EPA-600/3-77-028
March 1977
Ecological Research Series
HETEROGENEOUS REACTIONS OF NITROGEN
OXIDES IN SIMULATED ATMOSPHERES
Environmental Sciences Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
-------
EPA-600/3-77-028
March 1977
HETEROGENEOUS REACTIONS OF NITROGEN OXIDES
IN SIMULATED ATMOSPHERES
by
H.S. Judeikis, S. Siegel, T.B. Stewart and H.R. Hedgpeth
The Aerospace Corporation
El Segundo, California 90245
R802687-01
Project Officer
Jack L. Durham
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 of commercial products constitute
endorsement or recommendation for use.
ii
-------
ABSTRACT
A laboratory study has been conducted on heterogeneous reactions of
nitrogen dioxide and nitric oxide to evaluate their potential role in
reactions in polluted urban atmospheres. The results of this study suggest
that nitrogen dioxide decomposes on a wide variety of solids likely to be
found in urban environments. Measured reaction rates indicate these pro-
cesses can be important in the atmosphere. Humidification of reaction
mixtures leads to increased reactivities. It is concluded that hetero-
geneous reactions in the atmosphere are unimportant for the oxidation of
nitric oxide.
This report was submitted in fulfillment of Grant No. R-802687-01 by
the Aerospace Corporation under the partial sponsorship of the U.S.
Environmental Protection Agency. This report covers a period from November
1973 to November 1976, and work was completed as of November 1976.
iii
-------
CONTENTS
Abstract iii
Figures and Tables vi
1. Introduction 1
2. Experimental 3
Materials 3
Reactors 3
Analysis of surface nitrogen compounds 6
3. Results . . . 7
Decomposition of NCL in dry reaction mixtures 7
Decomposition of N0» in moist reaction mixtures 16
Reactivity of NO 21
4. Discussion and Conclusions 22
References 24
-------
FIGURES
Number Page
1 Decomposition of NO- over charcoal 8
2 Decomposition of NO- over MnO- 10
3 Decomposition of NO- over PbO and cement 12
4 Surface nitrogen compounds measured after NO-
decomposition over Al?0, 17
5 Decomposition of NO- over A1-0- 18
6 Decomposition of NO- over selected solids 19
TABLES
1 Solids studied and physical adsorption surface areas 4
2 Decomposition of NO- over charcoal 9
3 Decomposition of NO- over various solids 14
4 Analysis of" surface nitrogen compounds 15
vi
-------
SECTION 1
INTRODUCTION
In recent years, considerable attention has been given to the study of
aerosols in polluted urban atmospheres. This interest has arisen because
of potential health effects, as well as the possibility that aerosols may
play an active role in the physical and chemical processes taking place in
the atmosphere. A well-known example of the latter phenomena is the oxida-
tion of sulfur dioxide in water droplets (1-4).
Previously, through the use of a simple model and data from the catalyst
literature, the conditions were outlined under which selected heterogeneous
reactions involving urban aerosols could compete with gas phase reactions
that are known or are postulated to occur (5). Although the model calculations
suggested heterogeneous reactions could compete with homogeneous gas phase
reactions, the major limitation to these conclusions lay in the fact that
the literature data used were generally obtained from experiments conducted
at considerably higher temperatures and reactant concentrations than those
of interest in polluted atmospheres.
In this paper, results are presented from experiments, conducted under
near-atmospheric conditions, on heterogeneous reactions of oxides of nitrogen.
A wide variety of solids are found to be capable of decomposing NO.. This
is especially true when the reaction mixture is humidified (40-50% relative
humidity). In fact, some materials found to be unreactive in the absence
of added moisture reacted readily when water vapor was added. In the case of
NO, except for reactions over MnO~, only minimal activity was observed,
and it was concluded that heterogeneous reactions of this gas are unimportant
in polluted atmospheres.
Products formed during the heterogeneous decomposition of N0« consisted
primarily of strongly adsorbed surface nitrogen compounds and gaseous NO.
-------
Overall, the heterogeneous processes led to a substantial reduction of NO
X
(i.e., NO + NO™) in the gas phase.
Quantitatively, projection of our laboratory results for N0~ to the
atmosphere leads to the conclusion that NO- decomposition on the surface of
airborne particles in polluted atmospheres can occur with a lifetime of
^0.5 to 5 hours. This compares favorably with observed NO- decay times of
several hours (6,7) and suggests the potential importance of heterogeneous
processes for the removal of N0~. Moreover, although our results indicate
only minimal heterogeneous activity toward NO, the latter species is rapidly
converted to NO- in polluted atmospheres (7,8). Consequently, heterogeneous
processes could also be important in the removal of NO emitted as NO.
