ASTM D6784-24

This international standard was developed in accordance with internationally recognized principles on standardization established in the Decision on Principles for the
Development of International Standards, Guides and Recommendations issued by the World Trade Organization Technical Barriers to Trade (TBT) Committee.
Designation: D6784 − 24
Standard Test Method for
Elemental, Oxidized, Particle-Bound and Total Mercury in
Flue Gas Generated from Coal-Fired Stationary Sources
(Ontario Hydro Method)
This standard is issued under the fixed designation D6784; the number immediately following the designation indicates the year of
original adoption or, in the case of revision, the year of last revision. A number in parentheses indicates the year of last reapproval. A
superscript epsilon (´) indicates an editorial change since the last revision or reapproval.
1. Scope serious medical issues. Mercury, or its vapor, has been
demonstrated to be hazardous to health and corrosive to
1.1 This test method applies to the determination of
materials. Use caution when handling mercury and mercury-
elemental, oxidized, particle-bound, and total mercury emis-
containing products. See the applicable product Safety Data
sions from stationary combustion sources.
Sheet (SDS) for additional information. The potential exists
1.2 This test method is applicable to elemental, oxidized,
that selling mercury or mercury-containing products, or both,
particle-bound, and total mercury concentrations ranging from
is prohibited by local or national law. Users must determine
3 3
approximately 0.5 μg ⁄Nm to 100 μg ⁄Nm .
legality of sales in their location.
1.3 This test method describes equipment and procedures 1.10 This standard does not purport to address all of the
safety concerns, if any, associated with its use. It is the
for obtaining samples from effluent ducts and stacks, equip-
ment and procedures for laboratory analysis, and procedures responsibility of the user of this standard to establish appro-
priate safety, health, and environmental practices and deter-
for calculating results.
mine the applicability of regulatory limitations prior to use.
1.4 This test method is applicable to sampling elemental,
1.11 This international standard was developed in accor-
oxidized, and particle-bound mercury in flue gases of coal-fired
dance with internationally recognized principles on standard-
stationary sources. It may not be suitable at all measurement
ization established in the Decision on Principles for the
locations, particularly those with high particulate loadings, as
Development of International Standards, Guides and Recom-
explained in Section 16.
mendations issued by the World Trade Organization Technical
1.5 Method applicability is limited to flue gas stream
Barriers to Trade (TBT) Committee.
temperatures within the thermal stability range of the sampling
probe and filter components.
2. Referenced Documents
1.6 The values stated in SI units are to be regarded as the
2.1 ASTM Standards:
standard. The values in parentheses are for information only.
D1193 Specification for Reagent Water
D1356 Terminology Relating to Sampling and Analysis of
1.7 This standard requires users to be familiar with EPA
Atmospheres
stack-gas sampling procedures as stated in EPA Methods 1–4,
D3154 Test Method for Average Velocity in a Duct (Pitot
Method 5, and Method 17.
Tube Method)
1.8 The method requires a high level of experience and
D3685/D3685M Test Methods for Sampling and Determina-
quality control both in the field testing and analytical proce-
tion of Particulate Matter in Stack Gases
dures to obtain high quality data.
D3796 Practice for Calibration of Type S Pitot Tubes
1.9 Warning—Mercury has been designated by many regu-
D4840 Guide for Sample Chain-of-Custody Procedures
latory agencies as a hazardous substance that can cause
D7036 Practice for Competence of Air Emission Testing
Bodies
This test method is under the jurisdiction of ASTM Committee D22 on Air
Quality and is the direct responsibility of Subcommittee D22.03 on Ambient
Atmospheres and Source Emissions. For referenced ASTM standards, visit the ASTM website, www.astm.org, or
Current edition approved March 1, 2024. Published April 2024. Originally contact ASTM Customer Service at service@astm.org. For Annual Book of ASTM
approved in 2002. Last previous edition approved in 2016 as D6784 – 16. DOI: Standards volume information, refer to the standard’s Document Summary page on
10.1520/D6784-24. the ASTM website.
Copyright © ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959. United States
D6784 − 24
E2251 Specification for Liquid-in-Glass ASTM Thermom- 3.2.8 sample train, n—complete setup including nozzle,
eters with Low-Hazard Precision Liquids probe, probe liner, filter, filter holder, impingers, and connec-
tors.
2.2 Other Standards:
EPA Method 1 Sample and Velocity Traverses for Stationary
3.2.9 total mercury, n—all mercury (solid-bound, liquid, or
Sources gaseous) however generated or entrained in the flue gas stream
EPA Method 2 Determination of Stack Gas Velocity and
(that is, summation of elemental, oxidized, and particle-bound
Volumetric Flow Rate (Type S Pitot Tube) mercury).
