After my first year project went so badly, and my second year project went about as well, the university has given me a break, and this year I will be doing four separate projects, and I won’t be looking at anything more than a micron across all year. The “mini-projects” are one essay and three open-ended lab experiments taking between two and four weeks each, and must be submitted with a two-page report on each at the end of next semester. These are not those reports, and I’m not sure those reports will ever exist. It’s way too easy to look at a deadline like that and think “They’re due in in May, whereas Worms 3D comes out next Friday”.

Scanning Tunnelling Microscopy (STM)

Using a scanning tunneling microscope is a lot like making love to a beautiful woman, in that I never quite know if I’m doing it right, but that’s okay because I haven’t got one.

The experiment is to take some pictures of a few samples using a scanning tunnelling microscope. A machine with a name like that is obviously going to be a large chrome machine made of lots of perpendicularly intersecting cylinders. It’s clearly what it should look like. Of course, it doesn’t. We were really rather disappointed when it turned out to be a rather squat beige box roughly half the size of a Nintendo GameCube on its side.

The machine works by moving a metal tip (ideally only an atom or two wide) about a micrometer or two across a sample and reading the height of the sample at that point. It generates a height map of just the sort Hogs Of War uses, and I for one would be entertained by the idea of someone trying to play strategy games on a diffraction grating. Actually, now I came to actually type the words, it doesn’t sound very entertaining at all. Since the tip is suspended an Angstrom* above the sample, it is very sensitive to vibrations of any kind, and it is very easy to break it by jumping up and down at the other side of the lab.

For this reason the microscope is mounted on a table designed to isolate it from any vibrations, and this table works very well provided Richard Heaton doesn’t sit on it again. The microscope is connected to two important objects. One is a tiny metal tip we had to make ourselves using wire cutters and pliers out of Platinum-Iridium wire. The other is a computer. Well, nominally a computer. To be specific, Elonex (who I thought only made sub-par mouse mats) combined a 50MB hard disk with a painfully slow chip and a version of Windows written over a decade ago, and it still crashes at random. I honestly believe that the metal tip is worth considerably more money than the computer could ever fetch.

And to make matters worse, the software the computer is running would score very poorly were it entered as an A-Level project. For a start, it has no validation code. Quite the reverse, in fact; the default settings don’t actually work. Once you get the hang of everything, it is, though, entirely possible to create a series of good images. The trick, apparently, is to make sure that the computer doesn’t override any valid settings with its default garbage, and to make sure the lab is very quiet. This, though, is made more difficult by the fact that the NMR scanner is located just next to the STM, and has noisy coolant running through it. Nor is it helped by the investigation into sound waves at the other side of us. Having established this unhelpful pattern, the university decided to place the pancake flipping catapult just across from us. To get around this, we came in when the lab was open we weren’t officially supposed to be in it. Exactly what the problem was with this I couldn’t say, since nobody else uses the STM, and we weren’t disturbing anyone, but apparently it isn’t allowed. When we explained this, we were told that we could come in on a lunchtime, when not only are we not supposed to be there, the lab is closed, often locked, and the supervising staff have left and we could be stealing the valuable Platinum-Iridium wire for all they know.

But not to worry, the demonstrator tells us that this is a difficult experiment, and the others will be better…

---

*Angstrom: A unit of length equal to 0.0000000001m. It is pointless, stupid, and I prefer microns. (A micron being equal to a micrometre, which is a unit of length a millionth as long as a metre. The word micron derives from the word micrometre because it sounds cool and because Americans can’t pronounce it wrongly.)

Raman Microscopy, or, Dynamite With A Laser Beam

At the start of this experiment, the sum total of what I understood about Raman microscopy is that it was probably invented by someone called Raman. For this reason, the first week of this experiment was purely research. This meant, for the first time since I got to University, I had to visit the library.

