Through Barrier Detection of Ethanol using Handheld Raman Spectroscopy - Conventional Raman versus Spatially Offset Raman Spectroscopy.

  • Fay Nicolson (Creator)

Dataset

Description

The purpose of this data set is to investigate the thicknesses to which two handheld spectrometers can detect Raman signal of ethanol through two plastic barriers - clear acrylic and blue polypropylene.

Abbreviations
SORS = spatially offset Raman spectrsocopy
PCA = Principal component analysis

Figure 2 – Measurements were performed on a Snowy Range Instruments CBEx handheld spectrometer (785 nm). The data was produced into SPC files using Ocean Optics software. These were then converted to Excel files for processing. The data was averaged in excel then baselined, normalised and smoothed in Matlab. The resulting data was then used to produce a graph in Matlab which produced figure 2.

Figure 3 – PCA scores plots discriminating between the plastic and ethanol at plastic thicknesses 9 mm and 10 mm using Raman spectroscopy. 15 spectra of plastic sample (3 samples 5 replicates) were collected at a thickness of 9 mm. In addition, 15 spectra (3 samples 5 samples) of ethanol obscured by 9 mm of plastic were also collected. Then, 15 spectra of plastic sample (3 samples 5 replicates) were collected at a thickness of 10 mm and 15 spectra (3 samples 5 samples) of ethanol obscured by 10 mm of plastic were collected. These data sets were then used to perform principle component analysis using specific Matlab scripts. Measurements were performed on a Snowy Range Instruments CBEx handheld spectrometer (785 nm). The data was produced into SPC files using Ocean Optics software. These were then converted to Excel files for processing.

Figure 4 –SORS spectra of ethanol obscured by Clear acrylic plastic at thicknesses of 5 mm, 10 mm and 15 mm. The spatial offset was increased from 1 mm to 8 mm in 1 mm increments and the offset spectra recorded. 15 spectra (3 samples 5 replicates) were collected at each offset. Reference spectra of the analyte (ethanol) and barrier (plastic) were also taken (3 samples 5 replicates, 15 spectra in total). Measurements were performed using a cobalt light systems resolve instrument 830 nm. Spectra were produced from a database file (labelled 2017-Mar-14 17-09 ResolveRES90033) using specialized software from Cobalt light systems. The resulting spectra can be seen in the excel spreadsheets. The customized software was purpose wrote and confidential.

Figure 5 – SORS spectra of ethanol obscured by 21 mm and 22 mm of clear acrylic plastic. 15 spectra (3 samples 5 replicates) were taken of 21 mm thick plastic and then 15 spectra (3 samples, 5 replicates) were taken of ethanol obscured by 21 mm thick plastic. PCA was then performed on this data set using MATLAB. 15 spectra (3 samples 5 replicates) were then taken of 22 mm thick plastic and then 15 spectra (3 samples, 5 replicates) were taken of ethanol obscured by 22 mm thick plastic. PCA was then performed on this data set using MATLAB. Measurements were performed using a cobalt light systems resolve instrument 830 nm. Spectra were produced from a database file (labelled 2017-Mar-14 17-09 ResolveRES90033) using specialized software from Cobalt light systems. The resulting spectra can be seen in the excel spreadsheets. The customized software was purpose wrote and confidential


Figure 6 - Spectra were taken of ethanol reference, ethanol obscured by 1, 2, 3 and 4 mm of plastic and of 4 mm of plastic. Each data set contains 15 spectra (3 samples 5 replicates). Measurements were performed on a Snowy Range Instruments CBEx handheld spectrometer (785 nm). The data was produced into SPC files using Ocean Optics software. These were then converted to Excel files for processing. The data was averaged in excel then baselined, normalised and smoothed in Matlab. The resulting data was then used to produce a graph in Matlab. PCA scores plots discriminating between the plastic and ethanol at plastic thicknesses 2 mm and 3 mm using Raman spectroscopy. 15 spectra of plastic sample (3 samples 5 replicates) were collected of plastic at a thickness of 2 mm and 15 spectra (3 samples 5 samples) of ethanol obscured by 2 mm of plastic were also collected. Then, 15 spectra of plastic sample (3 samples 5 replicates) of a 3 mm thickness were collected, followed by 15 spectra (3 samples 5 samples) of ethanol obscured by 3 mm of plastic. These data sets were then used to perform principle component analysis using customised Matlab scripts (PCA). Measurements were performed on a Snowy Range Instruments CBEx handheld spectrometer (785 nm). The data was produced into SPC files using Ocean Optics software. These were then converted to Excel files for processing.


