ASTM D8470-22

Standard Practice for Development and Implementation of Instrument Performance Tests for Use on Multivariate Online, At-Line and Laboratory Spectroscopic Based Analyzer Systems

SIGNIFICANCE AND USE
4.1 If ASTM Committee E13 has not specified an appropriate test procedure for a specific type of instrument, or if the sample specified by a Committee E13 procedure is incompatible with the intended instrument operation, then this practice can be used to develop practical performance tests.  
4.1.1 For instruments which are equipped with permanent or semi-permanent sampling accessories, the test sample specified in a Committee E13 practice may not be compatible with the instrument configuration. For example, for FT-MIR instruments equipped with transmittance or IRS flow cells, tests based on putting polystyrene films into the sample position are impractical. In such cases, this practice suggests means by which the recommended test procedures can be modified by changing the test material or the location of the recommended test material.  
4.1.2 For instruments used in process measurements, the choice of test materials may be limited due to process contamination and safety considerations. The practice suggests means of developing performance tests based on materials which are compatible with the intended use of the analyzer.  
4.2 Tests developed using the practice are intended to allow the user to compare the performance of an instrument on any given day with prior performance, and specifically to compare performance during calibration of the analyzer to performance during validation of the analyzer and during routine use of the analyzer. The tests are intended to uncover malfunctions or other changes in instrument operation, but they are not designed to diagnose or quantitatively assess the malfunction or change. The tests are not intended for the comparison of analyzers of different manufacture.  
4.3 Tests developed using this practice are also intended to allow the user to compare the performance of a primary analyzer used in development of a multivariate model to the performance of secondary analyzers used to apply that model for the analysis of process or p...
SCOPE
1.1 This practice covers basic procedures that can be used to develop instrument performance tests for spectroscopic based online, at-line, laboratory and field analyzers. The practice is intended to be applicable to Raman spectrometers and to infrared spectrophotometers operating in the near-infrared and mid-infrared regions.  
1.2 This practice is not intended as a replacement for specific practices, such as Practices E275, E925, E932, E958, E1421, or E1683 that exist for measuring performance of specific types of laboratory spectroscopic instruments. Instead, this practice is intended to provide guidelines as to how similar practices should be developed when specific practices do not exist for a particular instrument type, or when specific practices are not applicable due to sampling or safety concerns. This practice can be used to develop instrument performance tests for on-line process spectroscopic-based analyzers.  
1.2.1 The performance tests described in this practice typically only evaluate the performance of the infrared spectrophotometer or Raman spectrometer part of the analyzer system, referred to herein as the instrument.  
1.2.2 Instrument performance tests do not typically evaluate performance of analyzer sampling systems.  
1.3 This practice describes univariate level zero and level one tests, and multivariate level A and level B tests which can be implemented to measure instrument performance. These tests are designed to be used as rapid, routine checks of instrument performance. They are designed to uncover malfunctions or other changes in instrument operation, but do not specifically diagnose or quantitatively assess the malfunction or change. The tests are not intended for the comparison of instruments or analyzers of different manufacture.  
1.4 The instrument performance tests described in this practice are used during the development of multivariate calibrations via Practice D8321 to establ...

General Information

• Status: Published
• Publication Date: 30-Jun-2022
• ICS: 19.060 - Mechanical testing
• ICS: 71.040.50 - Physicochemical methods of analysis
• Technical Committee: D02 - Petroleum Products, Liquid Fuels, and Lubricants
• Drafting Committee: D02.25 - Performance Assessment and Validation of Process Stream Analyzer Systems

Relations

• Refers: ASTM D4175-23a - Standard Terminology Relating to Petroleum Products, Liquid Fuels, and Lubricants
• Effective Date: 15-Dec-2023
• Refers: ASTM D6122-23 - Standard Practice for Validation of the Performance of Multivariate Online, At-Line, Field and Laboratory Infrared Spectrophotometer, and Raman Spectrometer Based Analyzer Systems
• Effective Date: 01-Jul-2023
• Refers: ASTM D4175-23e1 - Standard Terminology Relating to Petroleum Products, Liquid Fuels, and Lubricants
• Effective Date: 01-Jul-2023
• Refers: ASTM D6122-19b - Standard Practice for Validation of the Performance of Multivariate Online, At-Line, and Laboratory Infrared Spectrophotometer Based Analyzer Systems
• Effective Date: 01-Jun-2019
• Refers: ASTM D6122-19a - Standard Practice for Validation of the Performance of Multivariate Online, At-Line, and Laboratory Infrared Spectrophotometer Based Analyzer Systems
• Effective Date: 01-May-2019
• Refers: ASTM D6122-19 - Standard Practice for Validation of the Performance of Multivariate Online, At-Line, and Laboratory Infrared Spectrophotometer Based Analyzer Systems
• Effective Date: 01-Jan-2019
• Refers: ASTM D6122-18 - Standard Practice for Validation of the Performance of Multivariate Online, At-Line, and Laboratory Infrared Spectrophotometer Based Analyzer Systems
• Effective Date: 01-Jul-2018
• Refers: ASTM D6122-13 - Standard Practice for Validation of the Performance of Multivariate Online, At-Line, and Laboratory Infrared Spectrophotometer Based Analyzer Systems
• Effective Date: 01-May-2013
• Refers: ASTM D6122-10 - Standard Practice for Validation of the Performance of Multivariate Online, At-Line, and Laboratory Infrared Spectrophotometer Based Analyzer Systems
• Effective Date: 01-May-2010
• Refers: ASTM E131-10 - Standard Terminology Relating to Molecular Spectroscopy
• Effective Date: 01-Mar-2010
• Refers: ASTM E925-09 - Standard Practice for Monitoring the Calibration of Ultraviolet-Visible Spectrophotometers whose Spectral Slit Width does not Exceed 2 nm
• Effective Date: 01-Oct-2009
• Refers: ASTM D6122-09 - Standard Practice for Validation of the Performance of Multivariate Process Infrared Spectrophotometers
• Effective Date: 01-Jun-2009
• Refers: ASTM E1421-99(2009) - Standard Practice for Describing and Measuring Performance of Fourier Transform Mid-Infrared (FT-MIR) Spectrometers: Level Zero and Level One Tests
• Effective Date: 01-Mar-2009
• Refers: ASTM E275-08 - Describing and Measuring Performance of Ultraviolet and Visible Spectrophotometers
• Effective Date: 15-Oct-2008
• Refers: ASTM E932-89(2007) - Standard Practice for Describing and Measuring Performance of Dispersive Infrared Spectrometers
• Effective Date: 01-Dec-2007

