ISOKINETIC
SAMPLINGBY EPA METHOD306
FOR COMPLlANCE DEMONSTRATIONS
AND MONITORING
Prepared for AESF Chrome Summit
May, 1995
Presented By:
Kevin Walls
of
INTEGRAL COMPLIANCE SYSTEM& hC.
INTEGRAL COMPLIANCE SYSTEMS, hC.
110 Laurentum Parkway
Abingdon, Maryland 21009
Phone: (410) 515-3248
PLANNING FOR THE METHOD 306 C P 6 SAMPLING
EFFORT USING ISOKINETIC METHODOLOGY
1.
IDENTIFY POTENTIAL SOURCES
2.
DRAFT AND SUBMIT SAMPLING PROTOCOL
3.
SCHEDULE DEMONSTRATION
PRODUCTION
4.
SELECT AN APPROVED LABORATORY
5.
MAKE SITE IMPROVEMENTS FOR SAMPLING
6.
ACQUIRE REAGENTS AND SAMPLING EQUIPMENT
ACCORDING
TO
IDENTIFY POTENTIAL SOURCES
The instrumentationand technique that is part of the isokinetic sampling methodology can
be used to demonstrate compliance of various substances. The Environmental Protection
Agency (EPA) has published in Title 40 of the Code of Federal Regulations for the
Protection of the Environment, Part 60, a number of emission guidelines and standards
for performance for stationary sources that cover various chemical compounds. The
following chart lists some of the major reference test methods for determining source
emissions:
Method1
MethodlA
-
Method2
-
Method5
Method6
Method7
Method8
Method18
Method24
Method306
-
Sample and velocity traverses for stationary sources.
Sample and velocity traverses for stationary sources with small
stacks.
Determinationof stack gas velocity and volumetric flow rate (Type
S pitot tube).
Determination of particulate emissions from stationary sources,
Determinationof sulfur dioxide emissions from stationary sources.
Determinationof nitrogen oxide emissions from stationary sources.
Determination of sulfuric acid mist and sulfur dioxide emissions
from stationary sources.
Measurement of gaseous organic compound emissions by gas
chromography
Determination of volatile matter content, water content, density,
volume solids, and weight solids of surface coatings.
Determination of Chromium Emissions from Decorative and Hard
Chromium Electroplating and Anodizing Operations.
What is important to realize when planning to demonstrate compliance to the new
National Emission Standard for Chromium Emissions from Hard and Decorative
Chromium Electroplating and Chromium Anodizing Tanks is that, in most plating-related
cases, the equipment used for Method 306 can also be used to determine other emissions
from the facility.
As the standard reads, facilities that are subject to the new Chromium MACT standard
have to identify all areas of the plant that produce chromium compound air emissions.
Emissions from every potential Chromium source must be quantified using either Method
306, 306A, or 306B. Therefore, as part of the planning activity, knowing which
operations are affected, and identifying the stacks that require sampling, are initial
requirements for compliance with the new regulation.
DRAFTING AND SUBMITTAL OF TESTING PROTOCOL
Performance tests shall be conducted and the data reduced in accordance with the
established test methods and procedures contained in 40 CFR 60 unless the
Administrator:
1)
2)
3)
4)
5)
Specifies or approves, in specific areas, the use of a reference method
with minor changes in methodology,
Approves the use of an equivalent method,
Approves the results of which he has determine to be adequate for
indicating whether a specific source is in compliance,
Waives the requirements for performance tests because the owner/operator
of a source has demonstrated by other means to the Administrator’s
satisfaction that the affected facility is in compliance with the standard,
Approves shorter sampling times and smaller sample volumes when
necessitated by process variables or other factors.
Performance tests will be conducted under such conditions as the Adminstrator shall
specify to the plant operator based on representative performance of the affected facility.
To assist the local authority in making this judgement, it is useful to develop a testing
protocol for the Administrator’s review and approval. The regulations are written to
make production records available to the Administrator, so it would be prudent to submit
the following information, proactively:
0
0
0
Brief process description(s)
Sampling location description(s)
Description of sampling and analytical procedures
Description of any modification made to the standard procedure
Any past or preliminary testing results (if available)
Previous estimating data
Proposed Quality Assurance Procedures
.-
..
SCHEDULE DEMONSTRATION ACCORDING TO PRODUCTION
The production manager must ensure that the compliance demonstration is not conducted
during off-periods, such as start-up, shutdown, and periods of equipment malfunction.
Any deviation from normal production activities whether it results in measured emissions
being either below or in excess of the level of the applicable emission limit will be
considered a violation of the applicable emission limit unless otherwise specified in the
applicable standard.
Unless otherwise specified by the Administrator, each performance demonstration will
consist of three separate test runs during the applicable test period. Each run shall be
conducted under the conditions specified in the Chromium MACT - Method 306
Standard. For the purpose of determining compliance with the standard, the arithmetic
means of results of the three runs shall apply. In the event that a sample is accidentally
lost or conditions occur in which one of the three runs must be discontinued because of
forced shutdown, failure of an irreplaceable portion of the sample train, or other
circumstances, beyond the operator’s or owner’s control, compliance may, upon the
Administrator’s approval, be determined using the arithmetic mean of the results of the
other two runs.
..
. .-.
SELECT AN APPROVED LABORATORY
Selection of a state-approved laboratory is an important part of the compliance
demonstration and the testing protocol. In the case of Method 306,the laboratory must
have experience with one of the three analytical techniques presented in the standard:
0
GFAAS
0
Inductively Coupled Plasma (ICP)
IC/PCR
0
Individual states will provide certifications to those laboratories that have passed a
number of Quality Control tests, using specific analytical techniques. Therefore!, when
looking for an approved laboratory for chrome sample analysis, check for their state
certification for the specific technique desired.
According to the standard, each technique has a specific analytical and in-stack
sensitivity:
Analvtical
ICP
7 pg Cr/L
0.0021 mg Cr/dscm
GFAAS
1 pg Cr/L
0.00015 mg Cr/dscm
ICPCR
0.05 pg Cr/L
O.oooO15 mg Cr/dscm
It will be mentioned later in the Analytical Procedure portion of this briefmg that if the
ICPCR analytical technique is chosen, it will be necessary to filter impinger contents
immediately after sampling. This extra sampling step may cause contamination of the
impinger volume.
MAKE SITE IMPROVEMENTS FOR SAMPLING
The owner or operator of an affected facility shall provide testing improvements as
follows:
0
Sampling ports adequate for test methods applicable to the facility
0
Suitable access for nozzle and probe is a 3" diameter port
0
Ensuring that the volumetric flow rates and pollutant emission rates can be
accurately determined by applicable test methods and procedures
0
Providing a stack or duct fie of cyclonic flow during performance tests.
0
Requires an assessment to install sample ports ideally at least 8 diameters
from the last obstruction and two diameters from the next obstruction.
0
Safe sampling platforms (Scaffolding, if required, according to OSHA
regulations).
0
Safe access to the sampling platform(s)
0
Utilities for sampling and testing equipment
ACQUIRE REAGENTS AND SAMPLING EQUIPMENT
Reagents
Unless otherwise indicated, all reagent must conform to the specifications established by
the Committee on Analytical Reagents of the American Chemical Society (ACSReagent
Grade). Where such specifications are not available, use the best available grade. If the
reagents used in the sampling effort are not normally used as part of normal production
activities, it is advisable to purchase only small amounts just for the sampling effort.
Water
Reagent water that conforms to ASTM Specification D1193-77, Type II. It is
recommended that water blanks be checked prior to preparing sampling reagents to
ensure that the Chrome content is less than the analytical detection limit.
ACTUAL SAMPLING EFFORT
1.
DETERMINE SAMPLING SITE
2.
DETERMINE TRAVERSE POINTS
3.
DETERMINATION OF STACK GAS VELOCITY
4.
DETERMINATION OF VOLUMETRIC FLOW L T E
5.
ISOKINETIC SAMPLING TRAIN
6.
VERTICAL AND HORIZONTAL TRAVERSING
7.
SAMPLE CASE SUPPORT SYSTEMS
8.
USE OF NOMOGRAPHS
9.
SAMPLE COLLECTION
-
10. DATA REDUCTION
DETERMINATION OF SAMPLING SITE AND TRAVERSE POINTS
A sampling site should be chosen where there are at least two diameters downstream and
0.5 diameters upstream from any expansion, contraction, bend, obstruction, visible
flame, or exit. This choice of location will require sampling at 48 different points within
the duct work as exhibited in Figure 1. The optimum location is at least eight diameters
downstream and two upstream, requiring only 12 sampling points. For rectangular cross
sections, determine an equivalent diameter from the following equation:
equivalent diameter = 2 flennth) (width1
length width
+-
To use Figure 1, first measure the distance from the chosen sampling location ot the
nearest upstream and downstream disturbances. Determine the correspondence number
of sampling points from each distance from Figure 1, and select the higher of the two
numbers. Then round off the number for a round stack to the nearest multiple of four.
For a square or rectangular stack, divide the cross section into as many rectangular areas
as sample points, such that the ratio of the length to width of each elemental area is
between one and two.
Figure 1:
Minimum Number of Sampling Points
m
0.3
20
.
r of Duct Dirmrterr Uprtrsrm.
