Innovative Ion Trap on a Semiconductor

Denix writes "MIT's TechnologyReview has an interesting article on a silicon-based "ion trap" in order to host a "qubit." The Ion Trap technology 'uses electric and magnetic fields to isolate a charged particle from its environment -- a prerequisite for exploiting the temperamental quantum properties of electrons."

Excellent

[Anonymous Coward]

Now we have to hope that quantum computing comes out in time for Windows Vista.

Re:Excellent

[Kickboy12]

We'll definetly need it!

Re:Excellent

[dascandy]

You swapped two words: Now we have to hope that [b]Windows Vista[/b] comes out in time for [b]quantum computing[/b].

Re:Excellent

[Bewbewbew]

The thing is, I count four words. Yes, I'm a pedant.

Re:Excellent

[jlebrech]

Not to mention DNF will have to have to be totally redesigned from scratch right before its released, puttin it back another 10 years.

Obligatory Cosby Quote...

[robvs68]

A qubit! What's a qubit?

How will they be programmed?

[ReformedExCon]

I wonder what sort of programming language would be needed to manipulate such esoteric devices.

Re:How will they be programmed?

[moon-monster]

I wonder what sort of programming language would be needed to manipulate such esoteric devices.

... and how long it will be before someone ports Doom to it.

Re:How will they be programmed?

[jcuervo]

Sure, but anyone with a computer newer than yours may have already killed you by the time you boot up.

Or they may not have. He won't know until he observes it.

Re:How will they be programmed?

[LiquidCoooled]

If its doom we are talking about, you can't observe it, because everyone knows you don't have duct tape in a quantum environment.

Re:How will they be programmed?

[m50d]

C. For the simple reason that if they don't support C, noone will use them. Look at what happened to lisp machines.

Re:How will they be programmed?

[m50d]

AIUI you lost all the advantages of the machines (memory management etc.) by doing that, partly because you can't use such things in C, but also because of the fundamental differences between the two languages. So when you compared them at running C programs, the lisp machines weren't worth it.

Re:How will they be programmed?

[Stan Vassilev]

Thing is, if you program it with the toolset of C, it'll not be better than IBM XT at blazingly fast 4MHz (approx).

The quantum computers require totally different approach to programming... So that's why you won't play QuantumQuake very soon.

Re:How will they be programmed?

[m50d]

Agreed. But if QCs ever become popular I'd bet on it being handled by bolt-on extensions to C (qint i = superposerange(0,1e6);) rather than a new way of programming.

Re:How will they be programmed?

[Ironballs]

BASIC will do it

Re:How will they be programmed?

[FleaPlus]

The quantum programming [wikipedia.org] wikipedia article has a link to a rather neat paper by Simon Gay, titled Quantum Programming Languages: Survey and Bibliography [gla.ac.uk]. This seems to give a pretty good overview of current thinking regarding what sorts of programming languages would be appropriate for quantum computation. The abstract:

The field of quantum programming languages is developing rapidly and there is a surprisingly large literature. Research in this area includes the design of programming languages for quantum computing, the application of established semantic and logical techniques to the foundations of quantum mechanics, and the design of compilers for quantum programming languages. This article justfies the study of quantum programming languages, presents the basics of quantum computing, surveys the literature in quantum programming languages, and indicates directions for future research.

He has the bibliography, complete with paper links available here [gla.ac.uk].

Re:How will they be programmed?

[FleaPlus]

I hope you wouldn't compile a quantum program on a quantum computer. Imagine bugs that aren't even in your source code...

Or a Schroedinbug [wikipedia.org]!

Schroedinbug is a term used in software programming to describe a computer bug that is not discovered, but shows up after somebody reads the source or uses the application in an unusual way. The program then stops working for everyone until the bug is fixed. Although this sounds rather impossible, some programs have carried latent schroedinbugs.

Re:How will they be programmed?

[Black.Shuck]

I wonder what sort of programming language would be needed to manipulate such esoteric devices.

