Double slit experiment with particles

[michael879]

ok so I read this book recently, timeline. In it they explain an experiment I've heard about a lot in physics. However, they gave a very strange look at it that I can't figure out. Basically, its the double slit experiment except instead of using light waves / many photons, you use particles. They use photons but I am pretty sure I remember hearing electrons work too. If you send a lot at once, whether they are particles or waves the pattern is explainable. However, if u send 1 particle at a time, for each one you get a single dot on the screen right? if you send 1000, 1 at a time, you get the exact same pattern. No interference in this case, so how is it possible? Michael Chrighton uses this to prove this the existence of an infinite amount of multiple universes that interfere with ours, but I find that kind of weird. Anyone know any explanation of this?

 

[jtbell]

if u send 1 particle at a time, for each one you get a single dot on the screen right?

Right. See this article for some pictures of the interference pattern as it builds up one dot at a time.

if you send 1000, 1 at a time, you get the exact same pattern. No interference in this case,

The pattern demonstrates that there is interference taking place.

Michael Chrighton uses this to prove this the existence of an infinite amount of multiple universes that interfere with ours,

That's the "many-worlds" interpretation of quantum mechanics, which some people like...

but I find that kind of weird.

So do a lot of other people.

There are other interpretations of what's going on, but they all have features that many people find kind of weird. There doesn't seem to be any way to avoid weirdness in some form.

Some people like to argue endlessly about interpretations of quantum mechanics, but so long as they're all based on the same mathematics and they all predict the same results for actual physical measurements, and agree with experiments, it's kind of hard to resolve such arguments. I suspect that most practicing physicists don't worry too much about interpretations; they belong to the "shut up and calculate" camp.

 

[Sherlock]

If you send a series of single-photon emissions, through a double-slit
barrier, then you will eventually build up an interference pattern one
photon at a time on a detection screen.

In 1989, A. Tonomura et al. did this with electrons. By sending one
electron at a time through an electron biprism, and then through a
system of electrostatic lenses, they used a position-sensitive
electron-counting system to ultimately produce a video image of
electrons arriving one by one -- gradually building up an
interference pattern. After 3000 arrivals, it looks like the
dots are evenly distributed on the monitor screen. After 20,000
the faintest interference pattern is beginning to emerge.
And, after 70,000 the interference pattern is clearly evident.

How can *particles* do this? A particle would have
to go through one slit or the other, wouldn't it? How could
particles passing one by one through a double-slitted
barrier be interfering with each other (or themselves)?

Ok, here's where it's important to pay close attention
to the language surrounding the observed phenomena and
the calculations. While quantum mechanics speaks
of the probability of finding "the particle" at a given
point, the actual calculation is about the behavior
of waves (via the Schroedinger equation).

If you send a wave through a double-slit barrier, then
you get two interfering waves on the other side.

So, one way to look at it is that the physical nature
of electrons and photons is that they are waves, and
these waves produce, via certain experimental procedures,
the discrete instrumental phenomena (the 'particles')
that are predicted by theory.

 

[joshuaw]

This has to do with coherence vs incoherence. This is where the problems really can set in.

If I measure a particle through one slit or the other which demands that it go through that particular slit, then we do not have the interference.

If you measure where the particle will hit beyond the slits, then you can send 1, 5, 1000 or however many you wish, the interference pattern will still appear.

josh

 

[michael879]

ok well in timeline he says that the photons NEVER appear in the parts of the interference patterns that donts show up. Is it wrong? because you said the electrons were evenly distributed for a while. Also, the fact that whatever particle you use hits in a certain spot proves that it is a particle right? If photons were waves they would each create their own interference pattern instead of creating a point on the screen. If you send 1000000 photons at a double slit, 1 at a time there is no interference. How can the photons interfer with each other when they arent sent at the same time?

 

[juju]

If you assume the particle and its fields go through both slits at the same time, and if you interpret the wavefunction/superposition as an evolving distribution of particle properties, then there is no problem. The interactions at the screen (resultant pattern) are the result of the stength (amplitude) and asymmetry (uniqueness) of the particle distribution's interaction with the screen.

In this viewpoint the point particle is just the convergence point (or zone) for the value/symmetry structure of the particles fields.

juju

 

[michael879]

so what, your saying it interferes with itself due to the fact that its a wave but hits the screen as a particle?

 

[gonzo]

Yes, that's one way to look at it. The wave goes through both slits and interferes with itself. But the wave is just a probability distribution.

