Highly selective chemical and biological sensors

• US 9,037,418 B2
• Filed: 09/19/2013
• Issued: 05/19/2015
• Est. Priority Date: 11/09/2010
• Status: Active Grant

• Current Assignee: GE Infrastructure Technology, LLC (General Electric Company)
• Original Assignee: General Electric Company
• Inventors: Potyrailo, Radislav Alexandrovich, Surman, Cheryl Margaret, Burns, Andrew Arthur Paul, Nagraj, Nandini
• Primary Examiner(s): Le, John H
• Application Number:
  US14/031,951
• Publication Number:
  US 20140025313A1
• Time in Patent Office: 607 Days
• Field of Search: 702/19, 702/23, 702/25, 702/30, 324/602, 324/633, 324/652, 324/655, 324/71.1, 73/310.5, 73/645.3, 310/360, 340/10.41
• US Class Current: 702/23
• CPC Class Codes: G01N 27/02 by investigating impedance
• CPC Class Codes: G01N 27/026 Dielectric impedance spectroscopy electrochemical impedance spectroscopy for measuring corrosion G01N17/02
• CPC Class Codes: G01N 33/483 Physical analysis of biological material
• CPC Class Codes: G01N 33/48792 Data management, e.g. communication with processing unit for in vivo diagnostics A61B5/0002; transmission systems for measured values G08C

Abstract

Methods and sensors for selective fluid sensing are provided. Each sensor includes a resonant inductor-capacitor-resistor (LCR) sensor that is coated with a sensing material. In order to collect data, an impedance spectrum is acquired over a relatively narrow frequency range, such as the resonant frequency range of the LCR circuit. A multivariate signature may be calculated from the acquired spectrum to discern the presence of certain fluids and/or fluid mixtures. The presence of fluids is detected by measuring the changes in dielectric, dimensional, resistance, charge transfer, and other changes in the properties of the materials employed by observing the changes in the resonant electronic properties of the circuit. By using a mathematical procedure, such as principal components analysis (PCA) and others, multiple fluids and mixtures can be detected in the presence of one another, even in a high humidity environment or an environment wherein one or more fluids has a substantially higher concentration (e.g. 10×, 1,000,000×) compared to other components in the mixture.

Claims

1. A method of detecting analytes in a fluid, comprising:

acquiring, with a network analyzer, an impedance spectrum over a resonant frequency range of a single resonant sensor circuit; and

calculating, using a processor configured to perform principal components analysis, a multivariate signature from the acquired impedance spectrum.

2. The method, as set forth in claim 1, wherein calculating a multivariate signature from the acquired impedance spectrum is performed using the real part of the impedance spectrum.

3. The method, as set forth in claim 1, wherein the resonant frequency range is within a range of 0.1 Hz -1000 THz .

4. The method, as set forth in claim 1, wherein acquiring the impedance spectrum is performed in a fluid having greater than 50% humidity.

5. The method, as set forth in claim 1, wherein the multivariate signature is calculated using principal components analysis, canonical correlation analysis, regression analysis, nonlinear regression analysis, discriminate function analysis, multidimensional scaling, linear discriminate analysis, logistic regression, or neural network analysis.

6. The method, as set forth in claim 1, wherein acquiring the impedance spectrum comprises detecting dielectric, dimensional, resistance, charge transfer or other changes of material properties by monitoring changes in properties of the resonant circuit.

7. The method, as set forth in claim 1, wherein the resonant circuit comprises an inductor-capacitor-resistor (LCR) circuit.

8. The method, as set forth in claim 1, comprising:

measuring a real part and an imaginary part of the impedance spectrum of a resonant sensor antenna of the single resonant sensor circuit, wherein the resonant sensor antenna is coated with a sensing material;

calculating at least six spectral parameters of the single resonant sensor circuit coated with the sensing material;

reducing the impedance spectrum to a single data point using multivariate analysis to selectively identify the analyte; and

determining one or more environmental parameters from the impedance spectrum.

9. The method, as set forth in claim 1, comprising:

powering the single resonant sensor circuit to at least two power levels to affect at least one of the dipole moment, the dielectric constant, and the temperature of a sensing material disposed on the single resonant sensor circuit;

collecting spectral parameters of a sensor response of the single resonant sensor circuit at the at least two power levels;

performing multivariate analysis of the spectral parameters from combined impedance spectral profiles of the single resonant sensor circuit at the different power levels; and

calculating values of environmental parameters to which the single resonant sensor circuit is exposed from data produced by performing the multivariate analysis.

