Who Says Light Behaves as a Particle?
Nathan Urban
> I'd just like to have someone show me proof, beyond the shadow of a doubt
> (pun intended), light behaving as a particle.
I'm told that spontaneous emission is pretty good proof.
[Note followups.]
Gregory L. Hansen
Of course, the photoelectric effect is what started it all.
For proof beyond a shadow of a doubt, he might need to get his own
photomultiplier and damp down a light source until the signal becomes
discrete.
--
No electrons were harmed in the posting of this message.
srp
No, but quite a bunch of them might be knocked silly if he follows
your instructions. :o]
André Michaud
Service de Recherche Pédagogique http://www.microtec.net/~srp/
Katie Schwarz
>Nathan Urban <nur...@vt.edu> wrote:
>>trans...@hotmail.com wrote:
>>
>>> I'd just like to have someone show me proof, beyond the shadow of a doubt
>>> (pun intended), light behaving as a particle.
>>
>>I'm told that spontaneous emission is pretty good proof.
>
>
>For proof beyond a shadow of a doubt, he might need to get his own
>photomultiplier and damp down a light source until the signal becomes
>discrete.
Actually neither of these is really solid proof. The photoelectric
effect can be explained by a semiclassical theory where the energy of
the atoms is quantized but the electromagnetic field is classical.
And spontaneous emission can be explained by a "neoclassical" theory
where there is a random classical electromagnetic field present
everywhere with some appropriate spectrum, inducing transitions in
atoms at random times. To rule out both of these theories, you need a
coincidence experiment, where a beam is split and detected by two
detectors:
detector #1 ----- coincidence
^ counter
| |
| |
| |
incident light-----------> / --------->detector #2
If exactly one photon falls on the beam splitter, it must go to
detector #1 or 2, but not both, so the coincidence count rate will be
zero. Classical theories predict that any wave can be split into two
smaller ones, so the coincidence rate would never be zero. This kind
of experiment supports the existence of photons.
--
Katie Schwarz
"There's no need to look for a Chimera, or a cat with three legs."
-- Jorge Luis Borges, "Death and the Compass"
Gregory L. Hansen
Katie Schwarz <k...@socrates.berkeley.edu> wrote:
>Gregory L. Hansen <glha...@steel.ucs.indiana.edu> wrote:
>>Nathan Urban <nur...@vt.edu> wrote:
>>>trans...@hotmail.com wrote:
>>>
>>>> I'd just like to have someone show me proof, beyond the shadow of a doubt
>>>> (pun intended), light behaving as a particle.
>>>
>>>I'm told that spontaneous emission is pretty good proof.
>>
>>Of course, the photoelectric effect is what started it all.
>>
>>For proof beyond a shadow of a doubt, he might need to get his own
>>photomultiplier and damp down a light source until the signal becomes
>>discrete.
>
>Actually neither of these is really solid proof. The photoelectric
>effect can be explained by a semiclassical theory where the energy of
>the atoms is quantized but the electromagnetic field is classical.
Are you sure? The reason the photoelectric effect was important because
they tried to explain it with classical electromagnetism by assuming
energy will build up over time until an electron is excited. But the
response of the electrons was immediate, and no intensity of light would
create the effect if it was the wrong color.
>And spontaneous emission can be explained by a "neoclassical" theory
>where there is a random classical electromagnetic field present
>everywhere with some appropriate spectrum, inducing transitions in
>atoms at random times. To rule out both of these theories, you need a
>coincidence experiment, where a beam is split and detected by two
>detectors:
>
> detector #1 ----- coincidence
> ^ counter
> | |
> | |
> | |
> incident light-----------> / --------->detector #2
>
>If exactly one photon falls on the beam splitter, it must go to
>detector #1 or 2, but not both, so the coincidence count rate will be
>zero. Classical theories predict that any wave can be split into two
>smaller ones, so the coincidence rate would never be zero. This kind
>of experiment supports the existence of photons.
Yeah, I think that would convince me.
Nathan Urban
> >Nathan Urban <nur...@vt.edu> wrote:
> >>> I'd just like to have someone show me proof, beyond the shadow of a doubt
> >>> (pun intended), light behaving as a particle.
> >>I'm told that spontaneous emission is pretty good proof.
> And spontaneous emission can be explained by a "neoclassical" theory
> where there is a random classical electromagnetic field present
> everywhere with some appropriate spectrum, inducing transitions in
> atoms at random times.
The last time this subject came up on sci.physics, I seem to recall
someone citing Milonni's book for an argument that sponteneous emission
could _not_ be explained by a "neoclassical" theory in this way.
However, not having read the book, I can't say whether this is true,
or whether I'm accurately representing Milonni's claim.
Bruce Richmond
glha...@steel.ucs.indiana.edu (Gregory L. Hansen) wrote:
> In article <7v5bjl$5gk$1...@agate-ether.berkeley.edu>,
> Katie Schwarz <k...@socrates.berkeley.edu> wrote:
> >>>
> >>>> I'd just like to have someone show me proof, beyond the shadow
of a doubt
> >>>> (pun intended), light behaving as a particle.
> >>>
> >>>I'm told that spontaneous emission is pretty good proof.
> >>
> >>
> >>For proof beyond a shadow of a doubt, he might need to get his own
> >>photomultiplier and damp down a light source until the signal
becomes
> >>discrete.
> >
> >Actually neither of these is really solid proof. The photoelectric
> >effect can be explained by a semiclassical theory where the energy of
> >the atoms is quantized but the electromagnetic field is classical.
>
> Are you sure? The reason the photoelectric effect was important
because
> they tried to explain it with classical electromagnetism by assuming
> energy will build up over time until an electron is excited. But the
> response of the electrons was immediate, and no intensity of light
would
> create the effect if it was the wrong color.
>
As far as the immediate response of the detector, isn't it possible
that a resonance built up so quickly that it could not be measured by
the equipment used?
How does particle theory explain the sensitivity to only certain
colors? It's easy for wave theory to explain. The atom's electrons
resonate at certain frequencies. Particle theory uses the number of
particles per second to establish the intensity. And we can have
different colors of the same intensity. So how can the same variable
also be used to explain color?
> >And spontaneous emission can be explained by a "neoclassical" theory
> >where there is a random classical electromagnetic field present
> >everywhere with some appropriate spectrum, inducing transitions in
> >atoms at random times. To rule out both of these theories, you need
a
> >coincidence experiment, where a beam is split and detected by two
> >detectors:
> >
> > detector #1 ----- coincidence
> > ^ counter
> > | |
> > | |
> > | |
> > incident light-----------> / --------->detector #2
> >
> >If exactly one photon falls on the beam splitter, it must go to
> >detector #1 or 2, but not both, so the coincidence count rate will be
> >zero. Classical theories predict that any wave can be split into two
> >smaller ones, so the coincidence rate would never be zero. This kind
> >of experiment supports the existence of photons.
>
> Yeah, I think that would convince me.
>
If the detectors are resonating and periodically discharging when they
reach a certain level, what are the chances that the same wave is going
to put them both over the edge at the same time? Also how do we know
that each detector is getting exactly half the energy from each wave?
--
Bruce Richmond
Sent via Deja.com http://www.deja.com/
Before you buy.
Gregory L. Hansen
The time it takes to build up enough energy has been calculated and didn't
agree with experiment. There's also the matter that when the electron
drops back down and releases energy, that energy is not released over time
and does not consist of a continuum of frequencies.
>How does particle theory explain the sensitivity to only certain
>colors?
The energy is proportional to the frequency.
>It's easy for wave theory to explain. The atom's electrons
>resonate at certain frequencies.
A resonance would give a response to one particular color, with a
decreasing response as the frequency increases or decreases. What's
observed in photoelectric experiments is that there is no photoelectric
effect at all when you shine light of too low a frequency on the plate, no
matter how bright the light is. But the smallest intensity of
sufficiently blue light creates a response that continues for increasingly
higher frequencies.
>Particle theory uses the number of
>particles per second to establish the intensity. And we can have
>different colors of the same intensity. So how can the same variable
>also be used to explain color?
Energy of a photon is E=hf, Planck's constant multiplied by the frequency.
Total energy of light in a region is U=nE, number of photons multiplied by
the energy of a photon. Intensity is I=U/A=nE/A=nhf/A, energy divided by
area.
If there are many different colors you need to sum over all the colors.
I = (n1 f1 + n2 f2 + n3 f3 + ...) h / A
>> >And spontaneous emission can be explained by a "neoclassical" theory
>> >where there is a random classical electromagnetic field present
>> >everywhere with some appropriate spectrum, inducing transitions in
>> >atoms at random times. To rule out both of these theories, you need
>a
>> >coincidence experiment, where a beam is split and detected by two
>> >detectors:
>> >
>> > detector #1 ----- coincidence
>> > ^ counter
>> > | |
>> > | |
>> > | |
>> > incident light-----------> / --------->detector #2
>> >
>> >If exactly one photon falls on the beam splitter, it must go to
>> >detector #1 or 2, but not both, so the coincidence count rate will be
>> >zero. Classical theories predict that any wave can be split into two
>> >smaller ones, so the coincidence rate would never be zero. This kind
>> >of experiment supports the existence of photons.
>>
>> Yeah, I think that would convince me.
>>
>
>If the detectors are resonating and periodically discharging when they
>reach a certain level, what are the chances that the same wave is going
>to put them both over the edge at the same time? Also how do we know
>that each detector is getting exactly half the energy from each wave?
Use a beam splitter that passes exactly half the light energy in each
direction. Set up the experiment with bright light so a continuous signal
is produced. And you should use several different types of detectors, and
swap detector A with detector B to check for a bias. Ultimately your
analysis will involve some statistics. You want to determine if there is
a time dependence as you'd expect for energy to build up and release, or
if the time dependence is entirely random. And you'll want to check that
the detector firings are anti-correlated. That is, if you're passing one
photon at a time at a sufficiently slow rate, the detectors should almost
never fire anywhere near the same time. But a wave theory lets them do
that.
Matt Kennel
:Katie Schwarz <k...@socrates.berkeley.edu> wrote:
:>Gregory L. Hansen <glha...@steel.ucs.indiana.edu> wrote:
:>>Nathan Urban <nur...@vt.edu> wrote:
:>>>trans...@hotmail.com wrote:
:>>>
:>>>> I'd just like to have someone show me proof, beyond the shadow of a doubt
:>>>> (pun intended), light behaving as a particle.
:>>>
:>>>I'm told that spontaneous emission is pretty good proof.
:>>
:>>Of course, the photoelectric effect is what started it all.
:>>
:>>For proof beyond a shadow of a doubt, he might need to get his own
:>>photomultiplier and damp down a light source until the signal becomes
:>>discrete.
:>
:>Actually neither of these is really solid proof. The photoelectric
:>effect can be explained by a semiclassical theory where the energy of
:>the atoms is quantized but the electromagnetic field is classical.
:
:Are you sure? The reason the photoelectric effect was important because
:they tried to explain it with classical electromagnetism by assuming
:energy will build up over time until an electron is excited. But the
:response of the electrons was immediate, and no intensity of light would
:create the effect if it was the wrong color.
But the build-up of energy over time in a
classical-E-coupled-to-quantized-atom theory is about the atomic
size/c which is approximately the same as the fully quantized answer.
I asked a colleague who is an optics professor at Georgia tech who
confirmed that it is possible to demonstrate quantization of fields as
well as electrons with a photo-electric type of experiment, but it has
to be done carefully and analyzed carefully in the circumstance of
very dim emission so that the photon number is low.
Which after you think about it makes sense because quantized E fields
only behave differently than classical ones in just that circumstance.
:>And spontaneous emission can be explained by a "neoclassical" theory
:>where there is a random classical electromagnetic field present
:>everywhere with some appropriate spectrum, inducing transitions in
:>atoms at random times. To rule out both of these theories, you need a
:>coincidence experiment, where a beam is split and detected by two
:>detectors:
:>
:> detector #1 ----- coincidence
:> ^ counter
:> | |
:> | |
:> | |
:> incident light-----------> / --------->detector #2
:>
:>If exactly one photon falls on the beam splitter, it must go to
:>detector #1 or 2, but not both, so the coincidence count rate will be
:>zero. Classical theories predict that any wave can be split into two
:>smaller ones, so the coincidence rate would never be zero. This kind
:>of experiment supports the existence of photons.
:Yeah, I think that would convince me.
Me too---almost; one could imagine semiclassical theories such that
the cross section between classical light and the quantized atoms in the
detector would not permit coincidence.
I.e. given conservation of energy if the energy of the E field is
sufficiently small then there will be no atomic transition hence no
possibility for detection.
I think the basic answer comes down to the fact that the equations of motion
are distinct for quantized fields vs classical fields in the limit of
low photon number and you have to detect that.
The point is that the typical freshman physics description of the
photoelectric effect is somewhat wrong.
What it does demonstrate is that stronger EM waves of low frequency
cannot substitute for dimmer EM waves of higher frequency in producing
an atomic transition. Something surely has to be quantized then, but
it may not be the field.
--
* Matthew B. Kennel
* Institute For Nonlinear Science/University of California, San Diego
I would not, could not SAVE ON PHONE,
I would not, could not BUY YOUR LOAN,
I would not, could not MAKE MONEY FAST,
I would not, could not SEND NO CA$H,
I would not, could not SEE YOUR SITE,
I would not, could not EAT VEG-I-MITE,
I do *not* *like* GREEN CARDS AND SPAM! MAD-I-AM!
Katie Schwarz
>k...@socrates.berkeley.edu (Katie Schwarz) wrote:
>
>
>> And spontaneous emission can be explained by a "neoclassical" theory
>> where there is a random classical electromagnetic field present
>> everywhere with some appropriate spectrum, inducing transitions in
>> atoms at random times.
>
>someone citing Milonni's book for an argument that sponteneous emission
>could _not_ be explained by a "neoclassical" theory in this way.
>However, not having read the book, I can't say whether this is true,
>or whether I'm accurately representing Milonni's claim.
I ought to really read that book. Thanks for the incentive. After
skimming that section, as far as I understand it, the neoclassical
theory (due to Jaynes in the seventies) got the Lamb shift wrong.
Katie Schwarz
>>
>>The photoelectric
>>effect can be explained by a semiclassical theory where the energy of
>>the atoms is quantized but the electromagnetic field is classical.
>
>Are you sure? The reason the photoelectric effect was important because
>they tried to explain it with classical electromagnetism by assuming
>energy will build up over time until an electron is excited. But the
>response of the electrons was immediate,
That's what it says in Eisberg and Resnick's textbook, but on thinking
about it, this doesn't make sense to me: if energy comes in packages
instead of continuously, you still have to wait for a photon to come
along. I couldn't find anything about a time delay in semiclassical
treatments in books such as Loudon or Mandel & Wolf; anyone have
details?
> and no intensity of light would
>create the effect if it was the wrong color.
This can be explained by a semiclassical theory: Fermi's golden rule
gives a delta function of the frequency of the incident light, even if
you treat the incident light classically.
c.h.thompson
There seem to be some sensible ideas here, but I wonder how many people
realise just how important it is to the current state of fundamental
physics?
I mean, it is obviously important, but the consequences of what I regard as
a totally false view stretch far beyond the immediate issue. The
assumption of particle behaviour is, I have found, at the root of the
current official line re "quantum entanglement", "action at a distance",
"EPR correlations" or whatever. These have opened the door to corruption
and magic ...
Experiments such as Alain Aspect's have been interpreted as supporting
quantum weirdness, but if you look at the actual details you find that this
is only if you refuse to investigate properly the possibility that you might
be dealing with classical waves, not photons! Aspect and his colleagues did
do an experiment (on much the same lines as Katie suggested) to investigate
this, but it does not bear close scrutiny (See Grangier, P, Roger, G and
Aspect, A, Europhysics Letters 1, 173-179 (1986) and criticism by realists:
Marshall, T W and Santos, E, Europhysics Letters 3,
293-6, (1987). There have been others, e.g. Mizobuchi, Yutaka and Ohtake,
JJAP Series 9, 201-204 (1993)).
Katie Schwarz <k...@socrates.berkeley.edu> wrote
> Gregory L. Hansen <glha...@steel.ucs.indiana.edu> wrote:
> >Katie Schwarz <k...@socrates.berkeley.edu> wrote:
> >>
> >>The photoelectric
> >>effect can be explained by a semiclassical theory where the energy of
> >>the atoms is quantized but the electromagnetic field is classical.
> >
> >Are you sure? The reason the photoelectric effect was important because
> >they tried to explain it with classical electromagnetism by assuming
> >energy will build up over time until an electron is excited. But the
> >response of the electrons was immediate,
>
> That's what it says in Eisberg and Resnick's textbook, but on thinking
> about it, this doesn't make sense to me: if energy comes in packages
> instead of continuously, you still have to wait for a photon to come
> along. I couldn't find anything about a time delay in semiclassical
> treatments in books such as Loudon or Mandel & Wolf; anyone have
> details?
I don't think you will find the details in any text book, but there are
hints of them in published papers if you read between the lines!
And there's quite a lot in Alain Aspect's thesis (Paris Sud, 1983) but this
is not readily available. Besides, it is in French. I've translated some
of the important sections and put them on my web site as PostScript files.
Perhaps I should put them in quant-ph, then people could hopefully see them
in other formats.
Anyway, the net result of my studies is that Aspect did not try and
investigate the exact time of detection. He used coincidence windows of
around 15-20 ns. It seems entirely possible, in view of the fact that he
states that the electric pulses emitted by his detector were vary variable
in shape and size, that much of this variation was due to varying intensity
of individual pulses, each of which was supposed to a single photon. From
the way in which the "discriminator" worked, it is inevitable that on
average a large pulse would be detected earlier than a small one, though
only very slightly.
I doubt, though, if this is the most important factor in determining time of
detection. The picture that has built up in my mind is of the presence of
considerable background of electromagnetic noise, and of detection happening
when random noise added to signal exceeds some threshold. A small amount of
resonance might be involved, but it does not seem to matter what frequency
the noise is. The detectors are sensitive to heat, to cosmic rays, probably
to radiation from electronic equipment, and to the voltage applied.
The picture that seems to fit Aspect's observed time spectra is of the
signals comprising pulses of e/m waves, each starting at high intensity and
trailing off exponentially, with half-life (for one of his frequencies)
about 4-5 ns. (The other frequency, detected on the other side of the
experiment, may have had much shorter pulses.) Noise might vary on a scale
less than this.
Detection can occur any time during the time when the signal is fairly
strong (and also when there is no signal - there is a "dark count"). So
long as the distribution of noise intensities is reasonable (i.e. broad and
smooth), the probability of detection at any point in time will be roughly
proportional to the amplitude of the wave.
Given that the detectors only register the FIRST detection in a period of
100 ns or so (they have a considerable dead time, and later electronics
imposes other dead times), it is clear that weak signals will tend to be
detected later than strong ones.
I do not believe this has been properly tested! The belief has always been,
ever since the 1920's, that experiments showed no variation, but I don't
think they investigated times of less than about 3 ns, and, besides, it is
quite clear that you can get a vast range of different answers depending on
both you light source and your detector!
Timing, incidentally, is just one of the grey areas in the EPR experiments.
Probably the most important one is the "detection loophole", and this is
intimately tied up with this same matter of whether we have waves or
photons.
If we are dealing with photons then it seems plausible to assume that none
are lost at a two-channel polariser though at the detector a (considerable!)
proportion fail to register. You can assume that at such a polariser the
photon must go one way or the other. Under this assumption, it seems
reasonable to assume that the number of detections stays a constant
proportion of the number of emissions, regardless of the polarisation
direction of the photons relative to the axis of the polariser.
But if light is pure wave it's another matter! It may look as if we should
get a constant proportion of the pulses detected, but as soon as you allow
for real experimental conditions you find that this is most unlikely. If
the detectors turned intensities EXACTLY into probabilities, with linear
resonse curves, then one might think that one would get "fair sampling"
(another term for "no detection loophole") since one would be dealing with
sin^2 + cos^2 = 1 as a result of Malus' Law. But if you think about my
description above of how real detectors work, it is not possible to get
exacly this behaviour. It would require an impossible distribution for the
noise intensity, which cannot be flat over its entire range!
Is the response curve ever investigated? Well, it was not one of the things
reported in a recent study on detectors: Ribordy, G et al, Performance of
InGaAs/InP avalanche photodiodes as gated-mode photon counters, Applied
Optics 37, 2272-77 (1998). One of the co-authors of this article was
Nicolas Gisin, who was on the team that did those Bell tests over 10k in
1997 http://xxx.lanl.gov/abs/quant-ph/9707042 ...
Enough for now! Do look at my web site:
I've spent the past 6 years studying these matters (scarcely looking at text
books, much more at published papers).
Caroline
Gregory L. Hansen
Katie Schwarz <k...@socrates.berkeley.edu> wrote:
>Gregory L. Hansen <glha...@steel.ucs.indiana.edu> wrote:
>>Katie Schwarz <k...@socrates.berkeley.edu> wrote:
>> and no intensity of light would
>>create the effect if it was the wrong color.
>
>This can be explained by a semiclassical theory: Fermi's golden rule
>gives a delta function of the frequency of the incident light, even if
>you treat the incident light classically.
I would have to review Fermi's golden rule, and I don't have my books
here. But the delta functions enforce conservation of energy and
conservation of momentum. They don't have anything else to do with how
the electrons respond to the color of a wave. Now that you mention it, I
think Fermi's golden rule predicts a photoelectric effect for all colors
no matter how red. Any step function would have to be imposed
artificially.
--
"That's not an avacado, that's a grenae!" -- The Skipper
Gregory L. Hansen
c.h.thompson <c.h.th...@newscientist.net> wrote:
>Hi
>
>There seem to be some sensible ideas here, but I wonder how many people
>realise just how important it is to the current state of fundamental
>physics?
>
>I mean, it is obviously important, but the consequences of what I regard as
>a totally false view stretch far beyond the immediate issue. The
>assumption of particle behaviour is, I have found, at the root of the
>current official line re "quantum entanglement", "action at a distance",
>"EPR correlations" or whatever. These have opened the door to corruption
>and magic ...
Before you change the theory to remove the corruption and magic, make sure
the corruption and magic are disproved experimentally. For instance, some
of the recent EPR experiments that show the corruption and magic exist.
And quantized fields are deeper than the photoelectric effect.
Semiclassically, excited atoms should never de-excite when their
temperature goes to zero, but they still do. And there's the whole of
quantum field theory. It's an internally consistent and eperimentally
verified theory that's derived from a small number of first principles.
It's not so hard to imagine replacing one phenomenon or another with
waves, but it's more difficult to imagine replacing the whole darn thing
with a classical theory that's internally consistent, experimentally
verified, and derived from a minimum number of first principles rather
than a whole bunch of ad hoc additions and special cases.
Jon Bell
>Gregory L. Hansen <glha...@steel.ucs.indiana.edu> wrote:
>>
>>they tried to explain it with classical electromagnetism by assuming
>>energy will build up over time until an electron is excited. But the
>>response of the electrons was immediate,
>
>That's what it says in Eisberg and Resnick's textbook, but on thinking
>about it, this doesn't make sense to me: if energy comes in packages
>instead of continuously, you still have to wait for a photon to come
>along.
Yes, but not nearly as long. As an example from Beiser, "Concepts of
Modern Physics":
| A detectable photoelectron current occurs when 10^-6 W/m^2 of
| electromagnetic energy is absorbed by a sodium surface. [...] if the
| incident light is absorbed in the uppermost atomic layer, each atom
| receives energy at an average rate of 10^-25 W. At this rate over a
| month would be needed for an atom to accumulate energy of the magnitude
| that photoelectrons from a sodium surface are observed to have.
A simple calculation shows that for light with photon energy 2 eV
(approximately the work function of sodium) on the order of 10^12 photons
strike the surface per square meter per second, so any delay between
photons is going to be very small in this case.