Esoteric? ReformedExCon, Ion Traps are not a food.

Re:How will they be programmed?

[n54]

http://www.dict.org/bin/Dict?Form=Dict1&Strategy=* &Database=*&Query=Esoteric [dict.org]

Sorry if I missed a joke :)

Re:How will they be programmed?

[Black.Shuck]

Aye, it was a Family Guy reference. :D

Important point

[Sockatume]

Ion traps aren't particularly clever in themselves, but making them small- and mass-producing them- is important for quantum computing, which is where the research in the article is pointing.

Don't we have enough qubit styles already?

[C.]

Instead of addressing mass-production issues, maybe it would be time to put up a working quantum computer and see if it turns out useful. I doubt the goal of "10'000 qubits before it is useful" is impossible with today's technology.

Anyone remembers those computers that filled a whole room, and cost the price of a small country?

Re:Don't we have enough qubit styles already?

[Sockatume]

True, but if you want a working 10000 qubit machine to experiment with, you'll save a lot of time by mass-producing them on a small scale instead of machining them one at a time.

Re:Don't we have enough qubit styles already?

[C.]

Thanks for the clarification... I hate it when a reporter says "can't mass produce" rather than saying "doesn't scale".

For those interested

[Big Nothing]

Read up on Ion traps at: http://www.ionsource.com/links/iontrap.htm [ionsource.com]

Also, Wikipedia has quite a bit of useful information, especially regarding Paul traps: http://en.wikipedia.org/wiki/Ion_trap [wikipedia.org]

Re:For those interested

[StikyPad]

Don't forget these! [sharperimage.com]

Quantum Computing and Ghost Busting

[mystic_mind]

This is cool stuff But when someone figures out how to trap a ghost with this technology......now we're talking. And, will someone please teach me how to write an equation for ghost so we can finally fix them.

Re:Quantum Computing and Ghost Busting

[meringuoid]

But when someone figures out how to trap a ghost with this technology......now we're talking.

You want to store ghosts on computer chips? Hmm. Things could get really very interesting.

Aeria gloris, aeria gloris... :)

Re:Quantum Computing and Ghost Busting

[mystic_mind]

Actually there's someone singing his soulful best on this machine who had actually passed away earlier this year, God Blesss Him.

So if you think about it: There is actually a Ghost-in-the-Machine!

Will he be trapped there forever ?

Well, this is a new machine.

My last one lasted about two years before it started dragging .........

What happens to the ghost if the memory is a flash drive but the system wont ever boot up again ? Will he be trapped for eternity in a toxic wate dump ? Does it matter

Is it useful for the masses

[Stan Vassilev]

You know I've read lots of interesting application for quantum PC-s, but they are all esoteric, like hacking encryption keys or in scientific supercomputers for .. well simulating quantum level events :)

The progress is certainly interesting, but I everything shows this won't be the future of the mass computing, where running multimedia and office application are more practical usage of technology, and not especially suited for quantum computers (cuz most of you know quantum computers aren't just super fast computers.. they're just .. well odd).

For personal computing we gotta be looking into nano-technology, which also would be compatible with today's PC architecture (i.e. nano RAM in a laptop or nano HDD with SCSI interface for example).

Re:Is it useful for the masses

[MichaelSmith]

The progress is certainly interesting, but I everything shows this won't be the future of the mass computing, where running multimedia and office application are more practical usage of technology

multimedia and office applications are old technology and I agree that they don't require new tech. Back when 8 bit CPU's gave way to high tech 16 bit stuff I read an article about how 8 bits were better because most data (characters) came in 7 bit chunks.

The world has moved on from those days and will continue to do so. How about a search engine which indexes literally every bit of data in the world and uses a massively parallel quantum search engine to continually run searches and give answers in milliseconds? How about simulating whole communities of scanned minds in a simple chunk of reliable hardware? I can see lots of applications but they won't be the killer apps for the future.

