ASTM D8460 2022 Standard Test Method for Quantification of Volatile Organic Compounds Using Proton Transfer Reaction Mass Spectrometry

Designation D8460 − 22 Standard Test Method for Quantification of Volatile Organic Compounds Using Proton Transfer Reaction Mass Spectrometry1 This standard is issued under the fixed designation D8460.

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: D8460 − 22 Standard Test Method for Quantification of Volatile Organic Compounds Using Proton Transfer Reaction Mass Spectrometry1 This standard is issued under the fixed designation D8460; 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 standard Reporting of test results in units other than SI shall not be regarded as nonconformance with this standard Scope 1.1 This test method describes a technique of quantifying the results from measuring various volatile organic compound contents using a chemical ionization mass spectrometer resulting in the production of positively charged target compound ions Depending on the nature of production of so-called primary ions, the associated instruments having the capability to perform such analyses are either named Proton Transfer Reaction Mass Spectrometers (PTR-MS), Selected Ion Flow Tube Mass Spectrometers (SIFT-MS) or, in the most generic term, Mid-pressure chemical ionization mass spectrometers (MPCI-MS) Within this standard, the term PTR-MS is used to represent any of these instrumentations 1.5 All observed and calculated values shall conform to the guidelines for significant digits and rounding established in Practice D6026 1.5.1 The procedures used to specify how data are collected/ recorded or calculated in the standard are regarded as the industry standard In addition, they are representative of the significant digits that generally should be retained The procedures used not consider material variation, purpose for obtaining the data, special purpose studies, or any considerations for the user’s objectives; and it is common practice to increase or reduce significant digits of reported data to be commensurate with these considerations It is beyond the scope of this standard to consider significant digits used in analysis methods for engineering data 1.6 This standard may involve hazardous materials, operations, and equipment This standard does not purport to address all of the safety concerns, if any, associated with its use It is the responsibility of the user of this standard to establish appropriate safety, health, and environmental practices and determine the applicability of regulatory limitations prior to use 1.7 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 1.2 Either of the instrument types can be used with the two main mass analyzers on the market, that is, with either quadrupole (QMS) or time-of-flight (TOFMS) mass analyzer This method relates only to the quantification portion of the analysis Due to large differences in user interfaces and operating procedures for the instruments of the main instrument providers, the specifics on instrument operation are not described in this method 1.3 Details on the theoretical aspects concerning ion production and chemical reactions are included in this standard as far as required to understand the quantification aspects and practical operation of the instrument in the field of vapor intrusion analyses Specifics on the operation and/or calibration of the instrument need to be identified by using the user’s manual of the individual instrument vendor A comprehensive discussion on the technique including individual mass-line interferences and in-depth comparison with alternate methods are given in multiple publications, such as Yuan et al (2017) (1) and Dunne et al (2018) (2)2 Referenced Documents 2.1 ASTM Standards:3 D653 Terminology Relating to Soil, Rock, and Contained Fluids D1357 Practice for Planning the Sampling of the Ambient Atmosphere D3740 Practice for Minimum Requirements for Agencies 1.4 Units—Values stated in SI units are to be regarded as standard No other units of measurement are included in this This test method is under the jurisdiction of ASTM Committee D18 on Soil and Rock and is the direct responsibility of Subcommittee D18.21 on Groundwater and Vadose Zone Investigations Current edition approved May 1, 2022 Published June 2022 DOI: 10.1520/ D8460-22 The boldface numbers in parentheses refer to a list of references at the end of this standard For referenced ASTM standards, visit the ASTM website, www.astm.org, or contact ASTM 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 D8460 − 22 3.3.5 FEP—fluorinated ethylene propylene 3.3.6 GC—gas chromatography 3.3.7 ICAL—initial multipoint calibration 3.3.8 IMR—ion molecule reactor 3.3.9 LCS—laboratory control sample 3.3.10 MDL—method detection limit 3.3.11 MS—mass spectrometer 3.3.12 NIST—National Institutes of Standards and Technology 3.3.13 PEEK—polyetheretherketone 3.3.14 PFA—polyfluoroalkoxy alkane 3.3.15 PTFE—polytetrafluoroethylene 3.3.16 PTR-MS—proton transfer reaction - mass spectrometer or spectrometry 3.3.17 QMS—quadrupole mass spectrometer 3.3.18 SDS—safety data sheet 3.3.19 SIFT-MS—selected ion flow tube - mass spectrometer or spectrometry 3.3.20 TOFMS—time-of-flight mass spectrometer 3.3.21 VI—vapor intrusion 3.3.22 VOC—volatile organic compound Engaged in Testing and/or Inspection of Soil and Rock as Used in Engineering Design and Construction D5314 Guide for Soil Gas Monitoring in the Vadose Zone (Withdrawn 2015)4 D5730 Guide for Site Characterization for Environmental Purposes With Emphasis on Soil, Rock, the Vadose Zone and Groundwater (Withdrawn 2013)4 D6026 Practice for Using Significant Digits and Data Records in Geotechnical Data D8408/D8408M Guide for Development of Long-Term Monitoring Plans for Vapor Mitigation Systems E2600 Guide for Vapor Encroachment Screening on Property Involved in Real Estate Transactions Terminology 3.1 Definitions: 3.1.1 For ease of reading, the term PTR-MS is used to reflect any variations of instrumentation as described in 1.1 and 1.2 3.1.2 For definitions of common technical terms used in this standard, refer to the guidelines in Practice D1357 and Terminology D653 3.2 Definitions of Terms Specific to This Standard: 3.2.1 gas analysis, n—involves multiple gas measurements including calibration and zero gas background subtraction, therefore involves multiple gas measurements 3.2.2 gas measurement, n—an analysis performed with a PTR-MS without calibration nor zero gas background subtraction.5 3.2.3 ion molecule reactor, n—the instrument part within the PTR-MS where ionization reactions of the target molecules using primary ions happen 3.2.4 ZeroAir, n—a gas determined to be free of any interfering substances at the reporting limit of the project 3.2.4.1 Discussion—For example, the PTR-MS can be used to perform the analysis or an equivalent methodology or the certificate can be used in case of a certified cylinder.6 3.4 Symbols Used in Equations: 3.4.1 A—a target compound (analyte) 3.4.2 AH+—a protonated target compound 3.4.3 R—reagent ion (primarily hydronium) 3.4.4 