III. Preparing the Sampling Train for Use
The sampling train must be prepared in the laboratory before it can be used in
the field.The train must be assembled and leak tested,the flow rate must be
set,and the train must be certified clean.All ofthe following information
should be documented for the chain ofcustody for the passive sampling train
and the sample collected with it.
Assemble, Leak Test, and Set the Flow Rate of the Passive Sampling Train
Choose the critical orifice (Table II,page 3) according to the sampling period
and flow rate you anticipate using (Table III).This will ensure an accurate and
valid sample.There should be a marking on the outside ofthe critical orifice fit-
ting indicating the size ofthe orifice.In a clean environment,assemble the sam-
pling train components as shown in Figure 2 (page 2).It is imperative that you
leak test the assembled train.Ifthe sampling train leaks during sampling,the
final pressure in the canister will not be the desired final pressure,making the
sample invalid.The most common reason for invalid samples is leaks within the
sampling train.There are two ways to leak test the train:
1. Pass helium gas through the flow controller and use a sensitive helium
leak detector to test for leaks (e.g.,Restek Leak Detector).
2. Cap the inlet,attach the sampling train to an evacuated canister,open the
valve on the canister and evacuate the sampling train.Then,close the
valve and monitor any pressure change in the static sampling train.Leaks
ofless than 1 mL/min.can be detected in 1-2 minutes.
This is a good practical test—the small internal volume ofthe passive sam-
pling train,combined with even a small leak,will produce a large change in
monitored pressure.According to EPA Method TO-15,the pressure change
should be less than 2 psig (13.8 kPa) over a 24-hour period.
After you are certain the sampling train is leak-free,set the desired sampling
To set the desired flow rate follow these steps:
1. Remove the protective cap from the back ofthe Veriflo® Flow Controller
2. Connect either an evacuated canister or a vacuum source to the outlet of
the sampling train.
3. Connect a high quality calibrated flow meter (i.e.,mass flow meter,rotame-
ter,GC-type flow sensor [e.g.,Restek ProFLOW 6000 Electronic Flowmeter,
cat.# 22656]) to the inlet ofthe train.
4. Apply vacuum by opening the canister or turning on the vacuum source.
5. With a 3 mm hex (Allen®) wrench,adjust the piston gap screw to achieve
the desired flow rate (Table III).Between adjustments allow the flow to
equilibrate for several minutes.See Figure 8.
6. Replace the protective cap onto the back ofthe Veriflo® Flow Controller body.
Cleanliness: Certifying the Sampling Train for Use
US EPA Compendium Method TO-14A/TO-15 requires that the sampling
train be certified clean prior to use.Certify the train by passing a humidified,
high-purity air stream through the train,concentrating the exit gas on a trap,
and analyzing the gas by gas chromatography/mass spectrometry or other
selective detector.For the sampling train to pass certification the analytical sys-
tem should not detect greater than 0.2 ppbv ofany target VOC.
The certified sampling train should be carefully packaged in aluminum foil or
in a clean container for storage or for shipment into the field.Care in packag-
ing is critical.Careless handling could affect the preset flow rate.When the
sampling train is ready for sampling,prepare the canister.
Figure 8 Setting flow rate on
•Only hand tighten knob to close valve.
Overtightening may damage seat causing
•Tighten compression fitting on valve inlet
turn past finger tight.
Overtightening will cause leakage.
•Use prefilter during sampling to prevent
particulate damage to valve.
•Do not disassemble valve—disassembly may
•Protect valve inlet by replacing brass cap
when not in use.
•Do not exceed canister maximum pressure
Table IIIFlow rates for integrated sampling,
using a 6-liter canister and sampling on the
flat portion of the flow curve for the flow
controller (Figure 5).
FFllooww RRaattee RRaannggee
Collected volume is 4–5 liters
(flow = volume in mL / sampling time in min.).
IV. Preparing the Canister for Sampling
Preparing a canister for sampling involves certifying the canister clean,evacuating the canister to final
pressure for use,and identifying the canister.All information acquired during these processes is needed
for the chain ofcustody.
