Shirley vs Tougaard background in XPS

The background you subtract directly determines your peak areas, so the choice matters as much as the peaks themselves. PikaXPS offers Linear, Shirley, Shirley + Linear, and Tougaard (U2), plus an active Shirley that co-fits with the peaks.

Linear

A straight line between the two endpoints. Use it for weak peaks on a nearly flat background (C 1s, N 1s, O 1s) and narrow windows. Avoid it for transition-metal 2p regions, where the step across the peak is large.

Shirley

The standard choice for most metal / transition-metal regions. The background rises toward higher binding energy in proportion to the integrated peak area above it, computed iteratively. It’s sensitive to where you place the endpoints — keep them on flat regions well away from the peak tails, and nudge them ±1 eV to check how much the area moves.

Active Shirley

Derives the background from the peak model instead of the raw data, and optimises it together with the peaks. Less sensitive to noise and to other structure inside the window — switch it on when a plain Shirley looks wrong.

Tougaard (U2)

The most physically rigorous option: it integrates the universal inelastic-loss cross-section. Use it when absolute quantification matters and you measured a wide window (30+ eV past the peak). It’s inaccurate over narrow windows, where Shirley is usually enough.

Practical summary

Situation	Recommended
Transition-metal 2p (Ni, Co, Fe…)	Shirley (or active Shirley)
C 1s, N 1s, O 1s	Shirley or Linear
Absolute quantification, wide window	Tougaard
Background looks wrong under plain Shirley	Active Shirley

Always report which background and energy range you used — changing it changes your areas.

Download PikaXPS → and try Fit BG to settle the background before adding peaks.