X
-------
SECTION 2
EXPERIMENTAL
MATERIALS
Solids used in the present study are listed in Table 1, along with
their BET surface areas. These solids were supported on various pyrex
cylinders or helices ranging from ^1 cm in diameter x 10 cm long to 3.2 cm
in diameter x 40 cm long. The helices were constructed from 3-mm pyrex rod
with ^3 mm between loops. The appropriate cylinder or helix was prepared
by spraying a fluid suspension of the desired solid (water, acetone, alcohol,
or mixtures of these were used as the fluid carrier) onto the cylinder or
helix and allowing the wet film to dry in air before use. Typically, the
net weight of the dry film was several hundred milligrams. The coated
cylinder or helix was inserted concentrically into one of two cylindrical
-2 1
flow reactors and further dried under vacuum (<10 Torr ) for 1 to 3 hours
or in some cases overnight. (In the case of H^SO,, a cylinder was coated
with the concentrated acid, inserted directly into the reactor, and used
immediately without extensive evacuation.)
Gases used in these studies were obtained from Matheson. Gases were
used as received, except for NO-N. or NO-Ar mixtures. The latter gases
were passed through a molecular sieve column before use to decompose any
residual NO,,. Various gas mixtures were prepared or humidified through the
use of procedures described elsewhere (9,10).
REACTORS
Two different cylindrical flow reactors were used in these studies
(10,11). The reactors differed principally in their modes of reactant and
product analysis. The first reactor (11), designated reactor F (fluo-
rescence detection), was used for detailed rate studies of the heterogeneous
1 2
Torr = 1.33 22 E + 02 Newton/meter
-------
TABLE 1. SOLIDS STUDIED AND PHYSICAL ADSORPTION SURFACE AREAS*
Material
Source
Area
Wood charcoal
powder
Cement
PbO
Mn02
Fly ash
ZnO
Sand1"
V2°5
Analabs ,
activated alumina
no. 214-77
J.T.
r,
T. Bakei
no. 0537
Allied Chemical,
no. 1567
California Portland
Cement
J.T. Baker,
no. 2338 yellow powder
ROC/RIC,
MN-37
196
233 ± 7*
40.7
17.9
20.1
109 ± 10*
Western Coal Burning 15.2
Power Plant
New Jersey Zinc,
SP500
ROC/RIC,
CU-33
red 99%
Commercial
jeweler's rouge
Baker and Adamson,
no. 2171
ROC/RIC V-15
22.2
5.5 ± 3.0
27.3
7.3
14.1
*BET measurements using N« at -196 C, except for the case of MnO. where
C0_ was used at -78C. The following materials were also used, but their
BET areas were not measured: (NH,)?SO,, Al (powdered metal), CaO,
PbCl2, crushed oleander leaves, H2SO,, and soil (loam).
tCrushed to a fine powder before use. Other materials used as received.
tErrors derived from estimates of error in reading pressure devices,
i.e., Matheson 0-760 and voltmeter used with pressure transducer plus
an 0.01 V/min drift noted.
-------
decomposition of N0~. Generally, NCL-Ar mixtures were used.in this reactor.
An argon ion laser (488 nm) excited NCL fluorescence along the cylinder axis.
This fluorescence was detected at right angles to the laser beam by a photo-
multiplier equipped with narrow slits. A narrow uncoated strip running the
length of the cylinder on which the solid was supported permitted the fluo-
rescence to reach the photomultiplier tube. The coated cylinder (with the
uncoated strip) was mounted on a platform inside the reaction chamber. This
platform was constructed such that the cylinder could be moved back and forth
along the direction of the cylinder axis (11). This feature, along with the
narrow laser excitation beam and the narrow slits on the photomultiplier,
permitted the determination of N0_ concentration gradients along the cylinder
axis. The latter data, in conjunction with a mathematical analysis of the
system (11, 12), were used to obtain the gas-solid reaction rate constants.
In the second reactor (10), designated reactor A (absorption detection),
NO- was monitored by its optical absorption (480 nm) along the cylinder axis.
A chopped dual beam optical scheme was used for this purpose. Unlike reactor
F, the coated cylinder (or helix, which was also used in this reactor) was
not movable, and spatial discrimination of N0? along the cylinder axis was not
obtained. However, other reactants or products (i.e., NO and 0_) and inert
gases (i.e., N_ and Ar) could be analyzed before entry into or after exit
from the cylinder. Reactant or product NO was determined through a chemi-
luminscent reaction with 0~, while gas chromatography was used for the other
gases (10).
