Fluorescence excitation spectrum of the 2-butoxyl radical and kinetics of its reactions with NO and NO2

Article abstract of DOI:10.1039/b101628h

The (A ← X) fluorescence excitation spectrum of the 2-C₄H₉O(X) (2-butoxyl) radical in the wavelength range 345-390 nm was obtained using a combined laser photolysis/laser-induced fluorescence (LIF) technique following the generation of the radicals by excimer laser photolysis of 2-butylnitrite at λ = 351 nm. The fluorescence excitation spectrum shows 5 vibronic bands, where the dominant progression corresponds to the CO-stretching vibration in the first electronically excited state with v′_CO = (560 ± 10) cm^-1. The transition origin was assigned at v₀₀ = (26768 ± 10) cm^-1 (λ₀₀ = (373.58 ± 0.15) nm). The kinetics of the reactions of the 2-butoxyl radical with NO and NO₂ at temperatures between T = 223-305 K and pressures between p = 6.5-104 mbar have been determined. The rate coefficients for both reactions were found to be independent of total pressure with k_NO = (3.9 ± 0.3) × 10^-11 cm³ s^-1 and k_NO2 = (3.6 ± 0.3) times; 10^-11 cm³ s^-1 at T = 295 K. The Arrhenius expressions have been determined to be k^NO = (9.1 ± 2.7) × 10^-12 exp((3.4 ± 0.6) kJ mol^-1/RT) cm³ s^-1 and k^NO2 = (8.6 ± 3.3) × 10^-12 exp((3.3 ± 0.8) kJ mol^-1/RT) cm³ s^-1. In addition, the radiative lifetime of the 2-C₄H₉O(A) radical after excitation at λ = 365.938 nm in the (0,1) band has been determined to be τ^rad(2-C₄H₉O(A)) = (440 ± 80) ns. Quenching rate constants of the 2-C₄H₉O(A) radical were measured to be k_q = (4.7 ± 0.3) × 10^-10 cm³ s^-1 and k_q = (5.0 ± 0.4) × 10^-12 cm³ s^-1 for 2-butylnitrite and nitrogen, respectively.

Full text of DOI:10.1039/b101628h

View Article Online / Journal Homepage / Table of Contents for this issue
Fluorescence excitation spectrum of the 2-butoxyl radical and
kinetics of its reactions with NO and NO
2
Ch. Lotz and R. Zellner
Institut fur Physikalische und T heoretische Chemie, Universitat Essen, D-45117 Essen, Germany
Received 19th February 2001, Accepted 1st May 2001
First published as an Advance Article on the web 11th June 2001
The (A ^ X) Ñuorescence excitation spectrum of the 2-C H O(X) (2-butoxyl) radical in the wavelength range
4
9
345È390 nm was obtained using a combined laser photolysis/laser-induced Ñuorescence (LIF) technique
following the generation of the radicals by excimer laser photolysis of 2-butylnitrite at j \ 351 nm. The
Ñuorescence excitation spectrum shows 5 vibronic bands, where the dominant progression corresponds to the
CO-stretching vibration in the Ðrst electronically excited state with l@ \ (560 ^ 10) cm~1. The transition
CO
origin was assigned at l \ (26 768 ^ 10) cm~1 (j \ (373.58 ^ 0.15) nm). The kinetics of the reactions of the
00
00
2-butoxyl radical with NO and NO at temperatures between T \ 223È305 K and pressures between
2
p \ 6.5È104 mbar have been determined. The rate coefficients for both reactions were found to be independent
of total pressure with k \ (3.9 ^ 0.3) ] 10~11 cm3 s~1 and kNO2 \ (3.6 ^ 0.3) ] 10~11 cm3 s~1 at T \ 295
NO
K. The Arrhenius expressions have been determined to be k \ (9.1 ^ 2.7) ] 10~12 exp((3.4 ^ 0.6)
NO
kJ mol~1/RT ) cm3 s~1 and kNO2 \ (8.6 ^ 3.3) ] 10~12 exp((3.3 ^ 0.8) kJ mol~1/RT ) cm3 s~1. In addition, the
radiative lifetime of the 2-C H O(A) radical after excitation at j \ 365.938 nm in the (0,1) band has been
4
9
9
determined to be qrad(2-C H O(A)) \ (440 ^ 80) ns. Quenching rate constants of the 2-C H O(A) radical were
4
4 9
measured to be k \ (4.7 ^ 0.3) ] 10~10 cm3 s~1 and k \ (5.0 ^ 0.4) ] 10~12 cm3 s~1 for 2-butylnitrite and
q
q
nitrogen, respectively.
e†ect on the tropospheric production of ozone during photo-
chemical smog episodes.
