# I want to create a 1 meter BEC

## Recommended Posts

18 minutes ago, studiot said:

Well, I am pretty much trying to recreate this thing @@, the magneto-optic trap (https://en.wikipedia.org/wiki/Magneto-optical_trap)

P.S. Tell me that is different from my setup

• Replies 203
• Created

#### Posted Images

Well, I am pretty much trying to recreate this thing @@, the magneto-optic trap (https://en.wikipedia.org/wiki/Magneto-optical_trap)

P.S. Tell me that is different from my setup

What you are doing reminds me of the man who said "I am trying too make a walnut into an oyster, both have a hard shell in two parts with something I want inside. Since they are 'kinda' the same I should be able to grow pearls inside a walnut"

Edited by studiot
##### Share on other sites

On 1/31/2021 at 2:15 PM, fredreload said:

I want to create a 1 meter sphere BEC(Bose Einstein Condensate) for slow light

Just curious; does it have to be a 100cm sphere? What experiment would require that size rather than something smaller and maybe more practical?

##### Share on other sites

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

magnetic trap and magneto-optic trap are two very different things (“optic” meaning photons are involved)

##### Share on other sites

33 minutes ago, swansont said:

magnetic trap and magneto-optic trap are two very different things (“optic” meaning photons are involved)

Ya, I see that the point of the magneto-optical trap is to have the gas ions absorb the photons but not to trap it in a magnetic field, you are right.

P.S. Well, how would you design a 1 meter BEC? With bigger lasers?

P.S. Bigger lasers sound good

##### Share on other sites

Ya, I see that the point of the magneto-optical trap is to have the gas ions absorb the photons but not to trap it in a magnetic field, you are right.

A MOT traps neutral atoms, not ions.

P.S. Well, how would you design a 1 meter BEC? With bigger lasers?

P.S. Bigger lasers sound good

That’s part of the problem. If you need a few mW/cm^2, and your optics are ~5 cm in diameter, that’s about 40 mW. Six beams, that’s 240 mW. Losses from acousto-optic modulators to tune the laser frequency is a factor of 2. Maybe another factor of 2 for other losses. Basically you need a watt for that.

Now you want to scale this up by a factor of 20, which means the area goes up by a factor of 400. You need a 400 watt laser. Huge problem #1

You need 1 meter windows for your vacuum system. Optical-quality flat. If you can find them, they would be super expensive. Problem #2

But it’s probably all for naught, because a 1 meter cloud of atoms will be optically thick, meaning the laser light won’t penetrate, and the re-radiation would cause heating. You would likely not be able to get a 1m cloud of cold atoms to begin the process of forming a BEC.

##### Share on other sites

1 hour ago, swansont said:

But it’s probably all for naught, because a 1 meter cloud of atoms will be optically thick, meaning the laser light won’t penetrate, and the re-radiation would cause heating. You would likely not be able to get a 1m cloud of cold atoms to begin the process of forming a BEC.

So basically, if you can start cooling the edges of the cloud, more laser light penetrates further into the cloud, but the further in, the less chance the re-emitted light has of escaping the cloud? The cooling effect occurs because of absorption, but it still requires the energy being put into the cloud to escape? The initial absorption reduces heat energy, but since you can't just increase the non-thermal energy of the atoms indefinitely, it's going to have to escape the cloud or end up increasing the heat.

Does this mean that the laser cooling process is effectively using light to "poke" the atoms just right so that they emit more light energy than they absorb, so that more light energy goes out of the cloud than what goes in?

PS. as a spectator up till now, I appreciate the patient informative responses. Replies that focus on discouraging amateurs don't help others who are reading.

Edited by md65536
##### Share on other sites

1 hour ago, swansont said:

A MOT traps neutral atoms, not ions.

That’s part of the problem. If you need a few mW/cm^2, and your optics are ~5 cm in diameter, that’s about 40 mW. Six beams, that’s 240 mW. Losses from acousto-optic modulators to tune the laser frequency is a factor of 2. Maybe another factor of 2 for other losses. Basically you need a watt for that.

Now you want to scale this up by a factor of 20, which means the area goes up by a factor of 400. You need a 400 watt laser. Huge problem #1

You need 1 meter windows for your vacuum system. Optical-quality flat. If you can find them, they would be super expensive. Problem #2

But it’s probably all for naught, because a 1 meter cloud of atoms will be optically thick, meaning the laser light won’t penetrate, and the re-radiation would cause heating. You would likely not be able to get a 1m cloud of cold atoms to begin the process of forming a BEC.

Hmm, I still think the radio wave technique is worth a shot before I switch to big lasers. Just 1 meter of gas chamber with 6 antennas emitting 500MHz radio wave. If you are worried that it would get re-thermalize than perhaps a magnetic field can be used to encage the gas ions.

##### Share on other sites

Meaning they accept and release a photon so that uses some energy of the laser I think.

