Chemical Analyses of Particle Filter Deposits* - Aerosols: Measurement, Dosimetry, and Health Effects - page 183

Filters can be stored in individual Petri slides (PD1504700; Millipore, Billerica, MA) that
accommodate 47, 37, and 25 mm diameter ilters, the sizes most commonly used for ambient and
source sampling. Bar-coded labels can be afixed to the Petri slide with duplicates placed on the
ilter holder and ield data sheet. It is best to use a ilter identiier (one for each type in Figure 7.1)
followed by a sample identiier that is the same for each of the seven ilters. The sample identiier
can contain information about the network, sampling site, and sampling date if desired. Not more
than one ilter should be removed from its holder or Petri slide at a time to minimize the risk of
swapping of ilters among different samples. The same ilter ID should be transferred to analysis
vials and Laboratory Information Management Systems (LIMS) so that the different analyses can
be assembled into a uniied database.
For unknown reasons, some batches of ringed Telon-membrane ilters have yielded variable
(by up to 100 μg/ilter over a few days) blank masses (e.g., Tombach et al., 1987). This may be due
to evaporation of the adhesive between the ring and Telon membrane soon after manufacture. A
1 month storage period in a controlled environment, followed by 1 week of equilibration, is recom-
mended before gravimetric analysis. A sample of each batch of 100 preired, quartz-iber ilters
should be tested for carbon blank levels prior to sampling, and ilter batches with carbon levels
exceeding 1.5 μg/cm 2 for OC, or 0.5 μg/cm 2 for EC, should be reired or rejected. All preired ilters
should be sealed and stored in a refrigerator (<4°C) prior to and after ield sampling.
Cellulose-iber ilters are immersed in the appropriate impregnating solution for approximately
30 min. These disks are then removed and placed in large Petri dishes for drying in a vacuum oven
for 5-10 min. One hundred of the dried impregnated ilters are immediately sealed in polyethylene
bags and placed under refrigeration for later loading into ilter holders. One sample from each lot of
ilters should be analyzed prior to ield sampling to assure that ilter batches have not been contami-
nated and that the impregnating solution concentration level has adequate capacity for speciic gas
sampling. Impregnated ilters also are used for passive gas sampling (Cox, 2003), and their blank
levels will increase with exposure to ambient air.
Some analyses require the ilter to be cut into sections for different extraction and analysis proce-
dures. This can be done with a standard paper cutter to which a half-circle or quarter-circle polycar-
bonate jig has been attached. The jig should have the diameter of the ilter to allow the exact fraction
desired to be cut. The blade is cleaned with methanol (MeOH) and a laboratory wipe between ilter
cuttings. The unanalyzed ilter portion is archived under refrigeration in its original Petri slide. The
remaining ilter section is placed in a polystyrene extraction vial with a screw cap (e.g., 17 × 120 mm
vial, Greiner #188271). Each vial is labeled with a bar code sticker containing the ilter ID code. The
extraction tubes are placed in tube racks and extraction solutions are added. The extraction vials
are capped and sonicated for 60 min, shaken for 60 min, and then aged overnight to assure complete
extraction of the deposited material in the solvent. The ultrasonic bath water is monitored to prevent
temperature increases from dissipation of ultrasonic energy. After extraction, these solutions are
stored under refrigeration prior to analysis.
After completion and validation of nondestructive x-ray luorescence (XRF) analysis, the entire
Telon-membrane ilter is submitted to a strong acid extraction (Anzano and Ruiz-Gil, 2005; Feng
and Barratt, 1994; Kaasalainen and Yli-Halla, 2003; Link et al., 1998; Melaku et al., 2005; Rao
et al., 2008; Silveira et al., 2006; Wang et al., 1995). U.S. EPA Method 200.8 (U.S.EPA, 1994)
is modiied with the addition of hydroluoric acid (HF) to separate the metals from their min-
eral oxides. The ilter is cut into pieces using ceramic scissors to allow the ilter to it inside a
68 mL digestion vessel. Ethanol (0.2 mL) is added to wet the ilter, along with 2 mL of a concen-
trated (68%) HNO 3 mixture (1:1 v/v) with distilled-deionized water (DDW), 5 mL of concentrated
(34%) hydrochloric acid (HCl) mixture with DDW (1:4 v/v), and 0.1 mL of concentrated (49%) HF
with DDW. The digestion vessels with a relux cap are then placed in a hot block (Environmental
Express, Mt. Pleasant, SC) located in a clean-air enclosure and gently heated from room tempera-
ture to 113°C, which maintains the sample at 95°C. After 90 min, the samples are removed, cooled