EPA Method 3 Gas Analysis for the Determination of Dry
3.3 Symbols:
Molecular Weight 2 2
3.3.1 A—cross-sectional area of stack, m (ft )
EPA Method 4 Determination of Moisture Content in Stack
3.3.2 B —water vapor in the gas stream, proportion by
ws
Gases
volume
EPA Method 5 Determination of Particulate Emissions from
3.3.3 ΔH—average pressure differential across the orifice
Stationary Sources
meter, kPa (in. H O)
EPA Method 12 Determination of Inorganic Lead Emissions
from Stationary Sources
3.3.4 Hg —concentration of mercury in sample filter ash,
ash
EPA Method 17 Determination of Particulate Emissions
μg/g
from Stationary Sources (In-Stack Filtration Method)
tp
3.3.5 Hg —concentration of particle-bound mercury, μg/
EPA Method 29 Determination of Metals Emissions from
Nm
Stationary Sources
0 3
3.3.6 Hg —concentration of elemental mercury, μg/Nm
EPA Method 101A Determination of Particle-Bound and
2+ 3
Gaseous Mercury Emissions from Sewage Sludge Incin- 3.3.7 Hg —concentration of oxidized mercury, μg/Nm
erators
3.3.8 IR—instrument reading from mercury analyzer, μg/L
EPA Method 301 Field Validation of Pollutant Measurement
3.3.9 L —leakage rate observed during the post-test leak
p
Methods from Various Waste Media
check, m /min (cfm)
EPA SW 846 7470A Mercury in Liquid Waste—Manual
3.3.10 L —maximum acceptable leakage rate
Cold Vapor Technique
a
EPA Water and Waste 600/4-79-020 Methods for Chemical
3.3.11 M —molecular weight of stack gas, wet basis g/g-
s
Analysis of Water and Wastes
mole (lb/lb-mole)
3.3.12 M —molecular weight of water, 18.0 g/g-mole (18.0
w
3. Terminology
lb/lb-mole)
3.1 Definitions other than those given below in 3.2 and 3.3
3.3.13 N—normal conditions, defined as 0 °C and 101.3 kPa
are listed in Terminology D1356.
(in the U.S. standard conditions 32 °F and 1 atmosphere)
3.2 Definitions of Terms Specific to This Standard:
3.3.14 P —barometric pressure at the sampling site, kPa
bar
3.2.1 elemental mercury, n—mercury in its zero-oxidation
(in. Hg)
state, Hg .
3.3.15 P —absolute stack gas pressure, kPa (in. Hg)
s
3.2.2 elemental mercury catch, n—mercury collected in the
3.3.16 P —standard absolute pressure, 101.3 kPa (29.92
std
acidified hydrogen peroxide (HNO –H O ) and potassium
3 2 2
in. Hg)
permanganate (H SO –KMnO ) impinger solutions employed
2 4 4
in this test method; this is gaseous Hg . 3.3.17 R—ideal gas constant, 0.008314 kPa-m /K-g-mole
(21.85 in. Hg-ft /°R-lb-mole)
3.2.3 front half of the sampling train, n—all mercury col-
lected on and upstream of the sample filter. 3.3.18 T —absolute average dry gas meter temperature, K
m
(°R)
3.2.4 impinger train, n—setup including only the impingers
and connectors. 3.3.19 T —absolute stack temperature, K (°R)
s
3.3.20 T —standard absolute temperature, 293 K
3.2.5 oxidized mercury, n—mercury in its mercurous or
std
2+ 2+
mercuric oxidation states: Hg and Hg , respectively.
3.3.21 V —total digested volume, mL
D
3.2.6 oxidized mercury catch, n—mercury collected in the
3.3.22 V —volume of gas sample as measured by dry gas
m
aqueous potassium chloride (KCl) impinger solutions em-
meter, m (dscf)
2+
ployed in this test method; this is gaseous Hg .
3.3.23 V —volume of gas sample measured by the dry
m(std)
3.2.7 particle-bound mercury catch, n—mercury associated
gas meter in Nm (dscf)
with the particulate matter collected in the front half of the
3.3.24 V —volume of water vapor in the gas sample in
w(std)
sampling train.
m (scf)
3.3.25 W —total mass of ash on sample filter, g
ash
EPA Methods 1 – 29 available from the U.S. Environmental Protection
3.3.26 W —total weight of liquid collected in impingers
lc
Agency’s Emission Measurement Technical Information Center or Code of Federal
and silica gel, g (lb)
Regulations (40 CFR Part 60, Appendix A), Method 101A in 40 CFR Part 61,
Appendix B, Method 301 in 40 CFR 63 Appendix A40 CFR Part 61, Appendix B. 3.3.27 Y—dry gas meter calibration factor
D6784 − 24
FIG. 1 Schematic of Mercury-Sampling Train in the Method 5 Configuration
3.3.28 θ—total sampling time, min 5.2 This test method was developed initially for the mea-
surement of mercury in coal-fired power plants and has been
3.3.29 θ —sampling time interval, from the beginning of a
extensively validated for that application. Since the introduc-
run until the first component change, min
tion of this method, it has been extensively used on other
combustion sources such as cement kilns and waste incinera-
4. Summary of Test Method
tors. With additional procedures given in this standard, it is also
4.1 A sample is withdrawn from the flue gas stream isoki-
applicable to sources having a flue gas composition with high
netically through a probe/filter system, maintained at 120 °C or
levels of hydrochloric acid, and low levels of sulfur dioxide
the flue gas temperature, whichever is greater, followed by a
(Section 16).
series of impingers in an ice bath. Particle-bound mercury is
collected in the front half of the sampling train. Oxidized
6. Interferences
mercury is collected in impingers containing a chilled aqueous
6.1 Chlorine and particulate matter will interfere in speciat-
potassium chloride solution. Elemental mercury is collected in
ing flue gas samples for oxidized and elemental mercury
subsequent impingers (one impinger containing a chilled
concentrations. These biases are addressed further in Section
aqueous acidic solution of hydrogen peroxide and three im-
16 of this test method.
pingers containing chilled aqueous acidic solutions of potas-
sium permanganate). Samples are recovered, digested, and
7. Apparatus
then analyzed for mercury using cold-vapor atomic absorption
7.1 Sampling Train—Similar to Test Methods D3685/
(CVAAS) or fluorescence spectrometry (CVAFS). To achieve
D3685M, EPA Method 5/EPA Method 17 and EPA Method 29
the accuracy and precision specified in this test method, it is
trains, as illustrated in Fig. 1 and Fig. 2.
