40cfr50.12 - 2003

Environmental Protection Agency Pt. 50, App.
air quality meeting the 1-hour stand- § 50.11 National primary and sec-
ard. Area designations and classifica- ondary ambient air quality stand-
tions with respect to the 1-hour standards for nitrogen dioxide.
ards are codified in 40 CFR part 81. (a) The level of the national primary
ambient air quality standard for nitro-
[62 FR 38894, July 18, 1997, as amended at 65
gen dioxide is 0.053 parts per million
FR 45200, July 20, 2000]
(100 micrograms per cubic meter), an-
EFFECTIVE DATE NOTE: At 68 FR 38163, June nual arithmetic mean concentration.
26, 2003, § 50.9 was amended by adding para- (b) The level of national secondary
graph (c), effective August 25, 2003. For the ambient air quality standard for nitro-
convenience of the user, the added text is set gen dioxide is 0.053 parts per million
forth as follows. (100 micrograms per cubic meter), an-
§ 50.9 National 1-hour primary and sec- nual arithmetic mean concentration.
ondary ambient air quality standards for (c) The levels of the standards shall
ozone. be measured by:
(1) A reference method based on ap-
* * * * * pendix F and designated in accordance
with part 53 of this chapter, or
(c) EPA’s authority under paragraph (b) of (2) An equivalent method designated
this section to determine that the 1-hour in accordance with part 53 of this chap-
standard no longer applies to an area based ter.
on a determination that the area has at- (d) The standards are attained when
tained the 1-hour standard is stayed until
the annual arithmetic mean concentra-
such time as EPA issues a final rule revising
tion in a calendar year is less than or
or reinstating such authority and considers
and addresses in such rulemaking any com-
equal to 0.053 ppm, rounded to three
ments concerning (1) which, if any, imple- decimal places (fractional parts equal
mentation activities for a revised ozone to or greater than 0.0005 ppm must be
standard (including but not limited to des- rounded up). To demonstrate attain-
ignation and classification of areas) would ment, an annual mean must be based
need to occur before EPA would determine upon hourly data that are at least 75
that the 1-hour ozone standard no longer ap- percent complete or upon data derived
plies to an area, and (2) the effect of revising from manual methods that are at least
the ozone NAAQS on the existing 1-hour 75 percent complete for the scheduled
ozone designations. sampling days in each calendar quar-
ter.
§ 50.10 National 8-hour primary and
secondary ambient air quality [50 FR 25544, June 19, 1985]
standards for ozone.
§ 50.12 National primary and sec-
(a) The level of the national 8-hour ondary ambient air quality stand-
primary and secondary ambient air ards for lead.
quality standards for ozone, measured National primary and secondary am-
by a reference method based on appen- bient air quality standards for lead and
dix D to this part and designated in ac- its compounds, measured as elemental
cordance with part 53 of this chapter, is lead by a reference method based on
Copia controlada No.__3___
0.08 parts per million (ppm), daily max- appendix G to this part, or by an equiv-
imum 8-hour average. alent method, are: 1.5 micrograms per
(b) The 8-hour primary and secondary cubic meter, maximum arithmetic
ozone ambient air quality standards mean averaged over a calendar quarter.
are met at an ambient air quality mon- (Secs. 109, 301(a) Clean Air Act as amended
itoring site when the average of the an- (42 U.S.C. 7409, 7601(a)))
Fecha: 14/11/2013
nual fourth-highest daily maximum 8- [43 FR 46258, Oct. 5, 1978]

hour average ozone concentration is
less than or equal to 0.08 ppm, as deter- APPENDIX A TO PART 50—REFERENCE
mined in accordance with appendix I to METHOD FOR THE DETERMINATION OF
this part. SULFUR DIOXIDE IN THE ATMOS-
PHERE (PARAROSANILINE METHOD)
[62 FR 38894, July 18, 1997]
1.0 Applicability.
VerDate Jan<31>2003 15:23 Jul 18, 2003 Jkt 200140 PO 00000 Frm 00009 Fmt 8010 Sfmt 8002 Y:\SGML\200140T.XXX 200140T
Pt. 50, App. A 40 CFR Ch. I (7–1–03 Edition)
1.1 This method provides a measurement of ides by the addition of sulfamic acid,(10, 11)
the concentration of sulfur dioxide (SO2) in heavy metals by the addition of ethylene-
ambient air for determining compliance with diamine tetracetic acid disodium salt
the primary and secondary national ambient (EDTA) and phosphoric acid,(10, 12) and
air quality standards for sulfur oxides (sulfur ozone by time delay.(10) Up to 60 µg Fe (III),
dioxide) as specified in § 50.4 and § 50.5 of this 22 µg V (V), 10 µg Cu (II), 10 µg Mn (II), and
chapter. The method is applicable to the 10 µg Cr (III) in 10 mL absorbing reagent can
measurement of ambient SO2 concentrations be tolerated in the procedure.(10) No signifi-
using sampling periods ranging from 30 min- cant interference has been encountered with
utes to 24 hours. Additional quality assur- 2.3 µg NH3.(13)
ance procedures and guidance are provided in 5.0 Precision and Accuracy.
part 58, appendixes A and B, of this chapter 5.1 The precision of the analysis is 4.6 per-
and in references 1 and 2. cent (at the 95 percent confidence level)
2.0 Principle. based on the analysis of standard sulfite
2.1 A measured volume of air is bubbled samples.(10)
through a solution of 0.04 M potassium 5.2 Collaborative test results (14) based on
tetrachloromercurate (TCM). The SO2 the analysis of synthetic test atmospheres
present in the air stream reacts with the (SO2 in scrubbed air) using the 24-hour sam-
TCM solution to form a stable monochloro- pling procedure and the sulfite-TCM calibra-
sulfonatomercurate(3) complex. Once tion procedure show that:
formed, this complex resists air oxidation(4,
• The replication error varies linearly with
5) and is stable in the presence of strong
concentration from ±2.5 µg/m3 at con-
oxidants such as ozone and oxides of nitro-
centrations of 100 µg/m3 to ±7 µg/m3 at con-
gen. During subsequent analysis, the com-
centrations of 400 µg/m3.
plex is reacted with acid-bleached para-
• The day-to-day variability within an indi-
rosaniline dye and formaldehyde to form an
vidual laboratory (repeatability) varies
intensely colored pararosaniline methyl sul-
linearly with concentration from ±18.1 µg/
fonic acid.(6) The optical density of this spe-
m3 at levels of 100 µg/m3 to ±50.9 µg/m3 at
cies is determined spectrophotometrically at
548 nm and is directly related to the amount levels of 400 µg/m3.
of SO2 collected. The total volume of air • The day-to-day variability between two or
sampled, corrected to EPA reference condi- more laboratories (reproducibility) varies
tions (25 °C, 760 mm Hg [101 kPa]), is deter- linearly with concentration from ±36.9 µg/
mined from the measured flow rate and the m3 at levels of 100 µg/m3 to ±103.5 µ g/m3 at
sampling time. The concentration of SO2 in levels of 400 µg/m3.
the ambient air is computed and expressed in • The method has a concentration-dependent
micrograms per standard cubic meter (µg/std bias, which becomes significant at the 95
m3). percent confidence level at the high con-
3.0 Range. centration level. Observed values tend to
3.1 The lower limit of detection of SO2 in 10 be lower than the expected SO2 concentra-
mL of TCM is 0.75 µg (based on collaborative tion level.
test results).(7) This represents a concentra- 6.0 Stability.
tion of 25 µg SO2/m3 (0.01 ppm) in an air sam- 6.1 By sampling in a controlled tempera-
ple of 30 standard liters (short-term sam- ture environment of 15°±10 °C, greater than
pling) and a concentration of 13 µg SO2/m3 98.9 percent of the SO2–TCM complex is re-
(0.005 ppm) in an air sample of 288 standard tained at the completion of sampling. (15) If
liters (long-term sampling). Concentrations kept at 5 °C following the completion of sam-
less than 25 µg SO2/m3 can be measured by pling, the collected sample has been found to
sampling larger volumes of ambient air; be stable for up to 30 days.(10) The presence
however, the collection efficiency falls off of EDTA enhances the stability of SO2 in the
rapidly at low concentrations.(8, 9) Beer’s TCM solution and the rate of decay is inde-
law is adhered to up to 34 µg of SO2 in 25 mL pendent of the concentration of SO2.(16)
of final solution. This upper limit of the 7.0 Apparatus.
analysis range represents a concentration of 7.1 Sampling.
1,130 µg SO2/m3 (0.43 ppm) in an air sample of 7.1.1 Sample probe: A sample probe meeting
30 standard liters and a concentration of 590 the requirements of section 7 of 40 CFR part
µg SO2/m3 (0.23 ppm) in an air sample of 288 58, appendix E (Teflon or glass with resi-
Fecha: 14/11/2013
standard liters. Higher concentrations can be dence time less than 20 sec.) is used to trans-
measured by collecting a smaller volume of port ambient air to the sampling train loca-
air, by increasing the volume of absorbing tion. The end of the probe should be designed
solution, or by diluting a suitable portion of or oriented to preclude the sampling of pre-
the collected sample with absorbing solution cipitation, large particles, etc. A suitable
prior to analysis. probe can be constructed from Teflon tub-
4.0 Interferences. ing connected to an inverted funnel.
4.1 The effects of the principal potential 7.1.2 Absorber—short-term sampling: An all
interferences have been minimized or elimi- glass midget impinger having a solution ca-
nated in the following manner: Nitrogen ox- pacity of 30 mL and a stem clearance of 4±1
10
Environmental Protection Agency Pt. 50, App. A
mm from the bottom of the vessel is used for from a scientific supply firm. Upon receipt,
sampling periods of 30 minutes and 1 hour (or the orifice test should be performed to verify
any period considerably less than 24 hours). the orifice size. The 50 mL volume level
Such an impinger is shown in Figure 1. These should be permanently marked on the ab-
impingers are commercially available from sorber. The assembled absorber is shown in
distributors such as Ace Glass, Incorporated. Figure 2.
7.1.3 Absorber—24-hour sampling: A poly- 7.1.4 Moisture trap: A moisture trap con-
propylene tube 32 mm in diameter and 164 structed of a glass trap as shown in Figure 1
mm long (available from Bel Art Products, or a polypropylene tube as shown in Figure
Pequammock, NJ) is used as the absorber. 2 is placed between the absorber tube and
The cap of the absorber must be a poly- flow control device to prevent entrained liq-
propylene cap with two ports (rubber stop- uid from reaching the flow control device.
pers are unacceptable because the absorbing The tube is packed with indicating silica gel
reagent can react with the stopper to yield as shown in Figure 2. Glass wool may be sub-
erroneously high SO2 concentrations). A stituted for silica gel when collecting short-
glass impinger stem, 6 mm in diameter and term samples (1 hour or less) as shown in
158 mm long, is inserted into one port of the Figure 1, or for long term (24 hour) samples
absorber cap. The tip of the stem is tapered if flow changes are not routinely encoun-
to a small diameter orifice (0.4±0.1 mm) such tered.
that a No. 79 jeweler’s drill bit will pass 7.1.5 Cap seals: The absorber and moisture
through the opening but a No. 78 drill bit trap caps must seal securely to prevent leaks
will not. Clearance from the bottom of the during use. Heat-shrink material as shown in
absorber to the tip of the stem must be 6±2 Figure 2 can be used to retain the cap seals
mm. Glass stems can be fabricated by any if there is any chance of the caps coming
reputable glass blower or can be obtained loose during sampling, shipment, or storage.

Fecha: 14/11/2013
11

Fecha: 14/11/2013
12
50-702.eps</GPH>

Fecha: 14/11/2013
13
50-703.eps</GPH>
7.1.6 Flow control device: A calibrated ro- meter is also recommended to determine the
tameter and needle valve combination capa- duration of the sampling period.
ble of maintaining and measuring air flow to 7.2 Shipping.
within ±2 percent is suitable for short-term 7.2.1 Shipping container: A shipping con-
sampling but may not be used for long-term tainer that can maintain a temperature of 5°
sampling. A critical orifice can be used for ±5 °C is used for transporting the sample
regulating flow rate for both long-term and from the collection site to the analytical
short-term sampling. A 22-gauge hypodermic laboratory. Ice coolers or refrigerated ship-
needle 25 mm long may be used as a critical ping containers have been found to be satis-
orifice to yield a flow rate of approximately factory. The use of eutectic cold packs in-
1 L/min for a 30-minute sampling period. stead of ice will give a more stable tempera-
When sampling for 1 hour, a 23-gauge hypo- ture control. Such equipment is available
dermic needle 16 mm in length will provide a from Cole-Parmer Company, 7425 North Oak
flow rate of approximately 0.5 L/min. Flow Park Avenue, Chicago, IL 60648.
control for a 24-hour sample may be provided 7.3 Analysis.
by a 27-gauge hypodermic needle critical ori-
7.3.1 Spectrophotometer: A spectrophotom-
fice that is 9.5 mm in length. The flow rate
eter suitable for measurement of absorb-
should be in the range of 0.18 to 0.22 L/min.
ances at 548 nm with an effective spectral
7.1.7 Flow measurement device: Device cali-
bandwidth of less than 15 nm is required for
brated as specified in 9.4.1 and used to meas-
ure sample flow rate at the monitoring site. analysis. If the spectrophotometer reads out
7.1.8 Membrane particle filter: A membrane in transmittance, convert to absorbance as
filter of 0.8 to 2 µm porosity is used to pro- follows:
tect the flow controller from particles dur-
ing long-term sampling. This item is op- A = log10 (1/T ) (1)
tional for short-term sampling. where:
7.1.9 Vacuum pump: A vacuum pump
equipped with a vacuum gauge and capable A = absorbance, and
of maintaining at least 70 kPa (0.7 atm) vac- T = transmittance (0<´T<1).
uum differential across the flow control de- A standard wavelength filter traceable to
vice at the specified flow rate is required for the National Bureau of Standards is used to
sampling. verify the wavelength calibration according
7.1.10 Temperature control device: The tem- to the procedure enclosed with the filter.
perature of the absorbing solution during The wavelength calibration must be verified
sampling must be maintained at 15° ±10 °C.
upon initial receipt of the instrument and
As soon as possible following sampling and
after each 160 hours of normal use or every 6
until analysis, the temperature of the col-
months, whichever occurs first.
lected sample must be maintained at 5° ±5 °C.
Where an extended period of time may elapse 7.3.2 Spectrophotometer cells: A set of 1-cm
before the collected sample can be moved to path length cells suitable for use in the visi-
the lower storage temperature, a collection ble region is used during analysis. If the cells
temperature near the lower limit of the 15 ± are unmatched, a matching correction factor
10 °C range should be used to minimize losses must be determined according to Section
during this period. Thermoelectric coolers 10.1.
specifically designed for this temperature 7.3.3 Temperature control device: The color
control are available commercially and nor- development step during analysis must be
mally operate in the range of 5° to 15 °C. conducted in an environment that is in the
Small refrigerators can be modified to pro- range of 20° to 30 °C and controlled to ±1 °C.
vide the required temperature control; how- Both calibration and sample analysis must
ever, inlet lines must be insulated from the be performed under identical conditions
lower temperatures to prevent condensation (within 1 °C). Adequate temperature control

when sampling under humid conditions. A may be obtained by means of constant tem-
small heating pad may be necessary when perature baths, water baths with manual
sampling at low temperatures (<7 °C) to pre- temperature control, or temperature con-
vent the absorbing solution from freez- trolled rooms.
ing.(17) 7.3.4 Glassware: Class A volumetric glass-
7.1.11 Sampling train container: The absorb- ware of various capacities is required for pre-
Fecha: 14/11/2013
ing solution must be shielded from light dur- paring and standardizing reagents and stand-
ing and after sampling. Most commercially ards and for dispensing solutions during
available sampler trains are enclosed in a analysis. These included pipets, volumetric
light-proof box. flasks, and burets.
7.1.12 Timer: A timer is recommended to 7.3.5 TCM waste receptacle: A glass waste re-
initiate and to stop sampling for the 24-hour ceptacle is required for the storage of spent
period. The timer is not a required piece of TCM solution. This vessel should be
equipment; however, without the timer a stoppered and stored in a hood at all times.
technician would be required to start and 8.0 Reagents.
stop the sampling manually. An elapsed time 8.1 Sampling.
14
EC08NO91.000</MATH>
8.1.1 Distilled water: Purity of distilled 8.2.7 Potassium iodate solution: Accurately
water must be verified by the following pro- weigh to the nearest 0.1 mg, 1.5 g (record
cedure:(18) weight) of primary standard grade potassium
• Place 0.20 mL of potassium permanganate iodate that has been previously dried at 180
solution (0.316 g/L), 500 mL of distilled °C for at least 3 hours and cooled in a
water, and 1mL of concentrated sulfuric dessicator. Dissolve, then dilute to volume in
acid in a chemically resistant glass bottle, a 500-mL volumetric flask with distilled
stopper the bottle, and allow to stand. water.
• If the permanganate color (pink) does not 8.2.8 Stock sodium thiosulfate solution (0.1 N):
disappear completely after a period of 1 Prepare a stock solution by dissolving 25 g
hour at room temperature, the water is sodium thiosulfate (Na2 S2 O3÷5H2 O) in 1,000
suitable for use. mL freshly boiled, cooled, distilled water and
• If the permanganate color does disappear, adding 0.1 g sodium carbonate to the solu-
the water can be purified by redistilling tion. Allow the solution to stand at least 1
with one crystal each of barium hydroxide day before standardizing. To standardize, ac-
and potassium permanganate in an all curately pipet 50 mL of potassium iodate so-
glass still. lution (Section 8.2.7) into a 500-mL iodine
8.1.2 Absorbing reagent (0.04 M potassium flask and add 2.0 g of potassium iodide and 10
tetrachloromercurate [TCM]): Dissolve 10.86 mL of 1 N HCl. Stopper the flask and allow
g mercuric chloride, 0.066 g EDTA, and 6.0 g to stand for 5 minutes. Titrate the solution
potassium chloride in distilled water and di- with stock sodium thiosulfate solution (Sec-
lute to volume with distilled water in a 1,000- tion 8.2.8) to a pale yellow color. Add 5 mL of
mL volumetric flask. (Caution: Mercuric starch solution (Section 8.2.5) and titrate
chloride is highly poisonous. If spilled on until the blue color just disappears. Cal-
skin, flush with water immediately.) The pH culate the normality (Ns) of the stock so-
of this reagent should be between 3.0 and 5.0 dium thiosulfate solution as follows:
(10) Check the pH of the absorbing solution
W
by using pH indicating paper or a pH meter.
If the pH of the solution is not between 3.0
NS = × 2.80 (2)
and 5.0, dispose of the solution according to M
one of the disposal techniques described in where:
Section 13.0. The absorbing reagent is nor- M = volume of thiosulfate required in mL,
mally stable for 6 months. If a precipitate and
forms, dispose of the reagent according to W = weight of potassium iodate in g (re-
one of the procedures described in Section corded weight in Section 8.2.7).
13.0.
8.2 Analysis. 10 3 (conversion of g to mg) × 0.1(fraction iodate used )
2.80 =
8.2.1 Sulfamic acid (0.6%): Dissolve 0.6 g sul- 35.67 (equivalent weight of potassium iodate)
famic acid in 100 mL distilled water. Perpare 8.2.9 Working sodium thiosulfate titrant (0.01
fresh daily. N): Accurately pipet 100 mL of stock sodium
8.2.2 Formaldehyde (0.2%): Dilute 5 mL thiosulfate solution (Section 8.2.8) into a
formaldehyde solution (36 to 38 percent) to 1,000-mL volumetric flask and dilute to vol-
1,000 mL with distilled water. Prepare fresh ume with freshly boiled, cooled, distilled
daily. water. Calculate the normality of the work-
8.2.3 Stock iodine solution (0.1 N): Place 12.7 ing sodium thiosulfate titrant (NT) as fol-
g resublimed iodine in a 250-mL beaker and lows:
add 40 g potassium iodide and 25 mL water.
Stir until dissolved, transfer to a 1,000 mL
volumetric flask and dilute to volume with
N T = N S × 0.100 (3)
distilled water. 8.2.10 Standardized sulfite solution for the

8.2.4 Iodine solution (0.01 N): Prepare ap- preparation of working sulfite-TCM solution:
proximately 0.01 N iodine solution by dilut- Dissolve 0.30 g sodium metabisulfite (Na2 S2
ing 50 mL of stock iodine solution (Section O5) or 0.40 g sodium sulfite (Na2 SO3) in 500
8.2.3) to 500 mL with distilled water. mL of recently boiled, cooled, distilled
8.2.5 Starch indicator solution: Triturate 0.4 water. (Sulfite solution is unstable; it is
g soluble starch and 0.002 g mercuric iodide therefore important to use water of the high-
Fecha: 14/11/2013
(preservative) with enough distilled water to est purity to minimize this instability.) This
EC08NO91.003</MATH>
form a paste. Add the paste slowly to 200 mL solution contains the equivalent of 320 to 400
of boiling distilled water and continue boil- µg SO2/mL. The actual concentration of the
ing until clear. Cool and transfer the solu- solution is determined by adding excess io-
tion to a glass stopperd bottle. dine and back-titrating with standard so-
8.2.6 1 N hydrochloric acid: Slowly and while dium thiosulfate solution. To back-titrate,
stirring, add 86 mL of concentrated hydro- pipet 50 mL of the 0.01 N iodine solution
EC08NO91.002</MATH>
chloric acid to 500 mL of distilled water. (Section 8.2.4) into each of two 500-mL iodine
Allow to cool and dilute to 1,000 mL with dis- flasks (A and B). To flask A (blank) add 25
tilled water. mL distilled water, and to flask B (sample)
15
EC08NO91.001</MATH>
pipet 25 mL sulfite solution. Stopper the tinue the titration until the blue color just
flasks and allow to stand for 5 minutes. Pre- disappears.
pare the working sulfite-TCM solution (Sec- 8.2.11 Working sulfite-TCM solution: Accu-
tion 8.2.11) immediately prior to adding the rately pipet 5 mL of the standard sulfite so-
iodine solution to the flasks. Using a buret lution (Section 8.2.10) into a 250-mL volu-
containing standardized 0.01 N thiosulfate metric flask and dilute to volume with 0.04
titrant (Section 8.2.9), titrate the solution in M TCM. Calculate the concentration of sul-
each flask to a pale yellow color. Then add 5 fur dioxide in the working solution as fol-
mL starch solution (Section 8.2.5) and con- lows:
( A − B)( N T )(32,000)
C TCM / SO 2 (µg SO 2 /mL ) = × 0.02 ( 4)
25
where: percent potassium iodide (KI) solution in a

A = volume of thiosulfate titrant required 50-mL separatory funnel and shake thor-
for the blank, mL; oughly. If a yellow color appears in the alco-
B = volume of thiosulfate titrant required hol phase, redistill the 1-butanol from silver
for the sample, mL; oxide and collect the middle fraction or pur-
NT = normality of the thiosulfate titrant, chase a new supply of 1-butanol.
from equation (3); 2. Weigh 100 mg of pararosaniline hydro-
32,000 = milliequivalent weight of SO2, µg; chloride dye (PRA) in a small beaker. Add 50
25 = volume of standard sulfite solution, mL; mL of the equilibrated acid (drain in acid
and from the bottom of the separatory funnel in
0.02 = dilution factor. 1.) to the beaker and let stand for several
This solution is stable for 30 days if kept at minutes. Discard the remaining acid phase in
5 °C. (16) If not kept at 5 °C, prepare fresh the separatory funnel.
daily. 3. To a 125-mL separatory funnel, add 50
8.2.12 Purified pararosaniline (PRA) stock so- mL of the equilibrated 1-butanol (draw the 1-
lution (0.2% nominal): butanol from the top of the separatory fun-
8.2.12.1 Dye specifications— nel in 1.). Transfer the acid solution (from 2.)
containing the dye to the funnel and shake
• The dye must have a maximum absorbance
carefully to extract. The violet impurity will
at a wavelength of 540 nm when assayed in
transfer to the organic phase.
a buffered solution of 0.1 M sodium ace-
tate-acetic acid; 4. Transfer the lower aqueous phase into
• The absorbance of the reagent blank, another separatory funnel, add 20 mL of
which is temperature sensitive (0.015 ab- equilibrated 1-butanol, and extract again.
sorbance unit/ °C), must not exceed 0.170 at 5. Repeat the extraction procedure with
22 °C with a 1-cm optical path length when three more 10-mL portions of equilibrated 1-
the blank is prepared according to the butanol.
specified procedure; 6. After the final extraction, filter the acid
• The calibration curve (Section 10.0) must phase through a cotton plug into a 50-mL
have a slope equal to 0.030±0.002 absorbance volumetric flask and bring to volume with 1
unit/µg SO2 with a 1-cm optical path length N HCl. This stock reagent will be a yellowish
when the dye is pure and the sulfite solu- red.
tion is properly standardized. 7. To check the purity of the PRA, perform

8.2.12.2 Preparation of stock PRA solution— A the assay and adjustment of concentration
specially purified (99 to 100 percent pure) so- (Section 8.2.12.4) and prepare a reagent blank
lution of pararosaniline, which meets the (Section 11.2); the absorbance of this reagent
above specifications, is commercially avail- blank at 540 nm should be less than 0.170 at
able in the required 0.20 percent concentra- 22 °C. If the absorbance is greater than 0.170
tion (Harleco Co.). Alternatively, the dye under these conditions, further extractions
Fecha: 14/11/2013
may be purified, a stock solution prepared, should be performed.

and then assayed according to the procedure 8.2.12.4 PRA assay procedure— The con-
as described below.(10) centration of pararosaniline hydrochloride
8.2.12.3 Purification procedure for PRA— (PRA) need be assayed only once after purifi-
1. Place 100 mL each of 1-butanol and 1 N cation. It is also recommended that commer-
HCl in a large separatory funnel (250-mL) cial solutions of pararosaniline be assayed
and allow to equilibrate. Note: Certain when first purchased. The assay procedure is
batches of 1-butanol contain oxidants that as follows:(10)
create an SO2 demand. Before using, check 1. Prepare 1 M acetate-acetic acid buffer
by placing 20 mL of 1-butanol and 5 mL of 20 stock solution with a pH of 4.79 by dissolving
16
EC08NO91.004</MATH>
13.61 g of sodium acetate trihydrate in dis- cally described, the conditions must be ad-
tilled water in a 100-mL volumetric flask. justed so that linearity is maintained be-
Add 5.70 mL of glacial acetic acid and dilute tween absorbance and concentration over the
to volume with distilled water. dynamic range. Absorbing reagent volumes
2. Pipet 1 mL of the stock PRA solution ob- less than 10 mL are not recommended. The
tained from the purification process or from collection efficiency is above 98 percent for
a commercial source into a 100-mL volu- the conditions described; however, the effi-
metric flask and dilute to volume with dis- ciency may be substantially lower when
tilled water. sampling concentrations below 25µgSO2/
3. Transfer a 5–mL aliquot of the diluted m3.(8,9)
PRA solution from 2. into a 50–mL volu- 9.2 30-Minute and 1-Hour Sampling. Place 10
metric flask. Add 5mL of 1 M acetate-acetic mL of TCM absorbing reagent in a midget
acid buffer solution from 1. and dilute the impinger and seal the impinger with a thin
mixture to volume with distilled water. Let film of silicon stopcock grease (around the
the mixture stand for 1 hour. ground glass joint). Insert the sealed im-
4. Measure the absorbance of the above so- pinger into the sampling train as shown in
lution at 540 nm with a spectrophotometer Figure 1, making sure that all connections
against a distilled water reference. Compute between the various components are leak
the percentage of nominal concentration of tight. Greaseless ball joint fittings, heat
PRA by shrinkable Teflon tubing, or Teflon tube
fittings may be used to attain leakfree con-
A×K ditions for portions of the sampling train
%PRA = ( 5) that come into contact with air containing
W SO2. Shield the absorbing reagent from direct
where: sunlight by covering the impinger with alu-
minum foil or by enclosing the sampling
A = measured absorbance of the final mix-
train in a light-proof box. Determine the
ture (absorbance units);
flow rate according to Section 9.4.2. Collect
W = weight in grams of the PRA dye used in
the sample at 1±0.10 L/min for 30-minute
the assay to prepare 50 mL of stock solu-
sampling or 0.500±0.05 L/min for 1-hour sam-
tion (for example, 0.100 g of dye was used to
pling. Record the exact sampling time in
prepare 50 mL of solution in the purifi-
minutes, as the sample volume will later be
cation procedure; when obtained from com-
determined using the sampling flow rate and
mercial sources, use the stated concentra-
the sampling time. Record the atmospheric
tion to compute W; for 98% PRA, W=.098
pressure and temperature.
g.); and
9.3 24-Hour Sampling. Place 50 mL of TCM
K = 21.3 for spectrophotometers having a
absorbing solution in a large absorber, close
spectral bandwidth of less than 15 nm and
the cap, and, if needed, apply the heat shrink
a path length of 1 cm.
material as shown in Figure 3. Verify that
8.2.13 Pararosaniline reagent: To a 250–mL the reagent level is at the 50 mL mark on the
volumetric flask, add 20 mL of stock PRA so- absorber. Insert the sealed absorber into the
lution. Add an additional 0.2 mL of stock so- sampling train as shown in Figure 2. At this
lution for each percentage that the stock as- time verify that the absorber temperature is
says below 100 percent. Then add 25 mL of 3 controlled to 15±10 °C. During sampling, the
M phosphoric acid and dilute to volume with absorber temperature must be controlled to
distilled water. The reagent is stable for at prevent decomposition of the collected com-
least 9 months. Store away from heat and plex. From the onset of sampling until anal-
light. ysis, the absorbing solution must be pro-
9.0 Sampling Procedure. tected from direct sunlight. Determine the
9.1 General Considerations. Procedures are flow rate according to Section 9.4.2. Collect
described for short-term sampling (30-minute the sample for 24 hours from midnight to
and 1-hour) and for long-term sampling (24- midnight at a flow rate of 0.200±0.020 L/min.
hour). Different combinations of absorbing A start/stop timer is helpful for initiating
reagent volume, sampling rate, and sampling and stopping sampling and an elapsed time
time can be selected to meet special needs. meter will be useful for determining the
For combinations other than those specifi- sampling time.
Fecha: 14/11/2013
17
EC08NO91.005</MATH>
9.4 Flow Measurement. For 24-hour samples, the standard flow rate
9.4.1 Calibration: Flow measuring devices is determined at the time the absorber is
used for the on-site flow measurements re- placed in the sampling train and again when
quired in 9.4.2 must be calibrated against a the absorber is removed from the train for
reliable flow or volume standard such as an shipment to the analytical laboratory with a
NBS traceable bubble flowmeter or cali- calibrated flow measuring device connected
brated wet test meter. Rotameters or crit- to the inlet of the sampling train. The flow
ical orifices used in the sampling train may rate determination must be made with all
be calibrated, if desired, as a quality control components of the sampling system in oper-
check, but such calibration shall not replace ation (e.g., the absorber temperature con-
the on-site flow measurements required by troller and any sample box heaters must also
Fecha: 14/11/2013
9.4.2. In-line rotameters, if they are to be be operating). Equation 6 may be used to de-
calibrated, should be calibrated in situ, with termine the standard flow rate when a cali-
the appropriate volume of solution in the ab- brated positive displacement meter is used
sorber. as the flow measuring device. Other types of
9.4.2 Determination of flow rate at sampling calibrated flow measuring devices may also
site: For short-term samples, the standard be used to determine the flow rate at the
flow rate is determined at the sampling site sampling site provided that the user applies
at the initiation and completion of sample any appropriate corrections to devices for
collection with a calibrated flow measuring which output is dependent on temperature or
device connected to the inlet of the absorber. pressure.
18
50-708.eps</GPH>
Pb − (1 − RH)PH 2 O 298.16.
Q std = Q act × × (6)
Pstd (Tmeter + 27316
. ).
where: volume standards only, i.e., bubble flow-

Qstd = flow rate at standard conditions, std L/ meter or wet test meter; for dry standards,
min (25 °C and 760 mm Hg); i.e., dry test meter, PH2O=0);
Qact = flow rate at monitoring site conditions, Pstd = standard barometric pressure, in the
L/min; same units as Pb (760 mm Hg or 101 kPa);
Pb = barometric pressure at monitoring site and
conditions, mm Hg or kPa; Tmeter = temperature of the air in the flow or
volume standard, °C (e.g., bubble flow-
RH = fractional relative humidity of the air
meter).
being measured;
PH2O = vapor pressure of water at the tem- If a barometer is not available, the fol-
perature of the air in the flow or volume lowing equation may be used to determine
standard, in the same units as Pb, (for wet the barometric pressure:
Pb = 760 −.076(H) mm Hg, or Pb = 101−.01(H)kPa ( 7)
where: ture is above 10 °C. Store the sample at 5° ±5

H = sampling site elevation above sea level °C until it is analyzed.
in meters. 10.0 Analytical Calibration.
10.1 Spectrophotometer Cell Matching. If un-
If the initial flow rate (Qi) differs from the matched spectrophotometer cells are used,
flow rate of the critical orifice or the flow an absorbance correction factor must be de-
rate indicated by the flowmeter in the sam- termined as follows:
pling train (Qc) by more than 5 percent as de- 1. Fill all cells with distilled water and des-
termined by equation (8), check for leaks and ignate the one that has the lowest absorb-
redetermine Qi. ance at 548 nm as the reference. (This ref-
erence cell should be marked as such and
Qi − Qc continually used for this purpose throughout
% Diff = ×100 ( 8) all future analyses.)
Qc 2. Zero the spectrophotometer with the ref-
erence cell.
Invalidate the sample if the difference be- 3. Determine the absorbance of the remain-
tween the initial (Qi) and final (Qf) flow rates ing cells (Ac) in relation to the reference cell
is more than 5 percent as determined by and record these values for future use. Mark
equation (9): all cells in a manner that adequately identi-
EC08NO91.010</MATH>
fies the correction.
Qi − Qf
% Diff = ×100 ( 9) The corrected absorbance during future
analyses using each cell is determining as
Qf follows:
9.5 Sample Storage and Shipment. Remove

A = A obs − A c (10 )
EC08NO91.009</MATH>
the impinger or absorber from the sampling
train and stopper immediately. Verify that where:
the temperature of the absorber is not above
A = corrected absorbance,
25 °C. Mark the level of the solution with a
Aobs = uncorrected absorbance, and
temporary (e.g., grease pencil) mark. If the
Ac = cell correction.
sample will not be analyzed within 12 hours
Fecha: 14/11/2013
of sampling, it must be stored at 5° ±5 °C 10.2 Static Calibration Procedure (Option 1).

EC08NO91.008</MATH>
until analysis. Analysis must occur within 30 Prepare a dilute working sulfite-TCM solu-
days. If the sample is transported or shipped tion by diluting 10 mL of the working sul-
for a period exceeding 12 hours, it is rec- fite-TCM solution (Section 8.2.11) to 100 mL
ommended that thermal coolers using with TCM absorbing reagent. Following the
eutectic ice packs, refrigerated shipping con- table below, accurately pipet the indicated
tainers, etc., be used for periods up to 48 volumes of the sulfite-TCM solutions into a
EC08NO91.007</MATH>
hours. (17) Measure the temperature of the series of 25-mL volumetric flasks. Add TCM
absorber solution when the shipment is re- absorbing reagent as indicated to bring the
ceived. Invalidate the sample if the tempera- volume in each flask to 10 mL.
19
EC08NO91.006</MATH>
Volume timer that has been set for 30 minutes. Bring

Volume Total µg
of sulfite- all flasks to volume with recently boiled and
Sulfite-TCM solution of TCM, SO2
TCM cooled distilled water and mix thoroughly.
mL (approx.*
solution
The color must be developed (during the 30-
Working ......................... 4.0 6.0 28.8 minute period) in a temperature environ-
Working ......................... 3.0 7.0 21.6 ment in the range of 20° to 30 °C, which is
Working ......................... 2.0 8.0 14.4 controlled to ±1 °C. For increased precision,
Dilute working ................ 10.0 0.0 7.2 a constant temperature bath is rec-
Dilute working ................ 5.0 5.0 3.6 ommended during the color development
0.0 10.0 0.0
step. After 30 minutes, determine the cor-
*Based on working sulfite-TCM solution concentration of 7.2 rected absorbance of each standard at 548 nm
µg SO2/mL; the actual total µg SO2 must be calculated using against a distilled water reference (Section
equation 11 below.
10.1). Denote this absorbance as (A). Distilled
To each volumetric flask, add 1 mL 0.6% water is used in the reference cell rather
sulfamic acid (Section 8.2.1), accurately than the reagant blank because of the tem-
pipet 2 mL 0.2% formaldehyde solution (Sec- perature sensitivity of the reagent blank.
tion 8.2.2), then add 5 mL pararosaniline so- Calculate the total micrograms SO2 in each
lution (Section 8.2.13). Start a laboratory solution:
µg SO 2 = VTCM / SO 2 × C TCM / SO 2 × D (11)
where: a low flow of dry carrier gas to a mixing

VTCM/SO2 = volume of sulfite-TCM solution chamber where it is diluted with SO2-free air
used, mL; to the desired concentration and supplied to
CTCM/SO2 = concentration of sulfur dioxide in a vented manifold. A typical system is shown
the working sulfite-TCM, µg SO2/mL (from schematically in Figure 4 and this system
equation 4); and and other similar systems have been de-
D = dilution factor (D = 1 for the working scribed in detail by O’Keeffe and Ortman; (19)
sulfite-TCM solution; D = 0.1 for the di- Scaringelli, Frey, and Saltzman, (20) and
luted working sulfite-TCM solution). Scaringelli, O’Keeffe, Rosenberg, and Bell.
A calibration equation is determined using (21) Permeation devices may be prepared or
the method of linear least squares (Section purchased and in both cases must be trace-
12.1). The total micrograms SO2 contained in able either to a National Bureau of Stand-
each solution is the x variable, and the cor- ards (NBS) Standard Reference Material
rected absorbance (eq. 10) associated with (SRM 1625, SRM 1626, SRM 1627) or to an
each solution is the y variable. For the cali- NBS/EPA-approved commercially available
bration to be valid, the slope must be in the Certified Reference Material (CRM). CRM’s
range of 0.030 ±0.002 absorbance unit/µg SO2, are described in Reference 22, and a list of
the intercept as determined by the least CRM sources is available from the address
squares method must be equal to or less than shown for Reference 22. A recommended pro-
0.170 absorbance unit when the color is devel- tocol for certifying a permeation device to
oped at 22 °C (add 0.015 to this 0.170 specifica- an NBS SRM or CRM is given in Section 2.0.7
tion for each °C above 22 °C) and the correla- of Reference 2. Device permeation rates of 0.2
tion coefficient must be greater than 0.998. If to 0.4 µg/min, inert gas flows of about 50 mL/
these criteria are not met, it may be the re-
min, and dilution air flow rates from 1.1 to 15
sult of an impure dye and/or an improperly

L/min conveniently yield standard
standardized sulfite-TCM solution. A calibra-
atmospheres in the range of 25 to 600 µg SO2/
tion factor (Bs) is determined by calculating
m3 (0.010 to 0.230 ppm).
the reciprocal of the slope and is subse-
quently used for calculating the sample con- 10.3.1 Calibration Option 2A (30-minute and
centration (Section 12.3). 1-hour samples): Generate a series of six
10.3 Dynamic Calibration Procedures (Option standard atmospheres of SO2 (e.g., 0, 50, 100,
Fecha: 14/11/2013
2). Atmospheres containing accurately 200, 350, 500, 750 µg/m3) by adjusting the dilu-
known concentrations of sulfur dioxide are tion flow rates appropriately. The concentra-
prepared using permeation devices. In the tion of SO2 in each atmosphere is calculated
systems for generating these atmospheres, as follows:
the permeation device emits gaseous SO2 at
3
a known, low, constant rate, provided the Pr × 10
temperature of the device is held constant Ca = (12 )
ER31AU93.014</MATH>
(±0.1 °C) and the device has been accurately Qd + Qp

calibrated at the temperature of use. The
SO2 permeating from the device is carried by where:
20
EC08NO91.011</MATH>
Ca = concentration of SO2 at standard condi- Qd = flow rate of dilution air, std L/min; and
tions, µg/m3; Qp = flow rate of carrier gas across perme-
Pr = permeation rate, µg/min; ation device, std L/min.