The effects of moisture and simulated solar ultraviolet and visible
radiation could be determined in reactor A. Reaction mixtures were humidified
through the use of techniques similar to those described by Cheng et al. (9)
and Hedgpeth et al. (10). A bank of daylight-type fluorescent lights
surrounding the (pyrex) reactor was used to simulate solar radiation. In the
latter experiments, the coated helix was used, since this configuration allowed
for a more even light distribution inside the chamber. Additional details on
this system are given by Hedgpeth et al. (10), while the procedures used for
data analysis are described by Siegel et al. (12).
-------
ANALYSIS OF SURFACE NITROGEN COMPOUNDS
In several experiments, the coated cylinders were removed from the reaction
chamber upon completion of the experiment, washed with distilled water, and
filtered, and the filtrate analyzed for N0~ (i.e., N0~ or N0~). A wet
X fc -J
chemical procedure or nitrate ion electrode (Beckman, model no. 39618) was
used for NO- analysis (12-13), while wet chemical techniques were employed for
N0? analysis (13). The latter test gives a positive result for both NO- and
gaseous N0? dissolved in solution. Consequently, the NO- test cannot dis-
tinguish between adsorbed NO- and NO-. An additional uncertainty in the
analyses of NO lies in the possibility 'that NO could be formed from
X X
adsorbed N0_ during the washing procedures.
In some cases, these experiments were run for short periods of time to
avoid poisoning heterogeneous activity. In these experiments, the solid was
removed in segments, washed, and filtered, and the individual filtrates
analyzed for NO .
-------
SECTION 3
RESULTS
DECOMPOSITION OF N02 IN DRY REACTION MIXTURES
Reactivities of Charcoal, A^O-^, and
A series of experiments was conducted through the use of reactor F and
N0_-Ar mixtures to determine quantitatively the heterogeneous decomposition
rates of NO- over charcoal, A1-0-, and MnO-. Earlier experiments in reactor
A had demonstrated that this reaction readily occurs in N--0--NO-NO- mixtures
yielding surface nitrates, nitrites, and gaseous NO. Gas mixtures consisted
of 0.05 to 5% NO- in Ar, with total pressures of 1 to 100 Torr. The NO-
concentration, detected by laser-excited fluorescence, was measured as a
function of distance along the axis of the cylindrical reactor.
Data from these experiments were analyzed through the use of model
described by Stewart and Judeikis (11) and Siegel et al. (12). In this model,
the rate of change of the NO- concentration at the walls is written in terms
of duf fusion of gaseous NO- to and from the walls of the coated cylinder,
the gaseous flow rate through the cylinder, and the ratio k /k , where k is
the rate constant for removal of NO- from the gas phase and k is the gas-
solid collision rate constant. (Thus, k /k is the fraction of gas-solid
collisions leading to removal of NO- from the gas phase.) Integration of
the resulting differential equation yields a Bessel series solution for the
spatial distribution of gaseous NO- in the cylinder. This series reduces to
a single exponential term for large distances along the cylinder axis.
Results from a typical experiment with charcoal are illustrated in
Figure 1. The data shown by the solid curve in the figure were calculated
through the use of experimental parameters and a value of 0.0016 for k /k .
An example of the exponential decay with distance at large distances is
illustrated in Figure 2 for data from an experiment with MnO-. The solid
-------
0.02
O MEASURED
CALCULATED
0.00
1 2 3
DISTANCE ALONG CYLINDER AXIS (cm)
Figure 1. Decomposition of NOo over charcoal. Experimental
conditions: 0.2% N02 in Ar (10 Torr total pressure),
23 - 1 C, 1.7 cm^/sec flow, and a cylinder radius
of 0.95 cm.
-------
curve in Figure 2 represents the limiting slope calculated for large distances
through the use of the experimental parameters, a value of k /k of 0.0003,
and the model described above.
