United States
 Environmental Protection
 Agency
 Environmental Sciences Research  ^
 Laboratory
 Research Triangle Park NC 27711   *^
 Research and Development
 EPA-600/S3-80-095  Mar. 1981
 Project Summary
 Aerosol Formation from Diesel
 Exhaust and  SO2

 A  Chamber  Study
  Auto exhaust emissions are one of
the most  omnipresent,  man-made
pollution sources. Especially within
the past decade, a significant quantity
of research has been devoted to the
impact of the emissions of gasoline-
fueled autos upon the health of man
and the environment, the information
on smog, and the composite levels of
various pollutants. These investiga-
tions have provided data upon which
both federal and state control strate-
gies are based. During this period, the
use of light duty diesel-fueled autos in
the United States was insignificant.
Total  on-highway diesel fuel usage
was about 10% by volume of the
amount of gasoline  used (Office of
Technology Assessment, 1977).
  With the recent escalation in the
price of fuel for autos, there is an
economic justification for light  duty
diesel-fueled  autos.  Several  foreign
auto makers and at least one domestic
auto maker offer such vehicles. It is
speculated  that light  duty  diesels
could  comprise 25% of the new car
fleet by 1985. The potential for such a
dramatic increase of light duty diesel-
fueled autos as well as development of
medium duty diesels, necessitates
that  investigations  anticipate   the
impact of such levels of diesel emis-
sions upon urban pollution levels (both
paniculate and gaseous), photochem-
ical and catalytic oxidation processes
responsible for significant rates of
gas-to-particle  conversion   in   the
urban  atmosphere, and the health of
the urban population.
  As a result of the significant dif-
 ferences  between  diesel and gas-
 fueled auto emissions and the poten-
 tial for widespread use of diesels in the
 near future, smog chamber studies of
 a  simulated  diesel  exhaust/urban
 system were performed  by Calspan
 Corporation under Contract No. 68-
 02-2987 from the U.S. Environmen-
 tal Protection Agency. The program,
 which was performed in Calspan's
 600 m3 indoor smog chamber, was
 aimed at assessing the role of diesel
 emissions  in future urban environ-
 ments.  The  experiments provided
 semi-quantitative  data  on  aerosol
 formation  and  growth,  chemical
 kinetics, and the evolution of muta-
 genicity in the diesel exhaust/SO2/
 hydrocarbon system.
  This Project Summary  was devel-
 oped  by  EPA's  Environmental
 Sciences  Research Laboratory.
 Research  Triangle   Park. NC. to
 announce key findings of the research
project that is fully documented in a
 separate report of the same title (see
Project Report ordering information at
back).
Experimental Results
  In recent years, a significant level of
effort has been devoted to the study of
the chemical  and physical  processes
responsible for smog formation in urban
environments. Experimental data pro-
vided by field and laboratory measure-
ments have allowed many of the more

-------
 important parameters to be identified
 and modeled. As a result, current photo-
 chemical  kinetic   models  generally
 provide at least the qualitative trends
 observed experimentally.
   One of the primary differences be-
 tween  the emissions  of  light  duty
 diesels and gasoline-fueled, catalyst-
 equipped autos is the great quantity of
 diesel primary particulates (soot). One
 of the purposes of the program was to
 investigate the influence of the primary
 particulates of  the diesel  emissions
 upon the rate of gas-to-particle conver-
 sion. If  the primary particulates do
 significantly  influence  the chemical
 reactions of the system, existing models
 will not provide an accurate assessment
 of the system.
   Condensible organics known  to be
 mutagenic are also associated with the
 primary particulates of the diesel emis-
 sions. Another aspect of this program
 was  to  investigate the fate of  these
 organics   under  various   simulated
 conditions to determine if the mutagen-
 icity  of  the emissions  is likely  to be
 modified with  exposure to an  urban
 environment.
   This experimental study of the  diesel
 emission/hydrocarbon  system   was
 performed to  assess the physical and
 chemical nature of the system. Particu-
 lar interest was vested in the primary
 particulates of the diesel emissions and
 their influence upon the system.
