JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 104, NO. D13, PAGES 15,929-15,940, JULY 20, 1999
Carbonaceous aerosols and their potential role
in atmospheric chemistry
D. J. Lary 1
Department of Geophysics and Planetary Sciences, Tel Aviv University, Tel Aviv, Israel
D. E. Shallcross
Centre for Atmospheric Science, Cambridge University, Cambridge, England, United Kingdom
R. Toumi
Department of Physics, Imperial College, London, United Kingdom
Abstract. This paper considers the nature of carbonaceous urfaces, the means by
which they are activated, the nature of some functional groups that they support,
and some reaction mechanisms that may be involved. Because of the strong affinity
of carbonaceous urfaces for organic species and because of the ease with which
compounds in a high oxidation state can oxidize the carbonaceous urface, it is
highly likely that carbonaceous aerosols are interacting chemically with a range
of organic species in ways that have, as yet, not been fully characterized but may
significantly affect the oxidizing capacity of our atmosphere. If HONO is formed on
the surface of carbonaceous aerosols then this could be a significant source of HOz
as HONO is readily photolyzed to give OH, and it could explain the large values
of HONO often observed in the troposphere. In general, the reduction of NO on
carbonaceous aerosols is an important consideration, and it is addressed here.
1. Introduction
Carbonaceous aerosols (soot, charcoal, elemental car-
bon, etc.) are one of the most ubiquitous materials in
our atmosphere. All combustion processes lead to the
formation of carbonaceous material. As a result, be-
tween 10 and 50% of tropospheric particulates are car-
bonaceous, with particularly high levels being found in
the urban atmosphere. Because of air traffic, carbona-
ceous aerosols (CA) are also found in the stratosphere.
In his excellent introduction to CA, Goldberg [1985] has
pointed out that the universality of CA is related to
their refractory nature (very high melting point) and
to their origin in burning processes. They can even
be formed at temperatures of 200øC or less. They are
generally very impure forms of carbon. Environmental
charcoals have H/C ratios of 0.25-0.69 and O/C ratios
of 0.08-0.33 [Cope, 1979]. In addition, they can contain
up to a percent or so of nitrogen and sulfur, as well as
a host of metals and elements such as A1, Fe, Ca, Mg,
K, and Si [Gay et al., 1983, 1984; Goss and Eisenreich,
1Also at Centre for Atmospheric Science, Department of
Chemistry, Cambridge University, Cambridge, CB2 1EW, Eng-
land, U.K.
Copyright 1999 by the American Geophysical Union.
Paper number 1998JD100091.
0148-0227/99/1998JD100091509.00
1997; Medalia and Rivin, 1982a,b; Medalia et al., 1982,
1983].
Active carbon is extensively used in various indus-
trial processes, often as a reducing agent. It is there-
fore not surprising that recent laboratory studies have
shown that HNOa, NO2 and Os are also all reduced
on CA [e.g., Smith and Chughtai, 1997; Atomann et
al., 1995, 1997a,b; Rogaski et al., 1997; Chughtai et al.,
1994; Tabor et al., 1993, 1994; Thlibi and Petit, 1994].
Modeling studies have shown that this could have a sig-
nificant impact on atmospheric hemistry [Lary et al.,
1997; Hauglustaine t al., 1996; Bekki, 1997]. Since the
atmosphere is so strongly oxidizing, any process that
can lead to the reduction of atmospheric constituents,
such as CA, may have a significant impact on atmo-
spheric chemistry as they oxidize the CA surface. The
adsorptive property of black carbons to remove impuri-
ties from gases and solutions has been known for cen-
turies. Such processes are therefore worthy of further
study. This is particularly true if the reducing agent is
also one of the most abundant particulates of the urban
environment and is even present in the stratosphere.
Examining the nature of these atmospheric processes is
therefore important.
Some useful information on carbonaceous surfaces
can be gained from the literature on activated carbon.
However, considerable caution has to be applied when
distinguishing atmospheric CA and activated carbon.
15,929

15,930 LARY ET AL.' CARBONACEOUS AEROSOLS AND ATMOSPHERIC CHEMISTRY
They may have similarities but also important differ-
ences. One should not necessarily assume that CA
are 100% carbon, or that they are crystalline with a
well-known structure, since they are often incompletely
burned biomass with a complex and varying composi-
tion of organic phases.
Although carbonaceous materials can vary consider-
ably in their chemical reactivity, they do have several
features in common. They are all carbonaceous, they
can all be oxidized, they have a wide range of surface
functional groups, and they typically have large surface
areas.
Sections 2-10 consider the nature of carbonaceous
surfaces, the means by which they are activated, some
functional groups they support, and some reaction mech-
anisms that may be involved.
2. Nature of the Surface
A surface of carbon, or carbon plus hydrogen, is
strongly hydrophobic. The presence of oxygen makes
the surface more hydrophilic. CA may be acidic or basic
depending on the way in which they are formed. Ex-
posure of black carbon to temperatures between 200 ø
and 400øC yields an acidic surface whereas treatment
at higher temperatures in CO2 followed by exposure to
oxygen produces a basic form [Goldberg, 1985].
CA found in the environment are usually formed at
high temperatures and come from a variety of sources,
such as the combustion of plants, woods, fossil fuels,
and industrially produced substances. The properties
of the CA, such as the surface morphology and size
distribution, reflect their origin and their history since
they left the place of formation. Small submicron par-
ticles have usually been formed from combustion in the
vapor phase (e.g., the combustion of a gas), and the
larger particles which are bigger than 10/m reflect the
structure of the burnt material, the so called char. Both
may come from the same source, such as the combus-
tion of plants and other biomass [Goldberg, 1985]. A
detailed review of the chemical and physical properties
of many carbonaceous materials can be found in the
work of Mantell, [1968].
In CA the elemental carbon may be considered to be
a disordered form of graphite (whereas the organic car-
bon content can not be thought of in this way). The
degree of ordering within the carbonaceous material is
reflected in a number of its properties. e.g., the density
of a single crystal of graphite is 2.25 gcm -a, the den-
sity of coke is 2.05 gcm -3, and the density of carbon
lamp black is 1.90 gcm -a [Mantell, 1968]. Another ex-
ample is the heat of oxidation of graphite to CO2 which
is 32621 J g-, whereas the heat of oxidation of amor-
phous to CO2 is more 33104 J g-, and the heat of
oxidation of wood charcoal to CO2 is greater still at
33807 J g- [Mantell, 1968].
X-ray analysis of carbonaceous material shows that
it has a graphite like structure with randomly orien-
tated microcrystallites, with each of the platelets about
1 nm thick [de Vooys, 1983]. The disorder arises in the
amorphous carbons from the smaller amount of carbon
atoms in the layer planes, which have a mean diameter
of the order of 2.5 nm [Goldberg, 1985]. Typically, only
a percentage of the carbon surface is active, so poison-
ing is not very difficult. Usually, the smallest graphite
crystallites are oxidized first thereby exposing more of
the larger crystallites, which are less easily oxidized [van
der Plas, 1970].
