Natural Science Forum / Physics / Research / May 2005

If you know the contained energy (e.g., calorimetry) and the (weighted
distribution of) frequency then you know the number of photons.

I have come up with a semi-classical heuristic that should give you a
ballpark figure on the number of photons.  The "static" electric field
is,

<E_o> = 2pi*sqrt(hbar*c*n)/lambda^2 in gaussian cgs units.

Where n is the number of photons and lambda is the wavelength of the EM
radiation.

FrediFizzx

Your power is going to be Watts/time.  Since E=hf, you can determine
the energy of a single photon if you know its frequency.  So h*f*(#
photons) will give you the total power for a given number of photons.
Set this equal to Watts/time.  Thus the number of photons = Power/h*f

There is a very good paper on this in Ann. Phys. 17 on page 132.

Or, if by power you mean flux, integrate over the surface area to get
the actual power.

Herbie

This is an extremely interesting article. It states:
"A SINGLE-PHOTON TURNSTILE, a device in which photons are
emitted one at a time under controlled circumstances, has been
created by a team of scientists from Stanford (US), Hamamatsu
Photonics (Japan), and NTT (Japan).  Essentially the researchers use
the quantization of electrical conductance to produce a quantization
of photon emission.  They put together a quantum well (the frontier
between two thin semiconductor layers) containing a single electron
(other electrons are dissuaded from entering because of a "Coulomb
blockade" effect) with a quantum well containing a lone
(comparably Coulomb blockaded) hole, and then cycle the voltage
across the whole stack of layers in such a way that the lone electron
and lone hole meet, mate, and make a lone photon.  The resulting
device, which operates at mK temperatures,  is typically a tiny post
some 700 nm tall and with a diameter of 200-1000 nm.  (J. Kim et
al., Nature, 11 February 1999.)"

I don't think that it is actually about the length of the wave train of
a single photon. It is about the ability to release a single photon at a
time.

This is a really useful trick, because it allows the possibility of
making a new version of the Hanbury Brown and Twiss experiment which I
take to prove that there is no such thing as a "photon in flight"
because they showed that every photon can potentially be detected more
than once, in fact the mean is once but with a poisson distribution.
However QM was altered to state that photons are emitted in a poisson
distribution to get around the embarrassment.

I would really love to see this device used with the Hanbury Brown and
Twiss experiment to see whether it still gets a poisson distribution. I
say that it would, and that the result proves there is no "photon in
flight", just the e/m field developing continuously according to
Maxwell's equations.