Jump to content

Molecule Diffraction Setup


Enthalpy

Recommended Posts

Hello you all!

Experiments are built to observe the wave nature of ever heavier objects. It was done around 1999 with C60 fullerene at the Vienna Center for Quantum Science and Technology, short description there
www.univie.ac.at/qfp/research/matterwave/c60/
from which this sketch is adapted:
Setup.png.8c790f1606b89c5b862686690f338f44.png

I resembles what one expects from a diffraction setup, but the wavelength differs from optics, hence so do the distances, grating's period, fringes separation. The light beam is strong and concentrated to ionize the molecules on its path, and a detector senses the charges, around 1 to 100 per second.

The description doesn't detail all subtleties. Science has progressed meanwhile, as interferences were obtained with heavier objects like proteins, but the experiment with fullerenes remains admirable.

====================

Maybe some improvements can be brought to similar experiments ?

At least on the sketch, the oven resembles a barrel with a hole. A true and good nozzle would expand and cool the fullerene better, to obtain a narrower velocity distribution. Authentic De Laval, at least down to a pressure where the mean free path is smaller than the divergent. The narrow throat may need special fabrication. Seek a high pressure ratio, by a big oven pressure if needed. Try optionally to mimic the gas temperature at the nozzle's walls.

Heat the fullerene after it sublimates, before the expansion. This shall limit its condensation during the expansion.

Add a gas to the fullerene in the oven to make the expansion more efficient? Argon, methane… Problem: I don't know how to remove the gas from the beam. Maybe the detector can discriminate the fullerene from the gas?

Add mechanical choppers on the path, especially near the collimation slits, to keep only the fullerene molecules with nearly the mean velocity. You lose some beam intensity but improve the diffraction pattern.

Use a mechanical speed to impart the 200m/s or more. Up to 500m/s are easily accessible to a rotating disk of metal, more with carbon fibres. Heat the fullerene very little, just enough for slow sublimation at the rotating part. The emission is more in the plane, so less heat achieves the same beam intensity, and the speed is more uniform.

Use microphones as detectors. The kinetic energy is 6kT at 300K, so cold microphones are better. With a piezoelectric or piezoresistive material, with micromachined silicon or with electrets, you can cover an area with microphones, so the experiment is faster than by scanning the diffraction pattern area. I feel microphones easier than the power laser too.

Marc Schaefer, aka Enthalpy

 

Link to comment
Share on other sites

Here's how a mechanical chopper could look like to pick only the molecules within a narrow speed distribution.

In optics, similar choopers are often pairs of teethed disks, 50% solid, with the proper distance and relative phase. Here I suggest instead a helix (and didn't check if I'm the first, of course). It can be long for increased speed selectivity without letting other speed domains pass through. The helix can be thin to let most molecules with adequate speed pass through. Just 5mm groove width and 200mm helix length would leave +-2.5% speed tolerance, and the corresponding fraction of the beam intensity, sure.

Chopper.png.9c17a83f1d82153174d5569ef277f65e.png

Alloys achieve peripheral speeds like 400 to >600m/s, and turbine superalloy can be baked for vacuum operation. The design must prevent dynamic flexural instability. Some ceramic bearings can run in vacuum, and magnetic bearings of course. The peripheral speed, angle and number of threads let accommodate varied molecule speeds. A high peripheral speed lets evacuate to the sides the molecules with bad speed.

After short thinking, building the chopper between the slits seems feasible. If not, it can fit before the slits.

Marc Schaefer, aka Enthalpy

Link to comment
Share on other sites

14 hours ago, Enthalpy said:

Here I suggest instead a helix (and didn't check if I'm the first, of course).

You aren't.

https://eprints.soton.ac.uk/180595/1/GetPDFServlet%3Ffiletype%3Dpdf%26id%3DRSINAK000081000010106107000001%26idtype%3Dcvips%26doi%3D10.1063%252F1.3499254%26prog%3Dnormal

I'm a little surprised that they didn't use some sort of velocity selecting filter

Link to comment
Share on other sites

19 hours ago, Enthalpy said:

Hello you all!

Experiments are built to observe the wave nature of ever heavier objects. It was done around 1999 with C60 fullerene at the Vienna Center for Quantum Science and Technology, short description there
www.univie.ac.at/qfp/research/matterwave/c60/
from which this sketch is adapted:
Setup.png.8c790f1606b89c5b862686690f338f44.png

It was done for atoms and dimer molecules a few years before that.

19 hours ago, Enthalpy said:

I resembles what one expects from a diffraction setup, but the wavelength differs from optics, hence so do the distances, grating's period, fringes separation. The light beam is strong and concentrated to ionize the molecules on its path, and a detector senses the charges, around 1 to 100 per second.

The description doesn't detail all subtleties. Science has progressed meanwhile, as interferences were obtained with heavier objects like proteins, but the experiment with fullerenes remains admirable.

