What happens as circuits get so small that ...

[MathWizard]

So if most of the conduction electrons moving around in a circuit come from the metal conductor, at a given temperature. How small can the circuits be, (or how low of a temp), before you just start having so few electrons to move around, that it's not like a river of electrons, but more like discrete events, like buying a bottle of water once a day.

At some point that just becomes quantum physics, and atomic experiments, atomic machines.

But at room temp, how small of a capacitor or other part could they make before it's just get pointless for what people would call a normal circuit, because there's so few conduction electrons ?

[ppTRN]

In the semiconductor world it's always been about quantum phisics, just think about avalanche and tunneling electron transportation in diodes. There already are just "a bounch" of electrons moving around.

About passive components i really do not know exactly the production process and materials used on cutting edge modern technology

[barshatriplee]

At the atomic scale, the behaviour of electrons is governed by quantum mechanics rather than classical mechanics. For practical purposes, there is a limit to how small circuit components can be made before the circuit ceases to function as intended. This limit is currently being pushed by researchers and engineers in the field of nanotechnology,

[Berni]

You already have circuits that are sensitive to individual electrons moving around.

When you get to really low currents (like sub femto amps) you actually do have handfuls of electrons moving down your wire. There are "photon counter" light detectors that mostly function trough a electron getting knocked off something and triggering a process that produces a measurable pulse of current.

Some types of signal noise we see are also just a macroscopic manifestation of the quantum nature of electrons.

However the challenge in making circuits very very small is not that we wouldn't have enough electrons to send trough them. The actual challenge are the quantum effects that govern how the electrons act. The idea of electrons being tiny balls that flow trough wires is not quite what they actually are. The electrons are actually a cloud of probability that determines how likely the electron is to be found there. There is a possibility an electron from inside your TV will suddenly appear on the moon (the probability of that is really really really REALY small, but is not 0). However once you bring wires close enough (nanometers close) you get more and more into an electrons probability cloud, up to where the chance of the electron being in the other wire starts becoming high enough for that to actually happen on a regular basis. This is called quantum tunneling and it basically makes electrons 'teleport themselves' over tiny gaps because the electron is not quite sure where exactly it is. So at some point a very very tiny circuit would have too much leakage between wires to be able to function correctly.

[radiolistener]

the issue is not electrons, but electromagnetic field. When circuit is too small it's hard to avoid cross interference from two circuit blocks placed together. Such limit is already reached in modern chips

[Gyro]

The number of electrons sitting on a floating gate in a Flash memory really isn't very many. Getting them on there and off again therefore only involves a very small electron flow, under the influence of an electrostatic field. Bring on Fowler-Nordheim tunneling.

[TimFox]

When the length of a conductor along the direction of current becomes short compared to the mean-free-path for electron scattering in the material, the linear approximation (Ohm's Law) for current density as a function of voltage gradient fails.  This is considered to be "ballistic conduction", as opposed to "diffusive conduction".  See:   https://en.wikipedia.org/wiki/Ballistic_conduction

[Terry Bites]

As geometries apraoch the atomic scale quatum effects, particulary "tunneling" make it impossible to close the gate.

Its worth noting that current is not actually electrons moving around.
This idea is the most oft repeated falacy about electicity.
It's a metaphor and just like the water analogy, but its not what's really going on.
Current is defined as charge per unit time. Not electrons per second.

It fine for visualising circuits, but it all falls apart when you down to the physics of current flow.

[coppercone2]

research quantum fet

[Jim from Chicago]

According to the standard model of particle physics, an electron is basically a perfect little sphere which has mass but zero volume. Well, that is just incoherent on its face. That can't be true. It's a useful model, but it's a fiction. Physicists also describe electrons as "clouds of probability".

My preferred theory is that what we call "electrons" are actually tiny little demons that haunt the world and occasionally act in your favor as an engineer. I mean, a "cloud of probability" is pretty spooky. Sounds a lot like a demon. This theory really comes in handy when you need to explain failures to management.

