Am I getting fatter, or is a kilogram getting lighter

[Dr. Tower]

Or maybe a combination of both. The basic gist of the article is that the standard used for measuring a kilogram is about 50 mg off of the average of other standards based upon it, and we don't know why. Granted, this is 5 parts off in 100000, but "[f]or scientists, the inconstant metric constant is a nuisance."

What do you guys think? Should we change the definition of the kilogram to something else? The mass of a certain volume of deionized water at a given temperature and pressure? The one suggested in the article is a sphere of single crystal silicon. This makes a bit of sense as we are able to silicon more pure than any other substance on earth.

[Binks]

The problem with the water one, which was brought up somewhere else, is that water, on average, contains like 5% Heavy Water (the decimal point may be off, might be .5% or something). So do we use pure, heavy water free, water? (Might be hard, definitely harder than silicon) Do we use water with exactly 1% heavy water? 2%? etc.

I like the idea for a Silicon based measurement. I was under the impression, however, that this loss was a loss of weight rather than mass, as their measuring system uses weight or something. Is that correct or do they use a system that doesn't rely on gravity? Because if they're relying on gravity then there are a lot of things that could cause this supposed loss in mass without the item actually losing any mass (someone mentioned a 'bubble' in the magma below France which would alter the gravity there enough to cause an effect like this)

[Dr. Tower]

When you do a direct scale measurement between two masses, as long as the gravity has the same pull on both (which they would within our ability to measure if they are in the same room) then you can determine the mass. The article said that the various other references are transported to the original mass for the comparison, so the gravity would be the same for both, at least that is what I got from it.

[ampersand]

How hard would it be to make an mass that is exactly 1 kilogram? And if the stuff it's made from is disintegrating, why couldn't they use a metal that has the least chance of reacting with anything, thus losing mass?

[Blaze]

They're already ahead of you, Tower.

Australian Lab crafting perfect silicon kilogram spheres.

[Dr. Tower]

It wasn't my idea in the first place, I saw in in the article I linked, which was why I measured it. Also, it isn't guaranteed that they will use silicon spheres, as it has to be approved by an international body, and sometimes tradition holds more weight than it should.

[Deacon]

...so to speak

[adciv]

Come now, everyone knows mass isn't the same as weight.

[Deacon]

Actually, I just looked it up because I wasn't really sure what the difference was, though I knew there had to be one somewhere, and here it is:

A property of matter equal to the measure of an object's resistance to changes in either the speed or direction of its motion. The mass of an object is not dependent on gravity and therefore is different from but proportional to its weight.

[Dr. Tower]

Come now, everyone knows mass isn't the same as weight.

Nobody said that it was. However, to physically compare the mass of two items, the easiest way is to bring them to the same place (which has the same gravity) and weigh them against each other.

[Martin Blank]

I just caught an article last night that suggests setting Avogadro's number as a true constant first, and then specifying a gram as one-twelfth of a mole of carbon-12 atoms. Avogadro's number would be set from its current approximate state of 6.02214179x10²³ to 84446886³, or 602,214,098,282,748,740,154,456, which is divisible by 12, and so sets a gram as a mass equivalent to 50,184,508,190,229,061,679,538 carbon-12 atoms.

I rather like this, because it sets a precise, countable quantity as a baseline for measurement using a common, stable isotope. This goes along with time (number of vibrations of a particular cesium atom) and distance (distance traveled in a vacuum by a particular wavelength of light in one second), among other things. The platinum mass was bound to change anyway as a result of the decay of the naturally-occurring platinum-190, which makes up .014% of platinum and which decays through alpha emissions.

[ampersand]

Question, oh Mister Knower of Things: How are they going to count out 50,184,508,190,229,061,679,538 carbon-12 atoms? And how do you know that Avogadro's number is divisible by 12?

(Granted, I had no idea that platinum was naturally radioactive.)

[Martin Blank]

As it stands right now, Avogadro's number is (6.022 141 79+/-0.000 000 30) x 10^-23. It's a range of values, which researchers have been trying to determine for decades. The inverse method is what is suggested here -- instead of trying to figure out what it should be, simply define it within the values that are allowed (and make it divisible by 12 so that a gram is defined as a non-fractional number of carbon-12 atoms) and make that the source for the mass of a gram. This mass becomes constant no matter where one is, and no matter what the gravity field is like, because it's perfectly constant.

The issue isn't counting out 50,184,508,190,229,061,679,538 atoms. That kind of precision isn't necessary at gram masses. But when dealing with masses that get down to the level of femtograms (10^-15g, or 50,184,508 carbon-12 atoms), the extra precision becomes helpful. In fact, by setting the mass out to 23 significant digits, it will become one of the most precise constants in science.

We did the same thing with light. For many years, researchers tried to figure out how fast light was to very, very precise levels. Finally, they flipped it around and defined the meter in terms of the second, which had been defined already, and now one meter is the distance that light travels in 1/299,792,458 of a second, meaning that light travels 299,792,458m in one second, and not an angstrom more or less. (I had earlier stated that it was based on a particular frequency, but this is outdated, and was from when light was measured in terms of the number of wavelengths of a particular frequency of light.) The same idea is presented here -- instead of trying to figure out how much a kilogram masses, define the mass itself and then (if you want) build a reference mass.

[leviathan3k]

How about http://en.wikipedia.org/wiki/Planck_units ?

Planck units are based on elements of physics that are valid in any region of space. They're not dependent on any prototypes or arbitrary elements where someone said "this is how big a foot is".

The base units themselves are mostly radically tiny, but we could just add exponents until they're easy to wield, and give them colloquial names. 1 Planck length x 10^38 is close enough to a mile that we could probably call it that.

[collegestudent22]

Why do we need to redefine the length measurement? It is already based off the constant speed of light in a vacuum. The only measurement that is ambiguous in any sense is the kilogram....