Posted on 06/14/2007 10:36:36 PM PDT by anymouse
They will be the earth's roundest spheres, crafted by Australian scientists as part of an international hunt to find a new global standard kilogram.
Ever since scientists discovered that the current standard -- a bar of platinum and iridium held in a French vault since 1889 -- was slowly deteriorating, the search has been on for a replacement.
Using a single crystal of silicon-28 grown by Russian and German scientists over three years, a team of Sydney scientists and engineers will grind and polish two silvery balls, each weighing precisely one kilogram, with imperfections of less than 35 millionths of a millimeter.
"We are doing everything to really create a perfect object. It's not only near-perfect in roundness, but also the crystal purity, the atomic species and so on," project leader Walter Giardini told Reuters on Friday.
"Silicon is a very nice material to use that we understand well, makes good crystals and can be worked," said Giardini, from Australia's National Measurement Institute.
The two balls will take 12 weeks to create and, because they are made from a stable element, they will not fall victim to moisture, corrosion and contamination like the current kilogram standard, known as the International Prototype.
The spheres will be a step along the perfect kilogram road, with the project's ultimate aim to re-define the kilogram in terms of numbers of atoms, rather than an object open to damage from earthquake or environmental changes.
"The aim is not to change the value of the kilogram, but to ensure its stability for all future times," Giardini said. "It will no longer depend on an actual physical object and this is going to allow us to relate the mass to the individual atoms."
The project is a collaboration involving scientists from Russia, Germany, Italy, Belgium, Japan, the United States and Australia, in cooperation with the International Bureau of Weights and Measures.
On completion, the spheres will be measured for volume in Australia, Germany and Japan, then measured for mass. Belgian scientists will look at the molar mass of the crystal used to calculate the number of molecules in each sphere.
Australian scientists have the most expertise in grinding near-perfect spheres, having been turning them out for clients including NASA since the early 1990s.
"We have developed technology so that we can see what we are getting, whether they are slightly oval or flat. We are trying for an accuracy of two parts in 100 million," Giardini said.
Yes, but it still depends on an actual physical object.
the balls may be taken on tour, rented out.
I hear Bill is doing that, and now Hillery! wants them back
in time for the campaign.
Back in 1966 I was reading a magazine and came across an ad for Timken bearings; it said, “It takes balls to make a revolution.”
“Or, the balls may be taken on tour, rented out.”
I thought that was illegal in most states????
There are a kind of machine shop calibration blocks, called jo blocks, that have surfaces so flat that surrounding air pressure will make them stick together like a suction cup sticks to glass.
Sorry, I'm pretty sure they used to do it in Latin when I was a kid!
We do! Haven't you bought soda pop or cigarettes lately??
That's still a few skillion atoms...
Measurement of balls, or sphericity, is something that bearing manufacturers do well and to very tight tolerances. The technology for manufacturing and measuring bearing balls has also been around for a long time. It sounds like Australia has taken this technology a step further.
Yes, they do.
I'm just wondering when we stopped using the brass monkey standard for measuring perfection. Or were those only used to measure temperature?
Two perfect balls, eh? Perhaps they could be loaned to our President.
A Silicon surface spontaneously oxidzes to SiO2 to a thickness of about 40 Angstroms at room temperature and then stops. This is a bit of a headache for semiconductor manufacturing processes. To get more SiO2 to grow on the Si you need to elevate the temperature and add steam. We used to use about 900 C and burn H2 in the furnace tube for the steam.
Si is fairly unreactive to most chemicals. To etch or dissolve Si with wet chemistry a mixture of HF and HNO3 some times with a little Iodine in the etchant.
The number of atoms of Si would be equal to 6.02x10^23 x (the number of Moles in 1Kg of Si) with a small error for the surface oxide and what ever gases are dissolved in the crystal lattice.
I ALREADY explained that I am NOT talking about the damned balls; but SILICON, the element.
The ELEMENT is reactive; the JOURNALIST makes the error of claiming it ISN'T, because he did not understand, that as you & others say, it won't have any practical effect on the balls.
Si is actually fairly non-reactive at normal temperatures and in a normal atmosphere once the native atmospheic SiO2 layer is formed and that forms very quickly, on the order of a few minutes. It takes a lot to make Si react. There are ways to make Si react but not with out a lot of effort using some fairly exotic techniques-high temeperatures with steam to add more SiO2 or high temperatures with pure HCL gas flowing across the Si surface or in an RF plasma with SiF6 gas enetering the plama chamber.
Single crystal Si is usually grown from a melt(~1350 C) of amorphus or polycrytalline Si into which a seed crystal of the specific lattice orientaion desired—1,1,1 or 1,0,0 or 1,0,1, or 1,1,0, etc —is lowered while kept spinning and the slowly withdrwn from the melt.Much trouble is taken to keep all impurities out of the melt but you still get some gasses and other elements trapped in the crystal lattice. The Si for the balls may be produced in a different manner but most semiconductor Si is produced as I described.
So, yes Si is reactive But not by just sitting around in the vault at room temperature that these standards will be kept or by normal handling. I doubt that anyone would be allowed to touch them with bare hands after they are “complete”.
Congratulations on your intellectual triumph.
SiF6 should be SF6
Forty angstroms is 40E-10 meters. The diameter of the ball is perhaps .1E0 meters. The ratio is then 4E-8.
That ratio is greater than the target accuracy mentioned in the article of "two parts in 100 million". The oxide will form on opposite sides of the sphere, making the possible contribution of the oxide perhaps four times greater than the target accuracy for fabricating the spheres.
One may safely conclude that the inherent accuracy of the standard will be limited at somewhere in the 1E-7 range or higher. Once fabricated, the stability of the standard will be pretty good with proper care to avoid contamination of the sphere.
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.