Coherence length of Hg 546 nm line

[Steve Eckhardt]

Edmund Scientific sells monochromatic light sources that consist of a mercury
lamp, a diffuser and a green narrow-band filter. Does anyone know what the
coherence length of such a device might be? (What's the spectral width of this
peak for a low pressure Hg lamp?)
--
Best regards,
Steve Eckhardt (skeck...@mmm.com)

Opinions expressed herein are my own and may not represent those of my employer.

[Bret Cannon]

Many years ago I worked in a lab that purchased at 1.5 meter monochromator
that could be double passed, and the acceptance test was to scan the green
line from a low pressure Hg lamp, resolve the different isotopic lines and
from the ratio of the measured width of a given peak to the distance between
a particular pair of the isotopic lines the resolution of the monochromator
was calculated. The minimum resolution for acceptance was 300,000. As I
recall, the full width of the hyperfine spectrum was about 10 times the
width of a single line, thus the linewidth was about 546 nm/30,000 = 0.02 nm
= 0.7 wavenumber.

I am certain that the width is determined by the isotope shifts rather than
the more common broadening mechanisms like pressure, Doppler, or
self-absorption.

Bret Cannon

Steve Eckhardt <skeck...@mmm.com.deletethis> wrote in message
news:a6jdh2$43p$1...@magnum.mmm.com...

[Andy Resnick]

Steve Eckhardt wrote:

> Edmund Scientific sells monochromatic light sources that consist of a mercury
> lamp, a diffuser and a green narrow-band filter. Does anyone know what the
> coherence length of such a device might be? (What's the spectral width of this
> peak for a low pressure Hg lamp?)

We are using a high-pressure Hg arc for microscope-based interferometry. In my
experience, the "raw" peak (<5 nm FWHM, I think) gives a coherence length less than
0.5 mm. Which is good, because we don't get secondary fringes from the microsope
optics themselves...

Don't know what the difference would be between the high- and low-pressure arcs. I
don't think the high pressure arc is high enough to cause pressure broadening.

If the filter is narrower than the Hg line, you get a proportional increase in the
coherence length at a decrease in output intensity.

I think the main use for the above instrument is for Newton's rings or Fizeau-type
fringes between two glass surfaces. As long as the space is small (and it usually
is for flats and large curvature lenses) it's great - some guys in the next lab over
have one.

--
Andy Resnick, Ph.D.
Optical Physicist
Northrop Grumman