1
Introduction
Alkoxyl radicals (RO) are key intermediates in the oxidative
degradation of volatile organic compounds (VOCs) in the
troposphere1 as well as in hydrocarbon combustion chem-
istry. It is well known that alkoxyl radicals show four distinct
competitive pathways under tropospheric conditions depend-
ing on their structure and environmental as well as exprimen-
tal conditions:2 (i) reaction with O , yielding an aldehyde or a
ketone and HO , (ii) unimolecular decomposition, forming an
aldehyde and an alkyl radical, (iii) isomerization via an n-
membered transition state (1,(n [ 1)-H-shift; n \ 5, 6) and (iv)
recombination reaction with NO . While primary and sec-
ondary alkoxyl radicals up to C mainly react with O , isom-
erization and unimolecular decomposition become more
dominant with increasing number of carbon atoms. The reac-
Although alkoxyl radicals represent a major branching
point in the VOC oxidation schemes, only a few kinetic data
are available in the literature. Our present understanding of
their kinetics is mainly based on product yields and/or relative
studies. Only a few direct studies3h24 have been published, in
which the highly sensitive LIF technique was used to detect
the simplest alkoxyl radicals, methoxyl,10h21 ethoxyl5,19h23
and n-/iso-propoxyl.5h7,24 The knowledge of the Ñuorescence
excitation spectra of these radicals has been extremely useful
for direct kinetic investigations. For a long time the lack of
2
2
these spectra for higher (PC ) alkoxyl radicals hindered direct
x
4
kinetic studies on these radicals. In the last two years,
3
2
tions with NO and NO may also become important for
2
alkoxyl radicals in highly polluted environments, but are very
important in the interpretation of smog chamber experiments
due to the much higher concentrations of these gases. For the
2-butoxyl radical these reactions are schematically shown in
Fig. 1. Note, however, that 2-butoxyl radicals cannot isomer-
ize via a strain-free six-membered transition state. Hence
isomerization in Fig. 1 is only shown for the sake of complete-
ness.
From Fig. 1 it is obvious that while reactions with O and
2
NO are chain terminating, decomposition (and iso-
x
merization) lead to the formation of carbon centered radicals.
These radicals are further oxidized, i.e. they rapidly add O
2
and react with NO to form another alkoxyl radical and NO .
2
Hence the kinetics of these di†erent possible reaction path-
ways have signiÐcant e†ects on the number of NO to NO
conversions that take place during the oxidation of the corre-
Fig. 1 Possible reactions of 2-butoxyl radicals under atmospheric
conditions. The 1,4-H-shift isomerisation is only shown to illustrate
an isomerisation pathway. Under atmospheric conditions it will not
occur for 2-butoxyl.
2
sponding parent hydrocarbon. As NO is easily photolysed,
2
the branching ratio of alkoxyl radical reactions has a direct
DOI: 10.1039/b101628h
Phys. Chem. Chem. Phys., 2001, 3, 2607È2613
This journal is ( The Owner Societies 2001
2607

View Article Online
however, several groups3,4,25h27 have obtained Ñuorescence
excitation spectra of the tert-butoxyl radical generated by
laser photolysis of tert-butylnitrite at 351/355 nm or of di-tert-
butylperoxide at 248 nm. Its unimolecular decomposition3,25
in 10 or 20 l glass bulbs in the dark. All other gases were
taken from pressurized gas cylinders.
2-Butylnitrite was prepared by dropwise addition of con-
centrated sulfuric acid to a saturated solution of NaNO and
2
as well as its reactions with NO3,26 and NO 26 have been the
butan-2-ol.30 2-Butylnitrite appears as a pale yellow liquid,
2
subject of direct kinetic studies using the LIF technique. Even
which was further puriÐed by several freezeÈpumpÈthaw
cycles followed by trap to trap distillation. IR and UV spectra
were used to verify the identity and purity of 2-butylnitrite.
Other gases and chemicals used had the following purities
stated by the manufacturer (Messer Griesheim, if not other-
wise stated): N , 99.999%; gas mixture NO, 99.8% in N ,
more recently, vibrationally resolved Ñuorescence excitation
spectra have been obtained for 1-butoxyl,27 2-butoxyl,3,27 3-
pentoxyl28 and tert-pentoxyl.28 Although these spectra
provide a convenient tool for direct kinetic studies, there has
been only one study by Deng et al.29 in which the kinetics of
2 2
99.999%; gas mixture NO , 98% in N , 99.999%. All gases
the 2-butoxyl radical with O and NO has been addressed.
2
2
2
In the present work we present the Ñuorescence excitation
were used without further puriÐcation.
spectrum of 2-butoxyl radicals generated by the photolysis of
2-butylnitrite at 351 nm as well as results of direct tem-
perature and pressure dependent measurements of the rate
3
Results and discussion
3.1 LIF detection of 2-butoxyl radicals
constants of the reaction of 2-butoxyl with NO and NO . To
2
our knowledge this is the Ðrst direct study of the reaction of
Fig. 2 shows the Ñuorescence excitation spectrum of 2-butoxyl
(A ^ X) at T \ 295 K in the range 345È390 nm as obtained in
the excimer laser photolysis at 351 nm of 1 ] 1015 cm~3 2-
butylnitrite in 13 mbar nitrogen. The delay time between the
photolysis laser pulse (\80 mJ pulse~1) and the probe laser
pulse (\8 mJ pulse~1) was set to 10 ls. The spectrum was
corrected for background and normalized for laser power
variations. From the concentration of the photochemical pre-
cursor, an absorption cross section of p B 1.1 ] 10~19 cm231
at 351 nm, the photolysis laser Ñuence of 40 mJ cm~2 as well
as an assumed quantum efficiency for ROÈNO bond disso-
ciation of one, the initial radical concentration can be esti-
mated to be 8 ] 1012 cm~3.