Figure 7 – SORS spectra collected at 8 mm offset at plastic thicknesses of 2, 4 and 6 mm, alongside plastic (PP) reference spectra and ethanol (EtOH) reference spectra. For each of the 5 data sets (2, 4, 6 mm, EtOH, and PP references) spectra were collected from 3 samples, 5 replicates. Measurements were performed using a cobalt light systems resolve instrument 830 nm. Spectra were produced from a database file (labelled 2017-Mar-14 17-09 ResolveRES90033) using specialized software from Cobalt light systems. The resulting spectra can be seen in the excel spreadsheets. The customized software was purpose wrote and confidential. PCA was also performed on data collected using 9 mm and 10 mm thick plastic.15 spectra (3 samples 5 replicates) were taken of 9 mm thick plastic and then 15 spectra (3 samples, 5 replicates) were taken of ethanol obscured by 9 mm thick plastic. 15 spectra (3 samples 5 replicates) were then taken of 10 mm thick plastic and then 15 spectra (3 samples, 5 replicates) were taken of ethanol obscured by 10 mm thick plastic. PCA was then performed on these data sets using MATLAB. Measurements were performed using a cobalt light systems resolve instrument 830 nm. Spectra were produced from a database file (labelled 2017-Mar-14 17-09 ResolveRES90033) using specialized software from Cobalt light systems. The resulting spectra can be seen in the excel spreadsheets. The customized software was purpose wrote and confidential

Supplementary information

Figure S1 – Spectra of ethanol obscured by 21 mm thick clear acrylic plastic (3 samples, 5 replicates) were collected to produce two data sets. One set of spectra (total 15) were collected at the zero position, the other set at the 8 mm offset position. These two data sets were used to carry out a scaled subtraction using custom scripts in written for MATLAB.

Following this spectra of ethanol obscured by 22 mm thick clear acrylic plastic (3 samples, 5 replicates) were collected to produce two data sets. One set of spectra (total 15) were collected at the zero position, the other set at the 8 mm offset position.
These two data sets were used to carry out a scaled subtraction using custom scripts in written for MATLAB.

All Measurements were performed using a cobalt light systems resolve instrument 830 nm. Spectra were produced from a database file (labelled 2017-Mar-14 17-09 ResolveRES90033) using specialized software from Cobalt light systems. The resulting spectra can be seen in the excel spreadsheets. The customized software was purpose wrote and confidential.

Figure S2 - SORS spectra of ethanol obscured by Blue polypropylene plastic at thicknesses of 2 mm, 4 mm and 6 mm . The spatial offset was increased from 1 mm to 8 mm in 1 mm increments and the offset spectra recorded. 15 spectra (3 samples 5 replicates) were collected at each offset. Reference spectra of the analyte (ethanol) and barrier (plastic) were also taken (3 samples 5 replicates, 15 spectra in total). Measurements were performed using a cobalt light systems resolve instrument 830 nm. Spectra were produced from a database file (labelled 2017-Mar-14 17-09 ResolveRES90033) using specialized software from Cobalt light systems. The resulting spectra can be seen in the excel spreadsheets. The customized software was purpose wrote and confidential.

A scaled subtraction was also performed. Spectra of ethanol obscured by 6 mm thick blue polypropylene plastic (3 samples, 5 replicates) were collected to produce two data sets. One set of spectra (total 15) were collected at the zero position, the other set at the 8 mm offset position. These two data sets were used to carry out a scaled subtraction using custom scripts in written for MATLAB.

Figure S3 - PCA was also performed on data collected using 9 mm and 10 mm thick plastic.15 spectra (3 samples 5 replicates) were taken of 9 mm thick plastic and then 15 spectra (3 samples, 5 replicates) were taken of ethanol obscured by 9 mm thick plastic. 15 spectra (3 samples 5 replicates) were then taken of 10 mm thick plastic and then 15 spectra (3 samples, 5 replicates) were taken of ethanol obscured by 10 mm thick plastic. PCA was then performed on these data sets using MATLAB. Measurements were performed using a cobalt light systems resolve instrument 830 nm. Spectra were produced from a database file (labelled 2017-Mar-14 17-09 ResolveRES90033) using specialized software from Cobalt light systems. The resulting spectra can be seen in the excel spreadsheets. The customized software was purpose wrote and confidential.
Date made available6 Apr 2017
PublisherUniversity of Strathclyde
Date of data production1 Nov 2015 - 31 Mar 2017

Cite this

Nicolson, F. (Creator). (6 Apr 2017). Through Barrier Detection of Ethanol using Handheld Raman Spectroscopy - Conventional Raman versus Spatially Offset Raman Spectroscopy. . University of Strathclyde. Data_for_pure(.zip). 10.15129/c43cc987-469c-4afe-ac09-de91a0bf5608