Making an aluminum cold finger with LN2

[mjjaques]

Hi, I've looked everywhere for some advice on this, I hope someone can help. We are making a cooling system for cooling thin films, and part of that setup is a liquid nitrogen-cooled cold finger. This cold finger will be inside an aluminum tube, and will move closer and farther away from the thin film, which is mounted on the tube.

We could buy a cold finger, but they are all Pyrex or glass, which could scratch, and they mostly have inconvenient shapes. If we could make our own out of aluminum (because it needs to be nonmagnetic and conduct heat), that should work great.

Does anybody have design ideas for making this? The basic idea is just to circulate the liquid nitrogen in and out of the cold finger, but we're not sure how to seal the aluminum and have the in/out tubes fit properly. Thanks!

 

[brewnog]

Why would Pyrex scratch where aluminium wouldn't?

 

[mjjaques]

Well, Pyrex scratching on aluminum could break the Pyrex, right? Whereas it doesn't really matter if an inner aluminum rod gets scratched from the outer one. I could be completely wrong, but our professor doesn't like the idea of having it made of Pyrex for some reason...

 

[Q_Goest]

I suspect your professor is concerned about the Pyrex getting extremely brittle and possibly cracking. I've never heard of Pyrex being used at cryogenic temperature and I don't think I've ever seen material properties at low temp, so I would agree with your professor that Pyrex probably isn't a good material.

Is your cold finger going to look like the one on the left in this picture?
http://upload.wikimedia.org/wikipedia/commons/a/a2/Cold_fingers.svg

You could certainly use aluminum which has very good cryogenic properties, but I would suggest using copper tube/pipe. Copper also remains very ductile at low temperature and there are lots of fittings that can be used to make the assembly.

 

[mjjaques]

Yeah, it will look like the one on the left. We don't want to use copper, because we are doing this under a magnetic field, and aluminum will interfere with the results less. The problem we're having is how to make those "nozzle" things that the tubes (with liquid nitrogen) attach to. If we could even just find a place to buy aluminum pipe with those on it, that would also work great. But it's not looking like we can find that.

 

[Q_Goest]

Yeah, it will look like the one on the left. We don't want to use copper, because we are doing this under a magnetic field, and aluminum will interfere with the results less.

Both copper and aluminum are non-magnetic and used in MRI machines for example where they have very strong magnetic fields. I don't understand why copper wouldn't work for you, but certainly you can use aluminum.

The problem we're having is how to make those "nozzle" things that the tubes (with liquid nitrogen) attach to. If we could even just find a place to buy aluminum pipe with those on it, that would also work great. But it's not looking like we can find that.

If you're using pipe, you can use NPT threaded connections. I'd recommend using tube however, along with compression fittings such as Swagelok or equivalent which are commonly used in cryogenic applications.

 

[mjjaques]

Well, we would want to use pipe for the actual cold finger because it's solid, and then tubing for the liquid nitrogen to go in and out.

What about a hose barb like this? http://www.westmarine.com/webapp/wcs/stores/servlet/Product_11151_10001_27772_-1?cid=chanintel_google&ci_src=14110944&ci_sku=27772#.T-sPwrU7X4s

Would it work to screw one side into the pipe, with the tubing attached to the barbed side?

 

[M Quack]

Copper is routinely used on the sample holders of superconducting magnets with fields of 10T and more and temperatures down to 2K (below liquide He). Al and stainless steel are commonly used for baths and so on. Stainless has very bad thermal conductivity, so it is always used to mount things to the (warm) outside.

Don't use Cu or Al if you plan to use pulsed magnetic fields, as you will get lots of eddy currents.

I agree with Goest that your best bet is probably to use Swagelok parts. On the LN2 supply side you need some kind of flow control, e.g. a needle valve. If you can open/close that in a controlled way, that is best way to regulate the temperature, together with a heating element on the Cu sample holder.

For the "nozzle" you probably need some kind of capillary coiled up over some length to create an impedance. On the exit side you probably want a small vacuum pump.

 

[Q_Goest]

What about a hose barb like this? http://www.westmarine.com/webapp/wcs/stores/servlet/Product_11151_10001_27772_-1?cid=chanintel_google&ci_src=14110944&ci_sku=27772#.T-sPwrU7X4s

Would it work to screw one side into the pipe, with the tubing attached to the barbed side?

You can't connect aluminum tubing to the barbed end. The barbed end is for plastic or rubber tubing, most of which will crack and leak at low temperature. You might try some different plastics if you need flexibility. I would try Teflon or PFA tubing first if that's the case. Before you pass pressurized liquid nitrogen through them, try dunking a sample in LN2 to see how brittle it becomes.

The same kind of fitting can be obtained from Swagelok and others. You can get them with a male or female pipe thread on one side and a compression end on the other to connect to tubing.

Also, tubing can be just as thick and even thicker than pipe, so you could use tubing for the entire thing.

 

[aquinn@UB]

I also work in a thin films lab. We have several procedures that involve Cyrogenic sample handling For TOF-SIMS and XPS analysis.

first, Everything we use is under a vacuum with a backfill supply of Dry N2 gas to avoid condensation build up. you can cause a lot of damage in a lab if you start building up ice on the equipment. It leads to contamination issues and it just messy to be exposing anything as cold as LN2 to atmospheric gases.



In order to provide enough strength to keep a vacuum and fit all the other requirements you listed it sounds like you need to use stainless steel. Everything that is part of the cooling system should be copper. Aluminum oxidizes faster and will fatigue and break under the extreme cooling/ cycles.

Cold fingers in industrial instrumentation are always copper.

go to companies like Physical Electronics and ION-TOF and look at cryogenic sample handling
for ideas.

Kurt J. Lesker and UHV suppliers should have everything you are looking for to