United States
Environmental Protection
Agency
Atmospheric Sciences Research —,.
Laboratory
Research Triangle Park NC 27711
Research and Development
EPA/600/S3-85/028 May 1985
Project Summary
Atmospheric Reaction
Products from Hazardous Air
Pollutant Degradation
Chester W. Spicer, Ralph M. Riggin, Michael W. Holdren, Fred L. DeRoos, and
Richard N. Lee
This research project was undertaken
by Battelle's Columbus Laboratories to
investigate the products of selected
hazardous air pollutants (HAPs) as they
react in the atmosphere. Many hazard-
ous or potentially hazardous organic
compounds are emitted into the atmos-
phere and are subject to chemical
change by photolysis, or by reaction
with a number of highly reactive spec-
ies, including free radicals, ozone, and
nitrogen oxides. It is important to under-
stand both the fate and the persistence
of a HAP in order to assess its health
significance. However, for many HAPs,
the major atmospheric reaction pro-
ducts have not been identified.
The first phase of this study entailed a
literature survey to determine what
reaction product information was avail-
able for an initial selection of 15 target
HAPs. In the second phase of the
research, a number of the target HAPs
were studied in Battelle's 17.3-m3 smog
chamber to determine major reaction
products and to derive an estimate of
atmospheric lifetime. Those HAPs ex-
amined experimentally include propyl-
ene oxide, epichlorohydrin, toluene,
acrylonitrile, trichloroethylene, ben-
. zene, p-dichlorobenzene, aniline, nitro-
benzene, chlorobenzene, o-cresol, and
phenol. Following the chamber studies,
the third phase of the research was
devoted to development of field-com-
patible sampling and analysis methods
for those HAP reaction products which
(1) represent a potential hazard and (2)
for which current methods are unavail-
able or unsuitable. This report presents
the results of the literature survey, the
preliminary reaction products results,
and the details of the sampling and
analysis methods developed for several
HAP reaction products.
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 ire-
formation at back).
Introduction
Organic compounds emitted into the
atmosphere are subject to chemical reac-
tions because of the presence of a number
of highly reactive species, including free
radicals, ozone, nitrogen oxides, and
similar materials. In certain cases the
reaction products of the organic com-
pounds may be even more hazardous
than the starting compounds. Sampling
and analysis methods have not been
tested and validated for the reaction
products of most hazardous air pollutants
(HAPs), and in many cases, current
methods are probably not suitable for
field measurements of these species.
Indeed, for many HAPs of interest, the
major atmospheric reaction products
have not even been identified.
In order to understand the impact of
chemical reaction products of HAPs, it is
necessary to define the nature of the
major reaction products and then assess
the degree of hazard by measuring the
concentrations of important products in
the field. The major objectives of the
study reported herein were as follows:
-------
1. Identification of major reaction pro-
ducts of certain HAPs of particular
interest
2. Development of sampling and anal-
ysis methods for field measurement
of the reaction products.
The specific HAPs deemed to be of
particular interest are listed in Table 1. A
three-phase approach was taken to
achieve the project objectives:
1. A literature review was conducted
to obtain background information
pertinent to chemical reactions of
each compound to predict the likely
atmospheric reaction products of
each compound.
2. When insufficient information was
available concerning the atmos-
pheric reaction products of a com-
pound, laboratory experiments
were conducted using a smog cham-
ber to simulate atmospheric condi-
tions. The major products were
determined by using screening tech-
niques, including cryogenic trap-
ping gas chromatography, HPLC,
and atmospheric pressure ioniza-
tion triple-stage massspectrometry.
3. As a final phase, methods were
developed for the determination of
some of the reaction products iden-
tified in Phases 1 and 2. Since only
limited number of methods could be
developed within the project re-
sources, emphasis was placed on
those products believed to be of
general concern.
Procedures
A literature survey was conducted in
the initial stages of the project to deter-
mine what reaction product information
was available for an initial list of 15 HAPs.
The conclusion from the survey was that
insufficient reaction product data were
available for many of the target HAP
compounds. As a consequence, smog
chamber experiments were undertaken
io derive reaction product information.
