Spacer Layer Effect

Earlier in this chapter we pointed out that the substrate plays an important role in metallic nanostructure-enhancement of fluorescence. In this section we consider the role of coupling between arrays of fluorescent molecule-coated Ag nanostructures and an underlying metal film. 

The measurements done by Guo et al. [25] showed decay in the enhancement beyond approximately 80 nm. Their FDTD calculations instead showed a continued oscillation in the average field-squared at the position of the fluors out to at least 600 nm these oscillations correlate with Fabry-Perot resonances within the oxide, an observation which has been made previously by Andrewartha [59] for a similar type of structure (a bottle grating interestingly calculations of the reflectivity from this related structure showed similar sharp oscillations [59]. The calculated oscillations originate from the interference between diffraction from the nanowire array and multiple reflections from the aluminum 

Earlier observations by Cesario et al. [60] of a decay in fluorescence for arrays of Au nanoparticles spaced above a Ag film by a Si02 layer of increasing thickness, were interpreted as due to the finite vertical extent of the evanescent fields associated with a surface plasmon. In this model the coupling results in an enhanced interaction between individual localized plasmons at individual nanostructures [61] and thus an enhancement in the radiative efficiency increasing the spacer layer thickness moves the nanowires out of the evanescent field of the surface plasmon. A possible physical mechanism for the experimentally observed decay involves nonradiative decay of the excited states. The aluminum oxide deposited in these experiments was likely to be nonstoichio-metric, and defects in the oxide could act as recombination centers. Thicker oxides would result in higher areal densities of defects, and decay in fluorescence. A definitive assignment of the mechanism for the observed fall off of fluorescence would require determination of the complex dielectric function of our oxides (as deposited onto an Ag film), and inclusion in the field-square calculations. Finally it should be noted that at very small thicknesses quenching of the fluorescence is expected [38,62] consistent with observations of an optimum nanowire-substrate spacer thickness. 

The author is pleased to acknowledge contributions to the examples presented in this chapter from D.G. Britti, R. Chiuri, T.D. Corrigan, S. D Agostino, F. Della Safa, S.-H. Guo, J.J. Heetderks, H.-C. Kan, P.P. Pompa, H. Szmacinski, S.-J. Tsai, R. Rinaldi, and R. Cingolani, and support from The Laboratory for Physical Sciences, the CNR-INFM, a NSF International Grant (O1SE0242579) and a NSF-MRSEC (DMR-0080008). 

Raether, H. [ed.] [1988] Surface Plasmons on Smooth and Rough Surfaces and on Gratings, Springer Tracts on in Modern Physics, vol. Ill, Springer, Beriin. 

---

Guo, S.-H., Britti, D. G., Heetderks, J. J., Kan, H-C., and Phaneuf, R. J. (2009) Spacer layer effect in fluorescence enhancement from silver nanowires over a silver film Switching of optimum polarization. Nano Lett, 9, 2666-2670. 

When using the thin silica spacer layer, however, it was found that the results from the above-mentioned methods did not agree with the direct measurements from the Taly-surf profilemeter, as shown in Fig. 4(a). This was tentatively ascribed to the effect of penetration of the reflecting beam into the substrate. With a very thin silica layer, the depth of penetration and thus the phase change would depend upon the thickness of the silica spacer layer and also upon that of any oil film present. 

Usually, the front of the film-pack is covered with black paper and additional sheets may be introduced between the films to act as "spacers. 1 Effects due to absorption in these extra layers must also be taken into account. If the absorption coefficient for a spacer is k, a fraction N2(29) of the radiation incident at an angle 29 to the nornPal will be transmitted, where... 

The properties desired of an ideal spacer layer are that it be stiff in shear, but that the spacer itself contribute minimally to the bending stiffness of the base structure, shifting the neutral plane as little as possible. We note that for the spaced constrained layer, the combined function of the viscoelastic layer and spacer is to provide a thick, dissipative and appropriately stiff (in shear) layer between the constraining and base layers. Therefore the order of the viscoelastic and spacer elements is arbitrary and they may be subdivided as long as the desired properties are preserved. These possibilities give additional freedom in adapting viscoelastic materials for effective damping. 