Fecha: 14/11/2013
21
50-712.eps</GPH>
Be sure that the total flow rate of the
standard exceeds the flow demand of the Vb × C s
t= (14 )
sample train, with the excess flow vented at −3
atmospheric pressure. Sample each atmos- C a × Q s × 10
phere using similar apparatus as shown in where:
Figure 1 and under the same conditions as t = sampling time, min;
field sampling (i.e., use same absorbing rea- Vb = volume of absorbing solution used for
gent volume and sample same volume of air sampling (50 mL);
at an equivalent flow rate). Due to the Cs = desired concentration of SO2 in the ab-
length of the sampling periods required, this sorbing solution, µg/mL;
method is not recommended for 24-hour sam- Ca = concentration of the standard atmos-
pling. At the completion of sampling, quan- phere calculated according to equation 12,
titatively transfer the contents of each im- µg/m3 ; and
pinger to one of a series of 25-mL volumetric Qs = sampling flow rate, std L/min.
flasks (if 10 mL of absorbing solution was At the completion of sampling, bring the
used) using small amounts of distilled water absorber solutions to original volume with
for rinse (<5mL). If >10 mL of absorbing solu- distilled water. Pipet a 10-mL portion from
tion was used, bring the absorber solution in each absorber into one of a series of 25-mL
each impinger to orginal volume with dis- volumetric flasks. If the color development
tilled H2 O and pipet 10-mL portions from steps are not to be started within 12 hours of
each impinger into a series of 25-mL volu- sampling, store the solutions at 5° ± 5 °C. Add
metric flasks. If the color development steps the remaining reagents for color develop-
are not to be started within 12 hours of sam- ment in the same manner as in Section 10.2
pling, store the solutions at 5° ± 5 °C. Cal- for static solutions. Calculate the total µg
culate the total micrograms SO2 in each so- SO2 in each standard as follows:
lution as follows: −3
C a × Q s × t × Va × 10
−3 µgSO 2 = (15)
C a × Q s × t × Va × 10
µgSO 2 = (13) Vb
Vb
where:
where: Va = volume of absorbing solution used for
Ca = concentration of SO2 in the standard at- color development (10 mL).
mosphere, µg/m3 ; All other parameters are defined in equation
Os = sampling flow rate, std L/min; 14.
t=sampling time, min; Calculate a calibration equation and a
Va = volume of absorbing solution used for calibration factor (Bt) according to Section
color development (10 mL); and 10.2 adhering to all the specified criteria.
Vb = volume of absorbing solution used for 11.0 Sample Preparation and Analysis.
sampling, mL. 11.1 Sample Preparation. Remove the sam-
ples from the shipping container. If the ship-
Add the remaining reagents for color de- ment period exceeded 12 hours from the com-
velopment in the same manner as in Section pletion of sampling, verify that the tempera-
10.2 for static solutions. Calculate a calibra- ture is below 10 °C. Also, compare the solu-
tion equation and a calibration factor (Bg) tion level to the temporary level mark on
according to Section 10.2, adhering to all the the absorber. If either the temperature is
specified criteria. above 10 °C or there was significant loss
10.3.2 Calibration Option 2B (24-hour sam- (more than 10 mL) of the sample during ship-
ples): Generate a standard atmosphere con- ping, make an appropriate notation in the
taining approximately 1,050 µg SO2/m3 and record and invalidate the sample. Prepare
calculate the exact concentration according the samples for analysis as follows:
to equation 12. Set up a series of six absorb- 1. For 30-minute or 1-hour samples: Quan-
ers according to Figure 2 and connect to a titatively transfer the entire 10 mL amount
of absorbing solution to a 25-mL volumetric
common manifold for sampling the standard
flask and rinse with a small amount (<5 mL)
Fecha: 14/11/2013
atmosphere. Be sure that the total flow rate

of distilled water.
ER31AU93.017</MATH>
of the standard exceeds the flow demand at 2. For 24-hour samples: If the volume of the
the sample manifold, with the excess flow sample is less than the original 50-mL vol-
vented at atmospheric pressure. The absorb- ume (permanent mark on the absorber), ad-
ers are then allowed to sample the atmos- just the volume back to the original volume
phere for varying time periods to yield solu- with distilled water to compensate for water
tions containing 0, 0.2, 0.6, 1.0, 1.4, 1.8, and 2.2 lost to evaporation during sampling. If the
ER31AU93.016</MATH>
µg SO2/mL solution. The sampling times re- final volume is greater than the original vol-
quired to attain these solution concentra- ume, the volume must be measured using a
tions are calculated as follows: graduated cylinder. To analyze, pipet 10 mL
22
ER31AU93.015</MATH>
of the solution into a 25-mL volumetric is used to calculate a calibration equation in
flask. the form of:
11.2 Sample Analysis. For each set of deter-
minations, prepare a reagent blank by add- y = mx + b (16 )
ing 10 mL TCM absorbing solution to a 25-
where:
mL volumetric flask, and two control stand-
ards containing approximately 5 and 15 µg SO2, y = corrected absorbance,
respectively. The control standards are pre- m = slope, absorbance unit/µg SO2,
pared according to Section 10.2 or 10.3. The x = micrograms of SO2,
analysis is carried out as follows: b = y intercept (absorbance units).
1. Allow the sample to stand 20 minutes The slope (m), intercept (b), and correla-
after the completion of sampling to allow tion coefficient (r) are calculated as follows:
any ozone to decompose (if applicable).
2. To each 25-mL volumetric flask con- n ∑ xy − ( ∑ x)( ∑ y)
taining reagent blank, sample, or control m= (17)
standard, add 1 mL of 0.6% sulfamic acid n ∑ x 2 − ( ∑ x )2
(Section 8.2.1) and allow to react for 10 min.
3. Accurately pipet 2 mL of 0.2% formalde- ∑y − m∑x
b= (18 )
hyde solution (Section 8.2.2) and then 5 mL
n
of pararosaniline solution (Section 8.2.13)
into each flask. Start a laboratory timer set
at 30 minutes. m ( ∑ xy − ∑ x ∑ y / n )
r= (19 )
4. Bring each flask to volume with recently 2 2
∑ y − (∑ y) / n
boiled and cooled distilled water and mix
thoroughly. where n is the number of calibration points.
5. During the 30 minutes, the solutions A data form (Figure 5) is supplied for eas-
must be in a temperature controlled environ- ily organizing calibration data when the
ment in the range of 20° to 30 °C maintained slope, intercept, and correlation coefficient
to ±1 °C. This temperature must also be with- are calculated by hand.
in 1 °C of that used during calibration. 12.2 Total Sample Volume. Determine the
6. After 30 minutes and before 60 minutes, sampling volume at standard conditions as
determine the corrected absorbances (equa- follows:
tion 10) of each solution at 548 nm using 1-cm
optical path length cells against a distilled Qi + Qf
Vstd = ×t ( 20 )
water reference (Section 10.1). (Distilled water
is used as a reference instead of the reagent 2
blank because of the sensitivity of the reagent where:
blank to temperature.) Vstd = sampling volume in std L,
7. Do not allow the colored solution to Qi = standard flow rate determined at the
ER31AU93.022</MATH>
stand in the cells because a film may be de- initiation of sampling in std L/min,
posited. Clean the cells with isopropyl alco- Qf = standard flow rate determined at the
hol after use. completion of sampling is std L/min, and
8. The reagent blank must be within 0.03 t = total sampling time, min.
absorbance units of the intercept of the cali-
bration equation determined in Section 10. 12.3 Sulfur Dioxide Concentration. Calculate
and report the concentration of each sample
ER31AU93.021</MATH>
11.3 Absorbance range. If the absorbance of
the sample solution ranges between 1.0 and as follows:
2.0, the sample can be diluted 1:1 with a por- 3
tion of the reagent blank and the absorbance 3 ( A − A o )( B x )(10 ) Vb
redetermined within 5 minutes. Solutions µg SO 2 /m = × ( 21)
with higher absorbances can be diluted up to Vstd Va

sixfold with the reagent blank in order to ob-
tain scale readings of less than 1.0 absorb- where: ER31AU93.020</MATH>
ance unit. However, it is recommended that A = corrected absorbance of the sample solu-
a smaller portion (<10 mL) of the original tion, from equation (10);
sample be reanalyzed (if possible) if the sam- Ao = corrected absorbance of the reagent
ple requires a dilution greater than 1:1. blank, using equation (10);
Fecha: 14/11/2013
11.4 Reaqent disposal. All reagents con- Bx = calibration factor equal to Bs, Bg, or Bt
ER31AU93.018</MATH> ER31AU93.019</MATH>
taining mercury compounds must be stored depending on the calibration procedure

and disposed of using one of the procedures used, the reciprocal of the slope of the cali-
contained in Section 13. Until disposal, the bration equation;
discarded solutions can be stored in closed Va = volume of absorber solution analyzed,
glass containers and should be left in a fume mL;
hood. Vb = total volume of solution in absorber (see
12.0 Calculations. 11.1–2), mL; and
12.1 Calibration Slope, Intercept, and Correla- Vstd = standard air volume sampled, std L
tion Coefficient. The method of least squares (from Section 12.2).
23
EC08NO91.012</MATH>
DATA FORM 5. After 24 hours, allow the solution to

[For hand calculations] stand without stirring to allow the mercury
amalgam (solid black material) to settle to
Calibra- the bottom of the waste receptacle.
Micro- Absor-
tion point grams So 6. Upon settling, decant and discard the su-
no. 2 bance units
pernatant liquid.
(x) (y) x2 xy y2 7. Quantitatively transfer the solid mate-
1 ........... .................. ................... .......... .......... ........ rial to a container and allow to dry.
2 ........... .................. ................... .......... .......... ........ 8. The solid material can be sent to a mer-
3 ........... .................. ................... .......... .......... ........ cury reclaiming plant. It must not be dis-
4 ........... .................. ................... .......... .......... ........ carded.
5 ........... .................. ................... .......... .......... ........ 13.3 Method Using Aluminum Foil Strips.
6 ........... .................. ................... .......... .......... ........ 1. Place the waste solution in an uncapped
vessel in a hood.
S x=lll S y=lll S x =lll Sxylll 2
2. For each liter of waste solution, add ap-
Sy2lll proximately 10 g of aluminum foil strips. If
n=lll (number of pairs of coordinates.) all the aluminum is consumed and no gas is
llllllllllllllllllllllll evolved, add an additional 10 g of foil. Repeat
FIGURE 5. Data form for hand calculations. until the foil is no longer consumed and
allow the gas to evolve for 24 hours.
12.4 Control Standards. Calculate the ana- 3. Decant the supernatant liquid and dis-
lyzed micrograms of SO2 in each control card.
standard as follows: 4. Transfer the elemental mercury that has
( )
settled to the bottom of the vessel to a stor-
C q = A − A o × Bx (22) age container.
5. The mercury can be sent to a mercury
where: reclaiming plant. It must not be discarded.
Cq = analyzed µg SO2 in each control stand- 14.0 References for SO2 Method.
ard, 1. Quality Assurance Handbook for Air Pol-
A = corrected absorbance of the control lution Measurement Systems, Volume I,
standard, and Principles. EPA–600/9–76–005, U.S. Environ-
Ao = corrected absorbance of the reagent mental Protection Agency, Research Tri-
blank. angle Park, NC 27711, 1976.
2. Quality Assurance Handbook for Air Pol-
The difference between the true and ana-
lution Measurement Systems, Volume II,
lyzed values of the control standards must
Ambient Air Specific Methods. EPA–600/4–77–
not be greater than 1 µg. If the difference is
027a, U.S. Environmental Protection Agency,
greater than 1 µg, the source of the discrep-
Research Triangle Park, NC 27711, 1977.
ancy must be identified and corrected.
3. Dasqupta, P. K., and K. B. DeCesare. Sta-
12.5 Conversion of µg/m3 to ppm (v/v). If de- bility of Sulfur Dioxide in Formaldehyde and
sired, the concentration of sulfur dioxide at
Its Anomalous Behavior in
reference conditions can be converted to ppm
Tetrachloromercurate (II). Submitted for
SO2 (v/v) as follows:
publication in Atmospheric Environment, 1982.
µg SO 2 4. West, P. W., and G. C. Gaeke. Fixation of
−4
ppm SO 2 = × 3.82 × 10 ( 23) Sulfur Dioxide as Disulfitomercurate (II) and
3 Subsequent Colorimetric Estimation. Anal.
m
Chem., 28:1816, 1956.
13.0 The TCM absorbing solution and any 5. Ephraim, F. Inorganic Chemistry. P. C.
reagents containing mercury compounds L. Thorne and E. R. Roberts, Eds., 5th Edi-
must be treated and disposed of by one of the tion, Interscience, 1948, p. 562.
methods discussed below. Both methods re- 6. Lyles, G. R., F. B. Dowling, and V. J.
move greater than 99.99 percent of the mer- Blanchard. Quantitative Determination of
cury. Formaldehyde in the Parts Per Hundred Mil-
13.1 Disposal of Mercury-Containing Solu- lion Concentration Level. J. Air. Poll. Cont.
tions. Assoc., Vol. 15(106), 1965.
13.2 Method for Forming an Amalgam. 7. McKee, H. C., R. E. Childers, and O.
1. Place the waste solution in an uncapped Saenz, Jr. Collaborative Study of Reference
Fecha: 14/11/2013
vessel in a hood. Method for Determination of Sulfur Dioxide

2. For each liter of waste solution, add ap- in the Atmosphere (Pararosaniline Method).
proximately 10 g of sodium carbonate until EPA–APTD–0903, U.S. Environmental Pro-
neutralization has occurred (NaOH may have tection Agency, Research Triangle Park, NC
to be used). 27711, September 1971.
3. Following neutralization, add 10 g of 8. Urone, P., J. B. Evans, and C. M. Noyes.
granular zinc or magnesium. Tracer Techniques in Sulfur—Air Pollution
ER31AU93.023</MATH>
4. Stir the solution in a hood for 24 hours. Studies Apparatus and Studies of Sulfur Di-
Caution must be exercised as hydrogen gas is oxide Colorimetric and Conductometric
evolved by this treatment process. Methods. Anal. Chem., 37: 1104, 1965.
24
EC08NO91.070</MATH>
Environmental Protection Agency Pt. 50, App. B
9. Bostrom, C. E. The Absorption of Sulfur (MD–77), Research Triangle Park, NC 27711,
Dioxide at Low Concentrations (pphm) Stud- January 1981.
ied by an Isotopic Tracer Method. Intern. J.
[47 FR 54899, Dec. 6, 1982; 48 FR 17355, Apr. 22,
Air Water Poll., 9:333, 1965. 1983]
10. Scaringelli, F. P., B. E. Saltzman, and
S. A. Frey. Spectrophotometric Determina- APPENDIX B TO PART 50—REFERENCE
tion of Atmospheric Sulfur Dioxide. Anal. METHOD FOR THE DETERMINATION OF
Chem., 39: 1709, 1967.
SUSPENDED PARTICULATE MATTER IN
11. Pate, J. B., B. E. Ammons, G. A. Swan-
THE ATMOSPHERE (HIGH-VOLUME
son, and J. P. Lodge, Jr. Nitrite Interference
in Spectrophotometric Determination of At- METHOD)
mospheric Sulfur Dioxide. Anal. Chem., 1.0 Applicability.
37:942, 1965. 1.1 This method provides a measurement of
12. Zurlo, N., and A. M. Griffini. Measure- the mass concentration of total suspended
ment of the Sulfur Dioxide Content of the particulate matter (TSP) in ambient air for
Air in the Presence of Oxides of Nitrogen and determining compliance with the primary
Heavy Metals. Medicina Lavoro, 53:330, 1962. and secondary national ambient air quality
13. Rehme, K. A., and F. P. Scaringelli. Ef- standards for particulate matter as specified
fect of Ammonia on the Spectrophotometric in § 50.6 and § 50.7 of this chapter. The meas-
Determination of Atmospheric Concentra- urement process is nondestructive, and the
tions of Sulfur Dioxide. Anal. Chem., 47:2474, size of the sample collected is usually ade-
1975. quate for subsequent chemical analysis.
14. McCoy, R. A., D. E. Camann, and H. C. Quality assurance procedures and guidance
McKee. Collaborative Study of Reference are provided in part 58, appendixes A and B,
Method for Determination of Sulfur Dioxide of this chapter and in References 1 and 2.
in the Atmosphere (Pararosaniline Method) 2.0 Principle.
(24-Hour Sampling). EPA–650/4–74–027, U.S. 2.1 An air sampler, properly located at the
Environmental Protection Agency, Research measurement site, draws a measured quan-
Triangle Park, NC 27711, December 1973. tity of ambient air into a covered housing
15. Fuerst, R. G. Improved Temperature and through a filter during a 24-hr (nominal)
Stability of Sulfur Dioxide Samples Col- sampling period. The sampler flow rate and
lected by the Federal Reference Method. the geometry of the shelter favor the collec-
EPA–600/4–78–018, U.S. Environmental Pro- tion of particles up to 25–50 µm (aerodynamic
tection Agency, Research Triangle Park, NC diameter), depending on wind speed and di-
27711, April 1978. rection.(3) The filters used are specified to
16. Scaringelli, F. P., L. Elfers, D. Norris, have a minimum collection efficiency of 99
and S. Hochheiser. Enhanced Stability of percent for 0.3 µm (DOP) particles (see Sec-
Sulfur Dioxide in Solution. Anal. Chem., tion 7.1.4).
42:1818, 1970. 2.2 The filter is weighed (after moisture
equilibration) before and after use to deter-
17. Martin, B. E. Sulfur Dioxide Bubbler
mine the net weight (mass) gain. The total
Temperature Study. EPA–600/4–77–040, U.S.
volume of air sampled, corrected to EPA
Environmental Protection Agency, Research
standard conditions (25 °C, 760 mm Hg [101
Triangle Park, NC 27711, August 1977.
kPa]), is determined from the measured flow
18. American Society for Testing and Mate-
rate and the sampling time. The concentra-
rials. ASTM Standards, Water; Atmospheric
tion of total suspended particulate matter in
Analysis. Part 23. Philadelphia, PA, October
the ambient air is computed as the mass of
1968, p. 226.
collected particles divided by the volume of
19. O’Keeffe, A. E., and G. C. Ortman. Pri- air sampled, corrected to standard condi-
mary Standards for Trace Gas Analysis. tions, and is expressed in micrograms per
Anal. Chem., 38:760, 1966. standard cubic meter (µg/std m3). For sam-
20. Scaringelli, F. P., S. A. Frey, and B. E. ples collected at temperatures and pressures
Saltzman. Evaluation of Teflon Permeation significantly different than standard condi-
Tubes for Use with Sulfur Dioxide. Amer. tions, these corrected concentrations may
Ind. Hygiene Assoc. J., 28:260, 1967. differ substantially from actual concentra-
21. Scaringelli, F. P., A. E. O’Keeffe, E. tions (micrograms per actual cubic meter),
Fecha: 14/11/2013
Rosenberg, and J. P. Bell, Preparation of particularly at high elevations. The actual

Known Concentrations of Gases and Vapors particulate matter concentration can be cal-
With Permeation Devices Calibrated Gravi- culated from the corrected concentration
metrically. Anal. Chem., 42:871, 1970. using the actual temperature and pressure
22. A Procedure for Establishing during the sampling period.
Traceability of Gas Mixtures to Certain Na- 3.0 Range.
tional Bureau of Standards Standard Ref- 3.1 The approximate concentration range
erence Materials. EPA–600/7–81–010, U.S. En- of the method is 2 to 750 µg/std m3. The upper
vironmental Protection Agency, Environ- limit is determined by the point at which the
mental Monitoring Systems Laboratory sampler can no longer maintain the specified
25
Pt. 50, App. B 40 CFR Ch. I (7–1–03 Edition)
flow rate due to the increased pressure drop using a calibrated, continuous flow rate re-
of the loaded filter. This point is affected by cording device to record the actual flow rate
particle size distribution, moisture content during the samping period and integrating
of the collected particles, and variability the flow rate over the period, or (3) any other
from filter to filter, among other things. The means that will accurately measure the
lower limit is determined by the sensitivity total air volume sampled during the sam-
of the balance (see Section 7.10) and by in- pling period. Use of a continuous flow re-
herent sources of error (see Section 6). corder is recommended, particularly if the
3.2 At wind speeds between 1.3 and 4.5 m/sec sampler is not equipped with a constant flow
(3 and 10 mph), the high-volume air sampler controller.
has been found to collect particles up to 25 to 6.3 Loss of volatiles. Volatile particles col-
50 µm, depending on wind speed and direc- lected on the filter may be lost during subse-
tion.(3) For the filter specified in Section 7.1, quent sampling or during shipment and/or
there is effectively no lower limit on the par- storage of the filter prior to the
ticle size collected. postsampling weighing.(5) Although such
4.0 Precision. losses are largely unavoidable, the filter
4.1 Based upon collaborative testing, the should be reweighed as soon after sampling
relative standard deviation (coefficient of as practical.
variation) for single analyst precision (re- 6.4 Artifact particulate matter. Artifact par-
peatability) of the method is 3.0 percent. The ticulate matter can be formed on the surface
corresponding value for interlaboratory pre- of alkaline glass fiber filters by oxidation of
cision (reproducibility) is 3.7 percent.(4) acid gases in the sample air, resulting in a
5.0 Accuracy. higher than true TSP determination.(6 7)
5.1 The absolute accuracy of the method is This effect usually occurs early in the sam-
undefined because of the complex nature of ple period and is a function of the filter pH
atmospheric particulate matter and the dif- and the presence of acid gases. It is generally
ficulty in determining the ‘‘true’’ particulate believed to account for only a small percent-
matter concentration. This method provides age of the filter weight gain, but the effect
a measure of particulate matter concentra- may become more significant where rel-
tion suitable for the purpose specified under atively small particulate weights are col-
Section 1.0, Applicability. lected.
6.0 Inherent Sources of Error. 6.5 Humidity. Glass fiber filters are com-
6.1 Airflow variation. The weight of mate- paratively insensitive to changes in relative
rial collected on the filter represents the (in- humidity, but collected particulate matter
tegrated) sum of the product of the instanta- can be hygroscopic.(8) The moisture condi-
neous flow rate times the instantaneous par- tioning procedure minimizes but may not
ticle concentration. Therefore, dividing this completely eliminate error due to moisture.
weight by the average flow rate over the 6.6 Filter handling. Careful handling of the
sampling period yields the true particulate filter between the presampling and
matter concentration only when the flow postsampling weighings is necessary to avoid
rate is constant over the period. The error errors due to loss of fibers or particles from
resulting from a nonconstant flow rate de- the filter. A filter paper cartridge or cassette
pends on the magnitude of the instantaneous used to protect the filter can minimize han-
changes in the flow rate and in the particu- dling errors. (See Reference 2, Section 2).
late matter concentration. Normally, such 6.7 Nonsampled particulate matter. Particu-
errors are not large, but they can be greatly late matter may be deposited on the filter by
reduced by equipping the sampler with an wind during periods when the sampler is in-
automatic flow controlling mechanism that operative. (9) It is recommended that errors
maintains constant flow during the sampling from this source be minimized by an auto-
period. Use of a contant flow controller is matic mechanical device that keeps the fil-
recommended.* ter covered during nonsampling periods, or
6.2 Air volume measurement. If the flow rate by timely installation and retrieval of filters
changes substantially or nonuniformly dur- to minimize the nonsampling periods prior
ing the sampling period, appreciable error in to and following operation.
the estimated air volume may result from 6.8 Timing errors. Samplers are normally
using the average of the presampling and controlled by clock timers set to start and
Fecha: 14/11/2013
postsampling flow rates. Greater air volume stop the sampler at midnight. Errors in the
measurement accuracy may be achieved by nominal 1,440-min sampling period may re-
(1) equipping the sampler with a flow consult from a power interruption during the
trolling mechanism that maintains constant sampling period or from a discrepancy be-
air flow during the sampling period,* (2) tween the start or stop time recorded on the
filter information record and the actual
*At elevated altitudes, the effectiveness of start or stop time of the sampler. Such dis-
automatic flow controllers may be reduced crepancies may be caused by (1) poor resolu-
because of a reduction in the maximum sam- tion of the timer set-points, (2) timer error
pler flow. due to power interruption, (3) missetting of
26
the timer, or (4) timer malfunction. In gen- d. (†) Manually adjust the flow rate to ac-
eral, digital electronic timers have much commodate variations in filter pressure drop
better set-point resolution than mechanical and site line voltage and altitude. The ad-
timers, but require a battery backup system justment may be accomplished by an auto-
to maintain continuity of operation after a matic flow controller or by a manual flow
power interruption. A continuous flow re- adjustment device. Any manual adjustment
corder or elapsed time meter provides an in- device must be designed with positive
dication of the sampler run-time, as well as detents or other means to avoid uninten-
indication of any power interruption during tional changes in the setting.
the sampling period and is therefore rec- 7.2.2 Minimum sample flow rate, heavily load-
ommended. ed filter: 1.1 m3/min (39 ft3/min).‡
6.9 Recirculation of sampler exhaust. Under 7.2.3 Maximum sample flow rate, clean filter:
stagnant wind conditions, sampler exhaust 1.7 m3/min (60 ft3/min).‡
air can be resampled. This effect does not ap- 7.2.4 Blower Motor: The motor must be ca-
pear to affect the TSP measurement sub- pable of continuous operation for 24-hr peri-
stantially, but may result in increased car- ods.
bon and copper in the collected sample. (10) 7.3 Sampler shelter.
This problem can be reduced by ducting the 7.3.1 The sampler shelter shall:
exhaust air well away, preferably downwind, a. Maintain the filter in a horizontal posi-
from the sampler. tion at least 1 m above the sampler sup-
7.0 Apparatus. porting surface so that sample air is drawn
(See References 1 and 2 for quality assur- downward through the filter.
ance information.) b. Be rectangular in shape with a gabled
roof, similar to the design shown in Figure 1.
NOTE: Samplers purchased prior to the ef-
c. Cover and protect the filter and sampler
fective date of this amendment are not sub-
from precipitation and other weather.
ject to specifications preceded by (†).
d. Discharge exhaust air at least 40 cm
7.1 Filter. (Filters supplied by the Environ- from the sample air inlet.
mental Protection Agency can be assumed to e. Be designed to minimize the collection
meet the following criteria. Additional speci- of dust from the supporting surface by incor-
fications are required if the sample is to be porating a baffle between the exhaust outlet
analyzed chemically.) and the supporting surface.
7.1.1 Size: 20.3 ± 0.2 × 25.4 ± 0.2 cm (nominal 7.3.2 The sampler cover or roof shall over-
8 × 10 in). hang the sampler housing somewhat, as
7.1.2 Nominal exposed area: 406.5 cm2 (63 in2). shown in Figure 1, and shall be mounted so
7.1.3. Material: Glass fiber or other rel- as to form an air inlet gap between the cover
atively inert, nonhygroscopic material. (8) and the sampler housing walls. † This sample
7.1.4 Collection efficiency: 99 percent min- air inlet should be approximately uniform on
imum as measured by the DOP test (ASTM– all sides of the sampler. † The area of the
2986) for particles of 0.3 µm diameter. sample air inlet must be sized to provide an
7.1.5 Recommended pressure drop range: 42–54 effective particle capture air velocity of be-
mm Hg (5.6–7.2 kPa) at a flow rate of 1.5 std tween 20 and 35 cm/sec at the recommended
m3/min through the nominal exposed area. operational flow rate. The capture velocity
7.1.6 pH: 6 to 10. (11) is the sample air flow rate divided by the
7.1.7 Integrity: 2.4 mg maximum weight inlet area measured in a horizontal plane at
loss. (11) the lower edge of the cover. † Ideally, the
7.1.8 Pinholes: None. inlet area and operational flow rate should
be selected to obtain a capture air velocity
7.1.9 Tear strength: 500 g minimum for 20
of 25 ±2 cm/sec.
mm wide strip cut from filter in weakest di-
7.4 Flow rate measurement devices.

mension. (See ASTM Test D828–60).
7.4.1 The sampler shall incorporate a flow
7.1.10 Brittleness: No cracks or material sep- rate measurement device capable of indi-
arations after single lengthwise crease. cating the total sampler flow rate. Two com-
7.2 Sampler. The air sampler shall provide mon types of flow indicators covered in the
means for drawing the air sample, via recalibration procedure are (1) an electronic
duced pressure, through the filter at a uni- mass flowmeter and (2) an orifice or orifices
Fecha: 14/11/2013
form face velocity.

7.2.1 The sampler shall have suitable means
(†) See note at beginning of Section 7 of
to:
this appendix.
a. Hold and seal the filter to the sampler
‡ These specifications are in actual air vol-
housing.
ume units; to convert to EPA standard air
b. Allow the filter to be changed conven- volume units, multiply the specifications by
iently. (Pb/Pstd)(298/T) where Pb and T are the baro-
c. Preclude leaks that would cause error in metric pressure in mm Hg (or kPa) and the
the measurement of the air volume passing temperature in K at the sampler, and Pstd is
through the filter. 760 mm Hg (or 101 kPa).
27
located in the sample air stream together without disassembly of the unit. Use of a
with a suitable pressure indicator such as a conventional, orifice-type transfer standard
manometer, or aneroid pressure gauge. A is assumed in the calibration procedure (Sec-
pressure recorder may be used with an ori- tion 9). However, the use of other types of
fice to provide a continuous record of the transfer standards meeting the above speci-
flow. Other types of flow indicators (includ- fications, such as the one shown in Figure 2c,
ing rotameters) having comparable precision may be approved; see the note following Sec-
and accuracy are also acceptable. tion 9.1.
7.4.2 † The flow rate measurement device 7.9 Filter conditioning environment
must be capable of being calibrated and read 7.9.1 Controlled temperature: between 15° and
in units corresponding to a flow rate which 30 °C with less than ±3 °C variation during
is readable to the nearest 0.02 std m3/min equilibration period.
over the range 1.0 to 1.8 std m3/min. 7.9.2 Controlled humidity: Less than 50 per-
7.5 Thermometer, to indicate the approxi- cent relative humidity, constant within ±5
mate air temperature at the flow rate meas- percent.
urement orifice, when temperature correc- 7.10 Analytical balance.
tions are used. 7.10.1 Sensitivity: 0.1 mg.
7.5.1 Range: ¥40° to +50 °C (223–323 K). 7.10.2 Weighing chamber designed to accept
7.5.2 Resolution: 2 °C (2 K). an unfolded 20.3×25.4 cm (8×10 in) filter.
7.6 Barometer, to indicate barometric pres- 7.11 Area light source, similar to X-ray film
sure at the flow rate measurement orifice, viewer, to backlight filters for visual inspec-
when pressure corrections are used. tion.
7.6.1 Range: 500 to 800 mm Hg (66–106 kPa). 7.12 Numbering device, capable of printing
7.6.2 Resolution: ±5 mm Hg (0.67 kPa). identification numbers on the filters before
7.7 Timing/control device. they are placed in the filter conditioning en-
7.7.1 The timing device must be capable of vironment, if not numbered by the supplier.
starting and stopping the sampler to obtain 8.0 Procedure.
an elapsed run-time of 24 hr ±1 hr (1,440 ±60 (See References 1 and 2 for quality assur-
min). ance information.)
7.7.2 Accuracy of time setting: ±30 min, or 8.1 Number each filter, if not already num-
better. (See Section 6.8). bered, near its edge with a unique identifica-
7.8 Flow rate transfer standard, traceable to tion number.
a primary standard. (See Section 9.2.) 8.2 Backlight each filter and inspect for
7.8.1 Approximate range: 1.0 to 1.8 m3/min. pinholes, particles, and other imperfections;
7.8.2 Resolution: 0.02 m3/min. filters with visible imperfections must not
7.8.3 Reproducibility: ±2 percent (2 times co- be used.
efficient of variation) over normal ranges of 8.3 Equilibrate each filter in the condi-
ambient temperature and pressure for the tioning environment for at least 24-hr.
stated flow rate range. (See Reference 2, Sec- 8.4 Following equilibration, weigh each fil-
tion 2.) ter to the nearest milligram and record this
7.8.4 Maximum pressure drop at 1.7 std m3/ tare weight (Wi) with the filter identification
min; 50 cm H2 O (5 kPa). number.
7.8.5 The flow rate transfer standard must 8.5 Do not bend or fold the filter before col-
connect without leaks to the inlet of the lection of the sample.
sampler and measure the flow rate of the 8.6 Open the shelter and install a num-
total air sample. bered, preweighed filter in the sampler, fol-
7.8.6 The flow rate transfer standard must lowing the sampler manufacturer’s instruc-
include a means to vary the sampler flow tions. During inclement weather, pre-
rate over the range of 1.0 to 1.8 m3/min (35– cautions must be taken while changing fil-
64 ft3/min) by introducing various levels of ters to prevent damage to the clean filter
flow resistance between the sampler and the and loss of sample from or damage to the ex-
transfer standard inlet. posed filter. Filter cassettes that can be
7.8.7 The conventional type of flow transfer loaded and unloaded in the laboratory may
standard consists of: An orifice unit with be used to minimize this problem (See Sec-
adapter that connects to the inlet of the tion 6.6).
sampler, a manometer or other device to 8.7 Close the shelter and run the sampler
measure orifice pressure drop, a means to for at least 5 min to establish run-tempera-
Fecha: 14/11/2013
vary the flow through the sampler unit, a ture conditions.

thermometer to measure the ambient tem- 8.8 Record the flow indicator reading and,
perature, and a barometer to measure ambi- if needed, the barometric pressure (P 3) and
ent pressure. Two such devices are shown in the ambient temperature (T 3) see NOTE fol-
Figures 2a and 2b. Figure 2a shows multiple lowing step 8.12). Stop the sampler. Deter-
fixed resistance plates, which necessitate mine the sampler flow rate (see Section 10.1);
disassembly of the unit each time the flow if it is outside the acceptable range (1.1 to 1.7
resistance is changed. A preferable design, il- m3/min [39–60 ft3/min]), use a different filter,
lustrated in Figure 2b, has a variable flow re- or adjust the sampler flow rate. Warning:
striction that can be adjusted externally Substantial flow adjustments may affect the
28
calibration of the orifice-type flow indica- 9.1 Calibration of the high volume sam-
tors and may necessitate recalibration. pler’s flow indicating or control device is
8.9 Record the sampler identification infor- necessary to establish traceability of the
mation (filter number, site location or iden- field measurement to a primary standard via
tification number, sample date, and starting a flow rate transfer standard. Figure 3a illus-
time). trates the certification of the flow rate
8.10 Set the timer to start and stop the transfer standard and Figure 3b illustrates
sampler such that the sampler runs 24-hrs, its use in calibrating a sampler flow indi-
from midnight to midnight (local time). cator. Determination of the corrected flow
8.11 As soon as practical following the sam- rate from the sampler flow indicator, illus-
pling period, run the sampler for at least 5 trated in Figure 3c, is addressed in Section
min to again establish run-temperature con- 10.1
ditions. NOTE: The following calibration procedure
8.12 Record the flow indicator reading and, applies to a conventional orifice-type flow
if needed, the barometric pressure (P 3) and transfer standard and an orifice-type flow in-
the ambient temperature (T 3). dicator in the sampler (the most common
NOTE: No onsite pressure or temperature types). For samplers using a pressure re-
measurements are necessary if the sampler corder having a square-root scale, 3 other ac-
flow indicator does not require pressure or ceptable calibration procedures are provided
temperature corrections (e.g., a mass flow- in Reference 12. Other types of transfer
standards may be used if the manufacturer
meter) or if average barometric pressure and
or user provides an appropriately modified
seasonal average temperature for the site are
calibration procedure that has been approved
incorporated into the sampler calibration
by EPA under Section 2.8 of appendix C to
(see step 9.3.9). For individual pressure and
part 58 of this chapter.
temperature corrections, the ambient pres-
9.2 Certification of the flow rate transfer
sure and temperature can be obtained by on-
standard.
site measurements or from a nearby weather
9.2.1 Equipment required: Positive displace-
station. Barometric pressure readings ob- ment standard volume meter traceable to
tained from airports must be station pres- the National Bureau of Standards (such as a
sure, not corrected to sea level, and may Roots meter or equivalent), stop-watch, ma-
need to be corrected for differences in ele- nometer, thermometer, and barometer.
vation between the sampler site and the air- 9.2.2 Connect the flow rate transfer stand-
port. For samplers having flow recorders but ard to the inlet of the standard volume
not constant flow controllers, the average meter. Connect the manometer to measure
temperature and pressure at the site during the pressure at the inlet of the standard vol-
the sampling period should be estimated from ume meter. Connect the orifice manometer
weather bureau or other available data. to the pressure tap on the transfer standard.
8.13 Stop the sampler and carefully remove Connect a high-volume air pump (such as a
the filter, following the sampler manufactur- high-volume sampler blower) to the outlet
er’s instructions. Touch only the outer edges side of the standard volume meter. See Fig-
of the filter. See the precautions in step 8.6. ure 3a.
8.14 Fold the filter in half lengthwise so 9.2.3 Check for leaks by temporarily clamp-
that only surfaces with collected particulate ing both manometer lines (to avoid fluid
matter are in contact and place it in the fil- loss) and blocking the orifice with a large-di-
ter holder (glassine envelope or manila fold- ameter rubber stopper, wide cellophane tape,
er). or other suitable means. Start the high-vol-
8.15 Record the ending time or elapsed time ume air pump and note any change in the
on the filter information record, either from standard volume meter reading. The reading
the stop set-point time, from an elapsed time should remain constant. If the reading
indicator, or from a continuous flow record. changes, locate any leaks by listening for a
The sample period must be 1,440 ± 60 min. for whistling sound and/or retightening all con-
a valid sample. nections, making sure that all gaskets are
8.16 Record on the filter information record properly installed.
any other factors, such as meteorological 9.2.4 After satisfactorily completing the
conditions, construction activity, fires or leak check as described above, unclamp both
dust storms, etc., that might be pertinent to manometer lines and zero both manometers.
Fecha: 14/11/2013
the measurement. If the sample is known to 9.2.5 Achieve the appropriate flow rate
be defective, void it at this time. through the system, either by means of the
8.17 Equilibrate the exposed filter in the variable flow resistance in the transfer
conditioning environment for at least 24-hrs. standard or by varying the voltage to the air
8.18 Immediately after equilibration, re- pump. (Use of resistance plates as shown in
weigh the filter to the nearest milligram and Figure 1a is discouraged because the above
record the gross weight with the filter iden- leak check must be repeated each time a new
tification number. See Section 10 for TSP resistance plate is installed.) At least five
concentration calculations. different but constant flow rates, evenly dis-
9.0 Calibration. tributed, with at least three in the specified
29
flow rate interval (1.1 to 1.7 m3/min [39–60 ft 3/ Record Qstd to the nearest 0.01 std m3/min
min]), are required. in column 6 of Figure 4.
9.2.6 Measure and record the certification 9.2.15 Repeat steps 9.2.9 through 9.2.14 for
data on a form similar to the one illustrated at least four additional constant flow rates,
in Figure 4 according to the following steps. evenly spaced over the approximate range of
9.2.7 Observe the barometric pressure and 1.0 to 1.8 std m3/min (35–64 ft3/min).
record as P1 (item 8 in Figure 4). 9.2.16 For each flow, compute
9.2.8 Read the ambient temperature in the √DDH (P1/Pstd)(298/T1)
vicinity of the standard volume meter and (column 7a of Figure 4) and plot these value
record it as T1 (item 9 in Figure 4). against Qstd as shown in Figure 3a. Be sure to
9.2.9 Start the blower motor, adjust the use consistent units (mm Hg or kPa) for bar-
flow, and allow the system to run for at least ometric pressure. Draw the orifice transfer
1 min for a constant motor speed to be at- standard certification curve or calculate the
tained. linear least squares slope (m) and intercept
9.2.10 Observe the standard volume meter (b) of the certification curve:
reading and simultaneously start a stop- √DDH (P1/Pstd)(298/T1)
watch. Record the initial meter reading (Vi) =mQstd+b. See Figures 3 and 4. A certification
in column 1 of Figure 4. graph should be readable to 0.02 std m3/min.
9.2.11 Maintain this constant flow rate 9.2.17 Recalibrate the transfer standard an-
until at least 3 m3 of air have passed through nually or as required by applicable quality
the standard volume meter. Record the control procedures. (See Reference 2.)
standard volume meter inlet pressure ma- 9.3 Calibration of sampler flow indicator.
nometer reading as DP (column 5 in Figure
4), and the orifice manometer reading as DH NOTE: For samplers equipped with a flow
(column 7 in Figure 4). Be sure to indicate controlling device, the flow controller must
the correct units of measurement. be disabled to allow flow changes during
9.2.12 After at least 3 m3 of air have passed calibration of the sampler’s flow indicator,
through the system, observe the standard or the alternate calibration of the flow con-
troller given in 9.4 may be used. For sam-
volume meter reading while simultaneously
plers using an orifice-type flow indicator
stopping the stopwatch. Record the final
downstream of the motor, do not vary the
meter reading (Vf) in column 2 and the
flow rate by adjusting the voltage or power
elapsed time (t) in column 3 of Figure 4.
supplied to the sampler.
9.2.13 Calculate the volume measured by
the standard volume meter at meter condi- 9.3.1 A form similar to the one illustrated
tions of temperature and pressures as in Figure 5 should be used to record the cali-
Vm=Vf¥Vi. Record in column 4 of Figure 4. bration data.
9.2.14 Correct this volume to standard vol- 9.3.2 Connect the transfer standard to the
ume (std m3) as follows: inlet of the sampler. Connect the orifice ma-
nometer to the orifice pressure tap, as illus-
P1 − ∆P Tstd trated in Figure 3b. Make sure there are no
Vstd = Vm leaks between the orifice unit and the sam-
Pstd T1 pler.
9.3.3 Operate the sampler for at least 5
where: minutes to establish thermal equilibrium
Vstd = standard volume, std m3; prior to the calibration.
Vm = actual volume measured by the stand- 9.3.4 Measure and record the ambient tem-
ard volume meter; perature, T2, and the barometric pressure, P2,
P1 = barometric pressure during calibration, during calibration.
mm Hg or kPa; 9.3.5 Adjust the variable resistance or, if
DP = differential pressure at inlet to volume applicable, insert the appropriate resistance

meter, mm Hg or kPa; plate (or no plate) to achieve the desired
flow rate.
Pstd = 760 mm Hg or 101 kPa;
9.3.6 Let the sampler run for at least 2 min
Tstd = 298 K;
to re-establish the run-temperature condi-
T1 = ambient temperature during calibrations. Read and record the pressure drop
tion, K. across the orifice (DH) and the sampler flow
Fecha: 14/11/2013
Calculate the standard flow rate (std m3/min) rate indication (I) in the appropriate col-
as follows: umns of Figure 5.
9.3.7 Calculate √DDH(P2/Pstd)(298/T2) and de-
Vstd
Q std = termine the flow rate at standard conditions
t (Qstd) either graphically from the certifi-
cation curve or by calculating Qstd from the
where: least square slope and intercept of the trans-
EC08NO91.013</MATH>
Qstd = standard volumetric flow rate, std m3/ fer standard’s transposed certification curve:
min Qstd=1/m √DH(P2/Pstd)(298/T2)¥b. Record the
t = elapsed time, minutes. value of Qstd on Figure 5.
30
ER31AU93.024</MATH>
9.3.8 Repeat steps 9.3.5, 9.3.6, and 9.3.7 for 9.3.11 For a sampler equipped with a flow
several additional flow rates distributed over controller, the flow controlling mechanism
a range that includes 1.1 to 1.7 std m3/min. should be re-enabled and set to a flow near
9.3.9 Determine the calibration curve by the lower flow limit to allow maximum con-
plotting values of the appropriate expression trol range. The sample flow rate should be
involving I, selected from table 1, against verified at this time with a clean filter in-
Qstd. The choice of expression from table 1 de- stalled. Then add two or more filters to the
pends on the flow rate measurement device sampler to see if the flow controller main-
used (see Section 7.4.1) and also on whether tains a constant flow; this is particularly im-
the calibration curve is to incorporate geo- portant at high altitudes where the range of
graphic average barometric pressure (Pa) and the flow controller may be reduced.
seasonal average temperature (Ta) for the 9.4 Alternate calibration of flow-controlled
site to approximate actual pressure and tem- samplers. A flow-controlled sampler may be
perature. Where Pa and Ta can be determined calibrated solely at its controlled flow rate,
for a site for a seasonal period such that the
provided that previous operating history of
actual barometric pressure and temperature
the sampler demonstrates that the flow rate
at the site do not vary by more than ±60 mm
is stable and reliable. In this case, the flow
Hg (8 kPa) from Pa or ±15 °C from Ta, respec-
indicator may remain uncalibrated but
tively, then using Pa and Ta avoids the need
should be used to indicate any relative
for subsequent pressure and temperature cal-
culation when the sampler is used. The geo- change between initial and final flows, and
graphic average barometric pressure (Pa) the sampler should be recalibrated more
may be estimated from an altitude-pressure often to minimize potential loss of samples
table or by making an (approximate) ele- because of controller malfunction.
vation correction of ¥26 mm Hg (¥3.46 kPa) 9.4.1 Set the flow controller for a flow near
for each 305 m (1,000 ft) above sea level (760 the lower limit of the flow range to allow
mm Hg or 101 kPa). The seasonal average maximum control range.
temperature (Ta) may be estimated from 9.4.2 Install a clean filter in the sampler
weather station or other records. Be sure to and carry out steps 9.3.2, 9.3.3, 9.3.4, 9.3.6, and
use consistent units (mm Hg or kPa) for bar- 9.3.7.
ometric pressure. 9.4.3 Following calibration, add one or two
9.3.10 Draw the sampler calibration curve additional clean filters to the sampler, re-
or calculate the linear least squares slope connect the transfer standard, and operate
(m), intercept (b), and correlation coefficient the sampler to verify that the controller
of the calibration curve: [Expression from maintains the same calibrated flow rate; this
table 1]= mQstd+b. See Figures 3 and 5. Cali- is particularly important at high altitudes
bration curves should be readable to 0.02 std where the flow control range may be re-
m3/min. duced.

