phase consistency of photons in a stream
Neil
because of uncertainty issues. (Indeed, for "a photon" it is hard to imagine
just what "absolute phase" would mean.) However, does anything forbid a
stream of single photons (say one every ms) from having rather consistent
relative phases in some sense? (I mean, each photon is somehow no more than
some fraction of radian of effective phase different from its neighbors,
each of which is beyond the average coherence length for single photons.)
After all, the parts of split wave functions of single photons can interfere
with each other, or even attenuated beams from separate lasers containing
only one or two photons on average at a time. Maybe we could test such a
stream by pulling out every other photon, then use appropriate optical path
length adjustments to get the separated streams to interfere. (A very fast
electro-optical mirror to divert every other photon off to the side, to be
recombined with what's left.) How much theory/experiment has been directed
to this issue?
Repeating Rifle
<neil_...@hotmail.com> wrote:
I doubt that you know what you are talking about. Please explain to me how
you can tell (measure) the phase you are talking about. Then we can talk.
Bill
Louis Slotin
"Repeating Rifle" <salm...@sbcglobal.net> wrote in message
news:BEFC909E.266%salm...@sbcglobal.net...
Stream then through a mag field, would phase be same on all?
Ron Baker, Pluralitas!
"Neil" <neil_...@hotmail.com> wrote in message
news:11de6ah...@corp.supernews.com...
> We can't arrange for a photon to be emitted with a definite absolute phase
> because of uncertainty issues. (Indeed, for "a photon" it is hard to
> imagine
> just what "absolute phase" would mean.) However, does anything forbid a
> stream of single photons (say one every ms) from having rather consistent
> relative phases in some sense?
No.
> (I mean, each photon is somehow no more than
> some fraction of radian of effective phase different from its neighbors,
> each of which is beyond the average coherence length for single photons.)
Coherent but beyond the coherence length?
Then you've misunderstood coherence length.
> After all, the parts of split wave functions of single photons can
> interfere
> with each other, or even attenuated beams from separate lasers containing
> only one or two photons on average at a time. Maybe we could test such a
Test it for what?
> stream by pulling out every other photon, then use appropriate optical
> path
> length adjustments to get the separated streams to interfere. (A very fast
> electro-optical mirror to divert every other photon off to the side, to be
> recombined with what's left.) How much theory/experiment has been directed
> to this issue?
Lots.
--
rb
Josef Matz
"Neil" <neil_...@hotmail.com> schrieb im Newsbeitrag
news:11de6ah...@corp.supernews.com...
Good question.
Linear polarized light, elliptical and circular light have a defined phase.
Only thermally emitted light contains two components which have an
independent phase.
The absolute phase is not measureable. That means the value is not
determinable due to a measurement because of the uncertainty principle which
gives Delta Phi bigger or equal 2 pi. But we know it exists.
In thermal radiation there exist two phases. One for the left turning and
one for the right turning photon streams. Both not measureable. All
measureable quantities of light therfore must be independent on phase.
Therefore the cos and sin functions of a plane wave, which describe the
rotation of the EM field can not be viewed. If you modulate the amlitude
with a sin and cos - function you have something which is independent with
the sin and cos oscillations of the plane wave factor in the background and
therfore this phases can be measured. Even specialists in signaling do not
know this difference. This makes physics so hard to understand.
Hope i helped a little.
Josef jose...@arcor.de
John Sefton
Josef Matz wrote:
>
> Linear polarized light, elliptical and circular light have a defined phase.
> Only thermally emitted light contains two components which have an
> independent phase.
>
> The absolute phase is not measureable. That means the value is not
> determinable due to a measurement because of the uncertainty principle which
> gives Delta Phi bigger or equal 2 pi. But we know it exists.
>
> In thermal radiation there exist two phases. One for the left turning and
> one for the right turning photon streams. Both not measureable. All
> measureable quantities of light therfore must be independent on phase.
>
> Therefore the cos and sin functions of a plane wave, which describe the
> rotation of the EM field can not be viewed. If you modulate the amlitude
> with a sin and cos - function you have something which is independent with
> the sin and cos oscillations of the plane wave factor in the background and
> therfore this phases can be measured. Even specialists in signaling do not
> know this difference. This makes physics so hard to understand.
>
> Hope i helped a little.
>
> Josef jose...@arcor.de
>
>
Hi Josef'
What emits the linear polarized light, elliptical and circular light?
What does thermally-emitted light include?
John
Josef Matz
"John Sefton" <veg...@accesscomm.ca> schrieb im Newsbeitrag
news:42d7dcdb$1...@news.accesscomm.ca...
I think a dipole antenna radiates linear polarized EM waves. Interesting
would be the question what a laser emitts. But this depends probably on the
design. I do not know if laserlight is like thermal emitted light.
Interesting question.
> What does thermally-emitted light include?
Now if you analyze light from the sun or a lamp at a certain frequency with
a linear polarizer you get the same intensity for each angle of your linear
polarizer. But the same you would get if the sun light would be circular
polarized. But sun light is not simple circular polarized light. If you take
a circular polarizer which lets throught left circular polarized light you
get approx. half the intensity. But the same for a circular polarizer which
lets through right turning light.
If sun light would be circular polarized light, you would get 100 %
transmission in the one case and 0% transmission in the other case.
Therefore light from the sun or a lamp contains two statistical independent
components, which transport photons with spin +1 and -1.
Thermal emission from non black grey or white surfaces, for example a
metallic mirror contains two elliptic polarized independent components at
angles not normal to the surface.
Thats what i know. Maybe other contributions ?
> John
>
Andy Resnick
<snip?
>
> Good question.
>
> Linear polarized light, elliptical and circular light have a defined phase.
> Only thermally emitted light contains two components which have an
> independent phase.
That's not exactly true. At any instant in time, at any point in space,
the electric field has a precise singly-valued [*] phase. How the
phase changes in time and space as compared to other points in time and
space defines the coherence of the field. This is why I prefer the term
"randomly polarized" to "unpolarized". Sunlight (thermal emission) has
a short coherence time and transverse length. However, if the light is
chromatically filtered, one may observe speckle- the transverse
coherence length at 555 nm is around 1 mm or so.
>
> The absolute phase is not measureable. That means the value is not
> determinable due to a measurement because of the uncertainty principle which
> gives Delta Phi bigger or equal 2 pi. But we know it exists.
Also not exactly true- the absolute phase (mod 2*pi) is routinely
measured in radio-wave and lower frequencies, AFAIK.
<snip>
[*] this is not true for zero-amplitude singularities of the field,
which do physically exist and can be created in the lab.
--
Andrew Resnick, Ph.D.
Department of Physiology and Biophysics
Case Western Reserve University
Josef Matz
"Andy Resnick" <andy.r...@op.case.edu> schrieb im Newsbeitrag
news:db8p2k$crj$1...@eeyore.INS.cwru.edu...
> Josef Matz wrote:
>
> <snip?
> >
> > Good question.
> >
> > Linear polarized light, elliptical and circular light have a defined
phase.
> > Only thermally emitted light contains two components which have an
> > independent phase.
>
> That's not exactly true. At any instant in time, at any point in space,
> the electric field has a precise singly-valued [*] phase. How the
> phase changes in time and space as compared to other points in time and
> space defines the coherence of the field. This is why I prefer the term
> "randomly polarized" to "unpolarized". Sunlight (thermal emission) has
> a short coherence time and transverse length. However, if the light is
> chromatically filtered, one may observe speckle- the transverse
> coherence length at 555 nm is around 1 mm or so.
>
Yes i know this picture. But during the so called coherence lenght or time
which type of polarisation has thermal radiation, any possible - i doubt
this ? You also have to conider that all points on the suns surface
contribute. The roomly interference of all of these points delivers the
speckle pattern at a certain frequency.
So the coherence picture is in my opinion not fully sufficient. The half of
the emission of the atoms would be blocked during the coherence time and
resulting from this during the complete time. If the coherence model would
be the only truth bodies would radiate only the half of thermal energy than
we have it.
Therefore i really think that we have two statistical independent fields
radiated from bodies. In the upmost cases (diffuse surfaces which emit)
therefore you have also only an overlying right and left circular
polarization during the cohärence time. Thats my opinion. Only if this
modified picture is true you get the full thermal emission as it is
observed.
> >
> > The absolute phase is not measureable. That means the value is not
> > determinable due to a measurement because of the uncertainty principle
which
> > gives Delta Phi bigger or equal 2 pi. But we know it exists.
>
> Also not exactly true- the absolute phase (mod 2*pi) is routinely
> measured in radio-wave and lower frequencies, AFAIK.
>
I do not doubt that a pulsars rotation phase is measureable. But its a
modulated amplitudes phase you measure or ? It is not the phase of the
rotation of the the electric field, i assume.
> <snip>
>
> [*] this is not true for zero-amplitude singularities of the field,
> which do physically exist and can be created in the lab.
>
ok that might be, i do see what you mean with zero - amplitude singularity
of a field - weak field ? Is that universal phase smoothly changing with
frequency ?
If it is really measureable, what does quantum mechanics specialists say ?
Is Heisenbergs uncertainty principle violated in this case ?
Andy Resnick
<snip>
> Yes i know this picture. But during the so called coherence lenght or time
> which type of polarisation has thermal radiation, any possible - i doubt
> this ? You also have to conider that all points on the suns surface
> contribute. The roomly interference of all of these points delivers the
> speckle pattern at a certain frequency.
> So the coherence picture is in my opinion not fully sufficient. The half of
> the emission of the atoms would be blocked during the coherence time and
> resulting from this during the complete time. If the coherence model would
> be the only truth bodies would radiate only the half of thermal energy than
> we have it.
>
> Therefore i really think that we have two statistical independent fields
> radiated from bodies. In the upmost cases (diffuse surfaces which emit)
> therefore you have also only an overlying right and left circular
> polarization during the cohärence time. Thats my opinion. Only if this
> modified picture is true you get the full thermal emission as it is
> observed.
I'm sorry to say I am unable to understand any of this. Maybe it's just
the grammar. But, I can say this much- radiometry and the equation of
radiative energy transfer, the foundation of optics, is inherently
inconsistent, as shown by Marchand and Wolf in the 1960's. Applying
coherence theory (the mutual coherence function, based on the so-called
ambiguity function and the Wigner distribution) to the problem brought a
resolution. Coherence theory is at the foundation of optics, and a
proper understanding is required.