--
Jon Bell <jtb...@presby.edu> Presbyterian College
Dept. of Physics and Computer Science Clinton, South Carolina USA
[ Information about newsgroups for beginners: ]
[ http://www.geocities.com/ResearchTriangle/Lab/6882/ ]
c.h.thompson
> >There seem to be some sensible ideas here, but I wonder how many people
> >realise just how important it is to the current state of fundamental
> >physics?
> >
> >I mean, it is obviously important, but the consequences of what I regard
as
> >a totally false view stretch far beyond the immediate issue. The
> >assumption of particle behaviour is, I have found, at the root of the
> >current official line re "quantum entanglement", "action at a distance",
> >"EPR correlations" or whatever. These have opened the door to corruption
> >and magic ...
>
> Before you change the theory to remove the corruption and magic, make sure
> the corruption and magic are disproved experimentally. For instance, some
> of the recent EPR experiments that show the corruption and magic exist.
If you took the trouble to read the rest of my message you'd find that for
the past 6 years I have been studying these experiments! They do NOT show
magical action at a distance. If the various realist objections to the
experiments had been followed up instead of ignored, the world would by now
have an entirely different attitude! Magic would have been laughed off the
scene.
The experiments have simply been misinterpreted, largely as a result of the
false assumption that light is a particle.
Of course high-energy radiation possibly does come in packets of rather more
definite energy. The "magic" (quantum entanglement) has only actually been
"demonstrated", though, using moderate frequencies and very low intensities.
These low energies can only be detected after the addition of e/m noise.
There may have been one of two supposed demonstrations of entanglement using
whole atoms. A case in point is Hagley, E et al, "Generation of
Einstein-Podolsky-Rosen Pairs of Atoms", Physical Review Letters 79, 1
(1997). But if you read this paper you will find that the "visibility" of
the coincidence curves was well within the range allowed under local
realism. Some experimenters seem to forget that local realism is quite
happy with correlations that arise as a result of a common cause. Even in
the simplest of models, these correlations can achieve 50% visibility.
> And quantized fields are deeper than the photoelectric effect.
> Semiclassically, excited atoms should never de-excite when their
> temperature goes to zero, but they still do. And there's the whole of
> quantum field theory. It's an internally consistent and eperimentally
> verified theory
I suspect that that can be challenged, but I am not qualified to do so
myself.
> that's derived from a small number of first principles.
> It's not so hard to imagine replacing one phenomenon or another with
> waves, but it's more difficult to imagine replacing the whole darn thing
> with a classical theory that's internally consistent, experimentally
> verified, and derived from a minimum number of first principles rather
> than a whole bunch of ad hoc additions and special cases.
I've made a good start!
The first thing to do is satisfy oneself that the "verifications" are not as
conclusive as they make out. They depend on a great pyramid of assumptions.
If enough people were to accept that there is a need to start afresh, and if
physicists had the same humble attitude toward Nature that biologist do -
recognising that there is more to it than we are likely to be able to model
in our elegant mathematics - then progress would soon start to be made. It
would not matter that at first the new system looked a mess! With enough
brains working on it, and a ban on monopolies, models that enabled
understanding (as opposed to just prediction) would gradually emerge.
Caroline
<http://www.aber.ac.uk/~cat>
c.h.thompson
Jon Bell <jtb...@presby.edu> wrote in message news:FKD9K...@presby.edu...
> Katie Schwarz <k...@socrates.berkeley.edu> wrote:
> >Gregory L. Hansen <glha...@steel.ucs.indiana.edu> wrote:
> >>
> >> The reason the photoelectric effect was important because
> >>they tried to explain it with classical electromagnetism by assuming
> >>energy will build up over time until an electron is excited. But the
> >>response of the electrons was immediate,
> >
> >That's what it says in Eisberg and Resnick's textbook, but on thinking
> >about it, this doesn't make sense to me: if energy comes in packages
> >instead of continuously, you still have to wait for a photon to come
> >along.
>
> Yes, but not nearly as long. As an example from Beiser, "Concepts of
> Modern Physics":
>
> | A detectable photoelectron current occurs when 10^-6 W/m^2 of
> | electromagnetic energy is absorbed by a sodium surface. [...] if the
> | incident light is absorbed in the uppermost atomic layer, each atom
> | receives energy at an average rate of 10^-25 W. At this rate over a
> | month would be needed for an atom to accumulate energy of the magnitude
> | that photoelectrons from a sodium surface are observed to have.
But we don't KNOW that the photoelectrons are emitted by individual atoms!
It seems much more likely to me that they are produced by the whole
electromagnetic field of a large region when, as a result of incoming waves
bouncing around a bit and suffering self-interference, or the addition of
noise and general chance events, some natural threshold is exceeded. Never
mind the details - they, I admit, are speculation. The fact is that a wave
theory does not have to assume an even spread of energy among the atoms.
Light can come in pulses, and it can suffer self-interference so that
locally there are high intensities. And "lumpy" noise can be added from
the environment, or along with applied voltages.
> A simple calculation shows that for light with photon energy 2 eV
> (approximately the work function of sodium) on the order of 10^12 photons
> strike the surface per square meter per second, so any delay between
> photons is going to be very small in this case.
Under a wave model, local hot-spots could emit with negligible time-delay.
Average energy figures tell us little.
Caroline
<http://www.aber.ac.uk/~cat>
Ken Muldrew
>Gregory L. Hansen <glha...@steel.ucs.indiana.edu> wrote:
>>Katie Schwarz <k...@socrates.berkeley.edu> wrote:
>>>
>>>The photoelectric
>>>effect can be explained by a semiclassical theory where the energy of
>>>the atoms is quantized but the electromagnetic field is classical.
>>
>>Are you sure? The reason the photoelectric effect was important because
>>they tried to explain it with classical electromagnetism by assuming
>>energy will build up over time until an electron is excited. But the
>>response of the electrons was immediate,
>That's what it says in Eisberg and Resnick's textbook, but on thinking
>about it, this doesn't make sense to me: if energy comes in packages
>instead of continuously, you still have to wait for a photon to come
>along. I couldn't find anything about a time delay in semiclassical
>treatments in books such as Loudon or Mandel & Wolf; anyone have
>details?
If I understand the semiclassical theories (and I have had only the
most superficial contact with these) then matter can only absorb
energy from the field in discrete quanta so there will be a time
delay.
Possibly you can find some answers in:
Jaynes, E. T., and F. W. Cummings, 1972, ``Comparison of Quantum and
Semiclassical Radiation Theories with Application to the Beam Maser,''
in Laser Theory, F. S. Barnes (ed.), IEEE Press, pp. 173-203.
Ken Muldrew
kmul...@acs.ucalgary.ca
Joe Rongen
message news:3819f...@news1.vip.uk.com...>
> > >Gregory L. Hansen <glha...@steel.ucs.indiana.edu> wrote:
> > >>
> > >> The reason the photoelectric effect was important because
> > >>they tried to explain it with classical electromagnetism by
> > >>assuming
> > >>energy will build up over time until an electron is excited.
> > >>But the response of the electrons was immediate,
> > >
> > >That's what it says in Eisberg and Resnick's textbook, but on
> > >thinking
> > >about it, this doesn't make sense to me: if energy comes in
> > >packages
> > >instead of continuously, you still have to wait for a photon to
> > >come along.
> >
Support for the photon interpretation is provided by the time delay
between the instant the light beam is turned on and the onset of a
photoelectric current.
According to the wave picture; the energy of the light beam is
uniformly distributed in both space and time. On the other hand,
according to the photon hypothesis, the radiant energy arrives as
individual quanta, randomly distributed, and there is some chance
that the very first photon to arrive will liberate a photoelectron.
Experiment confirms the photon picture: Lawrence and Beams
showed in 1928 that photoelectrons are sometimes emitted less
than 3*10^(-9) sec after initial illumination, even with an incident
light beam so weak that the expected time delay according to
the wave picture would be much longer.
See: E.O.Lawrence and J.W.Beams, Phys. Rev. 32, 478 (1928)
Regards Joe
Katie Schwarz
>>
>>[Photoelectric effect]
>>This can be explained by a semiclassical theory: Fermi's golden rule
>>gives a delta function of the frequency of the incident light, even if
>>you treat the incident light classically.
>
>I would have to review Fermi's golden rule, and I don't have my books
>here. But the delta functions enforce conservation of energy and
>conservation of momentum. They don't have anything else to do with how
>the electrons respond to the color of a wave.
No, the delta function comes directly from an approximate solution of
Schroedinger's equation. You don't have to assume anything about
conservation of energy, but if you already know that the light energy
comes in units of hnu, you can then *interpret* the result as energy
conservation.
Fermi's golden rule is really simple, you can probably find it in
whatever quantum texts are available. It's just a first-order
approximation; integrate Schroedinger's equation once to get the
coefficient of the excited state, square the modulus, and you get
|H'|^2 sin^2 [(wfi - w)t/2]
probability of excitation = ---- --------------------
hbar^2 (wfi- w)^2
where wfi is the electron energy difference, and w is the frequency of
the *classical* electromagnetic field. The sin^2 is sharply peaked at
w = wfi for t large compared to the inverse frequency.
> Now that you mention it, I
>think Fermi's golden rule predicts a photoelectric effect for all colors
>no matter how red. Any step function would have to be imposed
>artificially.
It predicts a photoelectric effect for hnu = energy difference between
the initial and final states of the electron. If the *electron*'s
energy is quantized, so that there is a gap between the bound state
and the lowest possible free state, then Fermi's golden rule says
there will be no photoelectric effect for frequencies that are too
low. The incident light doesn't have to be quantized to get that
result.
c.h.thompson
Joe Rongen <joer...@whisp.com> wrote
> c.h.thompson <c.h.th...@newscientist.net> wrote
> > Jon Bell <jtb...@presby.edu> wrote
> > > Katie Schwarz <k...@socrates.berkeley.edu> wrote:
> > > >Gregory L. Hansen <glha...@steel.ucs.indiana.edu> wrote:
> > > >>
> > > >>they tried to explain it with classical electromagnetism by
> > > >>assuming energy will build up over time until an electron is
> > > >>excited.
> > > >
You don't seem to have read what I said: there is not just one "wave
theory". You can have "lumpy wave theories"! For a start, you can get
regions of high and low intensity due to interference; you can allow for
emission in pulses; you can assume that the receptor is "primed" by local
(lumpy) noise. One way or another, these influences can cause an
effectively immediate response.
> On the other hand,
> according to the photon hypothesis, the radiant energy arrives as
> individual quanta, randomly distributed, and there is some chance
> that the very first photon to arrive will liberate a photoelectron.
Now QT allows for a "work function", does it not? This means that it does
not in practice assume that every photon is converted to an electron of
fixed energy. There is not so very much difference between the theories. I
don't see time of detection as a barrier to a wave theory.
> Experiment confirms the photon picture: Lawrence and Beams
> showed in 1928 that photoelectrons are sometimes emitted less
> than 3*10^(-9) sec after initial illumination, even with an incident
> light beam so weak that the expected time delay according to
> the wave picture would be much longer.
Yes, I know of Lawrence and Beams work. To my shame, I have not looked up
the paper and don't know what energies they looked at. All that I am saying
is based on careful inspection of experimental details of recent
"single-photon" experiments such as Alain Aspect's. I can find nothing to
contradict the idea that detection in these cases is a matter of signal plus
noise exceeding a threshold. The appearance of "photons" is caused
artificially, in Aspect's experiments, by the "discriminator", which simply
thresholds the variable pulses and counts those whose voltage exceeds a
limit set by the experimenter. Aspect takes several pages of his PhD thesis
to explain just how he decided how to chose him setting of all the factors
involved. Though he rationalised this, there was a lot of personal
judgement involved.
Modern detectors, I gather (from correspondence with experimenters) try and
relieve the experimenter from this responsibility. Apparently they produce
pulses of very uniform size. I would deduce that the thresholding is done
within the detector. (Alternatively, that they are not dealing with such
low light intensities as Aspect was: a strong light pulse might be able to
induce emission with no addition of noise.
> See: E.O.Lawrence and J.W.Beams, Phys. Rev. 32, 478 (1928)
Yes, 1928! I believe there have been checks since, but there is scope for
more! If you look at my web site (or at, say,
http://xxx.lanl.gov/abs/quant-ph/9611044) you will find that I believe that
(in Bell test experiments and similar) weak signals that have passed through
a polariser whose axis was almost at right angles to their polarisation
direction are, on average, detected later than those with polarisation
parallel to the polariser axis. This may or may not be a significant
factor. I think it is the most likely cause of the Bell test violations of
one of the early tests, Freedman and Clauser's of 1972 (Freedman, S J and
Clauser, J F, Physical Review Letters 28, 938 (1972)), in which a rather
short coincidence window was used.
Of course, the QT model of polarisation is so peculiar that it may be hard
for anyone who has accepted it to even see what I mean! What I'm trying to
say is that the crucial difference between QT and wave theories is not the
minimum time to detection, as lumpiness can make this very small. The
average time is different. This is associated with the fact that classical
light does not emerge from a polariser in units of "photons" but in waves of
the same frequency as those that went in. They simply emerge with lower
amplitude.
A repeat of the kind of experiment Katie was suggesting, with beamsplitters
and checking for coincidences, is urgently needed, as so long as people
continue to believe that these demonstrate particle behaviour we have
deadlock.
So we need:
(a) beamsplitter experiment
(b) analysis of "response curves" of detectors as we alter light intensity
(c) analysis of detection times
If we were to have a rational (local realist) approach to EPR experiments,
the setup of these could be adapted to aid in these investigations.
Caroline
<http://www.aber.ac.uk/~cat>
c.h.thompson
Ken Muldrew <kmul...@acs.ucalgary.ca> wrote
> If I understand the semiclassical theories (and I have had only the
> most superficial contact with these) then matter can only absorb
> energy from the field in discrete quanta so there will be a time
> delay.
> Possibly you can find some answers in:
>
> Jaynes, E. T., and F. W. Cummings, 1972, ``Comparison of Quantum and
> Semiclassical Radiation Theories with Application to the Beam Maser,''
> in Laser Theory, F. S. Barnes (ed.), IEEE Press, pp. 173-203.
Jaynes had some very interesting things to say about the whole state of
fundamental physics in an article in 1990, mainly on the electron:
Jaynes, E, Scattering of light by free electrons, To be published by Kluwer
in proceedings of workshop on "The electron 1990" held at St Francis Xavier
University, Antigonish, Nova Scotia. Editors: A Weingartshofer and D
Hestenes.
Presumably this has now been published. If now I can probably get hold of
an electronic copy in a brand of Latex or as a ps file.
My notes from the introductory sections include:
Practical men often have no need for theories.
QT is deeply into epicycles, so that one can hardly blame them for ignoring
it.
Dirac's quantisation of the e/m field leads to some right answers but also
some "horrendously wrong" ones.
The "claim that all phenomena must be described in terms of Hilbert spaces,
energy levels etc. .. [has] captured the minds of physicists for over sixty
years." In practice, they have been content with phenomenology, and
deprecated all efforts at studying the real, nonlinear, underlying systems.
QT is not restrained by any physical principles. "The present QM is only an
empty mathematical shell in which a future physical theory may, perhaps, be
built."
P4: QM is claimed to be logically complete, yet has to admit that it is
observationally incomplete - in all experiments one can observe things the
theory can't predict.
"Even the EPR paradox failed to force retraction of this claim [of
completeness], and so currently taught QT still contains the schizoid
elements of local acausality on the one hand - and instantaneous action at
a distance on the other! We find it astonishing that anyone could seriously
advocate such a theory."
"Not surprisingly, there has been no really significant advance in basic
understanding since the 1927 Solvay Congress ."
" . for 60 years Bohr's teachings have been perverted into attempts to
deprecate and discourage any further thinking aimed at finding the causes
underlying microphenomena. Such thinking is termed 'obsolete mechanistic
materialism' ."
p5: This workshop might be held to be out of the mainstream, but "there is
no mainstream today; it had long since dried up and our vessel is grounded."
" . our efforts are much closer to the traditional mainstream of science
than much of what is done in theoretical physics today ."
CT: But Jaynes was interested in mathematical models, and personally I find
these too restrictive. He has produced mathematical semi-classical theories
of light, but I don't think it reasonable to talk as if these were the only
possible ones. His idea of the electron, incidentally, does not seem to
have been very definite!
The idea of absorption happening in discrete quanta is not forced on us by
any facts that I know of. The energy absorbed can always come from many
sources at once, parts of many emissions (plus "noise", which is what is
left that is too messy to account for or cannot be assumed to come from the
same general source).
Caroline
<http://www.aber.ac.uk/~cat>
Gregory L. Hansen
c.h.thompson <c.h.th...@newscientist.net> wrote:
>You don't seem to have read what I said: there is not just one "wave
>theory". You can have "lumpy wave theories"! For a start, you can get
>regions of high and low intensity due to interference; you can allow for
>emission in pulses; you can assume that the receptor is "primed" by local
>(lumpy) noise. One way or another, these influences can cause an
>effectively immediate response.
How does the lumpy wave theory hold up experimentally? It seems to me
that the lumpy wave theory still doesn't lead to a cut-off when the light
is too red.
--
"That's not an avacado, that's a grenade!" -- The Skipper
Joe Rongen
in message news:<381aa...@news2.vip.uk.com
> Joe Rongen <joer...@whisp.com> wrote
> >
> > Experiment confirms the photon picture: Lawrence and Beams
> > showed in 1928 that photoelectrons are sometimes emitted less
> > than 3*10^(-9) sec after initial illumination, even with an
> > incident light beam so weak that the expected time delay according
> > to the wave picture would be much longer.
>
> Yes, I know of Lawrence and Beams work. To my shame,
> I have not looked up the paper and don't know what energies
> they looked at.
See: E.O.Lawrence and J.W.Beams, Phys. Rev. 32, 478 (1928)
Your time looking for their work may be well spend.
Also, keep in mind that QM is pervaded by one great theme:
'The predictions of quantum mechanics are expressed
in terms of probabilities.'
Regards Joe
c.h.thompson
Gregory L. Hansen <glha...@steel.ucs.indiana.edu> wrote in message
news:7vepte$99a$2...@flotsam.uits.indiana.edu...
> >theory". You can have "lumpy wave theories"! For a start, you can get
> >regions of high and low intensity due to interference; you can allow for
> >emission in pulses; you can assume that the receptor is "primed" by local
> >(lumpy) noise. One way or another, these influences can cause an
> >effectively immediate response.
>
> that the lumpy wave theory still doesn't lead to a cut-off when the light
> is too red.
Admittedly it has to be quite a complicated theory, but who said the
universe was simple?
Insofar as it really is true that you don't get emissions unless the
frequency exceeds a threshold, my lumpy wave theory will need two factors.
Remember, it applies to light that requires to be boosted by noise as
otherwise individual pulses are just too weak to be detected. To get a
detection we must need a combination of:
(a) coupling of the field of the cathode to the incoming field, and
(b) an extra "kick" from noise.
It is possible that (a) just does not happen if the frequency is too low,
but I'm not God! I don't know.
"There is more in heaven and earth than is dreamt of in your philosopy!"
Anyway, if you want to try and visualise (a) and (b) in detail, try thinking
of a toy swing, very light, that can be set in motion by puffs of wind.
Obviously this works best if the frequency of the puffs equals the natural
frequency of the swing. So far we have part (a). But now along comes a
larger puff. This is noise, (b). It does not matter what its frequency is
as it need only be a single burst. So long as it happens to come at a
suitable time it may push the swing beyond its limits and break it! Hey
Presto: your electron is released!
But in point of fact the rule of no emission below a certain frequency is
not hard and fast. What about that "dark count"? It does not require any
input pulse at all to produce that!
Caroline
<http://www.aber.ac.uk/~cat>
c.h.thompson
Joe Rongen <joer...@whisp.com> wrote in message
news:IdFS3.319$u45...@198.235.216.4...
> c.h.thompson <c.h.th...@newscientist.net> wrote
> in message news:<381aa...@news2.vip.uk.com
> > Joe Rongen <joer...@whisp.com> wrote
>
probabilities.'
And it can't even get those right!
Look, probabilities are things you measure, but they are not physical
objects. You can't have a genuine fundamental theory that only deals in
such things! Such a theory is, as Einstein, Podolsky and Rosen so rightly
said, incomplete.
Underlying those parts of QM that do give the right answers (interference
patterns etc) there are real waves behaving as real waves always have done.
The detectors that are used artificially turn these into probabilities. In
the cases I've studied (the EPR experiments using very low intensity light)
it seems clear that it would be possible to use essentially the same
detectors and extract more information. There is some quantatitive
information relating to each detection that is simply thrown away by the
"discriminator". If all you want is to study an individual interference
pattern, though, it does not matter if you throw a lot of info away. There
are always so many "photons" that the pattern will emerge anyway. It may be
distorted, but who cares? One is generally only interested in whether or
not it is there and what scale it is on.
As I said, the probabilities of QM are not even right all the time. They
are wrong in the EPR case, and experiments have unjustifiably been
interpreted as if they are right. As a consequence it is generally believed
that the real world does not allow a theory based on "hidden variables",
such as the real amplitudes, phases and frequencies of the waves that are
claimed to be mere probability waves.
Caroline
Nathan Urban
> > > Joe Rongen <joer...@whisp.com> wrote
> > 'The predictions of quantum mechanics are expressed in terms of
> > probabilities.'
> And it can't even get those right!
> Look, probabilities are things you measure, but they are not physical
> objects. You can't have a genuine fundamental theory that only deals in
> such things!
What incredible arrogance, to claim to be able to dictate the laws that
the universe can and cannot follow.
c.h.thompson
Nathan Urban <nur...@crib.corepower.com> wrote in message
news:7vfjmr$urc$1...@crib.corepower.com...
> In article <381b3001$1...@news1.vip.uk.com>, "c.h.thompson"
<c.h.th...@newscientist.net> wrote:
>
> > > > Joe Rongen <joer...@whisp.com> wrote
>
> > > 'The predictions of quantum mechanics are expressed in terms of
> > > probabilities.'
> > You can't have a genuine fundamental theory that only deals in
> > such things!
>
> What incredible arrogance, to claim to be able to dictate the laws that
> the universe can and cannot follow.
Perhaps I should have quoted higher authority in this. Einstein, for
example. Do you just dismiss him for arrogance for writing a paper entitled
"Can Quantum-Mechanical Description of Physical Reality be Considered
Complete?" (His famous one with Podolsky and Rosen, Physical Review 47,
777-780 (1935)).
What he was saying was that a fundamental theory would not just predict
probabilities but would describe physical entities that caused the observed
probabilities. Quantum Mechanics, as we have so often been told, is not
compatible with their existence, but his argument was nevertheless valid.
It is QM that is wrong. The community made a big mistake scientifically
when they accepted Niels Bohr's rejoinder to the EPR paper.
But presumably you are happy with the status quo? Everything in the
garden's lovely? QM predictions never fail? The fact that nobody
understands it (not my words but Richard Feynman's, Niels Bohr's and many
others!) does not matter?
I'm sorry if I come over as arrogant but this whole farce makes me angry. I
don't like to see people wasting such huge amounts of mental energy trying
to understand the "quantum mysteries" when they could just look at the
actual experiments and say to themselves: "But isn't there a perfectly
ordinary explanation?"
This is what I've been doing for the past few years. In every case I've
found that there is. There are always "hidden variables" that can exist
comfortably beneath the probabilities. Not the predicted probabilities,
mind you, or not always. Hidden variables are completely compatible with
the experimentally estimated probabilities - a subtly different matter!
It was well understood in "the good old days" that a theory based on the
underlying variables would be stronger, have more predictive power, be
comprehensible, even (dare I say it?) have a chance of being a reasonable
model of "reality"!
You can derive the probabilities from such a theory, but you can't do the
reverse. Once you've reduced your data to just a probability estimate
you've lost information on individual entities.
Do you dispute this?
Caroline
<http://www.aber.ac.uk/~cat>
Nathan Urban
> Nathan Urban <nur...@crib.corepower.com> wrote in message
> news:7vfjmr$urc$1...@crib.corepower.com...