(Mttlnca A)
1.0
-
1.3
2.0
2.3
-9/
OISNRBANCT
ir
U r r o f Duct Dfrmrtcrr DormstreuP
(Distmcr 8)
For circular stacks, locate the sampling points on at least two diameters according to
Figures 2 and 3. Make sure that the traverse axes divide the stack into equal parts. For
rectangular stacks, locate the points at the centroid of each equal area as shown in Figure
5.
Another factor to be considered in selecting the sampling site is that the distance from
the probe to the bottom of the sample case is about 13 inches. Although the equipment
was designed to fit into 2.5" holes, it has been found that 3" holes allow easier entrance
and removal without nicking the nozzle or picking up deposited dust.
Mini" number of traverse points for particulate traverses
Figure 2:
OUCT OIAMETERS U m R E A M FROY FLOW OlSlURUNCE'(01STAMCE A)
as
1'
I
I
1.8
1.K
I
I
28
I
I
I
1
I
WGnER WUM8ER II ;.I RECTAWGULARSTACKS OR DUCTS
..
,I
t8E10. tX?AWION. COIIRALTION. €TU
I
a
I
4
S
I
I
I
t
8
I
8
8
I
DUCT DIAUETERS OOWNSTREAM FROM FLOW OILTURMWCE' (OISTANCE U
Mini" number of traverse points for velocity (nonparticulate) traverses
Figure 3:
DUCT OIAMETERS UPSTREAM FROM FLOW OmUR8ANCE (OUTAWCE
I
a
'HIOHER N U Y I E R U FOR RECTANOULAR STACKS OR O U L n
.
18
'4
R A C K OIAMETER >a61 II24 id
?
12
!?
18-
-
I
STACK OIAMETER 8.30 1.10 0.81
8>
2
I
J
I
4
I
I
I
S
I
7
80RY
-
-
(12 Z4id
(I
8I
OUCT OIAMETERS OOWNSTAEAM FROM FLOW OIRUAIANCE IOISTANCE 8)
I
S
18
Table 1:
Location of sampling points in circular stacks
(% of stack diameter from inside wall to sampling point)
I
-
~~
1
NUMBER OF SAMPLlNG POINTS ON A D l A h i m R
2
4
6
a
10
12
14
16
ia
20
22
24
1
14.6
6.1
4.4
3.3
2.5
2.1
1.8
1.6
1.4
1.3
1.1
1.1
2
85.4
25.0
14.1
10.5
8.2
6.1
5.7
4.9
4.4
3.9
3.5
3.2
3
75.0
29.5
19.4
14.6
11.0
9.9
8.5
7.5
6.1
6.0
5.5
4
93.3
10.5
32.3
22.6
11.7
14.6
12.5
10.9
9.7
0.1
1.9
5
85.3
61.1
34.2
25.0
20.1
16.9
14.6
12.9
11.6
10.5
6
95.6
80.6
65.8
35.5
26.9
22.0
18.8
16.5
14.6
13.2
7
89.5
77.4
64.5
36.6
28.3
23.6
20.4
18.0
16.1
a
96.7
as.4
63.0
63.4
37.3
29.6
3.0
21.8
19.4
9
91.8
a23
73.1
62.5
38.2
30.6
26.1
23.0
IO
91.3
811.2
19.9
71.7
61.8
38.8
31.5
27.2
11
93.3
as.4
78.0
70.4
61.2
39.3
32.3
12
91.9
90.1
83.1
76.4
60.7
39.8
13
94.3
87.5
81.2
75.0
68.5
a.2
14
98.2
91.5
as.4
79.6
73.9
67.7
15
95.1
89.1
83.5
78.2
n.a
16
96.4
92.5
87.1
82.0
77.0
17
9S.6
90.3
8S.4
80.6
la
98.6
93.3
88.4
83.9
19
96.1
91.3
86.8
20
98.7
94.0
89.3
963
92.1
98.9
94.5
21
~
22
~
~
~~~~
1
69.4
.
~~
23
%.a
24
90.9
Figure 4:
Cross section of circular stack divided into 12 equal areas, showing location
of sampling points at centroid of each
TRAVERSE
mwi
' 1
2
3
4
8
I
OISTANCE.
m.14.4
14.8
29.0
70.4
ab4
SO8
.
Cross section of rectangular stack divided into 12 equal areas, with sampling
points at the centroid of each
Figure 5:
0
0
0
0
0
0
0
0
0
0
0
DETERMINATION OF STACK GAS VELOCITY AND VOLUMETRIC FLOW RATE!
The average gas velocity in a stack is determined from the gas density and from
measurement of the average velocity head with a Type S pitot tube.
The velocity head and temperature are measured at the traverse points specified by
Method 1. The static pressure is then measured (one reading will be adequate). The
stack gas dry molecular weight can be estimated using a value of 29.0.
The following figure can be used to collect the velocity traverse data:
PLANT:
STACK DIAMETER/DIMENSIONS, (in)
BAROMETRIC PRESSURE (in Hg)
CROSS SECTIONAL AREA
(m
OPERATORS
P l ” TUBE I.D. NO.
AVG. COEFFICIENT, C,
SCREMATIC OF
STACK CROSS
SECTION
LAST DATE CALIBRATED
Stack Temperature
Traverse
h i n t No.
Velocity Head,A Q
(in
H,O)
AVERAGE
5, “c
T,, “K
PI
(in.
Hg)
dAP
,
Determine the moisture content of the stack gas using either wet-bulb/dry-bulb procedure
or the condenser method. Determine the cross-sectional area of the stack at the sampling
location. When possible, physically measure the stack dimensions rather than using blue
prints.
Calculations
Nomenclature
A
BnJs
CP
-
-
K P =
(e).
Cross sectional area of the stack, m2
Water vapor in the gas stream, proportion by volume
Pitot tube coefficient, dimensionless
Pitot tube constant, use either:
34.97 d s e c
for the metric systems
85.49 ft/sec
for the english systems
Molecular weight of stack gas, dry basis g/g-mole (Ib/lb-mole)
Molecular weight of stack gas, wet basis g/g-mole (lb/lb-mole)
& (1 - B,J + 18.0 B,
Barometric pressure at measurement site, mm Hg (in. Hg)
Stack Static Pressure, mm Hg (in. Hg)
Absolute state gas pressure, mm Hg (in. Hg)
pbar + Po
Standard absolute pressure, 760 mm Hg (29.92 in. Hg)
Dry volumetric stack gas flow rate corrected to standard conditions,
dscdhr (dscflhr)
Stack temperature, "C ("F)
Absolute stack temperature, OK ("R)
273
t,
for metric
460 t,
for english
+
+
3,600 =
18.0
=
Standard absolute temperature, 293" K (528" R)
Average stack gas velocity d s e c (Wsec)
Velocity head of stack gas, mm H20 (in. H20)
Conversion factor, seconds/hour
Molecular weight of water, g/g-mole (lb/lb-mole)
Average Stack Gas Velocity
Average Stack Gas Drv Volumetric Flow Rate
-
Q,a = 3,600 (1 B,J UA T
,,
T
I (PIP)
P.
pstd
SAMPLING TRAIN
VERTICAL AND HORIZONTAL TRAVERSING
Simpling should always be done traversing perpendicular to the flow of the stack gases.
A vertical stack requires a horizontal traverse. A horizontal stack m a y be traversed
horizontally or vertically. For stacks or ducts at offset angles, horizontal traversing is
recommended. This is not always practical when sampling circular ducts.
'
The sampling device must not only meet the requirements of the least disturbance to
flow, but also allow accessibility of the sampling equipment.
In general, the following will hold true for most sampling projects:
1.
Horizontal traversing is the least complicated, requires only standard supporting
apparatus, and allows the easiest access to the probe and the sample case.
2.
Vertical down traversing is preferable to vertical up. Ports on the bottom side
of a stack may emit accumulated dust or particulate.
3.
Support of the sample case and probe during a vertical traverse is best
accomplished by a custom built frame for each sampling site.
SAMPLE CASE SUPPORT
Support of the sample case is of primary consideration in any stack sampling job. The
requirement for a traversing platform equal in length to the stack diameter has brought
about the development of several sample case support systems. The two most common
systems are the support rail, the monorail, and the slide box. Examples of the monorail
and the slide box are shown in the next two figures.
Figure 6:
Monorail support system
l / T Hole
I
T
314" Eye Bolt
H
SIDE VIEW
3"
I
Figure 7:
Slide box
Rooftop
optional.tray support 2' x 4"Iegs or a chain looped around the port.
NOMOGWHS
The correction factor nomograph (Figure g) and the operating nomograph (Figure h) have
been designed for use with the sampling train as aids for rapid isokinetic sampling rate
adjustment and to facilitate selection of a convenient nozzle size. Nomograph are
available from sampling train leasing companies.
'
Figure 8:
Nomograph for correction factor C
..
:**
40 -
fiND
c s 1.d
Figure 9:
Nomograph for sampling operation
C
4
I '
--
-0.e
-0.a
L o
&*e*
-*
01
*
0.
.
~
-'
. .
-1
,.I
'3 ....
IO.
'
. ,
. .
- 4
In order to determine the correction factor, C, on the nomograph, the following
information is required.
1.
Orifice calibration factor, H@
This is determined from a laboratory calibration and should be provided by the
vendor supplying the sampling equipment.
2.
Percent moisture, %H,O
This may be determined from a previous test or presurvey, or before the sample
run. The most common methods for determining moisture are the wet-bulb, drybulb technique and the condenser technique. The wet-bulb, dry-bulb technique
can be used when the stack temperature is less than 250"F, and the moisture is
less than 25% by volume. The velocity should be less than 35 feet per second.
