EPA/625/R-96/01 Ob
Compendium of Methods
for the Determination of
Toxic Organic Compounds
in Ambient Air
Second Edition
Compendium Method TO-11A
Determination of Formaldehyde in Ambient Air
Using Adsorbent Cartridge Followed by High
Performance Liquid Chromatography (HPLC)
[Active Sampling Methodology]
Center for Environmental Research Information
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, OH 45268
January 1999
-------
Method TO-11A
Acknowledgements
ThisMcthodvvasprcparcdforpublicationin facCompendiumofMethodsfbrtheDeterminationofroxicOrganicCompounds
in Ambient Air, SecondEdition (EPA/625/R-96/010b), which was prepared under ContractNo. 68-C3-0315, WANo. 3-10,
by Midwest Research Institute (MRI), as a subcontractor to Eastern Research Group, Inc. (ERG), and under the
sponsorship of the U.S. Environmental Protection Agency (EPA). Justice A. Manning, John O. Burckle, and Scott Hedges,
CenterforEnvironmentalResearchInformation(CERI),andFrankF.McElroy,National ExposureResearchLaboratory
(NERL), all in the EPA Office of Research and Development, were responsible for overseeing the preparation of this
method. Additional support was provided by other members of the Compendia Workgroup, which include:
• John O. Burckle, U.S. EPA, ORD, Cincinnati, OH
James L. Cheney, Corps of Engineers, Omaha, NB
Michael Davis, U.S. EPA, Region 7, KC, KS
• Joseph B. Elkins Jr., U.S. EPA, OAQPS, RTP, NC
Robert G. Lewis, U.S. EPA, NERL, RTP, NC
Justice A. Manning, U.S. EPA, ORD, Cincinnati, OH
• William A. McClenny, U.S. EPA, NERL, RTP, NC
Frank F. McElroy, U.S. EPA, NERL, RTP, NC
• Heidi Schultz, ERG, Lexington, MA
William T. "Jerry" Winberry, Jr., EnviroTech Solutions, Cary, NC
Method TO-11 was originally published in March of 1989 as one of a series of peer reviewed methods in the second
supplement to "Compendium ofMethodsfor the Determination ofTjxic Organic Compounds in Ambient Air, "EPA 600/4-89-
018. In an effort to keep these methods consistent with currenttechnology, Method TO-11 has been revised and updated
as Method TO-11A in this Compendium to incorporate new or improved sampling and analytical technologies.
This Method is the result of the efforts of many individuals. Gratitude goes to each person involved in the preparation
and review of this methodology.
Author(s)
William T. "Jerry" Winberry, Jr., EnviroTech Solutions, Cary, NC
• Silvestre Tejada, U.S. EPA, NERL, RTP, NC
Bill Lonneman, U.S. EPA, NERL, RTP, NC
• Ted Kleindienst, ManTech, RTP, NC
Peer Reviewers
Robert G. Lewis, U.S. EPA, NERL, RTP, NC
Sucha S. Parmar, Atmospheric Analysis and Consulting, Ventura, CA
Joette Steger, Eastern Research Group, Morrisville, NC
• Lauren Drees, U.S. EPA, NRMRL, Cincinnati, OH
Finally, recognition is given to Frances Beyer, Lynn Kaufman, Debbie Bond, Cathy Whitaker, and Kathy Johnson of Midwest
Research Institute's Administrative Services staff whose dedication and persistence during the development of this
manuscript has enabled it's publication.
n
-------
DISCLAIMER
This Compendium has been subjected to the Agency's peer and administrative review, and it has been
approved for publication as an EPA document. Mention of trade names or commercial products does
not constitute endorsement or recommendation for use.
111
-------
Method TO-11A
Determination of Formaldehyde in Ambient Air Using Adsorbent Cartridge Followed by High Performance Liquid
Chromatography (HPLC) [Active Sampling Methodology]
TABLE OF CONTENTS
Page
1. Scope 11A-1
2. Applicable Documents 11A-3
2.1 ASTM Standards 11A-3
2.2 Other Documents 11A-3
2.3 Other Documents 11A-3
3. Summary of Method 11A-3
4. Significance 11A-4
5. Definitions 11A-6
5.1 C18 11A-6
5.2 HPLC 11A-6
5.3 Method Detection Limit (MDL) 11A-6
5.4 Photochemical Reaction 11A-6
5.5 Photochemical Smog 11A-6
5.6 ppbv 11A-6
5.7 ppmv 11A-6
5.8 Silica Gel 11A-6
5.9 Denuder 11A-7
5.10 Certification Blank 11A-7
5.11 Cartridge Blank 11A-7
5.12 Scrubber 11A-7
6. Extended Methodology and Common Interferences 11A-7
7. Apparatus 11A-8
7.1 Isocratic HPLC 11A-8
7.2 Cartridge sampler 11A-8
7.3 Sampling system 11A-9
7.4 Stopwatch 11A-10
7.5 Polypropylene shipping container
with polyethylene-air bubble padding 11 A-10
7.6 Thermometer 11A-10
7.7 Barometer (optional) 11A-10
7.8 Volumetric flasks 11A-11
7.9 Pipets 11A-11
7.10 Erlenmeyer flask, 1 L 11 A-11
7.11 Graduated cylinder, 1L 11 A-11
7.12 Syringe, 100-250 mL 11A-11
7.13 Sample vials 11A-11
IV
-------
TABLE OF CONTENTS (continued)
Page
7.14 Melting point apparatus 11 A-11
7.15 Rotameters 11A-11
7.16 Calibrated syringes 11 A-11
7.17 Soap bubble meter or wet test meter 11 A-11
7.18 Mass flow meters and mass flow controllers 11 A-11
7.19 Positive displacement 11 A-11
7.20 Cartridge drying manifold 11A-11
7.21 Liquid syringes 11A-11
7.22 Syringe rack 11A-11
7.23 Luer® fittings/plugs 11A-11
7.24 Hot plates, beakers, flasks, measuring and
disposable pipets, volumetric flasks, etc 11 A-11
7.25 Culture tubes (20 mm x 125 mm) with polypropylene screw caps 11 A-11
7.26 Polyethylene gloves 11A-11
7.27 Dry test meter 11A-12
7.28 User-prepared copper tubing for ozone scrubber 11 A-12
7.29 Cord heater and Variac 11A-12
7.30 Fittings 11 A-12
8. Reagents and Materials 11 A-12
8.1 2,4-Dinitrophenylhydrazine (DNPH) 11 A-12
8.2 DNPH coated cartridges 11 A-12
8.3 High purity acetonitrile 11 A-12
8.4 Deionized-distilled water 11 A-12
8.5 Perchloric acid 11A-12
8.6 Ortho-phosphoric acid 11 A-12
8.7 Formaldehyde 11A-12
8.8 Aldehydes and ketones, analytical grade, best source 11A-12
8.9 Carbonyl hydrazone 11 A-12
8.10 Ethanol or methanol 11A-13
8.11 Nitrogen 11A-13
8.12 Charcoal 11A-13
8.13 Helium 11A-13
8.14 Potassium Iodide 11A-13
9. Preparation of Reagents and Cartridges 11A-13
9.1 Purity of the Acetonitrile 11 A-13
9.2 Purification of 2,4-Dinitrophenylhydrazine (DNPH) 11 A-14
9.3 Preparation of DNPH-Formaldehyde Derivative 11A-15
9.4 Preparation of DNPH-Formaldehyde Standards 11 A-15
9.5 Preparation of DNPH-Coated Cartridges 11 A-15
9.6 Equivalent Formaldehyde Cartridge Concentration 11A-18
10. Sampling Procedure 11A-18
v
-------
TABLE OF CONTENTS (continued)
Page
11. Sample Analysis 11A-21
11.1 Sample Preparation 11 A-21
11.2 Sample Extraction 11A-21
11.3 HPLC Analysis 11A-22
11.4 HPLC Calibration 11A-23
12. Calculations 11A-24
13. Performance Criteria and Quality Assurance 11A-27
13.1 Standard Operating Procedures (SOPs) 11A-27
13.2 HPLC System Performance 11A-27
13.3 Process Blanks 11A-28
13.4 Method Precision and Accuracy 11A-28
13.5 Method Detection Limits 11A-28
13.6 General QA/QC Requirements 11A-29
14. Detection of Other Aldehydes and Ketones 11A-29
14.1 Introduction 11A-30
14.2 Sampling Procedures 11A-30
14.3 HPLC Analysis 11A-30
15. Precision and Bias 11A-31
16. References 11A-32
VI
-------
METHOD TO-11A
Determination of Formaldehyde in Ambient Air Using Adsorbent
Cartridge Followed by High Performance Liquid
Chromatography (HPLC)
[Active Sampling Methodology]
1. Scope
1.1 Thisdocumentdescribesamethodforthedeterminationofformaldehydeandothercarbonyl compounds (aldehydes
and ketones) in ambient air utilizing a coated-solid adsorbent followed by high performance liquid chromatographic
detection. Formaldehyde has been found to be a major promoter in the formation of photochemical ozone. In particular,
short term exposure to formaldehyde and other specific aldehydes (acetaldehyde, acrolein, crotonaldehyde) is known
to cause irritation of the eyes, skin, and mucous membranes of the upper respiratory tract.
1.2 Over the last several years, numerous methods have been developed for the sampling and analysis of carbonyl
compounds. Because of the role which formaldehyde plays in photochemistry, most of the more recent methods were
designed to quantitate formaldehyde specifically. Early methods centered around wet chemical technology involving
a bubbler or impinger containing a reactive reagent (1). In some cases the reactive reagent produced a color in the
presenceofformaldehyde. Examplesofthemore commonly used reagentswere:3-methyl-2-benzothiazolonehydrazone
(MBTH), sodium sulfite, 4-hexylresorcinol, water, sodium tetrachloromercurate, and chromatropic acid. These reagents
demonstrated high collection efficiency (>95%), provided fairly stable non-volatile products and minimized formation
of undesirable by-products. Indeed, as part of U. S. Environmental Protection Agency's (EPA's) effort to quantitate
atmospheric concentrations of formaldehyde, the National Air Sampling Network utilized the impinger technique for
severalyears containing chromatrophic acid specifically for formaldehyde. However, impinger sampling had numerous
weaknesses which eventually lead to its demise. They were:
• Labor intense.
• Used acidic/hazardous reagents.
• Lacked sensitivity.
• Prone to interferences.
• Poor reproducibility at ambient concentration levels.
As EPA's interestfocusedupon formaldehyde andit's sources, the development ofpassive personal sampling devices
(PSDs) developed (2). These devices were mainly used by industrial hygienists to assess the efforts of respiratory
exposure for formaldehyde on workers. However, because of the design and flow rate limitation, they require long
exposures (up to 7 days) to the atmosphere to meet traditional bubbler technique sensitivities. Consequently, the
passive PSD had limited application to ambient monitoring.
To address the need for amonitoring method to sample carbonyl compounds in the air at sensitivities needed to reach
health-base detection limits (10~6 risk level), a combination of wet chemistry and solid adsorbent methodology was
developed (3-6). Activating or wetting the surface of an adsorbent with a chemical specific for reacting with carbonyl
compounds allowed greater volumes of air to be sampled, thus enabling better sensitivity in the methodology. V arious
chemicals and adsorbents combinations have been utilized with various levels of success. The most commonly used
technique is based on reacting airborne carbonyls with 2,4-dinitrophenylhydrazine (2,4-DNPH) coated on an adsorbent
cartridge followed by separation and analysis of the hydrazone derivative by high performance liquid chromatography
(HPLC) with ultraviolet (UV) detection.
1.3 Historically, Compen&\}mMslihodTO-5"MethodFortheDeterminationofAlckhydesandKetonesinAmbientAirUsing
High Performance LiquidChromatography (HPLC)" wasuscdloquanlilalc formaldehyde in ambient air. Thismethod
involved drawing ambient air through a midget impinger sampling train containing 10 mL of 2N HC1/0.05% 2,4-DNPH
reagent. Formaldehyde (and other aldehydes and ketones) readily formed a stable derivative with the DNPH reagent,
and the DNPH derivative is analyzed for aldehydes and ketones utilizing HPLC. Compendium Method TO-11 modifies the
January 1999
Compendium of Methods for Toxic Organic Air Pollutants
Page 11A-1
-------
Method TO-11A
Formaldehyde
sampling procedures outlined in Method TO-5 by introducing a coated adsorbent. Compendium Method TO-11 is based on
the specific reaction of organic carbonyl compounds (aldehydes and ketones) with DNPH-coated silica gel cartridges
in the presence of a strong acid, as a catalyst, to form a stable color hydrazone derivative according to the following
reaction:
CARBONYL GROUP 2,4-DINITROPHENYLHYDRAZINE STABLE COLOR WATER
(ALDEHYDES AND KETONES) (DNHP) HYDRAZONE DERIVATIVE
where R and R1 are organic alkyl or aromatic group (ketones) or either substituent is a hydrogen (aldehydes). The
reaction proceeds by nucleophilic addition to the carbonyl followed by 1,2-elimination of water to form the 2,4-
diphenylhydrazone derivative. The determination of formaldehyde from the DNPH-formaldehyde derivative is similar
to Method TO-5 in incorporating HPLC as the analytical methodology.
1.4 Due to recent requirements in atmospheric carbonyl monitoring, EPA has determined a need to update the present
methodology found in Compendium Method TO-11. The revised Compendium Method TO-11A, as published here, incl
• Guidance on collocated sampling.
• Addition of ozone denuder or scrubber to reduce interferences.
• Sampler design update to allow heated-inlet and sequential sampling.
• Update HPLC procedure for column alternatives.
• Use of commercially prepared low pressure drop DNPH-coated cartridges.
Thetargetcompoundforthismethodisformaldehyde;however,atleastl4othercarbonylcompoundscanbe detected
and quantified.
1.5 The sampling method gives a time-weighted average (TWA) sample. It can be used for long-term
(1-24 hr) sampling of ambient air where the concentration of formaldehyde is generally in the low ppb (v/v) or for
short-term (5-60 min) sampling of source-impacted atmospheres where the concentration of formaldehyde could reach
the ppm (v/v) levels.
1.6 The method instructs the user to purchase commercially pre-coated DNPH cartridges. The method still includes the
instructions ofCompendium Method TO-11 forthepreparationofDNPH-coated cartridges. However due to the tedious
preparation and clean room requirements, the method recommends the purchase of pre-coated DNPH cartridges that are
now commercially available from at least three major suppliers. Different from previous cartridges identified in
Compendium Method TO-11, the pressure drop across the newer low-pressure drop cartridges are less than 3 7 inches of
water at a sampling flow of up to 2.0 liters/minute, allowing compatibility with pumps used in personal sampling
equipment. These pre-coated commercial cartridges have generally lower and more consistent background (7)
concentration of carbonyls than cartridges prepared under normal chemical laboratory environment, as specified in
the original Compendium Method TO-11.