CVN(A)= [A]—number concentration (molecules/mL) of a neutral A in the ion molecule reactor 3.4.5 CVV(A)—mixing ratio or concentration of a constituent ion sample air (mL/L) 3.4.6 I(AH+)—signal intensity, that is, ion count rates (ions/s) 3.4.7 E(AH+)—ion transmission efficiencies through the mass spectrometer 3.4.8 k—ion–molecule reaction rate constant (molecules mL–1 s–1) 3.4.9 t—reaction time (s) 3.4.10 τ—dwell time (s) 3.4.11 [air]R—CVN (air) number concentration of air in the ion molecule reactor (molecules/mL) 3.4.12 Ccal—calibration concentration (nL/L) 3.4.13 pR—pressure of the ion molecule reactor (mbar) 3.4.14 TR—temperature of the ion molecule reactor (K) 3.4.15 UR—voltage of the ion molecule reactor (V) 3.4.16 µ—ion mobility (m2 V–1 s–1) 3.4.17 µ0—reduced ion mobility (cm2 V–1 s–1) at standard conditions of p0 and T0 3.4.18 p0—air pressure in standard conditions (mbar) 3.4.19 T0—temperature in standard conditions (K) 3.4.20 NA—Avogadro constant 3.3 Abbreviations: 3.3.1 CI-MS—chemical ionization - mass spectrometer or spectrometry 3.3.2 DOD—United States Department of Defense 3.3.3 DOE—United States Department of Energy 3.3.4 EPA—United States Environmental Protection Agency The last approved version of this historical standard is referenced on www.astm.org Background—Signal is caused by contaminations in the sampling system and the ionizer This is different from the base line signal, which is caused by electronic noise, stray ions, and/or peak tails of very abundant compounds In practice this means that a gas mixture can have 20 components present at 10 nL/L (ppb) each These components shall not produce interfering signals or contribute significantly to the consumption of the reagent ion Commercially sold ZeroAir cylinders and generators usually guarantee the content to have 10,000) which allows separation of many isobars which is very useful to compensate the lack of chromatographic separation Their high mass 6.6 Chemical ionization means that a compound A is ionized via the chemical reaction with a reagent R In most cases the reagent is an ion, which is indicated as Rz where z is the charge state of the reagent In some cases, the reagent can be a neutral, metastable molecule or element, which is indicated as R● 6.7 The reaction can be of many different types Common reaction types are proton transfer ionization, electron transfer ionization, or adduct ionization 6.7.1 PTR-MS uses the reagent ion H3O+ and therefore ionizes organic analytes (A) via the following proton transfer reaction: A1H O →AH 1H O 6.7.1.1 Only compounds that have a proton affinity greater than that of water (693 kJ/mol) can be ionized when using the hydronium mode of ionization 6.7.1.2 This reaction may also take place with water cluster ions (H2O)nH3O+ as reagent ions or adduct ions, which can be helpful in untargeted analytical approaches 6.7.2 Electron transfer ionization (ETI) can also result in positive ionization: A1R →A 1R Mass resolution: ∆M50% = peak width at 50% height, which is approximately the smallest difference between two peaks M1 and M2 so that they can be identified as separate signals 6.7.3 Adduct chemical ionization is sometimes preferred to H3O+ because it is even “softer”: D8460 − 22 6.9.3.2 De-isotoping means accounting for the mass spectral signal of the various isotopes of a given compound during peak integration, and potentially assigning the integrated signal of less-abundant isotope ions to the monoisotopic ion pertinent to that compound accuracy enables identification of compounds without fragment libraries They measure all masses simultaneously and therefore are quite sensitive In addition, they can be quite compact and robust 6.8.2.1 TOF analyzers require a pulsed and cyclic ion extraction into the field free region of the MS All ions are measured in each extraction cycle which repeat at a rate of typically 10 to 50 kHz dependent upon instrument specifics Data from multiple extractions are accumulated into spectra for predefined time periods (typically 0.01 to 10 seconds) to improve signal to noise in the spectra 6.10 Based on reaction kinetics, the number concentration (in molecules/mL) of neutral VOC [A], in the IMR can be determined by the following equation: I ~ AH ! E ~ H O ! @ A # kt I H O E AH ~ ! ~ ! (1) where k is the ion—molecule reaction rate constant (molecules/mL s–1), t is the reaction time (s), I(AH+) and I(H3O+) are the respective ion counts rates (ions/s), and E(AH+) and E(H3O+) are the ion transmission efficiencies through the ion optics and the mass spectrometer The mixing ratio or concentration of the organic A in the sample air is then determined by the following equation: 6.9 The PTR-MS with TOF analyzers are ideal when rapid changes (bolus events, fugitive emissions) in vapor concentrations are anticipated which require high temporal resolution Other mass analyzers such as Fourier transform ion cyclotron resonance mass analyzers are used primarily in academic research settings and are not used in field deployments Data acquisition system (DAQ) usually includes electronics for recording the signals from the mass analyzer and a computing unit 6.9.1 The recording electronics can be either a time-todigital converter (TDC) or an analog-to-digital converter (ADC) Whereas TDCs count the ions individually, ADCs measure the current produced by the ions TDCs are faster, less expensive but have a limited dynamic range With ADCs becoming faster and mimicking TDC properties, the ADCs gradually replace the TDCs Modern ADCs have on-board processing, which means some data analysis can be done on-board 6.9.2 The main processing steps done in the computing unit are listed in Fig 1: 6.9.2.1 Peak selection can be done in two different ways: • Peaks are selected from a pre-defined peak list This is referred to as “targeted analysis.” • Peaks are selected using a peak finder algorithm in addition to the pre-defined peak list or from scratch This is referred to as “non-targeted analysis.” Many standards include a list of compounds to be measured, which amounts to a “targeted analysis.” In many cases, the “peak finding” of peaks that are not in the peak list is done in post-processing and even manually The new peaks can then be added to the predefined list and the complete data analysis can be repeated This blurs the line between targeted and non-targeted analysis 6.9.2.2 Peak integration collects all signal of an ion species into a single intensity for that species This process includes mass calibration, integration of a signal peak sometimes using peak fitting, and massspectral baseline correction These processes can be done either real-time during recording or in postprocessing 6.9.3 Transmission correction means accounting for the fact that the total ion transmission depends on the mass/charge of an ion This step does not need to be done when a compound is quantified using a calibration gas with a known concentration of the compound 6.9.3.1 De-fragmenting means assigning the signal of multiple fragment ions to their precursor ion X~A! @A# @A# @A# 109 nL⁄L 109 