Science desk
< June 11	<< May \| June \| Jul >>	June 13 >

Welcome to the Wikipedia Science Reference Desk Archives
The page you are currently viewing is an archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.

June 12

Gas centrifuge connections

Reading about the Gas centrifuge which is used to separate heavier and lighter isotopes of some gas, I understand how an electric motor can spin a tall cylinder in an evacuated outer container at high speed, like a thousand revolutionsmper second or more, with magnetic bearings rather than mechanical ball bearings which would quickly fail. But the relevant articles say that each cylinder had an inlet tube and two outlet tubes. I cannot envision a gas-tight connection which connects a rapidly rotating cylinder to a fixed gas pipe. A gas tight connection for general normal use might have a rubber o ring, or metal against metal with grease to prevent gas loss. These would fail in the first second at over 1000 revs per second. Metal on metal would melt. So how is the connection achieved, in general? Edison (talk) 01:00, 12 June 2018 (UTC)[reply]

Don't envision extracting gas radially outward from a cylinder during centrifuge; think more about a centrifugal pump. The entire outflow product is then cascaded, first through a diffusion barrier, and then through multiple pump stages, to perform enrichment. Here's a diagram, and the original source is page 98 of this great history book from the Energy Department: The New World, 1939-1946. Volume I: A History of the Atomic Energy Commission. There's a great description of the pump cascade.

The really important technical details aren't published in any books I've ever seen.

Nimur (talk) 04:11, 12 June 2018 (UTC)[reply]

{{ec}} Our Zippe-type centrifuge article agrees with this being an imporant detail, noting: "The three gas lines must be concentric with the fixed axis as the outer rim is spinning very quickly, and the seal is very important." But then I can't find details of what the actual engineering solution is. Several Iraq/ISIS restricted-materials lists simply say "magnetic bearings" and they appear to be rings, but do not comment about the surface material or tolerance. DMacks (talk) 04:16, 12 June 2018 (UTC)[reply]

I don't know anything about gas centrifuges that's not written in our articles (and if I knew, I wouldn't be allowed to tell), but I do note that at a radius of 10cm and a speed of 1500 revolutions/second, the acceleration at the wall of the centrifuge is about 8900km/s². The molecular mass of uranium hexafluoride is 349u, which gives at room temperature a scale height of about a millimetre. Close to the centre of the centrifuge there's already a pretty good vacuum. Seals for extremely dilute gases may be a bit different from normal seals. PiusImpavidus (talk) 09:47, 13 June 2018 (UTC)[reply]

The gas-tight connections aren't rotating, they just go to the (fixed) casing. The gas connections to the rotating part are more like scoops. Yes, there is some leakage, but then gas centrifuging is an equilibrium process, not an absolute separation. Andy Dingley (talk) 11:10, 13 June 2018 (UTC)[reply]

Aren't the scoops fixed by virtue of being attached to the pipes, which are fixed to the casing, running from the casing down into the rotor? So there would be a rotating seal where the pipes penetrate into the top of the rotor. But PiusImpavidus makes a good point, which in conjunction with the fact that the casing is also evacuated, maybe the seal doesn't have to be that good. DMacks (talk) 18:58, 13 June 2018 (UTC)[reply]

There's no rubbing contact. The pipes are internal to the rotor, and they don't contact it. The only contact is at the bearings and they're usually magnetically suspended (or some other complex form). There's no rotating seal because the rotors, their bearings and armature, are all entirely within the casing. Andy Dingley (talk) 23:23, 13 June 2018 (UTC)[reply]

To clarify, there's no moving gas-tight seal for the inlet and outlet lines. They are part of the stationary center post of the centrifuge assembly. (The scoops for withdrawing enriched and depleted gas are also fixed to the center post.)

