Effect of native oxide on thin flim oxygen analyses

[Probeman]

A colleague recently asked me to quantify the effect of the native oxide layer on silicon when analyzing for oxygen in a thin film using multi-voltage EPMA.

Obviously this effect is quite dependent on the thin film thickness, composition and the physics details, but to a first order the effect is due to the thickness of the native oxide layer, for a given thin film layer thickness.  Obviously the effect is larger as the thin film thickness decreases.

Most silicon has a native oxide layer thickness of around 2 to 3 nm which increases over months at room temperature, while a freshly etched silicon wafer can have a very thin oxide layer for a few hours or days depending on the relative humidity. The following screen shots are assuming native oxide layers for 0.01 nm, 1.0 nm, 2 nm and 3 nm for a 61 nm thick Bi, Dy, Fe oxide film.

So here is a compositional calculation assuming a native oxide layer of only 0.01 nm which is essentially a zero thickness:



And here are the 1, 2 and 3 nm thickness calculations.







Please let me know if you have any questions.
john