The dry-bulb temperature is just the stack temperature, while the wet-bulb
temperature is determined as follows:
a)
Make a wet-bulb thermometer by fastening cloth wicking around the end
of a thermometer,
b)
Saturate the wicking with water,
c)
Insert the thermometer into the gas flow,and read it when it arrives at
equilibrium.
A psychrometric chart, shown in Figure i, is used to determine the moisture
chart. When the wet-bulb, dry-bulb technique can not be used, the condenser
technique is employed. Here a sample must be removed fiom the stack and
pulled through a condenser. The moisture content is determined by the amount
of gas sampled and the amount of water condensed. The water is normally
condensed using two impingers in an ice bath. The impingers should be partially
filled with water initially. The volume of liquid water collected is converted to
a volume of water vapor using the ideal gas law and the volume of gas sampled
is determined by using a dry gas meter after the gas pump. The following
equations can be used for the calculations:
where:
Volume of water vapor collected (standard conditions),
Final volume of impinger contents in ml
Initial volume of impinger contents in ml
Ideal gas constant, 21.83 inches Hg - cu ft/lb mole Ro
Density of water, 1 gram/ml
Absolute temperature at standard conditions, 530°R
Absolute pressure at standard conditions, 29.92 inches Hg
Molecular weight of water, 18 Ib/mole
where:
v
m
c
vm
pal
Pa
Td
Tm
=
=
=
=
=
=
Dry gas volume through meter at standard conditions, cubic feet
Dry gas volume measured by meter, cubic feet
Barometric pressure at the dry gas meter, inches of mercury
Pressure at standard conditions, 29.92 inches mercury
Absolute temperature at standard conditions 530" R
Absolute temperature at meter ("F 460), O R
+
Bwo
-
W
V
vwc
P
+ vmc
+
Bw, =
W
V
C
+
vwc
(0.025)
+ KIc
where:
Bwo
=
vwc
v,,
=
=
kill
=
Proportion by volume of water vapor in the gas stream,
dimensionless
Volume of water vapor collected (standard conditions), cubic feet
Dry gas volume through meter (standard conditions), cubic feet
Approximate volumetric proportion of water vapor in the gas
stream leaving the impingers, 0.025
Figure 10:
Psychrometric chart for air-water vapor mixtures at 1 ATM (29.92 inches
Hg)
3.
Meter Temperature, T,
Temperature at the meter rises above ambient temperature because of the pump
and can easily be read on the ten station LCD temperature readout.
4.
Stack Pressure, P,
"h,is is measured prior to the sample
run using a pitot tube and a manometer.
The sampling probe should be held with its opening at right angles to the flow.
If, due to space conditions, the sampling site is near the exit of the stack,
atmospheric pressure is used.
5.
Meter Pressure, P,,,
Same as atmospheric pressure.
To obtain correction factor, C (Figure h)
a to T, to obtain point "A" on reference line 1.
1.
Draw line from
2.
Draw line from point "A" to %H,O to obtain "B" on reference line 2.
3.
Draw line from point "B" to the calculated value P,/p, to obtain the
correction factor, C.
To select the nozzle size and set the K-factor on the operating nomograph,
the following information is first required:
1.
C factor. This is obtained using the above procedure.
2.
Stack temperature, T,. This is determined in O F by a rough temperature
traverse to within +/- 25" before the sample run.
3.
Average velocity pressure, AP. This .is determined by a rough
u
m
and maximum
preliminary pitot traverse, using the average of "
AP's in inches of water. A pitot tube and manometer are required,
measuring the AP'S at each of the traverse points chosen for sampling.
The pitot tube attached to the probe and manometer on the control console
can be used if separate ones are not available.
4.
Exact available nozzle sizes, D. This ih obtained from calibration of
available nozzles.
To select the nozzle size and to set the K-factor pivot point, use the following
procedure on the operating nomograph (Figure h)
1.
Set the correction factor, C, on the sliding scale to the reference mark,
"A".
2.
Align T, with average P, note probe tip diameter on the D scale, and
select the exact nozzle size closest to it.
3.
Align T,with the exact nozzle size selected and obtain a value on the AP
scale.
4.
Align the AP value with reference mark, "B", on the AH scale, and set
the K-factor pivot point.
SAMPLE COLLECTION
Sample Case Assembly
Once the sampling site has been improved to accommodate the sampling apparatus, the
sampling train can be assembled. The sample case should be properly fitted with the
impingers and glass connectors. For the Method 306 test, fill two of the impingers with
100 ml of 0.1 N NaOH or 0.1 N NaHCO, and one with 175 grams of silica gel. The
third impinger in the sampling train should remain empty, acting as a liquid trap. Weigh
and record the weight of all impingers for later justification of the isokinetic procedure.
Ensure that there is proper fit between the impingers and the impinger stems.
Figure 11:
Impinger configuration
I
Glassware screw joints require no silicone sealing grease. Firm, but not excessive,
pressure should be used when forming a seal. Generally, the screw joints need only be
loosened and the male glass stems pushed in place.
The use of a filter is not required by Method 306, however the filter holder will remain
as part of the sampling train. Tighten the phenolic screw joint around the filter holder
to prevent leakage around the silicon rubber gasket. Put the umbilical connector through
the holes in the handle bracket and into the last impinger arm. The sample case can now
be moved to the sampling site.
Figure 12:
-
Filter holder impinger connection
1 DOOR
2 MODULAR SAMPLE CASE
3 PROBESUPPORT
4 PROBE
5 PROBE HEATER CORD
6 MONORAIL HANDLEASSEMBLY
8 BALLJOIMCIAMP
9 RLTERASSEMBLY
10 FILTER-IMPINGER ELBOW
IA '~MPINGER
INTERCONNECT
'u=
12 IMPINGER OUTLET THERMOCOUPLE
13 UMBILICALADAPTER
14 IMPINGER CASE
15 CARRYINGHANDLE
16 DRAINCOCK
17 IMPINGER CASE SLIDE
18 SPAREACOUTLET
19 HEATER CONTROL KNOB
20 AMPHENOL RECEPTACLEAND COVER
21 FILTER BOX THERMOCOUPLE
22 PROBE HEATER OUTLET
23 CYCLONE BYPASS
Probe Assembly
After selecting a suitable probe length and quartz or borosilicate glass nozzle size, union
connections should be made as follows:
The probe utilizes a union for securing the nozzle to the probe sheath, as shown in
Figure 1. The nozzle is sealed with a front and rear ferrule and a union nut. The union
body is attached to the probe sheath by a permanently welded jut on the nozzle end of
the probe sheath. The probe liner is sealed inside the union body by an inverted rear
ferrule and a O-ring. The end of the probe liner should be flush with the front end of the
welded union nut. The union assembly should be detached from the probe sheath when
the probe liner is to be removed. Only hand tightening is necessary when the inverted
ferrule/o-ring/union body assembly is replaced. At no time should a wrench be used
while tightening the union body to the probe sheath. However, two wrenches are
necessary when tightening the nozzle nut to the union body. At stack temperatures over
500" F, it is suggested that asbestos string be utilized instead of the O-ring. It is
recommended that probe fittings of plastic such as Teflon, polypropylene, etc. be used
over metal fittings to prevent contamination.
The probe liner and nozzle should be covered with serum caps or saran wrap. Mark the
probe with heat resistant tape to denote the proper distance into the stack for each
sampling point.
Figure 13:
Probe Assembly
"Heater
'
1
Rear Ferrule
Nozzle
-
t
Power Cord
Quick Connects
P i t o t .Tube
Figure 14:
Chromium Sampling Train
Leak Test
Connect the umbilical cord to the sample case and control console. The sampling train
should now be completely assembled. Plug the probe nozzle with a cork stopper. Turn
the Coarse Adjust valve on the Monitoring Unit to the ON position and fully open the
Fine Adjust valve. Turn on the pump. Partially close the Fine Adjust valve until the
vacuum gauge measures 10 inches of mercury (Hg). Then check the dry gas meter flow.
If the flow through the dry gas meter exceeds 0.02 ft?/minute at 10 inches of mercury
gauge pressure, or $% of the average sampling rate, whichever is less, the leak or leaks
must be found or corrected. If the union assembly has been sealed with asbestos string
(recommended at high temperatures), it should be checked first if any leaks do exist.
When the check is complete, first remove the cork stopper from the inlet to the probe
nozzle, and then turn off the vacuum pump. This will prevent water from being forced
backward through the impingers into the fdter housing.
Sample Run
Record all necessary initial data shown on Table 2, including the initial dry gas meter
reading. Turn off the Coarse Adjust valve on the Monitor unit and fully open the Fine
Adjust valve. Remove the cover from the nozzle and place the probe at the first
sampling point. Record the clock time, read AP on the pitot tube gauge, and determine
AHfrom the operating nomograph.
Turn the pump on. Start the test timer while setting AHon the orifice gauge first be
adjusting the Coarse Adjust valve and then the Fine Adjust valve.
The sample run plan should consider the number of traverse points and the sampling time
at each point. Generally the length of sampling time at each traverse point is 2 to 10
minutes.
During the sampling traverse, the probe is moved from point to point without turning the
pump off except when changing to a different sampling port. As long as the probe
nozzle is within the stack, the pump should be on. Do not introduce fugitive emissions
to the sampling effort be leaving the pump on when changing sampling ports. The AP
should be monitored and adjustments made on the o,rifice gauge with the aid of the
nomograph, a calculator, or laptop computer. Besides the regular time interval
recordings, a set of readings should be recorded when the AP changes by more than 20
percent.