<snip>
>
> I do not doubt that a pulsars rotation phase is measureable. But its a
> modulated amplitudes phase you measure or ? It is not the phase of the
> rotation of the the electric field, i assume.
I'm not talking about radio astronomy, I mean the direct coherent
detection of radio (MHz-GHz) frequencies. And even lower- kHz, or the
60 Hz coming out of your wall socket (maybe it's 50 Hz by you....).
>
>><snip>
>>
>>[*] this is not true for zero-amplitude singularities of the field,
>>which do physically exist and can be created in the lab.
>>
>
>
> ok that might be, i do see what you mean with zero - amplitude singularity
> of a field - weak field ? Is that universal phase smoothly changing with
> frequency ?
> If it is really measureable, what does quantum mechanics specialists say ?
> Is Heisenbergs uncertainty principle violated in this case ?
Zero-amplitude regions of the electromagnetic field are quite
interesting and subtle, and I recommend checking out the work by J. F.
Nye and Sir Michael Berry (of Berry's phase). It is possible to have a
region of space where the amplitude of the field is precisely zero. In
this case, the phase is completely undetermined, and can be multivalued.
"optical vortices" and "Bessel beams" are some of the buzzwords.
Josef Matz
My question is: Within coherence time you have polarisations. You have
resolved the coherence time you said.
Is during the coherence time a situation, where nearly linear polarized
effects can be measured or not ? Thermal emission is dependent on the degee
of freedom of movement. This is really a important question. If it goes out
as i think there are two independent electromagnetic fields overlapping at
every point, not only one.
Now please: Who can measure the polarisations and absolute phase constant
during coherence time ? Is there a place in the USA ? Has Wigner measured it
?
Ok i discuss a lot and english is not my mother languange. So please take
grammar and typing errors not so serious, and i dont want to look into the
dictionary so often. If you take it maybe like witten texanian slang that
would be fine - even if its Joes german english.
> <snip>
> >
> > I do not doubt that a pulsars rotation phase is measureable. But its a
> > modulated amplitudes phase you measure or ? It is not the phase of the
> > rotation of the the electric field, i assume.
>
> I'm not talking about radio astronomy, I mean the direct coherent
> detection of radio (MHz-GHz) frequencies. And even lower- kHz, or the
> 60 Hz coming out of your wall socket (maybe it's 50 Hz by you....).
>
Good question free 60 Hz radiation - is that the frequency of the photons
which take part or is it the time varying amplitude of the current ? So your
60 Hz in USA - are it just 60 Hz photons or a continuous photon spectrum and
moving atoms ? Have you ever measured a photon with energy h(bar) times 2 Pi
60 Hz ? Free currents radiate cherenkow radiation - is it quantized like the
atomic emission ?
> >
> >><snip>
> >>
> >>[*] this is not true for zero-amplitude singularities of the field,
> >>which do physically exist and can be created in the lab.
> >>
> >
> >
> > ok that might be, i do see what you mean with zero - amplitude
singularity
> > of a field - weak field ? Is that universal phase smoothly changing with
> > frequency ?
> > If it is really measureable, what does quantum mechanics specialists say
?
> > Is Heisenbergs uncertainty principle violated in this case ?
>
> Zero-amplitude regions of the electromagnetic field are quite
> interesting and subtle, and I recommend checking out the work by J. F.
> Nye and Sir Michael Berry (of Berry's phase). It is possible to have a
> region of space where the amplitude of the field is precisely zero. In
> this case, the phase is completely undetermined, and can be multivalued.
> "optical vortices" and "Bessel beams" are some of the buzzwords.
>
Ok i heard of it. But that is only experimental. No theory for or am i wrong
?
Autymn D. C.
Autymn D. C.
Neil
"Ron Baker, Pluralitas!" <sto...@bellsouth.net.pa> wrote in message
news:sMHBe.7517$rF5....@tornado.socal.rr.com...
>
> "Neil" <neil_...@hotmail.com> wrote in message
> news:11de6ah...@corp.supernews.com...
> > We can't arrange for a photon to be emitted with a definite absolute
phase
> > because of uncertainty issues. (Indeed, for "a photon" it is hard to
> > imagine
> > just what "absolute phase" would mean.) However, does anything forbid a
> > stream of single photons (say one every ms) from having rather
consistent
> > relative phases in some sense?
>
> No.
>
> > (I mean, each photon is somehow no more than
> > some fraction of radian of effective phase different from its neighbors,
> > each of which is beyond the average coherence length for single
photons.)
>
> Coherent but beyond the coherence length?
> Then you've misunderstood coherence length.
>
No, I don't mean that the "beam" is coherent, I am just talking about the
photons being farther apart than the length of the wave packet created by
the energy spread. (Consider the characteristic Delta U for each photon
derived from the emission time, say the lifetime of an excited atomic state.
From Delta U * Delta t >= hbar/2 we get a characteristic spread of energy,
and thus Fourier spectrum creating a wave which is bounded and has therefore
a length. Effectively, the coherence length is the distance light travels
during that characteristic time. It can be measured on average at least for
sporadic photons with the usual interferometric split with recombination of
unaltered with retarded paths. I use U to avoid conflict with E for electric
field.)
> > After all, the parts of split wave functions of single photons can
> > interfere
> > with each other, or even attenuated beams from separate lasers
containing
> > only one or two photons on average at a time. Maybe we could test such a
>
> Test it for what?
>
Test for the relative phases being orderly instead of random. I mean, each
photon is on average 300 +/- .05 wavelengths apart.
> > stream by pulling out every other photon, then use appropriate optical
> > path
> > length adjustments to get the separated streams to interfere. (A very
fast
> > electro-optical mirror to divert every other photon off to the side, to
be
> > recombined with what's left.) How much theory/experiment has been
directed
> > to this issue?
>
> Lots.
>
I would like to see some examples.
----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure Usenet News==----
http://www.newsfeeds.com The #1 Newsgroup Service in the World! 120,000+ Newsgroups
----= East and West-Coast Server Farms - Total Privacy via Encryption =----
Neil
"Repeating Rifle" <salm...@sbcglobal.net> wrote in message
news:BEFC909E.266%salm...@sbcglobal.net...
> <neil_...@hotmail.com> wrote:
>
> > We can't arrange for a photon to be emitted with a definite absolute
phase
> > because of uncertainty issues. (Indeed, for "a photon" it is hard to
imagine
> > just what "absolute phase" would mean.) However, does anything forbid a
> > stream of single photons (say one every ms) from having rather
consistent
> > relative phases in some sense? (I mean, each photon is somehow no more
than
> I doubt that you know what you are talking about. Please explain to me how
> you can tell (measure) the phase you are talking about. Then we can talk.
>
phases: by interference of alternating photons with each other. The
respondent Ron Baker curiously claims that lots of effort has already been
put into this issue, but didn't reference any of it.
Neil
"Josef Matz" <jose...@arcor.de> wrote in message
news:42d7d6d1$0$25087$9b4e...@newsread4.arcor-online.net...
polarization type, it just works differently for each one. For circular, the
amplitude remains constant but cycles around in direction, which defines the
phase. (Also, each type can be considered a composition of the other bases:
linear from R and L circular, and either circular from x or y linear, etc.)
The issue is "photons" versus light as a multi-photon beam. In the latter,
we can study phase more easily, just like many other properties (such as
polarization) that are easier or possible to measure for beams but difficult
for single photons. I am already considering the string of discrete photons
in my example to have some consistent polarization so that property will not
be an issue.
Neil
"Andy Resnick" <andy.r...@op.case.edu> wrote in message
news:db8p2k$crj$1...@eeyore.INS.cwru.edu...
> Josef Matz wrote:
>
> <snip?
> >
> > Good question.
> >
> > Linear polarized light, elliptical and circular light have a defined
phase.
> > Only thermally emitted light contains two components which have an
> > independent phase.
>
> That's not exactly true. At any instant in time, at any point in space,
> the electric field has a precise singly-valued [*] phase. How the
> phase changes in time and space as compared to other points in time and
> space defines the coherence of the field. This is why I prefer the term
> "randomly polarized" to "unpolarized". Sunlight (thermal emission) has
> a short coherence time and transverse length. However, if the light is
> chromatically filtered, one may observe speckle- the transverse
> coherence length at 555 nm is around 1 mm or so.
>
> >
> > The absolute phase is not measureable. That means the value is not
> > determinable due to a measurement because of the uncertainty principle
which
> > gives Delta Phi bigger or equal 2 pi. But we know it exists.
>
> Also not exactly true- the absolute phase (mod 2*pi) is routinely
> measured in radio-wave and lower frequencies, AFAIK.
He means that absolute phase isn't measurable for single photons. Once they
collect together in a radio wave emission, there are enough photons
overlapping to define these properties. I don't know what he means by saying
we can measure the absolute phase, but we know it exists. In any case, what
concerns me here is not the absolute phase of a single photon, but the
relative phase of two photons: can it be known they have a *relative phase*
of say 0.17 lambda, etc?
...
Neil
> Josef Matz wrote:
> <snip>
> > Yes i know this picture. But during the so called coherence lenght or
time
> > which type of polarisation has thermal radiation, any possible - i doubt
Thanks for the input. What we are concerned with here though is individual
photons being compared with other individual photons, and I am asking if we
can get them to be within a small fraction of lambda of each other's phase.
That is not as easy to define or measure as the same questions about beams
of light with many photons.
Ron Baker, Pluralitas!
>
> "Ron Baker, Pluralitas!" <sto...@bellsouth.net.pa> wrote in message
> news:sMHBe.7517$rF5....@tornado.socal.rr.com...
>>
>> "Neil" <neil_...@hotmail.com> wrote in message
>> news:11de6ah...@corp.supernews.com...
>> > We can't arrange for a photon to be emitted with a definite absolute
> phase
>> > because of uncertainty issues. (Indeed, for "a photon" it is hard to
>> > imagine
>> > just what "absolute phase" would mean.) However, does anything forbid a
>> > stream of single photons (say one every ms) from having rather
> consistent
>> > relative phases in some sense?
>>
>> No.
>>
> Has this ever been attempted or measured?
Your questions are better than I had originally thought.
Let me withdraw that 'No'.
It may not be possible to produce a stream of separate
photons that are coherent (and thus, in that sense, it is forbidden).
I had been thinking that 'photons' could be shorter
than the coherence length. Now I am starting
to think they can be no shorter than the coherence
length.