Overview
   Dilute diesel emission systems were
 studied  in  a  600  m3  indoor  smog
 chamber under various conditions likely
 to be encountered in an urban environ-
 ment. The major emphases of the study
 were aerosol formation and growth,
 chemical kinetics, and the mutagenic
 activity of the diesel aerosol. Important
 parameters with respect to visibility
 degradation and the formation of new
 aerosol  material  included   relative
 humidity, S02 concentration, HC/NOx
 ratio,  irradiation  interval,   and   the
 presence of  primary diesel aerosol
 (primarily soot).
   Aerosol formation and growth in UV-
 irradiated diesel  exhaust/SOz/propy-
 lene experiments were primarily attri-
 buted to sulfate aerosol formation. The
 rate of SO2  oxidation to form sulfate
 aerosol was  5% hr"1 maximum during
 the Oa production period. After the peak
 ozone production period, the first 3 to 6
 hours of a  24-hour experiment,  the
 ozone concentration remained relative-
 ly constant and the  SO2 oxidation rate
 decreased to approximately 1% hr"1.
 These observations could be tentatively
 attributed to the S02/hydroxyl  radical
 reaction during the initial hours of the
 experiments.   The  hydroxyl  radical
 concentration is high  during the first
 few hours of ozone production and the
 gas phase SOa/hydroxyl reaction pre-
 dominates.
   The  effect of  the  primary  diesel
 aerosol on new particle formation was
 investigated by conducting experiments
 with  and without  prefiltered  diesel
 exhaust.   The  presence   of   diesel
 aerosols  precluded  the formation of
 new  aerosol  particles;  instead,  the
 existing aerosols grow in diameter by
 the condensation of nuclei on them. The
 presence of the primary diesel aerosols
 appeared to increase  the rate  of S02
 oxidation and  the amount  of  sulfate
 formed via unknown heterogeneous or
 catalytic reactions. Increases in aerosol
 sulfate of from  25 to 75 percent were
 attributed to the presence of the primary
 diesel aerosol in contrast to what was
 attributed to the SO2/hydroxyl  radical
 gas phase reaction.
   In  most irradiated experiments, the
 production   of  organic   and  nitrate
 aerosols represented minimal contribu-
 tions to aerosol formation and growth.
 Nitrate aerosols  (<1000 //g/m3) did
 represent a substantial fraction of the
 total aerosol growth in  an experiment in
 which NH3 was added  to the system.
   The mutagenic  activity of the diesel
 aerosol was observed to be affected by
 simulated  atmospheric   exposure.
 Containment of dilute diesel emissions
 under  non-irradiated  conditions
 resulted in an increase  in the mutagenic
 activity of the diesel aerosol. While irra-
 diation alone appeared to have no effect
 upon  the  mutagenic  activity,
 photochemically  reactive systems
 exhibited  decreases   in   mutagenic
 activity  of  the diesel aerosols with
 irradiation. At  least a portion  of the
 observed decrease in mutagenicity was
 attributed to the presence of O3 (a
 product of the photochemistry).
Conclusions
   For UV-irradiated tests, the activity of
 the  diesel  emission/SOz/propylene
 system (with respect to aerosol forma-
 tion  and  growth)  had three distinct
 levels which correlated well with the
 rate  of change of ozone concentration
with time. Although new particle forma
tion was frequently associated with th
initiation of  irradiation,  the  rate  c
increase in  aerosol volume remaine
low until substantial amounts of nitre
gen dioxide were formed by the oxida
tion  of nitric  oxide.  The  ozon
concentration was very low during thi
period (<10 ppb). Following an increas
in the concentration of nitrogen dioxide
the  ozone  concentration  began  t
increase.  This period of  increasing
ozone  concentration  represented thi
most active  portion of the experimen
for aerosol growth. Sincethere general
ly were sufficient particles to serve  a
sites  for  condensation  of the  nev
matter, few new particles were formed
The  final  level   of  system  activit
occurred after the ozone peak, whicl
generally coincided with  the depletioi
of propylene (the predominant reactivi
hydrocarbon  species).  Subsequent  t<
the ozone peak, there was a gradual bu
persistent decay in the ozone concen
tration. Little further aerosol growth oi
visibility  degradation  was  associatec
with this period.