Overview

ASTM D8470-22 is the Standard Practice for Development and Implementation of Instrument Performance Tests for Use on Multivariate Online, At-Line, and Laboratory Spectroscopic Based Analyzer Systems. Issued by ASTM International, this standard provides a practical framework for developing performance tests for Raman and infrared spectroscopic analyzers-particularly when existing ASTM Committee E13 procedures are not available or are incompatible with the instrument or application.

Focused on supporting online, at-line, laboratory, and field spectroscopic systems, ASTM D8470-22 ensures the consistency, reliability, and comparability of analyzer performance throughout their lifecycle. The guidance is especially valuable in process industries such as petrochemicals, where spectroscopic analyzers are critical for quality assurance and process monitoring.

Key Topics

• Scope of Application

  • Applicable to Raman spectrometers and near-infrared (NIR) and mid-infrared (MIR) spectrophotometers.
  • Designed for instruments used in process measurement, laboratory analysis, and field applications.

• Performance Test Development

  • Offers procedures to create custom performance tests for analyzers without existing ASTM methods.
  • Addresses compatibility concerns when standard test samples are not suitable due to instrument configuration or process safety.

• Test Types

  • Details on univariate level zero and level one tests (routine and quantitative instrument checks).
  • Multivariate level A and level B tests for comprehensive assessment using calibration and validation samples.

• Test Materials

  • Guidelines on selecting or creating instrument qualification samples, performance verification samples, and optical reference materials.
  • Emphasizes sample compatibility with both process and analyzer configuration.

• Performance Monitoring

  • Performance tests help detect instrument malfunctions, changes in operation, or deviations during calibration, validation, and routine use.
  • Designed for rapid, routine checking rather than diagnosing specific malfunctions or comparing analyzers from different manufacturers.

Applications

ASTM D8470-22 is essential for:

• Process Analyzers: Ensures that online and at-line spectroscopic analyzers maintain their reliability for process monitoring in oil, gas, chemical, and petrochemical industries.
• Laboratory Spectroscopy: Provides a standardized approach when existing methods do not fit modern instrument setups or sample requirements.
• Model Transfer and Calibration: Facilitates the comparison of primary analyzers (used for initial model creation) and secondary analyzers (implementing the model in routine analysis) to maintain quality and consistency.
• Instrument Qualification and Requalification: Assists in qualifying new instruments, requalifying after maintenance, and validating performance during the lifecycle of the analyzer system.
• Adaptable Testing: Offers procedures adaptable to unique instrumentation, such as those with permanent sampling accessories or specific process safety needs.

Related Standards

Professionals using ASTM D8470-22 may also reference the following standards for comprehensive spectroscopic performance assessment:

• ASTM D6122: Validation of Multivariate Infrared and Raman Analyzer Systems
• ASTM E275, E925, E932, E958, E1421, E1683: Methods for evaluating laboratory spectroscopic instrument performance
• ASTM D8321: Development and Validation of Multivariate Analyses for Predicting Product Properties
• ASTM D7940: Analysis of LNG by Fiber-Coupled Raman Spectroscopy
• ASTM E1840, E1866: Guides for spectrophotometer and Raman shift standardization

In summary, ASTM D8470-22 is a crucial standard for anyone involved in the operation, maintenance, or quality assurance of spectroscopic analyzer systems. By enabling the development of effective performance tests tailored to specific instruments and processes, it ensures ongoing analyzer reliability and data integrity in diverse industrial settings.

Standards Content (Sample)

