Methods for characterizing, classifying, and identifying unknowns in samples

• US 6,606,567 B2
• Filed: 03/01/2001
• Issued: 08/12/2003
• Est. Priority Date: 08/10/1999
• Status: Expired due to Term

• Current Assignee: Battelle Memorial Institute, Eigenvector Research Incorporated
• Original Assignee: Battelle Memorial Institute, Eigenvector Research Incorporated
• Inventors: Grate, Jay W., Wise, Barry M.
• Primary Examiner(s): Hoff, Marc S.
• Assistant Examiner(s): Miller, Craig Steven
• Application Number:
  US09/797,162
• Publication Number:
  US 20010029774A1
• Time in Patent Office: 894 Days
• Field of Search: 702/22-24, 702/27, 702/90, 73/232, 73/232.1, 73/233.5, 73/233.6
• US Class Current: 702/22
• CPC Class Codes: B82Y 30/00 Nanotechnology for materials or surface science, e.g. nanocomposites
• CPC Class Codes: G01N 2291/0256 Adsorption, desorption, surface mass change, e.g. on biosensors
• CPC Class Codes: G01N 2291/0422 Shear waves, transverse waves, horizontally polarised waves
• CPC Class Codes: G01N 2291/0423 Surface waves, e.g. Rayleigh waves, Love waves
• CPC Class Codes: G01N 2291/0426 Bulk waves, e.g. quartz crystal microbalance, torsional waves
• CPC Class Codes: G01N 2291/0427 Flexural waves, plate waves, e.g. Lamb waves, tuning fork, cantilever
• CPC Class Codes: G01N 2291/106 one or more transducer arrays
• CPC Class Codes: G01N 29/022 Fluid sensors based on microsensors, e.g. quartz crystal-microbalance [QCM], surface acoustic wave [SAW] devices, tuning forks, cantilevers, flexural plate wave [FPW] devices microdevices per se B81B
• CPC Class Codes: G01N 33/0031 comprising two or more sensors, e.g. a sensor array electrochemical electrode arrays G01N27/27

Abstract

Disclosed is a method for taking the data generated from an array of responses from a multichannel instrument, and determining the characteristics of a chemical in the sample without the necessity of calibrating or training the instrument with known samples containing the same chemical. The characteristics determined by the method are then used to classify and identify the chemical in the sample. The method can also be used to quantify the concentration of the chemical in the sample.

Claims

1. A method of characterizing an unknown sample comprising:

obtaining a plurality of responses from a multi channel instrument,

modeling each of said plurality of responses as a function of a plurality of descriptors, and

determining said plurality of descriptors from said plurality of responses,

wherein said plurality of responses is equal to or greater than said plurality of descriptors and,

wherein said sample is modeled with said plurality of descriptors.

2. The method of claim 1 wherein:

said responses are modeled as a function of C, S_V, V′

, P′

, and S_P where,

S_V contains any sample specific parameters that influence the response independent of the specific interactions of the sample with each channel

V′

contains said plurality of descriptors,

P′

contains parameters specific to the properties of detector channels,

S_P contains channel specific sensitivity parameters, and

C contains sample concentration information.

3. The method of claim 2 wherein:

said responses are included in matrix R equal to S_V C Z<sup>(V′

P′

)</sup> S_P, where Z is a scalar.

4. The method of claim 3 wherein:

said plurality of sample parameters are determined from V′

_a equal to

5. The method of claim 4 wherein S_V and C are diagonal matrices.

6. The method of claim 5 wherein the sample is a vapor.

7. The method of claim 2 wherein said multi channel instrument includes a plurality of diverse sensors that output a signal that depends on the amount of said sample that interacts with said sensor.

8. The method of claim 7 wherein said amount is selected from the group consisting of mass, volume, and mass plus volume.