ppb @ A I R # IMR @ A I R # IMR @ A I R # IMR (2) where [AIR]IMR is the number concentration of air (molecules/mL) in the IMR; this equation may also be adjusted to take water cluster ion reactions into account 6.11 In practice the sensitivity of the PTR-MS to various VOCs is determined by using multicomponent compressed gas standards to establish the sensitivity s = I/S (signal intensity per concentration); this sensitivity s is measured in (ions/s)/(nL/L) = cps/ppb 6.12 In practice, due to differences in ion-molecule reaction rate constant and transmission efficiency, and different degree of fragmentation, different species have different sensitivities For example, sensitivities are typically larger for polar oxygenated compounds Special Skills 7.1 This method aims at post-analytical quantification aspects Personnel must be competent in the operation of the PTR-MS instrument, calibration and blank procedure 7.2 The user must be educated in the steps to calculate the normalized sensitivity of VOCs using data collected from the PTR-MS and calibration and zero system Ultimately, this requires the knowledge to determine ambient concentrations of VOCs from the calculated sensitivities Personnel should also be able to estimate the concentrations of tentatively identified compounds (TIC’s) using calculated sensitivities and proton transfer reaction rate constant data Safety 8.1 Components of the PTR-MS are at a high voltage and protected from accidental human contact However, care should be taken to avoid contact with energized parts and only qualified PTR-MS technicians should attempt repair or maintenance within potentially energized areas of the instrument 8.2 The multi-component VOC blend is stored inside a pressurized aluminum bottle with an attached regulator Before D8460 − 22 movement of the bottle from the security straps, the regulator should be removed and the bottle head should be covered with the supplied cap Safety Data Sheets (SDS) for chemicals, such as analytes and solvents, should be consulted before use The user of this test method should also be aware of the hazards associated with the operation of the multicomponent VOC blend that contains many toxic compounds Therefore, the exhaust of the calibration and zero system and PTR-MS should be vented outside the analytical workspace to avoid contamination of the air with the compounds of the multi component VOC mixture In case of primary ion sources other than hydronium, such as O2, standard safety procedures are to be consulted for handling gas cylinders with such content 8.3 Turbomolecular vacuum pumps can fail catastrophically if suddenly exposed to high pressure while they are operating, which could present a hazard to humans or property Turbomolecular pumps should be turned off and allowed to come to a complete stop before the instrument is vented Setup, Sample Collection, and Handling 9.1 Fig illustrates the schematic layout of a basic PTR-MS system Due to the connection with ZeroAir, a dilution of the actual sample can be performed in case of large amounts of VOC emissions that can overwhelm the instrument An example could be the investigation of alternative pathways For calibration the 2-way valve is switched to the calibration gas, while for measurements the valve is switched to the sample inlet side The sample inlet side can be either a single line of tubing or could be a multi-valve that switches between multiple sampling lines Due to the relatively low flow rate of the PTR-MS, which is in the range of 100 sccm, it is usually beneficial to use a secondary pump and subsample from that main flow 9.2 More sophisticated setups have been shown to be adequate for specific problem settings, such as GC-PTR-MS.8 FIG Basic Configuration of a Calibration and Sampling System for PTR-MS Analysis 9.3 The PTR-MS does not require any pre-conditioning of the sample While filters can be used to remove larger dust particles, these can also interfere with the vapor content of a sample A virtual-impactor setup is recommended, in which the PTR-MS samples a small flow orthogonally from a much larger flow supplied by an external pump (see Fig 2) Depending on the ambient air conditions, some advantages can also be gained through different sampling techniques such as the use of cold traps, nafion dryers, thermal desorption or sample dilution using either a mass flow controller or flow orifices, however, this is not a requirement for general indoor sampling and analyses 9.5 To provide the optimal sample to the instrument guidelines are provided by several ASTM standards, such as, D5314 and D5730 Minimal calibration requirements are shown in Table 9.6 The sampling line is to be kept at a stable temperature into the instrument, ideally with increasing temperature from the point of sampling to the IMR This avoids the so-called cold spots, which are areas within the sampling line colder than the ambient temperature and which potentially produce false results due to condensation on the walls However, due to the pressure difference between the ambient pressure and in the IMR, the temperature within the chamber can be reduced up to 20°C in comparison to the inlet tube temperature while still preventing condensation of sampling constituents This is beneficial to further reduce the amount of fragmentation for labile compounds during ionization 9.4 The sampling line can be extended to the length required by location Standard tubing diameters in the U.S are 1⁄4 in (6.4 mm) or 3⁄8 in (9.5 mm) OD; PFA or FEP are materials with a very good (that is, low) retention and price Sampling lines of up to 100 feet (30.5 m) can be set up 10 Operating Procedure Such systems have a reduced ability for real-time monitoring but an additional layer of separation which can be beneficial in tracking very low concentrations of target analytes; a side benefit is that this setup would fulfill the criteria to apply U.S EPA method 18 For comparison of such methods see Warneke et al (2015) (4) 10.1 Startup and Operating Steps—The individual steps on how to setup a PTR-MS run are highly dependent on the D8460 − 22 calibrant gas with a larger flow of zero air, such that the signals for the ions pertinent to the compounds in the calibrant mixture dominate any neighboring interferences As delineated in the chapter on mass calibration, two points or more are to be used, in the low (21.0232 Th for H3O+ isotope or NO+ at 29.9987 Th) and upper range (for QMS, alpha-pinene at 137 Th, for TOFMS 203.9940 Th from the fragmentation of 1,3-diiodobenzene if present or an equivalent standard in the range of analysis); a simple validation is to briefly breathe into the inlet and check for the mass of protonated acetone, which is 59.0865 Th individual instrument’s operating software The general steps described below serve to assure quality control For details on how to start the instrument and how to setup the parameters for analysis, such as IMR temperature, pressure, and voltage, sample inlet temperature, characteristics of the detectors and ion optics modules (if present) and of the