Electron emission from metals

gareth

Hi all,

I'm keen to find out more about electron ejection from metals. Eg multi-photon photoelectric effect and thermionic emission.

I've come across a few texts (Kittel) which deal with them in a general sense but can anyone recommend a more detailed analysis?

Some questions which I hope to resolve are as follows;

What determines the angular distribution of the emitted electrons?

What kind of velocities are associated with the electrons? (I understand their kinetic energy depends on the extra amount of energy they have after using up their energy to escape the metal)

How does it differ with different materials, i.e. I know the work function of a metal determines the energy required for an electron to overcome the vacuum potential, but is this all that matters? Do metals with a different conductivity, but the same work function behave differently under the same emission conditions?

ZapperZ

You need to narrow down a bit of the scope here. What process are you interesting in? Photoelectric effect? Multi-photon photoemissioni? Thermionic? These do not necessarily result in the same distribution and are governed by different mechanisms.

See what I mentioned above.

If "it" refers to the velocity (or, more precisely, the kinetic energy) of the emitted electrons, then again, it depends on the process - photoemission, thermionic, field emission, etc. The emitted energy certainly is a function of the temperature in thermionic emission, and the photon energy in photoemission processes. You also don't get just ONE energy, but a wide range of energy distribution.

So you need to narrow down a particular emission process first.

Zz.

gareth

OK,

Say if wanted to look into multi photon photoemission processes, where do I start?

ZapperZ

What kind of literature search have you done?

Zz.

gareth

I've read numerous papers, all of which give slighlty different results/interpretations of the angular distribution of the ejected electrons. Checked the library and gone through a few books on solid state theory (Kittel, Haug) they touch on emission processes but do not have an involved discussion of the directionality/multi photon emission.

So maybe I should just keep looking.

ZapperZ

You should have given the citations to the papers that you've read.

I am not aware of any angular studies on the emitted photoelectrons. Since most of the multi-photon photoemission studies were done on metallic surfaces in which the Rydberg energy levels are valid for the intermediate states, I don't see why one would want to measure the angular dependence. The closest that I can come up with that has any form of angular studies is the one on the angle of the incoming photons with s and p polarization[1].

Most studies using multiphoton photoemission are more interested in the time-resolved component of it, not the angular dependence of the photoelectrons. The angular dependence of the emitted photoelectrons are usually of critical importance in angle-resolved photoemission spectroscopy (ARPES) on single-crystal samples. This is one of the few situations where such angular resolution is needed.

Zz.

[1] A. Damascelli et al. Phys. Rev. B v.54, p.6031 (1996).

gareth

I'm pretty new to this but it seems like their is a lot of interest in highly collimated, energetic electron emission from metals, for possible use during ignition in a nuclear process (open to correction here).

Here are a couple of papers which detail the interest in the emission angles;
Li, Phys. Rev. Lett. 96, 165003 (2006)

Cai, PHYSICS OF PLASMAS VOLUME 10, NUMBER 8, page 3265, AUGUST 2003,

As you said there is much interest in temporal behaviour of the electrons, such as those
emitted during ultrafast laser pulses, these seem to be very hard to measure understandably.

ZapperZ

That would be VERY strange to use multiphoton photoemission to get that.

This appears to have nothing to do with multiphoton photoemission. Why is this relevant here?

Are you sure we are dealing with the same phenomenon here? Do you know what a "multiphoton photoemission" is?

Zz.

gareth

I think I'm starting to see the distinction, the previous two papers I cited are dealing with electron acceleration from the plasma state it seems. And regarding ignition experiments you were correct, the focus seems to be more on the laser heating of DT pellets.

Here is a paper that deals with mulitphoton emission [1], it seems it occurs quite readily at intensities below the plasma threshold, but the mechanism is still open for much discussion.

I understand a multi photon-photo emission process to occur when an electron manages to overcome the work function of the metal by the simultaneous absorption of several photons below the actual work function, but I am unclear as to what 'simultaneous' actually is in this case. I assume the electron must absorb the required amount of photons before it can give up its energy elsewhere, i.e. to the lattice. So would the requirement be that the photons must be absorbed during a time shorter than the electron phonon coupling time?

[1] Banfi et al., PHYSICAL REVIEW B 67, 035428, 2003

ZapperZ

Did you read the reference that I gave above? It gives you a lot of background info on multiphoton photoemission. I've done a lot of work on photoemission, and I've also done multiphoton photoemission, but "not on purpose".

When an electron absorbs a photon that is less than the work function, it gets excited to a quasistable state in the conduction band above the Fermi level, but it can't escape because it is still below the vacuum level. It has a lifetime in that state on the order of ... oh.... picoseconds. Now, in a normal photoemission/photoelectric effect, this electron will decay back to the Fermi level. However, if the light source being used in very intense (high photon density per unit area), then there's a non-negligible probability that another photon can be absorbed by the excited electron state before it decays. This is the multiphoton process. It isn't a simultaneous absorption of several photons at once. That's why there's an interest in the temporal studies of this effect, because there is a time delay being the absorption of the first photon with the subsequent photons (this temporal delay, btw, is another STRONG evidence for the existence of "photons". Wave mechanics has no ability to explain this.). The paper that I referenced has this study.

Zz.

gareth

Yes, the paper you cited was of great interest, thanks.

I can't quite grasp the concept of how the polarisation of the light effects the ejection yield and also the ejection angle. My thinking is that it's related to the crystal structure of the metal. If this is the case would rotating the metal sample by 90 degrees have the same effect? (I could be way out here but your comments are appreciated)

ZapperZ

Of course, this would only make sense if they're using single-crystal samples, which they did.

Even in regular photoemission, such as angle-resolved photoemission, the polarization of the light source can certainly change the result. You are trying to get electrons that has a certain momentum in a certain direction, depending on the crystal orientation. One can look at the First Brillouin zone, for example, to see how such crystal momentum can be anisotropic in various crystallographic direction. So the direction of the E-field in the slight source certainly can affect the photoelectron intensity.

How this actually works out in multiphoton photoemission is something I haven't quite understood yet.

Zz.

gareth

Thanks Zz,

Another quick (trivial) question if you don't mind; you metion single-crystal samples in your last post. I always had trouble with this, surely a piece of pure metal arranges itself in its correct crystal orientation at room temperatures, be it fcc, bcc etc. But why then are some samples referred to as single crystals? Is it that some crystals are not perfectly periodic over long distances?

ZapperZ

Single crystal usually implies that there's only one uniform, contiguous crystal throughout the sample. One can have a crystalline material, but with grains in various orientations. So you have a clump here oriented in one direction, and another clump oriented in another direction, etc. So the crystal isn't contiguous.

Zz.