This paper reports on the application of sub-wavelength structured single layer reflectors in a Fabry-Perot-Interferometer
(FPI) that are used in order to replace distributed Bragg reflectors (DBR). A pair of two-dimensional arrays of ring
resonators was analyzed. A 100 nm thin Al layer is regularly patterned to form a meta-surface structure. It shows high
reflectance in a sufficiently wide wavelength range.
This design approach has the advantage that an optimization can be done by varying geometry parameters of lateral
structures only. Moreover, the material is highly compatible to standard MEMS processes. The structures used here are
rings that are arranged in a two-dimensional array. Thus, parameters to be varied are the inner and the outer ring
diameters and the array pitch. The optimum dimensions of the metal rings have been found iteratively.
Samples were fabricated by structuring of two silicon wafers and subsequent wafer bonding. Deep dry etching of the
reflector carriers from the back side in the areas of the resonator arrays results in free standing silicon nitride membranes
that carry the resonators. The carrier membranes elastically suspend the reflecting ring resonators for variation of the
cavity width. Finally, the substrates are assembled by a wafer bonding technique utilizing a SU-8 polymer layer with a
very definite thickness.
A peak transmittance of 55%, a bandwidth FWHM = 100 nm and a modulation contrast of M = 50:1 were achieved. The
optical performance was measured by fourier transform infrared spectrometer and compared to the simulation results. It
shows a widely good agreement of calculation and measurement.
This work deals with in-line measurement techniques for quantification of important microsystems parameters and
related scattering caused by the process conditions. Material properties, mechanical stress but also geometrical
dimensions and their tolerances are characterized by indirect method, based on specially designed test-structures. This
method involves a data fusion process that combines numerically calculated and experimentally determined information
to estimate sought parameters. Laser Doppler Vibrometrie is used to determine the frequency response function (FRF) of
the test-structure and find out their Eigenfrequencies. For the numerical simulation of the test-structures a parametrical
finite element (FE) model is used and a series of pre-stressed modal analyses have been performed. Hence the
dependence of the Eigenfrequencies on parameters of interest is obtained. The comparison to the measured frequencies
yields the values of the desired parameters. The test-structures are designed, produced and used for microsystems
manufacturing monitoring in Bonding and Deep Reactive Ion Etching (BDRIE) processes. An optimization of the teststructures'
form for a nontrivial goal function is shown. Measurement results of the presented technique are comparable
with results of common characterization methods. The presented technique is both in-situ and non-destructive.
The scanning laser display technology is one of the most promising technologies for highly integrated projection display
applications (e. g. in PDAs, mobile phones or head mounted displays) due to its advantages regarding image quality,
miniaturization level and low cost potential. As a couple of research teams found during their investigations on laser
scanning projections systems, the image quality of such systems is - beside from laser source and video signal
processing - crucially determined by the scan engine, including MEMS scanner, driving electronics, scanning regime
and synchronization. Even though a number of technical parameters can be measured with high accuracy, the test
procedure is challenging because the influence of these parameters on image quality is often insufficiently understood.
Thus, in many cases it is not clear how to define limiting values for characteristic parameters.
In this paper the relationship between parameters characterizing the scan engine and their influence on image quality will
be discussed. Those include scanner topography, geometry of the path of light as well as trajectory parameters.
Understanding this enables a new methodology for testing and characterization of the scan engine, based on evaluation
of one or a series of projected test images. Due to the fact that the evaluation process can be easily automated by digital
image processing this methodology has the potential to become integrated into the production process of laser displays.