Sure, but what is meant by "use"? The energy is not reduced through use.

Naively I can think of it like, photons are absorbed by atoms moving toward the laser, and the kinetic energy of the two partially cancel each other out. But really I think it's more like, an atom absorbs a photon with energy E in the atom's rest frame, changes speed in the process, then emits the same energy E (as one photon? or several over time?) in its new rest frame. In the lab frame, the energy absorbed is less (on average???) than the energy emitted due to the atom's change in speed. --- However, if the light is re-emitted in a random direction, I'm not sure this idea makes any sense.

The issue is, the real cooling corresponds with energy coming out of the cloud, not energy going in.

Edited by md65536
##### Share on other sites

7 minutes ago, md65536 said:

The issue is, the real cooling corresponds with the energy coming out of the cloud, not the energy going in.

Unless the cloud can convert KE to PE somehow, since temperature is a function of KE, not PE  don't you think ?

##### Share on other sites

2 minutes ago, studiot said:

Unless the cloud can convert KE to PE somehow, since temperature is a function of KE, not PE  don't you think ?

I assume it doesn't. If it did how would it do that? Atoms in a higher energy state? Even if it were possible, it couldn't be done forever. That would mean the more time spent cooling the cloud, the more energetic it would get.

##### Share on other sites

6 minutes ago, md65536 said:

I assume it doesn't. If it did how would it do that? Atoms in a higher energy state? Even if it were possible, it couldn't be done forever. That would mean the more time spent cooling the cloud, the more energetic it would get.

I have absolutely no idea how this would happen, but isn't that exactly what the word condense means?

Isn't that what happens when a gas condenses to a liquid or a liquid to a solid ?

A new form of PE is aquired - surface energy.

##### Share on other sites

Hmm, I still think the radio wave technique is worth a shot before I switch to big lasers. Just 1 meter of gas chamber with 6 antennas emitting 500MHz radio wave. If you are worried that it would get re-thermalize than perhaps a magnetic field can be used to encage the gas ions.

This reminds me of those folks who insist that their perpetual motion machine will work, and won’t listen to anyone who tells them different.

47 minutes ago, md65536 said:

Sure, but what is meant by "use"? The energy is not reduced through use.

Naively I can think of it like, photons are absorbed by atoms moving toward the laser, and the kinetic energy of the two partially cancel each other out. But really I think it's more like, an atom absorbs a photon with energy E, changes speed in the process, then emits the same energy E (as one photon? or several over time?). In the lab frame, the energy absorbed is less (on average???) than the energy emitted due to the atom's change in speed. --- However, if the light is re-emitted in a random direction, I'm not sure this idea makes any sense.

The issue is, the real cooling corresponds with energy coming out of the cloud, not energy going in.

Yes, the doppler shift will be changed slightly. That’s why you need millions of photon scatters to slow an atom to close to zero velocity - the imparted momentum from a single photon is small. p=E/c, and c is a big number.

1 hour ago, md65536 said:

So basically, if you can start cooling the edges of the cloud, more laser light penetrates further into the cloud, but the further in, the less chance the re-emitted light has of escaping the cloud? The cooling effect occurs because of absorption, but it still requires the energy being put into the cloud to escape? The initial absorption reduces heat energy, but since you can't just increase the non-thermal energy of the atoms indefinitely, it's going to have to escape the cloud or end up increasing the heat.

If the cloud is big, light won’t penetrate unless it’s being re-emitted, and that’s in a random direction. Light from random direction won’t cause slowing; there’s no net force being applied

1 hour ago, md65536 said:

Does this mean that the laser cooling process is effectively using light to "poke" the atoms just right so that they emit more light energy than they absorb, so that more light energy goes out of the cloud than what goes in?

It’s usually described in terms of the force, which is in the direction of the laser (looking at a 1-D example). But yes, the emitted light would be slightly more energetic.

1 hour ago, md65536 said:

PS. as a spectator up till now, I appreciate the patient informative responses. Replies that focus on discouraging amateurs don't help others who are reading.

##### Share on other sites

32 minutes ago, swansont said:

Yes, the doppler shift will be changed slightly. That’s why you need millions of photon scatters to slow an atom to close to zero velocity - the imparted momentum from a single photon is small. p=E/c, and c is a big number.

As a toy model just to describe where the energy's going, is the following a reasonable description?

Say you start with one million photons moving to the right, and one atom moving to the left.

After absorption and re-emitting, you end up with one million photons, scattered in a spherically uniform distribution of directions. The re-emitted photons would have variance in Doppler shift (because they're not necessarily immediately re-emitted?), but the photons re-emitted to the left would be blue-shifted on average, and the photons directly to the right would have no shift on average (basically, if a photon enters the atom from the left, and is re-emitted to the right, that photon hasn't changed the atom's final speed).