Data from experiments in reactor F were analyzed through the use of the
method indicated in Figure 2; i.e., the exponential decay with distance was
fitted through the use of limiting slopes calculated from the model described
by Stewart and Judeikis (11) and Siegel et al. (12). In general, slopes
differing by about ±25% from the best fitting slope would give noticeable
deviations from the experimental limiting slope. Examples of experimental
reproducibility for NO- decomposition over charcoal as a function of cylinder
radius, total pressure, and flow rate are given in Table 2. The average
_3
value of k /k for charcoal in these experiments was found to be 1.6 x 10
r C -3
(standard deviation = ±0.5 x 10 ). Results from similar experiments with
MnO» exhibited somewhat greater variations, and it is concluded that k /k =
-1 re
0.3 to 3 x 10 , whereas for Al-0- (Analabs) it is found that k /k = (3 ± 1)
i £ J L \~
x 10" .
TABLE 2. DECOMPOSITION OF NO- OVER CHARCOAL*
Cylinder
Radius (cm)
0.50
0.95
Ptotal
(Torr)
1.0
9.9
10.2
1.0
1.0
9.8
10.3
Flow Rate
/ 3, ,
(cm /sec)
17.5
25.0
16.5
0.4
0.4
1.7
1.7
3 kr
c
2.6
1.2
1.4
1.5
1.0
1.6
1.6
^Reaction mixture: 0.2% N02 in A (23 ± 1 C).
-------
O.OZOo
0.010
O
CM
o
Q.
0.004
0.002
0.001
o
MEASURED
CALCULATED
1234
DISTANCE ALONG CYLINDER AXIS (cm)
Figure 2. Decomposition of NC>2 over MnC^. Experimental
conditions: 0.2% N02 in Ar (10 Torr total pressure),
23 i 1 C, 1.0 cm^/sec flow, and a cylinder radius
of 0.95 cm.
10
-------
Additional Experiments in Reactor A
An additional sequence of experiments with dry NO_-N_ or N0_-Ar mixtures
was carried out in reactor A. The purpose of these experiments was to
screen other solids for reactivity toward NCL, analyze gaseous and adsorbed
reaction products, measure overall capacities for reaction, and examine the
effects of simulated solar radiation. Because of pressure and flow rate
limitations, however, only minimum reactivities could be determined for solids
that efficiently removed NCL. This was because, for k /k > 10 , reactions
were limited by diffusion of gaseous N0« to the cylinder walls in this system.
In addition to those materials discussed in the preceding section, the
following solids were found to exhibit substantial reactivity (k /k > 10 )
toward NCL decomposition: Al.,0,, (Baker), cement, PbO, fly ash, and CaO.
The remaining materials in Table 1 exhibited only minimal activity (k /k < 3
-7 re
x 10 ) or no activity within experimental error toward dry NCL reaction
mixtures.
Results from representative experiments in reactor A are illustrated
in Figure 3. These results are from experiments in which glass helices
(3.2 cm in diameter x 40 cm long) were used to support PbO and cement. In
addition, a 1-mil sheet of DuPont Kapton was placed between the fluorescent
lamps and reactor to filter wavelengths responsible for direct photolysis
of N02 (10).
Experimentally, dry nitrogen was passed through the chamber at the
desired flow and pressure until the gas flow stabilized. Then valves were
rapidly switched to admit NO~-N~ mixtures at the same pressures and flows.
This procedure avoided pressure surges in the system and required only a
few minutes for the NO^-N,, flow to stabilize (see blank run in Figure 3) .
In the case of PbO and cement, the initial nonzero values for the
average N0~ pressured after stabilization are due to the reaction mixtures
filling the space between the gas inlet and leading edge of the helices, as
well as partially penetrating into the helices before N09 diffuses to the
helix walls and reacts. Over long periods of time (^ 400 min in the experi-
ments illustrated), the N0~ concentrations slowly increased as the activities
of the solids gradually diminished, and ultimately (t > 400 min) reached the
11
-------
CEMENT
PbO
Figure 3. Decomposition of N02 over PbO and cement. Experi-
mental conditions: 0.2% N02 in N2 [total pressures
of 191 Torr (PbO) and 200 Torr (cement)], 23 + 1 C,
and flow rates of 9.5 cm3/sec (PbO) and 11.6 cm3/sec
(cement). The remaining curve in the figure is for
a blank experiment conducted in the absence of added
solids with the same experimental conditions as for
PbO. Arrows indicate where fluorescent lamps
equivalent suns) were turned on ( + ) or off (-0-
12
-------
levels obtained in the absence of added solids when the activities of the PbO
and cement were completely destroyed (not shown) .
From the flow rates, weights of solids, and time to poisoning, the
capacities of the solids to destroy NO- can be calculated. The results for
the experiments discussed above, as well as experiments in which other solids
were investigated are given in the first column of Table 3. Once poisoned,
exposure of the solid to vacuum for periods up to several hours did not
restore any substantial activity. Also, introduction of 0? during or after
the experimental had little effect on the results, which indicated poisoning
was not due to depletion of surface oxygen.
