EPA-600/2-78-009
February 1978
Environmental Protection Technology Series
       SAMPLING AND  ANALYTICAL  METHODOLOGY  FOR
                                   ATMOSPHERIC NITRATES
                            Interim Report.   Evaluation  of
                                         Sampling  Variables
                                    Environmental Sciences Research Laboratory
                                         Office of Research and Development
                                        U.S. Environmental Protection Agency
                                   Research Triangle Park, North Carolina  27711

-------
                RESEARCH REPORTING SERIES

Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:

      1.  Environmental Health  Effects Research
      2.  Environmental Protection Technology
      3.  Ecological Research
      4.  Environmental Monitoring
      5.  Socioeconomic Environmental Studies
      6.  Scientific and Technical Assessment Reports (STAR)
      7.  Interagency Energy-Environment Research and Development
      8.  "Special" Reports
      9.  Miscellaneous Reports

This report has  been assigned  to the ENVIRONMENTAL PROTECTION TECH-
NOLOGY series. This series describes research performed to develop and dem-
onstrate instrumentation, equipment, and methodology to repair or prevent en-
vironmental degradation from point and non-point sources of pollution. This work
provides the new or improved technology required for the control and treatment
of pollution sources to meet environmental quality standards.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.

-------
                                             EPA-600/2-78-009
                                             February 1978
        SAMPLING AND ANALYTICAL METHODOLOGY
             FOR ATMOSPHERIC NITRATES

 Interim Report.  Evaluation of Sampling Variables

                      by
    Chester W. Spicer and Philip M. Schumacher

                   BATTELLE
             Columbus Laboratories
                505 King Avenue
             Columbus, Ohio  43201
          Contract No. 68-02-2213
              Project Officer

               James D. Mulik
 Atmospheric Chemistry and Physics Division
 Environmental Sciences Research Laboratory
Research Triangle Park, North Carolina 27711
 ENVIRONMENTAL SCIENCES RESEARCH LABORATORY
     OFFICE OF RESEARCH AND DEVELOPMENT
    U.S. ENVIRONMENTAL PROTECTION AGENCY
RESEARCH TRIANGLE PARK, NORTH CAROLINA 27711

-------
                                  DISCLAIMER

          This report has been reviewed by the Environmental Sciences
Research Laboratory, U.S. Environmental Protection Agency, and approved for
publication.  Approval does not signify that the contents necessarily reflect
the views and policies of the U.S. Environmental Protection Agency, nor does
mention of trade names or commercial  products constitute endorsement or
recommendation for use.

-------
                                 ABSTRACT

     The first phase of a study to develop sampling and analytical  methodolo-
gy for atmospheric nitrates is described.  Experiments were carried out to
determine the effect of nitrogen-containing gases on a number of different
filter materials.  Gases studied included NO, N02, HN03» and PAN.  Experi-
ments were also conducted to determine the effect of sampling time  and
sampling rate on atmospheric nitrate collection.   Studies of filter storage,
and gas-filtrate and gas-soot interactions were also undertaken.  In many
cases serious interference with the collection of atmospheric nitrate was
found.
                                     111

-------
                                   CONTENTS

 Abstract	111
 Figures	    V
 Tables 	    V
            Introduction	   1
            Results and Discussion	   8
                 Gas-Filter Interactions	   8
                 Gas-Filtrate Interactions	23
                 Gas-Soot Interactions	23
                 Effect of Sampling Rate	25
                 Effect of Sampling Time	25
                 Effect of Filter Storage on Particulate Nitrate. ...  28
            Summary	  30
References	31
Appendix
   A.   Experimental Condition 	  33

-------
                                  FIGURES
Number                                                                Page
  1   Phase I sampling and analysis apparatus	     9
  2   Close-up of sampling manifold and filter holdser 	     9
  3   Schematic of sampling systems for eight  simultaneous  filters  .    10
  4   Gas-filter interactions	15-16
  5   Gas-filter interactions	    18

                                   TABLES

Number                                                                 Page
   1   Gas Analysis Instrumentation	   11
   2   Physical/Chemical Properties of Phase I Filter Media	   12
   3   Methods of Analysis  	 	   13
   4   Results of Low Concentration Nitric Acid Experiment 	   19
        (350 ppb HN03)
   5   Filter Adsorption Results for Nitric Acid 	   20
   6   Pan-Filter Interaction Study Results	   21
   7   Presoiled Filter Analyses (mg/Filter) 	   24
   8   Soot Interaction Study Results^  (mg/Filter) 	   25
   9   Sampling rate Study  (mg/Filter)  	   26
 10   Results of the Sampling Time Study	   27
 11   Compound Filter Results 	   28
 12   Results of the Storage-Time Study	   29
                                     vi

-------
                               INTRODUCTION

          It has been known for many years that participate nitrate is a
fundamental  constituent of our atmosphere and that the nitrate burden
increases considerably as one approaches our major urban centers.  Ques-
tions as to the physiological impact of particulate nitrates were raised
at the time of the Chattanooga epidemiological study and have been asked
with increasing frequency since that time.  However, it has only been in
recent months that the correlations between particulate nitrates and certain
types of morbidity have become available.  Certainly a strong indication
that nitrates cause detrimental health effects is the recent finding^ '
that airborne nitrates are associated with aggravation of asthma, even in
areas where primary ambient air quality standards are not exceeded.
          Particulate nitrate has been determined for many years through-
out the United States by standard high volume sampling techniques using
glass-fiber filters.  Robinson and Robbinsv ' have estimated the global
background nitrate concentration to be on the order of 0.2 yg/m  in the
lower atmosphere.  Measurements of nonurban nitrate levels by the National
Air Surveillance Network^ ' generally exhibit a range of annual averages
between 0.1 and 1.0 yg/m , with an overall mean of approximately 0.5 yg/m .
The lowest nitrate values are found in such areas as Glacier National  Park
in Montana and Black Hills National Forest in South Dakota, well away from
industrial and population centers.  Nitrate as nitric acid, has been shown1  '
to exist in the stratosphere at concentrations on the order of 0.003 ppm,
and is closely associated with the stratospheric ozone layer.  There is
             (5)
some evidence^ ' that this stratospheric nitric acid can be transported
across the tropopause and thus contribute to the background tropospheric
nitrate burden.