It is found that activated carbon absorbs hydrocar-
bon vapors in preference to water vapor [Mantell, 1968].
Heat is released when commercially available activated
carbon samples are wetted. The heat release increases
when the surfaces are highly oxidized. e.g., van Driel
[1983] showed that when two different activated carbon
samples were wetted by water between 34 and 35 J g-
of carbon was released. However, after oxidation this
increased to almost double, 66 J g- of carbon. This
implies that part of the surface was of a more polar
nature after oxidation. This explains why CA act as
atmospheric cloud condensation nuclei, and it implies
that atmospheric CA is oxidized.
In contrast, when the surface was wetted with toluene,
instead of water, the heat released was between 129
and 131 J g- of carbon, with only a slight increase
to between 135 and 138 J g-X of carbon for the oxi-
dized surface. The small increase in energy release was
attributed to a possible increase in surface area. The
heat release on wetting can provide energy for surface
reactions.
The affinity of carbonaceous urfaces for hydrocar-
bons may have significant implications for atmospheric
chemistry. If the effect of the HNO reduction on CA
can be observed in atmospheric observations, as it seems
from the work of Lary et al. [1997] and Hauglustaine t
al. [1996], then it is also likely that atmospheric inter-
actions of organic compounds with CA should be ob-
served, because they should be taken up preferentially
on CA. Goss and Eisenreich [1997] measured the sorp-
tion of polar and nonpolar volatile organic compounds
(VOCs) to particles from a combustion source. A de-
crease in the sorption with increasing relative humidity
was observed for all VOCs. The calculated sorption
enthalpies suggested stronger sorption for polar com-
pounds compared with nonpolar compounds of compa-
rable volatility. Existing field data indicate that soot
may significantly affect the environmental speciation of
polycyclic aromatic hydrocarbons (PAHs) [Gustafsson
et al., 1997]. The interactions between organic species
and CA are likely to be quite complex. e.g., McDow
et al. [1996] have shown that 4 out of 10 of the ma-
jor organic compound classes found in organic aerosols
include compounds that accelerated the photodegrada-
tion of benz[a]anthracene.
The most reactive areas of the surface are likely
to be where carbon is not exerting its full valency.
These will often occur where the soot particles have

LARY ET AL.: CARBONACEOUS AEROSOLS AND ATMOSPHERIC CHEMISTRY 15,931
exposed aliphatic and aromatic chains. These chains
are stripped off as the CA is oxidized leading to the for-
mation of CO, CO2, and water. The time taken to strip
off these chains is prolonged by an increased H content
in the soot [Smith and Chughtai, 1996].
Carbonaceous urfaces absorb hydrocarbon vapors in
preference to water vapor [Mantell, 1968]. It is therefore
very likely that many atmospheric interactions are oc-
curring between hydrocarbons and CA. We suggest hat
an important class of reactions may be with organic per-
oxides which can oxidize the carbonaceous aerosols and
which are reduced in the process.
3. Carbonaceous Aerosols as Cloud
Condensation Nuclei
There have been several studies of the state of wa-
ter adsorbed on carbonaceous surfaces. e.g., active car-
bons and soot were studied by Berezin et al. [1997]
who found that the state of adsorbed water is similar
to the state of a stretched liquid. CA can act as cloud
condensation uclei (CCN). e.g., Konopka and Vogels-
berger [1997] studied possible water condensation pro-
cesses during the formation of aircraft contrails for dif-
ferent kinds of condensation centers. They found that
soot particles with or without sulfuric acid can be more
easily activated than small H20/H2SO4 droplets for at-
mospheric situations.
Condensation properties of ultrafine carbon parti-
cles in the Aitken range (particle diameters between
20 and 100 nm) were investigated by Kotzick et al.
[1997]. They studied condensation phenomena with dry
monodisperse soot aerosols. After reaction with ozone,
activation of soot particles occurred at lower supersat-
urations. Fourier Transform Infrared (FTIR) studies
revealed that oxygen-containing functional groups are
generated on the particle surface during oxidation, fa-
cilitating water uptake in the condensation process.
Lainreel and Novakov [1995] found that the nucle-
ation ability of chemically modified carbonaceous parti-
cles increased with increasing soluble mass fraction and
was comparable to that of (NH4)2SO4 when the soluble
mass fraction exceeded about 10%. The hygroscopicity
of particles generated by combustion of diesel fuel in a
diffusion flame increased when a sulfur-containing com-
pound was added to the fuel. The CCN characteristics
of diesel soot appear to be comparable to that of wood
smoke aerosol. Karcher et al. [1996] analysed the pri-
mary contrail particles (aqueous solution droplets nu-
cleated in situ, emitted insoluble combustion aerosols,
and entrained background aerosols) and found that soot
must be involved as ice-forming nuclei if the visibility
criterion is to be fulfilled.
4. Surface Activation
Industrially, the basic method of activation is the in-
troduction of further amounts of oxygen into the sur-
face of carbon. This is done by submitting the carbon
surface to oxidizing agents. These are either gaseous ox-
idizing agents, such as oxygen, ozone, air, water vapor,
carbon dioxide, and nitrogen oxides, or solutions of oxi-
dizing agents, such as nitric acid, a mixture of nitric and
sulfuric acid, hydrogen peroxide, acidic potassium per-
manganate, chlorine water, sodium hypochlorite, and
ammonium persulphate. Our atmosphere is a strongly
oxidizing atmosphere that contains both categories of
oxidizing agents in significant quantities. Therefore it
should be perfectly able to activate, and keep activated,
any CA that are present. Jankowska et al. [1991] re-
ports that even a carbonaceous surface activated with
ariel oxygen has many oxygen-containing surface func-
tional groups.
Carbon oxidation is a complex heterogeneous process.
Though the exact mechanism for the overall process is
not fully understood, the reaction of carbon with oxy-
gen can be written stoichiometrically as the following
two exothermic reactions:
C+O2
2C + 02
, CO2 AHR=-387kJmol - (1)
, 2CO AHR=-226kJmol-  (2)
It is thought that both CO and CO2 are primary prod-
ucts [$misek and Cerny, 1970] and that the CO/CO2
ratio increases with temperature.
5. Surface Functional Groups
All carbonaceous materials, even pure materials like
diamond and graphite, contain surface functional groups
[van Driel, 1983]. A range of independent investigations
has shown that a large variety of surface compounds are
formed on the carbon surface. A characterization of the
functional groups produced by the combustion of hex-
ane has led to a more detailed understanding of the
overall surface structure [Keifer et al. 1981, Akhter et
al., 1984, 1985abc, 1986, Chughtai et al. 1994, Smith
and Chughtai 1995, 1996, 1997].
Those containing oxygen compounds are particularly
important because of their universal occurrence [Janko-
wska et al., 1991, p.82, and references therein]. The
origin of these functional groups can be the starting
material from which the CA were formed (this is par-
ticularly true for CA produced by raw materials rich in
oxygen), or they can be introduced later. All commer-
cial activated carbons also contain mineral matter. It is
very likely that this applies to CA too, and these may
have a significant effect on the rates of heterogeneous
reactions on CA.