Two di†erent dyes were used to cover the scanning range.
The spectra obtained with each dye showed the same struc-
ture in the overlapping region indicating that the peaks rep-
resent vibronic structure rather than noise. Therefore the
spectra obtained with di†erent dyes could be normalized to
each other.
2-butoxyl with NO .
2
2
Experimental
2-Butoxyl radicals were generated by the photolysis of 2-
butylnitrite at 351 nm using the radiation from an excimer
laser (Lambda Physik, LPX 105E). With a variable delay rela-
tive to the initial photolysis laser pulse the 2-butoxyl radicals
were excited by an excimer laser pumped dye laser (Lambda
Physik EMG 102 MSC, 308 nm (XeCl) and Lambda Physik
FL2002). The relative concentration of these radicals was
probed by monitoring the laser-induced Ñuorescence. Details
of the experimental set-up have been described elsewhere7,26
and only major aspects will be repeated here.
The experiments were performed under slow Ñow condi-
tions in a stainless steel cell. The photolysis excimer laser
beam and the probe excimer pumped dye laser beam pass
coaxially but counterpropagating through this cell. In order to
measure the relative pulse laser energies, reÑections of the
laser beams from the entrance windows of the cell were mea-
sured using pyroelectric energy meters (PEM 8, 245 and 310
V/J, Radiant Dyes). The Ñuorescence light is collected perpen-
dicular to the main axis in the center of the cell by a system of
lenses, passed through a cut-o† Ðlter (Schott GG 395/400 nm
3 mm) and is focussed onto the photocathode of the photo-
multiplier (Thorn EMI 9789 QB). The time dependent photo-
multiplier signal is integrated by a gated integrator/boxcar
averager and fed into a personal computer. The delay time
between the photolysis laser and the dye laser was variable
within the range 5È8000 ls in steps of 5 ls.
The Ñuorescence excitation spectrum of 2-butoxyl shows
5 intense vibronic bands. In analogy to the spectra of
smaller alkoxyl radicals
(CH O,32h39 C H O,40h43n-/iso-
3
2 5
C H O6,7,43) the dominant progression consisting of 3 bands
3
7
can be assigned to the CO-stretching vibration in the elec-
tronically excited state with spacings of (560 ^ 10) cm~1 and
with the (0,0) band origin at 26 768 cm~1. Two other intense
vibrational bands, identiÐed as peak a and a with a spacing
1
2
of (614 ^ 10) cm~1 and with a located B311 cm~1 above the
(0,0) band origin, are visible in Fig. 2. Since the spacing
1
between these two peaks is clearly larger than that of the CO
stretch progression, these peaks could either reÑect an
The Ñuorescence excitation spectrum of 2-butoxyl was
obtained by varying the wavelength of the dye laser in the
range j \ 345È390 nm and detecting the integrated Ñuores-
cence intensity. In order to cover this wavelength range,
DMQ and BiBuQ were used as laser dyes. The delay between
the photolysis and dye laser shot was set to 10 ls. The dye
laser step width and the scan frequency were 2.88 ] 10~2 nm
and 0.1 Hz, respectively, and lasers were operated with a repe-
tition rate of 5 Hz. The signals were averaged over 25 pulses.
Kinetic investigations were performed with a repetition rate
of 5 Hz. A typical 2-butoxyl decay curve consisted of 15È20
points corresponding to di†erent delay times. The kinetic
experiments were performed under pseudo-Ðrst order condi-
tions with NO and NO in at least 100 fold excess. For
2
acquiring a Ñuorescence time proÐle the photomultiplier
signal was averaged over 400 measurements.
The Ñow rates of the reactants were controlled using cali-
brated mass Ñow meters. The concentration of all gases in the
cell were calculated from their mole fractions and the absolute
pressure as determined by a capacitance manometer. 2-
Butylnitrite was diluted in nitrogen (x \ 0.05È0.08) and stored
Fig. 2 (A ^ X) Ñuorescence excitation spectrum of 2-butoxyl follow-
ing the photolysis of 2-butylnitrite at j \ 351 nm. Pump to probe
delay
time:
10
ls,
[2-C H ONO] \ 1 ] 1015
cm~3,
4
9
[2-C H O(X)] B 8 ] 1012 cm~3, p \ 13 mbar (nitrogen), T \ 295 K.
