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

Hi Professor Xi,

I think that the main reason APDs are used in that application is that
they are fast detectors, not affected by the potential high ambient
light exposure and there are plenty of photons available so sensitivity
is less an issue.  Below I am pasting in some text from a Hamamatsu paper:
http://sales.hamamatsu.com/assets/applications/Combined/lightlevelnoise.pdf

I don't know if this is any help. I think that ~0.5A/W is all that 
silicon is capable of yielding, and then it is a matter of signal, 
speed, or noise level, or damage by high ambient light.

But I have to admit I do not work with this in detail
and there may be other issues.


Dale


The avalanche photodiode (Figure 3) has features of both the photodiode
and the photomultiplier tube. It is a solid-state device that generates
electron-hole pairs upon exposure to light. A reverse bias of 100 to 
1000 V is placed on the PN junction, creating an internal electric field 
large enough to accelerate the electrons. When the electrons collide 
with the crystal lattice, they generate electron-hole pairs. This 
process cascades, amplifying the initial signal by a factor of 50 or more.

The statistical fluctuations in the number of collisions
and the yield of electron-hole pairs create more noise than
the dynodes in a photomultiplier tube but less noise than
a photodiode’s external amplifier. Thus, avalanche photodiodes
are often used when light levels are too high for photomultiplier
tubes but not high enough for photodiodes.

Avalanche photodiodes are inadequate for some applications,
such as those in which the lifetime of the photon source is short, a
measurement must be made rapidly, few photons arrive at the detector or 
the photons are spread over a large area. These situations will usually 
require a photomultiplier tube.

Peng Xi wrote:
> Search the CONFOCAL archive at 
> http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal
> 
> Hi Dale, Yes, I contacted them in the first place, as they have the
> best PMT (in my opinion). But the APD they recommend me is really not
> good (C5460) -- 2-3 times weaker signal than the PMT. I think APD
> should perform better than PMT in 800nm region, and that's why many
> reflective confocal setup uses APD (well, also APD should be
> cheaper).
> 
> Best regards, Peng Xi Associate Professor Institute for Laser
> Medicine and Biophotonics Shanghai Jiao Tong University 800 Dongchuan
> Rd. Shanghai 200240, China Tel: (86) 21-3420-4076 
> http://biophotonics.sjtu.edu.cn/
> 
> 
> 
> Dale Callaham wrote:
>> Search the CONFOCAL archive at 
>> http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal
>> 
>> Have a look at hamamatsu? I'm sure they are high quality. 
>> http://sales.hamamatsu.com/en/products/solid-state-division/si-photodiode-series/si-apd.php
>> 
>> 
>> 
>> I have no connection with Hamamatsu.
>> 
>> Dale
>> 
>> Peng Xi wrote:
>>> Search the CONFOCAL archive at 
>>> http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal
>>> 
>>> Hi,
>>> 
>>> Anybody know which company produces high-quality  Avalanche 
>>> photodiode (APD)? I am planning to use it to detect 800nm signal.
>>> I prefer high sensitivity, low noise, analog one used in confocal
>>>  reflective devices. Thank you for your advice!
>>> 
>>> 
>>> 
>>> Best regards, Peng Xi Associate Professor Institute for Laser
>>> Medicine and Biophotonics Shanghai Jiao Tong University 800
>>> Dongchuan Rd. Shanghai 200240, China Tel: (86) 21-3420-4076 
>>> http://biophotonics.sjtu.edu.cn/
>>> 
>>> 
>>>