•United States
Environmental Protection
Agency
Great Lakes National
Program Office
230 South Dearborn Street
Chicago, Illinois 60604
EPA-905/4-90-002
GLIMPO Report No 01 -90
£EPA
Field Intercomparison of
Precipitation Samplers
For Assessing Wet
Deposition of Organic
Contaminants
-------
pr Yf- - <7/-. c ..'//_ y^ -Or*" EPA-905/4-90-002
' J- • • GLNPO Report No. 01-90
FIELD INTERCOMPARISON OF PRECIPITATION SAMPLERS FOR
ASSESSING VET DEPOSITION OF ORGANIC CONTAMINANTS
S.J. Eisenreich, T.P. Franz and M.B. Svanson
Environmental Engineering Sciences
Department of Civil and Mineral Engineering
University of Minnesota
Minneapolis, MN 55455
Edward Klappenbach
Project Officer
Grant No. R005840-01
U.S. Environmental Protection Agency
Great Lakes National Program Office
230 S. Dearborn Street
Chicago. IL 60604
U S Environmental Protection Agency
GLNPO Library Collection (PL-12J)
77 West Jackson Boulevard.
Chicago, IL 60604-3590
-------
OUTLINE
Summary 1
Introduction 4
Statement of the Problem
Objectives
Strategy
Processes of Wet Deposition 6
Experimental Section 11
Site Description
Sampler Design
Sampling Protocol
Analytical Protocol
Quality Assurance/Quality Control
Results and Discussion 39
Criterea for Precipitation Collectors
Rain Collection Efficiency
Organic Contaminant Collection Efficiency
Sampling Protocol Affecting Collection Efficiency
Mechanical Reliability and Operational Characteristics
Conclusions 82
Acknowledgements 85
References 86
-------
Appendices
A. PAH Concentrations in Rain - 1986 91
B. PCB Concentrations in Rain - 1986 96
C. Cl-Pesticide Concentrations in Rain - 1986 104
D. Cl-Benzene Concentrations in Rain - 1986 109
E. Propagation of Error Analysis Ill
F. Field Notes on Precipitation Samplers - 1986 116
-------
LIST OF FIGURES
1. Location of Precipitation Samplers
2. Layout of Precipitation Collectors at Cedar Creek
3. Schematic of Solvent MIC Sampler (SM)
(Chan and Perkins, 1986)
4. Schematic of Resin MIC Sampler (RM) (Strachan and Huneault, 1984)
5. Schematic of XAD-2 Resin Adsorbent Cartridge
6. Schematic of Analytical Procedure for the Analysis of
of PAHs, PCBs, Cl-Pesticides and Cl-Benzenes
7. Chromatogram of 17 PAHs Using GC-Mass Selective Detector
8. Glass Capillary Gas Chroraatograms With Electron Capture Detection
of PCB Aroclor Mixtures (1:1:1 Aroclor 1242:1254:1260)
9. Rain Collection Efficiency of the SM, RM, FM and RA
Wet-Only Rain Samplers
10. Comparison of Theoretical to Actual Collected Rain Volume for
the Solvent MIC (SM) Sampler
11. Comparison of Theoretical to Actual Collected Rain Volume for
the Resin MIC (RM) Sampler
12. Comparison of Theoretical to Actual Collected Rain Volume for
the Filter MIC (FM) Sampler
13. Comparison of Theoretical to Actual Collected Rain Volume for
the Resin Aerochem Metrics (RA) Sampler
14. Volume-Weighted Mean Concentrations (VWM) and Propagated Errors
of Four PAHs for the SM, RM, FM and RA Samplers
15. Volume-Weighted Mean Concentrations (VWM) and Propagated Errors
of Six Chlorinated Hydrocarbons for the SM, RM, FM and RA
Samplers
16. Comparison of Volume-Weighted Mean Concentrations (VWM) for 17
PAHs Between the RM, FM and RA Samplers (A) and Between the
SM and All Samplers (B)
17. Comparison of Volume-Weighted Mean Concentrations (VWM) for 27
PCB Congeners for the RM, FM, and RA Samplers
18. Comparison of Volume-Weighted Mean Concentrations (VWM) for 27 PCB
Congeners Between the SM Sampler and All Samplers
19. Temporal Concentration Variations for Selected PAH Compounds
Observed in the SM, RM, FM and RA Samplers
(A): Phenanthrene and Chrysene; (B): Benzo[a]pyrene and
Benzo[ghi]perylene; (C): £ PAHs
20. Temporal Concentration Variations for Selected PCB Congeners
Observed in the SM, RM, FM and RA Samplers
(A): Congeners #31 and #10; (B) Congeners #110 and #138;
(C): Congener #180 and £ PCBs
21. Percentage of 14 Selected PAHs and Chlorinated Hydrocarbons in
the Funnel Rinse for the SM, RM, FM and RA Samplers
22. Percentage of (A) 17 PAHs and (B) 38 PCB Congeners in the
Funnel Rinse of the RM Sampler on Julian Date 195, 1986
23. Percentage of (A) 17 PAHs and (B) 38 PCB Congeners in the 1st and
2nd Bottles of the SM Sampler on Julian Date 190, 1986
24. Percentage of (A) 17 PAHs and (B) 38 PCB Congeners in the Resin
Adsorbent, Filter and Funnel Rinse of the FM Sampler on
Julian Date 195, 1986.
-------
LIST OF TABLES
1. Operational Variables of Precipitation Samplers
2. Selective Ion Monitoring Program for PAH on GC-MSD
3. Chromatographic Conditions for Analysis of PAHs and
Chlorinated Hydrocarbons
4. PCB Congener Numbers and Structures
5. Organic Compounds Analyzed in Wet-Only Precipitation
6. Summary of Analytical Quality Control Data
7. Recovery of Procedural Surrogates
8. Rain Sampler Collection Efficiency
9. Volume-Weighted Mean Concentrations for Fourteen Organic
Compounds in Wet-Only Precipitation - 1986
10. ANOVA Comparison of Fourteen Compounds in Four
Precipitation Samplers
11. Percentage of Compound Mass in Funnel Rinse
12. Percentage of Total Compound Mass on the Filter (FM Sampler)
13. Problems in Field Operation
-------
SUMMARY
The atmosphere is recognized as an important contributor of
organic contaminants to oceanic and lacustrine environments. In the
case of the Laurentian Great Lakes, it has been estimated that
atmospheric deposition represents an important if not dominant
fraction of total inputs of many chemicals from all sources.
Atmospheric deposition results from wet and dry inputs. The
objectives of this project were to perform a field intercomparison of
four wet-only precipitation samplers in an assessment of their ability
to efficiently collect rain and selected organic contaminants. The
samplers are evaluated and compared on the basis of their ability to
efficiently collect rainfall, exhibit mechanical reliability,
demonstrate adequate operational characteristics and provide precise
measures of wet-only inputs. The samplers differed in collection
surface area (0.08 to 0.21 m2), type of collection surface (stainless
steel; Teflon coated), mode of organic compound isolation (resin
adsorbent; batch extraction) and operational characteristics. We
found that the most significant difference between the four samplers
was their mechanical reliability in the field. The samplers performed
equally well in assessing organic concentrations in rain.
The four samplers were deployed in May 1986 at a site about 50 km
northwest of the Twin Cities. During the 1986 field season, wet-only
precipitation samplers collected rainfall integrated over periods of 5
-------
to 20 days (average - 14 days). Samples were analyzed for
approximately 60 organic contaminants including polycyclic aromatic
hydrocarbons (PAHs) and chlorinated hydrocarbons including
polychlorinated biphenyls (PCBs), chlorinated pesticides and
chlorinated benzenes. Fourteen of these compounds selected on the
basis of differing physical-chemical properties and occurrence in
rainfall were used as the basis of the intercomparison.
The four rain samplers captured from 82 to 90% of the rain gauge
precipitation. When periods of sampler malfunctioning were removed,
all samplers collected at ~ 95% efficiency. Problems encountered
varied from blown fuses and loose bolts on the movable arm to
vandalism. The key to proper collection of precipitation is to
monitor and properly maintain all samplers.
Volume-weighted mean (VWM) concentrations of the fourteen
compounds and propagated errors for each were calculated. The simple
criterion applied to the question of whether the samplers behaved
differently was whether the error bars overlapped for individual
compounds observed for different samplers. With few exceptions, there
was little or no significant difference between the four samplers
based on VWMs. For all compounds, the samplers containing adsorbent
resins provided an average 5 to 10% higher VWM concentration than
observed for the sampler using batch extraction. A one-way ANOVA
comparison of the 14 compounds was conducted to test the null
hypothesis that there is no difference between sampler behavior based
on event-to-event variations. With few exceptions, these data support
the null hypothesis that there is no or little significant difference
-------
in the collection efficiency of these 14 compounds. Where differences
were significant, no consistent pattern emerged.
One characteristic exhibited by all compounds and all samplers is
the retention of organic contaminants on the collection surface and/or
sampling train. For example, an average of 26% of the PAH and 40% of
the £ PCB mass in the sample occurred in the solvent rinse of the
funnel and sampler train. Although not a problem in assessing total
concentrations, this phenomenon makes the determination of speciation
in rain samples and of atmospheric removal processes all but
impossible. Collection of total compound mass requires the rinsing of
funnel surfaces with solvent and analyzing the rinse with the sample.
The intercomparison of wet-only, integrating rain samplers was
conducted in part to select the preferred characteristics of a rain
sampler that must be deployed in the field unattended for up to two
weeks. The MIC sampler, properly maintained, is suitable
for such a purpose. Of the two modes of compound isolation tested,
the resin adsorbent (XAD-2) exhibited modestly higher concentrations
than the solvent MIC but had the disadvantage of higher blanks and
clogging. Alternatively, the solvent MIC sampler had the advantage of
ease of sample handling and lower blanks. Both could be operated with
proper maintenance to provide precise data. The stainless steel and
Teflon coated funnel surfaces provided comparable data.
Three of the rain samplers had equal collection surface areas of
0.21 m2. Side-by-side operation permitted comparison of sampler
variability. These data suggest that the expected uncertainty in
loading estimates of organic contaminants at trace concentrations may
be no better than « 20%.
-------
INTRODUCTION
The atmosphere is now recognized as an important contributor of
anthropogenic organic compounds to oceanic (Bidleman and Olney, 1974;
Atlas and Giam, 1981, 1985; Tanabe et al., 1983) and to freshwater
ecosystems (Murphy and Rzeszutko, 1977; Andren, 1983; Murphy, 1984;
Eisenreich et al., 1981; Strachan and Huneault, 1979; Strachan, 1984).
The Laurentian Great Lakes are recognized as being particularly
susceptible to atmospheric inputs of organic contaminants because they
are near and generally downwind of major/industrial centers, have
large surface area to basin area ratios and have long water residence
times (Eisenreich et al., 1981). Estimates of wet and dry deposition,
especially for contaminants such as polychorinated biphenyls (PCBs)
using input-output budgets show that the upper lakes (Superior,
Michigan, Huron) receive the majority of total inputs from the
atmosphere (Strachan and Eisenreich, 1987). The lower lakes (Erie,
Ontario) receive a lower but significant percentage of their total
input from atmospheric deposition. The organic contaminants-are
removed from the atmosphere and deposited on water by wet deposition
(rain, snow), dry particle deposition and vapor absorption at the air-
water interface. The latter two constituting dry deposition are at
best difficult to infer from environmental measurements and models
(Doskey and Andren, 1981; Slinn et al., 1978; Mackay et al., 1986).
However, precipitation inputs may be properly assessed using wet-only
integrating or event samplers of various designs (e.g., see Strachan
-------
and Huneault, 1984; Pankow et al., 1984) employing either in-situ
compound isolation or bulk water extraction. The Great Lakes
National Program Office (GLNPO) of the U.S. Environmental Protection
Agency (EPA) seeks to modify its existing atmospheric monitoring
network to include measurement of atmospheric inputs of organic
contaminants (Eisenreich, 1986; Murphy, 1987). To that end, the
objective of this study was to intercompare four precipitation
samplers co-located at a site 50 km north-northwest of Minneapolis,
Minnesota. It is anticipated that precipitation samplers will be
deployed in the modified network for periods up to two weeks and must
be capable of unattended operation. The samplers are evaluated and
compared on the basis of their ability to efficiently collect
rainfall, exhibit mechanical reliability, demonstrate good operational
characteristics and provide precise measures of wet-only inputs of
selected organic contaminants. The samplers differed in collection
surface area (0.08 to 0.21 m2), type of collection surface (stainless-
steel; Teflon-coated), mode of compound isolation (resin adsorbent;
batch extraction) and other operational characteristics. Fourteen
compounds were selected out of about 60 compounds analyzed tor
determine the difference in sampler behavior. The atmospheric
processes governing wet deposition can influence collection results
and will be described here.
-------
Processes Governing Wet Deposition
The mechanisms of chemical removal from the atmosphere are very
different for particle associated compounds than for gas phase
compounds. The relative importance of these two processes depends on
the distribution of the organic compound between vapor and aerosol,
particle size distribution and Henry's Law constant (H). Non-reactive
organic gases will be scavenged by rain according to H if equilibrium
between the gas and aqueous phases is achieved (Slinn et al., 1978;
Ligocki et al., 1985). The overall resistance to vapor absorption by
rain is a result of resistances in the air phase, in the liquid phase
and a surface resistance (Peters, 1983). Assuming surface resistance
is negligible, air resistance depends on the relative velocity between
the raindrop and air-phase. The liquid-phase resistance depends on
molecular diffusion in the hydrometeor and internal circulation in the
droplet. In the absence of chemical reactions in the droplet, an
atmospheric gas should attain equilibrium with a falling raindrop in
about 10m of fall (Slinn et al., 1978; Scott, 1981; Ligocki et al.,
1985) . The position of equilibrium defined by H is a function of
temperature as it increases by about a factor of two for each 10°C
increase in temperature (Ligocki et al., 1985). For PCBs, H increases
by a factor of = 4 for each 10°C rise in temperature (Burkhard et al.,
1985). Perhaps more importantly, the H values decrease with falling
temperatures suggesting temperature-dependent changes in removal
efficiency from summer to winter seasons.
-------
The total extent of organic compound scavenging by falling rain
may be given as (Ligocki et al., 1985):
WT - Wg (1-*) + Wp(rf) (1)
where WT - overall scavenging efficiency of gases and particles by
hydrometeors
[rain,total]
WT (2)
[air, total]
Wg is the gas scavenging efficiency
[rain.diss]
Wg (3)
[air, gas]
Wp is the particle scavenging efficiency
[rain,particle]
Wp (4)
[air, particle]
and 4> is the fraction of the total atmospheric concentration occurring
in the particle phase.
An atmospheric organic vapor attaining equilibrium with a falling
raindrop is scavenged from the atmosphere inversely proportional to
H:
RT
Wg a (5)
H
where R is the universal gas constant (atm m3/mol °K), T is
temperature (°K), H - Henry's Law constant (atm m3/mol), and a -
solubility coefficient. Surface flux of a vapor-phase organic
compound removed by rain becomes
Fg - Q . p . Cg - Wg • P • Cg (6)
-------
where P is annual rainfall intensity and Cg is the concentration of
organic vapor in the atmosphere, also given by (1-<£)C Field
A X R , T
determined WT values are often substantially larger than Wg values
based on H for many organic compounds having Pv < ~ 10" 5 torr
suggesting particle scavenging by precipitation is an important flux
term. Slinn et al. , (1978), Scott (1981), Eisenreich et al. , (1981),
Bidleman and Foreman (1987), Peters (1983) and Atlas and Giam (1985)
present estimated Wg values. Ligocki et al., (1985) have recently
reported gas scavenging efficiencies for a variety of nonpolar organic
compounds measured in the field in Portland, OR. They compared
field-determined Wg values to those estimated from consideration of H
and ambient temperatures. Wg values ranged from 3 to 105
(dimensionless) for tetrachloroethylene and dibutylphthalate ,
respectively. Field Wg values (Wg) were calculated as:
[rain, dissolved] (ng/L)
[air, gas] (ng/m3)
Wg - (103L/m3) (7)
These values for Wg were underestimated by factors of 3 to 6 using H
data at 25°C applying the relationship provided earlier - Wg - a -
RT/H. Correcting published H values for ambient temperatures of 5 to
9°C, equilibrium between the atmospheric gas and dissolved constituent
in rain was demonstrated for several PAHs and other low MW compounds.
Based on these results, temperature-corrected Wg values (estimated
from H) may be used to estimate organic vapor concentrations in the
atmosphere, temperature-specific H values and/or wet vapor flux if
atmospheric vapor concentrations are known.
-------
Precipitation scavenging of particles containing sorbed organic
or inorganic species (Wp) permit the calculation of surface fluxes
(Ligocki et al., 1985):
Fp - Wp • P • Cp - Wp • P • CT (8)
where Fp is the wet particle flux and Cp is the concentration of
atmospheric particulate-bound species. Slinn et al., (1978), Gatz
(1975) and Slinn (1983) estimate Wp values for below-cloud scavenging
as « 103 to 10s for .01 to 1.0 /im particles. Scott (1981) suggests
that in-cloud scavenging may produce Wp values on the order of 10 .
Depending on particle size, precipitation intensity, and type of
meteorological event, Wp may range from 103 to 106. The higher value
implies that the aerosol is readily incorporated into cloud water and
is hygroscopic. The lower value implies a non-hygroscopic, probably
carbonaceous particle that is not readily incorporated into cloud
water. Aerosol collected over Lake Michigan was about 7 to 50%
organic carbon and sub-micrometer in size (Andren and Strand, 1981).
Slinn (1983) argues that even carbonaceous particles age into more
hygroscopic particles during transport. The relationship of
concentration in rain to precipitation intensity also implies whether
in-cloud or below-cloud scavenging is operative. In a convective
system, the cloud processes air and particles drawn into it. In this
way, the concentration of scavenged particles in rain reaching the
surface is independent of duration and amount of precipitation (Hicks,
1986). Below-cloud scavenging of particles by rain reduces the number
of particles below the cloud, and additional rain dilutes the
concentration of previously deposited chemicals in rain.
-------
10
The most comprehensive study of particle-bound chemical
scavenging by precipitation was conducted by Ligocki et al. (1985).
They list Wp values of 102 to 105 for a series of PAHs, alkanes and
phthalates. In general, Wp values are consistent with below-cloud and
in-cloud scavenging for PAHs, in which the compounds with higher
scavenging ratios were associated more frequently with large
particles. Particle scavenging ratios for trace metals as reported by
Gatz (1975), Talbot and Andren (1983), Settle and Paterson (1982),
Slinn (1983), Arimoto and Duce (1987) and others are on the order of
10* to 106. To adequately predict Wp, detailed information on
particle-size distribution, atmospheric concentrations in particle-
size ranges and detailed meteorological parameters are needed.
The total wet surface flux (vapor + particle) of organic
compounds in the atmosphere may be estimated from:
FT.W - WT ' c.ir - Wgd-«C.lr + Wp*Cair (9)
FT - Cair(Wg(l-4)) + Wp^ (10)
RT
FT d-^Cair + WP*Cair <1]->
H
Total wet atmospheric flux (FT w) is calculated as the sum of vapor
and particle contributions. The relative importance of each depends
in part on the distribution of the organic compound in the atmosphere
between the vapor and particle phases. The research of Junge (1977),
Yamasaki et al. (1982), Bidleman and Foreman (1982), and Pankow (1987)
provide a theoretical and predictive framework for estimating
atmospheric distributions. Research thus far indicates that PCBs,
DDT, low molecular weight (MW) hydrocarbons and low MW PAHs exist
-------
11
primarily in the vapor phase in "clean" or rural airsheds while higher
MW PAHs, PCBs and dioxins occur primarily in the particle phase. In
"dirty" or urban/industrial airsheds, a greater fraction of the total
atmospheric burden for a particular chemical will occur in the
particle phase. Wet-only sampling systems must be capable of
efficiently collecting both dissolved and particulate species in
rainfall to accurately assess total wet flux.
EXPERIMENTAL SECTION
Site Description
The precipitation samplers were deployed at the Cedar Creek
Natural History Area (CCNHA) in northern Anoka County, east-central
Minnesota (45°25' N, 93°10' W). The Natural History Area (Figure 1)
consists of 5460 acres of abandoned agricultural fields, uplands,
wetlands and lakes. The site was first under the auspices of the
Minnesota Academy of Science and given to the University of Minnesota
for use as a field research area.
The precipitation samplers were located on the western edge of
the CCNHA in a flat, grass-covered field. The field is flanked by
wooded areas to the north, east and south with private land, mostly
cultivated fields to the west. The nearest trees are = 40 m northwest
of the sampling platform. The approach to the site is from the south
by a one lane dirt road. State Highway 65 runs = 1.6 km to the west
and Anoka County Highway 24 runs = 0.8 km to the south.