This international standard was developed in accordance with internationally recognized principles on standardization established in the Decision on Principles for the
Development of International Standards, Guides and Recommendations issued by the World Trade Organization Technical Barriers to Trade (TBT) Committee.
Designation: D8470 − 22
Standard Practice for
Development and Implementation of Instrument
Performance Tests for Use on Multivariate Online, At-Line
and Laboratory Spectroscopic Based Analyzer Systems
This standard is issued under the fixed designation D8470; 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 calibrations via Practice D8321 to establish the performance
levelatthetimethecalibrationisdeveloped.Thesametestsare
1.1 Thispracticecoversbasicproceduresthatcanbeusedto
used during validation of analyzers via Practice D6122 to
develop instrument performance tests for spectroscopic based
qualify the working analyzer by demonstrating comparable
online, at-line, laboratory and field analyzers. The practice is
performance.
intended to be applicable to Raman spectrometers and to
1.4.1 Instrument performance tests are used to requalify
infrared spectrophotometers operating in the near-infrared and
instruments after analyzer maintenance.
mid-infrared regions.
1.4.2 Instrument performance tests are used to qualify
1.2 This practice is not intended as a replacement for
instruments in secondary analyzers to which calibrations are
specific practices, such as Practices E275, E925, E932, E958,
being transferred after development on a primary analyzer.
E1421,or E1683 that exist for measuring performance of
1.5 This standard does not purport to address all of the
specific types of laboratory spectroscopic instruments. Instead,
safety concerns, if any, associated with its use. It is the
this practice is intended to provide guidelines as to how similar
responsibility of the user of this standard to establish appro-
practices should be developed when specific practices do not
priate safety, health, and environmental practices and deter-
existforaparticularinstrumenttype,orwhenspecificpractices
mine the applicability of regulatory limitations prior to use.
are not applicable due to sampling or safety concerns. This
1.6 This international standard was developed in accor-
practice can be used to develop instrument performance tests
dance with internationally recognized principles on standard-
for on-line process spectroscopic-based analyzers.
ization established in the Decision on Principles for the
1.2.1 The performance tests described in this practice typi-
Development of International Standards, Guides and Recom-
cally only evaluate the performance of the infrared spectropho-
mendations issued by the World Trade Organization Technical
tometer or Raman spectrometer part of the analyzer system,
Barriers to Trade (TBT) Committee.
referred to herein as the instrument.
1.2.2 Instrument performance tests do not typically evaluate
2. Referenced Documents
performance of analyzer sampling systems.
2.1 ASTM Standards:
1.3 This practice describes univariate level zero and level
D4175 Terminology Relating to Petroleum Products, Liquid
one tests, and multivariate levelAand level B tests which can
Fuels, and Lubricants
be implemented to measure instrument performance. These
D6122 Practice for Validation of the Performance of Multi-
tests are designed to be used as rapid, routine checks of
variate Online, At-Line, Field and Laboratory Infrared
instrument performance. They are designed to uncover mal-
Spectrophotometer, and Raman Spectrometer Based Ana-
functions or other changes in instrument operation, but do not
lyzer Systems
specifically diagnose or quantitatively assess the malfunction
D7940 Practice for Analysis of Liquefied Natural Gas
or change. The tests are not intended for the comparison of
(LNG) by Fiber-Coupled Raman Spectroscopy
instruments or analyzers of different manufacture.
D8321 Practice for Development and Validation of Multi-
1.4 The instrument performance tests described in this
variate Analyses for Use in Predicting Properties of
practice are used during the development of multivariate Petroleum Products, Liquid Fuels, and Lubricants based
on Spectroscopic Measurements
This practice is under the jurisdiction ofASTM Committee D02 on Petroleum
Products, Liquid Fuels, and Lubricants and is the direct responsibility of Subcom-
mittee D02.25 on Performance Assessment and Validation of Process Stream For referenced ASTM standards, visit the ASTM website, www.astm.org, or
Analyzer Systems. contact ASTM Customer Service at [e-mail]. For Annual Book of ASTM
Current edition approved July 1, 2022. Published August 2022. DOI: 10.1520/ Standards volume information, refer to the standard’s Document Summary page on
D8470-22. the ASTM website.
Copyright © ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959. United States
D8470 − 22
D8340 Practice for Performance-Based Qualification of probe located in the process wherein the light interacts with a
Spectroscopic Analyzer Systems process sample, and collecting the transmitted or scattered
E131 Terminology Relating to Molecular Spectroscopy light, returning it to the analyzer where it is analyzed by the
E275 Practice for Describing and Measuring Performance of instrument to produce a spectrum of the process sample.
Ultraviolet and Visible Spectrophotometers
3.2.6.1 Discussion—For some NIR and nearly all Raman
E925 Practice for Monitoring the Calibration of Ultraviolet-
analyzers,thelightwilltypicallybetransportedtoandfromthe
Visible Spectrophotometers whose Spectral Bandwidth
analyzer using fiber optics.
does not Exceed 2 nm
3.2.7 instrument, n—for spectroscopic analyzers used in
E932 Practice for Describing and Measuring Performance of
petrochemical service, the spectrometer or spectrophotometer,
Dispersive Infrared Spectrometers
associated electronics and computer, spectrometer or spectro-
E958 Practice for Estimation of the Spectral Bandwidth of
photometer cell, and if utilized, transfer optics. D6122
Ultraviolet-Visible Spectrophotometers
3.2.8 instrument performance verification sample, n—for
E1421 Practice for Describing and Measuring Performance
multivariate spectroscopic analyzers used in the analysis of
of Fourier Transform Mid-Infrared (FT-MIR) Spectrom-
liquid petroleum products and fuels which employ extractive
eters: Level Zero and Level One Tests
sampling, a material representative of the product being
E1683 Practice for Testing the Performance of Scanning
analyzed which is adequately stored in sufficient quantity to be
Raman Spectrometers
used as a check on instrument performance; instrument perfor-
E1840 Guide for Raman Shift Standards for Spectrometer
manceverificationsamplesareusedininstrumentperformance
Calibration
tests and as checks on calibration transfer, but the samples and
E1866 Guide for Establishing Spectrophotometer Perfor-
their spectra are generally not reproducible long term D6122
mance Tests
3.2.8.1 Discussion—In E1866 and previous versions of
3. Terminology
D6122 and this practice, an instrument performance verifica-
3.1 For terminology related to molecular spectroscopic tion sample was referred to as a test sample.
methods, refer to Terminology E131. For terminology relating
3.2.9 instrument qualification sample, n—for multivariate
to petroleum products, liquid fuels and lubricants, refer to
