How to realize the stern gerlach experiment ?

[zodian]

Hey,

Ok so basically what i want to do is to realize the stern gerlach experiment, I know it's going to be a lot of work , but I'm highly motivated... oh and btw I'm only 16 so I don't know everything about physics and I'm not a native speaker, so sorry for possible errors concerning my english :)

-So here I've got the first problem:
Should I use atoms or electrons? I don't have an atom furnace to accelerate those, but maybe it would be possible to build one ? Whereas electrons have charge and so the lorentz force would disturb the experiment... but what if I would create an E-field that cancels exactly the lorentz force? So what would you use?

-Second problem:
The magnet: Ok so all I know is that the magnetic field must be inhomogeneous, but how can I build such a magnet ?

I would be glad for every advice :)

 

[xts]

For my intuition the electrons seem to be much easier to deal with them than atoms (production, detection, beam intensity, collimation...) despite of Lorenz force which need to be compensated. But I am not quite sure here - maybe atoms are better?

The magnet is not a problem at all - you may just cut a ferrite or iron core to an appropriate shape.

But I see you forgot about one more problem - in my opinion - the most serious difficulty: the vacuum system.
Such experiment must be conducted in pretty high quality vacuum, which is rather unfeasible to achieve it in home-lab constraints.
The problems is not only to pump out the air from a vessel, but you'd have to use all materials vacuum friendly (not emmiting residual gases) - it is not that easy to make vacuum friendly electromagnet...

 

[Doc Al]

Should I use atoms or electrons?

Atoms. You'll have a harder time dealing with electrons.

I don't have an atom furnace to accelerate those, but maybe it would be possible to build one ?

Building a thermal oven is not so hard. Basically a metal box with cavity and a hole. Put the substance in the cavity, heat it up, and the atoms stream out of the hole.

The vacuum is the main problem that I see.

 

[xts]

Doc Al:
Atoms - I may aggree with you - I have no experience building such device in home conditions, but I still see some problems with atoms:
- detection. How to detect them - without waiting long enough to collect so many hits on a glass to be visible with a naked eye?
- deflection angle: magnetic moments of atoms and electrons are of the same order of magnitude, while electrons are tens thousand times lighter, so with electrons you may use weaker magnetic field, give them higher energies, and use smaller magnet to achieve the same deflection angle.
- sharp image: it is easier to give well defined initial speed/energy to electrons - you just put such and such voltage to electron cannon. Atoms from the oven have thermal continuum of speeds, so in the magnet they will be deflected under different angles - forming large spot rather than sharp dot. For electrons you may utilize a carefully designed electron cannon from a CRT tube.

 

[Doc Al]

It's certainly easy to create an electron beam, but good luck observing the splitting. As far as detecting the atomic beam, there are several ways to do so besides waiting for deposits to accumulate on a plate. Using a hot wire detector to ionize the beam is one.

While I admit that I've never done such an experiment at home, I've certainly done it plenty of times in the lab. (Long enough ago that the details have faded from memory, I'm afraid.) Turning the magnet on and off and seeing the splitting is one of the coolest things ever.

 

[zodian]

Ok, so before thinking about electrons or atoms, I should think about the vacuum.
Is it really impossible to make a good vacuum chamber in a home-lab? In this case, I could ask a teacher at my school, because I know that we've got one at school... but it wasn't very powerful, could you tell me approximately how powerful should such a vacuum be?

 

[xts]

Doc Al has much more experience with atomic experiments than I have - listen to his advices more than to mine ;)
For my experience - you should have a vacuum at least such quality that free way is significantly bigger than path your atoms must travel - let's say 1m sounds reasonable.
1m free path <=> 1mPa - is a minimum required vacuum quality. But maybe there are some issues I am not aware of, which may push this limit further...
Similar experiments, performed in univ labs by 1st year students usually utilize vacuum systems able to reach about 50 µPa.
Original experiment was performed at 1 µTorr = 130 µPa.

You may want to browse the description of a student's lab exercise - pretty well written, covers such issues as you asked about: vacuum, magnet shape, atomic oven, detector.
See: http://www.hep.wisc.edu/~prepost/407/stern-gerlach/stern.pdf

 

[zodian]

Hey guys :)
Ok so basically I can't find a good vacuum chamber :(
But does anyone know a formula to find the maximum pressure to avoid deflections of particles of let's say mass m and velocity v? If I knew such a formula, I could maybe design an experiment which would be possible to realize with my material

 

[_antunes]

Theoretically, if you enter your CRT monitor in a inhomogeneous magnetic field, you will see the effect. The vacuum exists in tube, electrons are randomly distributed and the detection is made on the screen.

 

[_antunes]

No, my intuition failed. Electrons not are electrically neutrals, like atoms. You need cancel the effect of Lorentz force, but i think it isn't so simple.

See Joachim Kessler's book Polarized Electrons (Springer, Heidelberg 1976).

 

[zodian]

That was actually my idea :)
Ok I know that creating an E-field that exactly cancels the Lorentz force is going to be very difficult, maybe even impossible, since the Lorentz force, created by the inhomogeneous magnetic field, is not going to be the same at every point of the field, but I guess that's going to be my challenge :)

 

[xts]

OK: I am going under Doc Al fire on my own will... (try to search web for electron based S-G experiments - there must be lots of issues and don't blame me for wrong advice - I never did it myself with electrons)

My first idea was to use CRT (or better - oscilloscope tube), open it under argonium atmosphere, install magnet inside, seal (weld) the tube, pump the gas out.
In order to get reasonable magnetic field gradient, the gap between poles must be small (single milimetres) - you won't get it without opening the tube.

You must calculate the bending for your chosen geometry of magnet: it may happen that you have not to compensate Lorentz force at all - you'd just get the image in a corner of the screen. If you have to compensate: my first guess is to use only magnets in the sequence: N/S(flat poles) S/N(asymmetric) N/S(flat).

But - honestly - I never did such experiment and I read no reports about such setup working. Make more research anmd don't blame me if my idea is totally wrong...

 

[zodian]

Hey :)
There's one thing I want to ask. I might be wrong, but if I've well understood this formula : F=(gradient)*(m*B) explains the Stern Gerlach experiment. So what I need is a gradient. Is gradient = inhomogeneous magnetic field ? If this is true, I just have to create a magnetic field which changes over time. If I had for example a magnetic field induced by a current, I could just connect the coil to an high-frequency AC current generator and I would get an inhomogeneous magnetic field; is that right?

 

[jtbell]

You need a field that is constant with respect to time but varies with respect to position.

 

[Vanadium 50]

You're focusing on the wrong part of the problem. The easiest part is to get the right field. The next easiest is to get the source. What's hard is the detector and the vacuum.

 

[zodian]

Hey Vanadium 50,
Well basically I've just received a cathode ray tube, so the vacuum and the detection should no longer be a problem, or at least I hope so ^^

 

[xts]

Vacuum still will be the major problem!
In order to install a magnet inside you must cut the tube.
You must do it in argonium atmosphere, then carefully seal the tube back and pump the gas out, and renew the getter.
Anyway - I am not quite sure if my idea about experiment with electrons was right. Have you checked if anybody made a DIY electron based Stern-Gerlach experiment and what issues you may encounter? Better listen to DocAl - he has more experience with such experiments than I have...