  Aerosol formation and growth in the
diesel emission/SO2/propylene system
was primarily due to sulfate formation.
Prior to the period of rapidly increasing
ozone (during the decay of nitric oxide),
S02 oxidation rates (Rso2)on the order oj
0.1 % hr"1 were observed. S02 oxidatidl
rates were determined from measured
increases in aerosol sulfate with time
and the concentration  of available S02.
During the period  of increasing ozone
concentration,  Rso2 generally
approached 5% hr"1. Subsequent to the
formation of  the ozone peak,  Rso2
decreased to a low value (<1 % hr"1) and
occasionally was undetectable due  to
losses of sulfate aerosols in excess  of
the rate of formation  of sulfate. Such
aerosol losses were due to sedimenta-
tion and diffusion to the chamber walls.
The  fact that sulfate  formation  was
greatly reduced, even though ozone and
SO2  continued  to coexist,  provided
evidence that  the  SOz-ozone reaction
was not the primary  path  for sulfate
formation during the period of high Rso2.
Comparisons of aerosol sulfur as deter-
mined by 1C (soluble sulfate) and XRF
(total elemental sulfur) demonstrated
good  agreement,  indicating  that the
concentration  of  sulfonated  organics
was small with respect to total sulfates.
  The primary particulates of the diesel
emissions were principally composed of
elemental   carbon   and  various
condensible  organic compounds.

-------
 Analyses of the primary particulates of
 diluted diesel  emissions  revealed an
 elemental  to  organic carbon  ratio of
 about 5 to 1 . One of the major effects of
 the primary particulates was to increase
 the mass mediandiameter(MMD)of the
 aerosol (e.g., 0.3 to 0.4 /urn) through a
 redistribution  of mass versus  particle
 size. As a result, the presence of primary
 particulates yielded an increased effec-
 tiveness of the aerosol to scatter light.
 The increased MMD might be indicative
 of an increasing fraction of material
 deposited deep within the human lungs
 during respiration, as a larger portion of
 the  individual  aerosol  particles  falls
 within the respirable size range (0.5 to 3
 fjm diameter).
  Greater quantities of aerosol sulfate
 were consistently  produced by irradi-
 ated  systems containing the  primary
 diesel aerosol  than by  comparable
 systems which did not contain  the
 primary diesel aerosol. While the actual
 quantity of   increased  sulfate  was
 difficult to assess (due to few duplicate
 experiments),   increases  in  aerosol
 sulfate of from 25  to 70 percent were
 attributed to  the primary particulates.
 This  apparent  increase  in  sulfate
 formation may have been due to hetero-
 geneous reactions on the surface of the
 carbonaceous particles  or may have
 been  associated with the high molecu-
 lar weight organics known to be carried
 by the primary aerosols of the diesel
 exhaust. No significant sulfate forma-
 tion was observed in the non-irradiated
 diesel  exhaust/SOa/hydrocarbon
 system  at low humidity. However, at
 high humidities (—92%) SO2 oxidation
 fates  as high  as  1.5% hr"1 were
 observed  under dark  conditions.  To
 minimize error in the measurement of
calibrated at the  beginning  of  each
experiment using carrier air with same
RH as that to be encountered in the
experiment.
