ASTM D8431-22

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: D8431 − 22
Standard Test Method for
Detection of Water-soluble Petroleum Oils by A-TEEM
Optical Spectroscopy and Multivariate Analysis
This standard is issued under the fixed designation D8431; 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 responsibility of the user of this standard to establish appro-
priate safety, health, and environmental practices and deter-
1.1 This test method covers the (1) detection of trace level
mine the applicability of regulatory limitations prior to use.
(µg/L range) of oil and petroleum (water-soluble fraction)
1.5 This international standard was developed in accor-
pollutants in surface and ground drinking water sources, (2)
dance with internationally recognized principles on standard-
identification of the compounds, and (3) alerting analysts with
ization established in the Decision on Principles for the
a contaminant concentration prediction. This test method
Development of International Standards, Guides and Recom-
facilitates identification and quantification from 20 to 1000
mendations issued by the World Trade Organization Technical
µg/L of target contaminants, including: water-soluble fraction
Barriers to Trade (TBT) Committee.
of aromatic compounds from the BTEX family (benzene,
toluene, ethylbenzene, and xylenes) and naphthalene from the
2. Referenced Documents
polycyclic aromatic hydrocarbon (PAH) group, referred to as
2.1 ASTM Standards:
BTEXN in this test method, in water samples with up to 15
D1129 Terminology Relating to Water
mg/L of dissolved organic carbon (DOC). The main approach
D1193 Specification for Reagent Water
involves analyzing and characterizing key water intake loca-
D2777 Practice for Determination of Precision and Bias of
tions before the treatment and developing the contaminant
Applicable Test Methods of Committee D19 on Water
library. The water-soluble (BTEXN) contaminants are associ-
D3650 Test Method for Comparison of Waterborne Petro-
ated with, but not limited to petroleum oils and fuels including
leum Oils By Fluorescence Analysis (Withdrawn 2018)
commercial diesel fuel, gasoline, kerosene, heavy oil, fuel oil
D3694 Practices for Preparation of Sample Containers and
and lubricate oil, etc.
for Preservation of Organic Constituents
1.2 Thedatasetsareanalyzedusingmultivariatemethodsto
D4841 Practice for Estimation of Holding Time for Water
test contaminant identification and quantification. The multi-
Samples Containing Organic and Inorganic Constituents
variate methods include classification and regression algo-
D6046 Classification of Hydraulic Fluids for Environmental
rithms to analyze fluorescence EEM data acquired in the
Impact
laboratory. The common goal of these algorithms is to reduce
D6161 Terminology Used for Microfiltration, Ultrafiltration,
multidimensionality and eliminate noise of fluorescence and
Nanofiltration,andReverseOsmosisMembraneProcesses
background signals. Automated identification-quantification
E169 PracticesforGeneralTechniquesofUltraviolet-Visible
methods linked directly to the instrument acquisition-analysis
Quantitative Analysis
software are commercially available.
E2617 Practice for Validation of Empirically Derived Mul-
tivariate Calibrations
1.3 The values stated in SI units are to be regarded as
E2719 Guide for Fluorescence—Instrument Calibration and
standard. No other units of measurement are included in this
Qualification
standard.
E2891 Guide for Multivariate Data Analysis in Pharmaceu-
1.4 This standard does not purport to address all of the
tical Development and Manufacturing Applications
safety concerns, if any, associated with its use. It is the
For referenced ASTM standards, visit the ASTM website, www.astm.org, or
This test method is under the jurisdiction of ASTM Committee D19 on Water contact ASTM Customer Service at service@astm.org. For Annual Book of ASTM
andisthedirectresponsibilityofSubcommitteeD19.06onMethodsforAnalysisfor Standards volume information, refer to the standard’s Document Summary page on
Organic Substances in Water. the ASTM website.
Current edition approved May 1, 2022. Published July 2022. DOI: 10.1520/ The last approved version of this historical standard is referenced on
D8431-22. www.astm.org.