The question as to 'how long is a photon' has come up
in this news group before. There wasn't a concensus
or definitive answer. There are those who argue it has
zero length. There are those who argue that the length
is less than 3 wavelengths.
How did you come up with that 300 +/- .05 wavelengths?
>
>> > stream by pulling out every other photon, then use appropriate optical
>> > path
>> > length adjustments to get the separated streams to interfere. (A very
> fast
>> > electro-optical mirror to divert every other photon off to the side, to
> be
>> > recombined with what's left.) How much theory/experiment has been
> directed
>> > to this issue?
>>
>> Lots.
>>
> I would like to see some examples.
The question that occurs to me is whether coherence length
is effectively 'photon' length.
If they were not the same then double slit and interferometer
results would vary depending on whether multiple or single
photons were measured. I'm not aware of any experiments
specific to that have been performed but it seems to me that
such an effect would have been noticed in other experiments
to date. That would tend to indicate
that coherence length is at least approximately photon length.
--
rb
Josef Matz
thas it exactly and that is covered by classical physics. The phase factor
is responsible for interferences. The absolte phase of interference patterns
is not determinable. Modulated amplitudes (the overlap of many or at least
some different phases delivers a measureable phase if you have a periodicity
in your amplitude modulation.
Exactly this picture is used in classical physics to get the group velocity
of signals.
Y.Porat
by their phase is impossible .... for paractical rasons
(the limit between practical and theorethic is vague in this case)
ie
photons are products of some 'machine gun' device
ie they are produced not by just one Atom
they are produced by *many Atoms'! tha tmigh tbe very 'similar' or even
identically
thorethically but not practically
it is not dependant only on the specific element that emmits it
that atom has some dgrees of freedom of movement!!
so the number of possible stsrt positions at themoment of emmision- is
very big
especially while the number of prodicers of photons is big
it is actually 'a whole lattice' that makes that output of photons
so the probability of all of them to be (absolutely) identical is
small
ATB
Y.Porat
------------------------
Josef Matz
"Y.Porat" <map...@012.net.il> schrieb im Newsbeitrag
news:1121489567....@o13g2000cwo.googlegroups.com...
Josef Matz
"Y.Porat" <map...@012.net.il> schrieb im Newsbeitrag
news:1121489567....@o13g2000cwo.googlegroups.com...
Fully right. There are no two photons (bosons) with exactly the same
frequency. But because their number is so incredible big the frequencies may
differ by a so called differntial dw. The mathematics for that is covered by
classical physics.
Very nice picture your machine gun !
Jim Klein
Who cares and why?
Sounds like you are off your meds. :-)
Jim Klein
Timo Nieminen
> "Andy Resnick" <andy.r...@op.case.edu> schrieb:
> >
> > Zero-amplitude regions of the electromagnetic field are quite
> > interesting and subtle, and I recommend checking out the work by J. F.
> > Nye and Sir Michael Berry (of Berry's phase). It is possible to have a
> > region of space where the amplitude of the field is precisely zero. In
> > this case, the phase is completely undetermined, and can be multivalued.
> > "optical vortices" and "Bessel beams" are some of the buzzwords.
> >
>
> Ok i heard of it. But that is only experimental. No theory for or am i wrong
> ?
There is a lot of theory on it. Michael Berry has published extensively on
the matter; his papers are available from his home page. There are some
interesting papers by Miles Padgett in New Journal of Physics
(experimental work). Lots of stuff by Les Allen, though mostly dealing
with angular momentum rather than singularities as such. Some of our
contributions can be found at e-prints link below.
But the simplest theory is simply that of paraxial laser modes. All of the
higher order Hermite-Gauss modes have line (or plane, if you're
considering 3D rather than a cross-section of the beam) singularities
between adjacent bright spots in the pattern, and Laguerre-Gauss modes
have ring-shaped line singularities between successive bright rings (for
higher radial modes) and point-singularities on the beam axis (for higher
azimuthal modes). Very often, no attention is drawn to these
singularities, but they sit there staring out at you from the equations.
--
Timo Nieminen - Home page: http://www.physics.uq.edu.au/people/nieminen/
E-prints: http://eprint.uq.edu.au/view/person/Nieminen,_Timo_A..html
Shrine to Spirits: http://www.users.bigpond.com/timo_nieminen/spirits.html
Neil
"Jim Klein" <james...@earthlink.net> wrote in message
news:po3id1hnr5rca50d5...@4ax.com...
Why care about anything at all? What decides what is "interesting"? Why do
you care whether it is worth caring about? Who really needs meds? Actually,
if the photons are in phase, then some interesting interference experiments
can be done. One example is to use one beam as a reference beam, and let the
other one be split (HS mirror) and let one portion be monitored by a
detector. Then we combine the remaining beams in a BS. If the detector is
farther from emissions than the BS, the interference is different than if it
is closer, because in the latter case photon detections remove them totally
from the stream instead of their maintaining a consistent attenuated
amplitude. This happens no matter how large the distances are, and thus the
movement of the detector a small distance will cause changes at a distant
location if we apply basic rules to the interference, which causes FTL
signaling. (You will have to mess with this on paper if you care - I don't
have time to write about it much, really.)
Neil
> The question as to 'how long is a photon' has come up
> in this news group before. There wasn't a concensus
> or definitive answer. There are those who argue it has
> zero length. There are those who argue that the length
> is less than 3 wavelengths.
>
First of all, it isn't the length of the photons which is the direct issue,
but rather whether they have the same phase as each other. BTW, the "length"
is indeed the coherence length - what else do you think it could be? The
coherence length is contexed out in some different ways, but the
construction of a wave packet of a certain length is created by the
superposition of its possible wavelengths, which distribution is in turn
decided by the characteristic Delta t of creation of the packet (simply
put.) It is going to be more than only 3 wavelengths, unless you want an
energy spread of 30-40% or so! It can be measured in a literalistic way,
showing it is not just an abstraction, by splitting the light and
recombining it after one path has been delayed. If the delay is too long,
the train of wavicles won't match up and the in-phase null detector channel
will get hits. From a high-quality laser, the photons (statistically, but
remember one hit in a null channel shows it isn't coherent to that length!)
have lengths of a few kilometers!
This though is basically moot to the question, because what I mean is: if
we could measure from the peaks of one wave packet to the peaks of *another*
wave packet (farther away than the "length" of the wave packet as described
here), the difference would be very close to an integer of the average
wavelength of the packets.
...
> The question that occurs to me is whether coherence length
> is effectively 'photon' length.
> If they were not the same then double slit and interferometer
> results would vary depending on whether multiple or single
> photons were measured. I'm not aware of any experiments
> specific to that have been performed but it seems to me that
> such an effect would have been noticed in other experiments
> to date. That would tend to indicate
> that coherence length is at least approximately photon length.
...
Why approximately, and not identically? Like I said, could you explain what
photon length would be if not coherence length?
Ron Baker, Pluralitas!
I'm not disputing that but do you have any references
on that?
>
>> The question as to 'how long is a photon' has come up
>> in this news group before. There wasn't a concensus
>> or definitive answer. There are those who argue it has
>> zero length. There are those who argue that the length
>> is less than 3 wavelengths.
>>
> First of all, it isn't the length of the photons which is the direct
> issue,
> but rather whether they have the same phase as each other.
It seems to me that 'same phase' is 'coherent'.
> BTW, the "length"
> is indeed the coherence length - what else do you think it could be?
I guess it gets back to your original question.
It seems to me that if you had a stream or burst of photons much longer
than the length of a photon but they were all in phase then that
would exhibit coherence over the full length of the burst.
In that case coherence length and photon length would be different.
I don't see why that would not be possible but I don't
know that it ever happens.
It has perhaps already occured to you that such a stream would
yield different results in an interferometer or double slit depending
on whether the beam was attenuated down to one photon every
10 photon lengths or if there were thousands of photons per
photon length. I have never heard of such a difference being
noticed but then I have never heard of anyone specifically looking
for it. Single photon experiments don't seem to be concerned with
coherence length usually.
>The
> coherence length is contexed out in some different ways, but the
> construction of a wave packet of a certain length is created by the
> superposition of its possible wavelengths, which distribution is in turn
> decided by the characteristic Delta t of creation of the packet (simply
> put.) It is going to be more than only 3 wavelengths, unless you want an
> energy spread of 30-40% or so!
Yup.
> It can be measured in a literalistic way,
> showing it is not just an abstraction, by splitting the light and
> recombining it after one path has been delayed. If the delay is too long,
> the train of wavicles won't match up and the in-phase null detector
> channel
> will get hits. From a high-quality laser, the photons (statistically, but
> remember one hit in a null channel shows it isn't coherent to that
> length!)
> have lengths of a few kilometers!
If that result holds even when the beam has been attenuated to
one photon every 10 km or so then I'd say that shows
that the photon length is also that coherence length.
>
> This though is basically moot to the question, because what I mean is: if
> we could measure from the peaks of one wave packet to the peaks of
> *another*
> wave packet (farther away than the "length" of the wave packet as
> described
> here), the difference would be very close to an integer of the average
> wavelength of the packets.
If you don't call two photons of similar phase (even though
well separated in distance) 'coherent', what do you call them?
>
> ...
>> The question that occurs to me is whether coherence length
>> is effectively 'photon' length.
>> If they were not the same then double slit and interferometer
>> results would vary depending on whether multiple or single
>> photons were measured. I'm not aware of any experiments
>> specific to that have been performed but it seems to me that
>> such an effect would have been noticed in other experiments
>> to date. That would tend to indicate
>> that coherence length is at least approximately photon length.
> ...
>
> Why approximately, and not identically? Like I said, could you explain
> what
> photon length would be if not coherence length?
You asked about photons separated by greater than their
length having the same phase. To me 'having the same phase'
is 'coherent'. In such a case the coherence length would
be longer than the photon length.
--
rb
Jürgen Appel
Andy Resnick schrieb:
>> Linear polarized light, elliptical and circular light have a defined
>> phase. Only thermally emitted light contains two components which have
>> an independent phase.
>
> That's not exactly true. At any instant in time, at any point in space,
> the electric field has a precise singly-valued [*] phase.
This is not correct. The problem here lies again in not considering the
quantum nature of light. In quantum mechanics there exists observables
which cannot be sharply defined at the same time (Heisenbergs uncertainty
relation). Probably most have already heared that Position and Momentum of
a free particle are such observables.
The same is true for the elctric field strength, the magnetic field, the
photon number and also for the phase.