necessary that quality control and quality assurance procedures
associated with each step of the method be scrupulously
NOTE 1—It is recommended that an in-stack filter method (Method 1,
performed. Successful performance of the method by air Figure 2) be used if possible. The requirement of the method, that the filter
be maintained at the temperature of the flue gas, is ensured in this
emission testing bodies is best achieved by following Practice
configuration. In addition, the in-stack filter method has the added
D7036.
advantage that, only a small portion of the probe/nozzle collects ash that
needs to be brushed onto the filter. Method 5 procedures must be used
5. Significance and Use
when the temperature of the flue gas is below the water dew point (wet
stack). In this case an out-of-stack filter must be used and maintained at a
5.1 The measurement of particle-bound, oxidized,
temperature of 120 °C.
elemental, and total mercury in stationary-source flue gases
NOTE 2—If sampling is conducted in a wet stack where water droplets
provides data that can be used for emissions assessments and
are present, and the nozzle is positioned into the flow, water droplets will
reporting, the certification of continuous mercury monitoring
be collected, and mercury contained in the droplets will be measured.
When water droplets are present, the isokinetic sampling rate and percent
systems, regulatory compliance determinations and research
isokinetic must be calculated accordingly.
programs associated with dispersion modelling, deposition
evaluation, human health and environmental impact assess- 7.1.1 Probe Nozzle (Probe Tip)—Glass nozzles are required
ments. Particle-bound, oxidized, and elemental mercury mea- unless alternate nozzles are constructed of materials that are
surements before and after control devices may be necessary free from contamination and will not interact with the sample.
for optimizing and evaluating the mercury removal efficiency Probe fittings constructed of polytetrafluoroethylene (PTFE),
of emission control technologies. fluorinated ethylene propylene (FEP), etc., are required instead
D6784 − 24
FIG. 2 Schematic of Mercury-Sampling Train in the Method 17 Configuration
of metal fittings to prevent surface losses of mercury and 7.1.7.2 EPA Method 17 Configuration—For EPA Method 17
contamination. Coated metal fittings that overcome adsorption configuration, the sample filter is located in the duct and
losses of mercury may be used.
therefore, naturally maintained at the flue gas temperature. The
7.1.2 Probe Liner—If the sample train is to be in EPA
heating system is only required to maintain the probe and
Method 5 configuration (out-of-stack filtration), the probe liner
connecting umbilical cord to at least 120 °C. If the flue gas
must be constructed of quartz or borosilicate glass. If an EPA
temperature is less than 120 °C, then EPA Method 5 configu-
Method 17 (in-stack filtration) sampling configuration is used,
ration must be used.
the probe/probe liner may be constructed of borosilicate glass,
7.1.8 Condensing/Absorbing System, consists of eight im-
quartz or, depending on the flue gas temperature, for example,
pingers immersed in an ice bath and connected in series with
260 °C for PTFE.
leak-free ground glass fittings or other non-contaminating
7.1.3 Pitot Tube, Type S pitot tube. Refer to Section 2.2 of
leak-free fittings. (At no time are silicone grease or other
EPA Method 2 for a description.
greases to be used for this test method). The first, second,
7.1.4 Differential Pressure Gauges, inclined manometers or
fourth, fifth, sixth, and eighth impingers are of the Greenburg-
equivalent devices. Refer to Section 2.1 of EPA Method 2 for
Smith design modified by replacing the standard tip with a
a description.
1.3 cm (0.5 in.)-ID straight glass tube extending to about
7.1.5 Filter Holder, constructed of borosilicate glass or
1.3 cm (0.5 in.) from the bottom of the flask. The third and
PTFE-coated stainless-steel with a PTFE filter support or other
seventh impingers are also Greenburg-Smith design, but with
non-metallic, non-contaminating support. Do not use a glass
the standard tip including the glass impinging plate. The first,
frit or stainless-steel wire screen. A silicone rubber, viton, or
W
second, and third impingers contain aqueous 1 N (7.46 % ⁄V)
PTFE gasket, designed to provide a positive seal against
potassium chloride (KCl) solution. The fourth impinger con-
leakage from outside or around the filter, may be used.
V
tains an aqueous solution of 5 % ⁄V nitric acid (HNO ) and
7.1.6 Connecting Umbilical Tube, heated PTFE tubing. This
V
10 % ⁄V hydrogen peroxide (H O ). The fifth, sixth, and
2 2
tube must be heated to a minimum of 120 °C to help prevent
W
seventh impingers contain an aqueous solution of 4 % ⁄V
water and acid condensation. (The umbilical tube is defined as
V
potassium permanganate (KMnO ) and 10 % ⁄V sulfuric acid
any tubing longer than 0.5 m that connects the filter holder to
(H SO ). The last impinger contains silica gel or an equivalent
the impinger train). 2 4
desiccant. Refer to Note 4.
7.1.7 Probe and Filter Heating System:
7.1.7.1 EPA Method 5 Configuration—For EPA Method 5
NOTE 3—When flue gas streams are sampled with high moisture
configuration, the temperature of the flue gas, sample probe,
content (>20 %), additional steps must be taken to eliminate carryover of
and the exit of the sample filter must be monitored using
impinger contents from one sample type to the next. These steps must
include use of oversized impinger(s) or use of an empty impinger between
temperature sensors capable of measuring temperature to
the KCl and HNO –H O . If a dry impinger is used, it must be rinsed as
within 3 °C (5.4 °F). The heating system must be capable of 3 2 2
discussed in 13.2 of this test method and the rinse added to the preceding
maintaining the sample gas temperature of the probe and exit
impinger.
of the sample filter to within 615 °C (627 °F) of the flue gas
7.1.9 Metering System, vacuum gauge, leak-free pump,
temperature. Regardless of the flue gas temperature, to prevent
water and acid condensation, the probe temperature, sample thermometers capable of measuring temperature to within 3 °C
(5.4 °F), and a dry gas meter or controlled orifice capable of
filter exit gas temperature, or the temperature of the connecting
umbilical cord shall at no time be less than 120 °C. measuring volume to within 2 %.