The number of oxygen containing surface groups will
depend on the oxidizing agent and the type of surface.
Nonetheless, two points should be noted. First, oxygen-
containing surface functional groups typically represent
90% of the total amount of bound oxygen [van der Plas,
1970]. Second, several types of groups are found Jr-

15,932 LARY ET AL.' CARBONACEOUS AEROSOLS AND ATMOSPHERIC CHEMISTRY
respective of the type of active carbon or how it was
modified.
It has been found that there are various types of car-
bons: those produced at low temperatures that typi-
cally have a negative surface charge and acidic behavior,
and those produced at high temperatures that typically
have a positive surface charge and basic behavior. The
surface groups are found mostly at the edges of graphite
type basal planes, which supports the idea of electronic
interaction between surface groups. The acidity of one
group influences neighboring groups and an overlap is
possible between acidic strengths.
On air oxidation the carbonaceous surface becomes
more acidic and therefore more polar. This is probably
because of the formation of monocarboxylic or dicar-
boxylic acid groups [van Driel, 1983]. The reducing
properties of activated carbon are slightly decreased by
oxidation, perhaps owing to the formation of hydro-
quinone groups [van Driel, 1983]. Carbon groups at the
rim of the carbon plane are destroyed, and C=0 groups
are formed in a yet unknown configuration. Figure 1
shows the principal types of acidic groups formed on
carbonaceous urfaces; Figures 2 and 3 show models of
carbonaceous surfaces. It can be seen that the surface
OH
Carboxylic Carboxylic anhydride
o H
Normal actone Fluorescein-type lactone
Phcnolic Quinonoid
Figure 1. Principal types of acidic surface groups [after
dankowska et al., 1991; Mattson and Mark, 1971 ].
groups present on the activated carbon surface are iden-
tical to those found on carbonaceous aerosol surfaces.
The carboxyl surface groups make the CA acidic, and so
they can take up greater quantities of ammonia, some-
thing that may be relevant in the troposphere. Nob
howed
that there was a relation between ammonia adsorption
and the presence of surface carboxyl groups.
Smith and hugtai [1995] used FTIR spectroscopy
to study the structure and reactivity of black carbon
(in the form of n-hexane soot). They found that the
soot structure, as produced by high-temperature incom-
plete combustion is predominantly aromatic with a sur-
face coverage by oxygen-containing functional groups
of about 50%. $ergides et al. [1987] also found that
50% of the CA surface was covered by oxygen contain-
ing groups and that the ratio of aromatic to aliphatic
carbon in n-hexane soot is at least 9:1.
6. Mechanisms for Surface Reactions
The active sites on carbonaceous urfaces are thought
to consist of carbon atoms that are not exerting their
full valency. For example, the carbon-water reaction
was explained by Long and Sykes [1948] by carbon
atoms attached to the rest of the lattice by only three
bonds. Such sites are likely to be found at the free edges
of either the lamellar or the cross-linked aromatic rings,
which are the basis o the carbonaceous structure.
Some processes occurring on the CA will be catalytic,
and some will not. When considering the likely im-
pact of noncatalytic processes, knowing the maximum
number of carbon atoms that can be involved is impor-
tant. Figure 4 shows the equivalent gas phase mixing
ratio of carbon atoms if all the CA mass is assumed
to be carbon for 1, 10 and 100 ng m -3 as a function
of the altitude. At the earth's surface (1000 mb) I ng
m -3 of carbon is equivalent o about 2 pptv of car-
bon. During the Mauna Loa Observatory Photochem-
istry Experiment (MLOPEX) II campaign, values up
to 200 ng m -3 of carbon can be observed with values
of around 10 ng m -3 being quite typical, equivalent to
between 20 and 200 pptv of carbon. As pressure drops
exponentially with height this has increased to 10 pptv
at 15 km (100 rob). Blake and Karo [1995] report a typ-
ical Northern Hemisphere lower stratosphere value of
1.8 ng m -3 that corresponds to about 20 pptv of car-
bon.
Smith and Chugtai [1995] have shown that FTIR
methods are particularly well suited for following net
changes in surface groups, and gas phase reactant/pro-
duct concentrations. FTIR has been the key tech-
nique in determining the kinetics and mechanisms of
some important heterogeneous reactions of black car-
bon with gas phase oxidant molecules. e.g., the reaction
of NO2/N204 with soot follows a dual path mechanism,
down to 2 ppm, which is reflected in the rate law, initial
rate Ri'

LARY ET AL.: CARBONACEOUS AEROSOLS AND ATMOSPHERIC CHEMISTRY 15,933
H
COOH COOH OCo HC---\"C--COOH
HO c//'ø OH O_  OH
o o%/oo//o ø\"ø\"ø   ,, .,,ooo.r  OCH
Figure 2. Model of a fragment of an oxidized active carbon surface (after Tarkovskaya et al., [1977]
reproduced in English by Jankowska et al. [ 1991 ]).
Ri- (kl q- k2[soot] 1/2) PN02 (3)
On the other hand, catalytic decomposition initiates
the reaction with ozone, followed by the formation of
surface carboxylic groups and gaseous CO2 and H20.
The evidence suggests that dissociation of ozone yields a
steadystate concentration of the excited oxygen atom,
which is actually the oxidant. FTIR combined with
chemical measurements has proven that a high solubil-
ity observed for carbon particles exposed to ozone has
its origin in the hydrolysis of the surface carboxylics.
Significant effects of simulated solar radiation on the
reactions, especially in the soot/SO2/H20/O2 system,
has been revealed by FTIR. Infrared will continue its
central role in the examination of increasingly complex
systems containing black carbon, particularly through
its interface with ancillary techniques.
7. Carbon and Ozone
Smith and Chughtai [1996] report that the reaction
of ozone with a carbonaceous surface has three distinct
stages. First, a rapid reaction with the surface produces
functional groups such as hydroxyls and carboxylates.
O3 begins the degradation of the soot particles by strip-
ping them of aliphatic and aromatic chains with the for-
i o
Figure 3. Two-dimensional model of n-hexane soot as formed in a flame [after $ergides et al., 1987].

15,934 LARY ET AL.' CARBONACEOUS AEROSOLS AND ATMOSPHERIC CHEMISTRY
1 E-8 1 E-7 1 E-6
Black carbon Loading
1 ng/m3
() 10 ng/m3
---- 100 ng/m3
1E-12 1E-11 1E-10 1E-9
1 ' \"''\"'1 , ,,,,,,,I , ,,,,
2-
_
_
5-
_
_
lO
2-
_
_
5-
_
_
_
lOO
2-
_
5-
_
1,000 -  \"'\"'1 ; \".'\"'1 ' \"'\"'1 ' \"'\"'1 & \"'\"'1 0
1E-12 1E-11 1E-10 1E-9 1E-8 1E-7 1E-6
Equivalent v.m.r. of Carbon
4O
3O
(D
20 '
,, ....
lO
Figure 4. The equivalent gas phase mixing ratio of carbon atoms if all the CA mass is assumed to be carbon
for 1, 10, and 100 ng m '3 as a function of altitude and atmospheric pressure. This is useful when considering
whether or not chemical processes on CA will be significant even if they are not catalytic. v.m.r. means
volume mixing ratio. Read 1E-12 as lx10 -]2.
mation of CO2 and water. This stage is prolonged by
an increased H content in the soot.