> > In article <381b3001$1...@news1.vip.uk.com>, "c.h.thompson"
> <c.h.th...@newscientist.net> wrote:
> > > > > Joe Rongen <joer...@whisp.com> wrote
> > > > 'The predictions of quantum mechanics are expressed in terms of
> > > > probabilities.'
> > > You can't have a genuine fundamental theory that only deals in
> > > such things!
> > What incredible arrogance, to claim to be able to dictate the laws that
> > the universe can and cannot follow.
> Perhaps I should have quoted higher authority in this.
Perhaps you should leave argument-by-authority out of this altogether;
it only weakens your position.
> Einstein, for
> example. Do you just dismiss him for arrogance for writing a paper entitled
> "Can Quantum-Mechanical Description of Physical Reality be Considered
> Complete?"
I don't look highly on him for rejecting a perfectly possible viewpoint
simply because it didn't sit well with his own philosophy. It's well
known that Einstein never accepted non-deterministic quantum mechanics.
Most people don't view it as the highlight of his career, either.
> What he was saying was that a fundamental theory would not just predict
> probabilities but would describe physical entities that caused the observed
> probabilities.
That is what _he_ wanted in a fundamental theory. The universe is not
required to obey.
> Quantum Mechanics, as we have so often been told, is not
> compatible with their existence, but his argument was nevertheless valid.
He didn't have an argument. His "argument" was "I don't like the
following implication of quantum mechanics, therefore there's a problem
with quantum mechanics".
> It is QM that is wrong.
Too bad you can't prove that.
> But presumably you are happy with the status quo? Everything in the
> garden's lovely? QM predictions never fail?
They haven't yet.
> The fact that nobody
> understands it (not my words but Richard Feynman's, Niels Bohr's and many
> others!) does not matter?
Nope. I don't expect to be comfortable with a fundamental theory.
> I'm sorry if I come over as arrogant but this whole farce makes me angry.
That sounds like someone who's far too emotionally attached to their
own worldview. Like Einstein.
> This is what I've been doing for the past few years. In every case I've
> found that there is. There are always "hidden variables" that can exist
> comfortably beneath the probabilities.
Yes, hidden variables theories can exist compatible to standard quantum
theory (e.g. Bohmian QM). This is well known. It's simply that people
view them as having to go through far more contortions to make sense than
biting the bullet and accepting nondeterminism. (Plus, to my knowledge,
no one has ever presented a convincing _relativtistic_ hidden variables
theory.)
> It was well understood in "the good old days" that a theory based on the
> underlying variables would be stronger, have more predictive power, be
> comprehensible, even (dare I say it?) have a chance of being a reasonable
> model of "reality"!
The point of a hidden variables theory is that it _isn't_ any stronger
or more predictive than quantum mechanics. It is identical in its
predictions. (Unless you're proposing an alternative theory that has
testable differences?) When you have two interpretations, it's possible
to pick either, and most people today pick nondeterminism because they
find it _less_ weird than deterministic QM.
> You can derive the probabilities from such a theory, but you can't do the
> reverse.
That doesn't make a hidden variables theory "more fundamental", because
by definition the hidden variables are hidden. In terms of what you
can actually _observe_, there is no gain if the only difference is the
presence of deterministic hidden variables. It's like special relativity
vs. Lorentz aether theory -- you can derive SR from LET, but you can't
derive an aether from SR. Guess one which people picked, and why?
The aether has no observable consequence; it is unnecessary in the model.
> Once you've reduced your data to just a probability estimate
> you've lost information on individual entities.
You can't get that information anyway in a hidden variables theory.
Physics is the science of the _observable_.
Gregory L. Hansen
c.h.thompson <c.h.th...@newscientist.net> wrote:
>
>Gregory L. Hansen <glha...@steel.ucs.indiana.edu> wrote in message
>news:7vepte$99a$2...@flotsam.uits.indiana.edu...
>> In article <381aa...@news2.vip.uk.com>,
>> c.h.thompson <c.h.th...@newscientist.net> wrote:
>>
>> >theory". You can have "lumpy wave theories"! For a start, you can get
>> >regions of high and low intensity due to interference; you can allow for
>> >emission in pulses; you can assume that the receptor is "primed" by local
>> >(lumpy) noise. One way or another, these influences can cause an
>> >effectively immediate response.
>>
>> that the lumpy wave theory still doesn't lead to a cut-off when the light
>> is too red.
>
>Admittedly it has to be quite a complicated theory, but who said the
>universe was simple?
>
>Insofar as it really is true that you don't get emissions unless the
>frequency exceeds a threshold, my lumpy wave theory will need two factors.
>Remember, it applies to light that requires to be boosted by noise as
>otherwise individual pulses are just too weak to be detected. To get a
>detection we must need a combination of:
>(a) coupling of the field of the cathode to the incoming field, and
>(b) an extra "kick" from noise.
>
>It is possible that (a) just does not happen if the frequency is too low,
>but I'm not God! I don't know.
If the energy of light were not quantized, I don't see why you shouldn't
be able to demonstrate the photoelectric effect for waves of arbitrarily
long wavelength, just as long as the intensity is high enough. There's
certainly nothing in the pseudo-classical treatment of the photoelectric
effect that suggests a cut-off frequency. I've just checked in Shankar,
and Fermi's golden rule gives a frequency dependence but the rate never
goes to zero or has anything suggesting even a smoothed-out step function.
>Anyway, if you want to try and visualise (a) and (b) in detail, try thinking
>of a toy swing, very light, that can be set in motion by puffs of wind.
>Obviously this works best if the frequency of the puffs equals the natural
>frequency of the swing. So far we have part (a). But now along comes a
>larger puff. This is noise, (b). It does not matter what its frequency is
>as it need only be a single burst. So long as it happens to come at a
>suitable time it may push the swing beyond its limits and break it! Hey
>Presto: your electron is released!
>
>But in point of fact the rule of no emission below a certain frequency is
>not hard and fast. What about that "dark count"? It does not require any
>input pulse at all to produce that!
For one thing, the noise, or thermal motion of the atoms, could cause a
bona fide photon to be blueshifted into a knockout energy. And, again due
to temperature, some atoms are going to have electrons in excited states
so they'll be easier to knock out, even boil off on their own.
Measurements done while varying the temperature and bringing it as close
to zero as you can should help you eliminate the temperature dependence.
I think using very thin films will help you check for interference
effects.
Gregory L. Hansen
>news:IdFS3.319$u45...@198.235.216.4...
>> c.h.thompson <c.h.th...@newscientist.net> wrote
>> in message news:<381aa...@news2.vip.uk.com
>>
>> 'The predictions of quantum mechanics are expressed in terms of
>probabilities.'
>
It can't?
>Look, probabilities are things you measure, but they are not physical
>objects. You can't have a genuine fundamental theory that only deals in
>such things! Such a theory is, as Einstein, Podolsky and Rosen so rightly
>said, incomplete.
Einstein objected to certain interpretations of quantum mechanics, but he
had no problem with actually doing calculations with it. Personally, I
tend to follow the "shut up and calculate" interpretation. Einstein is
also the one that suggested light energy is quantized.
>As I said, the probabilities of QM are not even right all the time. They
>are wrong in the EPR case, and experiments have unjustifiably been
>interpreted as if they are right. As a consequence it is generally believed
Uh, which experiments, and what's incorrect about the interpretatins? Are
you talking about the EPR experiments that have been done in the last year
or so?
Bruce Richmond
glha...@steel.ucs.indiana.edu (Gregory L. Hansen) wrote:
> In article <7vattc$qlc$1...@nnrp1.deja.com>,
> Bruce Richmond <bsr...@my-deja.com> wrote:
> >Particle theory uses the number of
> >particles per second to establish the intensity. And we can have
> >different colors of the same intensity. So how can the same variable
> >also be used to explain color?
>
> Energy of a photon is E=hf, Planck's constant multiplied by the
frequency.
> Total energy of light in a region is U=nE, number of photons
multiplied by
> the energy of a photon. Intensity is I=U/A=nE/A=nhf/A, energy
divided by
> area.
>
Thank you for trying to clarify this for me, but I still have a few
questions.
How is the frequency defined? It must include n/t and I would think
the n would have to be per unit of area.
Does Planck's constant have any units attached to it?
What is the difference in energy for individual photons of different
frequencies attributed to? They all have the same velocity.
c.h.thompson
"c.h.thompson" <c.h.th...@newscientist.net> wrote:
> > > > > > Joe Rongen <joer...@whisp.com> wrote
>
> > > > > probabilities.'
>
> > > > such things!
>
that
> > > the universe can and cannot follow.
>
> > Perhaps I should have quoted higher authority in this.
>
> Perhaps you should leave argument-by-authority out of this altogether;
> it only weakens your position.
Fair enough, as Einstein was both hero and villain of the EPR story. Hero
in standing up for the law of (local) cause and effect - in point of fact he
was not so much against the indeterminate nature of QM as against its
"nonlocal" predictions - but villain in inventing the "photon".
> I don't look highly on him for rejecting a perfectly possible viewpoint
Now lets get down to hard facts. For more please do look at my web site,
and you might do worse than to start by glancing at "The Tangled Methods of
Quantum Entanglement", http://www.aber.ac.uk/~cat/Tangled/tangled.html ,
which describes the way in which my attempts to publicise KNOWN weaknesses
in the EPR experiments have failed to reach the pages of PRL and PRA. Some
of my papers are in http://xxx.lanl.gov/abs/quant-ph: 9611037, 9711044,
9903066.
Parts of QM are perfectly OK, but the part that Einstein was rejecting
predicted nonlocality. He interpreted this as implying that it was wrong.
If science is taken as being the identification of the causes of things then
he was right. (You seem to be saying, with Bohr, that science is just the
manufacture of prediction formulae.)
Perhaps I'll leave you to look at my web site rather than say much more now.
The point is that in order to explain the actual experiments, from Clauser
and Freedman's of 1972 to Aspect's of 1981-2 and the present, the "hidden
variables" you need are just the classical variables of polarisation, phase,
amplitude etc.. No new physics - or none of any significance - is needed!
You just need slight updating of old ideas, to accomodate the new
information that the dominance of QM has caused to be omitted.
"My" hidden variable explanations (shared in all essential aspects by
Stochastic Electrodynamics) is totally common sense, nothing to do with
Bohm's ideas. Bohm, as so many others, was trying to make a theory that
predicted exactly the same formulae as QM. This is not necessary. All that
science demands is explanations for observed facts, which is not the same
thing!
The "experts" know this, but don't seem able to accept that it really does
rock the boat.
> > It is QM that is wrong.
>
> Too bad you can't prove that.
If you take the trouble to read a couple of my papers - and do ask for more
explanation if necessary - you will see that in this instance the hidden
variable explanation is so obviously right that QM must be wrong!
Of course, you are unlikely to accept this straight away, and there are a
number of experiments that could be done to help you on the way. Some
critical ones would have the aim of proving that the light used, though in
most cases coming in pulses, was not in the form of "photons". When light
is split at a "two-channel polariser" it does not divide in the form of
whole photons, of fixed energy. It keeps its same frequency, yes, but the
amplitude (the ordinary, classical, wave amplitude) is decreased.
> > QM predictions never fail?
>
> They haven't yet.
Only (in the EPR case) because (a) it is not possible to get the "failed"
results published and (b) there is a certain amount of spontaneous, perhaps
subconscious, massaging of experimental parameters and/or the data! In
support of (a), take Holt and Pipkin's experiment, back in 1974, which is
only available as a Harvard University preprint.
> > You can derive the probabilities from such a theory, but you can't do
the
> > reverse.
>
> That doesn't make a hidden variables theory "more fundamental", because
> by definition the hidden variables are hidden. In terms of what you
> can actually _observe_, there is no gain
The gain is in understanding. Without the possibility of this you are
liable to make gross errors.
> It's like special relativity
> vs. Lorentz aether theory -- you can derive SR from LET, but you can't
> derive an aether from SR. Guess one which people picked, and why?
> The aether has no observable consequence; it is unnecessary in the model.
Not true, for the same reason. Given an aether, it is clear that light
exists as a wave in it (by definition, one might say) and hence that you
must have a theory in which its speed is always that relative to the aether.
If Einstein had stuck consistently to this idea he would never have
invented SR! (See also "The Einstein Hoax" thread.)
> Physics is the science of the _observable_.
Your physics may be, mine is not. The physics of the observable can make
gross errors if the instruments are not properly calibrated! And how can
you objectively calibrate a light detector?
Caroline
c.h.thompson
> >Gregory L. Hansen <glha...@steel.ucs.indiana.edu> wrote
> >>
> >> How does the lumpy wave theory hold up experimentally? It seems to me
> >> that the lumpy wave theory still doesn't lead to a cut-off when the
light
> >> is too red.
> >Insofar as it really is true that you don't get emissions unless the
> >frequency exceeds a threshold, my lumpy wave theory will need two
factors.
> >Remember, it applies to light that requires to be boosted by noise as
> >otherwise individual pulses are just too weak to be detected. To get a
> >detection we must need a combination of:
> >(a) coupling of the field of the cathode to the incoming field, and
> >(b) an extra "kick" from noise.
> >
> >It is possible that (a) just does not happen if the frequency is too low,
> >but I'm not God! I don't know.
I repeat, I don't know! And thinking about it the above is not very
satisfactory as it places too much emphasis on a resonance effect. It does
not explain why you get emission at all frequencies above a minimum.
But I have not time to read this all up, and have no lab to do experiements,
so I can't say more than that I simply doubt whether we have any
satisfactory theory yet. And until we get a clearer idea of whether or not
photons (and electrons?) exist, we are just a tiny bit stuck. We're in a
circular argument, escape from which will demand a leap of faith.
> If the energy of light were not quantized, I don't see why you shouldn't
> be able to demonstrate the photoelectric effect for waves of arbitrarily
> long wavelength, just as long as the intensity is high enough.
Nor do I, for that matter!
> >But in point of fact the rule of no emission below a certain frequency is
> >not hard and fast. What about that "dark count"? It does not require
any
> >input pulse at all to produce that!
>
> For one thing, the noise, or thermal motion of the atoms, could cause a
> bona fide photon to be blueshifted into a knockout energy.
Hmm ... I've nothing against Doppler shifts ...
> Measurements done while varying the temperature and bringing it as close
> to zero as you can should help you eliminate the temperature dependence.
Aha! But if you reduce the temperature I've a shrewd suspicion that the
kind of detector used in "single-photon" work goes on strike! It outputs
nothing.
Temperature is one of the parameters that the experimenter is free to choose
so as to get behaviour that appears to obey QM predictions!
As you may gather, I'm pretty ignorant of general work on the photoelectric
effect. I just know a little about what goes on in the detectors Aspect and
later workers (Kwiat, Tittel, Zeilinger ....) used and quite a lot about the
implications for the "demonstrations of nonlocality". For more on this (re
your other message) see my reply to Nathan Urban.
Caroline
<http://www.aber.ac.uk/~cat>
c.h.thompson
> >>
> >> 'The predictions of quantum mechanics are expressed in terms of
> >probabilities.'
> >
>
> It can't?
>
> >are wrong in the EPR case, and experiments have unjustifiably been
> >interpreted as if they are right.
>
> you talking about the EPR experiments that have been done in the last year
> or so?
All of them! See more in my reply to Nathan Urban. They use Bell tests
(or, in many recent experiments, simple "visibility" tests) that are not
valid.
Take visibility, which is an easy concept. We are dealing with sine waves,
or almost sine waves. The visibility is just (max-min)/(max+min). Now this
quantity is critically dependent on the min! If min is zero, visibility is
1. Yet just by altering the settings of the detector you can alter the
minimum, and you most certainly produce a drastic change if you "subtract
accidentals"! Ignoring the latter for the present (see for example
http://xxx.lanl.gov/abs/quant-ph/9903066 or various papers with titles such
as "EPR, Magic and the Nature of Light"
http://www.aber.ac.uk/~cat/Html/vigier.htm on my web site if you are
interested), the point is that Bell tests are critically dependent on the
shape of the response curve of the detector as you vary the input intensity.
Under QM, there is no interest in this response curve, since (by an act of
faith) we have inputs that are all identical. The Bell tests used in real
experiments are weak in many respects as they don't use quantities that
anybody pretends are true unbiased estimates of probabilities. Actual
detector "efficiencies" being low, the relative frequencies observed are
tiny and would not have a ghost of a chance of causing a violation of a true
test. Therefore they use different tests that use ratios that are not
probability estimates - or not unless certain assumptions are true.
The visibility test is one of these, and depends among other things on the
detectors having exactly linear reponses to variations in intensity. (By
this, I mean classical intensity, not "photon number".) To investigate it
one needs to take "weakened photons" that have passed through a polariser or
similar.
Nobody seems interested in doing this. I have discussed it with people at
Innsbruck (now in Vienna). They have offered to do experiments but nothing
has come of it ...
Caroline
<http://www.aber.ac.uk/~cat>
Charles Francis
ewscientist.net> writes
>
>Look, probabilities are things you measure, but they are not physical
>objects. You can't have a genuine fundamental theory that only deals in
>such things! Such a theory is, as Einstein, Podolsky and Rosen so rightly
>said, incomplete.
Absolutely.
>
>Underlying those parts of QM that do give the right answers (interference
>patterns etc) there are real waves behaving as real waves always have done.
And you accuse other people of being illogical! There is absolutely no
way you can deduce any such thing from "interference" patterns. So
called interference comes from hypothetical statements made in quantum
mechanics about what would happen if an experiment were to be done, not
from physical waves.
>
>As I said, the probabilities of QM are not even right all the time. They
>are wrong in the EPR case,
You may have evidence that the experiment is inconclusive in the EPR
case. If you are correct it is scientifically important, and should have
be published. But to suggest that you can show the predicted
probabilities are wrong is nothing short of a lie, which, unfortunately
destroys your credibility, along with your refusal to acknowledge
experiments which demonstrate the particulate nature of light.
Experiments far more conclusive than interference patterns, which only
demonstrate the validity of a mathematical formula, not the existence of
a wave.
>
--
Charles Francis
cha...@clef.demon.co.uk
c.h.thompson
Charles Francis <cha...@clef.demon.co.uk> wrote
c.h.thompson <c.h.th...@newscientist.net> writes
> >
> >Underlying those parts of QM that do give the right answers (interference
> >patterns etc) there are real waves behaving as real waves always have
done.
>
> And you accuse other people of being illogical! There is absolutely no
> way you can deduce any such thing from "interference" patterns. So
> called interference comes from hypothetical statements made in quantum
> mechanics about what would happen if an experiment were to be done, not
> from physical waves.
Have you never seen interference patterns on the surface of a pond?
Are these due to "hypothetical statements made in QM??????" How ridiculous
can you get! It is QM that invented the crazy idea of interference of
probability waves ....
But isn't it about time you looked at my web site? I thought you had.
> >As I said, the probabilities of QM are not even right all the time. They
> >are wrong in the EPR case,
>
> You may have evidence that the experiment is inconclusive in the EPR
> case. If you are correct it is scientifically important, and should have
> be published.
My dear Charles, may I suggest that you read my essay (published in the
journal Accountability in Research) on the subject of my difficulties in
getting the facts I and other realist have unearthed published! See
http://www.aber.ac.uk/~cat/Tangled/tangled.html and my contributions to
sci.physics in other threads in the past week - and two years ago, if you
can find them. Though most of my work is not published, I have written to
many of the experimenters concerned and had considerable discussion with
them.
> But to suggest that you can show the predicted
> probabilities are wrong is nothing short of a lie,
Please read my paper "The Chaotic Ball, An Intuitive Analogy for EPR
Experiments", published as Found. Phys. Lett. 9, 357 (1996), and available
at http://xxx.lanl.gov/abs/quant-ph/9611037 .
Quite a number of sane people in this world think I'm right! I've
presented papers on the subject at conferences of quantum theorists and been
applauded with considerably more enthusiasm than most!
> destroys your credibility, along with your refusal to acknowledge
> experiments which demonstrate the particulate nature of light.
Have you studied the experimental reports yourself?
> Experiments far more conclusive than interference patterns, which only
> demonstrate the validity of a mathematical formula, not the existence of
> a wave.
Uh? Brainwashed? Sorry, I give up! I shall not respond to any more of
your messages unless you do actually read what I say and discuss it
intelligently. In the unlikely event that you succeed in "destroying my
credibility" you will not have done any service to the long-term future of
physics. (Sorry, that is going a bit further than perhaps you meant, but I
really am exasperated!)
Caroline
z@z
| c.h.thompson wrote:
| >As I said, the probabilities of QM are not even right all the time. They
| >are wrong in the EPR case,
|
| You may have evidence that the experiment is inconclusive in the EPR
| case. If you are correct it is scientifically important, and should have
| be published. But to suggest that you can show the predicted
| probabilities are wrong is nothing short of a lie, which, unfortunately
| destroys your credibility, along with your refusal to acknowledge
| experiments which demonstrate the particulate nature of light.
If Caroline's claim is "nothing short of a lie" then Bohr's solution
of the EPR paradox is even worse.
Prior to the EPR paper, the orthodox exponents of QM such as Bohr
and Heisenberg had subscribed to these principles:
1) no actions at a distance
2) the polarisation direction of emerging photons is undetermined
3) photon pairs with the same "polarisation" are possible
That these statements are logically inconsistent was shown in
the EPR paper. Bohr's rather enigmatic reply was a masterpiece
of rhetoric and sophistry, but not much more.
In the meanwhile physicists have become accustomed to these strange
EPR actions at a distance, and the EPR correlations are sold as
experimentally confirmed, original QM predictions.
I ask you, Francis, does it make sense to sacrifice at first
so much of physical simplicity and of common sense to the
belief that actions at a distance are impossible, and later
reintroduce actions at distance in such a special context.
I think it was only face saving of those who had claimed that
QM is the best and most complete description of reality ever
possible.
If actions at distance are part of nature (and there is a lot
theoretical and empirical evidence), then we must restart
physics from its state at Maxwell's time.
Albert Einstein, 'Das Fundament der Physik', 1940, around p.2
"... Während aber bei einem schweren Sturm oder einer Springflut
ein Gebäude schwer beschädigt werden mag, ohne dass das Fundament
schaden erleidet, ist in der Wissenschaft das logische Fundament
in grösserer Gefahr, durch neue Erfahrungen oder sonstige neue
Erkenntnisse erschüttert zu werden als die in engster Fühlung mit
den Erfahrungstatsachen gewachsenen Teildisziplinen. In der
Verbundenheit mit allen Teilen liegt die Bedeutung des Fundamentes,
aber auch seine gefährdete Stellung allem Neuen gegenüber. Hält
man sich dies lebhaft vor Augen, so kann man sich nur darüber
wundern, dass die sogenannten Revolutions-Perioden der Physik das
Fundament nicht öfter und in stärkerem Masse verändert haben, als
es tatsächlich der Fall gewesen ist ..."
Cheers, Wolfgang
Jon Bell
>In article <7vavmc$i8n$1...@jetsam.uits.indiana.edu>,
> glha...@steel.ucs.indiana.edu (Gregory L. Hansen) wrote:
>>
>> Energy of a photon is E=hf, Planck's constant multiplied by the
>>frequency. Total energy of light in a region is U=nE, number of photons
>>multiplied by the energy of a photon. Intensity is I=U/A=nE/A=nhf/A,
>>energy divided by area.
>
>the n would have to be per unit of area.
Frequency is the number of cycles per second. When a light wave passes a
point, the electric and magnetic fields at that point oscillate back and
forth at a rate of f cycles per second. You can write it as f = n/t, but
keep in mind that this n is different from the n in U = nE above (which is
the number of photons in a given volume of space).
>Does Planck's constant have any units attached to it?
Yes, it has units of energy*time e.g. Joule*seconds. This makes Planck's
formula for the energy of a photon work out properly. E = hf so Joules =
(Joule*seconds)*(1/seconds). Cycles don't "count" when doing dimensional
analysis.
>What is the difference in energy for individual photons of different
>frequencies attributed to? They all have the same velocity.
