Granted Patent B1

US 12,644,842 · App. 19/291,087 · Granted Jun 2, 2026

Inventors: Vernon LaLone (Williamsburg, MI); John Woodruff (Traverse City, MI)

Assignee: Wave Lumina Inc.

G01N21/658G01N2201/0221

Abstract

Claims (82)

1 . A method of quantifying a concentration of a chemical entity in a matrix, the method comprising:

combining:

an aqueous sample buffer;

a reporter reagent separate from the chemical entity and including at least one reporter molecule selectively bindable to the chemical entity;

an organic, non-polar extraction solvent including an internal standard entity separate from the chemical entity;

a wash buffer; and

the matrix;

to thereby prepare a liquid precursor;

liquid-liquid extracting a sample of the liquid precursor, wherein the sample includes at least one of:

the internal standard entity; and

a plurality of complexes formed from the chemical entity and the at least one reporter molecule;

scanning the sample with a surface-enhanced Raman spectrometer to produce a dataset including at least one of:

a plurality of target spectra of the chemical entity; and

a plurality of reference spectra of the internal standard entity; and

processing the dataset to thereby quantify the concentration of the chemical entity in the sample formed from the matrix.

2 . The method of claim 1 , wherein combining includes selectively binding the at least one reporter molecule to the chemical entity to thereby enhance an intensity of the plurality of target spectra.

3 . The method of claim 1 , wherein combining includes:

adding the aqueous sample buffer and the reporter reagent to the matrix to form a first intermediate in a first container;

adding the organic, non-polar extraction solvent to the first intermediate to form a second intermediate; and

mixing the second intermediate for a first duration.

4 . The method of claim 3 , wherein liquid-liquid extracting includes:

after mixing the second intermediate, settling the second intermediate for a second duration;

after settling the second intermediate, transferring an extract supernatant of the organic, non-polar extraction solvent from the first container to a second container containing the wash buffer to thereby mix the extract supernatant of the organic, non-polar extraction solvent and the wash buffer and form a third intermediate;

mixing the third intermediate for the first duration; and

after mixing the third intermediate, settling the third intermediate for the second duration.

5 . The method of claim 4 , wherein liquid-liquid extracting further includes:

after settling the third intermediate, depositing the extract supernatant of the organic, non-polar extraction solvent onto a pre-calibrated surface-enhanced Raman spectroscopy (SERS) substrate that includes a surface and a plurality of nanoparticles disposed on the surface; and

after depositing, evaporating the extract supernatant of the organic, non-polar extraction solvent for a third duration that is longer than the second duration to thereby dispose the sample on the plurality of nanoparticles.

6 . The method of claim 5 , wherein depositing includes drop casting the extract supernatant of the organic, non-polar extraction solvent onto the pre-calibrated SERS substrate in a first region.

7 . The method of claim 6 , wherein scanning includes optically scanning the first region and a plurality of additional regions adjacent to the first region on the pre-calibrated SERS substrate to produce at least one of:

from 5 target spectra to 10,000 target spectra; and

from 5 reference spectra to 10,000 reference spectra.

8 . The method of claim 1 , wherein processing includes detecting whether the chemical entity is present in the matrix in from 1 minute to 2 hours.

9 . The method of claim 1 , wherein processing includes detecting the concentration of the chemical entity in the matrix at a level of from 3 parts per trillion of the chemical entity to 100 parts per million of the chemical entity in from 1 minute to 2 hours.

10 . The method of claim 1 , wherein the surface-enhanced Raman spectrometer has at least one acquisition parameter including an integration time, an excitation intensity, and a focal positioning; and

further including verifying the at least one acquisition parameter with an artificial neural network before scanning; and

optionally adjusting the at least one acquisition parameter before scanning.

11 . The method of claim 1 , wherein processing includes transforming the dataset with an artificial neural network to thereby predict the concentration of the chemical entity in the matrix.

12 . The method of claim 11 , wherein transforming includes training the artificial neural network with a training dataset.