If you prepare a certain light state, for example a coherent state, like
they come approximately out of a laser and measure the electric field
strength, in this case you will find, that you always will get slightly
different results. The mean value however changes sinusoidially with time.
The same is true for the magnetic field strengh.
If you have a Fock state (that is one with a sharply defined photon number:
If you prepare this light state identically many times you will always
find the same photon number), the electric field strength at a particular
time is not sharp, however on average you will find a field strength of
zero. The same is true for the phase-observable. It is the conjugate of
the photon number and as such maximally undefined if the number is sharp
(Each possible phase measurement outcome is of equal probability).
> How the
> phase changes in time and space as compared to other points in time and
> space defines the coherence of the field. This is why I prefer the term
> "randomly polarized" to "unpolarized".
This is a different matter: Here you do not prepare the light state
identically (pure state) but instead you generate a _mixture_ of states
with random polarization.
>> The absolute phase is not measureable. That means the value is not
>> determinable due to a measurement because of the uncertainty principle
>> which gives Delta Phi bigger or equal 2 pi. But we know it exists.
>
> Also not exactly true- the absolute phase (mod 2*pi) is routinely
> measured in radio-wave and lower frequencies, AFAIK.
For an absolute phase measurement you need a reference point in time, you
have to mark one point t=0. Once you have defined that, you can measure
"absolute Phases" _relative_ to that point -- which can be choosen
arbitrarily.
Best Regards,
Jürgen
--
GPG key:
http://pgp.mit.edu:11371/pks/lookup?search=J%FCrgen+Appel&op=get
Jürgen Appel
> Zero-amplitude regions of the electromagnetic field are quite
> interesting and subtle, and I recommend checking out the work by J. F.
> Nye and Sir Michael Berry (of Berry's phase). It is possible to have a
> region of space where the amplitude of the field is precisely zero.
One has to distinguish whether the intensity of a particular mode is
precisely zero (then the light state is a special fock state with photon
number zero and the phase is indeed completely undefined) or whether the
electric field strength/amplitude is precisely zero. The latter case
describes a maximally squeezed state where the phase is quite well
defined.
You cannot have both at the same time (Heisenberg's uncertainty relation)
since the photon number and electric field strength observables do not
commute.
> "optical vortices" and "Bessel beams" are some of the buzzwords.
This refers to certain superpositions of light modes; I assume you refer to
an ideal laser beam which just contains photons in the "donut mode" with a
dark spot in the center, there are no photons in the coaxial gauss-mode
and the photon number in this mode is zero. The electric field strength
_amplitude_ in this mode however is not sharp then (there are the famous
vacuum fluctuations).
Best regards,
Jürgen Appel
>> The question that occurs to me is whether coherence length
>> is effectively 'photon' length.
[...]
> Why approximately, and not identically? Like I said, could you explain
> what photon length would be if not coherence length?
There is no censensus since the question simply does not make sense in
generality.
In theory a single photon state is an eigenstate of the electromagnetic
field with a sharp energy. As such, it has to be absolutely monochromatic
and extend to infinity in both time and space. Such states do not exist in
reality and for that reason physicists call a state a single photon state
if it is a superposition
|1> := integral_k w(k) |1>_k dk.
Here w(k) is a complex normalized weighing function over all light field
modes k and |1>_k refers to the idealized one-photon state in mode k. So
your photon is in a superposition of being in some of these modes.
Usually w(k) has a support around a specific k-vector, so that a photon is
associated with a light pulse (e.g. from a downconversion crystal from a
femtosecond laser). In this case you could attribute the mean length of
this pulse to the spatial length of a photon.
However a photon does not have to be located in a single pulse. In quantum
communication devices as they are available commercially nowadays,
time-bin encoding has advantages over polarization encoding. Here the
photons are delocalized over two pulses, the complex coefficients of the
superposition carrying the quantum information. I can't see how you would
define the spatial length of such a delocalized photon.
Y.Porat
'a single photon'??
TIA
Y.Porat
-----------------------------
redbelly
Ron Baker, Pluralitas! wrote:
> If you don't call two photons of similar phase (even though
> well separated in distance) 'coherent', what do you call them?
However, it is not possible to have two coherent photons.
>From quantum optics:
If you know exactly how many photons there are (in the system being
studied), their phases are completely random.
Conversely, if you know the phase with any degree of precision, the
number of photons will have an uncertainty.
Mark
Andy Resnick
<snip>
>
> My question is: Within coherence time you have polarisations. You have
> resolved the coherence time you said.
> Is during the coherence time a situation, where nearly linear polarized
> effects can be measured or not ? Thermal emission is dependent on the degee
> of freedom of movement. This is really a important question. If it goes out
> as i think there are two independent electromagnetic fields overlapping at
> every point, not only one.
> Now please: Who can measure the polarisations and absolute phase constant
> during coherence time ? Is there a place in the USA ? Has Wigner measured it
Ok, let's work through this...
First, the concepts of "coherence time" and 'polarization' are not
dependant on each other. The polarization state of an electormagnetic
field (including a single photon) is perfectly well-defined, although
when we speak of photons people generally say 'helicity' and the sign is
reversed- I think that's done to intentionally confuse simple-minded
people like myself.
Anyhow, "coherence time" and more generally "coherence" really only have
meaning for a wave-like description of the electromagnetic field. There
is a quantum version of coherence theory, laid down by Roy Glauber, but
I don't understand a word of it.
Ok, now for a real field that we can measure and alter, when lots of
photons exist, in order to have a well-polarized state, the light must
also be coherent- not nessessarily coherence time [frequency] but more
usually coherence area [space]. This is best seen in the original
papers regarding Mueller Matrices, which are measurable properties of
the electromagnetic field based on the Stokes vector. Partially
coherent beams will also be at best partially polarized.
Now thermal emission is both incoherent and randomly polarized, except
in special cases: one may spatial filter the thermal light to obtain
spatially coherent light, or use a linear polarizer to polarize the
light, etc. But, the 'raw' emission is incoherent in the regime where
the wavelength is much smaller than the emitter.
It does not make sense to say there are two overlapping fields or not;
the field at a point has only one value (with the zero field exception),
and if the field is monitored over an extended region in space or in
time, it can be decomposed however you may like: a Fourier
decomposition, a Poincare sphere tracing, etc. etc.
<snip>
> Good question free 60 Hz radiation - is that the frequency of the photons
> which take part or is it the time varying amplitude of the current ? So your
> 60 Hz in USA - are it just 60 Hz photons or a continuous photon spectrum and
> moving atoms ? Have you ever measured a photon with energy h(bar) times 2 Pi
> 60 Hz ? Free currents radiate cherenkow radiation - is it quantized like the
> atomic emission ?
60 Hz radiation in free space has a wavelength of 5 *10^6 meters, and a
photon of that frequency has an energy of 4 * 10^-32 J. Not a lot there
to work with.
Free currents don't generate Cerenkov radiation- that arises from
charged particles moving faster than c/n, where n is the refractive
index of the media. In particle detection, it's used as a mass
discriminator- heavy particles don't produce it. The spectrum is
continuous.
Andy Resnick
> Hello,
>
> Andy Resnick schrieb:
>
>
>>>Linear polarized light, elliptical and circular light have a defined
>>>phase. Only thermally emitted light contains two components which have
>>>an independent phase.
>>
>>That's not exactly true. At any instant in time, at any point in space,
>>the electric field has a precise singly-valued [*] phase.
>
>
> This is not correct. The problem here lies again in not considering the
> quantum nature of light. In quantum mechanics there exists observables
> which cannot be sharply defined at the same time (Heisenbergs uncertainty
> relation). Probably most have already heared that Position and Momentum of
> a free particle are such observables.
IIRC, the uncertainty principle defines how well conjugate variables can
be measured simultaneously, not how well-valued the conjugate variables
'are' in some hidden-variables sense of the word.
In so far as the continuum picture holds, that we can define a
smoothly-varying field, the field is single-valued everywhere. If I
place a point detector anywhere in the field, I can query either the
intensity or the phase and get a value to any arbitrary precision.
Based on that, that the electric field is an analytic field, I can
obtain things like the Kramers-Kronig relation. Now, does a continuum
concept like 'index of refraction' etc, hold identically in the quantum
picture? No, and that's why purely quantum derivations of the index of
refraction/suseptibility/permittivity are so spotty and non-general.
> The same is true for the elctric field strength, the magnetic field, the
> photon number and also for the phase.
>
> If you prepare a certain light state, for example a coherent state, like
> they come approximately out of a laser and measure the electric field
> strength, in this case you will find, that you always will get slightly
> different results. The mean value however changes sinusoidially with time.
> The same is true for the magnetic field strengh.
Of course, but that does not change the fact that at any instant of
time, there is an exact single value for the amplitude. Because the
field intensity and photon number are non-commuting variables, there is
no clean correspondance between the (measured) intensity of the field
and the number of photons comprising the field.
<snip>
>
> This is a different matter: Here you do not prepare the light state
> identically (pure state) but instead you generate a _mixture_ of states
> with random polarization.
Right-this is the essence of coherence.
<snip>
> For an absolute phase measurement you need a reference point in time, you
> have to mark one point t=0. Once you have defined that, you can measure
> "absolute Phases" _relative_ to that point -- which can be choosen
> arbitrarily.
Yes, but we like our fields to be stationary- so if I need to get more
coffee, I can reset the clock when I come back and have the new results
be comparible with the pre-coffee break results.
Ron Baker, Pluralitas!
"redbelly" <redbe...@yahoo.com> wrote in message
news:1121689027....@g47g2000cwa.googlegroups.com...
Really? What is the physics behind that? Do you
have any references?
--
rb
Ron Baker, Pluralitas!
"Jürgen Appel" <jap...@linux01.gwdg.de> wrote in message
news:2296559.x...@linux01.gwdg.de...
> Neil schrieb:
>
>>> The question that occurs to me is whether coherence length
>>> is effectively 'photon' length.
>
> [...]
>
>> Why approximately, and not identically? Like I said, could you explain
>> what photon length would be if not coherence length?
>
> There is no censensus since the question simply does not make sense in
> generality.
>
> In theory a single photon state is an eigenstate of the electromagnetic
> field with a sharp energy. As such, it has to be absolutely monochromatic
> and extend to infinity in both time and space. Such states do not exist in
> reality and for that reason physicists call a state a single photon state
> if it is a superposition
>
> |1> := integral_k w(k) |1>_k dk.
