Photon detection efficiency of common PMTs

[Michael Giacomelli-2]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Hi,

I've seen warnings about confusing the photon detection efficiency
(PDE) of PMTs with the quantum efficiency (QE) of the cathode, but
aside to some references to the effects of electrons lost after the
photocathode I have not seen any PDE numbers for commonly used PMTs
like the H7422, H10770, etc.  Digging through the Hamamatsu datasheets
likewise gives the cathode sensitivity and the anode sensitivity, but
not the exact gain of the device being measured.

Are there any papers out there characterizing PDE of imaging PMTs?
I'm curious exactly how many photons actually contribute to signal in
a PMT-based microscope.

Mike

[Zdenek Svindrych-2]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Hi Mike,

very interesting question, indeed. After digging some old papers (such as
this one https://sci-hub.tw/10.1364/AO.8.001975 from 1961, I only looked at
"Summary" at page 1984) it seems there is a number of contributions on top
of the photocathode quantum efficiency. First, not all photoelectrons cause
amplification on the first dynode (50 % loss? mentioned in the paper), then
there is the excess noise of the amplification process similar to excess
noise in EMCCD cameras (I remember something like 20% larger variance of
the output than what would correspond to ideal poissonian statistics), and
at last (probably a variation on the previous point), there is some loss
during pulse discrimination... Altogether 1% counting efficiency for the
S20 photocathodes.

But that was 1961. Now, looking at the Hamamatsu PMTs (
https://www.hamamatsu.com/resources/pdf/etd/PMTmodules_TPMO0011E.pdf ) the*
Count Sensitivity* of H7421 is 10^6 s^-1pW^-1 (1 MHz/pW in more sensible
units) at 600 nm (corresponding to 0.33 x 10^-6 pJ per photon according to
this site: http://www.calctool.org/CALC/other/converters/e_of_photon ). So
there are 3x10^6 photons per second in a picoWatt. On average, the GaAsP
PMT module fires on every third photon, that gives you overall quantum
counting efficiency of 33%. Not bad at all! (hope I did not make a mistake
of couple orders of magnitude :-)
(Interestingly, in some old preliminary datasheets both the Count
Sensitivity and Dark Counts were lower by a factor of 4/5, probably just
different discriminator threshold?)

Just a note, of course, then there are pulses that do not correspond to any
photons, apart from the dark counts those are due to afterpulsing (could
not find any specific numbers, Becker and Hickl mention in one of their
white papers the afterpulsing probability is "high" in H7422 at maximum
gain; in the old papers, 5% afterpulsing probability was reported).

Hope this helps!

Best, zdenek

----------
Zdenek Svindrych, Ph.D.
Research Associate - Imaging Specialist
Department of Biochemistry and Cell Biology
Geisel School of Medicine at Dartmouth


On Sat, Mar 30, 2019 at 8:10 PM Michael Giacomelli <
[hidden email]> wrote:

> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> Post images on http://www.imgur.com and include the link in your posting.
> *****
>
> Hi,
>
> I've seen warnings about confusing the photon detection efficiency
> (PDE) of PMTs with the quantum efficiency (QE) of the cathode, but
> aside to some references to the effects of electrons lost after the
> photocathode I have not seen any PDE numbers for commonly used PMTs
> like the H7422, H10770, etc.  Digging through the Hamamatsu datasheets
> likewise gives the cathode sensitivity and the anode sensitivity, but
> not the exact gain of the device being measured.
>
> Are there any papers out there characterizing PDE of imaging PMTs?
> I'm curious exactly how many photons actually contribute to signal in
> a PMT-based microscope.
>
> Mike
>

--
--
Zdenek Svindrych, Ph.D.
Research Associate - Imaging Specialist
Department of Biochemistry and Cell Biology
Geisel School of Medicine at Dartmouth