D6784 − 24
7.1.10 Barometer, capable of measuring atmospheric pres- 7.3.2 Pipetters—All analysis should be performed with
sure to within 0.33 kPa (0.1 in. Hg). In many cases, the pipetters having accuracy within 60.5 % of the true value, and
barometric reading may be obtained from a nearby National precision ≤0.5 %. A repeater pipetter is recommended to
Weather Service station, in which case, the station value reduce the time required for sample preparation and analysis.
(which is the absolute barometric pressure) shall be requested. Air displacement pipetters are not recommended.
An adjustment for elevation differences between the weather 7.3.3 Transfer pipets, low-density polyethylene disposable
station and sampling point shall be applied at a rate of negative transfer pipets.
0.33 kPa (0.1 in. Hg) per 30 m (100 ft) elevation increase or 7.3.4 Balance, analytical grade, capable of weighing to
vice versa for elevation decrease. within 0.1 μg.
7.1.11 Thermometers, Precision digital thermometers based
7.4 Ancillary equipment, depending upon the application,
on resistance temperature detectors (RTDs), thermistors,
other flue gas parameters may need to be obtained to convert
thermocouples, or organic liquid-in-glass thermometers (such
the mercury measurements into appropriate units. This equip-
as Thermometer S18C in Practice E2251) meeting the require-
ment may include sampling equipment and O or CO analyz-
2 2
ments of specific applications in this test method may be used.
ers.
7.1.12 Gas Density Determination Equipment, temperature
7.5 Spare Parts—Enough sampling equipment must be
sensor and pressure gauge, as described in Section 2.3 and 2.4
brought to the site so that common spare parts are available.
of EPA Method 2. The temperature sensor shall, preferably, be
Arrangements should be made so that, if necessary, parts can
permanently attached to the pitot tube or sampling probe in a
also be shipped next day to the site.
fixed configuration, such that the sensor tip extends beyond the
leading edge of the probe sheath and does not touch any metal.
8. Reagents and Materials
Alternative temperature sensor configurations are described in
8.1 Purity of Reagents—Reagent-grade chemicals shall be
Section 2.1.10 of EPA Method 5. If necessary, a gas analyzer
used in all tests. Unless otherwise indicated, it is intended that
can be used to determine dry molecular weight of the gas (refer
all reagents conform to the specifications of the Committee on
to EPA Method 3).
Analytical Reagents of the American Chemical Society, where
7.2 Digestion Apparatus: 4
such specifications are available. Other grades may be used,
7.2.1 Dry Block Heater or Hot Water Bath, a heater capable
provided it is first ascertained that the reagent is of sufficiently
of maintaining a temperature of 95 °C is required for digestion
high purity to permit its use without lessening the accuracy of
of samples, similar to that described in EPA SW 846 Method
the determination.
7470A.
8.2 Purity of Water—Unless otherwise indicated, references
7.2.2 Ice Bath.
to water shall be understood to mean reagent water as defined
7.2.3 Digestion Flasks—Use 50 mL to 70 mL glass tubes or
by Type II in Specification D1193.
flasks with screw caps that will fit a dry block heater. For a
water bath, 300 mL biological oxygen demand glass bottles for 8.3 Reagents:
SW 846 Method 7470A are to be used. In addition, borosilicate 8.3.1 Boric Acid (H BO ), purified reagent grade.
3 3
glass test tubes, 35 mL to 50 mL volume, with rack are needed. 8.3.2 Hydrochloric Acid (HCl), trace metal-grade concen-
7.2.4 Microwave or Convection Oven and PTFE Digestion trated hydrochloric acid, with a specific gravity of 1.18.
Vessels, 120 mL, or equivalent digestion vessels with caps 8.3.3 Hydrofluoric Acid (HF), concentrated hydrofluoric
equipped with pressure relief valves for the dissolution of ash, acid, 48 % to 50 %.
V
along with a capping station or the equivalent to seal the 8.3.4 Hydrogen Peroxide (H O ), 30 % ⁄V hydrogen perox-
2 2
digestion vessel caps. Use a vented microwave or convection ide.
oven for heating. In addition, polymethylpentene (PMP) or 8.3.5 Hydroxylamine Sulfate ((NH OH) · H SO ), solid.
2 2 2 4
equivalent volumetric flasks are recommended for the digested 8.3.6 Hydroxylamine Hydrochloride (NH OH) · HCl),
2 2
ash solutions. 10 % solution.
8.3.7 Sodium Chloride (NaCl), solid.
7.3 Analytical Equipment:
8.3.8 Mercury Standard Solution, a certified (1000 μg/mL)
7.3.1 Mercury Analyzer, dedicated mercury analyzer or
mercury standard.
equivalent apparatus for the analysis of mercury via CVAAS.
8.3.9 Nitric Acid (HNO ), trace metal-grade concentrated
Alternatively, CVAFS may be used. CVAAS is a method based
V
nitric acid with a specific gravity of 1.42. 20 % ⁄V nitric acid.
on the absorption of radiation at 253.7 nm by mercury vapor.
