Jump to content

I want to create a 1 meter BEC


fredreload
 Share

Recommended Posts

13 hours ago, swansont said:

I was referring to the magnetic field for MRI, to point out that it’s not what you need for cooling.

For Rb? I expect the photon transition rate would be lower, and the photon momentum is smaller, as compared to laser cooling. I doubt you get significant cooling.

 

Right, it might take longer to cool down, but since there is nothing to heat up the rubidium(in a dark vacuum), it should gradually reach a minimum threshold, well in theory. I am curious about the laser frequency they use in the experiment to cool rubidium. Shouldn't it be somewhere in the radio frequency as well? Or does the laser encompass from infrared to radio wave = =?

https://en.wikipedia.org/wiki/Saturated_absorption_spectroscopy

Since the absorption spectrum for rubidium states 500MHz, which is the radio wave wavelength.

Link to comment
Share on other sites

30 minutes ago, fredreload said:

Right, it might take longer to cool down, but since there is nothing to heat up the rubidium(in a dark vacuum), it should gradually reach a minimum threshold, well in theory.

The transition that you use in MRI is not the same as you use in laser cooling. In MRI you cause Larmor precession at a frequency that's resonant because you have put on a magnetic field to match the frequency of the RF. The emission pattern of the radiation is laser cooling is isotropic, which is why the force from emission averages out to zero. As far as I know this is not the case for MRI interactions. If the incoming and outgoing photon are in opposite direction, so net momentum change occurs — the atom doesn't slow down.

Cooling and trapping requires that you scatter enough photons before the atom has passed through the interaction region. If you don't do this, the atom will not be trapped. If it passes through it then hots the vacuum chamber wall or, since the "dark vacuum" still has background gas, some other atom, and re-thermalizes.

Even if you managed to trap an atom or two, these collisions will liberate them from the trap.

 

Quote

I am curious about the laser frequency they use in the experiment to cool rubidium. Shouldn't it be somewhere in the radio frequency as well? Or does the laser encompass from infrared to radio wave = =?

https://en.wikipedia.org/wiki/Saturated_absorption_spectroscopy

Since the absorption spectrum for rubidium states 500MHz, which is the radio wave wavelength.

You use the optical frequency of the D2 transition (which minimizes the optical pumping I mentioned earlier) which for Rb is at ~780 nm

The 500 MHz they mention in the wikipedia article is ∆w, which is the Doppler-broadened width (FWHM = full width at half-maximum) of the transition, not the transition frequency. IOW, the laser has to be within about 500 MHz of the center to be resonant with one of the atoms in the thermal distribution (the Maxwell-Boltzmann distribution that is mentioned). They mention the natural line width of 6 MHz, which is true for any Rb atom at rest. That's how closely you have to tune the 780 nm light to be most likely to be absorbed by a particular atom.

 

 

Here is a trace from this paper

https://advancedlab.physics.gatech.edu/labs/SaturationSpectroscopy/SatSpecManual.pdf

It shows the saturated absorption signal of the D2 transitions in Rb for both isotopes (around 500 MHz FWHM) and the individual structure because they are doing Doppler-free saturated absorption, which can show the transitions for the atoms with v=0 in the ensemble. Those are nominally 6 MHz wide, but broadened by other effects, like laser power.

 

Screen Shot 2021-02-03 at 7.20.09 AM.png

Link to comment
Share on other sites

1 hour ago, swansont said:

The transition that you use in MRI is not the same as you use in laser cooling. In MRI you cause Larmor precession at a frequency that's resonant because you have put on a magnetic field to match the frequency of the RF. The emission pattern of the radiation is laser cooling is isotropic, which is why the force from emission averages out to zero. As far as I know this is not the case for MRI interactions. If the incoming and outgoing photon are in opposite direction, so net momentum change occurs — the atom doesn't slow down.

Cooling and trapping requires that you scatter enough photons before the atom has passed through the interaction region. If you don't do this, the atom will not be trapped. If it passes through it then hots the vacuum chamber wall or, since the "dark vacuum" still has background gas, some other atom, and re-thermalizes.

Even if you managed to trap an atom or two, these collisions will liberate them from the trap.

 

You use the optical frequency of the D2 transition (which minimizes the optical pumping I mentioned earlier) which for Rb is at ~780 nm

The 500 MHz they mention in the wikipedia article is ∆w, which is the Doppler-broadened width (FWHM = full width at half-maximum) of the transition, not the transition frequency. IOW, the laser has to be within about 500 MHz of the center to be resonant with one of the atoms in the thermal distribution (the Maxwell-Boltzmann distribution that is mentioned). They mention the natural line width of 6 MHz, which is true for any Rb atom at rest. That's how closely you have to tune the 780 nm light to be most likely to be absorbed by a particular atom.