I assume you're thinking of the classical kinetic energy formula K =
0.5mv^2. Relativistically the general relationship between mass, energy
and momentum of a particle is E^2 = (pc)^2 + (mc^2)^2 where m is what many
people call the "rest mass" of the particle. For a photon m = 0 so this
becomes E = pc. Photons (and other massless particles) have energy and
momentum, even though they have no (rest) mass.
Jon Bell
>
>The difference in energy for photons of different frequencies is
>attributed to the frequency!
Right, that was Planck's big *assumption*, borne out by the results of his
theory.
>It's a result of classical wave mechanics
>that waves with higher frequency have higher energy.
??? For a classical electromagnetic wave at least, the average energy
density of the wave depends only on the amplitude of the electric and
magnetic fields: u = 0.5 * epsilon_0 * c * E_max^2 (and a similar
equation with B_max).
Bruce Richmond
jtb...@presby.edu (Jon Bell) wrote:
> Bruce Richmond <bsr...@my-deja.com> wrote:
> >In article <7vavmc$i8n$1...@jetsam.uits.indiana.edu>,
> > glha...@steel.ucs.indiana.edu (Gregory L. Hansen) wrote:
> >>
> >> Energy of a photon is E=hf, Planck's constant multiplied by the
> >>frequency. Total energy of light in a region is U=nE, number of
photons
> >>multiplied by the energy of a photon. Intensity is
I=U/A=nE/A=nhf/A,
> >>energy divided by area.
> >
> >How is the frequency defined? It must include n/t and I would think
> >the n would have to be per unit of area.
>
> Frequency is the number of cycles per second. When a light wave
> oscillate back and forth at a rate of f cycles per second.
> You can write it as f = n/t, but keep in mind that this n is
> different from the n in U = nE above (which is the number of
> photons in a given volume of space).
>
We were discussing light being modeled as individual particles, and a
single particle as we normally think of a particle cannot have a
frequency. That's why I assumed f represented a count of the particles
hitting a given area per unit of time. The area must be taken into
account or you could vary the frequency by changing the area of your
detector.
In practice I imagine that frequency is not determined with a detector
actually counting photons but by using a prism to select a particular
frequency for study. IMHO though, if light is to be described as a
particle it should be done without any reference to waves or fields.
The only way I could see wave like descriptions as being acceptable
would if the model included the particles being emitted in bursts, like
volleys of gunfire. This model would also need some way of keeping the
particles in bunches, otherwise the frequency would quickly drop with
distance.
> >Does Planck's constant have any units attached to it?
>
> Yes, it has units of energy*time e.g. Joule*seconds. This makes
> Planck's formula for the energy of a photon work out properly.
> E = hf so Joules =(Joule*seconds)*(1/seconds). Cycles don't
> "count" when doing dimensional analysis.
>
> >What is the difference in energy for individual photons of different
> >frequencies attributed to? They all have the same velocity.
>
> I assume you're thinking of the classical kinetic energy formula K =
> 0.5mv^2. Relativistically the general relationship between mass,
> energy and momentum of a particle is E^2 = (pc)^2 + (mc^2)^2 where
> m is what many people call the "rest mass" of the particle. For a
> photon m = 0 so this becomes E = pc. Photons (and other massless
> particles) have energy and momentum, even though they have no
> (rest) mass.
>
If photons have no mass why are they affected by a gravitational field?
Gregory L. Hansen
> Gregory L. Hansen <glha...@steel.ucs.indiana.edu> wrote:
>>
>>The difference in energy for photons of different frequencies is
>>attributed to the frequency!
>
>Right, that was Planck's big *assumption*, borne out by the results of his
>theory.
And it seems to have worked out pretty well. The photons can be frequency
seperated by a diffraction grating and their energies measured.
>
>>that waves with higher frequency have higher energy.
>
>??? For a classical electromagnetic wave at least, the average energy
>density of the wave depends only on the amplitude of the electric and
>magnetic fields: u = 0.5 * epsilon_0 * c * E_max^2 (and a similar
>equation with B_max).
The energy of everyone's favorite prototypical wave, the wave on a string,
depends on frequency and amplitude. That result carries all over the
place in wave mechanics. For instance if a cork were bobbing up and down
in the water, it seems evident that its energy would depend on how quickly
it's moving, which in turn depends on the frequency. In electrodynamics
remember that the gradient of one field depends on the time derivative of
the other, and that the power radiated by an oscillating electric dipole
goes as the fourth power of frequency.
c.h.thompson
> >Gregory L. Hansen <glha...@steel.ucs.indiana.edu> wrote
>
> the experiment only obeys QM predictions at a certain temperature, you
> need to find out why.
Agreed! But even by careful reading of both Aspect's and Freedman's PhD
theses (the former in French, which was hard going) I have been unable to
find out what happened to the Bell test statistics at different
temperatures. I can guess that the temperatures actually used - room
temperature for some detectors, liquid nitrogen or some such for others -
were necessary in order to keep dark rates reasonably low and at the same
time maintain the "right" level of sensitivity.
Can you access PostScript files? All the technical details from Aspect's
thesis are in the part of Section 6 that I translated and put on my web
site. See http://www.aber.ac.uk/~cat/Aspect/thesis.ps .
> Sometimes particular temperatures are chosen, but
> not arbitrarily so that the experiment will match predictions.
True. In Aspect's case it seems to have been an important criterion that,
when one of the beams from the source was passed first through a polarising
filter then through the experimental one, the orientation of which was
varied, the resulting curve for the "singles counts" was as expected if
Malus' Law were being obeyed.
So far as wave theory is concerned, one would expect Malus' Law to be obeyed
as regards intensities, no matter what range these cover. Output intensity
from the experimental polariser should be cos^2 theta times input, where
theta is the angle between the axes of the initial filter and the moveable
polariser.
But this will only translate into a high-visibility sine curve for
PROBABILITIES if the detector is mimicking the QT prediction. To get the
desired result, the probability of detection has to be - as QT assumes it
automatically is - proportional to input intensity. QT assumes this because
it assumes the input intensity can only vary when the number of photons
varies. Classical theory says it can vary from a combination of several
causes, including number of pulses and, most importantly, amplitude of each
pulse. In the case in question, it is the amplitudes that vary.
So Aspect thought he had set his system to obey Malus' Law, but in reality
he had calibrated his instrument using this beam that had first passed
through a polarising filter. It therefore, according to classical theory,
consisted of a mixture of pulses of different amplitudes, since it is
assumed that the source produced a mixture of pulses of different
polarisations! Weak inputs could be producing one curve, strong ones
another, and we only see the average. This is one reason the calibration is
suspect.
Another is that it is not 100% clear that he kept the same intensity of
source all the time ...
The grave omission in Aspect's and all the EPR experiments is that only one
result is published for each. In some cases absolutely no information is
published about the actual data, only the derived Bell test statistic. It
is rare indeed to see more than a graph of summarised data, in all
likelihood "normalised" in some way. It is not possible for any but the
most dedicated to find out what really happened.
Caroline
c.h.thompson
z@z <z...@z.lol.li> wrote
> Charles Francis wrote:
>
> If Caroline's claim is "nothing short of a lie" then Bohr's solution
> of the EPR paradox is even worse.
>
> Prior to the EPR paper, the orthodox exponents of QM such as Bohr
> and Heisenberg had subscribed to these principles:
>
> 1) no actions at a distance
> 2) the polarisation direction of emerging photons is undetermined
> 3) photon pairs with the same "polarisation" are possible
>
> That these statements are logically inconsistent was shown in
> the EPR paper.
Hmm ... it was not exactly that. They were talking about position and
momentum, but I expect you're roughly on the right lines.
>Bohr's rather enigmatic reply was a masterpiece
> of rhetoric and sophistry, but not much more.
Agreed! Much of Bohr's writing was like this. It impressed but without
conveying much meaning. I've got some beautiful quotes lying around
somewhere ...
> In the meanwhile physicists have become accustomed to these strange
> EPR actions at a distance, and the EPR correlations are sold as
> experimentally confirmed, original QM predictions.
>
> I ask you, Francis, does it make sense to sacrifice at first
> so much of physical simplicity and of common sense to the
> belief that actions at a distance are impossible, and later
> reintroduce actions at distance in such a special context.
> I think it was only face saving of those who had claimed that
> QM is the best and most complete description of reality ever
> possible.
> If actions at distance are part of nature (and there is a lot
> theoretical and empirical evidence), then we must restart
> physics from its state at Maxwell's time.
Now you've got me mystified! In another message (EPR Paradox - explanation
requested. 25 October) you said "There is a huge difference between "common
sense" actions at a
distance and "spooky" EPR actions."
I took this as meaning that you rejected all spooky ones. So to what kind
does the above statement that they are a part of nature refer? "Common
sense" ones, presumably - ones that are not really instantaneous effects but
the result of waves at some finite speed, though they may be modelled by
approximate formulae that ignore speed of propagation.
> Albert Einstein, 'Das Fundament der Physik', 1940, around p.2
>
> "... Während aber bei einem schweren Sturm oder einer Springflut ..."
Any chance of a translation?
Cheers
Caroline
Jim Carr
} theory". You can have "lumpy wave theories"! For a start, you can get
} regions of high and low intensity due to interference; you can allow for
} emission in pulses; you can assume that the receptor is "primed" by local
} (lumpy) noise. One way or another, these influences can cause an
} effectively immediate response.
In article <7vepte$99a$2...@flotsam.uits.indiana.edu>
glha...@steel.ucs.indiana.edu (Gregory L. Hansen) writes:
>
>How does the lumpy wave theory hold up experimentally? It seems to me
>that the lumpy wave theory still doesn't lead to a cut-off when the light
>is too red.
It would be equally important to ask how the proposed theory differs
from classical E+M in dealing with the Compton Effect, which cannot
be attributed to bound-state energy levels (it concerns continuum
states) and where the angular distribution of scattered photons is
not what you get from standard classical E+M theory.
--
James A. Carr <j...@scri.fsu.edu> | Commercial e-mail is _NOT_
http://www.scri.fsu.edu/~jac/ | desired to this or any address
Supercomputer Computations Res. Inst. | that resolves to my account
Florida State, Tallahassee FL 32306 | for any reason at any time.
Jim Carr
>
>Now lets get down to hard facts. For more please do look at my web site,
>and you might do worse than to start by glancing at "The Tangled Methods of
>Quantum Entanglement", http://www.aber.ac.uk/~cat/Tangled/tangled.html ,
>which describes the way in which my attempts to publicise KNOWN weaknesses
>in the EPR experiments have failed to reach the pages of PRL and PRA. Some
>of my papers are in http://xxx.lanl.gov/abs/quant-ph: 9611037, 9711044,
>9903066.
Your complaint here is not, _strictly_ speaking, accurate.
Your preprint in quant-ph/9711044 was cited in Tittel's Physical
Review Letter [PRL 81, 3563 (1998)] in reference 10 and there is
a 'live' link to your preprint from the PRL-online version of the
Tittle paper. So it is "in" the journals in that sense.
Some of the loopholes you (and others) have talked about have been
addressed -- IMO the Weihs experiment, which shows no need to do
background subtraction deals with one of yours -- while others are
harder to deal with in the real world of experiment. Tittel's
latest (??) paper in Phys. Rev. A [PRA 59, 4150 (1999)] talks
about this extent.
Is your "tangled methods" article just a cutely titled rehash of the
things you have written about before, or do you have something specific
to say about, for example, the Weihs experiment and the need for the
subtraction of background to get non-local effects?
Jim Carr
} probabilities.'
In article <381b3001$1...@news1.vip.uk.com>
"c.h.thompson" <c.h.th...@newscientist.net> writes:
>
>And it can't even get those right!
Overstatement to the point where your assertion is nonsense does
not help your case. For example, in Rep. Prog. Phys. 51, 299
(1988) you can see an example of exactly one such probability
graph compared to data that does an extremely good job of
predicting the probability distribution of the electron in the
hydrogen atom. Note that since this was the first wavefunction
calculated by Schroedinger, the application of the interpretation
of the wavefunction as a probability applitude to it would make
this a test of the _first_ such prediction.
Note further that all of the 'Bell' tests are further examples,
and that QM does better than your chaotic ball.
>Look, probabilities are things you measure, but they are not physical
>objects.
Not a particularly important distinction. And I would say that
in the sense of the words you are using, probabilities are not
_things_ that you measure. They are precisely what the name
says they are -- the accumulated statistics of some set of
experimental measurements.
>You can't have a genuine fundamental theory that only deals in
>such things!
You would argue that you cannot have a fundamental theory that
only deals "in such things" as the probabilities of dice combinations
or of drawing a particular card? Because you are, arbitrarily,
defining statistics as not being fundamental? Fine. Then just
drop the loaded words and focus on what experiment says about
local hidden variable theories.
>Such a theory is, as Einstein, Podolsky and Rosen so rightly
>said, incomplete.
Maybe, but if they are right about that, they were wrong about
what experiment would show when certain kinds of tests were done.
The EPR argument, the Bell inequality, and the CHSH special case
that is particularly amenable to precision experimental tests, are
about whether an entire class of theories can be used as a replacement
for QM. The answer is pretty clear that they cannot.
>Underlying those parts of QM that do give the right answers (interference
>patterns etc) there are real waves behaving as real waves always have done.
Only if those real waves are complex probability amplitudes.
I include such things as BEC, neutron interferometry, 'Bell' tests,
and scattering experiments as falling in that category.
>The detectors that are used artificially turn these into probabilities.
In the same way that recording a series of dice throws is some sort
of "artificial" act? I don't see the point of your casual use of
such terms. Sounds more like propaganda than physics.
People have _counted_ events since long before QM came along. It is
called a spectrum.
>As I said, the probabilities of QM are not even right all the time.
Examples, please, and please be careful to separate the cases
where one knows the Hamiltonian from those where H is still not
well understood. However, this _very_ strong statement is not
at all consistent with the weaker remark you follow with as a
point of support:
>They
>are wrong in the EPR case, and experiments have unjustifiably been
>interpreted as if they are right.
They are not "wrong" in the published experiments I have seen,
where the probabilities agree within uncertainties with QM and
clearly exclude local hidden variable theories.
>As a consequence it is generally believed
>that the real world does not allow a theory based on "hidden variables",
>such as the real amplitudes, phases and frequencies of the waves that are
>claimed to be mere probability waves.
That is not so. Note the word "local" above. Now what you say
may be what is _generally_ believed in some subset of the population
that reads only certain popularizations of physics, but let us try
to talk about specific experiments here.
Jim Carr
>number of experiments that could be done to help you on the way. Some
>critical ones would have the aim of proving that the light used, though in
>most cases coming in pulses, was not in the form of "photons".
So examine the experiments with neutrons that show interference.
In what way do you think that the results of scattering particles
from a crystal differ from those where you scatter photons from a
crystal?
>When light
>is split at a "two-channel polariser" it does not divide in the form of
>whole photons, of fixed energy. It keeps its same frequency, yes, but the
>amplitude (the ordinary, classical, wave amplitude) is decreased.
So you think the same thing happens to neutrons? BECs?
If so, why do you think this view is "simpler" or somehow
less exotic than QM? (I'm also curious what you think of
the Sansbury theory that detectors know how far away the
target is.)
If not, how do you explain the results of those experiments?
Jim Carr
| c.h.thompson wrote:
|
| >are wrong in the EPR case,
|
| case. If you are correct it is scientifically important, and should have
| be published. But to suggest that you can show the predicted
| probabilities are wrong is nothing short of a lie, which, unfortunately
| destroys your credibility, along with your refusal to acknowledge
| experiments which demonstrate the particulate nature of light.
In article <7vhhgm$qs6$1...@pollux.ip-plus.net>
"z@z" <z...@z.lol.li> writes:
>
>If Caroline's claim is "nothing short of a lie" then Bohr's solution
>of the EPR paradox is even worse.
>
>Prior to the EPR paper, the orthodox exponents of QM such as Bohr
>and Heisenberg had subscribed to these principles:
>
> 1) no actions at a distance
> 2) the polarisation direction of emerging photons is undetermined
> 3) photon pairs with the same "polarisation" are possible
>
>That these statements are logically inconsistent was shown in
>the EPR paper. Bohr's rather enigmatic reply was a masterpiece
>of rhetoric and sophistry, but not much more.
Yes, which is why his philosophical arguments are pretty much
ignored today. You need to go searching in scientific biographies
for references to those remarks. Better to focus attention on
what QM predicts, and formal theorems like Bell's or the CHSH
relation, than all of those "interpretations".
>In the meanwhile physicists have become accustomed to these strange
>EPR actions at a distance, and the EPR correlations are sold as
>experimentally confirmed, original QM predictions.
You mean that they have gotten used to the agreement between
the predictions of QM and the results of experiment? Yes.
Jim Carr
"c.h.thompson" <c.h.th...@newscientist.net> writes:
>
> ... I can guess that the temperatures actually used - room
>temperature for some detectors, liquid nitrogen or some such for others -
>were necessary in order to keep dark rates reasonably low and at the same
>time maintain the "right" level of sensitivity.
Yes. Some detectors will not work, will just spew noise, or even
be damaged, if not kept within some temperature range.
> ... To get the
>desired result, the probability of detection has to be - as QT assumes it
>automatically is - proportional to input intensity.
The efficiency of detectors is measured using a known source.
>The grave omission in Aspect's and all the EPR experiments is that only one
>result is published for each. In some cases absolutely no information is
>published about the actual data, only the derived Bell test statistic. It
>is rare indeed to see more than a graph of summarised data, in all
>likelihood "normalised" in some way. It is not possible for any but the
>most dedicated to find out what really happened.
This is not true for _all_ EPR experiments.
Some (e.g. Weihs) have put the raw data files on the web and others
include much more than one result.
z@z
| z@z (Wolfgang) wrote:
| > If actions at distance are part of nature (and there is a lot
| > theoretical and empirical evidence), then we must restart
| > physics from its state at Maxwell's time.
|
| Now you've got me mystified! In another message (EPR Paradox - explanation
| requested. 25 October) you said "There is a huge difference between "common
| sense" actions at a distance and "spooky" EPR actions."
|
| I took this as meaning that you rejected all spooky ones. So to what kind
| does the above statement that they are a part of nature refer? "Common
| sense" ones, presumably - ones that are not really instantaneous effects but
| the result of waves at some finite speed, though they may be modelled by
| approximate formulae that ignore speed of propagation.
I reject EPR actions at a distance. But I think that nobody
should accept EPR actions at a distance and reject the actions
at a distance of pre-Maxwellian physics, because the latter are
more general than the former.
We can think of "material" models leading to the well-known
instantanous inverse distance square laws. We only have to
assume incompressible fluids. If the whole universe were filled
with such a fluid, the simple assumption that matter draws
in (and annihilates) fluid proportionally to the mass leads to
instantanous attraction obeying the inverse distance square law.
The important point is not whether all forces are mediated by
material causes, but whether they act instantanously or at the
speed of light.
If one calculates the actual vectors by which the earth is
accelerated by other planets, we find out that these vectors
point to the locations where these planets are now, and not
where they have been when photons arriving now were emitted.
So if we explain gravitation by gravitons (or by a field
propagating at c), we must assume that gravitons originating
from e.g. Jupiter must remain in telepathic connection with
Jupiter. Only such a telepathic link (or complicated calculations
in advance) can garantee that the gravitons accelerate the
earth in the direction where Jupiter is now.
See once again http://www.deja.com/=dnc/getdoc.xp?AN=538880440
Gruss, Wolfgang
Nathan Urban
> If photons have no mass why are they affected by a gravitational field?
FAQ:
http://www.corepower.com/~relfaq/light_mass.html
[Note followups.]
Bruce Richmond
"c.h.thompson" <c.h.th...@newscientist.net> wrote:
>
> Gregory L. Hansen <glha...@steel.ucs.indiana.edu> wrote
> > c.h.thompson <c.h.th...@newscientist.net> wrote:
> > >Gregory L. Hansen <glha...@steel.ucs.indiana.edu> wrote
> > >>
> > >> How does the lumpy wave theory hold up experimentally? It seems
to me
> > >> that the lumpy wave theory still doesn't lead to a cut-off when
the
> light
> > >> is too red.
>
the
> > >frequency exceeds a threshold, my lumpy wave theory will need two
> factors.
> > >Remember, it applies to light that requires to be boosted by noise
as
> > >otherwise individual pulses are just too weak to be detected. To
get a
> > >detection we must need a combination of:
> > >(a) coupling of the field of the cathode to the incoming field, and
> > >(b) an extra "kick" from noise.
> > >
> > >It is possible that (a) just does not happen if the frequency is
too low,
> > >but I'm not God! I don't know.
>
> I repeat, I don't know! And thinking about it the above is not very
> satisfactory as it places too much emphasis on a resonance effect.
It does
> not explain why you get emission at all frequencies above a minimum.
>
Just speculation here, but might it be possible that at low frequencies
the whole atom tends to move (vibrate) together so the electrons don't
get shaken free? Think of trying to shake berries off a bush. It
doesn't matter how far you push back and forth, without the a sudden
reversal of the motion the berries wont shake loose.
As far as the continued emission at frequencies higher than the
threshold, what says all the electrons are coming from the same shell?
Using the bush shaking analogy, if you grab the bush at the base and
shake, the berries at the top receive the greatest acceleration and
will shake free first. It takes more vigorous shaking (higher
frequency) to free the berries further down.
> your other message) see my reply to Nathan Urban.
>
> Caroline
> <http://www.aber.ac.uk/~cat>
Jim Carr
Katie Schwarz <k...@socrates.berkeley.edu> wrote:
}
} Gregory L. Hansen <glha...@steel.ucs.indiana.edu> wrote:
} >Of course, the photoelectric effect is what started it all.
} >
} >For proof beyond a shadow of a doubt, he might need to get his own
} >photomultiplier and damp down a light source until the signal becomes
} >discrete.
}
} Actually neither of these is really solid proof. The photoelectric
} effect can be explained by a semiclassical theory where the energy of
} the atoms is quantized but the electromagnetic field is classical.
In article <7v5q1g$cle$3...@flotsam.uits.indiana.edu>
glha...@steel.ucs.indiana.edu (Gregory L. Hansen) writes:
>
>Are you sure? The reason the photoelectric effect was important because
>they tried to explain it with classical electromagnetism by assuming
>energy will build up over time until an electron is excited. But the
>response of the electrons was immediate, and no intensity of light would
>create the effect if it was the wrong color.
You have to remember the historical context. The photoelectric effect
was explained (along with Planck's radiation law derivation) with a
photon derived by using a statistical mechanics analysis that treated
the electromagnetic field as akin to a macroscopic ensemble. There
was no quantum mechanics to justify the Planck assumption (which is
dicey in the "old" QM of Bohr and Sommerfeld anyway, as it ignores
any features specific to a particular atom) or to explain the work
function in the photo-effect.
The semi-classical explanations could be justified later, after
quantum mechanics was fully formed. An historical approach to
teaching the various elements that contributed to the development
of QM and, eventually, QED does not backtrack to mention this. It
may not even mention the compelling argument Compton made for the
photon with his observation and derivation.
Charles Francis
scientist.net> writes
>
>Charles Francis <cha...@clef.demon.co.uk> wrote
>c.h.thompson <c.h.th...@newscientist.net> writes
>> >
>> >patterns etc) there are real waves behaving as real waves always have
>done.
>>
>> way you can deduce any such thing from "interference" patterns. So
>> called interference comes from hypothetical statements made in quantum
>> mechanics about what would happen if an experiment were to be done, not
>> from physical waves.
>
>Have you never seen interference patterns on the surface of a pond?
>
>Are these due to "hypothetical statements made in QM??????" How ridiculous
>can you get!
It is only ridiculous to think that interference patterns on a pond have
anything to do with quantum effects. This is like observing that three
apples plus four apples is seven apples and three oranges plus four
oranges is seven oranges, and concluding that apples is oranges.
>It is QM that invented the crazy idea of interference of
>probability waves ....
>
>But isn't it about time you looked at my web site? I thought you had.