> Here w(k) is a complex normalized weighing function over all light field
> modes k and |1>_k refers to the idealized one-photon state in mode k. So
> your photon is in a superposition of being in some of these modes.
Sounds a lot like a Fourier sum.
>
> Usually w(k) has a support around a specific k-vector, so that a photon is
> associated with a light pulse (e.g. from a downconversion crystal from a
> femtosecond laser).
'support'? Is that like 'moment'? Like S k*w(k) dk ?
(Using 'S' as the integral sign.)
> In this case you could attribute the mean length of
> this pulse to the spatial length of a photon.
>
> However a photon does not have to be located in a single pulse. In quantum
> communication devices as they are available commercially nowadays,
> time-bin encoding has advantages over polarization encoding. Here the
> photons are delocalized over two pulses, the complex coefficients of the
> superposition carrying the quantum information.
You lost me there. What does that mean?
What does 'delocalized over two pulses' mean?
> I can't see how you would
> define the spatial length of such a delocalized photon.
You sound like you actually know stuff.
Welcome to sci.physics. I hope you stick around.
--
rb
redbelly
Ron Baker, Pluralitas! wrote:
> Really? What is the physics behind that? Do you
> have any references?
One reference is the book Introductory Quantum Optics by Christopher
Gerry.
He mentions that phase and photon number are, in a sense, complementary
in much the way energy & time, or momentum & position, are.
It's at the bottom of p. 16, which you can look at for free if you have
an ID at Amazon.com:
(If this link doesn't work directly, then:
1. Go to http://amazon.com
2. search Amazon.com for gerry quantum optics
3. Click on the first choice in the search results, Introductory
Quantum Optics by Christopher Gerry
4. Scroll about halfway down the web page, to the box after the words
"Search Inside This Book"
5. Type the number "16" in the box (no quotes) and click on "Go"
6. In the search results, click on the words "on page 16"
7. Once page 16 of the book is displayed, start reading the text after
Equation 2.49.
8. Click on "Next page" to continue reading on p. 17.
In this text, "n" refers to the number of photons. |n> refers to a
state of a system where the number of photons is precisely known to be
n.
Finally, <capital letter Delta>n is the uncertainty in the number of
photons, and <capital Delta> <lowercase phi> is the uncertainty in the
phase.
HTH, Mark
p.s. Gerry mentions that the there is some contention about the concept
of a quantum phase operator. I guess this means there is disagreement
as to whether phase is truly a measureable quantity, but I'm not
really up on why that is a problem.
redbelly
You can search Google on "number-phase uncertainty" (with the quotes)
to find more.
Mark
Neil
> >> The absolute phase is not measureable. That means the value is not
> >> determinable due to a measurement because of the uncertainty principle
> >> which gives Delta Phi bigger or equal 2 pi. But we know it exists.
> >
> > Also not exactly true- the absolute phase (mod 2*pi) is routinely
> > measured in radio-wave and lower frequencies, AFAIK.
>
> For an absolute phase measurement you need a reference point in time, you
> have to mark one point t=0. Once you have defined that, you can measure
> "absolute Phases" _relative_ to that point -- which can be choosen
> arbitrarily.
>
> Best Regards,
> Jürgen
Thanks, I was looking for some upscale commentary on all this. Per "absolute
phase": sure, you can have a reference point in time set to a fine accuracy,
but how will you know what that is for a *single photon* ? There is nothing
to measure that won't absorb the photon entire (?). As for the phase between
two photons, (in effect, how far the amplitude peaks of photon 1 are from
the peaks of photon 2) we can get them to interfere, and at least rule out
their being in phase by hits in the "dark" (out-of-phase) channel, etc. For
several photons, enough interference events gives us a rough idea. That's
what I mean. The question is, how does that square with the genuinely more
complicated nature of light, in which photon number itself is undefined?
Indeed, consider post above by redbelly, re "number-phase uncertainty": the
problem is, there aren't well defined photons if we know the phase to high
accuracy. I'll have more to say about that later. Any more about that from
other posters?
Ron Baker, Pluralitas!
>
> Ron Baker, Pluralitas! wrote:
>
>> Really? What is the physics behind that? Do you
>> have any references?
>
> One reference is the book Introductory Quantum Optics by Christopher
> Gerry.
> He mentions that phase and photon number are, in a sense, complementary
> in much the way energy & time, or momentum & position, are.
>
> It's at the bottom of p. 16, which you can look at for free if you have
> an ID at Amazon.com:
>
> http://www.amazon.com/gp/reader/052152735X/ref=sib_vae_pg_16/103-8976455-8880661?%5Fencoding=UTF8&keywords=16&p=S00W&twc=50&checkSum=Rln1YA%2FoxFaCJnC9I4CyxvAaSN0isj%2B9BrMscdxrMk8%3D#reader-page
>
> (If this link doesn't work directly, then:
> 1. Go to http://amazon.com
> 2. search Amazon.com for gerry quantum optics
> 3. Click on the first choice in the search results, Introductory
> Quantum Optics by Christopher Gerry
> 4. Scroll about halfway down the web page, to the box after the words
> "Search Inside This Book"
> 5. Type the number "16" in the box (no quotes) and click on "Go"
> 6. In the search results, click on the words "on page 16"
> 7. Once page 16 of the book is displayed, start reading the text after
> Equation 2.49.
It asked for a credit card number before showing the page.
It didn't get what it asked for.
> 8. Click on "Next page" to continue reading on p. 17.
>
> In this text, "n" refers to the number of photons. |n> refers to a
> state of a system where the number of photons is precisely known to be
> n.
> Finally, <capital letter Delta>n is the uncertainty in the number of
> photons, and <capital Delta> <lowercase phi> is the uncertainty in the
> phase.
It sounds like you are confusing 'photon number' with 'number of
photons'. Photon number is, practically speaking, wavelength, not
number of photons.
>
> HTH, Mark
>
> p.s. Gerry mentions that the there is some contention about the concept
> of a quantum phase operator. I guess this means there is disagreement
> as to whether phase is truly a measureable quantity, but I'm not
> really up on why that is a problem.
>
--
rb
redbelly
Ron Baker, Pluralitas! wrote:
>
> It asked for a credit card number before showing the page.
> It didn't get what it asked for.
Perhaps since I have ordered from Amazon before, they may already have
my credit card number and so didn't need me to resupply it. Hopefully
I have given a reasonable summary of what was in the book, or as I said
before you can search at Google on "number-phase uncertainty" and find
further references, eg. from scientific journals.
> It sounds like you are confusing 'photon number' with 'number of
> photons'. Photon number is, practically speaking, wavelength, not
> number of photons.
The author of the book I mentioned uses the terms interchangably, and
is in fact talking about the number of photons.
I've never heard wavelength refered to as photon number. Could you be
confusing it with the 'wave number' (= 1/wavelength)?
Mark
Ron Baker, Pluralitas!
>
> Ron Baker, Pluralitas! wrote:
>
>>
>> It asked for a credit card number before showing the page.
>> It didn't get what it asked for.
>
> Perhaps since I have ordered from Amazon before, they may already have
> my credit card number and so didn't need me to resupply it. Hopefully
> I have given a reasonable summary of what was in the book, or as I said
> before you can search at Google on "number-phase uncertainty" and find
> further references, eg. from scientific journals.
Which I previously did, and read the abstracts. The usage of
the term 'photon number' appeared similar to the term 'wave number'.
On further reading I see that it does indeed seem as you say.
They appear to actually be refering to numbers of photons.
>
>> It sounds like you are confusing 'photon number' with 'number of
>> photons'. Photon number is, practically speaking, wavelength, not
>> number of photons.
>
> The author of the book I mentioned uses the terms interchangably, and
> is in fact talking about the number of photons.
>
> I've never heard wavelength refered to as photon number. Could you be
> confusing it with the 'wave number' (= 1/wavelength)?
Sorry. That's what I meant. 1/wavelength.
Question for you. How do you think delta theta * delta n = 0.5
relates or does not relate to the interference pattern in the
double slit experiment?
--
rb
Gregory L. Hansen
Neil <neil_...@hotmail.com> wrote:
>We can't arrange for a photon to be emitted with a definite absolute phase
>because of uncertainty issues. (Indeed, for "a photon" it is hard to imagine
>just what "absolute phase" would mean.) However, does anything forbid a
>stream of single photons (say one every ms) from having rather consistent
>some fraction of radian of effective phase different from its neighbors,
>each of which is beyond the average coherence length for single photons.)
Wouldn't that be a laser?
>After all, the parts of split wave functions of single photons can interfere
>with each other, or even attenuated beams from separate lasers containing
>only one or two photons on average at a time. Maybe we could test such a
>stream by pulling out every other photon, then use appropriate optical path
>length adjustments to get the separated streams to interfere. (A very fast
>electro-optical mirror to divert every other photon off to the side, to be
>recombined with what's left.) How much theory/experiment has been directed
>to this issue?
Lots. It's important in interferometry. E.g. Maxwell (I think it was)
created a stellar interferometer that measured a star's diameter by
measuring the transverse coherence length of its light.
--
"For every problem there is a solution which is simple, clean and wrong."
-- Henry Louis Mencken
redbelly
Ron Baker, Pluralitas! wrote:
> Question for you. How do you think delta theta * delta n = 0.5
> relates or does not relate to the interference pattern in the
> double slit experiment?
I don't think it applies. In the double-slit experiment, each photon
can be thought of as interfering with itself. For example, the dark
lines of the interference pattern correspond to where the two possible
paths of each photon are 180 degrees out of phase, thus cancelling to
give an electric field of zero. Adding a random phase to the photon
would not change this.
The earlier posts, to which I responded by bringing up the phase
uncertainty, were talking about either the relative phase between
different photons, or the phase of a photon relative to some outside
reference.
Mark
Neil
A general question: I presume Delta theta * Delta n means that "the phase"
of the wave is indeterminate to an extent conjugate with the photon number.
However, does that really limit "relative phase" of one photon with respect
to another, if we don't need to determine "phase" per se? (Note that the
definition would involve the probabilities of results of interference, and
thus is defined in a way that "absolute" (actually, laboratory reference)
phase would not be.
Neil
Actually, we can get two separate lasers to interfere with each other for
brief periods of time. Sure, we lose which-way information, so you can be
snarky and say one photon "came from both laser", but in effect you have
interfered different photons.
