United States
Environmental Protection
Agency
Environmental Sciences Research
Laboratory
Research Triangle Park NC 27711
Research and Development
EPA-600/S3-83-105 Jan. 1984
Project Summary
Intercomparison of Periodic
Fine Particle Sulfur and
Sulfate Concentration Results
David C. Camp, Richard J. Paur, and Robert K. Stevens
A one-week study was conducted in
August 1979 to evaluate the compar-
ative ability of representative aerosol
sampling systems to monitor fine partic-
ulate sulfur and sulfate concentrations
periodically in situ. Participants in the
study operated their samplers simulta-
neously in the same location for the
duration of the study. Samplers tested
included five modified flame photomet-
ric detection instruments and two di-
chotomous samplers, including one
sampler designed to yield near-real-time
sulfur concentrations from on-line, wave-
length dispersive x-ray fluorescence
analysis of fine particles collected on
Teflon substrate. All but one of the
samplers were operated from a common
duct located within the same laboratory;
the other was operated on the roof of
the building.
Normalized results of all sulfur con-
centration levels measured for five of
the six samplers were 1.0 ± 0.05. The
sixth set gave a result of 0.60 ±0.14.
After revised results were resubmitted
for the sixth set, normalized results for
the six sets of hourly sulfur results
agreed to within ± 8% of their composite-
mean values, with standard deviations
varying from 9 to 22%. One participat-
ing group operated three similar sam-
plers, each calibrated by a different
method, that gave normalized results of
1.06 ± 0.12, 0.93 ± 0.22, and 1.54±
0.25; these results point out the need
for a standard calibration procedure.
This Project Summary was developed
by EPA's Environmental Sciences Re-
search Laboratory, Research Triangle
Park, NC, to announce key findings of
the research project that is fully docu-
mented in a separate report of the same
title (see Project Report ordering infor-
mation at back).
Introduction
In August 1979, a one-week study was
conducted at Washington University, St.
Louis, to compare a family of air pollution
monitoring instruments that are capable
of measuring in near-real-time atmos-
pheric concentrations of aerosol sulfur.
The main objective of this study was to
compare different procedures to measure
aerosol sulfur that use flame photometric
detectors (FPD's) and an automated semi-
real-time x-ray fluorescence (XRF) aero-
sol sample-analyzer. The instruments
that use the FPD's have been used in a
variety of air pollution studies, but be-
cause each instrument uses a different
method of sample conditioning prior to
the sulfur measurement, bias in the
measurements may exist. This study
compared measurements of different
automated aerosol sulfur monitors on
both simulated and real ambient air
samples over a period of 7 days to
establish the range of differences be-
tween instruments and the basis for
variations in response characteristics.
Procedure
Analyzers brought to Washington Uni-
versity by the seven participating groups
included five modified FPD instruments
and two dichotomous samplers. One
dichotomous sampler was designed to
report near-real-time sulfur concentra-
tions from an on-line, wavelength dis-
persive, XRF analysis of the fine particles
collected on a filter substrate. All sam-
-------
piers were operated simultaneously, and
all but one operated from a common duct
leading to a common laboratory. The inlet
to the common duct contained an impac-
tor plate particle-sizing device that re-
jected all particles greater than 3.0 (jm in
aerodynamic diameter. The sampler not
using the common duct was operated on
the roof above the laboratory.
Results
Concentration of the sulfur measured
during the study by the participants
consisted of eight sets of hourly fine
pa rticulate sulfur values. Only six of eight
sets were intercompared. One of the
participating groups reported three sets
of sulfur data based on three different
calibrations methods. Ambient sulfur
concentrations measured during the study
ranged from just below 1 /ug/m3 to 9
fjg/m3. Above 6 //g/m3 the five FPD
instruments and the automated XRF
sulfur analyzer intercompared agreed to
within ± 25% of their combined compos-
ite-mean concentrations. Five of the six
agreed to within ± 5% of the composite-
mean values. All of the automated FPD
instruments were able to resolve and
measure the relatively sharp sulfur con-
centration peaks that occurred daily. The
source and nature of these short, some-
times multi-peaked, excursions were not
investigated.
Three sets of 6-h sulfate concentra-
tions were reported and intercompared.
Their concentrations varied from about
2 fjg/m3 to 20 jug/m3. At relatively high
sulfate concentrations (> 14 /jg/m3) all
three samplers agreed to within ± 10% of
their composite 6-h means. Regression
analysis of 6-h composite-mean sulfate
concentrations, derived from ion chromat-
ographic analysis of fine particles col-
lected with a dichotomous sampler, vs.
equivalent 6-h averages for the hourly
sulfur composite-mean values gave a
sulfate-to-sulfur ratio of 2.2 for the St.
Louis study vs. 3.0 observed in two earlier
studies. This ratio remained at 2.2 for
both clean and stagnant air and during
the transient sulfur excursions. Thus,
either all of the sulfur did not occur in
sulfate form or the sulfate could not be
extracted completely from the filter-types
used by all three independent groups.
Three sampling systems operated by
the host institution provided results based
on three different calibration methods.
Their results based on an (NH4)2S04
calibration were closest to the composite
mean average (1.06 ± 0.12—an average
of all hourly results normalized to the
hourly composite-mean values); whereas
the sulfur monitors that used S02 for
calibration differed substantially from the
composite mean (1.54 ± 0.25). If either of
the two worst-result sets of the eight sets
submitted had been the only set of sulfur
concentrations measured, the report*
hourly sulfur concentrations would somi
times have been as much as a factor of
higher or lower than the consensi
composite-mean concentrations from tr
six sets intercompared.
Conclusions and
Recommendations
Measurements obtained from FPDsy;
terns modified to measure aerosol sulfi
and calibrated by appropriate procedure
agreed within ± 8% with the composii
mean. Data from FPD's calibrated usir
(NH4)2SO4 had less scatter and were i
closer agreement to the composite mea
average than data from FPD analyzei
calibrated using SO2. A standard ASTN
type aerosol calibration procedure neec
to be developed and documented f<
routine calibration of instruments usedl
measure in near-real-time ambient cor
centrations of aerosol sulfur.
David C. Camp is a Consultant at 258 Paraiso Drive, Danville, CA 94526; the EPA
authors, Richard J. Paur and Robert K. Stevens (also the EPA Project Officer,
see below) are with Environmental Sciences Research Laboratory, Research
Triangle Park, NC 27711.
The complete report, entitled "Intercomparison of Periodic Fine Particle Sulfur
and Sulfate Concentration Results," (Order No. PB84-119 940; Cost: $10.00,
subject to change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Environmental Sciences Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
«US. GOVERNMENT PRINTING OFFICE 1984-759-015/7285
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Official Business
Penalty for Private Use $300
PS 0000329
------- |