output files are to be identified using the manufacturer’s guidelines 10.2 Leak Detection—Upon start-up it is necessary to tune the ion source and identify the presence of a leak in the instrument Leaks should not occur during normal use of the instrument In case the vacuum chamber pressure doesn’t reach the appropriate range within regular time frames of initial startup (typically 15-45 minutes for QMS, 1-3 hours for TOFMS), a vacuum leak is the cause for such a delay Should the system fail to pump to the required vacuum, the leak must be found and corrected 11 Interferences 11.1 The PTR-MS identifies compounds as the molecular mass of the chemical species plus the mass of one proton when using hydronium ions for ionization The technique is therefore limited by isobaric interferences for PTR-QMS and isomeric interferences for PTR-TOFMS with higher than mass unit resolution One approach to identify interferences is to use different reagent ions, such as O2+ or NO+ and use the potentially different reaction mechanisms in the IMR as a separator Also, some species fragment upon ionization Another way to separate isomers is to use GC, see 9.2) 10.3 Tune Ion Source—The ion source is tuned to optimize the H3O+ count rate and keep the O2+ count rate less than % of the H3O+ count rate by using dry VOC-free air To tune the ion source the following ions are measured: H3O+, O2+, NO+, H2O+, (H2O)2 H+ The ion source is tuned by adjusting the H2O flow through the ion source, by adjusting the ion source current, and by adjusting the voltages of the secondary IMR lenses At this point the detector voltage can be increased to get H3O+ count rates into the desired range (actual rates of ions/s depend on the individual instrument model and are usually provided by the manufacturer) Equivalently, the ion ratios of O2+ to H3O+, (H2O)2 H+ / H3O+, and NO+ / O2+ are performance indicators, but the actual numbers of these ratios are instrument-dependent and vary between manufacturers 11.2 An important contributor to analyte fragmentation is the reaction with O2+; this ion is produced along with the hydronium ion (H3O+), but the IMR is tuned to increase the concentrations of the hydronium ion and reduce the concentrations of the O2+ ion As the ion source ages, the abundance of interference ions such as O2+ slowly increases (see 10.3 on tuning of the source) O2+ ionizes the VOCs of interest mainly through charge transfer reactions The reaction is a form of hard ionization and typically fragments the VOCs of interest which can lead to either overestimation of some compound concentrations through the interference by fragment ions or the underestimation of some VOC concentrations due to the loss of the primary ion The O2+ concentration should be monitored and recorded at a minimum daily and if found out of control based on the manufacturer’s specifications, the IMR retuned according to the manufacturer’s guidelines 10.4 Tuning of Alternative Ion Sources—If an ion source different to hydronium is chosen, the source ion needs to be optimized Due to the large number of potential source ions, only hydronium is specifically described within this guideline An individual optimization protocol shall be developed within the sampling plan In addition, many of these ion sources ionize the analyte by reactions other than proton-transfer These include the use of NO+ and O2+ as reagent ion 10.5 Mass Calibration (Internal Standard)—Before measurements are to be made, the mass-scale calibration must be verified The mass calibration verifies that the ion peaks are centered over the correct value of the ion mass 11.3 NO+ is also produced in the source, but to a lesser extent than O2+ This ion undergoes soft ionization reaction with several common analytes resulting in detectable interferences The ion can also fragment some VOC species resulting in further interferences 10.6 Mass drift can occur for various reasons, the most important being temperature changes and vibrations during transportation A good practice is to perform a quick mass calibration verification check after every transport Several instruments provide internal “continuous” mass calibration By injection of an inert substance such as 1,3-Di-iodobenzene into the IMR a permanent signal is generated that the instrument can target With such an omnipresent signal, software algorithms can validate the accuracy of the peak center every minute or less; these autocorrection features have limitations 11.4 Water dimers and larger clusters formed through the hydration of the reagent ion can also positively interfere with the quantification of polar species such as ketones, aldehydes and organic acids If a species has a proton affinity greater than the water dimer, then the organic compound will be ionized through proton transfer reaction from the water dimer Polar species can also be ionized through ligand switching reactions with the water dimers Because the basic calculation of the sample compounds is a function of the reagent ion only and not from ionization from any other means, the quantification of the sample compound will be positively biased due to the presence of water dimers The formation of water dimers is controlled through tuning the IMR voltage across the IMR The drift 10.7 In case the calibration is off by more than one mass in the target region, the algorithms usually cannot identify the appropriate peak In this case, a manual calibration with a known standard gas mix is advised by mixing a small flow of D8460 − 22 voltage controls the velocity the ions travel down the IMR The water dimers break apart through random collisions with other molecules in the flight path Increasing the voltage results in a lower abundance of water dimers through forced fragmentation but may also decrease the abundance of ionized sample VOCs through loss of the proton by random collisions The IMR voltage is tuned to minimize the water dimer interference while maintaining the sensitivity to VOCs 12 Quality Control Measures 12.1 Table provides the recommended quality control TABLE Quality Control Protocol for Continuous Monitoring with PTR-MS Activity Ion source tune Initial Multipoint Calibration (ICAL) Frequency Prior to ICAL and prior to each 24-hour period of sample analysis After movement of the instrument to the test site At the beginning of a sampling campaign Initial Calibration Verification (ICV) Once after each ICAL to verify source standard Continuing Calibration Verification (CCV) Daily before sample analysis, if continuous, after every 24 hours of analyses, and at the end of the batch run System (Method) Blank Once after the first CCV, and prior to starting field analysis In addition, after sampling gasses of high concentration or high humidity Mass Calibration (represents the internal standard) Continuous Ionization Softness (based on fragmentation of alphapinene/isoprene) Daily Comments See 10.3 Minimum of concentrations, one of them being at the CCV level and the lowest being at or below the LOD Acceptable if linear least square regression for each analyte is $0.99 