Previous Topic Back Forward Next Topic
Print Page Frank Dieterle
Ph. D. ThesisPh. D. Thesis 9. Results  All Data Sets9. Results All Data Sets 9.3. Methanol, Ethanol and 1-Propanol by the RIfS Array and the 4l Setup9.3. Methanol, Ethanol and 1-Propanol by the RIfS Array and the 4l Setup 9.3.1. Signals and Data Preparation9.3.1. Signals and Data Preparation
About Me
Ph. D. Thesis
  Table of Contents
  1. Introduction
  2. Theory Fundamentals of the Multivariate Data Analysis
  3. Theory Quantification of the Refrigerants R22 and R134a: Part I
  4. Experiments, Setups and Data Sets
  5. Results Kinetic Measurements
  6. Results Multivariate Calibrations
  7. Results Genetic Algorithm Framework
  8. Results Growing Neural Network Framework
  9. Results All Data Sets
    9.1. Methanol and Ethanol by SPR
    9.2. Methanol, Ethanol and 1-Propanol by SPR
    9.3. Methanol, Ethanol and 1-Propanol by the RIfS Array and the 4l Setup
      9.3.1. Signals and Data Preparation
      9.3.2. Mixtures by the RIfS Array
      9.3.3. Mixtures by the 4l Setup
      9.3.4. Conclusions
    9.4. Quaternary Mixtures by the SPR Setup and the RIfS Array
    9.5. Quantification of the Refrigerants R22 and R134a in Mixtures: Part II
  10. Results Various Aspects of the Frameworks and Measurements
  11. Summary and Outlook
  12. References
  13. Acknowledgements
Research Tutorials
Downloads and Links
Site Map
Print this Page Print this Page

9.3.1.   Signals and Data Preparation

The RIfS principle detects changes of the optical thickness nd of the sensitive layer whereby in most cases the changes of the thickness d are dominant [260]. As Makrolon is a hard polycarbonate with a high glass transition temperature, mainly the refractive index n and not the thickness d changes during exposure to analyte resulting in bad signal to noise ratios for the measurements using the 2 RIfS setups (see figure 69). Although the signal to noise ratio can be improved by the use of thicker sensitive layers and longer times of exposure to analyte (for the bigger molecules), both approaches drastically increase the time needed for measurements and between measurements (see figure 25), which is not desired for sensor applications. Additionally, the thickness of the sensitive layer of the 4l setup can be varied only within certain limits [261]. Therefore, it is investigated if the reduction of noise by the use of a smoothing technique is beneficial for the calibration. In figure 69, the sensor signals of the 80 nm Makrolon layer are shown for 1-propanol and for methanol before and after the application of an FFT filter for smoothing. It is visible that 1-propanol has the poorest signal to noise ratio, as not all micropores are occupied by the analyte in contrast to methanol. Thus, the sensor signals for propanol benefit most from smoothing. On the other hand, the sensor response of methanol shows a counterproductive effect of smoothing. The rectangular sensor signal of methanol before smoothing changes into a rounder sensor profile after smoothing whereas the shape of the wave-like sensor signal of 1-propanol is practically not affected by smoothing. This means that the shapes of the sensor responses of the different analytes are made more similar by smoothing. Thus, the quantification of the analytes is rendered more difficult as the quantification is based on the differences of the shapes. To investigate the effects of smoothing the data are evaluated separately with and without smoothing. Additionally, the effect of smoothing for sensitive layers with a different thickness is investigated as the thickness influences the signal to noise ratio.

figure 69: Sensor responses before and after filtering with an FFT-Filter for the 80 nm layer.

Page 128 © Frank Dieterle, 03.03.2019 Navigation