Previous Topic Back Forward Next Topic
Print Page Dr. Frank Dieterle
 
Ph. D. ThesisPh. D. Thesis 4. Experiments, Setups and Data Sets 4. Experiments, Setups and Data Sets 4.5. Data Sets 4.5. Data Sets 4.5.2. Homologous Series of the Low Alcohols4.5.2. Homologous Series of the Low Alcohols 4.5.2.2. Methanol, Ethanol and Propanol by the SPR Setup4.5.2.2. Methanol, Ethanol and Propanol by the SPR Setup
Home
News
About Me
Ph. D. Thesis
  Abstract
  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
    4.1. The Sensor Principle
    4.2. SPR Setup
    4.3. RIfS Sensor Array
    4.4. 4l Miniaturized RIfS Sensor
    4.5. Data Sets
      4.5.1. Refrigerants R22 and R134a
      4.5.2. Homologous Series of the Low Alcohols
        4.5.2.1. Methanol and Ethanol by the SPR Setup
        4.5.2.2. Methanol, Ethanol and Propanol by the SPR Setup
        4.5.2.3. Methanol, Ethanol and Propanol by the RIfS Array and the 4l Setup
        4.5.2.4. Quaternary Mixtures of Alcohols by the RIfS Array and the SPR Setup
  5. Results Kinetic Measurements
  6. Results Multivariate Calibrations
  7. Results Genetic Algorithm Framework
  8. Results Growing Neural Network Framework
  9. Results All Data Sets
  10. Results Various Aspects of the Frameworks and Measurements
  11. Summary and Outlook
  12. References
  13. Acknowledgements
Publications
Research Tutorials
Links
Contact
Search
Site Map
Guestbook
Print this Page Print this Page

4.5.2.2.   Methanol, Ethanol and Propanol by the SPR Setup

For the multicomponent analysis of ternary alcohol mixtures, 2 data sets were recorded [193] using the SPR setup with a 300 nm Makrolon layer (more details in [194]). The calibration data set is a 6-level equidistant full factorial design whereby the relative saturation pressures of methanol, ethanol and 1-propanol were varied between 0 and 0.035. 18 concentration combinations were measured two more times for an estimation of the experimental signal inaccuracies. The validation data set is a 5-level equidistant full factorial design with relative saturation pressures between 0.0035 and 0.0315. Additional 15 measurements were performed with the pure analytes at the five concentration levels. 9 measurements were identified as outliers according to [192] whereby 7 measurements of the calibration data set and 2 measurements of the validation data set were removed. During all measurements the polycarbonate was exposed to the analyte mixtures for 600 seconds and afterwards to dry synthetic air during 4760 seconds with 28 time points recorded during exposure to analyte and 22 time points recorded during exposure to air resulting in 50 input variables in total. The sensor responses are shown in figure 16 for different concentrations of the single analytes. All measurements were performed in random order within the two data sets. The validation data set was recorded one month after the calibration data set and the signals of this data set were averaged using two measurements.

 

figure 16: Signals of the SPR device during exposure to analyte and then to synthetic air for different concentrations of methanol ethanol and 1-propanol.

Page 57 © Dr. Frank Dieterle, 14.08.2006 Navigation