  In irradiated  diesel emission/SO2/
hydrocarbon  systems, aerosol nitrate
concentrations  of  about  1  pg/m"
(approximately the blank level of the
filter.) and gaseous ammonia concen-
trations  of  less  than  10  ppb  were
typically   observed.  With   the
introduction of ozone at low RH (<80%)
into the  system, the concentration of
aerosol nitrate appeared to  increase
temporarily. At high  humidity (94%),
significant  aerosol nitrate was formed
(~30 pg/m3) and  persisted with the
introduction of ozone. Aerosol nitrate
represented a substantial  fraction of
total  aerosol   growth  in  only  one
experiment (where 10 ppm of NH3 had
been  added to  the chamber).  The
reaction of NH3 with gaseous nitric acid
to produce the more stable ammonium
nitrate apparently was responsible for
the  presence  of  a   significant
concentration  of aerosol nitrate (in
excess of  1000 fjg/m3) in this particu-
lar experiment.

Recommendations
   This program was wide-ranging  in
scope and as  a result, survey-type  in
nature. The  goals of the  program were
to  identify  potential problem  areas
associated with diesel emissions in an
urban atmosphere to isolate the specific
parameters responsible.
  The  most significant  trends  in the
tests  of  the  diesel emission/SOa/
hydrocarbon system were  associated
with primary particulates. The primary
particulates were  apparently respon-
sible for an  increase in the amount  of
sulfate formed  during irradiated tests
and during high humidity conditions  in
the  absence  of  irradiation.  Further
experiments should be  performed  to
determine  the  expected increase  in
sulfate in  contrast with the loading  of
primary  particulates.  Experiments
should also be performed to identify the
component of the primary particulates
responsible  for the  increased sulfate
formation   (i.e.,  the  carbon  or  the
organics).
  The experiments  performed in  this
program  concentrated on  simple
systems (i.e., the addition of propylene
and butane to diluted diesel emissions)
to simulate the diesel emission/urban
atmosphere. This system was chosen
initially to aid in modeling the observed
photochemistry. Only a limited number
of experiments were performed with a
more  representative  mixture of diesel
and gasoline-fueled engine emissions.
While the photochemical reactivities of
the simple and more complex systems
are similar,  there are definite  differ-
ences in specific aspects of the system.
To assure optimum relevance of experi-
mental data, future experiments should
be performed with the more representa-
tive  diesel/gasoline-fueled  engine
emission system.
  The potential for nitrates to contribute
substantially to aerosol formation and
growth in the  presence of  high  NHs
concentrations has been demonstrated.
Further investigations  are  needed to
determine the  role of  NH3 and liquid
water  in promoting  the formation of
aerosol  nitrate  from  an   apparently
abundant source of gaseous nitrates.
  This program began to  explore the
possible effects of age and atmospheric
exposure  upon  the mutagenicity of
diesel aerosols. There are many unan-
swered questions pertaining  to the
mutagenicity of diesel  aerosols  and
aerosols  in  general.  The questions
concern:  the chemical species respon-
sible  for the  mutagenicity  of  diesel
emissions;  the effects  of  irradiation
upon  specific  mutagenic compounds;
the  overall   lifetime  of   mutagenic
materials in  the atmosphere; mechan-
isms for  the elimination of mutagenic
compounds  in   the  atmosphere;
potential mechanisms  by  which
mutagenic compounds  may be formed
by  atmospheric  reactions;  and,  the
mutagenicity of gaseous constituents of
diesel emissions. With the potential for
an increasing  atmospheric burden of
mutagenic aerosols to adversely affect
the public health, basic studies should
be performed to determine the lifetimes
and possible formation  mechanisms of
mutagenic  compounds  in  the
atmosphere.
   This Project Summary was authored by Staff of the Center for Environmental
    Research Information, USEPA, Cincinnati. OH 45268.
   H. M. Barnes is the EPA Project Officer (see below).
   The complete report, entitled "Aerosol Formation from Diesel Exhaust andS02,"
    (Order No. PB81-121 147; Cost: $11.00, 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:
          Environmental Sciences Research Laboratory
          U.S. Environmental Protection Agency
          Research Triangle Park, NC 27711
                                                                                      > U.S. GOVERNMENT FWNTINO OFFICE: Ml -757-012/7009

-------
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Postage and
Fees Paid
Environmental
Protection
Agency
EPA 335
Official Business
Penalty for Private Use S300
      "«  i9°OQ329

-------