9. The method of claim 8 wherein the amount is a volume and S_V is a diagonal matrix with sample specific volumes.

10. The method of claim 8 wherein the amount is a mass and S_V is the identity matrix.

11. A method for characterizing an unknown sample comprising:

determining a plurality of descriptors used to model the sample by modeling a plurality of responses obtained from a multichannel instrument as a function of each of the plurality of descriptors wherein the number of responses is greater than the number of descriptors.

12. The method of claim 11 wherein said plurality of descriptors arc molecular interaction characteristics of said unknown sample, molecular properties of said unknown sample, molecular structural features of said sample, or combinations thereof.

13. The method of claim 11 wherein said plurality of descriptors are related to a plurality of solubility properties of said samples.

14. The method of claim 11 wherein said plurality of descriptors are vapor solvation parameters.

15. The method of claim 11 wherein said plurality of descriptors are parameters in a linear free energy relationship.

16. The method of claim 11 wherein said plurality of descriptors are parameters in a linear solvation energy relationship.

17. The method of claim 11 wherein said plurality of descriptors are descriptors in a quantitative structure activity relationship.

18. The method of claim 11 wherein said plurality of descriptors are parameters in a principle components equation.

19. The method of claim 11 wherein the response of each channel of said multichannel instrument is modeled by an equation including a term that is related to said plurality of descriptors.

20. The method of claim 11 wherein the plurality of responses of said multichannel instrument are related to the thermodynamic partitioning of said unknown sample between phases.

21. The method of claim 11 wherein the plurality of responses of said multichannel instrument are related to the partitioning of said unknown sample between the ambient environment and a plurality of sorbent phases.

22. The method of claim 11 wherein said multichannel instrument utilizes a plurality of gas chromatographic columns.

23. The method of claim 11 wherein said multichannel instrument utilizes a plurality of sensors having sorbent phases.

24. The method of claim 23 wherein the sorbent phase is selected from the group comprising a solid surface, a self assembled monolayer, a molecular multilayer, an amorphous solid phase, a liquid, a membrane and a thin film.

25. The method of claim 11 wherein the sorbent phase is a stationary sorbent phase.

26. The method of claim 23 wherein the sorbent phase is a polymer.

27. The method of claim 23 wherein said multichannel instrument utilizes a plurality of acoustic wave sensors selected from the group of thickness shear mode devices, surface acoustic wave devices, Leaky surface acoustic wave devices, surface transverse wave devices, Love wave devices, shear-horizontal acoustic plate mode devices, flexural plate wave devices, thin film resonators, and thin rod flexural devices.

28. The method of claim 11 wherein said multichannel instrument utilizes a plurality of acoustic wave sensors coated with polymers and stationary phases.

29. The method of claim 11 wherein said multichannel instrument utilizes a plurality of optical Sensors.

30. The method of claim 11 wherein said multichannel instrument utilizes a plurality of chemiresistor sensors.

31. The method of claim 11 wherein said multichannel instrument utilizes a plurality of chemiresitor sensors having a sorbent layer phase and a solid electronic conductor.

32. The method of claim 11 wherein said multichannel instrument utilizes a plurality of sensors selected from the group of electrochemical and field effect transistor sensors.

33. The method of claim 11 wherein said multichannel instrument utilizes a plurality of sensors selected from the group of microbeam, microbar and microcantilever sensors.

34. The method of claim 11 wherein said multichannel instrument includes a plurality of diverse sensors that output a signal that depends on the amount of said sample that interacts with said sensor.

35. The method of claim 34 wherein said amount is mass, volume, or mass plus volume.

36. Computer-readable media tangibly embodying a program of instructions executable by a Computer to perform the method of claim 11.

37. The method of claim 11 further comprising utilizing one or more of said descriptors to classify said unknown sample as belonging to a class of chemicals with certain properties.

38. The method of claim 11 further comprising utilizing one or more of said descriptors to classify said unknown sample as belonging to a class of chemicals with certain structural features.

39. The method of claim 11 further comprising comparing said descriptors to a table of descriptors of known chemicals to determine the identity of said unknown sample.