Speaking of which, a general question: I presume Delta theta * Delta n means
that "the phase" of the wave is indeterminate to an extent conjugate with
the photon number that can be observed there. However, does that really
limit "relative phase" of one photon with respect to another, if we don't
need to determine "phase" per se? (Note that the definition would involve
the probabilities of results of
interference, and thus is defined in a way that "absolute" (actually,
laboratory reference) phase would not be.)
redbelly
Neil wrote:
> ... a general question: I presume Delta theta * Delta n means
> that "the phase" of the wave is indeterminate to an extent conjugate with
> the photon number that can be observed there. However, does that really
> limit "relative phase" of one photon with respect to another, if we don't
> need to determine "phase" per se? (Note that the definition would involve
> the probabilities of results of
> interference, and thus is defined in a way that "absolute" (actually,
> laboratory reference) phase would not be.)
Hello Neil,
You raise a good question, and I'm not sure what (if anything) the
number-phase uncertainty relation says about relative phases of
photons within a collection. Perhaps one of the others has an answer.
When I originally replied to this thread, it seemed there was a
question
about the phase between two separate photons, each of which consituted
a
system of exactly one (with zero uncertainty) photon, hence the
implications of number-phase uncertainty seemed fairly straightforward.
Regards,
Mark
Repeating Rifle
1122068427....@g49g2000cwa.googlegroups.com, "redbelly"
<redbe...@yahoo.com> wrote:
Once in a while, I peek at this useless thread. It drives me crazy.
The "relative phases of photons within a collection" concept is crazy.
Within the Copenhagen interpretation of quantum mechanics statements of fact
are only meaningful if a measurement can be performed. How do you measure
"relative phases of photons within a collection?"
What this form of the uncertainty principle means is that the error
measuring the phase on an electromagnetic wave packet depends upon the
energy in the packet. That is, the error set by the Heisenberg uncertainty
principle on the phase of the packet is inversely proportional to the number
of photons of energy in the packet. It says nothing about individual
photons.
Bill
Neil
"Repeating Rifle" <salm...@sbcglobal.net> wrote in message
news:BF06D5D6.F81%salm...@sbcglobal.net...
> On 7/22/05 2:40 PM, in article
> 1122068427....@g49g2000cwa.googlegroups.com, "redbelly"
> <redbe...@yahoo.com> wrote:
>
> >
> > Neil wrote:
> >> ... a general question: I presume Delta theta * Delta n means
> >> that "the phase" of the wave is indeterminate to an extent conjugate
with
> >> the photon number that can be observed there. However, does that
really
> >> limit "relative phase" of one photon with respect to another, if we
don't
> >> need to determine "phase" per se? (Note that the definition would
involve
> >> the probabilities of results of
> >> interference, and thus is defined in a way that "absolute" (actually,
> >> laboratory reference) phase would not be.)
> Once in a while, I peek at this useless thread. It drives me crazy.
>
> The "relative phases of photons within a collection" concept is crazy.
> Within the Copenhagen interpretation of quantum mechanics statements of
fact
> are only meaningful if a measurement can be performed. How do you measure
> "relative phases of photons within a collection?"
>
No, it isn't crazy. As I implied in the OPP, we can define the relative
phases of two photons as the relative probabilities of their getting
interference through the A-channel ("bright fringe") or B-channel ("dark
fringe") of an interferometer designed to bring them together. The only
problem I see is whether the coherence lengths match, which is a
complicating factor.
> What this form of the uncertainty principle means is that the error
> measuring the phase on an electromagnetic wave packet depends upon the
> energy in the packet. That is, the error set by the Heisenberg uncertainty
> principle on the phase of the packet is inversely proportional to the
number
> of photons of energy in the packet. It says nothing about individual
> photons.
...
No, not directly, but we can't measure or define the phase of a single
photon. Can you think of a way to do it?
Repeating Rifle
<neil_...@hotmail.com> wrote:
> No, not directly, but we can't measure or define the phase of a single
> photon. Can you think of a way to do it?
You are asking me! I am the one who is driven crazy by statements that make
no sense within Copenhagen interpretation. And you want me to prove that I
am wrong.
Bill
Y.Porat
does anyone known what is a single photon??
TIA
Y.Porat
--------------
Josef Matz
"Y.Porat" <map...@012.net.il> schrieb im Newsbeitrag
news:1122178420.0...@g47g2000cwa.googlegroups.com...
Josef Matz
"Neil" <neil_...@hotmail.com> schrieb im Newsbeitrag
news:11e5r7a...@corp.supernews.com...
The double silt experiment with single photons shows, that two or few
photons emitted at any time from the same source are in phase and interfere.
I think that this is fully ignored here. If you measure the phase of a
single photon in one of the silts, no interference takes place and the
photons seem to come from independent sources. Also Richard Feynman assumed
in his famous book about QED, that photons generally are in phase or have
the same phase. And he pointed out that this has been tested precisely and
he believes that there is no distance limit for this.
On the other hand the simple polarization experiment with sun light has to
be considered. There are left - and right handed photon streams in thermal
radiation.
These are independent from each other and each of these two overlying fluxes
is circular polarized. left handed circular polarization has a clock wise
rotation of the electric field and transports spin, while the composite
thermal radiation transports no spin. So even if the picture with a
statistical variation of polarisation with the so called coherence time or
lenght is suited to explain the polarisation properties quite well, it is
not suited to explain the inteference effects as we observe them.
Therefore these pictures have to be overthought. And following Richard
Feynman i think that there is not at all a coherence lenght definition in
nature as it is discussed here.
Now Richard Feynman did not know that there is an electromagnetic spin
associated with the transport of electromagnetic waves. I think that he
would have come to the same conclusion than me:
There are two circular polarized light beams with their fields in thermal
radiation.
If lets say the right circular polarized beam has the electric field E
(vector) the
left circular polarized has the field E*, the conjungated field amplitude.
But the phases of both are in no fixed relation to each other. If they would
be in phase, you would get linear polarized light, but thermal radiation of
shure is not linear polarized. If there would be a fixed mathematical
relation between theses phases
of the left and right circular polarized parts, sunlight would at any time
be linear polarized, but the orientation changing with time rather
irregular. It is also one possible picture but if this picture is right we
do not know yet. It would not be so difficult to determine if this picture
is true or not because you would get other formulas for the reflection
coefficient of thermal radiation at non normal incident angles as we assume
them presently to be true. No question that a complicated mathematical
relation between the two bare phases does in no case solve the Maxwell
equations. Therefore i think we must assume two fully independent fields and
we can not deal with them in the same equation ! But we get the second
independent component easily if we conjungate the complex macroscopic
maxwell equations. I just want to say that this was a question in Jacksons
Classical Electrodynamics ( Issues around 1980) where Jackson asked about
the physical meaning of the conjungated maxwell equation: It is the the wave
with opposite spin in thermal radiation, thats the ultimative answer i
think. And we should forget and cancel the words coherence lenght and
coherence time in physics. But that of shure is not the opinion of the
majority but i think i have good arguments for.
So the question is: If we discuss about relative phases: Can we measure the
phase difference of a photon with spin 1/2 and with spin -1/2 in thermal
radiation ?
The answer probably is no. And as i sayed in another post here a little
above i doubt that the absolute bare phase is measureable wether for spin
1/2 flux nor for the spin - 1/2 flux because of Heisenbergs uncertainty
principle.
The photon number can not be a input to the uncertainty principle. Maybe the
photon density could. But then you get dimensional prroblems. Therefore i
think that phase considerations with a photon number as it was discussed
here by some guys cant be good models at all.
Josef jose...@arcor.de
>
Repeating Rifle
42e3b383$0$18005$9b4e...@newsread4.arcor-online.net, "Josef Matz"
<jose...@arcor.de> wrote:
<snip>
> On the other hand the simple polarization experiment with sun light has to
> be considered. There are left - and right handed photon streams in thermal
> radiation.
> These are independent from each other and each of these two overlying fluxes
> is circular polarized. left handed circular polarization has a clock wise
> rotation of the electric field and transports spin, while the composite
> thermal radiation transports no spin. So even if the picture with a
> statistical variation of polarisation with the so called coherence time or
> lenght is suited to explain the polarisation properties quite well, it is
> not suited to explain the inteference effects as we observe them.
This shows possible confusion between probability and probability amplitude.
The key to understanding quantum mechanics is to UNDERSTAND the distinction
between probability an probability amplitude. As long as that is not second
nature to you, you will not be capable of understanding quantum mechanics.
>
> Therefore these pictures have to be overthought. And following Richard
> Feynman i think that there is not at all a coherence lenght definition in
> nature as it is discussed here.
>
> Now Richard Feynman did not know that there is an electromagnetic spin
> associated with the transport of electromagnetic waves. I think that he
> would have come to the same conclusion than me:
> There are two circular polarized light beams with their fields in thermal
> radiation.
<snip>
Let me assure you, from personal knowledge, that Feynman knew about
electromagnetic spin for at least a decade (and probably more than two
decades) before The Feynman Lectures on Physics was published.
Bill
Josef Matz
"Repeating Rifle" <salm...@sbcglobal.net> schrieb im Newsbeitrag
news:BF09252B.1330%salm...@sbcglobal.net...
But i did not speak of quantum mechanics. I speak upon electrodynamics and
electromagnetic waves. But there is also a amplitude and a phase. And i
think i know the difference.
> >
> > Therefore these pictures have to be overthought. And following Richard
> > Feynman i think that there is not at all a coherence lenght definition
in
> > nature as it is discussed here.
> >
> > Now Richard Feynman did not know that there is an electromagnetic spin
> > associated with the transport of electromagnetic waves. I think that he
> > would have come to the same conclusion than me:
> > There are two circular polarized light beams with their fields in
thermal
> > radiation.
>
> <snip>
>
> Let me assure you, from personal knowledge, that Feynman knew about
> electromagnetic spin for at least a decade (and probably more than two
> decades) before The Feynman Lectures on Physics was published.
>
Ok. He knew that photons have spin. But he could not knew the experiments
first published around 1991 by USA scientists and recently by indish
scientists, which make insight deeper for electromagnetic waves. In fact it
is known qualitatively and written in old Berkeley physics courses, that
circular polarized light transports spin. But this is a little too little.