-------
          The level of participate nitrate in and around major urban centers
can be considerably higher than background levels.   Based on an 8-year
study^ ' of the mean concentrations of selected particulate contaminants
in the atmosphere of the United States, it appears  that nitrate on the
average contributes somewhat less than 2 percent of the total  suspended
particulate weight.  The figures vary depending on  location.  Several
                                                   o                     o
representative urban areas were Atlanta at 2.0 yg/m, Chicago at 2.5 yg/m ,
and Pittsburgh showing 3.0
          In certain areas, such as the California southcoast basin, the
                                                       3
nitrate levels are even higher, averaging nearly 5 yg/m  in  the vicinity
                                                 ( 3)
of downtown Los Angeles according to NASN results v  .   It is interesting
to note however, that Grodon and Bryan^ ' have reported yearly average
nitrate levels in downtown Los Angeles ranging between 9.8 and 15.4 yg/m .
Short-term nitrate levels in the eastern basin have been reported as high
as 247 yg/m3(8).
          The major source of atmospheric particulate nitrate is thought
to be oxidation of natural and anthropogenic NO and NOo.  The major sinks
for particulate nitrate are precipitation scavenging and dry deposition,
                                          (2)
with the precipitation mechanism estimatedv ' to be three times as important
as dry deposition on a global basis.
          Between the emission of gaseous NO or NOo and the ultimate scaveng-
ing of the particulate nitrate, there is a highly complex series of reactions
which may result in a variety of reaction products prior to the ultimate
formation of particulate nitrate.  In a recent smog-chamber study of nitrogen
oxides reactions conducted by Spicer and Miller^ ' at Battelle- Columbus, the
major initial products of nitrogen oxides reactions in simulated photo-
chemical smog were peroxyacetyl nitrate (PAN) and nitric acid.  Excellent
nitrogen mass balances were maintained throughout the experiments and the
complex mechanisms leading to the formation of organic and inorganic nitrates
were investigated.  There  is little doubt that in the actual  atmosphere
several other forms of nitrate exist.  For example, we have observed^   '
low concentrations of alkyl nitrates and peroxypropionyl nitrate  in urban
atmospheres.  We have also detected low levels of particulate organic
nitrate in atmospheric aerosol samples^    .  Heuss and Glasson^   ' have

-------
observed another organic nitrate,  peroxybenzoyl  nitrate  (PBzN), in smog
simulations.   There is also reason to suspect  important, albeit low,  levels
of NpO,-, a nitrate precursor,  in  urban atmospheres.   It  is  almost certain
that all of these forms of gaseous nitrate  ultimately end up as particulate
nitrate.
          Because of differences  likely to be associated with the physio-
logical impact of gaseous and particulate nitrates, it is important to
distinguish between the two.  The most common means of collecting particulate
nitrate is filtration.  However,  there are a number of known problems
involved in filtration sampling of particulate nitrate, and there are also
several potentially serious problems which are not so widely recognized.
As with any system designed to collect samples by filtration, a most
                                                               /13)
important variable is  the filter medium itself.  Pate and Taborv  ' have
described the characteristics of a wide variety of glass-fiber filters
which have been employed by NASN and other for nitrate collection.   The
manufacture of such glass-fiber filters required at least six separate
steps and usually four participating manufacturers.  It is therefore not
surprising that substantial variations in the composition and character-
istics of these filters often occur.  Such variations can affect the
accuracy or efficiency of particulate nitrate sampling by
          (1)   Changing the nitrate  filter blank
          (2)  Affecting the efficiency of particulate
               collection
          (3)  Affecting the degree to which particulate
               nitrate may react with the filter and become
               unavailable for leaching and analysis
          (4)  Affecting the extent of extraneous nitrate
               formation on the filter by nitrate precursors.
                         (13)
           Pate  and Taborv   '  have reported that the  nitrate blank can
 typically  comprise up to  10 percent  of the nitrate collection in urban
 areas,  and a  considerably  higher  fraction  of  nonurban  filter samples.
 However, as  long  as  frequent  blank determinations are  carried out,  this
 should  not be a major difficulty  except when  sampling  extremely low levels
 of particulate  nitrate.

-------
          A potentially serious problem in the collection of participate
nitrates was pointed out in ig/^10'14'15'.   This  problem relates  to the
collection of artifact nitrate on filters due to the interactions  of
gaseous nitrogen compounds with certain filter materials.  Two different
studies at Battelle-Columbus have revealed the problem of artifact nitrate
collection on certain filter materials.  In one study^  ' investigating
nitrate in auto exhaust, it was found that glass-fiber filters collected
almost twice the quantity of nitrate in exhaust as did quartz-fiber filters.
Nitrate also appeared on backup filters for both quartz and glass, providing
an additional indication of artifact nitrate formation.
          In a separate investigation at Battelle, the interaction between
nitric acid and a variety of filter materials was studied.  The impetus
for this study came from the discovery of discrepancies in our atmospheric
nitrate data" ' collected on quartz as opposed to glass filters.   The
results of the study and the implications in terms of atmospheric chemistry
and past particulate nitrate data have been reported^  ' and presented'  ''
elsewhere.  Briefly we find that both the absolute concentrations  and also
the assumed size distributions of ambient particulate nitrate from many past
studies may be in error, due to gaseous nitric acid interference.
          One possible explanation for the different collection efficiencies
of gaseous nitric acid by quartz and glass filters involves the filter pH.
Studies of filter characteristics at Battelle-Columbus^  ' have shown that
quartz filters are nearly neutral (100 ml filter extracts yields pH of 5-7)
while glass filters are often distinctly alkaline (100 ml filter extract
yields pH of 9.4).  Thus, neutralization and trapping of nitric acid and
other acid gases may occur to a greater extent on glass filters than on
quartz.
                         (18)
          O'Brien, et al.v  ' have described a study of photochemical
aerosols in the Los Angeles basin in which high-volume samplers and a
cascade impactor were employed to determine concentrations and size dis-
tributions for NO^, NH4, and SO^, among other species.  The high-volume
sampler and each stage of the cascade impactor employed glass-fiber filters
as the collection medium.  The results of the study yielded very unusual
nitrate size distributions which appeared to be dependent on sampling site.

-------
The investigators reported that the strange results could be attributed to
formation of artifact nitrate on the glass filters by some gaseous pre-
cursor such as nitric acid, and that the true nitrate size distribution
was masked in their study by the conversion of gaseous nitrate precursors
on the filter.
          In addition to interactions between gaseous nitrate precursors
and filter media, several other potential interferences with particulate
nitrate collection should be considered.  For example, certain nitrate pre-
cursors may be stable with respect to the filter medium but may be converted
by interaction with some component of the aerosol collected on the filter.
An example of this type of interaction might be the formation of NaNO^ by
the reaction of NOo with NaCl collected on the filter.  In addition, there
may be certain conditions of relative humidity, temperature, atmospheric
composition, etc., under which species such as NH^. NC^, PAN, N^Og, or ^0
could form and/or react with filters or collected aerosol on the filters.
Some precursors may be held on the filter by absorption long enough to be
oxidized to artifact nitrate by ozone or some other oxidizing agent.
          Another type of interference might involve release of particulate
nitrate collected on a filter by conversion to some volatile form.  An
example might be the reaction of sulfuric acid aerosol with particulate
nitrate already collected on the filter to form nitric acid, which could
then be lost by volatilization.  In the same manner NH.NO^, which has a
significant vapor pressure, could be lost from particulate collection
filters.
          Two recent studies which touch on the kinds of interferences
                                                         (19)
just discussed have been reported by Lovelock and Penkettv  ' and Chang
and Novakov*  '.  The former investigators found that PAN and PPN do not
exist in the clean air over the Atlantic ocean but that clean air has the
potential for forming PAN and PPN when exposed to glass surfaces.  The
potential for PAN formation was greatest on days of high solar intensity,
with maximum production during the afternoon hours.  An important aspect
of this study is the possibility that there are precursors even over the
oceans which will form gaseous nitrates given the proper reaction surface.