-------
(>dnr Creek Natural History Area
MINNESOTA
to MN
A
N
X sampling site
|~~) open field
^IH wooded
— cultivated field
j brush-covered
___ dirt road
i paved road
Figure 1. Location of Precipitation Samplers
-------
13
The samplers were deployed on two Cedar decks about 0.6m above
the ground. They are parallel to the prevailing wind direction
(westerly) with the funnels upwind (Figure 2) and are = 2 m apart.
All trees or structures in the area subtend at an angle < 15 degrees
with the horizon. Two 4-inch rain gauges are located on diagonally-
opposite ends of the platform at the same height as the collector
funnels.
Sampler Description
All precipitation samplers used in this study are integrating,
wet-only precipitation collectors and their characterists are given in
Table 1. Three of the precipitation samplers were constructed by
M.I.C. Co. (Thornhill, Ontario) and individually modified in their
mode of organics isolation, type of collection surface, enclosure of
the sample compartment for all-weather adaptation and in-line
filtration. Two other samplers constructed by Aerochem Metrics
(Bushnell, FL) were also deployed. One was modified for in-situ
isolation of organic compounds in rain using XAD-2 resin and the other
was used for the collection of inorganic components and organic carbon
in rain.
All of the MIC collectors have a square funnel and a cover
constructed of stainless steel and have collection surface areas of
about 0.21 m2. The funnel is 0.46 m on a side with a surface slope of
20 degrees toward the center and a 10 cm vertical lip. The legs and
chassis are constructed of 0.3 cm cast aluminum. With the cover
closed, sampler dimensions are: 1.2m high x 1.0 m long x 0.5 m wide.
A stainless steel screen covers the horizontal surface next to the
-------
N
2.0 m
LJ\
FM
<*>
N'
2.4 m
I 1
1 m
Figure 2. Layout of Precipitation Collectors at Cedar Creek
-------
Table 1
Operational Variables of Precipitation Samplers
Parameter
Collection Surface
surface area
(sq. cm)
surface material
shape
Collection Efficiency
(ave. %) overall
Mode of Organics
Isolation
Reservior
Filter
Weather Suitability
Collection Capacity
Solvent MIC
(SM)
2060
stainless
steel
square
90.4
solvent
extraction
no
no
all-weather
8 L
Resin MIC
(RM)
2040
teflon
square
92.2
XAD-2
resin
yes
no
warm only
20 L
Filter MIC
(FM)
2060
teflon
square
88.7
XAD-2
resin
no
yes
warm only
20 L
Resin Aerochem
(RA)
814
stainless
steel
circular
81.6
XAD-2
resin
yes
no
warm only
20 L
-------
16
funnel to reduce rain splash into the funnel. A moisture sensor
consisting of two faces at 20 ° from the horizontal is held by an arm
slightly above and 0.5 m to the side of the sampler. The sensor
controls the automated opening and closing of the cover and is driven
by a 1/50 hp gear-motor. The electronics and motor are enclosed in
the compartment next to the funnel. The basic MIC sampler has been
described in detail by Strachan and Huneault (1984).
One MIC sampler was modified by Chan and Perkins (1986) to
include compound isolation by solvent extraction (SM sampler) in the
following manner. Organic compounds are isolated in-situ by passive
solvent extraction rather than resin adsorption (Figure 3). The lower
part of the sampler is enclosed, insulated with 1/2 inch styrofoam and
warmed by two thermostatically-controlled heaters during cold-weather
sampling. This sampler has a stainless steel funnel surface with
heating cable attached to the underside of the funnel. Precipitation
flows from the funnel through 1/4 inch Teflon tubing into a four litre
solvent bottle. A short section of the tubing leading to the bottle
is flexible and can be closed by a pinch clamp controlled by a
solenoid valve; this isolates the sample from the atmosphere during
dry periods. The tubing extends to the bottom of the bottle where
there is a 200 mL layer of dichloromethane (DCM) of 3-4 cm thickness.
The water must first pass through the DCM layer which is more dense
and remains on the bottom as the bottle fills. This serves as the
initial extraction step. Teflon tubing leads to a second bottle which
collects the overflow. Another solenoid valve controls venting and
flow between the two bottles. The total collection capacity is 8 L.
-------
17
MECHANICAL ARMS
5m i 0.5m
STAINLESS STEEL
FUNNEL
40 LITRE
SOLVENT BOTTLES
Figure 3. Schematic of Solvent MIC Sampler (SM)
(Chan and Perkins, 1986)
-------
18
When exceeded, water spills out of the second 4-L bottle through a
hole in the cap and the actual volume of precipitation collected is
unknown. The SM sampler was on loan to this project by C.H. Chan of
the National Water Research Institute, Burlington, Ontario.
The resin MIC sampler (RM) has a Teflon-coated funnel surface
having a collection area of 0.21 m2 (Figure 4). A two-litre glass
reservoir covered with aluminum foil is located directly below the
funnel. This is followed by a Teflon cartridge containing XAD-2
resin. The cartridge is made of 0.5 cm thick Teflon pipe, 15 cm long
with an inside diameter of 1.5 cm (Figure 5). The tubing is connected
by stainless-steel Swagelok fittings on either end. Glass wool plugs
on each end hold the resin in place. Water flows from the funnel into
the reservoir; since the flow through the resin cartridge averages
about 40 mL/min under gravity, the reservoir holds the sample to
ensure that it is not exposed to open air or direct sunlight before
extraction by the resin. After the water flows through the resin
cartridge, it drains into a 20-L plastic carboy where it is held for
volume measurement.
The Filter MIC sampler (FM) also has a Teflon-coated funnel
surface and has a collection area of 0.21 m2. The space below the
funnel is enclosed but not insulated or heated. The FM sampler is
similar in operation to the RM collector except there is an in-line
filter installed just above the resin cartridge. The type of filter
used was a 47 mm Gelman AE glass fibre filter held in a Nacom Teflon
filter assembly. The FM sampler was on-loan to this project by Maris
Lucis of the Ontario Ministry of the Environment, Toronto.
-------
19
Resin MIC Sampler
Precipitation Sensor
2 L Glass Reservoir
-20 L Nalgene Carboy
1.1 m
Figure 4. Schematic of Resin MIC Sampler (RM)
(Strachan and Huneault, 1984)
-------
20
to vent
Copper Vent Tube
to 2-Liter
Glass Reservoir
IT)
,A Stainless-steel
^L Fittings
il.Scmi
2.5cm
Teflon Cartridge
Stainless-steel
Fittings
1/4 in. 'Swagelok'x
Fittings Teflon Tubing
•\
to Collection Bottle
Figure 5. Schematic of X.AD-2 Resin Adsorbent Cartridge
-------
21
The Aerochem Metrics sampler (RA) is an automated wet-only
precipitation collector constructed of 0.4 cm thick aluminum. The
electric motor and mechanism for operating the cover are located
underneath the collection surface. Overall dimensions are: 1.3m high
to the top of the bucket or funnel, 0.4 m wide and 0.9 m long. The
moisture sensors have one face and are held closer to the body of the
collector than for the MIC samplers. The Aerochem Metrics sampler was
modified for in-situ organics isolation by resin extraction (similar
to Figure 4). A round stainless-steel funnel of 0.08 m2 surface area
was installed in place of the normal plastic bucket. The area below
the funnel is not enclosed. Collection and isolation are similar to
operation of the RM sampler. The other Aerochem Metrics sampler was
operated with a plastic bucket to collect precipitation for inorganic
ion and organic carbon analyses.
Sampling Protocol
Precipitation was collected in periods ranging from 5 to 30 days
depending on the amount of rainfall at the site (average - 14 days).
All of the sampling materials such as solvents, sample bottles, resin
cartridges and tools were transported to the site in a large, covered
polycarbonate tote box.
At the sampling site, the amount of precipitation collected by
the twin 4 inch rain gauges was measured and recorded to the nearest
0.01 inch. The rain samplers were then checked for standing water in
the funnels or reservoirs; if all water had not passed through the
resin cartridge, a peristaltic pump was installed in-line and the
standing water pumped through the cartridge and into the 20-L carboys.
-------
22
The SM sampler was checked for overflow; if the second bottle was
completely full, it was noted that sample overflow may have occurred
and an accurate measure of the sample volume was not possible.
The volume of water collected in the carboys attached to the
resin samplers was measured with a 2 L graduated cylinder, and the
water discarded after measuring. The volume collected in the SM
sampler bottles and inorganic bucket was measured later in the
laboratory.
The procedure for replacing the resin cartridge was similar for
all three resin samplers. The cartridge was disconnected, capped,
labeled and wrapped in aluminum foil. Then the funnel surface was
rinsed with 200 mL of solvent (either acetone or methanol) which was
collected in 250 mL amber glass bottles at the point where the
cartridge connects to the sampler. The sampler was flushed with 250
mL of Milli-Q water (not collected) and the replacement cartridge
installed. The resin was wetted by pouring 250 mL Milli-Q water into
the sampler funnel.
The SM sampler bottles were changed in a similar manner; the 4 L
amber glass bottles were disconnected and capped, and the funnel was
rinsed with solvent. Then the sampler was flushed with Milli-Q water,
the replacement bottles attached and the system wetted with 250 mL of
Milli-Q water.
The inorganic sampler required only replacing the sample bucket
with a clean, empty bucket and covering the first bucket with a tight-
fitting plastic lid. All of the samples (i.e., resin cartridges, SM
bottles, rinse bottles, inorganic species bucket) were labeled with
-------
23
sampler ID and date of removal from the sampler.
The motor drive for each MIC sampler was checked for proper
operation and loose connections; frequently the alien bolt holding the
drive chain sprocket to the motor shaft required tightening. Finally,
the moisture sensors for all samplers were activated to ensure correct
operation of the covers.
Analytical Procedure
All materials were thoroughly cleaned before use to avoid
contamination. The glassware was washed with soap and water, rinsed
with solvents and dried to 110 °C. Pesticide grade solvents
(Omnisolve, EM Science; High Purity Solvent, American Burdick and
Jackson) and cleaned XAD-2 resin gave low blanks which did not
interfere with analyte identification or quantification. The XAD-2
resin was cleaned in sequential 48 hour extractions using the solvent
series: methanol, acetone, hexane, petroleum ether, acetone, methanol
and then rinsed with and stored in Milli-Q water. Care was taken to
reproduce this procedure exactly as XAD-2 resin is known to present
difficulties in the analysis of chlorinated hydrocarbons by electron
capture gas chromatography. Florisil (60-100 mesh) was extracted for
24 hours in a Soxhlet apparatus with 1:1 (v/v) hexane/acetone,
activated at 550-600 °C overnight and stored at 110°C.
The overall analytical scheme shown in Figure 6 describes the
general procedures by which the polyaromatic hydrocarbons (PAHs) and
chlorinated hydrocarbons (CHs) in precipitation samples were analyzed.
The bulk rain samples containing DCM from the SM sampler and the XAD-2
resin from the RM, FM and RA samplers were spiked with surrogate
-------
FIGURE 6.
Schematic of Analytical Procedure for the Analysis of PAHs,
PCBs, Cl-Pesticides, and Cl-Benzenes.
SM
RAIH
Bulk
tfatar/DCM •
RA. RM. FM
Surrogate Spike ^ XAD-2 Realn
Adsorption
Cartridge
OCX
Separator?
Funnel
Extraction
MtOH/DCM
DCM Extract .
Soxhlet
Extraction
Back-Extraction
(Water)
25» Vol.
I.S. Splk«
Gas Blow-Doun (~200 uL)
Solvtnt Evaporation
(Kudurna-Danish)
Solvent Exchange
(Htxana)
Htxan* Concintrat*
^ I ^ 75> Vol
Org»nochlorln««
I
Clian-up/Fcmctionation
I FloriJll
60al HEX_
GC-MSD
HP5890 (GO
HP5970 (USD)
25n HRCC (HP19091B)
5% Ph«nyl Methyl Silicon*
JU
PCS*
Cl-
?«icicldas
50ml HEX:DEE
I (9:1)
Pescicldaa
I S. SpUa
=• J
olvent Evapocacion m^J
1:
Solvent Evaporatic
Kudurr.a- Danish
Cis 31ow-Dovm (LOO • 2CO -L)
CC-£C3
H.P5SM3
25m HRCC (HP19091B)
j» Phenyl Methyl Silicon*
-------
25
standards 1,4 dibromobenzene, mirex and d-10 anthracene to obtain
analytical recoveries. The SM water samples were extracted in a 5 L
separatory funnel with 400 mL additional DCM while the resin samples
were sequentially extracted for 24 hours with 1:1 methanol/DCM in a
Soxhlet apparatus. The methanol was then removed from the extracts by
back extraction with Milli-Q water. The field rinses of the sampler
surfaces were either analyzed separately (early in the project) or
incorporated into the rain extract. The resulting DCM extracts were
dried over anhydrous Na2S04 and concentrated to about 10 ml in a
Kurdurna-Danish apparatus with a concomitant solvent change to hexane.
The hexane extract was sub-sampled for PAH analysis removing about 25
% of the volume. The remaining extract containing the CHs for
analysis was fractionated on a 1.25 % (w/w) water-deactivated Florisil
column (13g; 2 cm i.d.) to remove interfering polar compounds and to
separate the compounds into a PCB and pesticide fraction. The column
was eluted with 60 mL hexane followed by 50 mL of 10 % (v/v) diethyl
ether in hexane. These two fractions were concentrated in a Kurduna-
Danish apparatus and then blown down under purified N2 to about 1 mL.
PAHs were quantified in the subsample without further cleanup by
isotope dilution gas chromatography - mass spectrometry (GC-MS). The
subsample was spiked with an internal standard containing d-8
naphthalene, d-10 phenanthrene, d-12 benzofa]anthracene, d-12
benzo[a]pyrene and d-12 benzo[ghi]perylene and concentrated to about
100 /iL with a gentle stream of purified N2. The PAHs were analyzed
using selective ion monitoring on a HP 5890 GC equipped with a HP 5970
Mass Selective Detector and a HP 59970B computer workstation. The GC-
-------
26
MSD was operated in the selective ion monitoring mode and scanned for
up to seven ions in one of five chromatographic windows (Table 2).
Each time window exhibited in Figure 7 and noted in Table 2 contained
an internal standard from which response factors were calculated. The
run time was about 20 minutes.
The organochlorine fractions were spiked with internal standards
2,3,5,6-tetrachlorobiphenyl and 2,2'3,4,5,6'-hexachlorobiphenyl, and
blown down with a gentle stream of N2 to about 100 jiL. The PCBs and
other CHs were analyzed on a HP 5840A GC equipped with a 63Ni electron
capture detector and a HP 7672A automatic sampler. Chromatographic
conditions for the PAHs and CHs are listed in Table 3.
Quantification of all CHs was based on standards of each of the
pesticides, PAHs, chlorinated benzenes and PCB congeners. The PCB
congeners were represented by those present in standard Aroclor
mixtures of 1242, 1254 and 1260. GC-ECD chromatograms of a standard
PCB mixture and a rain sample from the SM sampler are shown in Figure
8 detailing the identity of PCB congeners quantified in this study.
Table 4 relates IUPAC PCB congener numbers to structure. Capel et al.
(1985) list the PCB congener weight fraction in each of the standards.
The pesticide standard was a chlorinated pesticide mixture (Supelco)
to which hexachlorobenzene (HCB) and mirex were added. The PAH
standard was a mixture of 16 priority pollutant PAHs to which
benzo[e]pyrene was added (Supelco). Samples 6133-6190 were quantified
by externally comparing the area counts of the samples with those of
the standards. The chlorinated benzenes in all samples were
quantified by comparison to external standards (Ultra Scientific).
-------
27
Table 2
Selective Ion Monitoring program for PAH on GC/MSD
SIM
Group #
1
2
3
4
SIM
Ions
127,128,129,
152,153,154
136
165,166,167,
176,178,179,
188
200,202,203,
226,228,229,
240
126,250,252,
Quan
Compound Ion
Naphthalene
Acenaphthylene
Acenaphthene
D-8 Naphthalene
Fluorene
Phenanthrene
Anthracene
D-10 Phenanthrene
D-10 Anthracene
Fluoranthene
Pyrene
Benzo [ a ] anthracene
Chrysene
D-12 Benzo [a] anthracene
Benzo [b ] f luoranthene
tification
128
152
154
136
166
178
n
188
n
202
n
228
n
240
252
253,264 Benzo[k]fluoranthene "
Benzo[ejpyrene "
Benzo[a]pyrene "
D-12 Benzo[a]pyrene 264
138,276,277, Indeno[l,2,3-cd]pyrene 276
278,279,288 Benzo[ghijperylene
Dibenzo[ah]anthracene 278
D-12 Benzo[ghijperylene 288
-------
r\\j i
NAP
L
cea
ACE 4M
IBB
2BB
FL" PHEN ANT '"
\ |
11 u
H |T
n
jt
BE
BBF
»YR
BAA CHR |
. 1
1 1 BKF
1 1
aaa
P
IIB
•a
RAP tm
In
DBA
1
IDP |
_ K
ecu IP
IB II I) II
d 8 NAP
> i
d-10 PHEN
d-12 BAP
d-12 BAA
d-12 BCH1P
lM (Bin)
Figure 7. Chromatogram of 17 PAHs Using GC-Mass Selective Detector
ro
oo
-------
29
TABLE 3.
Chromatographic Conditions for Analysis of PAHs and Chlorinated
Hydrocarbons
COLUMN
Stationary Phase
Length
t.d.
Flln Thickness
TEMPERATURE PROGRAM
Initial Temp. (*C>
Initial Tine (Bin.)
Temp . Ramp ( *C/min)
Final Temp. CC)
Run Tin* (mln. )
CAS
Carrier
Makeup
Carrier Flow (mL/min)
Makeup Flow (mL/oin)
Injection Temp. (*C)
Detector Temp. (*C)
Detector Mode
5 % cross -linked phenyl
25 meters
0.31 mm
O.S2 ua
FCBc &
Pesticides
150
1.0
10 for 3 min.
1.3 to Final Temp.
275
70
Nitrogen
Nitrogen
0.5
30
250
325
ECD
methyl silicon. (Hewlett-Packard 19091S)
Chlorinated
Benzenes PAHs
50 125
3.0 0.5
10 for 2 mln. 20 co Final Temp.
2.5 for 20 Bin.
3 for 30 min.
20 for 5 min.
275 280
60 22
Nitrogen Heliua
Nitrogen N/A
0.5 ' 0.6
30 N/A
250 250
325 280
ECO Mass Spectrometer
Operated in Selective
Ion .".onltoring Mode
-------
FIGURE 8
Glass Capillary Gas Chromatograms with Electron Capture Detection of PCB Aroclor Mixtures
(1:1:1 Aroclor 1242:1254:1260)
(Sample SM 6265)
«T*,4I
«€>•,'«
-------
TABLE 4.
PCB Congener Numbers and Structures
31
(
Peak * (a) :
1
2
3
4
5
6
7
8
9
10
11
1 ")
a
13
14
15
16
17
18
Congener *
[UPAC (b)
8
18
17
16*. 32
31
28
33
22
52
49
47.48
44
37,42
41*. 64
74
70
66
60*, 56
(16)
(32)
(47)
(48)
(37)
(42)
(41)
(64)
V, w*» /
(60)
(56)
Structure
2.4'
2. 2'. 5
2,2' ,4
2,2' .3
2,4' ,6
2.4' ,5
2,4,4'
2'. 3, 4
2.3.4'
2.21 .5.5'
2,2' ,4,5'
2, 2', 4,4'
2,2' ,4,5
2,2' ,3,5'
3, 4^4'
2, 2', 3.4'
2.2' .3,4
2,3,4' ,6
2.4,4' ,5
2.3' .4' ,5
2,3' ,4,4'
2.3,4,4'
2.2' .3' .4'
Peak *
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
Congener #
IUPAC (b)
101
99
97
87
110
82
144
118*. 108
146
153
141
138
175
187*. 159
185
174
180
170
196
201
Structure
2, 2'. 4.5.5'
2.2' .4,4' .5
2.2' .3' .4.5
2,2',3.415'
2.3.3>.4'.6
2.2' .3.3' .4
2.2' .3. 4. 5'. 6
(118) 2. 3'. 4. 4', 5
(108) 2. 3, 3'. 4, 5'
2.2'. 3. 4'. 5.5'
2,2' ,3, 3'. 5, 5'
2. 2'. 3, 4. 5, 5'
2,2' ,3. 4. 4', 5'
2.2' .3. 3', 4. 5' .6
(187) 2.2'. 3, 4'. 5, 5', 6
(159) 2.3. 3'. 4.5. 5'
2. 2'. 3. 4. 5, 5'. 6
2,2' .3,3' ,4,5,6'
2,2' , 3. 4, 4', 5.5'
2,2' .3, 3', 4. 4', 5
2,2' .3.3- .4.4' ,5' .6
2, 2'. 3, 3', 4'. 5. 5'. 6
(a) Peak number corresponds to the chronological order of eluclon under
the chromacographic conditions used by this lab.