====================

Maybe some improvements can be brought to similar experiments ?

At least on the sketch, the oven resembles a barrel with a hole. A true and good nozzle would expand and cool the fullerene better, to obtain a narrower velocity distribution. Authentic De Laval, at least down to a pressure where the mean free path is smaller than the divergent. The narrow throat may need special fabrication. Seek a high pressure ratio, by a big oven pressure if needed. Try optionally to mimic the gas temperature at the nozzle's walls.

You probably don't want to expand the fullerene gas. That would reduce the signal, since the gratings are of finite size.

19 hours ago, Enthalpy said:

 

Use a mechanical speed to impart the 200m/s or more. Up to 500m/s are easily accessible to a rotating disk of metal, more with carbon fibres. Heat the fullerene very little, just enough for slow sublimation at the rotating part.

Slow sublimation probably means a vastly reduced signal. Heating a material in a molecular beam is done to get sufficient flux. The vapor pressure rises by more than two orders of magnitude by heating it to 900K

https://www.sesres.com/physical-properties/

19 hours ago, Enthalpy said:

The emission is more in the plane, so less heat achieves the same beam intensity, and the speed is more uniform.

Any justification to make one think this is true? Sublimation temperature is 800K. How much of a beam flux would one expect below that?

19 hours ago, Enthalpy said:

Use microphones as detectors. The kinetic energy is 6kT at 300K,

6 kT?

19 hours ago, Enthalpy said:

so cold microphones are better. With a piezoelectric or piezoresistive material, with micromachined silicon or with electrets, you can cover an area with microphones, so the experiment is faster than by scanning the diffraction pattern area. I feel microphones easier than the power laser too.

The fringe separation of the diffraction pattern is tens of microns. So, even if this worked, your microphone needs to be a few microns wide, in order to resolve the fringes, and several mm tall in order to maximize signal. But how would you keep the molecules from adsorbing on to your microphone, or the ones that bounce off from scattering off the beam and disrupting the signal?

Link to comment
Share on other sites

On 4/8/2018 at 1:57 PM, John Cuthber said:

You aren't [the first to propose a helix]

https://eprints.soton.ac.uk/180595/1/GetPDFServlet%3Ffiletype%3Dpdf%26id%3DRSINAK000081000010106107000001%26idtype%3Dcvips%26doi%3D10.1063%252F1.3499254%26prog%3Dnormal

I'm a little surprised that they didn't use some sort of velocity selecting filter

Thank you JC!

I was surprised too that they used no velocity filter, that's why I proposed one. But the idea is immediate enough, sure.

Possibly the group in Vienna hoped only to see the first sidelobes, and for that the unfiltered speed distribution is good enough.

On 4/8/2018 at 2:30 PM, swansont said:

It was done for atoms and dimer molecules a few years before that.

You probably don't want to expand the fullerene gas. That would reduce the signal, since the gratings are of finite size.

Slow sublimation probably means a vastly reduced signal. Heating a material in a molecular beam is done to get sufficient flux. The vapor pressure rises by more than two orders of magnitude by heating it to 900K

https://www.sesres.com/physical-properties/

Any justification to make one think this is true? Sublimation temperature is 800K. How much of a beam flux would one expect below that?

6 kT?

The fringe separation of the diffraction pattern is tens of microns. So, even if this worked, your microphone needs to be a few microns wide, in order to resolve the fringes, and several mm tall in order to maximize signal. But how would you keep the molecules from adsorbing on to your microphone, or the ones that bounce off from scattering off the beam and disrupting the signal?

Hi Swansont, thanks for your interest!

The essential factor the determines the number of molecules observed at the detector is the angle filter composed of both 10µm slits at 1m interval (+-10µm, not +-20µm, my bad). If a means lets reduce the emission angle, like a nozzle or a moving emission surface, a bigger proportion of the molecules have the proper direction, so fewer can be emitted, at a lower temperature - by how much remains to be checked.

A nozzle would reduce much the temperature of the emitted molecules, both as a lengthwise random speed (which influences the fringe quality) and as a transverse speed (which determines what proportion of the flux can be kept). But it does increase the width of the emitting source, yes.

A paper by chemists on fullerene properties:
http://www.dtic.mil/get-tr-doc/pdf?AD=ADA292726

Adsorption on the microscopic microphones isn't a worry to my eyes, with 103 molecules impinging per second. The molecules can just stay where they arrived. Collisions are so improbable that they're negligible.

6kT (Boltzmann's constant) is just to compare with the achievable thermal noise at the microphone. kT can be a noise energy at a microphone; some designs put this energy outside the measurement frequency band. So comparing the expected event energy with kT gives a first idea if the signal can be discriminated from the noise.

Link to comment
Share on other sites

1 hour ago, Enthalpy said:

  Hi Swansont, thanks for your interest!

