The speed of a photon
Phil Gardner
results obtained would seem to offer some support to the hypothesis
that:
Any single photon moving in a non-absorbent dielectric medium moves at a
speed that is continuously less than c. Its instantaneous speed is some
unknown function of the wave functions of all the atoms and ions in the
medium. In an isotropic medium its mean speed is c/n.
Is there any hard experimental evidence for the alternative of
intermittent motion at a speed of c with repeated absorption and
emission which always preserves a "memory" of the trajectory before each
absorption?
--
Phil Gardner <pej...@oznetcom.com.au>
Hans Aberg
(If I understood the question correctly...)
The photons always travel with speed c, the speed of light in vacuum. In a
medium where the speed of light v is said to be lower than c, the photons
repeteadly hit an atom, causing an electron to be exited for a moment
(absorbing the photon), whereafter it after a short while drops down,
re-emitting a photon with exactly the same energy-momentum. -- There is a
law of preservation of the energy-momentum (if the gravitation is weak),
amking it appear as it is the same photon, moving slower. -- According to
the laws of physics, it is impossible to put a mark on the original
photon, and to check whether it was really that one, or a new one that
arrived at the final spot.
Was this what you was asking about?
Hans Aberg * Anti-spam: remove "remove." from email address.
* Email: Hans Aberg <remove...@member.ams.org>
* Home Page: <http://www.matematik.su.se/~haberg/>
* AMS member listing: <http://www.ams.org/cml/>
Phil Gardner
>
>
> The photons always travel with speed c, the speed of light in vacuum. In a
> medium where the speed of light v is said to be lower than c, the photons
> repeteadly hit an atom, causing an electron to be exited for a moment
> (absorbing the photon), whereafter it after a short while drops down,
> re-emitting a photon with exactly the same energy-momentum. -- There is a
> law of preservation of the energy-momentum (if the gravitation is weak),
> amking it appear as it is the same photon, moving slower. -- According to
> the laws of physics, it is impossible to put a mark on the original
> photon, and to check whether it was really that one, or a new one that
> arrived at the final spot.
>
> Was this what you was asking about?
>
I am well aware that this is what is commonly aaserted, invariably
without any citation of any experiments that support it.
I ask agsin, are there any at all that support the claim that Nature has
really opted for the messiest, most commplicated mechanism ever dreamed
up by human minds for slowing down the speed of a photon in a medium,
when far simpler alteernatives were available to her?
--
Phil Gardner <pej...@oznetcom.com.au>
Armin Lambacher
> I ask agsin, are there any at all that support the claim that Nature has
> really opted for the messiest, most commplicated mechanism ever dreamed
> up by human minds for slowing down the speed of a photon in a medium,
> when far simpler alteernatives were available to her?
> --
Hmm, I once saw a calculation of the speed of a plan wave in a
dielectricum based solely on the polarizability of the dielectricum,
i. e. no absorption and reemission processes were necessary. The
incoming wave generates a polarization at a small volume element, this
element then generates a secondary wave which interferes with the
incoming wave. If you add up all the waves you end with a plane wave
with the same amplitude as the incoming wave but which has a slight
delay. This delay is the same as one would calculate just from the
speed of light of the medium. So there are no assumptions made at all
about the nature of the medium, it just has to be homogenous on the
scale of wavelengths and has to have a polarizability, so that means
no 'messy' mechanisms.
Armin
Harry Johnston
> I am well aware that this is what is commonly aaserted, invariably
> without any citation of any experiments that support it.
>
> I ask agsin, are there any at all that support the claim that Nature has
> really opted for the messiest, most commplicated mechanism ever dreamed
> up by human minds for slowing down the speed of a photon in a medium,
> when far simpler alteernatives were available to her?
(Such as?)
I understand (without, admittedly, having actually seen it done) that
using this model we can calculate from first principles the effective
speed of light in various mediums, and this always matches (to within
the appropriate error bounds) the speed actually measured. I believe
that this can be considered to be experimental confirmation.
It should also be mentioned that for the effective speed of light to
*not* be affected by a medium in this manner, an awful lot of other
theories would have to be wrong - most of which have been quite
thoroughly tested over the years.
Put in the most basic (and approximate) terms: we know that atoms do
absorb photons, because otherwise everything would be transparent; we
know that in a transparent substance the atoms reemit the photons
again, otherwise the substances wouldn't be transparent; and it is
physically unreasonable to suppose that the photons can be absorbed
and reemitted *simultaneously*. The only way out would be to propose
a mechanism by which atoms in a transparent substance don't absorb the
photons, even though the *exact same atoms* when formed into a
non-transparent substance *do* - now *that* would be a truly messy
model!
Harry.