Fecha: 14/11/2013
31
50-723.eps</GPH>
10.0 Calculations of TSP Concentration. may be achieved by dividing the sampling

10.1 Determine the average sampler flow period into intervals and calculating an av-
rate during the sampling period according to erage reading before determining Qstd.
either 10.1.1 or 10.1.2 below. 10.2 Calculate the total air volume sampled
10.1.1 For a sampler without a continuous as:
flow recorder, determine the appropriate ex-
V¥Qstd× t
pression to be used from table 2 cor-
responding to the one from table 1 used in where:
step 9.3.9. Using this appropriate expression, V = total air volume sampled, in standard
determine Qstd for the initial flow rate from volume units, std m3/;
the sampler calibration curve, either graphi- Qstd = average standard flow rate, std m3/min;
cally or from the transposed regression
t = sampling time, min.
equation:
Qstd = 10.3 Calculate and report the particulate
1/m ([Appropriate expression from table matter concentration as:
2]¥b) 6
( Wf − Wi ) × 10
Similarly, determine Qstd from the final flow TSP =
reading, and calculate the average flow Qstd V
as one-half the sum of the initial and final
where:
flow rates.
TSP = mass concentration of total suspended
10.1.2 For a sampler with a continuous flow
particulate matter, µg/std m3;
recorder, determine the average flow rate de-
vice reading, I, for the period. Determine the Wi = initial weight of clean filter, g;
appropriate expression from table 2 cor- Wf = final weight of exposed filter, g;
responding to the one from table 1 used in V = air volume sampled, converted to stand-
ard conditions, std m3,
Fecha: 14/11/2013
step 9.3.9. Then using this expression and the

average flow rate reading, determine Qstd 106 = conversion of g to µg.
from the sampler calibration curve, either 10.4 If desired, the actual particulate mat-
graphically or from the transposed regres- ter concentration (see Section 2.2) can be
sion equation: calculated as follows:
Qstd =
(TSP)a=TSP (P3/Pstd)(298/T3)
1/m ([Appropriate expression from table
2]¥b) where:
50-724.eps</GPH> ER31AU93.025</MATH>
If the trace shows substantial flow change (TSP)a = actual concentration at field condi-
during the sampling period, greater accuracy tions, µg/m3;
32
TSP = concentration at standard conditions, 11. EPA Test Procedures for Determining
µg/std m3; pH and Integrity of High-Volume Air Filters.
P3 = average barometric pressure during QAD/M–80.01. Available from the Methods
sampling period, mm Hg; Standardization Branch, Quality Assurance
Pstd = 760 mn Hg (or 101 kPa); Division, Environmental Monitoring Sys-
T3 = average ambient temperature during tems Laboratory (MD–77), U.S. Environ-
sampling period, K. mental Protection Agency, Research Tri-
11.0 References. angle Park, NC 27711, 1980.
1. Quality Assurance Handbook for Air Pol- 12. Smith, F., P. S. Wohlschlegel, R. S. C.
lution Measurement Systems, Volume I, Rogers, and D. J. Mulligan. Investigation of
Principles. EPA–600/9–76–005, U.S. Environ- Flow Rate Calibration Procedures Associ-
mental Protection Agency, Research Tri- ated with the High-Volume Method for De-
angle Park, NC 27711, 1976. termination of Suspended Particulates.
EPA–600/4–78–047, U.S. Environmental Pro-
tection Agency, Research Triangle Park, NC,
027a, U.S. Environmental Protection Agency, June 1978.
Research Triangle Park, NC 27711, 1977.
3. Wedding, J. B., A. R. McFarland, and J.
E. Cernak. Large Particle Collection Charac-
teristics of Ambient Aerosol Samplers. Envi-
ron. Sci. Technol. 11:387–390, 1977.
4. McKee, H. C., et al. Collaborative Test-
ing of Methods to Measure Air Pollutants, I.
The High-Volume Method for Suspended Par-
ticulate Matter. J. Air Poll. Cont. Assoc., 22
(342), 1972.
5. Clement, R. E., and F. W. Karasek. Sam-
ple Composition Changes in Sampling and
Analysis of Organic Compounds in Aerosols.
The Intern. J. Environ. Anal. Chem., 7:109,
1979.
6. Lee, R. E., Jr., and J. Wagman. A Sam-
pling Anomaly in the Determination of At-
mospheric Sulfuric Concentration. Am. Ind.
Hygiene Assoc. J., 27:266, 1966.
7. Appel, B. R., et al. Interference Effects
in Sampling Particulate Nitrate in Ambient
Air. Atmospheric Environment, 13:319, 1979.
8. Tierney, G. P., and W. D. Conner. Hygro-
scopic Effects on Weight Determinations of
Particulates Collected on Glass-Fiber Fil-
ters. Am. Ind. Hygiene Assoc. J., 28:363, 1967.
9. Chahal, H. S., and D. J. Romano. High-
Volume Sampling Effect of Windborne Par-
ticulate Matter Deposited During Idle Peri-
ods. J. Air Poll. Cont. Assoc., Vol. 26 (885),
1976.
10. Patterson, R. K. Aerosol Contamination
from High-Volume Sampler Exhaust. J. Air
Poll. Cont. Assoc., Vol. 30 (169), 1980.

Fecha: 14/11/2013
33
50-725.eps</GPH>

Fecha: 14/11/2013
34
50-726.eps</GPH>

Fecha: 14/11/2013
35
50-727.eps</GPH>

Fecha: 14/11/2013
36
50-728.eps</GPH>
Environmental Protection Agency Pt. 50, App. C
[47 FR 54912, Dec. 6, 1982; 48 FR 17355, Apr. 22, 1983]
APPENDIX C TO PART 50—MEASUREMENT through a cell containing the gas sample to

PRINCIPLE AND CALIBRATION PROCE- be analyzed, and the quantitative absorption
DURE FOR THE MEASUREMENT OF of energy by CO in the sample cell is meas-
Fecha: 14/11/2013
CARBON MONOXIDE IN THE ATMOS- ured by a suitable detector. The photometer

is sensitized to CO by employing CO gas in
PHERE (NON-DISPERSIVE INFRARED
either the detector or in a filter cell in the
PHOTOMETRY) optical path, thereby limiting the measured
MEASUREMENT PRINCIPLE absorption to one or more of the char-
acteristic wavelengths at which CO strongly
1. Measurements are based on the absorp- absorbs. Optical filters or other means may
tion of infrared radiation by carbon mon-
oxide (CO) in a non-dispersive photometer.
Infrared energy from a source is passed
37
50-729.eps</GPH>
Pt. 50, App. C 40 CFR Ch. I (7–1–03 Edition)
also be used to limit sensitivity of the pho- trix if the zero air dilution ratio is not less
tometer to a narrow band of interest. Var- than 100:1. The assay of the cylinder(s) must
ious schemes may be used to provide a suit- be traceable either to a National Bureau of
able zero reference for the photometer. The Standards (NBS) CO in air Standard Ref-
measured absorption is converted to an elec- erence Material (SRM) or to an NBS/EPA-ap-
trical output signal, which is related to the proved commercially available Certified Ref-
concentration of CO in the measurement erence Material (CRM). CRM’s are described
cell. in Reference 2, and a list of CRM sources is
2. An analyzer based on this principle will available from the address shown for Ref-
be considered a reference method only if it erence 2. A recommended protocol for certi-
has been designated as a reference method in fying CO gas cylinders against either a CO
accordance with part 53 of this chapter. SRM or a CRM is given in Reference 1. CO
3. Sampling considerations. gas cylinders should be recertified on a reg-
The use of a particle filter on the sample ular basis as determined by the local quality
inlet line of an NDIR CO analyzer is optional control program.
and left to the discretion of the user or the 3.2 Dilution gas (zero air). Air, free of con-
manufacturer. Use of filter should depend on taminants which will cause a detectable re-
the analyzer’s susceptibility to interference, sponse on the CO analyzer. The zero air
malfunction, or damage due to particles. should contain <0.1 ppm CO. A procedure for
CALIBRATION PROCEDURE generating zero air is given in Reference 1.
4. Procedure Using Dynamic Dilution Method.
1. Principle. Either of two methods may be 4.1 Assemble a dynamic calibration system
used for dynamic multipoint calibration of such as the one shown in Figure 1. All cali-
CO analyzers: bration gases including zero air must be in-
(1) One method uses a single certified troduced into the sample inlet of the ana-
standard cylinder of CO, diluted as necessary lyzer system. For specific operating instruc-
with zero air, to obtain the various calibrations refer to the manufacturer’s manual.
tion concentrations needed. 4.2 Insure that all flowmeters are properly
(2) The other method uses individual cer- calibrated, under the conditions of use, if ap-
tified standard cylinders of CO for each con- propriate, against an authoritative standard
centration needed. Additional information such as a soap-bubble meter or wet-test
on calibration may be found in Section 2.0.9 meter. All volumetric flowrates should be
of Reference 1. corrected to 25 °C and 760 mm Hg (101 kPa).
2. Apparatus. The major components and
A discussion on calibration of flowmeters is
typical configurations of the calibration sys-
given in Reference 1.
tems for the two calibration methods are
4.3 Select the operating range of the CO an-
shown in Figures 1 and 2.
2.1 Flow controller(s). Device capable of alyzer to be calibrated.
adjusting and regulating flow rates. Flow 4.4 Connect the signal output of the CO an-
rates for the dilution method (Figure 1) must alyzer to the input of the strip chart re-
be regulated to ± 1%. corder or other data collection device. All
2.2 Flow meter(s). Calibrated flow meter adjustments to the analyzer should be based
capable of measuring and monitoring flow on the appropriate strip chart or data device
rates. Flow rates for the dilution method readings. References to analyzer responses in
(Figure 1) must be measured with an accu- the procedure given below refer to recorder
racy of ± 2% of the measured value. or data device responses.
2.3 Pressure regulator(s) for standard CO 4.5 Adjust the calibration system to deliver
cylinder(s). Regulator must have nonreac- zero air to the output manifold. The total air
tive diaphragm and internal parts and a suit- flow must exceed the total demand of the an-
able delivery pressure. alyzer(s) connected to the output manifold
2.4 Mixing chamber. A chamber designed to to insure that no ambient air is pulled into
provide thorough mixing of CO and diluent the manifold vent. Allow the analyzer to
air for the dilution method. sample zero air until a stable respose is ob-
2.5 Output manifold. The output manifold tained. After the response has stabilized, ad-
should be of sufficient diameter to insure an just the analyzer zero control. Offsetting the
insignificant pressure drop at the analyzer analyzer zero adjustments to +5 percent of
connection. The system must have a vent de- scale is recommended to facilitate observing
Fecha: 14/11/2013
signed to insure atmospheric pressure at the negative zero drift. Record the stable zero
manifold and to prevent ambient air from air response as ZCO.
entering the manifold. 4.6 Adjust the zero air flow and the CO flow
3. Reagents. from the standard CO cylinder to provide a
3.1 CO concentration standard(s). Cyl- diluted CO concentration of approximately
inder(s) of CO in air containing appropriate 80 percent of the upper range limit (URL) of
concentrations(s) of CO suitable for the se- the operating range of the analyzer. The
lected operating range of the analyzer under total air flow must exceed the total demand
calibration; CO standards for the dilution of the analyzer(s) connected to the output
method may be contained in a nitrogen ma- manifold to insure that no ambient air is
38
pulled into the manifold vent. The exact CO verify linearity) by decreasing FCO or in-
concentration is calculated from: creasing FD. Be sure the total flow exceeds
the analyzer’s total flow demand. For each
[ CO ]STD × FCO concentration generated, calculate the exact
[ CO ]OUT = (1) CO concentration using Equation (1). Record
FD + FCO the concentration and the analyzer’s re-
sponse for each concentration. Plot the ana-
where:
lyzer responses versus the corresponding CO
[CO]OUT = diluted CO concentration at the concentrations and draw or calculate the
output manifold, ppm; calibration curve.
[CO]STD = concentration of the undiluted CO 5. Procedure Using Multiple Cylinder Method.
standard, ppm; Use the procedure for the dynamic dilution
FCO = flow rate of the CO standard corrected method with the following changes:
to 25 °C and 760 mm Hg, (101 kPa), L/min; 5.1 Use a multi-cylinder system such as the
and typical one shown in Figure 2.
FD = flow rate of the dilution air corrected to
5.2 The flowmeter need not be accurately
25 °C and 760 mm Hg, (101 kPa), L/min.
calibrated, provided the flow in the output
Sample this CO concentration until a sta- manifold exceeds the analyzer’s flow de-
ble response is obtained. Adjust the analyzer mand.
span control to obtain a recorder response as 5.3 The various CO calibration concentra-
indicated below: tions required in Steps 4.6 and 4.7 are ob-
Recorder response (percent scale) = tained without dilution by selecting the ap-
propriate certified standard cylinder.
[ CO ]OUT REFERENCES
× 100 + Z CO (2)
URL 1. Quality Assurance Handbook for Air Pol-
where: lution Measurement Systems, Volume II—
Ambient Air Specific Methods, EPA–600/4–77–
URL = nominal upper range limit of the ana-
027a, U.S. Environmental Protection Agency,
lyzer’s operating range, and
Environmental Monitoring Systems Labora-
ZCO = analyzer response to zero air, % scale.
tory, Research Triangle Park, NC 27711, 1977.
If substantial adjustment of the analyzer 2. A procedure for Establishing
span control is required, it may be necessary Traceability of Gas Mixtures to Certain Na-
to recheck the zero and span adjustments by tional Bureau of Standards Standard Ref-
repeating Steps 4.5 and 4.6. Record the CO erence Materials. EPA–600/7–81–010, U.S. En-
concentration and the analyzer’s response. vironmental Protection Agency, Environ-
4.7 Generate several additional concentra- mental Monitoring Systems Laboratory
tions (at least three evenly spaced points (MD–77), Research Triangle Park, NC 27711,
across the remaining scale are suggested to January 1981.

Fecha: 14/11/2013
ER31AU93.027</MATH>
39
ER31AU93.026</MATH>
Pt. 50, App. C 40 CFR Ch. I (7–1–03 Edition)

Fecha: 14/11/2013
40
50-732.eps</GPH>

Fecha: 14/11/2013
[47 FR 54922, Dec. 6, 1982; 48 FR 17355, Apr. 22, 1983]
41
50-733.eps</GPH>
Pt. 50, App. D 40 CFR Ch. I (7–1–03 Edition)
APPENDIX D TO PART 50—MEASUREMENT

1
PRINCIPLE AND CALIBRATION PROCE- c(atm) = − (1n I / I O ) (2a )
DURE FOR THE MEASUREMENT OF αl
OZONE IN THE ATMOSPHERE or
106
c( ppm ) = − (1n I / I O ) (2 b)
MEASUREMENT PRINCIPLE
αl
1. Ambient air and ethylene are delivered The calculated O3 concentrations must be
simultaneously to a mixing zone where the corrected for O3 losses which may occur in
ozone in the air reacts with the ethylene to the photometer and for the temperature and
emit light, which is detected by a pressure of the sample.
photomultiplier tube. The resulting 2. Applicability. This procedure is applicable
photocurrent is amplified and is either read to the calibration of ambient air O3 ana-
directly or displayed on a recorder. lyzers, either directly or by means of a
transfer standard certified by this procedure.
2. An analyzer based on this principle will
Transfer standards must meet the require-
be considered a reference method only if it ments and specifications set forth in Ref-
has been designated as a reference method in erence 8.
accordance with part 53 of this chapter and 3. Apparatus. A complete UV calibration
calibrated as follows: system consists of an ozone generator, an
output port or manifold, a photometer, an
CALIBRATION PROCEDURE appropriate source of zero air, and other
1. Principle. The calibration procedure is components as necessary. The configuration
based on the photometric assay of ozone (O3) must provide a stable ozone concentration at
concentrations in a dynamic flow system. the system output and allow the photometer
to accurately assay the output concentra-
The concentration of O3 in an absorption cell
tion to the precision specified for the pho-
is determined from a measurement of the
tometer (3.1). Figure 1 shows a commonly
amount of 254 nm light absorbed by the sam- used configuration and serves to illustrate
ple. This determination requires knowledge the calibration procedure which follows.
of (1) the absorption coefficient (a) of O3 at Other configurations may require appro-
254 nm, (2) the optical path length (l) priate variations in the procedural steps. All
through the sample, (3) the transmittance of connections between components in the cali-
the sample at a wavelength of 254 nm, and (4) bration system downstream of the O3 gener-
the temperature (T) and pressure (P) of the ator should be of glass, Teflon, or other rel-
sample. The transmittance is defined as the atively inert materials. Additional informa-
ratio I/I0, where I is the intensity of light tion regarding the assembly of a UV photo-
which passes through the cell and is sensed metric calibration apparatus is given in Ref-
by the detector when the cell contains an O3 erence 9. For certification of transfer stand-
sample, and I0 is the intensity of light which ards which provide their own source of O3,
passes through the cell and is sensed by the the transfer standard may replace the O3
detector when the cell contains zero air. It is generator and possibly other components
shown in Figure 1; see Reference 8 for guid-
assumed that all conditions of the system,
ance.
except for the contents of the absorption
3.1 UV photometer. The photometer consists
cell, are identical during measurement of I of a low-pressure mercury discharge lamp,
and I0. The quantities defined above are re- (optional) collimation optics, an absorption
lated by the Beer-Lambert absorption law, cell, a detector, and signal-processing elec-
tronics, as illustrated in Figure 1. It must be
I − acl capable of measuring the transmittance, I/I0,
Transmittance = =e (1)
at a wavelength of 254 nm with sufficient

IO precision such that the standard deviation of
where: the concentration measurements does not
exceed the greater of 0.005 ppm or 3% of the
a = absorption coefficient of O3 at 254 nm=308 concentration. Because the low-pressure
±4 atm¥1 cm¥1 at 0 °C and 760 mercury lamp radiates at several wave-
torr. (1, 2, 3, 4, 5, 6, 7) lengths, the photometer must incorporate
Fecha: 14/11/2013
c = O3 concentration in atmospheres suitable means to assure that no O3 is gen-

l = optical path length in cm erated in the cell by the lamp, and that at
least 99.5% of the radiation sensed by the de-
In practice, a stable O3 generator is used to tector is 254 nm radiation. (This can be read-
produce O3 concentrations over the required ily achieved by prudent selection of optical
range. Each O3 concentration is determined filter and detector response characteristics.)
from the measurement of the transmittance The length of the light path through the ab-
ER31AU93.029</MATH>
(I/I0) of the sample at 254 nm with a photom- sorption cell must be known with an accu-
eter of path length l and calculated from the racy of at least 99.5%. In addition, the cell
equation, and associated plumbing must be designed to
42
ER31AU93.028</MATH>
Environmental Protection Agency Pt. 50, App. D
minimize loss of O3 from contact with cell tometer is established, the frequency of
walls and gas handling components. See Ref- these steps may be reduced consistent with
erence 9 for additional information. the documented stability of the photometer.
3.2 Air flow controllers. Devices capable of 5.2.1 Instruction manual: Carry out all set
regulating air flows as necessary to meet the up and adjustment procedures or checks as
output stability and photometer precision described in the operation or instruction
requirements. manual associated with the photometer.
3.3 Ozone generator. Device capable of gen- 5.2.2 System check: Check the photometer
erating stable levels of O3 over the required system for integrity, leaks, cleanliness,
concentration range. proper flowrates, etc. Service or replace fil-
3.4 Output manifold. The output manifold ters and zero air scrubbers or other
should be constructed of glass, Teflon, or consumable materials, as necessary.
other relatively inert material, and should 5.2.3 Linearity: Verify that the photometer
be of sufficient diameter to insure a neg- manufacturer has adequately established
ligible pressure drop at the photometer con- that the linearity error of the photometer is
nection and other output ports. The system less than 3%, or test the linearity by dilu-
must have a vent designed to insure atmos- tion as follows: Generate and assay an O3
pheric pressure in the manifold and to pre- concentration near the upper range limit of
vent ambient air from entering the manifold. the system (0.5 or 1.0 ppm), then accurately
3.5 Two-way valve. Manual or automatic dilute that concentration with zero air and
valve, or other means to switch the photom- reassay it. Repeat at several different dilu-
eter flow between zero air and the O3 con- tion ratios. Compare the assay of the origi-
centration. nal concentration with the assay of the di-
3.6 Temperature indicator. Accurate to ±1 °C. luted concentration divided by the dilution
3.7 Barometer or pressure indicator. Accurate ratio, as follows
to ±2 torr.
4. Reagents. A1 − A 2 / R
4.1 Zero air. The zero air must be free of E= × 100% ( 3)
contaminants which would cause a detect- A1
able response from the O3 analyzer, and it
should be free of NO, C2 H4, and other species where:
which react with O3. A procedure for gener- E = linearity error, percent
ating suitable zero air is given in Reference A1 = assay of the original concentration
9. As shown in Figure 1, the zero air supplied A2 = assay of the diluted concentration
to the photometer cell for the I0 reference R = dilution ratio = flow of original con-
measurement must be derived from the same centration divided by the total flow
source as the zero air used for generation of The linearity error must be less than 5%.
the ozone concentration to be assayed (I Since the accuracy of the measured flow-
measurement). When using the photometer rates will affect the linearity error as meas-
to certify a transfer standard having its own ured this way, the test is not necessarily
source of ozone, see Reference 8 for guidance conclusive. Additional information on
on meeting this requirement. verifying linearity is contained in Reference
5. Procedure. 9.
5.1 General operation. The calibration pho- 5.2.4 Intercomparison: When possible, the
tometer must be dedicated exclusively to use photometer should be occasionally intercom-
as a calibration standard. It should always pared, either directly or via transfer stand-
be used with clean, filtered calibration gases, ards, with calibration photometers used by
and never used for ambient air sampling. other agencies or laboratories.
Consideration should be given to locating 5.2.5 Ozone losses: Some portion of the O3
the calibration photometer in a clean labora- may be lost upon contact with the photom-
tory where it can be stationary, protected eter cell walls and gas handling components.
from physical shock, operated by a respon- The magnitude of this loss must be deter-
sible analyst, and used as a common stand- mined and used to correct the calculated O3
ard for all field calibrations via transfer concentration. This loss must not exceed 5%.
standards. Some guidelines for quantitatively deter-
5.2 Preparation. Proper operation of the mining this loss are discussed in Reference 9.
photometer is of critical importance to the 5.3 Assay of O3 concentrations.
Fecha: 14/11/2013
accuracy of this procedure. The following 5.3.1 Allow the photometer system to warm
steps will help to verify proper operation. up and stabilizer.
The steps are not necessarily required prior 5.3.2 Verify that the flowrate through the
to each use of the photometer. Upon initial photometer absorption cell, F allows the cell
operation of the photometer, these steps to be flushed in a reasonably short period of
should be carried out frequently, with all time (2 liter/min is a typical flow). The preci-
quantitative results or indications recorded sion of the measurements is inversely re-
in a chronological record either in tabular lated to the time required for flushing, since
form or plotted on a graphical chart. As the the photometer drift error increases with
performance and stability record of the pho- time.
43
ER31AU93.030</MATH>
5.3.3 Insure that the flowrate into the out- more ozone standards as determined accord-
put manifold is at least 1 liter/min greater ing to section 5.3. The exact procedure varies
than the total flowrate required by the pho- depending on the nature and design of the
tometer and any other flow demand con- transfer standard. Consult Reference 8 for
nected to the manifold. guidance.
5.3.4 Insure that the flowrate of zero air, Fz, 5.5 Calibration of ozone analyzers. Ozone
is at least 1 liter/min greater than the analyzers are calibrated as follows, using
flowrate required by the photometer. ozone standards obtained directly according
5.3.5 With zero air flowing in the output to section 5.3 or by means of a certified
manifold, actuate the two-way valve to allow transfer standard.
the photometer to sample first the manifold 5.5.1 Allow sufficient time for the O3 ana-
zero air, then Fz. The two photometer read- lyzer and the photometer or transfer stand-
ings must be equal (I=Io). ard to warmup and stabilize.
NOTE: In some commercially available 5.5.2 Allow the O3 analyzer to sample zero
photometers, the operation of the two-way air until a stable response is obtained and
valve and various other operations in section adjust the O3 analyzer’s zero control. Offset-
5.3 may be carried out automatically by the ting the analyzer’s zero adjustment to +5%
photometer. of scale is recommended to facilitate observ-
5.3.6 Adjust the O3 generator to produce an ing negative zero drift. Record the stable
O3 concentration as needed. zero air response as ‘‘Z’’.
5.3.7 Actuate the two-way valve to allow 5.5.3 Generate an O3 concentration stand-
the photometer to sample zero air until the ard of approximately 80% of the desired
absorption cell is thoroughly flushed and upper range limit (URL) of the O3 analyzer.
record the stable measured value of Io. Allow the O3 analyzer to sample this O3 con-
5.3.8 Actuate the two-way valve to allow centration standard until a stable response
the photometer to sample the ozone con- is obtained.
centration until the absorption cell is thor- 5.5.4 Adjust the O3 analyzer’s span control
oughly flushed and record the stable meas- to obtain a convenient recorder response as
ured value of I. indicated below:
5.3.9 Record the temperature and pressure recorder response (%scale) =
of the sample in the photometer absorption
cell. (See Reference 9 for guidance.)  [O 3 ]OUT 
5.3.10 Calculate the O3 concentration from  × 100 + Z ( 5)
 URL 
equation 4. An average of several determina-
tions will provide better precision. where:
URL = upper range limit of the O3 analyzer,
 −1 I   T   760)  10 6
ppm
[O3 ]OUT =  ln    × ( 4)
 al I O   273   P  L Z = recorder response with zero air, % scale
where: Record the O3 concentration and the cor-

responding analyzer response. If substantial
[O3]OUT = O3 concentration, ppm adjustment of the span control is necessary,
a = absorption coefficient of O3 at 254 nm=308 recheck the zero and span adjustments by re-
atm¥1 cm¥1 at 0 °C and 760 torr peating steps 5.5.2 to 5.5.4.
l = optical path length, cm 5.5.5 Generate several other O3 concentra-
T = sample temperature, K tion standards (at least 5 others are rec-
P = sample pressure, torr
ommended) over the scale range of the O3 an-
L = correction factor for O3 losses from
alyzer by adjusting the O3 source or by Op-
5.2.5=(1-fraction O3 lost).
tion 1. For each O3 concentration standard,
NOTE: Some commercial photometers may record the O3 and the corresponding analyzer
automatically evaluate all or part of equa- response.

tion 4. It is the operator’s responsibility to 5.5.6 Plot the O3 analyzer responses versus
verify that all of the information required the corresponding O3 concentrations and
for equation 4 is obtained, either automati- draw the O3 analyzer’s calibration curve or
cally by the photometer or manually. For calculate the appropriate response factor.
‘‘automatic’’ photometers which evaluate 5.5.7 Option 1: The various O3 concentra-
the first term of equation 4 based on a linear tions required in steps 5.3.11 and 5.5.5 may be
Fecha: 14/11/2013
approximation, a manual correction may be obtained by dilution of the O3 concentration

required, particularly at higher O3 levels. generated in steps 5.3.6 and 5.5.3. With this
See the photometer instruction manual and option, accurate flow measurements are re-
Reference 9 for guidance. quired. The dynamic calibration system may
5.3.11 Obtain additional O3 concentration be modified as shown in Figure 2 to allow for
standards as necessary by repeating steps dilution air to be metered in downstream of
5.3.6 to 5.3.10 or by Option 1. the O3 generator. A mixing chamber between
ER31AU93.033</MATH>
5.4 Certification of transfer standards. A the O3 generator and the output manifold is
transfer standard is certified by relating the also required. The flowrate through the O3
output of the transfer standard to one or generator (Fo) and the dilution air flowrate
44
ER31AU93.032</MATH>
Environmental Protection Agency Pt. 50, App. D
(FD) are measured with a reliable flow or vol- Monoxide, and Argon’’, J. Phys. Chem., 68, 412
ume standard traceable to NBS. Each O3 con- (1964).
centration generated by dilution is cal- 4. M. Griggs, ‘‘Absorption Coefficients of
culated from: Ozone in the Ultraviolet and Visible Re-
gions’’, J. Chem. Phys., 49, 857 (1968).
  5. K. H. Becker, U. Schurath, and H. Seitz,
FO
[ O 3 ]′OUT = [ O 3 ]OUT   (6) ‘‘Ozone Olefin Reactions in the Gas Phase. 1.
  Rate Constants and Activation Energies’’,
 FO + FD  Int’l Jour. of Chem. Kinetics, VI, 725 (1974).
6. M. A. A. Clyne and J. A. Coxom, ‘‘Ki-
where:
netic Studies of Oxy-halogen Radical Sys-
[O3]′OUT = diluted O3 concentration, ppm tems’’, Proc. Roy. Soc., A303, 207 (1968).
F0 = flowrate through the O3 generator, liter/ 7. J. W. Simons, R. J. Paur, H. A. Webster,
min and E. J. Bair, ‘‘Ozone Ultraviolet Pho-
FD = diluent air flowrate, liter/min tolysis. VI. The Ultraviolet Spectrum’’, J.
Chem. Phys., 59, 1203 (1973).
REFERENCES
8. Transfer Standards for Calibration of
1. E.C.Y. Inn and Y. Tanaka, ‘‘Absorption Ambient Air Monitoring Analyzers for
coefficient of Ozone in the Ultraviolet and Ozone, EPA publication number EPA–600/4–
Visible Regions’’, J. Opt. Soc. Am., 43, 870 79–056, EPA, National Exposure Research
(1953). Laboratory, Department E, (MD–77B), Re-
2. A. G. Hearn, ‘‘Absorption of Ozone in the search Triangle Park, NC 27711.
Ultraviolet and Visible Regions of the Spec- 9. Technical Assistance Document for the
trum’’, Proc. Phys. Soc. (London), 78, 932 Calibration of Ambient Ozone Monitors, EPA
(1961). publication number EPA–600/4–79–057, EPA,
3. W. B. DeMore and O. Raper, ‘‘Hartley National Exposure Research Laboratory, De-
Band Extinction Coefficients of Ozone in the partment E, (MD–77B), Research Triangle
Gas Phase and in Liquid Nitrogen, Carbon Park, NC 27711.

Fecha: 14/11/2013
45
ER31AU93.031</MATH>

Fecha: 14/11/2013
46
50-738.eps</GPH>
Environmental Protection Agency Pt. 50, App. F
[44 FR 8224, Feb. 8, 1979, as amended at 62 FR NO2 measurements and therefore filters
38895, July 18, 1997] should be changed frequently.
3. An analyzer based on this principle will
APPENDIX E TO PART 50 [RESERVED] be considered a reference method only if it
has been designated as a reference method in
APPENDIX F TO PART 50—MEASUREMENT accordance with part 53 of this chapter.
PRINCIPLE AND CALIBRATION PROCE-
DURE FOR THE MEASUREMENT OF NI- CALIBRATION
TROGEN DIOXIDE IN THE ATMOSPHERE 1. Alternative A—Gas phase titration (GPT)
(GAS PHASE CHEMILUMINESCENCE) of an NO standard with O3.
Major equipment required: Stable O3 gener-
PRINCIPLE AND APPLICABILITY ator. Chemiluminescence NO/NOX/NO2 ana-
1. Atmospheric concentrations of nitrogen lyzer with strip chart recorder(s). NO con-
dioxide (NO2) are measured indirectly by centration standard.
photometrically measuring the light inten- 1.1 Principle. This calibration technique is
sity, at wavelengths greater than 600 based upon the rapid gas phase reaction be-
nanometers, resulting from the chemilumi- tween NO and O3 to produce stoichiometric
nescent reaction of nitric oxide (NO) with quantities of NO2 in accordance with the fol-
ozone (O3). (1,2,3) NO2 is first quantitatively lowing equation: (8)
reduced to NO(4,5,6) by means of a converter.
NO, which commonly exists in ambient air NO + O 3 → NO 2 + O 2 (1)
together with NO2, passes through the con-
verter unchanged causing a resultant total The quantitative nature of this reaction is
NOX concentration equal to NO+NO2. A sam- such that when the NO concentration is
ple of the input air is also measured without known, the concentration of NO2 can be de-
having passed through the converted. This termined. Ozone is added to excess NO in a
latter NO measurement is subtracted from dynamic calibration system, and the NO
the former measurement (NO+NO2) to yield channel of the chemiluminescence NO/NOX/
the final NO2 measurement. The NO and NO2 analyzer is used as an indicator of
NO+NO2 measurements may be made concur- changes in NO concentration. Upon the addi-
rently with dual systems, or cyclically with tion of O3, the decrease in NO concentration
the same system provided the cycle time observed on the calibrated NO channel is
does not exceed 1 minute. equivalent to the concentration of NO2 pro-
2. Sampling considerations. duced. The amount of NO2 generated may be
2.1 Chemiluminescence NO/NOX/NO2 ana- varied by adding variable amounts of O3 from
lyzers will respond to other nitrogen con- a stable uncalibrated O3 generator. (9)
taining compounds, such as peroxyacetyl ni- 1.2 Apparatus. Figure 1, a schematic of a
trate (PAN), which might be reduced to NO typical GPT apparatus, shows the suggested
in the thermal converter. (7) Atmospheric configuration of the components listed
concentrations of these potential inter- below. All connections between components
ferences are generally low relative to NO2 in the calibration system downstream from
and valid NO2 measurements may be ob- the O3 generator should be of glass, Teflon,
tained. In certain geographical areas, where or other non-reactive material.
the concentration of these potential inter- 1.2.1 Air flow controllers. Devices capable of
ferences is known or suspected to be high maintaining constant air flows within ±2% of
relative to NO2, the use of an equivalent the required flowrate.
method for the measurement of NO2 is rec- 1.2.2 NO flow controller. A device capable of
ommended. maintaining constant NO flows within ±2%
2.2 The use of integrating flasks on the of the required flowrate. Component parts in
sample inlet line of chemiluminescence NO/ contact with the NO should be of a non-reac-
NOX/NO2 analyzers is optional and left to tive material.
couraged. The sample residence time be- 1.2.3 Air flowmeters. Calibrated flowmeters
tween the sampling point and the analyzer capable of measuring and monitoring air
should be kept to a minimum to avoid erro- flowrates with an accuracy of ±2% of the
neous NO2 measurements resulting from the measured flowrate.
reaction of ambient levels of NO and O3 in 1.2.4 NO flowmeter. A calibrated flowmeter
Fecha: 14/11/2013
the sampling system. capable of measuring and monitoring NO

2.3 The use of particulate filters on the flowrates with an accuracy of ±2% of the
sample inlet line of chemiluminescence NO/ measured flowrate. (Rotameters have been
NOX/NO2 analyzers is optional and left to the reported to operate unreliably when meas-
discretion of the user or the manufacturer. uring low NO flows and are not rec-
Use of the filter should depend on the ana- ommended.)
lyzer’s susceptibility to interference, mal- 1.2.5 Pressure regulator for standard NO cyl-
function, or damage due to particulates. inder. This regulator must have a nonreac-
Users are cautioned that particulate matter tive diaphragm and internal parts and a suit-
concentrated on a filter may cause erroneous able delivery pressure.
47
EC08NO91.075</MATH>
Pt. 50, App. F 40 CFR Ch. I (7–1–03 Edition)
1.2.6 Ozone generator. The generator must
be capable of generating sufficient and stable PR = [NO] RC × t R 2.75 ppm-minutes (2)
levels of O3 for reaction with NO to generate
NO2 concentrations in the range required.
 FNO 
Ozone generators of the electric discharge
[ NO] RC = [ NO]STD   (3)
type may produce NO and NO2 and are not
recommended.  FO + FNO 
1.2.7 Valve. A valve may be used as shown
in Figure 1 to divert the NO flow when zero VRC
air is required at the manifold. The valve tR = < 2 minutes (4)
should be constructed of glass, Teflon, or FO + FNO
other nonreactive material.
where:
1.2.8 Reaction chamber. A chamber, con-
structed of glass, Teflon, or other nonreac- PR = dynamic parameter specification, deter-
tive material, for the quantitative reaction mined empirically, to insure complete re-
of O3 with excess NO. The chamber should be action of the available O3, ppm-minute
of sufficient volume (VRC) such that the resi- [NO]RC = NO concentration in the reaction
dence time (tR) meets the requirements spec- chamber, ppm
ified in 1.4. For practical reasons, tR should R = residence time of the reactant gases in
be less than 2 minutes. the reaction chamber, minute
1.2.9 Mixing chamber. A chamber con- [NO]STD = concentration of the undiluted NO
structed of glass, Teflon, or other nonreac- standard, ppm
tive material and designed to provide thor- FNO = NO flowrate, scm 3/min
ough mixing of reaction products and diluent FO = O3 generator air flowrate, scm 3/min
air. The residence time is not critical when VRC = volume of the reaction chamber, scm 3
the dynamic parameter specification given
1.4.2 The flow conditions to be used in the
in 1.4 is met.
GPT system are determined by the following
1.2.10 Output manifold. The output manifold
procedure:
should be constructed of glass, Teflon, or
(a) Determine FT, the total flow required at
other non-reactive material and should be of
the output manifold (FT=analyzer demand
sufficient diameter to insure an insignificant
plus 10 to 50% excess).
pressure drop at the analyzer connection.
The system must have a vent designed to in- (b) Establish [NO]OUT as the highest NO
sure atmospheric pressure at the manifold concentration (ppm) which will be required
EC08NO91.081</MATH>
and to prevent ambient air from entering the at the output manifold. [NO]OUT should be
manifold. approximately equivalent to 90% of the
1.3 Reagents. upper range limit (URL) of the NO2 con-
1.3.1 NO concentration standard. Gas cyl- centration range to be covered.
inder standard containing 50 to 100 ppm NO (c) Determine FNO as
in N2 with less than 1 ppm NO2. This standard
[ NO]OUT × FT
EC08NO91.080</MATH>
must be traceable to a National Bureau of
Standards (NBS) NO in N2 Standard Ref-
FNO = (5)
erence Material (SRM 1683 or SRM 1684), an
[ NO]STD
NBS NO2 Standard Reference Material (SRM (d) Select a convenient or available reac-
1629), or an NBS/EPA-approved commercially tion chamber volume. Initially, a trial VRC
available Certified Reference Material may be selected to be in the range of ap-
50-754.eps</MATH>
(CRM). CRM’s are described in Reference 14, proximately 200 to 500 scm3.
and a list of CRM sources is available from (e) Compute FO as
the address shown for Reference 14. A rec-
ommended protocol for certifying NO gas
cylinders against either an NO SRM or CRM
EC08NO91.079</MATH>
is given in section 2.0.7 of Reference 15. Ref-
erence 13 gives procedures for certifying an
NO gas cylinder against an NBS NO2 SRM
and for determining the amount of NO2 im- (f) Compute tR as
purity in an NO cylinder.
1.3.2 Zero air. Air, free of contaminants VRC
tR =
Fecha: 14/11/2013
which will cause a detectable response on the (7)

EC08NO91.078</MATH>
NO/NOX/NO2 analyzer or which might react FO + FNO

with either NO, O3, or NO2 in the gas phase Verify that tR < 2 minutes. If not, select a re-
titration. A procedure for generating zero air action chamber with a smaller VRC.
is given in reference 13. (g) Compute the diluent air flowrate as
1.4 Dynamic parameter specification.
1.4.1 The O3 generator air flowrate (F0) and FD = FT ' FO ' FNO (8)
EC08NO91.077</MATH>
NO flowrate (FNO) (see Figure 1) must be ad-

justed such that the following relationship where:
holds: FD = diluent air flowrate, scm 3/min
48
EC08NO91.076</MATH>
(h) If FO turns out to be impractical for the no ambient air is pulled into the manifold
desired system, select a reaction chamber vent. Allow the analyzer to sample zero air
having a different VRC and recompute FO and until stable NO, NOX, and NO2 responses are
FD. obtained. After the responses have stabilized,
NOTE: A dynamic parameter lower than adjust the analyzer zero control(s).
2.75 ppm-minutes may be used if it can be de- NOTE: Some analyzers may have separate
termined empirically that quantitative reac- zero controls for NO, NOX, and NO2. Other
tion of O3 with NO occurs. A procedure for analyzers may have separate zero controls
making this determination as well as a more only for NO and NOX, while still others may
detailed discussion of the above require- have only one zero control common to all
ments and other related considerations is three channels.
given in reference 13. Offsetting the analyzer zero adjustments
1.5 Procedure. to +5 percent of scale is recommended to fa-
1.5.1 Assemble a dynamic calibration sys- cilitate observing negative zero drift. Record
tem such as the one shown in Figure 1. the stable zero air responses as ZNO, ZNOX,
1.5.2 Insure that all flowmeters are cali- and ZNO2.
brated under the conditions of use against a
1.5.8 Preparation of NO and NOX calibration
reliable standard such as a soap-bubble
curves.
meter or wet-test meter. All volumetric
flowrates should be corrected to 25 °C and 760 1.5.8.1 Adjustment of NO span control. Adjust
mm Hg. A discussion on the calibration of the NO flow from the standard NO cylinder
flowmeters is given in reference 13. to generate an NO concentration of approxi-
1.5.3 Precautions must be taken to remove mately 80 percent of the upper range limit
O2 and other contaminants from the NO pres- (URL) of the NO range. This exact NO con-
sure regulator and delivery system prior to centration is calculated from:
the start of calibration to avoid any conver-
FNO × [ NO ]STD
sion of the standard NO to NO2. Failure to do [ NO ]OUT = (9)
so can cause significant errors in calibration.
This problem may be minimized by (1) care- FNO + FO + FD
fully evacuating the regulator, when pos- where:
sible, after the regulator has been connected
to the cylinder and before opening the cyl- [NO]OUT = diluted NO concentration at the
inder valve; (2) thoroughly flushing the regu- output manifold, ppm
lator and delivery system with NO after Sample this NO concentration until the NO
opening the cylinder valve; (3) not removing and NOX responses have stabilized. Adjust
the regulator from the cylinder between cali- the NO span control to obtain a recorder re-
brations unless absolutely necessary. Fur- sponse as indicated below:
ther discussion of these procedures is given
recorder response (percent scale) =
in reference 13.
1.5.4 Select the operating range of the NO/
NOX/NO2 analyzer to be calibrated. In order  [ NO ] 