8.3.10 Potassium Chloride (KCl), solid.
CVAFS is a technique based on the absorption and re-emission
8.3.11 Potassium Dichromate (K Cr O ), solid.
2 2 7
of radiation at 253.7 nm by mercury vapor. The mercury is
8.3.12 Potassium Perchlorate (KClO ), solid.
reduced to the elemental state and sparged from solution in a
8.3.13 Potassium Permanganate (KMnO ), solid.
closed system. The mercury vapor passes through a cell
positioned in the light path of an atomic absorption or
fluorescence spectrometer. Absorbance or fluorescence is mea-
ACS Reagent Chemicals, Specifications and Procedures for Reagents and
Standard-Grade Reference Materials, American Chemical Society, Washington,
sured as a function of mercury concentration. A soda-lime trap
DC. For suggestions on the testing of reagents not listed by the American Chemical
and a magnesium perchlorate trap must be used to precondition
Society, see Analar Standards for Laboratory Chemicals, BDH Ltd., Poole, Dorset,
the gas before it enters the absorption cell. Alternatively, a
U.K., and the United States Pharmacopeia and National Formulary, U.S. Pharma-
hydroscopic membrane to precondition the gas can be used. copeial Convention, Inc. (USPC), Rockville, MD.
D6784 − 24
V
8.3.14 Potassium Persulfate (K S O ), solid. 10 % ⁄V H SO . Dissolve, with stirring, 40 g of KMnO into
2 2 8 2 4 4
V V
8.3.15 Soda Lime (Ca(OH) , NaOH, KOH), solid. 10 % ⁄V H SO , and add 10 % ⁄V H SO , with stirring, to
2 2 4 2 4
8.3.16 Sodium Thiosulfate (Na S O · 5H O) (for high make 1 L. To prevent autocatalytic decomposition of the
2 2 3 2
chloride applications). permanganate solution, filter the solution through filter paper.
8.3.17 Stannous Chloride (SnCl · 2H O), solid. (Warning—See 9.1.1.) H SO –KMnO absorbing solution
2 2 2 4 4
8.3.18 Sulfuric Acid (H SO ), trace metal-grade concen- must be made daily.
2 4
trated sulfuric acid, with a specific gravity of 1.84.
8.5.4 Potassium Permanganate Rinse Solution
W
8.3.19 Tin (Sn) Mossy.
(5 % ⁄V)—Mix 5 g KMnO into water, dilute to 100 mL, and
stir vigorously.
8.4 Materials:
8.4.1 Indicating Silica Gel, with a size of 6-16 mesh.
8.6 Rinse Solutions for Sample Train:
8.4.2 Crushed or Cubed Ice.
8.6.1 0.1 N HNO Solution—A certified reagent grade 0.1 N
8.4.3 Sample Filters, quartz fiber filters, without organic
HNO solution can be purchased directly or can be made by
binders, exhibiting at least 99.95 % efficiency (<0.05 % pen-
slowly adding 12.5 mL of concentrated HNO to a 2000 mL
etration) for 0.3 μm dioctyl phthalate smoke particles and
volumetric flask containing approximately 500 mL of water,
containing less than 0.2 μg/m of mercury. Test data provided
then diluting with water to volume.
by filter manufacturers and suppliers stating filter efficiency
V
8.6.2 10 % ⁄V HNO Solution—Mix carefully, with stirring,
and mercury content are acceptable. Filter material must be
100 mL of concentrated HNO into approximately 800 mL of
5 3
unreactive to sulfur dioxide (SO ) or sulfur trioxide (SO ).
2 3
water. When mixing, be sure to follow standard acid to water
8.4.4 Filter Papers, for filtration of digested samples. The
addition procedures and safety precautions associated with
filter paper must have a particle retention of >20 μm and
strong acids. Then add water, with stirring, to make 1 L.
filtration speed of >12 s.
W
8.6.3 10 % ⁄V Hydroxylamine Solution—Add 100 g hydrox-
8.4.5 Nitrogen Gas (N ), carrier gas of at least 99.998 %
ylamine sulfate and 100 g sodium chloride to a 1000 mL
purity. Alternatively, argon gas may be used.
volumetric flask containing approximately 500 mL of water.
8.4.6 Soda Lime, indicating 4 mesh to 8 mesh absorbent for
After the hydroxylamine sulfate and sodium chloride has been
trapping carbon dioxide.
dissolved, dilute with water to volume. As an alternative a
8.4.7 Sample Containers, glass or PTFE with PTFE-lined
10 % hydroxylamine hydrochloride solution can be used in all
lids.
cases as a replacement for the hydroxylamine sulfate/sodium
NOTE 4—It is recommended that glass amber bottles be used to prevent
chloride solution.
possible deterioration by ultraviolet (UV) light.
8.7 Sample Digestion Reagents:
8.5 Sampling Reagents:
W
8.7.1 Boric Acid Solution (4 % ⁄V)—Dissolve 4 g H BO in
3 3
8.5.1 KCl Absorbing Solution 1 N (1 mol/L)—Dissolve
water and dilute to 100 mL.
74.56 g of KCl in 500 mL of reagent water in a 1000 mL
8.7.2 Aqua Regia (HCl:HNO 3:1)—Add 3 parts concen-
volumetric flask, swirl to mix, and dilute to volume with water.
trated HCl to 1 part concentrated HNO . Note that this should
Mix well. A new batch of solution must be made prior to each
be made up in advance and allowed to form a dark orange
field test.
color. This mixture should be loosely capped, as pressure will
NOTE 5—For applications with high chlorine concentrations: KCl
build as gases form.