-------
Since the proper reaction surface was glass,  the same material  used for
filters in most high-volume samplers, the possible importance of this
reaction mechanisms for particulate nitrate formation must be considered.
          Chang and Novakov^  '  have studied the formation of particulate
nitrogen species due to gas/solid interactions using ESCA.  They report
the formation of several types of particulate nitrogen due to the reaction
of gaseous NO and NH3 with soot (carbon) particles.   Since soot, along with
other forms of carbon, is collected on high-volume filter samples, the
possibility of forming artifact nitrate from NO and NH^ reactions exists.
          Because of the potentially serious effect of the interferences
described above on the accuracy of particulate nitrate data, the U.S.
Environmental Protection Agency has initiated a 2-year study at Battelle-
Columbus to investigate the impact of these factors on nitrate sampling
procedures.
          The objectives of this program are threefold:
          (1)  To investigate the effects of environmental
               variables on the sampling and analysis of
               particulate atmospheric nitrate
          (2)  To develop an improved method for the
               analysis of particulate atmospheric nitrate
          (3)  To conceive, develop, validate, and optimize
               a sampling and analysis methodology for
               atmospheric nitrate.
          The experimental aspects of the program have been broken into
four phases.  A description of each phase is shown below:
          Phase I    A laboratory investigation of the factors
                     affecting atmospheric particulate nitrate
                     sampling
          Phase II   Development of analytical methodology for
                     atmospheric particulate nitrate
          Phase III  Development and evaluation of a sampling
                     procedure for atmospheric particulate
                     nitrate

-------
          Phase IV   Optimization, simplification, and delivery
                     to EPA of a validated sampling and analysis
                     methodology.
          The first year of the program has been devoted almost entirely
to Phase I.   This Interim Report describes the Phase I effort.  Subsequent
interim reports will describe Phase II and Phase III.  At the conclusion
of the program a final report will summarize the results of the entire
project.
          The main tasks that were undertaken in Phase I include inves-
tigations into
          •  Gas-filter interactions
          •  Gas-filtrate interactions
          •  Gas-soot interactions
          •  Effect of sampling time
          •  Effect of sampling rate
          •  Effect of storage time.
          The remainder of this report will be devoted to a discussion of
these tasks.

-------
                             RESULTS AND DISCUSSION
GAS-FILTER INTERACTIONS
          The principal  goal  of this task is to determine the effect of
several nitrogen-containing gases on common filter media and to identify
types of filters which might be suitable for participate nitrate sampling
in the atmosphere.   The experimental procedure has involved exposing
selected filter materials to exaggerated concentrations of nitrogen-containing
gases in clean air.   Nitrogen-containing gases are first diluted with clean
cylinder air in a 500-cu-ft Teflon^ chamber.  The chamber is then evacuated
through 47-mm filters of the various materials chosen for study.  The
pressure and concentration of the nitrogen gas are monitored above and
below each filter during the experiment; after exposure the filters are
analyzed for NOZ. N0p» NH*. and total  N.  The experimental apparatus is
shown in Figures 1  and 2.  A schematic of the apparatus is shown in Figure 3.
Each filter is typically exposed to more than 1 cubic meter of the dilute
nitrogen-containing gas.  The concentration of the nitrogen gas is adjusted
(low ppm) so that the filter is exposed to approximately the same mass of
nitrogen compound as a standard high-volume filter collected in an urban
area.  The face velocity is also quite similar to a standard Hi Vol.
          The nitrogen-containing gases examined thus far include NO, N02>
HNO^, PAN, NFL, and N20.  The analytical techniques used for these gases
are listed in Table 1.  N20 was not determined directly but was prepared
by known dilutions  of the pure gas.  Both dry and humidified conditions
have been employed.   We have investigated the effect of these nitrogen-
containing gases on artifact nitrate formation on a number of filter types.
Most of our experiments were conducted with the following filter types:

-------
      Figure  1.   Phase  I  sampling  and  analysis  apparatus.

Figure 2.   Close-up of sampling manifold and filter holders,




                              9

-------
              To insfruments
              and manometer
                       To filters
                        MM
                    Sample port
                         t u
                        To filters
500 cu ft
Teflon chamber
                             Filter holder
                                  To instruments
                                  and manometer
                          Critical
                          orifice
              To vacuum pump
Figure  3.  Schematic of sampling systems for eight simultaneous filters.
                                10

-------
           Glass  Fiber  -  Gelman  A
           Glass  Fiber  -  Gelman  E
           Glass  Fiber  -  Gelman  AE
           Teflon - Millipore Mitex
           Polycarbonate  - Nuclepore
           Nylon - Millipore DuraIon
           Cellulose Acetate - Mi Hi pore Celotate
           Quartz Fiber - ADL.

                 TABLE 1.  GAS  ANALYSIS INSTRUMENTATION

                  Gas               Analysis Method

                NO, N02         Chemiluminescence (low
                                temperature carbon con-
                                verter for NOo)
                HN03            Microcoulometry
                PAN             Electron Capture Gas
                                Chromatography
                NH3             Chemiluminescence (dual
                                temperature catalytic con-
                                verter

          Toward the end of the Phase I effort a limited number of experi-
ments with some additional filters were conducted.   These experiments
involved
          Quartz Fiber - Pal Iflex QAST
          Glass Fiber - Gelman AA
          Glass Fiber - Gelman Spectrograde
          Cellulose-Backed Quartz - Pallflex E 70-2075 U.
A brief description of each of these filter types is shown in Table 2.   The
procedures employed for analysis of the filter samples are summarized in
Table 3.
                                     11

-------
              TABLE 2.  PHYSICAL/CHEMICAL PROPERTIES OF PHASE I FILTER MEDIA
Filter Type
Cellulose Acetate
Polycarbonate
Teflon-Mitex
Glass-AA
Glass-A
Glass-AE
Glass-E
Glass-Spectro
Nylon
Quartz-ADL
Quartz- QAST
Quartz- E 70-2075
(a) Pore size in
(b) ASTM-D-202;
Pore Sizi
1.0
0.8
5.0
NA
NA
NA
NA
NA
1.0
NA
NA
W NA
micrometers
pH of 100 ml
la) (h) lc) - (ri)
2V ' pHv ' Alkalinityv ' NOo Blankv '
6.65
6.0
7.0
8.9
8.3
9.4
8.5
7.2
5.3
8.1
8.1
6.2
where applicable
H?0 extract.
(1.
(9

3.
4.
4
3.
1
(3.
1
3.
(7.
^

8 x 10-3)
x ID'4)
0.0
24 x 10-2
2 x 10-3
x 10-2
8 x 10- 3
x 10-4
6 x lO'3)
x TO'4
8 x 10-3
6 x 10'3)


<0.005
<0.006
<0.005
<0.005
<0.005

<0.005
<0.005
S0.007
<0.005
<0.005
<0.005


Supplier
Millipore
Nuclepore
Millipore
EPA/Gel man
Gel man
Gel man
Gel man
Gel man
Mi Hi pore
EPA/ADL
Pallflex
Pallflex


(c)  Milliequivalents  of  acid  or base  required  to  titrate  to  neutral  point per gram of
     filter.   Parentheses  indicate acidic  filter.

(d)  mg/47 mm filter.