Exposure of samples to visible light during experiments led to brief
decreases or increases in N0_, which suggested a slight restoration of solid
activity in the former case, or photodesorption of adsorbed NO- in the latter.
However, these changes were short-lived (Figure 3) and had little effect on
the overall activity or capacity of the solid. Exposure to visible light
after the solid had been poisoned frequently led to brief, large increases
in N02, probably because of photodesorption. Here also, the overall effects
on reactivity and capacity were small.
Reaction Products
In selected experiments, solids were removed from the reactor upon com-
pletion of the experiment, washed with distilled water, and filtered, and
the filtrate analyzed for NO (see Section 2 for discussion of uncertainties) .
X
Results from these experiments are given in Table 4. The results indicate
that both NO- and NO,, are formed, the dominant specie depending on the
solid used. The identity of the NO formed appeared to be independent of the
X
length of time the experiment was run.
In the case of Al?0_, the Analabs material gave predominantly N02, while
the Baker Al-O,. gave mostly N0_. Elemental and x-ray analysis of these two
solids indicated slight differences. Thus, elemental analysis gave higher
percentages of Si, Fe, and Na in the Analabs material (0.1 - 0.3% versus
0.002 - 0.08%). In addition, although the dominant x-ray patterns from both
materials were similar (boehmite) , the Analabs material gave several extra
(unidentified) peaks not observed for the Baker AlO-.
13
-------
TABLE 3. DECOMPOSITION OF N02 OVER VARIOUS SOLIDS
*t
PbO
Cement
A1203§
Charcoal
Mn02
Dry Reaction Mixtures
Capacity N0~ Formed
0.05
0.05
0.07 ± 0.02 0.03 ± 0.01
0.12 ± 0.03 0.03 ± 0.01
0.04
Humidified
Capacity
>0.23
>0.37
0.21 ± 0.
0.24 + 0.
>0.41
Reaction Mixtures1"1"
NO" Formed
X
-
-
01 0.11 ± 0.01
09 0.09 ± 0.05
tt
§
Capacities refer to the grams of N0~ decomposed per gram of solid, while
NO formed refers to the grams of N09 converted to surface NO per gram
X i. X
of solid.
Uncertainties are standard deviations from 3 to 5 experiments. Results
expressed without uncertainties are from single experiments (see Figure 6)
44% relative humidity
Baker.
14
-------
TABLE 4. ANALYSIS OF SURFACE NITROGEN COMPOUNDS
Surface Nitrogen Compounds
Solid
*
A12°3
Al203f
Charcoal
Mn02
PbO
NO"
3
Small
Large
Large
Large
Small
N0~
2
Large
Small
Small
Small
Large
Total as N0_ (mg)
N0~ N00 Lost
x 2
24 24
19 14
0.8 1.0
-
-
-
-
t
Analabs.
Baker.
15
-------
Experiments were also conducted to quantitatively determine NO . Experi-
X
ments run for short periods of time to avoid poisoning indicated a quantitative
conversion of NO- to NO . This is illustrated for Al-0- in the last two
£• X <- O
columns of Table 4. The NO- lost was determined from the gas pressures, flow
rates, and duration of the experiment. In several of these experiments,
adsorbed NO was analyzed as a function of distance along the cylinder.
X
Results from a typical experiment are shown in Figure 4. After the experi-
ment, the Al-0- was removed in sections, and the NO was analyzed in each
^ j X
section. The model described by Stewart and Judeikis (11) and Siegel et al.
(12) was used to determine the calculated concentration profile, which was
normalized through the use of the gas flow data and under the assumption that
-4
k /k > 10 , which is the diffusion limit in this experiment. Thus, the
calculated and experimental results may be quantitatively compared. The
major discrepancy, which occurs near z = 0, is probably due to edge effects.
Additional experiments were conducted for longer periods of time to
investigate changes that might occur as the solid activity poisons. Results
from these experiments are given in the third column of Table 3 and indicate
that the NO- -»• NO conversion bdcomes less than quantitative. In order to
establish a material balance for the N0_ lost, several experiments were con-
ducted with Al_0- (Baker) and MriO_ to determine if gaseous products were
formed. Results from one such experiment for NO™ decomposition over Al-0- are
shown in Figure 5. The production of gaseous NO (measured in the exit
stream) is apparent. Integration of the NO produced and NO- lost indicates
a 62% conversion of N0_ to NO. The remaining 38% can be accounted for, to
within experimental error, by NO formed on the Al-0- surface. Similar
X £m -J
results were obtained for MnO- with a 71% NO- -»• NO conversion. In the case
of MnO-, a distinct induction period was observed for NO formation even
though NO- was destroyed. The latter observation may be due in part to
removal of product NO by MnO- (see Section 3—Reactivity of NO).
