plasmons

None to speak of.

I'm not a big fan of the term "plasmon", since it refers to an electromagnetic surface wave, so its elementary excitation is a photon just like any other EM wave. However, grantsmanship being what it is, we seem to be stuck with it.

True EM surface waves exist only at an interface between materials with different signs of Re{epsilon}, because that's the only way you can patch together two solutions that both decay exponentially perpendicular to the interface.

The coinage metals (copper, silver, and gold) exhibit free-electron behaviour, i.e. they have a region in the thermal IR to the red where the real part of epsilon goes negative--the dielectric response goes oppositely to the applied E field.

This seems very weird, and indeed, if it happened at DC these metals would all spontaneously turn to lava. (It doesn't, so they don't.) The reason is that the electron damping is so weak. In the Drude model, the electric polarizability of a metal comes from collective motion of the electron cloud, which for any little element of it is a second-order ODE:

d**2 x/dt**2 + (1/tau) dx/dt - k E x = 0

When you apply a sinusoidal field, the derivatives turn into (j omega)'s, so the first term is negative. If the damping is weak enough, there's a spectral region where the overall response is nearly a negative real number, and that's where plasmons can exist.

So since your microwave antenna doesn't work in the infrared, there aren't any plasmons to speak of.

Cheers

Phil Hobbs

(Who used to make gold plasmonic antennas for the infrared.)

Hi,

From what I read on wikipedia the plasmons can be surface plasmons, or also inside the metal:

For an antenna ie. 2.4GHz RF or THz nano-optical, there should be surface and bulk metal oscillations of electrons, I don't see why the frequency determines if there are plasmons or not.

I think an RF antenna must have EM surface waves in order to function.

This is like a 2.4GHz wifi antenna surface, the bulk electrons (described as plasmons) create a surface AC plasmon field / electric field on the antenna surface which emits "photons"/electromagnetic waves. A THz or infrared optical antenna could be described to work in the same way I think.

Is this due to the metals spectral resonances? ie. like this graph:

Nearly a negative real number is ~0.00001, so this number is the interaction between the metal surface e-field and the externally applied e-field, and if it is zero, there is no interaction at that frequency?

I am not convinced yet! :)

cheers, Jamie

The constitutive relations (i.e. n and k vs omega) determine it, because Re(epsilon) has to be negative and large compared to Im(epsilon). You could have surface waves at much lower frequencies at the boundary of a low-density, low temperature plasma, for instance. Of course in plasmas you get mass motions too, which makes everything an order of magnitude more complicated.

No. An antenna needs to couple into propagating waves on the air side, or else it isn't an antenna. Plasmons are _surface_ waves, like a SAW filter except not as leaky. Try listening to a SAW device if you don't believe me. ;)

Nope. That isn't the physics of the problem.

No. At 1.6 um, where I did most of my work, gold has an epsilon of about -100+j12. It has to be nearly real as well as negative--i.e. have a phase angle of nearly pi.

You're kidding. ;)

Cheers

Phil Hobbs

Hi Phil,

I finally understand a plasmon is like a ripple on the surface of water (or a tsunami I suppose), but on the molecular/atomic surface of a material and composed of electric charge and not water. So the plasmon is special because it doesn't evaporate into a photon when the sun hits it. It all is clear as daylight now! :)

cheers, Jamie

Beside the surface plasmons mentioned by Phil, there are bulk plasmons in films, so alkali metal films may appear transparent at UV wavelength.

Rene