-------
                         TABLE 3.   METHODS OF ANALYSIS
                     Specie                  Method of Analysis

                 NH^                   Gas  sensing electrode
                 NO^                   Diazotiazation-colorimetric
                 H0~~                   Brucine sulfate colorimetric/
                                       Ion  chromatographic
                 Total  Nitrogen        Modified Kjeldahl  digestion
          All of the filters exposed thus far have been analyzed for nitrate
and total nitrogen and many have also been analyzed for ammonium and nitrite,
depending on the gas being studied.  Out of all the filters analyzed to date,
none have shown any significant increase in the nitrite concentration under
any circumstance.  Out of all the filters exposed to gaseous NH^, only the
nylon filters have shown substantial increases in particulate ammonium levels,
          The results of the nitrate analyses conducted to date are shown
in Figures 4 and 5.  These figures show in bar-graph format the quantity
of artifact nitrate found on the filter after exposure to ppm concentra-
tions of the gases shown on the left side of the figure.  The experiment
number shown at the left of the figure is keyed to a complete tabulation
of the experimental conditions contained in the Appendix.  In most cases,
two or more different concentrations of the nitrogen-containing gas have
been employed, and frequently both dry and humidified conditions have been
examined.
          For purposes of this discussion, we will arbitrarily set the
level  of significant nitrate interferences as >100 yg.   Since we are
dealing with a surface effect and a standard 8" x 10" Hi-Vol. filter has
40 times the effective collecting area of the 47-mm filters used in this
study, a standard Hi-VoT.  filter might be expected to collect about 40 times
as much artifact nitrate as our filters.   If our filters collect 100  g of
                                     13

-------
bogus nitrate, a standard Hi-Vol. might collect 4000 yg of artifact.  Assum-
                                     o
ing a 24-hour sample volume of 2000 m  for a standard Hi-Vol. filter, the
level of nitrate interference would be 2 ug/m3.  This concentration approaches
the average NOg concentration measured in many urban areas and thus con-
stitutes a major interference.
          Figure 4a shows the results of our experiments with cellulose
acetate filters.  Two different experiments with nitric acid exhibit sub-
stantial interferences on this filter type, thus precluding its use as a
                                     \
particulate nitrate .sampling medium.
          The-results of our studies of polycarbonate filters are shown
in Figure 4b.  It is clear from this figure that none of the gases studied
to date yield significant artifact nitrate on polycarbonate filters.
          The results with Teflon filters, pictured in Figure 4c, are
similar to the polycarbonate results in that no important interferences
are apparent from any of the gaseous species stuied to data.
          The results of our investigation of nylon filters are shown in
Figure 4d.  Nylon was chosen for study in the hopes that it might provide
quantitative collection of gaseous nitrates and thus serve as a gaseous
nitrate sampling technique.  It is clear from the figure that large
quantitites of nitric acid are collected by the nylon media.  Judging from
the final experiment, high levels of N02 (30 ppm) at high humidity also
lead to very high levels of artifact nitrate.  The first NH^ experiment
indicates formation of considerable artifact NOo, while the second and
third NH3 runs show no such effect.  We suspect that some NH^NO-j may have
formed in our Telfon chamber during this experiment from trace quantitites
of HNO^ remaining in the chamber from the previous experiment.  Thus the
results of the first NH3 experiment are probably in error.
          The results of our investigations of two types of glass-fiber
filters (Gelman A and E) are shown in Figures 4e and 4f.  Both filter
materials show substantial interferences from nitric acid and also from
high concentrations of NOo at high relative humidity.  The interference
by N02 is extremely important due to the high concentrations of N02 which
frequently occur in urban areas.  The interferences with particulate nitrate
collection on these alkaline-surfaced glass filters makes them rather poor
choices for nitrate sampling in the atmosphere.
                                    14

-------
Experiment Cellulose Acetate
1 Cleon Air
2 Clcr.n Air
3 NOj
< NO?
5 NO,(RH)
6 HNOj
7 HNOj
8 HNOj
9 HNOj
10 HNOj(RH)
11 NHj
Blonk
12 NHS
13 PAN
14 N,0
15 N,0(F(H)
16 NHj(RH)
17 NO; (KM)






L,
	 1
~l
1
J






1
i i i — i — i — i — i —
Expcr iment
HO. Polycarbonate
1 Cleon Air
* Cleon Air
3 NO?
4 NO,
5 NO,(RH)
6 HNOj
7 HNOj
8 HNOj
9 HNOj
"> HNO, (RH)
11 NHj
Dlo-ik
x '2 NHj
13 PAN
14 N,0
15 N]0(RH)
15 NKj(WH)
17 NO,(RH)





































0 00 300 JOO 400 bOO bOO _ KAJ bU. 0 OO ?io 3OC 400 -.TO GOO 1m 600
Micrograms Nitrate on Filter Micrograms Nitrate on Filter
4a. 4b.
Experiment Experiment
NO. Teflon No. Nylon
1 Clcun An
? Cleon Air


4 * 1

5 NO, (R H ) 1
« HNO, ft
7 HN°i
8 HNOj
9 HNOj
10 HNOj(RH)
11 NH,
Blonk
12 NHj
13 PAN
14 N,0
15 N,0(R.H)
16 NHj(RH)
17 NO,(RH)



J







1 Cleun Air
t Cleon Air
3 NOZ
< NO,
5 NO? (R H )
6 HNOj
7 HNOj
8 Hf.'O,
9 HNOj
10 HNOj (R K)
11 NHj





2500 /.!) j

9700 /iq j
ESOOjig J
'S,300,.g \
\
Blank jj
12 NH, j
13 PAN n
14 NZ0
15 N,0|DH)
16 NH,(RH)
17 NO,(RH.)

~~\
\
Z?00;ig






n
n
n
rj







0
100  ?00  300  400  !>00  600   70O
Micrograms Nitrate on Filter

           4c.
CO  ?OO  JCO  «(jO  1OO  OOO   7O3  600
Micrograms Nitrate on Filter

            4d.
                Figure 4.   Gas-filter interactions.
                            15

-------
Expe>' 1 men t
No.
1 CleeM Air
^ Clcon Air
3 NO?
4 NO,
& NOj(RM)
' HNOj
1 UNO,
8 HNOj
9 HNOj
10 HNO, (RH)
11 NH,
Blor.ii .
12 NH,
13 PAN
14 N,0
15 NjO(RH)
16 NHj(RH)
17 NOj(RH)
Glass Fiber - Gelman A