DECOMPOSITION OF NO- IN MOIST REACTION MIXTURES
The effects of moisture on the decomposition of NO- were quite dramatic
and are illustrated in Figure 6. . In the experiments illustrated, only
the first several centimeters of the cylinder were coated with the indicated
solids. Dry reaction mixtures were passed through the chamber until
16
-------
12.5
§ 10.0
OJ
V)
D
I X
O
N
7.5
5.0
2.5
*J19.6
II EXPERIMENTAL
• CALCULATED
Hfl
2.5 5.0 7.5
DISTANCE ALONG CYLINDER, z (cm)
10.0
Figure 4. Surface nitrogen compounds measured after NC^
decomposition over Al^C^ (Baker). Experimental
conditions: 0.2% NC>2 in N2 (700 Torr total
pressure), 23 + 1 C, flow rate = 10 cm3/sec, and
cylinder radius = 2.0 cm.
17
-------
N02 AVERAGE
Figure 5. Decomposition of N02 over
(Baker).
Experimental conditions: 0.05% N02 in N2
(180 Torr total pressure), 23 + 1 C, flow
rate = 9.2 cm-Vsec, and cylinder radius =
2.0 cm.
18
-------
0.6
I ' I ^ I r
-44% RELATIVE HUMIDITY ADDED
o
O
z
Ou
UJ
O
U
0.4
0.2
A CEMENT
OPbO
DMnO,,
0.0
100
200
t (min)
300
400
Figure 6. Decomposition of NC>2 over selected solids.
Effects of added moisture. Experimental
conditions: 0.2% N02 in Ar (^200-300 Torr
total pressures), 23 t 1 C, flow rates
* 10 cm-Vsec, and cylinder radius = 2.0 cm.
19
-------
reactivity vas destroyed. At that point moisture was added (44% relative
humidity), and the activity was restored. Separate experiments in the absence
of added solids indicated gas phase reactions between NCL and H~0 were
unimportant below 50% relative humidity. From Figure 6, it will be noted that
the total activity (before poisoning) is substantially increased when moisture
is added. Similar results were obtained when moisture was present initially
in the reaction mixture. Qunatitative values for capacities in moist systems
are given in the last two columns of Table 3. Diffusion limited reaction
rates were obtained in the presence of moisture (k /k > 10 ) for the first
seven materials listed in Table 1 and for CaO, all of which were reactive in
dry systems.
An equally significant result was that materials found to be unreactive
in the absence of moisture reacted readily when moisture was added (44%
relative humidity). The following values were measured for k /k in these
-4 -5 -5 r C
experiments: > 10 (ZnO), 3 x 10 (Cu-O, Fe-0-), 1 x 10 (sand), and
—ft
1 x 10 fV2°5» (NH4^2S04' H2SOJ• The remaining materials in Table 1 were
not examined in these experiments.
Glass filter paper of the type used in high-volume samplers for collecting
airborne particles was also examined. While this material was unreactive
with dry N09-N? mixtures, the addition of moisture led to extensive decom-
/
position of N0« (k /k > 10 ). This could have serious consequences in
the interpretation of particulate data if surface NO compounds are formed.
X
Some discrepancies between particles collected on various types of filters
have been noted (14).
In one case (Fe.O.,), reactivity was examined as a function of relative
-7
humidity (rh). The results obtained for k /k were: < 10 (0-7% rh),
^5 x 10~6 (14% rh), VL.6 x 10"5 (28% rh), and ^3 x lo"5 (44% rh). Attempts
to conduct experiments at higher relative humidities were unsuccessful
because of gas phase (or wall) reactions involving NO- and H-O.