OUT
× 100  + Z NO (10 )
to obtain maximum precision and accuracy 
for NO2 calibration, all three channels of the  URL 
analyzer should be set to the same range. If where:
operation of the NO and NOX channels on
higher ranges is desired, subsequent re- URL = nominal upper range limit of the NO
calibration of the NO and NOX channels on channel, ppm
the higher ranges is recommended. NOTE: Some analyzers may have separate
NOTE: Some analyzer designs may require span controls for NO, NOX, and NO2. Other
identical ranges for NO, NOX, and NO2 during analyzers may have separate span controls
operation of the analyzer. only for NO and NOX, while still others may
1.5.5 Connect the recorder output cable(s) have only one span control common to all
of the NO/NOX/NO2 analyzer to the input ter- three channels. When only one span control
minals of the strip chart recorder(s). All ad- is available, the span adjustment is made on
justments to the analyzer should be per- the NO channel of the analyzer.
formed based on the appropriate strip chart If substantial adjustment of the NO span
readings. References to analyzer responses in control is necessary, it may be necessary to
Fecha: 14/11/2013
the procedures given below refer to recorder recheck the zero and span adjustments by re-
responses. peating steps 1.5.7 and 1.5.8.1. Record the NO
1.5.6 Determine the GPT flow conditions concentration and the analyzer’s NO re-
required to meet the dynamic parameter sponse.
specification as indicated in 1.4. 1.5.8.2 Adjustment of NOX span control. When
1.5.7 Adjust the diluent air and O3 gener- adjusting the analyzer’s NOX span control,
ator air flows to obtain the flows determined the presence of any NO2 impurity in the
ER31AU93.045</MATH>
in section 1.4.2. The total air flow must ex- standard NO cylinder must be taken into ac-
ceed the total demand of the analyzer(s) con- count. Procedures for determining the
nected to the output manifold to insure that amount of NO2 impurity in the standard NO
49
ER31AU93.044</MATH>
cylinder are given in reference 13. The exact concentrations using equations (9) and (11)
NOX concentration is calculated from: respectively. Record the analyzer’s NO and
NOX responses for each concentration. Plot
[ NO X ]OUT =
(
FNO × [ NO ]STD + [ NO 2 ]IMP ) (11) the analyzer responses versus the respective
calculated NO and NOX concentrations and
FNO + FO + FD draw or calculate the NO and NOX calibra-
tion curves. For subsequent calibrations
where: where linearity can be assumed, these curves
[NOX]OUT = diluted NOX concentration at the may be checked with a two-point calibration
output manifold, ppm consisting of a zero air point and NO and
[NO2]IMP = concentration of NO2 impurity in NOX concentrations of approximately 80% of
the standard NO cylinder, ppm the URL.
1.5.9 Preparation of NO2 calibration curve.
Adjust the NOX span control to obtain a re- 1.5.9.1 Assuming the NO2 zero has been
corder response as indicated below: properly adjusted while sampling zero air in
recorder response (% scale) = step 1.5.7, adjust FO and FD as determined in
section 1.4.2. Adjust FNO to generate an NO
 [ NO ]  concentration near 90% of the URL of the NO

X OUT
× 100  + Z NO (12 ) range. Sample this NO concentration until
 the NO and NOX responses have stabilized.
 URL  X
Using the NO calibration curve obtained in
NOTE: If the analyzer has only one span section 1.5.8, measure and record the NO con-
control, the span adjustment is made on the centration as [NO]orig. Using the NOX calibra-
NO channel and no further adjustment is tion curve obtained in section 1.5.8, measure
made here for NOx. and record the NOX concentration as
If substantial adjustment of the NOX span [NOX]orig.
control is necessary, it may be necessary to 1.5.9.2 Adjust the O3 generator to generate
recheck the zero and span adjustments by re- sufficient O3 to produce a decrease in the NO
peating steps 1.5.7 and 1.5.8.2. Record the NOX concentration equivalent to approximately
concentration and the analyzer’s NOX re- 80% of the URL of the NO2 range. The de-
sponse. crease must not exceed 90% of the NO con-
1.5.8.3 Generate several additional concentration determined in step 1.5.9.1. After
centrations (at least five evenly spaced the analyzer responses have stabilized,
points across the remaining scale are sug- record the resultant NO and NOX concentra-
gested to verify linearity) by decreasing FNO tions as [NO]rem and [NOX]rem.
or increasing FD. For each concentration 1.5.9.3 Calculate the resulting NO2 con-
generated, calculate the exact NO and NOX centration from:
FNO × [NO 2 ]IMP

[ NO 2 ]OUT = [ NO]orig − [ NO]rem + (13)
FNO + FO + FD
where: If substantial adjustment of the NO2 span

[NO2]OUT = diluted NO2 concentration at the control is necessary, it may be necessary to
output manifold, ppm recheck the zero and span adjustments by re-
[NO]orig = original NO concentration, prior to peating steps 1.5.7 and 1.5.9.3. Record the NO2
addition of O3, ppm concentration and the corresponding ana-
[NO]rem = NO concentration remaining after lyzer NO2 and NOX responses. ER31AU93.048</MATH>
addition of O3, ppm 1.5.9.4 Maintaining the same FNO, FO, and
FD as in section 1.5.9.1, adjust the ozone gen-
Adjust the NO2 span control to obtain a re- erator to obtain several other concentrations
corder response as indicated below: of NO2 over the NO2 range (at least five even-
Fecha: 14/11/2013
recorder response (% scale) = ly spaced points across the remaining scale

EC08NO91.082</MATH>
are suggested). Calculate each NO2 con-

 [ NO ]  centration using equation (13) and record the
2 OUT
× 100  + Z NO

(14 )
 corresponding analyzer NO2 and NOX re-
 URL  2
sponses. Plot the analyzer’s NO2 responses

NOTE: If the analyzer has only one or two versus the corresponding calculated NO2 con-
span controls, the span adjustments are centrations and draw or calculate the NO2
ER31AU93.047</MATH>
made on the NO channel or NO and NOX calibration curve.

channels and no further adjustment is made
1.5.10 Determination of converter efficiency.
here for NO2.
50
ER31AU93.046</MATH>
1.5.10.1 For each NO2 concentration gen- calibration curve (see section 1.5.9) calculate
erated during the preparation of the NO2 the concentration of NO2 converted from:
[ NO 2 ]CONV = [NO 2 ]OUT ' ([NO X ]orig ' [NO X ]rem ) (15)
where: 2.2.4 NO flowmeter. A calibrated flowmeter

[NO2]CONV = concentration of NO2 converted, capable of measuring and monitoring NO
ppm flowrates with an accuracy of ±2% of the
[NOX]orig = original NOX concentration prior measured flowrate. (Rotameters have been
to addition of O3, ppm reported to operate unreliably when meas-
[NOX]rem = NOX concentration remaining uring low NO flows and are not rec-
after addition of O3, ppm ommended.)
2.2.5 Pressure regulator for standard NO cyl-
NOTE: Supplemental information on cali- inder. This regulator must have a non-reac-
bration and other procedures in this method tive diaphragm and internal parts and a suit-
are given in reference 13. able delivery pressure.
Plot [NO2]CONV (y-axis) versus [NO2]OUT (x- 2.2.6 Drier. Scrubber to remove moisture
axis) and draw or calculate the converter ef- from the permeation device air system. The
ficiency curve. The slope of the curve times use of the drier is optional with NO2 perme-
100 is the average converter efficiency, EC. ation devices not sensitive to moisture.
The average converter efficiency must be (Refer to the supplier’s instructions for use
greater than 96%; if it is less than 96%, re- of the permeation device.)
place or service the converter. 2.2.7 Constant temperature chamber. Cham-
2. Alternative B—NO2 permeation device. ber capable of housing the NO2 permeation
Major equipment required: device and maintaining its temperature to
Stable O3 generator. within ±0.1 °C.
Chemiluminescence NO/NOX/NO2 analyzer 2.2.8 Temperature measuring device. Device
with strip chart recorder(s). capable of measuring and monitoring the
NO concentration standard. temperature of the NO2 permeation device
NO2 concentration standard. with an accuracy of ±0.05 °C.
2.1 Principle. Atmospheres containing accu- 2.2.9 Valves. A valve may be used as shown
rately known concentrations of nitrogen di- in Figure 2 to divert the NO2 from the perme-
oxide are generated by means of a permeation device when zero air or NO is required
ation device. (10) The permeation device at the manifold. A second valve may be used
emits NO2 at a known constant rate provided to divert the NO flow when zero air or NO2 is
the temperature of the device is held con- required at the manifold.
stant (±0.1 °C) and the device has been accu- The valves should be constructed of glass,
rately calibrated at the temperature of use. Teflon, or other nonreactive material.
The NO2 emitted from the device is diluted 2.2.10 Mixing chamber. A chamber con-
with zero air to produce NO2 concentrations structed of glass, Teflon, or other nonreac-
suitable for calibration of the NO2 channel of tive material and designed to provide thor-
the NO/NOX/NO2 analyzer. An NO concentra- ough mixing of pollutant gas streams and
tion standard is used for calibration of the diluent air.
NO and NOX channels of the analyzer. 2.2.11 Output manifold. The output manifold
2.2 Apparatus. A typical system suitable for should be constructed of glass, Teflon, or
generating the required NO and NO2 con- other non-reactive material and should be of
centrations is shown in Figure 2. All connec- sufficient diameter to insure an insignificant
tions between components downstream from pressure drop at the analyzer connection.
the permeation device should be of glass, The system must have a vent designed to in-
Teflon, or other non-reactive material. sure atmospheric pressure at the manifold
2.2.1 Air flow controllers. Devices capable of and to prevent ambient air from entering the
maintaining constant air flows within ±2% of manifold.
Fecha: 14/11/2013
the required flowrate. 2.3 Reagents.

2.2.2 NO flow controller. A device capable of 2.3.1 Calibration standards. Calibration
maintaining constant NO flows within ±2% standards are required for both NO and NO2.
of the required flowrate. Component parts in The reference standard for the calibration
contact with the NO must be of a non-reac- may be either an NO or NO2 standard, and
tive material. must be traceable to a National Bureau of
2.2.3 Air flowmeters. Calibrated flowmeters Standards (NBS) NO in N2 Standard Ref-
capable of measuring and monitoring air erence Material (SRM 1683 or SRM 1684), and
flowrates with an accuracy of ±2% of the NBS NO2 Standard Reference Material (SRM
measured flowrate. 1629), or an NBS/EPA-approved commercially
51
EC08NO91.083</MATH>
available Certified Reference Material 2.4.3 Install the permeation device in the
(CRM). CRM’s are described in Reference 14, constant temperature chamber. Provide a
and a list of CRM sources is available from small fixed air flow (200–400 scm 3/min) across
the address shown for Reference 14. Ref- the device. The permeation device should al-
erence 15 gives recommended procedures for ways have a continuous air flow across it to
certifying an NO gas cylinder against an NO prevent large buildup of NO2 in the system
SRM or CRM and for certifying an NO2 per- and a consequent restabilization period.
meation device against an NO2 SRM. Ref- Record the flowrate as FP. Allow the device
erence 13 contains procedures for certifying to stabilize at the calibration temperature
an NO gas cylinder against an NO2 SRM and for at least 24 hours. The temperature must
for certifying an NO2 permeation device be adjusted and controlled to within ±0.1 °C
against an NO SRM or CRM. A procedure for or less of the calibration temperature as
determining the amount of NO2 impurity in monitored with the temperature measuring
an NO cylinder is also contained in Ref- device.
erence 13. The NO or NO2 standard selected 2.4.4 Precautions must be taken to remove
as the reference standard must be used to O2 and other contaminants from the NO pres-
certify the other standard to ensure consist- sure regulator and delivery system prior to
ency between the two standards. the start of calibration to avoid any conver-
2.3.1.1 NO2 Concentration standard. A perme- sion of the standard NO to NO2. Failure to do
ation device suitable for generating NO2 con- so can cause significant errors in calibration.
centrations at the required flow-rates over This problem may be minimized by
the required concentration range. If the per- (1) Carefully evacuating the regulator,
meation device is used as the reference when possible, after the regulator has been
standard, it must be traceable to an SRM or connected to the cylinder and before opening
CRM as specified in 2.3.1. If an NO cylinder is the cylinder valve;
used as the reference standard, the NO2 per- (2) Thoroughly flushing the regulator and
meation device must be certified against the delivery system with NO after opening the
NO standard according to the procedure cylinder valve;
(3) Not removing the regulator from the
given in Reference 13. The use of the perme-
cylinder between calibrations unless abso-
ation device should be in strict accordance
lutely necessary. Further discussion of these
with the instructions supplied with the de-
procedures is given in reference 13.
vice. Additional information regarding the
2.4.5 Select the operating range of the NO/
use of permeation devices is given by
NOX NO2 analyzer to be calibrated. In order
Scaringelli et al. (11) and Rook et al. (12).
to obtain maximum precision and accuracy
2.3.1.2 NO Concentration standard. Gas cyl- for NO2 calibration, all three channels of the
inder containing 50 to 100 ppm NO in N2 with analyzer should be set to the same range. If
less than 1 ppm NO2. If this cylinder is used operation of the NO and NOX channels on
as the reference standard, the cylinder must higher ranges is desired, subsequent re-
be traceable to an SRM or CRM as specified calibration of the NO and NOX channels on
in 2.3.1. If an NO2 permeation device is used the higher ranges is recommended.
as the reference standard, the NO cylinder NOTE: Some analyzer designs may require
must be certified against the NO2 standard identical ranges for NO, NOX, and NO2 during
according to the procedure given in Ref- operation of the analyzer.
erence 13. The cylinder should be recertified 2.4.6 Connect the recorder output cable(s)
on a regular basis as determined by the local of the NO/NOX/NO2 analyzer to the input ter-
quality control program. minals of the strip chart recorder(s). All ad-
2.3.3 Zero air. Air, free of contaminants justments to the analyzer should be per-
which might react with NO or NO2 or cause formed based on the appropriate strip chart
a detectable response on the NO/NOX/NO2 an- readings. References to analyzer responses in
alyzer. When using permeation devices that the procedures given below refer to recorder
are sensitive to moisture, the zero air pass- responses.
ing across the permeation device must be dry 2.4.7 Switch the valve to vent the flow from
to avoid surface reactions on the device. the permeation device and adjust the diluent
(Refer to the supplier’s instructions for use air flowrate, FD, to provide zero air at the
of the permeation device.) A procedure for output manifold. The total air flow must ex-
generating zero air is given in reference 13. ceed the total demand of the analyzer(s) con-
Fecha: 14/11/2013
2.4 Procedure. nected to the output manifold to insure that

2.4.1 Assemble the calibration apparatus no ambient air is pulled into the manifold
such as the typical one shown in Figure 2. vent. Allow the analyzer to sample zero air
2.4.2 Insure that all flowmeters are cali- until stable NO, NOX, and NO2 responses are
brated under the conditions of use against a obtained. After the responses have stabilized,
reliable standard such as a soap bubble adjust the analyzer zero control(s).
meter or wet-test meter. All volumetric NOTE: Some analyzers may have separate
flowrates should be corrected to 25 °C and 760 zero controls for NO, NOX, and NO2. Other
mm Hg. A discussion on the calibration of analyzers may have separate zero controls
flowmeters is given in reference 13. only for NO and NOX, while still others may
52
have only one zero common control to all
three channels.
[ NO X ]OUT =
(
FNO × [ NO ]STD + [ NO 2 ]IMP ) (18)
Offsetting the analyzer zero adjustments to
+5% of scale is recommended to facilitate ob- FNO + FD
serving negative zero drift. Record the stable
zero air responses as ZNO, ZNOX, and ZNO2. where:
2.4.8 Preparation of NO and NOX calibration [NOX]OUT = diluted NOX cencentration at the
curves. output manifold, ppm
2.4.8.1 Adjustment of NO span control. Adjust [NO2]IMP = concentration of NO2 impurity in
the NO flow from the standard NO cylinder the standard NO cylinder, ppm
to generate an NO concentration of approxi- Adjust the NOX span control to obtain a con-
mately 80% of the upper range limit (URL) of venient recorder response as indicated below:
the NO range. The exact NO concentration is
calculated from: recorder response (% scale)
FNO × [ NO ]STD  [ NO ] 
[ NO ]OUT = (16 ) = X OUT
× 100  + Z NO (19 )
FNO + FD  
 URL  X
where: NOTE: If the analyzer has only one span

[NO]OUT = diluted NO concentration at the control, the span adjustment is made on the
output manifold, ppm NO channel and no further adjustment is
FNO = NO flowrate, scm3/min made here for NOX.
[NO]STD=concentration of the undiluted NO If substantial adjustment of the NOX span
standard, ppm control is necessary, it may be necessary to
FD = diluent air flowrate, scm 3/min recheck the zero and span adjustments by re-
peating steps 2.4.7 and 2.4.8.2. Record the NOX
Sample this NO concentration until the NO concentration and the analyzer’s NOX re-
and NOX responses have stabilized. Adjust sponse.
the NO span control to obtain a recorder re- 2.4.8.3 Generate several additional con-
sponse as indicated below: centrations (at least five evenly spaced
recorder response (% scale) = points across the remaining scale are sug-
gested to verify linearity) by decreasing FNO
 [ NO ]  or increasing FD. For each concentration
= OUT
× 100  + Z NO (17 ) generated, calculate the exact NO and NOX
 
 URL  concentrations using equations (16) and (18)
respectively. Record the analyzer’s NO and
NOX responses for each concentration. Plot
 [ NO X ]  the analyzer responses versus the respective
= OUT
× 100 + Z NO X
ER31AU93.054</MATH>
(19) calculated NO and NOX concentrations and
 URL  draw or calculate the NO and NOX calibra-
tion curves. For subsequent calibrations
where:
where linearity can be assumed, these curves
URL = nominal upper range limit of the NO may be checked with a two-point calibration
channel, ppm consisting of a zero point and NO and NOX
NOTE: Some analyzers may have separate
ER31AU93.053</MATH>
concentrations of approximately 80 percent
span controls for NO, NOX, and NO2. Other of the URL.
analyzers may have separate span controls 2.4.9 Preparation of NO2 calibration curve.
only for NO and NOX, while still others may 2.4.9.1 Remove the NO flow. Assuming the
have only one span control common to all NO2 zero has been properly adjusted while
three channels. When only one span control sampling zero air in step 2.4.7, switch the
is available, the span adjustment is made on
ER31AU93.052</MATH>
valve to provide NO2 at the output manifold.
the NO channel of the analyzer. 2.4.9.2 Adjust FD to generate an NO2 con-
If substantial adjustment of the NO span centration of approximately 80 percent of the
control is necessary, it may be necessary to URL of the NO2 range. The total air flow
recheck the zero and span adjustments by re- must exceed the demand of the analyzer(s)
peating steps 2.4.7 and 2.4.8.1. Record the NO under calibration. The actual concentration
Fecha: 14/11/2013
concentration and the analyzer’s NO re- of NO2 is calculated from:

ER31AU93.051</MATH>
sponse.
2.4.8.2 Adjustment of NOX span control. When R×K
adjusting the analyzer’s NOX span control, [ NO 2 ]OUT = ( 20 )
the presence of any NO2 impurity in the FP + FD
standard NO cylinder must be taken into ac-
count. Procedures for determining the where:
ER31AU93.050</MATH>
amount of NO2 impurity in the standard NO [NO2]OUT = diluted NO2 concentration at the
cylinder are given in reference 13. The exact output manifold, ppm
NOX concentration is calculated from: R = permeation rate, µg/min
53
ER31AU93.049</MATH>
K = 0.532µl NO2/µg NO2 (at 25 °C and 760 mm REFERENCES
Hg)
1. A. Fontijn, A. J. Sabadell, and R. J.
Fp = air flowrate across permeation device, Ronco, ‘‘Homogeneous Chemiluminescent
scm3/min Measurement of Nitric Oxide with Ozone,’’
FD = diluent air flowrate, scm3/min Anal. Chem., 42, 575 (1970).
Sample this NO2 concentration until the NOX 2. D. H. Stedman, E. E. Daby, F. Stuhl, and
and NO2 responses have stabilized. Adjust H. Niki, ‘‘Analysis of Ozone and Nitric Oxide
the NO2 span control to obtain a recorder re- by a Chemiluminiscent Method in Labora-
sponse as indicated below: tory and Atmospheric Studies of Photo-
chemical Smog,’’ J. Air Poll. Control Assoc.,
recorder response (% scale) 22, 260 (1972).
3. B. E. Martin, J. A. Hodgeson, and R. K.
 [ NO ]  Stevens, ‘‘Detection of Nitric Oxide
= 2 OUT
× 100  + Z NO 2 ( 21)
  Chemiluminescence at Atmospheric Pres-
 URL  sure,’’ Presented at 164th National ACS
Meeting, New York City, August 1972.
NOTE: If the analyzer has only one or two
span controls, the span adjustments are 4. J. A. Hodgeson, K. A. Rehme, B. E. Mar-
tin, and R. K. Stevens, ‘‘Measurements for
made on the NO channel or NO and NOX
Atmospheric Oxides of Nitrogen and Ammo-
channels and no further adjustment is made
nia by Chemiluminescence,’’ Presented at
here for NO2.
1972 APCA Meeting, Miami, FL, June 1972.
If substantial adjustment of the NO2 span 5. R. K. Stevens and J. A. Hodgeson, ‘‘Ap-
control is necessary it may be necessary to plications of Chemiluminescence Reactions
recheck the zero and span adjustments by re- to the Measurement of Air Pollutants,’’
peating steps 2.4.7 and 2.4.9.2. Record the NO2 Anal. Chem., 45, 443A (1973).
concentration and the analyzer’s NO2 re- 6. L. P. Breitenbach and M. Shelef, ‘‘De-
sponse. Using the NOX calibration curve ob- velopment of a Method for the Analysis of
tained in step 2.4.8, measure and record the NO2 and NH3 by NO-Measuring Instruments,’’
NOX concentration as [NOX]M. J. Air Poll. Control Assoc., 23, 128 (1973).
2.4.9.3 Adjust FD to obtain several other 7. A. M. Winer, J. W. Peters, J. P. Smith,
concentrations of NO2 over the NO2 range (at and J. N. Pitts, Jr., ‘‘Response of Commer-
least five evenly spaced points across the re- cial Chemiluminescent NO-NO2 Analyzers to
maining scale are suggested). Calculate each Other Nitrogen-Containing Compounds,’’ En-
NO2 concentration using equation (20) and viron. Sci. Technol., 8, 1118 (1974).
record the corresponding analyzer NO2 and 8. K. A. Rehme, B. E. Martin, and J. A.
NOX responses. Plot the analyzer’s NO2 re- Hodgeson, Tentative Method for the Calibra-
sponses versus the corresponding calculated tion of Nitric Oxide, Nitrogen Dioxide, and
NO2 concentrations and draw or calculate Ozone Analyzers by Gas Phase Titration,’’
the NO2 calibration curve. EPA–R2–73–246, March 1974.
2.4.10 Determination of converter efficiency. 9. J. A. Hodgeson, R. K. Stevens, and B. E.
Martin, ‘‘A Stable Ozone Source Applicable
2.4.10.1 Plot [NOX]M (y-axis) versus [NO2]OUT
as a Secondary Standard for Calibration of
(x-axis) and draw or calculate the converter
Atmospheric Monitors,’’ ISA Transactions,
efficiency curve. The slope of the curve
11, 161 (1972).
times 100 is the average converter efficiency,
10. A. E. O’Keeffe and G. C. Ortman, ‘‘Pri-
EC. The average converter efficiency must be
mary Standards for Trace Gas Analysis,’’
greater than 96 percent; if it is less than 96 Anal. Chem., 38, 760 (1966).
percent, replace or service the converter.
11. F. P. Scaringelli, A. E. O’Keeffe, E.
NOTE: Supplemental information on cali- Rosenberg, and J. P. Bell, ‘‘Preparation of

bration and other procedures in this method Known Concentrations of Gases and Vapors
are given in reference 13. with Permeation Devices Calibrated Gravi-
3. Frequency of calibration. The frequency of metrically,’’ Anal. Chem., 42, 871 (1970).
calibration, as well as the number of points 12. H. L. Rook, E. E. Hughes, R. S. Fuerst,
necessary to establish the calibration curve and J. H. Margeson, ‘‘Operation Characteris-
and the frequency of other performance tics of NO2 Permeation Devices,’’ Presented
Fecha: 14/11/2013
checks, will vary from one analyzer to an- at 167th National ACS Meeting, Los Angeles,
other. The user’s quality control program CA, April 1974.
should provide guidelines for initial estab- 13. E. C. Ellis, ‘‘Technical Assistance Docu-
lishment of these variables and for subse- ment for the Chemiluminescence Measure-
quent alteration as operational experience is ment of Nitrogen Dioxide,’’ EPA–E600/4–75–
accumulated. Manufacturers of analyzers 003 (Available in draft form from the United
should include in their instruction/operation States Environmental Protection Agency,
manuals information and guidance as to Department E (MD–76), Environmental Moni-
these variables and on other matters of oper- toring and Support Laboratory, Research
ation, calibration, and quality control. Triangle Park, NC 27711).
54
ER31AU93.055</MATH>
14. A Procedure for Establishing 15. Quality Assurance Handbook for Air Pol-
Traceability of Gas Mixtures to Certain Na- lution Measurement Systems, Volume II, Ambi-
tional Bureau of Standards Standard Ref- ent Air Specific Methods. The U.S. Environ-
erence Materials. EPA–600/7–81–010, Joint mental Protection Agency, Environmental
publication by NBS and EPA. Available from Monitoring Systems Laboratory, Research
the U.S. Environmental Protection Agency, Triangle Park, NC 27711. Publication No.
Environmental Monitoring Systems Labora- EAP–600/4–77–027a.
tory (MD–77), Research Triangle Park, NC
27711, May 1981.

Fecha: 14/11/2013
55
50-761.eps</GPH>
Pt. 50, App. G 40 CFR Ch. I (7–1–03 Edition)
[41 FR 52688, Dec. 1, 1976, as amended at 48 FR 2529, Jan 20, 1983]
APPENDIX G TO PART 50—REFERENCE 1.4 The ultrasonication extraction with

METHOD FOR THE DETERMINATION OF HNO3/HCl will extract metals other than lead
from ambient particulate matter.
LEAD IN SUSPENDED PARTICULATE
2. Range, sensitivity, and lower detectable
MATTER COLLECTED FROM AMBIENT limit. The values given below are typical of
AIR the methods capabilities. Absolute values
will vary for individual situations depending
1. Principle and applicability.
on the type of instrument used, the lead line,
1.1 Ambient air suspended particulate mat- and operating conditions.
ter is collected on a glass-fiber filter for 24
2.1 Range. The typical range of the method
hours using a high volume air sampler. The is 0.07 to 7.5 µg Pb/m3 assuming an upper lin-
analysis of the 24-hour samples may be per- ear range of analysis of 15 µg/ml and an air
formed for either individual samples or com- volume of 2,400 m3.
posites of the samples collected over a cal- 2.2 Sensitivity. Typical sensitivities for a 1
endar month or quarter, provided that the percent change in absorption (0.0044 absorb-
compositing procedure has been approved in ance units) are 0.2 and 0.5 µg Pb/ml for the
accordance with section 2.8 of appendix C to 217.0 and 283.3 nm lines, respectively.
part 58 of this chapter—Modifications of meth-
2.3 Lower detectable limit (LDL). A typical

ods by users. (Guidance or assistance in re- LDL is 0.07 µg Pb/m3. The above value was
questing approval under Section 2.8 can be calculated by doubling the between-labora-
obtained from the address given in section tory standard deviation obtained for the low-
2.7 of appendix C to part 58 of this chapter.) est measurable lead concentration in a col-
1.2 Lead in the particulate matter is solu- laborative test of the method.(15) An air vol-
bilized by extraction with nitric acid (HNO3), ume of 2,400 m3 was assumed.
Fecha: 14/11/2013
facilitated by heat or by a mixture of HNO3 3. Interferences. Two types of interferences

and hydrochloric acid (HCl) facilitated by are possible: chemical and light scattering.
ultrasonication. 3.1 Chemical. Reports on the absence (1, 2, 3,
1.3 The lead content of the sample is ana- 4, 5) of chemical interferences far outweigh
lyzed by atomic absorption spectrometry those reporting their presence, (6) therefore,
using an air-acetylene flame, the 283.3 or no correction for chemical interferences is
217.0 nm lead absorption line, and the opti- given here. If the analyst suspects that the
mum instrumental conditions recommended sample matrix is causing a chemical inter-
by the manufacturer. ference, the interference can be verified and
corrected for by carrying out the analysis
56
50-762.eps</GPH>
Environmental Protection Agency Pt. 50, App. G
with and without the method of standard ad- 5.2.2.3 Pipettes. To deliver 50, 30, 15, 8, 4, 2,
ditions.(7) 1 ml.
3.2 Light scattering. Nonatomic absorption 5.2.2.4 Cleaning. All glassware should be
or light scattering, produced by high con- scrupulously cleaned. The following proce-
centrations of dissolved solids in the sample, dure is suggested. Wash with laboratory de-
can produce a significant interference, espe- tergent, rinse, soak for 4 hours in 20 percent
cially at low lead concentrations. (2) The in- (w/w) HNO3, rinse 3 times with distilled-de-
terference is greater at the 217.0 nm line ionized water, and dry in a dust free manner.
than at the 283.3 nm line. No interference 5.2.3 Hot plate.
was observed using the 283.3 nm line with a 5.2.4. Ultrasonication water bath, unheated.
similar method.(1) Commercially available laboratory ultra-
Light scattering interferences can, how- sonic cleaning baths of 450 watts or higher
ever, be corrected for instrumentally. Since ‘‘cleaning power,’’ i.e., actual ultrasonic
the dissolved solids can vary depending on power output to the bath have been found
the origin of the sample, the correction may satisfactory.
be necessary, especially when using the 217.0 5.2.5 Template. To aid in sectioning the
nm line. Dual beam instruments with a con- glass-fiber filter. See figure 1 for dimensions.
tinuum source give the most accurate cor- 5.2.6 Pizza cutter. Thin wheel. Thickness
rection. A less accurate correction can be ob- 1mm.
tained by using a nonabsorbing lead line that 5.2.7 Watch glass.
is near the lead analytical line. Information 5.2.8 Polyethylene bottles. For storage of
on use of these correction techniques can be samples. Linear polyethylene gives better
obtained from instrument manufacturers’ storage stability than other polyethylenes
manuals. and is preferred.
If instrumental correction is not feasible, 5.2.9 Parafilm ‘‘M’’.1 American Can Co.,
the interference can be eliminated by use of Marathon Products, Neenah, Wis., or equiva-
the ammonium pyrrolidinecarbodithioate- lent.
methylisobutyl ketone, chelation-solvent ex- 6. Reagents.
traction technique of sample preparation.(8) 6.1 Sampling.
4. Precision and bias. 6.1.1 Glass fiber filters. The specifications
4.1 The high-volume sampling procedure given below are intended to aid the user in
used to collect ambient air particulate mat- obtaining high quality filters with reproduc-
ter has a between-laboratory relative stand- ible properties. These specifications have
ard deviation of 3.7 percent over the range 80 been met by EPA contractors.
to 125 µg/m3.(9) The combined extraction- 6.1.1.1 Lead content. The absolute lead con-
analysis procedure has an average within- tent of filters is not critical, but low values
laboratory relative standard deviation of 5 to are, of course, desirable. EPA typically ob-
6 percent over the range 1.5 to 15 µg Pb/ml, tains filters with a lead content of 75 µg/fil-
and an average between laboratory relative ter.
standard deviation of 7 to 9 percent over the It is important that the variation in lead
same range. These values include use of ei- content from filter to filter, within a given
ther extraction procedure. batch, be small.
4.2 Single laboratory experiments and col- 6.1.1.2 Testing.
laborative testing indicate that there is no 6.1.1.2.1 For large batches of filters (>500
significant difference in lead recovery be- filters) select at random 20 to 30 filters from
tween the hot and ultrasonic extraction pro- a given batch. For small batches (>500 fil-
cedures.(15) ters) a lesser number of filters may be taken.
5. Apparatus. Cut one 3⁄4″×8″ strip from each filter any-
5.1 Sampling. where in the filter. Analyze all strips, sepa-
5.1.1 High-Volume Sampler. Use and cali- rately, according to the directions in sec-
brate the sampler as described in appendix B tions 7 and 8.

to this part. 6.1.1.2.2 Calculate the total lead in each fil-
5.2 Analysis. ter as
5.2.1 Atomic absorption spectrophotometer.
100 ml 12 strips
Equipped with lead hollow cathode or
electrodeless discharge lamp.
Fb = µg Pb/ml × ×
5.2.1.1 Acetylene. The grade recommended
strip filter
Fecha: 14/11/2013
by the instrument manufacturer should be where:

used. Change cylinder when pressure drops Fb = Amount of lead per 72 square inches of
below 50–100 psig. filter, µg.
5.2.1.2 Air. Filtered to remove particulate,
oil, and water. 6.1.1.2.3 Calculate the mean, Fb, of the val-
5.2.2 Glassware. Class A borosilicate glass- ues and the relative standard deviation
ware should be used throughout the analysis.
5.2.2.1 Beakers. 30 and 150 ml. graduated, 1 Mention of commercial products does not
Pyrex. imply endorsement by the U.S. Environ-

5.2.2.2 Volumetric flasks. 100-ml. mental Protection Agency.
57
EC08NO91.084</MATH>
(standard deviation/mean × 100). If the rel- 6.3.1 Master standard, 1000 µg Pb/ml in
ative standard deviation is high enough so HNO3. Dissolve 1.598 g of Pb(NO3)2 in 0.45 M
that, in the analysts opinion, subtraction of HNO3 contained in a 1 l volumetric flask and
Fb, (section 10.3) may result in a significant dilute to volume with 0.45 M HNO3.
error in the µg Pb/m3, the batch should be re- 6.3.2 Master standard, 1000 µg Pb/ml in
jected. HNO3/HCl. Prepare as in section 6.3.1 except
6.1.1.2.4 For acceptable batches, use the use the HNO3/HCl solution in section 6.2.7.
value of Fb to correct all lead analyses (sec- Store standards in a polyethylene bottle.
tion 10.3) of particulate matter collected Commercially available certified lead stand-
using that batch of filters. If the analyses ard solutions may also be used.
are below the LDL (section 2.3) no correction 7. Procedure.
is necessary. 7.1 Sampling. Collect samples for 24 hours
6.2 Analysis. using the procedure described in reference 10
6.2.1 Concentrated (15.6 M) HNO3. ACS rea- with glass-fiber filters meeting the specifica-
gent grade HNO3 and commercially available tions in section 6.1.1. Transport collected
redistilled HNO3 has found to have suffi- samples to the laboratory taking care to
ciently low lead concentrations. minimize contamination and loss of sam-
ple. (16).
6.2.2 Concentrated (11.7 M) HCl. ACS rea-
7.2 Sample preparation.
gent grade.
7.2.1 Hot extraction procedure.
6.2.3 Distilled-deionized water. (D.I. water). 7.2.1.1 Cut a 3⁄4″×8″ strip from the exposed
6.2.4 3 M HNO3. This solution is used in the filter using a template and a pizza cutter as
hot extraction procedure. To prepare, add 192 described in Figures 1 and 2. Other cutting
ml of concentrated HNO3 to D.I. water in a 1 procedures may be used.
l volumetric flask. Shake well, cool, and di- Lead in ambient particulate matter col-
lute to volume with D.I. water. Caution: Ni- lected on glass fiber filters has been shown
tric acid fumes are toxic. Prepare in a well to be uniformly distributed across the fil-
ventilated fume hood. ter.1, 3, 11 Another study 12 has shown that
6.2.5 0.45 M HNO3. This solution is used as when sampling near a roadway, strip posi-
the matrix for calibration standards when tion contributes significantly to the overall
using the hot extraction procedure. To pre- variability associated with lead analyses.
pare, add 29 ml of concentrated HNO3 to D.I. Therefore, when sampling near a roadway,
water in a 1 l volumetric flask. Shake well, additional strips should be analyzed to mini-
cool, and dilute to volume with D.I. water. mize this variability.
6.2.6 2.6 M HNO3+0 to 0.9 M HCl. This solu- 7.2.1.2 Fold the strip in half twice and place
tion is used in the ultrasonic extraction pro- in a 150-ml beaker. Add 15 ml of 3 M HNO3 to
cedure. The concentration of HCl can be var- cover the sample. The acid should com-
ied from 0 to 0.9 M. Directions are given for pletely cover the sample. Cover the beaker
preparation of a 2.6 M HNO3+0.9 M HCl solu- with a watch glass.
tion. Place 167 ml of concentrated HNO3 into 7.2.1.3 Place beaker on the hot-plate, con-
a 1 l volumetric flask and add 77 ml of contained in a fume hood, and boil gently for 30
centrated HCl. Stir 4 to 6 hours, dilute to min. Do not let the sample evaporate to dry-
nearly 1 l with D.I. water, cool to room tem- ness. Caution: Nitric acid fumes are toxic.
perature, and dilute to 1 l. 7.2.1.4 Remove beaker from hot plate and
6.2.7 0.40 M HNO3 + X M HCl. This solution cool to near room temperature.
is used as the matrix for calibration stand- 7.2.1.5 Quantitatively transfer the sample
ards when using the ultrasonic extraction as follows:
procedure. To prepare, add 26 ml of con- 7.2.1.5.1 Rinse watch glass and sides of
centrated HNO3, plus the ml of HCl required, beaker with D.I. water.
to a 1 l volumetric flask. Dilute to nearly 1 7.2.1.5.2 Decant extract and rinsings into a
l with D.I. water, cool to room temperature, 100-ml volumetric flask.

and dilute to 1 l. The amount of HCl required 7.2.1.5.3 Add D.I. water to 40 ml mark on
can be determined from the following equa- beaker, cover with watch glass, and set aside
tion: for a minimum of 30 minutes. This is a crit-
ical step and cannot be omitted since it al-
77 ml × 0.15 × lows the HNO3 trapped in the filter to diffuse
y= into the rinse water.
Fecha: 14/11/2013
0.9 M 7.2.1.5.4 Decant the water from the filter

where: into the volumetric flask.
y = ml of concentrated HCl required. 7.2.1.5.5 Rinse filter and beaker twice with
x = molarity of HCl in 6.2.6. D.I. water and add rinsings to volumetric
0.15 = dilution factor in 7.2.2. flask until total volume is 80 to 85 ml.
7.2.1.5.6 Stopper flask and shake vigor-
6.2.8 Lead nitrate, Pb(NO3)2. ACS reagent ously. Set aside for approximately 5 minutes
grade, purity 99.0 percent. Heat for 4 hours at or until foam has dissipated.
120 °C and cool in a desiccator. 7.2.1.5.7 Bring solution to volume with D.I.
6.3 Calibration standards. water. Mix thoroughly.
58
EC08NO91.085</MATH>
7.2.1.5.8 Allow solution to settle for one 9.1 Working standard, 20 µg Pb/ml. Prepared
hour before proceeding with analysis. by diluting 2.0 ml of the master standard
7.2.1.5.9 If sample is to be stored for subse- (section 6.3.1 if the hot acid extraction was
quent analysis, transfer to a linear poly- used or section 6.3.2 if the ultrasonic extrac-
ethylene bottle. tion procedure was used) to 100 ml with acid
7.2.2 Ultrasonic extraction procedure. of the same concentration as used in pre-
7.2.2.1 Cut a 3⁄4″×8″ strip from the exposed paring the master standard.
filter as described in section 7.2.1.1. 9.2 Calibration standards. Prepare daily by
7.2.2.2 Fold the strip in half twice and place diluting the working standard, with the
in a 30 ml beaker. Add 15 ml of the HNO3/HCl same acid matrix, as indicated below. Other
solution in section 6.2.6. The acid should lead concentrations may be used.
completely cover the sample. Cover the
Concentra-
beaker with parafilm. Volume of 20 µg/ml working stand- Final vol- tion µg Pb/
The parafilm should be placed over the ard, ml ume, ml ml
beaker such that none of the parafilm is in
contact with water in the ultrasonic bath. 0 ..................................................... 100 0
1.0 .................................................. 200 0.1
Otherwise, rinsing of the parafilm (section 2.0 .................................................. 200 0.2
7.2.2.4.1) may contaminate the sample. 2.0 .................................................. 100 0.4
7.2.2.3 Place the beaker in the 4.0 .................................................. 100 0.8
ultrasonication bath and operate for 30 min- 8.0 .................................................. 100 1.6
utes. 15.0 ................................................ 100 3.0
7.2.2.4 Quantitatively transfer the sample 30.0 ................................................ 100 6.0
50.0 ................................................ 100 10.0
as follows:
100.0 .............................................. 100 20.0
7.2.2.4.1 Rinse parafilm and sides of beaker
with D.I. water. 9.3 Preparation of calibration curve. Since
7.2.2.4.2 Decant extract and rinsings into a the working range of analysis will vary de-
100 ml volumetric flask. pending on which lead line is used and the
7.2.2.4.3 Add 20 ml D.I. water to cover the type of instrument, no one set of instruc-
filter strip, cover with parafilm, and set tions for preparation of a calibration curve
aside for a minimum of 30 minutes. This is a can be given. Select standards (plus the rea-
critical step and cannot be omitted. The gent blank), in the same acid concentration
sample is then processed as in sections as the samples, to cover the linear absorp-
7.2.1.5.4 through 7.2.1.5.9. tion range indicated by the instrument man-
NOTE: Samples prepared by the hot extrac- ufacturer. Measure the absorbance of the
tion procedure are now in 0.45 M HNO3. Sam- blank and standards as in section 8.0. Repeat
ples prepared by the ultrasonication proce- until good agreement is obtained between
dure are in 0.40 M HNO3 + X M HCl. replicates. Plot absorbance (y-axis) versus
8. Analysis. concentration in µg Pb/ml (x-axis). Draw (or
8.1 Set the wavelength of the compute) a straight line through the linear
monochromator at 283.3 or 217.0 nm. Set or portion of the curve. Do not force the cali-
align other instrumental operating condi- bration curve through zero. Other calibra-
tions as recommended by the manufacturer. tion procedures may be used.
8.2 The sample can be analyzed directly To determine stability of the calibration
from the volumetric flask, or an appropriate curve, remeasure—alternately—one of the
amount of sample decanted into a sample following calibration standards for every
analysis tube. In either case, care should be 10th sample analyzed: Concentration ≤ 1µg
taken not to disturb the settled solids. Pb/ml; concentration ≤ 10 µg Pb/ml. If either
8.3 Aspirate samples, calibration standards standard deviates by more than 5 percent
and blanks (section 9.2) into the flame and from the value predicted by the calibration
record the equilibrium absorbance. curve, recalibrate and repeat the previous 10
8.4 Determine the lead concentration in µg analyses.
Pb/ml, from the calibration curve, section 10. Calculation.
9.3. 10.1 Measured air volume. Calculate the
8.5 Samples that exceed the linear calibra- measured air volume at Standard Tempera-
tion range should be diluted with acid of the ture and Pressure as described in Reference
Fecha: 14/11/2013
same concentration as the calibration stand- 10.

ards and reanalyzed. 10.2 Lead concentration. Calculate lead con-
9. Calibration. centration in the air sample.
59
where: 1. Scott, D. R. et al. ‘‘Atomic Absorption

C = Concentration, µg Pb/sm3. and Optical Emission Analysis of NASN At-
µg Pb/ml = Lead concentration determined mospheric Particulate Samples for Lead.’’
from section 8. Envir. Sci. and Tech., 10, 877–880 (1976).
100 ml/strip = Total sample volume. 2. Skogerboe, R. K. et al. ‘‘Monitoring for
12 strips = Total useable filter area, 8″×9″. Ex- Lead in the Environment.’’ pp. 57–66, Depart-
posed area of one strip, 3⁄4″×7″. ment of Chemistry, Colorado State Univer-
Filter = Total area of one strip, 3⁄4″×8″. sity, Fort Collins, CO 80523. Submitted to
Fb = Lead concentration of blank filter, µg, National Science Foundation for publica-
from section 6.1.1.2.3. tions, 1976.
VSTP = Air volume from section 10.2. 3. Zdrojewski, A. et al. ‘‘The Accurate
Measurement of Lead in Airborne Particu-
11. Quality control. lates.’’ Inter. J. Environ. Anal. Chem., 2, 63–77
⁄ ″×8″ glass fiber filter strips containing 80
34
(1972).
to 2000 µg Pb/strip (as lead salts) and blank
4. Slavin, W., ‘‘Atomic Absorption Spec-
strips with zero Pb content should be used to
troscopy.’’ Published by Interscience Com-
determine if the method—as being used—has
any bias. Quality control charts should be pany, New York, NY (1968).
established to monitor differences between 5. Kirkbright, G. F., and Sargent, M.,
measured and true values. The frequency of ‘‘Atomic Absorption and Fluorescence Spec-
such checks will depend on the local quality troscopy.’’ Published by Academic Press,
control program. New York, NY 1974.
To minimize the possibility of generating 6. Burnham, C. D. et al., ‘‘Determination of
unreliable data, the user should follow prac- Lead in Airborne Particulates in Chicago
tices established for assuring the quality of and Cook County, IL, by Atomic Absorption
air pollution data, (13) and take part in Spectroscopy.’’ Envir. Sci. and Tech., 3, 472–
EPA’s semiannual audit program for lead 475 (1969).
analyses. 7. ‘‘Proposed Recommended Practices for
12. Trouble shooting. Atomic Absorption Spectrometry.’’ ASTM
1. During extraction of lead by the hot ex- Book of Standards, part 30, pp. 1596–1608 (July
traction procedure, it is important to keep 1973).
the sample covered so that corrosion 8. Koirttyohann, S. R. and Wen, J. W.,
products—formed on fume hood surfaces ‘‘Critical Study of the APCD–MIBK Extrac-
which may contain lead—are not deposited tion System for Atomic Absorption.’’ Anal.
in the extract. Chem., 45, 1986–1989 (1973).
2. The sample acid concentration should 9. Collaborative Study of Reference Method
minimize corrosion of the nebulizer. How- for the Determination of Suspended Particulates
ever, different nebulizers may require lower in the Atmosphere (High Volume Method). Ob-
acid concentrations. Lower concentrations tainable from National Technical Informa-
can be used provided samples and standards tion Service, Department of Commerce, Port
have the same acid concentration. Royal Road, Springfield, VA 22151, as PB–205–
3. Ashing of particulate samples has been 891.

found, by EPA and contractor laboratories,
10. [Reserved]
to be unnecessary in lead analyses by atomic
11. Dubois, L., et al., ‘‘The Metal Content
absorption. Therefore, this step was omitted
from the method. of Urban Air.’’ JAPCA, 16, 77–78 (1966).
4. Filtration of extracted samples, to re- 12. EPA Report No. 600/4–77–034, June 1977,
move particulate matter, was specifically ex- ‘‘Los Angeles Catalyst Study Symposium.’’
Fecha: 14/11/2013
cluded from sample preparation, because Page 223.

some analysts have observed losses of lead 13. Quality Assurance Handbook for Air Pol-
due to filtration. lution Measurement System. Volume 1—Prin-
5. If suspended solids should clog the ciples. EPA–600/9–76–005, March 1976.
nebulizer during analysis of samples, cen- 14. Thompson, R. J. et al., ‘‘Analysis of Se-
trifuge the sample to remove the solids. lected Elements in Atmospheric Particulate
13. Authority. Matter by Atomic Absorption.’’ Atomic Ab-
(Secs. 109 and 301(a), Clean Air Act, as sorption Newsletter, 9, No. 3, May-June 1970.
amended (42 U.S.C. 7409, 7601(a))) 15. To be published. EPA, QAB, EMSL,
14. References. RTP, N.C. 27711
60
50-766.eps</MATH>
16. Quality Assurance Handbook for Air Pol- ent Air Specific Methods. EPA–600/4–77/027a,
lution Measurement Systems. Volume II—Ambi- May 1977.

