Skeptoid PodcastSkeptoid on Facebook   Skeptoid on Twitter   Skeptoid on Stitcher   iTunes   Google Play

Members Portal



Get a Free Book



The standard that needs some cleaning

by Bruno Van de Casteele

February 3, 2013

Share Tweet Reddit

Donate One of the last "arbitrary" standards in physics is securely stored in a lab in Paris. I say "arbitrary" as it is not based on a physical constant. Instead it is "just" a chunk of platinum, or more correctly, a precisely manufactured piece of platinum and iridium that defines by itself the "kilogram".

It's called the International Prototype Kilogram, and was made in 1879. At the time, 40 exact copies were made and distributed across the world. This helped to keep track of the standard and of the calibration of instruments, without having to handle the standard itself too much. Handling means contamination, because even under very controlled circumstances and with the necessary precautions, stuff will stick to it during the process.

It was already known that the copies, with one exception, were gaining weight in respect to the "real" standard, presumably because they are handled more than the original. However, after careful investigation of the surface of the standard itself, it too was gaining weight due to contamination. It amounted to just a couple of grains of sand (50 micrograms per century), but that is enough difference so that action needs to be taken.

But how do you clean such a thing, in a standardized way? Just wiping it with baby wipes will likely a) leave trace amounts and b) might take off some of the original material. Now, after 20 years of study, Peter Cumpson, with the incredibly cool job as a metrologist, and others have devised a standardized method that can be applied to the standard and all copies with the same reproducible effect. As described in Metrologia and reported by Scientific American, the metal gets exposed to ozone and ultraviolet light, removing all carbon contamination. Not your average cleaning procedure ...

It's a start nevertheless, but it does not remove for instance mercury contamination (previously used in lab thermometers) which binds with the platinum. A better solution to avoid making and safeguarding an arbitrary chunk of metal would be to define it based on physical constants, as the second is now defined. For a good overview of current attempts, please check Wikipedia. Candidates include a watt balance (measuring the electric power needed to oppose a given weight) or counting a certain number of atoms (Avogadro Constant) but they are as of yet not precise enough. To put that in perspective, the error on the standard kilogram is 20 parts per billion (or 20 micrograms), and proposed methods do not yet improve on that.

So it seems that the cleaning method is a nice stopgap for a certain time to come. It has not yet been applied to the standard, and neither to one of its copies. The standard itself comes out only once every 50 years, and there are three independently kept keys necessary to get access to it in the vaults of the International Bureau for Weights and Measures. Just to show that our scientific measurements are maybe as important and valued as our gold reserves...

by Bruno Van de Casteele

Share Tweet Reddit

@Skeptoid Media, a 501(c)(3) nonprofit