Interesting and maybe relevant bit of science news:
A 100-year-old physics problem has been solved
and paper...
Breaking Lorentz reciprocity to overcome the time-bandwidth limit in physics and engineering
Seems the resonance limit on Q isn't inviolable.
Attached is a snippet from the paper. All credit where it is due.
Some discussion on the paper...
So wave packets may occasionally propagate in one direction and simply gain amplitude when hitting a boundary with magnetic fields blocking back-propagation. The entire information of the wave is conserved at the boundary. The authors used a combination of a Si dielectric, InSb semiconductor and Ag conductor. This heterostructure had a "static"* magnetic field applied in the -y direction. This would be akin to
increasing the electron pressure on the outermost boundary. In fact it is the E field (see orthogonality in diagram) which provides the directional force to prevent the wave packet from escaping its resonant oscillation at the point of incidence. The wave becomes sandwiched between two e fields (electron seas/phonon soup) until it decays. Supposing the power to the repulsive B field is turned off, then with back scattering and back propagation it can of course assume a longer closed path again.
Here's yet another analogy then of how I currently view the cavity in light of this renewed confirmation of the discrete and unidirectional nature of photonic propagation given the repulsive (TE directional aligned, remember B does not repel E but instead overlaps) E field. The inside of the cavity is like a plasma onion. Each electron sheet has photons bouncing (until absorption, dissipative losses or propagation (tunneling/however) through the wall) along paths not too dissimilar from that of a vacuum filled cavity. As each electron is excited, it in turn can emit a photon or collide to continue the propagation of the energy.
At certain points, however, due to magnetohydrodynamic shearing and torque from the time evolution of the resonant fields, and of course some splatter and chaos, specific areas will experience magnetic reconnection and transfer energy directionally. Before the reconnection occurs, individual photons, carrying information as a wave packet, will resonate between layers. This buffer of sorts will release a cascade of energy down to the weaker e field/onion layer upon reconnection and create more asymmetry further ensuring energy propagation towards the walls. The more excited the medium gets the stronger the E field (OBVIOUSLY) in each of the modal peaks and the greater the difference and therefore the statistical likelihood of prevention of back propagation between layers. Wave packets do not ignore the internal electrons, and internal electrons do not ignore the wave packets. Each area of high energy density will act just like any other ExB field in relation to light. This paper just confirms that fields can concentrate an arbitrary wave packet into a dense volume of space and therefore increases the entire system's capacity for energy storage. Instead of Q being a limitation based on the entire wave path, we understand that Q is determined by any number of arbitrary temporary resonances the packet enters during its lifetime (before leaving the system). When will we accept that the same physical laws governing plasmonic propagation in metals also govern plasmonic propagation in air just on a different time and energy magnitude?
Two topics which I brushed on earlier come to mind: plasmonics in zero dimensional metals/nanowires and (de)coherency of the system. Upon trapping the light it is actually made temporarily coherent as the path is closed. In theory, decohering a large amount of the photons stuck between sheets/wall (the wall is just a denser sheet) by no longer introducing new waves into the cavity, will lead to a reactionary force in the exact opposite direction of the relaxation events and the pressure gradient because the gradient is time dependent and forms a sequence of discrete quanta which repel weaker (younger) fields. If the thrust is caused from wall to wall reactions and largely ignores internal atmosphere, this reactionary force should be close to or equal to the instantaneous forward force. If the thrust is caused by electron/ionic pressures, then this reactionary force should be less than the forward force due to vortices, eddies, evanescence and more. The only issue is whether it decoheres instantaneously or over time and whether it has a tangible effect on system acceleration: a question of how important the gain medium actually is. If the gain medium is a cold plasma then we can begin to consider the different layers as discrete. Plasmonics in nanowires are so interesting because they also show asymmetry and pseudo-momentum naturally. Waves which resonate along such a structure naturally shift the mass and charge to one end of the structure, these resonances and oscillations seem eerily similar to field vectors and electrical currents inside a cavity.
I leave you with a quote for some of the other theorists from the paper:
The small increase in the total optical losses that we observed in our simulations for this latter case is
because the slope of the band (i.e., the pulse’s group velocity) reduces with increasing B0, leading to higher overall optical losses(27**)
*As proven before, true static fields do not exist! Minor oscillations are unavoidable, anything else is a mathematical flight of fancy.
** 27. R. W. Boyd, J. Opt. Soc. Am. B 28, A38–A44 (2011)