My graduate thesis was on an atom source for an atom interferometer, so this is familiar to me.

1 hour ago, Enthalpy said:

The essential factor the determines the number of molecules observed at the detector is the angle filter composed of both 10µm slits at 1m interval (+-10µm, not +-20µm, my bad). If a means lets reduce the emission angle, like a nozzle or a moving emission surface, a bigger proportion of the molecules have the proper direction, so fewer can be emitted, at a lower temperature - by how much remains to be checked.

The beam flux is heavily dependent on the vapor pressure of the material, which depends on temperature.

1 hour ago, Enthalpy said:

A nozzle would reduce much the temperature of the emitted molecules, both as a lengthwise random speed (which influences the fringe quality) and as a transverse speed (which determines what proportion of the flux can be kept). But it does increase the width of the emitting source, yes.

You also have to worry about the nozzle clogging up. Longer tubes will give better collimation, but also tend to be more prone to clogging. And clearing that means opening up the vacuum system. You tend to try and minimize that.

1 hour ago, Enthalpy said:

 Adsorption on the microscopic microphones isn't a worry to my eyes, with 103 molecules impinging per second. The molecules can just stay where they arrived. Collisions are so improbable that they're negligible.

But then you change the response of the microphone as the molecules build up. These are about 1 nm in diameter, so 1000 of them is 1 micron by 1 micron. That's a monolayer after only a few seconds of detection if they stick, for a microphone that's a few microns in diameter. And if they bounce, I don't see how you can claim that collisions would be improbable.

1 hour ago, Enthalpy said:

6kT (Boltzmann's constant) is just to compare with the achievable thermal noise at the microphone. kT can be a noise energy at a microphone; some designs put this energy outside the measurement frequency band. So comparing the expected event energy with kT gives a first idea if the signal can be discriminated from the noise.

I'd be worried about the sensitivity of the response of the microphone. If one can make a micriophone that small.

Link to comment
Share on other sites

On 4/11/2018 at 8:12 PM, swansont said:

But then you change the response of the microphone as the molecules build up. These are about 1 nm in diameter, so 1000 of them is 1 micron by 1 micron. That's a monolayer after only a few seconds of detection if they stick, for a microphone that's a few microns in diameter. And if they bounce, I don't see how you can claim that collisions would be improbable.

The experiment in Vienna had bins some 2µm wide and 10mm high which got around 1000 hits of 1nm molecules. This makes around 5*10-8 molecular layer. Even over many runs, there is margin.

I'm confident that 100 molecules per second, each filling 5*10-11 of the target area, don't collide.

On 4/11/2018 at 8:12 PM, swansont said:

I'd be worried about the sensitivity of the response of the microphone. If one can make a microphone that small.

I made microscopic accelerometers during the paleomonolithic era, and we got good sensitivities. The noise energy drops with the size, and achieving kT is natural for small sensors. Confining the kT noise energy outside the measurement frequency band needs engineering and thinking. Converting the movements of the sensing element into a signal without added noise tends to be feasible. That's why I suggested microphones. But it needs more detailed thoughts with figures, sure.

I have to think more at the rest.

Edited by Enthalpy
Link to comment
Share on other sites

1 hour ago, Enthalpy said:

The experiment in Vienna had bins some 2µm wide and 10mm high which got around 1000 hits of 1nm molecules. This makes around 5*10-8 molecular layer. Even over many runs, there is margin.

I'm confident that 100 molecules per second, each filling 5*10-11 of the target area, don't collide.

then you need a microphone of that size.

Quote

I made microscopic accelerometers during the paleomonolithic era, and we got good sensitivities.

How good, and how small were they?

And what is the "paleomonolithic" era? You are basically the only person on the internet using that term, according to Google.

Link to comment
Share on other sites

  • 1 month later...

Nozzles are long known as sources of molecular beams, and moving sources too are known. As one example, the 1969 PhD thesis by Les Sterling Sheffield:
https://oaktrust.library.tamu.edu/handle/1969.1/154986

 

On 4/15/2018 at 1:07 AM, swansont said:

then you need a microphone of that size. [2µm * 10mm]

How good, and how small were they?

The noise was around kT, but distributed mainly around the resonance frequency, leaving less noise in the useful band. Typical dimensions then were squares like 100µm wide, and reducing to 2µm would be easy now, at least for the lithography. The 10mm height would be covered by several (many) microphones probably.

Link to comment
Share on other sites

1 hour ago, Enthalpy said:

 The noise was around kT, but distributed mainly around the resonance frequency, leaving less noise in the useful band. Typical dimensions then were squares like 100µm wide, and reducing to 2µm would be easy now, at least for the lithography. The 10mm height would be covered by several (many) microphones probably.

What about sensitivity?

 

Edited by swansont
Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • 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.