On the Bottom Half

As the flat, square little building was taking shape over the past year, everyone has had a comment or two, usually disparaging, about this odd, squat little place with big poles positioned at two of its corners. To wit: "It looks bent." "Are those lightning attractors?" " What's with all the 60s-esque glass panels and crappy brown wood everywhere?" "Boy, that's a lot of cement trucks." "Why is it taking so long?" "Surely it doesn't take that long just to pour the foundation for a single story building." ...

Well, for all that name calling, everything, oddly, is odd for a reason.

First, a primer on electron microscopy: take a sample of just about anything and shave it down to about 1 or 2 square cm in size. If it's hard, like metal (not the music), or rock (also not the music), polish it to a mirror finish. Coat the item in an almost atomically thin layer of somthing conductive like gold, carbon, or chromium. Insert the item into the microscope sample chamber, which is kept at vacuum, and turn off the lights. Turn on the microscope and watch its TV. The microscope fires a stream of energised electrons at the sample through an electron gun, which stimulate the emission of secondary electrons when they smash into the surface of the sample. A CCD, very much like the one in your digital camera, picks up these secondary electrons and a computer interprets them into an image. Rather than optical contrast, what you see is contrast based on charge density. There are other modes that give images based on back-scattered electrons, not secondary electrons. Back-scattered electrons bounce off your sample, typically have higher energies, and hit a different sensor. The rate that these are produced is related to the atomic density of the sample. There are different types of electron guns, too, with different advantages. Thermal emission filaments are just like elements in your oven, but better. The best are cold-filaments, which are caused to release streams of electrons by using really high voltages.

Just like a normal microscope, you see an image on a screen, which is magnified anywhere from a few times to a few million, or billion times. For my purposes, I use a sturdy, old, well functioning SEM that can zoom up to about 10000 times magnification. For basic chemical composition, an SEM can be configured to collect X-ray emission from your sample. X-rays are produced by atoms in your sample absorbing some of the electrons from the gun and then re-emitting the absorbed energy as an X-ray. The energy of these emitted X-rays is proportional to the element which emitted them and therefore can be used to determine elemental composition. This is known to industry wonks as "energy dispersive spectroscopy" or EDS. EDS gives you a general, but not precise, idea about the elements that make up the material under the electron gun. There is a more precise version of this method called "wavelength dispersive spectroscopy" which measures the emitted energy a mite more precisely and is therefore quantitative (don't you just love the jargon?), meaning that you can publish the data as established elemental compositional fact. I call this "zapping" the sample.

There are two general methods of electron microscopy: scanning and transmission. Scanning is what I just described, where energy is measured as it bounces off or is re-emitted from the sample, hence SEM. Transmission is where the electrons fired from the gun are transmitted through the sample and measured on the other side, hence TEM. Samples for this method must be polished very thinly. Primer complete.

On with the building. This building is so impressive. They have built three different classes of labs: A, B, and C. C-class microscope labs house the non-flashy workhorses like my little SEM. They have thin concrete slabs for floors, aren't as precisely temperature controlled, and are less impressive to talk about. Moving on. B-class labs are something. They have 60-cm thick concrete slab floors separated not only from the main building's foundation, but also separated from the lab's surrounding blockwork walls. The block work walls, in turn, are on separate footings and so are isolated from the rest of the building. The lab roofs sit on the block work walls and are independent of the main building roof which is itself supported separately. The walls are lined with acoustic dampening sheets. To minimise mechanical disturbance from air-conditioning airflows, room temperature is controlled using water-cooled radiative panels on the ceiling, rather than air exchange systems. Airlocks have been used to minimise pressure gradients when doors into the lab are opened and for acoustic isolation.

The A-class labs, in addition to the above features, have 1m thick slab floors, stainless-steel doors grounded to their door frames with copper bushings, double-thick acoustic linings, and operator stations placed outside the microscope room. Temperature stability is managed to less than 0.1°c/30min and less than 0.005°c/sec. The building is made out of wood and glass whereever possible to stop currents from flowing through the structure. No elevators are present in or near the building, because electric motors in elevators are a large source of stray electromagnetic field. To allow access for those less fortunate, there is a wild and crazy wheelchair ramp that goes all the way around the outside of the building. They don't use any fluorescent lighting because the ballasts required for these create stray fields.

More features: they rewired 6 electrical substations in and around the university to reduce currents grounding into the Earth; shielded, twisted pair cables are used throughout to minimise stray fields generated by power services within the building; wiring to earth is in a star configuration to minimise risk of ground loops; the path of cables into the building is designed to maximise the distance of cable bundles from the main instrument labs and minimise the number of cables into the labs; chairs made from non-magnetic materials are used in front of instrument columns; the main switchboard and power supplies are located in a separate plant building several metres from the main building; any structural steel is earthed to a common point; items such as airconditioning duct work has non-conducting spacers at frequent intervals to block current formation.

The prize microscope under installation right now is a 300,000 volt FEG TEM. That means that the electron gun is a field emission gun (FEG) type and that the microscope is a transmission type: it's electron stream is caused to flow by charging the tip of the electron gun up to 300,000 volts and these electrons are transmitted through the sample before analysing them. This is a 3 or 4 week installation job: the microscope came in about a dozen wooden crates, cost more than a million dollars, and gets an (A-class) room to itself. Much like a moody teenager. There are maybe 2 or 3 of them in the world including this one. When it's running, it will be able to image things down to a couple of Angstroms, with is on the scale that lets you see how individual atoms in the atomic lattice of a substance organise themselves and form the material. This is important in the study of alloys and new composites. You will soon be flying, driving, doing, and living with materials built >50% from composites and understanding how these substances construct themselves, with inhomogeneities at the nano-scale, is fundamentally important.

This microscope place is overwhelmingly cool. I wish I had a reason to use the million-dollar baby.