Designing LN2 Manifold

Ken--

I thought others would have jumped on this already. Your post implies that your crystal is already capable of being cooled using LN₂. I did a little research before writing this reply. At 199j/g for vaporizing N₂, you are using about 2g per second. With a volume expansion of near 650:1 and a liquid density of about 0.8g/ml, you are creating about 1.7 liters of gaseous nitrogen per second. That is a pretty high flow rate, so I hope that your machined channels are capable of handling that volume and that your source of liquid is able to handle the back-pressure from the expanding gas in the manifold. Also, I would consider the possibility of noise caused by the exiting stream(s) of cold gas, as well as the residual cooling and possible ice formation they can cause. Perhaps you may need to blanket this manifold in such a way that it is protected from the moisture content of air.

If your manifold is operating closer to room temperature than the 77ºK cryogenic temperature of LN₂, then these concerns of ice formation and blanketing may be minor, but the volume expansion will still be significant and need to be accommodated.

Also, you may have to consider the impact this volume of gaseous N₂ has on the safety of the room's air for breathing. You could be dropping the O₂ percentage significantly, and the body may not sense this type of oxygen deprivation before unconsciousness occurs.

Back to your question on the post, I suspect that silver brazing will be better because it forms a stronger bond than soldering, and will be less susceptible to cracking from the temperature changes. Equally important, probably is the method of bonding your crystal to this manifold, or otherwise transferring the 400W of heat to it.

You are kind of sketchy on this side of your post, so I suspect the details are something you may want to keep proprietary. The closest I came to this in my work a generation ago was designing a LN₂ sourced cool stream for a university's X-ray diffraction crystallography research, for crystals unstable at room temperatures, and having to jacket the cool stream with a concentric room-temperature N₂ stream. For that work, the apparatus was designed using glass. The university's professional glassblower became very proficient at silvering the complex Dewar shapes we required. It worked quite well, but our volume flow rate was about 10x less than yours.

Good luck--John M.

John,

thank you for you insight, I was wondering if you are input on this question as well. I am designing a manifold (2 piece) to run the LN2 thru and in the first peice I will machining all the channels and the second piece I was panning to furnace braze on to cover the channels and make it one final part. what do you think about frurnace brazing or should I use a vacuum braze instead? this would go for both manifolds that are on either side of the crystal. Just to let you know that this whole system is vacuum sealed, even the holes outside the system are going to be vacuum tubing to cut down on icing. What is your bac ground in LN2 systems as well if you dont mind me asking.

Thank you

Ken Stebbins

Ken,

My father was a physical chemist and college professor. On days when he was working with LN₂ he brought the extra back in a stainless steel thermos bottle with a piece of aluminum foil over the cork. This was back in the 1950's, when that stuff was fairly exotic. He watched us carefully to ensure our safety from freeze burns, etc. The standard "gee-whiz" demonstrations were pretty impressive.

I haven't worked with it since 1971, when I was a graduate student in P-Chem. So, my hands-on experience is quite dated. However, the main concerns, such as differential rates of expansion for the components, ice buildup from the moisture content of the surrounding air, etc. are things that are hard to forget.

I read somewhere that semiconductors have problems with their band gaps, etc. and quit functioning if they are too cold. This may or may not be a worry for you. I have no background or experience with the different brazing techniques--sorry. Depending on the quality of vacuum you want to have, there may be some issues with the fluxes used and cleanliness of the finished components. If your crystal is bathed with the LN₂, then I foresee some problems with sealing it to the manifold. I'm sure you are familiar with the uses for MoS₂ as a sealing material on threaded connections in a vacuum.

I wish you well in your work. Don't break any crystals! I suspect they are expensive.

--John M.