ASTM E2310-04(2009)

This document is not an ASTM standard and is intended only to provide the user of an ASTM standard an indication of what changes have been made to the previous version. Because
it may not be technically possible to adequately depict all changes accurately, ASTM recommends that users consult prior editions as appropriate. In all cases only the current version
of the standard as published by ASTM is to be considered the official document.
Designation:E2310–04 Designation:E2310–04(Reapproved2009)
Standard Guide for
Use of Spectral Searching by Curve Matching Algorithms
with Data Recorded Using Mid-iInfrared Spectroscopy
This standard is issued under the fixed designation E 2310; the number immediately following the designation indicates the year of
original adoption or, in the case of revision, the year of last revision. A number in parentheses indicates the year of last reapproval. A
superscript epsilon (´) indicates an editorial change since the last revision or reapproval.
1. Scope
1.1 Spectral searching is the process whereby a spectrum of an unknown material is evaluated against a library (database) of
digitally recorded reference spectra. The purpose of this evaluation is classification of the unknown and, where possible,
identification of the unknown. Spectral searching is intended as a screening method to assist the analyst and is not an absolute
identification technique. Spectral searching is not intended to replace an expert in infrared spectroscopy. Spectral searching should
not be used without suitable training.
1.2 The user of this document should be aware that the results of a spectral search can be affected by the following factors
described in Section 5: (1) Baselines, baselines, (2) sample purity, (3) Absorbance linearity (Beer’s Law), (4) sample thickness,
(5) sample technique and preparation, (6) physical state of the sample, (7) wavenumber range, (8) spectral resolution, and (9)
choice of algorithm.
1.2.1 Many other factors can affect spectral searching results.
1.3 The scope of this document is to provide a guide for the use of search algorithms for mid-infrared spectroscopy. The
methods described herein may be applicable to the use of these algorithms for other types of spectroscopic data, but each type of
data search should be assessed separately.
1.4 TheEuclideandistancealgorithmandthefirstderivativeEuclideandistancealgorithmaredescribedandtheirusediscussed.
The theory and common assumptions made when using search algorithms are also discussed, along with guidelines for the use and
interpretation of the search results.
1.5 The values stated in SI units are to be regarded as standard. No other units of measurement are included in this standard.
2. Referenced Documents
2.1 ASTM Standards:
E 131 Terminology Relating to Molecular Spectroscopy
E 334 Practice for General Techniques of Infrared Microanalysis
E 573 Practices for Internal ReflectanceReflection Spectroscopy
E 1252 Practice for GeneralTechniques of Qualitative InfraredAnalysis Practice for GeneralTechniques for Obtaining Infrared
Spectra for Qualitative Analysis
E 1642 Practice for General Techniques of Gas Chromatography Infrared (GC/IR) Analysis
E 2105 Practice for General Techniques of ThermogravimetricAnalysis (TGA) Coupled withWith InfraredAnalysis (TGA/IR)
E 2106 PracticeforGeneralTechniquesofLiquidChromatography—-Infrared(LC/IR)andSizeExclusionChromatography—-
Infrared (SEC/IR) Analyses
3. Terminology
3.1 Definitions—For general definitions of terms and symbols, refer to Terminology E 131.
3.1.1 Euclidean distance algorithm—the Euclidean distance algorithm measures the Euclidean distance between each library
spectrum and the unknown spectrum by treating the spectra as normalized vectors.The closeness of the match, or hit quality index
(HQI), is calculated from the square root of the sum of the squares of the difference between the vectors for the unknown spectrum
and each library spectrum.
This guide is under the jurisdiction of ASTM Committee E13 on Molecular Spectroscopy and is the direct responsibility of Subcommittee E13.03 on Infrared
Spectroscopy.
Current edition approved Feb. 1, 2004. Published Feb. 2004.
Precision and Bias This guide is under the jurisdiction of ASTM Committee E13 on Molecular Spectroscopy and Separation Science and is the direct responsibility of
Subcommittee E13.03 on Infrared and Near Infrared Spectroscopy.
Current edition approved March 1, 2009. Published March 2009. Originally approved in 2004. Last previous edition approved in 2004 as E 2310 – 04.