Certifying the cleanliness ofthe canister is important toward ensuring that results reported are solely from
the site sampled,and not contaminated with residue from a previous site or volatiles in laboratory air.To
certify a canister clean,fill the canister with humidified air,pass the air from the canister through an adsor-
bent trap and analyze the adsorbent for target VOCs by GC/MS or other selective detector.Two US EPA
methods discuss canister certification:EPA Compendium TO-12 and EPA Compendium TO-14A/TO-15.
To comply with EPA Compendium Methods TO-14A/TO-15,the analytical system should not detect
greater than 0.2 ppbv ofany target VOC.To comply with EPA Compendium Method TO-12 the analyti-
cal system,GC/FID,should not detect greater than 0.02 ppmC hydrocarbons.Although batch certification
ofcanister cleanliness is a relatively common practice,we recommend certifying and documenting each
canister individually.Detailed cleaning instructions are presented in Section VIII.Cleaning the Canister
Some laboratories certify a canister for VOC stability by introducing a low concentration test mixture into
the canister and measuring degradation over a specified time period.Ifthe canister meets the specifica-
tion,it is certified for use.We recommend using such studies to ensure the effectiveness ofa canister or
group ofcanisters for a proposed application.
Once the canister is certified clean,evacuate the canister to a final vacuum of10-50 mtorr,using either the
canister cleaning system or a clean final vacuum system.This vacuum is critical to ensure the correct
amount ofsample is collected.Use an accurate test gauge (shown in Figure 7b,page 4) or digital pressure
tester to ensure final vacuum has been reached and to document the final vacuum reading for the chain
ofcustody.Install a brass cap nut onto the canister valve to ensure no contamination can enter the sam-
ple pathway during shipment to the field.
Apply an individual identity to the canister,either with a label and serial number or with a bar code.
Some analysts prefer to introduce surrogate standards into the canister prior to sampling.Debate on this
practice revolves around theories that there are potential loss issues due to low humidity and inadequate
surface passivation by water.Neither Restek chemists nor our consulting experts recommend adding sur-
rogates to the canisters.Ifyou choose to introduce surrogates into your canisters prior to sampling,be sure
to recheck and record the vacuum reading for each canister after adding the surrogates.
V. Field Sampling, Using a Passive Sampling Train and Canister
It is important to mention again that the sampling train and canister must be leak tested and certified
clean prior to use.To properly begin field sampling,we recommend bringing a “practice”evacuated can-
ister and a flow measuring device with you to the field.Use this canister to verify the flow rate through the
passive sampling train prior to using the train to obtain samples ofrecord.To verify the flow rate,connect
the passive sampling train to the “practice”canister.Attach a flow meter to the inlet ofthe sampling train.
Open the canister and measure the flow rate through the sampling train.Ifthe flow rate is within ± 10%
ofthe flow rate set in the lab,the train is ready to be used on the formal sampling canister.Ifthe flow rate
is not within these limits,adjust the flow rate by adjusting the piston gap screw.
When the flow rate is confirmed,record the rate as the canister flow rate for the chain ofcustody form.
did you know?
Our light-weight tripod holds
2 canisters securely without
Pressure Conversion Table
in Hg =
Multiply units in the left-most column by the conversion factors listed in the columns to the right.
e.g., 10PSI x 0.068 = 0.68atm, 10 bar x 29.5300 = 295.300 inches Hg
To begin sampling, using the formal sampling canister, follow these steps:
1. Remove the brass cap nut from the canister valve.
2. Ifyou are using a test gauge,attach the gauge to the canister and record the vacuum reading.Ifyou choose not to use a test
gauge under field conditions,record the reading on the vacuum gauge that is part ofthe passive sampling train.
3. Attach the verified passive sampling train to the canister.
4. Record the sampling start time and necessary meteorological data.
5. Open the canister valve and begin sampling.
6. Periodically check the canister throughout the sampling period to ensure the pressure reading is accurate and sampling is pro-
ceeding as planned.
7. Once the sampling period is complete,close the valve and remove the sampling train.Check the final pressure within the canis-
ter,using the test gauge or the vacuum gauge in the sampling train.
There are four possible scenarios:
A.Ideally there will be a vacuum of-7"to -4" Hg in the canister (e.g.,Table IV).