I have. There is a difference between finding experimental loopholes and
asserting things which are not true. If you want to challenge the
experimental foundations of quantum mechanics you will have to do better
than tackle a somewhat reserche experiment (Aspect) of little more than
pedagogical value, and of extreme subtlety in both its execution and
interpretation. This is like tackling the whole of Newtonian mechanics
on the basis of the anomalous orbit of mercury.
>
>> >As I said, the probabilities of QM are not even right all the time. They
>> >are wrong in the EPR case,
>>
>> You may have evidence that the experiment is inconclusive in the EPR
>> case. If you are correct it is scientifically important, and should have
>> be published.
>
>My dear Charles, may I suggest that you read my essay (published in the
>journal Accountability in Research) on the subject of my difficulties in
>getting the facts I and other realist have unearthed published!
I am a realist myself, and I know from hard experience that realist
theories will not even be sent by the editors of journals to a referee.
> See
>http://www.aber.ac.uk/~cat/Tangled/tangled.html and my contributions to
>sci.physics in other threads in the past week - and two years ago, if you
>can find them. Though most of my work is not published, I have written to
>many of the experimenters concerned and had considerable discussion with
>them.
>
It would be more interesting if your discussions could have been with
the people responsible for the experimental foundation of quantum
mechanics, Heisenberg, Schrodinger, or, if you had actually come to
grips with the experimental basis for quantum mechanics as described by
Dirac and Von Neumann.
>
>Have you studied the experimental reports yourself?
>
As I say, the problem is that you are not studying the significant
experiments, or the interpretation of experimental results from the
point of view of general measurement theory, which you declare to be
irrelevent. As far as I can tell, you do not have an idea of what
quantum mechanics is or what it says when it is put into a reasonably
rigorous form, but are hung up on some trite, though perhaps commonly
held misinterpretation of it, rooted in Schrodinger's wave equation. I
can only urge you to study the theory properly before shooting off on
these misinterpretations.
>> Experiments far more conclusive than interference patterns, which only
>> demonstrate the validity of a mathematical formula, not the existence of
>> a wave.
>
>Uh? Brainwashed? Sorry, I give up! I shall not respond to any more of
>your messages unless you do actually read what I say and discuss it
>intelligently.
I have read what you have to say. But until you read what I have to say,
and read extensively on the conceptual foundations of quantum mechanics
(I can lend you some books as you live so near) I doubt there is any
chance of an intelligent discussion.
>In the unlikely event that you succeed in "destroying my
>credibility" you will not have done any service to the long-term future of
>physics.
I do not wish to destroy your credibility. On the face of it, I think
you have something important to say and I am trying to stop you
destroying your own credibility, so that you may get the chance to put
it into a proper context and say it in such a way that you will be taken
seriously. That would be of far greater long-term service to physics.
--
Charles Francis
cha...@clef.demon.co.uk
Charles Francis
>Charles Francis wrote:
>| c.h.thompson wrote:
>
>| >are wrong in the EPR case,
>|
>| You may have evidence that the experiment is inconclusive in the EPR
>| case. If you are correct it is scientifically important, and should have
>| probabilities are wrong is nothing short of a lie, which, unfortunately
>| destroys your credibility, along with your refusal to acknowledge
>| experiments which demonstrate the particulate nature of light.
>
>of the EPR paradox is even worse.
>
>Prior to the EPR paper, the orthodox exponents of QM such as Bohr
>and Heisenberg had subscribed to these principles:
>
> 1) no actions at a distance
> 2) the polarisation direction of emerging photons is undetermined
> 3) photon pairs with the same "polarisation" are possible
>
>That these statements are logically inconsistent was shown in
>the EPR paper. Bohr's rather enigmatic reply was a masterpiece
>of rhetoric and sophistry, but not much more.
I agree with you that a lot of very inaccurate things have been said
about quantum mechanics, and I am actually particularly against the
Copenhagen school.
>
>In the meanwhile physicists have become accustomed to these strange
>EPR actions at a distance, and the EPR correlations are sold as
>experimentally confirmed, original QM predictions.
>
>I ask you, Francis, does it make sense to sacrifice at first
>so much of physical simplicity and of common sense to the
>belief that actions at a distance are impossible, and later
>reintroduce actions at distance in such a special context.
>I think it was only face saving of those who had claimed that
>QM is the best and most complete description of reality ever
>possible.
>
I agree with Einstein, that it is not complete and a better description
is possible. That is what I aim to show in my papers.
>If actions at distance are part of nature (and there is a lot
>theoretical and empirical evidence), then we must restart
>physics from its state at Maxwell's time.
>
I do not accept action at a distance either, but I do accept the
mathematical theory in the form given us by Dirac. All that is really
necessary is to understand that the mathematics of Dirac's particulate
theory of qed (as opposed to the field version) can be fixed, and that
despite Dirac's misgivings, there were only ever technical mathematical
adjustments necessary to the basis of the theory for it to make complete
and total sense.
--
Charles Francis
cha...@clef.demon.co.uk
c.h.thompson
Jim Carr <j...@ibms48.scri.fsu.edu> wrote in message
news:7vi7j3$llm$1...@news.fsu.edu...
Hi Jim
I'm glad you've joined in, and thanks. It's nice to know that there is a
glimmer of recognition of my work - albeit embedded in a list of other
refs - in that Tittel article in Physical Review Letters (PRL 81, 3563
(1998)).
> >Now lets get down to hard facts. For more please do look at my web site,
> >and you might do worse than to start by glancing at "The Tangled Methods
of
> >Quantum Entanglement", http://www.aber.ac.uk/~cat/Tangled/tangled.html ,
> >which describes the way in which my attempts to publicise KNOWN
weaknesses
> >in the EPR experiments have failed to reach the pages of PRL and PRA.
Some
> >of my papers are in http://xxx.lanl.gov/abs/quant-ph: 9611037, 9711044,
> >9903066.
My Tangled Methods paper includes a fair amount on the subtraction of
accidentals, but this is better covered in quant-ph/9711044. Possibly the
most important section, though, is the appendix,
http://www.aber.ac.uk/~cat/Tangled/appendix.html, listing a number of
different loopholes, one or more of which is open (and known to be open!)
in every Bell Test experiment conducted so far.
> Some of the loopholes you (and others) have talked about have been
> addressed -- IMO the Weihs experiment, which shows no need to do
> background subtraction deals with one of yours -- while others are
> harder to deal with in the real world of experiment. Tittel's
> latest (??) paper in Phys. Rev. A [PRA 59, 4150 (1999)] talks
> about this extent.
Yes, various experiments do close various loopholes, but only one at a time!
We have a most gigantic logical/scientific error here if we accept any
experiment as having demonstrated nonlocal effects!
Is the following logical or not?
Step 1 (Optional) : Select a particular "loophole" to be kept under control.
Step 2: Conduct an EPR experiment and calculate (one of the possible) Bell
test statistics. The use of this test will involve making certain
assumptions regarding ALL the loopholes. Let us be quite clear: if those
assumptions are false then the test has no meaning as local realist models
can equally well violate it.
Step 3: Apply the selected test and observe that it is violated.
Step 4: Deduce that QM is correct.
The correct deduction, as the experimenters and others in the know are
aware, is that EITHER QM is correct OR one or more of the loopholes is open
(or, other words, one or more assumptions are false).
The psychology (if I dare mention such a word in a serious physics
discussion) that leads to the above false logic goes something like this:
(a) We observe a result that agrees with QM
(b) This result depends on some assumptions
(c) We know from other experiments that QM is correct
(d) So the assumptions must be correct!
I could go on to discuss social factors, but perhaps I can leave these to
others.
As a general point of scientific method, I would suggest that the
experiments to date would have been of very much more use to the community
had they been conducted so as to explore wider parameter ranges. If, say, a
range of different detector voltages, or a range of different coincidence
windows, had been used, the likely magnitude of biases due to loopholes
might have become clear.
Many of the loopholes are tied up with the QM hypothesis that light has a
particle nature (hence my strong interest in the thread "Who says light
behaves as a particle?"). Surely investigation of this should be top
priority?
I have, as I believe I have told you, challenged Zeilinger to try an
experiment using a pair of beamsplitters
(http://www.aber.ac.uk/~cat/Suggestions/two_bs.htm). I have asked several
times now for news of progress on this, with no response. The idea of the
experiment is to show that if you set detectors so as to appear to get QM
predictions (zero coincidences, as a "photon" can only exit on way or the
other) at one beamsplitter, these same settings will result in wrong answers
at the next. The coincidences will, I believe, depend strongly on the
detector characteristics.
So the response characteristics of the detectors used needs to be studied.
Perhaps a direct study is not feasible, as no calibration method is likely
to satisfy both quantum theorists and realists. A method that I am myself
very interested in is the indirect one of using very carefully controlled
EPR coincidence curves to DEDUCE the characteristics. This method, though,
depends on accepting the hypothesis of local realism!
> Is your "tangled methods" article just a cutely titled rehash of the
> things you have written about before, or do you have something specific
> to say about, for example, the Weihs experiment and the need for the
> subtraction of background to get non-local effects?
I say something specific about two of Aspect's experiments, in which the
figures speak for themselves: subtraction of accidentals produced large
changes in the test statistics. With no subtraction we get no Bell test
violation. (I have, incidentally, sent this paper and many others to
Aspect. He has not commented.)
The article was written before the Weihs experiment was published.
Of course, as recent experiments have shown, not all EPR experiments involve
subtraction of accidentals. There have been many violations without it. In
the vast majority of these it will be found that the experimenters realise
that the "detection loophole" (alias "fair sampling assumption") leaves
room for possible local realist interpretations.
Incidentally, though I still have an affection for my first paper, The
Chaotic Ball (quant-ph/9611037), this applies much more directly to those
hypothetical Stern-Gerlach experiments than to the majority of real ones.
Real ones use light, and the natural local realist model for these is given
in quant-ph/9903066. I have nearly completed another paper (on rotational
invariance and some new ideas on the EPR source that is so popular these
days - the degenerate case of type II parametric down-conversion) that
covers this more fully. A draft is available in Latex on request.
Cheers
Caroline
c.h.thompson
> >number of experiments that could be done to help you on the way. Some
> >critical ones would have the aim of proving that the light used, though
in
> >most cases coming in pulses, was not in the form of "photons".
>
Have there been EPR experiments using neutrons? If not, then I'd rather
stick to discussing just light, as used in the optical experiments I've
studied.
> (I'm also curious what you think of
> the Sansbury theory that detectors know how far away the
> target is.)
I haven't met this but can say with no further ado that is sounds nonsense.
Caroline
c.h.thompson
> Joe Rongen <joer...@whisp.com> wrote
> in message news:IdFS3.319$u45...@198.235.216.4...
> }
> } 'The predictions of quantum mechanics are expressed in terms of
> } probabilities.'
>
> In article <381b3001$1...@news1.vip.uk.com>
> "c.h.thompson" <c.h.th...@newscientist.net> writes:
> >
> >And it can't even get those right!
>
> Overstatement to the point where your assertion is nonsense does
> not help your case.
Point taken.
I was talking in the context of EPR correlations, though, in which I'm 100%
sure of my facts. I was not trying to say ALL QM predictions were wrong.
> Note further that all of the 'Bell' tests are further examples,
> and that QM does better than your chaotic ball.
See my reply under "Bell Test Loopholes":
(i) My later papers talk in terms of a different realist model, well
accepted as suitable for experiments that use light. The Chaotic Ball,
though correct as a demonstration of principle, is not useful in practice as
Stern-Gerlach EPR experiments have never been done.
(ii) It is hard for me to accept that QM "does better" as I am so accutely
aware of the importance of the assumptions and the actual experimental
conditions! QM has never been shown to give good predictions of the true
EPR setup as a valid EPR experiment has never been done! Almost valid,
maybe, but always with room for realism. Physics is not forced to
incorporate magic.
> >Look, probabilities are things you measure, but they are not physical
> >objects.
>
> Not a particularly important distinction. And I would say that
> in the sense of the words you are using, probabilities are not
> _things_ that you measure. They are precisely what the name
> says they are -- the accumulated statistics of some set of
> experimental measurements.
>
> >You can't have a genuine fundamental theory that only deals in
> >such things!
>
> You would argue that you cannot have a fundamental theory that
> only deals "in such things" as the probabilities of dice combinations
> or of drawing a particular card? Because you are, arbitrarily,
> defining statistics as not being fundamental? Fine.
What I'm saying is that if the world is in reality continuous, and, in
particular, if light is purely wave, then you have thrown away a lot of
information by using instruments that only record it in the crude
presence/absence form that QM's detectors do. A theory that attempts to
reconstruct the real world from the resulting counts can only do so by
obscure methods. It is forced to talk always of ensembles. Its conclusions
are going to be ambiguous, as different detectors will give different
answers. The latter may always be true, but if we didn't pretend we were
dealing with particles it would not be a mystery!
> >Such a theory is, as Einstein, Podolsky and Rosen so rightly
> >said, incomplete.
>
> Maybe, but if they are right about that, they were wrong about
> what experiment would show when certain kinds of tests were done.
No Jim. They were right. The experiments have been misinterpreted, and if
Einstein had been around to defend his case possibly we would not now be in
this quandary.
> The EPR argument, the Bell inequality, and the CHSH special case
> that is particularly amenable to precision experimental tests, are
> about whether an entire class of theories can be used as a replacement
> for QM. The answer is pretty clear that they cannot.
Again, not true. The papers (some of them) may make this claim, but unless
an experiment is conducted that blocks all loopholes simultaneously the
claim is false. (See under "Bell Test Loopholes".)
>Underlying those parts of QM that do give the right answers (interference
>patterns etc) there are real waves behaving as real waves always have done.
> Only if those real waves are complex probability amplitudes.
There is a fairly successful realist theory called Stochastic
Electrodynamics (SED) that replaces QM's probability amplitudes by real
waves in a great number of cases. As far as I can gather, it takes over the
mathematics of QED.
> I include such things as BEC, neutron interferometry, 'Bell' tests,
> and scattering experiments as falling in that category.
Bell tests are different!
In all genuine interference cases, you are dealing with, as QM puts it,
"indistinguishable paths". Light (or whatever) can reach the point in
question by one of two or more paths. So what "really happens" is that the
light (let's forget neutrons etc for the time being) has been split at some
point and physically reunites at our test point. So we have real ordinary
interference. The Bell setup is different. Though QM uses much of the same
language, it is talking about interference between "photons" at different
locations. Totally different, and there is no evidence (propaganda again)
that it happens!
> >The detectors that are used artificially turn these into probabilities.
>
> In the same way that recording a series of dice throws is some sort
> of "artificial" act? I don't see the point of your casual use of
> such terms. Sounds more like propaganda than physics.
Dice really do have 6 distinct possible states. The result of a throw is a
discrete observable. The need to treat light in the same way has (witness
recent correspondence) never been satisfactorily established. The detectors
used in QM experiments record just counts, but they evolved from instruments
that recorded analogue information. QM, in "single-photon" experiments,
choses to use them in "Geiger mode". This, I am saying, is artificially
digitising the information.
> >As I said, the probabilities of QM are not even right all the time.
>
> Examples, please,
I am talking of only one QM probability, the one that controls the outcome
of EPR experiments.
> >They
> >are wrong in the EPR case, and experiments have unjustifiably been
> >interpreted as if they are right.
>
> They are not "wrong" in the published experiments I have seen,
> where the probabilities agree within uncertainties with QM and
> clearly exclude local hidden variable theories.
No! Results obtained ignoring loopholes don't count! And besides, most EPR
experiments talk as if they are using estimates of probabilities but are in
fact using ratios of counts that are not valid estimates of probabilities
and don't pretend to be. The standard test, of form -2<=S <= 2, uses ratios
in which the denominator is the total number of observed coincidences. The
visibility test uses the sum of the max and min. Have a look some day at
Clauser and Horne's 1974 paper (Physical Review D, 10, 526 (1974)). This
states categorically that the standard test can only be used if the number
of emissions is known and is used as the denominator.
A "normalised coincidence rate" is not at all necessarily a fair estimate of
a probability!
I do not know of many experiments that have used the improved test,
presented in the 1974 paper. This test uses ratios of counts, but it is
based on a different proof and does not pretend that they are probabilities.
(See quant-ph/9903066)
> Now what you say
> may be what is _generally_ believed in some subset of the population
> that reads only certain popularizations of physics, but let us try
> to talk about specific experiments here.
I've lost the thread here, but you are right. There is a definite
distinction between what the people in the field believe and what a certain
"subset" believes. Experimenters I've corresponded with will say things
such as "Oh, we weren't trying to demonstrate nonlocality! We just quoted
the Bell test as that's what people expect. It's the usual way of measuring
the "quality" of this kind of experiment!"
From what I can gather, though, the subset that believes that nonlocality
really happens is much larger than is justified by the evidence. It is not
just amateurs. It includes a great number of people in fields such as
quantum computing, a great number of PhD students doing research into some
consequence or other of quantum entanglement. If you yourself are at
present part of this subset, I do hope that I shall some day soon (before
the next millennium?) persuade you otherwise!
Cheers
Caroline
jmfb...@aol.com
> Bruce Richmond <bsr...@my-deja.com> wrote:
>>In article <7vavmc$i8n$1...@jetsam.uits.indiana.edu>,
>> glha...@steel.ucs.indiana.edu (Gregory L. Hansen) wrote:
>Yes, it has units of energy*time e.g. Joule*seconds. This makes Planck's
>formula for the energy of a photon work out properly. E = hf so Joules =
>(Joule*seconds)*(1/seconds). Cycles don't "count" when doing dimensional
>analysis.
<snip>
Curiously, not including cycles in dimensional analysis drove
me bonkers :-).
/BAH
Subtract a hundred and four for e-mail.
Gregory L. Hansen
Me, too. Especially since cycles and radians are technically unitless,
but they're not the same units.
Martin Green
the photo-electric effect continues to be put forward as evidence for the
existence of photons. Let me repeat what I posted a year ago, and which
has been extensively confirmed by published literature.
Given the quantized nature of the atomic energy levels, there is NO NEED
to invoke "photons" to explain the photo-electric effect. A quantum atom
placed
in a classical electromagnetic field will display ALL the standard
behaviors
of the photo-electiric effect...the cut-off frequency, electron emissions
in very
weak light, etc.
Martin Green
c.h.thompson
Jim Carr <j...@ibms48.scri.fsu.edu> wrote
> "c.h.thompson" <c.h.th...@newscientist.net> writes:
> >
> > ... I can guess that the temperatures actually used - room
> >temperature for some detectors, liquid nitrogen or some such for others -
> >were necessary in order to keep dark rates reasonably low and at the same
> >time maintain the "right" level of sensitivity.
>
> Yes. Some detectors will not work, will just spew noise, or even
> be damaged, if not kept within some temperature range.
>
> > ... To get the
> >desired result, the probability of detection has to be - as QT assumes it
> >automatically is - proportional to input intensity.
>
> The efficiency of detectors is measured using a known source.
Yes, maybe, yet at some point one has to have an absolute standard. It's a
real difficulty. Aspect included in his thesis a reprint of a paper
entitled "Absolute measurement of an atomic cascade rate using a two photon
coincidence technique" (Aspect, Imbert and Roger, Optics Communications 24,
p46 (1980)). Presumably he did not simply assume that he could count
photons and deduce the cascade rate.
Anyway, I should be interested to know how they calibrate detectors these
days. Do you know what the "known source" is?
> >The grave omission in Aspect's and all the EPR experiments is that only
one
> >result is published for each. In some cases absolutely no information is
> >published about the actual data, only the derived Bell test statistic.
It
> >is rare indeed to see more than a graph of summarised data, in all
> >likelihood "normalised" in some way. It is not possible for any but the
> >most dedicated to find out what really happened.
>
> This is not true for _all_ EPR experiments.
>
> Some (e.g. Weihs) have put the raw data files on the web and others
> include much more than one result.
I'm afraid my claim holds for Weihs' experiment as well as others. Though
there is a large quantity of data available it does not answer my questions
as it is only related to the specific conditions used in the report. It
does not tell me what happened for different parameter choices.
Caroline
<http://www.aber.ac.uk/~cat>
c.h.thompson
Jim Carr <j...@ibms48.scri.fsu.edu> wrote
> c.h.thompson <c.h.th...@newscientist.net> wrote:> }
> } You don't seem to have read what I said: there is not just one "wave
> } theory". You can have "lumpy wave theories"! For a start, you can get
> } regions of high and low intensity due to interference; you can allow for
> } emission in pulses; you can assume that the receptor is "primed" by
local
> } (lumpy) noise. One way or another, these influences can cause an
> } effectively immediate response.
>
> In article <7vepte$99a$2...@flotsam.uits.indiana.edu>
> >that the lumpy wave theory still doesn't lead to a cut-off when the light
> >is too red.
>
> from classical E+M in dealing with the Compton Effect, which cannot
> be attributed to bound-state energy levels (it concerns continuum
> states) and where the angular distribution of scattered photons is
> not what you get from standard classical E+M theory.
Perhaps I don't need my own theory here, as the main thing I'm concerned
with is the need to treat the radiation field classically. (Though I shall
continue to believe personally that this is not the right description, as
the evidence for the emission of individual electrons is not that strong.
Compton, for example, never tested this part of his theory!)
See Martin Green's contribution, 31 October: he says
"Given the quantized nature of the atomic energy levels, there is NO NEED
to invoke "photons" to explain the photo-electric effect. A quantum atom
placed in a classical electromagnetic field will display ALL the standard
behaviors of the photo-electiric effect...the cut-off frequency, electron
emissions
in very weak light, etc."
Grangier, Roger and Aspect says much the same thing in their paper on
anticorrelations (Europhysics Letters 1, 173-179 (1986)).
Caroline
<http://www.aber.ac.uk/~cat>
c.h.thompson
Charles Francis <cha...@clef.demon.co.uk> wrote
c.h.thompson <c.h.th...@newscientist.net> writes
> >
> >Charles Francis <cha...@clef.demon.co.uk> wrote
> >c.h.thompson <c.h.th...@newscientist.net> writes
> >> >
(interference
> >> >patterns etc) there are real waves behaving as real waves always have
> >> >done.
> >>
> >> way you can deduce any such thing from "interference" patterns. So
> >> called interference comes from hypothetical statements made in quantum
> >> mechanics about what would happen if an experiment were to be done, not
> >> from physical waves.
> >
> >Have you never seen interference patterns on the surface of a pond?
> >
> >Are these due to "hypothetical statements made in QM??????" How
ridiculous
> >can you get!
>
> It is only ridiculous to think that interference patterns on a pond have
> anything to do with quantum effects. This is like observing that three
> apples plus four apples is seven apples and three oranges plus four
> oranges is seven oranges, and concluding that apples is oranges.
Charles, we're talking about two different things! You're talking about
interactions of particles or something and all I'm concerned with is optics.
In optics, I'm claiming, interference is ordinary classical interference!
> >It is QM that invented the crazy idea of interference of
> >probability waves ....
> >
> >But isn't it about time you looked at my web site? I thought you had.
>
> I have. There is a difference between finding experimental loopholes and
> asserting things which are not true.
It is those who claim that the experiments rule out all possible local
realist models who in the wrong!
> If you want to challenge the
> experimental foundations of quantum mechanics you will have to do better
> than tackle a somewhat reserche experiment (Aspect) of little more than
> pedagogical value,
My personal aim is nothing like as grand as this! I am quite content to let
in a breath of fresh air by informing the world that they are once again
free to rule out magical instantaneous nonlocal effects. They have been
misinformed. Once this garbage is out of the way, other minds, with better
access to laboratories and teams of researchers to assist them, can explore
the consequences. QM will be open to challenge in a way to which it has
become unaccustomed.
> >Though most of my work is not published, I have written to
> >many of the experimenters concerned and had considerable discussion with
> >them.
> >
> It would be more interesting if your discussions could have been with
> the people responsible for the experimental foundation of quantum
> mechanics, Heisenberg, Schrodinger, or, if you had actually come to
> grips with the experimental basis for quantum mechanics as described by
> Dirac and Von Neumann.