If ICAL fails, rerun, if still fails, check dilution apparatus, check if zero air source is functional, verify there is no leak in the system (that is, no diluting with ambient air) Analytes should cover as many targets as possible, however reaction kinetics approach does not require all analytes being present in calibration (see 14.2.1.3) All reported analytes of the laboratory control sample (LCS) within ±30 % of certified value (either certified gas cylinder or pre-made canister) If ICV fails, rerun ICV, if still fails, repeat ICAL Concentration of the mid-point level of ICAL All analytes within ±30 % of the true value If CCV fails, analyze two consecutive samples of at least seconds each If both pass in comparison with last CCV but fail with ICV, check for drastic changes in humidity Some analytes have strong ties to humidity levels, such as formaldehyde If humidity had drastic changes, explain in Case Narrative In any case, since measurements are continuous and cannot be repeated, apply Q-flag to all results for the specific analytes for the duration of failure Data can be reported but must be explained in the Case Narrative The method blank is zero air – either provided through a certified canister/cylinder or through generator system No analytes shall be detected higher than 1⁄2 LOQ or 1⁄10 of the amount measured in any sample or 1⁄10 the regulatory limit, whatever is greater Common interferences must not be detected larger than LOQ If it fails, perform investigation on source and take appropriate corrective actions In some cases, running the instrument overnight, capped, under high vacuum is sufficient to remove contamination from the IMR If contamination is found in sampling system, exchange tubing if feasible; pull zero air at elevated temperatures through sampling system to clear out If MB fails and reanalysis cannot be performed, report data with a “B”-flag to all results Continuous mass calibration verification is performed by monitoring the masses that always exist within the mass spectra, such as the primary ions 19 Th, 21 Th, or 55 Th or by using the instrument specific sources as continuous internal standard, such as di-iodobenzene or chlorinated fluorocarbons For pass, the area response must be within 40 % of the mean area response The ionization softness is reflected by the fragmentation ratio of selected substances such as isoprene or alphapinene The recommended fragmentation ratio of alphapinene should not be greater than 55 % for the ratio of ions/s at 81 Th/(81 Th + 137 Th) In case the fragmentation ratio is too high, the results shall be flagged with “IS.” D8460 − 22 measures for continuous monitoring.9 TABLE Calibration Gas Mixture Recommended for Ambient Air Measurements (each analyte ~200 nL/L = 200 ppb) 13 Calibration and Standardization Molecular ID CH4O C2H3N C3H6O C2H3Cl C4H8O C6H6 C7H8 C2H2Cl2 C8H8 C8H10 C9H12 C2HCl3 C10H14 C2Cl4 13.1 Blanks are used for background subtraction Performing routine analytical blanks is important for quantifying the ion counts in the absence of analyte VOCs A blank can be performed by overflowing the inlet with zero air, which is typically provided from a generator or from a zero air gas cylinder Sometimes zero gas requires additional cleaning from remaining VOCs using a scrubber/filter that contains activated carbon 13.2 Analytical blanks are conducted before and after a sample is taken, after sampling gasses of high concentration (causing saturation), high humidity variations and after the first CCV When conducting continuous sampling, analytical blanks are conducted every eight hours at a minimum More frequent analytical blanks should be conducted when sampling gasses with high concentrations of volatile compounds such as acetone are anticipated or experienced An analytical blank must contain 10 sampling points (10 cycles for QMS or 10 seconds for TOFMS) or, after a saturation event, one continues until background count rates have returned to the original levels Saturation is determined by the lack of primary ions in the spectrum, so for the hydronium one uses peak intensity of m/z = 21.0224 (O18 isotope of H3O+) Name Methanol Acetonitrile Acetone Vinyl Chloride 2-Butanone Benzene Toluene 1,2-Dichloroethylene Styrene p-Xylene 1,3,5-Trimethylbenzene Trichloroethylene 1,2,3,5 Tetramethylbenzene Tetrachloroethylene provide the vapor that is then mixed with zero air for dynamic or static dilutions The LCS is used for the initial and continuing calibration verification (ICV and CCV) 13.5 An instrument calibration curve is typically derived from a zero-calibration-zero sequence; this sequence begins by sampling five sampling points or more of zero air followed by sampling sampling points or more of the calibration gas with decreasing concentrations The starting point of the calibration gas mix shall be around the anticipated maximum concentration of analytes and gradually diluted to 100-fold or more; to achieve the needed concentrations an appropriate standard gas mixture has to be used and gradually diluted with zero air The example contents of one such standard are shown in Table The sequence ends by sampling an adequate volume of zero air to flush the instrument and calibration system 13.3 The timescale of the zero measurement should be timed with the timescale of variability required of the measurement, as zeroing the instrument disrupts the equilibrium between the instrument surfaces and the sample air flow It follows that after long timescales of measuring clean air, the instrument is “cleaner” (has lower background) than shortly after measuring a polluted air sample For example, measurements of a vapor intrusion hot spot, which requires second measurements should not base the background on hour zero measurements as the background will be measured systematically low 13.6 The sensitivity s is calculated using the following equation: s~A! I cal ~ AH ! I zero ~ AH ! I cal ~ AH ! I zero ~ AH ! C cal ~ A ! C cal ~ A ! 109 ppb ~ nL ⁄ L ! 13.4 Calibrations are used to determine the sensitivities of compounds They can be performed by producing a laboratory control sample (LCS) from a metered flow of a NIST traceable multi component gas standard and a metered flow of the zero air mixed through dynamic dilution using an apparatus as sketched in Fig A calibration mixture at various concentrations is produced by altering the flow rates of the calibration mixture and/or zero air; see Table for the composition of a recommended calibration mixture for ambient measurements For the composition of a typical calibration gas mixture to be used in the dilution series, see Table The two gas flows are metered using mass flow controllers This mixture is introduced to the PTR-MS at various concentrations to conduct a multipoint calibration; a minimum of calibration points shall be performed The concentration range should bound the expected concentration of the analytes under evaluation Alternatively, liquid calibration systems (LCS) are available that use liquid standards and nebulizers of various kinds to (3) 13.7 The sensitivity of