PB85-173433
Chemical Transformations in Acid Rain
Volume 2, Investigation of Kinetics and
Mechanism of Aqueous-Phase Peroxide Formation
Brookhaven National Lab., Upton, NY
Prepared for
Environmental Protection Agency
Research Triangle Park, NC
Mar 85
-------
EPA/600/3-85/017
March 1985
CHEMICAL TRANSFORMATIONS IN ACID RAIN
Volume II. Investigation, of Kinetics and Mechanism
of Aqueous-Phase Peroxide Formation
Yin-Nan Lee
Environmental Chemistry Division
Department of Applied Science
Brookhaven National Laboratory
Upton, New York 11973
Interagency Agreement DW 930256
Project Officer
Marcia C. Dodge
Atmospheric Chemistry and Physics Division
Atmospheric Sciences Research Laboratory
Research Triangle Park, NC 27711
ATMOSPHERIC SCIENCES RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
RESEARCH TRIANGLE PARK, NC 27711
-------
TECHhJICAL REPORT DATA
(Pletue read Instructions on ths reverse before completing)
1. REPORT NO.
EPA/600/3-85/01*
2.
3. RECIPIENTS ACCESSION NO.
P?S 5 1 7 3 4 3 3
•;. TITLE AND SUBTITLE , , T-
CHEMICAL TRANSFORMATIONS IN ACID RAIN Volume II,
Investigation of Kinetics and Mechanism of
Aqueous-Phase Peroxide Formation
5. REPORT DATE
March 1985
6. PERFORMING ORGANIZATION CODE
7. AUTKOR(S)
Yin-Nan Lee
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Environmental Chemistry Division
Brookhaven National Laboratory
Upton, NY 11973
10. PROGRAM ELEMENT NO.
CCVN1A/02-3167 (FY-85)
11. CONTRACT/GRANT NO.
Interagency Agreement DW930£
56
12. SPONSORING AGENCY NAME AND ADDRESS
Research Laboratory-RTP.NC
Atmospheric Sciences
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina
13.
OS
rypE Q5 REh'ORT.AND-PEHJO
Final (8/B3-
27711
14. SPONSORING AGENCY CODE
EPA/600/09
15. SUPPLEMENTARY NOTES
16. ABSTRACT
- The aqueous-phase reactions of 63 with a number of species have been
studied in an effort to identify pathways leading to the production of hydrogen
peroxide in solution. The aqueous-phase systems studied included the
decomposition of 03 in pure water and the interaction of 03 with (1) N02,
(2) PAN, (3) ethylene, (4) formic acid, (5) formaldehyde and (6) formaldehyde
in the presence of N02. Except for the 03-ethylen^ reaction, peroxide was not
found as a reaction product. From the results obtained, it is concluded
that the reactions studied in this research effort are not significant
with respect to atmospheric peroxide formation. •-
17.
KFY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS C. COSATI Field/Group
18. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19. SECURITY CLASS (This Report/
UNCLASSIFIED
21. NO. OF PAGES
60
20. SECURITY CLASS (This page)
UNCLASSIFIED
22. PRICE
EPA Fwm 2220-1 (R«v. 4-77) PREVIOUS EDITION is OBSOLETE
i
-------
NOTICE
The information in this document has been funded by
the United States Environmental Protection Agency
under Interagency Agreement DW 930256 to Brookhaven
National Laboratory. It has been subject to the
Agency's peer and administrative review, and it has
been approved for publication as an tPA document.
Mention of trade names or rommercial products does
not constitute endorsement or recommendation for use.
-------
ABSTRACT
The aqueous-phase reaction kinetics of dissolved 03 with a number of
atmospheric components have been investigated, with special attention
focused on the formation of H202 or organic peroxi-Je as reaction products.
The reagent concentrations employed, rate laws, rate constants, and peroxide
yields (y) determined for the specified substrates are:
H20: -d[C3]/dt <= k[03] , k = 2.1 x 1(T4 s'1 (pK ~6 ) , y £ 0.5%;
H2°2 (~5 x 10~5 M^: -d[03]/dt = -d[H202]/dt = k[03nH202],
k = 2.6 x 103 M"1 s'1 (pH ~6);
HCO2H (~1 x 10"5 M): -d[03l/dt - k[03][HC02-],
k » 4.3 x 103 M"1 s'1, y < 0.5%;
H2CO (~1 x 10-5 K): -d.[03J/dt = k[03l1/7- [f^CO]1/2,
k = 1.?. x lO"3 s-l, y < 2%;
C2H4 (~15 ppm): d[peroxide]/dt = k[03l[C2H4],
k - 3.0 x 105 M*1 s-'-;
PAN (~100 ppb): d[peroxide]/dt = k[?AN][03), k <_ 3 x 103 M"1 s"1;
KO2 (40 ppb): dlperoxide]/dt = k HNO p^g,»
k !INO <_ 4 x 10"3 M atm'1 s"1;
RO2 + aC02 (I x 10'4 M): d[ peroxide] /d'; = k HNQ3 [H2CO]pNC,3 ,
k HNO _< 6 atrn"'- fi'1.
With the use a* these data, the rates of the aqueous-phase peroxide produc-
tion of these reactions under typical atmospheric conditio is are calculated
to be ~1 x 10~6 M hr~l or smaller. It is therefore concluded that the
reactions studied in this work contribute insignificantly to the formation
of peroxides in atmospheric water.
- iii -
-------
CONTENTS
ABSTRACT. Ill
TABLES vl
FIGURES vl i
ACKNOWLEDGMENT vi 11
INTRODUCTION 1
EXPERIMENTAL 4
Material 4
Ozone Generation 4
Gas-phase Oj Concentration 5
Prepartion of 0-j Solu on 5
Aqueous-phase Og Concentration , ,. 5
H202 and Peroxide Concentrations 7
Kinetic Measurements 7
RESULTS 12
03 Decomposition in Optical Cell....... 12
°3 " H2°2 Reaction « 12
°3 " HC02H Reaction.... 15
02 - H2CO Reaction 19
03 - C2H^ Reaction „ 22
Aqueous-phase Reactions of PAN 27
03 - N02 Reaction 29
03 - N02 - H2CO Reactions 30
V ••
DISCUSSION 31
03 Decomposition 31
03 - H202 Reaction 33
03 - HC02H Reaction 34
03 - H2CO Reaction 36
Oo - ^?HA Reaction <. 37
Aqueous-phase PAN Reactions 38
03 - N02 - H2CO Reactions 39
Aqueous-phase H202 Formation 41
SUMMARY 45
REFERENCES 47
-------
TABLES
Number Page
1 Second-order Rate Constant of 03 - HC02H
Reaction at 25±2°C 17
2 Second-order Rate Constant of Aqueous 0-j - ^2^^ Reaction 26
3 Summary of the Reaction Kinetics and Product Analysis
of Some Aqueous-phase Oj Reactions • ^6
- vi -
-------
FIGURES
Hunuer Page
1. Schematic diagram of the experimental setup... 6
2. Aqueous-phase 0-> spectra, taken during the course
of saturation process 8
3. Time dependence of the 03 saturation In a 2-liter
bubbler. Flow of 02 = 70 cc mln~l.,.,
with ^202. Initial concentrations of ^C^ and
were 3.5 x 10"5 M and ~4 x 10"6 M, respectively
4. Schematic of the construction of a bubbler-type
gas-liquid reaction vessel 11
5. Kinetics of 03 decomposition in a 5 cm cylindrical
quartz optical call 13
Time dependence of 0-j disappearance due to reaction
and 0-j
14
7. Plot of a second-order kinetics of 03 - HCC^H reaction
at neutral pH and [0310 ^ [HC02K]0 = 2 x 1CT5 M .......... 16
8. Plot of the pH dependence of the second-order rate
constants for 0-j - HCC^H reaction ...................... 18
9. Time dependence of 03 disappearance due to the
reaction with I^CO. Initial concentrations
of H.2CO and 03 were 1.65 x 10'4 M and 1.7 >-. 10"5 M,
respectively ........................................... 20
10. Reaction order determination for I^CO-Oj reaction.
Plots were made according to eq . 8 and eq . 9.