By the way, I believe they have done the two slit experiment with objects as large as balls made of 60 carbon atoms and still get an inteference pattern. So even a large molecule like that can be shown to be behave quantum mechanically under the right conditions.

 

[misskitty]

In 1989, A. Tonomura et al. did this with electrons. By sending one
electron at a time through an electron biprism, and then through a
system of electrostatic lenses, they used a position-sensitive
electron-counting system to ultimately produce a video image of
electrons arriving one by one -- gradually building up an
interference pattern. After 3000 arrivals, it looks like the
dots are evenly distributed on the monitor screen. After 20,000
the faintest interference pattern is beginning to emerge.
And, after 70,000 the interference pattern is clearly evident.

This experiment was originally done by Young ( I can't recall his first name I think it was Thomas) and was later repeated using electrons by Davisson and Germer in 1927. By achieving the same results you stated they had proved De Broglie right in his hypothesis of everything follow the rule of complimentarity. Schrodinger's Psi ^2 basically stated that the universe exsists in a set of probablities and there is this instantaneous exsistance-nonexsistance mode in which we all live.

I agree with what you are saying Sherlock. I just wanted to expand on it a little in case anyone wanted to do a bit of backround research on the topic.

~Kitty

 

[misskitty]

Yes, that's one way to look at it. The wave goes through both slits and interferes with itself. But the wave is just a probability distribution.

By the way, I believe they have done the two slit experiment with objects as large as balls made of 60 carbon atoms and still get an inteference pattern. So even a large molecule like that can be shown to be behave quantum mechanically under the right conditions.

This has to do with De Broglie applying everything to the rule of complimentarity. He was the one who proposed everything has an associated wavelength. Hence why we call it the De Broglie wavelength.

Interesting that this same experiment was performed with carbon balls. Do you know when this was done?

 

[misskitty]

Michael, if you are feeling a bit confused and lost, its ok. There are many scientists that are still trying to answer some of the questions you have been asking. So you are not alone in your confusion.

I'm still confused on why you get two interferring waves when you shoot one particle at two slits and it only goes through one slit. One of the explanations I got from my reading (text book assignment) was that electrons defract like light does. The other was light isn't a particle or a wave...it's light. You need both models to explain the behavior of light. Must stop now before I get more confused...

 

[joshuaw]

I think this may help, it works for me but I am not promising anything:
Every "particle" (and photons can be considered this since they also can be particle anti-particle pairs at any time) has a wavefunction. This wavefunction gives the probability to find information about that particle. There are spin wave functions, momentum, position, etc... Now when we measure a particle's position it must be in a given finite region. If we measure where the particle hits behind the slits, we have a probability to measure it in a given region. Once our statistics are high enough, we can see this probability distribution which is the interference spectrum.

If we measure (which means we are able to distinguish which slit the particle passeses) which slit the particle passes through, the interference is no longer seen. We have "forced the particle to pick" which slit it passes through. So we have equal distributions for each slit(given enough statistics).

Once a particle is forced into a state that we measure, the wavefunction collapses to that state for a time after we measure, then it "goes back to normal" in a sense(that part is hard to explain). This is covered pretty well in Griffiths Quantum Mechanics textbook in the first 2 chapters if you have access to it.

Later in the book it discusses the idea of self interaction. But that is a bit too much for this post. Remember, the Dirac notation of Quantum Mechanics were used before the Schroedinger Wave Equation. I mean you could ask why does a spinnor have to be a symmetric or anti-symmetric wave function(ie how can a spinnor be a wave). But once you get used to the vocabulary, it becomes a little clearer.

Josh

 

[michael879]

I think this may help, it works for me but I am not promising anything:
Every "particle" (and photons can be considered this since they also can be particle anti-particle pairs at any time) has a wavefunction. This wavefunction gives the probability to find information about that particle. There are spin wave functions, momentum, position, etc... Now when we measure a particle's position it must be in a given finite region. If we measure where the particle hits behind the slits, we have a probability to measure it in a given region. Once our statistics are high enough, we can see this probability distribution which is the interference spectrum.

If we measure (which means we are able to distinguish which slit the particle passeses) which slit the particle passes through, the interference is no longer seen. We have "forced the particle to pick" which slit it passes through. So we have equal distributions for each slit(given enough statistics).

Once a particle is forced into a state that we measure, the wavefunction collapses to that state for a time after we measure, then it "goes back to normal" in a sense(that part is hard to explain). This is covered pretty well in Griffiths Quantum Mechanics textbook in the first 2 chapters if you have access to it.