Page 11A-2
Compendium of Methods for Toxic Organic Air Pollutants
January 1999
-------
Formaldehyde
Method TO-11A
1.7 The commercially-prepared pre-coated cartridges are used as received and are discarded after use. The collected
and uncollected cartridges are stored in culture tubes with polypropylene caps and placed in cold storage when not in
use.
1.8 This methodmay involve hazardous materials, operations, and equipments. This method does not purport to address
all the safety problems associated with its use. It is the responsibility of whoever uses this method to consult and
establish appropriate safety and health practices and determine the applicability of regulatory limitations prior
to use.
2. Applicable Documents
2.1 ASTM Standards
• D1193 Specification for Reagent Water
• D1356 Terminology Relating to Atmospheric Sampling and Analysis
• D3195 Practice for Rotameter Calibration
• D3631 Method for Measuring Surface Atmospheric Pressure
• D5197 Determination of Formaldehyde and Other Carbonyl Compounds in Air (Active Sampler Methodology)
• E177 Practice for Use of the Terms Precision and Bias in ASTM Test Methods
• E682 Practice for Liquid Chromatography Terms and Relationships
2.2 Other Documents
• Technical Assistance Documentfor Sampling and Analysis Toxic Organic Compounds in Ambient Air, U. S.
Environmental Protection Agency, EPA-600/4-83-027, June 1983.
• QualityAssuranceHandbookforAirPollutionMeasurement Systems, U. S. Environmental Protection Agency, EPA-
600/R-94-D38b, May 1994.
• CompendiumoJMethodsfortheDeterminationofToxicOrganicCompoundsinAmbientAir:MethodTO-ll, Second
Supplement, U. S. Environmental Protection Agency, EPA-600/4-89-018, March 1989.
2.3 Other Documents
• Existing Procedures (8-10).
• Ambient Air Studies (11-15).
3. Summary of Method
3.1 Aknownvolumeofambientairis drawn throughaprepackedcartridge coatedwith acidified DNPH at a sampling rate
of100-2000mL/minforanappropriateperiodoftime.Samplingrateandtimearedependentuponcarbonyl concentration
in the test atmosphere.
3.2 After sampling, the sample cartridges and field blanks are individually capped and placed in shipping tubes with
polypropylene caps. Sample identifying tags and labels are then attached to the capped tubes. The capped tubes are
then placed in a polypropylene shipping container cooled to subambient temperature (~40 C), and returned to the
laboratory for analysis. Alternatively, the sample vials can be placed in a thermally-insulated styrofoam box with
appropriate padding for shipment to the laboratory. The cartridges may either be placed in cold storage until analysis
or immediately washed by gravity feed elution with 5 mL of acetonitrile from a plastic syringe reservoir to a graduated
test tube or a 5 mL volumetric flask.
January 1999
Compendium of Methods for Toxic Organic Air Pollutants
Page 11A-3
-------
Method TO-11A
Formaldehyde
3.3 The eluate is then diluted to a known volume and refrigerated until analysis.
3.4 For determining formaldehyde, the DNPH-formaldehyde derivative can be determined using isocratic reverse phase
HPLC with an ultraviolet (UV) absorption detector operated at 360 nm. To determine formaldehyde and 14 other
carbonyls, the HPLC system is operated in the linear gradient program mode.
3.5 For quantitative evaluation of formaldehyde and other carbonyl compounds, a cartridge blank is likewise desorbed
and analyzed.
3.6 Formaldehyde and other carbonyl compounds in the sample are identified and quantified by comparison of their
retention times and peak heights or peak areas with those of standard solutions. Typically, C,-C- carbonyl compounds,
including benzaldehyde, are measured effectively to less than 0.5 ppbv.
4. Significance
4.1 Formaldehyde is a major compound in the formation of photochemical ozone (16). Short term exposure to
formaldehyde and other specific aldehydes (acetaldehyde, acrolein, crotonaldehyde) is known to cause irritation of
the eyes, skin, and mucous membranes of the upper respiratory tract (19). Animal studies indicate that high
concentrations can injure the lungs and other organs of the body (19). In polluted atmospheres, formaldehyde may
contribute to eye irritation and unpleasant odors that are common annoyances.
4.2 Overthe last severalyears,carbonyl compounds includinglowmolecularweight aldehydes andketoneshavereceived
increased attention in the regulatory community. This is due in part to their effects on humans and animals as primary
irritation of the mucous membranes of the eyes, the upper respiratory tract, and the skin. Animal studies indicate
that high concentrations of carbonyl compounds, especially formaldehyde, can injure the lungs, may contribute to eye
irritation and effect other organs of the body. Aldehydes, either directly of indirectly, may also cause injury to
plants. Sources of carbonyl compounds into the atmosphere range from natural occurrences to secondary formation
through atmospheric photochemical reactions. Consequently, carbonyl compounds are both primary (directly emitted)
and secondary (formed in the atmosphere) air pollutants (19).
4.2.1 Natural Occurrence. Natural sources of carbonyls do not appear to be important contributors to air
pollution. Acetaldehyde is found in apples and as a by-product of alcoholic fermentation process. Other lower
molecular weight aliphatic aldehydes are not found in significant quantities in natural products. Olefinic and
aromatic aldehydes are present in some of the essential oils in fruits and plants. These include citronella, in rose
oil; citral, in oil of lemongrass; benzaldehyde, in oil of bitter almonds; and cinnamaldehyde, in oil of cinnamon.
4.2.2 Production Sources. Aldehydes are commercially manufactured by various processes, depending on the
particular aldehyde. In general, they are prepared via oxidation reactions of hydrocarbons, hydroformulation of
alkenes, dehydrogenation of alcohols, and addition reactions between aldehydes and other compounds. Formaldehyde
is manufactured from the oxidation of methanol as illustrated in the following equation:
[cat.]
CH3 OH - CHj O + Hj
Formaldehyde and other aldehyde production in the United States has shown a substantial growth over the last several
years. This is due, in part, to their use in a wide variety of industries, such as the chemical, rubber, tanning, paper,
perfume, and food industries. The major use is as an intermediate in the synthesis of organic compounds, including,
alcohols, carboxylic acids, dyes, and medicinals.
Page 11A-4
Compendium of Methods for Toxic Organic Air Pollutants
January 1999
-------
Formaldehyde
Method TO-11A
4.2.3 Mobile Combustion Sources. Amaj or source ofcarbonyl compounds in the atmosphere may be attributed tomotor
vehicle emissions. In particular, formaldehyde is the major carbonyl in automobile exhaust, accounting for 50-
7 0 percent of the total carbonyl burden to the atmosphere (19). Furthermore, motor vehicles emit reactive hydrocarbons
that undergo photochemical oxidation to produce formaldehyde and other carbonyls in the atmosphere.
4.3 Secondary Pollutant. As a secondary pollutant (formed in the atmosphere), carbonyls are formed by very complex
photo-oxidation mechanism involving volatile organic compounds (VOCs) with nitrogen oxide (20,21). Both
anthropogenic and biogenic (e.g., isoprene) hydrocarbons leads to in situ formation of carbonyls, especially
formaldehydecompounds.Aldehydesarebothprimary pollutants andsecondaryproductsofatmosphericphotochemistry.
The complete photo-oxidation mechanism is indeed complex and not well understood. However, a brief discussion is
warranted (22). When V OCs and oxides of nitrogen (NO.) arc in the atmosphere and are irradiated with sunlight, their
equilibrium in the photostationary state is changed. The photostationary state is defined by the equilibrium between
nitrogen dioxide (N02), nitrous oxide (NO) and ozone (O \ This equilibrium is theoretically maintained until VOCs
are introduced. V arious reactions occur to produce OH radicals. The V OCs react with the OH radicals and produce R02
radicals that oxidizes NO to N02, destroying the photostationary state. Carbonyls react with OH to produce R02
radicals. Likewise carbonyls, particularly formaldehyde in sunlight, are sources of the OH radicals.
The results of these processes lead to the following:
• Accumulation of ozone.
• Oxidation of hydrocarbons (HCs) to aldehydes and ketones which lead to the continued production of H02- and OH-
radicals, the real driving force in photochemistry smog.
Consequently, the determination of formaldehyde and other carbonyl compounds in the atmosphere is of interest because
of their importance as precursors in the production of photochemical smog, as photochemical reaction products and as
major source of free radicals in the atmosphere.
4.4 Historically, DNPH impinger techniques have been widely used to determine atmospheric carbonyls. However, due
to the limitation of applying this technique to remote locations, the solid adsorbent methodology has become a
convenient alternative to impinger sampling. A number of solid adsorbents have been used over the years to support
the DNPH coating. They are: glass beads, glass fiber filters, silica gel, Chromosorb® P, Florisil®, Carbopack® B,
XAD-2, and CI 8. Several of these adsorbents are available commercially as pre-packed cartridges. The commercially
available cartridges provide convenience of use, reproducibility and low formaldehyde blanks. Two of the more widely
used pre-packed adsorbents are silica gel and CI8.
4.4.1 Silica Gel. Silica gel is a regenerative adsorbent, consisting of amorphous silica (Si02) with surface OH
groups, making it a polar material and enhancing surface absorption. DNPH-coated silica gel cartridges have been used
by numerous investigators since 1980 for sampling formaldehyde in ambient air. Tejada (3,4) evaluated several
adsorbents, including CI8, Florsil, silanized glass wool, and silica gel as possible supports for the DNPH coating.
Results indicated that silica gel provided the best support with minimum interferences. The studies did document that
olefinic aldehydes such as acrolein and crotonaldehyde degraded partially and formed unknown species. For stable
carbonyls such as formaldehyde, acetaldehyde, propionaldehyde, benzaldehyde, and acetone, correlation with an DNPH-
impinger technique was excellent. However, further studies by Arnts and Tejada identified a severe loss of carbonyl-
DNPH derivative due to the reaction of atmospheric ozone on DNPH-coated silicagel cartridges while sampling ambient
air. This bias was eliminated when sampling continued with the application of an ozone scrubber system (KI denuder)
preceding the cartridge.
4.4.2 C18 Cartridge. C18 is an octadecylsilane bonded silica substrate which is non-polar, hydrophobic, and
relatively inert, whose surface has been passivated with non-polar paraffinic groups. Because of these qualities,
January 1999
Compendium of Methods for Toxic Organic Air Pollutants
Page 11A-5
-------
Method TO-11A
Formaldehyde
C18 has been used historically as an adsorbent trap for trace organics in environmental aqueous samples through
hydrophobic interactions. The adsorbed trace organic molecules are then eluted from the adsorbent with various
organic solvents. In early 1990, C18 was used in an ambient air study as the support for DNPH. While C18 showed
promising results (23), it's use today as the support for DNPH is limited.
4.5 Both adsorbents have historically performed adequately as the support for the DNPH coating. The comparison
between silica gel and C18 as the adsorbent for the DNPH is illustrated in Table 1. The user is encouraged to review
the weaknesses and strengths outlined in Table 1 for using silica gel or CI8 as the adsorbent for the DNPH coating.
5. Definitions
[Note: Definitions used in this document and any user-prepared Standard Operating Procedures (SOPs) should be
consistent with those used in ASTMDl356. All abbreviations and symbols are defined within this document at the point
of first use.]
5.1 C18— C18 is an octadecylsilane bonded silica substrate, which is non-polar, hydrophobic, and relatively inert.
5.2 HPLC— high performance liquid chromatography.
5.3 Method Detection Limit (MDL)— the minimum concentration of an analyte that can be reported with 95% confidence
that the value is above zero, based on a standard deviation of at least seven repetitive measurements of the analyte
in the matrix of concern at a concentration near the low standard.
5.4 Photochemical Reaction— any chemical reaction that is initiated as a result of absorption of light.
5.5 Photochemical Smog— air pollution resulting from photochemical reactions.
5.6 ppbv— a unit of measure of the concentration of gases in air expressed as parts of the gas per billion (109) parts
of the air-gas mixture, normally both by volume.
5.7 ppmv— a unit of measure of the concentration of gases in air expressed as parts of the gas per million (106) parts
of the air-gas mixture, normally both by volume.
5.8 Silica Gel—silica gel is a regenerative adsorbent consisting of amorphous silica (Si02) with OH surface
groups making it a polar material and enhancing surface reactions.
5.9 Denuder— A device designed to remove gases from an air sampling stream by the process of molecular diffusion to
a collecting surface.
5.10 Certification Blank— certification blank is defined as the mean value of the cartridge blank plus three standard
deviations. For Compendium Method TO-11A, the Certification Blank should be less than 0.15 |ig/cartridge for
formaldehyde.
5.11 Cartridge Blank— cartridge blank is the measured value of the carbonyl compounds on an unsampled, DNPH-coated
cartridge. This is the value used in the calculations delineated in section 12.
5.12 Scrubber— to remove a specific gas from the air stream by passing through a pack bed.
Page 11A-6
Compendium of Methods for Toxic Organic Air Pollutants
January 1999
-------
Formaldehyde
Method TO-11A
6. Extended Methodology and Common Interferences
6.1 This procedure has been written specifically for the sampling and analysis of formaldehyde. Other carbonyl
compounds found in ambient air are also observed in the HPLC analysis. Resolution of these compounds depend upon
column and mobile phase conditions during HPLC analysis. Organic compounds that have the same retention time and
significant absorbance at 360 nm as the DNPH derivative of formaldehyde will interfere. Such interferences (24) can
often be overcome by altering the separation conditions (e.g., using alternative HPLC columns or mobile phase
compositions). In addition, other aldehydes and ketones can be detected with a modification of the basic procedure.
In particular, chromatographic conditions can be optimized to separate acetone and propionaldehyde and 12 other higher
molecular weight aldehydes and ketones (within an analysis time of about one hour), as identified below, by utilizing
one or two Zorbax ODS columns in series under a linear gradient program:
Formaldehyde
Acetaldehyde
o-Tolualdehyde
Acetone
Isovaleraldehyde
Valeraldehyde
Butyr aldehyde
m-Tolualdehyde
Propionaldehyde
Crotonaldehyde
2,5-Dimethylbenzaldehyde
Benzaldehyde
p-Tolualdehyde
Hexanaldehyde
Methyl ethyl ketone
The linear gradientprogram varies the mobile phase composition periodically to achieve maximum resolution ofthe C-3,
C-4, and benzaldehyde region of the chromatogram.
6.2 Formaldehyde may be a contamination ofthe DNPH reagent. If user- prepared cartridges are employed, the DNPH must
be purified by multiple recrystallizations in UV grade carbonyl-free acetonitrile. Recrystallization is accomplished
at 40-600 C by slow evaporation of the solvent to maximize crystal size. The purified DNPH crystals are stored under
UV grade carbonyl-free acetonitrile until use. Impurity levels of carbonyl compounds in the DNPH are determined by
HPLC prior to use and should be less than the Certification Blank value of 0.15 ; ig/cartridgc.
6.3 The purity of acetonitrile is an important consideration in the determination of allowable formaldehyde blank
concentration in the reagent. Background concentrations of formaldehyde in acetonitrile will be quantitatively
converted to the hydrazone, adding a positive bias to the ambient air formaldehyde concentration. Within the project
quality control procedures, the formaldehyde in the acetonitrile reagent should be checked on a regular basis (see
Section 9.1).
