁/Kα₂ doublet in refinement.

Refinement to obtain IRF parameters

In FullProf/WinPLOTR (or equivalent):

Tip: Validate by refining a second clean standard with the new .irf; it should require negligible sample broadening.

“Generic” starting values (only to help the first fit)

These are common ballparks for a lab Cu Kα Bragg–Brentano with decent optics; replace with your fitted values:

• U ≈ 0.005–0.02, V ≈ −0.005–0.002, W ≈ 0.002–0.01 (degrees²)
• X, Y ≈ 0–0.01 (Lorentzian terms, degrees)
• FCJ asymmetry: small positive values; start from S/L ≈ 0.02, H/L ≈ 0.01, then refine.

Example of a FullProf .irf (for illustration only)

⚠️ File format lines vary with profile choice; best is to let WinPLOTR/FullProf create this for you. This shows the kind of content you’ll see.

! FullProf IRF: Cu Kα, TCHZ + FCJ, Bragg–Brentano ! (Replace numbers with your fitted values; keep format from exporter) 1.540593 1.544427 0.50000 ! lambda1 lambda2 I(Kα2)/I(Kα1) 0.00020 ! Zero shift (deg) 0.01050 -0.00120 0.00580 ! U V W (deg^2) 0.00080 0.00010 ! X Y (deg) 0.0200 0.0100 ! FCJ asymmetry parameters (S/L H/L)

If your setup uses a Kα₁ monochromator (no Kα₂), set λ = 1.540593 Å and disable the doublet (ratio = 0), then refit U, V, W, etc.

If you’re using different software

• GSAS-II: refine Instrument Parameters from the standard; export/save the .instprm. (GSAS-II doesn’t use .irf, but the concept is identical.)
• TOPAS: use fundamental parameters or an .instd instrument file; no .irf needed.
• MAUD/HighScore: analogous workflow; both can export instrument profiles (or pass parameters to FullProf).