Plasmonic effects associated with metallic nanostructures have been widely studied for color generation. It became apparent that highly saturated and bright colors are hard to obtain, and very small nanostructures need to be fabricated. To address this issue, in this study, we employ metal-insulator-metal sandwich nanodisks that support enhanced in-phase electric dipole modes, which are blue-shifted with respect to a single metal disk. The blue shift enables the generation of short wavelength colors with larger nanostructures. The radiation modes hybridize with the Wood's anomaly in periodic structures, creating narrow and high-resonance peaks in the reflection and deep valleys in the transmission spectra, thus producing vivid complementary colors in both cases. Full colors can be achieved by tuning the radius of the nanodisks and the periodicity of the arrays. Good agreement between simulations and experiments is demonstrated and analyzed in CIE1931, sRGB, and HSV color spaces. The presented method has potential for applications in imaging, data storage, ultrafine displays, and plasmon-based biosensors.

Li, Zhao, and Hu Reply: In our Letter [1], we pointed out two important things. First, the disorder can induce a finite thermal conductivity in lower temperature regime. Second, our numerical calculations with Nosé-Hoover thermostats show that a unique nonequilibrium stationary state may not exist in a disorder harmonic chain.In his Comment, Dhar [2] claims that the second point is not true. He mentioned that the existence and uniqueness of a nonequilibrium stationary state has been proved by Lebowitz et al. [3]. We should stress here that our observation does not contradict the proof of Lebowitz et al. In Ref. [3], the authors proved only that when such a (mass) disordered harmonic chain is placed in contact with stochastic reservoirs of the Langevin type, a nonequilibrium stationary state can be reached. However, they did not prove the existence and uniqueness of the nonequilibrium stationary state for the general case, namely, when the chain is in contact with general thermostated reservoirs. As pointed out recently by Bonetto, Lebowitz, and Rey-Bellet [4] "for general thermostated reservoirs the problem seems to be mathematically out of reach at the present time." So, the existence and uniqueness of a stationary state in a disordered harmonic chain for a general thermostated reservoir is still an open question. Most recently, Parisi [5] shows that the time needed to equilibrate in a harmonic chain is infinite. Our numerical results agree with this conclusion; viz., at any finite time, one cannot obtain a definite steady state starting from different initial conditions.Dhar's claim that our results might be caused by the insufficient equilibration time also seems to be unlikely. In Fig. 4(a) [1], we have shown that the temperature profile at two different time scales t 10 6 and t 10 7 are almost identical (see also the inset of this figure) when starting from the same initial condition.

No abstract

No abstract

No abstract