Preliminary product screening experi-
ments were conducted in a 180-1 irradia-
tion chamber. Subsequently, more com-
prehensive experiments to identify reac-
ion products were undertaken in a
17.3-m3 smog chamber. Concentrations
amployed in these chamber experiments
were nominally 5 ppm(v) of the target
HAP and 0.5 ppm NO2 in a background of
jltrahigh purity air. Standard measure-
ments made during these experiments
ncluded those for ozone, nitric oxide,
Table 1. Hazardous Air Pollutants
Targeted for this Study
Acetone
Acrylonitrile
Benzene
Chloroform
Methylene Chloride
Propylene Oxide
Perchioroethylene
Toluene
1,4-Dichlorobenzene Trichloroethylene
Epichlorohydrin Vinylidene Chloride
Ethylene Oxide Phenol
Ethylene Dichloride Cresols
Dibenzofuran Nitrobenzene
Aniline Chlorobenzene
nitrogen dioxide, carbon monoxide, meth-
ane, total hydrocarbon, temperature, and
relative humidity. More specialized meas-
urements were made for (1) the target
HAP, peroxyacetylnitrate, and organic
reaction products by gas chromatography
(GC); (2) individual aldehydes, dialde-
hydes, and ketones by a derivatization/
HPLC analysis procedure; (3) certain
chlorinated carbonyl compounds by an
HPLC method developed during this pro-
gram; and (4) a variety of organic reaction
products by mass spectrometry/mass
spectrometry (MS/MS) with a Sciex, Inc.,
TAGA instrument which was connected
to the smog chamber.
In the later stages of the project, five
compounds were added to the list of
target HAPs, aniline, nitrobenzene, chloro-
benzene, o-cresol, and phenol. These
species were screened for atmospheric
reaction products in the 17.3-m3 cham-
ber, but under a different set of reaction
conditions. Because these compounds
were expected to be rather unreactive,
they were studied in mixture with a
reactive two-component hydrocarbon mix-
ture. Nominal concentrations for these
experiments were 1.0 ppm of the target
HAP, 1.0 ppm propylene, 3.0 ppm butane,
and 0.5 ppm N02. The same sampling and
analysis methods were employed as in
the earlier experiments, with the excep-
tion that the MS/MS, which was unavail-
able, was replaced by a GC/MS system. A
new method for determining phenolic
compounds also was developed and
implemented during these experiments.
In addition, high-volume filter samples
were collected from the chamber at the
end of experiments which produced sig-
nificant quantities of particulate matter.
In one case (aniline), the particulate
sample was extracted and the extract
analyzed by HPLC and GC/MS. For those
experiments in which filters were collect-
ed, the filter mass loading was used to
infer the fraction of reacted HAP material
which was present in the condensed
phase.
Results and Discussion
The results of the product screening
experiments are shown in Tables 2
through 4. Table 2 shows the initial
concentrations of the target HAP and
NOX, the irradiation time, the peak ozone
concentration, and the time to maximum
ozone. It is important to note that the final
five experiments were conducted in a
reactive mixture which generates more
than 0.7 ppm 03 in the absence of the
HAP compounds. Tables 3 and4 show the
reaction products which were tentatively
identified in the screening experiments.
The sampling/analysis methods provid-
ing the tentative identification are noted.
The tentative nature of these identifica-
tions is stressed. Further research is
needed to confirm many of the products
shown in Tables 3 and 4 to identify
additional products.
Simultaneously with the product
screening experiments, a sampling and
analysis methods development effort was
undertaken to develop and evaluate meas-
urement techniques for pertinent HAP
reaction products.
Primary emphasis was placed on the
development of an analytical method for
phosgene, since (1) this compound was a
reaction product of several different chlor-
inated hydrocarbons and (2) existing
methods were not considered to be
sufficiently reliable at trace (sub-parts-
per-billion) levels. A simple derivatization
HPLC technique was developed and eval-
uated in detail. This method was shown
to give good recovery and precision down
to the 50-ppt level.
Modifications to the phosgene method
were made to allow determination of
formyl chloride and acylchlorides. How-
ever, further work is needed to completely
validate these methods. In addition, mod-
ifications were made to NIOSH Method
S330 to allow determination of lower
levels of phenolic compounds. This meth-
od was used to determine phenols in
smog chamber studies, and may be useful
for ambient analysis as well.