See text for definitions ............................... 23
11. Time dependence of the concentration of peroxides as
a reaction product in the 03 - C2H^ reaction;
?03 = °'5 PPm and PC2H4 = 28 ppm ................... 25
12. Contribution of the aqueous-phase H02 recombination
reaction to the generation of ^2^'2 ' calculated for
pH = 4.7. Line A represents the aqueous-phase rate
of 1^02 production and line B the characteristic
reaction time for gaseous HC>2 at L = 10"^. Line C
indicates the contribution to aqueous f^C^ from the
gas-phase recombination of H02', II2^2 produced is
assumed to be rapidly Incorporated into the liquid
water (L = 10'6) ......................... . ............. 44
- vii -
-------
ACKNOWLEDGMENT
The author acknowledges Mr, Paul J. Klotz for his assistance In carry-
Ing out part of the laboratory experiments, and Dr. Thomas J. Kelly for his
effort In making the ^-^2 detection technique available to us. The author
also wishes to thank Dr. Kelly and Dr. Roger L. Tanner for helpful
discussions.
- viii -
-------
INTRODUCTION
The major acids found in the rain and precipitation, namely, sulfuric
acid and nitric acid, are believed to be derived from S02 and NOX emitted
mainly from the burning of fossil fuel by power plants, smelters, mobile
sources, and space heating. The production of final acids from their pre-
cursors in the atmosphere involves chemical transformation in conjunction
with transport and deposition. In order to formulate an efficient strategy
to contiol the amount of acid deposition and to minimize its environmental
impact, a thorough understanding of each of these atmospheric processes has
to be acquired.
It has been recognized that the chemical reactions which produce the
strong acids can take place either in the gas phase or in the liquid phase.
This notion was established because convincing evidence has been collected
to indicate that the atmospheric oxidation of S02 is strongly affected by
aqueous-phase reactions, particularly by ozone and hydrogen peroxide
(Penkett et al., 1979; Martin, 1983; Kelly et al., 1984). In order to
assess the importance of these aqueous reactions, the atmospheric concentra-
tions of 03 and H202 have to be determined. Although the gas-phase concen-
tration of ozone can be accurately determined by various techniques such as
ethylene-chemiluminescence, no viable method is currently available for the
measurement of gas-phase concentrations of ^02- As a result, the tnajor
gas-phase routes for peroxide generation, i.e., the recombination of hydro-
peroxy radicals and the photolysis of formaldehyde, cannot be confidently
employed in a numerical model as the sole source for this species.
Furthermore, recent attempts to determine the gas-phase concentrations of
peroxide using bubbler series have revealed the existence of in-situ
-------
production of artifact peroxide (Heikes, 1984; ten Brink et al., 1984;
tteikes et al. , 1982). This observation suggests that aqueous pathways for
peroxide formation might exist, in addition to gas-phase counterparts.
Cle.arly, these pathways have to be identified and charact^rizad before it is
possible to accurately model the atmospheric budget of t^C^ and the rate of
S02 oxidation.
Among the various potential precursors of aqueous peroxide, it has been
speculated that Oj might be a plausible candid&te based on the following
considerations: (1) 0-^ produced peroxides upon reaction with certain
organic compounds, e.g., olefins (Gilbert, 1976), (2) aqueous-phase 03 reac-
tions were found to involve free radicals derived from 03 decomposition that
might serve as ^C^ precursors (Holgne and Bader, 1976; Bllhler et al.,
1984), and (3) the bubbler series experiments (Heikes et al. , 1982) demon-
strated that the levels of artifact ^02 do not diminish rapidly along the
bubbler train, consistent with the presumption that the precursor species
might be present in relatively high concentration and has a low aqueous
solubility. 03 appears to fit the description.
Although the aqueous-phase reactions of ozone have been the subject of
numerous studies for the past several decades, major gaps exist In the
understanding of the detailed features of these reactions. Additionally,
with the emphasis focused on drinking water treatment by Oj, no major
efforts have been directed to the product analysis, particularly HyC^. In
this current laboratory research we have examined a series of aqueous-phase
reactions of 03 and the formation of '^2®2 and organic peroxides as their
reaction products in an attempt to identify .the direct aqueous sources of
peroxides. The aqueous-phase reaction systems examined include: (1) 0^
- 2 -
-------
s»l f-decornpcsltion, (2) Oj reaction with formaldehyde (irttA and without the
presence, of NC^), formic acid, ethylene, and peroxyacetyl nitrate (PAN).
Except for the Oj-ethylene reaction, peroxide was not found as a reaction
product. From this study It Is concluded that the reactions studied here
are not significant with respect to atmospheric peroxide formation. As a
result, knowledge of direct tqueous-phase sources of ^02 remains highly
uncertain.
- 3 -
-------
'EXPERIMENTAL
Material
N2 (Ultra-High Purity, Llnc'e) and 02 (Ultra-Zero, Matheson) which con-
tained less than 0.5 ppm hydrocarbons (as CH^) were used as the diluent gas
and 03 source, respectively. N02 gas was prepared using a N02 permeation
device (Metronics, Inc., wafer type) enclosed in a constant temperature
oven. With a 15 cc/tnin flow of N2 carrier gas the concentration of N02 from
the permeation source was determined to be 6.0 ppm. Samples of gaseous PAN
(6-12 ppm) in N2 were prepared from concentrated PAN stock solution in
n-tridecane prepared according to a modified method of Nielsen (Nielsen et
al., 1982; Gaffney et si., 1984). Working stock of ethylene (0.140%) in N2
was prepared from pure ethylene (99.5%, Scott) and UHP N2. H202 (30%), H2CO
(37%) and HC02H (88%), all of reagent grade fron either Mallinckrodt or
Baker, were used without purification. EDTA, Trisma Base, and horseradish
peroxidase were of the highest purity available from Sigma Chemical
Company. Concentrated HC1 and NaOH and inorganic salts such as KC1 were all
of reagent grade And used without further purification. Fresh solutions of
standards and reagents were prepared for the same-day use. Distilled water
(resistance _>. 16 Mohm at 25°C), which had been further pi-rlfied by MillJpore
Milli-Q System, was used for all of the studies. (
Ozone Generation
Two ozone generators were used. For higher Oj output (up to ~200 ppm
at 1 '-/rain flow rate) a 10-inch Pen-Ray UV lamp was employed. This source
was used mainly for the preparation of saturated 03 solutions for batch-type
reactions. For lower 03 output, an AID Ozone Generator (Model 565) equipped
- 4 -
-------
with a continuously adjustable shutter for Oj output control was employed
(up to ~1 ppm at 1 l/mln flow). This source was used for the continuous-
flow reaction system. For both generators, high purity Q£ purchased from
Matheson was used in order to minimize possible interferences from NOX and
organic impurities.
Gas-Phase 03 Concentration
Gas-phase 0-j concentrations were determined either by a Dasibi 0^ moni-
tor (Model 1Q08-PC) operated based on 0V absorption (LOD ~2 ppb) or by a
Monitor Labs Ozone Analyzer (Model 8410) based on the C>3-ethylene chemilum-
inescence, with the former as the primary standard.
Preparation of Og Solution
For batch-type experiments, 0-j solutions were prepared in a 2-liter
bubbler through which 0-^ was continuously bubbled at a total 02 flow of 70
cc rain"'- (Fig. 1). Each Pyrex Oj bubbler (volume ~2 ?.) was equipped with a
coarse-sized frit for the enhancement of mixing and with two ports for
liquid transfer. The plumbing was constructed with parts made either of
stainless steel or Teflon for purity. Bubbler I was needed for the humidi-
fying of the gas-stream and the removal of any soluble substances; bubbler 2
supplied saturated 0^ solutions to be used for the ba';ch studies. For the
study of some continuous-flow reactions, valve M-l wf.s switched so that Oj
would flow through the gas-liquid reaction cell to it.-itiate the reactions.
Aqueous-Phase 0-j Concentration
The concentration of aqueous-phase 03 was measured by a UV-vis spectro-
photometer (Beckman Model DU-7) using either a 10 cm or a 5 cm cylindrical
- 5 -
-------
EXHAUST
A
METER
CONDUCTIVITY
METER
GAS-LIQUID
REACTION
CELL
EXHAUST
Figure 1. Schematic diagram of the experimental setup.