W
absorbing solution spiked with sodium thiosulfate (1 mol ⁄l KCl, 0.5 % ⁄V
8.7.3 Potassium Permanganate Solution
Na S O · 5H O) – 5 g Na S O · 5H O is dissolved in 1 L of 1 N KCl
2 2 3 2 2 2 3 2
W
(5 % ⁄V)—Mix 5 g KMnO into water, dilute to 100 mL, and
solution. This solution is used to charge each impinger (100 mL per 4
stir vigorously.
impinger). This solution should be made daily.
W
V 8.7.4 Potassium Persulfate Solution (5 % ⁄V)—Dissolve
8.5.2 HNO –H O Absorbing Solution (5 % ⁄V HNO ,
3 2 2 3
V 5 g K S O in water and dilute to 100 mL.
2 2 8
10 % ⁄V H O )—Add slowly, with stirring, 50 mL of concen-
2 2
trated HNO to a 1000 mL volumetric flask containing ap-
8.8 Analytical Reagents:
proximately 500 mL of water, and then add carefully, with
V
8.8.1 Hydrochloric Acid Solution (10 % ⁄V)—Add 100 mL
V
stirring, 333 mL of 30 % ⁄V H O . Dilute to volume with water.
2 2
concentrated HCl to water and dilute to 1 L. Be sure to follow
Mix well. A new batch of solution must be made prior to each
all safety precautions for using strong acids.
field test.
W
8.8.2 Stannous Chloride Solution (10 % ⁄V)—Dissolve
W
8.5.3 H SO –KMnO Absorbing Solution (4 % ⁄V KMnO ,
2 4 4 4 V V
100 g in 10 % ⁄V HCl, and dilute with 10 % ⁄V HCl to 1 L.
V
10 % ⁄V H SO )—Mix carefully, with stirring, 100 mL of
2 4
Difficulty in dissolving the stannous chloride can be overcome
concentrated H SO into approximately 800 mL of water.
2 4
by dissolving in a more concentrated HCl solution (such as
When mixing, be sure to follow standard acid to water addition
V
100 mL of 50 % ⁄V HCl) and diluting to 1 L with water. Note
procedures and safety precautions associated with strong acids.
that care must be taken when adding water to a strong acid
Then add water, with stirring, to make 1 L. This solution is
solution. Add a lump of mossy tin (~0.5 g) to this solution.
8.9 Mercury Standards:
Felix, L.G.; Clinard, G.I.; Lacey, G.E.; McCain, J.D. “Inertial Cascade
8.9.1 10 μg/mL Hg Stock Solution—Dilute 1 mL of
Impactor Substrate Media for Flue Gas Sampling,” U.S. Environmental Protection
V
1000 μg ⁄mL Hg standard solution to 100 mL with 10 % ⁄V
Agency, Research Triangle Park, NC 27711, Publication No. EPA-600/7-77-060;
June 1977, p. 83. HCl.
D6784 − 24
8.9.2 100 μg/L Hg Stock Solution—Dilute 1 mL of damage. Suitable safety equipment (gloves, goggles, etc.)
V
10 μg ⁄mL Hg stock solution to 100 mL with 10 % ⁄V HCl. This should be used when working with standards and samples
solution and the Working Hg Standards described below may containing mercury, or where exposure to mercury vapors is a
change concentration with time. As a minimum, stock solutions concern.
should be prepared weekly, and stored in glass or PTFE bottles.
9.1.5 Acetone is hazardous in case of skin contact (irritant,
8.9.3 Working Hg Standards—Prepare all working standards
of eye contact (irritant), of ingestion, and of inhalation.
by digesting along with the samples. Prepare digested stan- Acetone is highly flammable in the presence of open flames or
dards of 0.25 μg ⁄L, 0.5 μg ⁄L, 1.0 μg ⁄L, 2.5 μg ⁄L, 5.0 μg ⁄L,
sparks.
7.5 μg ⁄L, and 10.0 μg ⁄L as described in 13.4.1.1.
9.2 Precaution:
8.9.4 Quality Control Standard (QC)—A quality control
9.2.1 The determination of microquantities of mercury re-
standard is prepared from a separate Hg standard solution. The
quires meticulous attention to detail. Good precision is gener-
QC standard should be prepared at a concentration of approxi-
ally unattainable without a high level of experience with
mately one-half the calibration range. It is recommended to
stack-sampling procedures. Precision and accuracy may be
prepare a QC standard at a concentration of 5.0 μg/L in the
improved by knowledge of, and close adherence to, the
same manner as the 5.0 μg/L standard described in 8.9.3.
suggestions that follow.
8.10 Glassware Cleaning Reagents—Prior to any fieldwork,
9.2.1.1 All glassware used in the method must be cleaned
all glassware must be cleaned in accordance with the guide-
thoroughly prior to field use, as described in 8.10 of this test
lines outlined in EPA Method 29, Section 8.1.1 if the stated
method.
precision of this test method is to be met. This procedure
9.2.1.2 Use the same reagents and solutions in the same
requires that the sampling train glassware first be rinsed with
quantities for a group of determinations and the corresponding
hot tap water and then washed in hot soapy water. Then, rinse
solution blank. When a new reagent is prepared or a new stock
the glassware three times with tap water, followed by three
of filters is used, a new blank must be taken and analyzed.
additional rises with distilled water. Soak all glassware in a
V
10 % ⁄V nitric acid solution for a minimum of 4 h. Rinse three
10. Sampling
times with distilled water and rinse a final time with acetone.