Its funny. 20 years ago I agreed with that guy who said 16 bits would never catch on...

Re:Is it useful for the masses

[Stan Vassilev]

Well you're correct: working with databases WILL be one of the strong points of quantum computers. However as for the 8-bit 16-bit argument.. 16-bit is an upgrade to 8-bit technology that still can address content in bytes, you still run linear commands and get predictable results out of it.

Quantum computers are not evolution from computer technology, they're an entirely new beast. It's not even like PowerPC vs Intel or anything. You can't just "port" programs to it that are made for regular PC-s.

They are also not better in data throughput or speed as a technology on their own. They use aspects of quantum mechanics (which scientists still can't explain why they happen in first place) to run very specialised set of tasks through it and obtain results that'd take years of loops on a normal PC to compute.

I still don't see it in my mobile phone or PC, was my point.

They're not really super fast.

[Poromenos1]

Quantum computers can factor a number in polynomial time. That's not because they're really fast (they're not doing it in exponential time very fast, but in polynomial), it's because they have quantum properties. To factor a number, they convert the problem into a problem of finding the periodicity of a function, and then exploit the computer's ability to be in all states at the same time, and thus find the period very fast.

They have the potential to do stuff that we do slowly now much faster, but I don't t

Re:They're not really super fast.

[Stan Vassilev]

Indeed my point.

BTW this is why I always thought calling them "computers" is misleading. When you say computer you expect the full thing, being able to handle any PC tasks you throw at it.

I'd rather call it Quantum Processing Unit (QPU), and just like FPU before it it'll play together with the core CPU, not replace it, since you still need a "normal" CPU to display the interface, code editor, to compile code and to show the results of the quantum computations of the QPU.

You can't just grab a QPU and adapt it to do all tasks a moden CPU can. They're just .. different thing.

Re:They're not really super fast.

[Stan Vassilev]

With regular CPU's we have smart caching, brand predictions, optimizations, parsing of machine commands into simpler micro commands and what not and what not. This has all been designed and tested on the current generation of technology.

We're still fighting to get a QPU right enough so it can do math on just few qbits (i.e. few bits). Modern CPU simply does modern CPU job better than the first few generations of QPU will.

CPU's strengths are QPU weaknesses and vice versa.

Re:They're not really super fast.

[Buzz_Litebeer]

It doesnt invalidate his argument, since the basic thing (NAND / OR GATES) etc can be emulated in the quantum sphere, and probably on a larger scale, you could "emulate" processors in a quantum computer, or just go out and do something more natural as a new language, all he while keeping all the nice interfaces etc...

Just when is the issue, when...

Re:They're not really super fast.

[Stan Vassilev]

It's possible to run Linux on certain calculators as well but come on... I mean come on really :)

We're not talking theoretical capabilities here, sorry if I made you think so.

OT, you can see people trying to emulate the Doom enigne with canvas in Firefox.. It's certainly possible but the results are well.. not worth the hassle.

But will it run...

[wesley96]

For the obligatory 'but will it run Linux?' question, here's an answer I heard:

Being a Quantum computer, it can both run Linux, and NOT run Linux at the same time.

Re:But will it run...

[TheSpoom]

It dual boots?

Re:But will it run...

[C.]

Yes it dual boots - both OS'es simultaneously, but when you first look at it, it chooses a state and shows you a single OS.

(for those who aren't in the know - look up what happens to Schroedinger's cat ... http://en.wikipedia.org/wiki/Schroedinger's_cat [wikipedia.org] )

Re:But will it run...

[Hurricane78]

LOL! Finally someone else who gets it! That's why i love slashdot! Inside jokes ROCK! Even more when nobody here thinks that they're inside, but that the rest of the world just is stupid. (wich is pretty much the truth. ;)

Re:But will it run...

[StikyPad]

Right, but you never know what OS you're going to get until after it boots.

That, and you can either use your OS, or determine which OS you're using, but not both.