spectroscopic analyzers used in the analysis of liquid petro-
Terminology D4175.
leum products and fuels, a single pure compound, or a known,
3.2 Definitions: reproducible mixture of compounds whose spectrum is con-
3.2.1 absorbance (A), n—the logarithm to the base 10 of the stant over time such that it can be used in an instrument
reciprocal of the transmittance, (T). performance test. D6122
3.2.9.1 Discussion—In E1866 and previous versions of
A 5 log 1⁄ T 52log T
~ !
10 10
D6122 and this practice, an instrument qualification sample
E131
was referred to as a check sample.
3.2.1.1 Discussion—Absorbance is a measure of the capac-
ity of a substance to absorb light of a specific wavelength.
3.2.10 level A test, n—for spectroscopic analyzers used in
the analysis of liquid petroleum products and fuels, a pass/fail
3.2.2 action limit, n—for multivariate spectroscopic analyz-
ers used in the analysis of liquid petroleum products and fuels, instrument performance test in which the spectrum of a check
or instrument performance verification sample is compared
thelimitingvaluefromaninstrumentperformancetest,beyond
which the multivariate spectroscopic analyzer is expected to against historical spectra of the same sample via a multivariate
analysis.
produce potentially invalid results. D6122
3.2.3 calibration, n—in multivariate spectroscopic
3.2.11 level B test, n—for spectroscopic analyzers used in
measurement, a process for creating a multivariate model the analysis of liquid petroleum products and fuels, a pass/fail
relating component concentrations or sample properties to
instrument performance test in which the spectrum of a check
spectra for a set of known samples, referred to as calibration or instrument performance verification sample is analyzed
samples. D8321
using a multivariate model, and the results of the analysis are
compared to historical results for prior analyses of the same
3.2.4 control limits, n—limits on a control chart that are
sample.
used as criteria for signaling the need for action or for judging
whether a set of data does or does not indicate a state of
3.2.12 level one (1) test, n—for spectroscopic analyzers
statistical control.
used in the analysis of liquid petroleum products and fuels,a
simpleseriesofmeasurementsdesignedtoprovidequantitative
3.2.5 extractive sampling, n—for spectroscopic process
data on various aspects of spectrophotometer performance and
analyzers used in the analysis of liquid petroleum products and
information on which to base the diagnosis of problems.
fuels, the process of transporting a sample from the process,
performing any necessary sample conditioning and finally
3.2.13 level zero (0) test, n—for spectroscopic analyzers
presenting the sample to the analyzer so as to produce a
used in the analysis of liquid petroleum products and fuels,a
spectrum of the process sample.
routinecheckofinstrumentperformance,whichcanbedonein
3.2.6 in situ sampling, n—for spectroscopic process analyz- a few minutes, designed to visually detect significant changes
ers used in the analysis of liquid petroleum products and fuels, in instrument performance and provide a database to determine
the process of transporting light from the analyzer to a sample instrument performance over time.
D8470 − 22
3.2.14 liquid petroleum products and fuels, n—in relation to ible with the intended instrument operation, then this practice
analyzers, any single-phase liquid material that is produced at can be used to develop practical performance tests.
a facility in the petroleum and petrochemical industries and 4.1.1 For instruments which are equipped with permanent
will be in whole or in part of a petroleum product; it is orsemi-permanentsamplingaccessories,thetestsamplespeci-
inclusive of biofuels, renewable fuels, blendstocks, alternative fied in a Committee E13 practice may not be compatible with
blendstocks, and additives. D8340 the instrument configuration. For example, for FT-MIR instru-
ments equipped with transmittance or IRS flow cells, tests
3.2.15 model variables, n—the independent variables de-
based on putting polystyrene films into the sample position are
rived from the calibration spectra which are regressed against
impractical. In such cases, this practice suggests means by
the calibration sample properties to produce the multivariate
which the recommended test procedures can be modified by
model. D6122
changing the test material or the location of the recommended
3.2.15.1 Discussion—For MLR, the model variables would
test material.
bethespectralintensityattheselectedwavelengthsorfrequen-
4.1.2 For instruments used in process measurements, the
cies; for PCR or PLS, the model variables are the Principal
choice of test materials may be limited due to process
Components or latent variables.
contamination and safety considerations.The practice suggests
3.2.16 optical reference material, n—for spectroscopic ana-
means of developing performance tests based on materials
lyzers used in the analysis of liquid petroleum products and
which are compatible with the intended use of the analyzer.
fuels,anopticalfilterorotherdevicewhichcanbeinsertedinto
4.2 Tests developed using the practice are intended to allow
the optical path in the spectrometer, spectrophotometer or
the user to compare the performance of an instrument on any
probe producing a spectrum which is known to be constant
given day with prior performance, and specifically to compare
over time such that it can be used in place of a check or
performance during calibration of the analyzer to performance
instrument performance verification sample in a performance
during validation of the analyzer and during routine use of the
test.
analyzer. The tests are intended to uncover malfunctions or
3.2.17 primary analyzer, n—the analyzer(s) on which cali-
other changes in instrument operation, but they are not de-
bration spectra are collected for the purpose of building a
signed to diagnose or quantitatively assess the malfunction or
multivariate model. D8321
change. The tests are not intended for the comparison of
3.2.18 Raman Shift Standard, n—a physical sample having
analyzers of different manufacture.
known Raman shift characteristics used to determine the
4.3 Tests developed using this practice are also intended to
operating wavelength of laser module and correct for any
allow the user to compare the performance of a primary
short-term or accumulated error. The position of the spectra
analyzer used in development of a multivariate model to the
band(s) of this material can be used to calculate the operational
performance of secondary analyzers used to apply that model
wavelength of the laser:
for the analysis of process or product samples.
λ 5 (1)
i
5. Test Conditions
v¯ 1
S D
λ
s
5.1 When conducting the performance tests, the instrument
where λ is the incident wavelength (the operational wave-
i
should be operated under the same conditions as will be in
length of the laser) in cm, λ is the measured wavelength of
s
effect during its intended use. Sufficient warm-up time should
the Raman band in cm, and v¯ is the accepted standard posi-
be allowed before the commencement of any measurements.
tion of the Raman band of the reference material in
5.1.1 Ifpossible,theopticalconfigurationusedformeasure-
wavenumbers. D7940
ments of instrument performance verification and instrument
