• Organisation

Abstract:

This master thesis presents a work, which has been carried out on the characterization and optimization of ultraviolet (UV) depth lithography with SU-8 negative photoresist for fabricating high aspect ratio microstructures. The three major quality characteristics concentrated on in this thesis are high resolution, adhesion of SU-8 structures to the substrates as well as good sidewall profiles of the SU-8 structures. The SU-8 structures are fabricated on silicon substrates and have a thickness of 500µm.

Initially, an extensive literature research work has been carried out in order to summarize the state of the art of UV depth lithography and to discover the important process parameters and their influences. As a result, a table has been prepared indicating the SU-8 blend used and the suggested parameter values in each process step according to the different thicknesses considered. This table is used as a synopsis for parameter ranges.

UV depth lithography comprises of several process steps. These are spin coating, soft bake, UV exposure, post exposure bake, development, and hard bake. In each process step different parameters are considered, such as spin speed, spin time, baking time, baking temperature, exposure time, and development time. The decision on usage of SU-8 with cyclopentanone as a solvent is done by comparing its chemical properties with other SU-8 blends of different solvents. The photo resist used is defined as 75% of solid with high viscosity. A mask has been designed in AutoCAD for exposing resist. In order to characterize the parameters of each process step for defined thickness, i.e. 500µm, a decision has been made to perform spin curve and exposure experiments. With these experiments, the optimal spin coating and exposure parameter range are extracted.

In the second half of this thesis soft bake time, post bake time, and exposure time are chosen as the major influence factors for the subsequent parameter optimization. With the aid of a d-optimal design of experiments, a design plan consisting of 15 single experiments is generated using statistical functions of Matlab. After performing these 15 experiments in the clean room, the results are analyzed by defining a numerical value to its quality by a quantification process. This quantification is done in order to feed the output responses as input matrix to an optimization tool.

Finally, the parameters are optimized to get the best responses and their corresponding values are noted.

The verification experiments are done to investigate the optimized results and the verification results obtained are found promising.

EMK-spezifische Daten:

Lagerort Dokument: Archiv EMK, Kontakt über Sekretariate,

Bibliotheks-Sigel: 17/24 EMKM 1593

Art der Arbeit: Masterarbeit

Beginn Datum: 30-05-2005

Ende Datum: 01-12-2005

Querverweis: keiner

Studiengang: Elektrotechnik und Informationstechnik (ETiT)

Vertiefungsrichtung: Nachrichtentechnik (NT)

Abschluss: Master of Science (MSc.)

• Organisation