The optical waveguides and grating couplers can be made by a variety of routes. For fabricating the thin layers of high refractive index transparent oxides, either physical vapour deposition (e. g. evaporation or sputtering) or sol-gel pyrolysis are used. The optical quality requirements are extremely stringent: the absorption coefficient should generally be less than 10-5 cm-1, otherwise insufficient light will reach the end of the waveguide and the signal:noise ratio will be degraded. The gratings required for coupling can be made by either topographical or refractive index modulation. The former can be achieved by structuring the optical glass support (the S layer) using the conventional methods of photolithography; to obtain the fine gratings required (A of the order of 400 nm or less) for working with i = 1, holographic exposure of photoresist is required. Coarser gratings can be used with higher order diffraction peaks but the signal:noise ratio is lower. It is also possible to modulate the surface of the F layer: however, this is less desirable since the surface (that will interact with the proteins) may then become chemically contaminated. Shallow gratings (5-10 nm deep) are sufficient and have the great advantage that the F layer is thereby so slightly modulated in the grating region that the mode equations derived above retain their validity. A further advantage is that, from the hydrody-namic viewpoint, the surface may still be considered planar.
Perfect planarity is, however, only achievable with refractive index modulation. This can be achieved by exposing the substrate to a beam of energetic nitrogen through a mask (created, as above, by photolithography via holographic exposure of photoresist). Nitrogen atoms implanted in the silica modify its refractive index. The technology is non-trivial to perfect, however, mainly because of the difficulty of removing photoresist residues hardened by exposure to the energetic nitrogen beam.
Further fabrication steps concern the coating of the waveguide with whatever material is serving as adsorbent for the proteins under investigation. Lipid bilayers are conveniently deposited on the surface using the Langmuir-Blodgett technique (Ramsden 1999); alternatively a suspension of lipid vesicles can be used, which fuse with the surface to create the bilayer (Csucs and Ramsden 1998a). In both cases, a few molecular layers of water between the inner leaflet of the bilayer and the solid support ensure that the membrane (in its liquid crystalline state) has native fluidity, i.e. the lateral diffusivity of the lipid molecules is approximately the same as in the membrane of a living cell.
Polymer layers, including polyelectrolytes, can be deposited by spontaneous attachment from solution (e.g. Ramsden et al. 1995); the large number of contacts per molecule ensures irreversible attachment. Other materials can be deposited by physical or chemical vapour deposition. A layer 10 nm thick will generally be found to completely mask the underlying F layer (Kurrat et al. 1997). The only restriction is that the layer should not absorb the guided light; hence metals can only be deposited to a thickness of about 2 nm, which is often insufficient for a continuous layer to be formed; and materials (dyestuffs) with intensive absorption bands coinciding with the wavelength of the guided light should be avoided.
Was this article helpful?