Was 14V LED question / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

Was 14V LED question

first previous 2 3 4 5 next last

Move Thread

LAN_403

Conundrum

Wed Sept 30 2015, 02:06PM

Registered Member #96 Joined: Thu Feb 09 2006, 05:37PM
Location: CI, Earth
Posts: 4062

Some early examples of blue Nichia LEDs sold at Tandys around 1997 did emit quite a lot of UV, if overcurrent of about 0.5A for <50 us applied.
I actually tried this experiment to wipe a PIC12 and did manage to erase a few bits but gave up waiting for it.

Back in ye olden days it was customary to check erased chips at more than one voltage as sometimes "weak" erased bits had remanent memory of their previous state at some lower or higher voltage depending on the manufacturer.

Bored Chemist

Wed Sept 30 2015, 03:23PM

Registered Member #193 Joined: Fri Feb 17 2006, 07:04AM
Location: sheffield
Posts: 1022

Ash Small wrote ...

Bored Chemist wrote ...

The efficiency does, however, depend on the temperature in the way one would expect if it's relying on thermally excited electrons having the enrgy to get over the barrier, fall down, and emit light.

This is what the conventional, simplified, mathematical models tell us happens, BC, however I look at it slightly differently.

I argue that the 'threshold energy' is the energy required to 'pull' an electron from the already depleted region on the collector side of the junction (the 'base region'), this electron then becomes 'base current'. The presence of of the 'hole' left behind 'neutralises' the potential gradient at the junction, allowing electrons to cross the junction. This flow continues, in the conduction band, until an electron 'falls' into the hole left in the valence band, releasing a photon, and stopping the flow across the junction, thus re-introducing the 'potential gradient'.

I base this reasoning on the fact that no conduction band electrons flow until the base current starts flowing (it possibly becomes easier to visualise as ' charge flow' (coulombs) at this point)

I accept that my reasoning doesn't include 'quantum wells' at this point, so I'm expecting to be proved wrong (how else do you learn anything?)

EDIT: just to clarify, once an electron grosses the junction into the conduction band in the 'base region', this probably also re-established the potential gradient across the junction, until it either leave the base region in the conduction layer and flows to the collector, when, presumably, another electron can cross the junction. This stops when an electron falls from the conduction band into the 'hole' left in the valence band. The more 'holes' in the base region, the greater the current.

Again, just my ideas, but this reduces the process to effectively the same mechanism as a thermionic valve.

Why are you talking about a base and a collector?
Diodes (light emitting or not) don't have them.
And, thermionic valves have (at least) 3 electrodes.
Have you somehow got phototransistors muddled into this discussion?

Ash Small

Wed Sept 30 2015, 04:03PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

Bored Chemist wrote ...

Why are you talking about a base and a collector?
Diodes (light emitting or not) don't have them.
And, thermionic valves have (at least) 3 electrodes.
Have you somehow got phototransistors muddled into this discussion?

Surely a diode works on the same basic principle as a BJT, just with the base and collector combined into a single terminal?

You still have a conduction band and a valence band, surely?

The terminal itself actually forms another 'metal-semiconductor' junction, but that's not really relevant at the moment.

You can, in principle, use a BJT as a diode if you connect the collector and base together.

EDIT: It would appear from what I've read so far, that the 'quantum well' has something to do with 'valence band' holes and electrons, although I've only skimmed the surface so far.

Bored Chemist

Wed Sept 30 2015, 05:58PM

Registered Member #193 Joined: Fri Feb 17 2006, 07:04AM
Location: sheffield
Posts: 1022

Ash Small wrote ...

Surely a diode works on the same basic principle as a BJT, just with the base and collector combined into a single terminal?

Why start with the complicated version rather than just a single pn junction (OK, it's not that simple with an LED).

Ash Small

Wed Sept 30 2015, 06:43PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

Bored Chemist wrote ...

Why start with the complicated version rather than just a single pn junction (OK, it's not that simple with an LED).

Two reasons, firstly, we all have a pretty good idea of how a BJT works

And secondly, you have the same basic principle. Both the valence band and the conduction band play a part, unless the gain is zero, in which case you can use the base-emitter of a BJT as a comparison. I'm just using a BJT for comparison really, it's looking like there 'may' be a correlation between the base current in a BJT and the 'quantum well' in an LED, at least, from the perspective that I view a BJT from.

It just helps me to visualise the mechanism easier, if I can 'relate' to it. I was taught that before you attemp the maths, you should draw a picture (often a graph), as it's then much easier to apply the maths.