When testing has been completed, close the sampling valve, remove the probe from the
stack, and take a final set of readings. Take care to prevent impinger contents from
backing up into the fdter housing. Turn off the Filter Heat, the Probe Heat controls and
remove the probe from the sampling port. Cover the sampling nozzle as soon as it is
cool enough in order to prevent contamination or loss of sample.
Sample Recovery
Sample Storape Containers
Polyethylene, with leak-free screw cap, 500 ml or 1,OOO ml.
&"le
Recoverv Solution
Use 0.1 N NaOH or 0.1 N NaHCO,, whichever was used as the impinger absorbing
solution in place of acetone to recover the sample.
Probe-Liner and Probe-Nozzle Brushes
Brushes are not necessary for sample recovery. If a probe brush is used, it must be
nonmetallic.
Taking care to see that dust on the outside of the probe or other exterior surface does not
get into the sample, quantitatively recover any condensate from the probe nozzle, probe
fitting, probe liner, and the filter holder by washing the components with sample
recovery solution and placing the wash in a glass container (CONTAINER NO. 1).
Carefully remove the glass probe nozzle and clean the inside surface by rinsing with
sample recovery solution from a wash bottle and placing the wash in CONTAINER NO.
1.
Rinse the probe liner with the sample recovery solution, taking care to thoroughly wet
the entire inside diameter with solution to catch any condensate that may be within the
probe.
ImDinger SamDle Recoverv
Weigh all impingers used during the isokinetic sampling effort and record the weights
for later analysis.
Measure the volume of liquid in the first, second, and third impingers and quantitatively
transfer into CONTAINER NO. 1. Use approximately 200 to 300 ml of 0.1 N NaOH
or 0.1 N NaHCO, to rinse the probe nozzle, probe liner, three impingers, and connecting
glassware. Add this rinse to the same container.
In a separate container, CONTAINER NO. 2, place approximately 500 ml of 0.1 N
NaOH or 0.1 N NaHCO, absorbing solution in a labeled sample container.
If the laboratory selected to perform the analysis will be using IC/PCR to analyze for
CP6, it must be filtered immediately following recovery to remove any insoluble matter.
Nitrogen gas may be used as a pressure assist to the filtration process. Filter the entire
contents of the impinger collection container through a 0.45pm acetate filter, and collect
the filtrate in a 1,OOO ml graduated cylinder. Rinse the sample container with reagent
water three separate times, pass these rinses through the filter, and add the rinses to the
sample filtrate. Determine and record the f d volume of the filtrate and rinses and
return them to the rinsed polyethylene sample container for traosport to the laboratory.
Chain of Custody
Creation and maintenance of the evidence trail is necessary when compliance
demonstrations are performed. A detailed plan to ensure sample integrity should be part
of the sampling protocol submitted to the local authority prior to the isokinetic sampling
effort. At a minimum, the chain-ofcustody (CoC) procedure should involve sealing the
sample container with CoC tape, signed by the recovery chemist; and detailing sampling
time and date on an accompanying form. An example of a chain of custody form is
exhibited on the next page:
-
Chain of Custody #GO1
CHAIN OF CU!jTODY RECORD
SAMPLES SENT B Y
CONTACT:
TELEPHONE:
Integral Compliance S y n w ~Inc
,
Kevin Walk
(410) 515-3248
JOB REFERENCE:
COX CREEK REFINERY
ELECTROLYTIC TANK HOUSE
JOB LOCATION:
.***..*.
(410) 836-8685
JOB NAME
MULTIPLE METALS SAMPLING
~.~***~****.****oo*ooo****oo*o*o*o***o*oo*o*oo****oo**o*ooo*****o****oo****o*o*****~~*~w~***oo*******o***oo*o****o**
CLIENT SAMPLE NUMBER
1
SAMPLER:
FAX:
LAB I.D.
SAMPLING
DATYnME
SAMPLE
MATRIX
M-1
i600.3~1
FILTER
M-2
zoo0.3~1
FILTER
METALS
M-3
oi00.31~2
FILTER
METALS
M-1 A
1600.3~1
LIQUID
450 ml
MmALS
M-2 A
M-2
moo. 3/21
LIQUID
350 ml
METALS
M-3A
0100.3M
LIQUID
300 ml
METALS
.IN HNO,BLANK
0100,3N
LIQUID
METALS
KMNO,/H,SO, BLANK
0100.3M
LIQUID
MERCURY
8N HCI BLANK
0100.3N
LIQUID
MERCURY
HNOJH/HABLANK
0100.3M
LIQUID
METALS
M-1 PROBE RINSE
1600,3/21
LIQUID
M-2 PROBE RINSE
2000. 3/21
LIQUID
0100.3N
M-1 C
AMOUNT OF
SAMPLE
-
ANALYSIS
REQUESIED/REMARKS
METALS
36 ml
METALS
34ml
METAU
LIQUID
41 ml
METALS
METALS
1600.3ni
LIQUID
200 ml
MERCURY
M-2 C
2000. 3/21
LIQUID
“
I
MERCURY
M-3C
0100.3m
LIQUID
150 ml
MERCURY
M-3 PROBE RINSE
I
I
COMMENTS: SAMPLE METALS FOR pb, Zn. Cu. Bi. Ni. Cr. Se. Cd, As, R. Sb, AND Cr+‘.
Method 5 sample train utilized for air sampiing.
Permanganate sample (M-1 C. M-2 C, M-3 C) is for Mercury only.
METHOD
OF
PRESERVATION:
RELINQVrmn, BY:
RECEIVED B Y
Maintained
at
34-37‘F
DATE:
DATE:-
on
TIME:
TIME:
ice
at
sampling
SHIPPED B Y
and
sample
retrieval.
ISOI(IMETIC SAMPLING ANALYSIS EQUATIONS
ISOKINETIC SAMPLING ANALYSIS EQUATIONS
NOMENCLATURE
Isokinetic Sampling Rate (%)
Sample Time (minutes)
Total volume of liquid collected in impinger train (mL)
Volume of water vapor collected, corre~tedfor standard conditions (scf)
Volume of gas sampled through dry gas meter (dscf)
Proportion of water vapor in the gas stream (96)
Constant, 85.48 ft/sec * l(lb/Ib*moIe) (in HE)*
NOR)(in H m H
Pitot Tube Coefficient (Dimensionless)
Velocity head of stack gas (ii H20)
Average stack temperature (OR)
Absolute stack pressure (imches Hg)
Dry molecular weight of stack gas (lbfib * mole)
Stack gas velocity ($s)
Average dry gas meter temperature (OR)
Barometric Pressure (iiches Hg)
Pressure differential across the orifice meter (inches HzO)
Volume of gas sampled by the dry gas meter, conected to standard
conditions (scf)
VoIumetric flow rate corre~tedfor standard conditions (dscf7hr)
Conversion factor (2.205 E-9Ib/pg)
Concentration of constituent in stack gas, dry basis, corrected for standard
conditions (lb/dscf)
Mass Emission rate (lb/hr)
Meter pressure (inches Hg)
Static pressure (inches H20)
Dry gas meter coefficient (dimensionless)
Stack Area (ft3
Nozzle Area
Constant (sed (OR)
(hr) (inches Hg)
(e
ISOKINETIC SAMPLING ANALYSIS EQUATIONS
Preliminary Calculations (Meter pressure, stack pressure)
1.
P,,, = phr + n w i 3 . 6
P, = P b + PJ13.6
2.
Comted Dry Gas Volume Sampled
v
l
n
,
=
T,
3.
Volume of Water Vapor Collected
4.
Moisture Content in Stack Gas
5.
Average Stack VeIocity
6.
7.
.
Corrected Stack Gas Flow Rate
96 hkinetic
I
=
8.
Constituent Concentration
9.
MassEmissionRate
MER
=
1.667 T,[O.O0267 V b
ci*
+ V- *
4,
* PJ
-
Method 306
Determination of Chromium missions
from Decorative and Hard Chromium
Electroplating and Anodizing Operations
1. Applicability and Principle
1.1
Applicability. This method applies to the
determination of chromium (Cr) in emissions from decorative and
hard chrome electroplating facilities and anodizing operations.
1.2 Principle. A sample is extracted isokinetically from
the source using an unheated Method 5 sampling train (40 CFR Part
60, Appendix A ) , with a glass nozzle and probe liner, but with
the filter omitted. The Cr emissions are collected in an
alkaline solution: 0.1 N sodium hydroxide (NaOH) or 0.1 N sodium
bicarbonate (NaHC03). The collected samples remain in the
alkaline solution until analysis. Samples with high Cr
concentrations may be analyzed using inductively coupled plasma
emission spectrometry (ICP) at 267.72 nm. Alternatively, if
improved detection limits are required, a portion of the alkaline
impinger solution is digested with nitric acid and analyzed by
graphite furnace atomic absorption spectroscopy (GFAAS) at
357.9 nm.
If it is desirable to determine hexavalent chromium (Cr+6)
emissions, the samples may be analyzed using an ion chromatograph
equipped with a post-column reactor (IC/PCR) and a visible
wavelength detector. To increase sensitivity for trace levels of
Cr+6, a preconcentration system can be used in conjunction with
the IC/PCR.
2.
Range, Sensitivity, Precision, and Interferences
2.1 Range. The recommended analytical range for each of
the three analytical techniques is given below. The upper limit
of all three techniques can be extended indefinitely by
appropriate dilution.