10. A method of detecting chemical or biological species in a fluid, comprising:

measuring, with a network analyzer, a real part and an imaginary part of an impedance spectrum of a single resonant sensor antenna coated with a sensing material;

calculating, with the network analyzer, at least six spectral parameters of the single resonant sensor antenna coated with the sensing material;

reducing, with a first processor configured to perform canonical correlation analysis, regression analysis, nonlinear regression analysis, principal components analysis, discriminate function analysis, multidimensional scaling, linear discriminate analysis, logistic regression, or neural network analysis, the impedance spectrum to a single data point using multivariate analysis to selectively identify an analyte; and

determining, with the first processor or a second processor, one or more environmental parameters from the impedance spectrum.

11. The method, as set forth in claim 10, wherein measuring the impedance spectrum and calculating at least six spectral parameters comprises measuring over a resonant frequency range of the single resonant sensor antenna.

12. The method, as set forth in claim 10, wherein calculating at least six spectral parameters comprises calculating a frequency position of the real part of the impedance spectrum, and a magnitude of the real part of the impedance spectrum.

13. The method, as set forth in claim 10, wherein calculating at least six spectral parameters comprises calculating a resonant frequency of the imaginary part of the impedance spectrum, and an anti-resonant frequency of the imaginary part of the impedance spectrum.

14. The method, as set forth in claim 10, wherein determining comprises determining a resistance and a capacitance of the single resonant sensor antenna coated with a sensing material.

15. The method, as set forth in claim 10, wherein reducing the impedance spectrum to a single data point comprises calculating a multivariate signature.

16. The method, as set forth in claim 10, comprising:

acquiring the impedance spectrum over a resonant frequency range of a resonant sensor circuit having the single resonant sensor antenna; and

calculating a multivariate signature from the acquired impedance spectrum.

17. The method, as set forth in claim 10, comprising:

powering an IC chip having the single resonant sensor antenna to at least one power level to affect an impedance spectral profile of a sensor having the IC chip;

collecting spectral parameters of a sensor response of the sensor at the at least one power level;

performing multivariate analysis of the spectral parameters; and

calculating values of environmental parameters to which the sensor is exposed from data produced by performing the multivariate analysis and using stored calibration coefficients.

18. A method for controlling selectivity of a sensor response of a sensor having a single integrated circuit (IC) chip, comprising:

powering, with a diode rectifier, the single IC chip to at least one power level to affect an impedance spectral profile of the sensor;

collecting, with a network analyzer, spectral parameters of the sensor response at the at least one power level;

performing, with a first processor configured to perform canonical correlation analysis, regression analysis, nonlinear regression analysis, principal components analysis, discriminate function analysis, multidimensional scaling, linear discriminate analysis, logistic regression, or neural network analysis, multivariate analysis of the spectral parameters; and

calculating, with the first processor or a second processor, values of environmental parameters to which the sensor is exposed from data produced by performing the multivariate analysis and using stored calibration coefficients.

19. The method, as set forth in claim 18, wherein powering the single IC chip comprises powering the single IC chip to at least one power level in the range from −

50 dBm to +40 dBm.

20. The method, as set forth in claim 18, further comprising calculating a concentration of at least one analyte in the presence of at least one interferent.

21. The method, as set forth in claim 18, wherein performing the multivariate analysis of spectral parameters comprises principal components analysis, canonical correlation analysis, regression analysis, nonlinear regression analysis, discriminate function analysis, multidimensional scaling, linear discriminate analysis, logistic regression, or neural network analysis.

22. The method, as set forth in claim 18, wherein powering the single IC chip comprises powering the single IC chip to at least two different power levels and wherein performing multivariate analysis comprises performing multivariate analysis of the spectral parameters from combined impedance spectral profiles of the sensor at the at least two different power levels.

23. The method, as set forth in claim 22, wherein the impedance spectral profiles are different at each of the at least two different power levels.

24. The method, as set forth in claim 18, wherein powering the single IC chip comprises powering the single IC chip to at least two different power levels and wherein performing multivariate analysis comprises performing multivariate analysis of the spectral parameters, wherein differences in the spectral profiles at each of the at least two different power levels are pronounced in different values of Fp, Fl, F2, Fz, Zp, Z1, Z2, and calculated values of C and R.

25. The method, as set forth in claim 18, further comprising calculating concentration of at least one analyte in the presence of at least one interferent.

26. The method, as set forth in claim 18, comprising:

acquiring an impedance spectrum over a resonant frequency range of the single IC chip; and

calculating a multivariate signature from the acquired impedance spectrum.

27. The method, as set forth in claim 18, comprising:

measuring a real part and an imaginary part of the impedance spectrum of a resonant sensor antenna of the single IC chip, wherein the resonant sensor antenna is coated with a sensing material;

calculating at least six spectral parameters of the resonant sensor antenna coated with the sensing material;

reducing the impedance spectrum to a single data point using multivariate analysis to selectively identify the analyte; and

determining one or more environmental parameters from the impedance spectrum.