Conclusions and
Recommendations
During these experiments, consider-
able experience was gained in the anal-
ysis of HAPs and HAP reaction products
by atmospheric pressure ionization/mass
spectrometry/mass spectrometry. This
technique was especially valuable during
-------
tble 2. Initial Concentration and Maximum Ozone from HAP Product Screening Experiments
Initial Target
HAP Concentration Initial NO*
Target HAP ppmC ppm
Propylene Oxide
Epichlorohydrin
Toluene
Toluene
Acrylonitrile
Trichloroethylene
Benzene
p-Dichlorobenzene
Aniline™
Nitrobenzene™
Chlorobenzene™
o-Cresof°
Phenol
9.15
11.55
33.38
34.45
11.75
8.67
36.0
32.86
5.00
6.55
7.41
7.17
4.36
0.469
0.427
0.445
0.448
0.468
0.882
0.466™
0.421
0.422
0.391
0.405
0.405
0.398
Irradiation
Time Hours
19.5
24.4
5.0
3.1
17.5
6.4
22.7
24.0
5.5
5.5
4.8
5.6
6.2
Time to
(O 3) max,
(Oil max ppm Hours
0.188
0.226
0.442
0.404
0.645
1.046
0.253
0.064
0.107
0.663
0.757
0.314
0.380
77""
24.4'"
2.1
1.9
7.7
5.9
10.3
20.5""
1.4
1.5
1.5
0.8
0.7
""03 increasing when monitoring terminated.
""0.23 ppm NOZ injected after 6.8 h.
^Experiments carried out in the presence of 1 ppmV propylene and 3 ppmV butane.
the product screening experiments, and it
represents a major advance in the state-
of-the-art for HAP monitoring. Further
laboratory and field experiments are
warranted to document the capabilities of
this technique. The results of this study,
while somewhat preliminary in nature,
suggest that much important information
can be learned about the fate and per-
sistence of hazardous air pollutants
through environmental chamber exper-
iments. The knowledge gained in such
studies may be used to eliminate, or at
least minimize future risks associated
with exposure to airborne products- of
HAP reactions.
Propylene Epichloro-
Oxide hydrin Trichloroethylene Acrylonitrile
Formal- Acetone Phosgene' Formaldehyde1
dehyde' (art if act??*
Acetone Formyl chloride' PAN -type compound3
(artifact??*
Dichloroacetyl chloride' HO C(O) CN2
Chloroform3
Carbon monoxide3
PAN-type compound3
Toluene Benzene
Formaldehyde' Peroxyacetyl nitrate3
Benzaldehyde^ Phenol2
Glyoxal' NitrophenoP
Methyl glyoxal' Nitrobenzene2
Peroxyacetyl nitrate3 Glyoxal2
ButenediaP ButenediaP
4-Oxo-2-pentenaP 4-Oxo-2-butenoic
acid2
2-Oxo-2-butenoic
acid2
2-Oxo-2 -pentenoic
acid2
2-Butenal2
Nitrotoluene3
o-Cresol3
2-Butenoic acid2
NitrocresoP
Benzylnitrate2
p-Dichloro-
benzene
DichlorophenoP
Dichloronitro-
phenoP
'HPLC.
Atmospheric pressure ionization MS/MS.
-------
Table 4. Reaction Products Tentatively identified in Product Screening Studies
Aniline
Nitrobenzene Chlorobenzene
o-Cresol
Phenol
Nitrobenzene'
Nitrosobenzene'
Nitrophenol'
2-NitrophenoP
4-Nitrophenol2
Phenolic
Compounds2
Nitrocresol2 4-Nitrophenol2
(two isomers) 2-Nitrophenol2
2-Hydroxy benzaldehyde1
Aminophenol2
Nitroaniline^'2
Benzidine2
Phenol'
Isocyanato benzene'
Hydroxy
benzonitrile1
N-Phenyl
formamide^'2
'Tentatively identified by GC/MS.
tentatively identified by HPLC.
Chester W. Spicer, Ralph M. Riggin, Michael W. Holdren. Fred L. DeRoos, and
RichardN. Lee are with Battelle Columbus Laboratones. Columbus, OH43201.
Robert R. Arnts is the EPA Project Officer (see below).
The complete report, entitled "A tmospheric Reaction Products from Hazardous
Air Pollutant Degradation," (Order No. PB 85-185 841/AS; Cost: $11.50,
subject to change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA22161
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
•It U.S. GOVERNMENT PRINTING OFFICE: 1985-559-016/27060
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
.^NX^xJ-ac%i"-'-:-~T:;'
JUS'B'35 l|fe:;-n ??• !•
Official Business
Penalty for Private Use $300
OC00329 PS
------- |