In the first step of the LIGA process a resist layer of polymethylmethacrylate is patterned by deep X-ray lithography. The exposed parts are subsequently dissolved by an optimized organic developer. The quality of the microstructures is decisively determined by the development process. Because essential parts of the development process are not understood, it is necessary to carry out many extensive and costly experiments before a microstructure can be developed in an optimal way. The aim of this thesis was to determine the parameters which influence the development process in case of dip development as well as for the development with application of megasound and thus, to make the calculation of the development course possible. The development course during the development process was experimentally recorded by a mechanical measuring instrument with a probe tip less than 30 µm in diameter which therefore allows the measurement of microstructures up to a depth of 500 µm.

For dip development the geometry and their lateral dimensions have no ascertainable influence on the development rate. The dependency of the development rate on the dose could not only be affirmed, but also enlarged by an essential further dimension. The development rate above 5 kJ/cm³ depends on the value of the deposited dose and the depth of the dose deposition. In a dose area from 5 to 9 kJ/cm³ the development rate for the same dose value is in inverse proportion to the square root of the depth. For values in this dose area linear proportion exists between dose and development rate. In addition an exponential dependency of the development rate on the temperature in the studied area from 21 °C to 37 °C was observed. The combination of the three determing parameters - dose value, depth of dose deposition and temperature - leads to an "empirical" equation for the development rate. Using this equation it is possible - for a given dose profile and a given developer temperature - to calculate the development course in microstructures for dip development with an accuracy of 30 µm. The additional application of megasound accelerates the development process. In this case a significant dependency on the lateral dimension of the microstructures could be observed, but only for fundamental restrictions. A further typical phenomenon of the megasound development is the dependency of the development rate on the deposited dose but not on the depth of deposition. For the two investigated megasound intensities of 2 and 10 W/cm² the correlation of the dose and the development rate could be expressed by a polynomal equation. The comparison of calculated and real development course for different dose profils gives a correspondence with an deviation less than 50 µm.