analyte A is calculated by taking the difference between the instrument’s response Ical(AH+) and the instrument’s background Izero(AH+) and dividing it by the calibration concentration (Ccal) The sensitivity for each compound A should be calculated using the average of multiple, at least 5, known concentrations from diluting a gas standard (for example, the LCS) The range of the selected concentrations during the calibration should be selected in order to span the range of expected analyte concentrations 13.8 The transmission function describes the mass dependent efficiency between the transfer system, the actual mass separation and the detector The transmission is a function of the mass/charge, therefore E = E(m/Q) For the transmission calculation a gas standard mixture is used The compounds of the gas mixture must be selected in order to cover a wide range of masses and not to interfere with each other The QMS system has a higher transmission in the low masses while the TOFMS system has higher transmission in the higher masses In some instances, multiple transmission curves need to be prepared if there are optional ion funnel settings Such settings can immensely increase the sensitivity, however, in some cases This table is based on common laboratory practices, such as laid-out in (and adapted from) Table B-21 of the U.S DOD/DOE QSM 5.3, Appendix B, 2019 10 D8460 − 22 such increases cause the need for severe dilution of common VOC concentrations in indoor air 14.1.2 There are many other possible definitions of concentration, including any permutation of the three quantities used to measure mass m, volume V, abundance (= number) N 14.1.3 The signal S can also be defined or measured several ways It can be a voltage, a current, or a number of events In the case of mass spectrometers, the signal is usually transformed into number of ions, and because PTR-MS is usually recording data continuously in a cyclic method, the signal is actually regarded as a signal intensity I = signal per time (S/t) with units ions/s 14.1.4 Accordingly, the sensitivity can also come in many different forms For the purpose of this standard: 13.9 Method Detection Limits—The method detection limits (MDL) are determined from zero air The detection limit is different per compound A Limiting factors on precision and detection limits of PTR-MS measurements are the counting statistics of analyte ions, which follow a Poisson distribution: the 1-σ error of counting S ions is =s Thus, a compound A will have signal (S) equal to: S I ~ AH ! ·t c I zero ~ AH ! ·t c the noise or error N will be: N =S =I ~ AH ! ·t c I zero ~ AH · t c ! s~A! where tc is the time of each measurement cycle The limit of detection (LOD) of the analyte A will be I(AH+) when S/N = can be found by solving: S⁄N I LOD ~ A ! ·t c I zero ~ AH ! ·t c =I LOD ~ A ! ·t c I zero ~ AH ! ·t c 53 (4) 14.2 Signal Intensity—Even though PTR-MS provides real time qualitative measurements to the user, QC/QA validated data need to be thoroughly analyzed and validated for customer reporting This can be performed within automated software approaches and the real time data can thus be reported quantitatively The signal intensity is evaluated from the mass spectra in two steps: • Sample gas measurement, which includes all processing indicated in Fig • Background subtraction 14.2.1 A sample gas measurement, as shown in Fig 1, includes several processing steps which are usually done automatically 14.2.1.1 Mass Calibration—Although this step is performed during analysis, inspection of the resulting data can identify issues that are based on incorrect mass calibration The following step, peak integration, relies on a good mass calibration 14.2.1.2 Peak Integration—This step is only necessary when using TOFMS The QMS provides nominal mass data The raw data of the PTR-TOF are consecutive mass spectra 14.1 The quantification of PTR-MS data always uses the same basic principle, expressed in Eq The concentration of a substance A is calculated from the signal S of this substance and the sensitivity s of this substance: S~A! s~A! (7) 14.1.4.1 The unit of the sensitivity is either cps/ppb which is commonly used, but not SI compliant The SI compliant equivalent is (ions/s) / (nL/L) or Hz / (nL/L) 14.1.5 In the following steps, the signal intensity I and the sensitivity s are determined in order to find the concentrations C using Eq This is illustrated in Fig 14 Quantification Procedure C~A! I~A! I~A! C NN ~ A ! C VV ~ A ! (5) (see Fig 3) 14.1.1 Eq is a reversal of the definition of sensitivity, which is: sensitivity = signal / concentration: S~A! (6) C~A! NOTE 2—The concentration C can have many different meanings in this case, it can be: of A • A number density CVN (A) = N(A) / V where N(A) is the number of particles of A and V is the total volume • A particle concentration CNN = N(A) / N(air) where N(air) is the total number of particles • A volumetric concentration CVV = V(A) / V where V(A) is the partial volume of A Note that for ideal gases CNN = CVV, this is why the non-SI unit ppb can be replaced by the SI unit m3/m3 s~A! FIG Basic Workflow of a Quantitative Measurement to be Refined in Fig 11 D8460 − 22 FIG Workflow of a Quantitative Gas Analysis Using Zero Air Measurement as a Blank for Background Subtraction and Calibration Gas Measurement to Retrieve the Sensitivities shall be taken In case of large variations, the source for these variations needs to be identified within the time sequence and separate background values identified For instance, a previously sampled material could have been retained in either the sampling lines or even within the instrument due to cold spots and condensation effects In such cases, the replacement of the sampling lines is normally the only way to successfully remove the contamination that contain, depending on the mass resolution, multiple peaks per each nominal mass, representing isobars (= molecules with the same number of nucleons but different chemical composition) The signal of each peak (representing compounds) has to be integrated in order to extract quantitative information Some software programs offer an automatic integration, but the analyst should have a validating approach to such algorithms to ensure appropriate deconvolution of overlapping peaks Various peak fitting options are typically available; for details the instrument guidelines need to be consulted The peak fitting or integration process also includes a mass spectral baseline correction which accounts for signal from electronic noise, stray ions and peak tails 14.2.1.3 Ion Transmission Correction—The transmission function E(m/Q) describes the mass dependent efficiency between the transfer system, the actual mass separation and the detector For the transmission calculation the same gas standard mixture is used as for general calibration of the instrument The mixture contains multiple compounds (>10 preferably) in the mid nL/L (ppb) concentration range (see Table 2) These compounds shall be selected in order to cover a wide range of masses and not to interfere with each other (Table 3) The mixture is then diluted with zero air in ratios ranging from 1/10 to 1/100 and introduced to the PTR-MS Based on the response signal the transmission calculation can be performed I m ~ AH1 ! I ~ AH ! E ~ AH1 ! 