40. The method of claim 11 wherein said responses are modeled as a function of C, S_V, V′

, P′

, and S_P, where,

S_V contains any sample specific parameters that influence the response independent of the specific interactions of the sample with each channel

V′

contains said plurality of descriptors,

P′

contains parameters specific to the properties of detector channels,

S_P contains channel specific Sensitivity parameters, and

C contains sample concentration information.

41. The method of claim 40 wherein said responses are included in matrix R equal to S_V C Z^(V′

p′

) S_P, where Z is a scalar.

42. The method for characterizing un unknown sample comprising:

determining a plurality of descriptors used to model the sample from a plurality of responses obtained from a multichannel installment wherein the number of responses is at least equal to the number of descriptors and wherein the plurality of responses are mathematically modeled with coefficients for each descriptor.

43. The method of claim 42 wherein said coefficients are coefficients in a linear free energy relationship.

44. The method of claim 42 wherein said coefficients are coefficients in a linear solvation energy relationship.

45. The method of claim 42 wherein said coefficients are coefficients in a quantitative structure activity relationship.

46. The method of claim 42 wherein said coefficients are coefficients in a principle components equation.

47. The method of claim 42 wherein said coefficients are coefficients in a linear free energy relationship related to sorbent phase properties.

48. The method of claim 42 wherein said coefficients are coefficients in a linear free energy relationship based on thermodynamic partition coefficients.

49. The method of claim 42 wherein said coefficients are determined from instrument responses to known compounds.

50. Computer-readable media tangibly a program of instructions executable by a computer to perform the method of claim 42.

51. The method of claim 42 further comprising utilizing one or more of said descriptors to classify said unknown sample as belonging to a class of chemicals with certain properties.

52. The method of claim 42 further comprising utilizing one or more of said descriptors to classify said unknown sample as belonging to a class of chemicals with certain structural features.

53. The method of claim 42 further comprising comparing said descriptors to a table of descriptors of known chemicals to determine the identity of said unknown sample.

54. The method of claim 42 wherein said responses are modeled as a function of C, S_V, V′

, P′

, and S_P where,

S_V contains any sample specific parameters that influence the response independent of the specific interactions of the sample with each channel

V′

contains said plurality of descriptors,

P′

contains parameters specific to the properties of detector channels,

S_P contains channel specific sensitivity parameters, and

C contains sample concentration in formation.

55. The method of claim 54 wherein said responses are included in matrix R equal to S_V C Z^(V′

p′

) S_P, where Z is a scalar.

56. A method for characterizing an unknown sample comprising:

defining a matrix P containing coefficients for a plurality of descriptors used to model the sample; and

determining said plurality of descriptors from said matrix P and a plurality of responses obtained from a multichannel instrument wherein the number of responses is at least equal to the number of descriptors and wherein the plurality of responses are a function of the coefficients.

57. The method of claim 56 wherein matrix P contains LSER coefficients determined from measurements of thermodynamic partitioning.

58. The method of claim 56 wherein said plurality of descriptors are determined from said matrix P, said plurality of responses, and a matrix V containing solvation parameters for vapors.

59. The method of claim 56 wherein said plurality of descriptors are determined from said matrix P, said plurality of responses, and a matrix R containing responses of acoustic wave vapor sensors with sorbent interactor layers and wherein said matrix P contains LSER coefficients determined from measurements of responses of acoustic wave vapor sensors to known vapors.

60. The method of claim 56 further comprising the step of utilizing one or more of said descriptors to classify said unknown sample as belonging to a class of chemicals with certain properties.

61. The method of claim 56 further comprising the step of utilizing one or more of said descriptors to classify said unknown sample as belonging to a class of chemicals with certain structural features.

62. The method of claim 56 further comprising the stop of comparing said descriptors to a table of descriptors of known chemicals to determine the identity of said unknown sample.

63. Computer-readable media tangibly embodying a program of instructions executable by a computer to perform the method of claim 56.