 

 

Here is a trace from this paper

https://advancedlab.physics.gatech.edu/labs/SaturationSpectroscopy/SatSpecManual.pdf

It shows the saturated absorption signal of the D2 transitions in Rb for both isotopes (around 500 MHz FWHM) and the individual structure because they are doing Doppler-free saturated absorption, which can show the transitions for the atoms with v=0 in the ensemble. Those are nominally 6 MHz wide, but broadened by other effects, like laser power.

 

Screen Shot 2021-02-03 at 7.20.09 AM.png

Hmm, first of all, thanks for the explanation. I thought I can have the rubidium gas atoms flying around so it can be cooled by radio wave, but it seems a magnetic field line is needed to line up the molecules so they go into precession(rotation). This is in contrast to the laser cooling in which the molecules are "confined" in a region, to me it seems like the laser needs enough strength so that it produces a 500MHz frequency at the "center" of the magneto-trap. But since I have the molecules lined up with a magnetic field, all I need to worry about is the rotational frequency. As you can see below in the article that the water molecule rotational transition is at the microwave region 200 cm−1. But I do not know the rotational transition frequency for rubidium = =.

https://en.wikipedia.org/wiki/Electromagnetic_absorption_by_water

Of course I cannot think of a way to line up the rubidium gas atoms with the magnetic field line = =. So I will work on that(if someone knows feel free to let me know).

P.S. Many thanks for Dr. Swansnot's on the explanation on laser cooling. If I made a mistake I will go back and look into them

Link to comment
Share on other sites

15 minutes ago, dimreepr said:

You don't seem to understand that the explanation is, different wavelengths of the same field is at play, in the answer...

I don't get the magneto trap, it seems to trap(confined) the atoms rather then line them up with the magnetic field.

P.S. I'd build a mechanism to line up the atoms instead

Edited by fredreload
Link to comment
Share on other sites

6 minutes ago, fredreload said:

I don't get the magneto trap, it seems to trap(confined) the atoms rather then line them up with the magnetic field.

Trap and confine, are two very different words.

Link to comment
Share on other sites

58 minutes ago, fredreload said:

Hmm, first of all, thanks for the explanation. I thought I can have the rubidium gas atoms flying around so it can be cooled by radio wave, but it seems a magnetic field line is needed to line up the molecules so they go into precession(rotation).

Precession won't cool the atom.

58 minutes ago, fredreload said:

This is in contrast to the laser cooling in which the molecules are "confined" in a region,

Laser cooling by itself doesn't cause confinement.

58 minutes ago, fredreload said:

to me it seems like the laser needs enough strength so that it produces a 500MHz frequency at the "center" of the magneto-trap.

No, the 500 MHz is a feature of the atoms (owing to their innate width and thermal motion), not the laser.

The laser is at 780 nm, which is ~ 4 x 10^14 Hz

 

58 minutes ago, fredreload said:

But since I have the molecules lined up with a magnetic field, all I need to worry about is the rotational frequency. As you can see below in the article that the water molecule rotational transition is at the microwave region 200 cm−1. But I do not know the rotational transition frequency for rubidium = =.

Again: MRI and laser cooling are not the same thing. You can't mix and match them.

58 minutes ago, fredreload said:

Swansnot's 

You're still doing this?

47 minutes ago, fredreload said:

I don't get the magneto trap, it seems to trap(confined) the atoms rather then line them up with the magnetic field.

The confinement is from the light, but the presence of a particular magnetic field gradient (a quadrupole field) makes the force position-dependent, so the atoms will be at rest at the zero point of the magnetic field.

47 minutes ago, fredreload said:

P.S. I'd build a mechanism to line up the atoms instead

You can do that with the proper magnetic field gradient. It's a 2-D magneto-optic trap, sometimes called an atomic funnel.

Link to comment
Share on other sites

1 hour ago, swansont said:

Precession won't cool the atom.

Laser cooling by itself doesn't cause confinement.

No, the 500 MHz is a feature of the atoms (owing to their innate width and thermal motion), not the laser.

The laser is at 780 nm, which is ~ 4 x 10^14 Hz

 

Again: MRI and laser cooling are not the same thing. You can't mix and match them.

You're still doing this?