Second, a sharp increase in the CO2 released, which
increases to a maximum and then starts to decrease.
This unusual increase undoubtedly means several reac-
tions are occurring. After stripping off the H atoms in
the first stage the O3 oxidizes the CO to CO2 besides
its continued reaction with the most vulnerable parts of
the soot surface. This is consistent with a reduction in
CO release during this stage.
Thirdly, the CO2 release begins to decrease, probably
because the more reactive portions of the soot have now
been oxidized and the more organized parts of the soot
react at a slower rate.
Compared to soot that has not been exposed to
after 300 hours the ozonated soot experienced a de-
crease in particle size of 40% and a decrease in surface
area of 37% [Sergides et al., 1987]. Smith and Chugbrai
[1996] have summarized the individual reactions identi-
fied as shown in Figure 5. The catalytic portion of the
reaction can be written stoichiometrically as the highly
exothermic reaction
203 302 AHR - -284 kJ mo1-1 (4)
This catalytic reaction could have a profound impact on
the stratosphere [Lary et al., 1997; Bekki, 1997]. Cur-
rent assessment of the impact is hampered by our lack
of knowledge about the true \"average\" nature of the
soot. Bearing in mind that the soot lifetime in the lower
stratosphere is of the order of months, it seems likely
that very aged soot may have lost most of its reactiv-
ity as the surface degrades or becomes poisoned. We
certainly expect the available surface area to decrease
with time. Coal and amorphous carbon are rapidly at-
tacked by 03, producing primarily water soluble acids
similar to benzene carboxylic acids. It appears that
CO2 and oxalic acids result from the ozonation of aro-
matic structures [Sergides et al., 1987, and references
therein]. There is an approximately thirty fold increase
in carboxylic acid surface groups when soot is exposed
to 03, the increase in the particle's weight parallels
this increase [Sergides et al., 1987]. The proportion
of black carbon likely to be transformed into polycar-
boxylic acids decreases as the attack proceeds deeper
into the particle.
Smith and Chughtai [1997] have examined the soot-
ozone reaction at low concentrations to determine any
influence of solar radiation on its products and kinetics.
The effect of simulated solar radiation is to change the
product distribution toward CO2(g), CO(g), and H20(g)
at the expense of soot surface functional groups for-
mation. They concluded that the rapid diminution of
ozone in soot's presence is unaffected by solar radiation.
8. Carbon and NO2
Chughtai et al. [1994] found that-ONO,-NO0,-
NNO2, and oxygen surface groups are formed and that

LARY ET AL.: CARBONACEOUS AEROSOLS AND ATMOSPHERIC CHEMISTRY 15,935
I-t20
O
Oh
0
03 + CsurfHy
03 + Csurf
- 3Y/2 H20 + Csurf 03 + SS = SSO* + 02
SSO* + SSO* - 2 SS + 02
 CO 2
> CO 03 + SSO* - SS + 202
0 2 + Csurf  C '\" 0 2 abs
Figure 5. Individual reactions of ozone with CA from Smith and Chughtai [ 1996].
the presence of water plays a significant role leading
to the formation of NO and N20 [see also Pires et al.,
1996]. Tabor et al. [1993, 1994], Thlibi and Petit [1994],
and Rogaski et al. [1997] have found that the reduction
of NO2 to NO is very fast and therefore important. The
aerosol studies by Kalberer et al. [1996] showed that it is
slow when proceeding via a chemisorbed species which
has been characterized by Ammann et al. [1995]. The
chemisorption releases 100 kJ mol - and reversible ad-
sorption 30 kJ mol - JAmmann et al., 1995, 1997a,b].
The NO was produced only slowly with a firstorder rate
constant of around 10 -4 s-; yet even this small value
could be significant when high levels of CA are present.
Rogaski et al. [1997] studied the reactivity and hydra-
tion properties of amorphous carbon in a low-pressure
Knudsen cell reactor at room temperature (298 K). The
reaction with NO2 had an observed ' = 0.11+0.04.
Treating the amorphous carbon with NO2 and O3 does
not alter the H20 uptake, while treatment with SO2,
HNO3, and H2SO4 increases significantly the H20 up-
take.
HONO was first tentatively identified by Atomann et
al. [1997a]. Then Ammann et al. [1998] and Gerecke t
al. [1998] both showed that HONO is a major product
of the soot/NO2 interaction. The formation of HONO
involves several intermediate steps that are not easy to
separate.
If, as reported, HONO is formed, then this could be
a significant source of HOx as HONO is readily pho-
tolyzed to give OH. Even if this process is not catalytic
and only a fraction of the nitrogen incident on the CA
leads to HONO formation, this could be a significant
HOx source as at the Earth's surface 1 ng m -3 of car-
bon is equivalent o about 2 pptv of carbon. During
the MLOPEX II campaign, values up to 200 ng m -3
of carbon were observed, equivalent to about 400 pptv
of carbon. Tropospheric values of around 10 ng m -3
of black carbon are quite typical and are equivalent o
about 20 pptv of carbon. Blake and Karo [1995] report
a typical northern hemisphere lower stratosphere value
of 1.8 ng m -3 that corresponds to about 20 pptv of
carbon. Since OH and HO2 are present in the atmo-
sphere on the pptv scale these interactions could be a
significant source of HOx.
J. Kleffmann [personal communication, 1998] have
recently observed that the reaction of NO2 on dry soot
produces large quantities of HONO in the gas phase
(50% yields) at low pressure (2torr) with some NO and
adsorbed HNO3. At atmospheric pressure, NO yields
are high (50%) with only a small amount of HONO in
the gas phase and 3-5% adsorbed on the soot. In the
presence of water (50% humidity), significant increases
in HONO formation are observed (740 torr). They con-
clude that a consecutive reaction is occurring:
NO2 )
NO2 )
where the major channel is
HONO (5)
NO
NO2 q- soot/H20(ads) -- HONO + soot(ox.) (7)
and the minor channel is
2NO2 + H20(ads) -- HONO + HNO3 (8)

15,936 LARY ET AL.' CARBONACEOUS AEROSOLS AND ATMOSPHERIC CHEMISTRY
followed by
HONO + soot(reactive) - NO + soot(ox.) (9)
where soot(ox.) denotes oxidized soot.