6.4 Ozone at high concentrations has been shown to interfere negatively by reacting with both the DNPH and
its carbonyl derivatives (hydrazones) on the cartridge (25,26). The extent of interference depends on the temporal
variations of both the ozone and the carbonyl compounds and the duration of sampling. Significant negative
interference from ozone was observed even at concentrations of formaldehyde and ozone typical of clean ambient air
(i.e., 2 and 40 ppb, respectively).
6.5 Exposure ofthe DNPH-coated sampling cartridges to direct sunlight may produce artifacts and should be avoided.
6.6 The presence of ozone in the sample stream is readily inferred from the appearance of new compounds with retention
times different from the other carbonyl hydrazone compounds.
6.7 The most direct solution to the ozone interference is to remove the ozone before the sample stream reaches the
coated cartridge. This process entails constructing an ozone denuder (9) or scrubber and placing it in front of the
cartridge. The denuder can be constructed of 1 m of 0.64-cm outside diameter (O.D.) by 0.46-cm inside diameter (I.D.)
copper tubing, that is filled with a saturated solution of KI, allowed to stand for a few minutes, drained and dried
January 1999
Compendium of Methods for Toxic Organic Air Pollutants
Page 11A-7
-------
Method TO-11A
Formaldehyde
with a stream of clean air or nitrogen for about 1 h. The capacity of the ozone denuder as described is about 100,000
ppb-hour of ozone. Packed-bed granular potassium iodide (KI) scrubbers can also be used in place of the denuder and
are commercially available. Very little work has been done on long term usage of a denuder or KI scrubber to remove
ozone from the ambient air gas stream. The ozone removal devices should be replaced periodically (e.g., monthly) in
the sample train to maintain the integrity of the data generated.
6.8 Test aldehydes or carbonyls (formaldehyde, acetaldehyde, acrolein, propionaldehyde, benzaldehyde, and p-
tolualdehyde) that were dynamically spiked into an ambient sample air stream passed through the KI denuder with
practically no losses (7). Similar tests were also performed for formaldehyde (26).
6.9 Ozone scrubbers (cartridge filled with granular KI) are also available from suppliers of pre-coated DNPH
cartridges. These scrubbers are optimized when the ambient air contains a minimum of 15% relative humidity.
7. Apparatus
7.1 IsocraticHPLC. System consisting ofamobilephase reservoir ahighpressurepump; aninjection valve (automatic
samplcrwith an optional 25-|iLloop injector): aZorbaxODS (DuPontlnstruments, Wilmington, DE)reversephase (RP)
column, or equivalent (25-cm x 4.6-mm ID); a variable wavelength UV detector operating at 360 nm; and a data system,
as illustrated in Figure 1.
[Note: Most commercial HPLC analytical systems will be adequate for this application.]
12 Cartridge sampler. Prepacked, pre-coated cartridge (see Figure 2), commercially available or coated in situ with
DNPH according to Section 9.
[Note: This method was developed using the Waters Sep-Pak cartridge, coated in situ with DNPH on silica gel by the
users, as delineated in the original Compendium Method TO-11 as a guideline. EPA has experience in use of this
cartridge during various field monitoring programs over the last several years. Other manufacturer's cartridges
shouldworkaswell. However, modificationstotheseproceduresmaybenecessaryifanothercommerciallyavailable
cartridge is selected.]
Major suppliers of pre-coated cartridges are:
• Supelco, Supelco Park, Bellefonte, PA 16823-0048, (800) 247-6628.
• SKC Inc., 334 Valley View Road, Eighty Four, PA 15330-9614, (800) 752-8472.
• Milliporc/Waters Chromatography, P.O. Box 9162, Marlborough, MA 01752-9748,
(800) 252-4752.
• Atmospheric Analysis and Consulting (AAC) Inc., 4572 Telephone Rd., Suite 920, Ventura, CA 93003, (805) 650-
1642.
[Note: The SKC cartridge (see Figure 2) is an example of a dual bed tube. The glass cartridge contains afront bed of
300 mgDNPH-coated silica gel with the backbed of150 mgDNPH-coated silica gel. Airflow through the tube should be
from front to backbed, as indicated by the arrows enscribed on the cartridge. The dual bed tube cartridge may be used
in atmospheres containing carbony I concentrations in excess of the American Conference of Government Industrial
Hygienists (ACGTH) 8-hour exposure limit, where breakthrough of carbonyls on the adsorbent might occur. Ifused in
routine ambient air monitoring applications, the tube is recovered as one unit, as specified in Section 11.2.]
Page 11A-8
Compendium of Methods for Toxic Organic Air Pollutants
January 1999
-------
Formaldehyde
Method TO-11A
IfcommerciallypreparedDNPH-coatedcartridgesarepurchased, ensure ihata"CertificationBlankfor Formaldehyde"
is provided for the specific batch of which that cartridge is a member. For a commercial cartridge to be acceptable,
the following criteria must be met:
• Formaldehyde concentration: <0.15 (ig/cartridge.
If the enhanced carbonyl analysis is being performed, the following Certification Blank criteria must also be met:
• Speciated carbonyl concentration:
- Acetaldehyde: <0.10 (ig/cartridge
- Acetone: <0.30 (ig/cartridge
- Other: <0.10 (ig/cartridge
Typical physical and chemical characteristics of commercial cartridge adsorbents are listed in Table 2 and illustrated
in Figure 2.
7.3 Sampling system, the DNPH-cartridgeapproachis capable of accurately and precisely sampling 100-2000 mL/min
of ambient air. The monitoring of carbonyl compounds has recently been enhanced by the promulgation of new ambient
air quality surveillance regulations outlined in Title 40, Part 58. These regulations require States to establish
additional air monitoring stations as part of their existing State Implementation Plan (SIP) monitoring network as
partofEPA'sPhotochemicalAssessmentMonitoringStations(PAMS)toincludeprovisionsforenhanced(l)monitoring
of ozone and oxides of nitrogen (NO,). (2) monitoring of volatile organic compounds (V OCs), (3) monitoring of
meteorologicalparameters,and(4)monitoringselected carbonyl compounds (formaldehyde, acetone, andacetaldehyde).
Specifically, monitoring for carbonyl involves:
• 8, 3 h sequential samples starting at midnight.
• 1, 24 h time-integrated "reality check" sample.
Consequently, the samplermustbe able to accommodate numerous regulatory and practical needs. Practical needs would
include:
• Ability to sequence two cartridges in series for breakthrough volume confirmation for a 24-hour sampling event.
• Ability to collocate with any of the 8, 3 h samples.
Traditionally, three sampling approaches have been used to monitor carbonyl compounds in the ambient air. They are:
• Manual single-port carbonyl sampler.
• Programmable single-port carbonyl sampler.
• Automated multi-port sampler.
Components of the single-port carbonyl sampler, for both manual and semi-automatic, are illustrated in Figure 3.
Components usually include a heated manifold/sample inlet, a denuder/cartridge assembly, a flow meter, a vacuum
gauge/pump, atimer and apower supply. In operation, ambient air is drawn through the denuder/cartridge assembly with
a vacuum pump at a fixed flow rate between 0.1 to 2 Lpm. The vacuum gauge is used to measure the net vacuum in the system
for all flow-rate corrections. Controlling the system is usually a 7-day, 14-event timer to coordinate sampling events
to allow a sample to be extracted continuously or intermittently over a period of time. Finally, an elapsed-time
counter is employed to measure the actual time the sampling took place. This is particularly suitable for unattended
sampling when power fails for short periods.
The automated multi-port sampler is especially designed to collect numerous short-term (2 to 3 hours) sample
sequentially over a 24 hour, 7 day a week, nighttime and weekend monitoring period. This arrangement allows for the
sampling of short periods where the objectives of the project are to identify progress of atmospheric reactions
involving carbonyls. As illustrated in Figure 4, components of the fully automated multi-port carbonyl sampler
January 1999
Compendium of Methods for Toxic Organic Air Pollutants
Page 11A-9
-------
Method TO-11A
Formaldehyde
includes a heated inlet, ozone denuder (or scrubber) inlet manifold assembly, inlet check valves, DNPH multi-port
cartridge assembly, exhaust manifold, mass flow controller and sample pump. The multi-port sampler automatically
switches between sampling ports at preselected times, as programmed by the user. Typically, a sequential air sampler
contains a microprocessor timer/controller that provides precise control over each sampling event. The microprocessor
allows the user to program individual start date and time, sample duration, and delays between samples. The timer also
allows activation of the flow system prior (approximately 10 min) to sequencing to allow purging of the sampler inlet
with fresh sample. Finally, the automated sequential sampler can be operated from an external signal, such as an ozone
monitor, so that sampling starts above certain preset ozone levels or via a modem. As a final option, various
manufacturers provide wind sensor instrumentation (wind speed and direction) which is connected to the automated
sequential sampler so that sampling begins when the wind is from a preset direction and speed.
Major suppliers of commercially available carbonyl samplers are:
• Supelco, Supelco Park, Bellefonte, PA 16823-0048, (800) 247-6628.
• SKC Inc., 334 Valley View Road, Eighty Four, PA 15330-9614, (800) 752-8472.
• Millipore/Waters Chromatography, P.O. Box 9162, Marlborough, MA 01752-9748, (800) 252-4752.
• XonTech, Inc. 6862 Hayvenhurst Avenue, VanNuys, CA 91406, (818) 787-7380.
• ATEC Atmospheric Technology, P.O. Box 8062, Calabasas, CA 91372-8062, (310) 457-2671.
• Atmospheric Analysis and Consulting (AAC) Inc., 4572 Telephone Road, Suite 920, Ventura, CA 93003,
(805)650-1642.
• Scientific Instrumentation Specialists, P.O. Box 8941, Moscow, ID, (209) 882-3860.
7.4 Stopwatch.
7.5 Polypropylene shipping container (see Figure 5) with polyethylene-air bubble padding. To hold
sample cartridges.
7.6 Thermometer. To record ambient temperature.
7.7 Barometer (optional).
7.8 Volumetric flasks. Various sizes, 5-2000 mL.
7.9 Pipets. Various sizes, 1-50 mL.
7.10 Erlenmeyer flask, 1 L. For preparing HPLC mobile phase.
7.11 Graduated cylinder, 1 L. For preparing HPLC mobile phase.
7.12 Syringe, 100-250 jj.h. For HPLC injection, with capacity at least four times the loop value.
7.13 Sample vials.
7.14 Melting point apparatus (optional).
7.15 Rotameters.
7.16 Calibrated syringes.
Page 11A-10
Compendium of Methods for Toxic Organic Air Pollutants
January 1999
-------
Formaldehyde
Method TO-11A
7.17 Soap bubble meter or wet test meter.
7.18 Mass flow meters and mass flow controllers. For metering/setting air flow rate through sample cartridge of
100-2000 mL/min.
[Note: The massflow controllers are necessary because cartridges may develop a high pressure drop and at maximumflow
rates, the cartridge behaves like a "critical orifice." Recent studies have shown that critical flow orifices may be
usedfor 24-hour samplingperiods at a maximum rate of2L/minfor atmospheres not heavily loaded with particulates
without any problems.]
7.19 Positive displacement. Repetitive dispensing pipets (Lab-Industries, or equivalent), 0-10 mL range.
7.20 Cartridge drying manifold. With multiple standard male Luer® connectors.
7.21 Liquid syringes. 10 mL (polypropylene syringes are adequate) for preparing DNPH-coated cartridges.
7.22 Syringe rack. Made of an aluminum plate (0.16 cm x 36 cm x 53 cm) with adjustable legs on four corners. A matrix
(5 cm x 9 cm) of circular holes of diameter slightly larger than the diameter of the 10-mL syringes was symmetrically
drilled from the center of the plate to enable batch processing of 45 cartridges for cleaning, coating, and/or sample
elution.
7.23 Luer® fittings/plugs. To connect cartridges to sampling system and to cap prepared cartridges.
7.24 Hot plates, beakers, flasks, measuring and disposable pipets, volumetric flasks, etc. Used in the
purification of DNPH.
7.25 Culture tubes (20 mm x 125 mm) with polypropylene screw caps. Used to transport coated cartridges for field
applications (see Figure 5), Fisher Scientific, Pittsburgh, PA, or equivalent.
7.26 Polyethylene gloves. Used to handle cartridges, best source.
7.27 Dry test meter.
7.28 User-prepared copper tubing for ozone scrubber (see Figure 6a). A 36 inch length of %-inch O.D. copper tubing
is used as the body of the ozone scrubber. The tubing should be coiled into a spiral approximately 2 inches in O.D.
EPA has considerable field experience with the use of this denuder.
[Note: Ozone scrubbers (cartridge filled with granular KI) are also available from suppliers ofpre-coated DNPH
cartridges, as illustrated in Figure 6(b).]
7.29 Cord heater and Variac. A 24 inch long cord heater, rated at approximately 80 watts, wrapped around the outside
of the copper coil denuder, controlled by a Variac, to provide heat (~500 C) to prevent condensation of water or organic
compounds from occurring within the coil.
7.30 Fittings. Bulkhead unions are attached to the entrance and exit of the copper coil to allow attachment to other
components of the sampling system.
January 1999
Compendium of Methods for Toxic Organic Air Pollutants
Page 11A-11
-------
Method TO-11A
Formaldehyde
8. Reagents and Materials
[Note: Purity of Reagents—Reagent grade chemicals shall be used in all tests. Unless otherwise indicated, it is
intended that all reagents conform to the specifications of the Committee on Analytical Reagents of the American
Chemical Society where such specifications are available; Purity of Water—Unless otherwise indicated, references
to water shall be understood to mean reagent water as defined by Type II of ASTMSpecifications D 1193.]
8.1 2,4-Dinitrophenylhydrazine(DNPH). Aldrich Chemical or J. T. Baker, reagent grade or equivalent. Recrystallize
at least twice with UV grade acetonitrile before use.
8.2 DNPH coated cartridges. DNPH coated cartridge systems are available from several commercial suppliers.
8.3 High purity acetonitrile. UV grade, Burdick and Jackson "distilled-in-glass," or equivalent. The formaldehyde
concentration in the acetonitrile should be <1.5 ng/mL. It is imperative (mandatory) that the user establish the
purity of the acetonitrile before use (see Section 9.1).
8.4 Deionized-distilled water. Charcoal filtered.
8.5 Perchloric acid. Analytical grade, best source, 60%, specific gravity 1.51.
8.6 Ortho-phosphoric acid. Analytical grade, best source, 36.5-38%, specific gravity 1.19.
8.7 Formaldehyde. Analytical grade, best source, 37% solution (w/w).
8.8 Aldehydes and ketones, analytical grade, best source. Used for preparation of DNPH derivative
standards (optional).
8.9 Carbonyl hydrazones. Formaldehyde and other carbonyl hydrazones are available for use as standards from
commercial sources at various levels of purity.