The increased reactivity and capacity in the presence of moisture led
to the exploration of the possibility that surface nitrogen compounds formed
in the presence of water might volatilize as the corresponding acids. In
experiments with MnO_ and Al?0_ (Baker), effluent gas from the reactor was
20
-------
bubbled through a water solution, and the solution analyzed at intervals for
NO- and pH. (Under the experimental conditions, NO- was completely destroyed
in transit through the reactor being converted, in part, to gaseous NO.) In
the case of MnO-, NO- was observed even in dry reaction mixtures and increased
when moisture was added. (Separate experiments bubbling NO through the water
gave a negative NO- test). In the presence of moisture, the NO- analysis and
pH indicate that ^2 to 3% of the NO,., lost is ultimately converted to volatile
HNO- in the effluent stream. For A1-0-, no NO- was observed in the effluent
stream.
REACTIVITY OF NO
A number of experiments was conducted in reactor A with NO-0--N- mixtures
to determine the reactivity of NO toward various solids. The conditions of
these experiments were such that ^10 to 20% of the NO was converted to NO,,
during passage through the reactor by the termolecular gas phase reaction
between NO and 0- (7-8). Several of the materials listed in Table 1, notably
V-O,. and ZnO, appeared to exhibit some activity toward oxidation (photo-
oxidation in the case of ZnO) of NO to N0_, as evidenced by increased N09
-7
yields. However, k /k & 10 for these cases. In the case of MnO-,
charcoal, and A1-0-, NO- yields were sharply decreased. Separate experiments
with NO-N? mixtures indicated that, except for MnO_, this decrease was due
to N0--solid interactions and not NO-solid interactions. For MnO_, both NO
-4
and NO- were reactive with k /k > 10 . The remaining materials listed in
Table 1 exhibited no reactivity toward NO to within experimental error and
k /k < 10~7.
r c
Additional experiments were carried out with humidified (44% relative
humidity) NO-N- mixtures to determine the effects of moisture on reactivities.
The materials examined were A1-0-, charcoal, ZnO, and V-O^. Unlike the case
of NO-, moisture had no effect on the NO-solid interactions to within the
^ _7
limits of experimental detectability (k /k < 10 ).
21
-------
SECTION 4
DISCUSSION AND CONCLUSIONS
The results for N07 suggest that this pollutant could be heterogeneously
destroyed in the atmosphere on a wide variety of airborne solids likely to be
found in urban environments. The heterogeneous decomposition is greatly
facilitated by the presence of moisture. In fact, many solids that were
unreactive in the absence of moisture readily decomposed NO. when water vapor
was added. The striking effects of water suggest the involvement of surface
OH groups in the heterogeneous processes.
Initially, the decomposition reactions quantitatively convert NO- to
adsorbed NO . With time, the conversion becomes less than quantitative,
X
and gaseous NO is formed. Overall, ^20 to 30% of the NO- lost is converted
to adsorbed NO . Although volatilized HNO., was observed in one case (MnO~),
X
-------
•'7
the reaction continues. The latter species is rapidly converted to N0_ in
urban atmospheres during daylight hours (8).
The projected particle limited lifetimes compare favorably with observed
diurnal variations in NO (6-7). They are longer than the noontime photolytic
X
lifetime of 3 min for NO- in the atmosphere (8); however, the heterogeneous
processes lead to a net reduction in oxidant (NO- + 0-), as well as NO ,
Z. j X
while the photolytic process does not. Moreover, since the photolytic process
is not operable during the night, heterogeneous reactions could be the
dominant nighttime sink for NO-.
Only airborne particulate matter has been considered in our modeling.
Consideration of ground-level surfaces could greatly enhance total reactivity.
Moreover, reactions taking place on the latter surfaces could lead to con-
centration gradients near the ground, depending upon the amount of transport
and mixing taking place. These gradients could be significant in accurately
assessing pollution-related health hazards.
Finally, except for MnO-, NO exhibits only minimal activity toward
solids examined in this study. Based on quantitative rate measurements, it
is concluded that heterogeneous atmospheric processes of NO are probably
unimportant. However, as noted earlier, NO is rapidly converted to NO- in
urban atmospheres (8), and heterogeneous removal of the latter product could
act as the ultimate sink for NO emitted as NO.
x
23
-------
y ti
REFERENCES
1. Johnstone, H.F., and D.R. Coughanowr. Adsorption of Sulfur Dioxide from
Air. Ind. Eng. Chem., 50:1169-1172, 1958.
2. Urone, P., H. Lutsep, C.M. Noyes, and J.F. Parcher. Static Studies of
Sulfur Dioxide Reactions In Air. Environ. Sci. Technol., 2:611-618, 1968.
3. Matteson, M.J., W. StSber, and H. Luther. Kinetics of the Oxidation of
Sulfur Dioxide by Aerosols of Manganese Sulfate. Ind. Eng. Chem.