4
9
i
2608
Phys. Chem. Chem. Phys., 2001, 3, 2607È2613

View Article Online
Table 2 Summary of spectroscopic properties of alkoxyl radicals
unknown vibration or correspond to vibrational hot bands.
All observed bands together with their vibrational band inter-
vals are listed in Table 1.
l
/cm~1
COvstretch
The 2-butoxyl (A ^ X) spectrum has recently been observed
by Wang et al.4 and Carter et al.,27 who both generated the
2-butoxyl radicals in the 351 and 355 nm photolysis of 2-
butylnitrite. Carter et al.27 obtained the 2-butoxyl spectrum
under jet-cooled conditions. The spectrum obtained in the
present work is in good agreement with their studies as far as
the positions of the observed bands is concerned. The values
of 26 768 cm~1 and 560 cm~1 determined for the (0,0) tran-
sition and the CO-stretching frequency in the electronically
excited state, respectively, agree excellently with those
obtained by Carter et al.27 (26 757 cm~1 and 559 cm~1).
Besides the CO fundamental stretch these authors observed
three other cold bands and several vibrational hot bands.
Peaks a and a in our spectrum may be identiÐed as the
Radical
T
(AÈX)/cm~1
X
A
Ref.
00
CH O
31 540
31 540
31 530
31 614.5
29 181
29 200
29 204
28 637
29 000
28 634
27 167
27 123
27 140
1015
1022
1020
1047
1060
1080
1067
1000
1065
678
683
670
660
603
600
596
580
450
582
560
560
35
43
44
33, 34
41
44
40
7
45
27
7
6
45
43
27
27
3
C H O
2
5
n-C H O
3
7
i-C H O
900
960
3
7
D27 167
27 171
28 649
26 185
26 768
26 757
25 836
25 866
25 861
560
574,4
1
2
““hotÏÏ vibrational band at B27 070 cm~1 and band b in Fig. 4
in ref. 27, respectively. In contrast Wang et al.4 tentatively
assigned the (0,0) transition at 26 185 cm~1. This result,
however, cannot be conÐrmed by our measurements. In fact,
the authors have recently suggested reassigning the band
origin of the 2-butoxyl spectrum to 26 774 cm~1 based on the
trend for the origin frequencies of the di†erent alkoxyl rad-
icals.28 However, their reported value for the CO-stretching
frequency (567 cm~1) agrees well with the present work. Addi-
tionally, analogues to peaks a and a in our spectrum can
1-C H O
4
9
2-C H O
567
560
559
515
521
546
500
4
4
9
This work
27
26
4
27
3
tert-C H O
4
9
1
2
also be found in the 2-butoxyl spectrum obtained by Wang et
al.4
Table 2 lists spectroscopic constants for alkoxyl radicals
found in this work together with values from the literature. As
can be seen the transition origin is shifted to lower wavenum-
bers with increasing size of the alkoxyl radical. The same
observation is made when turning from primary to secondary
as well as to tertiary radicals. The 1-butoxyl radical, the spec-
trum of which has been recently obtained by Carter et al.,27
seems to be the only exception from this trend. Furthermore,
the CO stretching frequency of the electronic ground state is
always larger than that of the upper state, reÑecting the
increase of the CÈO bond length during the electronic tran-
sition. The decrease of the CO stretching frequency from
CH O to tert-C H O may reÑect the increase of the reduced
3
4 9
mass e†ect of the R CÈO (R \ H, CH , . . .) moiety.
3
3
Fig. 3 Typical logarithmic decay curve of 2-C H O(A) Ñuorescence
4
9
3.2 Lifetime and quenching behaviour of electronically excited
2-butoxyl
emission following the 351 nm photolysis of 2 ] 1015 cm~3 2-
butylnitrite in the presence of 6.5 mbar nitrogen and excitation at
365.938 nm at T \ 295 K. Pump to probe delay time: 10 ls.
Studies of the quenching behaviour and the radiative lifetime
of 2-butoxyl radicals were performed by exciting these radicals
in the (0,1) band in the CO stretching progression at
j \ 365.938 nm. N and 2-butylnitrite were used as quenching
2
partners, where the latter was applied as a mixture diluted in
nitrogen.
A typical logarithmic Ñuorescence decay curve of elec-
tronically excited 2-butoxyl radicals as shown in Fig. 3 reÑects
single exponential behaviour. SternÈVolmer plots obtained for
the decay constants of the Ñuorescence intensities in the pres-
ence of each collision partner are shown in Fig. 4. As
Table 1 Band assignments and positions in the Ñuorescence excita-
tion spectrum of 2-butoxyl. Transitions in the dominant progression
are assigned to CO-stretching vibrations (vA \ 0)
CO
v@
CO
\
0
1
2
l/cm~1
26 768
27 327
27 887
Fig. 4 SternÈVolmer plots for Ñuorescence quenching of 2-
C H O(A) radicals with di†erent collision partners at T \ 295 K; 351
*l/cm~1
559
614
560
4
9
nm photolysis of 2-butylnitrite followed by excitation at 365.938 nm;
Peak:
a
a
1
2
pump to probe delay time: 10 ls; …: [2-C H ONO] \ 4 ] 1015
l/cm~1
27 079
27 693
4
9
cm~3, variable total pressure; L: [2-C H ONO] \ variable,
*l/cm~1
4
9
p(N ) \ 6.5 mbar.