---
Harry Johnston, om...@ihug.co.nz
One Earth, One World, One People
Phil Gardner
>
> ..... I once saw a calculation of the speed of a plan wave in a
> dielectricum based solely on the polarizability of the dielectricum,
> i. e. no absorption and reemission processes were necessary. The
> incoming wave generates a polarization at a small volume element, this
> element then generates a secondary wave which interferes with the
> incoming wave. If you add up all the waves you end with a plane wave
> with the same amplitude as the incoming wave but which has a slight
> delay. This delay is the same as one would calculate just from the
> speed of light of the medium. So there are no assumptions made at all
> about the nature of the medium, it just has to be homogenous on the
> scale of wavelengths and has to have a polarizability, so that means
> no 'messy' mechanisms.
>
I thought it was now generally accepted that the "electomagnetic waves"
of classical optics are statistical descriptions of the time-averaged
distribution of photons, valid only when all the photons are of the same
energy, and nothing more.
My hypothesis runs as follows:
Like the fermions in Bohmian quantum mechanics (and any classical
particle) each photon has all the time a precisely defined position and
momentum but there is no pilot wave.
Every ground state atom has a wave function approximately of form, X =
exp(-r/a), exactly so for the H atom.
In its interaction with any such atom the speed, v, of any photon is
given by
c/v = 1 + a^2 del^2 (k*X),
where k is a dimensionless constant, an as yet undetermined function of
the atom's A and Z.
Absorption occurs only in the limit as r goes to zero, at which v = 0.
Phil Gardner <pej...@oznetcom.com.au>
Hans Aberg
<om...@ihug.co.nz> wrote:
>Phil Gardner <pej...@oznetcom.com.au> wrote:
>> I ask agsin, are there any at all that support the claim that Nature has
>> really opted for the messiest, most commplicated mechanism ever dreamed
>> up by human minds for slowing down the speed of a photon in a medium,
>> when far simpler alteernatives were available to her?
>
>(Such as?)
>
>I understand (without, admittedly, having actually seen it done) that
>using this model we can calculate from first principles the effective
>speed of light in various mediums, and this always matches (to within
>the appropriate error bounds) the speed actually measured. I believe
>that this can be considered to be experimental confirmation.
I think this is the very core of it: One can compute on it, and the
computations agree with experimental evidence.
Strictly speaking, physics deals only with theories about reality. One
requirement is that they can be "experimentally verified".
However, no theory of any generality can be verified in full, in view that
every experiment or computation using the theory is special. So there is
always the possibility that a theory is wrong.
Of course, it happens from time to time that one discovers that physical
theories, if not wrong, that they are flawed or must be augmented or
modified. For example, after Newtonian mechanics, one discovered
thermodynamics, quantum mechanics, and relativity.
Or one should perhaps better say that one created those theories: It is
always good to keep in mind that those are theories created by humans, so
that one may be aware of the fact that they might be in the need of
updating. The word "discover" only signifies an old 19th century view that
the physics theories created somehow are identified with reality.
So, returning to the original question, just go ahead, creating your own
favorite theory:
You could, for example, declare that the atoms in a media have small
fairies that slow down the photons. The interesting part is when you
should show that your theory has the same capacities as already existing
theories. That is, if you can carry out any better computations when
predicting experiments. And if the theory about fairies does not seem
necessary for the computations of the theory you made, people may say that
they prefer to apply Occam's razor, and excise that assumption.
So if you try to beat the existing theory by an alternative along the
lines that you suggest, I gather that you will find it difficult to do so.
Otherwise, there is nothing wrong with theories that do not correspond
exactly with what happens in reality (that is, relative experiments and
theories of better fine-tuned character): For example, in doped
semi-conductors, one may speak about electrons and "holes". The "holes"
are not really physical quantities, but the absence of an electron in an
orbit of an atom. The hole will behave as though there were positrons
present in the media, but this is not the case. It is also possible to
make computations using the hole theory. In fact, the hole theory is a
precursor to better QM theories.
Then, one does use the model that the light slows down in a media, and it
is possible to describe things like the Cerenkov effect by that (one gets
a light bang, just as one gets a sound bang of airplanes). But the photon
always travelling at the speed c in all time-space frames is a more
detailed theory.
Phil Gardner
>
> I understand (without, admittedly, having actually seen it done) that
> using this model we can calculate from first principles the effective
> speed of light in various mediums,
Does any statement of all the assumptions made in any such derivation
exist in the literature? Can anyone cite it?
>
> Put in the most basic (and approximate) terms: we know that atoms do
> absorb photons, because otherwise everything would be transparent; we
> know that in a transparent substance the atoms reemit the photons
> again,
"know" ??? Surely you mean assert, assert without any direct
exerimental support.