3.2.19 secondary analyzer, n—an analyzer not used in the
qualification samples should be identical to that used for
development of the multivariate model, but which will be used
normal operations. If identical optical configurations are not
for analysis of new materials. D8321
possible, the user should recognize that the performance tests
3.2.20 spectral intensity, n—a generic term referring to
may not measure the performance of the entire instrument.
either infrared absorbance or Raman scattering intensity.
5.1.2 Data collection and computation conditions should
D8321
generally be identical to those used in normal operation.
3.2.21 spectral position, n—a generic term referring to
However, instrument performance tests can use data collection
either wavelength or frequency position in a spectrum. D8321
and computation conditions that are more demanding than
those used in normal operation if advantageous for instrument
3.2.22 Spectrum Standard, n—light source of known spec-
standardization.
trum used to standardize/calibrate the detection module and
5.1.3 Spectral data used in performance tests should be date
providing correct mapping of scattered wavelength to physical
and time stamped, and the results of the tests should be stored
pixel coordinates/location. Typically, an atomic emission
in a historical database.
source such as a neon lamp is used for this purpose. D7940
4. Significance and Use 6. Materials Used for Performance Testing
4.1 IfASTM Committee E13 has not specified an appropri- 6.1 The types of materials used for performance testing will
ate test procedure for a specific type of instrument, or if the vary depending on whether the analyzer uses extractive or in
sample specified by a Committee E13 procedure is incompat- situ sampling.
D8470 − 22
6.1.1 Foranalyzersusingextractivesampling,itisgenerally rately pipetted or weighed at ambient temperature. It is
possible to introduce an instrument qualification or instrument recommended that mixtures be independently verified for
performance verification sample into the transmission/ composition prior to use.
scattering cell. The materials used for performance testing are
6.3.3 While mixtures can be used as instrument qualifica-
chosen to be compatible with the instrument configuration, and
tion samples, their spectra may be adversely affected by
to provide spectral features which are adequate for the tests
temperature-sensitive interactions that may manifest them-
being performed.
selves by wavelength (frequency) and spectral intensity
6.1.1.1 A liquid sample used for performance testing will
changes.
generally be in the same physical state (gas, liquid, or solid) as
6.4 Instrument Performance Verification Samples—An in-
the samples to be analyzed during normal operation of the
strument performance verification sample is a process or
instrument.
product sample or a mixture of process or product samples
6.1.1.2 Aliquid sample used for performance testing should
whose spectrum is expected to be constant for the time period
be physically and chemically compatible with the samples
it is used in performance testing. The instrument performance
analyzed during normal operation so as to not represent a
verification sample must be stored in bulk quantities in
safety or product quality issue if it ends up mixed into the
controlled conditions such that the material is stable over time.
process stream being analyzed.
6.4.1 Instrument PerformanceVerification Samples are used
6.1.1.3 A liquid sample used for performance testing pref-
to conduct Level B Test (see 9.2).
erably has a spectrum that is similar to the spectra which will
6.4.2 Sinceinstrumentperformanceverificationsamplesare
be collected during normal operation.
often complex mixtures which cannot be synthetically
6.1.1.4 For infrared systems, a liquid sample used for
reproduced, they can only be used for performance testing for
performance testing will preferably have several significant
limited time periods. If instrument performance verification
absorbances (0.3 < absorbance < 1.0) across the spectral range
samples are used for this purpose, collection of historical data
used for normal operation of the instrument.
onanewinstrumentperformanceverificationsampleshouldbe
6.1.1.5 In order to adequately determine the photometric
initiated before previous instrument performance verification
linearity of the infrared instrument, the peak absorbance for at
samples are depleted. It is recommended that new instrument
least one absorption band of a liquid sample will preferably be
performance verification samples be analyzed sequentially
similar to or slightly greater than the largest absorbance
with old instrument performance verification samples at least
expected for samples measured during normal operation.
15 times before they are used to replace the old instrument
6.1.1.6 A liquid sample used for performance testing will
performance verification sample. The 15 analyses must be
typicallybeintroduceddirectlyintothetransmission/scattering
performed over a time period that does not exceed one month
cell and not pass through any sample conditioning system. As
in duration.
such,itonlyteststheperformanceoftheinstrumentandnotthe
total analyzer system.
6.5 Optical Reference Materials—An optical reference ma-
6.1.2 For infrared analyzers using extractive sampling, in-
terial produces a spectrum which is known to be constant over
strument performance tests can be conducted by inserting an
time. This material may be automatically inserted into the
optical reference material into the optical path when the
optical path either within the instrument or in the fiber optic
transmission cell is empty.
sample probe to allow instrument performance tests to be
performed.
6.2 For analyzers using in situ sampling, introducing an
instrument qualification sample or an instrument performance 6.5.1 If an optical reference material is used routinely to
check or correct the spectral data collection or computation,
verification sample into the actual process sample probe can
only be done if the probe is removed from the process. For then the same material is preferably not used for instrument
performance testing. If the same filter is used, then the part of
analyzers equipped with optical multiplexers, a separate chan-
nel and probe may be used for the performance test. The the filter spectrum used in the performance testing should
preferably differ from that part used to check or correct the
performance test can be conducted on an instrument qualifica-
tion sample, an instrument performance verification sample, or instrument. For example, polystyrene filters are used to stan-
dardize (continuously check and correct) the wavelength scale
on an optical reference material. It should be recognized that
such tests do not verify the performance of the actual process of some dispersive NIR spectrophotometers and Raman ana-
lyzers. For such systems, polystyrene filters are preferably not
probe.
tobeemployedforwavelengthstabilityperformancetesting.If
6.3 Instrument Qualification Samples—Instrument qualifi-
polystyrenefiltersareused,thenthepeaksusedforwavelength
cation samples can be used to conduct performance tests.
stability testing should be different from those used for
Instrument qualification samples are single pure compounds or
standardizing the wavelength scale.
mixtures of compounds of definite composition.
