What Do You Need to Make Cold Atoms? Part 1: Vacuum Hardware

Over in the reader request thread, Richard asks for experimental details:

I'd be interested in (probably a series) of posts on how people practically actually do cold atoms experiments because I don't really know.

I needed to take some new publicity photos of the lab anyway, so this is a good excuse to bust out some image-heavy posts-- lab porn, if you will. There are a lot of different components that go into making a cold-atom experiment, so we'll break this down by subsystems, starting with the most photogenic of them, the vacuum system:


(Click on that for a much bigger version.)

This looks like a doomsday device of some kind. So what is all that stuff, and what is it for?

That picture is the beam line for my experiments at Union, and the whole thing together is an ultra-high vacuum system (UHV is a slightly nebulous term, but the pressure range in my system is between 10-9 and 10-7 torr, or around a ten-billionth of atmospheric pressure (780 torr, more or less)). The experiments need to be done under vacuum, because the laser cooling techniques used to make cold atoms affect only one particular atomic species, and only one specific isotope of that specific atom. Anything else that's around the lab will remain at or near room temperature, so the atoms in other objects have more than enough energy to knock atoms out of the trap they will eventually be held in. To avoid that, you need to make sure that none of the cold atoms you're making come in contact with any room-temperature atoms.

i-6c996a1804936e695663b774eb519ee2-sm_pumps.jpgThis process starts with the pumps. There are lots of different types of vacuum pumps, each with different advantages and disadvantages. What type of pump you use is largely determined by what you will be pumping with it. In my case, the target atoms are krypton atoms, which are chemically inert. This rules out nice, quiet ion pumps, and means that I have to use turbopumps to handle the gas load. The picture at right shows one of the two turbopump systems on the system.

The large thing at the bottom is an oil-free scroll pump, used to get a rough vacuum, and pump away the gas sucked out of the main chamber by the turbopump at the very top of the picture. These are both noisy pumps, but not as bad as some-- there's a big old Welch rotary pump that we use for some things that makes an incredible racket. The scroll pump is connected to the turbo through a long hose made up of two sections of flexible aluminum tubing, and a short section of red rubber hose (you can see another section of red rubber hose running off to the left, where it connects to the other turbo at the end of the beam line). This is intentional-- the aluminum hose has slightly better vacuum properties, but the red rubber hose damps out vibrations from the scroll pump somewhat, keeping the table a little more stable.

People who work with alkali metals-- rubidium, cesium, sodium, etc.-- can use ion pumps and thus much smaller vacuum systems, but they still need to get all the air out of the system, so any cold atom experiment will necessarily contain vacuum hardware.

The next important issue is the containment itself, and getting the atoms you want inside. The bulk of the chamber is stainless steel ConFlat hardware, which works like a baby bottle. There's one exception to this on my chamber, and that's the gas inlet seen here. Because krypton starts as a gas, we feed it into the chamber from a gas bottle. While it's nice to be starting with a vapor, rather than a solid that needs to be heated enough to vaporize, it's impossible to cool krypton in its ground state using current laser technology, so we need to excite the atoms into a metastable level using a plasma discharge, which is the purple glowing thing shown here. The discharge is sustained by a radio-frequency oscillating field, which means it has to be in a glass tube, rather than stainless steel like the rest of the system, so the gas is fed through a chain of valves and other connectors into a glass tube, which enters the chamber through a compression fitting-- there's a rubber gasket around the glass that's squeezed between two stainless steel plates, making a reasonable vacuum seal.

This compression fitting is the weakest point of the vacuum system. It's possible to get glass-to-metal seals that would go directly from glass tubing to ConFlat, but this is a whole lot cheaper, and works well enough for our current purposes.

People who work with alkali metal systems generally produce a vapor using some source of metal that is dumped into the vacuum system. When high temperatures are needed to get the metal into vapor form, the metal is generally inside an oven, which is just a vacuum can with a heater attached and a small hole in the side to let a beam of atoms out. The temperatures needed can be pretty high-- sodium ovens run at a couple hundred C, and some guys at NIST built a chromium trap whose source needed to be at 1000 C-- but it's not that difficult to arrange. Elements with low melting points, like rubidium and cesium (cesium will melt in your hand, right before it explodes from contact with water vapor) can be used in a vapor cell system, where you just dump some metal into your chamber, and trap the slowest atoms from the thermal vapor. Those systems are just a single chamber, or possibly two chambers, one for trapping atoms, and a second one with a lower vacuum pressure, where the atoms are transferred for experiments.