Agreed, but you miss the point! I do not pretend to know much physics. I
was trained as a statistician. I knew nothing of optics when I started
looking at the EPR experiments, but I found out sufficient for my purpose.
I simply wanted to assess the evidence for magic, and this I have done.
Those EPR experiments were designed as a test of the theory. Although it is
only the first few that were conducted in this spirit, their results have
been accepted as supporting magic. To include magic in science is the end
of science. It must be excluded.
> >Have you studied the experimental reports yourself?
> >
> As I say, the problem is that you are not studying the significant
> experiments, or the interpretation of experimental results from the
> point of view of general measurement theory, which you declare to be
> irrelevent. As far as I can tell, you do not have an idea of what
> quantum mechanics is
There is only one of me! I have found out something important and must
devote myself to publicising this until it is generally accepted. Till
then, the rest of QM will have to wait.
> >> Experiments far more conclusive than interference patterns, which only
> >> demonstrate the validity of a mathematical formula, not the existence
of
> >> a wave.
> >
> >Uh? Brainwashed? Sorry, I give up! I shall not respond to any more of
> >your messages unless you do actually read what I say and discuss it
> >intelligently.
>
> I have read what you have to say. But until you read what I have to say,
> and read extensively on the conceptual foundations of quantum mechanics
> (I can lend you some books as you live so near) I doubt there is any
> chance of an intelligent discussion.
I have read probably more than you realise. I just do not like QM. It has
been beset with conceptual difficulties from the start. I think we would be
better off with NO mathematical theory for a while. It is inevitable that
classical optics would come back into use in the area I've been studying,
but as to the rest of QM's domain, I just don't know. All I do know is that
in biology things are less dogmatic, less mathematical, much more
interesting and productive!
Still, you are right: I don't think we have much in common.
> >In the unlikely event that you succeed in "destroying my
> >credibility" you will not have done any service to the long-term future
of
> >physics.
>
> I do not wish to destroy your credibility. On the face of it, I think
> you have something important to say and I am trying to stop you
> destroying your own credibility,
Sorry, I realise this. Thanks.
Caroline
<http://www.aber.ac.uk/~cat>
Katie Schwarz
>
> You have to remember the historical context. The photoelectric effect
> was explained (along with Planck's radiation law derivation) with a
> photon derived by using a statistical mechanics analysis that treated
> the electromagnetic field as akin to a macroscopic ensemble. There
> was no quantum mechanics to justify the Planck assumption (which is
> dicey in the "old" QM of Bohr and Sommerfeld anyway, as it ignores
> any features specific to a particular atom) or to explain the work
> function in the photo-effect.
>
> The semi-classical explanations could be justified later, after
> quantum mechanics was fully formed.
True. The semiclassical explanation for the photoelectric effect was
published in 1927, if I recall correctly. It requires Schroedinger's
equation.
--
Katie Schwarz
"There's no need to look for a Chimera, or a cat with three legs."
-- Jorge Luis Borges, "Death and the Compass"
Katie Schwarz
> Katie Schwarz <k...@socrates.berkeley.edu> wrote:
>>Gregory L. Hansen <glha...@steel.ucs.indiana.edu> wrote:
>>>
>>> The reason the photoelectric effect was important because
>>>they tried to explain it with classical electromagnetism by assuming
>>>energy will build up over time until an electron is excited. But the
>>>response of the electrons was immediate,
>>
>>about it, this doesn't make sense to me: if energy comes in packages
>>instead of continuously, you still have to wait for a photon to come
>>along.
>
>Yes, but not nearly as long. As an example from Beiser, "Concepts of
>Modern Physics":
>
>| [...] if the
>| incident light is absorbed in the uppermost atomic layer, each atom
>| receives energy at an average rate of 10^-25 W. At this rate over a
>| month would be needed for an atom to accumulate energy of the magnitude
>| that photoelectrons from a sodium surface are observed to have.
Oh, right -- I forgot that the detector has more than one atom in it
(D'oh!) and it's the energy *per atom* per unit time that's too small.
Thanks.
Katie Schwarz
>>
>If the energy of light were not quantized, I don't see why you shouldn't
>be able to demonstrate the photoelectric effect for waves of arbitrarily
>certainly nothing in the pseudo-classical treatment of the photoelectric
>effect that suggests a cut-off frequency. I've just checked in Shankar,
>and Fermi's golden rule gives a frequency dependence but the rate never
>goes to zero or has anything suggesting even a smoothed-out step function.
Fermi's golden rule gives the sin^2 function which becomes a delta
function in frequency when the time gets long compared to the inverse
frequency. It sounds like you're missing a necessary assumption, that
the energy states of the *bound electrons* are quantized, with a gap
between the lowest free state and the highest bound state. If there
are no electrons with energies in this gap, then the delta function
condition, hnu = Efinal - Einitial, cannot be satisfied for light with
hnu too small. This provides the cutoff frequency.
The Fermi's golden rule derivation says there is a "beat frequency"
between the electron's initial and final wavefunctions, and the
perturbation must be in resonance with this beat frequency to cause
the transition. But it's not necessary, in this derivation, to
quantize the energy in the perturbing light.
>"That's not an avacado, that's a grenade!" -- The Skipper
Don't you mean avocado? :)
Charles Francis
Thanks for the posting.
In article <381e0...@news1.vip.uk.com>, c.h.thompson <c.h.thompson@new
scientist.net> writes
>
>Charles, we're talking about two different things! You're talking about
>interactions of particles or something and all I'm concerned with is optics.
>In optics, I'm claiming, interference is ordinary classical interference!
>
But the interference patterns in optics are quantum effects on particles
of light. See QED by Richard Feynman. As he says you can read this book
in a day and understand the principles better than many graduate
students after three years of training. But he does make a point of
explaining that he does not know why things work like this.
>
>It is those who claim that the experiments rule out all possible local
>realist models who in the wrong!
I am in 100% agreement on this. That is why I find QED so exciting. I
know that it is officially some bizarre second quantisation of a field,
but it is actually an explanation of everything we know in physics in
terms of a local realist model.
>
>> If you want to challenge the
>> experimental foundations of quantum mechanics you will have to do better
>> than tackle a somewhat reserche experiment (Aspect) of little more than
>> pedagogical value,
>
>My personal aim is nothing like as grand as this! I am quite content to let
>in a breath of fresh air by informing the world that they are once again
>free to rule out magical instantaneous nonlocal effects. They have been
>misinformed. Once this garbage is out of the way, other minds, with better
>access to laboratories and teams of researchers to assist them, can explore
>the consequences. QM will be open to challenge in a way to which it has
>become unaccustomed.
>
Yes, but I think the way to do this is to establish a theory of
measurement, and to analyse the mathematical rules which it ought to
obey. The magical instantaneous effects are due to the misinterpretation
of quantum mechanics, not quantum mechanics itself. Our universities do
not even teach the foundations of quantum mechanics, instead they teach
people not to think about it, and the establishment tries to suppress
work in the field as being "speculative".
>
>Agreed, but you miss the point! I do not pretend to know much physics. I
>was trained as a statistician. I knew nothing of optics when I started
>looking at the EPR experiments, but I found out sufficient for my purpose.
>I simply wanted to assess the evidence for magic, and this I have done.
>Those EPR experiments were designed as a test of the theory. Although it is
>only the first few that were conducted in this spirit, their results have
>been accepted as supporting magic. To include magic in science is the end
>of science. It must be excluded.
>
Yes. Science has actually signed its own death certificate by accepting
positivist doctrines to the effect that nothing can be known or
understood. They are left with magic.
>
>There is only one of me! I have found out something important and must
>devote myself to publicising this until it is generally accepted. Till
>then, the rest of QM will have to wait.
>
I know the problem. That is why I find it so frustrating when I cannot
communicate with people such as yourself who ought to be on the same
side, motivated by the same realist ideas, but with different areas of
expertise. I have also found out something important and must devote
myself to publicising it, which is that quantum mechanics does have a
local realist explanation. It is subtle, and you have to think about it,
but please don't dismiss without understanding what I am saying.
>
>I have read probably more than you realise. I just do not like QM. It has
>been beset with conceptual difficulties from the start.
I fully sympathise with you there. I walked out on the qm lectures two
years running in the tripos. I kind of think that almost every competent
mathematician does the same. I think that is why there has been no
progress in understanding quantum mechanics in seventy years. All the
major advances in science have been made by great mathematicians (I
cannot exclude Dirac, and Einstein was so dependent on Riemann that I
cannot exclude gtr from this rather sweeping generalisation).
>
>Still, you are right: I don't think we have much in common.
>
We have much more in common than our initial failure to communicate
indicates. I was really glad to read your posting because it is the
first indication we can break the communication barrier.
--
Charles Francis
cha...@clef.demon.co.uk
c.h.thompson
Charles Francis <cha...@clef.demon.co.uk>
c.h.thompson <c.h.th...@newscientist.net> writes
> >
> >Charles, we're talking about two different things! You're talking about
> >interactions of particles or something and all I'm concerned with is
optics.
> >In optics, I'm claiming, interference is ordinary classical interference!
> >
> of light. See QED by Richard Feynman.
Dear Charles, I've read it, and it's absurd!
Feynman is quoted as saying that nobody is as easy to deceive as oneself and
he was right! He seems to have deceived himself into thinking that
phenomena that were clearly wave phenomena, behaving entirely classically,
were in fact the result of photons behaving in quantum probabilistic manner.
What he'd done was to take some classical maths and translate it into QT.
This can be done. It, not surprisingly, gives the same answer. He kidded
himself that it was the QT translation that was "the real thing", but this
was just a matter of faith. He must have had some other reason (the
photoelectric effect, maybe) to be convinced that he was seeing particles.
He did not fool me!
> But he does make a point of
> explaining that he does not know why things work like this.
Not surprising ...
> >It is those who claim that the experiments rule out all possible local
> >realist models who in the wrong!
>
> I am in 100% agreement on this. That is why I find QED so exciting. I
> know that it is officially some bizarre second quantisation of a field,
> but it is actually an explanation of everything we know in physics in
> terms of a local realist model.
So we can deduce, can we not, that the QED prediction for the EPR setup
would not be the same as the QM one? Bell's test should not be infringed?
> >My personal aim is nothing like as grand as this! I am quite content to
let
> >in a breath of fresh air by informing the world that they are once again
> >free to rule out magical instantaneous nonlocal effects. They have been
> >misinformed. Once this garbage is out of the way, other minds, with
better
> >access to laboratories and teams of researchers to assist them, can
explore
> >the consequences. QM will be open to challenge in a way to which it has
> >become unaccustomed.
> >
> Yes, but I think the way to do this is to establish a theory of
> measurement, and to analyse the mathematical rules which it ought to
> obey.
Surely not everyone need get involved in the maths? In the last century,
when Faraday was giving public lectures on electricity and magnetism, I bet
he didn't talk maths! Yet the result was that people loved physics, and
understood wonderful new things that nobody had known only a few years
before. This is what can happen again. You do maths if you like, but don't
assume that this is what physics "is"!
> The magical instantaneous effects are due to the misinterpretation
> of quantum mechanics, not quantum mechanics itself.
If this is true (and I can quite believe that some of the founding fathers
would have been dead against nonlocality) then the mistake runs very deep
and started way back in the late 1920's, with Von Neumann's nonsense. And I
can't see any merit in Dirac! I've tried to read his basic text book.
Wasn't Dirac one of several inventors of QM who later doubted whether he'd
been on the right lines? If you want to get things right, you need to go
back to the original concepts, not the maths. The concepts somehow got
lost. I get the impression that this is a pretty inevitable consequence of
any but the simplest mathematical theory, and it is not a simple theory that
is needed here. Even if we - one or two geniuses such as your good self
(!!) - master the maths too few people will understand it. So there will be
no "quality control". No referee will understand enough to say "no" to any
mistakes that may creep in. They will become incorporated in the accepted
body of belief and off we go again on a hopeless, miserable, journey, lost w
ithout even knowing where we were trying to go!
> >To include magic in science is the end
> >of science. It must be excluded.
> >
> Yes. Science has actually signed its own death certificate by accepting
> positivist doctrines to the effect that nothing can be known or
> understood. They are left with magic.
>
> >There is only one of me! I have found out something important and must
> >devote myself to publicising this until it is generally accepted. Till
> >then, the rest of QM will have to wait.
> >
> I know the problem. That is why I find it so frustrating when I cannot
> communicate with people such as yourself who ought to be on the same
> side, motivated by the same realist ideas, but with different areas of
> expertise.
But why should I honour Dirac? So what, he "discovered" spin (or was that
Pauli?) and the positron. But he didn't really (this was just
rationalisation, a homo sapiens is good at). Personally, I have my own
views on spin and think the whole concept is wrong. As for the positron, it
seems to be just an electron in the wrong phase state ...
I want a new physics to be like Faraday's. It must tell me how lasers work
(and do you really think QED is the right approach for them? I don't!) and
how the fire beetle homes in on a fire many miles away. How a bird
navigates, how a spacecraft does, why the speed of gravity seems to be
infinite (I've got my own ideas on that, too, and they owe nothing to either
Einstein relativity or quantum theory!) ...
All QM does is predict spectra!
> I have also found out something important and must devote
> myself to publicising it, which is that quantum mechanics does have a
> local realist explanation.
The part concerned with entanglement in the EPR context does not!
> but please don't dismiss without understanding what I am saying.
I'm all ears. But I want concepts, please. Not maths other than high
school stuff.
> >Still, you are right: I don't think we have much in common.
> >
> We have much more in common than our initial failure to communicate
> indicates. I was really glad to read your posting because it is the
> first indication we can break the communication barrier.
We'll see. My first reaction to most of your postings is not favourable,
but certainly this one is OK! Apart from the Feynman bit.
Caroline
<http://www.aber.ac.uk/~cat>
c.h.thompson
z@z <z...@z.lol.li> wrote in message news:7vid4a$c1e$1...@pollux.ip-plus.net...
> c.h.thompson wrote:
> | z@z (Wolfgang) wrote:
>
> | > theoretical and empirical evidence), then we must restart
> | > physics from its state at Maxwell's time.
> |
explanation
> | requested. 25 October) you said "There is a huge difference between
"common
> | sense" actions at a distance and "spooky" EPR actions."
> |
> I reject EPR actions at a distance. But I think that nobody
> should accept EPR actions at a distance and reject the actions
> at a distance of pre-Maxwellian physics, because the latter are
> more general than the former.
>
> We can think of "material" models leading to the well-known
> instantaneous inverse distance square laws.
But Wolfgang, Newton's theory not really supposed to be instantaneous. It
was a sufficiently good approximation to treat the forces as if they were
transmitted instantaneously, but this was only because the applications of
the theories were restricted, I imagine, to the Sun and planets. The Sun is
much more massive so presumably does not move much.
Perhaps what you mean by pre-Maxwellian theories is later ones. I suppose
Tom van Flandern's paper
(HTTP://www.best.com/~dolphin/vanFlandern/gravityspeed.html ) is a revival
of some of these.
But I'm highly sceptical. I don't know what the errors might be. I suspect
that the answer is not very sensitive to the exact angle of the vector.
Alternatively, we may have totally misunderstood the forces involved. If
part of the reason for the Earth's motion is that it is embedded in a large
rotating aether region - the whole region rotating round the Sun, with
different velocities at different radial distances - then the effective
force need not act in a straight line.
I dispute the idea of instantaneous action in this context as well as all
others. Van Flandern calculates that the speed must be something phenomenal
(2*10^10c or so), but I would say it is much more likely that we've got the
basic model wrong. One in which we have some electromagnetic-type gravity
combined with a little of Steve Rado's aether whirlpool might be nearer the
truth.
Some interesting anomalies have been reported regarding apparent gravity
encountered by spacecraft. I think we should provisionally say that there
is no instantaneous action, but keep an eye on the anomalies. They should
eventually tell us what really does happen.
> with such a fluid, the simple assumption that matter draws
> in (and annihilates) fluid proportionally to the mass
Have you met Henry Lindner's ideas on this? See
http://www.aber.ac.uk/~cat/People/intro.htm . I don't find the idea
"simple" myself. It just does not agree with my intuition.
> If one calculates the actual vectors by which the earth is
> accelerated by other planets, we find out that these vectors
> point to the locations where these planets are now, and not
> where they have been when photons arriving now were emitted.
How sure are you that you've made the right assumptions before doing your
calculation?
> So if we explain gravitation by gravitons (or by a field
> propagating at c), we must assume that gravitons originating
> from e.g. Jupiter must remain in telepathic connection with
> Jupiter. Only such a telepathic link (or complicated calculations
> in advance) can garantee that the gravitons accelerate the
> earth in the direction where Jupiter is now.
No, there is something wrong. Could be various things but I'm not going to
admit telepathy into physics on such meagre and challengeable evidence!
>
> See once again http://www.deja.com/=dnc/getdoc.xp?AN=538880440
I don't agree with some of the faq answers there. In particular Einstein
had certain ideas about Faraday induction. This is not really my area, but
you could do worse than browse in the research documents of
http://www.omicron-research.com/. I think you'll find an interesting
article by Al Kelly ...
Cheers
Caroline
<http://www.aber.ac.uk/~cat>
Christof Pflumm
"c.h.thompson" <c.h.th...@newscientist.net> writes:
> > But the interference patterns in optics are quantum effects on particles
> > of light. See QED by Richard Feynman.
>
> Dear Charles, I've read it, and it's absurd!
>
> Feynman is quoted as saying that nobody is as easy to deceive as oneself and
> he was right! He seems to have deceived himself into thinking that
> phenomena that were clearly wave phenomena, behaving entirely classically,
> were in fact the result of photons behaving in quantum probabilistic manner.
> What he'd done was to take some classical maths and translate it into QT.
> This can be done. It, not surprisingly, gives the same answer. He kidded
> himself that it was the QT translation that was "the real thing", but this
> was just a matter of faith. He must have had some other reason (the
> photoelectric effect, maybe) to be convinced that he was seeing particles.
> He did not fool me!
And what about the double slit experiment with single photons? Can you
explain how the interference pattern is building up using classical
wave phenomena? Or what phenomena did you think of?
Bye,
Christof
c.h.thompson
Christof Pflumm <lti...@ltihp81.etec.uni-karlsruhe.de> wrote >
> "c.h.thompson" <c.h.th...@newscientist.net> writes:
>
> > > But the interference patterns in optics are quantum effects on
particles
> > > of light. See QED by Richard Feynman.
> >
> > Dear Charles, I've read it, and it's absurd!
> >
> > Feynman is quoted as saying that nobody is as easy to deceive as oneself
and
> > he was right! He seems to have deceived himself into thinking that
> > phenomena that were clearly wave phenomena, behaving entirely
classically,
> > were in fact the result of photons behaving in quantum probabilistic
manner.
> > What he'd done was to take some classical maths and translate it into
QT.
> > This can be done. It, not surprisingly, gives the same answer. He
kidded
> > himself that it was the QT translation that was "the real thing", but
this
> > was just a matter of faith. He must have had some other reason (the
> > photoelectric effect, maybe) to be convinced that he was seeing
particles.
> > He did not fool me!
>
There is no satisfactory evidence that the "single photon" exists! That is
what this thread is all about, or hadn't you noticed?
I have studied quite a few experiments that claim to have dealt with single
photons. None has been convincing. The illusion of particle behaviour can
always be explained by the operating characteristics of the detector. If
you want this put in the accepted terminology, see Martin Green's
contribution of 2nd November. I would prefer myself to see it in practical
terms, though. We simply have detection instruments that take in
continuous, analogue, information and output counts, because that is the way
they were manufactured.
Anyway, which double slit experiment did you have in mind? I should be glad
to study it. I have an idea that the great majority of such experiments
have existed in the mind only!
Cheers
Caroline
<http://www.aber.ac.uk/~cat>
Charles Francis
scientist.net> writes
>
>Feynman is quoted as saying that nobody is as easy to deceive as oneself and
>he was right! He seems to have deceived himself into thinking that
>phenomena that were clearly wave phenomena, behaving entirely classically,
>were in fact the result of photons behaving in quantum probabilistic manner.
>What he'd done was to take some classical maths and translate it into QT.
>This can be done. It, not surprisingly, gives the same answer. He kidded
>himself that it was the QT translation that was "the real thing", but this
>was just a matter of faith. He must have had some other reason (the
>photoelectric effect, maybe) to be convinced that he was seeing particles.
>He did not fool me!
>
one photon at a time. They have also been detected for electrons,
protons, neutrons and even atoms. There are by now numerous
demonstrations of "wave-particle duality" (which I agree is nonsense
btw) and in addition there are the phenomenally accurate predictions of
qm and qed in particular which cannot be explained any other way, such
as the gyromagnetic moment prediction of the electron.
>
>> >It is those who claim that the experiments rule out all possible local
>> >realist models who in the wrong!
>>
>> I am in 100% agreement on this. That is why I find QED so exciting. I
>> know that it is officially some bizarre second quantisation of a field,
>> but it is actually an explanation of everything we know in physics in
>> terms of a local realist model.
>
>So we can deduce, can we not, that the QED prediction for the EPR setup
>would not be the same as the QM one? Bell's test should not be infringed?
No, we can deduce that there is something fundamental wrong with the way
in which we habitually think of space, time, causality, and the results
of measurement. There is something wrong with our basic assumptions and
premises. Finding out precisely what is wrong is the central scientific
problem of our era (that is presumably why the institute of physics and
the APS do not consider papers in the subject). To put things in the
perspective of a Kuhnian interpretation of history of science, we are in
a period of scientific crisis, in which all we really know is that our
fundamental ideas are wrong, and then we may only know it if we are not
part of the establishment. Lavoisier is said to have converted almost
no-one to the oxygen theory of burning, giving rise to the notion that
old scientists do not change, but merely die off.
>
>> >
>> Yes, but I think the way to do this is to establish a theory of
>> measurement, and to analyse the mathematical rules which it ought to
>> obey.
>
>Surely not everyone need get involved in the maths? In the last century,
>when Faraday was giving public lectures on electricity and magnetism, I bet
>he didn't talk maths! Yet the result was that people loved physics, and
>understood wonderful new things that nobody had known only a few years
>before. This is what can happen again. You do maths if you like, but don't
>assume that this is what physics "is"!
Yes, you have hit on another problem with the way in which we do
science. I have tried to get away without actually doing the maths, on
the basis that the maths is really only the stuff they teach
undergraduates at Cambridge. It is not really so important to do it as
it is to know that it can be done. Once you have done a bit of studying
mathematical structures it is much easier to see what can be done than
it is to actually execute the construction. General relativity is a case
in point. The fundamental idea is really simple, the execution requires
almost superhuman mathematical ability. Very few relativists who have
grasped the technicalities are able to relate them back to the simple
idea that motivated it in the first place (especially as it is not
presented like that in the text books). Quantum mechanics is a bit like
that too. The real difference is that the fundamental idea of gtr
developed into mathematics before gtr, whereas the fundamental idea of
quantum mechanics was never properly developed.
>
>> The magical instantaneous effects are due to the misinterpretation
>> of quantum mechanics, not quantum mechanics itself.
>
>If this is true (and I can quite believe that some of the founding fathers
>would have been dead against nonlocality) then the mistake runs very deep
>and started way back in the late 1920's, with Von Neumann's nonsense. And I
>can't see any merit in Dirac! I've tried to read his basic text book.
>
The mistake is very deep, but I have to disagree with you. Von Neumann
and Dirac were the only two to make any headway with the problem at all,
and they didn't eliminate the mistake from their thinking either.
>Wasn't Dirac one of several inventors of QM who later doubted whether he'd
>been on the right lines?
Yes he did. He was of the generation that had to turn all their thinking
inside out, and he still recognised that it had not worked. That is far
more astute than those who claimed afterward that it was all okay. He
thought there would have to be another complete revolution in thinking
to resolve the problem, and that is perhaps true, but it is a revolution
in the way in which we think about the theory, not a change in the
mathematics of the theory.
>If you want to get things right, you need to go
>back to the original concepts, not the maths.