6.3.1 Instrument qualification samples are used to conduct
6.6 Spectrum Standards and Raman Shift Standards—Light
Level A tests (see 9.1). sources with known spectra are used for standardizing and
6.3.2 If mixtures are utilized as instrument qualification calibratingthefrequencyaxisofRamanspectrometers.Atomic
samples, they must be prepared in a repeatable manner and, if emission sources such as neon lamps are used for this purpose.
stored, stored such that the mixture is stable over long periods The peak positions of a Raman Shift Standard such as toluene
of time. In preparing mixtures, components should be accu- can be used in instrument performance testing (Practice
D8470 − 22
D7940). Practice E1840 lists various materials that can be used spectrum may be converted to an absorption spectrum by
as Raman shift standards. taking the negative log before the photometric noise calcu-
lations.
6.7 For Raman systems, the spectral features used in per-
7.2.2 For double beam instruments, a 100 % line spectrum
formance testing should be in the linear range for the detector,
is measured when the two beams are both empty, both contain
above the noise level but below the saturation level. If the
empty matched cells, or both contain reference samples in
linear range of the Raman detector has not previously been
matched cells.
established, a series of mixtures should be measured at various
7.2.3 Photometric noise is measured by fitting a line to the
exposure times to determine the range over which response is
spectrum over a short spectral region centered on the test
linear.
frequency (wavelength). The region should contain at least 11
datapoints,preferablycontains101datapoints,andshouldnot
7. Univariate Measures of Absorbance
exceed 2 % of the spectral range. The line is subtracted from
Spectrophotometer Performance
the spectral data, and the RMS noise is calculated as the square
root of the mean square residual.
7.1 Energy Level Tests—Energy level tests are intended to
detect changes in the radiant power in the instrument beam. 7.2.3.1 If T is the transmittance at the frequency v, then the
i i
slope, m, and intercept, b, of a line through the n data points
Decreases in energy levels may be associated with deteriora-
tion of the instrument source, with contamination or misalign- centered at test frequency v are given by the following:
ment of optical surfaces in the light path, or with malfunctions
nΣiT 2ΣTΣi ΣiT
i i i
m 5 5 (2)
of the detector. 2 2 2
nΣi 2 Σi Σi
~ !
7.1.1 For single beam instruments where background and
Σi ΣT 2ΣiΣiT ΣT
i i i
sample spectra are measured separately at different times, b 5 5 (3)
2 2
nΣi 2 ~Σi! n
energy level tests are generally conducted on a background
The photometric noise is calculated as follows:
spectrum. For double beam instruments where the ratio of
background and sample beam intensities is measured directly,
Σ T 2 mi 1 b
~ ~ !!
i
energy levels can be measured if it is possible to block the
Noise 5Π(4)
RMS
n 2 2
sample beam.
The index i in Eq 2-4 runs from – (n – 1)/2 to (n – 1)/2 (n
7.1.2 Energy levels should be measured at a minimum of
must be odd). The intercept represents the transmittance at test
two fixed frequencies (wavelengths). The frequencies (wave-
frequency v .
lengths) at which energy levels are measured should be chosen
7.2.3.2 If photometric noise is calculated on spectra, the
to avoid interferences due to atmospheric components (for
example, absorptions of water vapor and carbon dioxide) and spectral intensity values, A, are used in place of the transmit-
i
tance values, T,in Eq 2-4. If the abscissa for the spectral data
from interferences due to optical components (for example,
i
OH absorptions in SiO cells and fibers). Preferably, regions is wavelength, then wavelength values, λ, are used in place of
i
the frequency values, v,in Eq 2-4. Calculations should be
where the background spectrum is relatively flat and slowly
i
varying should be used for this test. consistently performed on the same data types.
7.2.4 Increases in the photometric noise can indicate a
7.1.3 To minimize the effects of photometric noise on the
misalignment of optical components, a source malfunction, or
energy level measurement, it is preferable to average the
a malfunction in the detector or electronics.
energy over a narrow frequency (wavelength) window.
Typically, the intensity at five points centered on the test
7.3 Short Term Baseline Stability Test—The transmittance is
frequency are averaged.
monitored at each of the test frequencies (wavelengths) used in
the energy level and photometric noise tests. The intercept
7.2 Photometric Noise Tests—Photometric noise is mea-
calculated in Eq 3 represents the transmittance averaged over
sured at the same frequencies (wavelengths) used for the
the n points around test frequency v . Deviation from 100 %
energy level tests. Preferably, photometric noise tests are
transmittance is an indication of short-term baseline instability
conducted on a 100 % line spectrum. Alternatively, photomet-
and may indicate a malfunction of the instrument.
ric noise tests may be conducted on the spectrum of an
7.3.1 If the tests are conducted on absorbance spectra,
instrument qualification or instrument performance verification
deviations from zero absorbance is used as an indication of
sample at regions where the spectrum is relatively flat, and the
baseline instability.
sample spectral intensity is minimal (<0.1).
7.3.2 If photometric noise tests are conducted on the spec-
7.2.1 For single beam instruments where background and
trum of an instrument qualification or instrument performance
sample spectra are measured separately at different times, a
verification sample, then variations in the absorbance spectrum
100 % line spectrum is obtained by ratioing two successive
at the test frequencies are taken as an indication of short-term
background measurements to obtain a transmittance spectrum.
baseline instability.
If, during normal operation of the instrument, backgrounds are
collected with a reference material in the optical path, then this
7.4 Optical Contamination Tests—The single beam back-
same configuration should be used for performance testing. ground scan which was used for the energy tests is examined
Photometric noise calculations are preferably done directly on
for absorptions which might indicate contamination of optical
the transmittance spectrum. Alternatively, the transmittance surfaces in the beam path.
D8470 − 22
7.4.1 Failure to clean cell or probe windows, IRS surfaces, half-maximum (FWHM) of the peak being measured divided
etc., are the most common source of optical contamination. by the digital resolution and rounded up to the nearest odd
Frequencies (wavelengths) at which typical samples exhibit integer.
maximum absorbance should generally be examined. For
7.6.3.2 Identify the zero-crossing associated with the peak
example, for systems used in hydrocarbon analysis, the regions
absorbance and compute its location by linear interpolation
wheretheC-Hstretchingvibrationsoccurshouldbeexamined.
between the two adjacent points straddling the zero crossing.
Significant increases above a nominal background level may
The zero crossing is taken as a measure of the peak position.
indicate contamination of windows and surfaces.
NOTE 1—Other peak finding algorithms can be used provided that they