The confinement is from the light, but the presence of a particular magnetic field gradient (a quadrupole field) makes the force position-dependent, so the atoms will be at rest at the zero point of the magnetic field.

You can do that with the proper magnetic field gradient. It's a 2-D magneto-optic trap, sometimes called an atomic funnel.

No you are trying to cool down the precession with atom's rotational frequency absorption of electromagnetic radiation, IOW make the atom spin slower.

Laser cooling does not cause confinement, but your magneto-trap does, it exhibits zero magnetic field at the center, so the atoms are not lined up but confined. A lion in a cage can move around. A lion lined up in a cage cannot move around.

Ya well professor Swansnot @@.

Well the point a about a 2-D magneto-optic trap is I am gonna use a magnetic coils, not lasers, so it has a wider coverage. But ya maybe opposite polarity so they stay at the center, good funnel

Link to comment
Share on other sites

19 minutes ago, fredreload said:

No you are trying to cool down the precession with atom's rotational frequency absorption of electromagnetic radiation, IOW make the atom spin slower.

What does that have to do with its temperature? (which is CoM motion)

Quote

Laser cooling does not cause confinement, but your magneto-trap does, it exhibits zero magnetic field at the center, so the atoms are not lined up but confined. A lion in a cage can move around. A lion lined up in a cage cannot move around.

Depends on what you mean by "lined up"

There will always be motion, and some spatial extent. But if you want more confinement, you should use a linear ion trap, which has a stronger restoring force, coupled with laser cooling.

Quote

Ya well professor Swansnot @@.

At this point I have to wonder if you are doing this deliberately. (after it's been brought to your attention twice in the last few days)

 

Quote

Well the point a about a 2-D magneto-optic trap is I am gonna use a magnetic coils, not lasers, so it has a wider coverage. But ya maybe opposite polarity so they stay at the center, good funnel

A 2-D MOT uses both magnetic fields and lasers.

Magnetic traps are not very deep, so they will only confine atoms that are already cold. It would not have a "wider coverage"

Link to comment
Share on other sites

39 minutes ago, swansont said:

What does that have to do with its temperature? (which is CoM motion)

Well if the atoms are lined up with the magnetic field then it has no movement(unless it jiggles around, i dunno), then what is left is that the atom is spinning, if you reduce the spin then I assume you lower the temperature. That is the same concept as laser cooling, essentially you are putting the atoms to rest. I mean for laser cooling you also stop the spins right? Or just the movement?

P.S. I think the magneto trap does exert a magnetic field on the sample cloud, but the magnetic field strength would be much higher in this case.

39 minutes ago, swansont said:

A 2-D MOT uses both magnetic fields and lasers.

Magnetic traps are not very deep, so they will only confine atoms that are already cold. It would not have a "wider coverage"

I think I will go with attraction of two magnetic coils on each side to pull the rubidium atoms, well it depends on the tesla strength of the magnet(might have to do with the jiggling too). Which is really hard to increase for the setup.

P.S. I will have to check at which temperature the BEC formation begins(which is 300nk)

P.S. So I have been running the same thing, the magneto trap does exert a magnetic field on the sample, with a much higher field strength

Edited by fredreload
Link to comment
Share on other sites

1 hour ago, fredreload said:

Well if the atoms are lined up with the magnetic field

They aren't. The spin of the proton is aligned with the magnetic field in MRI - pointing in that dierection. That doesn't mean the atoms are lined up in a line.

 

1 hour ago, fredreload said:

then it has no movement(unless it jiggles around, i dunno), then what is left is that the atom is spinning, if you reduce the spin then I assume you lower the temperature.

Spin is quantized. You can't reduce the spin.

 

1 hour ago, fredreload said:

That is the same concept as laser cooling, essentially you are putting the atoms to rest. I mean for laser cooling you also stop the spins right? Or just the movement?

You reduce the CoM motion.

Spin is quantized. All you can do is change the orientation of the spin.

 

1 hour ago, fredreload said:

P.S. I think the magneto trap does exert a magnetic field on the sample cloud, but the magnetic field strength would be much higher in this case.

Yes, there is a field, but the force is optical, not magnetic. The function of the field is to Zeeman shift the resonance of the atom.

Without the magnetic field you have what is called optical molasses. Cold atoms where the lasers overlap, but not confined. You don't get as many atoms in a molasses, but you can get them colder by letting the cloud expand (in a BEC, the next step is a magnetic trap, and then evaporative cooling)

 

 

 

Link to comment
Share on other sites

2 hours ago, fredreload said:

I mean for laser cooling you also stop the spins right?