(b) If two congeners listed, they coelute with GC/ECD conditions of this lab.
Dominant peak indicated with "*", if neither peak dominates both congeners listed.
-------
32
The PCBs and pesticides beginning with sample period 6195 and all of
the PAHs were quantified using the internal standard method. Response
factors were calculated from calibration curves generated from
standards which had been spiked with the same internal standards and
run on the GC at the same time as the samples. Total PCB
concentrations were calculated as the sum of the 38 PCB congeners of
greatest weight fraction in Aroclor 1242, 1254 and 1260 appearing in
the rain. Table 5 lists the compounds detected and analyzed in rain
samples for this study. Fourteen compounds were selected for
intercomparison of the four organic rain samplers based on their range
of physical-chemical properties and their consistent appearance in
collected precipitation.
-------
33
TABLE 5
ORGANIC COMPOUNDS ANALYZED IN WET-ONLY PRECIPITATION
PAH
Naphthalene
Acenaphthylene
Acenaphthalene
Fluorene
*Phenanthrene
Anthracene
Fluoranthene
Pyrene
Benzo[a]anthracene
*Chrysene
Benzo[b]fluoranthene
Benzo(k]fluoranthene
Benzo[e]pyrene
*Benzo[a]pyrene
Indeno[1,2,3 -cd]pyrene
* Benzo[ghi]perylene
Dibenzo[ah]anthracene
*Total PAHs
CHLORINATED BENZENES
1,2-Dichlorobenzene
1,3-Dichlorbenzenc
1,4-Dichlorobenzene
* 1,2,4-Trichlorobenzene
PCBs
38 PCB Congeners
including
Congener 31
Congener 70
Congener 110
Congener 138
Congener 180
Total PCBs
2,4',5-TriCB
2.3',4',5-TetCB
2,3.3'4',6-PentaCB
2,2',3,4,4',5'-HxCB
2.2,',3,4,4',5.5'-
HeptaCB
CHLORINATED HYDROCARBONS
Hexachlorobenzene - HCB
a - Hexachlorocyclohexane - HCH
- Hexachlorohexane
S • Hexachlorohexane
Heptachlor
Heptachlor epoxide
Aldrin
Dieldrin
Endrin
p.p'-DDE
p,p'-DDD; o.p'-DDD
p,p'-DDT; o,p'-DDT
HCH
- Lindane
Compounds selected for intercomparison of samplers
-------
Quality Assurance/Quality Control
The analytical procedure was characterized as to detection limits
of the fourteen compounds, procedural blanks carried through the whole
scheme, the recovery of compounds spiked into the resin or extracts,
the recovery of surrogate compounds spiked into each sample and the
ability of the XAD-2 resins to recover compounds of interest.
sDetecti-sra limits (DL) may be statistically generated by repeated
analysis
-------
35
Table 6
Summary of Analytical Quality Control Data
Compound Avg. Blank Cone. "Working" DL
XAD-2 SM-DCM
--fno/'M
Avg. XAD-2
Recovery
/»\
{_ng/Li; \T>J
Phenanthrene
Chrysene
Benzo [ a ] pyrene
Benzo [ ghi ] perylene
Total PAHs
1 , 2 ,4-TriClBenzene
HCB
p,p'-DDE
Congener 31
Congener 70
Congener 110
Congener 138
Congener 180
Total PCBs
0.9
0.3
0.7
ND
1.9
0.8
0.027
0.016
0.03
0.022
0.017
0.015
0.015
0.43
0.3
0.3
ND
0.1
1.0
0.1
0.006
0.007
0.014
0.01
0.017
0.012
0.007
0.21
0.210.17
0 . 210 . 15
0.1210.13
0.3710.31
3.913.0
0.01310.007
0.005
0.02
0.00910.006
0.00810.005
0.00610.004
0.00510.004
0.008±0.005
0.2210.15
69
73
75
95
86
92
Blanks based on sample volume - 8.0 L.
-------
36
represents DCM taken to the field and returned for analysis in the
normal sample runs. Table 6 lists the contaminant blanks for the 14
compounds of interest for the resin and DCM matrices. In general, the
blanks for SM sampler using batch DCM extraction were = 50% of those
for the XAD-2 cartridges. The range in blank values for the SM
sampler were as follows: PAHs, 0.1 - 1.4 ng/L; PCBs, 0.001 - 0.04 ng/L
for individual congeners and 0.15 - 0.3 ng/L for total PCBs;
chlorinated pesticides, 0.001 - 0.5 ng/1; chlorinated benzenes, 0.02 -
0.6 ng/L. The blanks observed for the XAD-2 resins occurred in the
range: PAHs, 0.1 - 1.7 ng/L; PCBs, 0.001 - 0.2 ng/L for individual
congeners and 0.2 - 0.7 ng/L for total PCBs; chlorinated pesticides,
0.001 - 0.4 ng/L; chlorinated benzenes, 0.01 - 1.9 ng/L. These values
assume a sample volume of 8.0 L, the average volume collected by the
MIC sampler. For the RA sampler, multiply these blanks by 3.2 to
correspond to a sample volume of 2.5 L. The naphthalene blank on the
XAD-2 resin was large and precluded any quantification. In general,
blanks were less than 10 to 20% of the analyte in the sample, and
often were much less. The blank values were sufficiently variable to
preclude subtraction from analyte mass in the sample.
Surrogate compounds were spiked into the DCM extract of each SM
sample and each XAD-2 resin from the RM., FM and RA samples to monitor
procedural "recoveries. Table 7 lists the recovery of the 3 surrrogate
compounds applied to the solvent (SM), resin and filter samples. In
general, the surrogate recoveries averaged 60 to 80% over the course
of the project. Analyte concentrations were not corrected for sample
recoveries. Samples and standards analyzed repeatedly yielded average
-------
37
Table 7
Recovery of Procedural Surrogates
Compounds
d-10 Anthracene
1 , 4 - D ib r omob enz ene
Mirex (ext. std.)
Mirex (int. std.)
Solvent
(SM)
60±9
6519
64±10
54±17
XAD-2
(RA.RM.FM)
60±14
74±16
73+14
62119
Filter
(FM-F)
65±14
7219
66+6
68127
-------
33
relative precision of ±5-20% depending on compound and concentration.
The ability of XAD-2 resin cartridges to concentrate CHs and PAHs
from rain water was evaluated in laboratory experiments. The organic
compounds of interest were added to duplicate, clean, empty 20-L glass
carboys and the solvent allowed to evaporate. Then the carboys were
filled with Milli-Q water, covered with aluminum foil and allowed to
equilibrate for almost five days at room temperature. The carboy at
the outset contained =100 ng/L PCBs as Aroclor 1242, 5-50 ng/L of the
chlorinated benzenes and 50-100 ng/L of the PAHs. The loss of some
organic compounds by volatilization and adsorption to the glass walls
was expected. Prior to the running of the experiment, each spiked
carboy was sampled for determination of the water concentration of
each CH and PAH. The spiked solutions were pumped thro;ugjh XAD-2 resin
cartridges connected in series at flow rates of afe-out 35 to 5-0 mL/mln
in replicate runs. Recoveries of siarrogAtes spiked into tite water
were: 1,4 Dibromobenzene, 67 ± 9%; Mirwt,, 82 ± 10% and d-10
Anthracene, €9 ± 7%. The chlorinated benzenes were quantitatively
recovered in the ^experiments with about 3 to 8% occurring in the
backup XAD-2 column. PCB congeners in Aroclor 1242 were also
quantitatively recovered with less than 2-5% occurring in the backup
column. Five PAHs in the spiked solution were recovered in the range
of 62 - 100-% on the first column and quantitatively recovered in the
combination of sequential columns. HCB, Lindane and p,p'-DDD were
recovered quantitatively on the first XAD-2 column but chromatographic
interferences precluded the analysis of p,p'-DDE and a-HCH in the
experiments. These experiments show that XAD-2 resins in the
-------
39
configuration used in the field and slightly elevated concentrations
are effective at isolating the compounds of interest.
RESULTS AND DISCUSSION
Criteria for Precipitation Collectors
Criteria to consider in constructing, modifying or selecting
precipitation samplers for assessing atmospheric inputs of trace
contaminants have been summarized in several publications (Eisenreich
et al., 1980; Strachan and Huneault, 1984; Pankow et al., 1984;
Murphy, 1987). Wet-only, precipitation-activated collectors should be
utilized for wet deposition measurement. The sample receptors should
be protected from sedimentary, turbulent and gaseous inputs when it is
not raining, and should be constructed of carefully selected materials
so that inadvertent loss or addition of a trace chemical is minimized.
Depending upon particular pollutants of interest, materials such as
stainless steel and glass may not always be appropriate. In general,
the use of non-contaminating polymeric surfaces is recommended for
many applications. For trace organic compounds in rain, use of glass,
stainless-steel or Teflon-coated surfaces have generally been
recommended.
Ideally, precipitation collectors for assessing organic inputs
from the atmosphere should have the following characteristics:
1. Collects wet-only precipitation.
-------
2. Collector possesses large surface area (= 0.2 to 1.0 m2).
3. Collection/storage surface should be non-contaminating, non-
adsorbing and made of stainless-steel, glass or Teflon.
4. Collector has a fast responding and tightly-sealing
covering mechanism.
5. Collector should be suitable for unattended operation
over periods of 2 to 4 weeks.
6. Collector will undoubtedly require access to electricity.
7. Initial stages of the rain events must be collected.
8. Collector should be versatile such that organic compounds
differing in concentration, physical/chemical properties
and speciation in rain may be efficiently collected and
isolated.
The MIC and Aerochem Metrics samplers selected and modified for
study in this project differ in their surface area, type of collection
surface, mode of organic compound isolation and in-line filtration.
All the samplers collect wet-only precipitation as water-sensitive
sensors activate movement of the mechanical arm to open the collection
funnel. .The samplers are intercompared on the basis of rain
collection efficiency, mechanical and operational characteristics and
ability to provide precise estimates of organic concentrations and
fluxes.
Rain Collection Efficiency
The ability of wet-only precipitation samplers to efficiently
collect rain under environmental conditions is crucial to determining
accurate concentrations and fluxes. The samplers were deployed in the
field on 8 May, 1986 with the first rain sample collected on 13 May,
1986 (Julian date 133) and the last rain sample collected on 13
October, 1986 (Julian date 286). The samplers were deployed in early
autumn of 1985 only as a test run for the 1986 field season. The
rainfall measured with the two rain gauges averaged 55.1 cm with less
than 2% difference between gauges. A continuous recording rain gauge
-------
at a Minnesota Pollution Control Agency site located within 50 m
showed cumulative rainfall of 55 cm over the same period.
The collection efficiency of each sampler was determined by
comparison to the theoretical collection efficiency based on the
duplicate rain gauges located on the sampling platform. Figure 9
shows the rain collection efficiencies of the SM, RM, FM and RA rain
samplers compared to the theoretical value. Excluding the periods
when mechanical malfunctions or vandalism occurred, the average rain
collection efficiency was = 95% (Table 8). Including these anomolous
periods, collection efficiencies were reduced to 80 - 90%. The RA
sampler suffered from early problems with the precipitation sensor.
Once replaced, the rain collection efficiency increased from 83% prior
to the sensor being changed to 101% after. Large deviations from 100%
collection efficiency for all samplers resulted from blown fuses,
lessened nuts on rotating arm and evaporation of water when flow
through the resin cartridge was hindered. A special problem exhibited
for the SM sampler occurs in rainfall volumes exceeding 8 L. In this
case, excess water is lost to drainage without being measured. This
problem can easily be solved by increasing storage volume in the SM
sampler.
-------
120
c
o
33
o
o
o
133 163
285 286
SM
dote of tdrnpflno
+ RM O FW
RA
Figure 9. Rain Collection Efficiency of the SM, RM, FM, and 8A Wet-Otlly Rain Samplers
(a) carboy empty, cause unknown (b) moisture sensor malfunction (c) blown fuse
ro
-------
Table 8
Rain Sampler Collection Efficiency
Collection Efficiency Rain Sampler
SM RM FM RA
Overall Average
Excluding malfunctions
Standard Deviation
90
94
10
92
96
10
89
96
6
82
95
14
Figures 10-13 directly show the collection efficiencies by
sampler for each rain sample period in 1986. The upper plot
represents a comparison of collected to theoretical rain volume with
the line depicting a 1:1 relationship. Points plotted below the 1:1
line represent those time periods in which collected rain volume was
less than that inferred from the average of the two rain gauges. The
lower plot represents the positive or negative deviation of each
sampler volume compared to the theoretical for each period. The
overall sampler performance occurs in the decreasing order RM > SM >
FM > RA. We conclude that properly operating precipitation samplers
of the MIC type are effective in the collection of falling rain. The
differences in performance are attributed to malfunctions in
operation.
-------
3
1
Precipitation volumes, 1986
oolltoUd v«. theoretical
(h«er«Ueal vel (L)
+ Solvent MIC
3
1
20
19
18
17
18
15
14
13
12
11
10
8
a
7
8
a
4
3
• 2
1
0
Precipitation volumes
collected v«. theoretical
I
I
I
I
183 174 182 180
185 203 212
Selwnt MIC
223 237 285 285 288 318
ooUwted > ttworaUoo*
1771 eoOMtedvol
Figure 10. Comparison of Theoretical to Actual Collected Rain
Volume for the Solvent MIC (SM) Sampler
-------
2
1
18
IB -
17 -
16 -
IB -
14 -
13 -
12 -
11 -
10 -
• -
a -
7 -
a -
s -
4 -
3
Precipitation volumes, 1986
collected v«- theoretical
8 11
theoretical vol (L)
+ Rerin MIC
13
IB
17
19
2
I
20
It -
18 -
17 -
16 -
IB -
14
13 -
12 -
11
10 -
8 -
8 -
7 -
8 -
t -
4
3 -
2 -
1 -
0
Precipitation volumes
collected vi. theoretical
XX
133 163 174 182 180
IBB 203
R«ln MIC
I7"71
212 223 237 2SO 286 286
collected
Figure 11. Comparison of Theoretical to Actual Collected Rain
Volume for the Resin MIC (RM) Sampler
-------
!
i
19
18 -
17 -
16 -
15 -
14 -
13 -
12 -
II -
10 -
g -
* -
f -i
4 -
3-
2 -
1 -
0
Precipitation volumes, 1986
coKocted v*. ttMontlcd
* M
UworaUcal val (L)
+ FBUrUlC
14
18
3
*•*
i
20
IB
IB
17
16
15
14
13
12
<11
10
8
8
7
6
5
4
3
2
1
0
Precipitation volumes
collected n. UMoratlcd
^
1
//
i
^S
\
Jp
i i i i i i i
133 163 171 174 162 190 IBS
i
I
\
\
i
!
1
y/.
I
//
\
<
\
r i i i i i i
203 212 223 237 255 266 286
FIKw MIC
Z23
Figure 12. Comparison of Theoretical to Actual Collected Rain
Volume for the Filter MIC (FM) Sampler
-------
Precipitation volumes, 1986
cotoctedvi. UMoratleal
theoretical val (L)
+ Ruin
Precipitation volumes
eeltoeUd
-------
Organic Contaminant Collection
The ability of the four samplers to collect trace organic
compounds in rainfall was evaluated for fourteen specific compounds.
Table 9 presents the volume-weighted mean concentrations of the 14
compounds collected and analyzed, and also the propagated error for
each. The volume-weighted mean (VWM) concentration was calculated as
follows:
VWMi - Z *u/ I Vj - X (C.xVj)/ I Vj (12)
where VWMt is the volume-weighted concentration of compound i, M is
* J
the mass of compound i in the jth rain interval, V is the volume of
the jth rain interval and C1 is the concentration of compound i in
each rain interval. The propagated errors listed in Table 9 were
calculated in the method outlined by Shoemaker et al. (1974); a full
derivation of the technique is presented in the Appendix.
The 14 compounds differ in their physical-chemical properties,
speciation in the atmosphere, and concentration in collected rain.
Total PAH VWM concentrations range from 44.4 to 60.5 ng/L depending on
sampler and total PCBs ranged from 2.3 to 2.8 ng/L. The propagated
errors varied with compound and occurred generally in the range of 10
to 30%.
The primary question addressed by this data set is whether VWM
concentrations of each of 14 compounds observed in the SM, RM, FM and
RA samplers were significantly different. A related question is
whether any significant differences could be attributed to
characteristics of the rain samplers and/or compounds. Given the
nature of the data set, the approach chosen to answer these questions
-------
Table 9
Volume-Weighted Mean Concentrations for Fourteen Organic
Compounds in Wet-Only Precipitation - 1986
(ng/L)
Compound
Phenanthrene
Chrysene
Benzo[ajpyrene
Benzo [ ghi ] perylene
Total PAHs
1,2,4-Trichloro
benzene
Hexachlorobenzene
p,p'-DDE
Total PCBs
Congener 31
Congener 70
Congener 110
Congener 138
Congener 180 0
SM
8 . 6±2 . 1
6.1+1.1
1.410.4
2 . 2±1 . 0
53.9
11.7±1.2
0.571.07
1.11.46
2 . 610 . 6
0.161.03
0.161.05
0.151.04
0.051.02
.0241.005
Sampler
RM FM
15.413.8
6.511.2
3 . 311 . 0
1 . 910 . 9
60.5
4 . 110 . 4
0.181.02
2.11.85
2 . 310 . 5
0.121.02
0.091.03
0.111.03
0.061.02
0.0311.006
8 . 812 . 2
5 . 110 . 9
2.910.9
1.410.7
44.4
10.711.1
0.121.02
0.331.13
2 . 710 . 6
0.191.04
0.091.03
0.0910.2
0.091.03
0.071.02
RA
11 . 212 . 8
5.611.0
3.511.1
1.510.7
54.2
10.711.1
0.481.06
0.921.38
2 . 810 . 6
0.131.03
0 . 141 . 04
0.111.03
0.041.01
0.081.02
-------
50
was to calculate the volume-weighted mean concentrations and
associated propogated errors. Parameters used to estimate the
propagated error were uncertainties in the measurement of rain volume,
determined masses of individual compounds, and precision of analytical
measurements. The simple criterion applied to the primary question is
whether the error bars calculated as propagated error overlapped for
individual compounds observed for different samplers,
fl^Bare JA ipitesaeats tthe VWM -concentrations and propagated errors
for fSbeaaaaatifeurene,, itifecrysffine, 'benzol a1]'py^aesie and benzo [ ghi ] perylene in
the ;SM, EM, 134 amd HA samplers. In general, t3ae error bars for -the
VWM concentrations of the compounds overlapped exhibiting no
significant differences. The only possible exception is
benzo[a]pyrene in the SM sampler exhibiting a VWM concentration about
50% of those observed for the other samplers. The primary difference
in the SM sampler compared to the others is the mode of compound
isolation; the SM sampler employs passive batch extraction with DCM in
the field whereas the others employ XAD-2 resin cartridges. The SM
has a stainless-steel surface (same as the RA sampler) which-did not
exhibit a widely different VWM concentration. Neither lost rain
volume nor occurrence of benzo[a]pyrene explain the difference since
substantial variations were not observed for other compounds.
Figure" 15 presents the VWM concentrations and propagated errors
for a group of chlorinated hydrocarbons including 1,2,4-
trichlorobenzene (TriCB), hexachlorobenzene (HCB), p,p'-DDE, two PCB
congeners having 4 and 5 chlorines and total PCBs. Again the VWM
concentrations for most of the compounds and samplers are not
-------
w VWM Concentrati
» P N
^ b in b
•P-
M< S-
n o
0 C
ol: 2'
Hi (D
H*
°"
S 5? z'
a.
!s «
up 5 .
H»
o c"i
1 .
» O -1
00
2
N
O^
"S"
5
i
a
" s
ft O
a- (D o M «/>
fl)0 b in 0
rt {
W H
•? rt s.
w rr a* *
M-
50 O
« CO
S? s"
5
5 ^-i. r*
pj 3 _
P o-
W »TJ
IS ' 5"
dT
>-+-i
>-+-*
00
0
N
S"
IT
1
(0 ft
on (ng/L;
000
B-
2-
2"
5-
•
•-«-«
i a i
i — a — i
TO
M
rr
i
0
«*
O
2"
i-
2-
5-
.