- 6 -
-------
optical cell. Spectra of the 03 solutions in the bubbler were taken during
the saturation process (Fig. 2) and the concentrations of 03 were determined
from the absorption peak at 260.0 nm using an extinction coefficient «=
2930 M-1 cm"1 (Kosak-Channing and Helz, 1983). The time dependence of 03
saturation In the 2- liter bubbler is shown in Figure 3. Using the 10-inch
Pen-ray UV ozonator at 02 flow rate of 70-90 cc ruin"1, a typical
aqueous-phase 03 concentration of 1-2 x 10~5 M was obtained after 40 min of
bubbling. The limit of detection for 03 with the use of the 10-cra cell is
1 x 10-7 M-
and Organic. Peroxide Concentrations
Concentrations of aqueous-phase ^02 and organic peroxides were deter-
mined by the horseradish peroxidase-f luorescence technique (HRPF) (Guilbault
et al., 1968) modified by NCAR (Lazrus et al., 1983). In our arrangement, a
Perkin-Elmer fluorometer (Model 204S) was employed in conjunction with a
liquid flow reaction system equipped with a rotary injection valve (Altex,
sample loop size 0.5 ml). The limit of detection of the HRPF technique is 1
x 10~7 M. Since this technique does not distinguish organic peroxides from
inorganic HoO^* the determination of the concentration of organic peroxides
was achieved by a difference method in which ^02 is preferentially
destroyed (or inactivated) by the enzyme catalase (Schonbaura and Chance,
1976).
Kinetic Measurements
Two different types of kinetic methods vere usod. In the batch-type
study, reactions were initiated by mixing the reagents with 03 solution in
an optical cell and the decrease of [03] accompanying the reactions were
- 7 -
-------
Displd'-| Hod*
[Overlay]
Function
\ n
': v.
\Y
\ X \
\ \ Vy
•-.
wave length, nm
Figure 2. Aqueous-phase 03 spectra, taken during the course of saturation
process.
-------
0.30
10.25
o
(£>
UJ
gO.15
<
CO
00.10
DQ
0.05
0
50
100
TIME.min
150
200
Figure 3. Time dependence of the 03 saturation in a 2-liter bubbler.
Flow of 02 = 70 cc min~ '
- 9 -
-------
followed spectrophotometrically at X - 260 nra. In the continous-flow reac-
tion, a reagent gas mixture containing constant concentrations of 0-j and
other gaseous reactants was continuously bubbled through a solution con-
tained In a bulk-type gas-llquld reactor. The detailed construction of the
gas-liquid reactor Is shown schematically In Figure A. The kinetics of H7C>
generation were followed by an aliquot method in which ^02 concentrations
were determined by the HRPF technique. Temperature of the reaction vessel
was maintained at 22.0 * 0.1°C for the latter method, but for the spectro-
photomatric technique the uncertainty was somewhat greater (i"20C/ due to th
lack of temperature control of the sample compartment for a cylindrical
cell.
- 10 -
-------
GAS OUTLET <—
THERMQSTATTED
WATER OUT
THERMISTOR WELL
THERMOSTATTED
WATER IN
DISK-FRIT
BAFFLES
LIQUID TRANSFER
PORT
ELECTRODES
GAS INLET
Figure A. Schematic of the construction of a bubbler-type
gas-liquid reaction vessel.
- 11 -
-------
RESULTS
03 Decomposition In Optical Cell
In order to study the reaction kinetics of 03 with other substrates
using a spectrophotometric method, it is necessary to determine the stabil-
ity of 03 solution in the optical cell employed. Figure 5 shows a typical
decay curve of 03 in a 5 cm cylindrical cell under neutral pH (no s'ltd or
base added). The first-order rate constant was found to be 2.1 x 10"* s"1.
When pH of the solution was adjusted to 2 by HCl the rate constant of 03
decomposition dropped to 1.4 x 10"^ s"'-. These values w snoul-d be quantified.
The kinetics of this reaction was studied under pseudo-first-order condi-
tion, i.e., [f^C^Jo >> l^lo- ^ typical trace of 03 decomposition under
such a condition is shown in Figure 6. The 03 decay kinetics was found to
conform to a pseudo-first-order reaction for ~3 half-lives. The second-
order rate constant, obtained by dividing the pseudo-firs t-order rate con-
stants by [H202]Q, which had been varied from 8 x 10~6 M to 6.4 x 10'5 M,
was determined to be (2.6 ± 0.4) x 103 M"1 s"1 at neutral pH. The good
agreement obtained between the second-order rate constants determined at
widely different H202 concenfcrat^ons permits the conclusion that the reac-
tion kinetics are also first order with respect to [H202]. At lower pH the
- 12 -
-------
Display ilode
[Overlay]
Function
[fibs]
fl
time, min
Figure 5. Kinetics of 03 decomposition in a 5 cm cylindrical
quartz optical cell.
- 13 -
-------
B, 180 T
DlSpl-i'-l Hodti
Function
[fibs]
• I
Tr
t ime, s
Figure 6. Time dependence of 63 disappearance due to reaction with
Initial concentrations of H202 and 03 were 3.5 x 10~-- M and
~ A x 10~6 Mt respectively.
-------
reaction rate decreased rapidly. A second-crder rate constant of 38 ± 8 M'1
s~l was determined at pH " 4, where the solutions contained [HCl] = 1.00 x
10-4 M.
03 - HC02H Reaction
Formic acid has been Identified as a major organic acid component In
rainwater (Keene et al., 1983) and therefore the sources and sinks of this
species may be of Importance to the understanding of acid rain formation
mechanisms. We examined the aqueous-phase reaction of 03 with IICC^H to
determine the reaction rate and tlvi yield of peroxide as a product.
*l'
03 + HC02H > Products (1)
The kinetic study of reaction (1) was made by monitoring the change of [0-j]
followed spectrophotometrtcally at X • 260 nm. Since the rate of this reac-
tion under current conditions Is too fast to be studied under pseudo-fIrst-
order condition, I.e., high concentrations of HCC^H, initial concentrations
of 03 and KC02H were made approximately equal st ca. 2 x 10"5 M. Treating
the reaction with an overall second-order kinetics, I.e.,
Rate - ll » k1'[03][HC02H] (2)
dt
the rate constant was determined by using a second-order plot
1 L_-k,'t (3)
[o3]t [o3]0
Plots of [03]t~l vs. time were found to be linear for at least three
half-lives. A typical run Is shown In Fig. 7. The rate constants obtained
are listed In Table 1 as a function of pH.
- 15 -
-------
20
40 60
TlME.s
80
Figure 7. Plot of a second-order kinetics of 03 - HC02H reaction
at neutral pH and I03]o ~ [HC02H'jO - 2 x 10 M.
- 16 -
-------
Table 1
Second-Order Rate Constant of 0? - !!C07K Reaction at 25 ± 2°C
3.00 ± 0.02 (2.5 ± 0.2) x 102
4.00 i" 0.05 (1.8 * 0.2) x 103
^.60 -fc 0.2 (3.3 t 0.5) x 103
The pH dependence of V.].1, increasing with higher pH, can bo fitted to a rate
law which assumes the rate determining step involves 03 and i-.he dissociated
formate ion, i.e.,
Rate = k4 [03)[HC02-]
) [0:,][HC02H]T
Ka
- k1l[03][HC02H]T (4)
where Xa Is the acid dissociation constant of formic aclf* and [HC02H]-p is
the total analytical concentration of formic acid. Fitting Eq. (4) to the
experimental data allowed the values of k^ and Ka to be determined; they
were found to be 4.3 x 103 M~l s"1 and 8.9 x 10~5 M"1, respectively.
Although the agreement between the data and the calculated curve
appears to be quite reasonable- (Fig. 8), the pKa value of formic ac5d deter-
mined here, 4.1 i" 0.1, Is nearly 0.3 unit higher than the literature value
of 3.V5 (Rlddick and Hunger, 1970). While this discrepancy may be accept-
able in view of the fact that the kinetic approach generally would yield a
larger uncertainty for pKfl determination (Slllen and Martell, 1964^ than
- 17 -
-------
PH
Figure 8. Plot of the pH dependence of the second-order rate constants
for 03 - HC02H reaction
- 18 -
-------
other methods (e.g., conductometrie), we nonetheless consider this base-
catalyzed mechanism tentative.
Final reaction mixtures of the Oj-HCC^H reaction were analyzed for per-
oxide by the HRPF method. The level of peroxide was found to be smaller than
the LOD of the instrument, namely, 1 x 10"^ M. Under the present reat ';ion
conditions, e.g., [O^]Q = [HC02H]o = 2 x 10"5 M, a value of 0.5% Is es,.!-
rcated as the upper limit for the peroxide yield of the C^-HCC^H reaction.
03 - H2CO Reaction
Since formaldehyde is a relatively abundant atmospheric organic consti-
tuent (Grosjean, 1982; Tanner and Meng, 1984) which is derived mainly from
combustion emission and the atmospheric oxidation of higher hydrocarbons,
Its chemistry is of considerable Interest to the understanding of atmo-
spheric oxldant formation mechanisms. The aqueous-phase reaction of 03 with
H2CO was studied In a similar fashion to the C^-HCO^U reaction, namely,
the extent of the reaction In a 5-cm optical cell was monitored by UV
absorption for [03] decrease.