14.3 Sensitivity—PTR-MS offers multiple different ways to retrieve the sensitivity: 14.3.1 Sensitivity by Direct Calibration—the concentration of the compound of interest is evaluated using a calibration gas analysis of the compound with a known concentration, and a zero-gas analysis, according to Eq This is the official method of this standard 14.3.1.1 Compound specific sensitivities are measured using a gas standard with the compounds of interest as mentioned in 11.2 In order to achieve reasonable precision on the sensitivities, a calibration curve with to points of dilution of the gas standard is recommended 14.3.1.2 The sensitivity is the slope of the calibration function: I ~ A ! s ~ A ! ·C NN ~ A ! (9) This method is the most accurate, but it is limited to the compounds contained within the gas standard used for calibration For this method there is no need for the calculation of the transmission efficiency of the instrument or the definition of the reaction rate constant The same approach can be used when a reagent ion different from H3O+ is used, such as O2+ or NO+ 14.3.2 Sensitivity by Interpolation to a Fitted Curve—If a compound needs to be quantified that is not in the calibration gas cylinder, but its reaction rate constant k is known, its sensitivity s can be estimated based on an interpolation between the measured sensitivities of the calibrant compounds This is illustrated in Fig This interpolation is not very precise However, the reaction rate values fall within a narrow range (within a factor 2), and so sensitivities also generally fall within this range (8) 14.2.1.4 Some software allows for a multiplier to account for potential fragmentation pathways, isotopes or by-products 14.2.2 Background Subtraction—In this step, the zero gas concentrations (blank) of the instrument is subtracted from the sample gas concentrations The background ion intensity for any VOC being measured by PTRMS will usually be more than zero due to electronic noise, sampling equipment contamination and instrument contamination The background signal differs per compound, instrument, and operational conditions The amount is empirically determined by evaluating the analytical blanks; if the concentrations of the targeted species show small variations within the blanks, the average 12 D8460 − 22 FIG Sensitivity Curve for Different Compounds (protonated) analyte Based on reaction kinetics the concentration of the analyte A will be equal to: 14.3.2.1 The sensitivities of the calibration compounds are plotted in a sensitivity versus reaction rate graph This requires that the reaction rate constants of the calibrants are known The sensitivity of compounds not present in the calibration gas can then be interpolated based on their reaction constant k 14.3.3 Kinetics Approach—This rather unique ability, in comparison to other analytical techniques, is the calculation based on reaction kinetics No gas calibration is required The sensitivity is calculated using instrument parameters and reaction rates k However, this method requires precise knowledge about both the transmission function E(m/Q) and IMR operating conditions, as well as accurate measurement of the reagent ion signals 14.3.3.1 This quantification method relies on well-defined conditions in the IMR and the previously determined kinetics of proton transfer reactions The great strength of this approach is the ability to calculate mass concentration for any compound for which the reaction rate constant is known or can be estimated by established laboratory techniques The mass concentration is calculated based on the physical conditions in and of the IMR (pressure, length, temperature, voltage, ion mobility), which are continuously recorded during the instrument operation 14.3.3.2 Based on the principle of operation, reagent ions (H3O+, O2+, NO+) are produced in the ion source and introduced in the IMR where they undergo collisions with the sample gas molecules When a reagent ion collides with an analyte A, the proton is transferred resulting in the ionized C VN ~ A ! and C VN ~ AH ! I ~ AH ! I ~ AH ! · N · 5 k·t C V ~ H O ! k·t I ~ H O ! s~A! (10) I m~ H O 1! E ~ H O 1! (11) s ~ A ! k ~ A ! ·t ~ H O ! ·I ~ H O ! k ~ A ! ·t· where k(A) is the reaction rate of A with H3O+ (molecule cm–3 s-1), t is the reaction time (s), I(AH+) and I(H3O+) are the respective signal intensities (ions/s), and E(AH+) and E(H3O+) are the ion transmission efficiencies 14.4 Concentration—The concentration of a sample is then calculated by dividing the background corrected ion signal of the sample by the sensitivity: C VN ~ A ! I ~ AH ! I zero ~ AH ! I ~ AH ! I zero ~ AH ! 109 ppb s~A! s~A! (12) 14.5 Concentration Conversion—If the number concentration is required, it must be calculated using the following equation: C VN ~ A ! C NN ~ A ! ·C VN ~ air! (13) N 14.5.1 The number concentration [air]R = CV (air) of air (molecules per cubic centimeter) in the IMR is equal to: N T p A R @ air# R C VN ~ air! V · T · p m R 13 (14) D8460 − 22 16 Maintenance where: NA = 6.022 × 10 23 molecules/mol = Avogadro constant, = absolute standard temperature = 273.15 K, T0 = absolute temperature in the reactor (in K), TR = 1013 mbar = 1013 hPa = standard pressure, p0 = pressure in reactor in hPa or mbar, pR Vm = molar volume = 22400 mL ⁄mol = 22.4 L/mol = volume per mol of gas at standard conditions CVN (air) = number concentration of air in the IMR (in molecules/mL = molecules ⁄cm3) 16.1 Basic Maintenance—If hydronium ions are the primary ion source, the hydronium source is fed from a water reservoir that generates water vapor Depending on the size of the water reservoir it needs to be refilled every one to six months The water level can be either visually assessed, or by the inability of the water vapor flow controller to reach its set value This can also happen when the operating temperature is too low, that is, close to water freezing temperature For refilling the water reservoir follow the steps in the vendor’s maintenance manual In addition, the instrument needs regular cleaning with a vacuum cleaner (once per two months or more frequently depending on the deployment areas) to prevent dust accumulation at the cooling fans 14.5.2 Therefore, the number concentration of compound A becomes: C VN ~ A ! C NN ~ A ! NA T0 pR · · Vm TR p0 (15) 16.2 Do not use antifreeze in the water source The volatile organic sources in antifreeze will highly interfere with the primary ion production 15 Report: Test Data Sheet(s)/Form(s) 15.1 Due to the nature of continuous monitoring the reporting requirements can vary highly Independent of additional regulatory requirements, record the following general information (data): 15.1.1 Project Identification/Location 15.1.2 Sample locations, sample types, sample identifications, sample definition.10 15.1.3 Test Numbers, Testing