Fecha: 14/11/2013
61
50-767.eps</GPH>
Pt. 50, App. H 40 CFR Ch. I (7–1–03 Edition)
(Secs. 109, 301(a) of the Clean Air Act, as amended (42 U.S.C. 7409, 7601(a)); secs. 110, 301(a) and
319 of the Clean Air Act (42 U.S.C. 7410, 7601(a), 7619))
[43 FR 46258, Oct. 5, 1978; 44 FR 37915, June 29, 1979, as amended at 46 FR 44163, Sept. 3, 1981;
52 FR 24664, July 1, 1987]
APPENDIX H TO PART 50—INTERPRETA- amples are contained in the ‘‘Guideline for

TION OF THE 1-HOUR PRIMARY AND Interpretation of Ozone Air Quality Stand-
Fecha: 14/11/2013
SECONDARY NATIONAL AMBIENT AIR ards.’’ For purposes of clarity in the fol-
QUALITY STANDARDS FOR OZONE lowing discussion, it is convenient to use the
term ‘‘exceedance’’ to describe a daily max-
1. GENERAL imum hourly average ozone measurement
This appendix explains how to determine that is greater than the level of the stand-
when the expected number of days per cal- ard. Therefore, the phrase ‘‘expected number
endar year with maximum hourly average of days with maximum hourly average ozone
concentrations above 0.12 ppm (235 µg/m3) is concentrations above the level of the stand-
equal to or less than 1. An expanded discus- ard’’ may be simply stated as the ‘‘expected
sion of these procedures and associated ex- number of exceedances.’’
62
50-768.eps</GPH>
Environmental Protection Agency Pt. 50, App. H
The basic principle in making this deter- ed. Any such waiver of the ozone monitoring
mination is relatively straightforward. Most requirement would be handled under provi-
of the complications that arise in deter- sions of 40 CFR, part 58. Some allowance
mining the expected number of annual should also be made for days for which valid
exceedances relate to accounting for incom- daily maximum hourly values were not ob-
plete sampling. In general, the average num- tained but which would quite likely have
ber of exceedances per calendar year must be been below the standard. Such an allowance
less than or equal to 1. In its simplest form, introduces a complication in that it becomes
the number of exceedances at a monitoring necessary to define under what conditions a
site would be recorded for each calendar year missing value may be assumed to have been
and then averaged over the past 3 calendar less than the level of the standard. The fol-
years to determine if this average is less lowing criterion may be used for ozone:
than or equal to 1. A missing daily maximum ozone value
2. INTERPRETATION OF EXPECTED may be assumed to be less than the level of
EXCEEDANCES the standard if the valid daily maxima on
both the preceding day and the following day
The ozone standard states that the ex- do not exceed 75 percent of the level of the
pected number of exceedances per year must standard.
be less than or equal to 1. The statistical Let z denote the number of missing daily
term ‘‘expected number’’ is basically an maximum values that may be assumed to be
arithmetic average. The following example
less than the standard. Then the following
explains what it would mean for an area to
formula shall be used to estimate the ex-
be in compliance with this type of standard.
pected number of exceedances for the year:
Suppose a monitoring station records a valid
daily maximum hourly average ozone value
for every day of the year during the past 3
e = v + [( v/n ) * ( N-n-z)] (1)
years. At the end of each year, the number of (*Indicates multiplication.)
days with maximum hourly concentrations
where:
above 0.12 ppm is determined and this num-
ber is averaged with the results of previous e = the estimated number of exceedances for
years. As long as this average remains ‘‘less the year,
than or equal to 1,’’ the area is in compli- N = the number of required monitoring days
ance. in the year,
n = the number of valid daily maxima,
3. ESTIMATING THE NUMBER OF EXCEEDANCES v = the number of daily values above the
FOR A YEAR
level of the standard, and
In general, a valid daily maximum hourly z = the number of days assumed to be less
average value may not be available for each than the standard level.
day of the year, and it will be necessary to
account for these missing values when esti- This estimated number of exceedances
mating the number of exceedances for a par- shall be rounded to one decimal place (frac-
ticular calendar year. The purpose of these tional parts equal to 0.05 round up).
computations is to determine if the expected It should be noted that N will be the total
number of exceedances per year is less than number of days in the year unless the appro-
or equal to 1. Thus, if a site has two or more priate Regional Administrator has granted a
observed exceedances each year, the stand- waiver under the provisions of 40 CFR part
ard is not met and it is not necessary to use 58.
the procedures of this section to account for The above equation may be interpreted in-
incomplete sampling. tuitively in the following manner. The esti-
The term ‘‘missing value’’ is used here in mated number of exceedances is equal to the
the general sense to describe all days that do observed number of exceedances (v) plus an
not have an associated ozone measurement. increment that accounts for incomplete sam-
In some cases, a measurement might actu- pling. There were (N-n) missing values for
ally have been missed but in other cases no the year but a certain number of these,
measurement may have been scheduled for namely z, were assumed to be less than the
that day. A daily maximum ozone value is standard. Therefore, (N-n-z) missing values
Fecha: 14/11/2013
defined to be the highest hourly ozone value are considered to include possible
recorded for the day. This daily maximum exceedances. The fraction of measured val-
value is considered to be valid if 75 percent of ues that are above the level of the standard
the hours from 9:01 a.m. to 9:00 p.m. (LST) is v/n. It is assumed that this same fraction
were measured or if the highest hour is applies to the (N-n-z) missing values and
greater than the level of the standard. that (v/n)*(N-n-z) of these values would also
In some areas, the seasonal pattern of have exceeded the level of the standard.
ozone is so pronounced that entire months
need not be sampled because it is extremely [44 FR 8220, Feb. 8, 1979, as amended at 62 FR
unlikely that the standard would be exceed- 38895, July 18, 1997]
63
EC08NO91.086</MATH>
Pt. 50, App. I 40 CFR Ch. I (7–1–03 Edition)
APPENDIX I TO PART 50—INTERPRETA- centration for each calendar day with ambi-
TION OF THE 8-HOUR PRIMARY AND ent ozone monitoring data. Because the 8-
SECONDARY NATIONAL AMBIENT AIR hour averages are recorded in the start hour,
QUALITY STANDARDS FOR OZONE the daily maximum 8-hour concentrations
from two consecutive days may have some
1. General. hourly concentrations in common. Gen-
This appendix explains the data handling erally, overlapping daily maximum 8-hour
conventions and computations necessary for averages are not likely, except in those non-
determining whether the national 8-hour pri- urban monitoring locations with less pro-
mary and secondary ambient air quality nounced diurnal variation in hourly con-
standards for ozone specified in § 50.10 are centrations.
met at an ambient ozone air quality moni- (b) An ozone monitoring day shall be
toring site. Ozone is measured in the ambi- counted as a valid day if valid 8-hour aver-
ent air by a reference method based on ap- ages are available for at least 75% of possible
pendix D of this part. Data reporting, data hours in the day (i.e., at least 18 of the 24
handling, and computation procedures to be averages). In the event that less than 75% of
used in making comparisons between re-
the 8-hour averages are available, a day shall
ported ozone concentrations and the level of
also be counted as a valid day if the daily
the ozone standard are specified in the fol-
maximum 8-hour average concentration for
lowing sections. Whether to exclude, retain,
or make adjustments to the data affected by that day is greater than the level of the am-
stratospheric ozone intrusion or other nat- bient standard.
ural events is subject to the approval of the 2.2 Primary and Secondary Standard-related
appropriate Regional Administrator. Summary Statistic. The standard-related sum-
2. Primary and Secondary Ambient Air Qual- mary statistic is the annual fourth-highest
ity Standards for Ozone. daily maximum 8-hour ozone concentration,
2.1 Data Reporting and Handling Conven- expressed in parts per million, averaged over
tions. three years. The 3-year average shall be com-
2.1.1 Computing 8-hour averages. Hourly av- puted using the three most recent, consecu-
erage concentrations shall be reported in tive calendar years of monitoring data meet-
parts per million (ppm) to the third decimal ing the data completeness requirements de-
place, with additional digits to the right scribed in this appendix. The computed 3-
being truncated. Running 8-hour averages year average of the annual fourth-highest
shall be computed from the hourly ozone daily maximum 8-hour average ozone con-
concentration data for each hour of the year centrations shall be expressed to three dec-
and the result shall be stored in the first, or imal places (the remaining digits to the
start, hour of the 8-hour period. An 8-hour right are truncated.)
average shall be considered valid if at least 2.3 Comparisons with the Primary and Sec-
75% of the hourly averages for the 8-hour pe- ondary Ozone Standards. (a) The primary and
riod are available. In the event that only 6 secondary ozone ambient air quality stand-
(or 7) hourly averages are available, the 8- ards are met at an ambient air quality moni-
hour average shall be computed on the basis toring site when the 3-year average of the
of the hours available using 6 (or 7) as the di- annual fourth-highest daily maximum 8-hour
visor. (8-hour periods with three or more average ozone concentration is less than or
missing hours shall not be ignored if, after equal to 0.08 ppm. The number of significant
substituting one-half the minimum detect- figures in the level of the standard dictates
able limit for the missing hourly concentra- the rounding convention for comparing the
tions, the 8-hour average concentration is
computed 3-year average annual fourth-high-
greater than the level of the standard.) The
est daily maximum 8-hour average ozone
computed 8-hour average ozone concentra-
concentration with the level of the standard.

tions shall be reported to three decimal
The third decimal place of the computed
places (the insignificant digits to the right of
value is rounded, with values equal to or
the third decimal place are truncated, con-
sistent with the data handling procedures for greater than 5 rounding up. Thus, a com-
the reported data.) puted 3-year average ozone concentration of
2.1.2 Daily maximum 8-hour average con- 0.085 ppm is the smallest value that is great-
centrations. (a) There are 24 possible running er than 0.08 ppm.
Fecha: 14/11/2013
8-hour average ozone concentrations for each (b) This comparison shall be based on three
calendar day during the ozone monitoring consecutive, complete calendar years of air
season. (Ozone monitoring seasons vary by quality monitoring data. This requirement is
geographic location as designated in part 58, met for the three year period at a moni-
appendix D to this chapter.) The daily max- toring site if daily maximum 8-hour average
imum 8-hour concentration for a given cal- concentrations are available for at least 90%,
endar day is the highest of the 24 possible 8- on average, of the days during the designated
hour average concentrations computed for ozone monitoring season, with a minimum
that day. This process is repeated, yielding a data completeness in any one year of at least
daily maximum 8-hour average ozone con- 75% of the designated sampling days. When
64
Environmental Protection Agency Pt. 50, App. J
computing whether the minimum data com- tion if the average annual fourth maximum
pleteness requirements have been met, mete- 8-hour concentration is greater than the
orological or ambient data may be sufficient level of the standard.
to demonstrate that meteorological condi- (d) Comparisons with the primary and sec-
tions on missing days were not conducive to ondary ozone standards are demonstrated by
concentrations above the level of the stand- examples 1 and 2 in paragraphs (d)(1) and (d)
ard. Missing days assumed less than the level (2) respectively as follows:
of the standard are counted for the purpose (1) As shown in example 1, the primary and
of meeting the data completeness require- secondary standards are met at this moni-
ment, subject to the approval of the appro- toring site because the 3-year average of the
priate Regional Administrator. annual fourth-highest daily maximum 8-hour
(c) Years with concentrations greater than average ozone concentrations (i.e., 0.084 ppm)
the level of the standard shall not be ignored is less than or equal to 0.08 ppm. The data
on the ground that they have less than com- completeness requirement is also met be-
plete data. Thus, in computing the 3-year av- cause the average percent of days with valid
erage fourth maximum concentration, cal- ambient monitoring data is greater than
endar years with less than 75% data com- 90%, and no single year has less than 75%
pleteness shall be included in the computa- data completeness.
EXAMPLE 1. AMBIENT MONITORING SITE ATTAINING THE PRIMARY AND SECONDARY OZONE STANDARDS
1st Highest 2nd Highest 3rd Highest 4th Highest 5th Highest
Percent Daily Max Daily Max Daily Max Daily Max Daily Max
Year Valid Days 8-hour 8-hour 8-hour 8-hour 8-hour
Conc. (ppm) Conc. (ppm) Conc. (ppm) Conc. (ppm) Conc. (ppm)
1993 ...................................................... 100% 0.092 0.091 0.090 0.088 0.085
1994 ...................................................... 96% 0.090 0.089 0.086 0.084 0.080
1995 ...................................................... 98% 0.087 0.085 0.083 0.080 0.075
Average .......................................... 98%
(2) As shown in example 2, the primary and concentration data for 1994 is used in these
secondary standards are not met at this computations, even though the data capture
monitoring site because the 3-year average is less than 75%, because the average fourth-
of the fourth-highest daily maximum 8-hour highest daily maximum 8-hour average con-
average ozone concentrations (i.e., 0.093 ppm) centration is greater than 0.08 ppm.
is greater than 0.08 ppm. Note that the ozone
EXAMPLE 2. AMBIENT MONITORING SITE FAILING TO MEET THE PRIMARY AND SECONDARY OZONE
STANDARDS
1st Highest 2nd Highest 3rd Highest 4th Highest 5th Highest
Percent Daily Max Daily Max Daily Max Daily Max Daily Max
Year Valid Days 8-hour 8-hour 8-hour 8-hour 8-hour
Conc. (ppm) Conc. (ppm) Conc. (ppm) Conc. (ppm) Conc. (ppm)
1993 ...................................................... 96% 0.105 0.103 0.103 0.102 0.102
1994 ...................................................... 74% 0.090 0.085 0.082 0.080 0.078

1995 ...................................................... 98% 0.103 0.101 0.101 0.097 0.095
Average .......................................... 89%
3. Design Values for Primary and Secondary average ozone concentration is also the air
Fecha: 14/11/2013
Ambient Air Quality Standards for Ozone. The quality design value for the site.
air quality design value at a monitoring site
is defined as that concentration that when [62 FR 38895, July 18, 1997]
reduced to the level of the standard ensures
that the site meets the standard. For a con- APPENDIX J TO PART 50—REFERENCE
centration-based standard, the air quality METHOD FOR THE DETERMINATION OF
design value is simply the standard-related PARTICULATE MATTER AS PM10 IN
test statistic. Thus, for the primary and sec- THE ATMOSPHERE
ondary ozone standards, the 3-year average
annual fourth-highest daily maximum 8-hour 1.0 Applicability.
65
Pt. 50, App. J 40 CFR Ch. I (7–1–03 Edition)
1.1 This method provides for the measure- mechanism, there may be no upper limit.
ment of the mass concentration of particu- For samplers that do not have an automatic
late matter with an aerodynamic diameter filter-changing mechanism, the upper limit
less than or equal to a nominal 10 microm- is determined by the filter mass loading be-
eters (PM1O) in ambient air over a 24-hour yond which the sampler no longer maintains
period for purposes of determining attain- the operating flow rate within specified lim-
ment and maintenance of the primary and its due to increased pressure drop across the
secondary national ambient air quality loaded filter. This upper limit cannot be
standards for particulate matter specified in specified precisely because it is a complex
§ 50.6 of this chapter. The measurement proc- function of the ambient particle size dis-
ess is nondestructive, and the PM10 sample tribution and type, humidity, filter type, and
can be subjected to subsequent physical or perhaps other factors. Nevertheless, all sam-
chemical analyses. Quality assurance proce- plers should be capable of measuring 24-hour
dures and guidance are provided in part 58, PM10 mass concentrations of at least 300 µg/
appendices A and B, of this chapter and in std m3 while maintaining the operating flow
References 1 and 2. rate within the specified limits.
2.0 Principle. 4.0 Precision.
2.1 An air sampler draws ambient air at a 4.1 The precision of PM10 samplers must be
constant flow rate into a specially shaped 5 µg/m3 for PM10 concentrations below 80 µg/
inlet where the suspended particulate matter m3 and 7 percent for PM10 concentrations
is inertially separated into one or more size above 80 µg/m3, as required by part 53 of this
fractions within the PM10 size range. Each chapter, which prescribes a test procedure
size fraction in the PM1O size range is then that determines the variation in the PM10
collected on a separate filter over the speci- concentration measurements of identical
fied sampling period. The particle size dis- samplers under typical sampling conditions.
crimination characteristics (sampling effec- Continual assessment of precision via collo-
tiveness and 50 percent cutpoint) of the sam- cated samplers is required by part 58 of this
pler inlet are prescribed as performance chapter for PM10 samplers used in certain
specifications in part 53 of this chapter. monitoring networks.
2.2 Each filter is weighed (after moisture 5.0 Accuracy.
equilibration) before and after use to deter- 5.1 Because the size of the particles making
mine the net weight (mass) gain due to col- up ambient particulate matter varies over a
lected PM10. The total volume of air sampled, wide range and the concentration of par-
corrected to EPA reference conditions (25 C, ticles varies with particle size, it is difficult
101.3 kPa), is determined from the measured to define the absolute accuracy of PM10 sam-
flow rate and the sampling time. The mass plers. Part 53 of this chapter provides a spec-
concentration of PM10 in the ambient air is ification for the sampling effectiveness of
computed as the total mass of collected par- PM10 samplers. This specification requires
ticles in the PM10 size range divided by the that the expected mass concentration cal-
volume of air sampled, and is expressed in culated for a candidate PM10 sampler, when
micrograms per standard cubic meter (µg/std sampling a specified particle size distribu-
m3). For PM10 samples collected at tempera- tion, be within ±10 percent of that calculated
tures and pressures significantly different for an ideal sampler whose sampling effec-
from EPA reference conditions, these cor- tiveness is explicitly specified. Also, the par-
rected concentrations sometimes differ sub- ticle size for 50 percent sampling
stantially from actual concentrations (in effectivensss is required to be 10±0.5 microm-
micrograms per actual cubic meter), particu- eters. Other specifications related to accu-
larly at high elevations. Although not re- racy apply to flow measurement and calibra-
quired, the actual PM10 concentration can be tion, filter media, analytical (weighing) pro-
calculated from the corrected concentration, cedures, and artifact. The flow rate accuracy
using the average ambient temperature and of PM10 samplers used in certain monitoring
barometric pressure during the sampling pe- networks is required by part 58 of this chap-
riod. ter to be assessed periodically via flow rate
2.3 A method based on this principle will be audits.
considered a reference method only if (a) the 6.0 Potential Sources of Error.
associated sampler meets the requirements 6.1 Volatile Particles. Volatile particles col-
specified in this appendix and the require- lected on filters are often lost during ship-
Fecha: 14/11/2013
ments in part 53 of this chapter, and (b) the ment and/or storage of the filters prior to
method has been designated as a reference the post-sampling weighing 3. Although ship-
method in accordance with part 53 of this ment or storage of loaded filters is some-
chapter. times unavoidable, filters should be re-
3.0 Range. weighed as soon as practical to minimize
3.1 The lower limit of the mass concentra- these losses.
tion range is determined by the repeatability 6.2 Artifacts. Positive errors in PM10 con-
of filter tare weights, assuming the nominal centration measurements may result from
air sample volume for the sampler. For sam- retention of gaseous species on filters 4, 5.
plers having an automatic filter-changing Such errors include the retention of sulfur
66
dioxide and nitric acid. Retention of sulfur d. Protect the filter and sampler from pre-
dioxide on filters, followed by oxidation to cipitation and prevent insects and other de-
sulfate, is referred to as artifact sulfate for- bris from being sampled.
mation, a phenomenon which increases with e. Minimize air leaks that would cause
increasing filter alkalinity 6. Little or no ar- error in the measurement of the air volume
tifact sulfate formation should occur using passing through the filter.
filters that meet the alkalinity specification f. Discharge exhaust air at a sufficient dis-
in section 7.2.4. Artifact nitrate formation, tance from the sampler inlet to minimize the
resulting primarily from retention of nitric sampling of exhaust air.
acid, occurs to varying degrees on many fil- g. Minimize the collection of dust from the
ter types, including glass fiber, cellulose supporting surface.
ester, and many quartz fiber filters 5, 7, 8, 9, 10. 7.1.2 The sampler shall have a sample air
Loss of true atmospheric particulate nitrate inlet system that, when operated within a
during or following sampling may also occur specified flow rate range, provides particle
due to dissociation or chemical reaction. size discrimination characteristics meeting
This phenomenon has been observed on Tef- all of the applicable performance specifica-
tions prescribed in part 53 of this chapter.
lon filters 8 and inferred for quartz fiber fil-
The sampler inlet shall show no significant
ters 11, 12. The magnitude of nitrate artifact
wind direction dependence. The latter re-
errors in PM10 mass concentration measure-
quirement can generally be satisfied by an
ments will vary with location and ambient
inlet shape that is circularly symmetrical
temperature; however, for most sampling lo-
about a vertical axis.
cations, these errors are expected to be 7.1.3 The sampler shall have a flow control
small. device capable of maintaining the sampler’s
6.3 Humidity. The effects of ambient humid- operating flow rate within the flow rate lim-
ity on the sample are unavoidable. The filter its specified for the sampler inlet over nor-
equilibration procedure in section 9.0 is de- mal variations in line voltage and filter pres-
signed to minimize the effects of moisture on sure drop.
the filter medium. 7.1.4 The sampler shall provide a means to
6.4 Filter Handling. Careful handling of fil- measure the total flow rate during the sam-
ters between presampling and postsampling pling period. A continuous flow recorder is
weighings is necessary to avoid errors due to recommended but not required. The flow
damaged filters or loss of collected particles measurement device shall be accurate to ±2
from the filters. Use of a filter cartridge or percent.
cassette may reduce the magnitude of these 7.1.5 A timing/control device capable of
errors. Filters must also meet the integrity starting and stopping the sampler shall be
specification in section 7.2.3. used to obtain a sample collection period of
6.5 Flow Rate Variation. Variations in the 24 ±1 hr (1,440 ±60 min). An elapsed time
sampler’s operating flow rate may alter the meter, accurate to within ±15 minutes, shall
particle size discrimination characteristics be used to measure sampling time. This
of the sampler inlet. The magnitude of this meter is optional for samplers with contin-
error will depend on the sensitivity of the uous flow recorders if the sampling time
inlet to variations in flow rate and on the measurement obtained by means of the re-
particle distribution in the atmosphere dur- corder meets the ±15 minute accuracy speci-
ing the sampling period. The use of a flow fication.
control device (section 7.1.3) is required to 7.1.6 The sampler shall have an associated
minimize this error. operation or instruction manual as required
6.6 Air Volume Determination. Errors in the by part 53 of this chapter which includes de-
air volume determination may result from tailed instructions on the calibration, oper-
errors in the flow rate and/or sampling time ation, and maintenance of the sampler.
measurements. The flow control device 7.2 Filters.

serves to minimize errors in the flow rate de- 7.2.1 Filter Medium. No commercially avail-
termination, and an elapsed time meter (sec- able filter medium is ideal in all respects for
tion 7.1.5) is required to minimize the error all samplers. The user’s goals in sampling de-
in the sampling time measurement. termine the relative importance of various
filter characteristics (e.g., cost, ease of han-
7.0 Apparatus.
dling, physical and chemical characteristics,
Fecha: 14/11/2013
7.1 PM10 Sampler. etc.) and, consequently, determine the

7.1.1 The sampler shall be designed to: choice among acceptable filters. Further-
a. Draw the air sample into the sampler more, certain types of filters may not be
inlet and through the particle collection fil- suitable for use with some samplers, particu-
ter at a uniform face velocity. larly under heavy loading conditions (high
b. Hold and seal the filter in a horizontal mass concentrations), because of high or
position so that sample air is drawn down- rapid increase in the filter flow resistance
ward through the filter. that would exceed the capability of the sam-
c. Allow the filter to be installed and re- pler’s flow control device. However, samplers
moved conveniently. equipped with automatic filter-changing
67
Pt. 50, App. J 40 CFR Ch. I (7–1–03 Edition)
mechanisms may allow use of these types of 8.1.2 Particle size discrimination by iner-
filters. The specifications given below are tial separation requires that specific air ve-
minimum requirements to ensure accept- locities be maintained in the sampler’s air
ability of the filter medium for measurement inlet system. Therefore, the flow rate
of PM10 mass concentrations. Other filter through the sampler’s inlet must be main-
evaluation criteria should be considered to tained throughout the sampling period with-
meet individual sampling and analysis objec- in the design flow rate range specified by the
tives. manufacturer. Design flow rates are speci-
7.2.2 Collection Efficiency. ´99 percent, as fied as actual volumetric flow rates, meas-
measured by the DOP test (ASTM–2986) with ured at existing conditions of temperature
0.3 µm particles at the sampler’s operating and pressure (Qa). In contrast, mass con-
face velocity. centrations of PM10 are computed using flow
7.2.3 Integrity. ±5 µg/m3 (assuming sampler’s rates corrected to EPA reference conditions
nominal 24-hour air sample volume). Integ- of temperature and pressure (Qstd).
rity is measured as the PM10 concentration 8.2 Flow Rate Calibration Procedure.
equivalent corresponding to the average dif- 8.2.1 PM10 samplers employ various types
ference between the initial and the final of flow control and flow measurement de-
weights of a random sample of test filters vices. The specific procedure used for flow
that are weighed and handled under actual rate calibration or verification will vary de-
or simulated sampling conditions, but have pending on the type of flow controller and
no air sample passed through them (i.e., fil- flow indicator employed. Calibration in
ter blanks). As a minimum, the test proce- terms of actual volumetric flow rates (Qa) is
dure must include initial equilibration and
generally recommended, but other measures
weighing, installation on an inoperative
of flow rate (e.g., Qstd) may be used provided
sampler, removal from the sampler, and final
the requirements of section 8.1 are met. The
equilibration and weighing.
general procedure given here is based on ac-
7.2.4 Alkalinity. <25 microequivalents/gram
tual volumetric flow units (Qa) and serves to
of filter, as measured by the procedure given
illustrate the steps involved in the calibra-
in Reference 13 following at least two
tion of a PM10 sampler. Consult the sampler
months storage in a clean environment (free
manufacturer’s instruction manual and Ref-
from contamination by acidic gases) at room
erence 2 for specific guidance on calibration.
temperature and humidity.
Reference 14 provides additional information
7.3 Flow Rate Transfer Standard. The flow
on the use of the commonly used measures of
rate transfer standard must be suitable for
flow rate and their interrelationships.
the sampler’s operating flow rate and must
be calibrated against a primary flow or vol- 8.2.2 Calibrate the flow rate transfer stand-
ume standard that is traceable to the Na- ard against a primary flow or volume stand-
tional Bureau of Standards (NBS). The flow ard traceable to NBS. Establish a calibration
rate transfer standard must be capable of relationship (e.g., an equation or family of
measuring the sampler’s operating flow rate curves) such that traceability to the primary
with an accuracy of ±2 percent. standard is accurate to within 2 percent over
7.4 Filter Conditioning Environment. the expected range of ambient conditions
7.4.1 Temperature range: 15 to 30 C. (i.e., temperatures and pressures) under
7.4.2 Temperature control: ±3 C. which the transfer standard will be used. Re-
7.4.3 Humidity range: 20% to 45% RH. calibrate the transfer standard periodically.
7.4.4 Humidity control: ±5% RH. 8.2.3 Following the sampler manufacturer’s
7.5 Analytical Balance. The analytical bal- instruction manual, remove the sampler
ance must be suitable for weighing the type inlet and connect the flow rate transfer
and size of filters required by the sampler. standard to the sampler such that the trans-
The range and sensitivity required will de- fer standard accurately measures the sam-
pend on the filter tare weights and mass pler’s flow rate. Make sure there are no leaks
loadings. Typically, an analytical balance between the transfer standard and the sam-
with a sensitivity of 0.1 mg is required for pler.
high volume samplers (flow rates >0.5 m3/ 8.2.4 Choose a minimum of three flow rates
min). Lower volume samplers (flow rates <0.5 (actual m3/min), spaced over the acceptable
m3/min) will require a more sensitive bal- flow rate range specified for the inlet (see
ance. 7.1.2) that can be obtained by suitable adjust-
Fecha: 14/11/2013
8.0 Calibration. ment of the sampler flow rate. In accordance

8.1 General Requirements. with the sampler manufacturer’s instruction
8.1.1 Calibration of the sampler’s flow manual, obtain or verify the calibration re-
measurement device is required to establish lationship between the flow rate (actual m3/
traceability of subsequent flow measure- min) as indicated by the transfer standard
ments to a primary standard. A flow rate and the sampler’s flow indicator response.
transfer standard calibrated against a pri- Record the ambient temperature and baro-
mary flow or volume standard shall be used metric pressure. Temperature and pressure
to calibrate or verify the accuracy of the corrections to subsequent flow indicator
sampler’s flow measurement device. readings may be required for certain types of
68
flow measurement devices. When such cor- 9.8 Set the timer to start and stop the sam-
rections are necessary, correction on an indi- pler at appropriate times. Set the elapsed
vidual or daily basis is preferable. However, time meter to zero or record the initial
seasonal average temperature and average meter reading.
barometric pressure for the sampling site 9.9 Record the sample information (site lo-
may be incorporated into the sampler cali- cation or identification number, sample
bration to avoid daily corrections. Consult date, filter identification number, and sam-
the sampler manufacturer’s instruction man- pler model and serial number).
ual and Reference 2 for additional guidance. 9.10 Sample for 24±1 hours.
8.2.5 Following calibration, verify that the 9.11 Determine and record the average flow
sampler is operating at its design flow rate rate (Q̄a) in actual m3/min for the sampling
(actual m3/min) with a clean filter in place. period in accordance with the instructions
8.2.6 Replace the sampler inlet. provided in the sampler manufacturer’s in-
9.0 Procedure. struction manual. Record the elapsed time
9.1 The sampler shall be operated in ac- meter final reading and, if needed, the aver-
cordance with the specific guidance provided age ambient temperature and barometric
in the sampler manufacturer’s instruction pressure for the sampling period (see note
manual and in Reference 2. The general pro- following step 9.6).
cedure given here assumes that the sampler’s 9.12 Carefully remove the filter from the
flow rate calibration is based on flow rates sampler, following the sampler manufactur-
at ambient conditions (Qa) and serves to il- er’s instruction manual. Touch only the
lustrate the steps involved in the operation outer edges of the filter.
of a PM10 sampler. 9.13 Place the filter in a protective holder
9.2 Inspect each filter for pinholes, par- or container (e.g., petri dish, glassine enve-
ticles, and other imperfections. Establish a lope, or manila folder).
filter information record and assign an iden- 9.14 Record any factors such as meteoro-
tification number to each filter. logical conditions, construction activity,
9.3 Equilibrate each filter in the condi- fires or dust storms, etc., that might be per-
tioning environment (see 7.4) for at least 24 tinent to the measurement on the filter in-
hours. formation record.
9.4 Following equilibration, weigh each fil- 9.15 Transport the exposed sample filter to
ter and record the presampling weight with the filter conditioning environment as soon
the filter identification number. as possible for equilibration and subsequent
9.5 Install a preweighed filter in the sam- weighing.
pler following the instructions provided in 9.16 Equilibrate the exposed filter in the
the sampler manufacturer’s instruction man- conditioning environment for at least 24
ual. hours under the same temperature and hu-
9.6 Turn on the sampler and allow it to es- midity conditions used for presampling filter
tablish run-temperature conditions. Record equilibration (see 9.3).
the flow indicator reading and, if needed, the 9.17 Immediately after equilibration, re-
ambient temperature and barometric pres- weigh the filter and record the postsampling
sure. Determine the sampler flow rate (ac- weight with the filter identification number.
tual m3/min) in accordance with the instruc- 10.0 Sampler Maintenance.
tions provided in the sampler manufacturer’s 10.1 The PM10 sampler shall be maintained
instruction manual. NOTE.—No onsite tem- in strict accordance with the maintenance
perature or pressure measurements are nec- procedures specified in the sampler manufac-
essary if the sampler’s flow indicator does turer’s instruction manual.
not require temperature or pressure correc- 11.0 Calculations.
tions or if seasonal average temperature and 11.1 Calculate the average flow rate over
average barometric pressure for the sam- the sampling period corrected to EPA ref-
pling site are incorporated into the sampler erence conditions as Q̄std. When the sampler’s
calibration (see step 8.2.4). If individual or flow indicator is calibrated in actual volu-
daily temperature and pressure corrections metric units (Qa), Q̄std is calculated as:
are required, ambient temperature and baro- Q̄std=Q̄a×(Pav/Tav)(Tstd/Pstd)
metric pressure can be obtained by on-site
measurements or from a nearby weather sta- where
Fecha: 14/11/2013
tion. Barometric pressure readings obtained Q̄std = average flow rate at EPA reference
from airports must be station pressure, not conditions, std m3/min;
corrected to sea level, and may need to be Q̄a = average flow rate at ambient conditions,
corrected for differences in elevation be- m3/min;
tween the sampling site and the airport. Pav = average barometric pressure during the
9.7 If the flow rate is outside the accept- sampling period or average barometric
able range specified by the manufacturer, pressure for the sampling site, kPa (or mm
check for leaks, and if necessary, adjust the Hg);
flow rate to the specified setpoint. Stop the Tav = average ambient temperature during
sampler. the sampling period or seasonal average
69
Pt. 50, App. K 40 CFR Ch. I (7–1–03 Edition)
ambient temperature for the sampling site, ies of Major Sampling Interferences. Atmos.
K; Environ., 13:543, 1979.
Tstd = standard temperature, defined as 298 K; 10. Appel, B.R. Letter to Larry Purdue,
Pstd = standard pressure, defined as 101.3 kPa U.S. EPA, Environmental Monitoring and
(or 760 mm Hg). Support Laboratory. March 18, 1982, Docket
No. A–82–37, II–I–1.
11.2 Calculate the total volume of air sam- 11. Pierson, W.R., W.W. Brachaczek, T.J.
pled as: Korniski, T.J. Truex, and J.W. Butler. Arti-
Vstd = Q̄std×t fact Formation of Sulfate, Nitrate, and Hy-
drogen Ion on Backup Filters: Allegheny
where Mountain Experiment. J. Air Pollut. Control
Vstd = total air sampled in standard volume Assoc., 30:30, 1980.
units, std m3; 12. Dunwoody, C.L. Rapid Nitrate Loss
t = sampling time, min. From PM10 Filters. J. Air Pollut. Control
Assoc., 36:817, 1986.
11.3 Calculate the PM10 concentration as:
13. Harrell, R.M. Measuring the Alkalinity
PM10 = (Wf¥Wi)×106/Vstd of Hi-Vol Air Filters. EMSL/RTP–SOP–QAD–
534, October 1985. Available from the U.S. En-
where
vironmental Protection Agency, EMSL/QAD,
PM10 = mass concentration of PM10, µg/std Research Triangle Park, NC 27711.
m3; 14. Smith, F., P.S. Wohlschlegel, R.S.C.
Wf, Wi = final and initial weights of filter col- Rogers, and D.J. Mulligan. Investigation of
lecting PM1O particles, g; Flow Rate Calibration Procedures Associ-
106 = conversion of g to µg. ated With the High Volume Method for De-
NOTE: If more than one size fraction in the termination of Suspended Particulates.
PM10 size range is collected by the sampler, EPA–600/4–78–047, U.S. Environmental Pro-
the sum of the net weight gain by each col- tection Agency, Research Triangle Park, NC
lection filter [S(Wf¥Wi)] is used to calculate 27711, 1978.
the PM10 mass concentration. [52 FR 24664, July 1, 1987; 52 FR 29467, Aug. 7,
12.0 References. 1987]
lution Measurement Systems, Volume I, APPENDIX K TO PART 50—INTERPRETA-
Principles. EPA–600/9–76–005, March 1976. TION OF THE NATIONAL AMBIENT AIR
Available from CERI, ORD Publications, QUALITY STANDARDS FOR PARTICU-
U.S. Environmental Protection Agency, 26 LATE MATTER
West St. Clair Street, Cincinnati, OH 45268.
2. Quality Assurance Handbook for Air Pol- 1.0 General.
lution Measurement Systems, Volume II, (a) This appendix explains the computa-
Ambient Air Specific Methods. EPA–600/4–77– tions necessary for analyzing particulate
027a, May 1977. Available from CERI, ORD matter data to determine attainment of the
Publications, U.S. Environmental Protection 24-hour and annual standards specified in 40
Agency, 26 West St. Clair Street, Cincinnati, CFR 50.6. For the primary and secondary
OH 45268. standards, particulate matter is measured in
3. Clement, R.E., and F.W. Karasek. Sam- the ambient air as PM10 (particles with an
ple Composition Changes in Sampling and aerodynamic diameter less than or equal to
Analysis of Organic Compounds in Aerosols. a nominal 10 micrometers) by a reference
Int. J. Environ. Analyt. Chem., 7:109, 1979. method based on appendix J of this part and
4. Lee, R.E., Jr., and J. Wagman. A Sam- designated in accordance with part 53 of this
pling Anomaly in the Determination of At- chapter, or by an equivalent method des-
mospheric Sulfate Concentration. Amer. Ind. ignated in accordance with part 53 of this
Hyg. Assoc. J., 27:266, 1966. chapter. The required frequency of measure-
5. Appel, B.R., S.M. Wall, Y. Tokiwa, and ments is specified in part 58 of this chapter.
M. Haik. Interference Effects in Sampling (b) The terms used in this appendix are de-
Particulate Nitrate in Ambient Air. Atmos. fined as follows:
Environ., 13:319, 1979. Average refers to an arithmetic mean. All
6. Coutant, R.W. Effect of Environmental particulate matter standards are expressed
Fecha: 14/11/2013
Variables on Collection of Atmospheric Sul- in terms of expected annual values: Expected