Copyright © ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959. United States
D8431 − 22
2.2 U.S. EPA Standards: 3.3.4 EEM—excitation emission matrix
Method 415.3 Determination of Total Organic Carbon and
3.3.5 ISS—intermediate stock solutions
Specific UV Absorbance at 254 nm in Source Water and
3.3.6 OD—optical density
Drinking Water
3.3.7 PAH—polycyclic aromatic hydrocarbon
Method 524.2 Measurement of Purgeable Organic Com-
pounds in Water by Capillary Column Gas 3.3.8 PCA—principal components analysis
Chromatography/Mass Spectroscopy
3.3.9 PLS—partial least squares analysis
Method 8270D Semivolatile Organic Compounds by Gas
3.3.10 PLSDA—partial least squares discriminant analysis
Chromatography/Mass Spectrometry
3.3.11 PPE—personal protective equipment
3. Terminology
3.3.12 PTFE—polytetrafluoroethylene
3.1 Definitions:
3.3.13 QC—quality control
3.1.1 For definitions of terms used in this standard, refer to
3.3.14 RSU—water Raman standard unit
Terminology D1129.
3.3.15 RU—Raman unit
3.2 Definitions of Terms Specific to This Standard:
3.3.16 SIMCA—softindependentmodelingbyclassanalogy
3.2.1 absorbance-transmittance and fluorescence excitation
3.3.17 SSS—standard stock solutions
emission matrix (A-TEEM), n—a spectroscopic technique that
simultaneously measures the absorbance, transmission and 3.3.18 SVM—support vector machine
fluorescence excitation-emission matrix spectra of samples in
3.3.19 SVMDA—support vector machine discriminant
solution.
analysis
3.2.2 colored dissolved organic matter (CDOM), n—the
3.3.20 XGB—extreme gradient boosted tree
optically measureable component of dissolved organic matter,
3.3.21 XGBDA—extreme gradient boosted tree discriminant
also known as chromophoric dissolved organic matter, that
analysis
strongly absorbs short wavelength light ranging from blue to
ultraviolet (UV).
4. Summary of Test Method
3.2.3 dissolved organic matter (DOM), n—the amount of
4.1 Fig. 1 provides a summary flow chart of this test
organic matter in a water sample passing through a 0.45 µm
method.
filter, reported as percent or fraction. D6161
4.2 Asampleofrawwaterbetween3mLto10mLisfiltered
3.2.4 fluorescent dissolved organic matter (FDOM), n—the
with 0.45 µm membrane filter and analyzed by an A-TEEM
fraction of CDOM that fluoresces and strongly absorbs in the
instrument for WSF oil contamination identification and quan-
UV spectrum.
tification.
3.2.5 raw water, n—water that has not been treated. D1129
4.3 Instrument calibration is achieved by first testing a
3.2.5.1 Discussion—Untreated water from wells, surface
negative control blank (either a sealed water standard per
sources, or public drinking water supplies.
13.2.2 or Type I water per 8.1), and then testing a set of target
3.2.6 water accommodated fraction (WAF), n—the pre-
contamination component (BTEXN) standards and assessing
dominately aqueous portion of a mixture of water and a poorly
them by multivariate analysis (see 12.2).
water-soluble material which separates in a specified period of
4.4 The classification or regression model is saved as a
time after the mixture has undergone a specified degree of
method in a predictor dashboard which then outputs reports of
mixing and includes water, dissolved components, and dis-
possiblecontaminantclassificationandorcontaminantconcen-
persed droplets of the poorly water-soluble material. D6046
trations in the raw water sample (see 13.3 and Section 14).
3.2.7 water-soluble fraction (WSF), n—water-soluble fluo-
rescent low-molecular weight aromatic compounds that are
5. Significance and Use
typically present in oil and petroleum products. D6046
5.1 Source water protection calls for a rapid and reliable
3.3 Abbreviations:
optical method to identify and quantify the oil spill
3.3.1 AFU—arbitrary fluorescence unit
contamination, such as water-soluble fraction of aromatic
3.3.2 BTEX—benzene, toluene, ethylbenzene, and xylenes
compounds from the BTEX family (benzene, toluene,
3.3.3 DOC—dissolved organic carbon ethylbenzene, and xylenes) and naphthalene from the polycy-
clic aromatic hydrocarbon (PAH) group.
5.2 This test method identifies the presence of contamina-
AvailablefromUnitedStatesEnvironmentalProtectionAgency(EPA),William
tion and quantifies the target contamination component(s) to
Jefferson Clinton Bldg., 1200 Pennsylvania Ave., NW, Washington, DC 20460,
provide a threshold-based alert signal.
http://www.epa.gov.
https://cfpub.epa.gov/si/si_public_file_download.cfm?p_download_
5.3 This test method can be used by drinking water treat-
id=525073&Lab=NERL
ment plant operators and decision makers as a first line of
https://www.epa.gov/sites/default/files/2015-06/documents/epa-524.2.pdf
defense for both initially detecting petroleum product spills, as
https://19january2017snapshot.epa.gov/sites/production/files/2015-12/docu-
ments/8270d.pdf well as tracking attenuation over time, in source water to
D8431 − 22
FIG. 1 Method Flowchart
prevent contaminant uptake into the processed water and Such an instrument should be capable of meeting the specifi-
treatment infrastructure. cations outlined in Annex A1, Table A1.1. See Ref (1).