> otherwise the substances wouldn't be transparent; and it is
> physically unreasonable to suppose that the photons can be absorbed
> and reemitted *simultaneously*. The only way out would be to propose
> a mechanism by which atoms in a transparent substance don't absorb the
> photons, even though the *exact same atoms* when formed into a
> non-transparent substance *do* - now *that* would be a truly messy
> model!
>
I was talking of photons moving in a non-absorbent dielectric medium, ie
of photons with energies well below the energy threshold for ground
state atoms (of the order of 75% of the ionization energy). Surely
quantum mechanics says that for all such atoms the absorption is exactly
zero. So the postulated absorption is presumably (this is never spelled
out) absorbed by a single atomic electron. This is done without any of
the scattering that would occur with a free electron. The electron
that absorbs the photon is intelligent enough to record in its memory
bank the photon's momentum and to start a timing mechanism. After an
interval determined by the A and Z of the atom it somehow manages to
emit a new photon with exactly the momentum it has recorded. You don't
consider that messy and complicated?
Phil Gardner <pej...@oznetcom.com.au>
kin...@hfwork1.tn.tudelft.nl
> I was talking of photons moving in a non-absorbent dielectric medium, ie
> of photons with energies well below the energy threshold for ground
> state atoms (of the order of 75% of the ionization energy). Surely
> quantum mechanics says that for all such atoms the absorption is exactly
> zero. So the postulated absorption is presumably (this is never spelled
> out) absorbed by a single atomic electron. This is done without any of
> the scattering that would occur with a free electron. The electron
> that absorbs the photon is intelligent enough to record in its memory
> bank the photon's momentum and to start a timing mechanism. After an
> interval determined by the A and Z of the atom it somehow manages to
> emit a new photon with exactly the momentum it has recorded. You don't
> consider that messy and complicated?
What I would think is happening is something akin to this: that the
incident photon is not absorbed and re-emitted, but that the photon
and atom couple together and become (temporarily) a
photon/excited-atom superposition, which in some sense moves slower
than c. The many transient superpositia that occur as the photon
moves through different atoms in the material has a net effect: that
the photon appears to move slower than c.
--
------------------------------+-----------------------------------------
Dr. Paul Kinsler |
Department of Applied Physics | Dr.Paul...@physics.org
Faculty of Applied Sciences |
Technical University Delft | http://terahertz.tn.tudelft.nl/kinsler/
Lorentzweg 1, 2628 CJ DELFT |
The Netherlands V
Harry Johnston
>> Put in the most basic (and approximate) terms: we know that atoms do
>> absorb photons, because otherwise everything would be transparent; we
>> know that in a transparent substance the atoms reemit the photons
>> again,
> "know" ??? Surely you mean assert, assert without any direct
> exerimental support.
If there are as many photons leaving the media as entering it, the
atoms *must* be emitting the same number of photons as they absorb.
That's just common sense.
> I was talking of photons moving in a non-absorbent dielectric medium, ie
> of photons with energies well below the energy threshold for ground
> state atoms (of the order of 75% of the ionization energy). Surely
> quantum mechanics says that for all such atoms the absorption is exactly
> zero. So the postulated absorption is presumably (this is never spelled
> out) absorbed by a single atomic electron. This is done without any of
> the scattering that would occur with a free electron. The electron
> that absorbs the photon is intelligent enough to record in its memory
> bank the photon's momentum and to start a timing mechanism. After an
> interval determined by the A and Z of the atom it somehow manages to
> emit a new photon with exactly the momentum it has recorded. You don't
> consider that messy and complicated?
I would, if that were how it worked!
My (rough) understanding is that the atom absorbs the photon, enters
an unstable higher energy state, and as a result emits a photon after
a brief delay. The new photon has the same energy as the original
(because of conservation of energy) but can be headed in any
direction.
Any of the atoms might have absorbed and reemitted the photon,
however, so all the possible paths have to be summed quantum
mechanically to get the overall result. The paths in which the photon
has changed direction cancel one another out, provided that the atoms
are arranged in a crystaline structure. Nice and simple.
Hans Aberg
<pej...@oznetcom.com.au> wrote:
>The electron
>that absorbs the photon is intelligent enough to record in its memory
>bank the photon's momentum and to start a timing mechanism. After an
>interval determined by the A and Z of the atom it somehow manages to
>emit a new photon with exactly the momentum it has recorded. You don't
>consider that messy and complicated?
The explanation is not all messy: Assuming that there is no strong
gravitation present, then the total energy-momentum of the atom plus
photon is preserved when the electron is absorbed. When a photon later is
emitted, it must by QM have the same energy, and preservation of
energy-momentum is preserved, so the effect is as that of a single photon
that is delayed for a moment by the atom. (The preservation of the
energy-momentum is a special relativity extension to Newtonian physics.)