Dates, Initials or names of person(s) performing the test 15.1.4 Sample preparation methods used (filters, length and condition of sampling lines, GC in front, concentration on adsorbent material) 15.1.5 Equipment Identification (quadrupole or time-offlight with mass resolving power) 15.1.6 Calibration Gas(es) used, along with batch or serial numbers 15.1.7 Humidity, temperature, barometric pressure, wind direction and wind speed (in particular for outdoor measurements) or other conditions that could affect the test results such as HVAC on/off, workday/weekend 15.1.8 Results of QA/QC 15.1.9 Results of the test (the data itself) This can be done in the form of averaged concentrations over the defined samples, or as required by the customer The presentation of these results can be in the form of the individual datapoints over time or time averaged data depending on the definition of a sample (see footnote) 15.1.10 Any sketches, maps or other graphic information that would be useful for presentation or evaluation of the data 16.3 The calibration gas cylinder needs to be checked regularly to determine whether the working pressure is still sufficient 16.4 Advanced Maintenance—For continuing optimal performance several parts require periodic servicing: the ion source, ion gauges, sampling lines, diaphragm pump, and turbomolecular pumps; service kits for the respective devices should be kept in stock Such maintenance should be performed every 12-24 month timeframe, depending on the usage of the instrument 17 Precision and Bias 17.1 Bias—There is no accepted reference value for this test method, therefore, bias cannot be determined to date In an effort to close this gap, the method is currently compared to discontinuous analyses techniques using the same standardized material The results of these inter-method comparison of PTR-MS with ASTM D5466 – 2, Standard Test Method for Determination of Volatile Organic Compounds in Atmospheres (Canister Sampling, Mass Spectrometry Analysis Methodology) will be used for a modification of this section upon completion 17.2 Precision—Test data on precision are not presented due to the nature of this test method It is either not feasible or too costly at this time to have ten or more agencies participate in an in-situ testing program at a given site Subcommittee D18.21 is seeking any data from the users of this test method that might be used to make a limited statement on precision 17.2.1 Peer reviewed published comparisons between PTRMS, GC-MS, GC-FID and other techniques are available, for example as in Yuan et al (2017) (1) and Dunne et al (2018) (2) and references therein; the latter reference also provides in-depth explanations on interference corrections for quadrupole-based measurements 17.2.2 Currently an interlaboratory study is initiated, ILS#1663, and results will be available in a published report upon completion 10 As for all continuous monitoring techniques, the definition of a “sample” in the field is not clearly defined and driven by the circumstances of each project The amount of data from multiple hours of continuous monitoring with a PTR-MS can be overwhelming information Clearly identified data quality objectives for what defines a sample by the customer need to be obtained prior to starting a sampling campaign Examples for definitions of samples include: (1) each room in a building, (2) each potential alternative pathway, (3) each building on a multiplex unit, (4) exceedance of specific internal temperature/pressure, (5) opening/closing of windows, doors etc., (6) air conditioning system on/off, (7) workday operation versus weekend, etc Guidance on criteria to be considered can be found in D8408/D8408M, Standard Guide for Development of Long-Term Monitoring Plans for Vapor Mitigation Systems 14 D8460 − 22 18 Keywords 18.1 air toxics; ambient atmospheric analysis; hazardous vapors; mass spectrometry; proton transfer reaction; real-time environmental monitoring; site characterization; vadose zone gases; vapor intrusion; VOC REFERENCES and J Conca, “Proton Transfer Reaction Mass Spectrometry as a Real-Time Method for Continuous Soil Organic Vapor Detection,” in Continuous Soil Gas Measurements: Worst Case Risk Parameters, ed L Everett and M Kram (West Conshohocken, PA: ASTM International, 2013), 32-44 https://doi.org/10.1520/ STP157020130026 (4) Warneke et al (2015) PTR-QMS versus PTR-TOF comparison in a region with oil and natural gas extraction industry in the Uintah Basin in 2013 Atmos Meas Tech., 8, 411-420 (1) Yuan et al (2017) Proton-Transfer-Reaction Mass-Spectrometry: Applications in Atmospheric Sciences Chem Rev 2017, 117, 1318713229 (2) Dunne et al (2018) Comparison of VOC measurements made by PTR-MS, adsorbent tubes-GC-FIDMS and DNPH derivatizationHPLC during the Sydney Particle Study, 2012 - a contribution to the assessment of uncertainty in routine atmospheric VOC measurements Atmos Meas Tech., 2018, 11, 141-159 (3) J Sears, T Rogers, J McCoskey, L Lockrem, H Watts, L Pingel, ASTM International takes no position respecting the validity of any patent rights asserted in connection with any item mentioned in this standard Users of this standard are expressly advised that determination of the validity of any such patent rights, and the risk of infringement of such rights, are entirely their own responsibility This standard is subject to revision at any time by the responsible technical committee and must be reviewed every five years and if not revised, either reapproved or withdrawn Your comments are invited either for revision of this standard or for additional standards and should be addressed to ASTM International Headquarters Your comments will receive careful consideration at a meeting of the responsible technical committee, which you may attend If you feel that your comments have not received a fair hearing you should make your views known to the ASTM Committee on Standards, at the address shown below This standard is copyrighted by ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959, United States Individual reprints (single or multiple copies) of this standard may be obtained by contacting ASTM at the above address or at 610-832-9585 (phone), 610-832-9555 (fax), or service@astm.org (e-mail); or through the ASTM website (www.astm.org) Permission rights to photocopy the standard may also be secured from the Copyright Clearance Center, 222 Rosewood Drive, Danvers, MA 01923, Tel: (978) 646-2600; http://www.copyright.com/ 15 ... compound = I/C Mass resolving power R = M ⁄∆M50%: for an isolated peak, observed mass divided by the peak width at 50 % height (FWHM, or full-width-at- half-maximum) B D8 460 − 22 obscured from... around the anticipated maximum concentration of analytes and gradually diluted to 100-fold or more; to achieve the needed concentrations an appropriate standard gas mixture has to be used and gradually... as laid-out in (and adapted from) Table B-21 of the U.S DOD/DOE QSM 5.3, Appendix B, 2019 10 D8 460 − 22 such increases cause the need for severe dilution of common VOC concentrations in indoor