Product study of the C2H5O2+HO2 reaction in 760 Torr of air at 284-312 K

Article abstract of DOI:10.1016/S0009-2614(00)00315-8

The title reaction was studied in 1 atmosphere pressure of synthetic air at 284, 298, and 312 K. FTIR spectroscopy was used to identify two products; ethyl hydroperoxide and acetaldehyde. Acetaldehyde was a minor product formed in a yield consistent with that expected from unwanted secondary reactions. In 1 atmosphere pressure of air at 284-312 K the title reaction produces ethyl hydroperoxide in a yield of (104±5)% and acetaldehyde in a yield of (2±4)%. We recommend that in atmospheric models the reaction of C₂H₅O₂ with HO₂ radicals be represented by a single-reaction pathway giving C₂H₅OOH.

Full text of DOI:10.1016/S0009-2614(00)00315-8

2
1 April 2000
Chemical Physics Letters 321 Ž2000. 57–61
www.elsevier.nlrlocatercplett
Product study of the C H O qHO reaction in 760 Torr of air at
2
5
2
2
2
84–312 K
a
a
a
b,)
M. Spittler , I. Barnes , K.H. Becker , T.J. Wallington
a
Physikalische Chemie, Fachbereich 9, Bergische UniÕersit a¨ t–GH Wuppertal, Gausstrasse 20, 42097 Wuppertal, Germany
b
Ford Research Laboratory, SRL-3083, Ford Motor, P.O. Box 2053, Dearborn, MI 48121-2053, USA
Received 24 January 2000; in final form 29 February 2000
Abstract
The title reaction was studied in 1 atmosphere pressure of synthetic air at 284, 298, and 312 K. FTIR spectroscopy was
used to identify two products; ethyl hydroperoxide and acetaldehyde. Acetaldehyde was a minor product formed in a yield
consistent with that expected from unwanted secondary reactions. In 1 atmosphere pressure of air at 284–312 K the title
reaction produces ethyl hydroperoxide in a yield of Ž104"5.% and acetaldehyde in a yield of Ž2"4.%. We recommend
that in atmospheric models the reaction of C H O with HO radicals be represented by a single-reaction pathway giving
2
5
2
2
C H OOH. q 2000 Published by Elsevier Science B.V. All rights reserved.
2
5
1
. Introduction
mental focus is shifting from local air quality issues
to a consideration of impacts on wider geographic
scales Žregional, national, and global.. Global atmo-
Peroxy radicals play a key role in atmospheric
chemistry. They are intimately involved in the for-
mation and destruction of ozone and in the photo-
oxidation of all organic compounds in the atmo-
sphere. In urban areas organic peroxy radicals, RO2 ,
and HO radicals react with NO and are part of the
spheric chemistry models are under development to
assess the impact of man-made emissions on the
global atmospheric environment. To provide reliable
results, such models must incorporate an accurate
description of peroxy radical chemistry. On local
2
x
chemical cycles that produce unwanted photochemi-
cal air pollution, ‘smog’. Recognition of the key role
of peroxy radical chemistry in the formation of
scales near to NO sources such as large urban areas,
x
the fate of organic peroxy radicals is reaction with
NO . On regional and global scales where the NO
x
x
‘
smog’ has led to many kinetic and mechanistic
studies of the reactions of peroxy radicals with NOx
the gross features of which are now reasonably
concentrations are lower the chemistry of organic
peroxy radicals is no longer dominated by reaction
–
with NO . Loss of organic peroxy radicals via reac-
x
well understood w1x. At the present time the environ-
tion with HO2 radicals become important in air
masses remote from pollution sources. Unfortu-
nately, our understanding of the reactions of organic
)
peroxy, RO , with HO radicals is far from com-
Corresponding author. Fax: q1-313-594-2923; e-mail:
2
2
twalling@ford.com
plete, and incorporation of these reactions into com-
0
009-2614r00r$ - see front matter q 2000 Published by Elsevier Science B.V. All rights reserved.