The kinetics, under conditions where initial [f^CO] was in 1..-. •.••ye excess
of [03], did not conform to a pseudo-first-order reection. In fact, the
effective second-order rate constant appeared to Increase as [03] was
decreasing, Indicating a reaction order of less than unity with respect to
03. This behavior is illustr&ted by a typical run shown in Figure 9 where
(H2CO]0 - 1.65 x 10~4 M and [03)o = 1-75 x 10~5 M.
Since it has been shown that formic acid Is produced as a product in
the 03-H2CO reaction (Kuo and Wen, 1977), i.e.,
03 + H2CO --> 02 + riC02H (5)
- 19 -
-------
\
Display !-bde
[Overlay]
Function
n r>•!
\
fl
~~*j™«—
19
time, min
Figure 9. Time dependence of 03 disappearance due to the reaction with
Initial concentrations of H2CO and 03 were" 1.65 x 10"^ M and
1.7 x 10~5 M, respectively.
- 20 -
-------
the 03 decomposition may be enhanced by the subsequent 03-HC02H reaction,
thereby manifesting a non-first-order kinetics. Comparing the initial rates
of the 03- H2CO reaction with that for Oj- HCO2H reaction indicated that the
0-j decomposition due to the secondary reaction with the intermediate
product, HCC^H, can be significant, If HC02H is indeed produced stoichio-
metrically. However, since the product analysis reported by Kuo and Wen
"1977) did not appear to be quantitative and it is not clear to us whether
or not formic acid Is produced from an elementary reaction step, we analyzed
our experimental data without Invoking the reaction of HC02H, but strictly
in terms of 03 and H2CO. It should be noted that the general kinetic
behavior so deduced may be applied only to reaction conditions similar to
that employed In this study.
To determine the reaction order with respect to each of the reagent
concentrations, we consider the equation
Rate = Id[°3] = k[03]n[H2CO]m (6)
dt
Under conditions that [H2CO]0 » [03]f), Eq. (6) is rearranged to yield
(Rate/[H2CO]m) - k [03]n (7)
where [f^CO]"1 is a constant throughout the reaction course. Taking
logarithm of Eq. (7), we obtain
log (Rate/[H2CO)m) = log k + n log [03] (8)
Fitting Eq. (8) to the data obtained from Figure 9 (from which a set of Rate
vs. [03] can be determined fir any specified time), the value of the slope,
- 21 -
-------
n, was found to be ~l/2 (Fig. 10). Using the same approach, an analogous
equation is obtained as
log (RateQ/fO,]1/2) «= log k + m log [H2GO)0 (9)
To determine ra, we measured the initial rate for three initial formaldehyde
concentrations: [H2CO]0 - 1.65 x 10'4 M (Fig. 9), 8.2 x 10'5 M, and
4.1 x 10" •* M. Fitting Eq . (9) to the rate data, ra was found to have a value
also close to 1/2 (Fig. 10). Eq. (6) is now tentatively identified as
Rate = k [O^1/2 [HjCO]1/2 (10)
with k = (1.15 -t 0.7) x 10"3 a"1, at t = 20ir2°C, for the reaction conditions
employed.
Product analysis of the fins! reaction mixtures showed that H202 was &
minor product. With [H2CO]0 - 8 x 10"5 M and [0310 = 8 x 10'6 M the H202
formed was found to be 1.5 x 10"? M at the conclusion, of the reaction.
°~<-'^ Reaction
Olefins represent an important class of organic constituents of the
atmosphere because of their reactivity towards 03 and OH oxidation (Niki et
al., 1983). Since the Criegee intermediates formed in 03-olefin reactions
may transform into peroxides in protlc solvents, we have examined the kin-
eticfi of the aqueous 03-ethylene reaction in connection to its rate of
peroxide production. Due to the low solr.bility of C2H^, i.e., HC^HA = ^.0
x 10~3 M atnT1 at 25°C (Wllhelm et al., 1977), the kinetics of this reaction
was studied by the use of the continuous-flow method. A constant flow of N2
(typically 2.0 I mln~l) was first allowed to flow through the gas- liquid
- 22 -
-------
10
-5
IO
r5
c
'e
UJ
H-
<
cr
IO
r6
10
-6
[HeCO] , M
10
-4
J L
I I L
J I I
IO"5
[03] , M
10'
c
"E
fO
O
UJ
<
cr
10
-4
Figure 10. Reaction order determination for H2CO-03 reaction. Plots were made
according to eq. 8 and eq. 9. See text for definitions.
- 23 -
-------
reacr.lon cell containing a known volume of liquid water to remove the dis-
solved C02. When the conductivity of the liquid water had stabilized,
reagent ^2^U was added to the gas stream; the concentra tiers range of £2^*1
employed was 7-28 ppm. After the reaction was initiated by the addition of
Oj (concentration range: 0.5-1.0 ppn) , the reaction mixture was analyzed for
peroxide concentration at known time intervals using the HRPF technique.
The concentration of peroxide of the reaction mixture was found to increase
linearly with time (Fig. 11) and the reaction rate defined as
Rate - d[per°xid-il (11)
dt
was determined.
Assuming that the aqueous-phase reaction of Oj and 02^4 is first order
with respect to each reagent, eq . (11) is given as
dtperoxide] ^ kiz [031]C2H4] (12)
Additionally, assuming that the liquid water is saturated with the reagent
gases according to Henry's Law, I.e.,
[X] = HxP>c (13)
which allows eq. (12) to be rewritten as
the second-order aqueous-phase rate constant kj 2 can readily be determined.
The values of kj 2 determined at various reagent concentrations and pHs are
-------
42
c
JD
0
Tir
Q
X
o
cc
LJ
f\
Q.
*
1 ' 1 « 1 '
I
,
1 ! 1 1 1 1 1 1
30
60
SO
120
Figure 11. Tin-e dependence of the concentration of peroxides as a reaction
product in the 03 - C2H4 reaction; Po3 - 0.5 PPm and pc2H4 " 28
- 25 -
-------
Hated in Table 2. It should be noted that the ctrerage values were obtained
.from repeating, runs as well as from runs with different reagent concentra-
tions. The fact that the rate constant determined froir. .°q. (14) remained
essentially constant while the pg and Pc2H4 were varied by c factor of
2 and 4, respectively, lends support to the assumed overall second-order
kinetics.
For a gas-liquid reaction taking place in bulk liquid, particular
attention has to be given to the mass transfer characteristics of the appar-
atus tn order to identify the extent of mass transfer limitations on the
Table 2
Second-Order Rate Constant of Aqueous 0-^-^2^^ Reaction*
pH 10-5 x k12, M'1 s'1
neutral 2.60 ± 0.26
(6 ~ 7)
5.0 3.0 -t 0.25
4.0 3.0 ± 0.23
3.0 •' 3.3 -t 0.41
*[KC1] = 1 x 10'3 M, except for runs at pH 3; T = 22.0 * 0.1CC.
overall reaction rates (Danckwerts, 1970). A simple approach to this ident-
ification is via the comparison of reaction time constant \Tr) to the mass
mass transfer time constant (Tm): when Tr 2. 10 T^ the reaction is purely
chemical reaction limited; when Tr / 10 Tm mass transfer will become a rate
limiting process. For our currently employed gas-liquid reactor, the
phase-mixing rate constant, k,,,, was determined to be 0.47 s'*- (at liquid
volume = 15 ml, total gas flow rate = 2.0 ft rain"1, cf. Lra and Schwartz,
1981). The time constants for the removal of aqueous Oj and
- 26 -
-------
other are estimated as TQ - (3 x 105 ^H/J)'1 - 24 s and TC H = (3 x
10 [03])"1 = 300 s. These reaction time constants are significantly-
longer than Tm (a km-1 - 2.14 s); It Is readily established that the chcmi-
cat reaction Is the rate limiting process and the reaction rnts Is not sig-
nificantly affected by mass transfer processes. The assumption of the
establishment of gas-liquid equilibrium is therefore validated.
From these results it is also seen that peroxide formation from the
aqueous €2^-03 reaction Is Independent of the solution pK,; the average
rate coefficient k12 = (3.0 ir 0.3) x 105 M"1 s"1, Is applicable over the pH
range 3-7. It should be noted that the fraction of organic peroxide
formed in the €2^-03 reaction was determined using the enzyme catalasc
technique if. which the HjC^ is preferentially destroyed (Kelly et al., 198A;
Sch^nbaum and Chance, 1976). The results of this test showed that >_60%
(average of 3 runs made at pH 3) of the total peroxide was present as
organic peroxide which appeared to be reasonably stable. Tha identity of
the organic peroxide was not determined, but from the mechanism advanced for
the forma-tJon of ozonide and Criegee radical, CHjC^H is considered a
possible candidate.