7.4.2 Instrument optical surfaces can be contaminated by
accurately track peak position. The procedure described in Annex A1
impuritiesinpurgegases.Forsystemsequippedwithflowcells
should be used to test peak finding algorithms to determine if they are
or circulating liquid temperature control, leaks in connecting
appropriate for this application. It is the user’s responsibility to demon-
lines can expose an optical surface to contamination. Users
strate that the peak finding algorithm is appropriate for monitoring
instrument frequency (wavelength) stability.
should consider possible sources of contamination and deter-
mine appropriate frequencies at which absorptions would
7.7 Resolution Stability Tests—The resolution stability of
result.
the instrument is monitored by measuring the bandwidths of
several absorption peaks in the spectrum of the instrument
7.5 Purge Contamination Tests—For instruments which are
qualification/instrument performance verification sample or
purged to minimize absorptions due to atmospheric
optical filter. At least three peaks are used for the test. If
components, the single beam spectrum used for energy tests
possible, the peaks should be in the upper, middle and lower
should be checked for variations in purge quality. Frequencies
third of the spectral range. Variations in the measured band-
(wavelengths) at which potential contaminants absorb should
widths are taken as an indication that the optical resolution of
be identified, as should baseline points where contaminant
the instrument is varying, suggesting a malfunction.
absorption would be minimal. The absorbance for contami-
nants is calculated as the negative log of the ratio of the peak 7.7.1 The spectral intensities for peaks used in this test are
intensity to the baseline intensity. preferably in the range from 0.37 to 0.75. For peak spectral
intensity outside this range, the resolution stability measure-
7.6 Frequency (Wavelength) Stability Tests—Frequency
ment may show increased sensitivity to photometric noise.
(wavelength) stability tests are conducted by monitoring the
7.7.2 Peaks used for the resolution stability test are prefer-
peak positions of several peaks across the absorption spectrum
ably symmetric in shape and well resolved from neighboring
of the instrument qualification or instrument performance
peaks. If such peaks are not available in the spectrum of the
verification sample or optical filter. At least three peaks are
instrument qualification/instrument performance verification
used for the test. If possible, the peaks should be in the upper,
sampleoropticalfilter,theresultsoftheresolutionstabilitytest
middle, and lower third of the spectral range.
may be variable.
7.6.1 The absorption for peaks used in this test are prefer-
7.7.3 It is recommended that the peak bandwidth be deter-
ably in the range from 0.37 to 0.75. For peak absorptions
mined by the following steps:
outside this range, the wavelength stability measurement may
7.7.3.1 Compute the second derivative of the spectrum by
show greater sensitivity to photometric noise.
applying the appropriate digital filter to the spectrum. A
7.6.2 Peaks used for the frequency stability test are prefer-
commonly used filter has been defined by Savitzky and Golay
ably symmetric in shape and well resolved from neighboring
(1)withcorrectionsbySteiner,Termonia,andDeltour (2),with
peaks. If such peaks are not available in the spectrum of the
application criteria discussed by Willson and Polo (3). The
instrument qualification/instrument performance verification
latterreferencediscussesoptimumfilterparametersbasedupon
sample or optical material, the user should be aware that
the relationship between spectral bandwidth and digitization
changes in instrument resolution will affect the measured peak
interval. A cubic filter is recommended. The number of points
position.
used in the filter should be the quotient of the FWHM of the
7.6.3 It is recommended that the peak position be deter-
peak being measured divided by the digital resolution and
mined by the following steps:
rounded up to the nearest odd integer.
7.6.3.1 Compute the first derivative of the spectrum by
7.7.3.2 Identify the zero crossing on each side of the peak
applying the appropriate digital filter to the spectrum. A
intensity and compute their locations by linear interpolation
commonly used filter has been defined by Savitzky and Golay
3 between the two adjacent points straddling the zero crossings.
(1) with corrections by Steiner, Termonia, and Deltour (2),
The difference in the frequencies of the interpolated zero
withapplicationcriteriadiscussedbyWillsonandPolo (3).The
crossings is taken as a measure of the peak bandwidth.
latterreferencediscussesoptimumfilterparametersbasedupon
the relationship between spectral bandwidth and digitization
7.8 Photometric Linearity Tests—Linearity of the instru-
interval. A cubic filter is recommended. The number of points
ment response is important for quantitative applications.
used in the filter should be the quotient of the full-width-at-
Unfortunately, the absolute photometric linearity cannot be
checked in a quick performance test. To do so would generally
requiretheuseofmultiplestandardsofknownabsorbance.The
test described here is intended only to measure changes in the
The boldface numbers in parentheses refer to a list of references at the end of
this standard. photometric linearity of a instrument.
D8470 − 22
7.8.1 Photometric linearity is tested using the ratio of the the wavelength positions and relative intensities of the various
spectral intensity of two or more peaks in the - spectrum. One spectralfeaturesassociatedwiththecomponents.Eachofthese
peak should have a spectral intensity at or near the maximum axes is calibrated by optical means.
spectral intensity that will be used for normal operations. The
8.2 WavelengthAxis—Forthewavelengthaxiscalibrationto
other peaks are preferably less intense than this maximum. If
be standardized, and thus to know which detected vibration
only two peaks are used, the second peak should be approxi-
frequencies correspond to given spectrum features, it is neces-
mately half the intensity of the first peak.
sarytodeterminethewavelengthfallingoneachgroupofCCD
7.8.2 Linear baselines for each peak are calculated from
pixels. This is done by exposing the detector to light that is
points of minimal spectral intensity on opposite sides of the
comprised of known individual wavelengths. Typically, this is
peaks. The maximum spectral intensity for each peak is
done by arranging for light from a neon bulb to be periodically
corrected for the baseline, and the ratio of the spectral
and automatically imaged onto the detector. The emission
intensities for the two peaks is calculated. The ratio is used to
spectra of neon is comprised of a large number of individual
track changes in the photometric linearity.
wavelengths that are known to great precision, and the physics
associated with neon light ensure wavelength stability through
8. Univariate Measures of Raman Spectrophotometer
time and over a wide range of environmental conditions, thus
Performance
establishing standardization. By recording where each neon
8.1 During the development of the analytic method, corre-
wavelength falls on the CCD, a map can be generated relating
lations are established between spectra and the sample species