10.1 Preparation for Test:
Allow the glassware to air dry and cover all glassware
10.1.1 Quality Assurance Plan—Develop a quality assur-
openings where contamination can occur until the sampling
ance plan (QAP) prior to conducting the tests. The basic
train is assembled for sampling.
elements of the QAP are sections that describe: (1) purpose of
NOTE 6—There are two ways to ensure clean glassware. The first is to
the project, (2) test methodologies, (3) project organization, (4)
bring enough glassware into the field to construct all needed sampling
description of test logistics and schedule, (5) data quality
trains. The second, is to clean the glassware in the field. This requires a
objectives, (6) quality control procedures, and (7) documenta-
large enough space to soak the glassware. In addition, depending on the
scope of the sampling program, an extra person may be required on site.
tion procedures. Note, a section on performance or data quality
assessments, or both, may be required in some applications.
9. Hazards
Each section comprises the following:
10.1.1.1 Purpose of the Project—Discusses why the test is
9.1 Warning:
being conducted and whether total or speciated mercury is to
9.1.1 Pressure may build up in the solution storage bottle
be determined.
because of a potential reaction between potassium permangan-
10.1.1.2 Project Organization—Provides the personnel
ate and acid. Therefore, these bottles should not be fully filled
structure and identify team members and qualifications of the
and should be vented to relieve excess pressure and prevent
team leader with respect to Practice D7036. Identifies the
explosion. Venting must be in a manner that is safe and will not
analytical laboratory that will analyze the samples. Identifies
allow contamination of the solution.
the Quality Assurance Directors of both the test team and
9.1.2 Hazards to personnel exist in the operation of the
laboratory and their respective responsibilities in the test
cold-vapor atomic absorption spectrophotometer. Refer to the
program.
manufacturer’s instruction manual before operating the instru-
10.1.1.3 Test Logistics—Details the test schedule, sampling
ment.
locations, equipment to be used, role of sampling and plant
9.1.3 Sample digestion with hot concentrated acids creates a
personnel during sampling. Addresses communication mecha-
safety problem. Observe appropriate laboratory procedures for
nisms within the team and with the plant. A site visit prior to
working with concentrated acids. Hydrofluoric acid used in the
the testing is recommended.
sample digestion procedures is highly corrosive and is very
toxic by inhalation or contact with the skin. Avoid exposure by 10.1.1.4 Quality Objective—States the quality objectives for
equipment calibration, test parameters, and precision and bias
contact with the skin or eyes, or by inhalation of HF vapor. It
is essential to use suitable personal protective equipment, of the results.
including impermeable gloves and eye protection when work- 10.1.1.5 Quality Control Procedures—Describes quality
ing with HF. Use a fume hood when working with concentrated
control procedures used for (1) equipment calibration, (2)
HF and when carry out open-vessel dissolution with HF. glassware cleaning and handling, (3) chain-of-custody describ-
9.1.4 Mercury standards at high concentrations (1000 μg/ ing sample management from the point of collection to analysis
mL) are toxic and can cause skin irritation and serious eye and final data reduction (see Guide D4840), (4) isokinetic
D6784 − 24
sampling, (5) field spikes, (6) sample blanks, and (7) laboratory were used for the velocity traverse as stated in 10.1.2 of this
analysis (including spikes, blanks, replicates, and calibration test method. Each traverse point must be sampled for a
procedures). minimum of 5 min.
10.1.1.6 Documentation Procedures—Includes the format
11. Preparation of Apparatus
of the test report, data sheet custody and integrity, and backup
for electronic files.
11.1 Pre-test Preparation:
10.1.1.7 Assessments—Assessments include but are not lim-
11.1.1 Weigh several 200 g to 300 g portions of silica gel in
ited to review of data to verify it meets its intended use as well
airtight containers to the nearest 0.5 g. Record the total weight
as analysis of audit samples and performance tests.
of the silica gel plus container on each container. Alternatively,
10.1.2 Preliminary Stack Measurements—Select the sam- the silica gel can be weighed directly in the impinger imme-
pling site, and determine the number of sampling points, stack diately prior to the train being assembled.
pressure, temperature, moisture, dry molecular weight, and 11.1.2 Desiccate the sample filters at 20 °C 6 5.6 °C (68 °F
range of velocity head in accordance with procedures of Test
6 10 °F) and ambient pressure for 24 h to 36 h, weigh at
Method D3154 or EPA Methods 1 through 4. intervals of at least 6 h to a constant weight (that is, <0.5 μg
change from previous weighing), and record results to the
NOTE 7—Prior to testing, it is good practice to remove mercury
nearest 0.1 μg. Alternatively, the filters may be oven-dried at
containing devices from both staging areas and testing areas (that is,
105 °C (220 °F) for 2 h to 3 h, desiccated for 2 h, and weighed.
mercury manometers, broken fluorescent lamps, mercury thermometers).
11.1.3 Clean all sampling train glassware as described in
10.1.3 Select the correct nozzle diameter to maintain isoki-
8.10 before each series of tests at a single source. Until the
netic sampling rates based on the range of velocity heads
sampling train is assembled for sampling, cover all glassware
determined in 10.1.2, and to provide adequate sample volume,
openings where contamination can occur.
without depleting the KMnO .
11.2 Preparation of Sampling Train:
NOTE 8—Too high of a flow rate will cause the KMnO to be depleted
11.2.1 Assemble the sampling train as shown in Fig. 1.
as it reacts with SO in the final set of impingers; as the KMnO is
2 4
11.2.2 Place 100 mL of the KCl solution (see 8.5.1) in each
depleted, it will turn brown then clear and will lose its ability to retain
of the first, second, and third impingers, as indicated in Fig. 1.
mercury.
NOTE 11—For applications with high chlorine concentrations, place 100
10.1.4 Ensure that the proper differential pressure gauge is
mL of the sodium thiosulfate-spiked KCl solution (see section 8.5.1) in
selected for the range of velocity heads (refer to EPA Method
each of the first, second, and third impingers, as indicated in Fig. 1.