I've always considered 'conventional semiconductor theory' to be oversimplified to the point where you have to use compensating mechanisms like 'leakage current', etc., although it does make the maths pretty simple, within limits

Ash Small

Sat Oct 03 2015, 09:28AM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

I suppose the reason I made the analogy with a BJT is mainly regarding the current flow reported above, where photons are visible down to ~2V with a blue LED, below which point current can still be detected flowing, but no photons can be detected.

While Wikipedia and other articles talk about 'electrons from the conduction band falling into quantum wells in the valence band', none of them clearly state whether or not ALL the electrons that pass through the LED fall into the well, or whether some continue through the LED in the conduction band.

If they all fall into the well as they traverse the LED then every electron passing through the LED must release a photon, where as if some remain within the conduction band, then not all the electrons that traverse the LED produce photons.

Any links to articles that 'shine any light' on this will be much appreciated

EDIT: I assume, using the BJT analogy again, some 'base current' has to flow in order to establish the well in the first place, assuming that 'holes' need to be 'injected' to create the well. The holes are obviously created by removing electrons from this region, and I'm assuming that no photons are produced while the well is being established.

If I'm incorrect can someone please correct me.

Bored Chemist

Sat Oct 03 2015, 05:56PM

Registered Member #193 Joined: Fri Feb 17 2006, 07:04AM
Location: sheffield
Posts: 1022

I think you may have found the problem with using a BJT as an analogy- there is no third electrode from which holies can be injected.
At a sufficiently low current there are other de-excitation mechanisms that compete with photon emission, but, at least as far as that paper measured, the fraction of electrons that give out light is constant (below a mA or so).

It's entirely possible that the reason you can't detect the photons from a blue LED at 2V is the they eye is pretty good, but it isn't perfect.

Ash Small

Sat Oct 03 2015, 11:42PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

The closest I've yet found of a simple description of how a quantum well works is Wikipedia, which states ' Charge-carriersâ€”electrons and holesâ€”flow into the junction from electrodes with different voltages.' Now, the only way 'holes' can 'flow into the junction' is by pulling electrons out. A hole, by definition, is a 'lack of an electron'

If it isn't a 'virtual third electrode' doing the pulling, (analagous to the base in a BJT), what is pulling out electrons, and thus injecting holes?

Bored Chemist

Sun Oct 04 2015, 09:50AM

Registered Member #193 Joined: Fri Feb 17 2006, 07:04AM
Location: sheffield
Posts: 1022

As far as I can tell, the quantum well only seems important in blue LEDs, but the behaviour of the other colours all seem pretty much the same, so, while these wells are interesting, they aren't that important to the issue.

Ash Small

Sun Oct 04 2015, 11:53AM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

From what I've read, all LED's have quantum wells. Wikipedia also states that ' The charge carriers recombine in a forward P-N junction as the electrons cross from the N-region and recombine with the holes existing in the P-region. Free electrons are in the conduction band of energy levels, while holes are in the valence energy band. Thus the energy level of the holes will be lesser than the energy levels of the electrons. Some part of the energy must be dissipated in order to recombine the electrons and the holes. This energy is emitted in the form of heat and light.' It also says that the only difference between LED's and normal diodes is that LED's emit light whereas diodes emit energy in the form of heat (you'd expect this, it corresponds with the forward voltage drop, blue is over 3V, red is under 2V, diodes with a lower voltage drop emit at below the visible spectrum).

My understanding is that when a potential difference is applied across the junction, holes are 'injected' into the valence band (electrons removed). All the textbooks say holes need to be injected before anything happens.

This allows electrons to cross the junction in the conduction band (it upsets the equilibrium that was established when the junction was formed, allowing electrons to cross)

Now, this is where things get a bit vague, all the sources say that conduction band electrons in this region can 'fall into the well established in the valence band, emitting energy'.

Nowhere does it say whether all the electrons that cross the junction in the conduction band fall into the well in the valence band, or if some continue in the conduction band, analagous with the collector current in a BJT.

Going back to my analogy with a BJT with the base and collector connected together, exactly the same mechanism occurs, you establish the 'well' by injecting holes (removing electrons) in the valence band, electrons then cross the junction in the conduction band, some electrons fall into the holes in the valence band, to become 'base current', while others with sufficient energy to traverse the region without falling into a hole continue in the conduction band as 'collector current'.

I'm just wondering if all the electrons fall into the well, analagous to base current in a BJT, or if some traverse the region entirely within the conduction band, analagous to the collector current in a BJT. Obviously, only electrons that fall from the conduction band into the well in the valence band emit photons.

first previous 2 3 4 5 next last

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob