Uptake of methanol vapor in sulfuric acid solutions

Article abstract of DOI:10.1021/jp001707a

The uptake of gas-phase methanol by liquid sulfuric acid has been investigated over the composition range of 40-85 wt% H₂SO₄ and the temperature range of 210-235 K. Laboratory studies were performed with a flow-tube reactor coupled to an electron-impact ionization mass spectrometer to detect trace gases. While reversible uptake was the primary mechanism at low acid concentrations, an irreversible reaction between methanol and sulfuric acid at low temperatures, forming methyl hydrogen sulfate and dimethyl sulfate, was observed at all concentrations. At compositions >65 wt% H₂SO₄, more than 90% of uptake was found to be reactive. On the basis of the uptake data and the calculated liquid-phase diffusion coefficients, the product of the effective Henry's law constant (H*) and the square root of the overall liquid-phase reaction rate (k₁) was calculated as a function of acid concentration and temperature. The results suggest that the reaction with sulfuric acid forming methyl hydrogen sulfate and dimethyl sulfate is the dominant loss mechanism of methanol and that the oxidation of methanol is only a minor source of hydroxyl radicals in the upper troposphere.

Full text of DOI:10.1021/jp001707a

J. Phys. Chem. A 2001, 105, 1411-1415
1411
ARTICLES
Uptake of Methanol Vapor in Sulfuric Acid Solutions†
Sean M. Kane and Ming-Taun Leu*
Earth and Space Sciences DiVision, Jet Propulsion Laboratory, California Institute of Technology,
Pasadena, California 91109
ReceiVed: May 5, 2000; In Final Form: July 20, 2000
The uptake of gas-phase methanol by liquid sulfuric acid has been investigated over the composition range
of 40-85 wt % H SO and the temperature range of 210-235 K. Laboratory studies were performed with a
2
4
flow-tube reactor coupled to an electron-impact ionization mass spectrometer to detect trace gases. While
reversible uptake was the primary mechanism at low acid concentrations, an irreversible reaction between
methanol and sulfuric acid at low temperatures, forming methyl hydrogen sulfate and dimethyl sulfate, was
observed at all concentrations. At compositions >65 wt % H
reactive. On the basis of the uptake data and the calculated liquid-phase diffusion coefficients, the product of
the effective Henry’s law constant (H*) and the square root of the overall liquid-phase reaction rate (k ) was
2 4
SO , more than 90% of uptake was found to be
l
calculated as a function of acid concentration and temperature. The results suggest that the reaction with
sulfuric acid forming methyl hydrogen sulfate and dimethyl sulfate is the dominant loss mechanism of methanol
and that the oxidation of methanol is only a minor source of hydroxyl radicals in the upper troposphere.
Introduction
the range of 200-240 K, sulfate aerosols are mainly composed
8,9
of between 40 and 80 wt % H2SO4. To understand the possible
importance of this system, we have examined the uptake and
reactivity of methanol in liquid sulfuric acid under the temper-
ature and acid concentration ranges of the upper troposphere
and lower stratosphere.
Oxygenated hydrocarbons, particularly acetone, play an
important role in atmospheric chemistry by contributing the
production of HOx free radicals and consequently increasing
the formation of ozone in the upper troposphere.1,2 Although
methanol is considered to be of secondary importance in these
aspects, the concentration of methanol has been found to be as
high as 700 ppt at 5-10 km.1 Hence, it is important to
investigate the production and loss mechanisms of methanol in
the atmosphere. Sources of methanol in the atmosphere include
secondary reactions of hydrocarbons, biomass burning, and
,3
Experimental Methods
Apparatus. Uptake measurements in this experiment were
performed using a fast flow-tube reactor coupled with an
electron-impact ionization mass spectrometer, which has previ-
4
direct biogenic and anthropogenic emissions. On the other hand,
the photolytic loss of methanol is believed to be insignificant.
1
0,11
ously been described in detail.
The reactor made of Pyrex
5
tubing was 25 cm long with an inner diameter of 1.8 cm. The
bottom of the reactor was recessed to form a trough (1.9 cm
wide and 0.3 cm deep) which held the liquid sulfuric acid. The
temperature during the experiments was controlled by flowing
cold methanol through the outer jacket of the reactor. Helium
carrier gas was admitted through a sidearm inlet, while methanol
in another helium carrier was added by a movable Pyrex injector.
Pressures in the reactor were monitored by a high-precision
capacitance manometer (MKS Instruments, Model 390 HA, 10
Torr full scale). Typically, a total pressure of 0.47 Torr was
used. Methanol was monitored at the fragmentation peak of m/e
Moreover, the reaction with ice particles inside cirrus clouds is
also very slow. It has been thought that the only significant
loss mechanism for methanol in the upper troposphere is the
reaction with hydroxyl radicals. However, the heterogeneous
reaction of methanol in liquid sulfuric acid has not, to this point,
been considered.
6
Sulfate aerosols are thought to be the dominant form of
aerosol in the upper troposphere. Very recently, organic acids
and hydroxymethanesulfonic acid (HMSA) have been identified
7
in situ in aerosols at altitudes of 5-19 km. These organic-
containing aerosols are particularly more pronounced in the
tropics because of convection from the troposphere. Thus, it is
intriguing to understand their formation mechanism, for ex-
ample, the interaction of gas-phase organic compounds with
liquid sulfuric acid.
)
31 amu for higher detection sensitivity.
Materials. Methanol (Fisher Scientific; 99.9%, reagent grade)
was used as received without further purification, and its purity
was confirmed by mass spectroscopy. A sample vial containing
the methanol was placed in a methanol/dry ice bath to control
the concentration of methanol inside the reactor. The partial
pressures of methanol in these experiments were in the range
In the upper troposphere, where ambient temperatures are in
†
Part of the special issue “Harold Johnston Festschrift”. Dedicate to
-
4
-6
of 1.7 × 10 to 1.7 × 10 Torr, depending on the type of
Professor Harold Johnston for his contributions to atmospheric chemistry.
*
Corresponding author.
experiment. Helium (Matheson Gas Company; 99.999%, ultra-
1
0.1021/jp001707a CCC: $20.00 © 2001 American Chemical Society
Published on Web 09/15/2000

1412 J. Phys. Chem. A, Vol. 105, No. 9, 2001
Kane and Leu
high purity grade) was used as shipped for both the methanol
carrier gas and the main flow gas. Sulfuric acid solutions of
known composition were prepared by diluting 96.2 wt % H2-
SO4 (J. T. Baker Chemical Co.) with distilled water. To ensure
the constant composition of H2SO4 over a long period of time,
the helium flow gas was humidified in a vessel with the same
H2SO4 composition and temperature as in the reaction cell.
Additionally, the acid reservoir was changed frequently, and
the composition of the acid was checked before and after each
set of experiments by determining the density of the acid
12
solutions as an expedient method to check H2SO4 composition.
Data Analysis. The uptake coefficient was determined from
the methanol data according to the equation1
0,11
γ ) (4k /ω)(V/S)
(1)
c
where V is the volume of the reaction cell, S is the geometric
area of the acid reservoir, ω is the mean thermal speed of the
molecule, and kc is the corrected first-order rate coefficient. This
rate coefficient is related to the fractional change of the gas-
10,11
phase concentration of methanol calculated by
2
k ) k (1 + k D /V )
(2)
c
g
g
g
where Dg is the diffusion coefficient of methanol in He (Dg )
2
-1
4
24/p Torr cm s at 295 K) and V is the average flow velocity.
1.75
A temperature dependence of T
was used for estimating Dg
at other temperatures. The observed first-order rate, kg, is
calculated by the equation
Figure 1. Uptake and desorption of methanol at 213.1 K, followed
2 4
by heating to room temperature, for 40, 75, and 85 wt % H SO (chosen
as most representative data). Heating curves for each experiment are
indicated by the dotted lines, scaled to the right axis. The 40 wt %
k ) (F /V)(∆n/n)
(3)
g
g
where Fg is the carrier gas flow rate and ∆n/n is the fractional
change in the gas-phase concentration of methanol after
exposure to sulfuric acid by moving the sliding Pyrex injector.
Since a symmetrical, cylindrical tube was not used for the uptake
coefficient measurements, correction for radial gas-phase dif-
fusion was not taken into account in the determination of kg
because this correction was considered to be rather imprecise.
However, we estimate that this correction is very small, less
than 10%.
Under conditions where less than 10% of methanol is
reversibly absorbed (for example, the experiments using g65
wt % H2SO4, see the next section), the observed uptake
coefficient can be approximately represented by the reactive
uptake coefficient
H
2
SO
trace amounts of methanol desorbed at 75 wt %, and no desorption
occurred at 85 wt % H SO at 213.1 K. Once the methanol flow was
4
shows significant low-temperature methanol desorption, only
2
4
shut off and the baseline level was achieved, the samples were heated
to room temperature.
acid
-
8
1/2
7
.4 × 10 (κsolvent)
c )
(6)
0.6
VA
where κsolvent is a solvent-dependent empirical factor (κsolvent )
13
64) and VA is the Le Bas molar volume of solute A (methanol)
3
15
at its normal boiling temperature (VA ) 37 cm /mol). We
-
8
calculated c to be 6.78 × 10 for methanol in H2SO4. In
general, Dl decreases with decreasing temperature and increasing
acid concentration. The square root of Dl is used in the
4
RTH*
ω
γ )
k1D1
(4)
x
1
/2
where R is the gas constant (0.082 L atm mol K-1), T is
-
1
determination of H*kl , and thus, the error associated with the
procedure of Dl estimation is about 10-20%.
temperature, H* is the Henry’s Law solubility constant, Dl is
the liquid diffusion constant, and kl is the overall rate constant
for liquid-phase reactions. Using eq 4, we are able to derive
Results and Discussion
1
/2
H*kl from γ. The details are given in the next section.
Liquid-Phase Diffusion Coefficients. The determination of
Methanol is found to exhibit some level of irreversible uptake
at all the concentrations examined, and it requires a long time
to reach equilibrium at concentrations dominated by reversible
the liquid-phase diffusion coefficient was performed using the
method suggested by Klassen et al.13 The diffusion coefficient
-4
uptake. A relatively high pressure of methanol (1.7 × 10 Torr)
was used in these experiments to facilitate the rate at which
equilibrium of methanol with liquid sulfuric acid could be
reached. A representative set of methanol uptake experiments
is shown in Figure 1. The uptake and desorption of methanol
is performed at 213.1 K for 40-85 wt % H2SO4. While 40 wt
% H2SO4 shows significant low-temperature methanol desorp-
tion, only trace amounts of methanol were observed at 75 wt
of methanol in liquid sulfuric acid is given by
D ) cT/η
(5)
1
where T is the temperature, η is the viscosity of sulfuric acid,
and c is a constant determined from the molar volume of
14
methanol (Le Bas additivity rules). Wilke and Chang empiri-
cally determined the value of c for the species in liquid sulfuric

Methanol Vapor in Sulfuric Acid Solutions
J. Phys. Chem. A, Vol. 105, No. 9, 2001 1413
2 4
Figure 3. Mass spectrum from mixing 15 mL of H SO with 10 mL
of methanol in a vacuum. Peaks at m/e ) 112 and 97 indicate the
formation of methyl hydrogen sulfate (MHS) under these conditions.
Additional peaks may be attributed to either MHS or sulfuric acid.
dependent reactive component for methanol absorption. As acid
concentration increases, the low-temperature (nonreactive up-
take) methanol fraction decreases from ∼0.5 at 40 wt % H2-
SO4 to 0 at 85 wt % H2SO4. Thus, under constant temperature
conditions, the physical uptake is approximately 50% at 40 wt
%
H2SO4. The desorbing methanol fraction, from heating to
room temperature (lower panel), increases from 0.25 at 40 wt
%
to a peak (∼0.6) at 75 wt %, followed by a steep drop at
higher concentrations.
The most likely reaction of methanol and sulfuric acid would
be the formation of methyl hydrogen sulfate (MHS)
Figure 2. Low-temperature (a) and high-temperature (b) fractions of
desorbing methanol at 213.1 K on a variety of sulfuric acid concentra-
tions. As acid concentration increases, the low-temperature (nonreactive
16-20
CH OH + H SO f CH SO H + H O
(7)
3
2
4
3
4
2
uptake) desorption fraction decreases from ∼0.5 at 40 wt % H
2 4
SO to
0
at 85 wt %. High-temperature (thermally reversible reactive uptake)
While this product was not directly observable under the flow-
tube experimental conditions, a series of experiments was
conducted by mixing liquid methanol with H2SO4 from 40 to
85 wt % at room temperature and monitoring the reaction
products by mass spectrometry. Figure 3 shows the results for
desorption increases from 0.25 for 40 wt % to a peak at 75 wt %,
followed by a steep drop at higher concentrations. The steep drop and
the total desorbing fraction of below 0.75 suggest a second reaction
that is irreversible under the experimental conditions.
7
5 wt % at 295 K. Peaks at m/e ) 112 (the parent mass of
%
, and no desorption occurred at 85 wt % H2SO4 at 213.1 K.
methyl hydrogen sulfate) and 97 and 81 (fragment peaks for
loss of methyl and methoxy, respectively) demonstrate that MHS
is a possible product from reaction at low temperature. The
change in the curve at compositions >75 wt % suggests that a
second reactive mechanism occurs at higher acid concentrations,
as shown in Figure 2b. This reaction is not readily reversible at
room temperature, and it is most likely the further reaction of
methyl hydrogen sulfate with methanol to produce dimethyl
To further identify the components in the sulfuric acid following
methanol exposure, we heated samples to room temperature after
the methanol flow was shut off and the baseline level was
achieved. The heating curves for each experiment are indicated
by the dotted lines, with the temperature on the right axis. Only
7
5 wt % H2SO4 shows significant high-temperature desorption
upon heating, although all samples did exhibit some methanol
desorption. While little methanol desorption at high temperature
is expected at 40 wt % H2SO4 since the majority of exposed
methanol is reversibly absorbed, the low yield at 85 wt % H2-
SO4 suggests a reaction, not observed at 75 wt %, that is not
thermally reversible at these conditions.
To better identify the reversible and reactive components of
methanol uptake by H2SO4, we determine the desorbing frac-
tions of methanol both at 213.1 K and following heating to
room temperature for a range of acid concentrations (40-85
wt %, see Figure 2). The desorbing fractions of methanol are
shown for low temperatures (upper panel) and high temperatures
21,22
sulfate
CH SO H + CH OH f (CH ) SO + H O
(8)
3
4
3
3 2
4
2
Low vapor pressure prevents the direct observation of this
product even in the mixed-liquid experiments. This reaction
mechanism is in good agreement with the chemistry previously
1
6-20
observed.
As discussed in Experimental Methods, the fractional change
in the methanol signal can be used to determine the uptake
coefficient for methanol on sulfuric acid (eqs 1-3). For acid
concentrations >65 wt %, all uptake can be considered
(lower panel). The low-temperature data support an acid-

1414 J. Phys. Chem. A, Vol. 105, No. 9, 2001
Kane and Leu
Figure 5. Determination of H*k 1
/2
l
2 4
for 65-80 wt % H SO . The
combined product is found to increase with acid concentration and
decrease as a function of temperature. See text for details.
Figure 4. Calculated values for γ plotted against temperature for 65-
0 wt % H SO . γ values range from 0.01 to 0.023 in the range of
8
2
4
concentrations and temperatures observed. Solid lines are linear
regression fits to the data.
the evidence that the solubility of peroxyacetyl nitrate (PAN)
25
in liquid sulfuric acid is nearly equal to that in water. The
irreversible. As the reactive uptake does not exhibit a recovery
curve (similar to the profile of 40 wt % in Figure 1), the
fractional change in the methanol signal is acquired by a
stepwise increase of the distance of the glass injector inside the
reaction cell allowing for sufficient time between each step to
collect a good signal.23 To better simulate atmospheric condi-
tions in these experiments, we reduced the partial pressure of
2
6-28
equation for H* of methanol in water
is
ln(H*) ) ln[KH(T0)] + (∆H /R)(1/T
- 1/T)
(9)
0
0
where ln[KH(T )] ) 5.39, ∆H0/R ) -4900 K and T0 ) 298.15
0
-
6
K. The overall rate constant for methanol in H2SO4 based on
this assumption is shown in Figure 6. The reaction rate increases
as acid concentration increases in the temperature range. The
rates for 65 and 75 wt % H2SO4, which exhibited thermal
reversibility as shown in Figure 2, increase with increasing
temperature. The data collected at 80 wt % H2SO4, however,
show a nearly constant reaction rate with respect to temperature.
This also supports the further reactions of methanol and sulfuric
acid to dimethyl sulfate, reactions 7 and 8, in this acidity range.
methanol to approximately 2 × 10 Torr. For each of three
acid compositions, the value of γ was determined over a
temperature range of 210-235 K. The results are plotted in
Figure 4, with linear regressions of these data and each curve
plotted separately. While the range of γ values is clustered
between 0.012 and 0.023 in the range of concentrations and
temperatures, a general trend can be observed. At very low
temperatures (∼210 K), the values for γ are the same regardless
of acid concentration. As temperature increases, however, γ
decreases for 65 wt % H2SO4, while it increases for 75 and 80
wt %. This also suggests a change in the reaction mechanism
involved in uptake, as previously discussed.
Atmospheric Implications. In Figure 6, the overall reaction
-1
rate, kl, in the range of 0.1-10 s suggests that it is potentially
important in the upper troposphere. We adopt 10 km as a
For concentrations greater than 65 wt %, almost all uptake
is reactive, and thus, the atmospherically important values of
H* and kl can be determined from eq 4. Initially, these values
are presented as a product. Figure 5 shows the results for the
calculation of H*kl1 for acid concentrations between 65 and
representative altitude. Using 65 wt % and 220 K, we calculate
1/2
that the diffusoreactive length, l ) (Dl/kl) , is about 7 µm by
-8
2
-1
assuming Dl ) 5 × 10 cm /s and kl ) 0.1 s . Since the size
of stratospheric aerosols is typically submicron or smaller, we
conclude that reactions 7 and 8 occur throughout the volume
of aerosols and that a portion of gas-phase methanol is
eventually converted into MHS and dimethyl sulfate.
/2
8
0 wt %. The combined product is found to increase with acid
concentration and decrease as a function of temperature.
To our knowledge, the effective Henry’s law constant has
not been measured for acid compositions between 65 and 80%.
The determination of kl can be made by assuming the Henry’s
law constant of methanol to be approximately that of the value
for methanol in water (based on the similar pKa values of
methanol and water).24 The assumption is further supported by
To illustrate the atmospheric importance of the reaction of
methanol in sulfuric acid, we need to detail the loss mechanisms
and their reaction rates under typical atmospheric conditions.
As noted in the Introduction, the photodissociative loss of
5
methanol has been found to be insignificant. Reactions with

Methanol Vapor in Sulfuric Acid Solutions
J. Phys. Chem. A, Vol. 105, No. 9, 2001 1415
Acknowledgment. This research was performed at the Jet
Propulsion Laboratory, California Institute of Technology, under
contract with the National Aeronautics and Space Administration
(NASA). The authors thank the members of the JPL Chemical
Kinetics and Photochemistry Group for helpful discussions and
the reviewers for useful comments.
References and Notes
(1) Singh, H. B.; Kanakidou, M.; Crutzen, P. J.; Jacob, D. J. Nature
1995, 378, 50.
(
(
(
(
2) Wennberg, P. O.; et al. Science 1998, 279, 49.
3) Singh, H. B.; et al. J. Geophys. Res. 1994, 99, 1805.
4) Holzinger, R.; et al. Geophys. Res. Lett. 1999, 104, 1705.
5) Harrison, A. J.; Cederholm, B. J.; Terwilliger, M. A. J. Chem. Phys.
1959, 30, 355.
(6) Zondlo, M. A.; Warshawsky, M. S.; Tolbert, M. A. Paper Number
A12C-01; AGU Fall Meeting in San Francisco, American Geophysical
Union, 1998.
(7) Murphy, D. M.; Thomson, D. S.; Mahoney, M. J. Science 1998,
282, 1664.
(8) Scientific Assessment of Ozone Depletion: 1994;Global Ozone
Research and Monitoring Project-Report No. 37; World Meteorological
Organization: Geneva, 1994; Chapter 3.
Figure 6. Rate constant for methanol uptake by 65, 75, and 80 wt %
(9) Steele, H. M.; Hamill, P. J. Aerosol Sci. 1981, 12, 517.
H
2
SO
4
. While the curves for 65 and 75 wt % show similar temperature
for 80 wt % is independent of temperature, indicating
(
10) Leu, M.-T.; Timonen, R. S.; Keyser, L. F.; Yung, Y. L. J. Phys.
dependence, k
l
Chem. 1995, 99, 13203.
that the overall reaction mechanism is different than that observed for
the lower acid concentrations.
(11) Choi, W.; Leu, M.-T. J. Phys. Chem. A 1998, 102, 7618.
(12) Handbook of Chemistry and Physics, 65th ed.; Weast, R. C., Ed.;
CRC Press: Boca Raton, FL, 1984.
