Why don't electrons leave a negatively charged metal in air?

[ZapperZ]

Also, if there's only 5 extra electrons in the charged metal (as in an initial simulation), I think the image charge would be a very inaccurate model. Yes?

Why don't you solve the Laplace equation and show me that this is the case?

And what if we heat a sphere with 5 extra electrons. I think they will leave eventually, and form air ions like in the discharge.

I have no idea what this is. Have you seen 5 electrons leave a surface in a thermionic emission? Do you even know what you are talking about here? It appears as if you're just making things up as you go along. If this is what you are doing, then I'm outta here. I am obviously severely mistaken that this thread has anything to do with something realistic.

Zz.

 

[Ulysees]

> Why don't you solve the Laplace equation and show me that this is the case?

I don't know how, I'm asking. I believe the 5 would balance on an hexahedron.

This does not look like the continously populated mirror surface you mentioned to establish image charges. So I expect no mirror charges here, I see no mirror surface.

> Have you seen 5 electrons leave a surface in a thermionic emission?

We said electrons can leave if they get more energy than the work function, right?

Doesn't heating the entire object possibly raise the energy level of electrons?

 

[jostpuur]

I don't understand why the sparking behavior goes the way it goes, so I slightly regret my tone on by previous post. I don't know this matter fully myself, but I have basic knowledge of solid conductors and insulators (QM, periodic potentials, statistics stuff...) so I can recognize overly heuristic explanations.

In a metal the electrons roam freely throughout the metal. Current begins to flow because electrons are being "stolen" from one end of the conductor and so electrons from the other end move toward the newly created charge imbalance. This doesn't happen easily in insulators because in those substances the electrons are much more tightly bound to the individual moleucles and therefore it requires a much higher potential to liberate them. You don't need quantum mechanics to understand this.

When you first talk about metals, and then insulators, it almost looks like you are talking about solid insulators, but judging by your explanation on molecules you were talking about gas insulator and not solid insulator. They are different things, and I must disagree with your last statement about QM.

Even in solid insulator, individual electrons are highly mobile. The lack of electric currents is a macroscopic issue. So with solids it goes like this: Individual electrons are always mobile. If the solid is insulator, it is so because the available momentum eigenstates for electrons are such, that no macroscopic currents can arise. The solid is a conductor if it is not an insulator.

So why indeed is the air insulator then? Why cannot there be electron gas among the air?

 

[peter0302]

The question was why don't electrons jump from a positively charged metal to an air molecule. The reason is the same as why electrons don't jump from a positively charged metal to a rubber insulator. The valence shells of the insulators are generally full, and so a higher potential would be required to put an electron in the next higher shell than is available.

You CAN have conduction even in an insulator such as rubber if the potential is high enough. You CAN have conduction in the air if potential is high enough. Touch a doorknob after brushing your feet on the carpet.

The reaosn metals are good conductors is because the valence electrons generally roam freely from atom to atom, because the valence shells are usually NOT full, and so when there is a charge imbalance, to which the electrons are attracted, the resistance is low. Insulators have some freedom of electron movement as well, of course - but much less. Hence - resistence.

Again, you really don't need QM to understand this!

The other question was about whether electrons just move through the air. They generally do not. When there is conduction in the air, it is because there is such a high potential that the electrons do jump from atom to atom like they would in a normal conductor.

Another question was about old amplifiers - which were called VACCUUM tubes. Why were they called vacuum tubes? Gee, could it be because there is no air in them? Is there conduction going on there? NO. That is one of the rare instances of electrons jumping across a long distance from one atom to another. That is not conduction.

There's really nothing left to be said on this topic.

 

[Ulysees]

Another question was about old amplifiers - which were called VACCUUM tubes. Why were they called vacuum tubes? Gee, could it be because there is no air in them?

So engineers have achieved perfect vacuum? No, it's thin air in the tube.

And if we slowly add air molecules to the tube, then there will be fewer and fewer electrons reaching the anode. This has to be explained with collisions and ionisation that opposes the external field.

Is there conduction going on there? NO. That is one of the rare instances of electrons jumping across a long distance from one atom to another. That is not conduction.

So current in perfect vacuum is not conduction for you. Alright. If we add a molecule of air to the vacuum, is it conduction?

 

[peter0302]

So engineers have achieved perfect vacuum? No, it's thin air in the tube.

