A questions for quantum electronics

[MobiusPizza]

I read a book about quantum dots; I was amused by the concept. As I haven't even go to university yet I am still a noob in this field.

As we know, electrons can 'tunnel' through seemly a impermeable barrier. An electron outside a quantum dot can tunnel into it and comes out the other side. an electron borrows energy to 'teleport' and repay it as soon as the tunnelling process is completed.
The process is almost instantaneous.

So I wonder, why can't we build an array of quantum dots close enough in vicinity of an electron wave. When dots are joint up it would give a sine wave of electron wave where peaks are inside each quantum dot. To those who doesn't know electron waves are probability of where a electron would be in if we are aware of the particle-wave duality of electrons. Hence if we have a electron in one end of this quantum dot array its existence can be detected at the other end instantaneously because of the tunneling effect. It's like the quantum array has extended the electron wave into a sine like wave.

If we then apply voltage to one end, the wave would be pushed to one and and electron would eventually come out the other end if we have a wire or something connected there and voltage is high enough.

Having said that, if we have bunch of electrons, an indenfinitely long quantum array, we effectively created a room temperature superconductor. (There's no resistance to tunneling as the effect is instantanoues and consume no energy) Am I wrong?

 

[CycloWizard]

The simple answer is that quantum dot production technology isn't that far yet. They are extremely hard to produce in a regular fashion, and the smaller you make them the harder it becomes. Getting perfectly uniform spacing is equally difficult. The scale you're talking about is much smaller than what most people are even considering with quantum dots at the moment, so it will be a while at least.

 

[MobiusPizza]

I heard peope had even been using quantum transistors?

 

[f95toli]

It is actually possible to build arrays which can almost do what you describe if you use tunnel junctions (I don't know if you can do it using quantum dots, i don't think it has been done), you can actually get solitons traveling down the array. This is used for counting electrons (current standards) and there was a paper in Nature about this a few weeks ago (the work was actually done in my research group, but I am not directly involved in that work).

However, you can not use arrays like this to transfer information faster than the speed of light, tunneling time (or to be more precise, the change in probability density) like everything else is limited to speeds below c. Hence, tunneling is not instantanous.

The last point is a bit tricky. I am not sure wheter or not you could build something lika an artificiall superconductor out of QD; however I am quite sure it wouldn't work at room temperature and not if you made the array longer than a few elements. The reason is that in order to have coherent tunneling you need to make sure that that the relevant energies in the system (what is the relevant energy depends on the system, it can e.g. be charging energy) are much higher than the thermal energy KbT. Hence you need to work at very low temperures, the work I mentioned above was done at about 0.03K.

Finally you have the problem of decoherence due to coupling to the enviroment, basically what this means that is that if you make things too big (or to be more correct: not sufficiently decoupled from enviromental degrees of freedom) all quantum effects will be smear and the system will start to act classically. This is the reason why quantum computing is so hard.
The reason why superconductors can be made as large as you want and still be superconducing is that the Cooper pairs are bosonic which means that they all can be in the same ground state (which is impossible for ordinary electrons), cooper pairs form when you cool the superconductor below the critital temperature. Furhermore, the pair-formation also creates a gap in the energy spectrum which protects the superconductor from decoherence.

 

[MobiusPizza]

Thanks for the deep analysis

hmm does that lead to quantum transistors works only in low temperatures?
If that's the case, my hype in this area is useless pointless .-.

I opt for studying Electronics Engineering just because of this.
Now I might switch to optoelectronics XD

 

[CycloWizard]

Originally posted by: AnnihilatorX
Thanks for the deep analysis

hmm does that lead to quantum transistors works only in low temperatures?
If that's the case, my hype in this area is useless pointless .-.

I opt for studying Electronics Engineering just because of this.
Now I might switch to optoelectronics XD

I would never recommend picking a field based on one particular case of anything, as the chances of you ever getting to work on that thing are very, very low. 😉

 

[f95toli]

Originally posted by: AnnihilatorX
Thanks for the deep analysis

hmm does that lead to quantum transistors works only in low temperatures?
If that's the case, my hype in this area is useless pointless .-.

I opt for studying Electronics Engineering just because of this.
Now I might switch to optoelectronics XD

It depends on what you mean by a "quantum transitor"?
You can (in principle) use quantum dots at room temperatures as well, as I mentioned the "trick" is to make sure that the relevant energies are higher than KbT. However, I am not sure wheater or not you can operate a QD based transitor at room temperature.
I know of some experiments on optical properties of QD that have been done at room temperature.

Maybe I should mention that even in optoelectronics people sometimes work with low temperatures. A device that works at e.g. 50K is still usefull, cryocoolers are cheap so if the performance is good enough cooling should not a problem.

Generally speaking the main problem for any new technology is to overcome the Si-obstacle, unless it is made form Silicon and CMOS based the industri is not interested. Even very well-established tecnologies like those based GaAs and GaN have the same problem, they are only used when there is no way to solve the same problem in Si (despite the fact that potentially the performance is much better than Si).

So quantum dots have the same problem as superconductors and nanotubes, being better is simply not enough.

 

[RaynorWolfcastle]

I know for a fact that there are optical devices based on quantum well lasers; these are usually built using some InGaAsP semiconductor. They are used in the fabrication of vertical cavity surface emitting lasers (VCSELs, pronounced vixels). Essentially, the quantum wells are the gain medium for a tiny semiconductor laser. I'm pretty sure that there are VCSELs that have been made using quantum dots as well since the they are very similar. Typically, the way they are used is to make several quantum well/dot lasers and operate them in parallel rather than using a single, larger device.

I'm not an expert on them but I the (very basic) theory behind quantum well lasers is the thickness is small very (comparable to the thermalized electrons' effective wavelength) so the device must be analyzed as a quantum device. For quantum dot lasers, as far as I understand all three dimensions are very small, which makes things even more complicated. In any event, the general advantage is that the threshold current needed for lasing in these devices is very low and they have high quantum efficiency compared to devices with much longer gain media.

VCSELs are getting a good deal of attention because they are relatively cheap to produce and so would work well in a LAN or MAN. In the longer term, they show promise as becoming cheap, fast devices for system level interconnects, and eventually they could even replace electric connections on printed circuit boards.

 

[patentman]

Originally posted by: f95toli
It is actually possible to build arrays which can almost do what you describe if you use tunnel junctions (I don't know if you can do it using quantum dots, i don't think it has been done), you can actually get solitons traveling down the array. This is used for counting electrons (current standards) and there was a paper in Nature about this a few weeks ago (the work was actually done in my research group, but I am not directly involved in that work).

F95toli:
I read that article. That was some nice work. Did you guys recently receive a patent for an ordered array of quantum dot tunnel junctions using ferromagnetic nanoparticles coated with a thin layer of carbon (with the carbon layer is the tunnel barrier)? Specifically, US Patent #6730395 (LINK
for "Magnetic tunnel junction using nanoparticle monolayers and applications therefor?"

If so, good work, that was one of the neatest things I examined while I was an examiner at the PTO.