United States
Environmental Protection
Agency
Atmospheric Sciences Research
Laboratory
Research Triangle Park NC 27711
Research and Development
EPA/600/S3-85/029 June 1985
&EPA Project Summary
Outdoor Smog Chamber
Experiments to Test
Photochemical Models:
Phase II
H. E. Jeffries, K. G. Sexton, R. M. Kamens, and M. S. Holleman
The smog chamber facility of the
University of North Carolina at Chapel
Hill (UNC) was used in a study to provide
experimental data for developing and
testing kinetic mechanisms of photo-
chemical smog formation. The smog
chamber, located outdoors in rural
North Carolina, is an A-frame structure
supporting Teflon film. Because the
chamber is partitioned into two sec-
tions, each with a volume of 156 m3,
two experiments can be conducted
simultaneously. The dual chamber is
operated under natural conditions of
solar radiation, temperature, and rela-
tive humidity.
In this study, 128 dual-experiments
were performed using NO« and various
hydrocarbons and hydrocarbon mix-
tures. The experiments performed dur-
ing this three-year project have been
added to the existing UNC database for
model validation testing, bringing the
total number of dual-experiments to
346.
This report presents an organizational
scheme for these 346 experiments. Six
attributes of each experiment were used
for classification. These are class of
experiment, experimental conditions,
quality, processing status, project that
produced the data, and membership in a
run series. Run series, a collection of
experiments performed in a certain
manner or addressing a particular ques-
tion, formed a major organizational
basis. Forty series were used, including
seven types of side-by-side series, four
types of chemical issues series, and five
types of characterization series. All dual
experiments, or runs, were classified
into one of the side-by-side series. Each
run could also appear in up to two addi-
tional series.
The report also discusses three ex-
amples of selecting runs to test mecha-
nisms. The examples are: runs to test an
explicit toluene chemistry mechanism;
runs to test mechanisms under dynamic
operating conditions; and runs to test
EKMA-type mechanisms for air quality
modeling. Recommendations for how
best to take advantage of the side-by-
side nature of the UNC runs in testing
mechanisms are given throughout the
discussion.
This Project Summary was developed
by EPA's Atmospheric Sciences Re-
search Laboratory, Research Triangle
Park, NC, to announce key findings of
the research project that is fully docu-
mented in a separate report of the same
title (see Project Report ordering infor-
mation at back).
Introduction
Photochemical kinetics reaction mod-
els are central components in EPA's
method for computing control require-
ments for organic emissions needed to
meet the National Air Quality Standard
for ozone. State and local control officials
are expected to use these models to esti-
mate control requirements. However, dif-
ferent kinetic models apply different
methods, often resulting in different des-
criptions of the same situation. Smog
-------
cha mber data must be used to test photo-
chemical mechanisms to determine the
adequacy of the chemical representation.
Presently, mechanism testing has not
used a large enough range of test data
encompassing the range of conditions
likely to occur in an urban control situa-
tion. This is because, in part, an organized
database containing the range of condi-
tions needed was not available before
now.
This report uses a hierarchy of chemi-
cal species, based on the number of
hydrocarbon and oxides of nitrogen
(HC/NOx) systems in which the species
occur, to determine the type of smog
chamber experiments needed for ideal
mechanism validation. This species hier-
archy included a hierarchy of operating
conditions. These four sets of conditions,
which include large dilution and continu-
ous injection, are ordered to proceed from
the standard smog chamber type experi-
ment to conditions that simulate the
urban environment. All experiments took
place under natural sunlight and ambient
temperature. An example of this hier-
archy of experimental conditions is shown
in Figure 1, in which a similar photochem-
ical system was performed under four dif-
ferent physical conditions. Two of these
experiments also show the unique side-
to-side nature of the UNC chambers.
Three other projects using the UNC
Outdoor Smog Chamber have contributed
data for model testing. The types of data
provided by these projects are described'
briefly in this report, and the project iden-
tifications are used in the classification
scheme.
Purpose
The purpose of this research project
was to: (1) measure react ant and product
concentrations as functions of time in
selected photochemical smog systems;
(2) process these data into a form readily
distributable to model developers and
testers; and (3) provide auxiliary informa-
tion needed to create mathematical model
descriptions of the systems. The 128
dual-experiments, when added to the
other experiments in the database, en-
compass a range of complexity from
single HC species in totally static situa-
tions to complex urban mixtures in fully
dynamic conditions. As a whole, the data
set is well suited to develop and test
models of chemical transformations in
polluted atmospheres.
Organizing Principles
The 346 runs were classified by "keys"
to allow easy selection. Six kinds of
attributes were used to organize the runs
in the data set. The C/ass of a run was
either a characterization run that ad-
dressed unique aspects of the UNC
chamber or organic/NOx runs to test
mechanism design. Further, either of
these types could be one-day or multi-day
runs. Experimental Conditions were
based on four major types of run condi-
tions. These conditions were: dilution
, (large or normal),,injection (initial or con-
tinuousK and type of HC system used, i.e.,
single HC species or mixture, and, if a
mixture, whether the composition was
constant or varying over time. Initial Con-
ditions included the exact NOX and HC
concentrations that occurred on both
sides, as well as the identity of the HC for
up to three HC species. For runs that had
more than three species, HC mixture
names were used. The Quality of a run
was rated on a scale of 0 to 9. The major
determining factors for quality were the
weather conditions and the number and
performance of the analytical instru-
ments. The Processing Status of a run
indicated the availability of validated data
on tape or in other forms such as plots. All
runs were included in the database, but
not all have yet been completely pro-
cessed to the final validated form. The
Project Name identified the runs by the
project that produced the experiment.