8.10 Ethanol or methanol. Analytical grade, best source.
8.11 Nitrogen. High purity grade, best source.
8.12 Charcoal. Granular, best source.
8.13 Helium. High purity grade, best source.
8.14 Potassium Iodide. Analytical grade, best source. Used for coating inside of copper tubing of denuder system
to remove ozone interference.
9. Preparation of Reagents and Cartridges
9.1 Purity of the Acetonitrile
Page 11A-12
Compendium of Methods for Toxic Organic Air Pollutants
January 1999
-------
Formaldehyde
Method TO-11A
9.1.1 The purity of acetonitrile is an important consideration in the determination of the formaldehyde blank
concentration. Formaldehyde in the reagent will be quantitatively converted to the hydrazone and measured as part
of the blank. The contribution to the blank from the reagent is dependent on the formaldehyde concentration in the
reagent and the amount of the reagent used for extraction. Some examples will illustrate these considerations.
Example A
• Silica gel DNPH cartridge has a blank level of 60 ng.
• Cartridge is eluted with 5-mL of acetonitrile reagent containing a formaldehyde of 3 ng/mL.
• Analyst measures a blank level of 75 ng of which 80% comes from the cartridge and 20% comes from the reagent.
Example B
• Silica gel DNPH cartridge has a blank level of 30 ng.
• Cartridge is eluted with 5 mL of acetonitrile reagent containing a formaldehyde of 6 ng/mL.
• Analyst measures a blank level of 60 ng of which 50% comes from the cartridge and 50% comes from the reagent.
9.1.2 As a quality control procedure, the formaldehyde in the acetonitrile reagent should be checked on a regular
basis. This can be done by mixingknown proportions ofthe acetonitrile reagent and a DNPH solution having ameasured
formaldehyde blank. (The extract from a blank cartridge can serve as the DNPH solution.) After analyzing the
resultant solution, a mass balance is performed on the observed formaldehyde level and the contribution from the DNPH
reagent as shown in the following example.
• 1 mL of a DNPH solution containing 2.1 ng/mL of formaldehyde (as carbonyl) is mixed with 9 mL of acetonitrile
reagent containing as unknown formaldehyde blank. The analyst measures a resultant solution concentration of
1.55 ng of formaldehyde. This data can be used to calculate the formaldehyde in the reagent:
HCHOng^mL = (1-55 ng/mLx 10 mL-2.1 ng/mLx 1 ml) = l 49j^mL
9 mL
The formaldehyde contribution to the cartridge blank should be as low as possible but certainly less than 20% of the
total measured blank. Using a cartridge blank level of 30 ng/cartridge, the formaldehyde concentration in the reagent
would have to be less than 1.5 ng/mL (i.e., 50 nM) to give a blank level less than 20% of the measured blank.
9.2 Purification of 2,4-Dinitrophenylhydrazine (DNPH)
[Note: This procedure shouldbe performedunderaproperlyventilatedhood, as inhalation of acetonitrile can result
in nose and throat irritation. Various health effects are resultant from the inhalation of acetonitrile. At 500 ppm
in air, brief inhalation has produced nose and throat irritation. At 160 ppm, inhalation for 4 hours has caused
flushing of the face (2 hour delay after exposure) and bronchial tightness (5 hour delay). Heavier exposures have
produced systemic effects with symptoms rangingfrom headache, nausea, and lassitude to vomiting, chest or abdominal
pain, respiratory depression, extreme weakness, stupor, convulsions and death (dependent upon concentration and
time).]
[Note: Purified DNPH, suitable for preparing cartridges, can be purchased commercially.]
January 1999
Compendium of Methods for Toxic Organic Air Pollutants
Page 11A-13
-------
Method TO-11A
Formaldehyde
9.2.1 Prepare a supersaturated solution of DNPH by boiling excess DNPH in 200 mL of acetonitrile for approximately
one hour.
9.2.2 After one hour, remove and transfer the supernatant to a covered beaker on a hot plate and allow gradual
cooling to 40-60°C.
9.2.3 Maintain the solution at this temperature (40-60 °C) until 95% of solvent has evaporated.
9.2.4 Decant solution to waste, and rinse crystals twice with three times their apparent volume of acetonitrile.
9.2.5 Transfer crystals to another clean beaker, add 200 mL of acetonitrile, heat to boiling, and again let
crystals grow slowly at 40-60 °C until 95% of the solvent has evaporated.
9.2.6 Repeat rinsing process as described in Section 9.2.4.
9.2.7 Take an aliquot of the second rinse, dilute 10 times with acetonitrile, acidify with 1 mL of 3.8 M perchloric
acid per 100 mL of DNPH solution, and analyze by HPLC.
[Note: An acidis necessary to catalyze the reaction ofthe carbonyls with DNPH. Most strong inorganic acids such as
hydrochloric, sulfuric, phosphoric, or perchloric acids will do the job. Perchloric or phosphoric acids are the
preferred catalystfor using acetonitrile solution of DNPH as the absorbing solution. The DNPH derivatives do not
precipitate from solution as readily as when hydrochloric or phosphoric acids are used as the catalyst. This is an
ideal situation for an HPLC analytical finish as this minimizes sample handling. For most ambient air sampling,
precipitation is not a problem because the carbonyl concentration is generally in the ppb range.]
9.2.8 An impurity level of <0.15 , «g/cartridgc of formaldehyde in DNPH-coated cartridge is acceptable (based on
the Certification Blank section 5.10). An acceptable impurity level for an intended sampling application may be
defined as the mass of the analyte (e.g., DNPH-formaldehyde derivative) in a unit volume of the reagent solution
equivalent to less than one tenth (0.1) the mass of the corresponding analyte from a volume of an air sample when the
carbonyl (e.g., formaldehyde) is collected as DNPH derivative in an equal unit volume of the reagent solution. An
impurity level unacceptable for a typical 10 L sample volume may be acceptable if sample volume is increased to 100 L.
If the impurity level is not acceptable for intended sampling application, repeat recrystallization.
9.2.9 If the impurity level is not satisfactory, pipet off the solution to waste, then add 25 mL of acetonitrile
to the purified crystals. Repeat rinsing with 20 mL portions of acetonitrile until a satisfactorily low impurity level
in the supernatant is confirmed by HPLC analysis.
9.2.10 If the impurity level is satisfactory, add another 25 mL of acetonitrile, stopper and shake the reagent
bottle, then set aside. The saturated solution above the purified crystals is the stock DNPH reagent.
9.2.11 Maintain only a minimum volume of saturated solution adequate for day to day operation. This will minimize
wastage of purified reagent should it ever become necessary to re-rinse the crystals to decrease the level of impurity
for applications requiring more stringent purity specifications.
9.2.12 Use clean pipets when removing saturated DNPH stock solution for any analytical applications. Do not pour
the stock solution from the reagent bottle.
9.3 Preparation of DNPH-Formaldehyde Derivative
[Note: Purified crystals or solutions of DNPH-derivatives can be purchased commercially.]
9.3.1 To a portion of the recrystallized DNPH, add sufficient 2N HC1 to obtain an approximately saturated
solution. Add to this solution formaldehyde (other aldehydes or ketones may be used if their detection is desirable),
in molar excess of the DNPH. Allow it to dry in air.
9.3.2 Filter the colored precipitate, wash with 2N HC1 and water and let the precipitate air dry.
9.3.3 Checkthe purity ofthe DNPH-formaldehyde derivative by melting point determination or HPLC analysis. The
DNPH-formaldehyde derivative should melt at 167°C ± 1 °C. If the impurity level is not acceptable, recrystallize the
Page 11A-14
Compendium of Methods for Toxic Organic Air Pollutants
January 1999
-------
Formaldehyde
Method TO-11A
derivative in ethanol. Repeat purity check and recrystallization as necessary until acceptable level of purity (e.g.,
99%) is achieved.
9.3.4 DNPH derivatives of formaldehyde and other carbonyls suitable for use as standards are commercially
available both in the form of pure crystals and as individual or mixed stock solutions in acetonitrile.
9.4 Preparation of DNPH-Formaldehyde Standards
9.4.1 Prepare a standard stock solution of the DNPH-formaldehyde derivative by dissolving accurately weighed
amounts in acetonitrile.
9.4.2 Prepare a working calibration standard mix from serial dilution of the standard stock solution. The
concentration of the DNPH-formaldehyde compound in the standard mix solutions should be adj usted to reflect relative
distribution in a real sample.
[Note: Individual stock solutions of approximately 100 mg/L are prepared by dissolving 10 mg of the solid derivative
in 100 mL of acetonitrile. The individual solution is used to prepare calibration standards containing the derivative
of interest at concentrations of0.5-20. g ml., which spans the concentration of interestfor most ambient air work. ]
9.4.3 Store all standard solutions in a refrigerator. They should be stable at least one month.
9.4.4 DNPH-formaldehyde standards can also be purchased from various commercial suppliers. If
purchased, ensure that a "Certification of Concentration" is provided.
9.5 Preparation of DNPH-Coated Cartridges
[Note: Thisprocedure must beperformed in an atmosphere with avery low aldehyde background. All glassware andplastic
waremustbescrupulouslycleanedandrinsedwithdeionizedwaterandcarbonylfreeacetonitrile. Contactofreagents
with laboratory air must be minimized. Polyethylene gloves must be worn when handling the cartridges. If the user
wishes to purchase commercially prepared DNPH-coated cartridges, they are available from various vendors. If
commercial prepared DNPH-coated cartridges are purchased, ensure that a "Certification Blankfor Formaldehyde" is
provided for the specific batch of which that cartridge is a member. For a commercial cartridge to be acceptable, the
following criteria must be met:
• Formaldehyde concentration: <0.15 fig/cartridge.
If the enhanced carbonyl analysis is being performed, the following Certification Blank criteria must also be met:
• Speciated carbonyl concentration:
- Acetaldehyde: <0.10 ftg/cartridge
- Acetone: <0.30 ^g/cartridge
- Other: <0.10 ft g/cartridge
One who is not experienced in the preparation of DNPH-coated cartridge is strongly advised to use certified
commercially available cartridges.]
9.5.1 DNPH Coating Solution
9.5.1.1 Pipet 30 mL of saturated DNPH stock solution to a 1000 mL volumetric flask, then add 500 mL acetonitrile.
9.5.1.2 Acidify with 1.0 mL of ortho-phosphoric acid (H3P04).
January 1999
Compendium of Methods for Toxic Organic Air Pollutants
Page 11A-15
-------
Method TO-11A
Formaldehyde
[Note: The atmosphere above the acidifiedsolutionshouldpreferablybefilteredthroughaDNPH-coatedcartridge to
minimize contamination from laboratory air. Shake solution, then make up to volume with acetonitrile. Stopper the
flask, invert and shake several times until the solution is homogeneous. Transfer the acidified solution to a reagent
bottle with a 0-10 mL range positive displacement dispenser.]
9.5.1.3 Prime the dispenser and slowly dispense 10-20 mL to waste.
9.5.1.4 Dispense an aliquot solution to a sample vial, and check the impurity level of the acidified solution
by HPLC according to Section 9.2.
9.5.1.5 The impurity level should be less than the Certification Blank of <0.15 ,ig/cartridgc for formaldehyde,
similar to that in the DNPH coating solution.
9.5.2 Coating of Cartridges
9.5.2.1 Open the pre-packed cartridge package, connect the short end to a 10-mL syringe, and place it in a
syringe rack (see Figure 7).
[Note: Prepare as many cartridges (-100) and syringes as possible.]
9.5.2.2 Using a positive displacement repetitive pipet, add 10 mL of acetonitrile to each of the syringes (see
Figure 7).
9.5.2.3 Let liquid drain to waste by gravity.
[Note: Remove any air bubbles that maybe trapped between the syringe and the silica cartridge by displacing them with
the acetonitrile in the syringe.]
9.5.2.4 Set the repetitive dispenser containing the acidified DNPH coating solution to dispense 7 mL into the
cartridges.
9.5.2.5 Once the effluent flow at the outlet of the cartridge has stopped, dispense 7 mL of the DNPH coating
reagent into each of the syringes (see Figure 7).
9.5.2.6 Let the coating reagent drain by gravity through the cartridge until flow at the other end of the
cartridge stops.
9.5.2.7 Wipe the excess liquid at the outlet of each of the cartridges with clean tissue paper.
9.5.2.8 Assemble a drying manifold with a scrubber or "guard cartridge" connected to each of the ports (see
Figure 7). These "guard cartridges" are DNPH-coated and serve to remove any trace of formaldehyde in the nitrogen gas
supply.
9.5.2.9 Insert cartridge connectors (flared at both ends, 0.64 by 2.5-cm outside diameter TFE-fluorocarbon FEP
tubing with inside diameter slightly smaller than the outside diameter of the cartridge port) onto the long end of the
scrubber cartridges.
9.5.2.10 Remove the cartridges from the syringes and connect the short ends to the exit end of the scrubber
cartridge.
9.5.2.11 Pass nitrogen through each of the cartridges at about 300-400 mL/min for 5-10 minutes.
9.5.2.12 Within 10 minutes of the drying process, rinse the exterior surfaces and outlet ends of the cartridges
with acetonitrile using a Pasteur pipet.
9.5.2.13 Stop the flow of nitrogen after 15 minutes, wipe the cartridge exterior free of rinsed acetonitrile
and remove the dried cartridge.
9.5.2.14 Plug both ends of the coated cartridge with standard polypropylene Luer® male plugs, place the plugged
cartridge in a shipping tube with polypropylene screw caps.
9.5.2.15 Put a serial number and a lot number label on each of the individual shipping tubes.
9.5.2.16 Store shipping tubes containing the DNPH-coated cartridges in a refrigerator at 4°C until use.
Page 11A-16
Compendium of Methods for Toxic Organic Air Pollutants
January 1999
-------
Formaldehyde
Method TO-11A
[Note: Plugged cartridges may also be placed in screw-capped glass culture tubes andplaced in a refrigerator until
use. Cartridges will maintain their integrity for up to 90 days stored in refrigerated, capped shipping tubes.]
9.5.2.17 Take a minimum of 3 blank cartridges from the cartridge batch and analyze for formaldehyde, as
delineated in Section 11. The batch of user-prepared DNPH-coated cartridges is acceptable if the following criteria
are met:
• Formaldehyde Certification Blank: <0.15 (ig/cartridge.
If the enhanced carbonyl analysis is being performed, the following certification criteria must also be met:
• Speciated carbonyl concentration:
- Acetaldehyde: <0.10 (ig/cartridge
- Acetone: <0.30 (ig/cartridge
- Other: <0.10 (ig/cartridge
9.5.2.18 If analysis meets the above criteria, provide documentation with all cartridges associated with
that batch involving "Certification Blank for Formaldehyde. " This certificate must be part of the project
records.
9.5.2.19 If the cartridge results are close to, but above the Certification Blank, run a few more blank
cartridges to check background level.
9.5.2.20 If analysis indicates failure of the cartridge, then all cartridges in that batch are unacceptable.
Prepare a new batch of cartridges according to Section 9.5 until certification is achieved.