There is, however, a moving interface at the gap where the spinning rotor moves around the stationary center post. While (as PiusImpavidus notes above) the pressure of gas is pretty low already near the central axis of the centrifuge, some gas can still leak out of the rotor at this gap. It is not uncommon for gas centrifuges to incorporate a molecular pump at this interface to pull escaping gas back into the rotor. (This is accomplished by the relatively simple expedient of cutting a spiral groove into one of the surfaces, producing a drag pump.) See, for example, this NRC publication, or this FAS article. TenOfAllTrades(talk) 14:06, 14 June 2018 (UTC)[reply]

Exactly. There is a seal, and it's between a rotating and a non-rotating part. It's not a tube that rotates in its middle (cut a tube, then one end rotates vs other without leaking at the cut) but instead concentric (stationary tube that penetrates through rotating panel). Perhaps Andy Dingley would be happer calling it a stuffing box rather than a rotating seal? DMacks (talk) 14:42, 14 June 2018 (UTC)[reply]

There is no seal. They're allowed to leak. This leak is small (density at the centre of the rotor is low) and it's contained overall by the casing so there's no loss. There is a loss of efficiency (there is some recirculation of enriched material to the unenriched stream), but this is pretty minor. Andy Dingley (talk) 17:34, 14 June 2018 (UTC)[reply]

For what function melanin normally found in the adrenal gland?

The function of melanin as I know is to filtrate the UV therefore it is in the skin and iris, but it is found also in the adrenal gland. For what function it is there? (and also in the eyes some people has a little quantity of it in the eyes, and it seems that it is not necessary to be there for normal function of the eyes. Isn't it?) --93.126.116.89 (talk) 18:28, 12 June 2018 (UTC)[reply]

Why do you think that melanin is present in adrenal glands? Ruslik_Zero 20:44, 12 June 2018 (UTC)[reply]

Maybe they read our article on melanin? That's supported by this ref, which says: "Melanin is also found in certain regions of the brain and adrenal gland of some mammals." which is not a great ref for stating that is exists in human adrenal glands. This is a borderline WP:RS, but states it clearly. This does not cover the human aspect, but offers some suggestions for what it's doing there (if you already have the requisite PhD in biochemistry to understand the abstract). Matt Deres (talk) 03:17, 13 June 2018 (UTC)[reply]

Eye_color#Medical_implications links to studies that show eye color is correlated with risk of certain degenerative conditions of the eye, so the melanin in the iris could be to protect from this. As for the adrenal glands, well, melanocytes are actually found in many tissues that are far removed from your skin, and it is not entirely clear why. It is now thought that melanocytes serve a role in the immune system (mentioned on the linked page), but this is still an area of active research. Someguy1221 (talk) 20:55, 12 June 2018 (UTC)[reply]

I think it's neuromelanin that's found in the adrenals. The adrenals interface with the nervous system to release epinephrine rapidly. Abductive (reasoning) 04:58, 13 June 2018 (UTC)[reply]

Relatively little seems to be known about extraepidermal melanocytes, due in substantial part to the failure of an air conditioner in Louisiana in 1960. [1] (perhaps I overstate the case, but the spontaneous emergence of the Pigmented Extraepidermal Tissues mutant and its subsequent divergence into five different lines of differently internally pigmented mice would have been a spectacular instance for understanding both cell migration and epigenetics) However, the variously pigmented nervous tissues mostly reflect a fact that catecholamines are not exceptionally stable and are prone to polymerize. Melanin is useful as a physical barrier, I think for electron transfer to the environment, and for protection against light; bananas, for example, produce a variant of it when wounded, much as insects do. In more recent evolution (perhaps a billion years or less) animals picked up the knack of converting epithelial cells into a signalling network using the sort of aromatic amino acid derivatives formerly secreted only for pigmentation. But they may still be prone to incidental pigment formation, and as with everything that happens incidentally in biology, that can become part of the normal functioning of an organism. The adrenal medulla is essentially a sympathetic ganglion that innervates the blood, hence part of the nervous system, and its use of catecholamines like dopamine, norepinephrine, and epinephrine mean this can happen. Analogous things happen in the locus coeruleus and substantia nigra ("blue spot" and "black stuff" of the brain). Something less directly analogous happens in the red nucleus with some sort of glutamic acid derivative, I think. Wnt (talk) 11:12, 15 June 2018 (UTC)[reply]

In this context (the biological significance of melanin), the works of Hercules Petrus Wassermann might be of interest: [2] [3]. --62.99.192.174 (talk) 23:02, 16 June 2018 (UTC)[reply]

One theory I heard recently is to protect stem cells from ultraviolet light. Graeme Bartlett (talk) 09:36, 17 June 2018 (UTC)[reply]