Aug 28, 2007
Two Slits Are Better Than One
Sciencebase Exclusive - Careful experimentation and theoretical analysis of a double-slit experiment have finally quashed a controversy in fundamental physics – the complementarity-uncertainty debate.
Ever since the catflap to the quantum world was opened up to us and Schrödinger’s feline friend was idiomatically let out of the bag, to mix a metaphor or two, there have been more questions and controversies raised than conundrums solved in the world of the very, very small. How can something be both particle and wave, for instance? What allows particles of matter to tunnel through solid objects? And, how is the interference pattern destroyed in a double-slit experiment when measurements are performed on the path traversed by a particle?
What is a double slit experiment, you ask? Well, traditionally, Young’s double-slit experiment consists of shining a light through two narrow, closely spaced slits and observing the results on a screen placed beyond the slits.
Intuitively, you might think that the result would simply be two bright lines, aligned with the slits, representing where the light passes through the slits and hits the card. However, this is not seen in practice, instead, the light is diffracted by the slits and produces fringes corresponding to wave-like interference pattern. The fringes of light and dark regions correspond to where either the light waves constructively (add) and destructively (subtract) from each other. Two peaks in the light wave meet to make a brighter fringe whereas a dark fringe is formed when a peak and a trough coincide. This result seemingly settles a three-century conundrum about whether light is particle or wave, showing apparently that it is a wave.
However, a similar experiment carried out with beams of electrons or atoms fired through the slits produces a very similar interference pattern. How could that be? Particles are solid objects, surely? Well, the double-slit experiment shows that they are not. They produce an interference pattern, which suggests that the particles behave as waves.
The double-slit experiments work perfectly well and reveals interference patterns with light, electrons, and beams of other particles, but only if the experimenter does not try to find out through which slit a particular wave-particle passed before hitting the screen. Try to fire particles through the slits one at a time and as illustratd in the 5-minute video below, you will still see an interference pattern. It is as if each particle passes through both slits simultaneously, each slit individually and together and neither slit all at the same time; behaving some as waves…
As if this were not complicated enough, physicists reasoned that if they could discover which slit the individual particle really goes through each time in this experiment, they could solve the problem. So, they put a measuring device next to one slot and observed what happens as particles are fired through the slits one at a time. Astoundingly, the interference pattern disappears, simply having a measuring device present to observe the route taken by the particles somehow disturbs their wave-like nature and they revert to being tiny, solid objects and produce just two bands on the screen as if they were tiny marbles rather than wave. How could the particles know they were being watched.
This loss of interference has been explained by several of the biggest names in twentieth century physics, among them Niels Bohr and Richard Feynman. They suggested that whenever the path is measured within the double-slit, the momentum of the wave-particle is uncontrollably and irreversibly disturbed. Think about it, it has to be affected by the observer somehow because the very act of observing involves some kind of sharing of information either via photons, charge, energy or matter. This process “washes out” the interference fringes.
Most physicists simply accept this as being precisely what happens. It is a little vague and some might say “handwaving” because it does not pin down the nature of this washing out nor say anything about how the momentum is disturbed by the transaction between observer and observed. More precisely, it is simply what happens because of the back-reaction resulting from the Heisenberg uncertainty relation that says we cannot know simultaneously both the energy and position of any quantum wave or particle with absolute precision. While that kind of folds the argument into a loop, Feynman famously pointed out that, “No one has ever thought of a way around the uncertainty principle.”
But, not everyone was happy with this. In 1991, 
Marlan Scully, Berthold-Georg Englert, and Herbert Walther (Nature 1991, 351, 111) suggested that a microscopic pointer could be used to carry out the observation in such a way that the very act of observation would not disturb the momentum of the particle and so bypass the uncontrollable and irreversible effects suggested by Bohr that leads to interference breakdown. However, Pippa Storey, Sze Tan, Matthew Collett, and Daniel Walls (Nature, 1994, 367, 626), countered this argument, demonstrating that no matter how small the observer nor how the measurements are made, momentum is affected and the interference pattern would disappear. A long and controversial debate has raged between the two scientific factions that back either the Scully or Walls teams.
