A stonking good start to our experimental physics paper. Not.

By Marcus Wilson 13/07/2013


 Just a couple of hours ago, I was thinking that I really need to do another blog entry for the week, but (a) can’t think what to do it on and (b) don’t have time to do it because I have a lab class for the afternoon. Well, the events in the lab class have (in)conveniently both of those problems. 

It’s the first lab class for our second year Experimental Physics paper. As the name suggest, this paper is based in the laboratory. There are a few lectures, but these are to support the teaching of lab techniques. The first session is always an introductory one. I talk about some general issues with doing experiments, some safety issues, what I expect from the students, and so on, and then we work through an experiment as a group. The idea is that I model what I would like to see in their experimental work – e.g. making several measurements of something, not just one, plotting a graph carefully, dealing with experimental uncertainty (both random and systematic), making good notes in a logbook, working to the right number of significant figures, and so on. I’ve found that doing this in the first session really helps students when they get set loose on some equipment for themselves the next session. They have some idea what I’m expecting of them.  

 The experiment I usually do with the class is measuring the charge-to-mass ratio of an electron, "e/m", using an electron beam in a magnetic field. It’s a lovely, historical experiment, and the theory behind the method is understandable just from school physics. An electron moves in a magnetic field. It has a force on it perpendicular to both the velocity of the electron and the magnetic field. A force always perpendiular to a velocity is what happens in circular motion – and this therefore results in the electrons moving in a circle (or, more accurately, a helix). The diameter of the circle depends on the strength of the force, which in turn depends on the strength of the magnetic field. So, in this experiment we change the magnetic field by changing the current through some coils of wire, and measure the resulting diameter of the circular electron beam. From the measurements we can calculate e/m. 

It’s a classic experiment. Historically, the ratio of the charge of the electron to its mass was measured before either the charge or the mass of the electron was measured individually. Several years later Millikan measured the charge on the electron with his famous oil drop experiment, and then the charge of the electron, and hence its mass (since e/m was already known), two very important values in physics, were known for the first time. Incidentally, we also have Millikan’s experiment in the lab, but I tend not to inflict it on students as it is exceptionally tedious and frustrating. It is true that a good scientist needs to be patient, but it is also true that there are limits on the numbers of students we can afford to lose from our courses due to extreme boredom and frustration. I have great admiration for Millikan’s patience and persistence. 

 How does one generate and ‘see’ electrons? We use a vacuum diode. A heated cathode emits electrons. An electric field accelerates them and a beam emerges from a slot in an anode. The electrons are ‘visible’ because the equipment is inside a sealed glass vessel containing a low density gas. This gas fluoresces as the gas molecules are hit by electrons. We see a nice circular beam. 

Or, in the case of this afternoon, we don’t. It appears that the tube has blown. Possibly the cathode has got too thin and has melted, rather like a ‘blown’ lightbulb filament. There’s the right voltage across it, suggesting there’s not a problem with the power supply, but there’s no glow coming from the filament. After a bit of investigation I realized I wasn’t going to solve the problem in a few minutes and I canned the rest of the class, which was rather unfortunate.  

A lovely example of "biology experiments wriggle, chemistry experiments smell, and physics experiments don’t work."