Aqueous-phase Reactions of PAN
Peroxyacetyl nitrate (PAN), formed from the reaction of CH3C(0)02 and
N02, is an important atmospheric species because of Its capacity as a reser-
voir for NC>2 and free radical species in transport processes. PAN was syn-
thesized according to the method described by Nielsen (1982) and was stored
In a heavy Hpld, i.e., n-trldecane. Gas samples of PAN were prepared by
mixing N2 with the distillate from PAN/n-trldecane solution. With PAN vapor
- 27 -
-------
pressure (over n-tvtdecane) In the order of 10 torr, 10 ppn gaseou-j PAN
standards can be readily prepared. Because it was found that PAN decomposes
on standing with a half-life of approximately 110 brs (at 25°C, In the
absence of HO, Senutn and Gaffney, 1984), PAW standards were always freshly
prepared for the same day use.
The three reactions of PAN examined include
PAN(a) ---- > peroxide (15)
k!6
PAN(a) + 02 ---- > peroxide (16)
*17
PAN(a) + 03 ---- > peroxide (17)
The typical concentrations used for PAN, 02 , and 03 Mere 100 ppb, 20%, and
1.0 ppm, respectively. Due to the low solubilities of 02 and 03, as well as
of PAN (HpAN " 3.6 M atm"1, cf. Lee, 1*584), these reactions were again
studied by the use of the gas-liquid reaction cell described above. Final
reaction mixtures (with total reaction time up to 1 hr) of these reactions
were analyzed for peroxide and it was found that the levels of peroxide were
all below the detection limit of our HRPF technique. Using a general rate
expression
^peroxide] = k [PAN] fOx] (18)
dt X
the upper limits for rate constants ^5, k^G' and kj^ were estimated to be 2
x 10"4 s"1, 3 x 102 M"1 s"1, and 3 x 103 M"1 s'1, respectively.
- 23 -
-------
03 - N'0? Reaction
In view of the suggestion that NO-j Is potentially Important in cloud
water chemistry (Helkes and Thompson, 1983), we have conducted sose prelimi-
nary studies of MOj + H2G. Since the solubility of N03 was not expected to
be large, the reaction was carried out in the bubbler-type gas-liquid reac-
tor into which the reagent gases were contiguously replenished. Study of
the background reaction involving H20 and 1.3 and 43 ppm of 03 showd chat
no detectable H202 wa« produced after 4'^ min of reaction. This result is
consistent with that observed for the pure aqueous-phase Oj decotaposition
study conducted in an optical cell (see above). To conduct the N02 experi-
ment a prt-reaction chamber of l-iiter size was placed upstream of the bub-
bler to allow the production of WD3 from the 03- N02 £as-phase reaction.
Since the second-order gas-phase rate constant for
03 + N02 --> 02 + N03 (19)
is 3 x 10'17 cm3 s"1 (Baulch et al. , 1982), the 1/e time for tJ03 formation
is calculated to be 31 s at pjj, - 43 ppn, the 0^ concentration employed.
With 30 s residence time in the mixing chamber (total gas flow rAtes =21
min"1), nearly 70% of the N(>2 should be converted to N03 (pfjo2 employed
was 40 ppb). It should be noted that the experlranets were carried out in
the dark in order to avoid the photolysis of N03. At the end of 40 min the
reaction mixture in the gas- liquid reactor was analyzed for H202 content.
Concentration of H202 was found to be ~1 K 1C"7 M. This low concentration
seema to indicate that N03 has a limited role in the direct aqueous produc-
tion of peroxide, at least In pure water and under neutral pH.
- 29 -
-------
Oj-NC^-t^CQi Reaction
Reactions of KOj with atmospheric constituents have been found to be
important in nighttinw chemistry (see, for exanple, Pitts et ei., 1934).
Since NOj can react with aldehydes to initiate free radical chain reactions,
we have tested the possibility of the formation of peroxide by the aqueous
reaction of NO-j and l^CO. In our approach, NC>2 (40 i 1 ppb) and 03 (40 i 1
ppm) In K2 (with 2% O^) were allowed to react in a pre-rftact!on chamber
(ca. 1 J. size) for ~30 s before entering the gas-liquid bubbler that con-
tained 20 snt of « 1.0 x 10"^ M H2CO solution. According to our calculation
given in the previous section, nearly 70% of the N02 should l-e converted to
NOj. The reaction extent of the formaldehyde solution was folloved by the
peroxide concentration determined by an aliquot method. Despite the long
reaction tirae (up to 100 min) no discernible peroxide was detected as A
product.
- 30 -
-------
DISCUSSION
A number of selected aqueous-phase reactions of O-j have been studied In
\
this research project. The main purpose of this work Is to Identify and
characterize those aqueous Oj reactions that would produce either hydrogen
peroxide or organic perloxldes as reaction products. Stju-.j the effort was
mainly directed at the identification of those reactions, the entire project
was scoping in nature; rather than spending a great length of time to char-
acterize the detailed kinetics of n reaction which does not produce
peroxide, we tried to naxlsnlze the number of reections to be examined for
peroxide production. As a consequence, the k'.netlc Information obtained
here with limited variations in reaction conditions arc necessarily not
comprehensive. For instance, reagent concentration variations were gen-
erally limited, the solution acidity was normally confined between neutral
and pH 3, and the temperature was fixed either at 22.0 i 0.1°C or room
temperature. Nevertheless, we have attempted to make use of the data
obtained in this work to qualitatively assess the atmospheric importance of
these aqueous-phase 0^ reactions to the formation of peroxide and to the
depletion of the reagent species.
C>3 Decomposition
Despite the fact that the kinetics of 03 decomposition in aqueous solu-
tions have been studied extensively, no agreement was reached with respect
to its reaction order. The reaction order reported in th« literature rangfd
from 1 (for exanple, Sullivan and Roth, 1970} to 2 (for example, Gurol and
Singer, 1982), with some as 3/2 (for example, Kllpatrlck «>t al., 1956).
This discrepancy was suggested to result from a number of possible sources
- 31 -
-------
such as mass transfer limitations, tonic strengths, impurities, as well as
effects of buffers with which the solution pH were adjusted. The main
fear.ures of Oj decomposition observed in this current study are in agreement
with some of the recently published results: (1) 03 deconposition is
insensitive to pH in the acidic region (Gurol and Singer, 1982), (2) decay
of 03 is first order in 03 concentration (Sullivan and Roth, 1979), and (3)
the first oro'er rate constants determined in this work, i.e., 2.1 x 10~^ s"1
at neutral pH and l.A x 10"^ s""- at pH = 2, are within a factor of 2 of
those determined by Sullivan and Soth (1979), i.e., A.I x 10~4 s"1 and 2.4 x
10""4 s~l for pH « A and 2, respectively. It should be noted that the
uq-tieous-phase concentrations of Oj used in this study were smaller than
those used in previous studies by a factor of at least 10, and, therefore,
may more closely resemble the actual atmospheric conditions. Using the
aqueous-phase decomposition rate expressed as the gas-phase reaction rate
(Lee. and Schwartz, 1981), i.e.,
PO,
- k Ho,po, LRT (20)
dt 3 3
where L is the liquid water content of the atmosphere (volume fraction), we
estimated the reaction time constant of Oj against aqueous-phase
decomposition to be 5 x 10° hr or longer for solution pH of neutral or lower
at L ** 10"^. If wall effects in the laboratory study are partially
responsible for the 0-j decomposition, then these estimated time constants
would represent lower limits, rendering the atmospheric self-decomposition
of 03 in aqueous droplets even less important.
- 32 -
-------
°3"H2°2
Oj decomposition Is known to be Initiated by OH" io^ under basic condi-
tions (Gurol and Singer, 1982; Sullivan and Roth, 1979; Staehelin and
Hoigne, 1982). According to Staehelin and Hoigne, the conjugate base of
H202, i.e., H02~, also reacts with Oj to produce free radicals such as OH-
and 02"~. FTom studies raade in alkaline solutions, the rate constant for
the Oj - HC>2~ reaction,, defined as
d 10,]
- - — - k2l [03] [H02-] (21)
dt
was given as (2.8 ~t .5) x 106 M'1 s"1. Replacing [H02~] with the total
analytical concentration of H202 , [H202]T, we obtain
(22)
- k21' [03] [H202]T
where. Ka is the acid dissociation constant of H202. For pKa = 11.6 and pH
6, k2]^' Is calculated to be 7.0 M"1 s"1. This value is significantly
rmaller than that determined here; it Is therefore concluded that the
second-order rate constant determined in this study may reflect the direct
reaction of 03 and H202. The lower rate constant detsrrained at pH 4 main-
tained with [HC1] = 1 x 10~A M is qualitatively consistent with a negative
[H"*"] dependence, although the possibility that the lower rate constant
resulted from Cl~ ion inhibition (Taube and Bray, 1940) can not be
excluded.