If there is air in the tube it is so little that the electrons do manage to avoid it. The air is not "conducting" electricity. The electrons are moving through space, attracted to the coloumb force of the positively charged anode, governed by Coloumb's law and the Schrodinger equation.

And if we slowly add air molecules to the tube, then there will be fewer and fewer electrons reaching the anode. This has to be explained with collisions and ionisation that opposes the external field.

Precisely. That's not conduction! That's air becoming ionized, i.e., _keeping_ those electrons. That's because the electrons would much rather be loosely bound to an air molecule than be unbound entirely. We also know this from QM.

Another way you can see this is obvious: when current runs through a metal conductor and then is shut off, the metal is no longer ionized. In your example, the air remains ionized.

So current in perfect vacuum is not conduction for you. Alright. If we add a molecule of air to the vacuum, is it conduction?

NO! Conduction is a macroscopic event defined by a large flow of electrons from point A to point B through a medium in which the electrons continuously bind to different atoms along their path of travel. But they still remain bound to their atoms the entire time.

That is not the same as an electron jumping from one piece of metal to another. The "electron cloud" in a vacuum tube is not "conduction."

And I want to make one more comment here. We don't need quantum mechanics to explain everything. Perhaps QM _can_ explain everything, but classical physics can explain certain things also, and this is one of the areas that classical physics is good at. And the reason for the confusion in this thread is because you are comparing electric current (which was a well understood phenominon before QM) with vacuum tubes, which do utilize QM concepts. You're mixing apples and oranges by bringing vacuum tubes into this. People were making electricity work before we even knew about electron states and themionic emission. Lightning and conduction are in one camp and creation of unbound electrons is an entirely different animal.

 

[jostpuur]

The reaosn metals are good conductors is because the valence electrons generally roam freely from atom to atom, because the valence shells are usually NOT full, and so when there is a charge imbalance, to which the electrons are attracted, the resistance is low. Insulators have some freedom of electron movement as well, of course - but much less. Hence - resistence.

Looks like cargo cult science.

 

[peter0302]

What does that mean? I'm a liar? What exactly is inaccurate about that statement? I suspect you don't even know the context of that phrase.

Sorry you don't like "overly hueristic" explanations but generally "why" questions call for such.

 

[Ulysees]

Hey jostpuur, no need to be so harsh. Physics is all about models, and a model is as good as the experimental measurements it matches.

Myself I like detail in models in order to imagine simulation software for them (ie dynamic equations where all motion is driven by forces, not abstract "tendencies" like electrons tend to do this, electrons tend to do that. Other people are satisfied with less detailed explanations, can't blame them, it's a personal preference.

---------------

Only thing is, Peter won't admit valence concepts from solids are of little use in thin air. Possibly any pressure of air can allow electrons to jump off the metal, a spark can occur at high pressure too. Eg in turbo engines.

Also, Peter you seem to have forgotten the issue of the topic, it's not naming convention (what we call conduction and what not), the issue is: "why doesn't corona discharge occur at all times?"

In other words, how close together do the air molecules have to be, and how fast moving, and how much voltage on a spherical conductor, for electrons to leave?

I think charge is lost through air at all times.

 

[Ulysees]

> I think charge is lost through air at all times.

Provided the work function is exceeded. What do you guys think?

 

[peter0302]

Small amounts of charge might be lost at all times but very little. If we're talking about a true "discharge" like lightning or turbo engines, there has to be a very high potential for this to happen. That is because _sigh_ _again_ air is a very good insulator.

[Edit]
Work function? Work function generally refers to the amount of energy needed before electrons will become UNBOUND from a metal. Such as thermionic radiation. Conduction / discharge is not free electrons. Electrons in conduction remain bound to the conductor.

Apples and oranges!

 

[jostpuur]

The reaosn metals are good conductors is because the valence electrons generally roam freely from atom to atom, because the valence shells are usually NOT full, and so when there is a charge imbalance, to which the electrons are attracted, the resistance is low. Insulators have some freedom of electron movement as well, of course - but much less. Hence - resistence.

Again, you really don't need QM to understand this!

According to the theory of energy bands, the electrons are assumed to be in form of delocalized Bloch waves. All electrons are mobile, but insulators are insulators because the average value of momentums of large number of electrons is forced to be zero due to the band structure. It doesn't make sense to speak about electrons moving from atom to atom in this model.