PII: S0009-2614Ž00.00315-8

5
8
M. Spittler et al.rChemical Physics Letters 321 (2000) 57–61
puter models introduces a significant degree of un-
certainty into the model results.
long-path mirror system and a Bruker IFS 88 FTIR
spectrometer w15x. C H O and HO radicals were
2
5
2
2
Mechanistic information for reactions of RO with
generated by the near-UV irradiation of C H r
2
2
6
HO2 radicals is largely confined to room tempera-
ture. Such reactions can proceed via two reaction
channels giving either hydroperoxide or carbonyl
products. The hydroperoxide channel is dominant for
reactions of unsubstituted peroxy radicals Že.g.,
CH O w2x, CD O w3x, C H O w4x, cyclo-C H O
CH OHrCl mixtures in 1000 mbar of air diluent at
3
2
284, 298, or 312 K.
Cl qhn 2Cl
Ž
Ž
Ž
Ž
Ž
2
3
4
5
6
.
.
.
.
.
2
ClqC H
C H qHCl
2
6
2
5
3
2
3
2
2
5
2
5
9
2
w5x, cyclo-C H O w5x, and ŽCH . CCH O w6x.
6
11
2
3
3
2
2
C H qO qM C H O qM
2
5
2
2
5
2
with HO . The carbonyl channel is generally impor-
2
tant, and can be dominant, for peroxy radicals carry-
CH
3
OHqCl CH
2
OHqHCl
ing electron withdrawing substituents Že.g., CH FO
2
2
CH OHqO
HO qHCHO
w7x, CH ClO w8x, CHCl O w9x, CH OCH O w10x..
2
2
2
2
2
2
2
3
2
2
There is no information concerning the effect of
temperature on the branching ratio of such reactions.
The kinetics of the reaction of C H O with HO
The decay of ethane and the formation of products
were analyzed using FTIR spectroscopy with an
optical pathlength of 484.7 m and a spectral resolu-
2
5
2
2
radicals have been the subject of 4 separate studies
y1
tion of 1 cm . Infrared spectra were derived from
w11–14x and are reasonably well established. The
6
4 co-added interferograms. Ethane Ž99.95%. and
overall reaction proceeds with a rate constant of
synthetic air were purchased from Messer–Gries-
sheim. The air was dried using a Messer–Griessheim
Hydrosorb absorption tube. Methanol was purchased
from Fluka and had a stated purity of )99.8%.
Reference spectra of C H , CH OH, and CH CHO
y12
3
y1
y1
approximately 8=10
room temperature. Measurements between 210 and
80 K show that the reaction has a negative tempera-
cm molecule
s
at
4
ture dependence with reported E rR values of be-
a
2
6
3
3
tween y650 K and y1260 K. The kinetic studies
suggest, but do not prove, that the reaction occurs by
a single mechanism over the range of atmospheri-
cally relevant temperatures Ž200–300 K.. The two
possible products of reaction Ž1. are ethylhydroper-
were acquired by introducing authentic samples into
the reaction chamber using calibrated volumes. The
reference spectrum of ethyl hydroperoxide was taken
y1
y18
from the library at Ford, s Ž842 cm .s3.3=10
2
y1
cm molecule . Reactants and products were moni-
oxide and acetaldehyde:
tored by comparing calibrated reference spectra over
y1
C H O qHO
C H OOHqO
Ž
Ž
1a
1b
.
.
the range 700–900 cm
with the recorded spectra.
2
5
2
2
2
2
5
2
Initial concentrations of the gas mixtures were:
C H O qHO
CH CHOqH OqO
2
5
2
3
2
2
14
14
wC H xs2, 4 or 8=10 , wCH OHxs7–8=10 ,
2
6
3
In 20y700 Torr of air at 295 K it has been reported
14
y3
and wCl xs4=10 molecule cm
of air. The mixtures were irradiated for 4–21 min
resulting in ethane consumptions of 6–36%.
in 1000 mbar
2
that the reaction of C H O qHO produces ethyl-
2
5
2
2
hydroperoxide in a yield of 102"6% w4x indicating
that under these conditions reaction Ž1. proceeds
predominately, if not exclusively, via pathway Ž1a..
To study the effect of temperature we have used
FTIR – smog chamber techniques to measure the
products of reaction Ž1. at 284–312 K in 1000 mbar
3. Results
of air. Results are reported herein.
Following the UV irradiation of C H rCH OHr
2
2
6
3
Cl rair mixtures reactions Ž1. and Ž7. com-
pete for the available C H O radicals.
