Laboratory science is not like working in a wood shop; you usually don’t have the option to just sand off a little more to cover up your blunders. Lab experiments should be done to a ridiculously high standard. Everything is calibrated, tared, titrated, compensated for… and then re-checked.

As the differences you are looking for are often so slim, the tiniest bias in one way or the other can efface results, or worse, give you false ones.

Standard operating procedures are the holy scriptures of lab land. These documents provide a step-by-step explanation of a rigorous, standard way of doing many types of experiments. Because they are ‘standard’, many other researchers will use them as well. Sharing methodologies is not just a short-cut from developing your own; it means that your results will all be comparable later on as you used the same procedure.

However, as those of you who have already spent significant time in lab land will know, things NEVER work out like this. Unlike the Hollywood portrayal of white, sterile environments with gleaming new machinery, the reality is usually much different. Lab equipment is insanely expensive, and as a consequence gets replaced as infrequently as possible. It’s normal to be using 30, 40, even 50 year old machines in some cases, and they all have their own idiosyncrasies.

Idiosyncrasies may be a mild word. In fact, I would say a better description would be that each machine hates you in a slightly different way. You have to spend time with it; get to know its temperament and quirks.

I believe that in the past I have alluded to an experiment I worked on which looked at pollination drops. Basically what this means is that I would go outside with a minuscule glass pipette hunting for juniper bushes. I would then use a flame to melt the pipette into an infinitesimally fine point and use that eyelash-thin shard to begin collecting tiny drops of liquid from the tips of the cones on the plant. When I say infinitesimal, I’m talking about fractions of a microliter. It took days to get a single eyedrop’s worth of sample. Having done this, I needed to transfer the sample into a well in my machine’s sample holder by blowing it out of the pipette. The well was so small that I would use a microscope to try to load my stuff in. On one occasion, I shifted my weight to the other foot and my diaphragm must have just moved a tiny bit, causing an almost imperceptible involuntary inhale.

I ate it.

I ate my sample.

I ate one week’s work.

Lately I have been dealing with a machine that has given me a strong desire to eat my sample out of frustration on many occasions.

A bomb calorimeter is a machine with 2 parts: a bucket with a thermometer in it and a metal cylinder that closes very tightly (the ‘bomb’). The basic idea is that you put a sample in the cylinder and lock it up. Then you immerse it in the bucket of water and burn the sample completely. Since we know how much energy it takes to heat up water, the change in temperature of the water in the bucket tells you the exact energy released from whatever you burned. Simple, huh?

So simple, in fact, that the machine has been around since it was invented by Marcellin Berthelot (curse his name) in the mid 1800s with essentially no major design improvements. And you know what? Its age shows.

This thing has been killing me. First you have to measure out EXACTLY 2 L of water and make it EXACTLY 23 degrees. Then you must press your sample into a pellet and weigh it to 0.0001g. But this is no simple feat my friends! It turns out you must press the handle of the pellet mill in ONE steady motion. Thou shalt not press down twice. Three times is right out. After doing your ONE steady press you must tap the handle with a hammer FIVE times. NOT FOUR OR SIX! The result should be a pellet that sticks together, but is uniform in density and not too hard (this was discovered after several frustrating days of having my supervisors watch as I screwed up sample after sample).

You then suspend the sample in a little metal holder and cut EXACTLY 10cm of fuse wire. The wire must be juuuuuust touching the sample, but not damaging it, and not touching the holder or cylinder or anything else. Then you pressurize the bomb to 30 atmospheres, put it in your 23 degree 2 L water bucket, and start the machine. Each sample takes 15 minutes.

So far this thing has managed to develop a habit of misfiring, 2 of the 3 bombs have developed a leak, and, best of all, my samples have developed a habit of leaping magnificently out of the sample holder as soon as I ignite them (hard to blame them for this I suppose). Imagine the joy of doing this whole rigamarole, waiting 15 minutes for the machine to run, and then opening the whole thing up only to find that your sample has decided to once again flee the holder, staring smugly out at you unburned in the bottom of the bomb. Again and again.

Many times I have thought that I’d prefer to be using the original calorimeter, a device built by Antoine Lavoisier which involved a panting guinea pig in a bucket with snow around it. I will grudgingly admit that Berthelot’s invention was a SLIGHT improvement on that. I suppose the bomb cal wasn’t such a horrible invention after all; Berthelot did get a bronze bust done by Rodin, so he must have been doing something right.

In any case, I will gladly make a bust or scultpture or portrait of anyone that can save me from processing the 200+ bomb calorimeter samples I still have to do…