9 Carbon and Nitric Acid
Rogaski et al. [1997] have studied the reduction of
HNO3 on a carbonaceous surface. They'report hat the
main products are NO, NO2, and H20. Constructing
some overall stoichiometric reactions that are exother-
mic is possible, e.g.,
HNOa +Oa CA}
CA
2HNOa + 2Oa 
HONO + 202 (10)
NO + NO. + H20 + 40. (11)
which are both exothermic at 298 K. No detailed re-
action mechanism is given by Rogaski, but on basis of
the findings of Chughtai et al. [1994], who studied the
interaction of soot with NO2 and N204, it is likely that -
ONO,-NOO, and -NNO2 surface groups will be formed.
Rogaski et al. [1997] studied the reactivity and hydra-
tion properties of amorphous carbon in a low-pressure
Knudsen cell reactor at room temperature (298 K).
They found that the reaction of HNOs had an observed
'7 = 0.038+0.008.
As mentioned for NO. reacting on the surface of CA,
if HONO is formed, then this could be a significant
source of HOx as HONO is readily photolyzed to give
OH. Even if this process is not catalytic and only a frac-
tion of the nitrogen incident on the CA leads to HONO
formation, Figures 6 and 7 show that this could be a sig-
nificant source of OH and HONO. This is the case even
if the CA surface area used is as low as 0.01/m 2 cm -3
and the reaction probability is as low as 10 -3 . This may
explain why large values of HONO [e.g., Harrison and
Kitto, 1994] are often observed in the troposphere in
regions where the numerical models do not predict such
large amounts of HONO. If hydrogen is abstracted from
HNOs then NOs, a nighttime oxidizing agent, could be
released.
Kinoshita [1988] and Mahajan et al. [1978] report
that treatment with HNOs is one of the most effective
ways to make the carbon surface acidic. The treatment
of a carbonaceous surface with HNOs does not affect
their physical morphology but rather alters their surface
chemical properties.
During the Transport and Atmospheric Chemistry
Near the Equatorial Atlantic (TRACE A) campaign,
$myth et al. [1996] observed high NO concentrations
in the upper troposphere, over a region characterized
by intense biomass burning. These workers have con-
cluded that one possible explanation for these results
is that HNO3 is being rapidly recycled back to NOx
in the upper troposphere. In addition, Jacob et al.
[1996] have also come to the same conclusion from the
TRACE A campaign. In the 8-12 km region these work-
1E-13 1E-12 1E-11
3 5 2 3 5 2 3 5
I I I II
No Ca rbonaceous aerosol j//
24 -- 0.01 '2 ore'3 ///'// 24
////
20 20
16 16
12 12
/// Mid-latitudes a?2nox
3 5 2 3 5 2 3 5
1E-13 1E-12 1E-11
Concentration (v,m,r,)
Figure 6. The calculated effect at midlatitudes at equinox on the concentration of OH if the postulated
reaction HNO3 + O3 -' HONO + 202 is included on carbonaceous aerosols with a reaction probability of only
10 '.

LARY ET AL.' CARBONACEOUS AEROSOLS AND ATMOSPHERIC CHEMISTRY 15,937
24--
20--
16-
12--
1E-14 1E-13 1E-12 1E-11
2 3 5 2 3 5 2 3 5 2 3 5
I I I I Illll , I I I Illll I I I I Illll I I I I II
No Carbonaceous aerosol
 0.01r cr3
---- 0. rcr.-3
+ lr cry3
HONO
at equinox
2 3 5 2 3
1E-10
24
-- 20
-- 16
-- 12
8 --8
2 3 5 2 3 5 5
1E-14 1E-13 1E-12 1E-11 1E-10
Concentration (v.m.r.)
Figure 7. The calculated effect at mid-latitudes at equinox on the concentration of HONO if the postulated
reaction HNO3 + O3 -- HONO + 202 is included on carbonaceous aerosols with a reaction probability of only
10 -3 '
ers note that the median conversion of NOx to HN03
from a model is 57 ppt day - whereas the median HN03
levels observed are 59 ppt. Therefore the lifetime of
HNO3 needs to be of the order of a day, whereas its
lifetime with respect to reaction with OH and pho-
tolysis is of the order of 12 days. During biomass
burning events, large quantities of black carbon are
produced, and it is extremely likely that during deep
convective events, elevated levels of black carbon are
brought up to the upper troposphere. The conversion of
HNO3 to HONO/NO2/NO and the conversion of NO2
to HONO/NO on black carbon could well be the missing
mechanism noted by Jacob et al. [1996] and Smyth et
al. [1996] to explain the elevated levels of NO observed
in this region.
The calculations shown in Figures 6 and 7 were made
using the chemical model AUTOCIEM for midlatitude
equinox conditions. No family or photochemical equi-
librium assumptions are made. All species are time
integrated separately with a 15 min time step. The
time integration scheme used is an adaptive timestep
Burlisch-Stoer [Stoer and Burlisch, 1980] scheme specif-
ically designed for integration of stiff systems after Press
et al. [1992]. The time integration package is as accu-
rate as the often used Gear [1971] package but faster.
Photolysis rates are calculated by using full spherical
geometry and multiple scattering as described by Lary
and Pyle [1992ab] after Meier et al. [1982], Nicolet et
al. [1982], and Anderson [1983]. The average photolysis
rate over a model time step is calculated using 10-point
Gaussian quadrature as described by Press et al. [1992].
AUTOCHEM has also been used to perform for the first
time four-dimensional variational analysis of chemical
species [Fisher and Lary, 1995]. The reactions on the
carbonaceous surface are treated in the same way as
reactions on the surface of polar stratospheric clouds
(PSCs) are, a surface area is specified, as well as a re-
action probability. Because this is a preliminary study,
no account is taken of surface ageing or poisoning, al-
though both of these are likely to be important. For this
to be treated accurately more kinetic data is required,
which, as yet is unavailable. The model contains a to-
tal of 56 species with a total of 320 reactions. Fifty-one
species are integrated, namely, o(xn), O(sP), Os, N,
NO, NO2, NOs, N205, HONO, HNOs, HO2NO2, C1,
C12, C10, C1OO, OClO, C1202, C1N02, C1ONO2, HC1,
HOC1, CHsOC1, Br, Br2, BrO, BrONO2, BrONO, HBr,
HOBr, BrC1, H2, H, OH, H02, H202, CHa, CHsO,
CHsO2, CHsOH, CHsOOH, CHsONO2, CHsO2N02,
HCO, HCHO, CH4, CHsBr, CF2C12, CO, N20, CO2,
and H20.
10. Entrainment Within Liquid Drops
Since oxidation of the CA makes them hydrophilic,
it is very likely that CA will become incorporated into
liquid drops, whether these are rain drops or sulphate
aerosols. This is what is observed [e.g., Karcher et al.,
1996]. It is also found that soot embedded in a liquid

15,938 LARY ET AL' CARBONACEOUS AEROSOLS AND ATMOSPHERIC CHEMISTRY
coating containing H2SO4 and HNOa triggers heteroge-
neous freezing of water ice and leads to visible contrails
[Karcher, 1996].