2.1.1
GFAA8 RANGE. A s reported in Method 7191 of SW-846
(Citation 5 in Bibliography), the optimum concentration range for
GFAAS is 5 to 100 pg Cr/l of concentrated analyte.
ICP Range. A linear response curve for ICP can be
2.1.2
obtained in the range of 10 to at least 500 pg Cr/l of absorbing
solution.
2.1.3
IC/PCR Range. IN EPA Method Cr+6 (40
CFR Part 266, Appendix IX) the lower limit of the detection range
for IC/PCR when employing a preconcentration procedure is
reported to be about 0.1 pg Cr+6/1 of absorbing solution.
2.2
Sensitivity
2.2.1
Analytical sensitivity.
2.2.1.1
ICP Analytical Sensitivity. The minimum detection
limit for ICP, as reported in Method 6 0 1 0 A of SW-846, is 7 1-14
Cr/l.
1
2.2.1.2
GFAAS Analytical Sensitivity. The minimum
detection limit for GFAAS, as reported in Method 7 1 9 1 of SW-846,
is 1 pg Cr/l.
2.2.1.3
IC/PCR Analytical Sensitivity. The minimum
detection limit for IC/PCR with a preconcentrator, as reported in
Method Cr+6, is 0 . 0 5 bq Cr+6/1.
2.2.2
In-stack 'Sensitivity. The in-stack sensitivity
depends upon the analytical detection limit, the volume of stack
gas sampled, and the total volume of the impinger absorbing
solution plus the rinses. Using the analytical detection limits
given in Sections 2 . 2 . 1 . 1 , 2 . 2 . 1 . 2 , and 2 . 2 . 1 . 3 ; a stack gas
sample volume of 1.7 dscm; and a total liquid sample volume of
5 0 0 ml; the corresponding in-stack detection limits are 0.0021 mg
Cr/dscm for ICP, 0 . 0 0 0 1 5 mg Cr/dscm for GFAAS, and 0 . 0 0 0 0 1 5 mg
Cr+6/dscm for IC/PCR with preconcentration. However, it is
recommended that the concentration of Cr in the analytical
solutions be at least five times the analytical detection limit
to optimize sensitivity in the analyses. Using this guideline
and the same assumptions for impinger sample volume and stack gas
sample volume ( 5 0 0 ml and 1 . 7 dscm, respectively), the
recommended minimum stack concentrations for optimum sensitivity
are 0 . 0 1 0 3 mg Cr/dscm for ICP, 0 . 0 0 0 7 4 mg Cr/dscm for GFAAS, and
0 . 0 0 0 0 7 4 mg Cr+6/dscm for IC/PCR with preconcentration. If
required, the in-stack detection limits can be improved by either
increasing the stack gas sample volume, reducing the volume of
the digested sample for GFAAS, improving the analytical detection
limits, or any combination of the three.
s
2.3
Precision. The following precision data have been
reported for the three analytical methods. In the case of the
GFAAS there is also bias data. In all cases, when sampling
precision is combined with analytical precision, the resulting
overall precision may be lower.
2.3.1
GFAAS Precision. A s reported in Method 7 1 9 1 of SW8 4 6 , in a single laboratory (EMSL), using Cincinnati, Ohio tap
water spiked at concentrations of 1 9 , 4 8 , and 77 pg Cr/l, the
standard deviations were fO.l, f0.2, and f0.8, respectively.
Recoveries at these levels were 9 7 % , 101%, and l 0 2 % ,
respectively.
2.3.2
ICP Precision. A s reported in Method 6010A of SW8 4 6 , in an EPA round-robin Phase 1 study, seven laboratories
applied the ICP technique to acid/distilled water matrices that
had been spiked with various metal concentrates. For true values
of 10, 50, and 150 pg Cr/l; the mean reported values were 1 0 , 5 0 ,
and 1 4 9 pg Cr/l; and the mean percent relative standard
deviations were 1 8 , 3 . 3 , and 3 . 8 % , respectively.
2.3.3
IC/PCR Precision. A s reported in Method Cr+6, the
precision of the IC/PCR with sample preconcentration is 5 to 10
%; the overall precision for sewage sludge incinerators emitting
120 ng/dscm of Cr+6 and 3 . 5 pg/dscm of total Cr is 25% and 9% for
Cr+6 and total Cr, respectively; and for hazardous waste
incinerators emitting 300 ng/dscm of Cr+6 the precision is 2 0 %.
2.4
Interferences.
2.4.1
GFAAS Interferences.
Low concentrations of calcium
and/or phosphate may cause interferences; at concentrations above
200 pg/l, calcium's effect is constant and eliminates the effect
of phosphate. Calcium nitrate is therefore added to the
concentrated analyte to ensure a known constant effect. Other
matrix modifiers recommended by the instrument manufacturer may
also be suitable. Nitrogen should not be used as the purge gas
due to cyanide band interference. Background correction may be
required because of possible significant levels of nonspecific
absorption and scattering at the 357.9 nm analytical wavelength.
Zeeman or Smith-Hieftje background correction is recommended to
correct for interferences due to high levels of dissolved solids
in the alkaline impinger solutions.
2.4-2
2.4.2.1
ICP Interferences.
ICP Spectral Interferences.
Spectral interferences
are caused by: (1) overlap of a spectral line from another
element; (2) unresolved overlap of molecular band spectra;
(3) background contribution from continuous or recombination
phenomena; and (4) stray light from the line emission of highconcentration elements. Spectral overlap may be compensated for
by computer correcting the raw data after monitoring and
measuring the interfering element. At the 267.72-nm Cr
analytical wavelength, iron, manganese, and uranium are potential
interfering elements. Background and stray light interferences
can usually be compensated for by a background correction
adjacent to the analytical line. Unresolved overlap requires the
selection of an alternative Cr wavelength. Consult the
instrument manufacturer's operation manual for interference
correction procedures.
2.4.2.2
ICP Physical Interferences, High levels of
dissolved solids in the samples may cause significant
inaccuracies due to salt buildup at the nebulizer and torch tips.
This problem can be controlled by diluting the sample or
providing for extended rinse times between sample analyses.
Standards are prepared in the same matrix as the samples (i.e.,
0.1 N NaOH or 0.1 N NaHC03)
2.4.2.3
ICP Chemical Interferences. These include
molecular compound formation, ionization effects and solute
vaporization effects, and are usually not significant in ICP,
especially if the standards and samples are matrix matched.
2.4.3
IC/PCR Interferences. Components in the sample
matrix ma cause Cr+6 to convert to trivalent chromium (Cr+3) Or
cause Cr+'
to convert to Cr+6. The chromatographic separation of
Cr+6 using ion chromatography reduces the potential for other
metals to interfere with the postcolumn reactim. For the IC/PCR
analysis, only compounds that coelute with Cr
and affect the
diphenylcarbazide reaction will cause interference. Periodic
analyses of reagent water blanks are used to demonstrate that the
analytical system is essentially free of contamination. Sample
cross-contamination that can occur when high-level and low-level
samples or standards are analyzed alternately is eliminated by
thorough purging of the sample loop. Purging can easily be
.
3
achieved by increasing the injection volume of the samples t o ten
times the size of the sample loop.
3.
Apparatus
3.1
Sampling Train. A schematic of the sampling train used
in this method is shown in Figure 306-1. The train is the same
as Method 5, Section 2.1, except that the filter is omitted, and
quartz or borosilicate glass must be used for the probe nozzle
and liner in place of stainless steel. It is not necessary to
heat the probe liner. Probe fittings of plastic such a s Teflon,
polypropylene, etc. are recommended over metal fittings to
prevent contamination. If desired, a single combined probe
nozzle and liner may be used, but such a single glass piece is
not a requirement of this methodology. Use 0.1 N NaOH or 0.1 N
NaHC03 in the impingers in place of water.
3.2
Sample Recovery. Same as Method 5, Section 2.2, with
the following exceptions:
3.2.1
Probe-Liner and Probe-Nozzle Brushes. Brushes are
not necessary for sample recovery. If a probe brush is used, it
must be non-metallic.
3.2.2
Sample Recovery Solution. Use 0.1 N NaOH or 0.1 N
NaHC03, whichever was used as the impinger absorbing solution, in
place of acetone to recover the sample.
3.2.3
Sample Storage Containers. Polyethylene, with leakfree screw cap, 500 ml or 1,000 ml.
3.2.4
Filtration Apparatus for IC/PCR. Teflon, or
equivalent, filter holder and 0.45 pm acetate, or equivalent,
filter.
3.3
Analysis. For analysis, the following equipment is
needed.
3.3.1
General.
3.3.1.1
Phillips Beakers.
(Phillips beakers are preferred,
but regular beakers can also be used.)
3.3.1.2
Hot Plate.
3.3.1.3
Volumetric Flasks. Class A, various sizes as
appropriate.
3.3.1.4
Assorted Pipettes.
3.3.2
Analysis by GFAAS.
Chromium Hollow Cathode Lamp or Electrodeless
3.3.2.1
Discharge Lamp.
3.3.2.2
Graphite Furnace Atomic Absorption
Spectrophotometer.
3.3.3
Analysis by ICP.
3.3.3.1
ICP Spectrometer. Computer-controlled emission
spectrometer with background correction and radio frequency
generator.
3.3.3.2
Argon G a s Supply. Welding grade or better.