Yes
>The concepts somehow got
>lost.
Yes. Quantum mechanics is nothing more than a model of relationships
found in measurement, not a fundamental description of matter.
>I get the impression that this is a pretty inevitable consequence of
>any but the simplest mathematical theory, and it is not a simple theory that
>is needed here.
I'm afraid not.
>Even if we - one or two geniuses such as your good self
>(!!) - master the maths too few people will understand it. So there will be
>no "quality control". No referee will understand enough to say "no" to any
>mistakes that may creep in.
So far I have hardly got a referee to look at what is right, let alone
find mistakes, and I make far too many of those. What we really need
though, is not more mastery of the maths, but more mastery of the
concepts. Orthodoxy says we must not study concepts, so it is difficult
to know even where to try to get published. I am investigating
philosophy journals, and have found one that looks promising. But that
means another rewrite, but it should let me concentrate more on
concepts.
>
>But why should I honour Dirac? So what, he "discovered" spin (or was that
>Pauli?) and the positron. But he didn't really (this was just
>rationalisation, a homo sapiens is good at). Personally, I have my own
>views on spin and think the whole concept is wrong. As for the positron, it
>seems to be just an electron in the wrong phase state ...
Basically, it is. Dirac found the equation that said this phase state
must exist. The maths is very tough, but it is right, and unlike almost
everything else in relativistic quantum mechanics it is based on sound
precepts. Mind you, the hole interpretation is dodgy. The Feynman
interpretation is better.
Dirac was also responsible for the most general and the most
conceptually sound version of quantum mechanics. Although I would not
want to knock the achievements of the other founding fathers, I don't
think it is far from the truth to say that quantum mechanics is Dirac's
theory, and every one else only contributed bits and pieces, many of
which were misleading or ill-conceived. (After all F=ma is attributed to
Newton, but it was actually Euler's rewrite, and Galileo was a major
contributor to Newtonian dynamics, we can't expect any attribution of
scientific merit to be 100% accurate).
>All QM does is predict spectra!
When reformulated as a discrete theory of particle interactions QED
predicts all electromagnetic phenomena, Newton's laws, quantum
mechanics, and even (using heuristic arguments) general relativity. Mind
you, I agree with you that you have to get rid of this idea of particles
passing through every point in space simultaneously. That is nonsense.
It is a way of describing one of the mathematical formulae, not
something which actually happens.
>
>> I have also found out something important and must devote
>> myself to publicising it, which is that quantum mechanics does have a
>> local realist explanation.
>
>The part concerned with entanglement in the EPR context does not!
>
It does. Entanglement is part of a bad description, not a real challenge
to locality or realism. The whole issue is wrapped up epistemology; what
does measurement really tell us, as distinct from what we think it tells
us, and what can we actually say about matter, as distinct from what we
would like to be able to say. The moment a false or a hidden assumption
creeps in, it is liable to clash with what actually happens, producing a
paradox, suggesting that we should throw the whole theory away, while
actually what we need to do is re-evaluate our most fundamental
assumptions.
The most obvious qed explanation of EPR is rooted in its complete
symmetry between fore and after states (the same symmetry is actually in
classical mechanics too, and is only broken by the statistical laws of
thermodynamics). This symmetry suggests that, in fundamental physical
law, the future has as strong an influence on the past as the past has
on the future. Changing apparatus A may therefore affect the original
pair production, which therefore affects the measurement at apparatus B.
That explanation violates neither locality nor realism, though it does
challenge conventional ideas about causality. Personally I would not
want to assume too much about causality anyway.
That is the explanation I am happiest with, But it is not the only
possible qed explanation. Another one is that the causal relationships
don't really work like that, but the description in terms of measured
states makes it appear as though they do. On its own quantum mechanics
lacks a description of an underlying reality. It is only a theory of
relationships found in measurements. In other words quantum mechanics is
not complete. Particles may be local, but measurements are not. In the
particular case of EPR, it looks like there has been faster than light
communication, just as it looks like the collapse of the wave function
is instantaneous, but in practice the relationships between measurements
cannot be analysed until much afterwards, when a description of both
measurements is possible.
I suspect that this second explanation is not adequate on its own. But
to take things any further you have to think about spin, the fact that
no such thing as spin can exist within conventional space-time. Spin
cannot be a property just of a particle, but has something to do also
with the environment. That is already non-local. But the existence of
things which are non-local and depend on the way in which we look at
them does not mean that fundamental reality is non-local.
>> but please don't dismiss without understanding what I am saying.
>
>I'm all ears. But I want concepts, please. Not maths other than high
>school stuff.
>
Which was the purpose of the manuscript I originally showed you. My
ideas have improved since then, and I haven't yet written them up in
this form. But I am inclined to think that if the concepts cannot be
explained at this level, then no one can understand them. This maths is
difficult even for mathematicians, and I don't believe in mental
abilities of even a mathematician to simultaneously do the maths and
understand the concepts.
--
Charles Francis
cha...@clef.demon.co.uk
Christof Pflumm
"c.h.thompson" <c.h.th...@newscientist.net> writes:
> > And what about the double slit experiment with single photons?
>
> There is no satisfactory evidence that the "single photon" exists! That is
> what this thread is all about, or hadn't you noticed?
Hm, I must admit, I have not been very clear. I am always mixing
different meanings for photon. To explain, think about this: You have
a beam of light and some means of diminishing it. Using a counter, you
will (at a high intensity) get so many counts that you can't
distinguish them. If you lower the intensity, you will eventually get
distinguishable events. These discrete events, to some approximation,
can be considered as being caused by single photons, I think. Now the
problem is that the quantized modes of the electromagnetic field are
also called photons. And those photons are quite different from what I
called a single photon first. But let us take the first definition.
> I have studied quite a few experiments that claim to have dealt with single
> photons. None has been convincing. The illusion of particle behaviour can
> always be explained by the operating characteristics of the
> detector.
In terms of QM, there is no particle behaviour. There is nothing that
could be compared to a classical particle. This can only be done in an
approximate sense.
> If you want this put in the accepted terminology, see Martin Green's
> contribution of 2nd November.
Can't find that. I only found one from 1 Nov. He says the
photoelectric effect can be explained by considering the incoming
light as a wave and the atoms as QM things. I don't know whether
that's true.
> I would prefer myself to see it in practical terms, though. We
> simply have detection instruments that take in continuous, analogue,
> information and output counts, because that is the way they were
> manufactured.
So you would say when we detect a "single photon" (one event in a
photomultiplier for example), that this might be caused by a wave and
not a single particle? That is certainly true.
> Anyway, which double slit experiment did you have in mind? I should be glad
> to study it. I have an idea that the great majority of such experiments
> have existed in the mind only!
I thought of the one where they sent (single) electrons through a
double slit and got an interference pattern. I must admit that I don't
know whether it has been done with photons. But have you looked at
some recent experiments in Quantum Optics? Some of them have been
explained in Scientific American, but I have to look for more exact
references.
When I got it right, you are concerned with the particle aspect of
light. You don't like the idea that light behaves as a particle. Now,
there's no problem as QM does tell you that there is nothing like a
classical particle. So I must admit I don't understand what exactly
your problem is. Perhaps you could explain it again, if you don't
mind?
Bye,
Christof
Charles Francis
Pflumm <lti...@ltihp81.etec.uni-karlsruhe.de> writes
> These discrete events, to some approximation,
>can be considered as being caused by single photons, I think. Now the
>problem is that the quantized modes of the electromagnetic field are
>also called photons. And those photons are quite different from what I
>called a single photon first.
No, they are precisely the same. It is just that there are different
ways of looking at qed. It is, in my view, better to think of the
electromagnetic field as being the expected action of many photons.
>
>In terms of QM, there is no particle behaviour. There is nothing that
>could be compared to a classical particle.
Again I think the adjustment is the wrong one. A classical particle is
normally considered in the context of classical space-time. Rather than
ditch the classical particle, we should ditch classical space-time.
>
>When I got it right, you are concerned with the particle aspect of
>light. You don't like the idea that light behaves as a particle. Now,
>there's no problem as QM does tell you that there is nothing like a
>classical particle.
Surely there is a serious problem, quantum mechanics tells us things
which, on the face of it, cannot be true of anything at all.
--
Charles Francis
cha...@clef.demon.co.uk
Martin Green
> photoelectric effect can be explained by considering the incoming
> light as a wave and the atoms as QM things. I don't know whether
> that's true.
>
There are two arguments based on the photo-electric effect that supposedly
demonstrate
the particle nature of light...the frequency effect, and the intensity
effect. Both these effects
are, in fact, consistent with a quantume atom interacting with a classical
field.
Any two electron states, e.g. one free and one bound, can only be coupled
by a disturbance
whose frequency corresponds to the energy difference between those two
states. This coupling
frequency can easily be provided by a classical wave.
The intensity argument goes more or less as follows: The power in a
classical wave is so spread
out that it could not possibly cause an electron to be ejected. Therefore
the energy must be
concentrated in little bundles.
This argument is based on a lack of understanding of classical
electromagnetic theory, in particular
the theory of how a field interacts with a receiver. If it were necessary
for the energy density of
the wave to be entirely concentrated on the cross-section of the electron,
then it is perhaps true
that the photo-electric effect would be impossible. But in that case, radio
and television would also
be impossible, because the energy density falling on a typical receiving
antenna is far too weak to
cause the "observed effects."
The real puzzle is why, after almost one hundred years, the photo-electric
effect continues to
be put forward to students as the main "proof" of the existence of photons.
Martin Green
Jim Carr
Christof Pflumm <lti...@ltihp81.etec.uni-karlsruhe.de> wrote >
}
} "c.h.thompson" <c.h.th...@newscientist.net> writes:
} >
} > and he was right! He seems to have deceived himself into thinking that
} > phenomena that were clearly wave phenomena, behaving entirely classically,
} > were in fact the result of photons behaving in quantum probabilistic
} > manner.
} > What he'd done was to take some classical maths and translate it into QT.
} > This can be done. It, not surprisingly, gives the same answer. He kidded
} > himself that it was the QT translation that was "the real thing", but
} > this was just a matter of faith. He must have had some other reason (the
} > photoelectric effect, maybe) to be convinced that he was seeing
} > particles. He did not fool me!
}
In article <38203...@news1.vip.uk.com>
"c.h.thompson" <c.h.th...@newscientist.net> writes:
>
>There is no satisfactory evidence that the "single photon" exists!
There appears to be none that will satisfy you, so I guess I
will ask if you think that a single electron or neutron exists.
>That is what this thread is all about, or hadn't you noticed?
The thread is actually about quantum mechanics, as the text above
attributed to you makes clear. Thus it is relevant to ask if you
think neutron or electron scattering from a crystal is somehow
different from photon scattering from a crystal despite giving
identical results for identical momenta. Your dismissal of quantum
effects attributed to photons cannot dismiss quantum mechanics
because there is more to quantum mechanics than photons.
>I have studied quite a few experiments that claim to have dealt with single
>photons. None has been convincing. The illusion of particle behaviour can
>always be explained by the operating characteristics of the detector.
Do you think that a proton, neutron, electron, or BEC cloud is
an "illusion"? Why or why not? Will you accept experimental
evidence of "quantum weirdness" provided in experiments done
with those 'particles'? Why or why not?
Jon Bell
>
>The real puzzle is why, after almost one hundred years, the photo-electric
>effect continues to be put forward to students as the main "proof" of the
>existence of photons.
Most introductory treatments of "modern" physics mimic the historical
development of the field. The photoelectric effect was originally considered
a strong piece of evidence for the existence of photons, via Einstein's
explanation of the effect.
However, while most textbooks I've seen do mention the photoelectric effect
first (after Planck's blackbody theory, of course), they do also go on to
discuss the Compton effect and other phenomena that support the idea of
photons.
--
Jon Bell <jtb...@presby.edu> Presbyterian College
Dept. of Physics and Computer Science Clinton, South Carolina USA
[ Information about newsgroups for beginners: ]
[ http://www.geocities.com/ResearchTriangle/Lab/6882/ ]
Joe Rongen
news:01bf2622$2ac106a0$0100a8c0@mgreen...
> The real puzzle is why, after almost one hundred years, the
> photo-electric effect continues to be put forward to students
> as the main "proof" of the existence of photons.
>
> Martin Green
Maybe, because it is "a basic process" ...the absorption
of a quantum of energy E=hv by an individual electron.
Regards Joe
c.h.thompson
Christof Pflumm <lti...@ltihp81.etec.uni-karlsruhe.de> wrote
"c.h.thompson" <c.h.th...@newscientist.net> writes:
> > > And what about the double slit experiment with single photons?
> >
> > There is no satisfactory evidence that the "single photon" exists! That
is
> > what this thread is all about, or hadn't you noticed?
>
> Hm, I must admit, I have not been very clear. I am always mixing
> different meanings for photon.
Yes, the term is not exactly used consistently, but I'm glad you're
interested. I'll try and answer your points but do get back to me at
c.h.th...@newscientist.net or go to <http://www.aber.ac.uk/~cat> if you
need more detail.
> To explain, think about this: You have
> a beam of light and some means of diminishing it. Using a counter, you
> will (at a high intensity) get so many counts that you can't
> distinguish them. If you lower the intensity, you will eventually get
> distinguishable events. These discrete events, to some approximation,
> can be considered as being caused by single photons, I think. Now the
> problem is that the quantized modes of the electromagnetic field are
> also called photons. And those photons are quite different from what I
> called a single photon first. But let us take the first definition.
>
> > I have studied quite a few experiments that claim to have dealt with
single
> > photons. None has been convincing. The illusion of particle behaviour
can
> > always be explained by the operating characteristics of the
> > detector.
>
> In terms of QM, there is no particle behaviour. There is nothing that
> could be compared to a classical particle. This can only be done in an
> approximate sense.
In the context that fascinates me (interpretation of actual EPR and related
quantum entanglement experiments), I have discovered that the interpretation
depends critically on whether or not you think light behaves as a particle
when it goes through a polariser, or when split at a beamsplitter.
In my view, it seems evident that it is the unquantised radiation field we
are dealing with here, but the assumptions used in the usual Bell tests
include, indirectly, the assumption that (whatever it is to start with) if a
photon impinges on a polariser, say, it behaves as a particle and exits by
EITHER the + OR the - channel.
If this assumption is not even an official part of QM (and I can well
believe it is not) then the vast majority of Bell tests must be rejected out
of hand, as they depend on it. They are not valid in any case to anyone who
thinks light is pure wave. As the experiments are meant to be establishing
a fundamental truth about the nature of the universe, I should have thought
it is rather important that only tests agreed by all to be valid should be
used!
(Incidentally, the few EPR experiments that use an improved Bell test suffer
from other weaknesses, some of which I explore in papers such as
http://xxx.lanl.gov/abs/quant-ph/9903066 )
> > If you want this put in the accepted terminology, see Martin Green's
> > contribution of 2nd November.
>
> Can't find that. I only found one from 1 Nov. He says the
> photoelectric effect can be explained by considering the incoming
> light as a wave and the atoms as QM things. I don't know whether
> that's true.
This bit of theory suited my purpose, though I doubt if anyone knows the
truth. What matters for the logic of Bell tests is whether or not light is
quantised at beamsplitters etc.. When it is detected, the instruments
digitise it but that does not matter. As you say, it may sometimes (and in
all these experiments is assumed to) come in pulses, but so long as the
pulses are merely waves and can be split into parts whose amplitude
(ordinary wave amplitude, not probability) is reduced, the statistics work
out OK. Local realism can cope.
> So you would say when we detect a "single photon" (one event in a
> photomultiplier for example), that this might be caused by a wave and
> not a single particle? That is certainly true.
Yes.
> > Anyway, which double slit experiment did you have in mind? I should be
glad
> > to study it. I have an idea that the great majority of such experiments
> > have existed in the mind only!
>
> I thought of the one where they sent (single) electrons through a
> double slit and got an interference pattern.
I have read accounts of this in popular books, and in Silverman's "And yet
it moves" (CUP, 1993), which cannot really be counted as popular, or not in
this context. Silverman was on the team that did the experiment. Some
rather complicated apparatus was used, in which quite a lot of theory must
be assumed in order to interpret the results. The whole experiment seems to
have been rather a PR exercise! Data was recorded then used to build up the
motion picture that impressed the world.
Oh, but that isn't what I meant to say! I meant to say "Please stick to the
subject. We're talking about photons!"
> I must admit that I don't
> know whether it has been done with photons. But have you looked at
> some recent experiments in Quantum Optics?
I most certainly have! I have my own ideas for realist interpretations of
many of them. I've recently been concentrating on Gregor Weihs' experiment
(quant-ph/9810080 and PRL 81, 5039 (1998)) and some related experiments at
Innsbruck and in the States. Paul Kwiat's are especially relevant.
> Some of them have been explained in Scientific American, but
> I have to look for more exact references.
They have also been discussed in Physics Today and Physics World. See
Physics World, March 1998, p33 for a supplement on the subject (if I've read
my notes correctly!)
> When I got it right, you are concerned with the particle aspect of
> light. You don't like the idea that light behaves as a particle. Now,
> there's no problem as QM does tell you that there is nothing like a
> classical particle. So I must admit I don't understand what exactly
> your problem is. Perhaps you could explain it again, if you don't
> mind?
My problem is that those Bell tests assume that the pulses of light in EPR
experiments and similar must either go one way or the other at a two-channel
polariser. I doubt if many people realise that this assumption has been
made! I can't blame them. I don't think it reasonable to expect a
physicist to battle with the probabilities and logic of EPR experiments.
This is a job for logicians and statisticians. But, as it is not reasonable
to ask people to believe anything without understanding it themselves, I
have now published several papers that try and put the argument as simply as
possible.
I'm putting my total energy into publicising what I know, as at present
public opinion, steered by both the scientific journals and the popluar
press, says that experiments demonstrate quantum nonlocality and we just
have to accept it! Ridiculous!
Do look at my web site ...
Caroline
<http://www.aber.ac.uk/~cat>
c.h.thompson
>
> will ask if you think that a single electron or neutron exists.
No, Jim, this is not relevant, and I think you know why: because it is only
light that has been used in EPR experiments.
> The thread is actually about quantum mechanics, as the text above
> attributed to you makes clear.
As a matter of fact, it didn't! I was talking about Feynmans little book
"QED" and as far as I can remember this is all about light. Well, the bits
that interested me were anyway.
> Your dismissal of quantum
> effects attributed to photons cannot dismiss quantum mechanics
> because there is more to quantum mechanics than photons.
Of course I can't "dismiss QM"!
But I can say that all papers that make statements such as "the results
cannot be explained by any local realist model", or conclude that the
experiment forces us to think again about fundamental concepts concerning
locality, are not stating scientific fact. These statements should always
be qualified. It is always a matter of "if all the assumptions are true".
The papers don't make the assumptions clear, and this is partly because the
experimenters themselves do not fully understand the Bell test. The test is
designed for realists, and makes little sense unless you believe that you
really can multiply independent probabilities to get joint ones in ALL
contexts. It is UNIVERSALLY true. If you can't then the things you are
multiplying aren't probabilities - or are not fair estimates of them. This
is what happens in all but a few Bell tests: the estimates are not fair
ones - UNLESS light is behaving as particles at polarisers and
beamsplitters. (See also my reply to Christof Pflumm's second message.)
> >I have studied quite a few experiments that claim to have dealt with
single
> >photons. None has been convincing. The illusion of particle behaviour
can
> >always be explained by the operating characteristics of the detector.
>
> Do you think that a proton, neutron, electron, or BEC cloud is
> an "illusion"?
I don't know.
> Why or why not?
I don't have enough information. In the case of light, I have read up about
photomultipliers. I am pretty confident I know how they behave. I have not
studied other detectors.
> Will you accept experimental
> evidence of "quantum weirdness" provided in experiments done
> with those 'particles'?
Not unless I am given a course on the instrumentation used! I think I'd
have to see it with my own eyes, too.
But what's the point? Why should anyone want to demonstrate quantum
weirdness? It is self-evident to me that it is just a mistake! Nobody
would lose anything by dropping the idea!
Even those people currently looking into applications may be relieved to
know that many of the interesting properties of the "entangled photons" of
recent experiments do not depend on QM at all. I've go my own theory,
almost written up in a Latex document. It is hinted at in a section of my
chapter in the Instantaneous Action at a Distance book by Chubykalo and
Pope. (In CONTEMPORARY FUNDAMENTAL PHYSICS - Valerie V. Dvoeglazov - Editor)
1999, 475 pages. ISBN 1-56072-698-9. $145.)
Caroline
c.h.th...@newscientist.net
<http://www.aber.ac.uk/~cat>
Luc Bourhis
> I have studied quite a few experiments that claim to have dealt with single
> photons. None has been convincing.
Did you have a look at high energy photons seen at LEP or any other
particles accelerators ?
--
Luc Bourhis
Center for Particle Physics / University of Durham
United Kingdom
Luc Bourhis
> > But the interference patterns in optics are quantum effects on particles
> > of light. See QED by Richard Feynman.
>
> Dear Charles, I've read it, and it's absurd!
>
> He seems to have deceived himself into thinking that
> phenomena that were clearly wave phenomena, behaving entirely classically,
> were in fact the result of photons behaving in quantum probabilistic manner.
> What he'd done was to take some classical maths and translate it into QT.
> This can be done. It, not surprisingly, gives the same answer.
There is much more than that in QED. How do you predict classicaly the
anomalous magnetic moment of the electron for example ? QED and
experimental results agreed with 11 significant digits. And I could have
come with dozen of cross-sections for high energy interactions between
electron and positron. On the contrary I do not know any relativistic
theory with hidden variables able to reproduce the success of QED, not
speaking of the whole Standard Model.
> Personally, I have my own
> views on spin and think the whole concept is wrong.
Could you elaborate ?
> As for the positron, it
> seems to be just an electron in the wrong phase state ...
Idem.
Luc Bourhis
> The grave omission in Aspect's and all the EPR experiments is that only one
> result is published for each. In some cases absolutely no information is
> published about the actual data, only the derived Bell test statistic. It
> is rare indeed to see more than a graph of summarised data, in all
> likelihood "normalised" in some way. It is not possible for any but the
> most dedicated to find out what really happened.
Concerning the problem about the treatment of "accidentals" in Aspect's
experiment, I wonder why you gave a plot comparing the raw data with
your non-quantum model in one of your paper [1] ? After all even with
this model there is still some accidentals, and a sketch in an other
paper of yours show that you know that [2]. It seems to me that your
comparison is therefore meaningless especially since you do not justify
the fact that you do not take into account experimental imperfections.
Also as a theorist I know very well how difficult it is to analyse
experiments if one has not the necessary "know-how" which can only be
obtained by making real experiments. Since you claim that you pointed
out many experimental flaws in EPR experiment, I would like to know what
kind of expertise you have in this experimental field of physics ? Did
you build such kind of experiments yourself for example ?
[1] quant-ph/9903066
By the way your accounting of Aspect's answer to Marshall et al is not
as fair as it could be in this paper : Aspect et al turned almost in
pieces the arguments of their contradictors giving the clear impression
that Marshall et al did not know very well the physics involved by this
experiment.
Furthermore you reproduce extensively in Appendix B the work of Clauser
and Horne and you provide useless details in Appendix C. This is quite
uncommon and I would suggest that it would be more interesting for you
to develop your own analysis in this paper.
[2] quant-ph/9711044
Christof Pflumm
"Martin Green" <test...@pangea.ca> writes:
> There are two arguments based on the photo-electric effect that supposedly
> demonstrate
> the particle nature of light...the frequency effect, and the intensity
> effect. Both these effects
> are, in fact, consistent with a quantume atom interacting with a classical
> field.
>
> Any two electron states, e.g. one free and one bound, can only be coupled
> by a disturbance
> whose frequency corresponds to the energy difference between those two
> states. This coupling
> frequency can easily be provided by a classical wave.
Yes. That is done in almost every textbook when it comes to
absorption/emission of light.
> The intensity argument goes more or less as follows: The power in a
> classical wave is so spread out that it could not possibly cause an
> electron to be ejected. Therefore the energy must be concentrated in
> little bundles.
