Technical Issues - Tooling Factor Correction

Optical monitor tooling (monitor/substrate M/S) can be calculated in MONITOR. One deposits a fairly thick film, say 15 quarter waves, on a monitor glass and subsequently measures the coated substrate in a spectrophotometer. In former times a slide rule helped to determine the number of quarter waves on the coated substrate. This technique, possibly developed at Bausch & Lomb in Rochester, has stood the test of time and is widely utilized.
 
M/S is actually the ratio of the apparent physical thickness on the optical monitor to the physical thickness on the substrate. 'Apparent' because M/S is based on the number of monitor quarter waves deposited at some nonzero angle of incidence. While calculations based on thick single layers will be fairly accurate, M/S for films in a stack  usually requires correction. We need a design which uniquely gives correction factors for two materials.

A technique known to thin film cognoscenti utilizes a half wave ripple design, say 1H 2L 2H 2L 2H 2L 1H. The method has been encapsulated in the Excel sheet shown below. Since tooling correction is a built-in DESIGN function, the circuitous Ripple Test is no longer but is nonetheless interesting and instructive. Click here to download RIPPLE2.xls. Click here for further explanation and support.

Is the half wave ripple design optimum for DESIGN tooling correction? How much can thicknesses deviate before tooling correction fails? We applied our Gedankenspektrum technique with a FilmStar BASIC program simulating tooling deviations from -30% to +30%. As it happens, the half wave ripple design is not the best candidate. In initial experiments we found 1H 1L 2H 1L 1H to be somewhat superior with correct solutions for ~20% deviation in both H and L layers.


'Optimize 1' indicates tooling factor refinement

The FilmStar BASIC program and corresponding FILM Archive file can be downloaded. As shown below, solutions are possible with fairly large thickness deviations.

Back to Technical Issues