2
Phys. Chem. Chem. Phys., 2001, 3, 2607È2613
2609

View Article Online
expected, the collisional quenching of 2-C H O(A) is particu-
4
9
larly fast for 2-butylnitrite (k \ (4.7 ^ 0.3) ] 10~10 cm3 s~1).
However, efficient quenching is also observed for 2-C H O(A)
4
9
by nitrogen (k \ (5.0 ^ 0.4) ] 10~12 cm3 s~1). This contrasts
with the quenching of electronically excited tert-butoxyl rad-
icals by nitrogen, which has been found to be at least Ðve
hundred times slower (k O 1 ] 10~14 cm3 s~1)26 and conse-
quently negligible. This result, however, is not inconsistent
with quenching efficiencies of nitrogen with other alkoxyl rad-
icals.7
From the extrapolations of the individual SternÈVolmer
plots to zero pressure a collision free lifetime of q B (440 ^ 80)
ns is derived. This value is close to those obtained for CH O
3
(q \ 1.5 ls46), for C H O (q \ 1.0 ls40), for n- and iso-
2
5
C H O (q \ 0.9 ls7 and 0.77 ls5) and for tert-C H O
3
7
4 9
(q \ 1.5 ls,26 1.0 ls25 and 150 ns4). In contrast, Wang et al.4
reported state speciÐc lifetimes of electronically excited 2-
butoxyl in the range 70È99 ns after excitation in three di†erent
Fig. 6 Pseudo-Ðrst-order rate constant for the reaction of 2-butoxyl
radicals with NO as a function of NO concentration at di†erent tem-
peratures. Bu†er gas: N , p \ 26 mbar, [2- C H ONO] \ 1.4 ] 1014
2
4 9
vibrational bands ((0,1), (0,2) and a ). The authors excluded
the onset of predissociation as a reason for the slight mode-
cm~3, [2-C H O(X)] B 4.5 ] 1011 cm~3.
1
4
9
i
dependence of the Ñuorescence lifetime. Since in the present
work 2-butoxyl radicals were only excited in the strongest (0,
1) band at j \ 365.938 nm, we are not able to extract a mode-
dependence of the Ñuorescence lifetimes. However, similar dis-
crepancies arose between Wang et al.Ïs4 and our studies
concerning the Ñuorescence lifetime of tert-butoxyl. In this
case the larger value of the lifetime as obtained in our study26
has recently been conÐrmed by Fittschen et al.25
vation volume, unimolecular decomposition and the reaction
with 2-butylnitrite as well as impurities. As can be seen from
Fig. 6 the contribution of these processes to the intercept
increases with temperature. Whereas di†usion is the main
channel of loss at low temperatures in the absence of NO,
unimolecular decomposition of 2-butoxyl becomes dominant
at temperatures above T \ 253 K (k1st [ 800 s~1). The reac-
0
tion of 2-butoxyl with 2-butylnitrite and its impurities (mainly
3.3 The reaction of 2-butoxyl radicals with NO
NO ) could be determined to be k B (1È2) ] 10~12 cm3 s~1 at
x
Pseudo-Ðrst order rate constants k1st for the reaction of 2-
butoxyl radicals with NO were derived from the slope of the
individual semi-logarithmic plots of the Ñuorescence inten-
sities vs. the delay time between the photolysis and excitation
laser. 2-Butoxyl radicals were excited at j \ 365.938 nm. In all
cases the decays were found to be exponential and could be
monitored over at least 3 lifetimes. Typical logarithmic decay
curves at T \ 263 K are shown in Fig. 5. In this case the NO
concentration was varied in the range (0È2.2) ] 1014 cm~3
with nitrogen as bu†er gas and the initial 2-butoxyl radical
T \ 295 K. As a result the contribution of this process to the
intercepts can be estimated to be small, especially at higher
temperatures where the concentration of 2-butylnitrite
(1.3 ] 1014 cm~3) in the present experiments is low. The rate
coefficient for the unimolecular decomposition of 2-butoxyl
will be presented in a forthcoming publication.47
Rate constants for the reaction of 2-butoxyl with NO were
measured in the temperature range 223È305 K at p \ 26 mbar
and at 295 K in the range p \ 6.5È104 mbar. The individual
rate coefficients are listed in Table 3. The results of the pres-
sure dependent measurements are summarized in Fig. 7. As
can be seen, the data do not exhibit a signiÐcant pressure
concentration was estimated to be B4.5 ] 1011 cm~3. k
NO
can thus be obtained from the slope of a plot of k1st vs. the
reactant concentration. Typical examples for the reaction of
2-butoxyl with NO at di†erent temperatures are shown in Fig.