- 33 -
-------
The fact that H202 can be destroyed by 03 requires the effect of this
reaction on ^02 concentrations be quantitatively assessed. To
evaluate the time constant for the removal of H2Q2 by the 03-11202 reaction,
we employ
K202 - (k23 [03] r
(23)
With PQ. =» 1 ppm, TJJ Q is calculated to be 10 hr at neutral pH and 660
hr at pH 4. These time constants are apparently too long to be important
for our current laboratory studies.
The time constant for removal of atmospheric gas-phase Oj by the
aqueous 03-^02 reaction is given as
T03 "
-------
i •>
_- K25 f03] [HC02H]1'2 (25)
dt
was determined as 2.6 x 102 and 8.7 x 1C3 M"1 s"1 at pH 2 and p>! 7, respec-
tively. These values, along with an additional measurement at pH 11, showed
a pH dependence In which the rate constant Is approximately proportional to
[OH"]1'^. Although the rate constant at pH 7 is in reasonable agrsamert
with that determined in this work, i.e., 4.3 x 103 M"1 s"1, the pH depen-
dence of the reaction rate Is quite different from that observed here.
Vhe results reported by Hoigne and Bader (1983a), on the other hand,
exhibited a pH dependence similar to that observed In this work, I.e., the
rate law Is consistent with a mechinisra for which the reacting species con-
sists of the dissociated formate Ion rather than the undissociated formic
acid. The rate constant reported by them for the O-j-HCC^" reaction was 140
M'l- s"*-, which Is a factor of 28 smaller than that determined in this work.
Their srnaller value may very well have resulted from the doping of the reac-
tion mixture with propanol, a free radical scavenging material. According
to Holgne and Bader, the presence of such a substance serves to inhibit the
Oj decomposition via free radical pathways. However, It may be somewhat
uncertain whether the scavenger Itself may enter the reaction pathways and
whether certain scavengers may be more efficient than others.
Using the expression given by Eq. (4), the reaction time c>. \star.ts
referred to gas phase for 0-j and HCC^H against reaction (1) are estimated
as
and
H0 HHCO „ PKCO H LRT]-I (26)
- 35 -
-------
THC02H - tk4 (— -— ) HO, HHCo2n P0, LRT]'1 (27)
[H+]+Ka J 2 3
Using HHC02H = 23° M atra"1 (at pH 4, Gaffney and Senum, 1984) and the fol-
lowing conditions: pH » A, L - 10~6, p^ = 50 ppb, Pnco2H "• 10 PPb» and
T - 25°C, Q3 and RC02H are calculated to be 2.4 x 105 hr a.id 4.8 x 104
hr, respectively. These values are clearly too long to make this reaction
significant.
03-H2CO Reaction
Kuo and Wen (1977) studied this reaction and reported an 03 decomposi-
tion rate which is dependent on [H2CO] to the first power and on [03] to the
1/2 power. The rate constant defined by
k28[03]l/2[H2CO] (28)
dt
changed from 2.2 x 10~2 M"1/2 S"1 to 0.53 M"1/2 s"1 as pH increased froa
2.7 to 7.2. On the other hand, Hoigne and Bader (1983b) suggested an over-
all second-order reaction, first order in each reagent, for which the rate
constant was determined to be 0.1 ± O.OJ M~^ s~l at pH 2. In contrast to
both of these rate laws, we have suggested a tentative rate law, namely,
that given by Eq. (10) where the power dependences of the rate on [03] and
on [H2CO] are assigned as one-half. The discrepancies observed in these
studies may result from the different reaction conditions employed. For
example, Hoigne and Bader used a l^CO concentration in the range of 6-70 x
10~2 M, which is higher than the concentrations used in this study by 3 to 4
orders of magnitude. Since it is known that some aliphatic alcohols can
serve as free radical scavengers, the hydrolyzeci formaldehyde, i.e.,
- 36 -
-------
dihydroxyme thane, night efficiently block the free radical pathvayp of this
reaction and diminish the overall observed rate.
The fact that different kinetic behavior is observed for varied reac-
tion conditions suggests that these observed rate laws may at best be
applied to those conditions used and should not be extrapolated to condi-
tions which are vastly different. Since the conditions employed in this
study involved the lowest concentrations for both 0-j and f^CO among these
studies, the kinetic information obtained in this study may be used to esti-
mate the reaction time constants of atmospheric 0-j and H^CO against the
aqueous Oj-HCHO reaction. For the following conditions, namely, pH •*
neutral, P^-CO " ^ PP^» PO-J ** *® PP^i an<* ^ ™ 10"^, these reaction time
constants are found to be 6 x 10° hrs or longer. Again, it may be concluded
that the aqueous-phase reaction of 0^ and l^CO affects very little the atmo-
spheric life times of these species.
0-0 Reaction
Due to the low solubility of C2^4» tn*s reaction was studied using a
continuous flow technique with the gas-phase reagent concentrations main-
tained constant throughout the reaction course. The second-ordrjr aqueous-
phase rate constant determined from the peroxide formation rate, i.e., 3 x
1()5 M-I a'1, was found to be of similar magnitude to those determined for
some substituted ethylenes such as l-hexene-4-ol (2 x KP M"'- 8~*) and
ctyrene (3 x lO-' M~^ a" ) studied as homogeneous aqueous-phase reactions
(Hoigne and Bader, l<»83b). It is interesting to note that the substituent
effect observed for the gas-phase O^-olefin reactions seems to be absent in
the aqueous phase.
- 37 -
-------
The rate of depletion of Oj and £2^4 from the gas phase due to aqueous
03-02^ reaction may be estimated from
T- (k H03 Hc2H4 LET p)-l (29)
For PC^HA " 20 ppb, PQ = 50 ppb, and L = 10"6, the values of TS are
found to be In the order of 1 x 107 hr. Evidently, these long reaction time
constants «*ould have minor consequences on the atmospheric residence times
of either 03 or C2H^.
The aqueous-phase rate of peroxide production from this reaction is
estimated to be 6 x iO~H M hr"1 for PQ, = 50 ppb and PC?HA "" ^0 ppl>.
It should be noted that the 0-j-olefin reaction will not constitute an
Important source of aqueous-phase peroxide unless the condition ("p)oiefin >
1 x 10"* M Is met. The current knowledge, however, Indicates that such a
species Is not to be found (NSF, 1977).
Aqueous-Phase PAN Reactions
The contribution of aqueous-phase PAN reactions to peroxide production
in the atmosphere can be estimated from the upper limits assigned to reac-
tions (15) to (17). For a typical ambient condition where p0 = 50 ppb
and PPA>] * 2 ppb, the upper limits of the rate of peroxide formation are
evaluated as 4 x 1.0'9, 1 x 10"6, and 4 x 10~u M hr'1 for PAN hydrolysis,
02-PAN, and 03-PAN reactions, respectively. The atmospheric residence times
of PAN against hydrolysis, 02-PAN, and O^-PAN reactions under the same con-
ditions and L=10'6 are- found to be 2 x 104, 1 x 102, and 7 x 107 hr, respec-
tively. Since th«se time constants are all significantly longer than the
- 38 -
-------
typical cloud lifetime of ~1 hr (Pruppacher and Klett, 1973) and also longer
than that for the gas-phase PAN-NO reaction, It is concluded that the atmo-
spheric lifetime of PAN Is not affected by these reactions.