fate. Environ. Sci. Technol., 11:873, 1977. number of exceedances per year for the 24-
7. Spicer, C.W., and P. Schumacher. Inter- hour standards and expected annual arith-
ference in Sampling Atmospheric Particu- metic mean for the annual standards.
late Nitrate. Atmos. Environ., 11:873, 1977. Daily value for PM10 refers to the 24-hour
8. Appel, B.R., Y. Tokiwa, and M. Haik. average concentration of PM10 calculated or
Sampling of Nitrates in Ambient Air. Atmos. measured from midnight to midnight (local
Environ., 15:283, 1981. time).
9. Spicer, C.W., and P.M. Schumacher. Par- Exceedance means a daily value that is
ticulate Nitrate: Laboratory and Field Stud- above the level of the 24-hour standard after
70
Environmental Protection Agency Pt. 50, App. K
rounding to the nearest 10 µg/m3 (i.e., values µg/m3 before comparison with the annual
ending in 5 or greater are to be rounded up). standards (fractional values equal to or
Expected annual value is the number ap- greater than 0.5 are to be rounded up).
proached when the annual values from an in- 2.3 Data Requirements.
creasing number of years are averaged, in (a) 40 CFR 58.13 specifies the required min-
the absence of long-term trends in emissions imum frequency of sampling for PM10. For
or meteorological conditions. the purposes of making comparisons with
Year refers to a calendar year. the particulate matter standards, all data
(c) Although the discussion in this appen- produced by National Air Monitoring Sta-
dix focuses on monitored data, the same tions (NAMS), State and Local Air Moni-
principles apply to modeling data, subject to toring Stations (SLAMS) and other sites
EPA modeling guidelines. submitted to EPA in accordance with the
2.0 Attainment Determinations. part 58 requirements must be used, and a
2.1 24-Hour Primary and Secondary Stand- minimum of 75 percent of the scheduled PM10
ards. samples per quarter are required.
(a) Under 40 CFR 50.6(a) the 24-hour pri-
(b) To demonstrate attainment of either
mary and secondary standards are attained
the annual or 24-hour standards at a moni-
when the expected number of exceedances
toring site, the monitor must provide suffi-
per year at each monitoring site is less than
cient data to perform the required calcula-
or equal to one. In the simplest case, the
tions of sections 3.0 and 4.0 of this appendix.
number of expected exceedances at a site is
The amount of data required varies with the
determined by recording the number of
exceedances in each calendar year and then sampling frequency, data capture rate and
averaging them over the past 3 calendar the number of years of record. In all cases, 3
years. Situations in which 3 years of data are years of representative monitoring data that
not available and possible adjustments for meet the 75 percent criterion of the previous
unusual events or trends are discussed in paragraph should be utilized, if available,
sections 2.3 and 2.4 of this appendix. Further, and would suffice. More than 3 years may be
when data for a year are incomplete, it is considered, if all additional representative
necessary to compute an estimated number years of data meeting the 75 percent cri-
of exceedances for that year by adjusting the terion are utilized. Data not meeting these
observed number of exceedances. This proce- criteria may also suffice to show attainment;
dure, performed by calendar quarter, is de- however, such exceptions will have to be ap-
scribed in section 3.0 of this appendix. The proved by the appropriate Regional Adminis-
expected number of exceedances is then esti- trator in accordance with EPA guidance.
mated by averaging the individual annual es- (c) There are less stringent data require-
timates for the past 3 years. ments for showing that a monitor has failed
(b) The comparison with the allowable ex- an attainment test and thus has recorded a
pected exceedance rate of one per year is violation of the particulate matter stand-
made in terms of a number rounded to the ards. Although it is generally necessary to
nearest tenth (fractional values equal to or meet the minimum 75 percent data capture
greater than 0.05 are to be rounded up; e.g., requirement per quarter to use the computa-
an exceedance rate of 1.05 would be rounded tional equations described in sections 3.0 and
to 1.1, which is the lowest rate for nonattain- 4.0 of this appendix, this criterion does not
ment). apply when less data is sufficient to unam-
2.2 Annual Primary and Secondary Stand- biguously establish nonattainment. The fol-
ards. Under 40 CFR 50.6(b), the annual pri- lowing examples illustrate how nonattain-
mary and secondary standards are attained ment can be demonstrated when a site fails
when the expected annual arithmetic mean to meet the completeness criteria. Non-
PM10 concentration is less than or equal to attainment of the 24-hour primary standards
the level of the standard. In the simplest can be established by the observed annual
case, the expected annual arithmetic mean is number of exceedances (e.g., four observed
determined by averaging the annual arith- exceedances in a single year), or by the esti-
metic mean PM10 concentrations for the past mated number of exceedances derived from
3 calendar years. Because of the potential for the observed number of exceedances and the
incomplete data and the possible seasonality required number of scheduled samples (e.g.,
in PM10 concentrations, the annual mean two observed exceedances with every other
Fecha: 14/11/2013
shall be calculated by averaging the four day sampling). Nonattainment of the annual
quarterly means of PM10 concentrations standards can be demonstrated on the basis
within the calendar year. The equations for of quarterly mean concentrations developed
calculating the annual arithmetic mean are from observed data combined with one-half
given in section 4.0 of this appendix. Situa- the minimum detectable concentration sub-
tions in which 3 years of data are not avail- stituted for missing values. In both cases, ex-
able and possible adjustments for unusual pected annual values must exceed the levels
events or trends are discussed in sections 2.3 allowed by the standards.
and 2.4 of this appendix. The expected annual 2.4 Adjustment for Exceptional Events and
arithmetic mean is rounded to the nearest 1 Trends.
71
Pt. 50, App. K 40 CFR Ch. I (7–1–03 Edition)
(a) An exceptional event is an uncontrol- Equation 1
lable event caused by natural sources of par-
[( ) ( )]
ticulate matter or an event that is not ex-
pected to recur at a given location. Inclusion eq = vq + vq nq × Nq − nq = vq × Nq nq
of such a value in the computation of
exceedances or averages could result in inap- where:
propriate estimates of their respective ex- eq = the estimated number of exceedances for
pected annual values. To reduce the effect of calendar quarter q;
unusual events, more than 3 years of rep- vq = the observed number of exceedances for
resentative data may be used. Alternatively, calendar quarter q;
other techniques, such as the use of statis- Nq = the number of days in calendar quarter
tical models or the use of historical data q;
could be considered so that the event may be nq = the number of days in calendar quarter
discounted or weighted according to the q with PM10 data; and
likelihood that it will recur. The use of such q = the index for calendar quarter, q=1, 2, 3
techniques is subject to the approval of the or 4.
appropriate Regional Administrator in ac- (b) The estimated number of exceedances
cordance with EPA guidance. for a calendar quarter must be rounded to
(b) In cases where long–term trends in the nearest hundredth (fractional values
emissions and air quality are evident, math- equal to or greater than 0.005 must be round-
ematical techniques should be applied to ac- ed up).
count for the trends to ensure that the ex- (c) The estimated number of exceedances
pected annual values are not inappropriately for the year, e, is the sum of the estimates
biased by unrepresentative data. In the sim- for each calendar quarter.
plest case, if 3 years of data are available
under stable emission conditions, this data Equation 2
should be used. In the event of a trend or
shift in emission patterns, either the most 4
recent representative year(s) could be used e = ∑ eq
or statistical techniques or models could be
q =1
used in conjunction with previous years of
data to adjust for trends. The use of less (d) The estimated number of exceedances
than 3 years of data, and any adjustments for a single year must be rounded to one dec-
are subject to the approval of the appro- imal place (fractional values equal to or
priate Regional Administrator in accordance greater than 0.05 are to be rounded up). The
with EPA guidance. expected number of exceedances is then esti-
3.0 Computational Equations for the 24-hour mated by averaging the individual annual es-
Standards. timates for the most recent 3 or more rep-
3.1 Estimating Exceedances for a Year. resentative years of data. The expected num-
ber of exceedances must be rounded to one
(a) If PM10 sampling is scheduled less fre-
decimal place (fractional values equal to or
quently than every day, or if some scheduled
greater than 0.05 are to be rounded up).
samples are missed, a PM10 value will not be
(e) The adjustment for incomplete data
available for each day of the year. To ac- will not be necessary for monitoring or mod-
count for the possible effect of incomplete eling data which constitutes a complete
data, an adjustment must be made to the record, i.e., 365 days per year.
data collected at each monitoring location (f) To reduce the potential for overesti-
to estimate the number of exceedances in a mating the number of expected exceedances,
calendar year. In this adjustment, the as- the correction for missing data will not be
sumption is made that the fraction of miss- required for a calendar quarter in which the
ing values that would have exceeded the first observed exceedance has occurred if:
standard level is identical to the fraction of (1) There was only one exceedance in the
measured values above this level. This com- calendar quarter;
putation is to be made for all sites that are (2) Everyday sampling is subsequently ini-
scheduled to monitor throughout the entire tiated and maintained for 4 calendar quar-
Fecha: 14/11/2013
year and meet the minimum data require- ters in accordance with 40 CFR 58.13; and
ments of section 2.3 of this appendix. Be- (3) Data capture of 75 percent is achieved
cause of possible seasonal imbalance, this during the required period of everyday sam-
adjustment shall be applied on a quarterly pling. In addition, if the first exceedance is
basis. The estimate of the expected number observed in a calendar quarter in which the
of exceedances for the quarter is equal to the monitor is already sampling every day, no
observed number of exceedances plus an in- adjustment for missing data will be made to
crement associated with the missing data. the first exceedance if a 75 percent data cap-
ER18JY97.181</GPH>
The following equation must be used for ture rate was achieved in the quarter in
these computations: which it was observed.
72
ER18JY97.180</GPH>
Environmental Protection Agency Pt. 50, App. K
Example 1 Equation 3
a. During a particular calendar quarter, 39
mq
( ) ( )
out of a possible 92 samples were recorded,
with one observed exceedance of the 24-hour eq = N q m q × ∑ v j k j
standard. Using Equation 1, the estimated
j =1
number of exceedances for the quarter is:
where:
eq=1×92/39=2.359 or 2.36.
eq = the estimated number of exceedances for
b. If the estimated exceedances for the the quarter;
other 3 calendar quarters in the year were Nq = the number of days in the quarter;
2.30, 0.0 and 0.0, then, using Equation 2, the mq = the number of strata with samples dur-
estimated number of exceedances for the ing the quarter;
year is 2.36=2.30=0.0=0.0 which equals 4.66 or vj = the number of observed exceedances in
stratum j; and
4.7. If no exceedances were observed for the 2
kj = the number of actual samples in stratum
previous years, then the expected number of
j.
exceedances is estimated by: (1/
3)×(4.7=0=0)=1.57 or 1.6. Since 1.6 exceeds the (c) Note that if only one sample value is re-
allowable number of expected exceedances, corded in each stratum, then Equation 3 re-
this monitoring site would fail the attain- duces to Equation 1.
ment test.
Example 3
Example 2 A monitoring site samples according to a
systematic sampling schedule of one sample
In this example, everyday sampling was
every 6 days, for a total of 15 scheduled sam-
initiated following the first observed exceed-
ples in a quarter out of a total of 92 possible
ance as required by 40 CFR 58.13. Accord- samples. During one 6-day period, potential
ingly, the first observed exceedance would episode levels of PM10 were suspected, so 5
not be adjusted for incomplete sampling. additional samples were taken. One of the
During the next three quarters, 1.2 regular scheduled samples was missed, so a
exceedances were estimated. In this case, the total of 19 samples in 14 sampling strata
estimated exceedances for the year would be were measured. The one 6-day sampling stra-
1.0=1.2=0.0=0.0 which equals 2.2. If, as before, tum with 6 samples recorded 2 exceedances.
no exceedances were observed for the two The remainder of the quarter with one sam-
previous years, then the estimated ple per stratum recorded zero exceedances.
exceedances for the 3–year period would then Using Equation 3, the estimated number of
be (1/3)×(2.2=0.0=0.0)=0.7, and the monitoring exceedances for the quarter is:
site would not fail the attainment test. eq=(92/14)×(2/6=0=. . .=0)=2.19.
3.2 Adjustments for Non-Scheduled Sampling
Days. 4.0 Computational Equations for Annual
(a) If a systematic sampling schedule is Standards.
4.1 Calculation of the Annual Arithmetic
used and sampling is performed on days in
Mean. (a) An annual arithmetic mean value
addition to the days specified by the system-
for PM10 is determined by averaging the
atic sampling schedule, e.g., during episodes
quarterly means for the 4 calendar quarters
of high pollution, then an adjustment must of the year. The following equation is to be
be made in the eqution for the estimation of used for calculation of the mean for a cal-
exceedances. Such an adjustment is needed endar quarter:
to eliminate the bias in the estimate of the
quarterly and annual number of exceedances Equation 4

that would occur if the chance of an exceed-
ance is different for scheduled than for non- nq
scheduled days, as would be the case with
episode sampling. ( )
x q = 1 nq × ∑ x i
(b) The required adjustment treats the sys- i =1
tematic sampling schedule as a stratified where:
Fecha: 14/11/2013
sampling plan. If the period from one sched- x̄q = the quarterly mean concentration for
uled sample until the day preceding the next quarter q, q=1, 2, 3, or 4,
scheduled sample is defined as a sampling nq = the number of samples in the quarter,
stratum, then there is one stratum for each and
scheduled sampling day. An average number xi = the ith concentration value recorded in
of observed exceedances is computed for each the quarter.
of these sampling strata. With nonscheduled (b) The quarterly mean, expressed in µg/m3,
ER18JY97.183</GPH>
sampling days, the estimated number of must be rounded to the nearest tenth (frac-
exceedances is defined as: tional values of 0.05 should be rounded up).
73
ER18JY97.182</GPH>
Pt. 50, App. L 40 CFR Ch. I (7–1–03 Edition)
(c) The annual mean is calculated by using Example 5
the following equation:
a. During one calendar quarter, 9 observa-
Equation 5 tions were recorded. These samples were dis-
tributed among 7 sampling strata, with 3 ob-
servations in one stratum. The concentra-
4
( )
tions of the 3 observations in the single stra-
x = 1 4 × ∑ xq tum were 202, 242, and 180 µg/m3. The remain-
q =1 ing 6 observed concentrations were 55, 68, 73,
92, 120, and 155 µg/m3. Applying the weighting
where: factors specified in Equation 6, the quarterly
x̄ = the annual mean; and mean is:
x̄q = the mean for calendar quarter q.
x̄q = (1/7) × [(1/3) × (202 = 242 = 180) = 155 = 68
(d) The average of quarterly means must = 73 = 92 = 120 = 155] = 110.1
be rounded to the nearest tenth (fractional
b. Although 24-hour measurements are
values of 0.05 should be rounded up).
rounded to the nearest 10 µg/m3 for deter-
(e) The use of quarterly averages to com-
minations of exceedances of the 24-hour
pute the annual average will not be nec-
standard, note that these values are rounded
essary for monitoring or modeling data
to the nearest 1 µg/m3 for the calculation of
which results in a complete record, i.e., 365
means.
days per year.
(f) The expected annual mean is estimated [62 FR 38712, July 18, 1997]
as the average of three or more annual
means. This multi-year estimate, expressed APPENDIX L TO PART 50—REFERENCE
in µg/m3, shall be rounded to the nearest in- METHOD FOR THE DETERMINATION OF
teger for comparison with the annual stand- FINE PARTICULATE MATTER AS PM2.5
ard (fractional values of 0.5 should be round- IN THE ATMOSPHERE
ed up).
1.0 Applicability.
Example 4 1.1 This method provides for the measure-
Using Equation 4, the quarterly means are ment of the mass concentration of fine par-
calculated for each calendar quarter. If the ticulate matter having an aerodynamic di-
quarterly means are 52.4, 75.3, 82.1, and 63.2 ameter less than or equal to a nominal 2.5
µg/m 3, then the annual mean is: micrometers (PM2.5) in ambient air over a 24-
hour period for purposes of determining
x̄ = (1/4)×(52.4=75.3=82.1=63.2) = 68.25 or 68.3. whether the primary and secondary national
4.2 Adjustments for Non-scheduled Sampling ambient air quality standards for fine partic-
Days. (a) An adjustment in the calculation of ulate matter specified in § 50.7 of this part
the annual mean is needed if sampling is per- are met. The measurement process is consid-
formed on days in addition to the days speci- ered to be nondestructive, and the PM2.5 sam-
fied by the systematic sampling schedule. ple obtained can be subjected to subsequent
For the same reasons given in the discussion physical or chemical analyses. Quality as-
of estimated exceedances, under section 3.2 sessment procedures are provided in part 58,
of this appendix, the quarterly averages appendix A of this chapter, and quality as-
would be calculated by using the following surance guidance are provided in references
equation: 1, 2, and 3 in section 13.0 of this appendix.
1.2 This method will be considered a ref-
Equation 6 erence method for purposes of part 58 of this
chapter only if:
(a) The associated sampler meets the re-
m k
 1 
( )
q j quirements specified in this appendix and
xq = × Σ Σ x k the applicable requirements in part 53 of this
 mq  j =1 i =1 ij j chapter, and
where: (b) The method and associated sampler
have been designated as a reference method
x̄q = the quarterly mean concentration for
in accordance with part 53 of this chapter.
quarter q, q=1, 2, 3, or 4;
1.3 PM2.5 samplers that meet nearly all
Fecha: 14/11/2013
xij = the ith concentration value recorded in

specifications set forth in this method but
stratum j;
have minor deviations and/or modifications
kj = the number of actual samples in stratum
of the reference method sampler will be des-
j; and
ignated as ‘‘Class I’’ equivalent methods for
mq = the number of strata with data in the
PM2.5 in accordance with part 53 of this chap-
quarter.
ter.
(b) If one sample value is recorded in each 2.0 Principle.
stratum, Equation 6 reduces to a simple 2.1 An electrically powered air sampler
ER18JY97.185</GPH>
arithmetic average of the observed values as draws ambient air at a constant volumetric
described by Equation 4. flow rate into a specially shaped inlet and
74
ER18JY97.184</GPH>
Environmental Protection Agency Pt. 50, App. L
through an inertial particle size separator ance, such an exceedance would be a valid
(impactor) where the suspended particulate exceedance for the sample period. When re-
matter in the PM2.5 size range is separated ported to AIRS, this data value should re-
for collection on a polytetrafluoroethylene ceive a special code to identify it as not to be
(PTFE) filter over the specified sampling pe- commingled with normal concentration
riod. The air sampler and other aspects of measurements or used for other purposes.
this reference method are specified either ex- 4.0 Accuracy.
plicitly in this appendix or generally with 4.1 Because the size and volatility of the
reference to other applicable regulations or particles making up ambient particulate
quality assurance guidance. matter vary over a wide range and the mass
2.2 Each filter is weighed (after moisture concentration of particles varies with par-
and temperature conditioning) before and ticle size, it is difficult to define the accu-
after sample collection to determine the net racy of PM2.5 measurements in an absolute
gain due to collected PM2.5. The total volume sense. The accuracy of PM2.5 measurements
of air sampled is determined by the sampler is therefore defined in a relative sense, ref-
from the measured flow rate at actual ambi- erenced to measurements provided by this
ent temperature and pressure and the sam- reference method. Accordingly, accuracy
pling time. The mass concentration of PM2.5 shall be defined as the degree of agreement
in the ambient air is computed as the total between a subject field PM2.5 sampler and a
mass of collected particles in the PM2.5 size collocated PM2.5 reference method audit
range divided by the actual volume of air sampler operating simultaneously at the
sampled, and is expressed in micrograms per monitoring site location of the subject sam-
cubic meter of air (µg/m3). pler and includes both random (precision)
3.0 PM2.5 Measurement Range. and systematic (bias) errors. The require-
3.1 Lower concentration limit. The lower de- ments for this field sampler audit procedure
tection limit of the mass concentration are set forth in part 58, appendix A of this
measurement range is estimated to be ap- chapter.
proximately 2 µg/m3, based on noted mass 4.2 Measurement system bias. Results of col-
changes in field blanks in conjunction with located measurements where the duplicate
the 24 m3 nominal total air sample volume sampler is a reference method sampler are
specified for the 24-hour sample. used to assess a portion of the measurement
3.2 Upper concentration limit. The upper system bias according to the schedule and
limit of the mass concentration range is de- procedure specified in part 58, appendix A of
termined by the filter mass loading beyond this chapter.
which the sampler can no longer maintain 4.3 Audits with reference method samplers to
the operating flow rate within specified lim- determine system accuracy and bias. According
its due to increased pressure drop across the to the schedule and procedure specified in
loaded filter. This upper limit cannot be part 58, appendix A of this chapter, a ref-
specified precisely because it is a complex erence method sampler is required to be lo-
function of the ambient particle size dis- cated at each of selected PM2.5 SLAMS sites
tribution and type, humidity, the individual as a duplicate sampler. The results from the
filter used, the capacity of the sampler flow primary sampler and the duplicate reference
rate control system, and perhaps other fac- method sampler are used to calculate accu-
tors. Nevertheless, all samplers are esti- racy of the primary sampler on a quarterly
mated to be capable of measuring 24-hour basis, bias of the primary sampler on an an-
PM2.5 mass concentrations of at least 200 µg/ nual basis, and bias of a single reporting or-
m3 while maintaining the operating flow rate ganization on an annual basis. Reference 2 in
within the specified limits. section 13.0 of this appendix provides addi-
3.3 Sample period. The required sample pe- tional information and guidance on these
riod for PM2.5 concentration measurements reference method audits.
by this method shall be 1,380 to 1500 minutes 4.4 Flow rate accuracy and bias. Part 58, ap-
(23 to 25 hours). However, when a sample pe- pendix A of this chapter requires that the
riod is less than 1,380 minutes, the measured flow rate accuracy and bias of individual
concentration (as determined by the col- PM2.5 samplers used in SLAMS monitoring
lected PM2.5 mass divided by the actual sam- networks be assessed periodically via audits
pled air volume), multiplied by the actual of each sampler’s operational flow rate. In
number of minutes in the sample period and addition, part 58, appendix A of this chapter
Fecha: 14/11/2013
divided by 1,440, may be used as if it were a requires that flow rate bias for each ref-
valid concentration measurement for the erence and equivalent method operated by
specific purpose of determining a violation of each reporting organization be assessed
the NAAQS. This value assumes that the quarterly and annually. Reference 2 in sec-
PM2.5 concentration is zero for the remaining tion 13.0 of this appendix provides additional
portion of the sample period and therefore information and guidance on flow rate accu-
represents the minimum concentration that racy audits and calculations for accuracy
could have been measured for the full 24-hour and bias.
sample period. Accordingly, if the value thus 5.0 Precision. A data quality objective of 10
calculated is high enough to be an exceed- percent coefficient of variation or better has
75
been established for the operational preci- greater. The filters shall be weighed under
sion of PM2.5 monitoring data. laboratory conditions and shall have had no
5.1 Tests to establish initial operational air sample passed through them, i.e., filter
precision for each reference method sampler blanks. Each test procedure must include
are specified as a part of the requirements initial conditioning and weighing, the test,
for designation as a reference method under and final conditioning and weighing. Condi-
§ 53.58 of this chapter. tioning and weighing shall be in accordance
5.2 Measurement System Precision. Collo- with sections 8.0 through 8.2 of this appendix
cated sampler results, where the duplicate and general guidance provided in reference 2
sampler is not a reference method sampler of section 13.0 of this appendix.
but is a sampler of the same designated 6.9.1 Test for loose, surface particle contami-
method as the primary sampler, are used to nation. After the initial weighing, install
assess measurement system precision ac- each test filter, in turn, in a filter cassette
cording to the schedule and procedure speci- (Figures L–27, L–28, and L–29 of this appen-
fied in part 58, appendix A of this chapter. dix) and drop the cassette from a height of 25
Part 58, appendix A of this chapter requires cm to a flat hard surface, such as a particle-
that these collocated sampler measurements free wood bench. Repeat two times, for a
be used to calculate quarterly and annual total of three drop tests for each test filter.
precision estimates for each primary sam- Remove the test filter from the cassette and
pler and for each designated method em- weigh the filter. The average change in
ployed by each reporting organization. Ref- weight must be less than 20 µg.
erence 2 in section 13.0 of this appendix pro- 6.9.2 Test for temperature stability. After
vides additional information and guidance weighing each filter, place the test filters in
on this requirement. a drying oven set at 40 °C ±2 °C for not less
6.0 Filter for PM2.5 Sample Collection. Any fil- than 48 hours. Remove, condition, and re-
ter manufacturer or vendor who sells or of- weigh each test filter. The average change in
fers to sell filters specifically identified for weight must be less than 20 µg.
use with this PM2.5 reference method shall 6.10 Alkalinity. Less than 25 microequiva-
certify that the required number of filters lents/gram of filter, as measured by the guid-
from each lot of filters offered for sale as ance given in reference 2 in section 13.0 of
such have been tested as specified in this sec- this appendix.
tion 6.0 and meet all of the following design 6.11 Supplemental requirements. Although
and performance specifications. not required for determination of PM2.5 mass
6.1 Size. Circular, 46.2 mm diameter ±0.25 concentration under this reference method,
mm. additional specifications for the filter must
6.2 Medium. Polytetrafluoroethylene be developed by users who intend to subject
(PTFE Teflon), with integral support ring. PM2.5 filter samples to subsequent chemical
6.3 Support ring. Polymethylpentene (PMP) analysis. These supplemental specifications
or equivalent inert material, 0.38 ±0.04 mm include background chemical contamination
thick, outer diameter 46.2 mm ±0.25 mm, and of the filter and any other filter parameters
width of 3.68 mm ( ±0.00, -0.51 mm). that may be required by the method of chem-
6.4 Pore size. 2 µm as measured by ASTM F ical analysis. All such supplemental filter
316–94. specifications must be compatible with and
6.5 Filter thickness. 30 to 50 µm. secondary to the primary filter specifica-
6.6 Maximum pressure drop (clean filter). 30 tions given in this section 6.0 of this appen-
cm H2O column @ 16.67 L/min clean air flow. dix.
6.7 Maximum moisture pickup. Not more 7.0 PM2.5 Sampler.
than 10 µg weight increase after 24-hour ex- 7.1 Configuration. The sampler shall consist
posure to air of 40 percent relative humidity, of a sample air inlet, downtube, particle size
relative to weight after 24-hour exposure to separator (impactor), filter holder assembly,
air of 35 percent relative humidity. air pump and flow rate control system, flow
6.8 Collection efficiency. Greater than 99.7 rate measurement device, ambient and filter
percent, as measured by the DOP test (ASTM temperature monitoring system, barometric
D 2986–91) with 0.3 µm particles at the sam- pressure measurement system, timer, out-
pler’s operating face velocity. door environmental enclosure, and suitable
6.9 Filter weight stability. Filter weight loss mechanical, electrical, or electronic control
shall be less than 20 µg, as measured in each capability to meet or exceed the design and
Fecha: 14/11/2013
of the following two tests specified in sec- functional performance as specified in this
tions 6.9.1 and 6.9.2 of this appendix. The fol- section 7.0 of this appendix. The performance
lowing conditions apply to both of these specifications require that the sampler:
tests: Filter weight loss shall be the average (a) Provide automatic control of sample
difference between the initial and the final volumetric flow rate and other operational
filter weights of a random sample of test fil- parameters.
ters selected from each lot prior to sale. The (b) Monitor these operational parameters
number of filters tested shall be not less as well as ambient temperature and pressure.
than 0.1 percent of the filters of each manu- (c) Provide this information to the sampler
facturing lot, or 10 filters, whichever is operator at the end of each sample period in
76
digital form, as specified in table L–1 of sec- appendix), the dimension of the impaction
tion 7.4.19 of this appendix. jet must be verified by the manufacturer
7.2 Nature of specifications. The PM2.5 sam- using Class ZZ go/no-go plug gauges that are
pler is specified by a combination of design traceable to NIST.
and performance requirements. The sample 7.3.4.2 Impactor filter specifications:
inlet, downtube, particle size discriminator, (a) Size. Circular, 35 to 37 mm diameter.
filter cassette, and the internal configura- (b) Medium. Borosilicate glass fiber, with-
tion of the filter holder assembly are speci- out binder.
fied explicitly by design figures and associ- (c) Pore size. 1 to 1.5 micrometer, as meas-
ated mechanical dimensions, tolerances, ma- ured by ASTM F 316–80.
terials, surface finishes, assembly instruc- (d) Thickness. 300 to 500 micrometers.
tions, and other necessary specifications. All 7.3.4.3 Impactor oil specifications:
other aspects of the sampler are specified by (a) Composition. Tetramethyltetraphenyl-
required operational function and perform- trisiloxane, single-compound diffusion oil.
ance, and the design of these other aspects (b) Vapor pressure. Maximum 2×10¥8 mm
(including the design of the lower portion of Hg at 25 °C.
the filter holder assembly) is optional, sub- (c) Viscosity. 36 to 40 centistokes at 25 °C.
ject to acceptable operational performance. (d) Density. 1.06 to 1.07 g/cm3 at 25 °C.
Test procedures to demonstrate compliance (e) Quantity. 1 mL ±0.1 mL.
with both the design and performance re- 7.3.5 Filter holder assembly. The sampler
quirements are set forth in subpart E of part shall have a sample filter holder assembly to
53 of this chapter. adapt and seal to the down tube and to hold
7.3 Design specifications. Except as indicated and seal the specified filter, under section 6.0
in this section 7.3 of this appendix, these of this appendix, in the sample air stream in
components must be manufactured or repro- a horizontal position below the downtube
duced exactly as specified, in an ISO 9001- such that the sample air passes downward
registered facility, with registration ini- through the filter at a uniform face velocity.
tially approved and subsequently maintained The upper portion of this assembly shall be
during the period of manufacture. See fabricated as indicated in Figures L–25 and
§ 53.1(t) of this chapter for the definition of L–26 of this appendix and shall accept and
an ISO-registered facility. Minor modifica- seal with the filter cassette, which shall be
tions or variances to one or more compo- fabricated as indicated in Figures L–27
nents that clearly would not affect the aero- through L–29 of this appendix.
dynamic performance of the inlet, downtube, (a) The lower portion of the filter holder
impactor, or filter cassette will be consid- assembly shall be of a design and construc-
ered for specific approval. Any such proposed tion that:
modifications shall be described and sub- (1) Mates with the upper portion of the as-
mitted to the EPA for specific individual ac- sembly to complete the filter holder assem-
ceptability either as part of a reference or bly,
equivalent method application under part 53 (2) Completes both the external air seal
of this chapter or in writing in advance of and the internal filter cassette seal such
such an intended application under part 53 of that all seals are reliable over repeated filter
this chapter. changings, and
7.3.1 Sample inlet assembly. The sample inlet (3) Facilitates repeated changing of the fil-
assembly, consisting of the inlet, downtube, ter cassette by the sampler operator.
and impactor shall be configured and assem- (b) Leak–test performance requirements
bled as indicated in Figure L–1 of this appen- for the filter holder assembly are included in
dix and shall meet all associated require- section 7.4.6 of this appendix.
ments. A portion of this assembly shall also (c) If additional or multiple filters are
be subject to the maximum overall sampler stored in the sampler as part of an auto-
leak rate specification under section 7.4.6 of matic sequential sample capability, all such
this appendix. filters, unless they are currently and di-
7.3.2 Inlet. The sample inlet shall be fab- rectly installed in a sampling channel or
ricated as indicated in Figures L–2 through sampling configuration (either active or in-
L–18 of this appendix and shall meet all asso- active), shall be covered or (preferably)
ciated requirements. sealed in such a way as to:
7.3.3 Downtube. The downtube shall be fab- (1) Preclude significant exposure of the fil-
Fecha: 14/11/2013
ricated as indicated in Figure L–19 of this ap- ter to possible contamination or accumula-
pendix and shall meet all associated require- tion of dust, insects, or other material that
ments. may be present in the ambient air, sampler,
7.3.4 Impactor. or sampler ventilation air during storage pe-
7.3.4.1 The impactor (particle size sepa- riods either before or after sampling; and
rator) shall be fabricated as indicated in Fig- (2) To minimize loss of volatile or semi-
ures L–20 through L–24 of this appendix and volatile PM sample components during stor-
shall meet all associated requirements. Fol- age of the filter following the sample period.
lowing the manufacture and finishing of each 7.3.6 Flow rate measurement adapter. A flow
upper impactor housing (Figure L–21 of this rate measurement adapter as specified in
77
Figure L–30 of this appendix shall be fur- 7.4.3.1 The volumetric flow rate, measured
nished with each sampler. or averaged over intervals of not more than
7.3.7 Surface finish. All internal surfaces ex- 5 minutes over a 24-hour period, shall not
posed to sample air prior to the filter shall vary more than ±5 percent from the specified
be treated electrolytically in a sulfuric acid 16.67 L/min flow rate over the entire sample
bath to produce a clear, uniform anodized period.
surface finish of not less than 1000 mg/ft2 7.4.3.2 The coefficient of variation (sample
(1.08 mg/cm2) in accordance with military standard deviation divided by the mean) of
standard specification (mil. spec.) 8625F, the flow rate, measured over a 24-hour pe-
Type II, Class 1 in reference 4 of section 13.0 riod, shall not be greater than 2 percent.
of this appendix. This anodic surface coating 7.4.3.3 The amplitude of short-term flow
shall not be dyed or pigmented. Following rate pulsations, such as may originate from
anodization, the surfaces shall be sealed by some types of vacuum pumps, shall be at-
immersion in boiling deionized water for not tenuated such that they do not cause signifi-
less than 15 minutes. Section 53.51(d)(2) of cant flow measurement error or affect the
this chapter should also be consulted. collection of particles on the particle collec-
7.3.8 Sampling height. The sampler shall be tion filter.
equipped with legs, a stand, or other means 7.4.4 Flow rate cut off. The sampler’s sample
to maintain the sampler in a stable, upright air flow rate control system shall terminate
position and such that the center of the sam- sample collection and stop all sample flow
ple air entrance to the inlet, during sample for the remainder of the sample period in the
collection, is maintained in a horizontal event that the sample flow rate deviates by
plane and is 2.0 ±0.2 meters above the floor or more than 10 percent from the sampler de-
other horizontal supporting surface. Suitable sign flow rate specified in section 7.4.1 of this
bolt holes, brackets, tie-downs, or other appendix for more than 60 seconds. However,
means should be provided to facilitate me- this sampler cut-off provision shall not apply
chanically securing the sample to the sup- during periods when the sampler is inoper-
porting surface to prevent toppling of the ative due to a temporary power interruption,
sampler due to wind. and the elapsed time of the inoperative pe-
7.4 Performance specifications. riod shall not be included in the total sample
7.4.1 Sample flow rate. Proper operation of time measured and reported by the sampler,
the impactor requires that specific air ve- under section 7.4.13 of this appendix.
locities be maintained through the device. 7.4.5 Flow rate measurement.
Therefore, the design sample air flow rate 7.4.5.1 The sampler shall provide a means
through the inlet shall be 16.67 L/min (1.000 to measure and indicate the instantaneous
m3/hour) measured as actual volumetric flow sample air flow rate, which shall be meas-
rate at the temperature and pressure of the ured as volumetric flow rate at the tempera-
sample air entering the inlet. ture and pressure of the sample air entering
7.4.2 Sample air flow rate control system. The the inlet, with an accuracy of ±2 percent.
sampler shall have a sample air flow rate The measured flow rate shall be available for
control system which shall be capable of pro- display to the sampler operator at any time
viding a sample air volumetric flow rate in either sampling or standby modes, and the
within the specified range, under section measurement shall be updated at least every
7.4.1 of this appendix, for the specified filter, 30 seconds. The sampler shall also provide a
under section 6.0 of this appendix, at any at- simple means by which the sampler operator
mospheric conditions specified, under sec- can manually start the sample flow tempo-
tion 7.4.7 of this appendix, at a filter pressure rarily during non-sampling modes of oper-
drop equal to that of a clean filter plus up to ation, for the purpose of checking the sample
75 cm water column (55 mm Hg), and over the flow rate or the flow rate measurement sys-
specified range of supply line voltage, under tem.