6
liquid water or ice are also found to be very slow. In addition
(13) Klassen, J. K.; Hu, Z.; Williams, L. R. J. Geophys. Res. 1998, 103,
to the reaction mechanism we have suggested, the only other
significant loss mechanism for methanol is reaction with the
OH radical. The estimated reaction rate for OH + CH3OH f
CH3O + H2O at 10 km and 220 K is k(OH + CH3OH) × [OH]
16197.
(
14) Wilke, C. R.; Chang, P. AlChE J. 1995, 1, 264.
15) Reid, R. C.; Prausnitz, J. M.; Poling, B. E. The Properties of Gases
(
and Liquids; McGraw-Hill: New York, 1987.
16) Liler, M. Reaction Mechanisms in Sulphuric Acid and Other Strong
Acid Solutions; Academic Press: New York, 1971.
17) Vinnik, M. I.; Kislina, I. S.; Kitaigorodskii, A. N.; Nikitaev, A. T.
-
7
-1
)
1.3 × 10 s , and the lifetime is nearly 80 days. The rate
(
coefficient is taken from the recommendation of the NASA Data
Evaluation Panel Report,28 and the diurnally averaged OH
(
5
3 28
concentration is assumed to be 3 × 10 molecules/cm .
To calculate the loss rate of methanol due to the reaction
with sulfuric acid, we estimate the first-order rate to be /4γωA.
The γ value is adopted from the data for 65 wt % H2SO4
reported in the previous section. We assume the surface-area
density of sulfate aerosol at an altitude of 10 km to be about 2
Bull. Acad. Sci USSR 1986, 35, 2447.
(18) Almstead, N.; et al. Tetrahedron Lett. 1987, 28, 1627.
1
(19) Hanson, L. D.; White, V. F.; Eatough, D. J. EnViron. Sci. Technol.
986, 20, 872.
1
(20) Japar, S. M.; Wallington, T. J.; Andino, J. M.; Ball, J. C. EnViron.
Sci. Technol. 1990, 24, 313.
-
7
2
3
-8
×
10 cm /cm for volcanic-perturbed conditions and 1 × 10
(21) Deno, N. C.; Wisotsky, M. J. J. Am. Chem. Soc. 1963, 85, 1735.
(22) Deno, N. C.; Newman, M. S. J. Am. Chem. Soc. 1950, 72, 3852.
2
3
29,30
cm /cm for quiescent conditions.
By using the mean thermal
velocity for methanol, ω, at 220 K, we estimate the rate to be
(
23) Kane, S. M.; Timonen, R. S.; Leu, M.-T. J. Phys. Chem. A 1999,
-5
-1
-6
4
× 10
s
under perturbed volcanic conditions and 2 × 10
103, 9259.
-
1
s
under quiescent conditions. Thus, the reaction rate with
(24) Fox, M. A.; Whitesell, J. K. Organic Chemistry, 2nd ed.; Jones
sulfuric acid is significantly greater than the reaction rate with
hydroxyl radicals. Under quiescent conditions, the lifetime of
methanol is about 5 days, as compared to the average lifetime
of 16 days reported by Singh et al.1
and Bartlett: Sudbury, MA, 1997.
(25) Zhang, R.; Leu, M.-T. J. Geophys. Res. 1997, 102, 8837.
(
26) Danckwerts, P. V. Gas-Liquid Reactions; McGraw-Hill: New
York, 1970.
(
27) Snider, J. R.; Dawson, G. A. J. Geophys. Res. 1985, 90, 3797.
28) DeMore, W. B.; et al. Chemical Kinetics and Photochemical Data
Although the calculations are somewhat crude, we conclude
that the reaction with sulfuric acid forming methyl hydrogen
sulfate and dimethyl sulfate is the dominant loss mechanism of
methanol and that the oxidation of methanol is only a minor
source of hydroxyl radicals in the upper troposphere. Further-
more, balancing the measured abundance of methanol and our
calculated removal rates, the global source of methanol is greater
(
for Use in Stratospheric Modeling; JPL Publication 97-4; Jet Propulsion
Laboratory: Pasadena, CA, 1997.
(29) Scientific Assessment of Ozone Depletion: 1998; Global Ozone
Research and Monitoring Project-Report No. 44; World Meteorological
Organization: Geneva, 1998; Chapter 3.
(
30) AViation and the Global Atmosphere; Cambridge University
1
than the estimated value of 45 Tg/yr.
Press: New York, 1999; p 72.