3.3.4
Analysis by IC/PCR
4
Type s
Pitot Tube
lmpinger Train
I
Figure 306-1. Chromium Sampling Train.
3.3.4.1
IC/PCR System. High performance liquid
zhromatograph pump, sample injection valve, post-column reagent
Aelivery and mixing system, and a visible detector, capable of
3perating at 520 mu, all with a non-metallic (or inert) flow
path. An electronic peak area mode is recommended, but other
recording devices and integration techniques are acceptable
provided the repeatability criteria and the linearity criteria
€or the calibration curve described in Section 6.4.1 can be
satisfied. A sample loading system will be required if
preconcentration is employed.
3.3.4.2
Analytical Column. A high performance ion
zhromatograph (HPIC) non-metallic column with anion separation
zharacteristics and a high loading capacity designed for
separation of metal chelating compounds to prevent metal
interference. Resolution described in Section 5.5 must be
Dbtained. A non-metallic guard column with the same ion-exchange
naterial is recommended.
3.3.4.3
Preconcentration Column. An HPIC non-metallic
zolumn with acceptable anion retention characteristics and sample
Loading rates as described in Section 5.5.
3.3.4.4
0.45-pm Filter Cartridge.
For the removal of
insoluble material. To be used just prior to sample
injection/analysis.
P.
Reagents
Unless otherwise indicated, all reagents shall conform to
the specifications established by the Committee on Analytical
Zeagents of the American Chemical Society (ACS reagent grade).
Vhere such specifications are not available, use the best
wailable grade.
4.1
Sampling.
4.1.1
Water. Reagent water that conforms to
4STM Specification D1193-77, Type I1 (incorporated by reference).
It is recommended that water blanks be checked prior to preparing
sampling reagents to ensure that the Cr content is less than the
nnalytical detection limit.
4.1.2
Sodium Hydroxide (NaOH) Absorbing Solution, 0 . 1 N or
godium Bicarbonate (NaECO,!
Absorbing Solution, 0.1 N. Dissolve
g.0 g of sodium hydroxide in 1 1 of water, or dissolve 8.5 g of
sodium bicarbonate in 1 1 of water.
4.2
Sample Recovery.
4.2.1
0 . 1 N NaOH or 0 . 1 N NaHCO,.
See Section 4.1.2.
Use
the same solution for recovery as was used in the impingers.
4.2.2
pH Indicator Strip, for IC/PCR. pH indicator capable
sf determining the pH of solutions between the pH range of 7 and
12, at 0.5 pH intervals,
4.3
Sample Preparation and Analysis.
4.3.1
Nitric Acid (HNo3), Concentrated, for GFAAS. Trace
netals grade or better HN03 must be used for reagent preparation.
4CS reagent grade HNO, is acceptable for cleaning glassware.
6
4.3.2
1 . 0 % ( v / v ) , for GFAAS.
Add, with stirring, 10
m l of concentrated HNO, to 800 ml of water. Dilute to 1,000 ml
with water. This reagent shall contain less than 0.001 mg Cr/l.
4.3.3
Calcium Nitrate Ca(N03),2 Solution (10 pg Ca/ml) for
GFAAS.
Prepare the solution by weighing 36 mg of Ca(N03)2 into a
1 1 volumetric flask. Dilute with water to 1 1.
4.3.4
Matrix Modifier, for G F M S . See instrument
manufacturer’s manual for suggested matrix modifier.
4.3.5
Chromatographic Eluent, for IC/PCR. The eluent used
in the analytical system is ammonium sulfate based. Prepare by
adding 6.5 ml of 29% ammonium hydroxide (NH,OH) and 33 g of
ammonium sulfate ((NH4)2S04)to 500 ml of reagent water. Dilute
to 1 1 with reagent water and mix well. Other combinations of
eluents and/or columns may be employed provided peak resolution,
as described in Section 5.5, repeatability and linearity, as
described in Section 6.4.1, and analytical sensitivity are
acceptable.
4.3.6
Post-Column Reagent, for IC/PCR. An effective postcolumn reagent for use with the chromatographic eluent described
in Section 4.3.5 is a diphenylcarbazide (DPC)-based system.
Dissolve 0.5 g of 1,5-diphenylcarbazide in 100 ml of ACS grade
methanol. Add 500 ml of reagent water containing 50 ml of 96%
spectrophotometric grade sulfuric acid. Dilute to 1 1 with
reagent water.
4.3.7
Chromium Standard Stock Solution (1000 mg/l)- Procure
a certified aqueous standard or dissolve 2.829 g of potassium
dichromate (K2CrZ0,,) in water and dilute to 1 1.
4.3.8
Calibration Standards for GFAAS. Chromium solutions
for GFAAS calibration shall be prepared to contain 1.0% (v/v)
Calibration
HNO,.
The zero standard shall be 1.0% (v/v) HNO,.
standards should be prepared daily by diluting the Cr standard
Use at least four
stock solution (Section 4.3.7) with 1.0% HNO,.
standards to make the calibration curve. Suggested levels are 0 ,
5, 50, and 100 pg Cr/l.
4.3.9
Calibration Standards for ICP or IC/PCR. Prepare
calibration standards for ICP or IC/PCR by diluting the Cr
standard stock solution (Section 4.3.7) with 0.1 N NaOH or 0.1 N
NaHCO,, whichever was used as the impinger absorbing solution, to
achieve a matrix similar to the actual field samples. Suggested
levels are 0, 25, 50, and 100 pg Cr/l for ICP, and 0, 0.5, 5, and
10 pg Crt6/1 for IC/PCR.
4.4
Glassware Cleaning Reagents.
4.4.1
ENO,, Concentrated. ACS reagent grade or equivalent.
4.4.2
Water. Reagent water that conforms to ASTM
Specification D1193-77, Type 11, (incorporated by reference).
4.4.3
ENOg, 10% ( v / v ) .
Add with stirring 500 ml of
concentrated HNO, to a flask containing approximately 4000 ml of
water. Dilute to 5000 ml with water. Mix well. The reagent
shall contain less than 2 pg Cr/l.
5.
Procedure
7
5.1
Sampling. Same as Method 5, Section 4.1, except omit
the filter and filter holder from the sampling train, use a glass
nozzle and probe liner, do not heat the probe, place 100 ml of
0.1 N NaOH or 0.1 N NaHC03 in each of the first two impingers,
and record the data for each run on a data sheet such as the one
shown in Figure 306-2.
Clean all glassware prior to sampling in hot soapy water
designed for laboratory cleaning of glassware. Next, rinse the
glassware three times with tap water, followed by three
additional rinses with reagent water. Then soak all glassware in
10% (v/v) HN03 solution for a minimum of 4 hours, rinse three
times with reagent water, and allowed to air dry. Cover all
glassware openings where contamination can occur with Parafilm,
or equivalent, until the sampling train is assembled for,
sampling.
If the sample is going to be analyzed for Cr+6 using IC/PCR,
determine the pH of the solution in the first impinger at the end
of the sampling run using a pH indicator strip. The pH of the
solution should be greater than 8.5. If not, the concentration
of the NaOH or NaHC03 impinger absorbing solution should be
increased to 0.5 N and the sample should be rerun.
5.2
Sample Recovery. Follow the basic procedures of Method
5, Section 4.2, with the exceptions noted below; a filter is not
recovered from this train.
5.2.1
Container No. 1. Measure the volume of the liquid in
the first, second, and third impingers and quantitatively
transfer into a labelled sample container. Use approximately 200
to 300 ml of 0.1 N NaOH or 0.1 N NaHC03 to rinse the probe
nozzle, probe liner, three impingers, and connecting glassware;
add this rinse to the same container.
5.2.2
Container No. 2 (Reagent Blank). Place approximately
500 ml of 0.1 N NaOH or 0.1 N NaHCO, absorbing solution in a
labeled sample container.
If the sample is to be
Sam le Filtration for IC/PCR.
5.2.3
analyzed for CrR by IC/PCR, it must be filtered immediately
following recovery to remove any insoluble matter. Nitrogen gas
may be used as a pressure assist to the filtration process.
Filter the entire contents of Container No. 1 through a 0.45-pm
acetate filter (or equivalent), and collect the filtrate in a
1,000 ml graduated cylinder. Rinse the sample container with
reagent water three separate times, pass these rinses through the
filter, and add the rinses to the sample filtrate. Determine the
final volume of the filtrate and rinses and return them to the
rinsed polyethylene sample container.
5.2.4
sample Preservation. Refrigerate samples upon
receipt. (Containers Nos. 1 and 2).
8
Plant
Location
Operator
Ambient temperature
Barometric pressure
Assumed moisture, %
Probe length, (ft.)
Nozzle identification No
Average calibrated nozzle diameter, (in.)
Leak rate,
. i fm
)fc(
Static pressure, (in. Hg)
Run No.
Sample box No.
Meter box No.
Meter AH@
C factor
Pitot tube coefficient, C,
SCHEMATIC OF STACK CROSS SECTION
Traverse
point
number
Sampling
time
min.
Stack
Pressure
Gas meter
Vacuum temperature Velocity head differential across reading
orifice meter
(in. Hg)
(T,) ( O F )
(Ac) (in. H20)
(in. H p )
(W
Gas sample temp.
at dW gas meter
Inlet
Outlet
(OF)
(OF)
Total
Average
I
Avg .
Figure 306-2. Chromium Field Data Sheet.