In a system like mine, the atoms form a beam, all traveling in more or less the same direction, which is from right to left in the picture at top. They're decelerated using light forces as they fly down the beam pipe, and end up in another chamber at the end, where they're hit by lasers and magnetic fields that trap the slow atoms. This chamber has a second, smaller, turbopump to keep it under vacuum, and looks like this:


I refer to this as the "WMD" chamber, because it looks like a doomsday weapon from a 1950's sci-fi flick. the turbopump is at the bottom-- you can see the hoses going into it-- and the big spherical chamber with lots of ports coming off it is where all the trapping will be done.

The huge number of ports on this chamber are there to provide lots of optical access. Technically, you only need six to do laser cooling, but it's always nice to be able to look at the atoms from other directions, and to be able to shine in additional laser beams to probe the atoms. The chamber that's on there is probably overkill, but I wanted to be as flexible as possible.

You can also see one of the two magnetic field coils that help make the trap-- the big coild of copper wire on the right side of the chamber. Accumulating cold atoms requires a smallish magnetic field inside the vacuum chamber, which can be obtained by either putting small current-carrying coils inside the chamber (which is an enormous pain in the ass), or larger ones outside. The metal of the chamber itself (electropolished stainless steel) screens the field a little bit, which is why people who need really big fields will use all-glass systems (which also provide lots of optical access more easily). That wouldn't work with some of the experiments I want to do, though, so I have a big metal chamber.

You can see some of the many windows on the chamber, and also some plastic covers (spray-painted black) that we use to block the windows for low-light experiments. The amorphous black mass to the left is a cloth shroud covering a photomultiplier tube that we were using to detect slow atoms in the chamber. when everything is running properly, that wouldn't be there. You can also see a few of the many mirrors used to steer lasers into the chamber, which will be the subject of the next post in this series, probably tomorrow.

So, that's the vacuum apparatus I use to keep my cold atoms from coming into contact with anything that would heat them up. Any questions?

More like this

In the previous post on this topic, I discussed the various types of noisy vacuum pumps, both clean and dirty varieties. This time out, we'll deal with the quiet pumps, the ones that don't deafen people working in the lab. Quiet and Dirty: The quintessential quiet and dirty pump is an oil…
Last week's series of posts on the hardware needed for laser cooling and trapping experiments dealt specifically with laser-cooling type experiments. It's possible, though, to make cold atoms without using laser cooling, using a number of techniques I described in two posts back in January. Those…
A great many physics experiments need to be conducted at low pressures, in order to avoid sample contamination, thermal effects, or dissipative forces produced by interaction with air. Some experiments don't require all that much in terms of vacuum, while others require pressures so low that they'…
We've continued plugging away at the optical excitation experiment discussed in the Week in the Lab series last year, and have finally managed to get a decent metastable signal out of the thing. The signals are at pressures that are considerably higher than I would like (and quite a bit higher than…

the pressure range in my system is between 10-9 and 10-7 torr, or around a ten-billionth of atmospheric pressure (780 torr, more or less)

Is your lab in, like, death valley or something?

I worked at a (very, very) small semiconductor research company for a few years, so I learned a bit about vacuum systems. For those who want to learn more, a good place is the web-sites of the suppliers. These two both have good descriptions of the various bits of metal and pumps needed for a high vacuum system:


There are others, of course, but these two were our main suppliers and they both have good websites.

Despite appearances, Chad's vacuum system actually seems fairly simple: there are no moving parts under vacuum, he doesn't have to introduce large objects on a regular basis, and since he works with krypton gas, I imagine he doesn't have to clean it out very often. I've never seen a microchip manufacturing plant, but I imagine they have some very cool vacuum toys.

By Grep Agni (not verified) on 11 Aug 2010 #permalink

ugh. right now 2.75" viewports are on my list of evil lab items. we have one we use for an optical temperature measurement of our wafers. it developed a spider webbing of cracks.

It's not pinin,' it's passed on! This viewport is no more! It has ceased to be! It's expired and gone to meet its maker! This is a late viewport! It's a stiff! Bereft of life, it rests in peace! If we hadn't bolted it to the port it would be pushing up the daisies! Its structural processes are of interest only to crystallographers! It's hopped the vacuum chamber! It's shuffled off this mortal coil! It's run down the curtain and joined the choir invisible! This.... is an EX-VIEWPORT!

however, it is a plucky little viewport. it is only leaking about 10^6 Torr, rather than imploding and shooting silicate shrapnel into our chamber, shredding everything in it's path.

replacing it will take 5 minutes, but then pumping down and baking the chamber will take days and days and days.

sad face.

You Leftpondians really are entrenched in your old-fashioned units, aren't you?

I can understand not linking Pa for meteorological work (yet we do), but it would seem that such a wee unit would be well suited for vacuum work.

What I like is the peg you use, near the containment chamber, to hang the tinfoil for your hat. Handy incase the atoms get out, no?