2
5
2
2
. Experimental
C H O qHO
products
products
2
Ž
Ž
1
7
.
.
2
5
2
2
Experiments were performed in Wuppertal using
a 1080-liter quartz glass reactor equipped with a
C H O qC H O
2
5
2
2
5

M. Spittler et al.rChemical Physics Letters 321 (2000) 57–61
59
Fortunately, the rate constant for reaction Ž1. is more
than two orders of magnitude larger than that for
y12
3
y1 y1
reaction Ž7. Žk s8=10
7
cm molecule
s
;
1
y14
3
y1 y1
k s5.9=10
cm molecule
s
w16x at 298
K. and it is relatively straightforward to establish
conditions where reaction Ž1. is the dominant fate of
C H O radicals. In the present work three sets of
2
5
2
experiments Žusing wCH OHx rwC H x s1, 2, or 4.
3
o
2
6 o
were performed at each of three different tempera-
tures Ž284, 298, or 312 K.. Using the Acuchem
chemical kinetic modeling package w17x with the
chemical mechanism given in Table 1 it can be
calculated that for experiments employing wCH -
3
OHx rwC H x s1, 2, or 4, reaction with HO radi-
o
2
6
o
2
cals accounts for 90.7, 98.6 or 99.6% of the loss of
the C H O radicals, respectively.
2
5
2
Fig. 1 shows typical IR spectra acquired before
ŽA. and after ŽB. a 6.5 min irradiation of a mixture
1
4
y3
containing 3.65=10 molecule cm
C H , 6.92
2 6
1
4
y3
14
=
10
molecule cm
CH OH, and 4=10
3
y3
molecule cm
Cl2 in 1000 mbar of air at 284 K.
Panel C shows the result of subtracting features
attributable to C H from panel B. Comparison of
2
6
panel C with the reference spectra of C H OOH and
Fig. 1. IR spectra acquired before ŽA. and after ŽB. a 6.5 min
2
5
1
4
y3
CH CHO given in panels D and E shows the forma-
irradiation of a mixture containing 3.65=10 molecule cm
3
1
4
y3
14
C H , 6.92=10
molecule cm Cl2 in 1000 mbar of air at 284 K. Panel C shows
the result of subtracting features attributable to C H from panel
molecule cm
CH OH, and 4=10
3
tion of these products. As with all product studies,
careful attention needs to be paid to possible sec-
ondary chemistry in the system. To check for possi-
ble heterogeneous losses of C H OOH and
2
6
y3
2
6
B. Reference spectra of C H OOH and CH CHO are given in
2
5
3
panels D and E.
2
5
CH CHO in the chamber, reaction mixtures were
3
allowed to stand in the dark at 284 and 298 K for 30
observed formation of C H OOH and CH CHO
2 5 3
min; there was no observable loss Ž-2%. of either
versus the loss of C H following the irradiation of
2 6
C H OOH and CH CHO. Fig. 2 shows a plot of the
C H rCH OHrCl rair mixtures.
2
5
3
2
6
3
2
Table 1
Reaction mechanism
3
y1
y1x
Reaction
Rate constant wcm molecule
s
Reference
y4 y1
Cl qhn 2 Cl
6.0=10
5.7=10
5.4=10
8.0=10
4.0=10
1.9=10
8.0=10
2.9=10
7.8=10
9.8=10
9.0=10
s
This work
w18x
2
y11
y11
y12
y14
y14
y15
y12
y11
y11
y12
ClqC H qO
C H O qHCl
2
2
6
2
2
5
2
CH OHqCl HO qHCHO
w19x
3
C H O qHO
C H OOHqO
w16x
2
5
2
2
2
5
2
C H O qC H O
2 C H OPqO
w16x
2
5
2
2
5
2
2
5
2
C H O qC H O
CH CHOqC H OH
w16x
2
5
2
2
2
5
2
3
2
5
C H O qO
CH CHOqHO
w20x
2
5
2
3
2
HO qHO
H O qO
w18x
2
2
2
2
2
ClqCH CHO products
w19x
3
a
ClqC H OOH CH CHOqproducts
w21,22x
2
5
3
a
ClqC H OOH other products
w21,22x
2
5
a
Kinetic data from w21x, mechanistic data by analogy to C H OH w22x.
