Determining the quantity of a taggant in a liquid sample

• US 9,995,681 B2
• Filed: 07/24/2015
• Issued: 06/12/2018
• Est. Priority Date: 09/28/2010
• Status: Active Grant

• Current Assignee: Authentix, Inc.
• Original Assignee: Authentix, Inc.
• Inventors: Conroy, Jeffrey L., Forshee, Philip B., Cronin, Paul J., Soyemi, Olusola
• Primary Examiner(s): Porta, David
• Assistant Examiner(s): Malevic, Djura
• Application Number:
  US14/808,041
• Publication Number:
  US 20150355091A1
• Time in Patent Office: 1,054 Days
• Field of Search: 2504591, 2504581
• CPC Class Codes: G01N 2021/6439 with indicators, stains, dyes, tags, labels, marks
• CPC Class Codes: G01N 21/643 non-biological material
• CPC Class Codes: G01N 33/2882 Markers

Abstract

Device and methods for detecting/quantifying a fluorescent taggant in a liquid sample. Generally, the liquid samples are fuels having low concentrations (measured in ppb) of a fluorescent taggant. The detection/quantification generates a predicted concentration of the fluorescent tagging compound using a process selected from the group of a multivariate process, a background subtraction process, or a combination of both. The invention addresses the detection of an adulteration of gasoline and diesel fuels.

Claims

1. A liquid sample analysis method comprising;

providing an excitation light and a broadband light to a liquid sample;

obtaining, by a fluorescence spectrometer, measured emission pre-spectrum from the liquid sample;

storing the measured emission pre-spectrum in a memory;

determining a peak of the measured emission pre-spectrum, wherein the peak comprises a first intensity;

adjusting, when the first intensity is outside a desired range, an integration time to produce a measured emission spectrum of intensity versus wavelength;

generating a predicted concentration of a fluorescent tagging compound in the liquid sample using the measured emission spectrum;

comparing the predicted concentration of the fluorescent tagging compound in the liquid sample to a preset concentration of the fluorescent tagging compound in the liquid sample; and

determining whether the measured emission spectrum is a member of a class for a sub-model based on spectral fitness metrics,

wherein the spectral fitness metrics comprise a spectrum fit based on the measured emission spectrum, a matrix of spectrum weights, and a matrix of spectrum loadings.

2. The method of claim 1, wherein generating the predicted concentration of the fluorescent tagging in the liquid sample comprises performing a multivariate analysis.

3. The method of claim 2, wherein performing the multivariate analysis comprises:

determining whether the measured emission spectrum is the member of the class for the sub-model based on spectral similarity metrics; and

classifying the measured emission spectrum.

4. The method of claim 1, wherein generating the predicted concentration of the fluorescent tagging in the liquid sample comprises performing a background subtraction analysis.

5. The method of claim 1, further comprising:

determining an absorption of a laser based on a second intensity of the excitation light, and

adjusting the measured emission spectrum based on the absorption of the laser.

6. The method of claim 1, further comprising:

determining an absorbance of a broadband light source based on the broadband light, and

adjusting the measured emission spectrum based on the absorbance of the broadband light source.

7. The method of claim 1, further comprising determining whether the liquid sample is adulterated based on the comparing of the predicted concentration of the fluorescent tagging compound in the liquid sample to the preset concentration of the fluorescent tagging compound in the liquid sample.

8. The method of claim 1, wherein the adjusting comprises:

determining that the first intensity is outside the desired range;

altering the integration time to bring the first intensity into the desired range;

recording a series of spectra using the integration time;

averaging the series to obtain an averaged spectrum; and

dividing intensities at each of a plurality of wavelengths by the integration time to obtain the measured emission spectrum.