2, Section 2.2).
11.2.3 Place 100 mL of the HNO –H O solution (see 8.5.2)
3 2 2
10.1.5 If the flue gas is stratified with respect to particulate
in the fourth impinger, as indicated in Fig. 1.
concentrations, gas concentrations, or both, the stack cross-
11.2.4 Place 100 mL of the H SO –KMnO absorbing
2 4 4
section is traversed, as specified by EPA Method 1. If the flue
solution (see 8.5.3) in each of the fifth, sixth, and seventh
gas is not stratified, sample at a fixed, representative location
impingers, as indicated in Fig. 1.
where the flue gas is well-mixed. It is recommended that an
11.2.5 Transfer approximately 200 g to 300 g of silica gel
EPA Method 17 configuration be used; however, if an EPA
from its container to the last impinger, as indicated in Fig. 1.
Method 5 setup is to be used, then select a suitable probe length
11.2.6 Prior to final train assembly, weigh and record the
such that when the stack cross-section is traversed, all traverse
weight of each impinger. This information is required to
points can be sampled. Consider sampling from opposite sides
calculate the moisture content of the sampled flue gas.
of the stack to minimize probe length when a large duct or
11.2.7 To ensure leak-free sampling train connections and to
stack is sampled.
prevent possible sample contamination problems, use PTFE
NOTE 9—Traversing may not be necessary, depending on the objectives
tape, PTFE-coated O-rings, or other non-contaminating mate-
of the test. In coal-fired power plants, gas-phase mercury has been found
rial.
to constitute on the order of 95 % of the total mercury. Errors introduced
11.2.8 Place a weighed filter in the filter holder using
by not traversing may be either positive or negative depending upon the
pattern and degree of stratification. Considering low-levels of particulate tweezers or clean disposable surgical gloves.
adsorbed mercury, bias introduced by sampling at a single point may be
11.2.9 Install the selected nozzle using a non-contaminating
negligible. For low dust applications where the flue gas is well-mixed and
rubber-type O-ring or equivalent when stack temperatures are
a sampling location representative of the stack flow can be found, single
less than 260 °C (500 °F) and an alternative gasket material
point sampling may be adequate for obtaining representative samples.
when temperatures are higher. Other connecting systems, such
NOTE 10—Traversing the stack may affect the performance, accuracy,
and precision of the method if not done carefully. When traversing, the as PTFE ferrules or ground glass joints, may also be used on
apparatus should be moved in a manner to avoid leaks or breakage of the
the probe and nozzle.
glassware.
11.2.10 Mark the probe with heat-resistant tape or by some
other method to denote the proper distance into the stack or
10.1.6 Sampling Time and Volume—The total sampling time
duct for each sampling point.
for this test method should be at least 2 h but not more than
11.2.11 Place crushed or cubed ice around the impingers.
3 h. Use a nozzle size that will guarantee an isokinetic gas
sample volume between 1.0 dry cubic metres corrected to 11.2.12 Leak-Check Procedures—Follow the leak-check
3 3
standard conditions (Nm ) and 2.5 Nm . If traverse sampling is procedures given in Section 4.1.4.1 (Pre-test Leak Check),
done (when required), use the same points for sampling that Section 4.1.4.2 (Leak Checks During the Sample Run), and
D6784 − 24
Section 4.1.4.3 (Post-test Leak Checks) of EPA Method 5 or 615 °C of the flue gas temperature at the sampling location.
17. When 50 kPa vacuum is applied, the leak rate must Alternatively, for reasons discussed in paragraph 16.2.3, the
<0.01 cfm. filter and probe may be operated at 120 °C. However, regard-
less of the sample configuration, the sample filter, probe, or
NOTE 12—If O-ring seal glassware is used, the leak rate should be
connecting umbilical cord temperature must not at any time be
<0.01 cfm.
lower than 120 °C.
NOTE 13—If the flue gas temperature at the sampling location is greater
than 260 °C (above the temperature where PTFE or rubber-type seals can
13.1.2 Record the data, as indicated in Fig. 3, at least once
be used), the post-test leak check is determined beginning at the front end
at each sample point but not less than once every 5 min.
of the probe (does not include nozzle or sample filter holder for EPA
13.1.3 Record the dry gas meter reading at the beginning of
Method 17).
a sampling run, the beginning and end of each sampling time
11.3 Preparation of the Field Blank—A field blank is
increment, before and after each leak check, and when sam-
performed by assembling a sample train, transporting it to the
pling is halted.
sampling location during the sampling period, and recovering
13.1.4 Level and zero the manometer. Periodically check
it as a regular sample. These data are used to ensure that there
the manometer level and zero, because it may drift during the
is no contamination due to sampling activities. See 13.4.3.2.
test period.
Conduct at least one field blank for each day of testing.
13.1.5 Clean the port holes prior to the sampling run.
11.4 Preparation of Field Spike—A field spike is similar to
13.1.6 Remove the nozzle cap. Verify that the filter and
the field blank, with the addition of a predetermined amount of
probe heating systems are up to temperature and that the pitot
mercury added to each of the three types of impinger solutions.
tube and probe are properly positioned.
Perform the field spike by assembling a sample train, trans-
NOTE 15—For an EPA Method 5 configuration, prior to starting the gas
porting it to the sampling location during the sampling period,
flow through the system, the sample filter exit gas temperature may not be
adding the spiked solutions and recovering it as a regular
at the hot box temperature. However, if the system is set up correctly, once
sample. These solutions are then
...