This is a prime example of why you should stop guessing and listen to some very knowledgeable person who has the patience to listen to you.

Alternatively you could actually read a textbook and find such things out for yourself.

1 hour ago, swansont said:

Spin is quantized. You can't reduce the spin.

+1 for your patience.

Link to comment
Share on other sites

1 hour ago, studiot said:

This is a prime example of why you should stop guessing and listen to some very knowledgeable person who has the patience to listen to you.

Alternatively you could actually read a textbook and find such things out for yourself.

+1 for your patience.

Too patient sometimes when people are just pulling stuff out of their derriere.  :)

Link to comment
Share on other sites

13 hours ago, StringJunky said:

Too patient sometimes when people are just pulling stuff out of their derriere.  :)

Just helping a troubling kid trying to pass the physics class lol. Hmm, in this case you need something with a magnetic moment on two sides. Ionic bond? Man you guys are crazy. Capacitor?

Edited by fredreload
Link to comment
Share on other sites

So to trap the sodium atoms. I would create 4 negative capacitor plates(top, down, left, right). That is because if you only have top, down capacitor plates the sodium atoms can still spin sideways. Now after the sodium atoms are situated, you use a radio wave to reduce the remaining CoM motion, thereby reducing the temperature.

Link to comment
Share on other sites

1 hour ago, fredreload said:

Just helping a troubling kid trying to pass the physics class lol. Hmm, in this case you need something with a magnetic moment on two sides. Ionic bond? Man you guys are crazy. Capacitor?

Guesswork is a very inefficient method of learning at the best of times. The less solid and more hocus pocus the background you base your guesses on the more inefficient it becomes.

It is no accident that the initials of Harry Potter are H.P.

Who is the kid and who is teaching him ?
Where do they teach magnetic cooling to kids in Physics ?

27 minutes ago, fredreload said:

So to trap the sodium atoms. I would create 4 negative capacitor plates(top, down, left, right). That is because if you only have top, down capacitor plates the sodium atoms can still spin sideways. Now after the sodium atoms are situated, you use a radio wave to reduce the remaining CoM motion, thereby reducing the temperature.

 

Electrostatic traps work on ions not atoms.

I wonder why ?

Edited by studiot
Link to comment
Share on other sites

17 minutes ago, studiot said:

Guesswork is a very inefficient method of learning at the best of times. The less solid and more hocus pocus the background you base your guesses on the more inefficient it becomes.

It is no accident that the initials of Harry Potter are H.P.

Who is the kid and who is teaching him ?
Where do they teach magnetic cooling to kids in Physics ?

 

Electrostatic traps work on ions not atoms.

I wonder why ?

Sodium atom are well, mostly ions cuz of salt you know @@(easily losing a valence electron in the outer shell). Now is just a matter if it can trap light(photons). Ionno what frequency range of light it traps and the temperature it needs to achieve sodium ions BEC.

Edited by fredreload
Link to comment
Share on other sites

47 minutes ago, fredreload said:

So to trap the sodium atoms. I would create 4 negative capacitor plates(top, down, left, right). That is because if you only have top, down capacitor plates the sodium atoms can still spin sideways.

How would this trap the atoms? I’m looking for a physics answer, not a WAG.

Put another way: provide a link to anyone who has trapped atoms like this.

47 minutes ago, fredreload said:

Now after the sodium atoms are situated, you use a radio wave to reduce the remaining CoM motion, thereby reducing the temperature.

I’ve told you this won’t work, and you haven’t detailed how it could work, so why are you asserting that it will? Physics is not magic. Wishing does not make it so. Invoking a few key words is not an incantation that will give you your desired result. 

Link to comment
Share on other sites

1 hour ago, swansont said:

How would this trap the atoms? I’m looking for a physics answer, not a WAG.

Put another way: provide a link to anyone who has trapped atoms like this.

I’ve told you this won’t work, and you haven’t detailed how it could work, so why are you asserting that it will? Physics is not magic. Wishing does not make it so. Invoking a few key words is not an incantation that will give you your desired result. 

A capacitor got a positive and a negative plate.

「capacitor schematicv」的圖片搜尋結果

I align 4 negative plate(top, down, left, right) with sodium atoms situated in the middle. Sodium atom is an ion with a +1 charge, so it will get pulled by the four capacitor plates.

875066901_.thumb.png.3ab00d1117c4b1e5c2ce8539a0068cdf.png

Thereby trapping the atoms at the center, if it still spins I will apply an extra radio wave to reduce the CoM movement.