A theoretical solution was posited by 
Howard Wiseman and colleagues in 2003 (Phys Rev A, 2003, 311, 285) and refined in 2004 (J. Opt. B: Quant. Semiclass. Opt. 2004, 6, S506-S517). Now, in a seminal paper published today in the New Journal of Physics, Aephraim Steinberg together with Wiseman and colleagues Mir, Lundeen, Mitchell, and Garretson have applied the theory in a novel double-slit setup. Their experimental results suggest that, as is the way with all things quantum, both camps are equally correct and equally wrong. Somehow, you can have your quantum cake and eat it.
They found that by using only weak measurements, they can directly observe the momentum transfer that causes interference breakdown but equally do so without disturbing the two-slit superposition. They effectively verify both the Scully and Walls views. In terms of the Scully position, the team shows that there is no change in the mean momentum, or the mean energy, whereas with respect to the Walls work, they show that the momentum is spread, as one would expect given the uncertainty inherent in the quantum world, according to Heisenberg’s principle.
Feynman always held that the double-slit setup was central to quantum theory, but would never be fully understood. This work by Wiseman and colleagues shows that the humble double-slit experiment can still throw up new quantum mysteries to baffle us.
David Bradley
Site FAQ
Contact me
Web 2.0 stuff
Nature Reviews Drug Discovery
Email updates
Podcast
My recent reading
Join my LinkedIn network
Connect on StumbleUpon
Poke me on Facebook
Follow my Plurks
My Flickr gallery
Sciencebase on Twitter
Be my Feed Friend
Ariana said,
August 28, 2007 at 8:54 pm
This was a lot of help for my Science prodject!!!!!
Shaun said,
August 28, 2007 at 9:41 pm
Does the waveform vary depending on the distance between the slits and the projector? Is it affected by the width and separation of the slits? How do additional slits affect the wave patterns? Is it possible for the waveform motion to be introduced by something environmental? Does changing the material of the slitted panel affect the waveform? Thicker panels, thinner panels, wider, thiner sots? Do other materials of similar mass to the sensors used break the wave pattern the same way the sensor does? Is the panel losing any energy?
David Bradley said,
August 29, 2007 at 9:33 am
Shaun - those are very good questions, many of which have been answered in experiments discussed elsewhere. If any Sciencebase reader with indepth knowledge of the double slit experiment cares to answer any or all of them, I’d be very grateful…
Ariana - you’re welcome, hope it gets you an A grade (by the way project has no “d”)
Wayne Smallman said,
August 29, 2007 at 8:48 pm
What I can’t quite figure out is the observational interference problem.
Yes, we know that there must be some kind of exchange in energy. But all a sensor does is see what’s there, not extract extra energy.
A small sensor is no more robbing energy from a particle of light than a sheet of cardboard does.
However, if there’s any doubt about how sensors take energy away, why doesn’t someone develop a type of sensor that puts the energy back in once it’s finished?
I know this is a totally naive idea, but while I’m no scientist, I like to think I’m fairly logical…
David Bradley said,
August 29, 2007 at 10:12 pm
Wayne, I suppose the problem of the idea of something being exchanged between observer and observed might be explained by analogy with how our eyes see things. For something to be visible photons of light must touch an object, they are absorbed by the particles at the surface of the object, which are nudged into an excited state, as they relax new photons are emitted at a different energy some of which then reach our eyes. In that photon-surface-emission process energy changes hands. There is no way to “see” something without an interaction of a similar kind taking place and so energy levels being disturbed. The idea of putting energy back into the system to compensate for some kind of losses along the way would be so unsubtle as to bring with it its own disturbances. Again by analogy it would be like trying to hit the object with lower energy photons to compensate for the energy losses coincident with the absorption-emission process. How could such a process be timed precisely without first measuring the original effect so that the detector would “know” by how much to compensate the system being observed?
Wayne Smallman said,
August 29, 2007 at 11:06 pm
Yes, but why is an eye any different to the card from which the slits are cut into?
David Bradley said,
August 30, 2007 at 8:40 am
I’m going to defer to Howard Wiseman on this as I may have argued myself into a corner using the eye-light analogy. I’ll post back as soon as I’ve spoken to him and found out what he has to say.