]
1
"~|
1
1
1
1 ,
1


~|
r_.
_j

i
11 - j — i — i — i — i —
Experiment
No.
1 Clcon Air
2 Cloon Air
3 NOz
4 NOZ
5 NO; (R H )
6 HNO,
J HNOj
8 UNO,
9 HNO,
10 HNOj (RH)
11 NH,
Blonk
12 NH,
13 PAN
14 NjO
15 N,0 (R.H)
16 NH,(R.H)
17 NOz(RH)
Glass Fiber - Gelman E




i
|
1

1
1
1





1
1
i i i i t i .
 OO   ZOO  JOO  400   500  fX>   700   80O
Micrograms Nitrate on  Filter
            4e.
                                                    200  300  
-------
          The results of our investigations using quartz-fiber filters are
pictured in Figure 4g.  In terms of artifact nitrate collection, the quartz
filters look quite good.  Only one experiment, at high nitric acid and
high relative humidity, shows any significant interference.   However, the
results from this one experiment are suspect since the total  nitrogen
analysis indicated no artifact nitrogen collection by the filter.
          The results of a limited number of experiments with four additional
filter materials compared with the ADL quartz filters are shown in Figure 5.
The Pallflex quartz filters show no evidence of serious interference.
However, the three remaining materials were all influenced by nitric acid,
and the Gelman AA filter by nitrogen dioxide.  The interference with
particulate nitrate collection on Gelman AA filters is particularly signif-
icant since the material is widely employed in the NASN program.  We strongly
suspect that the surface coating of the Spectrograde filter is attacked by
nitric acid, the interaction with the exposed surface resulting in very
high levels of bogus nitrate.
          All of the experiments described above have involved sampling
about 1 cubic meter of air containing ppm quantities of some nitrogeneous
gas through individual filters.   While the results of these experiments
will only be used to indicate types of filters suitable for particulate
nitrate sampling, still it would be useful to understand the relationships
among artifact nitrate formation, sample concentration and sample volume.
The results of a preliminary investigation of these relationships are
listed in Table 4.  The experiment (No. 22) shown in this table was designed
to investigage the effect on filters of sampling low nitric acid concentra-
                                                    3
tions over longer exposure times.  Approximately 4 m  of the sample mixture
(350 ppb nitric acid in air) passed through the filters.  Based on the
100 yg interference criterion put forward earlier, the glass filter shows
significant artifact nitrate.  The ADL microquartz was unaffected by the
dilute gaseous nitric acid, while the Pallflex tissuequartz collected
artifact nitrate at levels approaching the 100 pg interference criterion.
We plan to investigate these relationships further at the conclusion of
the Phase III field study.
                                     17

-------
               Experiment
oo
No.
23
24
25
23
24
25
23
24
25
23
24
25
23
24
25
N02
HN03
N02 (R. H.)
N02
HN03
N02 (R.H.)
N02
HN03
N02 (R.H.)
N02
HN03
N02(R.H.)
N02
HN03
N02 (R-H.)

Quartz - Microquartz
Quartz- Tissuequartz
1
^
Quartz - Fiber - Backed
1

	 |
Glass Fiber- Getman Spectro
1

Glass Fiber- Gelman AA
1

]
1 1 1
0 100 ZOO 300
                                                      Micrograms Nitrate on Filter
                                        Figure 5.   Gas-filter interactions

-------
                   TABLE  4.   RESULTS  OF  LOW CONCENTRATION
                             NITRIC ACID EXPERIMENT (350 ppb HN03)
Filter Material
Glass (Gelman AE)
Quartz (ADL Microquartz)
Quartz (Pallflex
Tissuequartz)
N0"3
0.23
<0.005
0.086
NHj
<0.003
<0.003
<0.003
Total N
0.04
0.01
0.03
           The  interaction  of the  gaseous  nitrates  PAN  and  nitric  acid with
 filter  materials  is  of particular concern in  this  study  because (1) we  have
 observed  these gases at relatively high concentrations ( 50  ppb)  in urban
 atmospheres  and (2)  they could  easily form artifact nitrate  by simple
 adsorption on  filters.   In addition to the experiments already reported
 with these two gases,  we have carried out several  experiments using a
 different technqiue.   In these  experiments either  PAN or nitric acid was
 generated at low  (sub-ppm) concentrations  in a 200 liter glass smog chamber.
 The mixture  was then pulled  through the filters under study.  The concentra-
 tion of PAN  or nitric  acid in the  chamber was determined at  the start of
 the experiment, and  then the concentration was monitored down stream of
 the filters.   After  an  initial  induction  period during which the gaseous
 nitrates  were  removed  by the filter, breakthrough occurred and the concen-
 tration downstream of  the filter slowly increased to the level in the chamber.
 The breakthrough response curve was integrated and used to calculate the
 mass of gaseous nitrate  removed by the filter.  A system blank (same
 apparatus but filter holders empty) was run for each gas  and subtracted
 from the  filter results.
          The results of the filter adsorption experiments for nitric acid
 are shown in Table 5.  These data are in substantial  agreement with the
nitric acid results presented earlier,  i.e.,  nitrate interference  due to
nitric acid adsorption on Teflon,  quartz and  polycarbonate filters is
negligible.  Interference due to adsorption on the two types of glass
filters  can be  serious.
                                    19

-------
        TABLE 5.   FILTER ADSORPTION RESULTS FOR
                  NITRIC ACID
                              Nitrate Removed by 47-mrn
    Filter Material              Diameter Filter, yg
Teflon                                   0.13
Quartz                                   0.11
Polycarbonate                            2.8
Glass (Gelman E and AE)     Very large (no breakthrough
                            after 8 hours of sampling)
                        20

-------
           PAN was generated at ambient concentrations for the breakthrough
 experiments by the photolysis of ethyl nitrite in dry air.   A gas chromato-
 graph equipped with an electron capture detector was  used to monitor the
 PAN.   The mass of PAN removed by the various  filters  is  shown in Table 6,
 along with a list of the potential  artifact nitrate which might be expected
 under actual ambient sampling conditions for  each filter type.   These data
 indicate, as did the results  from experiment  13,  that nitrate interference
 due to PAN adsorption is unimportant for the  filters  studied.

                  TABLE 6.  PAN-FILTER INTERACTION STUDY RESULTS
                                                       Potential Hi-Vol.
                                     PAN Adsorbed,   Nitrate Interference,
            Filter Material              yg                  yg
Quartz - Microquartz
Quartz - Pal If lex 2500 QAST
Quartz - Pal If lex E 70-2075 W
Glass - Gelman AA
Glass - Gelman A
Glass - Gelman E
Glass - Gelman AE
Glass - Gelman Spectrograde
Nylon - Duralon
Cellulose Acetate
Polycarbonate
Teflon
0.003
0.006
0.009
0.35
0.010
0.012
0.18
0.020
0.008
0.007
0.007
0.005
.065
.13
.20
7.6
.22
.26
3.9
.43
.17
1.3
.15
.11
          A check of the filter nitrogen balance for nylon, Gelman A and
Gelinan E filters was carried out using the results of experiment numbers
6, 7, 8, 9, 10 and 17.   These experiments were chosen because the quantity
of artifact nitate found on the filters was well above the filter blank.
The percentage of the total filter nitrogen which could be accounted for
as nitrate-nitrogen has been calculated for these three filter types,
                                     21

-------
taking both the nitrate and total  nitrogen blank into account.   The average
accountability for the six experiments was
          Nylon - 71  percent
          Gelman A -  127 percent
          Gelman E -  93 percent.
The value shown for nylon is probably not very accurate due to  the large
total nitrogen blank  correction.  The Gelman A percentage is unexpectedly
high.  The high percentage seems to result from low total nitrogen analyses
of Gelman A filters for two of the experiments.  The reason for the low
analytical results is unknown.
          With the laboratory data obtained in the Phase I effort, it is
extremely dangerous to attempt to quantitate the impact of artifact nitrate
formation on filters  under actual  ambient sampling conditions.   The greatest
danger lies in extrapolating directly from our laboratory results—obtained
under high concentration/low sample volume conditions, to ambient (low
concentration/high volume) conditions.  The Phase I study was not designed
to quantitate the interference so much as to screen prospective filter
materials in terms of their suitability for particulate nitrate sampling.
A subsequent phase of this program will investigate the impact  of artifact
nitrate formation under actual  ambient conditions and should provide a much
more accurate estimate of the extent of nitrate interference.  Additional
laboratory studies are also underway with the objective of defining the
relationships between nitrate interference and precursor concentration/
sample volume.  Indications from our laboratory data suggest that we
saturated the surface sites of many of the filters early in our experiments
by using ppm quantities of precursor gases.  Under such conditions the
apparent artifact collection efficiency is lower than expected  under ambient
sampling conditions.   In other words, we believe that the percentage of
artifact nitrate  interference reported here may only be the lower limit to
that expected under ambient sampling conditions.
                                    22