The thickness of the Cu spacer layer (Fig. 5.7.6) was chosen to fit the second antiferromagnetic coupling maximum (tCu=2.15 nm). Reducing the thickness to the first maximum (tCu = 1.1 nm) increases the effect even more and also results in a minimum hysteresis, but the slope is much flatter and thus the device is less sensitive, because, due to the stronger coupling, the saturated state is reached at higher external fields. In addition, thinner layers mean less temperature stability. The example stack shown here withstands ambient temperatures up to 270 °C for one hour. 

To better see the importance of the effect of the reduction of the critical thickness in stacked layers in future QD-based tunneling devices, we have realized a 10-bilayer stracture, in which the Ge deposited amount was kept constant in all layers. The Si spacer thickness between adjacent islands was chosen to be 2.5 nm, a value typically used in tuimeling devices. It is worth noting that as the height of capped islands in the first layer is about 7 run, the total thickness of the spacer layer is 9.5 nm. This amount was then kept constant in all layers. Fig. 3(b) shows a typical TEM of such a structure. In contrast to the image of Fig. 3(a), the present image clearly shows that the Si spacer layer is not thick enough to completely cover... 

Here, we have combined RHEED and TEM to study the effect of the island ordering. Motivated by our finding on the reduction of the critical thickness versus the number of deposited bilayers, we have systematically undertaken measurements of the Ge critical thickness in the second layer as a function of the thickness of the Si spacer layer. The growth temperature was 600°C for both Ge and Si. Fig. 4 shows a typical result on the variation of the critical thickness in the second layer (dc2) versus the Si spacer thickness (dsi). dd is the Ge critical thickness in the first layer, which is of 4 ML. 

The driving mechanisms for the island vertical correlation have been the subject of extensive studies over the past years. Because the buried islands produce a nonuniform strain field at the surface of the spacer layer, i.e. the regions above the islands are tensely strained while the regions in between islands remain compressed, exciting models have treated the island distribution at the spacer layer surface by considering the effect of such a strain field on surface diffusion [4] or on island nucleation [3]. Recent calculations have taken into account the effect of the elastic anisotropy of the materials [16], the surface energy [18] or the elastic interaction between the buried islands with newly deposited ones [19]. However, in all of the above models it was assumed that the surface of the spacer layer becomes perfectly flat before the deposition of a new layer. From the experimental point of view, this... 

To summarize, we have demonstrated that the phenomenon of vertical ordering is characterized not only by the aligmnent of islands along the growth direction but also by a reduction of the critical thickness in subsequent layers. The better the vertical ordering is, the more pronounced the reduction of the critical thickness will be. Such a decrease of the Ge critical thickness can be explained by elastic strain fields induced by buried layers and mediated by the spacer layers. We have shown that when the Ge deposited amount is no longer kept constant in all layers as frequently carried out in typical experiments but adjusted according to the effective... 

Jung, G.Y., et al. 2000. The effect of insulating spacer layers on the electrical properties of polymeric Langmuir-Blodgett film light emitting devices. / Phys D Appl Phys 33 1029-1035. 

In contrast, much more studies have been devoted to the ditfusion of small molecules in the multilayers, because of its importance in applications like permeation membranes or biosensing. Both IR spectroscopy [316] and fluorescence measurements [317] have shown the diffusion of protons in the multilayers, and thus an influence of the pH of the outer solution even far inside the films. The influence of water on the thickness of the multilayers is also well-documented [111,318]. Diffusion of radiolabeled salt ions has also been measured [125,312], Voltamperometry showed that PAH/PAA films had little effect on the diffusion of Fe(CN)g , but that PAH/PSS films could hinder its transport [116], 6-CF [96], acridine orange [81], daunomycin [251], 2 -3 cyclic adenosine monophosphate [252], bisulfite [313], and different diazonium salts [147,313] have been shown to permeate deeply in multilayers built by ESA. Immunoglobulin G (IgG) could permeate or not in a superlattice made of anti-IgG layers and PAH/PSS spacer layers, depending on the thickness of the spacer layer. The diffusion constants of rhodamine and of 2,2,6,6,-tetramethyl-4-piperidinol-l-oxide (TEMPOL) in PAH/PSS multilayers have been quantified [113,114],... 

---