2
5

6
0
M. Spittler et al.rChemical Physics Letters 321 (2000) 57–61
Fig. 3 shows the result of taking the data in Fig. 2
correcting the observed C H loss for that fraction
2
6
which does not react with HO radicals and correct-
2
ing the observed C H OOH and CH CHO product
2
5
3
yields for the effect of secondary reactions with Cl
atoms. At this point it should be stressed that the
chemical mechanism in Table 1 used to compute the
corrections is taken entirely from the literature and
has not been adjusted in anyway to ‘fit’ the data
trend shown in Fig. 2. It is gratifying that corrections
calculated using the chemical mechanism in Table 1
account for the curvature in Fig. 2 and result in the
linear plots shown in Fig. 3. As seen from Fig. 3
within the experimental uncertainties there is no
discernable effect of temperature, wCH OHx r
3
o
wC H x concentration ratio, or consumption of C H
2
6
o
2
6
upon the product yields. The lines through the data
in Fig. 3 are linear least squares fits which give
yields of Ž104"5.% for C H OOH and Ž2"4.%
Fig. 2. Plot of the observed formation of C H OOH Žfilled
2
5
symbols. and CH CHO Žopen symbols. versus the observed loss
3
of C H , curves are second order fits to aid visual inspection of
2
5
2
6
for acetaldehyde and provide values of k rŽk q
the data trends.
1a
1a
k . s 1.04 " 0.05 and k rŽk q k . s 0.02 "
1
b
1b
1a
1b
The product data in Fig. 2 contain information
concerning the branching ratio for reaction Ž1.. To
0.04 at 284–312 K. Quoted errors are 2 standard
deviations from the linear least-squares analysis.
extract such information appropriate allowance must
be made for two factors. First, it should be recog-
nized that while the loss of C H provides an
2
6
accurate value for the amount of C H O radicals
2
5
2
formed in the system, not all of these C H O
2
5
2
radicals react with HO . As discussed above for the
2
experiments using wCH OHx rwC H x s1, 2, or 4,
3
o
2
6 o
reaction with HO radicals accounts for 90.7, 98.6,
2
or 99.6% of the loss of the C H O radicals. Hence,
2
5
2
the observed loss of C H must be multiplied by
2
6
either 0.907, 0.986, or 0.996 to obtain the amount of
C H O radicals which are lost via reaction Ž1..
2
5
2
Second, and more importantly, the effect of sec-
ondary reactions of Cl atoms with C H OOH and
2
5
CH CHO needs to be considered. The curvature
3
shown in the yield plots in Fig. 2 indicate that
C H OOH is consumed and CH CHO is produced
2
5
3
by secondary reactions in the system. It is well
established that Cl atoms react rapidly with both
C H OOH and CH CHO. Fortunately, the kinetics
2
5
3
and mechanism of these reactions Žand of the com-
Fig. 3. Plot of the observed formation of C H OOH Žfilled
2
5
symbols. and CH CHO Žopen symbols. corrected for secondary
peting losses of Cl atoms via reaction with C H
3
2
6
reactions involving Cl atoms Žsee text. versus the observed loss of
and CH OH. are well understood and their effect
3
C H , corrected for the fraction which does not react with HO
2
6
2
can be explicitly calculated for each experiment us-
radicals. Experiments were conducted in 1000 mbar of air diluent
ing the chemical mechanism in Table 1.
at either 284 Žsquares., 298 Žcircles., or 312 K Žtriangles..

M. Spittler et al.rChemical Physics Letters 321 (2000) 57–61
61
w2x T.J. Wallington, S.M. Japar, Chem. Phys. Lett. 167 Ž1990.
13.
w3x T.J. Wallington, M.D. Hurley, Chem. Phys. Lett. 193 Ž1992.
4.
w4x T.J. Wallington, S.M. Japar, Chem. Phys. Lett. 166 Ž1990.
95.
4
. Discussion
5
The aim of the present work was to provide
8
mechanistic information to facilitate the accurate
representation of reactions of alkyl peroxy radicals
4
with HO radicals in atmospheric models. Within the
w5x D.M. Rowley, R. Lesclaux, P.D. Lightfoot, B. Noziere, T.J.