'
Temperature of
gas leaving
condenser or last
impinger
(OF)
5.3
Sample P r e p a r a t i o n and Analysis for GFAAS. For
analysis by GFAAS, an acid digestion of the alkaline impinger
solution is required. Two types of blanks are required for the
analysis. The calibration blank is used in establishing the
analytical curve, and the reagent blank is used to assess
possible contamination resulting from the sample processing. The
1.0% HNO, is the calibration blank. The 0.1 N NaOH solution or
the 0.1 N NaHC0, from Section 5.2.2 is the reagent blank. The
reagent blank must be carried through the complete analytical
procedure, including the acid digestion, and must contain the
same acid concentration in the final solution as the sample
solutions.
5 . 3 . 1 A c i d Digestion for GFAAS.
In a beaker, add 10 ml of
concentrated HNO, to a sample aliquot of 100 ml taken for
analysis. Cover the beaker with a watch glass. Place the beaker
on a hot plate and reflux the sample down to near dryness. Add
another 5 ml of concentrated HNO, to complete the digestion.
Carefully reflux the sample volume down to near dryness. Wash
down the beaker walls and watch glass with reagent water. The
final concentration of HNO, in the solution should be 1% (v/v).
Transfer the digested sample to a 50 ml volumetric flask. Add
0.5 ml of concentrated HNO,, and 1 ml of the 10 pg/ml of
Ca(N03)2. Dilute to 50 ml with reagent water. A different final
volume may be used, based on the expected Cr concentration, but
the HNO, concentration must be maintained at 1% (v/v).
5.3.2
Sample Analysis by GFAAS. The 357.9-nm wavelength
line shall be used. Follow the manufacturer's operating
instructions for all other spectrophotometer parameters.
Furnace parameters suggested by the manufacturer should be
employed as guidelines. Since temperature-sensing mechanisms and
temperature controllers can vary between instruments and/or with
time, the validity of the furnace parameters must be periodically
confirmed by systematically altering the furnace parameters while
analyzing a standard. In this manner, losses of analyte due to
higher-than-necessary temperature settings or losses in
sensitivity due to less than optimum settings can be minimized.
Similar verification of furnace parameters may be required for
complex sample matrices. Calibrate the GFAAS system following
the procedures specified in Section 6.
Inject a measured aliquot of digested sample into the
furnace and atomize. If the concentration found exceeds the
calibration range, the sample should be diluted with the
calibration blank solution (1.0% HN03) and reanalyzed. Consult
the operator's manual for suggested injection volumes. The use
of multiple injections can improve accuracy and help detect
furnace pipetting errors.
Analyze a minimum of one matrix-matched reagent blank per
sample batch to determine if contamination or any memory effects
are occurring. Analyze a calibration blank and a midpoint
calibration check standard after approximately every 10 sample
inject ions
.
10
Calculate the Cr concentrations (1) by the method of
standard additions (see operator's manual), (2) from the
calibration curve, or (3) directly from the instrument's
concentration readout. All dilution or concentration factors
must be taken into account. All results should be reported in
pg Cr/ml with up to three significant figures.
5.4
Sample Analysis by ICP. The ICP measurement is
performed directly on the alkaline impinger solution; acid
digestion is not necessary provided the samples and standards are
matrix matched. However, ICP should only be used when the
solution analyzed has a Cr concentration greater than 35 pg/l.
Two types of blanks are required for the analysis. The
calibration blank is used in establishing the analytical curve,
and the reagent blank is used to assess possible contamination
resulting from sample processing. Use either 0.1 N NaOH or 0.1 N
NaHC03, whichever was used for the impinger absorbing solution,
for the calibration blank. The calibration blank can be prepared
fresh in the laboratory; it does not have to be from the same
batch of solution that was used in the field. Prepare a
sufficient quantity to flush the system between standards and
samples. The reagent blank (Section 5 . 2 . 2 ) is a sample of the
impinger solution used for sample collection that is collected in
the field during the testing program.
Set up the instrument with proper operating parameters
including wavelength, background correction settings (if
necessary), and interfering element correction settings (if
necessary). The instrument must be allowed to become thermally
stable before beginning performance of measurements (usually
requiring at least 3 0 min of operation prior to calibration).
During this warmup period, the optical calibration and torch
position optimization may be performed (consult the operator's
manual).
Calibrate the instrument according to the instrument
manufacturer's recommended procedures, and the procedures
Before analyzing the samples,
specified in Section 6.3.
reanalyze the highest calibration standard as if it were a
sample. Concentration values obtained should not deviate from
the actual values by more than 5%, or the established control
limits, whichever is lower (see Sections 6 and 7). If they do,
follow the recommendations of the instrument manufacturer to
correct for this condition.
Flush the system with the calibration blank solution for at
least 1 min before the analysis of each sample or standard.
Analyze the midpoint calibration standard and the calibration
blank after each 10 samples. Use the average intensity of
multiple exposures for both standardization and sample analysis
to reduce random error.
Dilute and reanalyze samples that are more concentrated than
the linear calibration limit or use an alternate, less sensitive
Cr wavelength for which quality control data are already
established.
11
If dilutions are performed, the appropriate factors must be
applied to sample values. All results should be reported in pg
Cr/ml with up to three significant figures.
5.5
Sample Analyses by IcIPCR. The Cr+' content of the
sample filtrate is determined by IC/PCR. To increase sensitivity
for trace levels of chromium, a preconcentration system is also
used in conjunction with the IC/PCR.
Prior to preconcentration and/or analysis, filter all field
samples through a 0.45-pm filter. This filtration should be
conducted just prior to sample injection/analysis.
The preconcentration is accomplished by selectively
retaining the analyte on a solid absorbent (as described in
Section 3.4.3.3), followed by removal of the analyte from the
absorbent. Inject the sample into a sample loop of the desired
size (use repeated loadings or a larger size loop for greater
sensitivity). The Cr" is collected on the resin bed of the
column. Switch the injection valve so that the eluent displaces
the concentrated Cr+6 sample, moving it off the preconcentration
column and onto the IC anion separation column. After separation
from other sample components, the Cr+6 forms a specific complex
in the post-column reactor with the DPC reaction solution, and
the complex is detected by visible absorbance at a wavelength of
520 nm. The amount of absorbance measured is proportional to the
concentration of the Cr" complex formed. Compare the IC
retention time and the absorbance of the Cr+6 complex with known
Cr+6 standards analyzed under identical conditions to provide
both qualitative and quantitative analyses.
Two types of blanks are required for the analysis. The
calibration blank is used in establishing the analytical curve,
and the reagent blank is used to assess possible contamination
resulting from sample processing. Use either 0.1 N NaOH or 0.1 N
NaHC03, whichever was used for the impinger solution, for the
calibration blank. The calibration blank can be prepared fresh
in the laboratory; it does not have to be from the same batch of
solution that was used in the field. The reagent blank (Section
5.2.2) is a sample of the impinger solution used for sample
collection that is collected in the field during the testing
program.
Prior to sample analysis, establish a stable baseline with
the detector set at the required attenuation by setting the
eluent' flow rate at approximately 1 ml/min and the post-column
reagent flow rate at approximately 0.5 ml/min. Note: A s long as
the ratio of eluent flow rate to PCR flow rate remains constant,
the standard curve should remain linear. Inject a Sample of
reagent water to ensure that no Cr+6 appears in the water blank.
First, inject the calibration standards prepared, as
described in Section 4.3.9 to cover the appropriate concentration
range, starting with the lowest standard first. Next, inject, in
duplicate, the calibration reference standard (as described in
Section 7.3.1), followed by the reagent blank (Section 5.2.2),
and the field samples. Finally, repeat the injection of the
calibration standards to assess instrument drift. Measure areas
'
12
or heights of the Cr+6/DPC complex chromatogram peaks. The
response for replicate, consecutive injections of samples must be
within 5% of the average response, or the injection should be
repeated until the 5% criterion can be met. Use the average
response (peak areas or heights) from the duplicate injections of
calibration standards to generate a linear calibration curve.
From the calibration curve, determine the concentrations of the
field samples employing the average response from the duplicate
injections.
6.
Calibration
6.1
Sampling Train Calibration. Perform all of the
calibrations described in Method 5, Section 5. The alternate
calibration procedures described in Section 7 of Method 5 may
also be used.
6.2
GFAAS Calibration. Either (1) run a series of chromium
standards and a calibration blank and construct a calibration
curve by plotting the concentrations of the standards against the
absorbencies, or (2) using the method of standard additions, plot
added concentration versus absorbance. For instruments that read
directly in concentration, set the curve corrector to read out
the proper concentration, if applicable. This is customarily
performed automatically with most instrument computer-based data
systems.
6.2.1
GFAAB Calibration Curve. If a calibration curve is
used, it should be prepared daily with a minimum of a calibration
blank and three standards. Calibration standards for total
chromium should start with 1% v/v HNO, with no chromium for the
calibration blank, with appropriate increases in total chromium
concentration for the other calibration standards (see Section
Calibration standards should be prepared fresh daily.
4.3.9.).
6.3
ICP Calibration. Calibrate the instrument according to
the instrument manufacturer's recommended procedures, using a
calibration blank and three standards for the initial
calibration. Calibration standards should be prepared fresh
Be sure that samples and
daily, as described in Section 4.3.9.
calibration standards are matrix matched. Flush the system with
the calibration blank between each standard. Use the average
intensity of multiple exposures for both standardization and
sample analysis to reduce random error.
6.4
IC/PCR Calibration. Prepare a calibration curve using
the calibration blank and three calibration standards prepared
fresh daily as described in Section 4.3.9.