The Edward Boyle Library was built in a year and named after someone called Edward Boyle. As you enter, you pass between two metal devices which automatically delete your memories of the instructional meeting at the start of each year. Fortunately, the system for checking books out is fairly straightforward. Go to the computers, enter some keywords to find what book to get, it lists the appropriate books. Some of these will be marked “unavailable”, meaning that you cannot read this book. These are usually located in something called the Cage. I don’t know where (or what) that is. You then try to remember where it says the book can be found. You go to the right floor, then the right subject, then back to the computers because you’ve forgotten where it was, then to the right sub-section. You find the book you want isn’t there, so you check out a subtly different one you feel slightly surprised didn’t appear on the computer, and take it back to the lobby. You then have a choice between joining the queue of people waiting to have their books issued, or doing it yourself in ten seconds. It amazed me there was a queue.

In conclusion, I rather like the library; it’s populated almost entirely by girls and you’re not allowed to talk much, so it doesn’t matter that I daren’t. (For some reason I can’t seem to find the bar — I’d be alright then.)

After the research week, came the research week. This was due to a last-minute change in schedule not being properly communicated to anyone at all. On the Monday and the Friday we met Dr. Batcheldor, who told us a little about everything, and set us on our way. It turns out, apparently, that Raman microscopes cost about £75,000. (It may have been £750,000 in fact, but it doesn’t make a lot of difference to a student who is thousands of pounds short of being able to afford a pencil. Besides, it’s only one order of magnitude out, and that’s quite good for the experimental physicist.) For this reason, we aren’t allowed to touch the Raman microscope. To get around this, Dr. Batcheldor had built one. This was very much fitting with his general attitude — his first presentation to us was one he’d prepared for the judge in a case he was an expert witness in which had used Raman microscopy to determine what was in some pills or other.

Perhaps the most important thing he said, though, was how to turn the thing on. It seems the PSU on the computer is gone, and therefore it frequently turns on only about halfway, and has to be physically turned off and on again. Then he told us that there was a good chance we would be made to demonstrate the experiment to some prospective undergraduates on the Tuesday afternoon. This seemed alright, since we had about seven hours to get this thing working before then.

Of course, we didn’t know then that the computer had been set up to give us no access priveliges at all. This meant we were not allowed to monitor the CCD temperature. (A CCD array is exactly the same thing as a digital camera but sounds more scientific. It also needs to be cooled to a very low temperature.) The instruction manual suggests we check the CCD temperature, and when we tell it to, the software doesn’t do it. Well written software, one would think, would bring up a dialogue box saying “You do not have the access rights to do this,” or “Error. Cannot check temperature”. No. This software just ignores you completely.

Luckily, on the Tuesday things went rather better. We managed to calibrate everything without any major hitches, and even got a set of textbook Raman spectra out of the thing. The prospective undergraduates were not very impressed, though, because our demonstration has been moved to Thursday and therefore they missed our winning streak completely.

Thursday came and Thursday went, and since there were no prospective students on tour that day, we didn’t have to do anything. This was lucky, because we were a bit stuck. After we’d taken all the spectra (”I’m off to lab to focus a high power laser onto TNT. I may be some time.”), the instructions (which had already suggested we check the CCD temperature told us there was a button we should be pressing to split some files into smaller ones the software supported better. Unfortunately it didn’t exist. Apparently someone had replaced it with a different button that didn’t seem to do anything helpful. We decided to ga and ask Dr. Batchelder about it, but unfortunaltely, being semi-retired he only works three days a week, and well, we aren’t at all sure which ones. I ran over to his office, which is hidden away in the depths of the Research Deck. Fortunately I had the room number written in the lab book. Unfortunatley, this was not only the wrong number for his office, it wasn’t even close. It wasn’t even a real room, let alone the right one. Eventually I found a postgrad who helped me (by accidentally bumping into Dr. Voice who knew exactly where it was). When I eventually found it, I was told that he’d just set off and that if I ran I might catch up with him before he got there. I didn’t.

A lenthy trawl through the help files and a short email conversation later, we discovered two things about the elusive button:

1. How to put it back
2. It didn’t do what we wanted it to anyway.

We had found a way to do it without the button, though. We just had to load each file (we had about forty), split it manually, and close it again before doing the next one. Then we had to convert them to ASCII files. Then we had to combine the ASCII files into a particular sort of graph which is very awkward to make in Excel. And when I say “we” I mean “I”, because Chris can’t work computers.