In addition, the internal "clock" that the electron uses in order to
decide when to emit a photon, there is a statistical fluctuation, and QM
describes that. And I gather it is possible to measure that, too.
In addition, devises such as atomic clocks are built up around these
theories. Via QM, the frequencies that are absorbed can be computed, and
these show up as spectral lines in the in the light spectrum of the atoms:
one can heat them up, and they start to radiate photons on these spectral
lines. Alternatively, one can shine light on the atoms, and observe that
these frequencies are absorbed (so, evidently not all atoms re-emit a
photon). Etc., etc.
So these theories can explain many phenomena. They have over the last
hundred years or so successively built up to explain these physical
phenomena.
Can your, simpler theory, explain all that?
Steve McGrew
>Phil Gardner <pej...@oznetcom.com.au> wrote:
>> I was talking of photons moving in a non-absorbent dielectric medium, ie
>> of photons with energies well below the energy threshold for ground
>> state atoms (of the order of 75% of the ionization energy). Surely
>> quantum mechanics says that for all such atoms the absorption is exactly
>> zero. So the postulated absorption is presumably (this is never spelled
>> out) absorbed by a single atomic electron. This is done without any of
>> the scattering that would occur with a free electron. The electron
>> that absorbs the photon is intelligent enough to record in its memory
>> bank the photon's momentum and to start a timing mechanism. After an
>> interval determined by the A and Z of the atom it somehow manages to
>> emit a new photon with exactly the momentum it has recorded. You don't
>> consider that messy and complicated?
>
>What I would think is happening is something akin to this: that the
>incident photon is not absorbed and re-emitted, but that the photon
>and atom couple together and become (temporarily) a
>photon/excited-atom superposition, which in some sense moves slower
>than c. The many transient superpositia that occur as the photon
>moves through different atoms in the material has a net effect: that
>the photon appears to move slower than c.
There are other ways to look at this that might be more
informative, that from the single-photon perspective. If we think of
the photon as a classical plane wave with a given frequency, and an
electron it encounters as being a charged harmonic oscillator, the
oscillator's response to the driving frequency depends on whether the
driving frequency is above or below the resonant frequency of the
oscillator. The charge in the oscillator is driven by the incoming
plane wave, but responds out of phase by some amount. Because the
charge is accelerated, it radiates another wave with a frequency and
phase that depends on its response, not directly on the driving force.
Consequently, the radiated wave is out of phase with the incoming
plane wave by some amount unless it happens to be precisely in
resonance with the oscillator.
The absorption and re-emission of a photon is directly
analogous, with one crucial proviso: any experiment set up to measure
the exact time of emission of the photon will get a definite answer
that is probabilistic, rather than a spread-out time of the sort that
a classical model would predict.
Steve
Phil Gardner
> The explanation is not all messy: Assuming that there is no strong
> gravitation present, then the total energy-momentum of the atom plus
> photon is preserved when the electron is absorbed. When a photon later
> is emitted, it must by QM have the same energy,
If, as you assert, absorption occurs (by some process not known to QM)
and if the momentum of the atom+photon is non-zero, which is invariably
the case in a monatomic gas, energy-momentum conservation would force
changes in the momentum of the photon, and with it scattering in almost
all cases.
> ...........................................................
> So these theories can explain many phenomena. They have over the last
> hundred years or so successively built up to explain these physical
> phenomena.
>
> Can your, simpler theory, explain all that?
My hypothesis, posted on March 4 and carefully ignored by all readers of
this thread, makes no attempt to do what QM does well. But it does not
conflict with QM and it provides a far simpler mechanism for reducing
the speed of a photon in a dielectric medium than any other on offer.
Is no one prepared to refute it, to cite any grounds for rejecting it?
Phil Gardner
Jim Carr
Phil Gardner <pej...@oznetcom.com.au> writes:
>I ask again, are there any at all that support the claim that Nature has
>really opted for the messiest, most commplicated mechanism ever dreamed
>up by human minds for slowing down the speed of a photon in a medium,
>when far simpler alteernatives were available to her?
My answer would be the Lamb Shift, as confirmation of a prediction
of QED, and that I don't find the mechanism of absorption and
reemission all that complicated apart from the complications of
condensed matter itself.
My comment would be that I saw your answer to a question about
what you thought the "simpler alternative" would be and found
it quite unsatisfactory.
--
James Carr <j...@scri.fsu.edu> http://www.scri.fsu.edu/~jac/
Dopeler Effect: The tendency of stupid ideas to seem smarter when they
come at you rapidly. (anon source via e-chain-letter)
Phil Gardner
>
>
> If there are as many photons leaving the media as entering it, the
> atoms *must* be emitting the same number of photons as they absorb.
