Re: [SPAM] Fwd: Re: Importance of the tube lens NA

Posted by James Pawley on
URL: http://confocal-microscopy-list.275.s1.nabble.com/Fwd-Re-Importance-of-the-tube-lens-NA-tp2436781p2438527.html

Hi all,

We have all seen ray diagrams like.

http://www.antonine-education.co.uk/New_items/MUS/images/ray_diag_6.gif

Light from a point on the axis and in the focal plane diverges until some of it hits the (perfect) lens and emerges as a parallel ray bundle on the other side. However, things become a little more complex when the focal-plane emitter is moved off the axis, as in

http://www.phg.ulg.ac.be/Hololab/fichiers/Graphic1.jpg

Now the emergent parallel ray bundle is at an angle to the axis. So you would think that, as long at the next element in the optical system (the tube lens) was sufficiently large enough (or close enough) to collect the angled ray bundle created by light emerging from the focus plane at the edge of the field of view, then all the light from the field of view would be focused into the Intermediate Image. Then it would pass through the eyepiece/camera-coupler and end up in the recorded image.

But what about light emerging from above or below the focal plane? Although I couldn't find an on-line sketch, it is clear that instead of producing a parallel emerging ray bundle, it will produce one that either diverges or converges (resp). So some of these rays, particularly those from points both off axis and out of focus, will leave the objective at angles even larger than those from the edge of the field of view at the plane of focus. How much more will depend on how far off axis and away from the focal plane they are, but when one focuses tens of micrometers into a thick, fluorescent specimen, it can be quite a large angle.

As we want to place reasonable limits on the diameter of the black metal tube between the objective and the tube lens, (and on the diameter of the tube lens itself), then compromises must be made and some of the light from out-of-focus, off-axis sources will strike the wall and be lost. The general term for this loss is "Vignetting".

I know nothing about the actual changes that Olympus has apparently incorporated in their LV200 except what I found at

http://www.olympus-europa.com/corporate/1696_1948.htm

However, as I remember it, bioluminescence has a very low signal and the specimens are often quite 3-dimensional. To the extent that this is true and that a high NA objective is used to collect the signal, much of this signal may be out of focus. So I guess that having a tube lens with a large diameter will collect more of this light . Of course, it won't focus this extra light very well (the out of focus sources will still look like pale, diffuse blobs. You can't get around this.) but the total number of photons in the image will be larger with this large tube lens than with a "normal" one.

Two last notes:

1) If the tube lens has a larger diameter, the cone of in-focus-light that converges from it to focus at the image plane will have a larger half angle, and hence a higher NA. However, unless the magnification of the entire system is very low indeed (i.e. using a 1-3x infinity objective), then this tube NA has no affect on the sharpness of the final image (unless the large NA tube lens is not properly corrected to produce near-diffraction-limited performance at this NA: in which case the final image may be less sharp).

2) To some extent, all microscopes have this limitation (off-axis, out-of-focus light is preferentially lost) but it will be particularly severe when using high-NA objectives of lower magnification because these have larger fields of view. Indeed, vignetting and other limitations on the performance of objectives are discussed (with better figures!) in Chapters 7 and 11 of the Handbook of Biological Confocal Microscopy.

Cheers,

Jim Pawley

-- 

              **********************************************
Prof. James B. Pawley,                                          Ph.  608-263-3147 
Room 223, Zoology Research Building,                                  FAX  [phone]
1117 Johnson Ave., Madison, WI, 53706                                [hidden email]

3D Microscopy of Living Cells Course, June 13-25, 2009, UBC, Vancouver Canada
Info: http://www.3dcourse.ubc.ca/       Applications due by March 15, 2009
       "If it ain't diffraction, it must be statistics." Anon.