*' ' • '" *
—
i • i
1 * '
0
M
S
-------
1,2,4-TCB
HCB
fa -
10-
5
i i
0
1.0-
0.5-
I j
8
0
SM RM FM RA SM RM FM RA
hJ
C
C
0
u
(8 2-
Vi
4J
C
0)
o
c
£ 0-
,
i
1
SM R
s
3
>
P'P'*DDE 2,3',4',5-TetCB
i
1 1
9
u.*i
0.2-
T
1 i i f
M FM RA SM RM FM RA
2,3,3',4',6-PentaCB Tota| pees
0.1 -
0.0-
I
I ;
I 1
4 -
2 •
n .
1(11
SM RM FM RA
SM RM FM RA
Figure 15. Volume-Weighted Mean Concentrations (VWM) and Propagated
Errors of Six Chlorinated Hydrocarbons for the SM, KM,
FM and RA Samplers
-------
53
significantly different. Exceptions to this rule are TriCB in the RM
sampler and HCB for which VWM concentrations stratify into two groups
differing in concentration by a factor of = 2. The resin MIC sampler
occassionally had standing water in the funnel due to a backup in the
resin column. This may have resulted in enhanced losses of TriCB by
volatilization. The FM sampler, identical to the RM sampler except
for the presence of an in-line filter, does not exhibit similarly low
values. The low values of RM and FM compared to the SM and RA
samplers may be due to the clogging problem noted above permitting
losses to volatilization. Losses to the collector surface cannot give
the same results since the surface was rinsed and the rinse analyzed
with the resin. The RM and FM samplers both have Teflon surfaces
which may in some cases cause problems with sorption. Compounds
having lower aqueous solubilities and vapor pressures might be
expected to exhibit more severe sorption problems; they apparently do
not.
The VWM concentrations for p,p'-DDE, 2,3',4',5-TetCB,
2,3,3',4',6'-PentaCB and total PCBs across samplers are not .
significantly different. The lower VWM concentration for p,p'-DDE was
used in the sampler intercomparison since only one very high value
contributed to the difference in VWMs.
These"results are consistent with the hypothesis that the SM, RM,
FM and RA samplers are equally capable of providing comparable
concentration data when expressed as VWMs. We suggest that
significant differences in the VWMs between samplers may be mostly due
to rain water being trapped in the funnel rather than the rain
-------
reservoir and subjected to volatilization. This problem is largely
restricted to the resin samplers. Future samplers employing resin
cartridges should be equipped with small in-line peristaltic pumps
linked to the funnel covering mechanism or activated by a flow or
volume sensor.
Volume-weighted mean concentrations were calculated for all PAHs,
PCBs and chlorinated benzenes and pesticides analyzed in 1986 rain.
Figure 16A shows a comparison between the VWMs of 16 PAHs for the
samplers with XAD-2 resins as the mode of compound isolation. There
is generally less than a 20% difference between the VWM concentrations
of the RM.FM and RA samplers. The biggest difference in VWMs occurs
for phenanthrene where the RM, FM and RA samplers give concentrations
of 15.3, 8.8 and 11.4 ng/L. Differences are not attributed to the
relative concentrations of PAH in rain or physical-chemical properties
of the compounds. Figure 16B shows a comparison of the VWM of the
total sample population and the values exhibited for the solvent MIC
or SM sampler. The agreement in all cases is very good with the
resin-based samplers providing on average 5-10% higher VWM
concentration compared to the SM sampler. No statistical difference
was observed in this comparison between the solvent and resin samplers
for isolating PAHs from rain.
Figure 17 compares the VWM concentrations of 27 PCB congeners
ranging from 2 to 6 chlorines in the samplers employing only the resin
cartridges (RM, FM, RA). Observed VWM concentrations range from ~
0.01 - 0.03 ng/L for congener 201 to 0.1 - 0.26 ng/L for congener 33.
In general, the FM and RA samplers exhibit an equal number of highest
-------
55
ACY ACE FU» PHEN AWT FU4 PY»» BAA
[771 FtUr U1C
IX XI
A«i uuh«rn
Rmh UC
(B)
11 -
to -
9 -
a -
7 -
a -
s -
4 -
3 -
2 -i
_
~
y
y
/
/
/
y
y
y
y
y
y
/
y
y
y
y
/
y
/
y
y
y
/
\
\
\
\
\
\
\
\
\
\
\
\
\
\
\
\
\
\
\
\
\
\
\
i-Fi n
PNrN
ww
7
/
/
/
s|
^
S
^
y
y
y
y
y
y1
/
/
/
/
/
y
y
/
/
/
f
f
y
\
\
\
\
\
\
\
\
\
\
V
s
\
s
\
s
s
y
\
s
s
s
V
-CO.
7
/
/
/
/
/
/
/
/
/
y
y
/
/
y
y
y
y
—
\
S
S
\
\
\
s
\
s
s
s
s
V
s
V
s
s
s
s
7
/
/
/
y
y
/
y
y
y
y
y
S
s
\
V
s
s
s
s
\
\
s
s
s,
s
/
/
/
/
/
-^
s
s
s
X
s
\
7
/
y
/
/
/
/
y
/
/
y
y
y
/
\
V
s
V
V
s
V
s
V
s
s,
s
V
n
S pr. PL
Y ^ .-. n i- ^ ^ i ^ ^
r ^ P N Pli r ^ ^ i ^ ^
r ^ r i KIN r ^ ^ i rrn ^ ^
HAP ACY ACE FU? PHEN AMTFmPYRBAACHRBeFBKFBEPaAPDPDBA BOHf
1771 Sohwot UIC
Told VWU
Figure 16. Comparison of Volume-Weighted Mean Concentrations (VWM)
for 17 PAHs Between the RM, FM and RA Samplers (A) and
Between the SM and All Samplers (B)
-------
o
c
o
u
0.3
0.28 -
0.26 -
1
0.24 -
0.22 -
0.18 -
0.16 -
0.14 - \
0.12 - '
0.1 -":
0.08 -
0.06 -,,
.04 -
002 - " '
o -H"
•
n
r
" n
,, , ,, ,
, M / V • '
1 « ' » ' •
1 1
•
,
, P . . .
' '- ' '
, , , , ,
, t * . .
I
'<
1
V
' F v
' P
I
>
- ,
•P
i
, c.
:[•
; n ;
, , ,.;
* • ' ' ' '
i
i r
•
*
, , , , ,
' FH
,,. . - r . - '
,, , , , , L, . >
"'. ' :[: '
,
"<
'•
• 4>
1
n
PI
PI PI
rl M R
r*FTl PTI
' rfu riu
1 1
n
N N n
"i 1 1 1 1 L|
in M M rri
LI Jill 1 M
tl n FN H Hi
Li 1 1 1 1 1 1 1 I kl
H 1 1 n 1 1 h 1 n IH
fJlfHItnm nu J.
HIIHIIHIlH lln fr
W IrH rrH rffl fVi rfri rr
Wl ml HI HJHl H i-Plnr
flWIlwflFfflFfflrfflJBftt
)|)li|ll
Ifl 17 16 31 28 33 22 44 37 41 74 70 66 60 101 99 97 110144118146
62 64 56 108 '
175 180 201
185 196
Congener
V7\ RM
r *, i J
Figure 17. Comparison of Volume-Weighted Mean Concentrations
(VWM) for 27 PCB Congeners for the RM, FM, and RA
Samplers
-------
57
VWMs and the RM sampler the highest number of low VWMs. This pattern
suggests that the resin MIC (RM) sampler is less effective than the
other resin samplers in isolating PCB congeners from rain. The reason
for this observation is not clear. Although there are 50 to 150%
differences in VWMs of individual PCB congeners across the resin
samplers, the values of total PCBs are not significantly different
(Figure 15). The RM sampler has a Teflon collection surface, a glass
reservoir below the funnel and an XAD-2 resin cartridge. The FM
sampler also has a Teflon collection surface and an XAD-2 resin
cartridge but no glass reservoir and an in-line filter. The RA
sampler has a stainless steel collection surface, a glass reservoir
and an XAD-2 resin cartridge. The lower collection efficiency of PCB
congeners for the RM sampler compared to the FM sampler may be due to
passage of some fine particles through the resin cartridge along with
increased retention by the filter/resin cartridge combination.
Surface effects and lack of retention and recovery of dissolved PCB
congeners by the resin cartridges apparently do not explain the
results.
Figure 18 gives a direct comparison between the VWMs of
individual PCB congeners for the SM sampler and the sum of all
samplers. There are an approximate equal number of PCB congeners for
which the SM sampler provides higher values as when it provides lower
values. The difference in VWMs between the two types of samplers
varies in general between 20 and 50%. All rain samplers show a
predominance of 2 through 5 chlorinated PCBs at the expense of low
quantities of higher chlorinated species. This pattern is similar to
-------
0>
c
u
c
o
o
03 --
0.28 -
0.26 -
0.24 -
0.22 -
0.2 -
0.18 -
O.16 -
0.14 -
0.12 -
0.1 -
0.08 -
0.06 -
0.04 -
OrtO _
.U^
0 -
/
'
/
/
s
' '
> /
> <
,
\ ,
x /
s /
\ /
\ f
- »
V '
s /
s /
s / ^
s / /
s / s '
S • N /
'
s ,
S »- > '
s / S /
s / S ^
v X1 S '
s / \ '
\ S \ '
^
s 0.49
>
s
>
._^
v '
'
1
S
f
/
/
-
\
\
\
'
/
^
n
N ^
r N t
/ \ /
s v •
/
/
'
s
<
-- r \ t
\ / \ /
\ f V '
^ ' \ /
\ ' V '
«a
V / s
N / S ^
\ / \ /
S / v f
S / \ ^
V /S /
s / S /
N / S C
" " rti '
V ' > / S '
v / S / V t
^
s
v
y
s
N
s
>
^
s
^
\ /
\ /
s
s
V
<
^
/
s
s
/
s
rli
m
1
'
/ /
/ ^
'1 '
/ s /
^ N /
-1
N ^^^ /
S / V / N / pi
S / S / N / /
s - - ^ firing
18 17 16 31 28 33 22 44 37 41 74 70 66 60 101 99 97 110144118146 175 180 201
32 42 64 56 108 138 185 196
[771 All Samplers
Congener
SM Samplor
Figure 18. Comparison of Volume-Weighted Mean Concentrations
(VWM) for 27 PCB Congeners Between the SM Sampler
and All Samplers
-------
59
that observed in air samples collected over Lake Superior during 1986
and proximate surface waters (J.E. Baker and S.J. Eisenreich,
University of Minnesota, unpublished results).
Comparisons of VWM concentrations of organic compounds collected
by the four samplers provides some insight as to their overall
behavior. The results presented thus far suggest that the organic
compound collection efficiency for the four samplers is not
significantly different in assessing annual values and atmospheric
fluxes. However it does not account for variations in compound
collection efficiency in the same rain periods. To evaluate this
phenomenon, the concentrations of selected PAHs and PCB congeners in
1986 rain events are presented followed by a one-way ANOVA comparison
of compound behavior. The question being addressed is: do event-to-
event variations in compound concentration show a statistically
significant difference in a sampler's ability to collect organic
compounds in rain.
Figure 19 shows event- to-event variations in rain concentration
of phenanthrene, chrysene, benzo[a]pyrene, benzo[ghi]perylene and £
PAHs for the four samplers in 1986. Phenanthrene dominated the PAH
distribution representing a relatively constant 20 to 25% of the
£ PAH while chrysene represented a constant 10% of the total. This
distribution is not unlike distributions observed in rain samples
collected on Isle Royale in Lake Superior (McVeety and Hites, 1987)
and in air over Lake Superior (Baker and Eisenreich, unpublished
results). Phenanthrene and benzo[ghi]perylene concentrations are
relatively uniform over the sampling period whereas the other PAHs
-------
60
(A)
SM
"1
H
-I
^
V
)• -
If •
• -
RM
i
\
\
\
\
1
{//
\
j
I/X
\
1
RA
^
1
I
1
1
i
1
1
^
1
1
w
1
^
pi
II
g
s
V
•
I
(B)
M
(•
11
• •
m -
«4 •
M -
""
ia -
• -
a i
14 -
M •
!• •
11 -
• *
4.
SM £*
vA
f'/l
^
^ ^A
'/A "A
tyfyv.
fyy/'//
^iSHfl
RM
iM
// //
// //
// //
y Y (7/1
FM
^
w
A
p ^
^ ^YAYA^AYY;
?;.
V
v\i
'
I
y
j
P7
-------
61
(C)
SM
RA
1T4 IM 1M IM m lit BJ U» U* M« 1M
" umplt dat«
RM
i
I
I
!
FM
i
1
i
i
P3
^
I
l
iJ
I
J
(D)
SM
1
1
iJ
J
S
^
!
i
i
%
1
i
FM
RA
RM
I
\
I
I
m ^
S
1
f ample date
-------
62
(E)
SM
'"-I
< -
RM
FM
1
RA
Wr*-.
IJJ l« IT4 10 1M IM 1M 119 CU Uf
•ampte datt
y,
^Y\
-------
63
exhibit increasing concentrations from spring to autumn.
Figure 20 show shows event-to-event variations in rain
concentrations of five PCB congeners and £ PCS for the four samplers
in 1986. There are relatively large variations in PCB congener
concentrations from sampler-to-sampler sometimes approaching 100% and
variation in concentrations over the sampling period within each
sampler. These results will be examined in a subsequent paper on
concentrations and fluxes of PAHs and chlorinated hydrocarbons in
rain.
A one-way ANOVA comparison of the fourteen selected compounds was
conducted to test the null hypothesis that there is no difference
between sampler behavior based on the event-to-event variations. The
ANOVA calculations on the log transformed data are presented in Table
10. The mean square in column 1 represents the variance between
samplers and the mean square of column 3 designated residual
represents the variance within samplers. The F value of column 5
represents the ratio of the variance of the sample means to the
variance of all samples. The last column presents the results of the
ANOVA with respect to the null hypothesis. If the calculated F value
is less than the "critical" F value (2.23 for 90% confidence; 3
degrees of freedom in means; 40 to 50 degrees of freedom in samples; F
- 2.84 for" 95% confidence), then there is a greater variation between
samples than samplers. Table 10 (last column) indicates that the
behavior of phenanthrene, benzo[a]pyrene, and PCB congeners 138 and
180 suggest a statistically significant difference in the four
samplers. For phenanthrene, the difference is attributed to the
-------
FIGURE 20.
Temporal Concentration Variations for Selected PCB Congeners
Observed in the SM, RM, FM, and RA Samplers,
(A) Congener
SM
y
"J"
,u4
RM
3 J
.1
tl
u
e
o
U
FM
-1
«3
I
j RA
'""-I
-3
(B) Congener #70
1
SM
1
\ / A
1 '/*
1 * -^ Y-*
\ 'A //
RH
1
|
1
ea i' 1
P//J A' X/
" ' - ' ' ••.«!!.
Ffo
S
1
^1 ^ ^
ut a* t» i«
!• !• IT4 !• IK I*
Sample date
Sample date
-------
FIGURE 20.
Temporal Concentration Variations for Selected PCB Congeners
Observed in the SM, RM, FM, and RA Samplers.
(C) Congener #110 (D) Congener #138
M
c
O
V
u
a
U
SM
RM
FM
III nj UT 114 !• 1M
Sample date
SM
RM
I
i
FM
ilk
F
1
i
O-JJ"
RA
IM tit n* a* in
Sample date
-------
66
FIGURE 20.
Temporal Concentration Variations for Selected PCB Congeners
Observed in the SM, RM, FM, and RA Samplers.
(E) Congener #180
SM
z
M
c
v
u
C
O
U
RM
FM
J p
-1 RA
UZZL
1
f/
-------
6?
TABLE 10.
ANOVA Comparison of Fourteen Compounds in Four Precipitation Samplers
Mean Square
Main Effect Degrees of Degrees of
Compound (Sampler) (a) Freedom Resldual(b) Freedom "F*
Phen
Chr
B(a)P
B(ghl)P
local PAHs
HCB
p.p-DDE
Congener *31
Congener *70
Congener *110
Congener *138
Congener *180
Total PCBs
1,2,4-TCB
0.905
1.236
6.983
0.73
0.323
0.955
3.199
0.26
1.603
0.854
0.959
4.158
0.262
4.132
3
3
3
3
3
3
3
3
3
3
3
3
3
3
0.274
2.415
1.185
0.588
0.527
1.127
1.882
0.759
0.742
0.528
0.315
0.539
0.361
2.857
39
39
39
39
39
38
38
45
46
46
43
42
46
46
Difference
between
1 Value(c) Samplers 7 (d)
3.307
0.512
5.895
1.241
0.614
0.847
1.7
0.342
2.159
1.617
3.049
7.708
0.726
1.446
yes
no
yes
no
no
no
no
no
no
no
yes
yes
no
no
NOTES:
(a) Mean Square represents the variance between sampler means - (Var.0)"2 + 4(Var.l)"2,
where (Var.l)A2 Is the variance between samplers.
(b) Mean Square represents the variance within all samples - (Var.O)A2.
(c) "F" Value represents the ratio of the variance between the sampler means to the
variance within all the samples.
For 90% Confidence interval the critical "F" value - 2.23
with 3 degrees of freedom in means and 40-50 degrees of freedom in samples.
For 95% Confidence Interval the critical "F" value is 2.84 for
the same degrees of freedom.
(d) The null hypothesis is that there is no difference between samplers [(Var.l)*2 - Oj.
which is true if calculated "F" value < critical "F" value.
This suggests that there is greater variation between samples than between
samplers.
If the calculated "F" value > critical "F" value the null hypothesis is rejected,
which signifies chat chere is a difference between the samplers.
-------
68
variations in the resin MIC sampler. For benzo[a]pyrene, the
difference is attributed to variations in the SM sampler. The
difference in collection behavior of PCB congeners is attributed to
variations in the RA sampler. In general, these data support the null
hypothesis that there is no or little significant difference in the
collection efficiency of these 14 selected compounds. Where
differences are significant, no consistent pattern emerges suggesting
analytical causes may be responsible for some of the observed
variation. The RA sampler has a collection surface area approximately
30% less than than the MIC samplers. Consequently less mass of
compound was available for each sample potentially increasing the
uncertainty of these measurements.
Sampling Protocols Affecting Collection of Organic Compounds
Interaction with Collection Surfaces
Falling rain comes into contact with the collection surfaces
(i.e., funnels) and downstream parts of the sampling train. The
funnels have coatings of either stainless steel (SM and RA samplers)
or Teflon (RM and FM samplers). The sampling train consists-of Teflon
tubing and glass bottles (SM), glass reservoirs under the funnel
followed by Teflon cartridges holding the XAD-2 resin (RM and RA) or
an in-line filter holder and resin cartridge both made of Teflon (FM).
AtmospherieT particles scavenged from the atmosphere and organic vapors
partitioned into the falling rain results in organic compounds being
either associated with atmospheric particles or dissolved in aqueous
solution. The particles may become attached to the collection surface
and the dissolved species may be sorbed to the sampling train. To
-------
69
evaluate this phenomenon, collection funnels and lower parts of the
sampling train were rinsed with acetone or methanol following removal
of the resin cartridges or rain bottles. For samples collected early
in the 1986 field season, funnel rinses were analyzed separately.
For the remaining field season, funnel rinses were combined with the
resin or rain water extract.
Surprisingly, a significant if not dominant fraction of the total
sample mass occurred in the funnel rinses irrespective of sampler and
organic compound. Figure 21 presents a plot of the percentage of
total sample mass occurring in the rinse for the fourteen selected
compounds and four samplers. Table 11 presents the mean percentage of
total sample mass occurring in the rinse for all organic compounds
analyzed.
Table 11
Percentage of Compound Mass in Funnel Rinse
Sampler
SM
RM
FM
RA
All Samplers
FAHs
26
27
24
28
26
FCB Congeners £ FCBs
8-63 26
22-72 62
12-47 28
13-77 37
23-55 40
Cl-Pesticides
17
21
20
15
18
Cl-Benzenes
25
32
19
31
28
-------
70
5
o
K
t>
60 -
50 -
40 -
30 -^
20 -
10 -i
o-i
/ S
/ \
V ' \
' \
N&J
M
''*
> \.7
PI
PHEN CHR BAP BGHIP HCB DDE C31 C70 C110 C138 C180 TC8
FM
RA
Figure 21. Percentage of 14 Selected PAHs and Chlorinated Hydrocarbons in
the Funnel Rinse for the SM, RM, FM and RA Samplers
-o
o
-------
71
The PAHs exhibit a constant fraction of total sample mass, 26%, in the
funnel rinse whereas PCBs exhibit widely varying amounts. The Cl-
pesticides and Cl-benzenes also show a relatively constant fraction of
their total sample mass to be in the rinse fraction. In all cases, a
large percentage of the total sample mass occurs in the rinse
independent of sampler, composition of funnel surface and specific
compound. The RM sampler shows PCB congeners to be highly
concentrated in the rinse of its sample train. The RM sampler has a
Teflon collection surface and a 2 L glass reservoir in the sample
train. The FM sampler is nearly identical except it has no glass
reservoir. By intercomparison of the sampler behavior and sampler
characteristics, the funnel contributes on average about 50% of the
total mass in the rinse and the glass reservoir the other 50%.