pixel to wavelength. Any changes in the position of a given
by taking spectra of known samples. Because of the inherent
wavelengthonthedetectorasaresultofthermalormechanical
linearity of the Raman effect, the method will correctly
effects can thus be corrected by measuring the change with the
measure sample concentrations if the Raman spectra are valid.
neon spectra. Ongoing instrument qualification is done analo-
Thus, the primary task to ensure analyzer calibration is to
gously.
ensure the spectra are calibrated and standardized before the
8.2.1 Example Neon Spectrum—Y-axis: counts, X-axis: Ra-
analyzer is commissioned. There also needs to be a means to
-1
man shift (cm ) (see Fig. 1).
ensure this calibration remains valid over time by using
instrumentqualificationapproaches.Aspectrumisessentiallya 8.3 Intensity Axis Calibration—Optical Path Wavelength
two-dimensional chart with the number of detected photons on Transmission Effıciency—The detection efficiency of signal
the vertical or intensity axis and the wavelength of those photons traveling from the sample, through the probe optics,
photons on the horizontal axis. The mathematical method used through the fiber, and finally, to the detection module, varies
to generate analytic results from the spectra relies on knowing with wavelength. Thus, different molecular vibrations have
FIG. 1 Example Neon Spectrum
D8470 − 22
different detection efficiencies. These efficiencies are unique to 8.3.1 – 8.3.4.This is typically done annually or at a convenient
each instrument. Thus, for a method to be valid and accurate, time when the sample point is out of service. In some cases, it
these variations shall be corrected to yield accurate relative
is desired to check the intensity axis calibration without
abundancecalculations.Duringthesysteminstallationprocess,
removing the probe from its stream mounting. This can be
just before commissioning, reference light with a known
accommodatedbyconsideringthatthesystemcomponentsthat
intensity versus wavelength profile is transmitted through the
can affect intensity calibration are the probe, the fiber-optic
entire optical path by exposing it to the probe collection
cables, and the detection module. The detection module can be
aperture.Aspectrum is taken of this light source, which is then
checked for any changes in intensity calibration by comparing
divided by the known profile, and the result is then stored as a
the ratios of neon peak heights as discussed in 8.3.4. Note that
reference spectrum on the instrument being fabricated. This
this is done automatically and periodically. The fiber-optic
reference spectrum is then used to correct subsequent sample
cable and the probe are passive optical devices that are not
spectra to obtain spectra that are independent of the spectro-
subject to wear or aging absent external modification or
graph throughput. There are several ways to produce this
contamination and, thus, do not require ongoing instrument
reference light.
qualification.Any change that could affect intensity calibration
8.3.1 Broadband Light from Incandescent Bulb—The light
is typically caused either by contamination of the probe
source is typically a tungsten halogen bulb with an integral
windoworfiberconnections,orphysicaldamageofthefiberor
diffuser to ensure uniform illumination, driven by constant
probe. Either of these two scenarios would produce noticeable
current electronics to ensure stable operation. The bulb is
effects in the Raman spectra that can be detected and acted
typically calibrated relative to a National Metrology Institute
upon. The two most likely effects are a large drop in signal or
(NMI) traceable reference source such as National Institute of
a large rise in the baseline background of the spectra, or both.
Standards and Technology (NIST) thus creating a table of
values relating the radiative intensity of the bulb/diffuser
9. Multivariate Measures of Instrument Performance
combination at each wavelength over the spectral range of the
analyzer. The bulb/diffuser combination is presented to the
9.1 Level A Tests—ALevelAperformance test is a pass/fail
collection aperture of the probe and calibration spectra are
test that is sensitive to many of the univariate performance
taken as described above. The bulb must be recertified at the
parametersdiscussedinSection7.LevelAtestsdonotidentify
factory on a yearly basis. This is the preferred approach for
specific failure modes, but merely indicate if the instrument
performing intensity calibration.
performance is within historical bounds. In this test, the
8.3.2 Fluorescent Material—A number of materials readily
spectrum of an instrument qualification sample, an instrument
fluoresce when illuminated by laser light producing a broad-
performanceverificationsampleoranopticalfilteriscompared
band light that has a known spectral profile. NIST produces
to a historical spectrum of the instrument qualification sample,
several types of glass doped with dyes that are specifically
the instrument performance verification sample or the optical
engineered to calibrate Raman instruments. The glass is
filter by multivariate methods (least squares fitting or a
presented to the active probe beam and a reference spectrum is
PCR/PLS model; see Practice D8321 for descriptions of PCR
taken as described above.
and PLS). This procedure can provide some information about
8.3.3 Raman Spectra of Reference Material—Instead of a
specific instrument parameters, but essentially looks for devia-
broadband light source, the Raman spectrum of a known mix
tions in the residual spectrum as compared to the historical
ofmaterialscanbeusedasareference.Themixshouldcontain
residual spectra.
each of the materials, in gravimetrically established
9.1.1 Level A tests are generally applied on instruments
percentages, that are intended to be measured during analyzer
which are in use for multivariate, quantitative analysis. The
operation. The relative intensity of each band in the reference
spectral range used in Level A tests should be comparable to
spectrum will be established by the known Raman signature of
that used in the calibration model for the analysis being
the material and its known molar percentage of the mix.
performed. If the spectrum of the instrument qualification
8.3.4 Spectra from Alternative Sources—Although not used
sample, the instrument performance verification sample or the
forintensityaxiscalibrationandstandardization,thesamelight
optical filter used in the Level A test contains absorptions that
source used for wavelength calibration and standardization can
are significantly higher than those of the typical samples being
be used to validate the intensity axis. For example, each time
analyzed, then these peaks can be excluded from the Level A
a neon spectrum is collected, a series of ratios of the individual
fit.
peaks is calculated. The most robust practice is to generate a
ratioofeachpeakwithrespecttootherpeaksbeingusedforthe
9.1.2 Level A Tests Using a Least Squares Method:
calibration (typically, this is eight to ten peaks). A set of these
9.1.2.1 In this LevelAtest, a least squares fit of the current
ratio values are then stored on the analyzer during commis-
spectrum of the instrument qualification sample, instrument
sioning to be used as an instrument qualification reference.
performance verification sample or optical filter
...