B. Ifmore than -7" Hg vacuum remains,less sample was collected than ini-
tially anticipated.The sample will be valid,but the detection limit may be
higher than expected.You might have to pressurize the canister prior to
the analysis,which will dilute the sample and require you to use a dilution
factor to determine final concentrations oftarget compounds.
C.A vacuum ofless than -4" Hg indicates the sample might be skewed
toward the initial part ofthe sampling period.This assumption usually is
valid because the flow rate through the flow controller will fall once the
vacuum falls below -5" Hg (Figure 6,page 4),when the change in pressure
across the flow controller diaphragm becomes too small and the flow con-
troller is unable to maintain a constant flow.Although flow was not con-
stant over the entire sampling period,the sample may be usable because
sample was collected over the entire interval.
D.Ifthe ending vacuum is less than -1" Hg the sample should be considered
invalid because it will be impossible to tell when the sample flow stopped.
8. Record the final pressure in the canister and replace the cap nut.
Information that should be acquired at the sam-
pling site includes the start time and interval time,
the stop time,atmospheric pressure and tempera-
ture and,for ambient sampling,wind direction.
Include elevation ifit is a factor.These parameters
often prove very useful when interpreting results.
After sampling,the canisters are sent back to the
laboratory where the final vacuum is measured
again with a test gauge.Using the initial vacuum
and final vacuum,the sample volume collected can
be determined from Equation 1.
It is also good practice to recheck the flow rate after
sampling,because this will affect the sample volume
(Equation 2).Laboratories typically allow a maxi-
mum deviation of±10% to ±25% between the ini-
tial flow rate and the post-sampling flow rate.
Table IV Final vacuum and volume of
sample collected in 6-liter canister.
sample volume = [(initial flow rate + post-sampling flow rate)/2] x sampling time
Example: A flow controller was set at 3.3 mL/min. After obtaining a 24-hour
sample the flow rate was 3.0 mL/min.
sample volume = [(3.3 mL/min. + 3.0 mL/min.)/2] x 1,440 min. = 4,536mL
sample volume =
x canister volume
*initial pressure – final pressure
Example: A sample is collected in a 6-liter canister. The initial gauge pressure
reading when the canister left the lab was -29.92" Hg vacuum; the final gauge
pressure reading when the canister was returned to the lab was -7" Hg vacuum.
-29.92" Hg – (-7" Hg)
sample volume =
x 6 L = 4.59 liters collected
VI. Analysis of Collected Samples
Once received by the lab,each canister is identified from the information in the
chain of custody report.The final pressure is checked to ensure no leaks
appeared during transport.It might be necessary to pressurize a canister prior
to the analysis;do this by adding humidified nitrogen or air to the canister to
a pressure greater than 5 psig or higher,depending on the sample volume
needed for analysis or for suitably diluting the sample (e.g.,Table V).The need
to dilute is determined by the preconcentrator instrument.Some air precon-
centrators can be operated while the canister is under slight vacuum.Check
with your instrument manuals or with the manufacturer to determine ifyou
must dilute your samples prior to analysis.Dilution factors can be calculated
according to Equation 3.
To analyze the sample,withdraw an aliquot ofthe sample from the canister.For low level ambient air analysis,withdraw 250-500 mL
ofsample from the canister and concentrate the analytes by using a mass flow controller and a cryogenically cooled trap (e.g.,glass
beads and/or a solid sorbent).Desorb the concentrated analytes from the trap and deliver them to a cryofocuser to focus the sample
bandwidth prior to introduction onto the GC column.A 60 m x 0.32 mm ID x 1.0 µm Rtx®-1 column typically is used for EPA
Method TO-14A or Method TO-15 ambient air analysis;an MSD is a common detector.Figure 9 shows a typical TIC spectrum for
a TO-15 ambient air analysis.
Table VDilution factors to adjust final
sampling pressure to 14.7 psigfor a
dilution factor = (P
) / (P
The dilution factor is calculated from the post-sampling pressure (before dilu-
tion), the final pressure (after dilution), and the atmospheric pressure in the lab-
oratory. The factor for converting "Hg to psi = 0.491.
Example: At the end of a sampling period the gauge pressure in a canister was
-7 "Hg. The canister was pressurized with nitrogen to 14.7 psig (1 Atm.).
The dilution factor is (14.7 + 14.7) / (14.7 - (7 x 0.491)) = 2.61
Figure 9 US EPA TO-15 ambient air analysis.