Later in the book it discusses the idea of self interaction. But that is a bit too much for this post. Remember, the Dirac notation of Quantum Mechanics were used before the Schroedinger Wave Equation. I mean you could ask why does a spinnor have to be a symmetric or anti-symmetric wave function(ie how can a spinnor be a wave). But once you get used to the vocabulary, it becomes a little clearer.

Josh

wait WHAT? youre saying that if we send photons at the double slit one at a time, they will create the interference pattern unless we measure which hole they are going through and then they will only go through one or the other?

 

[joshuaw]

Yes, but you have to make sure you can measure which slit they pass. This destroys the information "stored" in the particle. Check out Griffiths, this website also gives an illustraction: http://boson.physics.sc.edu/~gothe/511-S05/werbung.html

For P1 and P2, we can distinguish which slit the particle passes(we can measure which slit the particle passes) This is the incoherent case. For P12, we get an interfernce spectrum, this is the coherent case. This is very important to understanding measurements as well as quantum mechanics,

Josh

 

[Sherlock]

ok well in timeline he says that the photons NEVER appear in the parts of the interference patterns that donts show up. Is it wrong?

What experiment are they referring to?

because you said the electrons were evenly distributed for a while. Also, the fact that whatever particle you use hits in a certain spot proves that it is a particle right?

The word photon refers, most precisely, to a quantum of a single
mode (single wavelength, direction, and polarization) of the
electromagnetic field. In this sense photons are, as has
been pointed out in other threads, theoretical creations that
are *neither* waves nor particles in any classically analogous
sense.

However, the nature of light is that it's waves in a medium
of unknown physical structure. At least that's how I think
about it.

Specific detection locations on a detecting screen correspond
to wavefront(s) of specific energy interacting with the
detecting screen at those 'points'. There are no 'particles'
in the classical sense involved -- just wavefronts moving
from the emitter to the detector.

If photons were waves they would each create their own interference pattern instead of creating a point on the screen. If you send 1000000 photons at a double slit, 1 at a time there is no interference. How can the photons interfer with each other when they arent sent at the same time?

If you send a certain number of wavefronts at a double slit, one
at a time, there is interference. As each wavefront goes through
the double-slit barrier, two wavefronts are created and they
interfere with each other. A certain number of cycles correspond
to the photon (or electron) that refers to the individual detection
event.

 

[michael879]

If you send a certain number of wavefronts at a double slit, one
at a time, there is interference. As each wavefront goes through
the double-slit barrier, two wavefronts are created and they
interfere with each other. A certain number of cycles correspond
to the photon (or electron) that refers to the individual detection
event.

ok, i get that. But if there are two wave fronts, why does the light only appear as a single dot on the screen? shouldn't it appear as a complete interference pattern?

 

[caribou]

You can think of the waves as "collapsing" to a particle in a definite place upon detection.

Or you can think of a photon or electron as a particle which acts as if it went from A to B by every possible path and that all of these possible histories for the particle interfere with each other, hence a somewhat wave-like interference effect influences where the particle is eventually found.

The second view is the one held by the more recent "big name" scientists as far as I know. Waves "collapsing" is now considered by some to be just a mathematical shortcut, not a physical effect. It can be useful, though.

 

[Sherlock]

... if there are two wave fronts, why does the light only appear as a single dot on the screen? shouldn't it appear as a complete interference pattern?

Why are the screen dots produced sequentially and
not simultaneously? The short answer is that the
disturbances that produce the light waves don't
all happen at the same time. They happen one after
another. *Very short* times, to be sure, but still
one by one.

But, maybe that's not exactly what you're getting at.

The two wavefronts produced by the double-slit interfere,
and the wavefronts produced by the interference hit the
detecting screen at different locations. The detecting
screen is like a lattice of wave structures (atoms) that
are the same as each other. For any given photon, or
'packet' of wavefronts, the kinetic energy imparted to the
screen is concentrated at the 'point' on the screen
where the first-arriving wavefront of each interference-produced
waveset hits the screen.

[I'm editing this now, because I see that it doesn't answer
your (*the*) question. Even if one location (atom?) is being
hit first by the interference-produced wavefronts of the
group comprising a single photon, why are the other
interference-produced wavefronts of the photon apparently
having no effect (imparting no energy) on the other
locations where wavefronts should be hitting. Well, maybe
they are, but it's just not enough to produce other detectable
effects (eg., ionize atoms at those other locations) during the
time interval defining the photon event. It takes a
certain amount of kinetic energy to excite an atom.
The other interference-produced wavefronts might be
contributing to this, but the energy imparted at the other
interaction locations during the given interval is just below
the threshold required to produce a screen-dot at those
locations. So, you get one screen-dot per photon.]

