Observation of dynamical Casimir effect

[The Geoff]

Plain english report here: http://www.sciencedaily.com/releases/20 ... 133050.htm
Original arxiv paper here: http://arxiv.org/pdf/1105.4714

In a nutshell, virtual photons have been coaxed into becoming real by oscillating a "mirror" (actually a SQUID) at relativistic speeds (25% c). Kinetic energy is converted into real particles, it's all very funky!

My question, and one that isn't obvious from either the Science Daily report or my limited understanding of the paper, is whether momentum is conserved - I would presume that the total momentum of the photon pair created equals zero, they both have equal momentum in opposite directions? Otherwise this could form the basis for a reactionless drive?

[Yakk]

Turning energy into momentum is easy: just shoot a laser.

It isn't efficient.

[Tass]

Indeed, any momentum of the photons would be taken from the mirror. Nothing strange in that, a flashlight will do the same. It isn't reactionless, the photons are the reaction mass

[The Geoff]

Yes, but the reaction mass has been created from energy, so in theory you could travel in any direction using light coming from any other direction/s? Like a solar sail, but without the "upwind/downwind" limitations? You could travel towards the source of energy. Take it to an extreme and you could use 360 degree solar panels (net incoming momentum=0) to produce massive particles, shoot them off in one direction and do away with the need to carry reaction mass?

It sounds a little too good to be true, which is why I suspect the total momentum of the photons must be zero.

[Yakk]

To soak up energy from (say) photons, you have to absorb their momentum.

To use photons to move yourself, you have to emit them (and their energy).

As photon momentum ~ energy (up to a constant and a unit conversion) (using ~ as "is proportional to"), no, you cannot move towards a source of photons then, emitting only energy you got from the incoming photons, accelerate towards the source. Regardless of using this method, or a laser.

[The Geoff]

Yeah, I see what you mean. What I had in mind is no different to using solar panels and a laser.

This experiment still works as a plain old Star Trek replicator though? (Well, presuming they can coax massive particles out of it)

[thoughtfully]

Yeah, I see what you mean. What I had in mind is no different to using solar panels and a laser.

This experiment still works as a plain old Star Trek replicator though? (Well, presuming they can coax massive particles out of it)

A source of protons, neutrons, and electrons is not the limiting factor in creating a replicator type device. There isn't a limitation on the supply, last I checked

[Sagekilla]

Yeah, I see what you mean. What I had in mind is no different to using solar panels and a laser.

This experiment still works as a plain old Star Trek replicator though? (Well, presuming they can coax massive particles out of it)

A source of protons, neutrons, and electrons is not the limiting factor in creating a replicator type device. There isn't a limitation on the supply, last I checked

To clarify this a bit: Can you imagine how you would order all your elements (assuming you have a special replicator device that can produce whole elements) in
such a fashion so that it would become say, a cheeseburger?

It's easy to lump a bunch of particles together. The hard part is forcing them to orient themselves in the manner that a hamburger might have it's particles oriented.
Can you easily write down in some systematic way where each particle is? I doubt it.

[yurell]

Yes, but the reaction mass has been created from energy, so in theory you could travel in any direction using light coming from any other direction/s? Like a solar sail, but without the "upwind/downwind" limitations? You could travel towards the source of energy. Take it to an extreme and you could use 360 degree solar panels (net incoming momentum=0) to produce massive particles, shoot them off in one direction and do away with the need to carry reaction mass?

It sounds a little too good to be true, which is why I suspect the total momentum of the photons must be zero.

A photon's momentum is p = ^E/_c, so if you want to increase your momentum by 1 kgms^-1, you need to pay 3E8 Joules. That becomes very expensive very quickly ... one Newton thrust is 3E8 Watts, roughly comparable to the peak power output of a Nimitz aircraft carrier (cite wikipedia).

[Soralin]

Yeah, I see what you mean. What I had in mind is no different to using solar panels and a laser.

This experiment still works as a plain old Star Trek replicator though? (Well, presuming they can coax massive particles out of it)

A source of protons, neutrons, and electrons is not the limiting factor in creating a replicator type device. There isn't a limitation on the supply, last I checked

Not to mention, if this process was 100% efficient, and you had 100% efficient matter->energy conversion, you'd still have to destroy a proton to get enough energy to produce a proton.

