fke logo Silicon Process Modeling
Process modeling plays an important role in the development of new generations of integrated circuits. It provides a better understanding of individual process steps, saves time and cost in the design and optimization of a new technology, and helps solve problems in manufacturing.

Overview of Research Areas

  • Ion implantation: Modeling may help predict and understand dopant distributions and damage formation in ion implanted materials. Our extensive studies in this field have resulted in a carefully verified Monte Carlo code called IMSIL. This code is used in the research laboratories of companies like IBM, Infineon, and Lucent Technologies.

  • Focused ion beam processing: A recent extension of our implant modeling activities is the simulation of focused ion beam (FIB) processing. The aim of our work is to predict the topography evolution and compositional changes of the target due to a localized ion beam, and to correlate the results with experiments done at our institute. A derivative code of IMSIL, FIBSIM, has been produced.

  • Transient enhanced diffusion and extended defect evolution: Defects introduced by ion implantation cause enhanced dopant diffusion during annealing. The amount and duration of the enhancement are determined by the number of defects introduced during implantation and the temporal evolution of extended defects such as {311} defects. Our models have been implemented in commercial TCAD tools like the new Taurus Process&Device code of Synopsys.

More Information

For more detailed information on individual projects follow these links:

Links

last modified on 14 Apr 2003 by Gerhard Hobler 		sidewall doping by FIB

Here you can see the topography of a silicon edge with a tungsten layer on top after erosion by a focused ion beam. The original geometry is indicated by the dashed line. The beam is indicated by the gray column. It has been scanned starting from the edge to the present position. The colored areas indicate the gallium doping from the beam. The simulation has been performed with our program FIBSIM.