Howard Wiseman said,
August 30, 2007 at 12:00 pm
First comment: the disturbance under question here all takes place *after* the wave has passed through the slits. The form of the wavefunction is not much affected by what the card is made of, how thick it is etc. It is very much affected by how wide the slits are and how far apart they are. But all of that is before the sensor that detects which slit the particle went through.
David Bradley is quite right: in general sensors *do* exchange energy with the system they sense. If you illuminate something with light, and it refelects that light, then both momentum and energy are exchanged. A sheet of card is (you could say) a sensor, but the particles it senses are just absorbed by it, and don’t play any part in the experiment. The particles that do play a part are those that pass through the slits. Everything interesting happens after the slits.
As for the question: can you make a sensor that doesn’t exchange energy with the particle, that is *exactly* what is special about the Scully-Englert-Walther scheme. What our experiment shows is that with a measurement scheme like theirs there is no change in the mean energy (as stated in David’s article above). The remarkable thing is that even though there is no net energy exchange, the momentum still gets disturbed by just enought to destroy the interference pattern. So you still can’t beat quantum mechanics: if you find out which slit the particle went through you necessarily lose the wave behaviour (the interference pattern).
Hope this clarifies things a bit.
David Bradley said,
August 30, 2007 at 3:02 pm
Thanks for the explanation, Howard. I’m sure readers will still have questions but I will endeavour to fend them off as best I can ;-)
Wayne Smallman said,
August 30, 2007 at 3:11 pm
Hi Howard, that helps tremendously.
Here’s another idea: is there a way to make the slits themselves the sensor?
If the slits are absorbing some of the flow of the particles, then surely if you’ve got a photosensitive material, wouldn’t that material show the variations in the volume of particles as they pas through each slit?
That way, you’re not messing around with the interesting stuff on the other side of the slits.
Howard Wiseman said,
August 30, 2007 at 10:27 pm
A slit is just a rectangular hole in a piece of card. So a slit is not made of anything. It doesn’t absorb anything. It lets particles through. The card around the slit absorbs the particles that don’t go through the slits, and they could be made of photographic film if you wanted. But that wouldn’t tell you anything about which slit the particle goes through if it *doesn’t* hit the card, but rather goes through the slits.
Wayne Smallman said,
August 31, 2007 at 10:12 am
But if you have something that acts as a wave, doesn’t that itself leave some kind of impression?
I know it’s a crude analogy, but much like during a morning stroll on the beach, the only proof you have of a the presence of water is the tide mark left behind.
So is it possible to detect the wave, even in the absence of the particle?
tony said,
August 31, 2007 at 5:02 pm
This is a little off topic. Reading this reminded me of an article i read in The Economist, around 1980-83. The article was about Quantum Physics. One experiment that was discussed has stuck with me. Here it is. I hope i remember it properly. Everything in our world is made up of atoms, electrons, protons etc. The scientists took one sheet of pure lithium. The sheet had one tiny hole in it. One electron was shot through the hole and two came out. From what i remember, the two were not simply one cut in half, but two complete electrons. This is one interesting theory came from the experiment. People who are schizophrenic see and hear things that are not there. These people are often treated with Lithium to control their condition; so they don’t hear or see these “imaginary” things. The theory that was postulated was, what if one of the electrons was from “our world/dimension” and the other was from a “dimension/world” the people with schizophrenia see. Are these people really imagining these things or are they real, only to be seen/heard by them? I hope i remembered this correctly, i hate to sound stupid. :) Any comments?
David Bradley said,
August 31, 2007 at 5:39 pm
Tony, it’s an interesting thought, but I think there is a world of difference between electrons being fired through a hole in a sheet of lithium metal and the behaviour of dissolved lithium ions in the human brain. If anything, wouldn’t you think that adding lithium to a schizophrenic brain would produce two signals from one and so actually make those double entry thought occur? Incidentally, I found nothing on this Googling and no entries for the terms “lithium schizophrenia quantum” on PubMed. Anyone else have references to support Tony’s suggestion.
