Re: optical filters

From: Howard Shapiro (hms@shapirolab.com)
Date: Sat Dec 22 2001 - 19:28:40 EST


Emanuel Gustin wrote, in response to Gib Otten's question:

> > Does anyone know if optical filters (bandpass, shortpass, longpass, etc)
>in
> > the flow cytometer fluorescence path degrade with time or with use
>(assuming
> > filters not mishandled) ?
>
>Filters in the fluorescence path are not exposed to high incident
>power, and are usually well protected. There will always be
>some accumulation of dirt on the optics, of course, depending
>on environmental factors (Smoking, powdered gloves, dust, ...).
>But the filters should last for a long time, unless you work in
>a very dirty environment or play around with corrosive chemicals.
>
> > Would the filters be more likely to degrade in an
> > epifluorescence microscope with an HBO100 light source?
>
>Certainly, at least the excitation filter and the beam splitter.
>Depending on your setup (possible presence of pre-filters)
>these may be exposed to a fair amount of heat and UV light.
>The filters are also far more exposed to dust and inaccurate
>handling.
>
>Errors in fabrication may cause optical components to
>delaminate, i.e. the thin layers of metal or dielectric coatings
>that give them their properties may peel off. It doesn't happen
>very often, but it happens.

When you've been around as long as I have, your estimate of the frequency
of delamination of filters changes.  Almost all interference filters are
subject to this, usually after years of use; you can tell by looking at or
through them, as they become blotchy when delaminated.  Dichroic beam
splitters are less prone to delamination because they typically have fewer
layers.

Poor quality interference filters may have pinholes in the coatings that
will let through light at wavelengths supposedly excluded by the filter.

Absorptive (color glass) filters don't delaminate, but you can bleach spots
in them if you put them in a laser beam.

-Howard



This archive was generated by hypermail 2b29 : Sun Jan 05 2003 - 19:01:44 EST