> That's just common sense.
>
Agreed. But it is equally true if there is no absorption at all.
> My (rough) understanding is that the atom absorbs the photon, enters
> an unstable higher energy state, and as a result emits a photon after
> a brief delay. The new photon has the same energy as the original
> (because of conservation of energy) but can be headed in any
> direction.
>
You keep insisting that a ground state atom caa absorb a low energy
phhoton, eg a helium atom (ionization energy 24+ ev) aabsorbing a 2 ev
photon, when quantum mechanics and all the experimental evidence say it
can not. What do you base this assertion on? Can you cite a single
reference that supports it?
Phil Gardner <pej...@oznetcom.com.au>
kin...@hfwork1.tn.tudelft.nl
> There are other ways to look at this that might be more
> informative, that from the single-photon perspective. If we think of
> the photon as a classical plane wave with a given frequency, and an
> electron it encounters as being a charged harmonic oscillator, the
> oscillator's response to the driving frequency depends on whether the
> driving frequency is above or below the resonant frequency of the
> oscillator. The charge in the oscillator is driven by the incoming
> plane wave, but responds out of phase by some amount. Because the
> charge is accelerated, it radiates another wave with a frequency and
> phase that depends on its response, not directly on the driving force.
> Consequently, the radiated wave is out of phase with the incoming
> plane wave by some amount unless it happens to be precisely in
> resonance with the oscillator.
You can't do just this: your plane wave extends throughout all
space, and hence is interacting with all the atoms at the same time,
and further, isn't starting from point A and then travelling to point
B; and it's the speed from point A to point B which is the issue here.
If you want photons slowed by a medium described in a photon+atom kind
of way, you need to have your photons in pulses. At the very least,
you need to work out your phase shifts over a range of plane wave
frequencies, then use that range of frequencies to build a photon
pulse, then see how fast the pulse would travel.
However, we might assume that to a good approximation a novel quantum
state like squeezed light will propagate through this medium without
getting its quantum coherences scrambled. This means we have to do
more than worry about what happens to the pulse of plane waves; we
have think about how if one photon is transferred briefly to an atom
(in some sense), then it gets back into the pulse just as it was.
You can't do that with a classical picture.
Hans Aberg
<pej...@oznetcom.com.au> wrote:
>> The explanation is not all messy: Assuming that there is no strong
>> gravitation present, then the total energy-momentum of the atom plus
>> photon is preserved when the electron is absorbed. When a photon later
>> is emitted, it must by QM have the same energy,
>
>If, as you assert, absorption occurs (by some process not known to QM)
There is a funny story which Feynman told that his dad wanted to know if
the college tuition money was well spent. So when Feynman explained what
he had learned, his dad asked him why the photon is absorbed and
re-emitted, Feynman could not explain that. So his dad started to wonder
if it was really worth to spend that college tuition, when his son could
not answer even such a simple question.
It is a fact that physics only deals developing theories about
observations, and one cannot provide any explanations beyond that.
It is however possible to observe absorption and remission of photons,
because one can send in photons singly into a media, and check if they
come out. A classical QM experiment consists of sending photons one by one
onto a slit, and observe the interference patterns. Without QM, it is not
possible to explain that phenomenon.
So it is not currently possible to give a better explanation of why the
photon is absorbed and re-emitted, because there is no theory explaining
exactly what happens in the absorption and re-emitting moment. -- This is
a shortage of the current QM theories.
But it is possible to observe the fact that the photons are indeed
absorbed and re-emitted.
>and if the momentum of the atom+photon is non-zero, which is invariably
>the case in a monatomic gas, energy-momentum conservation would force
>changes in the momentum of the photon, and with it scattering in almost
>all cases.
I am not sure what you are saying. Light does almost invariable scatter to
some degree when passing through a medium.
A funny thing is that it is also possible, to some degree reverse, the
process: The latest telescopes uses a laser beam so compute the current
distortions made by the atmosphere, and then one can compensate for that
to some degree.
>> So these theories can explain many phenomena. They have over the last
>> hundred years or so successively built up to explain these physical
>> phenomena.
>>
>> Can your, simpler theory, explain all that?
>
>My hypothesis, posted on March 4 and carefully ignored by all readers of
>this thread, makes no attempt to do what QM does well. But it does not
>conflict with QM and it provides a far simpler mechanism for reducing
>the speed of a photon in a dielectric medium than any other on offer.
>Is no one prepared to refute it, to cite any grounds for rejecting it?
Well, if you think that you have a theory that in alternate way can
provide computations, then it is a valid theory as anyone else. You can
write it up and make it public, making it available to others.