Rtx®-1, 60m, 0.32mm ID, 1.0µm (cat.# 10157)
TO-15 standard (cat.# 34436) humidified to 33% RH in a 6L SilcoCan®
canister (cat.# 24182)
Concentrator: Nutech 3550A Preconcentrator; 300mL sample concentrated at
-160°C, thermally desorbed at 150°C, cryofocused at -185°C,
thermally desorbed to column at 150°C
helium, constant flow
30°C (hold 4 min.) to 175°C @ 8°C/min., to 220°C @
20°C/min. (hold 2 min.)
20.methyl ethyl ketone
37.methyl isobutyl ketone
41.methyl butyl ketone
Procedures used in these chromatographic analyses generally include a multi-
point calibration,using gas standards.Therefore calculations oforganic com-
pounds in collected samples are straightforward—only volumes analyzed and
dilution rates are needed to determine sample concentrations.High concen-
tration calibration gas standards are commercially available (e.g.,1 ppmv or
100 ppbv).To prepare analytical standards,introduce an aliquot ofstock mate-
rial into a canister and dilute with humidified air or nitrogen.After analyzing
the calibration standards,determine the response factor for each analyte using
the peak area counts per concentration.
After analyzing the multipoint calibration standards and calculating peak
area/concentration response factors,analyze the “real world”samples.Ifan
“unknown”sample has not been diluted,apply the corresponding response
factor to each “unknown”analyte peak area to get the reporting limit concen-
tration ofthe “unknown”in the analysis (typically in ppbv).Ifyou have dilut-
ed the canister to get a positive pressure,you must apply the dilution factor to
the concentration values.This is done by multiplying the reporting limit by the
VII. Cleaning the Passive Sampling Train
The cleanliness ofthe sampling train is critical to collecting accurate and rep-
resentative samples.Practices followed for cleaning passive sampling equip-
ment between uses range from purging the sampling pathway with humidified
nitrogen or air for many hours,to heating the pathway during a purge,to dis-
assembling each component,sonicating the pieces in solvent (except for the
critical orifice),and oven baking the pieces prior to reassembly.The most
suitable mode ofcleaning depends on the concentrations ofanalytes ofinter-
est,and contaminants,in the previous sample collected.
The particle filter must be thoroughly cleaned between uses.Disassemble the
filter,then remove the larger particles from the frit by blowing particle-free
nitrogen through the frit from the outlet surface toward the inlet surface.After
the larger particles are removed,sonicate or rinse the filter parts in methanol
and then bake the parts in an oven at 130 °C to remove any residual organic
The critical orifice and flow controller can be cleaned in either oftwo ways.
The first method is to disassemble the flow controller and clean all the metal
parts with methanol.This will remove any high boiling point compounds that
have condensed onto the wetted areas ofthe controller.Heat the cleaned parts
in an oven at 130 °C to remove residual organic vapors.Do not sonicate the
critical orifice.Do not sonicate in solvent or bake any ofthe nonmetallic
parts,such as O-rings,or they will be damaged.Do not rinse the vacuum
gauge with methanol.The vacuum gauge may be heated,but do not exceed 80
°C;higher temperatures will damage the face and the laminated safety glass
lens.Heating to 80 °C will not affect the mechanical operation ofthe spiral
bourdon tube in the vacuum gauge.
A less involved method ofcleaning the flow controller is to use a heating jack-
et or heat gun to heat the components ofthe assembled sampling train,while
purging the system with nitrogen.As organic compounds are heated and des-
orbed from the interior surfaces,the nitrogen gas sweeps them out ofthe sam-
Preparing the Clean Passive Sampling Train for Re-use
After the sampling train components have been cleaned,reassemble the sys-
tem,check for leaks,set the desired flow rate,and certify the sampling system
clean.Follow the procedures described previously in this guide.Package the
clean sampling train to prevent contact with airborne contaminants.
Where can I find EPA
Air Toxic Methods?
pdf files of US EPA Air Toxic
Methods are available at this
ASTM Reference D5466 Standard
Test Method for Determination
of Volatile Organic Chemicals in
(Canister Sampling Methodology)
available at www.astm.org
Documents you may be interested
Documents you may be interested