If you had simply said that "in conductors electrons are allowed to move from atom to atom, and in insulators they are not", then I could have responded more simply by saying that you are wrong and that the theory of energy bands gives the correct answer. However, now you explained that "because the valence band is not full, hence the electrons are allowed to move from atom to atom". What can I say to this? I cannot instruct you to find out about the theory of energy bands, because you already know about them. Your explanation was cargo cult science after all: The concepts of the well accepted main stream theories were being mentioned, but not being understood.

What does that mean? I'm a liar?

I'll try to keep this attack as an attack against your explanation, and not against you. Science is hard, you know: No mercy to the explanations.

 

[Ulysees]

Conduction / discharge is not free electrons.

Peter, you're getting stuck in terminology again, please don't. Allow me to say it a little more abruptly (no offense): I don't care what conduction is. What matters is what force keeps an electron from leaving.

 

[ZapperZ]

Conduction / discharge is not free electrons.

Peter, you're getting stuck in terminology again, please don't. Allow me to say it a little more abruptly (no offense): I don't care what conduction is. What matters is what force keeps an electron from leaving.

Wait. What do you think that whole point on the discussion on the image charge was for? The "work function", which is what prevent an electron from "walking out" of a metal's surface includes a substantial portion that is due to the image charge!

Zz.

 

[Ulysees]

Wait. What do you think that whole point on the discussion on the image charge was for? The "work function", which is what prevent an electron from "walking out" of a metal's surface includes a substantial portion that is due to the image charge!

Exactly. Didn't I imply this by saying the work function has to be exceeded? The next step, is what other force is stopping the electron, once it has got far enough to exceed the work function.

 

[ZapperZ]

Exactly. Didn't I imply this by saying the work function has to be exceeded? The next step, is what other force is stopping the electron, once it has got far enough to exceed the work function.

There is none. If the emitted electron has overcome the image potential, it is then free!

Zz.

 

[Ulysees]

Unless another electron has previously got stuck to an air molecule. Ie ionisation.

Or even previous electrons in the space between molecules, repel or change the direction of new electrons.

 

[Ulysees]

So I'd expect a cloud of electrons and ions at all times, once the work function has been exceeded.

Maybe room air ionisers are doing just that, but they use ac currents, so it's more complicated I guess.

 

[Ulysees]

What was that about tunneling you mentioned earlier, Zapper?

Some quantum effect that renders some of what we have said inaccurate?

 

[ZapperZ]

Unless another electron has previously got stuck to an air molecule. Ie ionisation.

Or even previous electrons in the space between molecules, repel or change the direction of new electrons.

So I'd expect a cloud of electrons and ions at all times, once the work function has been exceeded.

Er... what?

I didn't realize that we are also including a gazillion other factors such as space-charge. Since when it this necessary? Are you working in a particle accelerator that produces more than 10 nC of charge within 10 ps? I do, and this is not very common for most particle accelerators, which means that under ordinary circumstances, space-charge effects right above the metal's surface is not a big deal.

And why are there ions? If the electrons are being emitted with barely eV scale energies, these are no sufficient to cause ionizations in the neutral gas. You need at least 100 eV or so to be able to sufficiently create a plasma, because that's when the ionization cross-section becomes substantial, at least for oxygen.

This thread is very confusing. It seems like there are several diverging issues being discussed at once, and they are all jumbled together. If you want to know about the factors involved in the "work function", then STICK to just that. Don't introduce external factors such as space-charge effects that simply adds to the confusion of what it is. The empirical measurement of the work function of a particular metal does NOT include such external factors.

Zz.

 

[ZapperZ]

What was that about tunneling you mentioned earlier, Zapper?

Some quantum effect that renders some of what we have said inaccurate?

And what about it? How likely do you think it is that you can depend on an electron to tunnel out of a metallic surface without any applied potential? Calculate this yourself if you wish. If you think this is very likely to actually factor into your "world view", then you should expect that a broken vase to spontaneously assemble back into its original shape after I throw the pieces onto the floor. When was the last time you saw that happening?

What this has anything to do with "work function", I have no idea. Again, thread going in ALL directions without any focus on anything.

Zz.

 

[Ulysees]

Er... what?

I didn't realize that we are also including a gazillion other factors such as space-charge. Since when it this necessary? Are you working in a particle accelerator that produces more than 10 nC of charge within 10 ps? I do, and this is not very common for most particle accelerators, which means that under ordinary circumstances, space-charge effects right above the metal's surface is not a big deal.

