Modeling and controlling topographical nonuniformity in thermoplastic micro- and nano-embossing

Abstract

The embossing of thermoplastic polymeric plates is valuable for manufacturing micro- and nanofluidic devices and diffractive optics. Meanwhile, the imprinting of sub-micrometer-thickness thermoplastic layers has emerged as a lithographic technique with exceptional resolution. Yet neither hot micro-embossing nor thermal nanoimprint lithography will be fully adopted without efficient numerical techniques for simulating these processes. This thesis contributes a computationally inexpensive approach to simulating the embossing of feature-rich patterns into thermoplastic polymeric materials. The simulation method employs a linear viscoelastic model for the embossed layer, and computes the distribution of contact pressure between the polymeric surface and an embossing stamp. An approximation to the embossed topography of the polymeric layer is thereby generated as a function of the material being embossed, the stamp's design, and the embossing process's temperature, duration, and applied load. For a stamp design described with an 800 x 800 matrix of topographical heights, simulation can be completed within 30-100 s using a computer with an Intel Pentium 4 processor and 2 GB RAM. This method is sufficiently fast for it to be employed iteratively when designing a pattern to be embossed or when selecting processing parameters. The method is able to build abstracted representations of feature-rich patterns, increasing the simulation speed still further. The viscoelastic properties of three materials - polymethylmethacrylate, polycarbonate, and Zeonor 1060R, a cyclic olefin polymer - have been experimentally calibrated as functions of temperature. For a test-pattern having features with diameters 5 [mu]m to 90 [mu]m, simulated and experimental topographies agree with r.m.s. errors of less than 2 [mu]m across all processing conditions tested, with absolute topographical heights ranging up to 30 [mu]m. In thermal nanoimprint lithography, the key challenge is to minimize spatial variation of the polymeric layer's residual thickness where stamp protrusions press down into the layer. The simulation method is therefore extended to incorporate elastic stamp deflections and their influence on residual layer thickness. Some design-rules are proposed that could help to minimize residual layer thickness variation. A way is also proposed for representing any shear-thinning of the imprinted layer.