Two New Elements Join the Periodic Table | Hacker News Reader

Two New Elements Join the Periodic Table

blogs.discovermagazine.com

30 pointsGarbage10 comments

>The new elements’ discoverers will get to pick permanent names for 114 and 116. And they can be just about anything the researchers want—”as long as it’s not something really weird,” says the Joint Working Party head

Can I hope that means no "unobtanium"? Please?

I have a question. Since they're trying to get to an "island of stability", what do they hope to find there? And since they seem (to me) to be likely to get there by smashing a few atoms together, it's not likely to be anything we can do anything with. The achievement would seem to provide a better understanding of particle physics, perhaps, but will the elements themselves ever be of use?

That's what's exciting - we have no idea what we'll find there.

But if it's stable it guaranteed to be interesting. True right now we have no production methods, but if the elements are useful enough we'll work on that.

But the science is probably the main driver, not applications. We don't even know where the island is! We have guesses, but no knowledge.

When physicists tell you that the Standard Model is incredibly complete, remember that they still can't predict the properties of elements.

I have always wondered, why is it so difficult to predict the properties of an element?

The quantum N-body problem is a 3N dimensional problem. If your element has 100 electrons then you need to model a 300 dimensional space. If you took a very coarse discretization of 10 cells per dimension, you'd need a computer with at least O(10^300) bits of memory.

Which means that you rapidly get into trying to develop a large number of approximation schemes. The validity of each scheme is difficult to determine, and there's always the possibility that the atom, molecule or crystal that you're studying relies upon some feature of the full 300 dimensional picture that you waved away as not being significant.

Not only that, but there's a disconnect between what we know about chemistry, and what we know how to interpret from the results of the simulation. Everyone knows that carbon can form 4 bonds, we've known this long before we were sure about electrons. However, trying to explain what this means in terms of electron probability densities is a research problem in itself.

And yet we can predict (or at least run a humanly-feasible amount of experiments to discover) the properties of new molecules that are orders of magnitude more complex than an atom, such as recombinant DNA structures and fusion proteins - which are being used to great effect in medicine etc. It blows my mind

Also remember that a model/theory being complete is only loosely correlated with your ability to make complex predictions, especially of a vastly different scale.

For some reason it occurred to me that someone has already registered the domain names, eg http://ununquadium.com , even though these are not the official names. I checked, and someone has. Amazing.

http://google.com/?q=ununquadium

If I was a domain squatter, I would crawl the internet looking for words/ngrams and register .com domains with them. Which is probably what someone did.

Reply on news.ycombinator.com