This phenomenon is further examined in Figure 22 which shows the
percentage of total sample mass in the rinse for PAHs (Figure 22A) and
PCB congeners (Figure 22B). The retention of PAH in the RM sampler
ranges from 10 to 50% and is independent of whether the compound is
thought to be dissolved (e.g., phenanthrene) or particulate (e.g.,
B[a]pyrene. The pattern of retention is more pronounced for the PCBs.
Figure 22B shows that nearly all PCB congeners are retained to some
extent with some approaching 100% retention in the funnel and glass
reservoir.-
The problems associated with the retention of organic compounds
in the funnel and sampling train may be easily solved by carefully
rinsing the surfaces with the appropriate wetting solvent. These
rinses may then be combined with the sample extract to yield an
-------
72
o
k
9
00
so -
40 -I
30
«>
N
S
f
X
^
/
s
p^
s
M n
s
>
l<
'H
V
^
\
i
-
-
X
^
X
'
X
'
'
^
'
^
s
s
"
,
'
p.
s
s
s
7
X
/
X
"
\
X
'
V
V
s
X
'<•
,
'
5nH
iRR
s
>
pi
T
s
V
7
/
X
X
V
N
\
S
V V
* V
'
X X
' '
R
3|
/ Hlxlb nKj
^ HrlP NIN
HMPRMNN r-,F
Congener #
R.iin
Figure 22. Percentage of (A) 17 PAHs and (B) 38 PCB
Congeners in the Funnel Rinse of the RM Sampler
on Julian Date 195, 1986
-------
73
accurate assessment of the concentration. However, the problem does
have significant implications with respect to quantifying atmospheric
removal processes. It will be impossible to determine either the
speciation of the organic compound in the collected rain or the form
of the compound in the atmosphere. No information may be extracted on
the mechanism with which particulate and gaseous species are removed
by rain. The mechanistic information cannot be obtained using wet-
only, integrating samplers. This type of information can only be
confidently obtained using event, wet-only samplers such as the Pankow
rain collector (Pankow et al., 1984) and taking special care to use
and monitor inert surfaces.
Efficiency of the Solvent MIC Sampler
The solvent MIC (or SM) sampler isolates the organic compounds in
rain using passive solvent extraction. The SM sampler has a Teflon
tube running from the funnel to the bottom of a sampling bottle. The
tube extends into a 200 mL layer of DCM which serves to isolate and
preserve organics. The efficiency of this procedure was tested by
collecting the contents of the first bottle and the contents.of the
reserve bottle and analyzing them separately for the first part of the
1986 study period. Figure 23 shows the fraction of the total mass of
each of 17 PAHs and numerous PCB congeners in each bottle for one
date. In general, the high MW PAHs presumably on particles were
effectively isolated by the first bottle. A lesser portion of the low
MW PAHs were isolated in the first bottle. On the order of 20 to 80%
of the PCB congeners were retained in the second bottle with the lower
MW species being least retained in the first bottle. Since DCM is
-------
1
ESI
u
u
1.1
I
u
OS
a?
as
0.4
OJ
at
o
s
s
s
s
s
s
V
s
s
s
7
f
t
s t
\ f
S X
^ J
. * ^
y '
t t.
f r
t t
\ \ \
?
^ S S
' - S
' / N
r i i
R N
/ i i i
s
s,
r
f
^^
I '<''<
*\i * s
$'/,
tiii
S
S
s
\
s
/ n
^ s
' s
/ W
;:g
/ ' '
/ / /
^
^
s
s
7
'
J
/
^
1
3
W
1 1 1
s
s
7
x
t
t
f
'
f
+
*
f
f
f
f
f
S
S
X
/
s
/
/
s
s
p
J 1
s
s
7
'
;
/
*t
1
s
r
^
^
'
s
/
^
'
I 1 I 1
^
s
s
s
s
'
'
/
^
nP
1 ' 1
S* —
* 0
ezi
Congener
BaMto #2
Figure 23. Percentage of (A) 17 PAHs and (B) 38 PCB
Congeners in the 1st and 2nd Bottles of the
SM Sampler on Julian Date 190, 1986
-------
75
soluble to the extent of 1.5% in water, some of the organic compounds
may have been solubilized and been transported into the second bottle.
Equally likely is that simple passage of the rain through the DCM may
not have been efficient. A solution to this problem is to place a
mixer in the bottom of the first bottle which is activated by the
moisture sensor. A larger water storage system is also needed to a
volume of 20 L.
Separation of Particulate and Dissolved Species
The FM sampler is equipped with a Teflon-coated collection
surface and an in-line filter assembly upstream of a Teflon cartridge
containing XAD-2 resin. The filter is used to separate the
particulate from the dissolved species so as to prevent resin
clogging, enhance collection of particulate organic species and to
provide data on atmospheric removal processes. As stated previously,
the collection funnel retains a significant portion of the total
sample mass of nearly all compounds tested and samplers tested. Since
both dissolved and particulate species seem to be retained to variable
degrees, it is unlikely that use of a filter is effective at-
establishing speciation. Figure 24 shows the percentage of the total
mass of 17 PAHs and 38 PCB congeners in the filter, rinse and resin
fractions for a single sample collected by the FM sampler. Of the
total mass^collected, approximately 10 to 30% of individual PAHs and
PCBs were isolated on the filter. Table 12 provides a summary for all
data collected.
-------
76
Miw*
1 -
0.9 -
0.8 -
0.7 -
0.6 -
0.5 -
0.4 -
0.3 -
0.2 -
0.1 -
2
3
7
^
V
/
/
/
/
-
7
\
/
/
-
7
',
'/
/
s
/
s
**.
r
^
S
S
S
\
r
/
',
~
\
p-
/
M "
a t
/ *
/•
^ ^
/
s t
s
/ J
i ,
\
/
^
/
s
/
t
B lei
~ x r^ \ /
/ L^ / /
v n/r'/'
' / H/k / /
S S 5 5 5 3 3 ^
:
1
\
\
7
. 2
'j
y
\
S
S
/
/
*
ft
^
r
S
\
/
'
§
BBS
£ » •
I
/
'y
3
^
nN
m 5
r
!
/
/
s
^
^
^
t
S
S 2
N
7
/
™
^
^
1 / KV\\\VO
/
^
s
R
s
/
s
S
\
S
r
^
?
\
S
S
s
£
1 p
^ s
_ V
/ 7
I ' I"'1
ssss
Congener
1771 R«ifl
Figure 24. Percentage of (A) 17 PAHs and (B) 38 PCS Congeners
in the Resin Adsorbent, Filter and Funnel Rinse
of the FM Sampler on Julian Date 195, 1986
-------
77
Table 12
Percentage of Total Compound Mass on the Filter
Sampler PAHs FOB Congeners £ FCBs Cl-Pesticides Cl-Benzenes
FM 30 9-67 29 18 27
One problem observed with this sampler in this study and also by Haris
Lucis is that the filter clogs when large quantities of soil dust or
pollen are scavenged from the atmosphere. In those instances, rain
water is trapped in the funnel. A solution to this problem is to
install an in-line peristaltic pump to assist in water passage through
the sampling train and to use a pre-filter. However it is not clear
whether filters can provide useful information in integrating
samplers.
Mechanical Reliability and Operational Characteristics
MECHANICAL RELIABILITY AND OPERATIONAL CHARACTERISTICS
In a monitoring network, a wet-only, integrating sampler must be
able to reliably operate unattended for periods of approximately two
weeks. AllT of the samplers demonstrated this ability or have the
potential to do so with proper maintenance, careful operation, or with
minor design modifications. During the course of this study, a variety
of problems were encountered which could adversely effect sample
quality. Table 13 lists the problems that occurred with each sampler
-------
78
Table 13
Problems in field operation
date SM RM
133 0
163 C F,C
174 0
182
190
195
203
212 C
223
237 R
255 0
265 0 R
286 R
FM
F
F
F
F
F
X
F
F.C.X
R,C
EA
7
S
R
R
R
C: cover open on arrival
F: standing water in funnel
0: overflow of collection bottles
R: standing water in reservior
S: moisture sensor malfunction
X: blown fuse
?: carboy empty, cause unknown
-------
79
by sampling date.
The MIC samplers exhibited the greater number of mechanical
difficulties. A funnel cover was found open on five occasions due to a
loose cotter pin (RM) or alien bolt (SM) on the motor shaft sprocket,
and twice a fuse blew (FM). This would allow dry deposition to be
incorporated into the sample. Conversely, no sample would be collected
if the cover failed to open, which did occur once. These malfunctions
are symptoms of the basic weakness of the MIC sampler; namely, the
mechanism of opening and closing the cover does not seem designed to
sustain the generated torque. The Aerochem Metrics sampler, which has
a weighted counterbalance for its cover, never exhibited this type of
mechanical failure. The only mechanical problem observed with the RA
sampler involved the moisture sensor, which became either insensitive
or slow to respond and was replaced.
Operational difficulties encountered could not be differentiated
according to manufacturer's design, but varied according to mode of
isolation: resin adsorption or solvent extraction. Since gravity flow
through the resin was approximately 40 mL/min, water would accumulate
in the reservoir or collection funnel during periods of intense or
prolonged rainfall. Standing water could allow volatilization to occur
prior to resin isolation, and enhance the possibility of adsorption to
sampling tfain surfaces. Debris such as leaf material or insects
occasionally clogged the fittings on the resin samplers and restricted
flow. The RA sampler exhibited few operational problems of this sort
due to the smaller volumes collected.
-------
80
Due to its design, the SM sampler did not experience any flow
restriction. However, since the sampling bottles have a bulk water
capacity of 8 L, overflow, probable loss of contaminant mass, and
inaccurate volume measurement occurred on four occasions when rainfall
exceeded 1.5 inches (3.9 cm). This diminished its ability to remain
unattended during especially wet periods of the year. In addition, the
lack of effective passive extraction by the solvent in the primary
sample bottle contributed to contaminant loss with overflow. A deeper
solvent layer with active stirring or agitation to enhance the contact
between solvent and water, and a larger volume capacity would increase
the efficiency of the extraction process and decrease overflow losses.
The headspace above the water in the sample bottles could lead to
evaporative losses or air/water exchange, although the pinch clamp
attempts to close the system to the atmosphere. The weight and bulk of
the sample bottles and presence of solvent could present transportation
problems and higher shipping costs in an extensive monitoring network.
Mechanically, the alien bolt on the motor shaft sprocket and the clutch
plate would loosen and required tightening with each sampling visit.
The advantages of the SM sampler include the sampling train design that
prevents flow restriction, contains less surface for adsorption, and
allows particles to be incorporated into the sample. The sampler,
which is he"ated and insulated, is suitable for all-weather operation.
Finally, the MIC sensor appeared to be superior in its sensitivity and
response to precipitation.
-------
81
The RM sampler suffered mechanical problems similar to the SM. In
this case, a cotter pin holding the sprocket to the motor shaft, a
modification to solve the alien bolt problem of the SM, was
periodically sheared off. Flow restriction due to the resin and
occasional plugs of insect or plant material caused water to back-up
into the reservoir or funnel. Material did adhere to the surface of
the glass reservoir but it was assumed that by rinsing the sampling
train with solvent the adsorbed organics were extracted. Resin
adsorption of particulate organics is questionable, since discolored
glass wool plugs at each end of the resin cartridge were noted and
suggested particle breakthrough. The beneficial characteristics of the
RM sampler include the large 20 L capacity; the MIC moisture sensor;
the lightweight and easily transported XAD-2 resin adsorption system;
and the aluminum foil covered reservoir, which serves to protect
compounds from photolysis and limit evaporative losses prior to resin
extraction.
The FM sampler did not exhibit any motor sprocket problems.
However, fuses did periodically need replacement which suggested a
short or strain on the electical system. The characteristics of the FM
are similar to the RM, except that it had an in-line filter and was
without a reservoir. The major problem encountered was associated with
the filter— During seven of the thirteen sampling periods, the FM
sampler had standing water in the funnel attributable to either plugged
filter pores or insufficient head pressure for proper flow. A
peristaltic pump activated by the moisture sensor or a volumetric
sensor would alleviate this problem.
-------
82
The RA sampler was the most mechanically reliable sampler
evaluated. The counterbalanced cover prevented the strain that
contributed to the MIC malfunctions. Disadvantages of the RA include
the moisture sensor, which did not seem as sensitive or responsive as
the MIC sensor, and the small funnel surface area. The collection of
less volume and contaminant mass can influence the detectability of
some trace organics. Since the sampling train was similar to the RM,
the problems and advantages discussed above also apply to the RA.
CONCLUSIONS
Four wet-only precipitation samplers were deployed at a site near
Minneapolis, MN in 1986 with the objective of comparing the ability of
each to provide precise estimates of trace organic concentrations and
fluxes in rain. The four samplers were evaluated and compared on the
basis of their ability to efficiently collect rainfall, exhibit
mechanical reliability, demonstrate adequate operational
characteristics and precise measures of wet-only inputs. The samplers
differed in collection surface area, type of collection surface, mode
of organic compound isolation and operational characteristics. The
samplers consisted of modified MIC rain collectors (Meterological
Instruments Co.) and a modified Aerochem Metrics collector.
-------
83
The overall conclusion of this study is that the four samplers
collected rain volumes and a mixture of chlorinated hydrocarbons and
polycyclic aromatic hydrocarbons equally well. The overall performance
of the samplers was more closely linked to mechanical and operational
characteristics than to ability to precisely assess wet-only
concentrations of organic compounds in rain.
The four samplers all collected 95% of rain gauge volume when
functioning properly. Mechanical malfunctions (blown fuses, loose
bolts, stuck arms) reduced rain collection efficiency to 82 to 90%.
Volume-weighted mean concentrations of 14 selected organic
compounds generally agreed within statistical propagated error limits.
One-way ANOVA comparison of the samplers using log-normalized
concentration data generally showed no difference in individual
sampler performance.
One characteristic exhibited by all samplers was the occurrence
of a significant (if not dominant) fraction of the total sample mass
in the solvent rinse of the funnel. This behavior occurred in all
samplers independent of whether the surface was stainless steel or
Teflon, but was exacerbated by use of a glass reservior. Attention to
this detail in the sampling protocol is important in assessing rain
concentrations. However this phenomenon makes determination of
atmospheric- removal processes and compound speciation all but
impossible.
The MIC sampler, having a surface area of 0.2 m2, is an
appropriate collector for unattended operation ( 2 weeks) in a
monitoring network. The use of XAD-2 resin cartridge or batch solvent
-------
extraction is adequate for characterizing concentrations of organic
compounds in rain. The former exhibits higher blanks but has the
advantage of in-situ isolation in the field. The MIC-solvent
combination requires incorporation of an active mixing process to be
more effective.
Although this project did not have as its objective to determine
the uncertainty in flux measurements, the propagated error within
samplers suggests that loading estimates of organic contaminants at
trace concentrations can be no less than ± 20%.
-------
85
ACKNOWLEDGMENTS
We thank C.H. Chan and Loran Perkins of the Canadian Centre for
Inland Waters for the loan of the solvent MIC sampler (SM) and Maris
Lucis of Canada's Department of Environment, Air Resources Branch for
the use of the filter MIC (FM) sampler. We are also indebted to our
colleagues Paul Capel for his help during the initial stages of
sampler and site development, and Joel Baker for his many helpful
discussions and assistance in developing the PAH analytical scheme.
Frank Martin of the University of Minnesota provided statistical
advise. Lastly, we express our appreciation to Rob Holzknecht for his
analytical contributions and laboratory operation.
-------
86
REFERENCES
1. Andren, A.W. Processes determining the flux of PCBs across the
air-water interface. In Physical Behavior of PCBs in the Great lakes.
D. Mackay; S. Paterson; S.J. Eisenreich; M. Simmons (Eds.), Ann Arbor
Science: Ann Arbor, MI, 1983. 127-140.
2. Andren, A.W.; Strand, J.W. Atmospheric deposition of particulate
organic carbon and PAHs to Lake Michigan. In Atmospheric Pollutants
in Natural Waters. S.J. Eisenreich (Ed.), Ann Arbor Science: Ann
Arbor, MI 1981. 459-479.
3. Arimoto, R.; Duce, R . Air-sea transfer of trace elements. In
Sources and Fates of Aquatic Pollutants. R.A, Kites; S.J. Eisenreich
(Ed.), Adv. Chem. Ser. # 216, American Chemical Society: Washington,
D.C., 1987. 131-150.
4. Atlas, E.; Giam, C.S. Global transport of organic pollutants:
ambient concentrations in the remote marine atmosphere. Science,
1981. 211, 163-165.
5. Atlas, E.; Giam, C.S. Sea-air exchange of high molecular weight
synthetic organic compounds. In The Role of Air-Sea Exchange in
Geochemical Cycling. P. Buat-Menard (Ed.), NATO ASI Series #185, D.
Reidel Publishing Co.: Dordrecht, 1985. 295-329.
6. Bidleman, T.F.; Olney, C.E. Chlorinated hydrocarbons in the
Sargasso Sea atmosphere and water. Science, 1974. 183, 517-518.
7. Bidleman T.F.; Foreman, W.T. Vapor-particle partitioning of semi-
volatile organic compounds. In Sources and Fates of Aquatic
-------
8?
Pollutants. R.A. Hites; S.J. Eisenreich (Eds.), Adv. Chem. Ser. #216,
American Chemical Society: Washington, D.C. 1987. 27-56.
8. Burkhard, L.P.; Armstrong, D.E.; Andren, A.W. Henry's Law
constants for the PCBs. Environ. Sci. Tech., 1985. 19, 590-96.
9. Chan, C.H.; Perkins, L. Wet deposition monitoring of
organochlorines and polychlorinated biphenyls in the Great Lakes.
Ontario Region Water Quality Branch, Environment Canada, Draft Report,
1986.
10. Doskey, P.V.; Andren, A.W. Modeling the flux of atmospheric PCBs
across the air-water interface. Environ. Sci. Tech., 1981. 15, 705.
11. Eisenreich, S.J.; Looney, B.B.; Thornton, J.D. Assessment of
airborne organic contaminants in the Great Lakes ecosystem. Science
Advisory Board, International Joint Commission, Windsor, Ontario,
1980. 150pp.
12. Eisenreich, S.J.; Looney, B.B.; Thornton, J.D. Airborne organic
contaminants in the Great lakes ecosystem. Environ. Sci. Tech., 1981.
15, 30-38.
13. Eisenreich, S.J. The chemical limnology of nonpolar organic
contaminants: PCBs in Lake Superior. In Sources and Fates of Aquatic
Pollutants. R.A. Hites; S.J. Eisenreich (Eds.), Adv. Chem. Ser. #216,
American Chemical Society: Washington, D.C., 1987. 393-469.
14. Gatz,-I).F. Pollutant aerosol deposition into southern Lake
Michigan. Water, Air and Soil Poll., 1975. 5, 239-251.
15. Hicks, B. NAPAP Workshop on Dry Deposition, Harper's Ferry, W.V.,
25-27 March, 1986.
-------
16. Junge, C.E. Basic considerations about trace constituents in the
atmosphere as related to the fate of global pollutants. In Fate of
Pollutants in the Air and Water Environment. Part I.. I.H. Suffet
(Ed.), Adv. Environ. Sci. Tech., Wiley-Interscience: New York, 1977.
7-25.
17. Ligocki, M.P.; Leuenberger, C.; Pankow, J.F. Trace organic
compounds in rain. II. Gas scavenging of neutral organic compounds.
Atmos. Environ., 1985a. 19, 1609-1617.
18. Ligocki, M.P.; Leuenberger, C.; Pankow, J.F. Trace organic
compounds in rain. III. Particle scavenging of neutral organic
compounds. Atmos. Environ., 1985b. 19, 1619-1626.
19. Mackay, D.; Paterson, S.; Schroeder, W.H. Model describing the
rate of transfer of organic chemicals between atmosphere and water.
Environ. Sci. Tech., 1986. 20, 810-816.