9. A liquid sample emission spectrum classification method comprising:

obtaining, by a fluorescence spectrometer, a liquid sample emission pre-spectrum for a liquid sample;

storing the liquid sample emission pre-spectrum in a memory;

determining a peak of the liquid sample emission pre-spectrum, wherein the peak comprises an intensity;

adjusting, when the intensity is outside a desired range, an integration time to produce a liquid sample emission spectrum of intensity versus wavelength;

determining whether the liquid sample emission spectrum is a member of a class for a sub-model based on spectral similarity metrics;

determining whether the liquid sample emission spectrum is the member of the class for the sub-model based on spectral fitness metrics, wherein the spectral fitness metrics comprise a spectrum fit based on the liquid sample emission spectrum, a matrix of spectrum weights, and a matrix of spectrum loadings; and

classifying the liquid sample emission spectrum as in-model. when the liquid sample emission spectrum is a member of the class for the sub-model based on the spectral similarity metrics and the spectral fitness metrics.

10. The method of claim 9, further comprising classifying the liquid sample emission spectrum as out-of-model when the liquid sample emission spectrum is not a member of the class for the sub-model based on the spectral similarity metrics.

11. The method of claim 9, further comprising classifying the liquid sample emission spectrum as out-of-model when the liquid sample emission spectrum is not a member of the class for the sub-model based on the spectral fitness metrics.

12. The method of claim 9, further comprising generating a predicted concentration of a fluorescent tagging compound using the liquid sample emission spectrum and the sub-model.

13. The method of claim 12, further comprising comparing the predicted concentration of the fluorescent tagging compound to a preset concentration of the fluorescent tagging compound.

14. The method of claim 9, further comprising:

obtaining an absorbance spectrum for the liquid sample; and

adjusting the liquid sample emission spectrum based on the absorbance spectrum.

15. The method of claim 9, wherein the sub-model is associated with an inclusive flag or an exclusive flag.

16. The method of claim 9, wherein determining whether the liquid sample emission spectrum is a member of the class for the sub-model based on the spectral similarity metrics comprises comparing the liquid sample emission spectrum to a discriminant function.

17. The method of claim 16, wherein the discriminant function defines a geometric shape in a principle component (PC) space, and wherein comparing the liquid sample emission spectrum to a discriminant function comprises determining a location of the liquid sample emission spectrum relative to the geometric shape.

18. The method of claim 9, wherein determining whether the liquid sample emission spectrum is a member of the class for the sub-model based on the spectral fitness metrics comprises comparing the liquid sample emission spectrum to a scaled sum of square error (SSSE) threshold.

19. A liquid sample analysis method comprising:

obtaining, by a fluorescence spectrometer, a measured emission pre-spectrum for a liquid sample;

storing the measured emission pre-spectrum in a memory;

determining a peak of the measured emission pre-spectrum, wherein the peak comprises an intensity;

adjusting, when the intensity is outside a desired range, an integration time to produce a measured emission spectrum of intensity versus wavelength;

selecting a sub-model for the measured emission spectrum;

determining whether the measured emission spectrum is a member of a class for the sub-model based on spectral fitness metrics, wherein the spectral fitness metrics comprise a spectrum fit based on the measured emission spectrum, a matrix of spectrum weights, and a matrix of spectrum loadings

determining whether the measured emission spectrum is in-model for the sub-model;

determining whether the sub-model is associated with an inclusive flag; and

performing a quantitative analysis on the measured emission spectrum when the measured emission spectrum is in-model for the sub-model and the sub-model is associated with the inclusive flag.

20. The method of claim 19, wherein performing a quantitative analysis comprises performing a multivariate analysis on the measured emission spectrum.

21. The method of claim 19, wherein determining whether the measured emission spectrum is in -model for the sub-model comprises:

determining whether the measured emission spectrum is the member of the class for the sub-model based on spectral similarity metrics; and

determining that the measured emission spectrum is in-model for the sub-model when the measured emission spectrum is a member of the class for the sub-model based on the spectral similarity metrics and the spectral fitness metrics.

22. The method of claim 19, further comprising:

obtaining an absorbance spectrum for the liquid sample; and

adjusting the measured emission spectrum based on the absorbance spectrum.