Good science, duff reporting

[Richard Kirk]

The devices are made of "Gallium, Aluminium, and Arsenide". The stuff may be called Gallium Aluminium Arsenide, but the element is called Arsenic. If they send you down to stores to get a jar of Arsenide, expect to get left-handed screwdrivers and tartan paint too.

So, the reporter doesn't know his periodic table? I bet he's red-hot at quantum physics, though. Really brilliant and highly trained minds sometimes skip over the basic stuff, yerknow.

Bah.

Re:

[account_deleted]

Comment removed based on user account deletion

Re:Regarding the story title...

[MattSausage]

I was wondering how people could work anti-MS comments into an article about ion traps.. but I'm new here and I should have known it could be done easily. Just add the word innovative and we're done.

Re:

[account_deleted]

Comment removed based on user account deletion

Re:

[account_deleted]

Comment removed based on user account deletion

Solar wind problems?

[digitaldc]

The chip is placed in a vacuum, which then gets injected with a vapor of cadmium ions. When the appropriate voltages are applied to the electrodes, a cadmium ion with a free electron becomes trapped, floating between the cantilevers above the etched hole. In order to actually use the atom's free electron for computation, Monroe explains, the ion must be probed by a laser beam that reads the electron's spin state.

http://science.msfc.nasa.gov/ssl/pad/solar/sun_win d.htm [nasa.gov]

Is it possible for solar wind to affect the ions and electrons making these calculations?

Re:Solar wind problems?

[Anonymous Coward]

Solar wind only affects machines with Windows.

What about photon bits

[EBFoxbat]

I was hoping for an ion drive for my spaceship before an ion bit in it's control system but we make due with what we have. Having not read any of the linked papers, is there a binifit to this ion bit over a photon based bit?

Re:What about photon bits

[X3J11]

http://www.space.com/scienceastronomy/solarsystem/ deepspace_propulsion_000816.html [space.com]

Ion drives. Apparently they've taken 60 years to get here, they don't go very fast, but they're very fuel efficient. And yes, I'm off-topic. :(

Shor's Algorithm

[eldavojohn]

One of the most important caveats of today's progress in quantum computing is Shor's Algorith [imsa.edu].

Why hasn't quantum computing gone further? Well, first you need to know that it requires your qubit to be tied to nearby qubits. When done with electrons, this is difficult because decoherence sets in very quickly.

In the end, they can "compute" with this string of qubits by bathing it in a certain frequency wavelength. What comes back are the multiple waves with the frequencies of all the prime factorizations of the initial frequency. The initial frequency cannot be greater than 2^(# of qubits).

The information I am relaying to you is from George Johnson's [talaya.net] book, A Shortcut Through Time [amazon.com]. Which is quite good.

I would also like to point out that the United States Government Lab in Los Alamos [lanl.gov] has done considerable research [lanl.gov] regarding this.

As a citizen of the U.S., you are funding this project so you have paid for and are entitled to read about their discoveries [fas.org] and I encourage you to do so if you have the time.

The reason for all this research?

"Many public key cryptosystems, such as RSA, will become obsolete if Shor's algorithm is ever implemented in a practical quantum computer."

~ From the Wiki Talk on Shor's Algorithm [wikipedia.org]

Re:Shor's Algorithm

[Luyseyal]

So... we just need a new quantum encryption algorithm called "anti-Shor". Problem solved! :)
-l

What's a qubit?

[JonKatzIsAnIdiot]

Riiight.

What's a qubit?

Zwo-pah ... zwo-pah ... zwo-pah

(apologies to Bill Cosby)

Re:What's a qubit?

[What me a Coward]

Well odviously it's someone these folks wanna host....

    Perhaps a relative of Qbert? maybe his son....

  qubit! The next generation :D

Other applications of cheaper ion traps

[brianf711]

There are other important applications for cheap ion traps as well, not least among them is in the field of biology / proteomics and perhaps chemistry. Ion traps are used in mass spectrometry and may be able to be adapted for desktop mass spectrometers.