The higher the photon flux, the more wavefronts hitting the
detecting screen per unit of time, and the more 'dots' you see
per unit time.

At least, this is how I picture it. But, don't take *my* word
for any of this. It might not be the best heuristic.

 

[NEOclassic]

If you collimate beams so as miss the edges of the

if u send 1 particle at a time, for each one you get a single dot on the screen right? if you send 1000, 1 at a time, you get the exact same pattern. No interference in this case, so how is it possible?

If you collimate beams whether photons or particles (or both for that matter) there cannot be any wave phenomena. It should be possible to flood (including the edges of the orifices) one opening with photons and the other with electrons and get a so-called wave-pattern. I suggest that your example is the epitome of collimation so that you have already proven my point. Look up Fresnel's Bright Spot experiment which has been repeated recently using laser beams where the bright spot shows that laser photons were bent as a consequence of passing near the silhouette edge of a spherical target. Remember that the sharp edge often found with slits actually has a finite radius of curvature and that's all you need to show that the wave pattern is a function of the edges of the slits or holes etc.

Cheers, Jim

 

[michael879]

If you collimate beams whether photons or particles (or both for that matter) there cannot be any wave phenomena. It should be possible to flood (including the edges of the orifices) one opening with photons and the other with electrons and get a so-called wave-pattern. I suggest that your example is the epitome of collimation so that you have already proven my point. Look up Fresnel's Bright Spot experiment which has been repeated recently using laser beams where the bright spot shows that laser photons were bent as a consequence of passing near the silhouette edge of a spherical target. Remember that the sharp edge often found with slits actually has a finite radius of curvature and that's all you need to show that the wave pattern is a function of the edges of the slits or holes etc.

Cheers, Jim

wow that actually rly helped even thought I barely understood the english lol. what does collimate mean?

 

[Sherlock]

If you collimate beams whether photons or particles (or both for that matter) there cannot be any wave phenomena. It should be possible to flood (including the edges of the orifices) one opening with photons and the other with electrons and get a so-called wave-pattern. I suggest that your example is the epitome of collimation so that you have already proven my point. Look up Fresnel's Bright Spot experiment which has been repeated recently using laser beams where the bright spot shows that laser photons were bent as a consequence of passing near the silhouette edge of a spherical target. Remember that the sharp edge often found with slits actually has a finite radius of curvature and that's all you need to show that the wave pattern is a function of the edges of the slits or holes etc.

wow that actually rly helped even thought I barely understood the english lol. what does collimate mean?

I thought we were talking about situations where there's only
one photon or electron in the apparatus at a time. If so, then
flooding the openings doesn't answer the question

Collimation basically means aiming the photons or electrons so
that they're directed at the same point (I think). I don't know
if the experiments that deal with single photon or electron
interference do that in the way that NEOclassic seems to be
suggesting.

Anyway, the question remains: why do the wave packets or
wave trains that are photons only produce one dot on a detecting
screen (if their physical nature is that they are wave phenomena)?

 

[michael879]

when I said what he said helped, I didnt mean with the original question. the wave collapsing theory seems to work although its almost as weird as the multiple universe one.

 

[Sherlock]

when I said what he said helped, I didnt mean with the original question. the wave collapsing theory seems to work although its almost as weird as the multiple universe one.

The multiple universe idea seems like too much of a stretch to
me also. There might be something to the wave collapse idea,
but then there's the problem with nonlocality/superluminality.
In any case, the *mechanism* of 'wave collapse' is unspecified.
I would like to be able to envision, largely in terms of analogy with
classical wave mechanics, what's happening at the level of
photon-atom interactions -- which doesn't seem possible at this
time. But, I've just begun thinking about it this way.
Anyway, I think it's an interesting question. There seems to be
more to this than what quantum mechanics allows one to 'see'.
The deep reality of the instrumental phenomena that we're
pondering is yet to be revealed.

I should note that I don't so far see any reason why wave
behavior at the level of individual quanta should be fundamentally
different than wave behavior at the macroscopic level. The details
are very difficult to sort out because we can't see what's
happening.

 

[NEOclassic]

I thought we were talking about situations where there's only
one photon or electron in the apparatus at a time. If so, then
flooding the openings doesn't answer the question

Collimation basically means aiming the photons or electrons so
that they're directed at the same point (I think). I don't know
if the experiments that deal with single photon or electron
interference do that in the way that NEOclassic seems to be
suggesting.

