Multicomponent quantitative analytical method and apparatus

• US 4,660,151 A
• Filed: 09/19/1983
• Issued: 04/21/1987
• Est. Priority Date: 09/19/1983
• Status: Expired due to Term

• Current Assignee: Beckman Instruments, Inc. (Danaher Corporation)
• Original Assignee: Beckman Instruments, Inc. (Danaher Corporation)
• Inventors: Chipman, Russell A., Brown, Christopher W., Obremski, Robert J.
• Primary Examiner(s): Gruber, Felix D.
• Application Number:
  US06/533,833
• Time in Patent Office: 1,310 Days
• Field of Search: 364/497, 364/498, 356/319-334
• US Class Current: 702/23
• CPC Class Codes: G01J 3/42 Absorption spectrometry; Double beam spectrometry; Flicker spectrometry; Reflection spectrometry beam switching arrangements G01J3/08
• CPC Class Codes: G01N 21/274 Calibration, base line adjustment, drift correction

Abstract

A multicomponent quantitative analytical method and apparatus wherein the method includes the steps of and the apparatus is capable of performing steps of obtaining a plurality of calibration spectra, transforming the calibration spectra using a transform with orthogonal basis vectors, obtaining a calibration matrix relating the transform spectra to concentrations of analytes in the calibration samples, obtaining a spectrum for an unknown sample, transforming the unknown sample spectrum, and relating the transformed unknown sample spectrum to the calibration matrix to thereby determine the concentration of analytes in the unknown sample.

Claims

1. A multicomponent quantitative analytical method for use with an analytical instrument comprising the steps of:

obtaining calibration values related to spectra for a select number of calibration samples containing predetermined concentrations of selected analytes such that information regarding spectra values proportional to the concentrations of the selected analytes is determined;

performing a transformation of the spectra calibration values using a transform to obtain orthogonal basis vectors;

performing a multivariate analysis to obtain a calibration value matrix relating known concentrations of analytes in the calibration sample to the transformed spectra calibration values;

obtaining values of an unknown spectrum for a sample containing unknown concentrations of analytes;

performing a transformation of the values of the unknown spectrum using said transform having the property of orthogonality;

determining the concentrations of the analytes in the sample using the transformed values of the unknown spectrum and the calibration value matrix; and

displaying or storing information of the concentrations of the analytes.

2. A method is defined in claim 1 additionally comprising the step of excluding from the transformed spectra values terms corresponding to values representing high frequency noise following the step of transforming the calibration values.

3. A method is defined in claim 1 additionally comprising the step of excluding from the transformed spectra values terms corresponding to values representing high frequency noise following the step of transforming the values of the unknown spectrum.

4. A method is defined in claim 1 additionally comprising the step of excluding a portion of the transformed spectra values corresponding to background absorbance following the step of transforming the calibration values.

5. A method is defined in claim 1 additionally comprising the steps of excluding a portion of the transformed spectra values corresponding to background absorbance following the step of transforming the values of the unknown spectra.

6. A method is defined in claim 1 additionally comprising the step of selecting a group of basis vectors on which to perform multivariate analysis in accordance with a predetermined selection function following the step of transforming the calibration values.

7. A method is defined in claim 6 additionally comprising the step of selecting a group of basis vectors using said predetermined selection function for use in relation to the calibration value matrix to determine concentration, following the step of transforming the values of the unknown spectra.

8. A multicomponent quantitative analytical apparatus comprising:

a first means for obtaining calibration values related to spectra for a select number of calibration samples containing predetermined concentrations of selected analytes such that information regarding spectra values proportional to the concentrations of the selected analytes is determined and;

for obtaining values of an unknown spectrum for a sample containing unknown concentrations of analytes;

a second means receiving information signals from said first means, for transforming the calibration values using a transform to obtain orthogonal basis vectors and

for performing a multivariate analysis to obtain a calibration value matrix relating known concentrations of analytes in the calibration sample to the transformed spectra calibration values and

for performing a transformation of the values of the unknown spectrum using said transform having the property of orthogonality and

for determining the concentrations of the analytes in the sample using the transformed values of the unknown spectrum and the calibration value matrix; and

a third means for displaying or storing information of the concentrations of the analytes.

9. An apparatus as in claim 8 wherein said second means additionally includes means for excluding from the transformed spectra values terms corresponding to values representing high frequency noise.

10. An apparatus as in claim 8 for in said second means additionally includes means for excluding a portion of the transformed spectra values corresponding to background absorbance.

11. An apparatus as in claim 8 wherein said second means additionally includes means for selecting a group of basis vectors from the orthogonal basis vectors obtained through transforming the calibration values or the values of an unknown spectrum in accordance with a predetermined selection function such that said select group of basis vectors may be used in relating the calibration value matrix with transformed values of the unknown spectrum to determine concentrations of the analytes.