And i think if Feynman would have found a modified complex refractive index
theory he would have published it because in one of his books he treated
this theme. Yeah he was a great scientist with a lot of good ideas, and
feelings on what is important and what not. And even where he did not know
the answers he had the right feeling on how they might look. Joe
> Bill
>
Repeating Rifle
42e3f9e1$0$25328$9b4e...@newsread2.arcor-online.net, "Josef Matz"
<jose...@arcor.de> wrote:
> Ok. He {Feynman} knew that photons have spin. But he could not knew the
>experiments
> first published around 1991 by USA scientists and recently by indish
> scientists, which make insight deeper for electromagnetic waves. In fact it
> is known qualitatively and written in old Berkeley physics courses, that
> circular polarized light transports spin. But this is a little too little.
> And i think if Feynman would have found a modified complex refractive index
> theory he would have published it because in one of his books he treated
> this theme. Yeah he was a great scientist with a lot of good ideas, and
> feelings on what is important and what not. And even where he did not know
> the answers he had the right feeling on how they might look. Joe
The angular momentum of circularly polarized light is a classical effect
that was understood well before it was attributed to photons. Measuring the
angular momentum was difficult. See
http://www.aip.org/pt/vol-53/iss-8/p66.html.
Bill
Neil
ability to define what relative phase between photons would be, and I
provided it (probabilities of interference.) Then, you said the HUP says
nothing about individual photons, and I provided a counterexample (it says,
AFAIK, that we can't know the phase of a single photon.) When I asked you if
we could, that was just a rhetorical question to show the point that we
couldn't.
Autymn D. C.
! -> !
? -> ?
can not -> cannot
shure -> sure
theses -> these
wether -> whether
Heisenbergs -> Heisenberg's
cant -> can't
Andy Resnick
Here's a link to the paper (1936 vintage):
http://prola.aps.org/abstract/PR/v50/i2/p115_1?qid=4eaed0e102d8448c&qseq=19&show=10
Mechanical Detection and Measurement of the Angular Momentum of Light
Richard A. Beth*
Worcester Polytechnic Institute, Worcester, Mass. and Palmer
Physical Laboratory, Princeton University
Received 8 May 1936
The electromagnetic theory of the torque exerted by a beam of polarized
light on a doubly refracting plate which alters its state of
polarization is summarized. The same quantitative result is obtained by
assigning an angular momentum of [h-bar] (- [h-bar] ) to each quantum
of left (right) circularly polarized light in a vacuum, and assuming the
conservation of angular momentum holds at the face of the plate. The
apparatus used to detect and measure this effect was designed to enhance
the moment of force to be measured by an appropriate arrangement of
quartz wave plates, and to reduce interferences. The results of about
120 determinations by two observers working independently show the
magnitude and sign of the effect to be correct, and show that it varies
as predicted by the theory with each of three experimental variables
which could be independently adjusted.
©1936 The American Physical Society
URL: http://link.aps.org/abstract/PR/v50/p115
DOI: 10.1103/PhysRev.50.115
The photon picture is alluded to here, but what is more relevant is that
this paper was written in 1936, well before QED and the Feynman lectures
existed.
Josef Matz
"Andy Resnick" <andy.r...@op.case.edu> schrieb im Newsbeitrag
news:dc2mvr$o14$1...@eeyore.INS.cwru.edu...
Thank you, Andrew for this link. So the guys have nearly been as far as i am
now 60 years ago. Interesting indeed. So it seems that the biggest problem
of physics is the filter and teach process. Maybe also that the stuff became
so big that its nearly impossible for book writers to have a complete
overview. And some of the new truths are already buried in history -
interesting.
And Beth already measured spin effects and experimentally verified a
conservation law between non conductors !!!! - Nowhere in standard physics
books ! All classical electrodynamics ! So what remains is a good feeling
that my thoughts have been right. And a bad feeling that my "new"
contributions reduce to a few but still interesting points.
But maybe also this points have been already thought and written in former
times. One never knows.
Good is also that i have an experimental verification of my theory - even if
it is from 1936. Sometimes i think that the second world war made a cut in
this tradition or the interest switched to other topics.
Thanks a lot. Joe
Maybe you can bury out something which is related with alternative ancient
theories related to the index of refraction ? Hope on your assistence in
future discussions here in this group.
Josef Matz
"Andy Resnick" <andy.r...@op.case.edu> schrieb im Newsbeitrag
news:dc2mvr$o14$1...@eeyore.INS.cwru.edu...
Something is wrong with Bethe´s experiments !
Consider the following case: You total reflect linear polarized light (spin
0) on a glass - air surface. The reflected light is circular polarized and
has spin.
Now we have two possibilities:
a) Spin is not conserved. This is a discrepancy with the measurements of
Bethe.
Those discrepancies would also exist at metallic surfaces. With more
sophisticated arrangements it is possible to construct (at least at an
theoretical level) machines with have no energy conservation ! We are living
in a non conservative world concerning certain radiation effects.
b) Our description of light is wrong. The reflected light contains two
left - and right
turning components each of these two elliptical polarized with the same
axe ratio and orientation of the main axes. In the extreme case of circular
polarized light you would have the polarization properties of thermal light.
Then our theory of light is fully wrong.
What of these two statements is true ? I think its the first one is true.
What was Bethes reflection and transmission theory ?
Repeating Rifle
42e59b21$0$6222$9b4e...@newsread2.arcor-online.net, "Josef Matz"
<jose...@arcor.de> wrote:
> Something is wrong with Bethe´s experiments !
>
> Consider the following case: You total reflect linear polarized light (spin
> 0) on a glass - air surface. The reflected light is circular polarized and
> has spin.
> Now we have two possibilities:
>
> a) Spin is not conserved. This is a discrepancy with the measurements of
> Bethe.
> Those discrepancies would also exist at metallic surfaces. With more
> sophisticated arrangements it is possible to construct (at least at an
> theoretical level) machines with have no energy conservation ! We are living
> in a non conservative world concerning certain radiation effects.
>
> b) Our description of light is wrong. The reflected light contains two
> left - and right
> turning components each of these two elliptical polarized with the same
> axe ratio and orientation of the main axes. In the extreme case of circular
> polarized light you would have the polarization properties of thermal light.
> Then our theory of light is fully wrong.
>
> What of these two statements is true ? I think its the first one is true.
>
> What was Bethes reflection and transmission theory ?
I am not totally sure of what you are driving at.
Nevertheless, when right-handed light reflects off of a perfect metal, it
converts into left-handed light. This means that the angular momentum vector
of the reflected light is the same as that of the incident light. There is
no net angular momentum transfer to the reflecting metal surface. There is,
however a transfer of linear momentum.
Bill
Josef Matz
> On 7/25/05 7:13 PM, in article
> 42e59b21$0$6222$9b4e...@newsread2.arcor-online.net, "Josef Matz"
> <jose...@arcor.de> wrote:
>
>
> Nevertheless, when right-handed light reflects off of a perfect metal, it
> converts into left-handed light. This means that the angular momentum
vector
> of the reflected light is the same as that of the incident light. There is
> no net angular momentum transfer to the reflecting metal surface. There
is,
> however a transfer of linear momentum.
>
> Bill
>
light. No spin.
If the light is not perpendicular or parallel polarized the reflection is an
elliptic polarized beam. In the ideal case is as circular polarized as
possible. The reflected light has spin (if it would not be of a composite
type with two components). Because of this reason the body gets a force and
if it is a machine begins to turn.
Now the construct with total reflection: With total reflection you have the
same situation. There exists one angle of incidence and one angle of the
orientation of the incident light, that you get circular polarized ligt
after two total reflections.
So the two prisms can be arranged to rotate ( a machine ). I know that the
forces are small. But in ideal case you would have a 100% conversion to the
machine, no light losses and the outcome light 100% circular polarized. You
take a black wheel, an idealized machine which makes again 100% efficiency.
now you have 200 % efficiency in total.
You can construct arrays of machines where even 200% is small. You could get
as much energy as you want from a single light beam without spin ! linear
polarized incident beam.
That would be the case if present theory is true.
Astonishing isnt it ?
Josef.
Neil
Repeating Rifle wrote:
> On 7/25/05 7:13 PM, in article
> 42e59b21$0$6222$9b4e...@newsread2.arcor-online.net, "Josef Matz"
> <jose...@arcor.de> wrote:
>
> > Something is wrong with Bethe´s experiments !
> >
> > Consider the following case: You total reflect linear polarized light (spin
> > 0) on a glass - air surface. The reflected light is circular polarized and
> > has spin.
> > Now we have two possibilities:
> >
> > a) Spin is not conserved. This is a discrepancy with the measurements of
> > Bethe.
> > Those discrepancies would also exist at metallic surfaces. With more
> > sophisticated arrangements it is possible to construct (at least at an
> > theoretical level) machines with have no energy conservation ! We are living
> > in a non conservative world concerning certain radiation effects.
...
No, there is no problem with the Beth experiment - but there are other
interesting paradoxes nevertheless about photon spin, energy, and
measurement. Beth directed CP light through a linearly birefringent
half-wave plate (actually, quarter-wave plate stuck to a mirror which
gives the same result.) A HW plate flips the spin of CP light and this
transfers spin to the plate.
However, two paradoxes:
1) Transferring angular momentum to the plate can increases its energy
(make it rotate a little.) Where does that energy balance out? Tigran
Galstian of Laval University thinks it would make the photon a bit
redder. However, what if the photon has very low energy, and the HW
plate is already spinning? There is a photon energy of course low
enough that we couldn't offset the increased energy of the plate. (I
know, probably a radio wave photon, but in principle...)
2) Can we measure the amount of circular polarization (proportions of
RH and LH components) and not just yes-no chances of hits on an
eigenstate sorter? Suppose we ran the same photon through a HWP many
many times, by reflecting it and running through another HWP to revert
it back to original handness. HW plates do not alter the type of
polarization (how elliptical, etc.), only the handness. After many
passes, there would be an amount of transferred angular momentum equal
to 2n*hbar*tau_c where tau_c is the circularity of the photon (e.g, +1
for LH, -1 for HR, zero for linear, in between for elliptical.) This
would give us a way to know that the photon had a wide or skinny
ellipse of polarization, etc., supposedly forbidden. (Notoriety buzz:
type "quantum measurement paradox" into Google (not even Groups) and a
thread I started about this comes up first.)
Repeating Rifle
1122419233....@g43g2000cwa.googlegroups.com, "Neil"
<para...@mailcity.com> wrote:
There is no paradox. To the extent that there is energy "lost," consider
(and I will coin some terms) an angular Compton or Doppler effect.