For referencedASTM standards, visit theASTM website, www.astm.org, or contactASTM Customer Service at service@astm.org. For Annual Book of ASTM Standards
volume information, refer to the standard’s Document Summary page on the ASTM website.
Copyright © ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959, United States.
E2310–04 (2009)
3.1.2 first derivative Euclidean distance algorithm—in the first derivative Euclidean distance algorithm the Euclidean distance
is also computed, except the derivative of each spectrum is calculated prior to the Euclidean distance calculation.
3.1.3 hit quality index (HQI)—a table which ranks the library spectra in the database according to their hit quality values (see
7.5).
3.1.4 hit quality value—thespectralsearchsoftwarecompareseachspectruminthedatabasetothatoftheunknownandassigns
a numeric value for each library entry demonstrating how similar the two spectra are.
3.1.4.1 Discussion—There are several methods for assigning hit quality values and either a high or low value can be assigned
as the best match. Refer to the software manufacturers documentation.
3.1.5 normalization—the mathematical technique used to compensate for an intensity difference between two spectra (see 5.1).
3.1.6 peak searching—the process whereby the peak table of the spectrum of an unknown material is evaluated against a library
of peak tables. Each reference spectrum in the library contains a peak table and the peak table is individually compared to the peak
table of the unknown, and assigned a numerical value as to the goodness of fit.
3.1.7 reference spectrum—an established spectrum of a known compound or chemical sample.
3.1.1.1
3.1.7.1 Discussion—This spectrum is typically stored in retrievable format so that it may be compared against the sample
spectrum of an analyte.
3.1.1.2
3.1.7.2 Discussion—This term has sometimes been used to refer to a background spectrum; such usage is not recommended.
3.1.2spectral searching—the process whereby a spectrum of an unknown material is evaluated against a library of digital
reference spectra. Each reference spectrum in the library is individually compared to the spectrum of the unknown, and assigned
a numerical value as to the goodness of fit. To perform this comparison, each data point in the unknown spectrum is compared
to each corresponding point in the reference spectrum.
3.1.3peak searching—the process whereby the peak table of the spectrum of an unknown material is evaluated against a library
of peak tables. Each reference spectrum in the library contains a peak table and the peak table is individually compared to the peak
table of the unknown, and assigned a numerical value as to the goodness of fit.
3.1.4spectral library—a collection of reference spectra stored in a computer readable form, also called a library, database, or
spectral database.
3.1.5
3.1.8 search algorithm—the mathematical formula used to make a point-by-point comparison of two spectra.
3.1.6hit quality value—the spectral search software compares each spectrum in the database to that of the unknown, and assigns
a numeric value for each library entry demonstrating how similar the two spectra are.
3.1.6.1Discussion—There are several methods for assigning Hit Quality values and either a high or low value can be assigned
as the best match. Refer to the software manufacturers documentation.
3.1.7hit quality index (HQI)—a table which ranks the library spectra in the database according to their Hit Quality values (see
7.5).
3.1.8Euclidean Distance algorithm—the Euclidean Distance algorithm measures the Euclidean distance between each library
spectrum and the unknown spectrum by treating the spectra as normalized vectors. The closeness of the match, or, HQI, is
calculated from the square root of the sum of the squares of the difference between the vectors for the unknown spectrum and each
library spectrum.
3.1.9 First Derivative Euclidean Distance algorithm—in the First Derivative Euclidean Distance algorithm the Euclidean
distance is also computed, except the derivative of each spectrum is calculated prior to the Euclidean distance calculation. spectral
library—a collection of reference spectra stored in a computer readable form, also called a library, database, or spectral database.
3.1.10 normalization—themathematicaltechniqueusedtocompensateforanintensitydifferencebetweentwospectra(see5.1).
spectral searching—the process whereby a spectrum of an unknown material is evaluated against a library of digital reference
spectra. Each reference spectrum in the library is individually compared to the spectrum of the unknown, and assigned a numerical
value as to the goodness of fit. To perform this comparison, each data point in the unknown spectrum is compared to each
corresponding point in the reference spectrum.
4. Theory
4.1 Beer’s Law—One of the basic principles that make spectral searching possible is Beer’s Law (see Terminology E 131),
which states that A = abc, where A is the absorbance, a is the absorptivity, b is the sample pathlength, and c is the concentration
of the analyte of interest. As long as Beer’s Law applies, two spectra of the same material recorded under similar conditions can
be made to appear the same by normalization of the data.
