Appen dix B: Gra vimetric An alysis Proc edure
Airmetrics
69
filters typically weigh approximately 150 mg.
2.
With nothing on the pans, press the "Tare" button on the balance after the
reading has stabilized to zero the balance.
3.
Place the 200 mg Class S weight on the balance. After the reading has
stabilized, calibrate the balance by pressing the "Calibrate" button.
4.
Remove the calibration weight from the sample pan.
5.
The balance is now ready for weighing.
E.
Record each use of the microbalance in its log book.
F.
Every three months, or when conditions warrant, check the working calibra-
tion weig ht against a second certified 20 0 mg w eight wh ose value is traceable
to NIST.
The purpose of our routine checks against the certified weight is to detect any
change s in the w orking sta ndard du e to wea r, dirt, or other damag e. Reco rd all
steps of the proced ure carefully in the balan ce log book. A lso record
barometric pressure, temperature, and relative humidity. Zero and calibrate the
balance with the working stand ards; then weig h the certified weigh t and record
the value. Repeat twice to provide three comparative values.
G. Data from procedure F above should be reviewed periodically to assure that
tare and calibration va lues do not chan ge without detec tion. Since our prim ary
interest is in w eight differe nces betw een clean and load ed filters, the a bsolute
value of th e balanc e tare is not a s critical as the constanc y of the va lue. Thu s it
is extremely important that frequent checks be performed to monitor changes
in the calibration weights.
Following the long-term trends of calibration weighings can furnish informa-
tion about the calibration standards.
H. Precision and Accuracy
A measure of the precision of the microbalance has been obtained by repetitive
weighings over a four-month period of the series of substitute weights. A total
of 75 meas urements m ade on six different w eights resulted in a rang e of ±1
Airmetrics MiniVol Users Guide
70
Airmetrics
microgram on the 200 milligram range, which is the range used in filter
weighing. This is in agreement with the man ufacturer's stated precision as a
percent of sample load for a 200 mg range.
Repetitive weighing actually reflects the precision of the microbalance and the
weights comb ined. Alth ough the class S m etal weig hts are reas onably s table
and constant, they are subject to environmental factors such as air buoyancy
and temperature and to corrosion, dirt, and handling damage.
Accura cy is define d as the ag reemen t of a balanc e reading with the tru e samp le
mass. The balance accuracy is limited by digital voltmeter linearity, torque
motor linearity, precision, and range agreement. As a percent of electrical
range, Cahn specifies the microbalance accuracy is ±0.005%.
The principal uncertainty is not in the balance, but in the calibration weights.
The Oregon State Metrology Laboratory states the overall uncertainty for our
primary certified w eight as ±1 2 microgram s.
On the 250 mg range, the total uncertainty of the system becomes ±14
micrograms. This figure can be used as the uncertainty in an apparent value
determination a fter the corrections discus sed earlier.
In weighing as performed on membrane filters, the uncertainties associated
with the weights are assumed to cancel out because of the calibration proce-
dures; but additional errors are introduced by the properties of the filters and
the handling procedures. Tracking of these errors is discussed in the section
detailing each sampling method.
Low Volume Suspended Particulate Matter
A. Equipment and Supplies
1.
Filters--Quartz Microfibre (QM-A) 47 mm, manufactured by Whatman
Corporation, catalog no. 1851-047.
2.
Filter Holders--Nuclepore Corp. Aerosol Holder, 47 mm Stock No.
430200.
3.
Multiple holder adapters, 47 mm Nuclepore Stock No. 470400, 8 per