-------
GAS-FILTRATE INTERACTIONS

          As mentioned in the introductory section of this report there are
several possible interactions between gases or aerosols in ambient air and
the particulate matter (filtrate) collected on high volume filters, which
could result in either positive or negative interference with particulate
nitrate determinations.  The interaction between gases or aerosols passing
through a filter and the collected filtrate was investigated by exposing
actual ambient filter collections to exaggerated concentrations of several
potential interferences.   The concentration of nitrate on the filters was
determined both before and after exposure, so that a simple comparison
should reveal any significant interferences.  In terms of experimental pro-
cedure, ambient Columbus aerosol was collected by Hi Vol sampling on several
142-mm filter media.  47-mm circles were cut from these filters and exposed
to candidate  gases in the same apparatus used for the gas-filter interaction
studies.   The results of nitrate analyses of these filters are shown in
Table 7.   The lack of consistent changes in NOo levels upon exposure indicates
that interactions between candidate substances and the collected particulate
are not significant.

GAS-SOOT INTERACTIONS

          The potential role of soot collected on Hi Vol filters in con-
verting gases such as NO? and NH, to nitrate has been discussed by Chang
            (20}
and Novakovv   .  The importance of gas-soot interactions was investigated
in this study by loading 142-mm Gelman A filters with 4 mg of finely dis-
                                                      •i
persed carbon-black and then passing approximately 2 m  of air containing
ppm levels  of NH3 or NO^ through several 47-mm diameter circles cut from
these filters.  Analysis of these filters gave the results shown in Table 8.
The nitrate  level increased after exposure to N02, however, the magnitude
of the increase is below the level of artifact nitrate formed during exposure
                                    23

-------
 TABLE 7.  PRESOILED FILTER ANALYSES (mg/FILTER)
Filter Material
NO:
NH,
Total N
Presoiled Filters (Before Exposure)
Glass-Gelman AE (1)^
Glass-Gelman AE (2)
Cellulose Acetate (3)
Cellulose Acetate (4)
Quartz-Microquartz (5)
Quartz-Tissuequartz (8)
Exposed Filters
Experiment No. 18 - 11.5 ppm NH3
Glass-Gelman AE (1)
Cellulose Acetate (3)
Quartz-Microquartz (5)
Experiment No. 19 - 97 yg/m3 H2S04
Glass-Gelman AE (1)
Cellulose Acetate (3)
Quartz-Microquartz (5)
Quartz-Tissuequartz (8) /.x
Quartz-Microquartz (unsoiledr
Experiment No. 20 - 3.4 ppm HNOs
Glass-Gelman AE (1)
Cellulose Acetate (3)
Quartz-Microquartz (5)
Experiment No. 21 - 19.5 ppm N02
Glass-Gelman AE (1)
Cellulose Acetate (4)
Quartz-Microquartz (5)
Quartz-Tissuequartz (8)
(a) Numbers in parentheses identify
(b) S04"2 was 0.087 for this filter.
0.36
0.26
0.037
0.047
0.038
0.12


0.35
0.033
0.025

0.28
0.030
0.055
0.12
<0.005

0.40
0.26
0.037

0.22
0.043
0.028
0.11
ports used

0.007
0.006
0.014
0.027
0.028
0.072


0.005
0.016
0.026

0.068
0.045
0.10
0.19
0.092

0.011
0.028
0.025

0.013
0.021
0.020
0.072
during

0.10
0.10
0.06
0.09
0.06
0.15


0.13
0.10
0.07

0.17
0.14
0.14
0.20
0.11

0.13
0.22
0.06

0.10
0.07
0.05
0.14
presoiling

                       24

-------
              TABLE 8.  SOOT INTERACTION STUDY RESULTS
                        (mg/FILTER)
Exposure Conditions
Filter before exposure
11.5 ppm NH3
19.5 ppm N02
NO-
0.008
0.010
0.017
NHj
<0.003
<0.005
<0.003
              (a)  Filter medium used was Gelman A with
                   4-mg carbon-black loaded on a 142-mm
                   circle.
of clean Gelman A filters alone.  Thus the gas-soot interaction does not
appear to contribute significant quantities of artifact nitrate.

EFFECT OF SAMPLING RATE
          To examine the effect of sampling rate on nitrate collection
efficiency and artifact nitrate formation, ambient Columbus aerosol was
collected simultaneously over 24 hours on two groups of four Gelman AE
142-mm filters, each at two flowrates.  The results of this study are
included in Table 9.  The average nitrate collected on the four filters
                                                 •^
run at 38.2 £/min flow rate was 2.65 ± 0.16, wg/m  while the average for
the 99.3 £/min rate was 2.45 ± 0.20 pg/m3.  Since this difference is not
statistically significant, nitrate collection is not affected by moderate
variation of sampling  rate.

EFFECT OF SAMPLING  TIME
           The  effect of sampling time has been examined by simultaneous
 collection  of  atmospheric aerosol  samples on parallel  samplers.   During
 the collections, the sample stream is split in half,  with each half pass-
 ing through an identical set of three filter types.  One set of filters
 continuously sampled the atmosphere for a 48-hour period, while the second
                                     25

-------
               TABLE 9.  SAMPLING RATE STUDY (mg/FILTER)
Filter
Material
Gelman AE (1)
" (2)
" (3)'
" (4)
11 (5)
" (6)
11 (7)
" (8)
1 iwo
3
138
380
146
360
153
315
146
345
K
29
82
29
89
42
97
29
124
Total N
70
160
70
170
70
200
70
200
m3
56
142
54
150
54
143
56
137
yg/NQ5
m3
2.46
2.68
2.70
2.40
2.83
2.20
2.61
2.52
set of filters was changed after the first 24 hours of the sampling period.
Comparison of the amount of nitrate collected by the 48-hour filter with
the sum of nitrate collected by the two 24-hour filters will indicate
whether smapling time affects the collection of particulate nitrate.  The
three filter types included quartz tissue (ADL), glass fiber (Gelman AE),
and a compound filter consisting of a quartz (ADL) and a nylon (Duralon)
filter inserted in the same filter holder.  This compound filter was used
to investigate whether particulate and gas-phase nitrate could be separated
and determined simultaneously by a filtration technique.  The quartz filter
has been shown to remove particulate nitrate but not gaseous nitrate, while
the nylon filter quantitatively removes gaseous nitric acid.  Thus the dual
filter might make the simultaneous separation and determination of the two
nitrate types feasible.
          The results of the total mass determinations and the aerosol
nitrogen analyses for these experiments are presented in Table 10.  If
sampling time has no effect on the aerosol collections, then the sum of
the two 24-hour filter collections should approximate the 48-hour filter
sample.  The data in Table 10 indicates that within the estimated experi-
mental uncertainty, the total mass, NH., NOo, and total nitrogen values