And why are there ions? If the electrons are being emitted with barely eV scale energies, these are no sufficient to cause ionizations in the neutral gas. You need at least 100 eV or so to be able to sufficiently create a plasma, because that's when the ionization cross-section becomes substantial, at least for oxygen.

This thread is very confusing. It seems like there are several diverging issues being discussed at once, and they are all jumbled together. If you want to know about the factors involved in the "work function", then STICK to just that. Don't introduce external factors such as space-charge effects that simply adds to the confusion of what it is. The empirical measurement of the work function of a particular metal does NOT include such external factors.

Zz.

Err?

Didn't you say you want real things before? Conversations will be more comprehensive if they are about real things like corona discharge (post number one of the OP :) )

 

[Ulysees]

> How likely do you think it is that you can depend on an electron to tunnel out of a metallic surface without any applied potential? Calculate this yourself if you wish.

Without any applied potential? The one due to the extra electrons remaining in the metal, not good enough? Did we put any limits to how many these electrons were?

 

[ZapperZ]

Err?

Didn't you say you want real things before? Conversations will be more comprehensive if they are about real things like corona discharge (post number one of the OP :) )

.. which I had already described the one particular mechanism for such a thing. So what's the issue left here?

You were asking about what a "work function" is, didn't you? I explained it. Somehow, you then started to include other external effects that have nothing to do with a work function. Have you figured out how the values of the work function are obtained? Do you think space-charge effects are included in this value? If not, then why did you bring it up in your understanding of what a work function is? That is what I don't understand!

This has nothing to do with being "real". It has everything to do with narrowing down the the exact topic of discussion. Going in a million directions isn't being real. It's being silly. If you want to know the model for breakdown, there are plenty of resources for that.

Zz.

 

[jostpuur]

I tried to google a little bit because I'm not familiar with experimental facts, and found this http://www.Newton.dep.anl.gov/Newton/askasci/1993/physics/PHY102.HTM

A good vacuum is a very good insulator. Much better than air because there
are no molecules to ionize and participate in an avalanche.

I'm slightly confused about the claim that vacuum would be insulator. After all, a free electron is pretty free if it is in vacuum, so vacuum is not really insulator in the sense that electrons could not move in it. Perhaps this sentence simply means that vacuum behaves as an insulator, because it is difficult for electrons to leave the metal into the vacuum due to the mirror charge effect?

But then they say that the vacuum is better insulator than air! Is this claim based on experimental fact? If the claim is true, and on the other hand vacuum is not really insulator, doesn't it mean that the air is not insulator at all, but instead the air actually only helps current conduction outside solids?

 

[ZapperZ]

> How likely do you think it is that you can depend on an electron to tunnel out of a metallic surface without any applied potential? Calculate this yourself if you wish.

Without any applied potential? The one due to the extra electrons remaining in the metal, not good enough? Did we put any limits to how many these electrons were?

Again, another example of things being clouded for no apparent reason. Why do you need to put extra electrons on anything in your understanding of electrical discharge/breakdown? Whose model are you using that require you to put extra electrons on anything? ALL model of breakdown, be it in vacuum systems and in air, requires no such starting point. Having high field-emission regions, yes, but extra charges? Nope!

Until you show me a model that requires such extra electrons placed on the metal to account for such discharge/breakdown effects, I would consider this as nothing more than a confusing and unnecessary distraction.

Zz.

 

[Ulysees]

> You were asking about what a "work function" is, didn't you?

No I wasn't. You provided this idea, and I understood it immediately.

> I explained it.

No, you just insisted that it's right when no one had said it was wrong. Just lack of intuitiveness, remember?

> Have you figured out how the values of the work function are obtained?

It doesn't matter, we're talking about corona discharge in this topic.

> Do you think space-charge effects are included in this value?

No, did I say otherwise?

> If not, then why did you bring it up" in your understanding of what a work function is?

I wasn't talking about work function, I was talking about corona discharge.

> This has nothing to do with being "real". It has everything to do with narrowing down the the exact topic of discussion.

But you're not talking about the exact topic (corona discharge), you're talking about work function, only part of the topic.

So I will

 

[Defennder]

I think he was asking was if the repulsive electrostatic force from the other electrons on the charged surface would increase the probability that a particular electron on the charge surface would be able to tunnel out of the metallic lattice.