Hans Aberg
wrote:
>> It is however possible to observe absorption and remission of photons,
>> because one can send in photons singly into a media, and check if they
>> come out. A classical QM experiment consists of sending photons one by one
>> onto a slit, and observe the interference patterns. Without QM, it is not
>> possible to explain that phenomenon.
>
I am not sure what you mean: Without QM (a physical theory), the
phenomenon can still be observed. Thus, one must develop some kind of
theory.
J. J. Lodder
> It is however possible to observe absorption and remission of photons,
> because one can send in photons singly into a media, and check if they
> come out. A classical QM experiment consists of sending photons one by one
> onto a slit, and observe the interference patterns. Without QM, it is not
> possible to explain that phenomenon.
Without QM the problem does not exist :-)
J. J. Lodder
> In article <1eqazyv.1yz...@de-ster.demon.nl>, j...@de-ster.demon.nl
> I am not sure what you mean: Without QM (a physical theory), the
> phenomenon can still be observed.
We just learned in this group that that is contrary to historical fact:
Nobody thought of investigating interference at low intensitity
before Einstein (1905) introduced photons,
and someone (GI Taylor) (communication of Ken Zetie in spr)
did it almost immediately after (1908)
Best,
Jan
Phil Gardner
>
> .......................................... you (Phil Gardner)claim that
> the photon has a precisely defined position and
> momentum, which is in fundamental conflict with both the principles of
> quantum mechanics and a *lot* of experimental evidence.)
>
What evidence? A well defined position and momentum (our knowledge of
which is always limited by the uncertainty principle) is implicit in the
Bohm interpretation of quantum mechanics which agreed with all the
experimental resultls just as well as any other interpretation.
Phil Gardner <pej...@oznetcom.com.au>
Harry Johnston
> So it is not currently possible to give a better explanation of why the
> photon is absorbed and re-emitted, because there is no theory explaining
> exactly what happens in the absorption and re-emitting moment. -- This is
> a shortage of the current QM theories.
This statement may be misleading to some readers. It is true that
there is currently no explanation at a fundamental level for the
ability of charged particles to absorb and/or emit photons, i.e., for
the fact that the charge and photon fields interact. Of course, the
fact that they *do* interact is trivially verifiable [1].
However, when dealing with a particular atom in a particular
electromagnetic field, e.g., the atoms in a piece of glass and a beam
of light, QED allows a detailed description of the process. This
description is sufficiently accurate to allow us to calculate the
effect of the glass on the beam of light and vice versa, which can be
experimentally verified.
Harry.
[1] The detailed nature of the interaction (within the confines of
special relativity, of course) can also be experimentally verified,
although it is much harder to do. I belive the Lamb shift has already
been mentioned.
Phil Gardner
> It is however possible to observe absorption and remission of photons,
> because one can send in photons singly into a media, and check if they
> come out. A classical QM experiment consists of sending photons one by
> one onto a slit, and observe the interference patterns. Without QM, it
> is not possible to explain that phenomenon.
I would claim that exactly the same diffraction/interference pattern is
obtained by assuming that there is no absorption/emission and that the
mean energy flow (time averaged photon flux) is given by the classical
(time-independent) wave equation. And predictions made that way call
for less mathematical effort and make better physical sense.
> But it is possible to observe the fact that the photons are indeed
> absorbed and re-emitted.
In forty years of reading physics literature I have found no mention of
any experiment that claims to have done this - to have observed the
absorption of a photon followed by delayed emission of another photon
with exactly the same energy and momentum. Can you cite one, just one?
Phil Gardner <pej...@oznetcom.com.au>
Hans Aberg
wrote:
>> I am not sure what you mean: Without QM (a physical theory), the
>> phenomenon can still be observed.
>We just learned in this group that that is contrary to historical fact:
>Nobody thought of investigating interference at low intensitity
>before Einstein (1905) introduced photons,
>and someone (GI Taylor) (communication of Ken Zetie in spr)
>did it almost immediately after (1908)
Are you aware of that what you are saying is that nature changes as we go
along inventing physical theories about it! -- Few believe that these
days. :-)
You probably mean that already developed physical theories can lead to
ideas of new experiments, but it is not the same thing as saying that
those new experiments cannot be observed without the old theories: In
fact, often the opposite happens, one tries to find experiments that
cannot be explained by the old theories. This then forces the need for
developing new theories that can also explain these new experiments.
Hans Aberg
<om...@ihug.co.nz> wrote:
>> So it is not currently possible to give a better explanation of why the
>> photon is absorbed and re-emitted, because there is no theory explaining
>> exactly what happens in the absorption and re-emitting moment. -- This is
>> a shortage of the current QM theories.
