Two-Photon Polymerization in Self-Assembled Photonic Crystals
Stephanie Pruzinsky
Former Graduate Student in Materials Science
There are numerous applications for which photonic bandgap materials have been proposed, including low-threshold lasers, low-loss waveguides, and on-chip optical circuitry. In order to realize these applications, it is necessary to incorporate pre-engineered aperiodic defect structures within otherwise periodic photonic crystals. One widely explored technique for the fabrication of three dimensional photonic crystals for use in the near-infrared begins with the self-assembly of microspheres into colloidal crystals. However, prior to our work[1], an unaddressed yet critical limitation to the utility of self-assembled photonic crystals for the realization of PBG-based applications was the lack of an inherent method to controllably incorporate aperiodic defect structures. We demonstrated the use of two-photon polymerization (TPP) to generate such embedded features within colloidal crystals.[1] Two-photon polymerization is additionally interesting because it can also be used to embed three dimensional features within holographically defined photonic crystals.
Figure 1. Fluorescence confocal micrographs of cross-sections from 3D TPP features written within a colloidal crystal. Four features were defined—one is a 2D double-bend structure and three are 3D triple-bend structures with varying bend angles in the horizontal plane. (a) Vertical cross-section through the 2D double bend structure. The arrows indicate the location of the horizontal cross-sections presented in (b-d). The lower images are horizontal cross-sections through the: (b) vertical components of the four features extending ~10 µm from the substrate into the crystal, (c) horizontal components containing 0°, 30°, 60°, or 90° bends, and (d) vertical components extending ~10 µm further into the colloidal crystal. The arrows in b-d indicate the location of the vertical cross-section (a). (Figure and text from reference 2.)
Figure 2. Scanning electron micrographs of TPP features written (a) in bulk AF-350:TMPTA on a coverslip and (b) within a colloidal crystal after removal of the silica via an aqueous HF etch. These demonstrate the similar resolution achieved when defining features within v. outside of colloidal crystals. The colloidal crystal provides a support network for features which might not otherwise remain free-standing—for example in (b) some portions of the thin features fall over upon removal of the colloidal crystal. An alternative imaging method is fluorescence confocal microscopy which enables the non-destructive imaging of non-self-supporting 3D TPP features written within colloidal crystals (Fig. 1). (Figure and text from reference 2.)
Additionally, we have also demonstrated that colloidal crystals with embedded features can be infiltrated with high index materials, for example, selenium or silicon and am currently working on the characterization of such structures (Fig. 3).
Figure 3. Linear air defects defined via TPP in a colloidal crystal which was then infiltrated with silicon via CVD. After removal of polymer and silica, the resulting structure is air defects in a silicon inverse opal (preliminary unpublished data, S. A. Pruzinsky, F. García-Santamaría, P. V. Braun).
Optical Diffraction Studies
We also investigated the effect of numerical aperture (NA) on the optical diffraction response of polystyrene colloidal crystals. It was found that for low NA (~ < 0.30), the peak parameters were not greatly impacted by the NA. This suggests that it may be valid to compare reflectance spectra that was experimentally collected (with a nonzero NA) with normal incidence calculations for low-index contrast photonic crystals. This comparison is extremely common in literature, so it was important to demonstrate its validity. For this project, I was involved in an advisory capacity with respect to the experimental details, and I performed the simulations and analyzed that data.
We also interrogated the optical diffraction response of glucose-sensitive inverse opal hydrogels.[4] Again, I was involved in a computational capacity. I carried out the simulations and developed the simple swelling model that we used in an effort to gain a simple qualitative understanding of the swelling mechanism. After comparing experiment and theory, it was concluded that the swelling response could not be accurately described by a simple swelling model. This prompted the subsequent work on the finite element modeling of swelling in inverse opal hydrogels.
1. W. Lee, S. A. Pruzinsky, P. V. Braun, Advanced Materials 2002,14, 271.
2. S. A. Pruzinsky, P. V. Braun, Advanced Functional Materials, in press.
3. Y. -J. Lee, S. A. Pruzinsky, P. V. Braun, Optics Letters, 2005, 30, 153-155.
4. Y. -J. Lee, S. A. Pruzinsky, P. V. Braun. Langmuir 2004, 20, 3096-3106.