Re: [ATM] Additional phase contrast and Foucault images

["Stephen Koehler" <s.c.koehler@gmail.com>]

Dominic,

> I have not spent much time investigating best ways to reduce this
> effect or its exact origin, but in my lab we have different phase
> contrast systems in which one may or may not have these multiple
> images, so I have not viewed this as unavoidable, albeit not hard
> to encounter. I have always associated this with the one clear
> difference between these systems, which was size and location
> of phase plate. BTW, in our systems, we have essentially
> monochromatic green light in both cases, resulting from filters.
> These are roughly optimal wavelengths for the optics in the
> system.

I did some experiments by simulation, and it looks like the double
image is affected by a combination of the frequencies filtered by the
phase plate and by the source slit width.  For these simulations, I
used a phase "knife" of .25 transmission.  It's possible that you get
different effects by using a narrow phase line, but Dale was using a
wide line, which is effectively a phase knife.

Here's the first graphic:

http://www.atmlist.net/contrib/s-dot-c-dot-koehler-at-gmail-dot-com/lyot_test/lyot_dx.png

In this one, I used a source slit of .1mm in width, and varied how
deeply the knife cut into the image (annotated in the figures).   I
put a small spot on the center of the mirror (like my center spot) to
highlight the diffraction.  As the knife is cut more deeply, higher
frequency detail is seen, as expected.  The center spot has the most
pronounced doubling at a knife cut of -.1, and grows slowly less as
the knife is cut deeper.  There's also some "ringing" at the edge that
gets finer and finer as the knife is cut deeper.

Here's the second graphic:

http://www.atmlist.net/contrib/s-dot-c-dot-koehler-at-gmail-dot-com/lyot_test/lyot_slit.png

In this one, I left the knife cut at -.1 and varied the width of the
source.   As the source becomes wider, the doubling of the center spot
goes away, but so does the ability to see higher frequency detail.

My conclusion is that you can affect the image doubling by changing
the source width or the knife position, but you may not be able to
both get rid of the doubling and get the detail you want in the phase
contrast image.  I realize that I'm making this statement with
absolutely no practical experience in making phase contrast images, so
I'm happy to be proven wrong :-)
-- 
Steve Koehler
_______________________________________________
ATM mailing list http://www.atmlist.net/