[MathWizard]

As geometries apraoch the atomic scale quatum effects, particulary "tunneling" make it impossible to close the gate.

Its worth noting that current is not actually electrons moving around.
This idea is the most oft repeated falacy about electicity.
It's a metaphor and just like the water analogy, but its not what's really going on.
Current is defined as charge per unit time. Not electrons per second.

It fine for visualising circuits, but it all falls apart when you down to the physics of current flow.

In terms of average motions of electrons interacting with the atoms, and other e's, with drift and diffusion, isn't the average speed of an electron like um/s or mm/s in a typical circuit ? And it's just a conversion from avg. carrier motion to saying x # of e's / or holes moving ?

I guess a Coloumb of e's isn't really that big. Yeah the gate charges inside some chips, really are getting so small, it's not like classical EM.

[Bud]

Infinitly small electronic parts are useless for the majority of applications because they would not possess properties values required , e.g. resistance, capacitance, inductance. Niche areas such as microwave applications may use some of it but still require sufficient quantities of a parameter. And you cant build an audio amplifier using nano particle parts.

[coppercone2]

but uh in general this is the realm of quantum electronics, quantum photonics.

Structures like you describe maybe kind of like carbon nanotube electronics too. Good place to start studying is quantum carbon nanotube electronic circuits

[Berni]

As geometries apraoch the atomic scale quatum effects, particulary "tunneling" make it impossible to close the gate.

Its worth noting that current is not actually electrons moving around.
This idea is the most oft repeated falacy about electicity.
It's a metaphor and just like the water analogy, but its not what's really going on.
Current is defined as charge per unit time. Not electrons per second.

It fine for visualising circuits, but it all falls apart when you down to the physics of current flow.

In terms of average motions of electrons interacting with the atoms, and other e's, with drift and diffusion, isn't the average speed of an electron like um/s or mm/s in a typical circuit ? And it's just a conversion from avg. carrier motion to saying x # of e's / or holes moving ?

I guess a Coloumb of e's isn't really that big. Yeah the gate charges inside some chips, really are getting so small, it's not like classical EM.

Yeah technically current is charge per second. But electrons have a charge so you can easily convert between the two. You can have current flow by moving protons around too, but those tend to be part of atoms and they have a lot more mass so they don't move around so easily. So when you think of electrical current in a wire, it is the electrons moving.

You can still have fast moving electrons too. Like inside a CRT the electrons are being thrown at the phosphor screen at speeds that are a significant fraction of the speed of light. The thing there is that each electron has to carry enough excess energy to excite the phosphorescent coating on the screen, so the solution is to throw them really really hard using a very strong electric field (hence high voltage on CRTs). However the number of electrons is small, so the average electrical current in the electron beam is still tiny.

Yeah the charge of a single electron is tiny at 1.6*10^-19 coulombs. But electrons are very small so you can have a lot of them in one place. The electromagnetic force is also very strong. Throwing something off a building, it is gravity taking a few seconds to accelerate it into the ground, but at the bottom the electromagnetic force holding the ground together stops it within miliseconds. So for this reason electrons only need to move very slowly in a wire to create a large current, while they don't experience the kind of "inertia" that balls would. The electrons are just near instantly rearranging in wires to conform to the electric field around them.

This is why you can think of electrons as not actually being the thing that carries power along wires, but instead it is the fields between wires that carry power. But best not go there as there are a lot of opinions on what is the "correct way" of describing it.

[Terry Bites]

I understand that QFT is the current model (since the 70's) and the electron is not a sphere and cannot be described as an object in the conventional sense. Its an excitation of quantum fields. Shape has no meaning, its a set of properties.

[m k]

I remember that 30k or 50k ferromagnetic atoms are needed for storing a bit, overall shape is not so relevant.
It's not that less is not enough, but then the definition must be changed.

Observing a single photon is also possible, but it also has a surrounding definition, like maybe a neutrino detector.

[tszaboo]

Just look into thermal noise, and see that we are already close to theoretical limits of what we can do, due to quantum effects.