6.
e†ect. As a result a rate coefficient at 295 K of k
^ 0.3) ] 10~11 cm3 s~1 is obtained. Fig. 8 shows the corre-
sponding Arrhenius plot, from which the expression:
\ (3.9
NO, 295 K
The intercepts reÑect the sum of all removal processes other
than the reaction with NO, namely di†usion out of the obser-
k
\ (9.1 ^ 2.7)
NO
] 10~12 exp((3.4 ^ 0.6) kJ mol~1/RT ) cm3 s~1
was derived.
Table 3 Summary of rate coefficients for the reactions (1) 2-C H O
4
9
] NO ] products, (2) 2-C H O ] NO ] products for di†erent tem-
4
9
2
peratures, bath gas: N , p \ 26 mbara
2
T /K
k
/10~11 cm3 s~1
k
NO2/10~11 cm3 s~1
NO
223
233
243
253
263
273
283
295
305
5.48 ^ 0.13
5.20 ^ 0.13
4.96 ^ 0.10
4.60 ^ 0.29
4.27 ^ 0.16
3.92 ^ 0.20
3.76 ^ 0.20
3.86 ^ 0.33
3.10 ^ 0.40
4.95 ^ 0.22
5.07 ^ 0.29
4.68 ^ 0.22
4.39 ^ 0.25
3.99 ^ 0.20
3.74 ^ 0.25
3.34 ^ 0.24
3.47 ^ 0.23
3.17 ^ 0.47
Fig. 5 Typical logarithmic 2-butoxyl decay traces following the 351
nm photolysis of 1.4 ] 1014 cm~3 2-butylnitrite in the presence of 26
mbar nitrogen and di†erent concentrations of NO at 263 K, [2-
C H O(X)] B 4.5 ] 1011 cm~3.
a Limits represent statistical 2p-errors. The systematic error can be
estimated to be less than 30%.
4
9
i
2610
Phys. Chem. Chem. Phys., 2001, 3, 2607È2613

View Article Online
(a) addition of NO and (b) abstraction of an a-hydrogen atom,
viz.
CH CH(O~)CH CH ] NO
3
2
3
`M
ÈÈÈ CH CH(ONO)CH CH
(1a)
3
2
3
ÈÈÈ CH C(O)CH CH ] HNO (1b)
3
2
3
In the present work no distinction between the two reaction
channels can be made, since we only monitored 2-butoxyl rad-
icals and hence the sum of the rate constants k \ k ] k
1
1a
1b
was measured. Moreover, no attempts have been made to
detect HNO as a reaction product in our experiments. The
absence of any pressure e†ect also does not provide a hint as
to which reaction channel is dominant. However, in analogy
to the smaller alkoxyl radicals it can be supposed that the
abstraction to addition ratio is very small and that the addi-
tion channel has reached its high pressure limit at pressures
O6.5 mbar.
The lack of a pressure dependence of the rate constant at
room temperature is consistent with the results of former
studies on the smaller alkoxyl radicals. As summarized in
Table 4 the results of these studies on the reaction of alkoxyl
radicals with NO show no pressure dependence for the rate
constants above 15 mbar for all smaller alkoxyl radicals
except methoxyl.
Fig. 7 Dependence of the second order rate constant for the reac-
tions 2-C H O ] NO ] products and 2-C H O ] NO ] products
4
9
4
9
2
on total pressure. T \ 295 K, bu†er gas: N , [2-C H ONO] \ 1.4
2
4 9
] 1014 cm~3, [2-C H O(X)] B 4.5 ] 1011 cm~3.
4
9
i
The rate coefficient for the reaction of 2-butoxyl radical
with NO has previously been determined by Deng et al.29 in
the range T \ 226È311 K using the laser photolysis/LIF tech-
nique and 2-butylnitrite as photochemical precursor. They
reported pressure independent kinetics and an Arrhenius
The slightly negative temperature dependence of the reac-
tion of 2-butoxyl radicals with NO is indicative of a recombi-
nation reaction since collision complexes formed with higher
internal energy are more likely to redissociate before stabiliza-
tion.49 Alternatively, the negative temperature dependence
may be determined by the (temperature-dependent) capture
rate constant as is common in barrierless recombination reac-
tions.50 The observed negative temperature dependencies
seem to increase with increasing size of the alkoxyl radical.
AtkinsonÏs2 recommendations for the reaction of all alkoxyl
radicals with NO, as based on relative rate data (for methoxyl,
ethoxyl, propoxyl, iso- and tert-butoxyl and tert-amyloxyl)
cited by Batt51 as well as direct kinetic studies on alkoxyl
expression of k \ (7.50 ^ 1.69) ] 10~12 exp((2.98 ^ 0.47) kJ
NO
mol~1/RT ) cm3 s~1, in good agreement with the result
obtained in the present work.