Reactions
NOj has been examined extensively as an Intermediate of the nighttime
chemical transformations of nitrogen oxides (Winer et al . , 1984). This spe-
cies Is suggested to be not only the source of HNC>3 but also a free radical
initiator (Calvert and Stockwell, 1983), even for cloudwater droplets during
the daytime (Thompson, 1983). According to the studies made in this work,
no peroxide was found to be produced In the N02-03-H20 reaction system
regardless of whether H2CO was present. In the case of NO-j reacting with
water the upper limit for the reaction
k30 , v
N03 + H20 — -> N02 + H202 (30)
can be estimated by equating the assumed rate expression
Rate - dlH2,°Z] . k30[N03l (31)
dt
to the maximum possible rate estimated from the limit of detection of the
HRPF method, i.e. ,
k30[N03! < 4 x 10'11 M s'1 (32)
It should be noted that the presence of high concentrations of 03 (A3 ppm)
in these experiments might affect the aqueous-phase concentrations of
as it has been shown that their reaction can be repld. From the rate
- 39 -
-------
constants determined for the aqueous-phasa °j-^2°2 reaction the characteris-
tic reaction times of H202 for [03] = 4.7 x 10'7 M are 13 min and 930 min at
pH «» neutral and 2, respectively. Although the 0^-^2°2 reaction is unimpor-
tant at pH 2, it may keep the [t^C^] from building up at higher pH, If it
is assumed that a steady-state concentration of V?QI is established from the
competing actions of 03-11202 reaction and reaction (30), we may observe the
following relationship, i.e.,
k30
<">
For ths conditions applied in this work, the inequality. k3Q H^Q - _
1.2 x lO"*-" M s"*- is obtained. Using the previously calculated value for
PNQ , the product k3Q H^Q is estimated to he 4 x 10"^ M atm~* s~* or
smaller. A.lthough this value Is approximately three times greater than that
calculated from Eq . (32), the contribution of reaction (30) to the peroxide
production in the cloudwater is still negligible, being only 5 x 10~° M hr"*-
at PM03 = 3 PPb-
Tn the exsralnatlon of the aqueous-phase reaction of NC>3 with I^CO^ the
reaction between 03 and f^CO may consume the available I^CO. According to
Eq. (10), the rate of l^CO disappearance due to reaction (5) is calculated
to be 1.5 x 10~7 M hr~l for PQ - 43 ppm and [H2CO] « 1.0 x 10-4 M. This
rate is too small to Influence the aqueous-phase concentration of J^CO.
An upper-limit of the rate of the aqueous-phase reaction
k34
N03 + H2CO ---- > Peroxide (34)
can therefore be estimated from
- 40 -
-------
IH2CO) (35)
a
* k34 HH03 PN03 ("2C°J
£ 1.7 x ID'11 M s"1
using the limit of detection of HRPF, I.e., 1 x 10"7 M. The product, 1:34
HNQ , waa calculated to be smaller than 6 attn~l s"1. For the condition
where 5^0 = 3ppb and p^-CO = 20 PP^, the rate of peroxide formation in
the aqueous-phase is estimated to be 8 x 10~9 M hr~l. Again, this xvste Is
found to be unimportant. It should be noted that the NOj concentration used
here was calculated from the known kinetic Information and has not been
verified by an analytical method such as a spec trophotome trie technique.
Although our approach night suffer fron interferences such as wall effects
or the formation of ^Oj, the experimental arrangement nonetheless mimics
the bubbler ^02 sampling methodology which manifested the artifact produc-
tion of peroxide.
Aqueous-Phase Hg*^ Formation
Although it has been suggested that aqueous- phase reactions of Oj might
be responsible for the In situ production of H202 in the bubbler experi-
ments, the reactions examined In this work thus far did not provide evidence
to support this contention. The maximum rate of peroxide formation of the
reactions studied in this work, under typical atmospheric conditions, is
found to be smaller than I x 10"6 M hr~l. Naturally, the calculations were
made for pure water systems, and the possible effects of, say, metal ions
have not been included. This conclusion may not be at variance with the
observations reported by Heikes (1984), although the surface catalyzes 03
-------
decomposition leading to K202 production was not observed In our pure ays-
teas, where pure gases, instead of acbieat air, were used.
One reaction which might produce H202 in t'te aqueous phase is the
recombination of the dissolved HGj and its conjugate base G2~, i.e. ,
\
k36
H02 + 02' --->. K02" + 02 (36)
fast
- >H202
\
Using the Kerry's law coefficient of H02 (2 x 103 M atra'1) end rate constant
kjg (1 x 10° M"'- 8*1-) calculated fron a thermocheraical cycle (Schwartz,
198/+) and pulse-rndiolysis studies (Bielski, 1978), respectively, the
aqueous-phase rate of H202 formation can be calculated from
fl. IH021 (37)
+ *
dt [H+]
" k36* f!H022 PH022
where Ka is the acid dissociation constant of H02, being 2.0 x 10~5 M
(Bielski, 1978). Accordingly, the gas-phase reaction time constant of H02
apainst reaction (36) Is obtained as
TH02 ' <2k36' HH022 PK02 LRT)'1 (38)
This quantity, calculated as a function of P[jo2> is shown in Tigure 12.
Thesa values, calculated under the assumption that the aqueous solution Is
saturated with the solute H02 according to Henry's law, suggest that the
aqueous-phas« recombination of H02 Is potentially important as a source for
the aqueous-phase H202 when p^g. is equal to or greater than 2 x 10~lz
-------
atm. It should be noted that the calculations made here are limited to a
single pU, i.e., pH - pKft, for simplicity. Additionally, it should be
pointed out that a steady-state concentration of H02 has been assumed in
these calculations. If the consumption rate of 1102 °^ reaction (36) is sig-
nificantly faster than the rale of HC>2 production, ':hen the steady-state
as.vuaption aiay not apply asd the rate of H2(>2 production will be prisarily
controlled by the source '.rate of HC^-
The aqueous-phase production rate of ^02 from the gas-phase recombina-
tion of H(>2 is also shown in Figure 12. Here we used a second-order rate
constant of 2.5 x 10~^2 Cm3 molecule"*- s~* and have aasuiaed that the incor-
poration of the 11202 into the liquid phase is instantaneous. The liquid
water content is again assumed as 10"^. It 'i£~interesting to note that the
acquiring of the aqueous-phase ^02 from the reactions of dissolved H02 is
nearly two orders of magnitude faster than that can be supplied by the con-
comitant gas-phase recombination reaction of W^. Clearly, laboratory
studies have to be carried out to examine this potentially important source
of aqueous
-------
pH02,atm
Figure 12. Contribution of the aqu-eous-phase H02 reconbination reaction to the
generation of H202, calculated for pH « 4.7. Line A represents the
aqueous-phase rate of H202 production and line B the characteristic
reaction time for gaseous H02 at L = 10~6. Line C indicates the
contribution to aqueous H202 from the gas-phase recombination of H02',
H202 produced is assumed to be rapidly incorporated into the liquid
water (L - 1Q-&)
-------
SUMMARY
Aqueous-phase reactions of G3 with a number of inportant atmospheric
coaponents have been exaalned. Reaction klnntlcs &nd product analysis for
hydrogen peroxide arc! organic peroxldos were determined for tha following
reaction systems: (l) 03 - H20, (2) 0-j - »2°2 • ^ °3 " HC02H« ^4) °3 "
K2CO, (5) 03 - C2H4, (6) Q3 - PAN, (7) 03 - H02, ->nd (8) 03 - K02 - H2CO.
The results obtained are listed in Table 3. Using these values calculations
were made to estimate the atmospheric itaportance of th«se rencfciona to the %
\
renoval of Oi and the reagent species, as we 11 as the production cf per- \
oxides In the liquid-phase. For typical atmospheric conditions (po3 ~ 50
ppb, PJ^CQ ~ PHC02H ^ PC2»4 " 20 PPb> PPAN < 2 PPb> PN03 < 2 PPb-
L - 10~6, and pH " 2 ~ fc) it is concluded that none of these aqueoua-phtse
reactions la Important as a source of aqueous-phase peroxides or a sink for
gaseous 03 and its corresponding reagents.
- 45 -
-------
•Cable 3
Suasnary of tte Reaction Kinetics and Product Analysis
of Sane Aqueous-Phase Oj Reactions (22 ±
P-eactton Reaction d[0^} , d[Pertodde] Rate
Systta Oorditioa ~ ~St or StConsfea
03 - I'^O [03] £ 2 x 10-5 M ^[03] ^0.5% 2.1 x 1CT* s"1, {« -6
1.4 x 10"4 s"1, (H - 2
°3 ~ H2°2 l°3l I2 x 10"5 M k
&-&* x !Cr6 M 38 M"1 a"1, i« - 4
Og - HCO-^H [03] « [ICCKH] k [03] [HCOf] <.f--5% 4J x 103 M"1 s~l
- 1 x 1CT5 M jiD - 4.6
03 - H2CO [03] £2 x 1(T5 M k [(ty^IH^OOp2 ^2% 1.2 x !Cr3 s-l
[H2CO] « 4-17 x 10~5 M
p - 0.5 - 1.0 ppn k [(^nc^] 3.0 x 105 M"1 s"1
7~28 H" ^ ^
03 - PAN PpM = 100 ppb k [PAN] [03] £3 x 103 hf1 s'1
"03 "1.0 ppn
03-^ P(b-43pim k klx 10-3 Maori
03-^02 IH^J ° 1 x lO-4 M k HJJ^ [K2QO] pj^ k %>} £ 6 ataT1 s"1
-------
REFERENCES
Baulch, D.L., Cox, R.A., Crutzen, P.J., Hampson, Jr., R.F., Kerr, J.A.,
Troe, J. and Watson, R.T., Evaluated kinetic and photochemical data for
atmospheric chemistry. J. Phys. Chera. Ref. Data 11, 327-496 (1982).