7.2 Fluorescence Cuvette Cells—Quartz cells with 4 clear
6. Interferences
windows, made from fluorescence-free synthetic quartz glass
6.1 Naturally occurring fluorescing materials, such as
with a path length of 10 mm and a height of 45 mm.
humic/fulvic acids and protein-like components, may interfere
7.2.1 The UV transmittance range should be ideally in the
with the identification of target chemicals with relatively low
operational range of 200 nm to 800 nm. The maximum
quantum yields especially those compounds with fluorescence
operational temperature of the quartz material should be at
exciting and emitting in the near ultraviolet (UV) region. A
least 600 °C.
ruggedness study conducted prior to ILS determined the range
7.2.2 Quartz flow cells with a path length of 10 mm are
of DOC up to 15 mg/Ldoes not have a significant effect on the
equipped with optional automatic sipper accessory which
test results.
provides flow-through capability and does not require cell
stirrer.
6.2 The typical upper limit of turbidity in drinking water
sources is site-specific. High turbidity in source water may be
7.3 Cuvette Cell Stirrer—A small magnetic cell stirrer with
expected to interfere with the fluorescence measurements
a controller. Place stirrer into the bottom of the cell compart-
because of the particulate light scattering effects that are
ment. The cell is then placed on top of the magnet drive.
enhanced at lower wavelengths (higher energies). This method
7.4 Magnetic Stir Bars—PTFE-coated magnetic stir bars
is designed to alleviate the effects of turbidity on the water-
with appropriate dimensions to fit within the quartz cuvette
soluble components detection by filtering out any particles
cells and provide rapid vertical and horizontal mixing.
greater than 0.45 µm. A ruggedness study conducted prior to
7.5 Water Bath—Aliquid circulating temperature controller.
ILS determined the range of turbidity up to 20 NTU does not
Includesallnecessarytubingforconnectiontosamplechamber
have a significant impact on the test results.
using quick-release pipe couplings. Recommend setting ex-
6.3 Solvent or blank contamination, or issues due to im-
periment temperature at 25 °C.
proper glassware cleaning may cause interferences.
7.6 Syringe Filters—0.45 µm membrane filters. EPA
Method 415.3 (2.2) recommends two hydrophilic filter mem-
7. Apparatus
brane materials, polyethersulfone and polypropylene.
7.1 Fluorescence Spectrophotometer (or
Spectrofluorometer) —An instrument recording in the spectral
8. Reagents
range of 200 nm to at least 800 nm for both excitation and
8.1 Purity of Water—The purity of reagent water should be
emission responses with Inner-filter effect correction function.
checked by running a water blank prior to running samples
(refer to Section 13, Procedure).Type I water is acceptable and
preferred in this test method (Specification D1193).
The sole source of supply of the apparatus known to the committee at this time
is HORIBAInstruments Incorporated, 20 Knightsbridge Rd, Piscataway, NJ 08854.
If you are aware of alternative suppliers, please provide this information to ASTM
International Headquarters. Your comments will receive careful consideration at a The boldface numbers in parentheses refer to the list of references at the end of
meeting of the responsible technical committee, which you may attend. this standard.
D8431 − 22
8.2 BTEXN Standards—Neat compounds or high purity 10.4 The test unit of sample absorbance is in optical density
–1
standards are required.These may be purchased from commer- (OD) in units of cm . The test unit of sample fluorescence
cial sources. The following BTEXN compounds are required: EEM is either in arbitrary fluorescence unit (AFU) or normal-
8.2.1 Benzene (CAS # 71-43-2) ized to water Raman unit (RU) (see Section 13).
8.2.2 Toluene (CAS # 108-88-3)
8.2.3 Ethylbenzene (CAS # 100-41-4) 11. Preparation of Apparatus
8.2.4 o-Xylene (CAS # 95-47-6)
11.1 Follow the manufacturer’s instructions for instrument
8.2.5 m-Xylene (CAS # 108-38-3)
start-up and warm-up.