20. McVeety, B.D.; Kites, R.A. Atmospheric deposition of PAHs to
water surfaces: a mass balance approach. Atmos. Environ., in press.
21. Murphy, T.J. Atmospheric inputs of chlorinated hydrocarbons to
the Great Lakes. In Toxic Contaminants in the Great lakes. J/0.
Nriagu; M.S. Simmons (Eds.), Adv. Environ. Sci. Tech. #14, Wiley-
Interscience: New York, 1984. 53-80.
22. Murphy, T.J. Design of a Great Lakes Atmospheric Inputs and
Sources (Gl&IS) Network. Report to US EPA (Grant number R005818-01)
Great Lakes National Program Office: Chicago, IL, 1987.
23. Murphy, T.J.; Rzeszutko, C.P. Precipitation inputs of PCBs to
Lake Michigan. J. Great Lakes Res., 1977. 3, 305-312.
-------
89
24. Pankow, J.F.; Isabella, L.M.; Asher, W.E. Trace organic compounds
in rain. I. Sampler design and analysis by adsorption/thermal
desorption (ATD). Environ. Sci. Tech., 1984. 18, 310-318.
25. Pankow, J.F. Atmos. Environ., 1987. in press.
I
26. Peters, L.K. Gases and their precipitation scavenging in the
marine atmosphere. In Air-Sea Exchange of Gases and Particles. P.S.
Liss; W.G.N. Slinn (Eds.), NATO ASI Series, D. Reidel Publishing:
Dordrecht, 1983. 173-240.
27. Scott, B.C. Modeling of atmospheric wet deposition. In
Atmospheric Pollutants in Natural Waters. S.J. Eisenerich (Ed.), Ann
Arbor Science: Ann Arbor, MI, 1981. 3-22.
28. Settle, D.M.; Patterson, C.C. Magnitude and sources of
precipitation and dry deposition fluxes of industrial and natural lead
to the north Pacific at Enewetak. J. Geophys. Res., 1982. 87, 8857-
8869.
29. Shoemaker, D.P., Garland, C.W., and Steinfeld, J.I. Experiments
in Physical Chemistry. 3rd Edition, McGraw Hill, New York, 1974.
30. Slinn, W.G.N., et al. Some aspects of the transfer of atmospheric
trace constituents past the air-sea interface. Atmos. Environ. 1978.
12, 2055-2087.
31. Slinn, W.G.N. Air to sea transfer of particles. In Air-Sea
Exchange of-'Gases and Particles. P.S. Liss; W.G.N. Slinn (Eds.), NATO
ASI Ser., D. Reidel Publishing: Dordrecht, 1983. 299-406.
32. Strachan, W.M.J.; Huneault, H. Polychlorinated Biphenyls and
organochlorine pesticides in Great Lakes precipitation. J. Great Lakes
Res., 1979. 5, 61-68.
-------
90
33. Strachan, W.M.J.; Huneault, H. Automated rain sampler for trace
organic substances. Environ. Sci. Tech., 1984. 18, 127-130.
34. Strachan, W.M.J.; Eisenreich, S.J. Mass Balancing of Organic
Contaminants in the Great Lakes: the Role of Atmospheric Deposition.
Report to the International Joint Commission: Windsor, Ontario, 1987.
35. Talbot, R.W.; Andren, A.W. Relationships between Pb and Pb-210
in aerosol and precipitation at a remote site in northern Wisconsin.
J. Geophys. Res. 1983. 9, 474-496.
36. Tanabe, S.; Hidaka, H. ; Tatsukawa, R. PCBs and chlorinated
hydrocarbon pesticides in Antarctic atmosphere and hydrosphere.
Chemosphere, 1983. 12, 277-288.
37. Yamasaki, H. ; Kuwata, K.; Yoshio, K. Nippon Kagaka Kaish, 1984.
1324-1329, as cited in Bidleman and Foreman, 1987.
38. Yamasaki, H. ; Kuwata, K.; Miyamoto, H. Effects of ambient
temperature on aspects of airborne polycyclic aromatic hydrocarbons.
Environ. Sci. Tech., 1982. 16, 189-194.
-------
91
APPENDIX A
PAH CONCENTRATIONS IN RAIN - 1986
-------
92
PAH concentrations in Precipicacion-1986
(ng/L)
--naphthalene-•
sampler
Solvent MIC
Resin MIC
Filter MIC
Resin Aerochem
all samplers
volume weighted
mean (ng/L)
10.2
10.2
arithmetic
mean (ng/L)
L0.9
10.9
standard
deviation
6.7
6.7
-•range--
min max
3.1 26.8
3.1 26.8
n
12
12
--acenaphthylene- -
volume weighted arithmetic standard
sampler mean (ng/L) mean (ng/L) deviation
Solvent MIC
Resin MIC
Filter MIC
Resin Aerochem
all samplers
1
0
1
1
1
.1
.9
.9
.8
.3
1.
1.
1.
1
1.
0
0
.9
,3
.3
0.
1.
1.
1.
1
7
6
.4
.3
.3
- -range --
min max
0.
0.
0.
0.
0.
1
1
5
3
1
2.
5.
4.
4.
5.
1
6
8
9
6
n
12
10
10
10
42
--acenaphthene--
volume weighted arithmetic standard
sampler mean (ng/L) mean (ng/L) deviation
Solvent MIC
Resin MIC
Filter MIC
Resin Aerochem
all samplers
0.
2.
0.
2,
1.
,7
,1
,7
.9
.3
0.
2.
0.
2
1
.7
.6
.8
.3
.6
0.
2.
0.
1.
1,
5
3
,4
,7
.7
- -range --
min max
0.
0.
0.
0.
0
2
3
,3
,5
.2
1
8
1
5
8
.6
.2
.9
.5
.2
n
12
10
10
10
42
--fluorene--
volume weighted arithmetic standard
sampler mean (ng/L) mean (ng/L) deviation
Solvent MIC
Resin MIC
Filter MIC
Resin Aerochem
all samplers
1.8
1.6
1.1
1.2
1.5
1.9
2.0
1.2
1.5
L.7
1.1
1.5
0.8
1.0
l.l
- -range --
min max
0.3
0.3
0.6
0.3
0.3
4.1
4.5
3.1
2.9
4.5
n
13
10
10
10
43
-------
-•phenanthrene--
volume weighted arithmetic standard
sampler mean (ng/L) mean (ng/L) deviation
Solvent MIC
Resin MIC
Filter MIC
Res In Aerochea
all samplers
8.6
15.3
8.8
11.2
10.6
9.1
19.6
8.9
12.4
12.3
4.2
16. 5
2.9
3.7
9.6
•- range --
rain max
2.2
5.1
4.9
6.0
2.2
17.7
62.8
15.1
18.8
62.8
n
13
10
10
10
43
--anthracene--
volume weighted arl three etc standard
sampler mean (ng/L) mean (ng/L) deviation
Solvent MIC
Resin MIC
Filter MIC
Resin Aerochea
•11 samplers
0.
0.
0.
0.
0.
2
3
3
3
3
0.
0.
0.
0.
0.
3
3
3
4
3
0
0
0
0
0
.2
.3
.2
.2
.2
- -range- •
rain max n
.0
.0
O.I
0.1
.0
0.
0.
0.
0.
0.
13
10
10
10
43
- -fluoranthene--
volume weighted arithmetic standard
sampler mean (ng/L) mean (ng/L) deviation
Solvent MIC
Resin MIC
Filter MIC
Resin Aerochem
all samplers
7.
10.
7.
8.
8.
9
,1
.8
,5
.5
7.
10,
7
8.
8.
9
5
0
,3
.4
9
9
5
7
8
.9
.9
.9
.2
.6
-- range ••
•in max
0
0
0
0
0
.2
.4
.5
.6
.2
32
36
18
24
36
.5
.4
.5
.7
.4
n
13
10
10
10
43
--pyrene--
volume weighted arithmetic standard
sampler mean (ng/L) mean (ng/L) deviation
Solvent MIC
Resin MIC
Filter MIC
Resin Aerochem
all samplers
5.
7.
5.
7.
6.
2
3
.8
.5
.1
5.
7.
5.
7,
6
3
.7
.2
.1
.2
7.
7.
4.
5.
6.
1
9
5
9
.6
•-range--
mln max
0.
0.
0.
0.
0.
1
3
3
6
1
25.
28.
14.
19,
28,
,0
,7
.0
.8
.7
n
13
10
10
10
43
93
•-benzo(a]anchracene--
volume weighted arithmetic standard
sampler mean (ng/L) mean (ng/L) deviation
Solvent MIC
Resin MIC
Filter MIC
Resin Aerochem
all samplers
2
3
3
3
3
.4
.7
.1
.1
.0
2
4
2.
3
3
.6
3
.9
1
.2
2
4,
2
2
3
.8
.4
.5
.6
.2
- - range - -
rain max
0
0
0
.0
.2
.2
.2
.0
7
14
7
8
14
.6
.9
.1
.6
.9
n
11
8
8
8
35
-------
- -chryjene--
volume weighted arithmetic standard
sampler mean (ng/L) mean (ng/L) deviation
Solvent MIC
Resin MIC
Filter MIC
Resin Aerochem
all samplers
6
6
5
5
5.
.1
.5
.1
.6
.9
6.
7.
4.
5.
5,
. 7
.0
.5
.3
,9
8
6.
4
5
6
.4
.9
.7
.0
.7
• -range --
rain max
0.
0.
0.
0.
0.
.2
3
.3
.4
.2
27.
23
14
15
27
.4
.4
.9
.5
.4
n
13
10
10
10
43
-benzo[b]fluoranthene- •
volume weighted arithmetic standard
sampler mean (ng/L) mean (ng/L) deviation
Solvent MIC
Resin MIC
Filter MIC
Resin Aerochea
•11 saaplers
1.
2.
1,
1
1.
.8
.3
.6
.9
.9
1
2
1
1
1
.9
.4
.5
.9
.9
1
2
1
1
1
.9
.2
.5
.9
.9
•- range --
rain max
0
0
0
0
0
.2
.2
.1
.1
.1
5.
8.
4.
6.
8
1
,0
.9
.8
.0
n
13
10
10
10
43
• -benzo (le] fluoranthene--
volume weighted arithmetic standard
sampler mean (ng/L) mean (ng/L) deviation
Solvent MIC
Resin MIC
Filter MIC
Resin Aerochem
all samplers
1
1
1
1
1
.3
.6
.3
.5
.4
1
1
1
1
1
.4
.7
.2
.5
.4
1.
1.
1.
1.
I.
6
6
3
2
5
- -range •-
mln max
0.
0.
0.
0.
0.
1
1
1
1
1
4
5
4
4
S
.2
.4
.4
.1
.4
n
13
10
10
10
43
--benz[ejpycene--
volume weighted arithmetic standard
sampler mean (ng/L) mean (ng/L) deviation
Solvent MIC
Res la MIC
Filter HIC
Resin Aerochem
all samplers
1.2
1.7
1.5
1.7
1.4
1.3
1.9
1.4
1.7
1.5
1.4
1.7
1.1
1.4
1.4
- -range- -
rain max
0.1
0.1
0.2
0.1
0.1
3.7
5.8
3.3
4.3
5.8
n
12
8
8
8
36
-benzo(a]pyrene-
volume weighted arithmetic standard
sampler mean (ng/L) mean (ng/L) deviation
Solvent MIC
Resin MIC
Filter MIC
Resin Aerochem
all samplers
1.4
3.3
2.9
3.5
2.5
1 5
3.5
3.0
3.8
2.9
I 7
2.6
1.8
2.1
2.2
- - range --
min max
0.1
0.4
0.7
0.9
0.1
4.7
9.4
6.4
7.0
9.4
n
13
10
10
10
43
-------
- -id«no[123-c,d]pyr«n«--
volume weighted arithmetic standard
sampler mean (ng/L) mean (ng/L) deviation
Solvent MIC
Resin MIC
Filter MIC
Resin Aerochea
all laaplers
2
2
I
2
1
.0
.0
.4
.1
.8
2.
2.
1.
2
2
.2
.1
.4
.2
.0
1.
2.
0.
1.
1.
,3
,1
.9
.5
.6
-- range --
rain max
0.
0.
0.
0.
0.
7
3
2
5
2
5.
6
3
5
6
3
.9
.0
.8
.9
n
13
10
10
10
43
--dibenzo[a.h|anthracene--
volume weighted arithmetic standard
sampler mean (ng/L) mean (ng/L) deviation
Solvent MIC
Resin MIC
Filter MIC
Resin Aerochea
•11 aaarplers
0.5
0.7
0.5
0.7
0.6
0.5
0.8
0.6
0.9
0.7
0.4
0.9
0.3
0.5
0.6
• -range- -
rain max
0.2
0.1
0.2
0.2
0.1
1.8
3.3
1.1
1.7
3.3
n
13
10
10
10
43
--benzo(g,h,i]perylene--
volume weighted arithmetic standard
sampler mean (ng/L) mean (ng/L) deviation
Solvent MIC
Resin MIC
Filter MIC
Resin Aerochem
all samplers
2.2
1.9
1.4
1.5
1.9
2.3
2.1
1.4
1.7
1.9
1.1
2.1
0.9
1.1
1.4
• -range --
•In max
0.4
0.3
0.2
0.4
0.2
4.9
8.1
3.5
3.9
8.1
n
13
10
10
10
43
95
•-total PAH--
volume weighted arithmetic standard
sampler mean (ng/L) mean (ng/L) deviation
Solvent MIC
Resin MIC
Filter MIC
Resin Aerochea
all samplers
53
60
44
54
53
.9
.5
.4
.2
.0
' 56
68
42
54
55
.0
.3
.3
.3
.3
41
49
27
31
39
.7
.9
.4
.2
.8
- - range - -
min max
9
13
15
22
9
.7
.6
.0
.9
.7
130.
198.
98.
121.
198.
7
6
9
3
6
n
13
10
10
10
43
-------
96
APPENDIX B
PCB CONCENTRATIONS IN RAIN - 1986
-------
Data Summary: PCBs - 1986 RAIN
97
I.U.P.A.C.
Congener 8
Volume Weighted Arithmetic Standard -- Range •-
Sampler Mean Cone. (ng/L) Mean (ng/L) Deviation Mlnlnua Maxlnua Number
Solvent MIC
Resin MIC
Filter MIC
Resin Aerochem
All Samplers
0.002
0.072
O.OS2
0.162
0.057
0.045
0.098
0.062
0.281
0.140
0.12
0.07
0.58
0.32
0.05
0.02
0.03
0.02
0.02
0.05
0.42
0.23
1.69
1.69
1
9
9
7
26
I.U.P.A.C. #
- - Congener 18
Volume Weighted Arithmetic Standard -- Range -•
Sampler Mean Cone. (ng/L) Mean (ng/L) Deviation Minimum Maximum Number
Solvent MIC
Resin MIC
Filter MIC
Resin Aerochem
All Samplers
0.06
0.11
0.12
0.20
0.11
0.07
0.12
0.14
0.25
0.14
0.04
0.13
0.08
0.26
0.17
0.02
0.02
0.03
0.04
0.02
0.17
0.52
0.36
1.07
1.07
12
12
12
12
48
I.U.P.A.C. *
Congener 17 --
Volume Weighted Arithmetic Standard
Sampler Mean Cone. (ng/L) Mean (ng/L) Deviation
Solvent MIC
Resin MIC
Filter MIC
Resin Aerochem
All Samplers
0.20
0.17
0.09
0.15
0.15
0.19
0.16
0.13
0.15
0.16
0.09
0.14
0.12
0.15
0.13
-- Range --
Minimum Maximum Number
0.053
0.007
0.007
0.009
0.007
0.33
0.44
0.41
0.53
0.53
12
12
12
13
49
I.U.P.A.C. *
-- Congener 16*.32 --
Volume Weighted Arithmetic Standard
Sampler Mean Cone. (ng/L) Mean (ng/L) Deviation
Solvent MIC
Resin MIC
Filter MIC
Resin Aerochem
AIL Samplers
0.038
0.077
0.067
0.052
0.060
0.04
0.09
0.09
0.08
0.07
0.02
0.1S
0.07
0.06
0.09
-- Range --
Minimum Maximum Number
0.016
0.011
0.008
0.015
0.008
0.08
0.58
0.24
0.24-
o.sa
12
12
12
12
48
I.U.P.A.C. *
•- Congener 31 --
Volume Weighted Arithmetic Standard
Sampler Mean Cone. (ng/L) Mean (ng/L) Deviation
Solvent MIC
Resin MIC
Filter MIC
Resin Aerochem
All Samplers
0.16
0.12
0.19
0.13
0.15
0.17
0.12
0.28
0.19
0.19
0.11
0.06
0.39
0.24
0.25
-- Range --
Minimum Maximum Number
0.056
0.038
0.024
0.038
0.024
0.39
0.22
1.53
0.93
1.53
12
12
13
12
49
I.U.F.A.C.
-- Congener 28 --
Volume Weighted Arithmetic Standard
Sampler Mean Cone. (ng/L) Mean (ng/L) Deviation
Solvent MIC
Resin MIC
Filter MIC
Resin Aerochem
All Samplers
0.08
0.06
0.10
0.07
0.08
0.09
0.07
0.13
0.11
0.10
0.09
0.05
0.16
0.11
0.12
-- Range --
Minimum Maximum Number
0.013
0.019
0.004
0.015
0.004
0.31
0.21
0.62
0.37
0.62
11
12
13
12
48
-------
Data Summary: PCBs - 1986 RAIN
98
I.U.P.A.C. »
-- Congener 33
Sampler
Solvent HIC
Resin MIC
Filter MIC
Resin Aerochea
All Samplers
I.O.P.A.C. #
Sampler
Solvent MIC
Resin MIC
Filter MIC
Resin Aerochem
All Samplers
Volume Weighted
Mean Cone. (ng/L)
0.49
0.10
0.26
0.14
0.27
Volume Weighted
Mean Cone. (ng/L)
0.019
0.024
0.068
0.052
0.039
Arithmetic
Mean (ng/L)
0.75
0.12
0.28
0.21
0.31
- - Congener
Arithmetic
Mean (ng/L)
0.04
0.03
0.10
0.07
0.06
Standard
Deviation
0.81
0.07
0.37
0.08
0.47
22 --
Standard
Deviation
0.06
0.02
0.1S
0.16
0.12
-- Range
.-
Minimum Maximum
0.014
0.063
0.01
0.101
0.01
-- Range
2.18
0.30
1.43
0.37
2.18
^ —
Minimum Maximum
0.002
0.007
0.004
0.004
0.002
0.18
0.08
0.47
0.51
0.51
Number
8
11
13
10
42
Number
7
11
10
9
37
I.U.P.A.C. *
Congener 52 --
Volume Weighted Arithmetic Standard -- Range --
Sampler Mean Cone. (ng/L) Mean (ng/L) Deviation Minimum Maximum Number
Solvent MIC
Resin MIC
Filter MIC
Resin Aerochem
All Samplers
0.027
0.031
0.098
0.085
0.057
0.14
0.20
0.32
0.96
0.35
0.14
0.17
0.28
0.144
0.065
0.112
0.961
0.06S
0.14
0.34
0.55
0.96
0.96
1
2
4
1
8
I.U.P.A.C. *
•- Congener 49
Volume Weighted Arithmetic Standard
Sampler Mean Cone. (ng/L) Mean (ng/L) Deviation
Solvent MIC
Resin MIC
Filter MIC
R*sln Aerochem
All Samplers
0.009
0.004
0.051
0.016
0.021
0.032
0.016
0.197
0.093
0.079
0.020
0.007
0.200
0.078
0.128
-- Range -•
Minimum Maximum Number
0.010
0.007
0.023
0.015
0.007
0.06
0.02
0.48
0.17
0.48
4
3
3
2
12
I.U.P.A.C. *
•- Congener 47*.48
Sampler
Solvent MIC
Resin MIC
Filter MIC
Resin Aerochem
All Samp UTS.