                                    26

-------
          TABLE 10.   RESULTS OF THE SAMPLING TIME STUDY
Total Mass,
Filter Type mg
48- hour filter
Sum of 24-hour filters
48- hour filter
Sum of 24-hour filters
48-hour filter
Sum of 24-hour filters
Quartz

Glass Fiber
(Gelman AE)
Quartz and Nylon
23.53
26.44
27.92
35.03
30.63
30.03
NHj,
mg
1.11
1.21
0.74
0.93
1.91
2.11
m~3,
mg
0.56
0.64
1.92
1.44
1.24
1.28
Total N,
mg
0.97
1.30
0.85
0.94
0.94
-------
          Table 11 shows the analyses of the individual quartz and nylon
filters which made up our compound filter.   Again, the sum of the 24-hour
filters compared to the 48-hour filters indicates no dramatic sampling
time effect.  The quartz prefliter results from Table 11 compare quite
well with the single quartz filter results shown in Table 10, serving as
a check of our precision.  The observation of NH. and NOl on the nylon
backup filter indicates that gaseous ammonium and nitrate precursors are
penetrating the quartz filter and are adsorbed by the nylon backup.  Thus,
both the nylon and the glass-fiber filter results suggest that a gaseous
nitrate precursor can strongly influence the apparent particulate nitrate
concentrations.  This is completely consistent with our earlier Phase I
experimental findings.

                   TABLE 11.  COMPOUND FILTER RESULTS
                                                NHj,    NOs,    Total N,
                             Filter Type         mg      mg        mg
48-hour filter Quartz (compound)
Sum of 24- hour filters
48-hour filter Nylon (Backup)
Sum of 24- hour filters
1.11
1.13
0.80
0.98
0.36
0.48
0.88
0.80
0.94
1.08
-
-
EFFECT OF FILTER STORAGE ON PARTICULATE NITRATE

          Filters collected in the field for particulate nitrate determination
must frequently be stored for days, weeks, or even months before the actual
analyses fire performed.  The effect of this storage period on particulate
nitrate is uncertain.  Therefore, a brief investigation of storage-time
effects was added to the Phase I effort.
          The results of several storage-time experiments are shown in
Table 12.  The filters were analyzed, stored for either 2 or 7 months in
                                  28

-------
              TABLE 12.  RESULTS OF THE STORAGE-TIME STUDY
Filter Materials
Gelman AE
Celotate
Quartz (Pallflex QAST)
Duralon (Two 24-hr collections)
Duralon (One 48-hr collection)
N0~-
Before
Storage
0.26
0.047
0.12
0.80
0.88
j
After
Storage
0.31
0.016
0.14
0.68
0.69
Storage
Time, Mos.
2
2
2
7
7
glassine envelopes within sealed plastic bags, and then reanalyzed.   There
does not appear to be any decay of nitrate on Gelman AE or Pallflex quartz
(QAST) during a 2-month storage period.  Loss of nitrate from the Cellotate
filter is indicated; however, the precision of the analysis at such low
levels is not good.  This may account for part of the apparent decay.   A
loss of nitrate was detected on the Duralon filters.  This loss may be
related to the longer storage of the Duralon filter samples.
          The results of this brief study suggest that storage of quartz
and glass (Gelman AE) filters for periods of at least 2 months prior to
analysis should have a minimal effect on particulate nitrate results.
                                    29

-------
                                SUMMARY

          A great deal  has been learned about particulate nitrate sampling
during this investigation, even though some of the individual  studies are
not yet complete.  Our investigation of the interaction between gaseous
nitrogen compounds and filter substrates indicates that nylon  filters,
cellulose acetate filters and many glass fiber filters are subject to signif-
icant particulate nitrate interference due to the formation of artifact
nitrate on the filter.   Teflon, polycarbonate and quartz fiber filters
showed only minimal interferences from gases studied.   Considering other
factors such as cost, handling characteristics, pressure drop, efficiency
for submicron particle collection and mass loading considerations, the
quartz filters appear most appropriate for particulate nitrate sampling,
especially for large sampling networks.
          Studies of the possible interaction between  gases or aerosols
being pulled through a filter and the particles already collected on the
filter indicated no major interferences, either positive or negative, with
particulate nitrate determination.  The interaction between NOo or NH-, and
soot collected on filters was also shown to result in  negligible artifact
nitrate formation under the conditions studied.
          Sampling time and rate were investigated in  this study and were
found to have no effect on particulate nitrate collection over the range
of rates and times studied.  The effect on nitrate determinations of storing
filter samples up to 2 months prior to analysis was found to be negligible
for glass (Gelman AE) and quartz filters.
          The results of this laboratory investigation will ultimately be
combined with data on the comparison of many of the same filter materials
under actual field sampling conditions.  Those data, which will complement
and augment these Phase I results, will be collected and reported in the
third phase of this program.
                                     30

-------
                                REFERENCES
 1.   "Summary Report on Atmospheric Nitrates", U.S. Environmental Protection
     Agency, Research Triangle Park, N.C., July 31, 1974.

 2.   Robinson, E.,  and Robbins, R.C., J. Air Poll. Control Assoc., 20 (5),
     303 (1970).

 3.   Air Quality Data for 1967 from the National Air Surveillance Networks
     (Revised 1971) PB 203 546, National Technical Information Center.

 4.   Murcray, D.G., Kyle, T.G., Murcray, F.H., and Williams, W.J., Nature,
     218, 78 (1968).

 5.   Lazrus, A.L.  and Gandrud, B.W., Proceedings of Third Conference on
     Climatic Impact Assessment Program, p. 161, Feb. 26,  1974.

 6.   Public Health  Service Publ. No. 978, U.S.  Dept. of Health, Education,
     and Welfare,  Washington, D.C.  (1962).

 7.   Gordon, R.J.  and Bryan, R.J.,  Environ. Sci. and Tech., 7. (7), 645
     (1973).

 8.   Hidy,  G.M.,  et al., "Characterization of Aerosols in California",
     Final  Report (Vo. IV), California Air Resources Board, Sept., 1974.

 9.   Spicer, C.W.,  and Miller, D.F., "Nitrogen Balance in Smog Chamber
     Studies", presented at the 67th Annual Meeting of the Air Pollution
     Control Association, Denver, Colorado (1974).

10.   Spicer, C.W.,  "The Fate of Nitrogen Oxides in the Atmosphere", Battelle-
     Columbus Laboratories Final Report to EPA/CRC, 1974.

11.   Miller, D.F.,  Schwartz, W.E.,  Jones, P.E., Joseph, D.W., Spicer, C.W.,
     Riggle, C.J.,  and Levy, A., "Haze Formation:  Its Nature and Origin —
     1973", Battelle-Columbus Laboratories report to EPA and CRC, June, 1973

12.   Heuss, J.M.  and Glasson, W.A., Environmental Sci. Tech., 2_, 1109 (1968).

13.   Pate,  J.B.  and Tabor, E.G., Ind. Hyg. Jour., March-April, 1962, p. 145.

14.   Spicer, C.W.,  Gemma, J., Joseph, D., and Levy, A., "The Fate of Nitrogen
     Oxides in Urban Atmospheres",  presented at the American Chemical Society
     Meeting, Atlantic City, New Jersey, Sept., 1974.
                                     31

-------
15.  Spcicer, C. W., "Non-regulated Photochemical Pollutants Derived from
     Nitrogen Oxides", presented at EPA Symposium on Autombotive Pollutants,
     Washington, D. C., Feb. 12, 1975.