For a reaction to occur on a CA embedded within
a droplet, several mass transfer steps must take place
[Seinfeld, 1986, Hanson et al., 1994]:
1. Gaseous species must diffuse from the bulk gas to
the surface of the droplet. This has a timescale given
by
2
rp (12)
rg = 4Dg
where the radius of the drop is rp and the typical molec-
ular diffusivity Dg is 0.1 cm 2 s -1. So for a droplet with
a radius of 0.1/m the timescale is just 25 ns.
2. Transfer across the gas-liquid interface must take
place. This depends on the Henry's law coefficient and
the accommodation coefficient via the relationship
rp Da(4RTH) 2 - (13) cc
where D is the aqueous phase diffusion coefficient of
the species initially in the gas phase, typically around
10 -5 cm 2 s -1. c is the average speed of gas molecules
given by kinetic theory (c - v/SRT/rM ). H is the
Henry's law coefficient; as H increases, so does rp. If we
take the accommodation coefficient c to have its maxi-
mum value of 1, then for HNO3 rp is of the order of a
second or so, whereas for 03 and NO2, rp is approxi-
mately 10 -15 s.
3. If ionization of the species occurs, it will be rapid
for strong acids such as HNO3.
4. The dissolved species in the aqueous phase diffuse.
The characteristic time for aqueous-phase diffusion is
2
rp (14) Tda -- 71.2Da
so for a radius of 1/m, v-a is just 10 -2 s.
5. If a chemical reaction occurs as these processes
happen sequentially, the overall rate is the rate of the
slowest step. In the case of CA for molecules such as
HNOa with large Henry's law coefficients, the slowest
step is likely to be the transfer across the gas-liquid
interface or the reaction at the CA surface embedded
in the drop. Even so, this is still quite rapid, occurring
on the timescale of a few seconds. For molecules such
as NO2 and Oa with small Henry's law coefficients, the
slowest step is likely to be the reaction at the CA surface
embedded in the drop. For droplets with a radius of less
than a/m, diffusion of the dissolved species within the
aqueous phase is rapid.
Therefore the fact that CA become entrained within
droplets does not necessarily mean that they will no
longer be effective in chemically processing the sur-
rounding air. However, a relevant question which needs
further kinetic investigation is how reactive will the
carbonaceous \"core\" be once entrained within a liquid
aerosol?
11. Summary and Suggestions
Carbonaceous aerosols are readily oxidized in the
atmosphere by both gas and aqueous phase oxidizing
agents. In the process the oxidizing agents are them-
selves reduced, something that in some cases does not
rapidly occur in the atmosphere. e.g., HNOa is reduced
to NO2 on carbonaceous aerosols, something that will
normally occur only slowly by reaction with OH or by
photolysis.
The oxidation of the carbonaceous aerosols leads to
the formation of a range of polar surface groups, such as
carboxylates, which cause the carbonaceous aerosols to
become more hydrophilic and acidic. As a result, they
can act as effective cloud condensation nuclei. Newly
formed carbonaceous aerosols are much more reactive
as they contain more aromatic and aliphatic chains and
sites where carbon is not exerting its full valency. This
oxidation is slower when more H is present. Once the
sites where carbon is not exerting its full valency are oxi-
dized (with the associated release of CO, CO2 and H20)
the less reactive, graphite like, sheets become more ex-
posed. As well as the mass loss Because of the formation
of CO and CO2, there is a mass gain by the formation
of groups such as surface carboxylates.
If HONO is formed on the surface of CA, then this
could be a significant source of HOx as HONO is read-
ily photolyzed to give OH. Even if this process is not
catalytic and only a fraction of the nitrogen incident on
the CA leads to HONO formation, this could be a sig-
nificant source of OH and HONO. This is the case even
if the CA surface area used is as low as 0.01/m 2 cm -3
and the reaction probability is as low as 10 -a. This may
explain why large values of HONO are often observed
in the troposphere in regions where the numerical mod-
els do not predict such large amounts of HONO. If the
hydrogen is abstracted then NO3, a nighttime oxidizing
agent, could be released.
Carbonaceous urfaces absorb hydrocarbon vapors in
preference to water vapor. It is therefore very likely that
many atmospheric interactions are occurring between
hydrocarbons and carbonaceous aerosols. An impor-
tant class of reactions may be with organic peroxides
which can oxidize the carbonaceous aerosols and which
are reduced in the process. Also, if part, or all, of the re-
duction sequences, such as carboxylic acid } aidehyde
> primary alcohol > alkane, ketone } secondary
alcohol  alkane, tertiary alcohol  alkane, occur
on carbonaceous urfaces, it could significantly affect
the oxidizing capacity of the atmosphere. It appears
that the carbonaceous aerosol studies to date have omit-
ted considering the possibility of such reactions taking
place. These reduction sequences clearly include the in-
creasingly high concentrations of the many halogenareal
carbonyl compounds produced by the degradation of

LARY ET AL.: CARBONACEOUS AEROSOLS AND ATMOSPHERIC CHEMISTRY 15,939
hydrochloroflurocarbons (HCFCs), hydroflurocarbons
(HFCs), halons, and chlorofluorocarbons (CFCs).
Acknowledgments. David Lary is an Alon Fellow and
wishes to thank the Government of Israel for its support;
he is also a Royal Society University Research Fellow and
wishes to thank the Royal Society for its support. He also
thanks J.A. Pyle for his support and Tony Cox for very
useful conversations. R. Toumi thanks NEDO (Japan) for
support. The Centre for Atmospheric Science is a joint
initiative of the Department of Chemistry and the Depart-
ment of Applied Mathematics and Theoretical Physics. This
work forms part of the NERC UK Universities Global At-
mospheric Modeling Programme.
References
Akhter, M. S., A. R. Chughtai, and D. M. Smith, Reaction
of hexane soot with NO2/N204, J. Phys. Chem., 88(22),
5334-5342, 1984.
Akhter, M. S., A. R. Chughtai, and D. M. Smith, Aromatic-
ity of elemental carbon (soot) by aC cp/mas and FTIR
spectroscopy, Carbon, 23(5), 593-594, 1985a.
Akhter, M. S., J. R. Keifer, and A. R. Chuhtai, and D. M. Smith, The absorption-band at 1590 cm- in the infrared- ß
spectrum of carbons, Carbon, 23(5), 589-591, 1985b.
Akhter, M. S., Chughtai, A. R., Smith, D. M.,The structure
of hexane soot-I; Spectroscopic studies, Appl. $pectros.,
39(1), 143-153, 1985c.
Akhter, M. S., A. R. Chughtai, and D. M. Smith, The struc-
ture of hexane soot-II; Extraction studies, Appl. Spectros.,
39(1), 154-167, 1985d.
Akhter, M. S., A. R. Chughtai, and D. M. Smith, Deter-
mination of the integrated molar absorptivity for C-ONO
and C-NO2 groups, using FTIR spectroscopy, Appl. $pec-
tros., $0(7), 1068-1069, 1986.
Ammann, M., Heterogeneous interaction of NO2 with car-
bonaceous aerosols at the ppbv level: Adsorption, conver-
sion to NO, production of HONO, paper presented at the
7 th European Symposium on Physico-Chemical behavior
of atmospheric pollutants: The oxidizing capacity of the
troposphere, Venice, Italy, October 2-4, 1996.