>>>This statement may be misleading to some readers. It is true that
there is currently no explanation at a fundamental level for the
ability of charged particles to absorb and/or emit photons, i.e., for
the fact that the charge and photon fields interact. Of course, the
fact that they *do* interact is trivially verifiable [1].
>>>However, when dealing with a particular atom in a particular
electromagnetic field, e.g., the atoms in a piece of glass and a beam
of light, QED allows a detailed description of the process. This
description is sufficiently accurate to allow us to calculate the
effect of the glass on the beam of light and vice versa, which can be
>experimentally verified.
Right. What I really mean is this: Current QM give descriptions of the kind
A _____________ B
Particles before --> | Interaction | --> Particles after
-------------
and there are no good descriptions of what happens at A and B.
What one really would want is a Schrodinger equation that in detail
describes what happens at time A and B. Then one plugs the electron moving
towards the atom into the equation, and see, by the solutions how it is
transformed into an electron at a higher energy state.
Harry Johnston
> In forty years of reading physics literature I have found no mention of
> any experiment that claims to have done this - to have observed the
> absorption of a photon followed by delayed emission of another photon
> with exactly the same energy and momentum. Can you cite one, just one?
Sigh. I've already explained that the emitted photon does NOT have
the same momentum as the absorbed photon. The wave equation takes
care of this for us! The effect is the same in either quantum or
classical electromagnetics: if you have a bunch of sources whose
(spherically symmetric) output is proportionate to the intensity of a
beam of light, they reproduce that same beam of light.
For example, this is how a hologram works; the atoms in the film don't
have to somehow send each individual photon in the same direction it
was going when the film was exposed! The interference pattern that is
formed reproduces the original image anyway. Much the same thing
happens to light passing through a transparent material.
Harry Johnston
Oh, yes, I suppose that's true - I forgot about Bohm. In my opinion,
though, this interpretation is pretty much washed up by Occam's Razor;
what is the value of giving particles a well defined position and
momentum which never has any effect on the observable results?
Hans Aberg
<pej...@oznetcom.com.au> wrote:
>> It is however possible to observe absorption and remission of photons,
>> because one can send in photons singly into a media, and check if they
>> come out. A classical QM experiment consists of sending photons one by
>> one onto a slit, and observe the interference patterns. Without QM, it
>> is not possible to explain that phenomenon.
>I would claim that exactly the same diffraction/interference pattern is
>obtained by assuming that there is no absorption/emission and that the
>mean energy flow (time averaged photon flux) is given by the classical
>(time-independent) wave equation. And predictions made that way call
>for less mathematical effort and make better physical sense.
It is know that the patterns agree with those of classical physics. The
problem though is that one discovered that photons do come in undividable
energy packets. Thus, according to classical physics, there is no way one
can get a diffraction pattern, as there is only one undividable energy
packet, that has to pass through one single slit.
QM solves this dilemma by the introduction of quantum fields, which is a
not directly observable quantity, but which is only used in explaining the
experiment via computations.
>> But it is possible to observe the fact that the photons are indeed
>> absorbed and re-emitted.
>In forty years of reading physics literature I have found no mention of
>any experiment that claims to have done this - to have observed the
>absorption of a photon followed by delayed emission of another photon
>with exactly the same energy and momentum. Can you cite one, just one?
One can check this independently, by absorbed and emitted spectral lines
which agree, and can be computed via QM.
It would however be interesting, to find direct experiment, like you are
asking for.
Harry Johnston
> Right. What I really mean is this: Current QM give descriptions of the kind
> A _____________ B
> Particles before --> | Interaction | --> Particles after
> -------------
> and there are no good descriptions of what happens at A and B.
>
> What one really would want is a Schrodinger equation that in detail
> describes what happens at time A and B. Then one plugs the electron moving
> towards the atom into the equation, and see, by the solutions how it is
> transformed into an electron at a higher energy state.
I'm slightly out of my depth now, but I think that you can use QED to
do something much like you describe. It is only the pointlike
interactions in the Feynman diagrams that are axiomatic; these
interactions are in a sense unreal anyhow because you always have to
sum over all the (relevant) diagrams to get a real answer. And, in a
way, there isn't a process to explain, precisely because these
interactions *are* pointlike! (Ignoring thorny issues like
renomalisation and quantum gravity, of course.)
If you start in a state with one electron and one photon, the QED
state can be described as a wavefunction Psi(e,p). The evolution
equation can probably be written in these terms, more or less: you
have to invent an extra value for p to represent the state where the
photon has been absorbed, and I guess you'd have to artificially
suppress any terms that would produce a second photon. It should
still be a good approximation.
Certainly if you were considering a particular experiment you could
solve the QED equations and then write a density matrix for the
electron (as a function of time) in order to visualise the effect of
the electromagnetic field on the electron.