9.5.2.21 Store all certified cartridges in a refrigerator at 4°C until use.
9.5.2.22 Before transport, remove the shipping container (or screw-capped glass culture tubes) containing the
adsorbent tubes from the refrigerator and place culture tubes in a friction-top metal can containing 1-2 inches of
charcoal for shipment to sampling location. Alternately, acidified DNPH-coated filters can be used in place of
charcoal filters to remove impurity carbonyl compounds in the air.
9.5.2.23 As an alternative to friction-top cans for transporting sample cartridges, the coated cartridges could
be shipped in their individual glass containers (see Figure 5a). A batch of coated cartridges may also be packed in
a polypropylene shipping container for shipment to the field (see Figure 5b). The container should be padded with
cleantissuepaperorpolyethylene-airbubblepadding. Donotusepolyurethanefoamornewspaperaspaddingmaterial.
9.5.2.24 The cartridges should be immediately stored in a refrigerator or freezer (<40 C) upon arrival in the
field.
9.6 Equivalent Formaldehyde Cartridge Concentration
9.6.1 Onecancalculatetheequivalentformaldehydebackgroundconcentration(ppbv)contributedfromacommercial
or user-prepared DNPH-coated cartridge following exposure to formaldehyde-free air.
9.6.2 The equivalent formaldehyde background concentration includes the contribution of formaldehyde from both
the acetonitrile and the cartridge.
9.6.3 Knowing the equivalent background concentration, as determined by the user (see Section 9.5.2) or supplied
by the commercial supplier (see Note. Section 9.5), of formaldehyde in the cartridge (ng/cartridge), the formaldehyde
background concentration contributedby the DNPH-coated cartridge (thus the method minimum detection limits) can be
related to the total sample volume, as identified in Table 3.
January 1999
Compendium of Methods for Toxic Organic Air Pollutants
Page 11A-17
-------
Method TO-11A
Formaldehyde
9.6.4 For example, if the averaged background formaldehyde concentration supplied by the manufacturer is 70
ng/cartridge, then that cartridge can add 0.95 ppbv of equivalent formaldehyde, to the final ambient air concentration
value, as delineated in Table 3 for a total air volume of 60 L.
9.6.5 The user should use DNPH-coated cartridges with the lowest background concentration to improve accuracy
and detection limits.
10. Sampling Procedure
10.1 The sampling system is assembled and should be similar to that shown in Figures 3 and 4.
[Note: Figures 3 and 4 illustrate different tube/pump configurations. The tester should ensure that the pump is
capable of constant flow rate throughout the sampling period.]
It is recommended that the sampling system employ a heated inlet (~500 C) coupled to an ozone denuder or scrubber to
minimize water and ozone interference associated with the DNPH-coated adsorbent tube. Historically, the coated
cartridges have been used as direct probes and traps for sampling ambient air when the ambient temperature was above
freezing.
[Note: As illustrated in Figure 8, the ozone denuder has been effective for up to 80 hours without
breakthrough at ozone levels of approximately 700ppb. Other studies have evaluated both denuders andscrubbers at
ozone concentrations between 125 and200ppbv andfound they have effectively removed ozonefrom the air stream for up
to 100,000ppb-hours; however, moisture was required (~ 10% RH) in the gas stream (26). The user should evaluate the
length of time of the application ofthe denuder or scrubber to hisfield work. Caution should be utilized when using
these devicesfor extensive periods oftime at high humidity (> 65 %). Regarding the 24 hour samples, special caution
shoudbe taken while samplingnighttimeperiods when relative humidities approaching 100% are frequently encountered.
It is recommended that routine schedule of ozone removal device replacement should be implemented as part ofthe
sampling program.]
[Note: For sampling ambient air belowfreezing, a short length (30-60 cm) of heated (50-60° F) stainless steel tubing
must be added to condition the air sample prior to collection on the DNPH-coated cartridges.]
10.2 Before sample collection, the system must be checked for leaks. Plug the inlet of the system so no flow is
indicated at the output end of the pump. The mass flow meter should not indicate any air flow through the sampling
apparatus.
10.3 Air flow through the DNPH-adsorbent cartridge may change during sampling as airborne particles deposit on the
front of the cartridge. The flow change could be significant when sampling particulate-laden atmospheres. Particle
concentrations greater than 50 ug/m3 are likely to represent a problem. For unattended or extended sampling periods,
a mass flow controller is highly recommended to maintain constant flow. The mass flow controller should be set at least
20% below the maximum air flow through the cartridge.
10.4 The entire assembly (including a "test" sampling cartridge) is installed and the flow rate checked at a value near
the desired sampling rate. In general, flow rates of 1,000-2,000 mL/min should be employed. The total sample volume
shouldbe selected toensurethatthecollectedformaldehydeconcentrationexceeds the backgroundformaldehydeDNPH-
cartridge concentration, as illustrated in Table 3. The total moles of carbonyl in the volume of air sampled should
Page 11A-18
Compendium of Methods for Toxic Organic Air Pollutants
January 1999
-------
Formaldehyde
Method TO-11A
not exceed that of the DNPH concentration (i.e., 2 mg cartridge). In general, a safe estimate of the sample size should
be 75% of the DNPH loading of the cartridge.
[Note: If the user suspects that there will be breakthrough ofaDNPH-coated cartridge during the sampling event, a
backup cartridge should be used during the first sampling event. One would analyze the back-up cartridge for
formaldehyde. If the back-up cartridge concentration exceeds 10% ofthe formaldehyde concentration on the front
cartridge, then continue to use back-up cartridges in the monitoring program. However, if formaldehyde is not detected
above the average blanklevel in the back-up cartridge after thefirstsamplingevent, then one can continue to use only
one cartridge under normal representative conditions.]
[Note: The SKC tube is a dual bed configuration, allowing one to analyze the backbed (see Figure 2) for quantifying
breakthrough.]
Generally, calibration is accomplished using a soap bubble flow meter or calibrated wet test meter connected to the
flow exit, assuming the system is sealed.
[Note: ASTMMethodD3686describes an appropriate calibration scheme that does notrequire a sealedHow system
downstream of the pump.]
10.5 The operator must measure and record the sampling flow rate at the beginning and end of the sampling period to
determinesamplevolume. Adrygasmetermay be includedin the system tomeasure total sample volume and to compare
against the in-line mass flow controller. Some commerical systems use flow monitors with data loggers to make these
measurements.
10.6 Before sampling, flush the inlet (denuder/manifold, etc.) for approximately 15 min at the established flow rate
to condition the system. Remove the glass culture tube from the friction-top metal can or styrofoam box. Let the
cartridge warm to ambient temperature in the glass tube before connecting it to the sample train.
10.7 Using polyethylene gloves, remove the DNPH-coated cartridge from the shipping container and connect it to the
sampling system with a Luer® adapter fitting. Most commercially available cartridges are bidirectional. However,
review manufacturer suggestions for orientation of the cartridge to the inlet of the sampler.
[Note: If using the SKC dual bed tube, ensure the ambient air is pulled through the tube in the direction enscribed
on the tube by an arrow.]
RecordthefollowingparametersonCompendiumMethodTO-llAfieldtestdatasheet(FTDS),asillustratedinFigure9:
date, sampling location, time, ambient temperature, barometric pressure (if available), relative humidity (if
available), dry gas meter reading (if appropriate), flow rate, rotameter setting, cartridge batch number, and dry gas
meter pump identification numbers.
10.8 The sampler is turned on and the flow is adjusted to the desired rate. A typical flow rate through one cartridge
is 1.0 L/min and 0.8 L/min for two tandem cartridges.
10.9 The sampler is operated for the desired period, with periodic recording of the variables listed in Figure 9.
10.10 Ifthe ambient air temperature during sampling is below 15 °C, a heated inlet probe is recommended. However,
no pronounced effect of relative humidity (between 25% - 90%) has been observed for sampling under various weather
January 1999
Compendium of Methods for Toxic Organic Air Pollutants
Page 11A-19
-------
Method TO-11A
Formaldehyde
conditions-cold, wet, and dry winter months and hot and humid summer months. However, a negative bias has been
observed when the relative humidity is <25%. At high humidity, the possibility of condensation must be guarded
against, especially when sampling is an air conditioned trailer.
10.11 At the end ofthe sampling period, the parameters discussed in Section 10.7 are recorded and the sample flow is
stopped. If a dry gas meter is not used, the flow rate must be checked at the end of the sampling interval. If the flow
rates at the beginning and end of the sampling period differ by more than 10%, the sample should be marked as suspect.
10.12 Immediately after sampling, remove the cartridge (using polyethylene gloves) from the sampling system, cap with
Luer® end plugs, and place it back in the original labeled glass shipping container or culture tube. Cap, seal with TFE-
fluorocarbon tape, and place it in appropriate padding. Refrigerate at 4°C until analysis. Refrigeration period prior
to analysis should not exceed 2 weeks. If a longer storage period is expected, the cartridge should be extracted with
5 mL of acetonitrile (see Section 11.2.4 and 11.2.5) and the eluant placed in a vial for long term storage.
[Note: If samples are to be shipped to a central laboratory for analysis, the duration ofthe non-refrigeratedperiod
should be kept to a minimum, preferably less than two days.]
10.13 If a dry gas meter or equivalent total flow indicator is not used, the average sample flow rate must be calculated
according to the following equation:
Qi + Q2 + Qn
A N
where:
Qa = average flow rate, L/min.
Qi. Q2. ... QN = flow rates determined at beginning, end, and intermediate points during sampling, L/min.
N = number of points averaged.
10.14 The total flow rate is then calculated using the following equation:
Vm = (T2-T,)xQA
where:
Vm =
total volume sampled at measured temperature and pressure, L.
T2 =
stop time, minutes.
T: =
start time, minutes.
T2 - Tj =
total sampling time, minutes.
Qa =
average flow rate, L/min.
10.15 The total volume (Vs)atEPA standard conditions, 25 °Cand760mmHg,iscalculatedfrom the following equation:
Page 11 A-20 Compendium of Methods for Toxic Organic Air Pollutants January 1999
-------
where:
Vs = total sample volume at 25 °C and 760 mm Hg pressure, L.
Vm = total sample volume at measured temperature and pressure, L.
PA= average ambient pressure, mm Hg.
Ty= average ambient temperature, °C.
11. Sample Analysis
11.1 Sample Preparation
11.1.1 The samples (trip blank, field blank and field samples) are returned to the laboratory in a shipping
container and stored in a refrigerator at (<4°C) until analysis. Alternatively, the samples may also be stored alone
in their individual containers.
11.1.2 The time between sampling and extraction should not exceed 2 weeks. Since background levels in the
cartridges may change due to adsorption during storage, always compare field samples to their associated field and
trip blank samples, stored under the same conditions.
11.2 Sample Extraction
[Note: Beware ofunintentional exposure of samplers andeluted samples to aldehyde andketone sources. Laboratory
air often holds high concentrations of acetone. Labelinginks, adhesives, and packaging containers (includingvials
with plastic caps) are all possible sources on contamination.]
[Note: Contamination is most likely to occur during sample extraction. Before eluting derivatives, clean all
glassware by rinsing with acetonitrile, then heating in a 60° C vacuum oven for at least 30 minutes. Eluting the samples
in a nitrogen-purged glove bag further reduces the risk of contamination.
The acetonitrile used to elute the DNPH derivatives is a typical source of contamination. Formaldehyde-free
acetonitrile used to elute samples should be used onlyfor this purpose, and stored in a carbonylfree environment.
A concentration of 10 fig/L of any aldehyde or ketone in the acetonitrile adds 0.05 fig of that carbonyl to sample blank
values if using 5 mL extraction volumes.]
11.2.1 Remove the sample cartridge from the labeled shipping tube or container. Connect the sample cartridge to
a clean syringe.(Some commerical cartridges do not require the addition of a syringe for elution.)
[Note: The liquid flow during desorption should be in the reverse direction of air flow during sample collection.]
January 1999
Compendium of Methods for Toxic Organic Air Pollutants
Page 11A-21
-------
Method TO-11A
Formaldehyde
11.2.2 Place the sample cartridge syringe in the syringe rack (see Figure 7).
[Note: If the two beds in the SKC tube are being recovered separatelyfor breakthrough studies, break the tube and
place the beds in separate vials. Add exactly 5 mL of acetonitrile to each vial. Proceed with recovery, as specified
in Section 11.2.4 through Section 11.2.5. Particulate in the relatively small number of samples used in the
breakthrough studies should not adversely impact the sample valve or back pressure.]
11.2.3 Backflush the cartridge (gravity feed) by passing 5 mL of acetonitrile from the syringe through the
cartridge to a 5-mL volumetric flask. The backflush elution approach may add particulate particles also collected on
the cartridge to the acetonitrile solution which can cause sample valve failure and increase column back pressure.
To minimize this, frontflush the cartridge contents with the acetonitrile reagent rather than blackflush. The use of
5mL of acetonitrile is sufficient for quantitative cartridge sample elution in either mode.
[Note: A dry cartridge has an acetonitrile holdup volume of about 0.3 mL. The eluant flow may stop before the
acetonitrile in the syringe is completely drained into the cartridge because of air trapped between the cartridge
filter and the syringe Luer® tip. If this happens, displace the trapped air with the acetonitrile in the syringe using
a long-tip disposable Pasteurpipet.]
11.2.4 Dilute to the 5-mL mark with acetonitrile. Label the flask with sample identification. Store in
refrigerated conditions until the sample is analyzed by HPLC. Pipet two aliquots into sample vials with TFE-
fluorocarbon-lined septa. Analyze the first aliquot for the derivative carbonyls by HPLC. Store the second aliquot
in the refrigerator until the results of the analysis of the first aliquot are complete and validated. The second
aliquot can be used for confirmatory analysis, if necessary.
11.2.5 Sample eluates are stable at 4°C for up to one month.
11.3 HPLC Analysis
11.3.1 The HPLC system is assembled and calibrated as described in Section 11.4. The operating
parameters are as follows when formaldehyde is the only carbonyl of interest:
Column Zorbax ODS (4.6-mm ID x 25-cm), or equivalent.
Mobile Phase:
Detector:
Flow Rate:
Retention Time:
Sample Injection Volume:
60% acetonitrile/40% water, isocratic.
ultraviolet, operating at 360 nm.
1.0 mL/min.
7 minutes for formaldehyde with one Zorbax ODS column. Thirteen minutes for
formaldehyde with two Zorbax ODS columns.
25 |iL.
Before each analysis, the detector baseline is checked to ensure stable conditions.
11.3.2 The HPLC mobile phase is prepared by mixing 600 mL of acetonitrile and 400 mL of water. This mixture is
filtered through a 0.22-/iin polyester membrane filter in an all-glass and Teflon® suction filtration apparatus. The
filteredmobile phase is degassed by purgingwithheliumfor 10-15 minutes(100 mL/min)or by heating to 60°C for 5-10
minutes in an Erlenmeyer flask covered with a watch glass. A constant back pressure restrictor (350 kPa) or short
length (15-30 cm) of 0.25-mm (0.01 inch) ID Teflon® tubing should be placed after the detector to eliminate further
mobile phase outgassing.