section 7.4.15.1 of this appendix. This flow 7.4.5.2 During each sample period, the sam-
control system shall allow for operator ad- pler’s flow rate measurement system shall
justment of the operational flow rate of the automatically monitor the sample volu-
sampler over a range of at least ±15 percent metric flow rate, obtaining flow rate meas-
of the flow rate specified in section 7.4.1 of urements at intervals of not greater than 30
this appendix. seconds.
Fecha: 14/11/2013
7.4.3 Sample flow rate regulation. The sample (a) Using these interval flow rate measure-
flow rate shall be regulated such that for the ments, the sampler shall determine or cal-
specified filter, under section 6.0 of this ap- culate the following flow-related parameters,
pendix, at any atmospheric conditions speci- scaled in the specified engineering units:
fied, under section 7.4.7 of this appendix, at a (1) The instantaneous or interval-average
filter pressure drop equal to that of a clean flow rate, in L/min.
filter plus up to 75 cm water column (55 mm (2) The value of the average sample flow
Hg), and over the specified range of supply rate for the sample period, in L/min.
line voltage, under section 7.4.15.1 of this ap- (3) The value of the coefficient of variation
pendix, the flow rate is regulated as follows: (sample standard deviation divided by the
78
average) of the sample flow rate for the sam- again at a later time at least 10 minutes
ple period, in percent. after the first pressure measurement.
(4) The occurrence of any time interval (ii) CAUTION: Following completion of the
during the sample period in which the meas- test, the adaptor valve should be opened
ured sample flow rate exceeds a range of ±5 slowly to limit the flow rate of air into the
percent of the average flow rate for the sam- sampler. Excessive air flow rate may blow
ple period for more than 5 minutes, in which oil out of the impactor.
case a warning flag indicator shall be set. (7) Upon completion of the test, open the
(5) The value of the integrated total sam- adaptor valve, remove the adaptor and plugs,
ple volume for the sample period, in m3. and restore the sampler to the normal oper-
(b) Determination or calculation of these ating configuration.
values shall properly exclude periods when (b) The associated leak test procedure
the sampler is inoperative due to temporary shall require that for successful passage of
interruption of electrical power, under sec- this test, the difference between the two
tion 7.4.13 of this appendix, or flow rate cut pressure measurements shall not be greater
off, under section 7.4.4 of this appendix. than the number of mm of Hg specified for
(c) These parameters shall be accessible to the sampler by the manufacturer, based on
the sampler operator as specified in table L– the actual internal volume of the sampler,
1 of section 7.4.19 of this appendix. In addi- that indicates a leak of less than 80 mL/min.
tion, it is strongly encouraged that the flow
(c) Variations of the suggested technique
rate for each 5-minute interval during the
or an alternative external leak test tech-
sample period be available to the operator
nique may be required for samplers whose
following the end of the sample period.
design or configuration would make the sug-
7.4.6 Leak test capability.
gested technique impossible or impractical.
7.4.6.1 External leakage. The sampler shall
The specific proposed external leak test pro-
include an external air leak-test capability
cedure, or particularly an alternative leak
consisting of components, accessory hard-
test technique, proposed for a particular can-
ware, operator interface controls, a written
procedure in the associated Operation/In- didate sampler may be described and sub-
struction Manual, under section 7.4.18 of this mitted to the EPA for specific individual ac-
appendix, and all other necessary functional ceptability either as part of a reference or
capability to permit and facilitate the sam- equivalent method application under part 53
pler operator to conveniently carry out a of this chapter or in writing in advance of
leak test of the sampler at a field moni- such an intended application under part 53 of
toring site without additional equipment. this chapter.
The sampler components to be subjected to 7.4.6.2 Internal, filter bypass leakage. The
this leak test include all components and sampler shall include an internal, filter by-
their interconnections in which external air pass leak-check capability consisting of
leakage would or could cause an error in the components, accessory hardware, operator
sampler’s measurement of the total volume interface controls, a written procedure in the
of sample air that passes through the sample Operation/Instruction Manual, and all other
filter. necessary functional capability to permit
(a) The suggested technique for the oper- and facilitate the sampler operator to con-
ator to use for this leak test is as follows: veniently carry out a test for internal filter
(1) Remove the sampler inlet and installs bypass leakage in the sampler at a field
the flow rate measurement adapter supplied monitoring site without additional equip-
with the sampler, under section 7.3.6 of this ment. The purpose of the test is to determine
appendix. that any portion of the sample flow rate that
(2) Close the valve on the flow rate meas- leaks past the sample filter without passing
urement adapter and use the sampler air through the filter is insignificant relative to
pump to draw a partial vacuum in the sam- the design flow rate for the sampler.
pler, including (at least) the impactor, filter (a) The suggested technique for the oper-
holder assembly (filter in place), flow meas- ator to use for this leak test is as follows:
urement device, and interconnections be- (1) Carry out an external leak test as pro-
tween these devices, of at least 55 mm Hg (75 vided under section 7.4.6.1 of this appendix
cm water column), measured at a location which indicates successful passage of the
downstream of the filter holder assembly. prescribed external leak test.
Fecha: 14/11/2013
(3) Plug the flow system downstream of (2) Install a flow-impervious membrane
these components to isolate the components material in the filter cassette, either with or
under vacuum from the pump, such as with a without a filter, as appropriate, which effec-
built-in valve. tively prevents air flow through the filter.
(4) Stop the pump. (3) Use the sampler air pump to draw a par-
(5) Measure the trapped vacuum in the tial vacuum in the sampler, downstream of
sampler with a built-in pressure measuring the filter holder assembly, of at least 55 mm
device. Hg (75 cm water column).
(6) (i) Measure the vacuum in the sampler (4) Plug the flow system downstream of the
with the built-in pressure measuring device filter holder to isolate the components under
79
vacuum from the pump, such as with a built- inlet and downtube) and external to the
in valve. vertical plane of the nearest side or protu-
(5) Stop the pump. berance of the sampler case or enclosure.
(6) Measure the trapped vacuum in the The maximum temperature measurement
sampler with a built-in pressure measuring error of the ambient temperature measure-
device. ment system shall be less than 1.6 °C at 1 m/
(7) Measure the vacuum in the sampler s wind speed and 1000 W/m2 solar radiation
with the built-in pressure measuring device intensity.
again at a later time at least 10 minutes 7.4.8.2 The ambient temperature sensor
after the first pressure measurement. shall be of such a design and mounted in
(8) Remove the flow plug and membrane such a way as to facilitate its convenient
and restore the sampler to the normal oper- dismounting and immersion in a liquid for
ating configuration. calibration and comparison to the filter tem-
(b) The associated leak test procedure perature sensor, under section 7.4.11 of this
shall require that for successful passage of appendix.
this test, the difference between the two 7.4.8.3 This ambient temperature measure-
pressure measurements shall not be greater ment shall be updated at least every 30 sec-
than the number of mm of Hg specified for onds during both sampling and standby (non-
the sampler by the manufacturer, based on sampling) modes of operation. A visual indi-
the actual internal volume of the portion of cation of the current (most recent) value of
the sampler under vacuum, that indicates a the ambient temperature measurement, up-
leak of less than 80 mL/min. dated at least every 30 seconds, shall be
(c) Variations of the suggested technique available to the sampler operator during
or an alternative internal, filter bypass leak both sampling and standby (non-sampling)
test technique may be required for samplers modes of operation, as specified in table L–1
whose design or configuration would make of section 7.4.19 of this appendix.
the suggested technique impossible or im- 7.4.8.4 This ambient temperature measure-
practical. The specific proposed internal leak ment shall be used for the purpose of moni-
test procedure, or particularly an alternative toring filter temperature deviation from am-
internal leak test technique proposed for a bient temperature, as required by section
particular candidate sampler may be de- 7.4.11 of this appendix, and may be used for
scribed and submitted to the EPA for spe- purposes of effecting filter temperature con-
cific individual acceptability either as part trol, under section 7.4.10 of this appendix, or
of a reference or equivalent method applica- computation of volumetric flow rate, under
tion under part 53 of this chapter or in writ- sections 7.4.1 to 7.4.5 of this appendix, if ap-
ing in advance of such intended application propriate.
under part 53 of this chapter. 7.4.8.5 Following the end of each sample pe-
7.4.7 Range of operational conditions. The riod, the sampler shall report the maximum,
sampler is required to operate properly and minimum, and average temperature for the
meet all requirements specified in this ap- sample period, as specified in table L–1 of
pendix over the following operational ranges. section 7.4.19 of this appendix.
7.4.7.1 Ambient temperature. -30 to =45 °C 7.4.9 Ambient barometric sensor. The sampler
(Note: Although for practical reasons, the shall have capability to measure the baro-
temperature range over which samplers are metric pressure of the air surrounding the
required to be tested under part 53 of this sampler over a range of 600 to 800 mm Hg ref-
chapter is -20 to =40 °C, the sampler shall be erenced as described in reference 3 in section
designed to operate properly over this wider 13.0 of this appendix; also see part 53, subpart
temperature range.). E of this chapter. This barometric pressure
7.4.7.2 Ambient relative humidity. 0 to 100 measurement shall have a resolution of 5
percent. mm Hg and an accuracy of ±10 mm Hg and
7.4.7.3 Barometric pressure range. 600 to 800 shall be updated at least every 30 seconds. A
mm Hg. visual indication of the value of the current
7.4.8 Ambient temperature sensor. The sam- (most recent) barometric pressure measure-
pler shall have capability to measure the ment, updated at least every 30 seconds,
temperature of the ambient air surrounding shall be available to the sampler operator
the sampler over the range of -30 to =45 °C, during both sampling and standby (non-sam-
with a resolution of 0.1 °C and accuracy of pling) modes of operation, as specified in
Fecha: 14/11/2013
±2.0 °C, referenced as described in reference 3 table L–1 of section 7.4.19 of this appendix.
in section 13.0 of this appendix, with and This barometric pressure measurement may
without maximum solar insolation. be used for purposes of computation of volu-
7.4.8.1 The ambient temperature sensor metric flow rate, under sections 7.4.1 to 7.4.5
shall be mounted external to the sampler en- of this appendix, if appropriate. Following
closure and shall have a passive, naturally the end of a sample period, the sampler shall
ventilated sun shield. The sensor shall be lo- report the maximum, minimum, and mean
cated such that the entire sun shield is at barometric pressures for the sample period,
least 5 cm above the horizontal plane of the as specified in table L–1 of section 7.4.19 of
sampler case or enclosure (disregarding the this appendix.
80
7.4.10 Filter temperature control (sampling ing sample validity when a warning flag oc-
and post-sampling). The sampler shall provide curs. It is further recommended (not re-
a means to limit the temperature rise of the quired) that the sampler be capable of re-
sample filter (all sample filters for sequen- cording the maximum differential between
tial samplers), from insolation and other the measured filter temperature and the am-
sources, to no more 5 °C above the tempera- bient temperature and its time and date of
ture of the ambient air surrounding the sam- occurrence during both sampling and post-
pler, during both sampling and post-sam- sampling (non-sampling) modes of operation
pling periods of operation. The post-sam- and providing for those data to be accessible
pling period is the non-sampling period be- to the sampler operator following the end of
tween the end of the active sampling period the sample period, as suggested in table L–1
and the time of retrieval of the sample filter of section 7.4.19 of this appendix.
by the sampler operator. 7.4.12 Clock/timer system.
7.4.11 Filter temperature sensor(s).
(a) The sampler shall have a programmable
7.4.11.1 The sampler shall have the capa-
bility to monitor the temperature of the real-time clock timing/control system that:
sample filter (all sample filters for sequen- (1) Is capable of maintaining local time
tial samplers) over the range of -30 to =45 °C and date, including year, month, day-of-
during both sampling and non-sampling peri- month, hour, minute, and second to an accu-
ods. While the exact location of this tem- racy of ±1.0 minute per month.
perature sensor is not explicitly specified, (2) Provides a visual indication of the cur-
the filter temperature measurement system rent system time, including year, month,
must demonstrate agreement, within 1 °C, day-of-month, hour, and minute, updated at
with a test temperature sensor located with- least each minute, for operator verification.
in 1 cm of the center of the filter down- (3) Provides appropriate operator controls
stream of the filter during both sampling for setting the correct local time and date.
and non-sampling modes, as specified in the (4) Is capable of starting the sample collec-
filter temperature measurement test de- tion period and sample air flow at a specific,
scribed in part 53, subpart E of this chapter. operator-settable time and date, and stop-
This filter temperature measurement shall ping the sample air flow and terminating the
have a resolution of 0.1 °C and accuracy of sampler collection period 24 hours (1440 min-
±1.0 °C, referenced as described in reference 3 utes) later, or at a specific, operator-settable
in section 13.0 of this appendix. This tem- time and date.
perature sensor shall be of such a design and
(b) These start and stop times shall be
mounted in such a way as to facilitate its
readily settable by the sampler operator to
reasonably convenient dismounting and im-
within ±1.0 minute. The system shall provide
mersion in a liquid for calibration and com-
a visual indication of the current start and
parison to the ambient temperature sensor
stop time settings, readable to ±1.0 minute,
under section 7.4.8 of this appendix.
7.4.11.2 The filter temperature measure- for verification by the operator, and the
ment shall be updated at least every 30 sec- start and stop times shall also be available
onds during both sampling and standby (non- via the data output port, as specified in table
sampling) modes of operation. A visual indi- L–1 of section 7.4.19 of this appendix. Upon
cation of the current (most recent) value of execution of a programmed sample period
the filter temperature measurement, up- start, the sampler shall automatically reset
dated at least every 30 seconds, shall be all sample period information and warning
available to the sampler operator during flag indications pertaining to a previous
both sampling and standby (non-sampling) sample period. Refer also to section 7.4.15.4
modes of operation, as specified in table L–1 of this appendix regarding retention of cur-
of section 7.4.19 of this appendix. rent date and time and programmed start
7.4.11.3 For sequential samplers, the tem- and stop times during a temporary electrical
perature of each filter shall be measured in- power interruption.
dividually unless it can be shown, as speci- 7.4.13 Sample time determination. The sam-
fied in the filter temperature measurement pler shall be capable of determining the
test described in § 53.57 of this chapter, that elapsed sample collection time for each PM2.5
the temperature of each filter can be rep- sample, accurate to within ±1.0 minute,
resented by fewer temperature sensors. measured as the time between the start of
Fecha: 14/11/2013
7.4.11.4 The sampler shall also provide a the sampling period, under section 7.4.12 of
warning flag indicator following any occur- this appendix and the termination of the
rence in which the filter temperature (any sample period, under section 7.4.12 of this ap-
filter temperature for sequential samplers) pendix or section 7.4.4 of this appendix. This
exceeds the ambient temperature by more elapsed sample time shall not include peri-
than 5 °C for more than 30 consecutive min- ods when the sampler is inoperative due to a
utes during either the sampling or post-sam- temporary interruption of electrical power,
pling periods of operation, as specified in under section 7.4.15.4 of this appendix. In the
table L–1 of section 7.4.19 of this appendix, event that the elapsed sample time deter-
under section 10.12 of this appendix, regard- mined for the sample period is not within the
81
range specified for the required sample pe- such power interruptions per sample period.
riod in section 3.3 of this appendix, the sam- (More than 10 such power interruptions shall
pler shall set a warning flag indicator. The invalidate the sample, except where an ex-
date and time of the start of the sample pe- ceedance is measured, under section 3.3 of
riod, the value of the elapsed sample time for this appendix.) The sampler shall provide for
the sample period, and the flag indicator sta- these power interruption data to be available
tus shall be available to the sampler oper- to the sampler operator following the end of
ator following the end of the sample period, the sample period, as specified in table L–1 of
as specified in table L–1 of section 7.4.19 of section 7.4.19 of this appendix.
this appendix. 7.4.16 Control devices and operator interface.
7.4.14 Outdoor environmental enclosure. The The sampler shall have mechanical, elec-
sampler shall have an outdoor enclosure (or trical, or electronic controls, control de-
enclosures) suitable to protect the filter and vices, electrical or electronic circuits as nec-
other non-weatherproof components of the essary to provide the timing, flow rate meas-
sampler from precipitation, wind, dust, ex- urement and control, temperature control,
tremes of temperature and humidity; to help data storage and computation, operator
maintain temperature control of the filter interface, and other functions specified. Op-
(or filters, for sequential samplers); and to erator-accessible controls, data displays, and
provide reasonable security for sampler com- interface devices shall be designed to be sim-
ponents and settings. ple, straightforward, reliable, and easy to
7.4.15 Electrical power supply.
learn, read, and operate under field condi-
7.4.15.1 The sampler shall be operable and
tions. The sampler shall have provision for
function as specified herein when operated
operator input and storage of up to 64 char-
on an electrical power supply voltage of 105
acters of numeric (or alphanumeric) data for
to 125 volts AC (RMS) at a frequency of 59 to
purposes of site, sampler, and sample identi-
61 Hz. Optional operation as specified at ad-
fication. This information shall be available
ditional power supply voltages and/or fre-
to the sampler operator for verification and
quencies shall not be precluded by this re-
quirement. change and for output via the data output
7.4.15.2 The design and construction of the port along with other data following the end
sampler shall comply with all applicable Na- of a sample period, as specified in table L–1
tional Electrical Code and Underwriters Lab- of section 7.4.19 of this appendix. All data re-
oratories electrical safety requirements. quired to be available to the operator fol-
7.4.15.3 The design of all electrical and lowing a sample collection period or ob-
electronic controls shall be such as to pro- tained during standby mode in a post-sam-
vide reasonable resistance to interference or pling period shall be retained by the sampler
malfunction from ordinary or typical levels until reset, either manually by the operator
of stray electromagnetic fields (EMF) as or automatically by the sampler upon initi-
may be found at various monitoring sites ation of a new sample collection period.
and from typical levels of electrical tran- 7.4.17 Data output port requirement. The
sients or electronic noise as may often or oc- sampler shall have a standard RS–232C data
casionally be present on various electrical output connection through which digital
power lines. data may be exported to an external data
7.4.15.4 In the event of temporary loss of storage or transmission device. All informa-
electrical supply power to the sampler, the tion which is required to be available at the
sampler shall not be required to sample or end of each sample period shall be accessible
provide other specified functions during such through this data output connection. The in-
loss of power, except that the internal clock/ formation that shall be accessible though
timer system shall maintain its local time this output port is summarized in table L–1
and date setting within ±1 minute per week, of section 7.4.19 of this appendix. Since no
and the sampler shall retain all other time specific format for the output data is pro-
and programmable settings and all data re- vided, the sampler manufacturer or vendor
quired to be available to the sampler oper- shall make available to sampler purchasers
ator following each sample period for at appropriate computer software capable of re-
least 7 days without electrical supply power. ceiving exported sampler data and correctly
When electrical power is absent at the oper- translating the data into a standard spread-
ator-set time for starting a sample period or sheet format and optionally any other for-
Fecha: 14/11/2013
is interrupted during a sample period, the mats as may be useful to sampler users. This
sampler shall automatically start or resume requirement shall not preclude the sampler
sampling when electrical power is restored, from offering other types of output connec-
if such restoration of power occurs before the tions in addition to the required RS–232C
operator-set stop time for the sample period. port.
7.4.15.5 The sampler shall have the capa- 7.4.18 Operation/instruction manual. The
bility to record and retain a record of the sampler shall include an associated com-
year, month, day-of-month, hour, and prehensive operation or instruction manual,
minute of the start of each power interrup- as required by part 53 of this chapter, which
tion of more than 1 minute duration, up to 10 includes detailed operating instructions on
82
the setup, operation, calibration, and main- precautions. The manual shall also include a
tenance of the sampler. This manual shall clear description of all procedures pertaining
provide complete and detailed descriptions of to installation, operation, periodic and cor-
the operational and calibration procedures rective maintenance, and troubleshooting,
prescribed for field use of the sampler and all and shall include parts identification dia-
instruments utilized as part of this reference grams.
method. The manual shall include adequate 7.4.19 Data reporting requirements. The var-
warning of potential safety hazards that may ious information that the sampler is re-
result from normal use or malfunction of the quired to provide and how it is to be provided
method and a description of necessary safety is summarized in the following table L–1.
TABLE L–1—SUMMARY OF INFORMATION TO BE PROVIDED BY THE SAMPLER
Availability Format
Appendix L
Information to be section
provided End of Visual
reference Anytime1 Data output4 Digital reading5 Units
period2 display3
Flow rate, 30- 7.4.5.1 ....... ✔ .................... ✔ * XX.X ................. L/min

second max-
imum interval.
Flow rate, aver- 7.4.5.2 ....... * ✔ * ✔ XX.X ................. L/min
age for the
sample period.
Flow rate, CV, 7.4.5.2 ....... * ✔ * ✔0 XX.X ................. %
for sample pe-
riod.
Flow rate, 5-min. 7.4.5.2 ....... ✔ ✔ ✔ ✔0 On/Off ..............
average out of
spec. (FLAG6).
Sample volume, 7.4.5.2 ....... * ✔ ✔ ✔0 XX.X ................. m3
total.
Temperature, 7.4.8 .......... ✔ .................... ✔ .................... XX.X ................. °C
ambient, 30-
second inter-
val.
Temperature, 7.4.8 .......... * ✔ ✔ ✔0 XX.X ................. °C
ambient, min.,
max., average
for the sample
period.
Baro pressure, 7.4.9 .......... ✔ .................... ✔ .................... XXX .................. mm Hg
ambient, 30-
second inter-
val.
Baro pressure, 7.4.9 .......... * ✔ ✔ ✔0 XXX .................. mm Hg
ambient, min.,
max., average
for the sample
period.
Filter tempera- 7.4.11 ........ ✔ .................... ✔ .................... XX.X ................. °C
ture, 30-sec-
ond interval.
Filter tempera- 7.4.11 ........ * ✔ ✔ ✔0 On/Off ..............
ture differen-
tial, 30-second
interval, out of
spec. (FLAG6).
Filter tempera- 7.4.11 ........ * * * * X.X, YY/MM/DD °C, Yr./Mon./
ture, max- HH:mm. Day Hrs. min
imum differen-
tial from ambi-
Fecha: 14/11/2013
ent, date, time

of occurrence.
Date and time ... 7.4.12 ........ ✔ .................... ✔ .................... YY/MM/DD Yr./Mon./Day
HH:mm. Hrs. min
Sample start 7.4.12 ........ ✔ ✔ ✔ ✔ YY/MM/DD Yr./Mon./Day
and stop time HH:mm. Hrs. min
settings.
Sample period 7.4.12 ........ .................... ✔ ✔ ✔0 YYYY/MM/DD Yr./Mon./Day
start time. HH:mm. Hrs. min
Elapsed sample 7.4.13 ........ * ✔ ✔ ✔0 HH:mm ............. Hrs. min
time.
83
TABLE L–1—SUMMARY OF INFORMATION TO BE PROVIDED BY THE SAMPLER—Continued

Availability Format
Appendix L
Information to be section
provided End of Visual
reference Anytime1 Data output4 Digital reading5 Units
period2 display3
Elapsed sample 7.4.13 ........ .................... ✔ ✔ ✔0 On/Off ..............

time, out of
spec. (FLAG6).
Power interrup- 7.4.15.5 ..... * ✔ * ✔ 1HH:mm, Hrs. min
tions ™1 min., 2HH:mm, etc
start time of ....
first 10.
User-entered in- 7.4.16 ........ ✔ ✔ ✔ ✔0 As entered .......
formation,
such as sam-
pler and site
identification.
✔ Provision of this information is required.
*Provision of this information is optional. If information related to the entire sample period is optionally provided prior to the end
of the sample period, the value provided should be the value calculated for the portion of the sampler period completed up to the
time the information is provided.
0 Indicates that this information is also required to be provided to the AIRS data bank; see § § 58.26 and 58.35 of this chapter.
1 Information is required to be available to the operator at any time the sampler is operating, whether sampling or not.
2 Information relates to the entire sampler period and must be provided following the end of the sample period until reset
manually by the operator or automatically by the sampler upon the start of a new sample period.
3 Information shall be available to the operator visually.
4 Information is to be available as digital data at the sampler’s data output port specified in section 7.4.16 of this appendix fol-
lowing the end of the sample period until reset manually by the operator or automatically by the sampler upon the start of a new
sample period.
5 Digital readings, both visual and data output, shall have not less than the number of significant digits and resolution speci-
fied.
6 Flag warnings may be displayed to the operator by a single-flag indicator or each flag may be displayed individually. Only a
set (on) flag warning must be indicated; an off (unset) flag may be indicated by the absence of a flag warning. Sampler users
should refer to section 10.12 of this appendix regarding the validity of samples for which the sampler provided an associated flag
warning.
8.0 Filter Weighing. See reference 2 in sec- 8.2.5 Conditioning time. Not less than 24
tion 13.0 of this appendix, for additional, hours.
more detailed guidance. 8.3 Weighing procedure.
8.1 Analytical balance. The analytical bal- 8.3.1 New filters should be placed in the
ance used to weigh filters must be suitable conditioning environment immediately upon
for weighing the type and size of filters spec- arrival and stored there until the pre-sam-
ified, under section 6.0 of this appendix, and pling weighing. See reference 2 in section 13.0
have a readability of ±1 µg. The balance shall of this appendix for additional guidance.
be calibrated as specified by the manufac- 8.3.2 The analytical balance shall be lo-
turer at installation and recalibrated imme- cated in the same controlled environment in
diately prior to each weighing session. See which the filters are conditioned. The filters
reference 2 in section 13.0 of this appendix for shall be weighed immediately following the
additional guidance. conditioning period without intermediate or
8.2 Filter conditioning. All sample filters transient exposure to other conditions or en-
used shall be conditioned immediately before vironments.
both the pre- and post-sampling weighings as
8.3.3 Filters must be conditioned at the

specified below. See reference 2 in section same conditions (humidity within ±5 relative
13.0 of this appendix for additional guidance. humidity percent) before both the pre- and
8.2.1 Mean temperature. 20 - 23 °C. post-sampling weighings.
8.2.2 Temperature control. ±2 °C over 24 8.3.4 Both the pre- and post-sampling
hours. weighings should be carried out on the same
8.2.3 Mean humidity. Generally, 30–40 per- analytical balance, using an effective tech-
Fecha: 14/11/2013
cent relative humidity; however, where it nique to neutralize static charges on the fil-
can be shown that the mean ambient relative ter, under reference 2 in section 13.0 of this
humidity during sampling is less than 30 per- appendix. If possible, both weighings should
cent, conditioning is permissible at a mean be carried out by the same analyst.
relative humidity within ±5 relative humid- 8.3.5 The pre-sampling (tare) weighing shall
ity percent of the mean ambient relative hu- be within 30 days of the sampling period.
midity during sampling, but not less than 20 8.3.6 The post-sampling conditioning and
percent. weighing shall be completed within 240 hours
8.2.4 Humidity control. ±5 relative humidity (10 days) after the end of the sample period,
percent over 24 hours. unless the filter sample is maintained at 4 °C
84
or less during the entire time between re- downtube. The generic procedure given here
trieval from the sampler and the start of the serves to illustrate the general steps in-
conditioning, in which case the period shall volved in the calibration of a PM2.5 sampler.
not exceed 30 days. Reference 2 in section The sampler operation/instruction manual
13.0 of this appendix has additional guidance required under section 7.4.18 of this appendix
on transport of cooled filters. and the Quality Assurance Handbook in ref-
8.3.7 Filter blanks. erence 2 in section 13.0 of this appendix pro-
8.3.7.1 New field blank filters shall be vide more specific and detailed guidance for
weighed along with the pre-sampling (tare) calibration.
weighing of each lot of PM2.5 filters. These 9.2.2 The flow rate standard used for flow
blank filters shall be transported to the sam- rate calibration shall have its own certifi-
pling site, installed in the sampler, retrieved cation and be traceable to a NIST primary
from the sampler without sampling, and re- standard for volume or flow rate. A calibra-
weighed as a quality control check. tion relationship for the flow rate standard,
8.3.7.2 New laboratory blank filters shall be e.g., an equation, curve, or family of curves
weighed along with the pre-sampling (tare) relating actual flow rate (Qa) to the flow rate
weighing of each set of PM2.5 filters. These indicator reading, shall be established that is
laboratory blank filters should remain in the accurate to within 2 percent over the ex-
laboratory in protective containers during pected range of ambient temperatures and
the field sampling and should be reweighed pressures at which the flow rate standard
as a quality control check. may be used. The flow rate standard must be
8.3.8 Additional guidance for proper filter re-calibrated or re-verified at least annually.
weighing and related quality assurance ac- 9.2.3 The sampler flow rate measurement
tivities is provided in reference 2 in section device shall be calibrated or verified by re-
13.0 of this appendix. moving the sampler inlet and connecting the
9.0 Calibration. Reference 2 in section 13.0 of flow rate standard to the sampler’s downtube
this appendix contains additional guidance. in accordance with the operation/instruction
9.1 General requirements. manual, such that the flow rate standard ac-
9.1.1 Multipoint calibration and single- curately measures the sampler’s flow rate.
point verification of the sampler’s flow rate The sampler operator shall first carry out a
measurement device must be performed peri- sampler leak check and confirm that the
odically to establish and maintain sampler passes the leak test and then verify
traceability of subsequent flow measure- that no leaks exist between the flow rate
ments to a flow rate standard. standard and the sampler.
9.1.2 An authoritative flow rate standard 9.2.4 The calibration relationship between
shall be used for calibrating or verifying the the flow rate (in actual L/min) indicated by
sampler’s flow rate measurement device with the flow rate standard and by the sampler’s
an accuracy of ±2 percent. The flow rate flow rate measurement device shall be estab-
standard shall be a separate, stand-alone de- lished or verified in accordance with the
vice designed to connect to the flow rate sampler operation/instruction manual. Tem-
measurement adapter, Figure L–30 of this ap- perature and pressure corrections to the flow
pendix. This flow rate standard must have rate indicated by the flow rate standard may
its own certification and be traceable to a be required for certain types of flow rate
National Institute of Standards and Tech- standards. Calibration of the sampler’s flow
nology (NIST) primary standard for volume rate measurement device shall consist of at
or flow rate. If adjustments to the sampler’s least three separate flow rate measurements
flow rate measurement system calibration (multipoint calibration) evenly spaced with-
are to be made in conjunction with an audit in the range of -10 percent to =10 percent of
of the sampler’s flow measurement system, the sampler’s operational flow rate, section
such adjustments shall be made following 7.4.1 of this appendix. Verification of the
the audit. Reference 2 in section 13.0 of this sampler’s flow rate shall consist of one flow
appendix contains additional guidance. rate measurement at the sampler’s oper-
9.1.3 The sampler’s flow rate measurement ational flow rate. The sampler operation/in-
device shall be re-calibrated after struction manual and reference 2 in section
electromechanical maintenance or transport 13.0 of this appendix provide additional guid-
of the sampler. ance.
9.2 Flow rate calibration/verification proce- 9.2.5 If during a flow rate verification the
Fecha: 14/11/2013
dure. reading of the sampler’s flow rate indicator

9.2.1 PM2.5 samplers may employ various or measurement device differs by ± 4 percent
types of flow control and flow measurement or more from the flow rate measured by the
devices. The specific procedure used for cali- flow rate standard, a new multipoint calibra-
bration or verification of the flow rate meas- tion shall be performed and the flow rate
urement device will vary depending on the verification must then be repeated.
type of flow rate controller and flow rate 9.2.6 Following the calibration or
measurement employed. Calibration shall be verification, the flow rate standard shall be
in terms of actual ambient volumetric flow removed from the sampler and the sampler
rates (Qa), measured at the sampler’s inlet inlet shall be reinstalled. Then the sampler’s
85
normal operating flow rate (in L/min) shall specific quality assurance program estab-
be determined with a clean filter in place. If lished for the sampler by the user.
the flow rate indicated by the sampler differs 10.7 The sampler’s timer shall be set to
by ±2 percent or more from the required sam- start the sample collection at the beginning
pler flow rate, the sampler flow rate must be of the desired sample period and stop the
adjusted to the required flow rate, under sec- sample collection 24 hours later.
tion 7.4.1 of this appendix. 10.8 Information related to the sample col-
9.3 Periodic calibration or verification of lection (site location or identification num-
the calibration of the sampler’s ambient ber, sample date, filter identification num-
temperature, filter temperature, and baro- ber, and sampler model and serial number)
metric pressure measurement systems is also shall be recorded and, if appropriate, entered
required. Reference 3 of section 13.0 of this into the sampler.
appendix contains additional guidance. 10.9 The sampler shall be allowed to collect
10.0 PM2.5 Measurement Procedure. The de- the PM2.5 sample during the set 24-hour time
tailed procedure for obtaining valid PM2.5 period.
measurements with each specific sampler 10.10 Within 96 hours of the end of the sam-
designated as part of a reference method for ple collection period, the filter, while still
PM2.5 under part 53 of this chapter shall be contained in the filter cassette, shall be
provided in the sampler-specific operation or carefully removed from the sampler, fol-
instruction manual required by section 7.4.18 lowing the procedure provided in the sampler
of this appendix. Supplemental guidance is operation or instruction manual and the
provided in section 2.12 of the Quality Assur- quality assurance program, and placed in a
ance Handbook listed in reference 2 in sec- protective container. The protective con-
tion 13.0 of this appendix. The generic proce- tainer shall contain no loose material that
could be transferred to the filter. The protec-
dure given here serves to illustrate the gen-
tive container shall hold the filter cassette
eral steps involved in the PM2.5 sample col-
securely such that the cover shall not come
lection and measurement, using a PM2.5 ref-
in contact with the filter’s surfaces. Ref-
erence method sampler.
erence 2 in section 13.0 of this appendix con-
10.1 The sampler shall be set up, calibrated,
tains additional information.
and operated in accordance with the specific, 10.11 The total sample volume in actual m3
detailed guidance provided in the specific for the sampling period and the elapsed sam-
sampler’s operation or instruction manual ple time shall be obtained from the sampler
and in accordance with a specific quality as- and recorded in accordance with the instruc-
surance program developed and established tions provided in the sampler operation or
by the user, based on applicable supple- instruction manual. All sampler warning
mentary guidance provided in reference 2 in flag indications and other information re-
section 13.0 of this appendix. quired by the local quality assurance pro-
10.2 Each new sample filter shall be in- gram shall also be recorded.
spected for correct type and size and for pin- 10.12 All factors related to the validity or
holes, particles, and other imperfections. Un- representativeness of the sample, such as
acceptable filters should be discarded. A sampler tampering or malfunctions, unusual
unique identification number shall be as- meteorological conditions, construction ac-
signed to each filter, and an information tivity, fires or dust storms, etc. shall be re-
record shall be established for each filter. If corded as required by the local quality assur-
the filter identification number is not or ance program. The occurrence of a flag warn-
cannot be marked directly on the filter, al- ing during a sample period shall not nec-
ternative means, such as a number-identified essarily indicate an invalid sample but rath-
storage container, must be established to er shall indicate the need for specific review
maintain positive filter identification. of the QC data by a quality assurance officer
10.3 Each filter shall be conditioned in the to determine sample validity.

conditioning environment in accordance 10.13 After retrieval from the sampler, the
with the requirements specified in section 8.2 exposed filter containing the PM2.5 sample
of this appendix. should be transported to the filter condi-
10.4 Following conditioning, each filter tioning environment as soon as possible
shall be weighed in accordance with the re- ideally to arrive at the conditioning environ-
quirements specified in section 8.0 of this ap- ment within 24 hours for conditioning and
Fecha: 14/11/2013
pendix and the presampling weight recorded subsequent weighing. During the period be-
with the filter identification number. tween filter retrieval from the sampler and
10.5 A numbered and preweighed filter shall the start of the conditioning, the filter shall
be installed in the sampler following the in- be maintained as cool as practical and con-
structions provided in the sampler operation tinuously protected from exposure to tem-
or instruction manual. peratures over 25 °C. See section 8.3.6 of this
10.6 The sampler shall be checked and pre- appendix regarding time limits for com-
pared for sample collection in accordance pleting the post-sampling weighing. See ref-
with instructions provided in the sampler op- erence 2 in section 13.0 of this appendix for
eration or instruction manual and with the additional guidance on transporting filter
86
samplers to the conditioning and weighing Va = total air volume sampled in actual vol-
laboratory. ume units, as provided by the sampler, m3.
10.14. The exposed filter containing the
NOTE: Total sample time must be between
PM2.5 sample shall be re-conditioned in the
1,380 and 1,500 minutes (23 and 25 hrs) for a
conditioning environment in accordance
fully valid PM2.5 sample; however, see also
with the requirements specified in section 8.2
section 3.3 of this appendix.
of this appendix.
13.0 References.
10.15. The filter shall be reweighed imme-
diately after conditioning in accordance
lution Measurement Systems, Volume I,
with the requirements specified in section 8.0
Principles. EPA/600/R–94/038a, April 1994.
of this appendix, and the postsampling
Available from CERI, ORD Publications,
weight shall be recorded with the filter iden-
U.S. Environmental Protection Agency, 26
tification number.
West Martin Luther King Drive, Cincinnati,
10.16 The PM2.5 concentration shall be cal- Ohio 45268.
culated as specified in section 12.0 of this ap- 2. Copies of section 2.12 of the Quality As-
pendix. surance Handbook for Air Pollution Meas-
11.0 Sampler Maintenance. The sampler urement Systems, Volume II, Ambient Air
shall be maintained as described by the sam- Specific Methods, EPA/600/R–94/038b, are
pler’s manufacturer in the sampler-specific available from Department E (MD-77B), U.S.
operation or instruction manual required EPA, Research Triangle Park, NC 27711.
under section 7.4.18 of this appendix and in 3. Quality Assurance Handbook for Air Pol-
accordance with the specific quality assur- lution Measurement Systems, Volume IV:
ance program developed and established by Meteorological Measurements, (Revised Edi-
the user based on applicable supplementary tion) EPA/600/R–94/038d, March, 1995. Avail-
guidance provided in reference 2 in section able from CERI, ORD Publications, U.S. En-
13.0 of this appendix. vironmental Protection Agency, 26 West
12.0 Calculations Martin Luther King Drive, Cincinnati, Ohio
12.1 (a) The PM2.5 concentration is cal- 45268.
culated as: 4. Military standard specification (mil.
PM2.5 = (Wf ¥ Wi)/Va spec.) 8625F, Type II, Class 1 as listed in De-
partment of Defense Index of Specifications
where: and Standards (DODISS), available from
PM2.5 = mass concentration of PM2.5, µg/m3; DODSSP–Customer Service, Standardization
Wf, Wi = final and initial weights, respec- Documents Order Desk, 700 Robbins Avenue,
tively, of the filter used to collect the Building 4D, Philadelphia, PA 1911–5094.
PM2.5 particle sample, µg; 14.0 Figures L–1 through L–30 to Appendix L.

Fecha: 14/11/2013
87

Fecha: 14/11/2013
88
ER18JY97.022</GPH>

Fecha: 14/11/2013
89
ER18JY97.023</GPH>

Fecha: 14/11/2013
90
ER18JY97.024</GPH>

Fecha: 14/11/2013
91
ER18JY97.025</GPH>

Fecha: 14/11/2013
92
ER17FE98.004</GPH>

Fecha: 14/11/2013
93
ER18JY97.027</GPH>

Fecha: 14/11/2013
94
ER18JY97.028</GPH>

Fecha: 14/11/2013
95
ER18JY97.029</GPH>

Fecha: 14/11/2013
96
ER18JY97.030</GPH>

Fecha: 14/11/2013
97
ER18JY97.031</GPH>

Fecha: 14/11/2013
98
ER18JY97.032</GPH>

Fecha: 14/11/2013
99
ER17FE98.005</GPH>

Fecha: 14/11/2013
100
ER18JY97.034</GPH>

Fecha: 14/11/2013
101
ER18JY97.035</GPH>

Fecha: 14/11/2013
102
ER18JY97.036</GPH>

Fecha: 14/11/2013
103
ER18JY97.037</GPH>

Fecha: 14/11/2013
104
ER18JY97.038</GPH>

Fecha: 14/11/2013
105
ER18JY97.039</GPH>

Fecha: 14/11/2013
106
ER18JY97.040</GPH>

Fecha: 14/11/2013
107
ER18JY97.041</GPH>

Fecha: 14/11/2013
108
ER18JY97.042</GPH>

Fecha: 14/11/2013
109
ER18JY97.043</GPH>

Fecha: 14/11/2013
110
ER18JY97.044</GPH>

Fecha: 14/11/2013
111
ER18JY97.045</GPH>

Fecha: 14/11/2013
112
ER18JY97.046</GPH>

Fecha: 14/11/2013
113
ER17FE98.006</GPH>

Fecha: 14/11/2013
114
ER18JY97.048</GPH>

Fecha: 14/11/2013
115
ER18JY97.049</GPH>

Fecha: 14/11/2013
116
ER18JY97.050</GPH>

Fecha: 14/11/2013
117
ER18JY97.051</GPH>
Pt. 50, App. M 40 CFR Ch. I (7–1–03 Edition)
[62 FR 38714, July 18, 1997, as amended at 64 For samplers that do not have an automatic
FR 19719, Apr. 22, 1999] filter-changing mechanism, the upper limit
is determined by the filter mass loading be-
APPENDIX M TO PART 50—REFERENCE yond which the sampler no longer maintains
METHOD FOR THE DETERMINATION OF the operating flow rate within specified lim-
PARTICULATE MATTER AS PM10 IN its due to increased pressure drop across the
THE ATMOSPHERE loaded filter. This upper limit cannot be
specified precisely because it is a complex
1.0 Applicability. function of the ambient particle size dis-
1.1 This method provides for the measure- tribution and type, humidity, filter type, and
ment of the mass concentration of particu- perhaps other factors. Nevertheless, all sam-
late matter with an aerodynamic diameter plers should be capable of measuring 24-hour
less than or equal to a nominal 10 microm- PM10 mass concentrations of at least 300 µg/
eters (PM1O) in ambient air over a 24-hour m3 while maintaining the operating flow rate
period for purposes of determining attain- within the specified limits.
ment and maintenance of the primary and 4.0 Precision.
secondary national ambient air quality 4.1 The precision of PM10 samplers must be
standards for particulate matter specified in 5 µg/m3 for PM10 concentrations below 80 µg/
§ 50.6 of this chapter. The measurement proc- m3 and 7 percent for PM10 concentrations
ess is nondestructive, and the PM10 sample above 80 µg/m3, as required by part 53 of this
can be subjected to subsequent physical or chapter, which prescribes a test procedure
chemical analyses. Quality assurance proce- that determines the variation in the PM10
dures and guidance are provided in part 58, concentration measurements of identical
Appendices A and B of this chapter and in samplers under typical sampling conditions.
references 1 and 2 of section 12.0 of this ap- Continual assessment of precision via collo-
pendix. cated samplers is required by part 58 of this
2.0 Principle. chapter for PM10 samplers used in certain
2.1 An air sampler draws ambient air at a monitoring networks.
constant flow rate into a specially shaped 5.0 Accuracy.
inlet where the suspended particulate matter 5.1 Because the size of the particles making
is inertially separated into one or more size up ambient particulate matter varies over a
fractions within the PM10 size range. Each wide range and the concentration of par-
size fraction in the PM1O size range is then ticles varies with particle size, it is difficult
collected on a separate filter over the speci- to define the absolute accuracy of PM10 sam-
fied sampling period. The particle size dis- plers. Part 53 of this chapter provides a spec-
crimination characteristics (sampling effec- ification for the sampling effectiveness of
tiveness and 50 percent cutpoint) of the sam- PM10 samplers. This specification requires
pler inlet are prescribed as performance that the expected mass concentration cal-
specifications in part 53 of this chapter. culated for a candidate PM10 sampler, when
2.2 Each filter is weighed (after moisture sampling a specified particle size distribu-
equilibration) before and after use to deter- tion, be within ±10 percent of that calculated
mine the net weight (mass) gain due to col- for an ideal sampler whose sampling effec-
lected PM10. The total volume of air sam- tiveness is explicitly specified. Also, the par-
pled, measured at the actual ambient tem- ticle size for 50 percent sampling effective-
perature and pressure, is determined from ness is required to be 10±0.5 micrometers.
the measured flow rate and the sampling Other specifications related to accuracy
time. The mass concentration of PM10 in the apply to flow measurement and calibration,
ambient air is computed as the total mass of filter media, analytical (weighing) proce-
collected particles in the PM10 size range di- dures, and artifact. The flow rate accuracy of
vided by the volume of air sampled, and is PM10 samplers used in certain monitoring
expressed in micrograms per actual cubic networks is required by part 58 of this chap-
meter (µg/m3). ter to be assessed periodically via flow rate
2.3 A method based on this principle will be audits.
considered a reference method only if the as- 6.0 Potential Sources of Error.
sociated sampler meets the requirements 6.1 Volatile Particles. Volatile particles col-
specified in this appendix and the require- lected on filters are often lost during ship-
ments in part 53 of this chapter, and the ment and/or storage of the filters prior to
Fecha: 14/11/2013
method has been designated as a reference the post-sampling weighing 3. Although ship-
method in accordance with part 53 of this ment or storage of loaded filters is some-
chapter. times unavoidable, filters should be re-
3.0 Range. weighed as soon as practical to minimize
3.1 The lower limit of the mass concentra- these losses.
tion range is determined by the repeatability 6.2 Artifacts. Positive errors in PM10 con-
of filter tare weights, assuming the nominal centration measurements may result from
air sample volume for the sampler. For sam- retention of gaseous species on filters 4, 5.
plers having an automatic filter-changing Such errors include the retention of sulfur
mechanism, there may be no upper limit. dioxide and nitric acid. Retention of sulfur
118
Environmental Protection Agency Pt. 50, App. M
dioxide on filters, followed by oxidation to (d) Protect the filter and sampler from pre-
sulfate, is referred to as artifact sulfate for- cipitation and prevent insects and other de-
mation, a phenomenon which increases with bris from being sampled.
increasing filter alkalinity 6. Little or no ar- (e) Minimize air leaks that would cause
tifact sulfate formation should occur using error in the measurement of the air volume
filters that meet the alkalinity specification passing through the filter.
in section 7.2.4 of this appendix, Artifact ni- (f) Discharge exhaust air at a sufficient
trate formation, resulting primarily from re- distance from the sampler inlet to minimize
tention of nitric acid, occurs to varying de- the sampling of exhaust air.
grees on many filter types, including glass (g) Minimize the collection of dust from
fiber, cellulose ester, and many quartz fiber the supporting surface.
filters 5, 7, 8, 9, 10. Loss of true atmospheric par- 7.1.2 The sampler shall have a sample air
ticulate nitrate during or following sampling inlet system that, when operated within a
may also occur due to dissociation or chem- specified flow rate range, provides particle
ical reaction. This phenomenon has been ob- size discrimination characteristics meeting
served on Teflonfilters 8 and inferred for all of the applicable performance specifica-
tions prescribed in part 53 of this chapter.
quartz fiber filters 11, 12. The magnitude of ni-
The sampler inlet shall show no significant
trate artifact errors in PM10 mass concentra-
wind direction dependence. The latter re-
tion measurements will vary with location
quirement can generally be satisfied by an
and ambient temperature; however, for most
inlet shape that is circularly symmetrical
sampling locations, these errors are expected
about a vertical axis.
to be small. 7.1.3 The sampler shall have a flow control
6.3 Humidity. The effects of ambient humid- device capable of maintaining the sampler’s
ity on the sample are unavoidable. The filter operating flow rate within the flow rate lim-
equilibration procedure in section 9.0 of this its specified for the sampler inlet over nor-
appendix is designed to minimize the effects mal variations in line voltage and filter pres-
of moisture on the filter medium. sure drop.
6.4 Filter Handling. Careful handling of fil- 7.1.4 The sampler shall provide a means to
ters between presampling and postsampling measure the total flow rate during the sam-
weighings is necessary to avoid errors due to pling period. A continuous flow recorder is
damaged filters or loss of collected particles recommended but not required. The flow
from the filters. Use of a filter cartridge or measurement device shall be accurate to ±2
cassette may reduce the magnitude of these percent.
errors. Filters must also meet the integrity 7.1.5 A timing/control device capable of
specification in section 7.2.3 of this appendix. starting and stopping the sampler shall be
6.5 Flow Rate Variation. Variations in the used to obtain a sample collection period of
sampler’s operating flow rate may alter the 24 ±1 hr (1,440 ±60 min). An elapsed time
particle size discrimination characteristics meter, accurate to within ±15 minutes, shall
of the sampler inlet. The magnitude of this be used to measure sampling time. This
error will depend on the sensitivity of the meter is optional for samplers with contin-
inlet to variations in flow rate and on the uous flow recorders if the sampling time
particle distribution in the atmosphere dur- measurement obtained by means of the re-
ing the sampling period. The use of a flow corder meets the ±15 minute accuracy speci-
control device, under section 7.1.3 of this ap- fication.
pendix, is required to minimize this error. 7.1.6 The sampler shall have an associated
6.6 Air Volume Determination. Errors in the operation or instruction manual as required
air volume determination may result from by part 53 of this chapter which includes de-
errors in the flow rate and/or sampling time tailed instructions on the calibration, oper-
measurements. The flow control device ation, and maintenance of the sampler.
serves to minimize errors in the flow rate de- 7.2 Filters.

termination, and an elapsed time meter, 7.2.1 Filter Medium. No commercially avail-
under section 7.1.5 of this appendix, is re- able filter medium is ideal in all respects for
quired to minimize the error in the sampling all samplers. The user’s goals in sampling de-
time measurement. termine the relative importance of various
filter characteristics, e.g., cost, ease of han-
7.0 Apparatus.
dling, physical and chemical characteristics,
Fecha: 14/11/2013
7.1 PM10 Sampler. etc., and, consequently, determine the choice