Run the standards
with the field samples as described in Section 5.5.
7.
Quality Control
7.1
GFAAS Quality Control
GFAAS Calibration Reference Standards. If a
calibration curve is used, it must be verified by use of at least
one calibration reference standard (made from a reference
7.1.1
13
material or other independent standard material) at or near the
mid-range of the calibration curve. The calibration reference
standard must be measured within 10% of it's true value for the
curve to be considered valid. The curve must be validated before
sample analyses are performed.
7.1.2
GFAAS Check Standards. Run a check standard and a
calibration blank after approximately every 10 sample injections,
and at the end of the analytical run. These standards are run, in
part, to monitor the life and performance of the graphite tube.
Lack of reproducibility or a significant change in the signal for
the check standard indicates that the graphite tube should be
replaced. Check standards can be the mid-range calibration
standard or the reference standard. The results of the check
standard shall agree within 10% of the expected value. If not,
terminate the analyses, correct the problem, recalibrate the
instrument, and reanalyze all samples analyzed subsequent to the
last acceptable check standard analysis.
The results of the calibration blank are to agree within
three standard deviations of the mean blank value. If not,
repeat the analysis two more times and average the results. If
the average is not within three standard deviations of the
background mean, terminate the analyses, correct the problem,
recalibrate, and reanalyze all samples analyzed subsequent to the
last acceptable calibration blank analysis.
7.1.3
GFAAS Duplicate Samples. Run one duplicate sample
for every 20 samples, (or one per source test, whichever is more
frequent). Duplicate samples are brought through the whole
sample preparation and analytical process separately. Duplicate
samples shall agree within 10%.
7.1.4
G F M S Matrix Spiking. Spiked samples shall be
prepared and analyzed daily to ensure that correct procedures are
being followed and that all equipment is operating properly.
Spiked sample recovery analyses should indicate a recovery for
the Cr spike of between 75 and 125%. spikes are added prior to
any sample preparation. Cr levels in the spiked sample should
provide final solution concentrations that fall within the linear
portion of the calibration curve.
7.1.5
GFAAS Method of Standard Additions.
Whenever sample
matrix problems are suspected and standard/sample matrix matching
is not possible or whenever a new sample matrix is being
analyzed, the method of standard additions shall be used for the
analysis of all extracts. Section 5.4.2 of Method 12 (40 CFR
Part 60, Appendix A) specifies a performance test to determine if
the method of standard additions is necessary.
7.1.6
GFAAS Reagent Blank Samples. Analyze a minimum of
one matrix-matched reagent blank (Section 5.2.2) per sample batch
to determine if contamination or memory effects are occurring.
The results should agree within three standard deviations of the
mean blank value.
7.2
ICP Quality Control.
7.2.1
ICP Interference Check. Prepare an interference
check solution to contain known concentrations of interfering
14
elements that will provide an adequate test of the correction
factors in the event of potential spectral interferences. Two
potential interferences, iron and manganese, may be prepared as
1000 pg/ml and 200 pg/ml solutions, respectively. The solutions
should be prepared in dilute HN03 (1-5%). Particular care must
be taken to ensure that the solutions and/or salts used to
prepare the solutions are of ICP grade purity (i.e., that no
measurable Cr contamination exists in the salts/solutions).
Commercially prepared interfering element check standards are
available. Verify the interelement correction factors every
three months by analyzing the interference check solution. The
correction factors are calculated according to the instrument
manufacturer's directions. If interelement correction factors
are used properly, no false Cr should be detected.
7.2.2
ICP Calibration Reference Standards. Prepare a
calibration reference standard in the same alkaline matrix as the
calibration standards; it should be at least 10 times the
instrumental detection limit. This reference standard should be
prepared from a different Cr stock solution source than that used
f o r preparation of the calibration curve standards and is used to
verify the accuracy of the calibration curve. Prior to sample
analysis, analyze at least one reference standard. The
calibration reference standard must be measured within 10% of
it's true value for the curve to be considered valid. The curve
must be validated before sample analyses are performed.
7.2.3
ICP Check Standards. Run a check standard and a
calibration blank after every 10 samples, and at the end of the
analytical run. Check standards can be the mid-range calibration
standard or the reference standard. The results of the check
standard shall agree within 10% of the expected value; if not,
terminate the analyses, correct the problem, recalibrate the
instrument, and rerun all samples analyzed subsequent to the last
acceptable check standard analysis. The results of the
calibration blank are to agree within three standard deviations
of the mean blank value. If not, repeat the analysis two more
times and average the results. If the average is not within
three standard deviations of the background mean, terminate the
analyses, correct the problem, recalibrate, and reanalyze all
samples analyzed subsequent to the last acceptable calibration
blank analysis.
7.2.4
ICP Duplicate Samples. Analyze one duplicate sample
for every 20 samples, (or one per source test, whichever is more
frequent). Duplicate samples are brought through the whole
sample preparation and analytical process. Duplicate samples
shall agree within 10%.
7.2.5
ICP Reagent Blank Samples. Analyze a minimum of one
matrix-matched reagent blank (Section 5 . 2 . 2 ) per sample batch to
determine if contamination or memory effects are occurring. The
results should agree within three standard deviations of the mean
blank value.
7.3
IC/PCR Quality Control.
15
7.3.1
IC/PCR Calibration Reference Standards. Prepare a
calibration reference standard in the same alkaline matrix as the
calibration standards at a concentration that is at or near the
mid-point of the calibration curve. This reference standard
should be prepared from a different Cr stock solution source than
that used for preparing the calibration curve standards. The
reference standard is used to verify the accuracy of the
calibration curve. Prior to sample analysis, analyze at least
one reference standard. The results of this analysis of the
reference standard must be within 10% of the true value of the
reference standard for the calibration curve to be considered
valid. The curve must be validated before sample analyses are
performed
7.3.2
IC/PCR Check Standards. Run the calibration blank
and calibration standards with the field samples as described in
Section 5.5.
For each standard, determine the peak areas
(recommended) or the peak heights, calculate the average response
from the duplicate injections, and plot the average response
against the Cr+6 concentration in pg/1
The individual responses
for each calibration standard determined before and after field
sample analysis must be within 5% of the average response for the
analysis to be valid. If the 5% criteria is exceeded, excessive
drift and/or instrument degradation may have occurred, and must
be corrected before further analyses are performed.
Employing linear regression, calculate a predicted value for
each calibration standard using the average response for the
duplicate injections. Each predicted value must be within 7% of
the actual value for the calibration curve to be considered
acceptable. If not acceptable, remake and/or rerun the
calibration standards. If the calibration curve is still
unacceptable, reduce the range of the curve.
7.3.3
IC/PCR Duplicate Samples. Analyze one duplicate
sample for every 2 0 samples, (or one per source test, whichever
is more frequent). Duplicate samples are brought through the
whole sample preparation and analytical process. Duplicate
samples shall agree within 10%.
7.3.4
ICP Reagent Blank Samples. Analyze a minimum of one
matrix-matched reagent blank (Section 5 . 2 . 2 ) per sample batch to
determine if contamination or memory effects are occurring. The
results should agree within three standard deviations of the mean
blank value.
.
.
8
.
M s s i o n Calculations
Carry out the calculations, retaining one extra decimal
figure beyond that of the acquired data. Round off figures after
final calculations.
8.1
Total Cr in Sample. Calculate ,
M
the total Pg Cr in
each sample, as follows:
Eq. 306-1
16
where:
vml
= Volume of impinger contents plus rinses, ml.
cs = Concentration of Cr in sample solution, pg Cr/ml.
F = Dilution factor.
= Volume of aliquot after dilution. ml
Volume of aliquot before dilution, ml
D = Digestion factor.
= Volume of samnle aliquot after diaestion, ml
Volume of sample aliquot submitted to
digestion, ml
Average Dry Gas Meter Temperature and Average Orifice
8.2
Pressure Drop. Same as Method 5, Section 6.2.
8.3
Dry Gas Volume, Volume o f Water Vapor, Moisture
Content. Same as Method 5, Sections 6.3, 6.4, and 6.5,
respectively.
8.4
Cr Emission Concentration. Calculate CCr, the
Cr concentration in the stack gas, in mg/dscm on a dry basis,
corrected to standard conditions, as follows:
Eq. 306-2
where:
'm(std)
--
Gas sample volume
measured by the dry gas
meter, corrected to dry
standard conditions,
dscm.
8.5
Isokinetic Variation, Acceptable Results.
Method 5, Sections 6.11 and 6.12, respectively.
9.
Same as
Bibliography
IlTest Methods for Evaluating Solid Waste,
1.
Physical/Chemical Methodst1, U.S. Environmental Protection Agency
Publication SW-846, 2nd Edition, July 1982.
2. Cox, X.B., R.W. Linton, and F.E. Butler. Determination
of Chromium Speciation in Environmental Particles
A
Multitechnique Study of Ferrochrome Smelter Dust. Accepted for
publication in Environmental Science and Technology.
Same as Bibliography of Method 5, Citations 2 to 5
3.
and 7.
4.
California Air Resources Board, IIDetermination of Total
Chromium and Hexavalent Chromium Emissions from Stationary
Sources.tt Method 425, September 12, 1990.
5.
"Test Methods for Evaluating Solid Waste, Physical/
Chemical Methodst8,U. S. Environmental Protection Agency
Publication SW-846, 3rd Edition, November 1986 as amended by
Update I , November 1990.
-
17