The reaction of primary and secondary alkoxyl radicals
with NO may generally proceed via two channels,31 namely
radicals O C , predict a slightly smaller temperature depen-
3
dence but a higher A-factor for the reaction of 2-butoxyl with
NO. Nevertheless our room temperature value for this reac-
tion is very close to the predicted one. Hence AtkinsonÏs2 rec-
ommendation does not overestimate the rate coefficent for the
reaction of NO with the 2-butoxyl radical at ambient tem-
peratures, as was suggested for the corresponding reactions of
the tert-butoxyl as well as higher alkoxyl radicals.26
3.4 The reaction of 2-butoxyl radicals with NO
2
Second order rate constants for the reaction of 2-butoxyl rad-
icals with NO were obtained as described above for the reac-
tion with NO. Again, in all cases the 2-butoxyl decays were
Fig. 8 Arrhenius representation of the rate coefficient of the reaction
2-C H O ] NO ] products, p \ 26 mbar (N ); …: this work, L:
2
4
9
tot
2
Deng et al.29
found to be exponential. The NO concentration was varied
2
Table 4 Arrhenius parameters and k
values of the reactions of alkoxyl radicals with NO from direct kinetic studies
295
Radical
A/10~11 cm3 s~1
E /kJ mol~1
k
/10~11 cm3 s~1
295
Ref.
a
CH O
3
È
È
3.6 ^ 1
19
5
5
5
6
26
3
29
C H O
2.0 ^ 0.7
1.2 ^ 0.2
0.89 ^ 0.02
1.22 ^ 0.28
0.76 ^ 0.12
0.78 ^ 0.18
0.750 ^ 0.169
0.91 ^ 0.27
2.3
[0.6 ^ 0.4
[2.9 ^ 0.4
[3.3 ^ 0.5
[2.59 ^ 0.59
[3.2 ^ 0.8
[2.85 ^ 0.29
[2.98 ^ 0.47
[3.4 ^ 0.6
[1.25
3.88
2
5
n-C H O
3.98
3.3
3
7
i-C H O
3
7
3.28 ^ 0.07
2.94 ^ 0.11
2.5
2.5 ^ 0.9
3.86 ^ 0.33
3.8
tert-C H O
4
9
2-C H O
4
9
This work
2, 48
All
Phys. Chem. Chem. Phys., 2001, 3, 2607È2613
2611

View Article Online
in the range (0È2.2) ] 1014 cm~3 with nitrogen as a bu†er gas.
The initial 2-butoxyl radical concentration was estimated to
be 4 ] 1011 cm~3. The total pressure was varied in the range
p \ 6.5È104 mbar at T \ 295 K. Fig. 7 summarizes the
resulting rate coefficients for which no signiÐcant pressure
e†ect could be observed. Consequently, a rate coefficient for
laser Ñuence was varied by at least a factor of two, but no
inÑuence on the measured Ðrst order rate constant at even the
highest NO concentrations was observed.
2
Fig. 10 shows the corresponding Arrhenius plot, from which
the expression:
k
NO2 \ (8.6 ^ 3.3)
the reaction of 2-butoxyl radicals with NO at 295 K of
2
k
\ (3.6 ^ 0.3) ] 10~11 cm3 s~1 is obtained.
] 10~12 exp((3.3 ^ 0.8) kJ mol~1/RT ) cm3 s~1
NO2, 295 K
Additionally, rate constants for the reaction of 2-butoxyl
was obtained. The individual rate coefficients are listed in
Table 3. Again, we observe a slightly negative temperature
dependence, which is typical of a recombination reaction. To
our knowledge this is the Ðrst direct study, in which the rate
radicals with NO were measured in the temperature range
2
223È305 K at p \ 26 mbar. At low temperatures the dimer-
ization of NO to N O had to be taken into account and the
2
2 4
calculated NO concentrations from the Ñow measurements
2
coefficient for the reaction of 2-butoxyl with NO has been
had to be corrected for the NO /N O equilibrium. The
actual NO concentration in the reaction cell corrected for
equilibrium of dimerization was calculated from the expres-
2
2
2 4
determined. Literature data are not available.
2
Similar to the reaction of 2-butoxyl with NO the reaction
with NO may proceed via two di†erent reaction pathways,
sion:
2
namely (a) addition of NO and (b) abstraction of a b-
2
(J8[NO ] K ] 1) [ 1
hydrogen atom, viz.
2 0 eq
[NO ] \
,
(2)
2
4K
CH CH(O~)CH CH ] NO
eq
3
2
3
2
where [NO ] is the NO concentration predicted from the
`M
2 0
2
Ñow measurement alone and K 52 is the equilibrium con-
ÈÈÈ CH CH(ONO )CH CH
(3a)
eq
3
2
2
3
stant. In order to verify if the Ñowing gas mixtures were given
enough time to equilibrate before entering the observation
volume in the cell, the linear gas velocity has been varied in
the range 10È20 cm s~1. No dependence of the measured Ðrst-
ÈÈÈ CH C(O)CH CH ] HNO (3b)
3
2
3
2
For the same arguments as for the reaction of 2-butoxyl
radicals with NO the addition channel (a) is considered to be
the main reaction channel. Attempts to detect HONO have
not been made.
order rate constant on this velocity, even at high NO concen-
2
trations and low temperatures, could be observed.