7.1.1 The sampler shall be designed to: among acceptable filters. Furthermore, cer-
(a) Draw the air sample into the sampler tain types of filters may not be suitable for
inlet and through the particle collection fil- use with some samplers, particularly under
ter at a uniform face velocity. heavy loading conditions (high mass con-
(b) Hold and seal the filter in a horizontal centrations), because of high or rapid in-
position so that sample air is drawn down- crease in the filter flow resistance that
ward through the filter. would exceed the capability of the sampler’s
(c) Allow the filter to be installed and re- flow control device. However, samplers
moved conveniently. equipped with automatic filter-changing
119
Pt. 50, App. M 40 CFR Ch. I (7–1–03 Edition)
mechanisms may allow use of these types of to calibrate or verify the accuracy of the
filters. The specifications given below are sampler’s flow measurement device.
minimum requirements to ensure accept- 8.1.2 Particle size discrimination by iner-
ability of the filter medium for measurement tial separation requires that specific air ve-
of PM10 mass concentrations. Other filter locities be maintained in the sampler’s air
evaluation criteria should be considered to inlet system. Therefore, the flow rate
meet individual sampling and analysis objec- through the sampler’s inlet must be main-
tives. tained throughout the sampling period with-
7.2.2 Collection Efficiency. ™99 percent, as in the design flow rate range specified by the
measured by the DOP test (ASTM–2986) with manufacturer. Design flow rates are speci-
0.3 µm particles at the sampler’s operating fied as actual volumetric flow rates, meas-
face velocity. ured at existing conditions of temperature
7.2.3 Integrity. ±5 µg/m3 (assuming sampler’s and pressure (Qa).
nominal 24-hour air sample volume). Integ- 8.2 Flow Rate Calibration Procedure.
rity is measured as the PM10 concentration 8.2.1 PM10 samplers employ various types
equivalent corresponding to the average dif- of flow control and flow measurement de-
ference between the initial and the final vices. The specific procedure used for flow
weights of a random sample of test filters rate calibration or verification will vary de-
that are weighed and handled under actual pending on the type of flow controller and
or simulated sampling conditions, but have flow rate indicator employed. Calibration is
no air sample passed through them, i.e., fil- in terms of actual volumetric flow rates (Qa)
ter blanks. As a minimum, the test proce- to meet the requirements of section 8.1 of
dure must include initial equilibration and this appendix. The general procedure given
weighing, installation on an inoperative here serves to illustrate the steps involved in
sampler, removal from the sampler, and final the calibration. Consult the sampler manu-
equilibration and weighing. facturer’s instruction manual and reference 2
7.2.4 Alkalinity. <25 microequivalents/gram of section 12.0 of this appendix for specific
of filter, as measured by the procedure given guidance on calibration. Reference 14 of sec-
in reference 13 of section 12.0 of this appendix tion 12.0 of this appendix provides additional
following at least two months storage in a information on various other measures of
clean environment (free from contamination flow rate and their interrelationships.
by acidic gases) at room temperature and hu- 8.2.2 Calibrate the flow rate transfer stand-
midity. ard against a primary flow or volume stand-
7.3 Flow Rate Transfer Standard. The flow ard traceable to NIST. Establish a calibra-
rate transfer standard must be suitable for tion relationship, e.g., an equation or family
the sampler’s operating flow rate and must of curves, such that traceability to the pri-
be calibrated against a primary flow or vol- mary standard is accurate to within 2 per-
ume standard that is traceable to the Na- cent over the expected range of ambient con-
tional Institute of Standard and Technology ditions, i.e., temperatures and pressures,
(NIST). The flow rate transfer standard must under which the transfer standard will be
be capable of measuring the sampler’s oper- used. Recalibrate the transfer standard peri-
ating flow rate with an accuracy of ±2 per- odically.
cent. 8.2.3 Following the sampler manufacturer’s
7.4 Filter Conditioning Environment. instruction manual, remove the sampler
7.4.1 Temperature range. 15 to 30 C. inlet and connect the flow rate transfer
7.4.2 Temperature control. ±3 C. standard to the sampler such that the trans-
7.4.3 Humidity range. 20% to 45% RH. fer standard accurately measures the sam-
7.4.4 Humidity control. ±5% RH. pler’s flow rate. Make sure there are no leaks
7.5 Analytical Balance. The analytical bal- between the transfer standard and the sam-
ance must be suitable for weighing the type pler.
and size of filters required by the sampler. 8.2.4 Choose a minimum of three flow rates
The range and sensitivity required will de- (actual m3/min), spaced over the acceptable
pend on the filter tare weights and mass flow rate range specified for the inlet, under
loadings. Typically, an analytical balance section 7.1.2 of the appendix, that can be ob-
with a sensitivity of 0.1 mg is required for tained by suitable adjustment of the sampler
high volume samplers (flow rates >0.5 m3/ flow rate. In accordance with the sampler
min). Lower volume samplers (flow rates <0.5 manufacturer’s instruction manual, obtain
Fecha: 14/11/2013
m3/min) will require a more sensitive bal- or verify the calibration relationship be-
ance. tween the flow rate (actual m3/min) as indi-
8.0 Calibration. cated by the transfer standard and the sam-
8.1 General Requirements. pler’s flow indicator response. Record the
8.1.1 Calibration of the sampler’s flow ambient temperature and barometric pres-
measurement device is required to establish sure. Temperature and pressure corrections
traceability of subsequent flow measure- to subsequent flow indicator readings may be
ments to a primary standard. A flow rate required for certain types of flow measure-
transfer standard calibrated against a pri- ment devices. When such corrections are
mary flow or volume standard shall be used necessary, correction on an individual or
120
Environmental Protection Agency Pt. 50, App. M
daily basis is preferable. However, seasonal flow rate to the specified setpoint. Stop the
average temperature and average barometric sampler.
pressure for the sampling site may be incor- 9.8 Set the timer to start and stop the sam-
porated into the sampler calibration to avoid pler at appropriate times. Set the elapsed
daily corrections. Consult the sampler manu- time meter to zero or record the initial
facturer’s instruction manual and reference 2 meter reading.
in section 12.0 of this appendix for additional 9.9 Record the sample information (site lo-
guidance. cation or identification number, sample
8.2.5 Following calibration, verify that the date, filter identification number, and sam-
sampler is operating at its design flow rate pler model and serial number).
(actual m3/min) with a clean filter in place. 9.10 Sample for 24±1 hours.
8.2.6 Replace the sampler inlet. 9.11 Determine and record the average flow
9.0 Procedure. rate (Q̄a) in actual m3/min for the sampling
9.1 The sampler shall be operated in ac- period in accordance with the instructions
cordance with the specific guidance provided provided in the sampler manufacturer’s in-
in the sampler manufacturer’s instruction struction manual. Record the elapsed time
manual and in reference 2 in section 12.0 of meter final reading and, if needed, the aver-
this appendix. The general procedure given age ambient temperature and barometric
here assumes that the sampler’s flow rate pressure for the sampling period, in note fol-
calibration is based on flow rates at ambient lowing section 9.6 of this appendix.
conditions (Qa) and serves to illustrate the 9.12 Carefully remove the filter from the
steps involved in the operation of a PM10 sampler, following the sampler manufactur-
sampler. er’s instruction manual. Touch only the
9.2 Inspect each filter for pinholes, par- outer edges of the filter.
ticles, and other imperfections. Establish a 9.13 Place the filter in a protective holder
filter information record and assign an iden- or container, e.g., petri dish, glassine enve-
tification number to each filter. lope, or manila folder.
9.3 Equilibrate each filter in the condi- 9.14 Record any factors such as meteoro-
tioning environment (see 7.4) for at least 24 logical conditions, construction activity,
hours. fires or dust storms, etc., that might be per-
9.4 Following equilibration, weigh each fil- tinent to the measurement on the filter in-
ter and record the presampling weight with formation record.
the filter identification number. 9.15 Transport the exposed sample filter to
9.5 Install a preweighed filter in the sam- the filter conditioning environment as soon
pler following the instructions provided in as possible for equilibration and subsequent
the sampler manufacturer’s instruction man- weighing.
ual. 9.16 Equilibrate the exposed filter in the
9.6 (a) Turn on the sampler and allow it to conditioning environment for at least 24
establish run-temperature conditions. hours under the same temperature and hu-
Record the flow indicator reading and, if midity conditions used for presampling filter
needed, the ambient temperature and baro- equilibration (see section 9.3 of this appen-
metric pressure. Determine the sampler flow dix).
rate (actual m3/min) in accordance with the 9.17 Immediately after equilibration, re-
instructions provided in the sampler manu- weigh the filter and record the postsampling
facturer’s instruction manual. weight with the filter identification number.
(b) Note: No onsite temperature or pres- 10.0 Sampler Maintenance.
sure measurements are necessary if the sam- 10.1 The PM10 sampler shall be maintained
pler’s flow indicator does not require tem- in strict accordance with the maintenance
perature or pressure corrections or if sea- procedures specified in the sampler manufac-
sonal average temperature and average baro- turer’s instruction manual.

metric pressure for the sampling site are in- 11.0 Calculations.
corporated into the sampler calibration, 11.1 Calculate the total volume of air sam-
under section 8.2.4 of this appendix. If indi- pled as:
vidual or daily temperature and pressure V = Q at
corrections are required, ambient tempera-
ture and barometric pressure can be obtained where:
Fecha: 14/11/2013
by on-site measurements or from a nearby V = total air sampled, at ambient tempera-

weather station. Barometric pressure read- ture and pressure,m3;
ings obtained from airports must be station Qa = average sample flow rate at ambient
pressure, not corrected to sea level, and may temperature and pressure, m3/min; and
need to be corrected for differences in ele- t = sampling time, min.
vation between the sampling site and the 11.2 (a) Calculate the PM10 concentration
airport. as:
9.7 If the flow rate is outside the accept-
PM10 = (Wf¥Wi)×106/V
able range specified by the manufacturer,
check for leaks, and if necessary, adjust the where:
121
Pt. 50, App. N 40 CFR Ch. I (7–1–03 Edition)
PM10 = mass concentration of PM10, µg/m3; 14. Smith, F., P.S. Wohlschlegel, R.S.C.
Wf, Wi = final and initial weights of filter col- Rogers, and D.J. Mulligan. Investigation of
lecting PM1O particles, g; and Flow Rate Calibration Procedures Associ-
106 = conversion of g to µg. ated With the High Volume Method for De-
termination of Suspended Particulates.
(b) Note: If more than one size fraction in
EPA–600/4–78–047, U.S. Environmental Pro-
the PM10 size range is collected by the sam-
tection Agency, Research Triangle Park, NC
pler, the sum of the net weight gain by each
27711, 1978.
collection filter [S(Wf¥Wi)] is used to cal-
culate the PM10 mass concentration. [62 FR 38753, July 18, 1997]
12.0 References.
1. Quality Assurance Handbook for Air Pol- APPENDIX N TO PART 50—INTERPRETA-
lution Measurement Systems, Volume I, TION OF THE NATIONAL AMBIENT AIR
Principles. EPA–600/9–76–005, March 1976.
QUALITY STANDARDS FOR PARTICU-
Available from CERI, ORD Publications,
U.S. Environmental Protection Agency, 26 LATE MATTER
West St. Clair Street, Cincinnati, OH 45268.
1.0 General.
(a) This appendix explains the data han-
dling conventions and computations nec-
essary for determining when the annual and
027a, May 1977. Available from CERI, ORD
24-hour primary and secondary national am-
Publications, U.S. Environmental Protection
bient air quality standards for PM specified
Agency, 26 West St. Clair Street, Cincinnati,
in § 50.7 of this chapter are met. Particulate
OH 45268.
matter is measured in the ambient air as
3. Clement, R.E., and F.W. Karasek. Sam-
PM10 and PM2.5 (particles with an aero-
ple Composition Changes in Sampling and
dynamic diameter less than or equal to a
Analysis of Organic Compounds in Aerosols.
nominal 10 and 2.5 micrometers, respec-
Int. J. Environ. Analyt. Chem., 7:109, 1979.
4. Lee, R.E., Jr., and J. Wagman. A Sam- tively) by a reference method based on ap-
pling Anomaly in the Determination of At- pendix M of this part for PM10 and on appen-
mospheric Sulfate Concentration. Amer. Ind. dix L of this part for PM2.5, as applicable, and
Hyg. Assoc. J., 27:266, 1966. designated in accordance with part 53 of this
5. Appel, B.R., S.M. Wall, Y. Tokiwa, and chapter, or by an equivalent method des-
M. Haik. Interference Effects in Sampling ignated in accordance with part 53 of this
Particulate Nitrate in Ambient Air. Atmos. chapter. Data handling and computation pro-
Environ., 13:319, 1979. cedures to be used in making comparisons
6. Coutant, R.W. Effect of Environmental between reported PM10 and PM2.5 concentra-
Variables on Collection of Atmospheric Sul- tions and the levels of the PM standards are
fate. Environ. Sci. Technol., 11:873, specified in the following sections.
1977.Spicer, C.W., and P. Schumacher. Inter- (b) Data resulting from uncontrollable or
ference in Sampling Atmospheric Particu- natural events, for example structural fires
late Nitrate. Atmos. Environ., 11:873, 1977. or high winds, may require special consider-
8. Appel, B.R., Y. Tokiwa, and M. Haik. ation. In some cases, it may be appropriate
Sampling of Nitrates in Ambient Air. Atmos. to exclude these data because they could re-
Environ., 15:283, 1981. sult in inappropriate values to compare with
9. Spicer, C.W., and P.M. Schumacher. Par- the levels of the PM standards. In other
ticulate Nitrate: Laboratory and Field Stud- cases, it may be more appropriate to retain
ies of Major Sampling Interferences. Atmos. the data for comparison with the level of the
Environ., 13:543, 1979. PM standards and then allow the EPA to for-
10. Appel, B.R. Letter to Larry Purdue, mulate the appropriate regulatory response.
U.S. EPA, Environmental Monitoring and Whether to exclude, retain, or make adjust-
Support Laboratory. March 18, 1982, Docket ments to the data affected by uncontrollable
No. A–82–37, II–I–1. or natural events is subject to the approval
11. Pierson, W.R., W.W. Brachaczek, T.J. of the appropriate Regional Administrator.
Korniski, T.J. Truex, and J.W. Butler. Arti- (c) The terms used in this appendix are de-
fact Formation of Sulfate, Nitrate, and Hy- fined as follows:
drogen Ion on Backup Filters: Allegheny Average and mean refer to an arithmetic
Fecha: 14/11/2013
Mountain Experiment. J. Air Pollut. Control mean.

Assoc., 30:30, 1980. Daily value for PM refers to the 24-hour av-
12. Dunwoody, C.L. Rapid Nitrate Loss erage concentration of PM calculated or
From PM10 Filters. J. Air Pollut. Control measured from midnight to midnight (local
Assoc., 36:817, 1986. time) for PM10 or PM2.5.
13. Harrell, R.M. Measuring the Alkalinity Designated monitors are those monitoring
of Hi-Vol Air Filters. EMSL/RTP–SOP–QAD– sites designated in a State PM Monitoring
534, October 1985. Available from the U.S. En- Network Description for spatial averaging in
vironmental Protection Agency, EMSL/QAD, areas opting for spatial averaging in accord-
Research Triangle Park, NC 27711. ance with part 58 of this chapter.
122
Environmental Protection Agency Pt. 50, App. N
98th percentile (used for PM2.5) means the tions of section 2.3 of this appendix) is great-
daily value out of a year of monitoring data er than the level of the standard.
below which 98 percent of all values in the (c) Situations may arise in which there are
group fall. compelling reasons to retain years con-
99th percentile (used for PM10) means the taining quarters which do not meet the data
daily value out of a year of monitoring data completeness requirement of 75 percent or
below which 99 percent of all values in the the minimum number of 11 samples. The use
group fall. of less than complete data is subject to the
Year refers to a calendar year.
approval of the appropriate Regional Admin-
(d) Sections 2.1 and 2.5 of this appendix
contain data handling instructions for the istrator.
option of using a spatially averaged network (d) The equations for calculating the 3-year
of monitors for the annual standard. If spa- average annual mean of the PM2.5 standard
tial averaging is not considered for an area, are given in section 2.5 of this appendix.
then the spatial average is equivalent to the 2.2 24-Hour PM2.5 Standard.
annual average of a single site and is treated (a) The 24-hour PM2.5 standard is met when
accordingly in subsequent calculations. For the 3-year average of the 98th percentile val-
example, paragraph (a)(3) of section 2.1 of ues at each monitoring site is less than or
this appendix could be eliminated since the equal to 65 µg/m3. This comparison shall be
spatial average would be equivalent to the based on 3 consecutive, complete years of air
annual average. quality data. A year meets data complete-
2.0 Comparisons with the PM2.5 Standards. ness requirements when at least 75 percent of
2.1 Annual PM2.5 Standard. the scheduled sampling days for each quarter
(a) The annual PM2.5 standard is met when have valid data. However, years with high
the 3-year average of the spatially averaged concentrations shall not be ignored just be-
annual means is less than or equal to 15.0 µg/ cause they are comprised of quarters with
m3. The 3-year average of the spatially aver-
less than complete data. Thus, in computing
aged annual means is determined by aver-
the 3-year average 98th percentile value, years
aging quarterly means at each monitor to
containing quarters with less than 75 percent
obtain the annual mean PM2.5 concentrations
at each monitor, then averaging across all data completeness shall be included in the
designated monitors, and finally averaging computation if the annual 98th percentile
for 3 consecutive years. The steps can be value (rounded according to the conventions
summarized as follows: of section 2.3 of this appendix) is greater
(1) Average 24-hour measurements to ob- than the level of the standard.
tain quarterly means at each monitor. (b) Situations may arise in which there are
(2) Average quarterly means to obtain an- compelling reasons to retain years con-
nual means at each monitor. taining quarters which do not meet the data
(3) Average across designated monitoring completeness requirement. The use of less
sites to obtain an annual spatial mean for an than complete data is subject to the ap-
area (this can be one site in which case the proval of the appropriate Regional Adminis-
spatial mean is equal to the annual mean). trator.
(4) Average 3 years of annual spatial means (c) The equations for calculating the 3-year
to obtain a 3-year average of spatially aver- average of the annual 98th percentile values is
aged annual means. given in section 2.6 of this appendix.
(b) In the case of spatial averaging, 3 years 2.3 Rounding Conventions. For the purposes
of spatial averages are required to dem- of comparing calculated values to the appli-
onstrate that the standard has been met. cable level of the standard, it is necessary to
Designated sites with less than 3 years of round the final results of the calculations de-
data shall be included in spatial averages for
scribed in sections 2.5 and 2.6 of this appen-
those years that data completeness require-

dix. For the annual PM2.5 standard, the 3-
ments are met. For the annual PM2.5 stand-
year average of the spatially averaged an-
ard, a year meets data completeness require-
nual means shall be rounded to the nearest
ments when at least 75 percent of the sched-
0.1 µg/m3 (decimals 0.05 and greater are
uled sampling days for each quarter have
valid data. However, years with high con- rounded up to the next 0.1, and any decimal
centrations and more than a minimal lower than 0.05 is rounded down to the near-
Fecha: 14/11/2013
amount of data (at least 11 samples in each est 0.1). For the 24-hour PM2.5 standard, the
quarter) shall not be ignored just because 3-year average of the annual 98th percentile
they are comprised of quarters with less values shall be rounded to the nearest 1 µg/
than complete data. Thus, in computing an- m3 (decimals 0.5 and greater are rounded up
nual spatially averaged means, years con- to nearest whole number, and any decimal
taining quarters with at least 11 samples but lower than 0.5 is rounded down to the nearest
less than 75 percent data completeness shall whole number).
be included in the computation if the result- 2.4 Monitoring Considerations.
ing spatially averaged annual mean con- (a) Section 58.13 of this chapter specifies
centration (rounded according to the conven- the required minimum frequency of sampling
123
for PM2.5. Exceptions to the specified sam- then computing the average of these values
pling frequencies, such as a reduced fre- across sites:
quency during a season of expected low con-
centrations, are subject to the approval of Equation 3
the appropriate Regional Administrator.
Section 58.14 of 40 CFR part 58 and section 2.8 ns
1
∑
of appendix D of 40 CFR part 58, specify
which monitors are eligible for making com- xy = x y ,s
parisons with the PM standards. In deter- ns s =1
mining a spatial mean using two or more where:
monitoring sites operating in a given year,
the annual mean for an individual site may x̄y = the spatially averaged mean for year y;
be included in the spatial mean if and only if x̄y,s = the annual mean for year y and site s;
the mean for that site meets the criterion and
specified in § 2.8 of appendix D of 40 CFR part ns = the number of sites designated to be
58. In the event data from an otherwise eligi- averaged.
ble site is excluded from being averaged with (2) In the event that an area designated for
data from other sites on the basis of this cri- spatial averaging has two or more sites at
terion, then the 3-year mean from that site the same location producing data for the
shall be compared directly to the annual same time periods, the sites are averaged to-
standard. gether before using Equation 3 by:
(b) For the annual PM2.5 standard, when
designated monitors are located at the same Equation 4
site and are reporting PM2.5 values for the
same time periods, and when spatial aver- nc
aging has been chosen, their concentrations 1
shall be averaged before an area-wide spatial
average is calculated. Such monitors will
x y,s* =
nc
∑ x y ,s
s =1
then be considered as one monitor. where:
2.5 Equations for the Annual PM2.5 Standard.
(a) An annual mean value for PM2.5 is de- x̄y,s* = the annual mean for year y for the
termined by first averaging the daily values sites at the same location (which will now
of a calendar quarter: be considered one site);
nc = the number of sites at the same location
Equation 1 designated to be included in the spatial av-
erage; and
nq x̄y,s = the annual mean for year y and site s.
1
x q , y ,s =
nq
∑ x i , q , y ,s (d) The 3-year average of the spatially
averaged annual means is calculated by
i =1 using the following equation:
where:
x̄q,y,s = the mean for quarter q of year y for Equation 5
site s;
nq = the number of monitored values in the 3
1
quarter; and
xi,q,y,s = the ith value in quarter q for year y
x=
3
∑ xy
for site s. y =1
ER18JY97.004</GPH>
(b) The following equation is then to be where:
used for calculation of the annual mean: x̄ = the 3-year average of the spatially aver-
aged annual means; and

Equation 2 x̄y = the spatially averaged annual mean for
year y. ER18JY97.003</GPH>
4
1
x y, s =
4
∑ x q, y, s Example 1—Area Designated for Spatial Aver-
aging That Meets the Primary Annual PM2.5
q =1
Fecha: 14/11/2013
Standard.
where:
a. In an area designated for spatial aver-
x̄y,s = the annual mean concentration for year
ER18JY97.001</GPH> ER18JY97.002</GPH>
aging, four designated monitors recorded

y (y = 1, 2, or 3) and for site s; and
data in at least 1 year of a particular 3-year
x̄q,y,s = the mean for quarter q of year y for
period. Using Equations 1 and 2, the annual
site s.
means for PM2.5 at each site are calculated
(c)(1) The spatially averaged annual mean for each year. The following table can be cre-
for year y is computed by first calculating ated from the results. Data completeness
the annual mean for each site designated to percentages for the quarter with the fewest
be included in a spatial average, x̄y,s, and number of samples are also shown.
124
ER18JY97.000</GPH>
TABLE 1—RESULTS FROM EQUATIONS 1 AND 2

Site ⊥1 Site ⊥2 Site ⊥3 Site ⊥4 Spatial mean
Year 1 ................... Annual mean (µg/m3) ....... 12.7 ...................... ...................... ...................... 12.7
% data completeness ....... 80 0 0 0 ......................
Year 2 ................... Annual mean (µg/m3) ....... 12.6 17.5 15.2 ...................... 15.05
Year 3 ................... Annual mean (µg/m3) ....... 12.5 18.5 14.1 16.9 15.50
3-year mean ......... ........................................... ...................... ...................... ...................... ...................... 14.42
b. The data from these sites are averaged culation of the spatial mean for year 3 and in
in the order described in section 2.1 of this the calculation of the 3-year average. The 3-
appendix. Note that the annual mean from year average is rounded to 14.4 µg/m3, indi-
site #3 in year 2 and the annual mean from cating that this area meets the annual PM2.5
site #4 in year 3 do not meet the 75 percent standard.
data completeness criteria. Assuming the 38
percent data completeness represents a quar- Example 2—Area With Two Monitors at the
ter with fewer than 11 samples, site #3 in Same Location That Meets the Primary An-
year 2 does not meet the minimum data com- nual PM2.5 Standard.
pleteness requirement of 11 samples in each a. In an area designated for spatial aver-
quarter. The site is therefore excluded from aging, six designated monitors, with two
the calculation of the spatial mean for year monitors at the same location (#5 and #6),
2. However, since the spatial mean for year 3 recorded data in a particular 3-year period.
is above the level of the standard and the Using Equations 1 and 2, the annual means
minimum data requirement of 11 samples in for PM2.5 are calculated for each year. The
each quarter has been met, the annual mean following table can be created from the re-
from site #4 in year 3 is included in the cal- sults.
TABLE 2—RESULTS FROM EQUATIONS 1 AND 2
Annual Average of Spatial
mean (µg/ Site ⊥1 Site ⊥2 Site ⊥3 Site ⊥4 Site ⊥5 Site ⊥6 ⊥5 and ⊥6 mean
m3)
Year 1 ....... 12.9 9.9 12.6 11.1 14.5 14.6 14.55 12.21
Year 2 ....... 14.5 13.3 12.2 10.9 16.1 16.0 16.05 13.39
Year 3 ....... 14.4 12.4 11.5 9.7 12.3 12.1 12.20 12.04
3-Year
mean ...... .................... .................... .................... .................... .................... .................... .................. 12.55
b. The annual means for sites #5 and #6 are Example 3—Area With a Single Monitor That
averaged together using Equation 4 before Meets the Primary Annual PM2.5 Standard.
the spatial average is calculated using Equa- a. Given data from a single monitor in an
tion 3 since they are in the same location. area, the calculations are as follows. Using
The 3-year mean is rounded to 12.6 µg/m3, in- Equations 1 and 2, the annual means for
dicating that this area meets the annual PM2.5 are calculated for each year. If the an-
PM2.5 standard. nual means are 10.28, 17.38, and 12.25 µg/m3,
then the 3-year mean is:
x = (1 / 3) × (10.28 +17.38 +12.25) = 13.303 µg /m 3 .

Fecha: 14/11/2013
b. This value is rounded to 13.3, indicating plished by the following steps. All the daily
that this area meets the annual PM2.5 stand- values from a particular site and year com-
ard. prise a series of values (x1, x2, x3, ..., xn), that
2.6 Equations for the 24-Hour PM2.5 Standard. can be sorted into a series where each num-
(a) When the data for a particular site and ber is equal to or larger than the preceding
year meet the data completeness require- number (x[1], x[2], x[3], ..., x[n]). In this case, x[1]
ments in section 2.2 of this appendix, cal- is the smallest number and x[n] is the largest
culation of the 98th percentile is accom- value. The 98th percentile is found from the
125
ER18JY97.005</GPH>
sorted series of daily values which is ordered Equation 7
from the lowest to the highest number. Com-
pute (0.98) × (n) as the number ‘‘i.d’’, where 3
‘‘i’’ is the integer part of the result and ‘‘d’’
is the decimal part of the result. The 98th per- ∑ P0.98, y
centile value for year y, P0.98, y, is given by y =1
Equation 6: P0.98 =
3
Equation 6 (c) The 3-year average 98th percentile is
rounded according to the conventions in sec-
P0.98, y = X[i +1] tion 2.3 of this appendix before a comparison
with the standard is made.
where:
P0.98,y = 98th percentile for year y; Example 4—Ambient Monitoring Site With
x[i=1] = the (i=1)th number in the ordered series Every-Day Sampling That Meets the Primary
of numbers; and 24-Hour PM2.5 Standard.
i = the integer part of the product of 0.98 and a. In each year of a particular 3 year pe-
n. riod, varying numbers of daily PM2.5 values
(b) The 3-year average 98th percentile is (e.g., 281, 304, and 296) out of a possible 365
then calculated by averaging the annual 98th values were recorded at a particular site
percentiles: with the following ranked values (in µg/m3):
TABLE 3—ORDERED MONITORING DATA FOR 3 YEARS

Year 1 Year 2 Year 3
j rank Xj value j rank Xj value j rank Xj value
275 57.9 296 54.3 290 66.0

276 59.0 297 57.1 291 68.4
277 62.2 298 63.0 292 69.8
b. Using Equation 6, the 98th percentile val-

ues for each year are calculated as follows:
0.98 × 281 = 275.38 ⇒ i + 1 = 276 ⇒ P0.98,1 = X[ 276 ] = 59.0 µg / m 3
0.98 × 304 = 297.92 ⇒ i + 1 = 298 ⇒ P0.98,2 = X[ 298] = 63.0 µg / m 3
ER18JY97.011</GPH>
0.98 × 296 = 290.07 ⇒ i + 1 = 291 ⇒ P0.98,3 = X[ 291] = 68.4 µg / m 3
ER18JY97.010</GPH>
c.1. Using Equation 7, the 3-year average
98th percentile is calculated as follows:
ER18JY97.009</GPH>
59.0 + 63.0 + 68.4

P0.98 = = 63.46 µg / m 3 , which rounds to 63µg / m 3 .
3
Fecha: 14/11/2013
ER18JY97.007</GPH> ER18JY97.008</GPH>
2. Therefore, this site meets the 24-hour erage of the annual means is determined by
PM2.5 standard. averaging quarterly means to obtain annual
3.0 Comparisons with the PM10 Standards. mean PM10 concentrations for 3 consecutive,
3.1 Annual PM10 Standard. complete years at each monitoring site. The
(a) The annual PM10 standard is met when steps can be summarized as follows:
the 3-year average of the annual mean PM10 (1) Average 24-hour measurements to ob-
concentrations at each monitoring site is tain a quarterly mean.
less than or equal to 50 µg/m3. The 3-year av-
126
ER18JY97.006</GPH>
(2) Average quarterly means to obtain an rounded up to the next whole number, and
annual mean. any decimal less than 0.5 is rounded down to
(3) Average annual means to obtain a 3- the nearest whole number). For the 24-hour
year mean. PM10 standard, the 3-year average of the an-
(b) For the annual PM10 standard, a year nual 99th percentile values of PM10 shall be
meets data completeness requirements when rounded to the nearest 10 µg/m3 (155 µg/m3
at least 75 percent of the scheduled sampling and greater would be rounded to 160 µg/m3
days for each quarter have valid data. How- and 154 µg/m3 and less would be rounded to
ever, years with high concentrations and 150 µg/m3).
more than a minimal amount of data (at 3.4 Monitoring Considerations. Section 58.13
least 11 samples in each quarter) shall not be of this chapter specifies the required min-
ignored just because they are comprised of imum frequency of sampling for PM10. Ex-
quarters with less than complete data. Thus, ceptions to the specified sampling fre-
in computing the 3-year average annual quencies, such as a reduced frequency during
mean concentration, years containing quar- a season of expected low concentrations, are
ters with at least 11 samples but less than 75 subject to the approval of the appropriate
percent data completeness shall be included Regional Administrator. For making com-
in the computation if the annual mean con- parisons with the PM10 NAAQS, all sites
centration (rounded according to the conven- meeting applicable requirements in part 58
tions of section 2.3 of this appendix) is great- of this chapter would be used.
er than the level of the standard. 3.5 Equations for the Annual PM10 Standard.
(c) Situations may arise in which there are (a) An annual arithmetic mean value for
compelling reasons to retain years con- PM10 is determined by first averaging the 24-
taining quarters which do not meet the data hour values of a calendar quarter using the
completeness requirement of 75 percent or following equation:
the minimum number of 11 samples. The use
of less than complete data is subject to the Equation 8
approval of the appropriate Regional Admin-
istrator. nq
(d) The equations for calculating the 3-year 1
average annual mean of the PM10 standard
are given in section 3.5 of this appendix.
x q, y =
nq
∑ x i, q, y
i =1
3.2 24-Hour PM10 Standard.
(a) The 24-hour PM10 standard is met when where:
the 3-year average of the annual 99th per- x̄q,y = the mean for quarter q of year y;
centile values at each monitoring site is less nq = the number of monitored values in the
than or equal to 150 µg/m3. This comparison quarter; and
shall be based on 3 consecutive, complete xi,q,y = the ith value in quarter q for year y.
years of air quality data. A year meets data (b) The following equation is then to be
completeness requirements when at least 75 used for calculation of the annual mean:
percent of the scheduled sampling days for
each quarter have valid data. However, years Equation 9
with high concentrations shall not be ig-
nored just because they are comprised of 4
1
∑
quarters with less than complete data. Thus,
in computing the 3-year average of the an- xy = x q, y
nual 99th percentile values, years containing 4 q =1
quarters with less than 75 percent data com- where:
pleteness shall be included in the computa-
tion if the annual 99th percentile value x̄y = the annual mean concentration for year
(rounded according to the conventions of sec- y, (y=1, 2, or 3); and

tion 2.3 of this appendix) is greater than the xq,y = the mean for a quarter q of year y.
level of the standard. (c) The 3-year average of the annual means
(b) Situations may arise in which there are is calculated by using the following equa-
compelling reasons to retain years con- tion:
taining quarters which do not meet the data
completeness requirement. The use of less Equation 10
Fecha: 14/11/2013
than complete data is subject to the ap-

proval of the appropriate Regional Adminis- 3
1
ER18JY97.013</GPH> ER18JY97.014</GPH>
trator.
(c) The equation for calculating the 3-year
x=
3
∑ xy
average of the annual 99th percentile values is y =1
given in section 2.6 of this appendix. where:
3.3 Rounding Conventions. For the annual
x̄ = the 3-year average of the annual means;
PM10 standard, the 3-year average of the an-
and
nual PM10 means shall be rounded to the
nearest 1 µg/m3 (decimals 0.5 and greater are x̄y = the annual mean for calendar year y.
127
ER18JY97.012</GPH>
Example 5—Ambient Monitoring Site That Does for PM10 are calculated for each year. If the
Not Meet the Annual PM10 Standard. annual means are 52.42, 82.17, and 63.23 µg/m3,
a. Given data from a PM10 monitor and then the 3-year average annual mean is:
using Equations 8 and 9, the annual means
x = (1 / 3) × (52.42 + 82.17 + 63.23) = 65.94, which is rounded to 66 µg / m 3 .
b. Therefore, this site does not meet the x[i=1] = the (i=1)th number in the ordered series
annual PM10 standard. of numbers; and
3.6 Equation for the 24-Hour PM10 Standard. i = the integer part of the product of 0.99 and
(a) When the data for a particular site and n.
year meet the data completeness require-
ments in section 3.2 of this appendix, cal- (b) The 3-year average 99th percentile value
culation of the 99th percentile is accom- is then calculated by averaging the annual
99th percentiles:
plished by the following steps. All the daily
values from a particular site and year com- Equation 12
prise a series of values (x1, x2, x3, ..., xn) that
can be sorted into a series where each num- 3
∑ P0.99,y
ber is equal to or larger than the preceding
number (x[1], x[2], x[3], ..., x[n]). In this case, x[1]
is the smallest number and x[n] is the larg- y =1
est value. The 99th percentile is found from P0.99 =
the sorted series of daily values which is or-
3
dered from the lowest to the highest number. (c) The 3-year average 99th percentile is
rounded according to the conventions in sec-
Compute (0.99) × (n) as the number ‘‘i.d’’,
tion 3.3 of this appendix before a comparison
where ‘‘i’’ is the integer part of the result
with the standard is made.
and ‘‘d’’ is the decimal part of the result.
The 99th percentile value for year y, P0.99,y, is
given by Equation 11: Example 6—Ambient Monitoring Site With Sam-
pling Every Sixth Day That Meets the Pri-
Equation 11 mary 24-Hour PM10 Standard.
a. In each year of a particular 3 year pe-
P0.99, y = X [ i +1] riod, varying numbers of PM10 daily values
(e.g., 110, 98, and 100) out of a possible 121
where: daily values were recorded at a particular
site with the following ranked values (in µg/
P0.99,y = the 99th percentile for year y; m3):
TABLE 4—ORDERED MONITORING DATA FOR 3 YEARS
Year 1 Year 2 Year 3
j rank Xj value j rank Xj value j rank Xj value
ER18JY97.019</GPH>
108 120 96 143 98 140
109 128 97 148 99 144

110 130 98 150 100 147
ER18JY97.018</GPH>
b. Using Equation 11, the 99th percentile

values for each year are calculated as fol-
lows:
Fecha: 14/11/2013
ER18JY97.016</GPH> ER18JY97.017</GPH>
0.99 × 110 = 108.9 ⇒ i + 1 = 109 ⇒ P0.99,1 = X[109] = 128µg / m 3
0.99 × 98 = 97.02 ⇒ i + 1 = 98 ⇒ P0.99,2 = X[ 98] = 150 µg / m 3
128
ER18JY97.015</GPH>
Environmental Protection Agency Pt. 51
0.99 × 100 = 99 ⇒ i + 1 = 100 ⇒ P0.99,3 = X[100 ] = 147µg / m 3
c. 1. Using Equation 12, the 3-year average

99th percentile is calculated as follows:
128 + 50 + 147
= 141.7 µg / m 3 rounds to 140 µg / m 3 .
3
2. Therefore, this site meets the 24-hour 51.102 Public hearings.

PM10 standard. 51.103 Submission of plans, preliminary re-
[62 FR 38755, July 18, 1997] view of plans.
51.104 Revisions.
51.105 Approval of plans.
PART 51—REQUIREMENTS FOR
PREPARATION, ADOPTION, AND Subpart G—Control Strategy
SUBMITTAL OF IMPLEMENTATION 51.110 Attainment and maintenance of na-
PLANS tional standards.
51.111 Description of control measures.
Sec. 51.112 Demonstration of adequacy.
51.113 [Reserved]
Subpart A—Emission Inventory Reporting 51.114 Emissions data and projections.
Requirements 51.115 Air quality data and projections.
GENERAL INFORMATION FOR INVENTORY 51.116 Data availability.
PREPARERS 51.117 Additional provisions for lead.
51.118 Stack height provisions.
51.1 Who is responsible for actions described 51.119 Intermittent control systems.
in this subpart? 51.120 Requirements for State Implementa-
51.5 What tools are available to help pre- tion Plan revisions relating to new
pare and report emissions data? motor vehicles.
51.10 How does my State report emissions
51.121 Findings and requirements for sub-
that are required by the NOX SIP Call?
mission of State implementation plan re-
SPECIFIC REPORTING REQUIREMENTS visions relating to emissions of oxides of
nitrogen.
51.15 What data does my State need to re- 51.122 Emissions reporting requirements for
port to EPA? SIP revisions relating to budgets for NOX
51.20 What are the emission thresholds that emissions.
separate point and area sources?
51.25 What geographic area must my State’s
Subpart H—Prevention of Air Pollution
inventory cover?
51.30 When does my State report the data to Emergency Episodes
EPA?
51.150 Classification of regions for episode

51.35 How can my State equalize the effort plans.
for annual reporting?
51.151 Significant harm levels.
51.40 In what form should my State report
51.152 Contingency plans.
the data to EPA?
51.45 Where should my State report the 51.153 Reevaluation of episode plans.
data?
Subpart I—Review of New Sources and
Fecha: 14/11/2013
APPENDIX A TO SUBPART A OF PART 51—TA-

BLES AND GLOSSARY
Modifications
APPENDIX B TO SUBPART A OF PART 51 51.160 Legally enforceable procedures.
[RESERVED]
51.161 Public availability of information.
51.162 Identification of responsible agency.
Subparts B—E [Reserved]
51.163 Administrative procedures.
Subpart F—Procedural Requirements 51.164 Stack height procedures.
51.165 Permit requirements.
ER18JY97.021</GPH>
51.100 Definitions. 51.166 Prevention of significant deteriora-

51.101 Stipulations. tion of air quality.
129
ER18JY97.020</GPH>

40cfr50.12 - 2003

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

40cfr50.12 - 2003

Uploaded by

Copyright:

Available Formats

Environmental Protection Agency Pt. 50, App.

nual fourth-highest daily maximum 8- [43 FR 46258, Oct. 5, 1978]

Copia controlada No.__3___

Copia controlada No.__3___

Copia controlada No.__3___

lower temperatures to prevent condensation (within 1 °C). Adequate temperature control

distilled water. 8.2.10 Standardized sulfite solution for the

where: percent potassium iodide (KI) solution in a

tion is properly standardized. 7. To check the purity of the PRA, perform

may be purified, a stock solution prepared, should be performed.

where: volume standards only, i.e., bubble flow-

Pb = 760 −.076(H) mm Hg, or Pb = 101−.01(H)kPa ( 7)

where: ture is above 10 °C. Store the sample at 5° ±5

9.5 Sample Storage and Shipment. Remove

of sampling, it must be stored at 5° ±5 °C 10.2 Static Calibration Procedure (Option 1).

Volume timer that has been set for 30 minutes. Bring

µg SO 2 = VTCM / SO 2 × C TCM / SO 2 × D (11)

where: a low flow of dry carrier gas to a mixing

sult of an impure dye and/or an improperly

(±0.1 °C) and the device has been accurately Qd + Qp

Copia controlada No.__3___

atmosphere. Be sure that the total flow rate

with higher absorbances can be diluted up to Vstd Va

taining mercury compounds must be stored depending on the calibration procedure

DATA FORM 5. After 24 hours, allow the solution to

vessel in a hood. Method for Determination of Sulfur Dioxide

Rosenberg, and J. P. Bell, Preparation of particularly at high elevations. The actual

7.4 Flow rate measurement devices.

form face velocity.

vary the flow through the sampler unit, a ture conditions.

DP = differential pressure at inlet to volume applicable, insert the appropriate resistance

Copia controlada No.__3___

10.0 Calculations of TSP Concentration. may be achieved by dividing the sampling

step 9.3.9. Then using this expression and the

Poll. Cont. Assoc., Vol. 30 (169), 1980.

Copia controlada No.__3___

Copia controlada No.__3___

Copia controlada No.__3___

Copia controlada No.__3___

[47 FR 54912, Dec. 6, 1982; 48 FR 17355, Apr. 22, 1983]

APPENDIX C TO PART 50—MEASUREMENT through a cell containing the gas sample to

CARBON MONOXIDE IN THE ATMOS- ured by a suitable detector. The photometer

Copia controlada No.__3___

Copia controlada No.__3___

Copia controlada No.__3___

[47 FR 54922, Dec. 6, 1982; 48 FR 17355, Apr. 22, 1983]

APPENDIX D TO PART 50—MEASUREMENT

at a wavelength of 254 nm with sufficient

c = O3 concentration in atmospheres suitable means to assure that no O3 is gen-

where: Record the O3 concentration and the cor-

automatically evaluate all or part of equa- response.

approximation, a manual correction may be obtained by dilution of the O3 concentration

Copia controlada No.__3___

Copia controlada No.__3___

the sampling system. capable of measuring and monitoring NO

cylinders against either an NO SRM or CRM

which will cause a detectable response on the (7)

NO/NOX/NO2 analyzer or which might react FO + FNO

NO flowrate (FNO) (see Figure 1) must be ad-

FNO × [NO 2 ]IMP

where: If substantial adjustment of the NO2 span

recorder response (% scale) = ly spaced points across the remaining scale

are suggested). Calculate each NO2 con-

Copia controlada No.3_

Copia controlada No.3_

Copia controlada No.3_

Copia controlada No.3_

Copia controlada No.3_

Copia controlada No.3_

Copia controlada No.3_

Copia controlada No.3_

Copia controlada No.3_

Copia controlada No.3_

Copia controlada No.3_

Copia controlada No.3_

Copia controlada No.3_

Copia controlada No.3_

Copia controlada No.3_

Copia controlada No.3_

Copia controlada No.3_

Copia controlada No.3_

Copia controlada No.3_

Copia controlada No.3_

Copia controlada No.3_

Copia controlada No.3_

Copia controlada No.3_

Copia controlada No.3_

Copia controlada No.3_

Copia controlada No.3_

Copia controlada No.3_

Copia controlada No.3_

Copia controlada No.3_

Copia controlada No.3_

Copia controlada No.3_

Copia controlada No.3_

Copia controlada No.3_

Copia controlada No.3_

Copia controlada No.3_

Copia controlada No.3_

Copia controlada No.3_

Copia controlada No.3_

Copia controlada No.3_