Atomann, M., U. Baltensperger, U. K. Bochert, B. Eichler,
H. W. Ggggeler, D. T. Jost, A. Tufter, and A.P. Weber,
Study of HI, HBr and NO2 adsorption on graphite and sil-
ver aerosol-particles using short-lived isotopes, J. Aerosol
$ci., 26(1), 61-70, 1995.
Atomann, M., Baltensperger, U., Scherrer, P., Heteroge-
neous formation of nitrous acid on carbonaceous aerosol
at very low nitrogen dioxide concentrations, Abst. Pap.
Am. Chem. $oc., 21$(Ptl), 6, 1997a.
Atomann, M., et al., Heterogeneous interaction of NO2 with
carbonaceous aerosols at the ppb level: absorption, con-
version to NO, production of HONO, in The Oxidizing
Capacity of the Troposphere, edited by B. Larsen, B.
Versino, and G. Angeletti , EUR 178œ, pp. 683-687, Eur.
Comm. 1997b.
Atomann, M., M. Kalberer, D.T. Jost, L. Tobler, E. Rossler,
D. Piguet, H.W. G/iggeler, and U. Baltensperger, Hetero-
geneous production of nitrous acid on soot in polluted air
masses, Nature, 395(6698), 157-160, 1998.
Anderson, D. E., The troposphere to stratosphere radiation
field at twilight: A spherical model, Planet. Space $ci.,
$1(12), 1517-1523, 1983.
Bastick, M., G. Dupupet, M. Marlot, J. M. Perrot, and J.
Weber, Carbon, 15(25), 1977.
Bekki, S., The possible role of aircraft-generated soot in
the middle latitude ozone depletion, J. Geophys. Res.,
10œ(D9), 10751-10758, 1997.
Berezin, G.I., G. A. Petukhova, N. S. Polyakov, and M.
L. Gubkina, Thermodynamic investigation of the state of
water adsorbed by carbonaceous adsorbents Russ. Chem.
Bull., 6(3), 443-447, 1997.
Blake, D. F., and K. Kato, Latitudinal distribution of black
carbon soot in the upper troposphere and lower strato-
sphere, J. Geophys. Res., 100(D4), 7195-7202, 1995.
Chughtai, A. R., S. A. Gordon, and D. M. Smith, Kinet-
ics of the hexane soot reaction with NO2/N204 at low
concentration, Carbon, 32(3), 405-416, 1994.
Cope, M. J., Physical and chemical properties of coalified
and charcoalified phytoclasts from some British Meso-
zoic sediments: An organic geochemical approach to pa-
leobotany. Paper presented at the 9 th International Meet-
ing on Organic Geochemistry, Newcastle-upon-Tyne, Eng-
land, 1979.
de Vooys, F., The pore size distribution of activated carbon,
in Activated carbon: A Fascinating Material, edited by A.
Capelle, F. de Vooys, 13, Norit N.V., Amersfoort, The
Netherlands, 1983.
Fisher, M., and D. J. Lary, Lagrangian four dimensional
variational assimilation of chemical species, Q. J. R. Me-
teorol. Soc., 1œ1(527), 1681-1704, 1995.
Gay, A. J. , R. F. Littlejohn, and P. J. Vanduin, Formation
of carbonaceous cenospheres during fluidized-bed combus-
tion of bituminous coals, Fuel, 6œ(10), 1224-1226, 1983.
Gay, A. J. , R. F. Littlejohn, and P. J. Vanduin, Studies of
carbonaceous cenospheres from fluidized-bed combustors,
Sci. Total Environ., 36, 239-246, 1984.
Gear, C.W., Numerical Initial Value Problems in Ordinary
Differential Equations, Englewood Cliffs, N.J.: Prentice-
Hall, chap. 9, 1971.
Gereck, A., A. Thielmann, L. Gutzwiller, and M.J.
Rossi, The chemical kinetics of HONO formation result-
ing from heterogeneous interaction of NO2 with flame
soot, Geophys. Res. Lett., œ5(13), 2453-2456, 1998.
Goldberg, E. D., Black carbon in the environment: Proper-
ties and distribution, John Wiley, New York, 1985.
Goss, K. U., and S. J. Eisenreich, Sorption of volatile or-
ganic compounds to particles from a combustion source
at different temperatures and relative humidities, A tmos.
Environ., $1(17), 2827-2834, 1997.
Gustafsson, O., F. Haghseta, C. Chan, J., MacFarlane, and
P.M. Gschwend, Quantification of the dilute sedimentary
soot phase: Implications for PAH speciation and bioavail-
ability, Environ. Sci. Technol., $1(1), 203-209, 1997.
Hanson, D.R., A.R. Ravishankara, and S. Solomon, Hetero-
geneous reactions in sulfuric-acid aerosols - a framework
for model-calculations, J. Geophys. Res., 99(D2), 3615-
3629, 1994.
Harrison, R.M., and A.M.N. Kitto, Evidence for a sur-
face source of atmospheric nitrous-acid, Atrnos. Environ.,
28(6), 1089-1094, 1994.
Hauglustaine, D. A., B. A. Ridley, S. Solomon, P.G. Hess,
and S. Madronich, HNOa/NO ratio in the remote tropo-
sphere during MLOPEX-2: Evidence for nitric-acid re-
duction on carbonaceous aerosols, Geophys. Res. Lett.,
23(19), 2609-2612, 1996.
Jacob, D. J., et al., Origin of ozone and NO in the tropical
troposphere- a photochemical analysis of aircraft obser-
vations over the south-atlantic basin, J. Geophys. Res.,
101(D19), 24235-24250, 1996.
Jankowska, H., A. Swiatkowski, and J. Choma, Active Car-
bon, Ellis Hotwood, Chichester, England, 1991.
Kalberer, M., Tabor, K., Ammann, M., Parrat, Y., Wein-
gartner, E., Piguet, D, Rosslet, E., Jost, D. T., Tufter,
A., G/iggeler, H. W., Baltensperger, U., Heterogeneous

15,940 LARY ET AL.: CARBONACEOUS AEROSOLS AND ATMOSPHERIC CHEMISTRY
chemical-processing of xaNO. by monodisperse carbon
aerosols at very-low concentrations, J. Phys. Chem.,
100(38), 15487-15493, 1996.
Karcher, B., Aircraft-generated aerosols and visible con-
trails, Geophys. Res. Left., 23(15), 1933-1936, 1996.
Karcher, B., T. Peter, U. M. Biermann, and U. Schumann,
The initial composition of jet condensation trails, J. At-
mos. Sci., 53(21), 3066-3083, 1996.
Keifer, J.R., M. Novicky, M.S. Akhter, A.R. Chughtai, and
D.M. Smith, The nature and reactivity of the elemen-
tal carbon (soot) surface as revealed by fourier-transform
infrared (FTIR) spectroscopy, Proc. Soc. Photo-Opt. In-
strum. Eng., 289, 184-188, 1981.