Also you end up with an atom with near zero velocity. The total energy of the system is the same. I would guess that nearly all of the kinetic energy of the atom at the start, goes into the scattered direction of the photons in the end, and a tiny portion of it goes into the Doppler shift of those photons?

Edited by md65536
##### Share on other sites

6 hours ago, studiot said:

I have absolutely no idea how this would happen, but isn't that exactly what the word condense means?

Isn't that what happens when a gas condenses to a liquid or a liquid to a solid ?

A new form of PE is aquired - surface energy.

It's a different state of matter, but the meaning (according to wikipedia) comes from, "Einstein proposed that cooling bosonic atoms to a very low temperature would cause them to fall (or "condense") into the lowest accessible quantum state". But even if it's analogous to condensing from gas to liquid to solid, those are typically going from higher energy states to lower energy states. Eg. water releases energy when it freezes, it doesn't absorb it. Anyway, a BEC involves atoms being in their lowest quantum state; they must get rid of that energy to form a BEC.

##### Share on other sites

6 hours ago, swansont said:

This reminds me of those folks who insist that their perpetual motion machine will work, and won’t listen to anyone who tells them different.

Just coming from the newbies point about view about the radio wave @@. If that does not work then might have to resort to the magnetron laser to get enough juice to power up the laser. But I'd give the radio wave method a shot before that happens

##### Share on other sites

8 hours ago, swansont said:

This reminds me of those folks who insist that their perpetual motion machine will work, and won’t listen to anyone who tells them different.

Well Swanson sir, what I don’t get is why do you need a nanometer laser to cool a gas atom that has an absorption spectrum of 500MHz which is radio wavelength? I might have missed something and I apologize if I do @@

On 2/3/2021 at 8:17 PM, swansont said:

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.

Not getting this part

See 500MHz is radio wave, did I miss something?

I see, so you are saying if I apply a 500MHz frequency it would subject to a lower frequency at the gas atom center.

##### Share on other sites

8 hours ago, md65536 said:

As a toy model just to describe where the energy's going, is the following a reasonable description?

Say you start with one million photons moving to the right, and one atom moving to the left.

After absorption and re-emitting, you end up with one million photons, scattered in a spherically uniform distribution of directions. The re-emitted photons would have variance in Doppler shift (because they're not necessarily immediately re-emitted?), but the photons re-emitted to the left would be blue-shifted on average, and the photons directly to the right would have no shift on average (basically, if a photon enters the atom from the left, and is re-emitted to the right, that photon hasn't changed the atom's final speed).

Right. The motion of the atom means it will absorb light at a lower frequency than its resonance frequency. For the emitted photons, for every redshifted photon you would get a blueshifted one, on average.

Well Swanson sir, what I don’t get is why do you need a nanometer laser to cool a gas atom that has an absorption spectrum of 500MHz which is radio wavelength? I might have missed something and I apologize if I do @@

The resonance is not at 500 MHz. I explained this. That’s how wide the resonance is - if you are within 500 MHz of the resonance, the photon has the highest chance of absorption.

##### Share on other sites

2 hours ago, swansont said:

The resonance is not at 500 MHz. I explained this. That’s how wide the resonance is - if you are within 500 MHz of the resonance, the photon has the highest chance of absorption.

So, when a 780nm laser hits a sodium gas item it would create a 500MHz resonance at the center of the atom @@? I am not getting the picture but I will try to draw it. You can correct it for me anyway you want

##### Share on other sites

So, when a 780nm laser hits a sodium gas item it would create a 500MHz resonance at the center of the atom @@? I am not getting the picture but I will try to draw it. You can correct it for me anyway you want

780 nm is the resonance for Rb.

The resonance frequency for Rb is ~3.85 x 10^14Hz. (that’s 780 nm) Because the atoms are moving, there is a Doppler shift, that resonance can be higher or lower, depending on that motion. The Doppler shift is up to 500 MHz for a significant fraction of the atoms*, so to be resonant with some of them, you need to be within 500 MHz of 3.85 x 10^14Hz.

*the atoms move with speeds that depend on temperature. It’s a distribution of speeds; some move fast, some slow.

##### Share on other sites

1 hour ago, swansont said:

you need to be within 500 MHz of 3.85 x 10^14Hz.

O, that makes sense @@, I guess we need a few big magnetron laser for this one if we want a 1 meter BEC

P.S. Thanks for clearing it up for me

##### Share on other sites

Not making a dent, am I?

##### Share on other sites

29 minutes ago, swansont said:

Not making a dent, am I?

No, gonna take a break and look into fusion/solar flare next

P.S. How much photons can the presumed laser cooled sodium ions store?

##### Share on other sites

P.S. How much photons can the presumed laser cooled sodium ions store?

You’re just trolling me now.

##### Share on other sites

6 minutes ago, swansont said:

You’re just trolling me now.

Time to knock it on the head.

##### Share on other sites

This topic is now closed to further replies.

×