P.S. The electric force is also much greater than the magnetic one if the magneto-optical trap is not doing this already.

Edited by fredreload
Link to comment
Share on other sites

This won’t work.

“Earnshaw's theorem states that a collection of point charges cannot be maintained in a stable stationary equilibrium configuration solely by the electrostatic interaction of the charges.”

https://en.wikipedia.org/wiki/Earnshaw's_theorem

Guesswork is not a replacement for knowledge (studiot expressed a similar sentiment earlier)

In this case, if the atom is not at rest at the exact center, it feels a force. And it will never be at rest. The atom will eventually leak out

Link to comment
Share on other sites

1 hour ago, swansont said:

This won’t work.

“Earnshaw's theorem states that a collection of point charges cannot be maintained in a stable stationary equilibrium configuration solely by the electrostatic interaction of the charges.”

https://en.wikipedia.org/wiki/Earnshaw's_theorem

Guesswork is not a replacement for knowledge (studiot expressed a similar sentiment earlier)

In this case, if the atom is not at rest at the exact center, it feels a force. And it will never be at rest. The atom will eventually leak out

You can't just say Earnshaw's theorem is not in favor of my electric field idea, but is in favor of the magneto-optic trap idea(which is kind of similar but using a magnetic field in this case).

Well what it does prove is that only the push force works but not the pull force = =, thanks for the correction, I will use the positive electric field plates in this case = =.

This is from Wikipedia about the Earnshaw's theorem:

"Informally, the case of a point charge in an arbitrary static electric field is a simple consequence of Gauss's law. For a particle to be in a stable equilibrium, small perturbations ("pushes") on the particle in any direction should not break the equilibrium; the particle should "fall back" to its previous position. This means that the force field lines around the particle's equilibrium position should all point inwards,"

Link to comment
Share on other sites

56 minutes ago, fredreload said:

You can't just say Earnshaw's theorem is not in favor of my electric field idea, but is in favor of the magneto-optic trap idea(which is kind of similar but using a magnetic field in this case).

Sure I can. Physics backs me up. 

“kinda similar” is not much of an argument; a quadrupole magnetic field + a laser is not “kinda similar” to the capacitor plates

 

56 minutes ago, fredreload said:

Well what it does prove is that only the push force works but not the pull force = =, thanks for the correction, I will use the positive electric field plates in this case = =.

Changing the sign does nothing. You still have an unstable equilibrium. If the atom is not at rest at the exact center, it feels a force. And it will never be at rest. In this case it will be attracted to a one of the plates.

 

56 minutes ago, fredreload said:

This is from Wikipedia about the Earnshaw's theorem:

"Informally, the case of a point charge in an arbitrary static electric field is a simple consequence of Gauss's law. For a particle to be in a stable equilibrium, small perturbations ("pushes") on the particle in any direction should not break the equilibrium; the particle should "fall back" to its previous position. This means that the force field lines around the particle's equilibrium position should all point inwards,"

And this is not the case for either of the electrostatic cases. You don’t have a stable equilibrium.

 

 

Link to comment
Share on other sites

2 minutes ago, swansont said:

Sure I can. Physics backs me up. 

“kinda similar” is not much of an argument; a quadrupole magnetic field + a laser is not “kinda similar” to the capacitor plates

 

Changing the sign does nothing. You still have an unstable equilibrium. If the atom is not at rest at the exact center, it feels a force. And it will never be at rest. In this case it will be attracted to a one of the plates.

 

And this is not the case for either of the electrostatic cases. You don’t have a stable equilibrium.

 

 

Kinda similar meaning the Earnshaw theorem also applies for your magneto-optic trap idea which is using magnetic field.

" It is usually referenced to magnetic fields, but was first applied to electrostatic fields. " Wikipedia

I changed it to a plus so the plates are now exerting a push force to the sodium ions(not an attraction force), which should keep the atoms in the center.

Link to comment
Share on other sites

13 minutes ago, fredreload said:

Kinda similar meaning the Earnshaw theorem also applies for your magneto-optic trap idea which is using magnetic field.

" It is usually referenced to magnetic fields, but was first applied to electrostatic fields. " Wikipedia

I changed it to a plus so the plates are now exerting a push force to the sodium ions(not an attraction force), which should keep the atoms in the center.

banghead.gif.68d156af3e1139bd90aaf91d59a3f2af.gif

Link to comment
Share on other sites

Guest
This topic is now closed to further replies.
 Share

×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.