Availability of reliable flight sensor data and knowledge of the structural behaviour are essential for safe operation of the Ariane launcher. The Ariane launcher is currently monitored by hundreds of electric sensors during test and qualification. Fibre optic sensors are regarded as a potential technique to overcome limitations of recent monitoring systems for the Ariane launcher [1]. These limitations include cumbersome application of sensors and harness as well as a very limited degree of distributed sensing capability. But, in order to exploit the various advantages of fibre optic sensors (high degree of multiplexing, distributed sensing capability, lower mass impact, etc.) dedicated measurement systems have to be developed and investigated. State-of-the-art fibre optic measurement systems often use free beam setups making them bulky and sensitive to vibration impact. Therefore a new measurement system is developed as part of the ESAstudy [2].
There are various methods for film thickness measurement. This paper aims at thin film measurement methods
which work with wavelength-dependent sensor signals, no matter how the signal was captured. These measurement
systems are called Thin Film Reflectometers (TFR). The resolution of thin film reflection measurement
is limited by the spectral resolution of the detection system, the spectral wavelength range and the analyzing
algorithm. Analytical theory of four different algorithms for thin film measurement is described and algorithms
are compared via numerical simulations. Most of these algorithms can be found both in the literature and in
different software-libraries (e.g. MATLAB, LabVIEW...)
To compare different algorithms, the reflected light intensities of a thin film have been simulated for an exemplary
thin glass film and a common off-the-shelf-spectrometer, a broadband visible light source and characteristic
noise levels. The same data was fed for four selected algorithms in order to compare the results. To characterize
algorithms in resolution, range and accuracy, the standard deviation of the output data has been computed for
different spectral windows and resolutions. As a result we can provide a concise recommendation for appropriate
use of the presented TFR algorithms.
Chromatic confocal microscopy is a common way to interrogate topologies and is well understood. Thin film
reflectometry (TFR), on the other hand, is a way to monitor film thicknesses. Semiconductor and optics producing
industries, e.g., require information about topological, film thickness or optical constants. We developed
a spectrometric measuring system which is capable of determining high precision thin film thickness and topographic
information of a specimen at the same time. The spectral intensity distribution reflected by a transparent
thin film differs from a spectroscopic confocal observation by a chromatic measurement head, since the spectral
interference fringes appear in the spectra. The spectrometer-based system interrogates both confocal, as well as
thin film signals employing an analytical model of the chromatic shift of the measuring head, film composition
and a least-square estimator.
Hence, the advantage of this combined measuring system is the concurrent determination of film thickness and
distance to the measuring head. By scanning the surface of a specimen laterally, a both tomo- and topological
image can be acquired. Spacial measurements at test objects were carried out to demonstrate this measurement
principle and the results are discussed.
Numerous processes, e.g. in semiconductor and optics producing industries require film thickness observation.
These measuring systems depend on different working principles, e.g. spectral reflectometry or ellipsometry. The
spectral reflectometry interrogation method can be evaluated by various algorithms depending on resolution and
measuring range demanded. All methods require a broad spectral distribution of the light source in order to
sample the signal sufficiently for parameter extraction. Spectral sampling is often realized using a spectroscope,
which produces equidistant sampling points in frequency space. In contrast to conventional spectrally broad
light sources, the one employed here emits several spectral lines, which are non-equidistantly distributed. It also
introduces problems like variations of intensity in the output spectrum and narrow wavelength bands, in which
the reflected spectrum can be investigated. Non-equidistant sampling points additionally imply problems in
conventional analysis algorithms, e.g. a FFT anticipates equidistant sampling points. Narrow wavelength bands
imply little information to interrogate at the same spectral resolution of the interrogator. Strong variations
of intensity lead to high noise levels at wavelengths with low intensities. Therewith, accuracy, resolution and
measuring range are limited. An interrogator based on a Hg-Ar light source, a fiber coupler and a commercial
spectroscope is described in this work. Both, accuracy and measuring range, are investigated by simulation and
are experimentally proven on a glass on silicon demonstrator. Introducing an advanced algorithm, uncertainties
invoked by the source's spectral and intensity distribution are minimized and resolution as well as measuring
range are increased.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.