I.U.P.A.C. *
Sampler
Solvent MIC
Resin MIC
Filter MIC
Resin Aerochem
All Samplers
Volume Weighted
Mean Cone. (ng/L)
0.021
0.014
0.059
0.067
0.036
Volume Weighted
Mean Cone. (ng/L)
0.055
0.030
0.114
0.052
0.065
Arithmetic
Meen (ng/L)
0.040
0.038
0.092
0.201
0.082
-- Congener
Arithmetic
Mean (ng/L)
0.067
0.037
0.138
0.073
0.081
Standard
Deviation
0.041
0.054
0.107
0.256
0.136
44 --
Standard
Deviation
0.050
0.026
0.243
0.060
0.140
-- Range
•-
Minimum Maximum
0.009
0.008
0.012
0.023
0.008
- - Range
0.14
0.15
0.33
0.64
0.64
--
Minimum Maximum
0.023
0.014
0.012
0.017
0.012
0.22
0.11
0.95
0.25
0.95
Number
8
5
8
4
25
Number
11
11
13
12
47
-------
99
Data Summary: PCBs - 1986 RAIN
I.U.P.A.C. #
Sampler
Solvent MIC
Res In MIC
Filter MIC
Resin Aerochem
All Samplers
I.U.P.A.C. #
Seapler
Solvent MIC
Resin MIC
Filter MIC
Resin Aerochem
All Samplers
Volume Uelghted
Mean Cone. (ng/L)
0.018
0.025
0.073
0.046
0.040
Volume Weighted
Mean Cone. (ng/L)
0.07
0.04
0.09
0.13
0.07
-- Congener
Arithmetic
Mean (ng/L)
0.029
0.029
0.099
0.072
0.059
-• Congener
Arithmetic
Mean (ng/L)
0.08
0.05
0.11
0.18
0.11
37.42 --
Standard
Deviation
0.028
0.031
0.145
0.042
0.087
41.64 --
Standard
Deviation
0.04
0.02
0.08
0.20
0.12
-- Range --
Minimum Maximum
0.011
0.009
0.011
0.021
0.009
0.11
0.12
0.54
0.17
0.54
-- Range --
Minimum Maximum
0.013
0.013
0.012
0.02
0.012
0.17
0.09
0.28
0.68
0.68
Number
9
10
11
10
40
Number
11
12
12
12
47
I.U.P.A.C.
•- Congener 74 --
Volume Weighted Arithmetic Standard -- Range -•
Saapler Mean Cone. (ng/L) Mean (ng/L) Deviation Minimum Maximum Number
Solvent MIC
Resin MIC
Filter MIC
Resin Aerochea
All Samplers
0.13
0.10
0.07
0.12
0.10
0.24
0.15
• 0.08
0.17
0.15
0.24
0.17
0.09
0.11
0.16
0.024
0.009
0.006
0.04
0.006
0.72
0.51
0.29
0.33
0.72
8
9
13
10
40
I.U.P.A.C. *
•- Congener 70
Volume Weighted Arithmetic Standard •- Range --
Sampler Mean Cone. (ng/L) Mean (ng/L) Deviation Minimum Maximum Number
Solvent MIC
Resin MIC
Filter MIC
Resin Aeroehem
All Samplers
0.16
0.09
0.09
0.14
0.12
0.18
0.10
0.12
0.17
0.14
0.13
0.07
0.10
0.11
0.11
0.059
0.014
0.013
0.021
0.013
0.52
0.23
0.39
0.40
0.52 .
12
12
13
13
50
I.U.P.A.C.
-- Congener 66
Volume Uelghted Arithmetic Standard
Sampler Mean Cone. (ng/L) Mean (ng/L) Deviation
Solvent MIC
Resin MIC
Filter MIC
Resin Aerochem
All Samplers-^"
0.
0.
0.
0.
0.
17
12
14
15
.14
0.
0.
0.
0.
0.
.18
.12
.17
,20
.17
0.
0.
0.
0,
0.
.13
13
.13
,13
.13
- - Range - -
Minimum Maximum Number
0.
0.
0.
0.
0.
058
021
020
042
.020
0
0
0
0
0
.46
.51
.37
.45
.51
12
12
13
12
49
I.U.P.A.C. * •- Congener 60*.56 --
Volume weighted Arichmecic Scandard
Sampler Mean Cone. (ng/L) Mean (ng/L) Deviation
-- Range --
Minimum Maximum Number
Solvent MIC
Resin MIC
Filter MIC
Resin Aerochem
All Samplers
0.043
0.034
0.053
0.051
0.044
0.050
0.035
0.063
0.065
0.053
0.042
0.027
0.042
0.053
0.044
0.016
0.012
0.006
0.014
0.006
0.17
0.11
0.16
0.22
0.22
12
12
13
13
50
-------
100
Data Summary: PCBs - 1986 RAIN
I.U.P.A.C. »
Sampler
Solvent MIC
Resin MIC
Filter MIC
Reiln Aerochan
All Samplers
Volume Weighted
Mean Cone. (ng/L)
0.19
0.10
0.10
0.12
0.13
-- Congener
Arithmetic
Mean (ng/L)
0.22
0.10
- 0.12
0.15
0.15
101 --
Standard
Deviation
0.15
0.05
0.08
0.14
0.12
-- Range --
Minimum Maximum
0.077
0.041
0.022
0.037
0.022
0.54
0.21
0.29
0.60
0.60
Number
12
12
13
13
50
I.U.P.A.C. *
-- Congener 99 --
Sampler
Solvent MIC
Re«ln MIC
Filter MIC
Re* in Aerochem
All Samplers
I U P A C. #
Saapler
Solvent MIC
Resin MIC
Filter MIC
Resin AerocViem
All Samplers
I.U.P.A.C. »
Saapler
Solvent MIC
Resin MIC
Filter MIC
Resin Aerochem
All Samplers
I.U.P.A.C. *
Sampler
Solvent MIC
Rasln MIC
Filter MIC
Resin Aerochem
All Samplers
Volume Weighted
Mean Cone. (ng/L)
0.042
0.037
0.020
0.100
0.041
Volume Weighted
Mean Cone. (ng/L)
0.028
0.016
0.008
0.047
0.021
Volume Weighted
Mean Cone. (ng/L)
0.046
0.013
0.007
0.010
0.021
Volume Weighted
Mean Cone. (ng/L)
0.15
0.11
0.09
0.11
0.11
Arithmetic
Mean (ng/L)
0.057
0.040
0.025
0.132
0.060
- - Congener
Arithmetic
Mean (ng/L)
0.034
0.023
0.011
0.064
0.034
-- Congener
Arithmetic
Mean (ng/L)
0.051
0.027
0.016
0.039
0.035
-- Congener
Arithmetic
Mean i.ng/L)
0.17
0.11
0.11
0.14
0.13
Standard
Deviation
0.038
0.051
0.020
0.293
0.145
97 --
Standard
Deviation
0.023
0.018
0.006
0.076
0.046
87 --
Standard
Deviation
0.035
0.019
0.011
0.009
0.028
110 --
Standard
Deviation
0.12
0.09
0.07
0.12
0.11
-- Rang* --
Minimum Maximum
0.015 0.13
0.009 0.19
0.003 0.08
0.007 0.96
0.003 0.96
• • Range - -
Minimum Maximum
0.007
0.003
0.002
0.008
0.002
0.08
0.06
0.02
0.23
0.23
-- Range --
Minimum Maximum
0.013
0.014
0.005
0.027
0.005
0.15
0.07
0.04
0.05
0.15
-- Range --
Minimum Maximum
0.056
0 023
0.027
0.026
0.023
0.51
0.31
0.26
0.47
0.51
Number
10
12
11
9
42
Number
11
9
10
11
41
Number
12
7
8
4
31
Number
12
12
13
13
50
I.U.P.A.C. *
-- Congener 32 --
Volume Weighted Arithmetic Standard -- Range --
Sampler Mean Cone. (ng/L) Mean (nj/L) Deviation Minimum Maximum Number
Solvent MIC
Resin MIC
Filter MIC
Resin Aerochea
All Samplers
0.004
0.009
0.009
0.013
0.008
0 007
0 009
0 010
0 025
0 013
0.006
0 007
0.007
0.034
0.020
0.001
0.003
0.002
0.002
0 001
0.02
0.03
0.03
0.12
0.12
6
12
12
10
40
-------
101
Data Summary: PCBs -
I.U.P.A.C. 0
Volv;
Sampler Mean
Solvent MIC
Resin MIC
Filter MIC
Resin Aarochea
All Samplers
I.U.P.A.C. »
Volt
Sampler Mean
Solvent MIC
Resin MIC
Filter MIC
Resin Aeroehem
All Samplers
1986 RAIN
me Weighted
Cone. (ng/L)
0.045
0.075
0.085
0.095
0.071
joe Weighted
Cone. (ng/L)
0.07
0.20
0.08
0.07
0.11
-- Congener
Arithmetic
Mean (ng/L)
0.059
0.093
0.090
0.153
0.099
- - Congener
Arithmetic
Mean (ng/L)
0.09
0.27
0.08
0.10
0.13
144 --
Standard
Deviation
0.043
0.062
0.055
0.107
0.078
118*. 108 --
Standard
Deviation
0.07
0.33
0.03
0.07
0.18
-- Range
Minimum Mi
0.019
0.020
0.023
0.022
0.019
-- Range
Minimum MJ
0.021
0.040
0.028
0.036
0.021
ixlaum
0.163
0.209
0.218
0.330
0.330
ixlmum
0.30
1.07
0.1S
0.26
1.07
Number
10
11
13
11
45
Number
11
9
13
9
42
I.U.P.A.C.
-- Congener 146 --
Sampler
Solvent MIC
Resin MIC
Filter MIC
Resin Aeroehem
All Samplers
I.U.P.A.C. *
Sampler
Solvent MIC
Resin MIC
Filter MIC
Resin Aeroehem
All Samplers
Volume Weighted
Mean Cone. (ng/L)
0.053
0.048
0.055
0.038
0.050
Volume Weighted
Mean Cone. (ng/L)
0.10
0.18
0.11
0.11
0.13
Arithmetic
Mean (ng/L)
0.059
0.057
0.065
0.054
0.059
-- Congener
Arithmetic
Mean (ng/L)
0.17
0.36
0.22
0.34
0.27
Standard
Deviation
0.040
0.029
0.035
0.023
0.033
153 --
Standard
Deviation
0.08
0.41
0.11
0.13
0.24
-- Range
Minimum Mi
0.020
0.020
0.019
0.017
0.017
-- Range
Minimum Hi
0.078
0.077
0.065
0.173
0.065
--
ix ilium
0.16
0.13
0.14
0.09
0.16
..
ixinum
0.28
1.34
0.42
0.49 -
1.34
Number
12
11
13
11
47
Number
6
7
8
5
26
I.U.P.A.C. *
•- Congener 141
Volume Weighted Arithmetic Standard -- Range --
Sampler Mean Cone. (ng/L) Mean (ng/L) Deviation Mlnloum Maximum Number
Solvent MIC
Resin MIC
Filter MIC ~
Resin Aeroehem
All Samplers
0.013
0.017
0.015
0.014
0.015
0.019
0.027
0.022
0.036
0.025
0.017
0.025
0.010
0.018
0.019
0.006
0.004
0.010
0.011
0.004
0.06
0.08
0.04
0.07
0.08
10
10
9
6
35
-------
Data Summary: PCBs - 1986 RAIN
I.U.P.A.C.
-- Congener 138 --
Voluae Uelghted Arithmetic Standard
Sampler Mean Cone. (ng/L) Mean (ng/L) Deviation
Solvent MIC
Reiln MIC
Filter MIC
Resin Aerochea
All Sampler*
0.053
0.060
0.088
0.040
0.064
0.062
0.066
0.093
0.046
0.068
0.047
0.046
0.062
0.017
0.050
-- Range --
Minimum Maximum Number
0.017
0.022
0.046
0.020
0.017
0.17
0.21
0.29
0.08
0.29
11
12
13
11
47
102
I.U.P.A.C. *
Congener 175 --
Volume Weighted
Sampler Mean Cane. (ng/L)
Solvent MIC
Re* In MIC
Filter MIC
Re* in Aerochem
All Sampler*
I.U.P.A.C. »
Sampler
Solvent MIC
Re* In MIC
Filter MIC
Re* in Aerochea
All Sampler*
I.U.P.A.C. *
Sampler
Solvent MIC
Resin MIC
Filter KIC
Resin Aerochem
All Samplers
I.U.P.A.C. »
Sampler
Solvent MIC
Resin MIC
Filter MIC
Resin Aerochem
All Samplers
0.026
0.106
0.038
0.026
0,052
Volume Weighted
Mean Cone. (ng/L)
0.006
0.010
O.OIO
0.009
0.009
Volume Weighted
Mean Cone. (ng/L)
0.009
0.014
0.025
0.021
0.016
Volume Weighted
Mean Cone. (ng/L)
0.005
0.007
0.023
0.028
0.014
Arithmetic
Mean (ng/L)
0.051
0.119
0.051
0.045
0.076
-- Congener
Arithmetic
Mean (ng/L)
0.008
0.015
0.012
0.018
0.013
-- Congener
Arithmetic
Mean (ng/L)
0.016
0.024
O.C41
O.CiJ
0.032
-- Congener
Arithmetic
Mean (ng/L)
0.014
0 021
0.053
0 147
0 053
Standard
Deviation
0.040
0.193
0.024
0.054
0.121
187*. 159 --
Standard
Deviation
0.005
0.018
0.012
0.008
0.012
185 --
Standard
Deviation
0.013
0.020
0.035
0.026
0.031
174 --
Standard
Deviation
0.009
0.016
0.058
0.192
0.108
-- Range --
Minimum Maximum
0.012
0.01S
0.012
0.026
0.012
0.11
0.71
0.09
0.18
0.71
-- Rang* --
Minimum Maximum
0.003
0.004
0.004
0.004
0.003
0.021
0.062
0.044
0.027
0.062
- - Range - -
Minimum Maximua
0.003
0.006
0.008
O.CC5
0.003
0.05
0.07
0.13
0.13
0.13
- - Range • -
Mlnlaum Maxiaum
0.003
0.006
0.003
0 025
0.003
0.03
0.04
0.18
0.53
0.53
Number
7
11
9
5
32
Number
11
8
10
7
36
Number
7
8
10
9
34
Number
5
5
8
5
23
-------
103
Data Summary: PCBs - 1986 RAIN
I.U.P.A.C. *
-- Congener 180 --
Volume Weighted
Sampler Mean Cone. (ng/L)
Solvent MIC
Resin MIC
Filter MIC
Res In Aarochem
All Staplers
I.U.P.A.C. »
Soapier
Solvent MIC
Resin MIC
Filter MIC
Resin Aerochea
All Sampler*
I.U.P.A.C. *
Stapler
Solvent MIC
Resin MIC
Filter MIC
Resin Aerochea
All Staplers
I.U.P.A.C. #
Stapler
Solvent MIC
Resin MIC
Filter MIC
Resin Aerochea
All Staplers
0.024
0.031
0.072
0.083
0.048
Volua* Weighted
Mean Cone. (ng/L)
0.001
0.002
0.002
0.002
0.002
Volua* Weighted
Metn Cone. (ng/L)
0.004
0.005
0.015
0.017
0.009
Voluao Weighted
Metn Cone. (ng/L)
0.012
0.017
0.027
0.032
0.020
Arithmetic
Metn (ng/L)
0.030
0.038
0.078
0.116
• 0.066
-- Congener
Arithmetic
Meen (ng/L)
0.019
0.018
0.018
0.066
0.028
-- Congener
Arithmetic
Ketn (ng/L)
0.006
0.006
0.029
0.057
0.023
-- Congener
Arithmetic
Meen (ng/L)
0.015
0.019
0.044
0.042
0.030
Standard
Deviation
0.025
0.026
0.041
0.111
0.070
170 --
Standard
Deviation
0.002
0.019
196 --
Standard
Deviation
0.003
0.004
0.039
0.101
0.054
201 --
Standard
Deviation
0.007
0.014
0.040
0.031
0.030
-- Range --
Minimum Maximum
0.006
0.014
0.025
0.016
0.006
0.097
0.116
0.184
0.352
0.352
-- Rang* --
Minimua Haxlaua
0.019
0.018
0.016
0.066
0.016
0.019
0.018
0.020
0.066
0.066
-- Rang* --
Mlnioua Maximum
0.002
0.002
0.002
0.002
0.002
0.012
0.015
0.111
0.304
0.304
-- Rang* •-
Minimum Maximum
0.005.
0.004
0.008
0.008
0.004
0.03
0.05
0.16
0.13
0.16
Number
11
11
13
11
46
Number
1
1
2
1
5
Number '
9
9
10
7
35
Mumber
10
11
10
12
43
--- Total PCBs •--
Volume Weighted Arithmetic Standard -- Range •-
Sampler Mean Cone. (ng/L) Metn (ng/L) Oevlttion Minimum Maximum Kumber
Solvent MIC
Resin MIC
Filter MIC
Resin Aerochea
All Samplers
2.
2.
2.
2.
2
.63
.28
.66
.79
.36
2
2
3
3
2
.33
.i2
.10
.26
.91
I.
I.
I.
1.
1.
,18
.52
.90
.86
.68
1.
0.
0.
0.
0,
09
65
90
30
.65
5.
5.
7.
7.
7.
35
.73
57
.39
.39
12
12
13
13
50
-------
104
APPENDIX C
CL-PESTICIDE CONCENTRATIONS IN RAIN - 1986
-------
Pesticide concentrations In Preclpltatlon-1986
(ng/L)
-- HCB --
-- a-HCH --
-- b-HCH --
105
volume weighted arithmetic
sampler mean (ng/L) mean (ng/L)
Solvent MIC
Resin MIC
Filter MIC
Resin Aerochea
all samplers
0.57
0.18
0.12
0.48
0.31
0.44
0.18
0.12
0.45
0.30
s tandard - - range - -
deviation mln max
0.73
0.35
0.17
1.05
0.70
0.019
0.016
0.007
0.014
0.007
2.73
1.36
0.58
3.95
3.95
n
13
13
11
13
50
volume weighted arithmetic standard
sampler mean (ng/L) mean (ng/L) deviation
Solvent MIC
Resin MIC
Filter MIC
Resin Aerochem
all samplers
40.
44.
3.
3.
28.
8
5
7
5
2
37
41
3
3
22
.3
.6
.3
.5
.4
57
113
2
4
69
.4
.0
.7
.0
.7
--range--
Bin max
0.
0.
0.
0.
0.
10
76
14
04
04
195
432
8
14
432
.2
.4
.2
.1
.4
n
10
13
9
12
44
volume weighted arithmetic standard
sampler mean (ng/L) mean (ng/L) deviation a
Solvent MIC
Resin MIC
Filter MIC
Resin Aerochem
all samplers
0
0
0
0
0
.3
.6
.6
.9
.5
0.
0.
0.
1.
0.
3
6
6
1
6
0
0
0
1
0
.2
.5
.8
.4
.9
0.
0.
0.
0.
0.
- -range --
tin max
03
04
08
02
02
0
1
2
3
3
.7
.5
.7
.8
.8
n
10
9
8
7
34
--llndane--
volume weighted arithmetic standard
sampler mean (ng/L) mean (ng/L) deviation
Solvent MIC
Resin MIC
Filter MIC
Resin Aerochera
all samplers
2
3
1.
1.
2
.8
.8
.2
.8
.7
2
3
1
1
2
.6
.4
.2
.5
.2
1.
3.
0
1.
2
.8
.6
.9
.9
.5
-- range --
rain max
0
0
0
0
0
.02
.06
.10
.01
.01
5
11
2
6
11
.7
.4
.5
.4
.It
n
10
11
8
10
39
-------
--hepcachlor-•
•-heptachlor epoxide--
--aldria--
106
volume weighted arithmetic standard
sampler mean (ng/L) mean (ng/L) deviation
Solvent NIC
Resin MIC
Filter MIC
Resin Aerochem
•11 samplers
0
0
0
0
0
.13
.58
.05
.95
.40
0.
0.
0.
1.
0.
1
6
1
3
6
0.
1.
0
3
2
.2
.8
.1
.8
.3
-- range --
mln
0.01
0.004
0.01
0.01
0.004
max
0.
6.
0.
13.
13.
5
,7
.2
3
.3
n
9
12
8
11
40
volume weighted arithmetic standard
sampler mean (ng/L) aean (ng/L) deviation
Solvent MIC
Resin MIC
Filter MIC
Resin Aerochem
all samplers
0.3
0.4
0.3
0.5
0.4
0.3
0.4
0.3
0.5
0.4
0.2
0.2
0.2
0.5
0.3
-- range --
mln max
0.02
0.08
0.08
0.01
0.01
0.7
0.8
0.8
1.5
1.5
n
10
13
9
12
44
volume weighted arithmetic standard
sampler mean (ng/L) mean (ng/L) deviation
Solvent MIC
Resin MIC
Filter MIC
Resin Aerochem
all samplers
4.5
1.0
0.5
2.7
1.9
3.6
1.1
0.5
2.7
1.9
4.9
1.5
0.6
3.6
3.2
• -range --
oln max
0.01
0.03
0.07
0.08
0.01
W.I
5.3
1.8
11.1
14.1
n
7
13
6
8
34
--dieldrln-
volume weighted arithmetic standard
sampler mean (ng/L) mean (ng/L) deviation
Solvent MIC
Resin MIC
Filter MIC
Resin Aerochem
all samplers
1.