13 . The method of claim 12 , wherein training includes creating the training dataset by:

preparing a plurality of samples each having a separate determined concentration of the chemical entity; and

scanning each of the plurality of samples with the surface-enhanced Raman spectrometer.

14 . The method of claim 13 , wherein creating the training dataset includes applying a centering filter to the dataset to reduce variation between each of the plurality of target spectra and a respective one of the plurality of reference spectra and produce a pre-processed centered dataset including at least one of a plurality of centered target spectra and a plurality of centered reference spectra.

15 . The method of claim 14 , wherein creating the training dataset includes normalizing the plurality of centered target spectra with respect to the plurality of centered reference spectra and producing a plurality of normalized target spectra.

16 . The method of claim 15 , wherein creating the training dataset includes partitioning the plurality of normalized target spectra into:

a first training portion configured for inputting into the artificial neural network to thereby produce an estimated concentration for each of the plurality of samples; and

a second validation portion configured for validating a result of the artificial neural network.

17 . A method of quantifying a concentration of per- and polyfluoroalkyl substances (PFAS) in a matrix, the method comprising:

combining:

an aqueous sample buffer;

a reporter reagent separate from the PFAS and including at least one reporter molecule selectively bindable to the PFAS;

an organic, non-polar extraction solvent including an internal standard entity separate from the PFAS and having a conjugated phenyl ring and a carbon-carbon triple bond;

a wash buffer; and

the matrix;

to thereby prepare a liquid precursor;

liquid-liquid extracting a sample of the liquid precursor in less than or equal to 90 minutes, wherein the sample includes at least one of:

the internal standard entity; and

a plurality of complexes formed from the PFAS and the at least one reporter molecule;

depositing the sample onto a pre-calibrated surface-enhanced Raman spectroscopy (SERS) substrate that includes a surface and a plurality of nanoparticles disposed on the surface;

after depositing, scanning the sample with a surface-enhanced Raman spectrometer having at least one acquisition parameter including an integration time, an excitation intensity, and a focal positioning to produce a dataset including at least one of:

a plurality of target spectra of the PFAS; and

a plurality of reference spectra of the internal standard entity;

prior to scanning, verifying the at least one acquisition parameter and with an artificial neural network;

optionally adjusting the at least one acquisition parameter; and

processing the dataset with the artificial neural network to thereby quantify the concentration of the PFAS in the sample formed from the matrix at a level of from 3 parts per trillion of the PFAS to 100 parts per million of the PFAS in from 1 minute to 2 hours.

18 . The method of claim 17 , wherein processing includes filtering out environmental interference within the matrix from at least one of humic acid, chloride, and background fluorescence within the surface-enhanced Raman spectrometer.

19 . A system for quantifying a concentration of a chemical entity in a matrix, the system comprising:

a sample preparation kit configured for preparing a sample solvent extracted from a liquid precursor formed from the matrix, wherein the sample preparation kit includes:

an aqueous sample buffer;

a reporter reagent including at least one reporter molecule selectively bindable to the chemical entity;

an organic, non-polar extraction solvent including an internal standard entity;

a wash buffer; and

a pre-calibrated surface-enhanced Raman spectroscopy (SERS) substrate including:

a surface; and

a plurality of nanoparticles disposed on the surface; and

an instrument configured for analyzing the sample disposed on the plurality of nanoparticles of the pre-calibrated SERS substrate and including:

a surface-enhanced Raman spectrometer configured for optically scanning the sample to produce a dataset including at least one of:

a plurality of target spectra of the chemical entity; and

a plurality of reference spectra of the internal standard entity;

an artificial neural network configured to process the dataset; and

a controller in communication with the surface-enhanced Raman spectrometer and including an instruction set that is executable to process the dataset with the artificial neural network to thereby quantify the concentration of the chemical entity in the matrix.

20 . The system of claim 19 , wherein the instrument is portable.

Assignments (1)

References Cited (88)