8.2.6 p-Xylene (CAS # 106-42-3)
11.2 Start water bath and ensure it is at equilibrium at room
8.2.7 Naphthalene (CAS # 91-20-3)
temperature (22 °C to 25 °C).
8.3 Cleaning Reagent—All glassware must be scrupulously
11.2.1 Ensuretheinstrument’ssamplechamberisconnected
clean. The necessary level of cleanliness can be achieved by
to the water bath.
performing all of the steps in Test Method D3650.
8.4 Methanol—Spectroscopic grade methanol (preferred)
12. Calibration and Standardization
shall be used in all tests unless otherwise stated.
12.1 Instrument Calibration:
NOTE 1—Every solvent has a UV-Vis absorbance cutoff wavelength.
12.1.1 Calibrate the instrument according to the manufac-
The solvent cutoff is the wavelength below which the solvent itself
absorbs all of the light. When choosing a solvent be aware of its
turer’s instructions, Practice E169, and Guide E2719.
absorbance cutoff and where the compound under investigation is thought
12.2 Method Calibration:
to absorb. If they are close, choose a different solvent. Refer to Practice
E169 for a table of solvent cutoffs. 12.2.1 Standard Solution Preparation:
12.2.1.1 Prepare analyte stock solutions using spectroscopic
8.5 Nitric Acid, 68 % to 70 %—Concentrated nitric acid for
grademethanol(see8.4).Dilutetheneatcommercialanalytical
cleaning cuvettes.
standards (see 8.2) to intermediate stock solutions (ISS)
9. Hazards containing mixtures of compounds of interest in methanol (see
8.4). ISS should be prepared fresh the day when a new
9.1 The analytes in this test method are known carcinogens.
calibration is carried out.
Neat standards and stock solutions should be handled and
12.2.1.2 Generally, prepare calibration standards using cali-
prepared in a fume hood. The lab personnel should don
brated volumetric pipettes to dilute the ISS into disposable
appropriate PPE, such as safety glasses, gloves, and lab coat to
glass test tubes using methanol (see 8.4). Place a piece of
minimize exposure to these chemicals used in this test method.
Parafilmoverthetopofthetesttubesandvortexvigorouslyfor
10. Sampling, Test Specimens, and Test Units approximately 30 s to ensure mixing. Table 1 shows the
example preparation of a BTEXN calibration standard with the
10.1 CollectsamplesinaccordancewithPracticesD3694as
final concentration of 100 µg/L for each analyte. The BTEXN
applicable.
standard stock solutions (SSS) are first prepared by spiking
10.2 Analyzesampleswithin2huponthetimeofcollection
neat standards into methanol and diluting to 10 mL for each
(Practices D3694).
analyte. The SSS concentrations are calculated using spike
NOTE 2—Samples may be preserved and stored according to Practices
volume and analyte density. A portion of SSS is spiked into
D3694 and D4841.
anotherflaskcontainingmethanolanddilutedwithmethanolto
10.3 For calibration samples, dilute BTEXN standards (see
10 mL to make intermediate standard solutions (ISS) with a
8.2) then spike into raw water for calibration standards as
concentration of 20 000 µg/L. Note that neat naphthalene is a
described in Section 12.
solid. Weigh 0.05 g neat naphthalene and add to a flask
containing 10 mL methanol. Dissolve and dilute to the 10-mL
10 mark with methanol.Aportion of this is diluted in methanol in
Reagent Chemicals, American Chemical Society Specifications, American
a separate flask to achieve an ISS of naphthalene of
Chemical Society, Washington, D.C. For suggestions on the testing of reagents not
listed by the American Chemical Society, see Annual Standards for Laboratory
20 000 µg⁄L. The ISS is spiked into each sample tube and
Chemicals, BDH Ltd., Poole, Dorset, U.K., and the United States Pharmacopeia
diluted to 9 mL with raw water. The final concentrations of
and National Formulary, U.S. Pharmacopeial Convention, Inc. (USPC), Rockville,
analytes are 100 µg/L in the calibration standards.
MD.
TABLE 1 Example Preparation of Calibration Standards
Vol. of ISS Final
Vol. of SSS Vol. of
Density of Vol. of neat spiked to Calibration
SSS conc. SSS conc. Vol. of ISS ISS conc. to spiked to Calibration
Analyte neat solution solution Calibration Standard
(mg/L) (µg/L) (mL) (µg/L) ISS Standard
(mg/µL) (µL) Standard conc.