Page 11A-22
Compendium of Methods for Toxic Organic Air Pollutants
January 1999
-------
Formaldehyde
Method TO-11A
11.3.3 The mobile phase is placed in the HPLC solvent reservoir and the pump is set at a flow rate of 1.0 mL/min
and allowed to pump for 20-30 minutes before the first analysis. The detector is switched on at least 30 minutes before
the first analysis, and the detector output is displayed on a strip chart recorder or similar output device. The
isocratic flow of 60% acetonitrile/40% water is adequate for the analysis of formaldehyde; however, sufficient time
between air sample analyses is required to assure that all other carbonyl compounds are eluted from the HPLC column
prior to the next sample. The gradient flow approach,mentioned later (see Section 14.3) is properly programmed to
elute other carbonyl compounds.
11.3.4 A 100-joL aliquot of the sample is drawn into a clean HPLC injection syringe. The sample injection loop
(25-|iL) is loaded and an injection is made. The data system, if available, is activated simultaneously with the
injection. If a strip chart recorder is used, mark the point of injection on the chart paper.
11.3.5 After approximately one minute, the injection valve is returned to the "load" position and the syringe and
valve are rinsed or flushed with acetonitrile/water mixture in preparation for the next sample analysis.
[Note: The flush/rinse solvent should not pass through the sample loop during flushing.]
The loop is cleaned while the valve is in the "load" mode.
11.3.6 After elution of the DNPH-formaldehyde derivative (see Figure 10), data acquisition is terminated and the
component concentrations are calculated as described in Section 12.
11.3.7 After a stable baseline is achieved, the system can be used for further sample analyses as described above.
Be sure to examinethechromatogram closely toensurethatbackgroundDNPH-formaldehydederivativepeaks are not on
the solvent slope of the DNPH peak.
[Note: After several cartridge analyses, background buildup on the column may be removed byflushingwith several
column volumes of 100% acetonitrile.]
11.3.8 If the concentration of analyte exceeds the linear range of the instrument, the sample should be diluted
with mobile phase, or a smaller volume can be injected into the HPLC.
11.3.9 If the retention time is not duplicated (±10%), the acetonitrile/water ratio may be increased or
decreased to obtain the correct elution time. If the elution time is too long, increase the ratio; if it is too short,
decrease the ratio. If retention time is not reproducing, the problem may be associated with the HPLC flow system. A
control chart is recommended to evaluate retention time changes.
[Note: The chromatographic conditions described here have been optimizedfor the detection of formaldehyde. Analysts
are advised to experiment with their HPLC system to optimize chromatographic conditions for their particular
analytical needs. If a solvent change is necessary, always recalibrate before running samples.]
11.4 HPLC Calibration
11.4.1 Calibration standards can be prepared by the user in acetonitrile from the solid DNPH-formaldehyde
derivative or liquid standards can be purchased from various manufacturers. From the solid compound, individual stock
solutions of 100 ug/mL are prepared by dissolving 10 mg of solid derivative in 100 mL of acetronitrile. Since the MW
ofHCHO-hydrazone is 210 g/mol, and the MW ofHCHO is 30 g/mol, the stock solution concentration converts to 14.3 ug/mL
as formaldehyde (30/210 x lOOmg/mL). The solid compound is weighed using a 5-place analytical balance and liquid
dilutions are made with volumetric glassware. Stock solutions obtained from commercial suppliers generally range from
1 to 50 ug/mL as the carbonyl compound. These stock solutions are typically provided in 1 mL ampules.
11.4.2 Using the stock solution, working calibration standards are produced. To generate the highest
concentration working standard, use a pipette to quantitatively transfer 1.00 ml of the stock solution to a 25 mL
volumetric flask. For example, using a 14.3 ug/mL stock solution produces a working standard solution of570 ng/mL
January 1999
Compendium of Methods for Toxic Organic Air Pollutants
Page 11 A-23
-------
Method TO-11A
Formaldehyde
( 14300 ng/mL x 1/25 ). The high concentration working standard diluted serially, using 1 to 5 mL pipettes and
volumetric flasks, can produce working standards ranging between 28.5 and 570 ng/mL.
11.4.3 Each calibration standard (at least five levels) is analyzed three times and area response is tabulated
against mass concentration injected (see Figure 11). All calibration runs are performed as described for sample
analyses in Section 11.3. The results are used to prepare a calibration curve, as illustrated in Figure 12. The slope
of the calibration curve gives the response factor, RF. Linear response is indicated where a correlation coefficient
of at least 0.999 for a linear least-squares fit of the data (mass concentration versus area response) is obtained.
The intercept of the calibration curve should pass through the origin. If it does not, check your reagents and standard
solutions preparation procedure for possible contamination. If the calibration curve does not pass through the origin,
the equation for the calibration curve should include the intercept.
11.4.4 Each new calibration curve should be verified by analyzing a standard prepared from material obtained from
a second source. This standard should show arecovery of 85 to 115%. Ifnot, corrective action is required to eliminate
the discrepancy between the two sources of the standard material.
11.4.5 Once linear response has been documented, a concentration standard near the anticipated levels of each
carbonyl component, but at least 10 times the detection limit, should be chosen for daily calibration. The day to day
response for the various components should be within 10% of the calibration value. If greater variability is observed,
prepare a fresh calibration check standard. If the variability using a freshly prepared calibration check standard
is greater than 15%, a new calibration curve must be developed from fresh standards. A plot of the daily values on a
Quality Control Chart (day versus concentration) is helpful to check for long term drift of the concentration value.
11.4.6 The response for each component in the daily calibration standard is used to calculate a response factor
according to the following equation shown for formaldehyde:
RF - ^
HCHO
r
HCHO
where:
RFhcho = response factor for formaldehyde given as area counts per ng/mL.
CHCho = concentration of analyte in the calibration standard in units of ng/mL.
P = peak area counts for the formaldehyde standard.
PQ = calibration curve intercept; in most cases this is zero.
11.4.7 The RF for each carbonyl compound is determined in the same way as that given for formaldehyde. The
concentration of HCHO and other carbonyl compounds is determined with the calibration curves for each component in
the analyzed sample. Example calculation for HCHO is given in section 12.
12. Calculations
Determination of the carbonyl compound air concentration requires three steps: (1) determination of the average
blank and the standard deviation of the blank; (2) determination of the collected carbonyl compound mass of the
cartridge; (3) calculation of the carbonyl compound air concentration. The following discusion provides these steps
for formaldehyde.
12.1 Blank Determination
Since the blank level for any arbitary cartridge is unknown, an average value for the blank is used in the
calculation. As noted earlier, the average blank value is determined for each lot of cartridges. For a given lot size,
N, a minimum of /N cartridge blanks (rounded to the next whole number) should be analyzed; i.e., for a lot size of
Page 11 A-24
Compendium of Methods for Toxic Organic Air Pollutants
January 1999
-------
Formaldehyde
Method TO-11A
200, a minimum of /200 or 14 cartridge blanks should be analyzed. A minimum of 3 of these blanks are used for the
Certification Blank, and the remaining 11 are used for field blanks. The mass of HCHO on each cartridge is determined
by multiplying the observed peak area for blank cartridge solution by the acetonitrile extract volume (typically 5
mL ) and dividing by the response factor as provided in the following equation:
^BL-HCHOj X ^
^BL-HCHOj =
HCHO
where:
MBL-HCHOi = the blank HCHO mass for cartridge , i.
RFhcho = HCHO response factor calculated in Section 11.4.5.
PBL-HCHOi = area counts for HCHO in blank sample extract.
VE =extract volume in mL (usually 5 mL).
Once all blank cartridges have been measured, the average blank value is determined by the following equation:
_ | i = 1
^BL-HCHO " TtXX/ ^BL-HCHO,
JN N
where:
M bl-hcho = the average HCHO mass for all cartridges.
MBL-HCHOi = blank HCHO mass for cartridge, i.
N = the number of blank cartridges.
[Note: Measurement of cartridge blanks should be distributed over the period that this particular cartridge lot is
usedfor ambient air sampling. It is recommended that a trend plot of blank results be constructed to evaluate
background carbonyl results over the period of cartridge lot utilization in the sampling program. If significant
drifting is observed, blank average values should be segmented to be more representative of carbonyl background.]
12.2 Carbonyl Analyte Mass
The calculation equation for the mass of the collected carbonyl compounds on an individual cartridge is the same
as that for the cartridge blanks. The gross measured carbonyl mass is determined with an equation analogous to that
given in section 12.1. The equation for formaldehyde is given as:
A i x "^E
SAi RF
HCHO
where:
MSAi = gross HCHO mass for cartridge, i.
PSAl = HCHO peak area counts for cartridge, I.
January 1999
Compendium of Methods for Toxic Organic Air Pollutants
Page 11 A-25
-------
Method TO-11A
Formaldehyde
RFhcho = the response factor for HCHO.
VE = acetonitrile extract volume in mL (typically 5 mL).
The net HCHO mass for an individual cartridge is determined by substracting the average blank value from the gross HCHO
mass obtained for sample i, and is given as:
^HCHOj " ^SAj ~ ^BL-HCHO
12.3 Carbonyl Compound Concentration
The sample air concentration for carbonyl compounds cannot be determined directly from the mass measurement
and requires conversion to units of volume. The conversion calculation for HCHO is determined using the ideal gas law
and is given by the following equation:
Mhcho, /T, ^ , 760
^HCHO i MW- X B ^ X p
AM B
where:
VHCHOi= gas volume of HCHO on cartridge, i.
MHchoi = mass of HCHO on cartridge, i.
MW = molecular weight of HCHO, 30.03 g/mole.
R = gas constant, 0.082 L-atm/mol-deg.
TAMh = ambient air temperature in degrees Kelvin, 273 + T (C°).
Pamb = ambient air pressure in torr.
For an ambient air temperature of 25°C and a pressure of 760 torr, the ideal law equation reduces to:
^HCHO ; = I' 2276 X Mhcho
In this equation, the HCHO mass in ng is converted to a volume in nL. The volume of air that was passed through the
cartridge was measured by either a mass flow controller or dry test meter calibrated at a known temperature and
pressure. To determine HCHO concentration in the units of ppbv, apply the following equation:
V
. HCHO
Chc.hoH** =
HCHO1v
VAI R
Page 11 A-26
Compendium of Methods for Toxic Organic Air Pollutants
January 1999
-------
Formaldehyde
Method TO-11A
where:
VHCHOi= volume of formaldehyde in nL
VMR = volume of sample air through the cartridge
13. Performance Criterial and Quality Assurance
This section summarizes required quality assurance measures and provides guidance concerning performance criteria
that should be achieved within each laboratory.
13.1 Standard Operating Procedures (SOPs).
13.1.1 Users should generate SOPs describing the following activities in their laboratory: (1) assembly,
calibration, and operation of the sampling system, with make and model of equipment used; (2) preparation,
purification, storage, and handling of sampling reagent and samples; (3) assembly, calibration, and operation of the
HPLC system, with make and model of equipment used; and (4) all aspects of data recording and processing including
lists of computer hardware and software used.
13.1.2 SOPs should provide specific stepwise instructions and should be readily available to and understood by
the laboratory personnel conducting the work.
13.2 HPLC System Performance
13.2.1 The general appearance of the HPLC system should be similar to that illustrated in Figure 1.
13.2.2 HPLC system efficiency is calculated according to the following equation:
N= 5.54
W
1/2/
where:
N = column efficiency, theoretical plates.
tr = retention time of analyte, seconds.
W1/2 = width of component peak at half height, seconds.
A column efficiency of >5,000 theoretical plates should be utilized.
13.2.3 Precision of response for replicate HPLC injections should be ±10% or less, day to day, for analyte
calibration standards at 150 ng/mL or greater levels (as the carbonyl compound). At 75 ng/mL levels and below,
precision of replicate analyses could vary up to 25%. Precision of retention times should be ±7% on a given day.
13.3 Process Blanks
13.3.1 At least one field blank should be used for each day of field sampling, shipped and analyzed with each group
of samples. The number of samples within a group and/or time frame should be recorded so that a specified minimum number
of blanks is obtained for a given cartridge lot used for field samples. The field blank is treated identically to the
samples except that no air is drawn through the cartridge. The performance criteria described in Section 9.2 should
be met for field blanks. It is also desirable to analyze trip and laboratory blank cartridges as well, to distinguish
between possible field and lab contamination.
January 1999
Compendium of Methods for Toxic Organic Air Pollutants
Page 11A-27
-------
Method TO-11A
Formaldehyde
[Note: Remember to use the field blank value for each cartridge lot when calculating concentration. Do not mix
cartridge lots in the blank value determinations ]
13.4 Method Precision and Accuracy
13.4.1 At least 50% of the sampling events should include a collocated sample. A collocated sample is defined as
a second sampling port off the common sampling manifold. If more than five samples are collected per sampling event,
a collocated sample should be collected for each sampling event. Precision for the collocated samples should be ±20%
or better. EPA historical data has demonstrated effectiveness in reaching ±20%, as illustrated in Figure 13.
13.4.2 Precision for replicate HPLC injections should be ±10% or better, day to day, for calibration standards.
13.4.3 Cartridges spiked with analytes of interest can be used in round-robin studies to intercompare several
laboratories performing carbonyl analyses. The spiked samples are prepared in the laboratory by spiking a blank
cartridge with a solution of derivatized carbonyls in acetonitrile. The laboratory preparing the spike samples should
analyze at a minimum 3 of the prepared spiked samples to evaluate the consistency of prepared samples.
13.4.4 Before initial use of the method, each laboratory should generate triplicate spiked samples at a minimum
of three concentration levels, bracketing the range of interest for each compound. Triplicate nonspiked samples must
also be processed. Spike recoveries of >80 ±10% and blank levels should be achieved.
13.4.5 For ambient air sampling, an ozone denuder must be used as part of the sampling system. As discussed in
Section 6.4, ozone effects the ultimate method precision and accuracy by reacting with its carbonyl derivative
(hydrazones) on the cartridge. To illustrate this point, Figure 14 documents the concentration of formaldehyde
captured on collocated DNPH-cartridges, one with a denuder (see Figure 14a) and the other without a denuder (see
Figure 14b). The formaldehyde peak is considerably higher with use of an ozone denuder.
13.5 Method Detection Limits
13.5.1 Determine method detection limits using the procedures in 40 CFR Part 136B. Prepare a low level standard
of the carbonyl derivatives at a concentration within two to five times the estimated method detection limit. Inject
the standard into the analytical system seven times.
13.5.2 Calculate the measured concentration using the calibration curve.
13.5.3 Determine the standard deviation for the seven analyses and use the standard deviation to calculate the
detection limit as described in 40 CFR Part 136B.
13.6 General QA/QC Requirements
13.6.1 General QA/QC requirements associated with the performance of Compendium Method TO-11A include:
Sampling
• Each sampling event, flow calibration with bubble meter, both pre- and post-checks.
• Mass flow meter calibration factor determined every quarter.