Anyway, the question remains: why do the wave packets or
wave trains that are photons only produce one dot on a detecting
screen (if their physical nature is that they are wave phenomena)?

Hi Sherlock
Thanks for explaining collimation and I would add that you are correct about the beam of photons or electrons being focused to a point but really it means that the photons in the bundle do not come anywhere close to the edges (regardless of the shape of the slit) of the opening - the diameter of the bundle as it passes through the slit is the same as the diameter where the bundle meets the target. (a laser beam) For electrons the bundle needs to be elecrostatically moving while being magnetically focused (as in the fashion of the electron beam inside a TV tube).
Your argument concerning a single electron or photon in the apparatus cannot form a wave pattern but only a dot on the screen, even a thousand or a billion in collimated beams that do not become diffracted cannot yield a wave pattern.
I suggested the laser test of the Fresnel "bright spot" experiment to show how a coherent parallel beam of photons were able to bend around the circular silhouette of an opaque sphere. Of course this was discovered a few hundred years ago when most photons were not beamed but divergent.
Perhaps photons are really particulant rather than wavy and the real problem is in the improper use of the slit experiments by flooding the orifices!
Cheers, Jim

 

[Sherlock]

For electrons the bundle needs to be elecrostatically moving while being magnetically focused (as in the fashion of the electron beam inside a TV tube).
Your argument concerning a single electron or photon in the apparatus cannot form a wave pattern but only a dot on the screen, even a thousand or a billion in collimated beams that do not become diffracted cannot yield a wave pattern.

Some interference is happening following the passing of the
photon or electron waves through the double slits. This seems evident
from the recorded patterns that eventually emerge. But the energy
of individual photons or electrons is focused in a vary small area.
I'm trying to see if it's possible to explain this using wave analogy.

I suggested the laser test of the Fresnel "bright spot" experiment to show how a coherent parallel beam of photons were able to bend around the circular silhouette of an opaque sphere.

This suggests a wave nature ... to me anyway.

Of course this was discovered a few hundred years ago when most photons were not beamed but divergent.
Perhaps photons are really particulant rather than wavy and the real problem is in the improper use of the slit experiments by flooding the orifices!
Cheers, Jim

Particles are wave bundles. Waves are disturbances in some medium
or other. A medium is a grouping of particles having similar properties.
Photons and electrons are themselves both media and disturbances
(or collections of disturbances) in some more fundamental medium
that we can't yet (and maybe never will be able to) ascertain the
structure of. Are these more fundamental media in any sense
particulate? They could be. Is there a fundamental medium (or
media) that isn't in any sense particulate? There could be, but
such a *continuous*, seamless, fundamental medium would, it seems,
never be amenable to our probings (ie., if everything is,
fundamentally, made of and moving as the same seamless 'stuff'
moves, then how would we detect *it*, the stuff itself, that is?).

String theory is an attempt to construct a fundamental particulate
medium underlying (and therefore unifying) all extant physical
and standard theoretical phenomena.

Anyway, I'm still stuck on the details of how interfering photon (or electron)
waves produce the (double-slit) effects that they do.

I don't think that collimation, or non-diffraction, or that the
openings are flooded is the answer.

 

[ZapperZ]

Anyway, I'm still stuck on the details of how interfering photon (or electron) waves produce the (double-slit) effects that they do.

You may want to catch up on what has been discussed many times already starting with this thread:

https://www.physicsforums.com/showthread.php?t=68917&page=1&pp=15

Zz.

 

[Sherlock]

You may want to catch up on what has been discussed many times already starting with this thread:

https://www.physicsforums.com/showthread.php?t=68917&page=1&pp=15

Zz.

Thanks for pointing that out. I probably would have
eventually found it, but sooner is better than later.
Though I've got a couple of qm treatments for similar
setups here -- am just looking for something that will give
me a general 'picture' (worth 1000 words? -- in the spirit
of emulating the approach of some great physicists) of the
behavior of each wavefront (assuming that, eg., from
photon to photon, the properties of the emitted waves
will be slightly different) as it passes through the
double slits, interferes, and then interacts with the
detecting plate or screen (ie., giving me a better idea
why a detection occurs in one maximal -- and, infrequently,
in a minimal -- band and not another.) And, a *picture*
of what is happening to produce a single, focused
energy transfer location (in one band) per photon, rather
than the several that my, so far naive, way of approaching
this (as purely a wave phenomenon) would seem to require.