(µL) (µL)
(µL) (µg/L)
Benzene 0.8765 28 2454 2 454 000 10 20 000 81 45 9000 100
Toluene 0.87 38 3306 3 306 000 10 20 000 60 45 9000 100
Ethylbenzene 0.8665 38 3293 3 293 000 10 20 000 61 45 9000 100
Xylenes (total) 0.864 38 3283 3 283 000 10 20 000 61 45 9000 100
Naphthalene – – 5000 5 000 000 10 20 000 40 45 9000 100
D8431 − 22
12.2.1.3 See Table 2 of example sample BTEXN final discriminant analysis (SVMDA), soft independent modeling
concentrations. by class analogy (SIMCA), and extreme gradient boosted tree
12.2.1.4 Prepare a baseline sample by filtering 3 mL to
discriminant analysis (XGBDA). Build regression models
10 mL raw water with a 0.45 µm membrane filter. Use this as
using techniques such as principal component analysis (PCR),
the 0 concentration sample (baseline). Insert the 0 concentra-
partial least squares analysis (PLS), support vector machine
tion sample. Measure EEM spectra following steps in Section
(SVM), and extreme gradient boosted tree (XGB) (2-5). The
13. An example is to start with 2.8 mL filtered raw water in a
spectral (X-block) calibration dataset is further preprocessed
clean cuvette with a clean stir bar.
using mean-centering and clutter removal using the full rank
extended mixture model. Concentration (Y-block) calibration
NOTE 3—To avoid cross-contamination or sample carryover problem, it
data is preprocessed using mean-centering. Evaluate the re-
isnotrecommendedtouseflowcellorflow-throughsystemforperforming
calibration spiking samples. The flow-through system is suitable for
gression model effectiveness based on calibration, cross-
routine monitoring of this test method as mentioned in Section 13,
validation, and validation set prediction correlation coeffi-
Procedure.
cients. Classification is based on the resulting confusion matrix
12.2.1.5 Prepare all calibration standards in the same man-
parameters for both the strict (p > 0.5) and the most probable
ner as their position arises in the batch operation. All samples
rule of the total number of positive identifications. Select the
are spiked then filtered before data acquisition. Turn on the
best-performing algorithm to be incorporated into the final
stirrerinthesamplecompartmenttoensurethoroughmixingof
method file.
the sample. Allow to stir for 60 seconds prior to data acquisi-
12.3 Models Validation:
tion to ensure homogenous mixing of the sample solution. The
stirring speed was determined to be an insignificant factor in
12.3.1 Refer to Practice E2617 for model validation. Model
the ruggedness study.
validation must be performed with an independent test set. No
12.2.1.6 When introducing solvent-based standard into wa-
model is ready for deployment until a proper validation has
terincuvette,avoiddilutingmorethan3 %volume,(thatis,90
been performed. In general, a minimum of 20 validation
µL standard to 3000 µL water). To do this, use the smallest
samplesisrecommended,andtheyshouldnotcontainthesame
calibrated pipette volume practical.
raw water matrix or replicates from the calibration samples.
Validation samples should span the ranges of the independent
NOTE 4—It is optimal to prepare each calibration standard in four
variable, that is, concentration values, over which the calibra-
aliquots and analyze each aliquot four times to ensure enough dataset for
an accurate multivariate model.
tion will be used.
NOTE 5—It is optimal to collect initial calibration dataset with different
12.3.2 Revalidate when the instrument conditions change,
raw water samples that cover sufficient variations in DOC and turbidity
such as hardware or software updates. Ongoing periodic
characteristics to ensure model robustness. Once an initial calibration
model is considered acceptable, and after testing, it should be subject to
revalidation should be monitored as more samples are
model validation before deployment (section 12.3).
collected, especially when there are potentially challenging
–1
NOTE 6—The absorbance OD should be less than 1 cm to be in
254 nm
samples, such as raw water samples with high DOC or high
thelinearrangewithfluorescence.DilutethesampleifOD isgreater
254 nm
–1
turbidity values.
than 1 cm . UseType I water for the dilution process. Record the dilution
factor.
12.4 Models Maintenance:
12.2.2 Calibration Classification and Regression Models:
12.4.1 Refer to Guide E2891 for guidance on models
12.2.2.1 Export the absorbance and processed EEM data
maintenance. It is recommended to build routines that check
files
...