Fundamentals, 8:677-687, 1969.
4. Carabine, M.D. Interactions in the Atmosphere of Droplets and Gases.
Chem. Soc. Rev., 1:411-432, 1972.
5. Judeikis, H.S., and S. Siegel. Particle-Catalyzed Oxidation of Atmospheric
Pollutants. Atmospheric Environment, 7:619-631, 1973.
6. Pitts, J.N. Environmental Appraisal: Oxidants, Hydrocarbons, and
Oxides of Nitrogen. J. Air Pollut. Control Ass., 19:658-667, 1969.
7. Altshuller, A.P., and J. Bufalini. Photochemical Aspects of Air Pollution:
A Review. Photochem. Photobiol., 4:97-146, 1965.
8. Shuck, E.A., and E.R. Stephens. Oxides of Nitrogen. In: Advances in
Environmental Sciences, Vol.1, J.N. Pitts and R.L. Metcalf, eds. Wiley-
Interscience, New York, New York, 1969. pp. 73-118.
9. Cheng, R.T., J.O. Frohliger, and M. Corn. Aerosol Stabilization for
Laboratory Studies of Aerosol-Gas Interactions. J. Air Pollut. Control
Ass., 21:138-142, 1971.
10. Hedgpeth, H., S. Siegel, T. Stewart, and H.S. Judeikis. Cylindrical
Flow Reactor for the Study of Heterogeneous Reactions of Possible
Importance in Polluted Atmospheres. Rev. Sci. Instrum., 45(3): 344-347,
1972.
11. Stewart, T.B., and H.S. Judeikis. Measurements of Spatial Reactant and
Product Concentrations in a Flow Reactor Using Laser-Induced Fluorescence.
Rev. Sci. Instrum., 45(12):1542-1545, 1974.
12. Siegel, S., H.S. Judeikis, and C. Badcock. The Role of Solid-Gas
Interactions in Air Pollution. EPA-650/3-74-007, U.S. Environmental
Protection Agency, Washington, D.C., 1974.
24
-------
13. Feigl, F. Spot Test in Inorganic Analysis. Ed. 5. El Sevier Publishing
Co., New York, New York, 1958. pp. 326-332.
14. Muller, P.K., S. Twiss, and G. Sanders. Selection of Filter Media: An
Annotated Outline. Paper presented before 13th Conference on Methods
Air Pollution and Industril Hygiene Studies, Berkeley, California,
October 30-31, 1972.
15. Foster, P.M. The Oxidation of Sulphur Dioxide in Power Station Plumes.
Atmospheric Environment, 3:157-175, 1969.
25
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/3-77-028
2.
3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUBTITLE
HETEROGENEOUS REACTIONS OF NITROGEN OXIDES IN
SIMULATED ATMOSPHERES
5. REPORT DATE
March 1977
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
H.S. Judeikis, S. Siegel, T.B. Stewart and
H.R. Hedgpeth
8. PERFORMING ORGANIZATION REPORT NO.
ATR-75(7441)-2
9. PERFORMING ORGANIZATION NAME AND ADDRESS
The Aerospace Corporation
P.O. Box 92957
Los Angeles, California 90245
10. PROGRAM ELEMENT NO.
1AA603 (1AA008)
11. CONTRACT/GRANT NO.
Grant No. 802687
12. SPONSORING AGENCY NAME AND ADDRESS
Environmental Sciences Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
13. TYPE OF REPORT AND PERIOD COVERED
Interim 11/73-11/76
14. SPONSORING AGENCY CODE
EPA/600/09
15. SUPPLEMENTARY NOTES
16. ABSTRACT
A laboratory study has been conducted on heterogeneous reactions o£ nitrogen
dioxide and nitric oxide to evaluate their potential role in reaction in polluted
urban atmosphere. The results of this study suggest that nitrogen dioxide
decomposes on a wide variety of solids likely to be found in urban environments.
Measured reaction rates indicate these processes can be important in the atmosphere.
Humidification of reaction mixtures leads to increased reactivities. It is
concluded that heterogeneous reactions in the atmosphere are unimportant for the
oxidation of nitric oxide.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS C. COSATI Field/Group
*Air pollution
*Nitrogen oxides
Aerosols
*Environmental simulation
13B
07B
07D
14B
18. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19. SECURITY CLASS (This Report)
UNCLASSIFIED
21. NO. OF PAGES
32
20. SECURITY CLASS (This page)
UNCLASSIFIED
22. PRICE
EPA Form 2220-1 (9-73)
26
------- |