Transistor-laser breaks Kirchhoff's Law, rewrites textbooks

5/14/2010 4:07 AM EDT

Big deal. III-V transistors emitted light when dinosaurs still walked the earth.

And there's no need to rewrite Kirchoff's law either, as that great self-promoter Dr Feng postulates.

I submit Feng's transistor model/symbol is faulty and is missing a current sink node that accounts for the E/O conversion of current, much the same as an optical detector is modeled by adding a current source to a node.

With that, don't burn your college texts just yet. It's just hype, IMO.

5/14/2010 6:09 AM EDT

> "New text books will be needed to incorporate the transistor laser," said Feng.
> "Kirchhoff's Current Law should be redefined as Kirchhoff 's Current and Energy Law."

This is nonsense. For most circuit design purposes, the seeming "loss" of energy is completely irrelevant, because it is the Kirchhoff Current Law and Kirchhoff Voltage Law that lie at the heart of (analog) circuit simulation and circuit analysis. Conversion into light is in this respect only a side-effect, just as the universal conversion into heat is. There is zero loss of charge in the generation of laser light or heat. Kirchhoff's Current Law still holds, and conventional analysis and simulation techniques still apply. Action-at-a-distance (e.g., via light) can be properly modeled through nonlinear controlled sources without violating Kirchhoff's Current Law in any way.

Regards

5/14/2010 10:37 AM EDT

I really hate the way some scientists use bogus hype to gain some media attention. This oversexing of results was one of the main reasons I wanted to go into industry instead of staying in academia. In industry you only really need to get something to work and understand enough about it to keep it working long enough to meet the needs of the application. With science reporting you get ridiculous statements like this, and the one earlier this weak claiming a single grad student could produce more logic with DNA than the entire semiconductor industry regardless of the fact that there was no plan in place to actually "wire up" these logic elements and put them into a package that could actually end up in a device.

Charge is always conserved even when it comes to black holes.

5/14/2010 11:06 AM EDT

Nothing new. Yet another reason to rename KCL to KQL (Kirchoff's Charge Law). Charge is always preserved even if it is converted to a laser beam. Simple accounting, no of electrons preserved in a closed circuit.

5/14/2010 3:38 PM EDT

Somehow we have managed all these years to analyze circuits containing LEDs and diode lasers without the need for new laws. I question the novelty of anything that needs this much hype.

5/14/2010 5:37 PM EDT

I was lured by the eye-catching title to finish reading in the hope of learning the biggest discovery in post-Einstein era, only to be bored and insulted by the self-acclaimed accomplishment that does not fool anyone beyond their high school years.

What a shame and negative example for diligent researchers and engineers.

5/14/2010 6:19 PM EDT

KCL is just a narrow special case solution of Maxwell's equations and there have always been situations in which KCL did not apply.

New textbooks needed? I don't think so -- just a need to remember what we all learned in those old textbooks.

5/14/2010 6:49 PM EDT

Just closed my eyes and tried to dream when Mr.Kirchhoff made his statement to world how powerful and irritating feedbacks he got, after reviewing same fact again and again his statement became a law. Its true title of the topic is astonishing however without hitting words i dont think theyd be able to get any attention neither budget. The universe is established on such laws, which is just waiting to be discovered, we as human being should work and get knowledge of it. Personally i am also having serious doubts about this research however always open to see new ideas around since they are provable again at any time... To counterfellows who shared their ideas above, i have a question... What if the research is guiding us to another law? Just imagine what we could shape at out future with this?

5/17/2010 6:35 PM EDT

Null, my point was simply that when dealing with multiple forms of energy in the same system -- in this case both electrical and optical energy -- Maxwell's equations are really what govern. The authors themselves understood and stated that what was important in analyzing this device was conservation of energy, not simply conservation of charge. To make our lives easy, we take the divergence of Ampere's Law and combine it with Gauss' Law to come up with the simplified equation we know as Kirchhoff's Current Law. It is only meant to be used at low frequencies -- where the operating wavelength is much larger than the physical dimensions of the circuit. It might at first glance appear that KCL is being violated by this new transistor, but on closer observation, it appears that charges are being accelerated at the junction by the E-field associated with the optical output. This does not imply that charge is being created out of nothing. It simply implies that the low-frequency simplification we refer to as KCL is insufficient to fully analyze the conservation of energy in this circuit. This is also not the first electro-optical circuit for which that is true. It is somewhat disingenuous to say this transistor breaks Kirchhoff's Law and requires a re-write of the textbooks. These guys know better...

5/19/2010 5:03 PM EDT

This "transistor" DOES NOT break KCL, the model of the transistor is insufficient to describe its behavior (including losses and conversions). Why are you looking to break a fundamental EE accounting rule for any reason other than HYPE. As I said, most III-V transistors emit light, in fact anything that gets hot emits IR...oopsie - have we been wrong all these years? BS.

5/20/2010 12:53 AM EDT

Does that mean some electrons are getting converted to photons? That would violate a lot more laws than just Kirchoff's.

5/20/2010 3:39 AM EDT

The article is some late for 1-st April joke.

5/20/2010 6:41 AM EDT

Violation of Kirchoff's law seems unlikely - unless the device is emitting a lot of neutrinos as well - in which case we have proof of violation of charge conservation. Step aside Pons and Fleischmann!

Actually, all Feng and Holonyak appear to be saying is that, in order to model this device properly, you need to account for the energy emitted as photons - so they've "extended" Kirchoff's current law to include energy conservation. They hav'na changed the laws o' physics cap'n.

5/20/2010 7:32 AM EDT

But that has been always so. Energy is conserved when current heats up a wire in a bulb and photons are emitted. So how is it different now?

5/20/2010 8:36 AM EDT

One of these articels which only serve to raise attention
and help gettings funds.

Remember the memristor? A device which was said
to be the 4th basic element, like C R L memristor.

If the law of charge conservation is violated, then.,
maybe by some extreme shorttime effects, where
charge is squeezed together . Don't know whether this is doable.

The Kirchhoff statement 'the charge flowing into the node in a certain
time interval equals the charge flowing out of this node in the same
interval' somehow implies that the said node should not have any
charge storage (by whatever means) capabilities.

If a node could store charge , like storing hydrogen in certain
special metallic compounds, then Kirchhoff must indeed be rewritten
'the charge flowing into the node in a certain
time interval equals the charge flowing out of this node in the same
interval OR LATER ' .

This assumes that the stored charge is released somewhat later.

By the way, what about a transistor inside a flash memory?
Isn't there some charge getting lost when flowing from source to drain
and stored on the floating gate?

The total charge pumped into the transistor source is never fully
arriving at the drain, some of it goes onto the gate.

5/20/2010 10:54 PM EDT

Or is it a stunt to get attention and some funding?