NOTE 1—In an ideal case, this is true for transmittance spectra, but there are differences in the spectral peak intensities when reflectance spectra are
compared to transmittance spectra.
5. Spectral Data Pre-Treatment
5.1 Normalization:
5.1.1 Normalization of spectra compensates for the differences in sample quantity (concentration or pathlength, or both) used
E2310–04 (2009)
to generate the reference spectra in the library and that of the unknown. The spectra are normalized over the complete spectral
range of the library.When searching less than the full spectral range of the library, the spectra must be re-normalized over the new
range before an accurate comparison can be made. Normalization of a spectrum for library searching is a two step process. First,
theminimumabsorbancevalueintheselectedspectralrangeissubtractedfromalltheabsorbancesinthesamerange.Theresulting
values are then scaled by dividing by the maximum result value in the range. The end result is a spectrum (or a sub-range portion
of a spectrum) where the minimum value is zero (0) and the maximum is one (1) absorbance. If the range chosen for normalization
has only one or two strong bands and a few medium intensity bands, the range of the spectrum must be reselected or the spectrum
will be dominated by the strong bands in the spectrum and the HQI will be insensitive to weaker fingerprint bands necessary for
identification of a specific compound. Successful compound identification may require the spectral match exclude the strongest
bands, then the normalization will be based on a medium intensity band, and weak fingerprint bands will be emphasized in the
HQI.
5.2 Data Point Matching:
5.2.1 The algorithms used for searching a spectrum against a library use a calculation that mathematically compares the data
points of the spectrum being searched to the data points of the spectra in the library. This requires that the data points in both the
sample and library spectra occur at the same frequency. If the data points in the sample and library spectra are not aligned in this
manner, then one of the spectra must be mathematically altered (interpolated) to make the data points match. Typically the
unknown spectrum being searched is altered to match the data point spacing of the spectra in the library.
5.2.2 Data point matching is commonly accomplished using a linear data point interpolation method. In this method, the slope
and offset of a line segment is calculated between the absorbances of every pair of data points in the spectrum. A new set of
absorbances is calculated by locating the values that occur on the line segments at positions corresponding to the datapoint
frequency of the library spectrum.
6. Conditions or Issues Affecting Results
6.1 Spectral quality is one of the primary conditions or issues that can affect search results.There is no substitute for a carefully
recorded spectrum. There are several conditions or issues that affect spectral quality as pertains to spectral searching. These
conditions or issues apply to both the spectra used to create the reference database and to the unknown spectrum.
6.2 Baselines:
6.2.1 A flat baseline is preferred for the Euclidean distance algorithm as the Euclidean distance algorithm compares each data
point in the unknown spectrum to the corresponding data point in the reference spectrum. The effect of an offset or slope in the
baseline is interpreted as a difference between the two spectra. Therefore, when a spectrum with a sloping baseline or offset is
evaluated using the Euclidean distance algorithm, a simple baseline correction should be used.
NOTE 2—Negative bands can also produce an offset in the baseline as a result of the data normalization process.
6.2.2 The first derivative Euclidean distance algorithm minimizes the effect of an offset or sloping baseline. In this algorithm,
the comparison is made between the difference of a pair of adjacent points in the unknown spectrum to the difference between the
corresponding pair of adjacent points in the reference spectrum. In effect, this causes the first derivative Euclidean distance
algorithm to look only at the differences in the slope of adjacent data points between the two spectra. Fig. 1 shows how the two
algorithms view the same two spectra.
NOTE 3—The first derivative algorithm converts a sloping baseline into an offset that is then eliminated by the normalization procedure.
6.3 Sample Purity:
6.3.1 The physical state of the sample should be as close as possible to the physical state of the reference materials used to
obtain the library. For example, a pure liquid sample would ideally be searched against a library of spectra of only liquid reference
materials.Asamplewhichisprobablyamixture,suchasacommercialformulation,shouldbecomparedtoalibraryofcommercial
formulations.
6.3.2 In some cases the nature of the sample may not be well understood. An unknown sample may be a pure material or a
mixture. It may have additional contaminants that will affect its spectrum by adding spurious ban
...