[The Geoff]

Can you easily write down in some systematic way where each particle is? I doubt it.

I'm a little uncertain - anyone got a Heisenberg Compensator handy?

Am I correct in thinking that this is essentially the opposite of a nuclear bomb: instead of converting mass into energy it could (theoretically) convert energy to mass? I'm aware that most methods of energy production (chemical, fusion etc) ultimately rely on E=mc^2, so using it to produce matter will ultimately have zero net effect, but there's a lot of "free" energy around in the form of light. So you could, in theory, have energy produced inside a star, pick it up light years away, and convert it back to matter instead of just exciting electrons as we do now?

[mfb]

I'm a little uncertain - anyone got a Heisenberg Compensator handy?

You don't need that. It is possible to create molecular bounds via moving individual atoms. And if that is not possible for some molecules, you always have the chemical way to do that. Removing surplus atoms elsewhere is easier then. And on a larger scale, you don't have to care about the uncertainty any more.
The "only" problem is that you have to arrange some 10^26 atoms for a single cheeseburger.

Converting energy into massive particles is done frequently at particle accelerators. However, entropy kills you if you try to convert starlight (~eV) into protons (~GeV). It is possible, as you can compensate that with a lot of heat radiation (~meV), but the efficiency is really crappy.

[Game_boy]

To soak up energy from (say) photons, you have to absorb their momentum.

To use photons to move yourself, you have to emit them (and their energy).

As photon momentum ~ energy (up to a constant and a unit conversion) (using ~ as "is proportional to"), no, you cannot move towards a source of photons then, emitting only energy you got from the incoming photons, accelerate towards the source. Regardless of using this method, or a laser.

http://arstechnica.com/science/news/201 ... -beams.ars

Or is that different?

[Yakk]

Slightly different. There, they are sort of tacking? Ie, a bunch of beams that aren't directly towards where they want to go.

It wouldn't work (unless there is a trick I can't think of) where all of the light comes directly in the direction you want to go.

With the light coming in at multiple different angles, the energy you can get can be greater than the momentum you lose in the direction you want to go. And turning (mass energy) into momentum is a solved problem.

[The Geoff]

The "only" problem is that you have to arrange some 10^26 atoms for a single cheeseburger.

If you were to suggest something as complex as my (fairly modest) computer to somebody 60 years ago when the transistor was invented then they'd probably say it was theoretically possible, but unlikely to be possible in the near future (decades). But, little bits at a time (no pun intended), we've found more efficient ways of making them. I would imagine any kind of replicator technology would follow a similar path; first you develop methods for making basic compounds like crystals, metals or amino acids, then use these templates to make more complicated structures like transistors or protein chains (I'll admit that protein folding problems may prove challenging), and so on, bootstrapping your way to complex creations.

Converting energy into massive particles is done frequently at particle accelerators. However, entropy kills you if you try to convert starlight (~eV) into protons (~GeV). It is possible, as you can compensate that with a lot of heat radiation (~meV), but the efficiency is really crappy.

Would this potentially provide a more targeted way of looking for particles than accelerators? Instead of piling energy into packets of protons (or whatever) and sifting the wreckage you focus it on a small "bit" of a vacuum and look at what pops out...eg, if you were looking for the Higgs you'd focus on 120GeV and see if anything pops up?

I realise I'm being wildly speculative here, feel free to tell me to go away if I'm stepping beyond the bounds of this bit of the forum

[Yakk]

So, in our universe, energy likes to take form. It isn't very picky about what form it takes.

If you make a hot spot, particles boil out of it. Most massive particles you can get is determined by how hot that spot is.

Accelerators are a way to generate a really hot spot. Out of it fountains a bunch of particles, which tend to immediately break down into other particles (because once they leave that hot spot, everything around them is really, really cold, and they "boil off" into other particles, sort of). We look at the tracks of the produced particles, and determine what kind of particle they boiled off of.