Howard Wiseman said,
September 3, 2007 at 5:31 pm
To answer Wayne’s latest question, the short answer is no: in the absence of the particle the wave leaves no impression. That fails to capture all the subtlety of quantum physics, but basically if a detector *doesn’t* detect a particle, that means it doesn’t detect anything. It just the collapses the wave, to a new wave which has *zero amplitude* wherever the detector is. In the case of the slits, that means that if the particle is not absorbed by the card, the wavefunction is collapsed to a new wavefunction that is zero everywhere except where the slits are. This is the initial wavefunction that is assumed at the start of the experiment. If that seems weird, well, that’s quantum mechanics.
David Bradley said,
September 6, 2007 at 10:10 am
In related news - Physicists at the University of Michigan have coaxed two separate atoms to communicate with a sort of quantum intuition that Albert Einstein called “spooky.” In doing so, the researchers have made an advance toward super-fast quantum computing. The research could also be a building block for a quantum internet.
SOURCE: UMich via Newswise
Adolf Erdmann said,
September 19, 2007 at 3:55 am
My brother and I have recently conducted some experiments in optics which you might find interesting. We found that the speed of an observer in relation to the speed of light can be measured by what is commonly known by astronomers as “stellar drift”.
The principle of the experiment is very simple and works as follows: Imagine you had a large cube on a rocket ship. On each corner of the cube you placed a light source of identical wavelength. If the observer (observing device) is exactly in the middle, the angle between any two lights would be 90 degrees when the space ship is not in motion. Now let’s assume the spaceship is moving at a high speed, the angle between the two lights in the front will be greater than 90 degrees, because the observer has moved from the position at which time the light signals left their sources. On the other hand, the angle between the two light in the back is now less than 90 degrees because the observer has moved farther away from those light sources. Even Einstein mentioned in his famous train experiment that the observer located in the middle of the train will see the front lightening of the forked lightening strike sooner.
You may say that there is nothing new about that effect, and that it is well known and therefore does not need to be proven. You may also what useful purpose does it have? The answer to that is that NASA will be able to measure the speed of rockets in outer space. Since we do not have a rocket at our disposal, we have used the speed of the earth around the sun by aiming the device forward or backward, and by taking readings, several times a day. And since, the angle difference is very small, we have used Young’s pinhole experiment to measure the shift in angle. It may seem very simple, but there are other factors to consider; not only do the angles between the light sources change, but also the angles at which the waves approach the observer. To illustrate the latter, let’s assume you are at one side of a fast flowing river, and a boat is crossing the river from the other side to meet you, you will see that the boat is approaching you in a straight line, but its front is facing upstream rather than you.
We have observed that there is a definite difference in the observed angles at different times of the day, or when the instrument is turned. But there are also unexplained changes in the angle occurring frequently. So far, we have not been able to come up with an exact measurement of the earth motion. All we can say at this point is that the instrument shows that the earth moves.
Sincerely,
Adolf Erdmann
David Bradley said,
September 19, 2007 at 7:15 am
Adolf, I am sure there are readers who will be able to spot any flaws in your argument. Superficially, it sounds like a reasonable thought experiment, but have you considered Lorenz contraction factors in your calculations on your cube moving at “high speed”? I assume by “high speed”, you mean a velocity close to the speed of light.
db
Adolf Erdmann said,
September 19, 2007 at 3:39 pm
David,
The Lorentz contraction is only a theory as far as I know, there may acttually be more to it than that. At this time I cannot comment too much on the apparatus I described, since we have registered it recently with the Canadian Intellectual Property Office, and are waiting for confirmation. The instrument uses also a laser and beam splitter. If anyone is interested, we will share the construction plans once we have received the registration number.
On the subject of the experiment “Two Slits Are Better Than One” I like to say that it is not that controversial to me if we consider that an electron when moving creates a magnetic field around its path. Is it not possible that the interference wave pattern is produced somehow by the interference in the magnetic fields.
David Bradley said,
September 19, 2007 at 4:32 pm
Okay, we’ll wait to hear more from you once you’ve asserted your IP rights. As to the magnetic interference aspect of the double slit experiment, perhaps you could elaborate…
db
David Bradley said,
December 5, 2007 at 2:29 pm
Young’s experiment has now been carried out using a hydrogen molecule