Figure 2 LSPR schematics. Schematic charge distribution, electric near-field amplitude distribution, and far-field scattering OICR-9429 research buy radiation pattern of a gold nanorod upon excitations of (a) its dipole mode (2,060 nm),

(b) quadrupole mode (1,030 nm), and (c) sextupole mode (734 nm). Red numbers in the scattering patterns indicate the angles with maximal scattering power. Sensitivities of quadrupole resonances In the following, we will investigate the extinction response of four types of gold nanorods and compare their RI sensing performance. The structures under study are as follows: type A, gold nanorod with a = 200 nm and d = 80 nm; type B, gold nanorod with a = 500 nm and d = 80 nm; type C, gold nanobipyramid with a = 200 nm and d = 100 nm; and type D, gold nanobipyramid with a = 200 nm and d = 42.5 nm. The dimensions of these nanorods are chosen such that the dipole resonance wavelength of types A and C and the quadrupole resonance wavelength of types B and D are all around 1,050 nm in order to compare their selleck chemicals llc RI sensing sensitivities

at the same wavelength. The geometry of nanobipyramids is selected because of its high FOM as Tipifarnib cost reported previously [7, 8]. To avoid numerical errors caused by the sharp tips and to be more realistic to the experimental samples, the edges of the two tips in nanobipyramids are blunted with a frustum shape. By changing the RI of the surrounding medium from 1.33 to 1.37 (supposing

a fixed incident angle = 60°), the extinction peak (λ sp) of each nanorod gradually redshifts towards a longer wavelength, as shown in Figure 3a,b,c,d. These results are summarized in Figure 3e in which the extinction peak for each nanorod is plotted as a function of the refractive index. It can be observed from Figure 3e that the slopes of the four curves – which directly represent the RI sensitivity dλ sp/dn – are not substantially different from each other, in an obvious Dimethyl sulfoxide contradiction to previous reports [3, 6–8]. This observation is due to the fact that the RI sensitivity of LSPRs is actually wavelength dependent, which means that the RI sensitivity will not depend much on the mode resonance of choice or the structure geometry once the sensing wavelength is fixed (consistent with previous theoretical results by quasi-static approximation [25, 26]). This also points out that it might be inappropriate to compare directly the RI sensitivities of LSPRs of different nanostructures at different wavelengths [3, 6–11, 13–17]. We also refer to the article [27], where the authors have argued that any single mode sensing of RIs such as LSPR sensing cannot surpass an upper limit of λ/n, where λ is the sensing wavelength and n is the surrounding RI – which means an upper limit of 1,050 nm/1.33 = 789.5 nanometer per RI unit (nm/RIU) for our case.