I have not done this kind of calculation, so I must ask some
questions: An EM wave hits an atom. The EM wave causes the atom to go
from one state to another, in this case from ground state to an
ionized state. Now you say that it is different for this calculation
whether the wave is interacting a long time and has a small amplitude
or if it has a big amplitude and interacts only shortly (I assume the
same energy in both cases)? Sounds interesting as it could perhaps
explain the photoelectric effect. You would need a short interaction
time to release an electron, which can only be achieved by a pulse. Is
that what you think?
> This argument is based on a lack of understanding of classical
> electromagnetic theory, in particular the theory of how a field
> interacts with a receiver.
Could be because the calculations are not easy (I think the
interaction is not classical em-theory as you need QM to describe
it. But it is "half-classical" because you use the classical fields in
the Hamiltonian).
> If it were necessary for the energy
> density of the wave to be entirely concentrated on the cross-section
> of the electron, then it is perhaps true that the photo-electric
> effect would be impossible.
When I remember it correctly, normally the atom is considered to be in
a time varying electric field with no spatial variation because of the
light's rather long wavelength (dipole approximation).
> The real puzzle is why, after almost one hundred years, the photo-electric
> effect continues to
> be put forward to students as the main "proof" of the existence of photons.
Perhaps the explantation in terms of photons is easier? But I must
admit, if the photoelectric effect can be explained classically, it is
not the best example to introduce the idea of a photon.
Bye,
Christof
Martin Green
> > The intensity argument goes more or less as follows: The power in a
> > classical wave is so spread out that it could not possibly cause an
> > electron to be ejected. Therefore the energy must be concentrated in
> > little bundles.
>
Christof Pflumm replied:
> I have not done this kind of calculation, so I must ask some
> questions: An EM wave hits an atom.
For the sake of simplicity, can we discuss the case of a hydrogen atom?
> The EM wave causes the atom to go
> from one state to another, in this case from ground state to an
> ionized state.
If you would like to go through the analysis, I would prefer to take the
simplest
case, i.e. the s-to-p transition from the ground state to the first excited
state.
> Now you say that it is different for this calculation
> whether the wave is interacting a long time and has a small amplitude
> or if it has a big amplitude and interacts only shortly (I assume the
> same energy in both cases)? Sounds interesting as it could perhaps
> explain the photoelectric effect. You would need a short interaction
> time to release an electron, which can only be achieved by a pulse. Is
> that what you think?
>
No. I think that this transistion can be driven by a very low intensity
wave of
long duration. There is a calculation people put forward based on
(power density of wave) x (cross-sectional area of atom) which purports
to show that this transition would be impossible with classical waves.
This calculation is wrong.
> Could be because the calculations are not easy (I think the
> interaction is not classical em-theory as you need QM to describe
> it. But it is "half-classical" because you use the classical fields in
> the Hamiltonian).
>
Right.
> When I remember it correctly, normally the atom is considered to be in
> a time varying electric field with no spatial variation because of the
> light's rather long wavelength (dipole approximation).
>
Also right.
> Perhaps the explantation in terms of photons is easier? But I must
> admit, if the photoelectric effect can be explained classically, it is
> not the best example to introduce the idea of a photon.
>
I do not object to using "photons" to simplify a few physics calculations,
any more
than I object to using "phasors" to simplify some electrical calculations.
What I would
strongly object to would be if the schools taught that "the success of
phasors in
explaining many electrical phenomena disproves once and for all the old
theory
that electrical power exists in the form of time-varying sinusoidal
voltages."
Martin Green
Christof Pflumm
"Martin Green" <test...@pangea.ca> writes:
> For the sake of simplicity, can we discuss the case of a hydrogen
> atom?
Yes.
> ...
> If you would like to go through the analysis, I would prefer to take the
> simplest
> case, i.e. the s-to-p transition from the ground state to the first excited
> state.
Agreed.
> > ...
> No. I think that this transistion can be driven by a very low intensity
> wave of
> long duration.
So you won't need short pulses (photons).
> There is a calculation people put forward based on (power density of
> wave) x (cross-sectional area of atom) which purports to show that
> this transition would be impossible with classical waves. This
> calculation is wrong.
I have never seen that one.
Do you have a web page with the calculation? Or would you prefer doing
it here in the ng?
Bye,
Christof
Martin Green
> > If you would like to go through the analysis, I would prefer to take
the
> > simplest
> > case, i.e. the s-to-p transition from the ground state to the first
excited
> > state.
>
> Agreed.
>
mixed state,
partly s and partly p, looks like a little oscillating electric dipole
antenna. You
may already know this...it's standard QM, but not always taught in courses.
The frequency of this antenna corresponds, of course, to 3/4 Rydberg, which
is the energy difference between the two states.
> > There is a calculation people put forward based on (power density of
> > wave) x (cross-sectional area of atom) which purports to show that
> > this transition would be impossible with classical waves. This
> > calculation is wrong.
>
The right way to do the calculation is to treat the light as an ordinary
classical wave,
and the atom as an ordinary, classical receiving antenna. Most people
assume that
the antenna can only absorb power that passes through the space in close
proximity
to the physical antenna. But it turns out that the antenna can actually
pull in power
over a much wider area. This is a suprising result that has nothing to do
with quantum
mechanics, but it helps to explain how weak, spread-out classical light can
pack
enough "punch" to knock an electron out of its orbit.
> Do you have a web page with the calculation? Or would you prefer doing
> it here in the ng?
>
If you're still in agreement so far, I will be glad to sketch the outline
of the antenna
calculation in the ng.
Regards,
Martin Green
c.h.thompson
> evidence of "quantum weirdness" provided in experiments done
Hi Jim
I've responded to your request! I couldn't find a suitable experiment
involving the kind of entanglement I've studied so far, but that fullerene
experiment was sufficiently weird. It's all very well for tiny particles to
be pushed around by E/M waves, but not C60 molecules.
So see my contributions to the thread "Fullerene diffraction", re
http://www.quantum.univie.ac.at/research/c60/ and Arndt et al,
"Wave-particle duality of C60 molecules", Nature 401, 680-682 (1999).
I've written to Arndt to ask his opinion of my classical hypothesis. These
fullerenes seem too big to be susceptible to the kind of interaction that
can produce NEGATIVE interference, but I've made a suggestion as to how they
could nevertheless show a KIND OF INTERFERENCE. An important consideration
is that the measuring instrument at the end of the day measures electric
current, not counts of molecules. If it responds the maximum amplitude
rather than to the integral over some short period, then that may be one way
to explain the observations, but I do need more facts than are available in
the paper.
What shall I tackle next? (No, I don't really want to know! I've got a
heavy schedule at present with revising my paper on entangled photons.)
Caroline
Christof Pflumm
"Martin Green" <test...@pangea.ca> writes:
> OK. The first important concept is to realize that a hydrogen atom in a
> mixed state,
> partly s and partly p, looks like a little oscillating electric dipole
> antenna. You
> may already know this...it's standard QM, but not always taught in courses.
I don't know that. Is this the same for other two-level systems? I
know I read something about general properties of two-level systems
and that there are transitions from one state to the other. If that's
it, I will look it up, but perhaps you don't mind to explain it,
especially in the context of the H-atom acting like an electric dipole.
> The frequency of this antenna corresponds, of course, to 3/4 Rydberg, which
> is the energy difference between the two states.
Sounds correct.
> The right way to do the calculation is to treat the light as an ordinary
> classical wave,
> and the atom as an ordinary, classical receiving antenna. Most people
> assume that
> the antenna can only absorb power that passes through the space in close
> proximity
> to the physical antenna. But it turns out that the antenna can actually
> pull in power
> over a much wider area. This is a suprising result that has nothing to do
> with quantum
> mechanics, but it helps to explain how weak, spread-out classical light can
> pack
> enough "punch" to knock an electron out of its orbit.
That sounds interesting. Do you say that the antenna will pull in some
em-wave, even if it's amplitude vanishes at the location of the
antenna? That would be a bit strange, but it does not sound
impossible. Or did I get you wrong?
> If you're still in agreement so far, I will be glad to sketch the outline
> of the antenna
> calculation in the ng.
Agreed.
Bye,
Christof
Martin Green
> > The first important concept is to realize that a hydrogen atom in a
> > mixed state,
> > partly s and partly p, looks like a little oscillating electric dipole
> > antenna. You
> > may already know this...it's standard QM, but not always taught in
courses.
>
> I don't know that. Is this the same for other two-level systems? I
> know I read something about general properties of two-level systems
> and that there are transitions from one state to the other. If that's
> it, I will look it up, but perhaps you don't mind to explain it,
> especially in the context of the H-atom acting like an electric dipole.
>
This is really important. Often professors give the impression that a
hydrogen atom
can be in either the s-state or the p-state but never "in-between". This is
wrong. The
atom cannot STAY in the mixed state indefinitely, because the mixed state
contains
accellerating charges which either lose or absorb energy...only the pure
states are
stable. But understanding the mixed state is the key to the semi-classical
explanation
of the photo-electric effect. (The term "semi-classical" refers to quantum
atoms and
classical radiation).
Here's how the atom looks like an antenna. The s-state is spherically
symmetric. The
p-state (let it be the one that lies along the z-axis) has a positive lobe
above the x-y plane,
and a negative lobe below the x-y plane. When you add the two states, there
is a net
cancellation at the bottom, and a net re-inforcement at the top. So the
wave function is
shifted northwards. The square of the wave function gives you the charge
density, and
this is your positive dipole moment.
That's half the story. To complete the picture, remember that the wave
functions are
each oscillating with their own frequency, corresponding to their energy
levels. So after
a certain period of time, corresponding to the difference in frequencies,
the relative polarities
of the two functions are reversed. Now the cancellation takes place ABOVE
the x-y plane,
and the re-inforcement below. So the charge density has moved to the other
side of the atom
....hence, the oscillating dipole. The atom is a little antenna.
> > The frequency of this antenna corresponds, of course, to 3/4 Rydberg,
which
> > is the energy difference between the two states.
>
Any light which shines on the atom with this frequency will exchange energy
with
it according to straightforward classical antenna theory.
Martin Green
Martin Green
> > The right way to do the calculation is to treat the light as an
ordinary
> > classical wave,
> > and the atom as an ordinary, classical receiving antenna. Most people
> > assume that
> > the antenna can only absorb power that passes through the space in
close
> > proximity
> > to the physical antenna. But it turns out that the antenna can actually
> > pull in power
> > over a much wider area. This is a suprising result that has nothing to
do
> > with quantum
> > mechanics, but it helps to explain how weak, spread-out classical light
can
> > pack
> > enough "punch" to knock an electron out of its orbit.
>
> That sounds interesting. Do you say that the antenna will pull in some
> em-wave, even if it's amplitude vanishes at the location of the
> antenna? That would be a bit strange, but it does not sound
> impossible. Or did I get you wrong?
My point is that the interaction takes place over a very wide region of
space, much bigger
than the physical dimensions of the antenna. The calculation goes something
like this:
Draw a series of parallel lines representing the wave-fronts of the
incoming wave. In the middle
of these lines, place a dot which represents the receving antenna. When the
incoming wave
forces the receiving antenna to oscillate, it re-radiates spherical
waves...draw these as concentric
circles, with the same in-between spacing as the incident parallel lines.
Notice that these out-going circles represent an energy flow which is
proportional to the
square of the antenna excitation. What is the source of this energy? It
must come from the
incoming wave. We can locate this missing energy if we examine the space in
the "shadow"
of the antenna.
Notice that behind the antenna, there is a zone where the two wave systems
(i.e. the paralled
incoming wave-fronts and the spherical out-going wavefronts) are close
enough togther that
they can reinforce each other coherently. You can see the shape of this
zone is a parabaloid.
How much energy is actually removed from the incoming wave?
Let the incident wave have unit strength, and the outgoing wave have
strength x << 1. The amplitude
of the wave in the shadow zone is 1-x, and its power is therefore
proportional to (1-x) squared.
The power which has been drawn out of the incident wave is 2x minus
x-squared....since x
is very small, we can ignore the x-squared.
Now look again at the spherical, out-going waves as a whole. Over most of
space, they represent
an outflow of power which is proportional to x-squared. But within the
parabaloid shadow region
behind the antenna, they represent a power GAIN (drawn from the incident
wave) which is proportional to x. As the antenna begins to oscillate in
response to the incoming wave, it (at
first) takes in more power than it gives off .... because the term in
x-squared will be much
smaller than the term in x. But as the antenna oscillates more strongly,
the x-squared term
(the re-radiated power) grows faster than the term in x (absorbed power)
and eventually there
is a balance point at which these two terms are equal. This is the maximum
level to which
the antenna can be excited.
At this point we can start to try and put numbers to everything and
actually calculate
results. But it's not really necessary to go through all that work. The
interesting thing to
observe is that NOWHERE in this calculation did we need to worry about the
size of the
antenna. The antenna interacts with the incident wave via its spherical
wave-pattern
which is pretty much the same no matter how big or small it is. If you do
the whole
calculation, you find out that the absorption cross-section of a small
electric dipole is
on the order of a square the size of the wave-length of the incident field.
For a hydrogen
atom making the s-p transition, this is much, much bigger than the size of
the atom...
by a factor of something on the order of 137^^2.
Conclusion: It's not necessary to concentrate the energy of light into a
tiny particle
in order for it to pack enough "punch" to knock an electron out of an atom.
Ordinary,
classical e-m radiation can do the job.
Martin Green
Jim Carr
}
} c.h.thompson <c.h.th...@newscientist.net> wrote:
} >As I said, the probabilities of QM are not even right all the time. They
} >are wrong in the EPR case, and experiments have unjustifiably been
} >interpreted as if they are right.
}
} Uh, which experiments, and what's incorrect about the interpretatins? Are
} you talking about the EPR experiments that have been done in the last year
} or so?
In article <381c0...@news2.vip.uk.com>
"c.h.thompson" <c.h.th...@newscientist.net> writes:
>
>All of them! ...
>
>Take visibility, which is an easy concept. We are dealing with sine waves,
>or almost sine waves. The visibility is just (max-min)/(max+min). Now this
>quantity is critically dependent on the min! If min is zero, visibility is
>1. Yet just by altering the settings of the detector you can alter the
>minimum, and you most certainly produce a drastic change if you "subtract
>accidentals"!
This criticism does not apply to the Weihs paper.
No background subtraction is needed to get high visibility.
Further, none of the experiments of the past year or so can be
described as producing data that disagree with the probabilities
predicted by QM.
Charles Francis
<j...@ibms48.scri.fsu.edu> writes
>Gregory L. Hansen <glha...@steel.ucs.indiana.edu> wrote
>}
>} c.h.thompson <c.h.th...@newscientist.net> wrote:
>} >As I said, the probabilities of QM are not even right all the time. They
>} >are wrong in the EPR case, and experiments have unjustifiably been
>} >interpreted as if they are right.
>}
>} Uh, which experiments, and what's incorrect about the interpretatins? Are
>} you talking about the EPR experiments that have been done in the last year
>} or so?
>
>In article <381c0...@news2.vip.uk.com>
>"c.h.thompson" <c.h.th...@newscientist.net> writes:
>>
>>All of them! ...
>>
>>Take visibility, which is an easy concept. We are dealing with sine waves,
>>or almost sine waves. The visibility is just (max-min)/(max+min). Now this
>>quantity is critically dependent on the min! If min is zero, visibility is
>>1. Yet just by altering the settings of the detector you can alter the
>>minimum, and you most certainly produce a drastic change if you "subtract
>>accidentals"!
>
> This criticism does not apply to the Weihs paper.
>
> No background subtraction is needed to get high visibility.
>
> Further, none of the experiments of the past year or so can be
> described as producing data that disagree with the probabilities
> predicted by QM.
I appreciate your checking this. While Caroline was on the question of
the subtraction of accidentals, I was able to consider that she might
have a point, but as time moved on it has become clear that she is
merely on a crusade to impose her own, rather narrow, notion of the
metaphysics of space-time, by fabricating 'evidence' and refusing to
look at any logical or mathematical argument which contradicts what she
wants to believe. Although the irrationality of her position that a
(light) wave is a local phenomenon ought to cause even her to reflect.
--
Charles Francis
cha...@clef.demon.co.uk
c.h.thompson
> Gregory L. Hansen <glha...@steel.ucs.indiana.edu> wrote
> } c.h.thompson <c.h.th...@newscientist.net> wrote:
> } >As I said, the probabilities of QM are not even right all the time.
They
> } >are wrong in the EPR case, and experiments have unjustifiably been
> } >interpreted as if they are right.
> }
> } Uh, which experiments, and what's incorrect about the interpretatins?
Are
> } you talking about the EPR experiments that have been done in the last
year
> } or so?
>
> In article <381c0...@news2.vip.uk.com>
> "c.h.thompson" <c.h.th...@newscientist.net> writes:
> >
> >All of them! ...
> >
> >Take visibility, which is an easy concept. We are dealing with sine
waves,
> >or almost sine waves. The visibility is just (max-min)/(max+min). Now
this
> >quantity is critically dependent on the min! If min is zero, visibility
is
> >1. Yet just by altering the settings of the detector you can alter the
> >minimum, and you most certainly produce a drastic change if you "subtract
> >accidentals"!
>
> This criticism does not apply to the Weihs paper.
Jim, you know perfectly well what I mean! Weihs did not subtract
accidentals but his Bell test (better than a visibility one but not by much)
was invalid because, as he states, he had to assume that "the pairs
registered are a fair sample of all pairs emitted". This is not a
reasonable assumption (see, for example, my Chaotic Ball paper,
http://xxx.lanl.gov/abs/quant-ph/9611037 and a later one that is perhaps
more relevant to recent experiments,
http://xxx.lanl.gov/abs/quant-ph/9903066). As I have explained elsewhere,
it is not a matter of the experimenter choosing a biased sample! He is not
taking a survey! The sample is not under his control. There are very few
real conditions in which the assumption of fairness is at all likely to be
true.
It is possible to conduct tests that could establish in some cases when it
is NOT true (though they cannot quite tell us when it is). What is needed
is a check on the observed total coincidences over the full range of
settings of both detectors. If there is any sign of sytematic variation
then we have not got fair sampling. Neither Weihs nor, to my knowledge,
anybody else who has made the assumption has done such checks to any
satisfactory standard. Aspect did some quite good tests, but I have not
been able to extract further information from him on just what angles he
looked at. All I know is that in his thesis he does mention a discrepancy
of 1 or 2 percent, and in the paper that resulted he mentions that,
presumably because of this, he did a modified Bell test as well as an
ordinary one. Someone, some day, really ought to look into this, as I
rather doubt if anyone has ever checked his modification ...
Anyway, as I'm sure you know, if in a given experiment a SINGLE loophole
exists then to claim that it matches QT predictions is not justifiable.
(Well OK, "misleading" might be a better way of putting it.) The results
will also match a realist model.
> Further, none of the experiments of the past year or so can be
> described as producing data that disagree with the probabilities
> predicted by QM.
Is this not arguing like a student who expects to pass a maths exam when
they've only got the answers right, not the methods? In my days, this used
to score zero!
Caroline
Charles Francis
>then we have not got fair sampling.
Now you are saying something sensible. It is not possible to conduct any
experiment on the fundamental quantities of nature without disturbing
them in some way with the measurement. It is not possible to obtain fair
sampling, and so it is not possible to use a classical, unmodified
probability theory in the model for experimental results.
> Neither Weihs nor, to my knowledge,
>anybody else who has made the assumption has done such checks to any
>satisfactory standard.
I have, and you have not studied it. The effect of the apparatus is that
the appropriate probability model is quantum logic. That is why we
should believe it, not because of any particular experiment.
http://xxx.lanl.gov/abs/physics/9909047
A Model of Classical and Quantum Measurement
You should also look at
http://xxx.lanl.gov/abs/physics/9909048
Conceptual Foundations of Special and General Relativity
Neither of these papers involves a level of maths you couldn't follow,
so you really have little excuse.
--
Charles Francis
cha...@clef.demon.co.uk
c.h.thompson
Charles Francis <cha...@clef.demon.co.uk> wrote in message >
[See]
> http://xxx.lanl.gov/abs/physics/9909047
> A Model of Classical and Quantum Measurement
for a model that explains why we should accept that the QM predictions for
EPR experiments are correct, despite the fact that "fair sampling" is not
possible.
Dear Charles, I've just looked at the abstract. Your say "We take the view
that physical quantities are values generated by processes in measurement,
not pre-existent objective quantities ..."
I can't accept that this is realist approach! I shall not read the article.
Besides, when you said that you had "made the necessary checks", I rather
doubt whether you meant you had actually done the tests I suggested! Pull
the other one! You don't have a lab hidden away in a back room do you?
The reason the QM prediction is unsatisfactory is that if you alter the
detectors a little you get a different answer. No QM formula I have seen
has shown any promise of being able to adapt to realistic changes in
detector characteristics. It cannot adapt correctly because it is committed
to the idea that light is a particle, so it is trapped into assuming that
all there is to say about a detector is contained in just one figure: its
"quantum efficiency".
Some investigations that I hope Gregor Weihs is currently carrying out may
help settle this matter. I am not interested in your theoretical approach.
> You should also look at
>
> http://xxx.lanl.gov/abs/physics/9909048
> Conceptual Foundations of Special and General Relativity
No, I don't think so! I'll manage without, thanks. If you don't accept the
independent existence of "pre-existent objective quantities" that are there,
unobserved, causing the apparatus to yield the values we see, I cannot
discuss physics with you.
Caroline
Jim Carr
jtb...@presby.edu (Jon Bell) wrote:
}
} Bruce Richmond <bsr...@my-deja.com> wrote:
} >In article <7vavmc$i8n$1...@jetsam.uits.indiana.edu>,
} >glha...@steel.ucs.indiana.edu (Gregory L. Hansen) wrote:
} >>
} >> Energy of a photon is E=hf, Planck's constant multiplied by the
} >>frequency. Total energy of light in a region is U=nE, number of
} >>photons multiplied by the energy of a photon. Intensity is
} >>I=U/A=nE/A=nhf/A, energy divided by area.
} >
} >How is the frequency defined? It must include n/t and I would think
} >the n would have to be per unit of area.
}
} Frequency is the number of cycles per second. When a light wave
} passes a point, the electric and magnetic fields at that point
} oscillate back and forth at a rate of f cycles per second.
} You can write it as f = n/t, but keep in mind that this n is
} different from the n in U = nE above (which is the number of
} photons in a given volume of space).
In article <7vikeb$uvu$1...@nnrp1.deja.com>
Bruce Richmond <bsr...@my-deja.com> writes:
>
>We were discussing light being modeled as individual particles, and a
>single particle as we normally think of a particle cannot have a
>frequency. That's why I assumed f represented a count of the particles
>hitting a given area per unit of time.
Except that the simplest corpuscular model for light is known to
be wrong, so what is being talked about above is the quantum model
for photons where you get macroscopic electromagnetic waves if
there are enough photons present. (Large N limit for QED.)
We also model neutrons and electrons as particles, but they also
diffract as if they propagate via wave-like rules.
>In practice I imagine that frequency is not determined with a detector
>actually counting photons but by using a prism to select a particular
>frequency for study.
The frequency of an individual photon is determined by its energy.
>If photons have no mass why are they affected by a gravitational field?
The General Relativity explanation (which explains the data) is that
gravity is due to the curvature of space-time. Just as the shortest
distance between two points on the surface of the earth is a Great
Circle route, a "straight line" in a curved space-time is not straight.
Thus light, traveling along the quickest path, must bend.
See the Relativity FAQ for information on this.
Jim Carr
"Martin Green" <test...@pangea.ca> writes:
>
>It's been something like a year since I've checked in on sci.physics, and
>the photo-electric effect continues to be put forward as evidence for the
>existence of photons.
See my posted comment on the historical approach used in textbooks,
most of which also mention the Compton Effect, and the even stronger
historical bias in the popular literature.
The real challenge for classical theories is the Compton Effect.