2
`
Wallington, M.D. Hurley, J. Phys. Chem. 96 Ž1992. 4889.
admittedly limited temperature range employed in
the present work, there was no evidence for any
effect of temperature on the products, and hence
mechanism, of reaction Ž1.. For purposes of atmo-
w6x D.M. Rowley, R. Lesclaux, P.D. Lightfoot, K. Hughes, M.D.
Hurley, S. Rudy, T.J. Wallington, J. Phys. Chem. 96 Ž1992.
7
043.
w7x T.J. Wallington, M.D. Hurley, W.F. Schneider, J. Sehested,
O.J. Nielsen, Chem. Phys. Lett. 218 Ž1994. 34.
w8x T.J. Wallington, M.D. Hurley, W.F. Schneider, Chem. Phys.
Lett. 251 Ž1996. 164.
spheric modeling, we recommend that the reaction of
C H O with HO radicals should be represented by
2
5
2
2
a single-reaction pathway giving hydroperoxide at all
temperatures. A similar recommendation has been
w9x V. Catoire, R. Lesclaux, W.F. Schneider, T.J. Wallington, J.
Phys. Chem. 100 Ž1996. 14356.
made recently for the reaction of CH O with HO
3
2
2
w10x T.J. Wallington, M.D. Hurley, J.C. Ball, M.E. Jenkin, Chem.
Phys. Lett. 211 Ž1993. 41.
radicals w16x. In the absence of information to the
contrary, it seems reasonable to extend this recom-
mendation to the reaction of all unsubstituted alkyl
w11x F.C. Cattell, J. Cavanagh, R.A. Cox, M.E. Jenkin, J. Chem.
Soc. Faraday II 82 Ž1986. 1999.
w12x P. Dagaut, T.J. Wallington, M.J. Kurylo, J. Phys. Chem. 92
Ž1988. 3836.
peroxy radical reactions with HO . It should be
2
noted that substituted peroxy radicals such as
w13x F.F. Fenter, V. Catoire, R. Lesclaux, P.D. Lightfoot, J. Phys.
CH ClO , CH CŽO.O , and CH OCH O are well
Chem. 97 Ž1993. 3530.
2
2
3
2
3
2
2
w14x M.M. Maricq, J.J. Szente, J. Phys. Chem. 98 Ž1994. 2078.
w15x I. Barnes, K.H. Becker, N. Mihalopoulos, J. Atmos. Chem.
known to react with HO via two reaction channels.
2
Further work is needed to establish the temperature
dependence of the branching ratios of such reactions.
1
8 Ž1994. 267.
w16x G.S. Tyndall, R.A. Cox, R. Lesclaux, G. Moortgat, M.J.
Pilling, A.R. Ravishankara, T.J. Wallington, J. Geophys.
Res., submitted.
Acknowledgements
w17x W. Braun, J.T. Herron, D.K. Kahaner, Int. J. Chem. Kinet.
2
0 Ž1988. 51.
T.J.W. thanks the Alexander von Humboldt ŽAvH.
Stiftung for an AvH fellowship, M.S. thanks the
Landesamt f u¨ r Graduiertenf o¨ rderung NRW’ for a
scholarship.
w18x W.B. DeMore, S.P. Sander, D.M. Golden, R.F. Hampson,
M.J. Kurylo, C.J. Howard, A.R. Ravishankara, C.E. Kolb,
M.J. Molina, Jet Propulsion Laboratory Publication 97-4,
Pasadena, CA, 1997.
‘
w19x G.S. Tyndall, J.J. Orlando, C.S. Kegley Owen, T.J. Walling-
ton, M.D. Hurley, Int. J. Chem. Kinet. 31 Ž1999. 776.
w20x D. Gutman, N. Sanders, J.E. Butler, J. Phys. Chem. 86
Ž1986. 63.
References
w21x T.J. Wallington, J.M. Andino, J.C. Ball, S.M. Japar, J.
w1x T.J. Wallington, O.J. Nielsen, J. Sehested, in: Z.B. Alfassi
ŽEd.., Peroxyl Radicals, John Wiley and Sons, Chap. 7,
Ž
.
Atmos. Chem. 10 1990 301.
w22
x
C.A. Taatjes, L.K. Christensen, M.D. Hurley, T.J. Walling-
ton, J. Phys. Chem. A. 103 Ž1999. 9805.
1
997.