> 2) Can we measure the amount of circular polarization (proportions of
> RH and LH components) and not just yes-no chances of hits on an
> eigenstate sorter? Suppose we ran the same photon through a HWP many
> many times, by reflecting it and running through another HWP to revert
> it back to original handness. HW plates do not alter the type of
> polarization (how elliptical, etc.), only the handness. After many
> passes, there would be an amount of transferred angular momentum equal
> to 2n*hbar*tau_c where tau_c is the circularity of the photon (e.g, +1
> for LH, -1 for HR, zero for linear, in between for elliptical.) This
> would give us a way to know that the photon had a wide or skinny
> ellipse of polarization, etc., supposedly forbidden. (Notoriety buzz:
> type "quantum measurement paradox" into Google (not even Groups) and a
> thread I started about this comes up first.)
All you have with linearly polarized photon is zero angular momentum before
passing through a HWP and zero angular momentum after it goes through the
plate. That does not require an exchange of angular momentum.
Bill
Josef Matz
paradoxes. And i think this are no paradoxes, let me try to explain this:
As i told in another article there exist now an index theory with energy
conservation on each surface. So energy conservation of the pointing vector
is fulfilled on all surfaces. The spin flux however is not conserved on
metallic surfaces and when total reflection takes place. This non
conservation of the spin flux has to do with the fact, that inhomgeneous
waves (waves with an exponential decay in amplitude) are involved. Now in
order to calculate the forces on a body
you must assume a conservation Law for the whole.
Example: The conservation Law where the Pointing vector is involved is the
Impulse conservation. A reflected photon or a wave transmits therefore a
little impulse to the makrobody. A absorbed photon on a black surface makes
only
the half force on the body. So we are used to that. The consequence: solar
radiation absorbed or reflected by the earth makes a small force which is
oriented away from the sun.
In order to calculate the forces (momenta) that turns bodies we have to
assume
a conservation Law for the total angular momentum. this contains the
Pointing vector via r x S and the spin flux via L = a E x E*. Those formulas
might be long if you also take into account absorption within the bodies.
But more important it also contains the term with broken conservation of the
spin flux whenever inhomogeneous waves are involved. So you have a broken
conservation law
for the spin flux but in total you need a conservation law to calculate the
forces on the earth where your laboratory is on.
As we have seen we can circumvent the first law of thermodynamics because
our broken conservation law for the spin flux also produces energy. Those
effects exist everywhere while their total amount is small. But do we really
know that absolute value already - no ? We just know that the forces
involved are normally small compared to other existing forces. Therefore -
under normal circumstances we do not remark anything. But the machinery of
the universe permanently produces small nonconservative effects somewhere
(or anywhere) related to radiation spin on surfaces. The local energy
increases therefore anywhere and therfore there must be other effects that
minder this increase - and a fact is that global red doppler shift is a
possible toy to compensate the local energy increase.
If we take this picture we have at least a model to understand why local
thermodynamics can be circumvented when spin effects are involved.
But now we have no paradox any more and understand where the energy comes
from. It comes out of the big pool of kinetic and potential energy of the
Makrobodies - and to the last out of the universe.
The world locally is non conservative. The approvement easily can be done by
everybody: Take two prisms and shift them together. Let linear polarized
light fall on this system and make a double total reflection so that the
outcomining light is circular polarized - the formulas are for example in
the Books of Max Born or
M Born and E.Wolf. Now you take a circular polaraizer plate for example from
the Berkeley Physics course. Let the double total reflected light beam pass
the circular polarizer. Turn the circular polarizer al let the beam pass
again. In one of the two cases the light beam passes and in the other case
you get nothing - if there are no spin flips. If it is so you have circular
polarized light wizh soin and no spinless two component light. You have the
approval that present spin theory is right and that the universe is non
conservative. You have the approval that termodynamics law can be
circumvented in this case and in other cases and that many paradoxes and
other effects might be true. You now understand why all this trouble with
physics.
If somebody could do this simple experiment for me ? I would be really
dissapointed if the expected result would not come out.
Josef
Neil
Repeating Rifle wrote:
> On 7/26/05 4:07 PM, in article
> 1122419233....@g43g2000cwa.googlegroups.com, "Neil"
> <para...@mailcity.com> wrote:
...
> There is no paradox. To the extent that there is energy "lost," consider
> (and I will coin some terms) an angular Compton or Doppler effect.
Yes, there is. Every photon has hbar of spin, but their energies vary.
If the photon energy is low enough, the amount of energy added to the
disk that's mandated by conservation of angular momentum exceeds the
energy of the photon (clearly, since photon energy can be as low as we
wish.) In that case, Galstian's concept of red shift of the photon
won't be enough to balance the total energy.
>>...
>
> All you have with linearly polarized photon is zero angular momentum before
> passing through a HWP and zero angular momentum after it goes through the
> plate. That does not require an exchange of angular momentum.
>
Yes, that's the whole point and the reason it creates a paradox. The
linear photon will fail to transfer angular momentum in this process
(which magnifies the effective spin of a single photon by passing it
many times through a HW plate.) Hence, we will know it was a linear
photon instead of either RH or LH circular due to no angular momentum
transferred to the plate. That violates the current QM postulates that
only yes-know answers to eignestate queries can be found (e.g., prism
that directs LH one way, RH another way, but linear has 50-50 chance of
going either way.)
Repeating Rifle
1122508620....@o13g2000cwo.googlegroups.com, "Neil"
<para...@mailcity.com> wrote:
>
>
> Repeating Rifle wrote:
>> On 7/26/05 4:07 PM, in article
>> 1122419233....@g43g2000cwa.googlegroups.com, "Neil"
>> <para...@mailcity.com> wrote:
> ...
>> There is no paradox. To the extent that there is energy "lost," consider
>> (and I will coin some terms) an angular Compton or Doppler effect.
>
> Yes, there is. Every photon has hbar of spin, but their energies vary.
> If the photon energy is low enough, the amount of energy added to the
> disk that's mandated by conservation of angular momentum exceeds the
> energy of the photon (clearly, since photon energy can be as low as we
> wish.) In that case, Galstian's concept of red shift of the photon
> won't be enough to balance the total energy.
That is true only for circularly polarized photons. The spin shows up only
if you measure it. If you have linearly polarized light, you have a
superposition of right and left circularly polarized photon states. Only
when you start to think in terms of amplitudes and interference, will you be
able to understand the key quantum concept. You are not there yet.
p.ki...@ic.ac.uk
> does anyone known how to define a 'single photon'????
> does anyone known what is a single photon??
This is easy. A photon is a single excitation of the quantised
harmonic oscillator inside a field mode. A field mode is a
normalised solution of Maxwells equations.
There have been (what seems to be) innumerable threads in
sci.physics.research about "what is a photon"; I suggest you
read them.
As regards "phase", the original topic of this thread,
I always refer to:
D.T. Pegg, S.M. Barnett
Phys. Rev. A39, (n4), (1989).
"Phase properties of the quantised single mode electromagnetic field"
--
---------------------------------+---------------------------------
Dr. Paul Kinsler
Blackett Laboratory (QOLS) (ph) +44-20-759-47520 (fax) 47714
Imperial College London, Dr.Paul...@physics.org
SW7 2BW, United Kingdom. http://www.qols.ph.ic.ac.uk/~kinsle/
Neil
Repeating Rifle wrote:
> On 7/27/05 4:57 PM, in article
> 1122508620....@o13g2000cwo.googlegroups.com, "Neil"
> <para...@mailcity.com> wrote:
>
> >
> >
> > Repeating Rifle wrote:
> >> On 7/26/05 4:07 PM, in article
> >> 1122419233....@g43g2000cwa.googlegroups.com, "Neil"
> >> <para...@mailcity.com> wrote:
> > ...
> >> There is no paradox. To the extent that there is energy "lost," consider
> >> (and I will coin some terms) an angular Compton or Doppler effect.
> >
> > Yes, there is. Every photon has hbar of spin, but their energies vary.
> > If the photon energy is low enough, the amount of energy added to the
> > disk that's mandated by conservation of angular momentum exceeds the
> > energy of the photon (clearly, since photon energy can be as low as we
> > wish.) In that case, Galstian's concept of red shift of the photon
> > won't be enough to balance the total energy.
>
> That is true only for circularly polarized photons. The spin shows up only
> if you measure it. If you have linearly polarized light, you have a
> superposition of right and left circularly polarized photon states. Only
> when you start to think in terms of amplitudes and interference, will you be
> able to understand the key quantum concept. You are not there yet.
> >
Not where? Actually, you are wrong since an LP photon has a 50% chance
of transferring (+) sping and a 50% chance of transferring (-) spin per
a given interaction. But for the sake of argument lets put LP photons
aside. In any case, the effect is certainly true for CP photons, so if
we use them there is a paradox, agreed? All it takes is for something,
not everthing, to produce an anomalous result and we have a paradox. So
if I send CP photons of very low energy through a disk that can flip
their spins, there is an excess of energy added to the disk that can't
be compensated for by losing even the entire energy of a photon - see?
> >>> ...
> >
> >>
> >> All you have with linearly polarized photon is zero angular momentum before
> >> passing through a HWP and zero angular momentum after it goes through the
> >> plate. That does not require an exchange of angular momentum.
> >>
> > Yes, that's the whole point and the reason it creates a paradox. The
> > linear photon will fail to transfer angular momentum in this process
> > (which magnifies the effective spin of a single photon by passing it
> > many times through a HW plate.) Hence, we will know it was a linear
> > photon instead of either RH or LH circular due to no angular momentum
> > transferred to the plate. That violates the current QM postulates that
> > only yes-know answers to eignestate queries can be found (e.g., prism
> > that directs LH one way, RH another way, but linear has 50-50 chance of
> > going either way.)
> >
This is a good paradox too - comments?
Neil
<p.ki...@ic.ac.uk> wrote in message
news:sm4nr2-...@delillo.lsr.ph.ic.ac.uk...
Paul - Thanks for the input. So, what do you think of my original question
and etc?
p.ki...@ic.ac.uk
> However, does anything forbid a
> stream of single photons (say one every ms) from having rather consistent
> relative phases in some sense? (I mean, each photon is somehow no more than
> some fraction of radian of effective phase different from its neighbors,
> each of which is beyond the average coherence length for single photons.)
It sounds to me like you want photon that are both phase coherent
and anti-bunched. Since phase coherent demands careful phase
control, and anti-bunching demands number-fluctuation control,
your two requirements are at odds with one another.