Bielaki, 8.H.J., Revaluation of the spectra and kinetic properties of HC>2
and 02~ free radicals. Photochem. Photobiol. 28, 645 (1978).
Buhler, R.E., Staehelin, J., and Hoigne, J., Ozone decomposition in water
studied by pulse radiolysis. 1, H02/02~ and 1163/03" as intermediates.
J. Phys. Chem. 88_, 2560 (1«84).
Calvert, J.G. and Stockwell, W.R., Acid generation in the troposphere by
gas-phase chemistry. Environ. Sci. Tcchnol. 17, 428A (1983).
Danckwerts. P.V., Gas-liquid Reactions, McGraw-Hill Book Company, New York,
(1970).
Caffney, J.S. and Senum, G.I., Peroxides, peracids, aldehydes and PANs and
their links to natural and anthropogenic organic sources. In Pro-
ceedings of the "Conference on Gas-Liquid Chemistry of Natural Waters",
Brookhaven National Laboratory, Upton, New York, April, 1984.
Gaffney, J.S., Fajer, R., and Senum, G.I., An improved procedure for high
purity gaseous peroxyacyl nitrate production: Use of heavy liquid
solvents. Atmos. Environ. 18, 215 (1984).
Gilbert, E., Reaction of ozone with trans-trans muconic acid in aqueous
solution. Water Research 14, 1637 (1980).
Grosjean, D., Formaldehyde and other carbonyls in Los Angeles ambient air.
Environ. Sci. Tech. 16^, 254 (1982).
Guilbault, G.G., Brignac, P.J., Jr., and Juneau, M., New substrates for the
fluorometric determination of oxldative enzymes. Anal. Chem. 40_, 1256
(1968).
- 47 -
-------
Gurol, M.D. and Singcu', P.C,S Kinetics of ozone decomposition: n dynamic
approach. Evirim. Sci. Techno.!. 16, 377 (1982).
Heikes, B., Aqueous H;;02 production from 03 in glass impingers. Atmos.
Environ. _18, U'A'i (1984).
Heikes, B.C. and ThoK[;won, A.M., Effects of heterogeneous processes on ^3,
HONO, and HN03 cheatistry in the troposphere. J. Geophys. Res. Rj^,
10,883 (1983).
Heikes, B.C., Lazrua, A.L., Kok, G.L., Kunen, S.M., Gandrud, B.W., Uttlin,
S.N., and Sperry, P.O., Fvridence for aqueous phase hydrogen p0fv>xide
synthesis in tha iropoephere. J. Geophys. Res. 87, 3045 (1? ozone
in aqueous solu'. ton. J. Amer. Chem. Soc. 78, 1784 (1956).
- 48 -
-------
Kosak-Charming, L.F. and Helz., G.R. , Solubility of ozone in aqueous solu-
tions of 0-0.6 M ionic strength at 5-30°C. Environ. Sci. Technol. 17,
145 (1983).
Kuo, C.H., and Wen, C.P., Oaonations of formic acid, formaldehyde, and
nethanol in aqueous solutions. AIChE Sympsoium, Series No. 166, 73,
272 (1977).
Lazrus, A.L., Kok, G.L., Lind, J.A., and Sperry, P.O., A fluorometric tech-
nique for H202 in precipitation and in cloudwater. EOS Trans. Am.
Geophys. Union 64_, 670 (1983).
Lee, Y.-N. and Schwartz, S.E., Evaluation of the rate of uptake of nitrogen
dioxide by atmospheric water and surface liquid water, J. Geophys.
Res. 86, 11971 (1981).
Lee, Y.-N. and Schwartz, 3.E., Reaction kinetics of nitrogen dioxide with
liquid water at low partial pressures, J. Phys. Chem. 85, 840 (1981).
Lfte, Y.-N., Kinetics of sotse aqueous-phase reactions of peroxyacetyl
nitrate. In Proceedings of "The Conference on Gas-Liquid Chemistry of
Natural Waters," Brookhaven National Laboratory, Upton, New York, April
1984.
Martin, L.R., Measurements of sulfate production in natural clouds. Atraos.
Environ. l]_, 1603 (1983).
National Academy of Sciences, Vapor-Phase Organic Pollutants, Committee on
Medical and Biological Effects of Environmental Pollutants, Washington,
D.C., 1976.
Nielsen, T., Hansen, A.M., and Thornsen, E.L., A convenient method for prep-
aration of pure standards of peroxyacetyl nitrate for atraospheric
analyses. Ar.mos. Environ. 16>, 2447 (1^82).
- 49 -
-------
Nikl, H,, Maker, P.O., Savage, C.M., and Breiteabaeh, L.P., Atmospheric
ozone-olefin reactions, Environ. Sci. Technol. 17, 312A (1983).
Penkett, S.A., Jcnes, M.R., Brice, K.A., and Eggleton, A.E.J., The Impor-
tance of atmospheric osone and hydrogen peroxide ia oxidizing sulfur
dioxide in cloud and rainwater. Atmos. Environ. 13, 123 (1979).
Pitts, J.N., Jr., Biermann, H.W., Atkinson, R., and Winer, A.M., Atmospheric
implications of simultaneous nighttime measurements of 1^)3 radicals and
HONO. Geophys. Res. Letts. U_, 557 (1984).
Pruppacher, H.R. and Klett, J.D., Microphystcs of Clouds and Precipitation,
D. Reidel Publishing Co., Boston, 1978.
Piddick, J.A. and Hunger, W.B., Techniques of Chemtctry, Vol. II; Organic
Solvents. Third ed., Wiley-Intersclence, New York, 1970.
Schonbaura, G.R. and Chance, B., Catalase. In The Enzymes, P.D. Boyer, ed.,
Vol. XIII, pp. 363-403, Academic Press, New York, 1976.
Schwartz, S.E., Cas~ arid aqueous-phase chetaistry of IK>2 i^ liquid-water
clouds, J. Geophys. Res. 89_. 11589 (1984).
Senuia, G.I. and Gaffney, J.S., Private communication (1984).
Sillen, L.G., and Martell, A.E., Stability Constants of Metal-Ion Com-
plexes. The Chemical Society, London, Special Publication No. 17,
1964.
Sullivan, D.E. and Roth, J.A., Kinetics of ozone self-decomposition in aque-
ous solution. AIChE Symposium Series, Water, 1979, PP. 142-149.
Tanner, R.L. and Meng, Z., Seasonal variations in ambient atmospheric levels
of formaldehyde and acetaldehyde. Environ. Sci. Technol. 18, 723
(1984).
- 50 -
-------
Taube, H. and Bray, V.C., Chain reactions IP aqueous solutions containing
ozone, hydrogen peroxide, and acid. J. Araer. Chera. Soc. 62, 3357
(1940).
ten Brink, H.H., K«lly, T. J., Lee, Y.-N., and Schwartz, S.E., Atteapted
measurement of gaseous 1*2^2 *n ^e ambient atmosphere. In Proceedings
of the Third European Sympsolura on the Physico-Chemical Behavior of
Atmospheric Pollutants, Varese, Italy, April H-13, 1984.
Thompson, A.M. Night-time oxidant chemistry as a source of free radicals In
cloudwater. Presented at the 5th CACGP Conference: Sympsolura on Trop-
ospherlc Chemistry, Oxford, UK, Aug. 28-Sept. 3, 1983.
Wllhelm, E., R. Battlno, and R.J. Wlllcock, Low pressure solubility of gases
in liquid water. Chera. Rev. 77^ 219 (1977).
Winer, A.M., Atkinson, R., and Pitts, J.M., Jr., Gaseous nitrate radical:
possible nighttime atmospheric sink for biogenlc organic compounds.
Science 22'-, 156 (1984).
- 51 -
-------
Page Intentionally Blank
------- |