Harry.
Paul Arendt
Harry Johnston <om...@ihug.co.nz> wrote:
[... about the picture of light slowing down in a medium due to
absorbtion and re-emission by the constituent atoms...]
>My (rough) understanding is that the atom absorbs the photon, enters
>an unstable higher energy state, and as a result emits a photon after
>a brief delay. The new photon has the same energy as the original
>(because of conservation of energy) but can be headed in any
>direction.
>
>Any of the atoms might have absorbed and reemitted the photon,
>however, so all the possible paths have to be summed quantum
>mechanically to get the overall result. The paths in which the photon
>has changed direction cancel one another out, provided that the atoms
>are arranged in a crystaline structure. Nice and simple.
Yes, that's a great way of putting it! The absorption events need
only be "virtual" to have an effect, and in fact if we took the pains
to observe every atom to see where and when the light was absorbed,
the light would take a very different path.
But your explanation needn't be so restrictive, actually. You don't
need a crystalline structure for it to work! For example, light
travels in straight lines in glass, too, which is pretty amorphous.
The catch is that visible light has a HUGE wavelength (tenths of
microns) compared with typical interatomic or intermolecular
distances in a solid (several tenths of *nanometers*, or slightly
more). So, many possible absorption sites have a chance to
constructively and destructively interfere over a single
wavelength of the light, even when there's no crystalline structure
involved. That's just the principle (Huygen's?) that says an
overall wavefront is the sum of a bunch of circular wavefronts
emitted from the individual atoms.
And in fact, when the wavelength gets short enough that the
crystal structure DOES matter, the refractive properties of the
material change dramatically, to diffractive/interfering patterns
(like you see in X-ray diffraction).
J. J. Lodder
> In article <1eqcrjx.oz0...@de-ster.demon.nl>, j...@de-ster.demon.nl
> wrote:
>
> >> I am not sure what you mean: Without QM (a physical theory), the
> >> phenomenon can still be observed.
>
> >We just learned in this group that that is contrary to historical fact:
> >Nobody thought of investigating interference at low intensitity
> >before Einstein (1905) introduced photons,
> >and someone (GI Taylor) (communication of Ken Zetie in spr)
> >did it almost immediately after (1908)
>
> Are you aware of that what you are saying is that nature changes as we go
> along inventing physical theories about it! -- Few believe that these
> days. :-)
Yes, I am aware, and no, I did not say that.
(BTW, are you aware of Van den Berg's Metabletica?)
You are talking about something different.
(unfortumately you snipped the context)
-Problems- are in the mind,
-phenomena- out there. (If you are a 'naive' realist)
If you never heard about quanta, it cannot make sense to investigate
how observed interference depends upon intensity, and the results of any
experiment along these lines cannot possibly be surprising.
Sure, you insert attenuating filters, and surely with a corresponding
increase of exposure time you still observe the same interference
pattern.
No 19th century physisist would be surprised,
nor anybody else before Einstein (1905).
In fact nobody even bothered to do the experiment, before 1905.
It is only -after- 1905 that the result -becomes- surprising.
> You probably mean that already developed physical theories can lead to
> ideas of new experiments, but it is not the same thing as saying that
> those new experiments cannot be observed without the old theories: In
> fact, often the opposite happens, one tries to find experiments that
> cannot be explained by the old theories. This then forces the need for
> developing new theories that can also explain these new experiments.
You have it entirely upside down:
it is new theory that gives rise to new experiment,
and to new wonder where on the old lines
things seemed too obvious to bother doing the experiment at all.
Best,
Jan
C. Cagle
<remove.haberg-2...@du131-226.ppp.su-anst.tninet.se>, Hans
Aberg <remove...@matematik.su.se> wrote:
> The photons always travel with speed c, the speed of light in vacuum. In a
> medium where the speed of light v is said to be lower than c, the photons
> repeteadly hit an atom, causing an electron to be exited for a moment
> (absorbing the photon), whereafter it after a short while drops down,
> re-emitting a photon with exactly the same energy-momentum. -- There is a
> law of preservation of the energy-momentum (if the gravitation is weak),
> amking it appear as it is the same photon, moving slower. -- According to
> the laws of physics, it is impossible to put a mark on the original
> photon, and to check whether it was really that one, or a new one that
> arrived at the final spot.
>
Considering that photons make their way through a material like glass
and then when received by a person's eye can be integrated into a
picture which is not materially different from that which would be
received without the intervening glass it seems that these are not new
photons that are being received but rather the original ones since they
contain the same information. Now if they aren't the same ones then
they are pretty good copies of the originals and are indeed able to
pass on the 'mark'. :-).