Search the CONFOCAL archive at
http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal

>Search the CONFOCAL archive at 
>http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal
>
>yes, there are all kinds of led,different wavelength, 
>375nm,405nm,460nm,585nm,....and the optical power is high, maybe up 
>to 100mw. I have ever bought different leds, for a 5w blue 
>led(460nm),the optical power is about 20mw(it is easy to adjust the 
>power by changing the current or voltage ),the price is about 14 RMB 
>in china (about 2 dollar).Compare with laser, the biggest problem is 
>led's big emission angle.So it is hard to couple light  to fiber or 
>focus into one point.I tried to do it, but most of the energy lost. 
>I think the led will replace the arc lamp in future, even the laser 
>for confocal or laser scanning microscopy.



Hi all,

I think that LED light sources are a very important development in 
fluorescence microscopy but the post above is the first one to really 
talk about the optics of the system and the limits that this places 
on the excitation intensity.

As LEDs are NOT as bright as arcs, the optics used to convey their 
output to the focus plane become very important.

Although this topic is covered in some detail in Chapters 6 and 10 of 
the Handbook, the main points are:

1. The intensity at the specimen can never be higher than that of the 
source and is usually much less (It depends on the NA of the 
collector and objective lenses, the total magnification of the image 
of the source, and secondarily on reflection losses and the fact that 
few sources are planar.).

2. If you want intensity uniformity, you use Kohler illumination, but 
you can get more brightness if you focus the source onto the part of 
the specimen that you wish to illuminate ("critical illumination") 
and this will work well if the source brightness is uniform. LEDs 
seem to be relatively uniform in brightness especially if their 
output has been scrambled by passage through an optical fiber.

3. As the brightest image of the source, will occur when the image of 
the source is the same size as the source (Assuming a fixed NA of the 
collector and objective lenses. You can change the total mag of the 
optics to make the image of the source larger or smaller, but this 
necessarily changes the relative EFFECTIVE NAs of the collector and 
objective lenses.)

4.  If you are working at high imaging magnification, (i.e., small 
field, say 150 micrometers diam) then the source need only be uniform 
over a similar area (assuming that all the optics including the 
objective give 1:1 imaging, not common, but used here just as an 
example).

5. A corollary here is that, if LEDs with higher power are also 
larger, the increased number of photons they produce won't help you 
because you cannot force them to fit into a fixed field of view on 
the specimen. They have more photons, but not more (and perhaps less) 
photons/micrometer squared

6. If you are working at lower imaging magnification, then your field 
of view will be larger and you will able to focus more of the light 
from a larger LED onto the part of the specimen that you are imaging. 
This is why normal WF fluorescence can sometimes look much brighter 
with lower mag objectives.

Since the Handbook came out a number of companies have jumped into 
the field. Choosing among them will be  greatly assisted if they can 
tell us something of the optical system they employ (Kohler or 
critical?, will both aperture and field diaphragms operate properly 
etc.), and the number of mW of light, with a given spectral 
distribution, that they can deliver to a circle of the specimen, say, 
100 micrometers in diameter (The size of this circle should be 
adjustable using the epi-illumination field diaphragm, and is 
important because with a source of given brightness, it should be 
possible to put 4x more photons through a 200 micrometer circle than 
through one of 100 micrometers. This will be true as long as the size 
of the illuminated field does not exceed the field of view of the 
objective lens.)

Please help us all by asking the manufacturers for this information.

Two other points: Because the spectral distribution of the light 
emitted by LEDs has tails on it, one usually still needs to use 
excitation filters unless the Stokes shift of the dye is large. As 
white LEDs are really only blue LEDs covered with a phosphor that 
converts some of the blue light into red and green light, they are 
intrinsically less "bright" in a given waveband than "colored" LEDs.

Cheers,

Jim P;.

-- 
               **********************************************
Prof. James B. Pawley,               		            Ph.  608-263-3147 
Room 223, Zoology Research Building,               
FAX  608-265-5315
1117 Johnson Ave., Madison, WI, 53706  
[log in to unmask]
3D Microscopy of Living Cells Course, June 14-26, 2008, UBC, Vancouver Canada
Info: http://www.3dcourse.ubc.ca/	     Applications due by March 15, 2008
	       "If it ain't diffraction, it must be statistics." Anon.