Oh, well. We’re all done now, and we even rigged it so the next group won’t need any stupid button. Aren’t we nice?

Nuclear Magnetic Resonance

Week two of the NMR experiment is drawing to a close, and the demonstrator still hasn’t told us what we’re doing. (You remember Mike Reis, right?) We know what to do, and how to do it, but until we’re told exactly what it all means, we don’t have to do any thinking. It could be done by a six year old, and he would have as much idea what he was doing as we do. It’s a waste of our time, spending 6 hours taking results unless we understand the process. We may as well be given the results. It would be quicker. In fact, though, it’s worse than that. Occasionally an error message would appear, and while they’re a welcome break from the routine, we haven’t a clue what to do about them, and we can’t ask because Mike was away for days on end. Lee is of the opinion they want us to feel busy.

It isn’t taking. On Tuesday I got so bored waiting for samples to cool to the right temperature and waiting for the NMR machine to finish scanning things I set every five-dial variable resistor in the lab to 24,601Ω. As it turned out there was only one of them, but I set it anyway. In the next few days I intend to do some research and discover exactly what I’ve been doing all day today. For all I know I might have spent my morning furthering Mike Reis’ evil scheme for world domination.

And the software to do it is another ancient, poorly-written affair. Having grown sick of Windows 3.1 in the first experiment, we now have to use DOS. DOS. I haven’t used actual DOS for years. This software is a little better than the STM software. While some of the menus cannot be escaped without setting the Ï„-values, there are at least some validation checks. They don’t work terribly well, though, and sometimes refuse input because it doesn’t like whereabouts in the edit box you typed it.

Part two of the experiment is to analyse the data you have collected. This means doing some research, and this means going to the library again. This time, as well as a couple of useful books, I found a battered, dusty old tome marked with a yellowing sticker “ANDREW — Nuclear Magnetic Resonance”. Unfortunately the best book has been reserved by an unknown party and until they pick it up, get bored of it, and return it, the University’s policy is “I’ll let you glance at it”. There is at least one NMR textbook that was created on a typewriter. You can tell because it is entirely in Courier and the equations have all been mangled to fit into one line of standard ASCII.

The Assessment

The lab module is assessed by looking at the laboratory notebooks. This requires them to collect the books in and mark them over Christmas. That wasn’t a problem. The viva wasn’t until two and a half weeks into the new semester and they gave the books back on the first day anyway. Well, that is to say they gave five of them back. Everyone else had to wait another week or two. I had to wait two full weeks, leaving me with two days to prepare for the viva. It was a push, but it could be done.

When I say they marked the lab books, that isn’t strictly true. I mean, really, that they kept them. The only difference between my book before and after Christmas was that they had inserted a sheet of paper which would have given me a detailed breakdown of my marks had they bothered to fill it in.

While I was waiting for the department to finish ignoring my lab book, I got an email from Dr. Marrows explaining that he hadn’t ignored my book yet, and would need to have it. His email was entirely in lower case. Out advanced laboratory module is being marked by a man who cannot capitalise. I emailed him back saying that I hadn’t got it, and when I did get it back I wanted to keep it for a while to sort out my viva. He replied, essentially saying “no book, no marks” and explained that he had been off sick and that was why he hadn’t marked my book, so I went to ask Mike Reis if I could postpone the viva. No. He said I should go and ask Dr. Marrows much the same thing. When I eventually found his office (which is located deep in the dark heart of the labyrinthine research deck, where strange creatures roam the corridors and space folds back on itself to make sure you can never reach your destination) there was a note on the door saying he was off sick for the week. At first I thought the note was out of date — he’d been off sick and hadn’t removed the note — but the date at the top of the note begged to differ. I kept my book. His secretary didn’t need to hold it for a week.

The viva wasn’t the usual farce, as it turned out. In fact it was a whole new type of farce I’d never even seen before. Essentially it consisted of Mike Reis donning a black cloak and a Time Gauntlet, and booming “JUSTIFY YOURSELF” at us, but without the convenience of being marked by ourselves. (I could bribe myself easily because I know what I like.)