Kinoshita, K., Carbon: Electrochemical and Physicochemical
Properties, John Wiley, New York, 1988.
Konopka, P., and W. Vogelsberger, Kohler equation for fi-
nite systems: A simple estimation of possible condensa-
tion mechanisms in aircraft contrails, J. Geophys. Res.,
102(D13), 16057-16064, 1997.
Kotzick, R., U. Panne, and R. Niessner, Changes in conden-
sation properties of ultrafine carbon particles subjected to
oxidation by ozone, J. Aerosol Sci., 28(5), 725-735, 1997.
Lammel, G., and T. Novakov, Water nucleation properties of
carbon-black and diesel soot particles, Atmos. Environ.,
29(7), 813-823, 1995.
Lary, D. J. and J. A. Pyle, Diffuse radiation, twilight and
photochemistry- I, J. Atmos. Chem., 13, 373-392, 1992a.
Lary, D. J. and J. A. Pyle, Diffuse radiation, twilight and
photochemistry-II, J. Atmos. Chem., 13, 393-406, 1992b.
Lary, D. J., R. Toumi, A.M. Lee, M. J. Newchurch, M.
Pirre, and J. B. Renard, Carbon aerosols and atmospheric
photochemistry, J. Geophys. Res., Atmospheres, 102(D3),
3671-3682, 1997.
Long, F.J., and K.W. Sykes, The mechanism of the steam-
carbon reaction, Proc. Roy. Soc. of London, Sec. A, 193,
377-399, 1948.
Mahajan, O. P., A. Youserf, and P. L. Walker, Separation
Sci. Technol., 13, 487, 1978.
Mantell, C. L., Carbon and Graphite Handbook, Inter-
science, New York, 1968.
Mattson, J. S., H. B., Mark, Activated Carbon, Marcel
Dekker, New York, 1971.
McDow, S.R., M.S. Jang, Y. Hong, and R.M. Kamens, An
approach to studying the effect of organic composition
on atmospheric aerosol photochemistry, J. Geophys. Res.-
Atmospheres, 101(D14), 19593-19600, 1996.
Medalia, A. I., Rivin, D., Forms of particulate carbon in
soot and carbon-black, Carbon, 20(2), 152, 1982.
Medalia, A. I., and D. Rivin, Particulate carbon and other
components of soot and carbon-black, Carbon, 20(6), 481-
ß 492, 1982.
Medalia, A. I., D. Rivin, and D.R. Sanders, A comparison
of carbon-black with soot, Rubber Chem. Technol., 55(1),
257-258, 1982.
Medalia, A. I., D. Rivin, and D.R. Sanders, A comparison of
carbon-black with soot, Sci. Total Environ., 31(1), 1-22,
1983.
Meier, R. R., D. E. Anderson, and M. Nicolet, The radiation
field in the troposphere and stratosphere from 240 nm to
1000 nm: General analysis, Planet. Space $ci., 30(9), 923-
933, 1982.
Nicolet, M., R. R. Meier, and D. E. Anderson, The radiation
field in the troposphere and stratosphere from 240 nm to
1000 nm: Numerical analysis, Planet. Space $ci., 30(9),
935-983, 1982.
Nob, J.S., and J.A. Schwarz, Effect of HNOa treatment on
the surface-acidity of activated carbons, Carbon, 28(5),
675-682, 1990.
Pires, M., H. VandenBergh, and M.J. Rossi, The heteroge-
neous formation of N.O over bulk condensed phases in
the presence of SO2 at high humidities, J. Atmos. Chem.,
25(3), 229-250, 1996.
Press W.H., S.A. Teukolsky, W.T. Vetterling, and B.P. Flan-
nery, Numerical Recipes in Fortran: The Art of Scien-
tific Computing, 2 \" edition, Cambridge Univ. Press, New
York 1992.
Rogaski, C.A., D.M. Golden, and L.R. Williams, Reactive
uptake and hydration experiments on amorphous carbon
treated with NO., SO., O3, HNO3, and H2804, Geophys.
Res. Left., 2j(4), 381-384, 1997.
Seinfeld, J. H., Atmospheric Chemistry and Physics of Air
Pollution, Wiley Interscience, 1986.
Sergides, C.A., J.A., Jassim, A.R., Chughtai, and D.M.
Smith, The structure of hexane soot, 3,Ozonation studies,
Appl. Spectros., dI(3), 482-492, 1987.
Smisek, M., and S. Cerny, Active Carbon, Elsevier, New
York, 1970.
Smith, D.M., and A.R. Chughtai, The surface-structure and
reactivity of black carbon, Colloids Surf. A, 105(1), 47-77,
1995.
Smith, D.M., and A.R. Chughtai, Reaction-kinetics ofozone
at low concentrations with n-hexane soot, J. Geophys.
Res., 101(D14), 19607-19620, 1996.
Smith, D.M., and A.R. Chughtai, Photochemical effects in
the heterogeneous reaction of soot with ozone at low con-
centrations, J. Atmos. Chem., 26(1), 77-91, 1997
Smyth, S. B., et al., Factors influencing the upper free tro-
pospheric distribution of reactive nitrogen over the South
Atlantic during the TRACE A experiment J. Geophys.
Res., 101(D19), 24165-24186, 1996.
Stoer, J., and R. Bulirsch, Introduction to Numerical Anal-
ysis, Springer-Verlag, New York, chap. 7, 1980.
Tabor, K., L. Gutzwiller, and M. J. Rossi, The heteroge-
neous interaction of NO. with amorphous-carbon, Geo-
phys. Res. Left., 20(14), 1431-1434, 1993.
Tabor, K., L. Gutzwiller, and M. J. Rossi, The heteroge-
neous interaction of NO. with amorphous-carbon at am-
bient temperature, J. Phys. Chem., 98(24), 6172-6186,
1994.
Thlibi, J., and J. C. Petit, A study of the NOy/Soot in-
teraction in the temperature range 303-1223 K, paper
presented at International Scientific Colloquium, Cologne,
Germany, April 18-20, 1994.
van der Plas, Th., The texture and the surface chemistry of
carbons, in Physical and Chemical Aspects of Adsorbents
and Catalysts, chap. 9, pp. 425-469, Academic Press, San
Diego, Calif., 1970.
van Driel, J., Surface chemistry of activated carbon, in
Activated carbon: A Fascinating Material, edited by
A. Capelie, F. de Vooys, Norit N.V., Amersfoort, The
Netherlands, 1983.
D. J. Lary, Department of Geophysics and Planetary Sci-
ences, Tel Aviv University, 69978, Tel Aviv, Israel.
D. E. Shallcross, Centre for Atmospheric Science, Depart-
ment of Chemistry, Cambridge University, Cambridge, CB2
1EW, England, U.K.
R. Toumi, Department of Physics, Imperial College, Lon-
don, SW7 2BZ, England, U.K.
(Received June 11, 1998; revised December 1, 1998;
accepted December 4, 1998.)