2.
0,
1.
1
.7
,0
.4
.6
.5
1.
1.
0.
I,
1.
.5
,7
,5
,8
.4
1.
1
0
1
1
.4
.8
.4
.8
.6
- - range - -
rain max
0.
0.
0.
0
0
12
29
.01
.02
.01
4.
5.
1.
5
5
.4
.8
.2
.4
.8
n
11
12
9
12
44
-------
-•endrin--
-- pp DDE -•
--op DDD --
10?
volume weighted arithmetic standard
sampler mean (ng/L) mean (ng/L) deviation
Solvent MIC
Resin MIC
Filter MIC
Re* In AerochaB
•11 samplers
0.2
0.4
0.2
0.5
0.3
0.2
0.4
0.2
0.6
0.4
0.3
0.8
0.2
0.9
0.7
- -range --
oln max
0.02
0.01
0.04
0.05
0.01
0.8
3.1
• 0.6
3.1
3.1
n
11
13
9
12
45
volume weighted arithmetic standard
sampler mean (ng/L) mean (ng/L) deviation
Solvent MIC
Resin MIC
Filter MIC
Resin Aerochem
all samplers
1.11
0.90
0.33
0.92
0.81
0.98
0.94
0.33
0.88
0.79
1.4
2.1
0.3
1.7
1.6
--range--
Bin max
0.04
0.07
0.06
0.09
0.04
3.8
7.8
1.1
6.7
7.8
n
11
12
11
13
47
volume weighted arithmetic standard
sampler mean (ng/L) mean (ng/L) deviation
Solvent MIC
Resin MIC
Filter MIC
Resin Aerochem
all samplers
0
0
0
0
0
.03
.25
.11
.16
.15
0
0
0
0
0
.03
.34
.15
.23
.21
0
0
0
0
0
.01
.59
.24
.49
.45
-- range •-
mln max
0.
0.
0
0
0
.01
.01
.01
.03
.01
0
1
0
1
1
.05
.87
.81
.81
.87
n
8
13
9
12
42
-- pp DDD --
volume weighted arithmetic standard
sampler mean (ng/L) mean (ng/L) deviation
Solvent MIC
Resin MIC
Filter MIC
Resin Aerochem
all samplers
0.
0.
0.
0.
0.
.03
.07 .
.05
.07
.05
0.
.03
0.07
0.
o.
0,
.08
,07
.06
0
0
0
0
0
.02
.10
.08
.03
.07
-- range --
rain max
0.
0.
0.
0,
0,
01
00
01
.03
.01
0.
0.
0.
0.
0,
09
38
27
.11
.38
n
11
12
9
12
44
-------
sampler
-- op DDT --
volume weighted arithmetic
mean (ng/L) mean (ng/L)
standard
deviation
--range--
min max
108
Solvent MIC
Resin NIC
Filter MIC
Resin Aerochem
all samplers
0.7
0.2
0.1
0.2
0.4
0.6
0.2
0.1
0.2
0.3
0.9
0.5
0.2
0.2
0.6
0.02
0.01
0.01
0.03
0.01
2.6
1.7
0.5
0.5
2.6
11
10
8
11
40
sampler
- - pp DDT - -
volume weighted arithmetic
mean (ng/L) mean (ng/L)
standard --range--
deviation min max
Solvent MIC
Resin MIC
Filter MIC
Resin Aerochem
all samplers
1.8
1.1
0.2
0.6
1.1
1.8
1.2
0.2
0.6
1.0
3.0
2.2
0.2
1.0
2.0
0.06
0.12
0.03
0.08
0.03
10.0
7.9
0.6
3.1
10.0
10
11
8
11
40
-------
109
APPENDIX D
CL-BENZENE CONCENTRATIONS IN RAIN
-------
110
Data summary: Chlorinated Benzenes In 1986 rain
-- 1.3-DICHLOROBENZENE--
volume weighted arithmetic standard
sampler mean (ng/L) mean (ng/L) deviation
Solvent MIC
Resin MIC
Filter MIC
Resin Aerochem
all samplers
0.61
0.48
1.26
1.63
0.87
0.90
0.62
1.60
2.01
1.27
0.59
0.53
0.79
1.03
0.94
-- range --
min max
0.48
0.12
0.12
0.60
0.12
2.60
2.12
2.92
3.68
3.68
n
12
13
12
12
49
-- 1.4-DICHLOROBENZENE--
volume weighted arithmetic standard
sampler mean (ng/L) mean (ng/L) deviation
Solvent MIC
Resin MIC
Filter MIC
Resin Aerochem
mil samplers
0.91
0.39
0.84
1.31
0.79
1.23
0.55
1.06
1.57
1.12
1.29
0.46
0.77
1.18
1.07
-- range --
min max
0.12
0.05
0.14
0.08
0.05
5.14
1.64
2.25
3.41
5.14
n
13
11
11
12
47
-- 1.2-DICHLOROBENZENE--
voluae weighted arithmetic standard
sampler mean (ng/L) mean (ng/L) deviation
Solvent MIC
Resin MIC
Filter MIC
Resin Aerochem
•11 samplers
0
0
1
0
0
.33
.59
.10
.97
.69
0.
0
1.
0
0
.42
.61
.99
.87
.81
0
0
3
0
1
.40
.81
.95
.74
.66
-- range --
min max
0
0
0
0
0
.02
.01
.03
.03
.01
1
2
10
2
10
.51
.41
.81
.17
.81
n
13
13
6
12
44
-- 1.2,4-TRICHLOROBENZENE--
volume weighted arithmetic standard
sampler mean (ng/L) mean (ng/L) deviation
Solvent MIC
Resin MIC
Filter MIC
Resin Aerochem
all samplers
11.
4.
10.
10.
9
.70
.05
.74
.67
.07
14.
6
20.
15
13
.21
.62
.28
.22
.96
22
12,
33
16
23
.66
.55
.95
.27
.21
- -range --
min max
0.
0.
0.
2,
0.
16
.28
.10
.19
.10
80.
48.
114.
56.
114
.93
,95
.74
.76
.74
n
13
13
12
12
50
--HEXACHLOROBENZENE--
volume weighted arithmetic standard
sampler mean (ng/L) mean (ng/L) deviation
Solvent MIC
Resin MIC~
Filter MIC
Resin Aerochem
all samplers
0.04
0.04
0.05
0.04
0.04
0.06
0.05
0.06
0.04
0.05
0.10
0.07
0.11
0.02
0.08
- - range --
min max
0.01
0.01
0.01
0.02
0.01
0.36
0.27
0.41
0.08
0.41
n
10
12
11
12
45
-------
Ill
APPENDIX E
PROPAGATION OF ERROR ANALYSIS
-------
112
PROPAGATION OF ERROR ANALYSIS
Quantification of a compound's mass in each sample analyzed by
the internal standard method involved the equation:
As x VL x CF
AIS x RF
where MS - Mass of compound in sample,
AS - Chromatographic peak area of sampled compound,
MIS - Mass of internal standard injected into sample,
AIS - Chromatographic peak area of internal standard,
CF - Correction factor for PAH subfraction,
RF - Response factor for compound relative to internal standard
from calibration curve.
Inherent in each term is some associated random error. Following the
method outlined by Shoemaker et al. (1974) for calculating the
propagated uncertainty in the sample mass
aMs - [($Ms/5As)2(aAs)2 + (SMS/5MIS) 2(aMIS)2 + (5Ms/5CF)2(aCF)2
(5Ms/SAIS)2(aAIS)2 + (5MS/5RF) 2(aRF)2 } ] 1/2
where a - s±andard deviation.
-------
113
Performing this calculation and finding aCF negligible, yields:
MsxCF AsxCF
]2(aAs)2 + [
AIsxRF
AgXCFxM.. AgXCFxM.
]2< 5) of
the uncertainty in the mass, it was found that the relative deviation
was nearly constant for any one compound ( * < ± 0.5%). Therefore, an
average relative deviation was applied to all other masses of the same
compound.
-------
The uncertainty in the concentration of a compound was also
calculated once the mass deviation was determined. Following multiple
calculations it was found that the error in the measured volume was
negligible and could be omitted. Thus, the uncertainty in the
concentration was due entirely to the uncertainty in the quantified
mass.
The uncertainty in the volume weighted mean concentration (VWM)
was calculated by summing the deviations of each mass and volume and
using the equation
aVWM - [(l/£Vol)2(aMs)2 + [lMs/(]Vol)2]2(aVol)2]1/2
Again, after multiple calculations, the deviation about the volume was
found to be negligible and the second term in the above equation could
be omitted.
Quantification of a compound's mass analyzed by the external
standard method involved the equation
Ms -
ASXCSTDXCFXVVOL
A
SID
where MS, AS, and CF are defined above,
CSTD - Concentration of standard solution which is being
externally compared to sample,
ASTD - Chromatographic peak area of standard solution,
VVQL - Volume of sample in autosampler vial.
-------
115
Assuming that the uncertainty associated with the subfraction
correction factor and the mass of the compound in the standard
solution were negligible, the following equation was derived for the
uncertainty in the mass.
CSTDxCFxVVOL CSTDxCFxVV()LxAs
t
]2
-------
116
APPENDIX F
FIELD NOTES ON PRECIPITATION SAMPLERS - 1986
-------
Summary of 1986 Rain Field Data
sampling
sampler
name front
(1986 Data)
Solvent MIC
Solvent MIC
Reservoir MIC
Resin Aerochem
Filter MIC
Inorp, Aerochem
Solvent MIC
Reservoir MIC
Resin Aerochrm
Fl Iter MIC
Inorp, Aerochem
Filter MIC
Solvent MIC
R.-scrvoir MIC
Rpsln Aerochem
Filter MIC
Inorp,. Aerochem
Solvent MIC
Rcseivoir MIC
Rpsln Arrochem
Filter MIC
Inorg. Aerochem
Solvent MIC
Reservoir MIC
Resin Aerochem
Filter MIC
Inorg. Aerochem
Solvent MIC
Reservoir MIC
Resin Aerochem
Filter MIC
Inorg. Aerochem
period ave. volume theoretical
aanple rainfall collected volume % collection Inorganic
to code (en) (ml) (ml) efficiency pH comments
*' w/new SA's
12-16
5-8
5-8
5-8
5-8
5-8
5-13
5-13
5-13
5-13
5-13
6-12
6-12
6-12
6-12
6-20
6-20
6-23
6-23
6-23
6-23
6-23
7-1
7-1
7-1
7-1
7-1
7-9
7-9
7-9
7-9
7-9
3-15 SN6074
5-13 SM6133
5-13 RM6133
5-13 RA6133
5-13 FM6133
5-13 IA6133
6-12 SM6163
6-12 RM6163
6-12 RA6163
6-12 FH6163
6-12 IA6163
6-20 FM6171
6-23 SM6174
6-23 RM6174
6-23 RA6174
6-23 FM6174
6-23 IA6174
7-1 SH6182
7-1 RM6182
7-1 RA6182
7-1 FM6182
7-1 IA6182
7-9 SM6190
7-9 RM6190
7-9 RA6190
7-9 FM6190
7-9 IA6190
7-14 SM6195
7-14 RM6195
7-14 RA6195
7-14 FW6195
7-14 IA6195
NA
6.83
6.83
6.83
6.83
6.83
3.86
3.86
3.86
3.86
3.86
0.38
4.57
4 57
4.57
4.19
4.57
1.575
1.575
1.575
1.575
1.575
2.985
2.985
2.985
2.985
2. -98 5
2.41
2.41
2.41
2.41
2.41
4140
8420 +
13380
5410
13800
4490
6960 *
6075 *
2315
6580 *
1487
650
8460 +
10090
2910
8560 *
3100
3140
3140
NA *
3000
810
6470
6220
1970
6510
1840
4810
5050
490
4850 *
1565
NA
14070
13933
5560
14070
4440
7952
7874
3142
7952
2509
783
9414
9323
3720
8631
2971
3245
3213
1282
3245
1024
6149
6089
2430
6149
1940
4965
4916
1962
4965
1567
NA
59.8 + Overflow
96.0
97.3
98.1
101. I 4 47/4.45
87.5 Cover open when arrived to sample
77.1 Water in funnel-lost lL(analyzed "> OHl.)
73.7 & Cover open when arrived to sample
82.8 Water in funnel-Lost 2 . 6L(analyzed 1 081.)
59.3 5.62/5.56
83.0 Water In funnel-Not enough to an.ilyr.p
89.9 + Overflow
108.2
78.2
99.2 Water In funnel-lost 2.1S1. (.iti.ily/i-il 6 '.II)
104.4 5.88/5.94
96.8
97.7
ERR * No Water In carboy-Vandals?
92.5
79.1 5.93/5.91
105.2
102.1
81.1
105.9
94.8 6.38/6.26/6.26
96.9
102.7
25.0
97.7 * Water In funnel-Pumped through resin
99.9 5.09/5.11
-------
Sunnary of 1<»6 Rain Field Data
, sMpllng perlo.
sampler »
n.iae ,«», , fVoai to
Solvent HIC^! Jfr
Reservoir M^K: ...
Rrsln Aerncnwvjj
t 1 1 tentH£fc' ^£Jf ii '
Inorjir^nJi-Jj,'
S..I ««•*,'&«:!£•&
"'••n\°$ If- 5-
l"i*r(*n?f^''p""
'"°»E> /^r&*1W
•,,,lvt'n&/)$ '*
Hi-s£in<&*r ^HJC£>
Res-W* jpjr «U2**eJJi|
H I.I e I ^M 1 Cftl* it.
liip'rfc •"fjeseifliiji
V *f ** 1 1. 1
>fi ff> ^- * * T
suiwiiifwiBi
Reservoir: nit ..i
"<•* l}> jAnrie^hrie-
1 iltci Mil" '"
Itioig Aerotlien
Solvent HIC
Reservoir HIC
Resin Aerochen
(liter HIC
Inorg Aerocheis
Solvent HIC
Reservoir HIC
Resin Aerochesi
(liter HIC
Inorg Aeroche*
Solvent HIC
Reservoir HIC
Resin Aerochesi
Filter HIC
Inorg. Aerochesi
Solvent HIC
Reservoir HIC
Resin Aerocheai
Filter HIC
Inorg. Aerochcn
7-
7-
7
7r
7-
7-
7-
7-
7
7
'•
7
7
,
a
8
a
8
a
t
8
j.
8
,.
9.
9-
9-
9-
9-
9-
9-
9-
9-
10-
9-
14
14
14
|4
U
22
22
22
2?
11
11
11
II
II
II
II
11
11
25
?•>
25
25
12
12
12
12
12
22
22
22
22
22
13
II
7-2}
7-22
7-21
7-22
7-22
7-31
7-31
7-31
7-31
7 31
8-11
8-11
• | |
* 1 1
811
811
8-25
8 25
8 25
8 25
8 25
9 12
9 12
9 12
9-12
9 12
9-22
9-22
9 22
9-22
9-22
10-13
10-13
10 13
10-13
10-13
11-14
11-14
d
aa*p la r
coda
SN6203
WM203
RA6201
FM6201
1A6203
SM6212
KM6212
FM212
IA4212
SH6223
KM6221
FH6221
IA6221
SH621/
RH6217
RA6717
111621'
IA6217
SH6255
RM6255
RA6255
FH6255
1A6255
SH6265
W)6265
M6265
FH6265
1A6265
SH6286
RH6286
RA6286
FH6286
IA6286
SH6318
IA6J18
ave . voluMe theoretical
2 79
2 79
2 79
2 79
2 79
4.105
4 105
4 105
4.105
4.305
4 064
4 064
•^ 06ft
4 064
4 064
3 127
1 127
J 177
1 127
1 127
4 267
4 267
4 267
4 267
4 267
966
966
966
966
966
.975
.975
.975
975
.975
1 664
5.639
(•D (•!>
6250
593O
2150
5690
1400
e no
7090 *
3750 *
8510 •
2690
8220
8020
2510 a
7690 *
2710
4910
5)50 a
2920
250 •
2100
6250 .
9410
15 JO
8880 *
2620
15040 »
16710 *
8100 a
16650 *
5610
7620
3910 *
3280
8350 *
3030
31tO
5747
5692
2271
5747
1814
1868
878?
))0ft
8868
2798
1372
8291
3308
im
2647
68*4
6717
2708
6154
2U3
8790
8705
3473
8790
2>74
18470
18291
7298
18470
5828
1189
1109
1214
8189
2584
1421
3665
efficiency
108
104
101
99
77
91
80
in?
IU 1
96
16
98
96
94
102
71
84
107
3
97
91
108
101
101
94.
81.
91.
Ill
90.
99
93.
48.
101.
102.
11.
It.
pH
7 »
2 •
5 •
0 •
2 5 51/5
7
7 «
04
"
2 •
1 >S 1,'l/b
2
7
if
•
2 *
6 6.40/6
9
7 •
8
6 «
1 4 50/4
9 »
1
6
0 •
5 4 82/4
4 »
4 *
0 a
I a
7 4 . 96/4
1
2 *
4
0 *
4
2 S.40/J
lie
coaMent s
two Intense rain
events, Including
one tornado approx
ten •! les south
51
.
Cover open wlien jirilvrri to sample
Water tn f mui«> 1 pumpeil through reslu
68
Water In 1 mine 1 pumped through resin
39
Water In teservolr pimped through rr*.lii
Fuse Blown
47
Ove r f 1 nw
Water In funnel -pimped through rr-.ln
80
Ove r f 1 OK
Watar In reservoir -drained through resin
Watar In reservoir -puatped through resin
Watar In funnel/Cover Open/Fuse Blown
94
Watar In reservoir ruaped through resin
Water in raaarvolr-puaiped/Cover open/Shut
Raln/SncM
laaovad Froai Fleld/Eltracted Reservoir
Raawved Froal Field/Extracted Raaarvolr
keaxrved Frosi Flald/titracted Reservoir
37**e»vad frost Field
Solvent HIC
11-14 12-4 SH6338
2000
No lain/Snow Gauge
00
-------
TECHNICAL REPORT DATA
(Please read lnstruct:ons on the reverse before completing)
1. REPORT NO.
EPA-905/4-90-002
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Field Intercomparison of Precipitation Samplers for
Assessing Wet Deposition of Organic Contaminants
5. REPORT DATE
March 1990
6. PERFORMING ORGANIZATION CODE
5GL
7. AUTHOR(S)
S.J. Eisenreich, T.P. Franz and M.B. Swanson
8. PERFORMING ORGANIZATION REPORT NO.
GLNPO Report No. 01-90
3. PERFORMING ORGANIZATION NAME AND ADDRESS
Environmental Engineering Sciences
Department of Civil and Mineral Engineering
University of Minnesota
Minneapolis, Minnesota 55455
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
R005840-01
12. SPONSORING AGENCY NAME AND ADDRESS
U.S. Environmental Protection Agency
Great Lakes National Program Office
230 South Dearborn Street
Chicago, Illinois 60604
13. TYPE OF REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY CODE
Great Lakes National Program
Office-USEPA, Region V
15. SUPPLEMENTARY NOTES
Project Officer - Edward Klappenbach
16. ABSTRACT
A field intercomparison of four wet-only precipitation samplers were performed
to assess their ability to efficiently collect rain and selected organic contaminants.
Samplers are evaluated and compared on the basis of their ability to efficiently
collect rainfall, exhibit mechanical reliability, demonstrate adequate operational
characteristics and provide precise measures of wet-only inputs. The most significant
difference between the four samplers was their mechanical reliability in the field.
The samplers performed equally well in assessing organic concentrations in rain.
This sampler intercomparison was conducted in part to select the preferred
characteristics of a rain sampler that must be deployed in the field unattended for
up to two weeks. The MIC sampler, properly maintained, is suitable for such a purpose.
Of the two modes of compound isolation tested, the resin adsorbent (XAD-2) exhibited
modestly higher concentrations that the solvent MIC but had the disadvantage of ease
of sample handling and lower blanks. Both could be operated with proper maintance
to provide precise data. The stainless steel and Teflon coated funnel surfaces pro
vided comparable data.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
c. COS AT I Field/Group
Atmospheric Chemistry
Wet Deposition
Atmosheric Organic Contaminants
Precipitation Samplers
Chicago, tl
18. DISTRIBUTION STATEMENT
Document is available through the National
Technical Information Service(NTIS)
Springfield, VA 22161
19. SECURITY CLASS {This Report I
21. NO. OF PAGES
20. SEC1 1ITY CLASS (This page)
22. PRICE
EPA Form 2220-1 (Rex. 4-77) PREVIOUS EDITION is OBSOLETE
-US GOVERNMENT PRINTING OFFICE 1990744-679
------- |