So what we do is we make a whole bunch of really hot spots, and look at the particles that come out of it, and add them to the particle zoo, and record their properties (derived usually from their decay products, which we find out about by surrounding the collision site with a whole bunch of detectors that track the cast-off particles as they fly through the detectors).

So yes, we can create matter from energy. This is, as noted, a solved problem. Generally there is plenty of matter of the more mundane sort around, and just grabbing it is much easier than creating more mundane matter. The hard part is arranging it, and it is harder to arrange the freshly created matter than it is to arrange humdrum everyday matter, so... The energy needed to create humdrum matter is really, really high -- instead of doing so, we just grab the humdrum matter that the universe already invested insane amounts of energy in and rearrange it.

[mfb]

The "only" problem is that you have to arrange some 10^26 atoms for a single cheeseburger.

If you were to suggest something as complex as my (fairly modest) computer to somebody 60 years ago when the transistor was invented then they'd probably say it was theoretically possible, but unlikely to be possible in the near future (decades). But, little bits at a time (no pun intended), we've found more efficient ways of making them. I would imagine any kind of replicator technology would follow a similar path; first you develop methods for making basic compounds like crystals, metals or amino acids, then use these templates to make more complicated structures like transistors or protein chains (I'll admit that protein folding problems may prove challenging), and so on, bootstrapping your way to complex creations.

That could be possible in some decades, indeed. We will see...

Would this potentially provide a more targeted way of looking for particles than accelerators? Instead of piling energy into packets of protons (or whatever) and sifting the wreckage you focus it on a small "bit" of a vacuum and look at what pops out...eg, if you were looking for the Higgs you'd focus on 120GeV and see if anything pops up?

If it would be possible to set the energy scale high enough - which is quite problematic. Between GHz photons and massive particles (not counting neutrinos), there are really many orders of magnitude.

[The Geoff]

Accelerators are a way to generate a really hot spot. Out of it fountains a bunch of particles, which tend to immediately break down into other particles (because once they leave that hot spot, everything around them is really, really cold, and they "boil off" into other particles, sort of). We look at the tracks of the produced particles, and determine what kind of particle they boiled off of.

I was more curious as to whether this would be a more efficient way of doing it. In accelerators there's a lot produced that we don't really care about - quarks, muons, electrons, neutrinos - we're really just smashing things to pieces and looking through an extraordinary amount of wreckage. One of the major problems in constructing something like the LHC is simply dealing with the sheer volume of data. I was wondering if this is a better approach, tweezers compared to a bulldozer if you want a bad analogy.

[Mr_Rose]

If you can tune it to produce lots of a particular particle, sure.
Would be interesting to see if you could even get the produced atoms to emerge in phase like in a laser.

[thoughtfully]

I was more curious as to whether this would be a more efficient way of doing it. In accelerators there's a lot produced that we don't really care about - quarks, muons, electrons, neutrinos - we're really just smashing things to pieces and looking through an extraordinary amount of wreckage. One of the major problems in constructing something like the LHC is simply dealing with the sheer volume of data. I was wondering if this is a better approach, tweezers compared to a bulldozer if you want a bad analogy.

The trouble with the LHC is that it's a hadron collider. That is, it collides big gooey balls of quarks, antiquarks, and gluons. A proton might have a hundred^* (the number depends on what the energy scale is) such particles (some virtual) buzzing around inside it at any given time. Electron colliders are much cleaner, but due to their low mass, they lose significant amounts of energy when moved in a curved path. Hence the preference for linear electron colliders (although LHC's predecessor, LEP, was an electron-positron collider, using the same circular tunnel). A large linear collider to succeed SLAC is in planning stages, but the really cool idea for new adventures in particle physics is a muon synchrotron.

Muons are basically really heavy electrons, about 200 times the mass of an electron, so the synchrotron radiation is much less of a concern. They are one of those "hot" particles that likes to decay quickly into lighter particles, but when they're moving quickly, their lifetimes are relativistically dilated, making them not so hard (but still tricky) to work with, especially since particle physics already works on extremely compressed timescales.

Thread split, anyone?

^* I think I got this number from an artistic rendering, so it might not be accurate. Any real particle physicists have a better number?