Data Set
The report describes how the runs
were distributed across the organization
keys set forth. The majority of the experi-
ments were one day, normal dilution,
initial injection runs performed between
June and October. The experiments were
summarized by HC type. All the species
and named mixtures used are listed. The
grading of experiments is explained:
important factors are sunlight, initial
conditions, analytical support, chamber
conditions, and need/usefulness. Most
runs were graded at 7 or 8.
All experiments in the data base are
listed two ways: sorted by series and by
species. The seven types of series lists
are characterization, matched conditions,
relative reactivity between sides, carbon
substitution, carbon addition, delta con-
centration of HC or NOx, and static-to-
dynamic transition experiments. The five
species lists are: formaldehyde, acetal-
dehyde, ethylene, propylene (other than
matched propylene) and toluene. These
lists include all experiments in which the
species appeared, alone or in mixtures,
on either side of the chamber. Within
each series and species list, the runs are
grouped by HC type beginning with the
simplest type that appears in the series
Within the HC type, the runs are in chro-
nological order. Guidance for the useful-
ness of a specific run for the purpose ol
model testing is also given in terms ol
general ranking categories.
Selecting Runs and Obtaining
Data
Using three examples, the report illus-
trates how experiments from the UNC
Smog Chamber Database might be se-
lected to test mechanisms. General rec-
ommendations and assumptions are de-
fined to guide modelers in general selec-
tion. The examples discussed involved
the testing of an explicit toluene reaction
mechanism, a mechanism under dynamic
conditions, and an EKMA-type mecha-
nism for urban conditions.
The data set described in this report is
neither complete nor static. New projects
are adding to this data set, and the exist-
ing runs are under constant review and
analysis. The possibility for, and the
means of, future revisions are explained.
This data set has been supplied to
modelers for analysis under EPA Contract
Nos. 68-02-3738 and 68-02-4104. The
purpose of this report is to describe the
data set and to provide guidance so others
in the scientific community can use it.
Fully processed runs and the experimen-
tal conditions database are available on
an ANSI formatted magnetic tape. Copies
of the tape and other supporting informa-
tion are available through the authors.
-------
No Dilution
Large Dilution
•2
.5
1
Q
|
n
o
i
o
1
n
O
pi
i
o
; .u
0.8
0.6
0.4
0.2
0.0
,'\'\'\'\'\'\'\'\'\'\'\'\'\'.
June 27. 1983 -
_ _
_ _
'- o3 -.
L X" :
7 NO NOi / ~
^x^^~/^~^^ PAN
, i , \ -\ i \ , \ -fr^L' • \ t i ' ' i i . i i
6 8 10 12 14 16 18
I.U
0.8
0.6
0.4
0.2
0.0
Hours, EOT
1.0
0.8
0.6
0.4
0.2
nn
_' 1 ' I ' I ' 1 ' 1 ' \ ' l ' I ' ] ' 1 ' \ ' \ ' \ ' .
- June 14, 1983 ~
_ _
•
~ °3 :
I / ]
/ —
NOZ /
~^-~7^—-—-^^ PAN
1.0
0.8
0.6
0.4
0.2
nn
|
n
o
i
i
S
6
cO
O
0
S
S
6
o
i
5
m
o
i
o
^
I.U
0.8
0.6
0.4
0.2
0.0
_'i'{'\'\'\'\'\'\'\'\'\'\'\'.
July 29, 1983 -
-
—
-
-
~ NO ^3 "-
': 1 \ t 1 r**\ • I-JW I t r"l" ' 1 1 1 1 1 I . ! - J .
6 8 10 12 14 16 18
/ .L/
0.8
0.6
0.4
0.2
0.0
Hours. EOT
1.0
0.8
0.6
0.4
0.2
nn
k ' I ' I ' I ' I ' I ' I ' I ' I ' I ' I ' I ' I ' I ' .
7 June 27. 1983 ~
— _
"
-
— _
"
03
^ff^-
---i**:?^^^^ .-
1.0
0.8
0.6
0.4
0.2
0.0
$_
n
0
i
1
S
|
o
o
o'
%
8 10 12 14 16 18
Hours, EOT
10 12 14 16 18
Hours, EOT
Figure 1.
Outdoor smog chamber runs as examples of hierarchial experimental conditions. The NOi concentration was <=*0.25 ppm, HC
concentration =2.6 ppmC of propylene/n-butane/toluene mixture. Large dilution means that 20% of initial mass is left after 10 hours
of dilution fequivalent to mixing height rise from 250 m to 1250 m).
H. E, Jeffries. K. G. Sexton. R. M. Kamens. and M. S. Holleman are with the
University of North Carolina. Chapel Hill. NC 27514.
Marcia C. Dodge is the EPA Project Officer (see below).
The complete report, entitled "Outdoor Smog Chamber Experiments to Test
Photochemical Models: Phase II," (Order No. PB 85-191 542/A S; Cost: $ 19.00,
subject to change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Atmospheric Sciences Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
•t, U.S. GOVERNMENT PRINTING OFFICE: 1985-559-016/27079
-------
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Official Business
Penalty for Private Use $300
OCOC329 PS
U S ENVIR PROTECTION AGENCY
REGION 5 LIBRARY
230 S DEARBCRN STPEET
CMCftGO It 60604
------- |