• Each sampling event, leak check, both pre- and post-checks.
• 10 percent of field samples collocated to help calculate method precision and evaluate biases.
• 10 percent of field samples operated with back-up cartridge to evaluate analyte breakthrough.
• Field and trip (optional) blank cartridges are included with each field sample collection program.
• Sample volumes calculated and reviewed project QA officer.
Reagents
• Coating solution prepared from concentrated stock solution immediately before each coating.
Page 11A-28
Compendium of Methods for Toxic Organic Air Pollutants
January 1999
-------
Formaldehyde
Method TO-11A
• Solution analyzed before each coating to determine acceptability (less than 0.15 (ig/cartridge for each
aldehyde), control chart of contaminant concentration maintained.
• Three blank cartridges per lot for immediate elution/analysis to determine Certification Blank for the
carbonyl compounds.
Analysis
• Multi point calibration curve performed each six months.
• Each initial calibration verified with a standard from a second source.
• Continuing calibration standard (mid-level) analyses every analytical run to evaluate precision, peak
resolution and retention time drift.
• Method detection limits (MDLs) verified annually or after each instrument change.
• Replicate analysis of approximately 10 percent of sample eluents to evaluate precision.
• Samples quantitated against least squares calibration line.
• Performance evaluation (PE) sample acquired from independent sources analyzed prior to and after field
samples.
• Randomcollocatedsamplesshippedtoindependentlaboratoryforanalysisandcomparedtoin-housecollocated
sample.
• Testing of acetonitrile used for sample extraction for background carbonyl evaluation.
Data Acquisition
• Sample chromatograms and standards checked daily for peak shape and integration quality, resolution of
carbonyls, overall sensitivity and retention time drift.
• Separate tape backups made of raw data immediately after completion of each analysis.
• Peaks in each sample checked for correct ID and integration using system software before export to ASCII
file.
• Final results checked and edited by project QA officer before producing final report.
• Tape backups of final data files produced.
13.6.2 All results should be reviewed by the project QA officer, independent of the field and laboratory
operations, to evaluate the overall adherence to the methodology in meeting the program data quality objectives
(DQOs).
14. Detection of Other Aldehydes and Ketones
14.1 Introduction
14.1.1 The procedure outlined above has been written specifically for the sampling and analysis of formaldehyde
in ambient air using an adsorbent cartridge and HPLC. Ambient air contains other aldehydes and ketones. Optimizing
chromatographic conditions by using two Zorbax ODS columns in series and varying the mobile phase composition through
a gradient program will enable the analysis of other aldehydes and ketones. Alternatively, other aldehydes and ketones
may also be analyzed using a single C-18, reverse phase column and a ternary gradient as described by Waters or Smith,
ctal. (./ Chromatography, 483,1989,431-436). Thus, other aldehydes and ketones can be detected with a modification
of the basic procedure.
14.1.2 In particular, chromatographic conditions can be optimized to separate acetaldehyde, acetone,
propionaldehyde, and some higher molecular weight carbonyls within an analysis time of about 1 h by utilizing two
Zorbax ODS columns in series, and a linear mobile phase program. Operating the HPLC in a gradient mode with one Zorbax
ODS columnmay also provide adequateresolutionandseparation. Carbonyl compounds covered within the scope ofthis
modification include:
January 1999
Compendium of Methods for Toxic Organic Air Pollutants
Page 11 A-29
-------
Method TO-11A
Formaldehyde
Formaldehyde Crotonaldehyde
o-Tolualdehyde
Aceteldehyde Butyraldehyde
m-Tolualdehyde
Acetone Benzaldehyde
/p-Tolualdehyde
Propionaldehdye Isovaleraldehyde
Hexanaldehyde
Valeraldehyde 2,5-Dimethylbenzaldehyde Methyl ethyl ketone
14.1.3 The linear gradientprogramvariesthemobilephase composition periodically to achieve maximumresolution
ofthe C-3, C-4 andbenzaldehyde region of the chromatogram. The following gradientprogram was found to be adequate
to achieve this goal: Upon sample injection, linear gradient from 65% acetonitrile (ACN)/35% water to 55% ACN/45%
waterin36min;to 100%ACNin20min; 100%ACNfor5min; reverse linear gradient from 100%ACNto60%ACN/40%water
in 1 min; maintain at 60% ACN/40% water for 15 min.
14.2 Sampling Procedures
Same as Section 10.
14.3 HPLC Analysis
14.3.1 The HPLC system is assembled and calibrated as described in Section 11. The operating parameters are as
follows:
Column: Zorbax ODS, two columns in series
Mobile Phase: Acetonitrile/water, linear gradient
Step 1. 60-75% acetonitrile/40-25% water in 30 minutes.
Step 2. 75-100% acetonitrile/25-0% water in 20 minutes.
Step 3. 100% acetonitrile for 5 minutes.
Step 4. 60% acetonitrile/40% water reverse gradient in 1 minute.
Step 5. 60% acetonitrile/40% water, isocratic, for 15 minutes.
Detector: Ultraviolet, operating at 360 nm
Flow Rate: l.OmL/min
Sample Injection Volume: 25 (iL
14.3.2 The gradientprogram allows for optimization of chromatographic conditions to separate acetaldehyde,
acetone, propionaldehyde, and other higher molecular weight aldehydes and ketones in an analysis time of about one
hour.
14.3.3 ThechromatographicconditionsdescribedherehavebeenoptimizedforagradientHPLC system equipped
withaUV detector (variable wavelength), an automatic sampler with a 25-|llL loop injector and two DuPontZorbax OD S
columns (4.6 x 250-mm), a recorder, and an electronic integrator. Analysts are advised to experiment with their HPLC
systems to optimize chromatographic conditions for their particular analytical needs. Highest chromatographic
resolution and sensitivity are desirable but may not be achieved. The separation of acetaldehyde, acetone, and
propionaldehyde should be a minimum goal of the optimization.
14.3.4 The carbonyl compounds in the sample are identified and quantified by comparing their retention times
and area counts with those of standard DNPH derivatives. Formaldehyde, acetaldehyde, acetone, propionaldehyde,
crotonaldehyde, benzaldehyde, and o-, m-, p-tolualdehydes can be identified with a high degree of confidence. The
Page 11A-30
Compendium of Methods for Toxic Organic Air Pollutants
January 1999
-------
Formaldehyde
Method TO-11A
identification of butyraldehyde is less certain because it coelutes with isobutyraldehyde and is only partially
resolved from methyl ethyl ketone under the stated chromatographic conditions. A typical chromatogram obtained with
the gradient HPLC system for detection of other aldehydes and ketones is illustrated in Figure 15.
14.3.5 The concentrations of individual carbonyl compounds are determined as outlined in Section 12.
14.3.6 Performance criteria and quality assurance activities should meet those requirements outlined in
Section 13.
15. Precision and Bias
15.1 This test method has been evaluated by round robin testing. It has also been used by two different laboratories
for analysis of over 1,500measurements of formaldehyde and other aldehydes in ambient air for EPA's Urban Air Toxics
Program (UATP), conducted in 14 cities throughout the United States.
15.2 The precision of 45 replicate HPLC injections of a stock solution of formaldehyde-DNPH derivative over a 2-month
period has been shown to be 0.85% relative standard deviation (RSD).
15.3 Triplicate analyses of each of twelve identical samples of exposed DNPH cartridges provided formaldehyde
measurements that agreed within 10.9% RSD.
15.4 A total of 16 laboratories in the U.S., Canada, and Europe participated in a round robin test that included 250
blank DNPH-cartridges, three sets of 30 cartridges spiked at three levels with DNPH derivatives, and 13 sets of
cartridges exposed to diluted automobile exhaust gas. All round robin samples were randomly distributed to the
participating laboratories. A summary of the round robin results is shown in Table 4.
15.5 The absolute percent differences between collocated duplicate sample sets from the 1988 UATP program were 11.8%
for formaldehyde (/?=405). 14.5% for acetaldehyde (w=386), and 16.7% for acetone (w=346).
15.6 Collocatedduplicatesamplescollectedinthel989UATPprogramandanalyzedbyadifferentlaboratory showed
a mean RSD of 0.07, correlation coefficient of 0.98, and bias of -0.05 for formaldehyde. Corresponding values for
acetaldehyde were 0.12,0.95 and -0.54, respectively. In the 1988 UATP program, single laboratory analyses of spiked
DNPH cartridges provided over the year showed an average bias of +6.2% for formaldehyde (n=14) and +13.8% for
acetaldehyde (/?= 13).
15.7 Single laboratory analyses of 30 spiked DNPH cartridges during the 1989 UATP program showed an average bias of
+1.0% (range -49 to +28%) for formaldehyde and 5.1% (range -38% to +39%) for acetaldehyde.
16. References
1. Riggin,R M., "DeterminationofAldehydesandKetonesinAmbientAir: Method TO-5," in Compendium ofMethods
for the Determination ofTjxic Organic Compounds in Ambient Air, U. S. Environmental Protection Agency, EPA-
600/4-84-041, Research Triangle Park, NC, April 1984.
2. Winberry, W. T. Jr., et al., "Determination of Formaldehyde and Other Aldehydes in Indoor Air Using Passive
Sampling Device, Method IP-6C," in Compendium ofMethods for the Determination of Air Pollutants in Indoor Air,
U. S. Environmental Protection Agency, EPA-600/4-90-010, May 1990.
3. Tejada,S.B., "StandardOperatingProcedureForDNPH-coatedSilicaCartridgesForSamplingCarbonylCompounds
In Air AndAnalysis by High-performance Liquid Chromatography, "Unpublished, U. S. Environmental Protection
Agency, Research Triangle Park, NC, March 1986.
January 1999
Compendium ofMethods for Toxic Organic Air Pollutants
Page 11A-31
-------
Method TO-11A
Formaldehyde
4. Tejada, S.B., "Evaluation of Silica Gel Cartridges Coated in situ with Acidified 2,4-Dinitrophenylhydrazine
for Sampling Aldehydes and Ketones in Air," Intern. J. Environ. Anal. Chem., Vol. 26:167-185, 1986.
5. Winberry,W. T. Jr.,etal., "DeterminationofFormaldehydeinAmbientAirUsingAdsorbentCartridgeFollowed
by HPLC: Method TO-11," in Compendium ofMethodsfor the Determination ofTjxic Organic Compounds in Ambient
Air, Second Supplement, U. S. Environmental Protection Agency, EPA-600/4-89-018, Research Triangle Park, NC,
March 1989.
6. Winberry, W. T. Jr., et al., "Determination of Formaldehyde and Other Aldehydes in Indoor Air Using a Solid
Adsorbent Cartridge: Method IP-6A," in (bmpendium ofMethodsfor the Determination of Air Pollutants in Indoor
Air, U. S. Environmental Protection Agency, EPA-600/4-90-010, May 1990.
7. Nolan,L.,etal.,"MonitoringCarbonylsinAmbientAirUsingtheNewSupelclean™LPD(Low Pressure Drop) DNPH
Caxtiidge,"mProceedingsofthel995EPA/AWMAInternationalSymposiumonMeasurementofToxicandRelated
Air Pollutants, VIP-50, pp. 279, May 1995.
8. Quality Assurance Handbookfor Air Pollution Measurement Systems, Volume II-Ambient Air Specific Methods,
EPA-600/R-94-038b, U. S. Environmental Protection Agency, Research Triangle Park, NC, May 1994.
9. Technical Assistance Document for Sampling and Analysis ofO precursors,\J. S. Environmental Protection
Agency, EPA-600/8-9-215, Research Triangle Park, NC, October 1991.
10. Ahonen, I., Priha, E., and Aijala, M-L, "Specificity of Analytical Methods Used to Determine the Concentration
of Formaldehyde in Workroom Air," Chemosphere, Vol. 13:521-525, 1984.
11. Levin, J. O., et al., "Determination of Sub-part-per-Million Levels of Formaldehyde in Air Using Active or
Passive Sampling on 2,4-Dinitrophenylhydrazine-Coated Glass Fiber Filters and High-Performance Liquid
Chromatography," Anal. Chem.,Nol. 57:1032-1035, 1985.
12. Perez, J. M., Lipari, F., and Seizinger, D. E., "Cooperative Development of Analytical Methods for Diesel
Emissions and Particulates - Solvent Extractions, Aldehydes and Sulfate Methods", presented at the Society of
Automotive Engineers International Congress and Exposition, Detroit, MI, February-March 1984.
13. Kring, E. V., et al., "Sampling for Formaldehyde in Workplace and Ambient Air Environments-Additional
Laboratory Validation and Field Verification of a Passive Air Monitoring Device Compared with Conventional
Sampling Methods," J. Am. Ind. Hyg. Assoc., Vol. 45:318-324, 1984.
14. Siiju, A., and Shepson, P. B., "Laboratory and Field Evaluation of the DNPH-Cartridge Technique for the
Measurement of Atmospheric Carbonyl Compounds," Environ. Sci. Technol. Vol. 29:384-392, 1995.
15. Chasz, E. et al., "Philadelphia Air Management Lab, Summary of Procedures and Analytical Data for Enhanced
Ambient Monitoring of PAMS Carbonyls," in Proceedings ofthe 1995 EPA/AWMA International Symposium on
Measurement of Toxic and Related Air Pollutants, VIP-50, pp. 293, May 1995.
16. Grosjean,D.,"AmbientLevelsofFormaldehyde,Acetaldehyde,andFormicAcidinSouthernCalifornia: Results
of a One-Year Base-Line Study," Environ. Sci. Technol., Vol. 25, 710-715, 1991.
Page 11 A-32
Compendium ofMethods for Toxic Organic Air Pollutants
January 1999
-------
Formaldehyde
Method TO-11A
17. Bufalini, J. J., and Brubaker, K.L., "The Photooxidation of Formaldehyde at Low Pressures." In: Chemical
Reaction in Urban Atmospheres, (C.S. Tuesday). American Elsevier Publishing Co.. New York. pp. 225-240.1971.
18. Altshuller, A.P., and Cohen, I.R., "Photooxidation of Formaldehyde in the Pressence of Aliphatic Aldehydes",
Science, Vol. 7:1043-1049, 1963.
19. "Formaldehyde and Other Aldehydes" Committee on Aldehydes, Board of Toxicology and Environmental Hazards,
National Research Council, National Academy Press, Washington, DC, 1981.
20. Altshuller, A. P., "Production of Aldehydes as Primary Emissions and Secondary Atmospheric Reactions of Alkenes
and Alkanes During the Night and Early Morning Hours," Atmos. Environ., Vol. 27A:21-31, 1993.
21. Tanner, R L., et al., "Atmospheric Chemistry of Aldehydes; Enhanced PAN Formation From Ethanol Fuel Vehicles,"
Environ. Sci. Technol., Vol. 22:1026-1034, 1988.
22. Ciccioli, P., and Cecinato, A., "Advanced methods for the Evaluation of Atmospheric Pollutants Relevant to
Photochemical Smog and Dry Acid Deposition: Chapter 11" in Gaseous Pollutants: Characterization and Cycling,
edited by Jerome O. Nriagu, ISBN 0-471-54898-7, John Wiley and Sons, Inc., 1992.
