Modular Solid State Technologies for a Multi-functional System Integration

Karlheinz Bock, University of Berlin (TU Berlin), TIB 4/2-1, Gustav-Meyer Allee 25, 13355 Berlin und Fraunhofer Research
Institution for Modular Solid State Technologies EMFT (Fraunhofer-Einrichtung für Modulare Festkörper-
Technologien EMFT), Hansastr. 27d, 80686 Munich, Germany, karlheinz.bock@emft.fraunhofer.de

Abstract

After decades of system integration technologies focusing on device technologies and their networks – from televisions
to computers and cell phones – future micro system integration is probably to center more on humans and the environment.
In keeping with this trend, system integration is focusing on multifunctional interfaces for smart systems in order
to develop tailored solutions. A modular merge of different technology concepts CMOS, MEMS, HF ICs (III-V, IIVI…)
3D system integration on interposer, wafer and board level takes place in order to design application oriented interfaces
amongst the major existing technologies for such a multi-functional systems integration.
With its trend of increasing system integration level, Moore’s Law will become less important in the future for such a
task. So-called “More Moore” will give way and will somehow merge with “Beyond CMOS” approaches towards modular
multi-functionality in the solid state technology development.

MOTT (Multifunctional On-Top Technologies) is to develop such modular system approaches for enhancing the functionality
of standard silicon chips, MEMS and CMOS technologies. Work is being carried out in two main directions. In
the first approach, functional layers are placed on previously made wafers and then structured during system integration.
These layers can contain other materials or non-CMOS-compatible materials, making integration of additional functions
(for example, sensor effects) possible. The second approach is vertical system integration (3D integration) on the wafer
level. This makes it possible to have a combination of chips from various functional groups, which cannot be sensibly
integrated in a monolithic way because of the material used or related costs involved. High-frequency ICs can be combined
with silicon or MEMS chips in a interposer or wafer level integration. Additional functional layers can be integrated
to support such on-top system integration processes. Merging chips in the frame of such MOTT approaches is perhaps
only the beginning of a multi-disciplinary fusion process.

In the beginning of the integrated microelectronics era, materials science and electrical engineering teamed up with
chemistry to create the microelectronics engineering. Over decades we developed this “hetero-system integration” to a
very high expertise towards a well-defined integrated silicon technology process. However, at the cost of further multidisciplinary
approaches such strict limitations have been route to success in silicon process technology and crystallized
the field into only a few main process concepts for processors, controllers and memory over these decades. Only in the
very last few years we have opened these strict rules when facing the stones of the red brick wall again and again. Now
we allow different new materials in silicon technologies partially also enabling new functions. The synonym heterosystem
integration is relative and valid only in the beginning, after some time of expertise and optimization such approaches
will always lead to a kind of homogeneous process technology if additionally the cost models behind can be
tailored to fit the product perspectives of the market.

A modular solid state technologies approach for future electronic systems could allow an additional degree of freedom
in a soft transition or mutation of silicon technologies into possible even new Nano device concepts or quantum electronics
as addressed in the Beyond CMOS approaches. It concentrates back on the strength of microelectronics engineering
in the early times being the merge of interdisciplinary topics (hetero-integration) into a technology process. It
allows us to address the new challenges of a human and environment centered technology for cost-efficient and highest
performance systems integration. We need now to add organics and biology to the materials science, electronics engineering
and chemistry to overcome the existing borders to the human and the environment centered micro systems integration.
Thinking about the possible development of systems technologies we obviously have to merge many different
functions from electronics over optics, towards, chemical and biological, fluidic and pneumatic functions. Systems of
the future have to be lowest power and energy autarkic and recyclable as well as sustainable. The pressing loads onto
the society to solve environmental, medical, social, nutrition related, security and logistical challenges can only be answered
by a more general sustainable concept of multi-functional modular technology development.

Paper 10a.4.pdf