Technical Issues - Designs on Multiple Substrates II

Continuation of Design on Multiple Substrates I

Consider a design with 0.02% transmittance. The same stack is deposited on the second parallel surface of the nonabsorbing substrate. The total transmittance is 0.01% instead of 0.0004%. This apparent anomaly is caused by multiple internal reflections. One way to increase blocking is to delete parallel from the above description. While the example below refers to coatings on a single substrate, it applies to assemblies with multiple substrates.

DESIGN 2.61.3101 adds new Side 2 Compensation mode: FWD Wedge. The mode applies only to designs with massive layers and is exactly the same as Mode 1 (FWD ignore Side 2) otherwise.

By eliminating reflections from secondary surfaces, a wedged substrate increases blocking of filters containing coatings on multiple surfaces. While thin film manufacturers can achieve similar improvement with a slightly absorbing substrate, that is not always possible.

Wedge mode assumes there are no reflections between massive surfaces. Reflectance is the reflectance of the first surface only while transmittance is the product of each element. The wedge angle must be sufficient for all reflections to walk off the assembly.

As example we consider the IR bandpass Fnum.faw design (included in the FilmStar installation) centered at 1640 nm. A second (Side 2) design is centered at 1630 nm. We set k = 0.0001 for both L and H materials.

Wedge mode is equivalent to multiplying two separate filters in FWD ignore Side 2 mode. Reflectance is Side 1 reflectance (R1) while transmittance is given by T1*T2. The Side 2 calculation sets substrate thickness to zero; otherwise substrate absorption would be counted twice. Example: Download Fnum-Diab.faw (Side 1), Fnum-1630.faw (Side 2) and run BASIC program TwoSideWedge.bas. Finally open Fnum-Diab-1630-1640.faw and evaluate in Side 2 mode 1 (Parallel) and mode 6 (Wedge). As seen in the plot below, Both Sides and Wedge traces overlap as expected.

TwoSideWedge.bas provides an alternative for users who do not need to simultaneously design coatings on both sides. Want to test absorption in the substrate (massive layer)? Set SUB k = 0.00001 in Fnum-Diab.faw and Fnum-1630.faw. Set massive material G k = 0.00001 in Fnum-Diab-1630-1640.faw and regenerate the curves shown below. Not sure what a massive layer is? No idea how to run a BASIC program? Contact FTG Software for an explanation using our Glance Networks facility. When substrate k>0 the BASIC program requires DESIGN 2.61.3058 or newer. (Contact FTG for a workaround.)

The green Both Sides plot is covered by the teal Wedge plot. Using FSPlot zoom (see below) we find that blocking is improved by a factor ~200. If you try the suggested exercise (substrate k = 0.00001) you will note much less improvement because blocking is already increased by the absorbing substrate.