23. Parmar, S. S., et al.," A Study of Ozone Interferences in Carbonyl Monitoring Using DNPH Coated C, 8 and Silica
Cartridges," inProceedingsofthel995EPA/AWMAInternationalSymposiumonMeasurementofToxicandRelated
Air Pollutants, VIP-50, pp. 306, May 1995.
24. Parmar, S. S., et al., "Effect of Acidity on the Sampling and analysis of Carbonyls Using DNPH Derivatization
Method,"mProceedingsofthel996EPA/AWMAInternationalSymposiumonMeasurementqfToxicandRelatedAir
Pollutants, VIP-64, pp. 311, May 1996.
25. Arnts, R. R, and Tejada, S. B., "2,4-Dinitrophenylhydrazine-Coated Silica Gel Cartridge Method for
Determination of Formaldehyde in Air: Identification of an Ozone Interference," Environ. Sci. Technol. Vol.
23:1428-1430, 1989.
26. Kleindienst, T. E., etal., "MeasurementofC1-C4 Carbonyls onDNPH-Coated Silica Gel and C partridges in the
PresznceofOzone,"mProceedingsofthel995EPA/AWMAInternationalSymposiumonMeasurementofToxicand
Related Air Pollutants, VIP-50, pp 29T, May 1995.
January 1999
Compendium of Methods for Toxic Organic Air Pollutants
Page 11 A-33
-------
Method TO-11A
Formaldehyde
TABLE 1. COM
PARISON OF DIN
fPH COATED CARTRIDGES: SILICA GEL VS. CI8
Topic
Comparison
Discussion
Background
Silica gel < C18
Silica gel is purer, therefore less background contamination from acetc
and formaldehyde as compared to CI8.
Breakthrough
Silica gel < C18
C18 allows carbonyl compounds to breakthrough easier with longer se
periods, thus causing bias results. C18 has a lower capacity for
carbonyls in general. Loading of DNPH on C18 plays an important ro
breakthrough for carbonyls.
Ozone interference
Silica gel C18
Ozone interference with silica gel is documented. Ozone interference
with C18 is not clear at this time. Therefore, must use denuder with b<
systems.
Extraneous chromato-
graphic peaks
Silica gel C18
Researchers have detected extraneous peaks in the chromatography of
C18 and silica gel when ozone is present.
TABLE 2. TY]
PICAL DNPH-CARTRIDGE SPECIFICATIONS
CaU'jion
T>pical Specifications
Adsorbent
chromatographic grade silica or C18 coated with 2,4-dinitrophenylhydrazine (I
Particle size
150-1000 (jm (60/100 mesh to 18/35 mesh)
DNPH loading1
0.3-0.9% (~l-3 mg/cartridge)
Bed weight2
approx. 350 mg
Capacity
approx. 75 jog formaldehyde, assuming a 50% consumption of DNPH
Background
(per cartridge)
<0.15 (ig formaldehyde
<0.10 (ag acetaldehyde
<0.10 ng other carbonyls
<0.30 ng acetone
Pressure drop
7 inches of water @0.5 L/min
15 inches of water @1.0 L/min
37 inches of water @ 2.0 L/min
Sampling temperature
10°C to 100°C
Collection efficiency
>95% for formaldehyde for sampling rates up to 2.0 L/min
Solvent hold-up volume
-1.0 mL
Tube dimensions
From ~2 inches to ~5 inches in length
~1 inch O.D. at widest point
'Loading is variable among commercial suppliers.
2The SKC tube is a dual bed cartridge with 300 mg of DNPH-coated silica gel in the front bed and 150 mg of
DNPH-coated silica gel in the back bed.
Page 11 A-34
Compendium of Methods for Toxic Organic Air Pollutants
January 1999
-------
Formaldehyde
Method TO-11A
TABLE 3. EQUIVALENT FORMALDEHYDE CONCENTRATION (ppbv) RELATED TO
BACKGROUND FORMALDEHYDE CONCENTRATION (ng/cartridge)
Kqiiivalenl formaldehyde concentration (ppbv]
Sample volume. 1.
60
120
180
1440
formaldehyde cartridge concentration
ng/cartridge 70
100
150
0.950
0.475
0.317
0.040
1.358
0.679
0.453
0.057
2.037
1.018
0.679
0.085
TABLE 4. ROUND ROBIN TEST RESULTS3
Sample Type
l-'ormaldclmlc
Acelaldehvde
IVopionaldchydc
licnzaldclmlc
Blank cartridges:
|ig aldehyde
(% RSD)
n
0.13
46
33
0.18
70
33
0.12
47
23
0.06
44
8
Spikedb cartridges:
% recovery (% RSE
low
medium
high
n
)
89.0 (6.02)
97.2 (3.56)
97.5 (2.15)
12
92.6(13.8)
97.8 (7.98)
102.2 (6.93)
13
108.7(32.6)
100.9(13.2)
100.1 (6.77)
12
114.7(36.1)
123.5 (10.4)
120.0(8.21)
14
Exhaust samples:
|ig aldehyde
% RSD
n
5.926
12.6
31
7.990
16.54
32
0.522
26.4
32
0.288
19.4
17
"Sixteen participating laboratories. Statistics shown after removal of outliers.
bNormal spiking levels were approximately 0.5, 5 and 10 |ig of aldehyde, designated as low, medium, and high in this table.
January 1999
Compendium of Methods for Toxic Organic Air Pollutants
Page 11 A-35
-------
Method TO-11A
Formaldehyde
INJECTION VALVE
WASTE
SAMPLE LOOP
GUARD COLUMN
AUTOSAMPLER
ANALYTICAL
COLUMN
ACN
WATER
DETECTOR
360 nm
RECORDER
INTEGRATOR
WASTE
Figure 1. Basic high-performance liquid chromatographic (HPLC) system used for carbonyl analysis.
Page 11 A-36
Compendium of Methods for Toxic Organic Air Pollutants
January 1999
-------
Formaldehyde
Method TO-11A
SKC
Precision-
Sealed Tips
Glass
Tube
High-Purity
Glass Wool
DNPH-Coated Adsorbent
(Back Bed)
DNPH-Coated Adsorbent
(Front Bed)
Precision
Lockspring
NIOSH-Approved
Sealing Caps
SUPELCO Liier®
Fitting
AAC
DNPH-C-18 Cartridge
Luer®
Fitting
DNPH - Coated Adsorbent
Luer®
Fitting
Luer®
Fitting
DNPH- Cartridge
DNPH - Coated Adsorbent
WATERS
DNPH-Silica Gel Cartridge
Luer®
Fitting
DNPH-
Luer®
Fitting
Cartridge
Figure 2. Example of commercially available DNPH-cartridges.
January 1999
Compendium of Methods for Toxic Organic Air Pollutants
Page 11A-37
-------
Formaldehyde
Method TO-11A
r
Mass Flow
Controller
Sampling Pump
DNPH-Coated
Sampling
Cartridge
o
Shut-off
Valve
Ozone
Denuder or
Scrubber
Flushing Pump
Needle Valve
2
.0
c
ro
T3
0)
ro
CD
X
Ambient Air
Figure 3. Example of configuration of a single-port carbonyl sampler using DNPH-coated cartridges.
Page 11 A-39
Compendium of Methods for Toxic Organic Air Pollutants
January 1999
-------
Method TO-11A
Formaldehyde
Sample Inlet
1
0.5 -1.0 LPM
Heated Sample Inlet
Heated Zone
Ozone ~~u
Denuder r~L
Purge
Pump
Inlet Manifold
DNPH Cartridge
Timer-controlled
Solenoid Valves
Exhaust Manifold
Pump
FC
> 1 LPM
Figure 4. Example of components of an automated multi-port sampler for
carbonyls monitoring using DNPH-coated cartridges.
Page 11A-40
Compendium of Methods for Toxic Organic Air Pollutants
January 1999
-------
Formaldehyde
Method TO-11A
Culture Tube
and Cap
Glass
Wool
DNPH
Adsorbent
Tube
(a) DNPH-cartridge in culture shipping tube
Packing
Foam
DNPH
Adsorbent
Tubes
Polypropylene
Shipping
Container
(b) DNPH-cartridge in polypropylene shipping container
Figure 5. Example of commercially available shipping containers for DNPH cartridges.
January 1999
Compendium of Methods for Toxic Organic Air Pollutants
Page 11A-41
-------
Method TO-11A
Formaldehyde
Heated Cord
Kr -H
3' Potassium
Iodide Coated
1/4" O.D. Copper
Tubing
Top View (cutaway)
(a) Cross-sectional view of EPA's ozone denuder assembly
Potassium Iodide
Cartridge Ozone
Scrubber
U
Female Luer
Polyethylene Frit
Male Luer
(b) Commercially available packed granular potassium iodide (Kl) ozone scrubber
Figure 6. Example of (a) cross-sectional view of EPA's ozone denuder assembly, and
(b) commercially available packed granular potassium iodide (KI) ozone scrubber.
Page 11A-42
Compendium of Methods for Toxic Organic Air Pollutants
January 1999
-------
Formaldehyde
Method TO-11A
10 mL Glass
Syringe
Uncoated
Sample Cartridges
Test Tube
Rack
-------
Method TO-11A
Formaldehyde
0
ozone at input end of denuder
ozone at output end of denuder
20 40 60
Sampling Time (hours)
80
100
Figure 8. Example of capacity of 3' x 0.25" O.D. x 4.6-mm I.D. copper KI ozone
denuder at 2 L/min flow.
Page 11A-44
Compendium of Methods for Toxic Organic Air Pollutants
January 1999
-------
Formaldehyde Method TO-11A
COMPENDIUM METHOD TO-11A
CARBONYL SAMPLING FIELD TEST DATA SHEET
COne Sample per Data Sheet)
I. GENERAL INFORMATION
PROJECT: DATES(S) SAMPLED:
SITE: TIME PERIOD SAMPLED:
LOCATION: OPERATOR:
CALIBRATED BY:
INSTRUMENT MODEL NO.: OZONE DENUDERUSE TIME (Hr):
PUMP SERIAL NO.: HEATED INLET: YES NO
ADSORBENT CARTRIDGE INFORMATION:
Type:
Adsorbent:
Serial Number:
Sample Number:
II. SAMPLING DATA INFORMATION
Start Time: Stop Time:
Time
Dry Gas
Meter
Reading
Rotameter
Reading
Flow Rate.
*QmI./min
Ambient
Temperature.
C
I barometric
Pressure,
mm 11 si
Relative
I Iumiditv. %
Comments
Avg.
* Flow rate from rotameter or soap bubble calibrator (specify which).
Total Volume Data (Vm) (use data from dry gras meter, if available)
Vm = (Final - Initial) Dry Gas Meter Reading, or = L
or
y _ Q\ + Ql + 63 ¦¦¦ Qn x 1 _
m N 1000 x (Sampling Time in Minutes)
III. COMMENTS
Figure 9. Example of Compendium Method TO-11A field test data sheet (FTDS).
January 1999
Compendium of Methods for Toxic Organic Air Pollutants
Page 11A-45
-------
Method TO-11A
Formaldehyde
OPERATING PARAMETERS
HPLC
Column: Zorbax ODS or C-18 RP
Mobile Phase: 60% Acetonitrile/40% Water
Detector: Ultraviolet, operating at 360 nm
Flow Rate: 1 mL/min
Retention Time: ~ 7 minutes for formaldehyde
Sample Injection Volume: 25 L
o
(D
a>
_c
75
w
CO
CO
0)
-Q
as
CO
0 10 20
TIME, min
Figure 10. Example of chromatogram of DNPH-formaldehyde derivative.
Page 11A-46
Compendium of Methods for Toxic Organic Air Pollutants
January 1999
-------
Formaldehyde
Method TO-11A
OPERATING PARAMETERS HPLC
Column: Zorbax ODS or C-18 RP
Mobile Phase: 60% Acetontrile/40% Water
Detector: Ultraviolet, operating at 360 nm
Flow Rate: 1 mL/min
Retention Time: ~ 7 minutes for formaldehyde
Sample Injection Volume: 25 |j.L
Peak
Cone.
jig/mL
Area Counts
a
0.61
226541
b
1.23
452186
c
6.16
2257271
d
12.32
4711408
e
18.48
6053812
e
d
Figure 11. Example of HPLC chromatogram of varying
concentration of DNPH-formaldehyde derivative.
January 1999
Compendium of Methods for Toxic Organic Air Pollutants
Page 11A-47
-------
Method TO-11A
Formaldehyde
o
o
X
I—
O
o
<
LLi
cr
<
§.
00
o
o-
rs
o
o-
C0
o
o
m
o
§
o
o
o-
CM
o
o
CORRELATION COEFFICIENT:
0.9999
OPERATING PARAMETERS
HPLC
Column: Zorbax ODS or C-18 RP
Mobile Phase: 60% Acetonitrile/40% Water
Detector: Ultraviolet, operating at 360 nm
Flow Rate: 1 mL/min
Retention Time: ~ 7 minutes for formaldehyde
Sample Injection Volume: 25 (jL
1
9
r
12
T
15
T
18
DNPH - Formaldehyde Derivative (pg/mL)
Figure 12. Example of calibration curve for formaldehyde.
Page 11A-48
Compendium of Methods for Toxic Organic Air Pollutants
January 1999
-------
Formaldehyde
Method TO-11A
CM
34:
tf
(0
O
+20% precision
Cart. #1
January 1999
Figure 13. Historical data associated with collocated samples for
formaldehyde (ppbv) in establishing 20% precision.
Compendium of Methods for Toxic Organic Air Pollutants
Page 11A-49
-------
Method TO-11A
Formaldehyde
x = Unknown
0 = DNPH
1 = Formaldehyde
2 = Acetaldehyde
3 = Acetone
0 2 4 6 8 10
Time (minutes)
Figure 14. Example of analysis demonstrating DNPH-coated cartridges sampling air with (A) and
without (B) ozone denuders, in the determination of formaldehyde.
Page 11 A-50
Compendium of Methods for Toxic Organic Air Pollutants
January 1999
-------
Formaldehyde
Method TO-11A
5
Q E
3 8
ooooooinooir>ir>ir>oo
¦^¦oooooooLncoT-OT-o
T-ooooomo-sf-srmmmo
o E Q.I cm
T-CSICO'd-lfiCOSCDraOT-NfO'd-lfi
I
0.
o
m
o
o
CO
o
CM
0
T3
"S
C<3
o
T3
cd
O
o
ffi
PLn
£
Q
cd
s
o
Cfl
e
o
+->
T3
2
u_
+-»
O
g
'n
c
s
"E
c3
s-h
,o
LU
1
1-
S-H
0D
O
s-h
Qh
e
• Sh
-3
c3
feb
s-h
c3
c3
o
§
u.
O
ts
G
O
s-h
o
C3
O
•a
H
•rl
S-H
E
January 1999
Compendium of Methods for Toxic Organic Air Pollutants
Page 11A-51
------- |