As for the applications that can be use for this type of studies, biochemical-sensors, Surface Enhance Raman Scattering, sending signals cross microchips for communications, and optical super-lens.
To use those applications and make them more efficient,this bring us to our objectives to remove the loss in metal (Au or Au) due to heat, and parametrically amplify our signal using the SPP, which we have demonstrated it in here.

I am really sorry for the late reply.
In second harmonic generation efficiency calculation, quantum efficiency is not what we look or calculate. It takes two 1080 nm photon to generate one 540 nm. So, what we calculate is the SHG efficiency, which is the E-(540nm)^2/E-(1080m)^4, and it is for our study: is 1.7E-17(M2/V^2). This efficiency may look small, but if we compare it the plain crystal MNA efficiency: its 2.06E-17(M2/V^2).
One would say, what is happening to the rest of the photons, the rest of the 99.999% of the 1080 nm photons? Well, they pass through the materials without doing anything.
One would still say, its still very small! Well, in the side of lasers, parametric amplification and SHG studies, this is very good and a modest value.
As for the 7*10^5 better, we found this to be better than the previous reported value using a different SHG material, and in a different system.