In other words, let us consider 2 cases, a uncharged metallic conductor and a exact same metallic conductor but with charges in it. Is the probability that an electron would tunnel out of the conducting surface higher in the 2nd case than in the first?

 

[ZapperZ]

I give up.

Zz.

 

[Ulysees]

> Why do you need to put extra electrons on anything in your understanding of electrical discharge/breakdown?

Because the topic is about corona discharge.

 

[ZapperZ]

I think he was asking was if the repulsive electrostatic force from the other electrons on the charged surface would increase the probability that a particular electron on the charge surface would be able to tunnel out of the metallic lattice.

In other words, let us consider 2 cases, a uncharged metallic conductor and a exact same metallic conductor but with charges in it. Is the probability that an electron would tunnel out of the conducting surface higher in the 2nd case than in the first?

But this is not the case that resulted in his "corona discharge". No model of such effects have been included in any of the models that I've looked at. And considering that I do extensive research work on this topic, I'd say that I've seen almost every single model of breakdown effects that is available in the literature.

That is why I am puzzled why this is even brought up, IF the whole point here is in trying to understand this phenomenon. I've outlined basically the most accepted mechanism that we know of. If people won't want to accept my word on it, then they can do their own leg work. I suggest starting with:

F.R. Schwirzke, "Vacuum breakdown on metal surfaces", IEEE Transaction on Plasma. Science v.19, p.690 (1991).

I thnk I'm done going around in circles with this thread.

Zz.

 

[Defennder]

Well I think this thread is quite interesting and informative, but Ulysees needs to be better able to rephrase and articulate his questions in a more formal manner if he wants to elicit proper replies from the experts here. There are genuine questions which are raised here, but it's a shame that the way they are presented is rather incoherent and confusing.

It happens a lot of time to myself when I'm asking my prof questions. Sometimes the conceptual questions leap to my mind and then to my mouth faster than I can think properly of how to articulate and present them in a more coherent manner. So what I suggest is that Ulysees sits back a little, think of a better way to present his questions in a lengthy coherent post rather than rapid shooting of 1-line posts which are cluttered and difficult to follow.

 

[Ulysees]

You're just stuck, the question is what I asked in the first post.

You have imagined something and we all have to accept that this is what was asked in this topic. Here's what was asked:

> I'm familiar with lightning rods taking advantage of the mutual repulsion of charges to shoot off a corona discharge off the sharp end and start a thunder, but why doesn't corona discharge happen to all charged metals? What makes air such a good insulator, when it's just gases, relatively few molecules moving all over the place bouncing on each other, how can this be a good insulator?

 

[Ulysees]

In fact what you have imagined even ignores the title of the topic! It says "negatively charged"! Mercy!

 

[ZapperZ]

You're just stuck, the question is what I asked in the first post.

You have imagined something and we all have to accept that this is what was asked in this topic. Here's what was asked:

> I'm familiar with lightning rods taking advantage of the mutual repulsion of charges to shoot off a corona discharge off the sharp end and start a thunder, but why doesn't corona discharge happen to all charged metals? What makes air such a good insulator, when it's just gases, relatively few molecules moving all over the place bouncing on each other, how can this be a good insulator?

1. the "corona discharge" has nothing to do with "mutual repulsion of charges". Again, this has been answered already when I presented to you the scenario of breakdown mechanism. Look at the list I gave. Where does it say "mutual repulsion"?

2. Where is the mechanism that involves putting in extra charges onto the metal? The mechanism that I outlined as no such extra charges being added. It would work even if the metal is isolated from ground, meaning it does not have to have any extra charges.

3. ALL metals can cause a discharge if I have a sharp-enough tip. If there are metals that don't cause a breakdown, we would have used it already in accelerating structures and achive a gazillion volts/meter of accelerating gradient. So asking why it doesn't happen in all "charged metals" is confusing. You're asking for an answer to a scenario that doesn't occur. What kind of an answer were you expecting?

4. The implication of "charge metals" as the requirement for a "corona discharge" is what I've been asking for you to produce. Show me a model in which a "charged metal" is required for such a discharge. This is the 3rd time I've mentioned this already and asked you to produce such evidence or model. Show me a model in which a corona discharge is caused by the addition of addtional charges to the metal. In the model that I had listed, no such addition is necessary. All that was required was regions of high field enhancement. No addition of extra charges at all! Until you can show me such a model for me to study, this "charge metal" scenario doesn't exist.

Zz.