16.  Pierson, W.R., Butler, J.W., and Trayser, D.A., Environ. Letters, _7 (3),
     267 (1974).

17.  Trayser, D.A., Blosser, E.R., Creswick, F.A., and Pierson, W.A., "Sulfuric
     Acid and Nitrate Emissions from Oxidation Catalysts", paper presented at
     SAE Congress and Exposition, Feb. 25, 1975.

18.  O'Brien, R.J., Holmes, J.R., Reynolds, R.J., Remoy, J.W., and Bockian,
     A.M., Paper No. 74-155, presented at 67th APCA Meeting, Denver, Colorado,
     (1974).

19.  Lovelock, J.E. and Penkett, S.A., Nature, 249, 434 (1974).

20.  Chang, S.G. and Novakov, T., "Formation of Pollution Particulate Nitrogen
     Compounds by NO-Soot and NH3-Soot Gas-Particle Surface Reactions", pre-
     print of paper submitted to Atm. Environ. (1974).
                                     32

-------
      APPENDIX A  -





EXPERIMENTAL CONDITIONS
           33

-------
                                                      TABLE A-l.
EXPERIMENTAL CONDITIONS AND SAMPLE VOLUMES
(VOLUMES IN m3)
CO
Experiment No. :
Experimental Conditions:
Filter Materials
Nylon (Dura Ion)
Teflon (Kitex)
Cellulose Acetate (Celotate)
Glass (Gelirun A)
Glass (Geliaan E)
Polycarbonate (Nuclepore)
Quartz (AOL)
Experiment No.:
Experimental Conditions:
Filter Materials
Nylon (Duralon)
Teflon (Mitex)
Cellulose Acetate (Celotate)
Glass (Gel man A)
Glass (Gelman E)
Polycarbonate (Nuclepore)
Quartz (ADL)
1
Clean Air

1.08
0.98
0.95(0.95)
1.12
1.12
1.07
1.12
10
3.0 ppm HNOi
301 RH

0.73
0.98
0.96
1.13
1.13
1.00
1.13
2
Clean Air

1.00
0.93(0.93)
0.89
1.05
1.03
1.00
1.04
11
80 ppm NH3

1.35 •
0.98
1.27
1.39
1.39
1.31
1.41
3
2.6 ppm N02

1.02(1.02)
0.92
0.89
1.02
1.03
1.00
1.05
12
5.5 ppm NH3

. 1.15
1.07(1.05)
1.05
1.18
1.20
1.12
1.20
4
2.0 ppm N02

0.99
0.91
0.89
1.02
1.03
0.97
1.01(1.03)
13
0.3 ppm PAN

0.93
0.82
0.85(0.82)
0.97
0.97
0.92
0.96
5
1.8 ppn N02
40% RH

0.87
0.79
0.68
0.79
0.79(0.79)
0.75
0.78
14
15.6 ppm NgO

1.61
1.52
1.45
1.66
1.67(1.64)
1.61
1.64
6
1.4 ppm HN'03

0.76
C.69
0.68
0.79
0.79(0.79)
0.75
0.78
15
18 ppm N20
•\-100* RH

1.33
1.25
1.17
1.41
1.39
1.31(1.33)
1.41
7
1.5 ppm HN03

0.99
0.89
0.85
1.00(1.01)
1.00
0.96
1.02
16
1.7 ppm NH3
40% RH

1.45
1.34(1.36)
1.29
1.47
1.52
1.40
1.50
8
8.0 ppsi HN03

1.29
1.22
1.17
1.35
1.35
1.31
1.35
17
30 ppn NO?
70S RH

0.99(1.01)
0.84
0.91
1.04
1.03
0.89
1.03
0
3.0 ppT. HNOj

1.32
1.22
1.18
1.34
1.37
1.28
1.35
18
11.5 pp.-; NH3

—
—
1.29
—
2.36
—
2.37

-------
TABLE A-2.   EXPERIMENTAL CONDITIONS AND SAMPLE VOLUMES
Experiment
No.
19




20


21



22


23




24




25




Experimental
Conditions Filter Material
0.024 ppm H2S04 Glass (Gelman AE)
Quartz (ADL)
Cellulose Acetate
Quartz (Pallflex)
Quartz (ADL), clean
3.4 ppm HN03 Quartz (ADL)
Glass (Gelman AE)
Cellulose Acetate
19.5 ppm N02 Glass (Gelman AE)
Cellulose Acetate
Quartz (ADL)
Quartz (Pallflex)
0.35 ppm HNO^ Quartz (Pallflex)
Quartz (ADL)
Glass (Gelman AE)
21 ppm N02 Quartz (Pallflex)
Quartz/Cellulose
Glass (Gelman AA)
Quartz (ADL)
Glass (Spectrograde)
17.5 ppm HN03 Quartz (ADL)
Glass (Gelman AA)
Quartz/Cellulose
Glass (Spectrograde)
Quartz (ADL)
16.5 ppm N0? Glass (Spectrograde)
17% RH Quartz (ADL)
Quartz (Pallflex)
Quartz/Cellulose
Glass (Gelman AA)
Volume
Sampled, m^
2.24
2.29
1.22
2.30
2.25
3.68
3.71
2.24
2.14
1.87
2.10
2.10
4.15
4.18
4.09
0.95
0.98
0.97
0.98
0.96
1.25
1.28
1.28
1.28
1.25
1.26
1.29
1.29
1.30
1.27
                        35

-------
                                   TECHNICAL REPORT DATA
                            (Please read Instructions on the reverse before completing)
 1. REPORT NO.
  EPA-600/2-78-009
                              2.
                                                            3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUBTITLE
SAMPLING  AND ANALYTICAL METHODOLOGY FOR ATMOSPHERIC
NITRATES
[nterim Report.   Evaluation of  Sampling Variables
             5. REPORT DATE
               February 1978
             6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)

Chester W.  Spicer and Philip M.  Schumacher
                                                            8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Sattelle,  Columbus Laboratories
505 King Avenue
:olumbus,  Ohio  43201
              10. PROGRAM ELEMENT NO.

               1AD712 BE-02(FY-77)
              11. CONTRACT/GRANT NO.
               68-02-2213
 12. SPONSORING AGENCY NAME AND ADDRESS
 Environmental  Sciences Research  Laboratory - RTP, NC
 Dffice of  Research and Development
 J.S. Environmental Protection Agency
 Research Triangle Park. N.C.  27711	
              13. TYPE OF REPORT AND PERIOD COVERED
              14. SPONSORING AGENCY CODE
               EPA/600/09
15. SUPPLEMENTARY NOTES
16. ABSTRACT
      The first phase of  a  study to develop  sampling and analytical  methodology for
atmospheric nitrates is  described.  Experiments  were carried out  to determine the
effect of nitrogen-containing 
-------