Fig. 9 illustrates the e†ect of the dimerization equilibrium
on the measured Ðrst-order rate constants. The solid points in
this Ðgure represent the Ðrst-order loss constant of 2-butoxyl
Table 5 summarizes the results of earlier direct measure-
ments on the reaction of smaller alkoxyl radicals with NO .
2
Apart from methoxyl the high pressure limit for the reaction
radicals vs. [NO ] . The open cycles are the Ðrst-order loss
2 0
rate constant is reached far below 15 mbar for all other
alkoxyl radicals. Hence, the lack of any pressure dependence
constant vs. the corrected NO concentration given by eqn.
2
(2). As can be seen there is a linear dependence between the
of the rate of 2-butoxyl with NO at room temperature as
Ðrst-order loss constant and the corrected NO concentration.
2
2
found in the present work is consistent with former studies on
Hence, it can be concluded, that (i) N O does not react as
2
4
the smaller alkoxyl radicals. Unfortunately, the information
on the temperature dependence of the reaction of alkoxyl rad-
icals with NO obtained in direct measurements and available
in the literature is very limited. Only Balla et al.6 reported an
Arrhenius expression for the reaction of iso-propoxy with
NO . These authors observed a dependence of the rate con-
stant on photolysis energy, which they explained with the
reaction of iso-propoxy, generated in the 351 nm photolysis of
iso-propylnitrite, with O(3P). Therefore this Arrhenius expres-
sion was obtained from extrapolations of the observed rate
rapidly as NO and (ii) the correction for N O can be carried
2
2 4
out quite accurately, since otherwise a non-linear dependence
should be expected. It should be mentioned that this correc-
tion only had to be carried out for the temperatures 223 and
233 K.
2
An additional source of error includes the reaction of 2-
2
butoxyl radicals with the photolysis products of NO , espe-
2
cially oxygen atoms. In order to exclude any secondary
chemistry arising from NO photolysis products, the excimer
2
constants to zero NO photolysis at three di†erent tem-
2
peratures in the range 295È384 K. However, compared to iso-
propoxyl the negative temperature dependence seems to
increase for the reactions of tert- and 2-butoxyl with NO .
2
The recommendations of Atkinson48 for the reactions of all
Fig. 9 E†ect of the NO /N O equilibrium at T \ 223 K on the
2
2 4
Ðrst-order 2-butoxyl loss constant. Solid points represent k1st vs.
[NO ] calculated from the gas Ñow outside the cell (i.e. uncorrected
2
for the dimerization equilibrium). Open circles represent k1st vs.
[NO ] in the reactor after accounting for NO /N O equilibrium.
2
2
2 4
The solid line is a non-linear least square Ðt to the equation shown in
the Ðgure.
Fig. 10 Arrhenius representation of the rate coefficient of the reac-
tion 2-C H O ] NO ] products, p \ 26 mbar (N ).
4
9
2
tot
2
2612
Phys. Chem. Chem. Phys., 2001, 3, 2607È2613

View Article Online
Table 5 Arrhenius parameters and k
values of the reactions of alkoxyl radicals with NO from direct kinetic studies
295
2
Radical
A/10~11 cm3 s~1
E /kJ mol~1
k
/10~11 cm3 s~1
295
Ref.
a
CH O
3
È
È
È
È
È
È
È
È
2.0 ^ 0.4
2.8 ^ 0.3
3.6 ^ 0.4
3.3 ^ 0.3
3.68 ^ 0.16
2.42 ^ 0.11
3.47 ^ 0.23
3.8
20
20
7
7
6
26
C H O
2
5
n-C H O
3
7
i-C H O
3
7
1.48 ^ 0.76
0.35 ^ 0.12
0.86 ^ 0.33
2.3
[2.18 ^ 1.34
[4.6 ^ 0.7
[3.3 ^ 0.8
[1.25
tert-C H O
4
9
2-C H O
All
This work
2, 48
4
9
19 M. J. Frost and I. W. M. Smith, J. Chem. Soc., Faraday T rans.,
1990, 86, 1757.
alkoxyl radicals with NO , which are also listed in Table 5
and based on relative rate data (for methoxyl, ethoxyl and
iso-propoxyl) cited by Batt,51 again predict a smaller tem-
perature dependence but a higher A-factor for the reaction of
2
20 M. J. Frost and I. W. M. Smith, J. Chem. Soc., Faraday T rans.,
1990, 86, 1751.
21 D. Gutman, N. Sanders and J. E. Butler, J. Phys. Chem., 1982, 86,
66.
2-butoxyl with NO . In contrast to tert-butoxyl our room
2
temperature value for the reaction of 2-butoxyl with NO is
22 D. Hartmann, J. Karthauser, J. P. Sawerysyn and R. Zellner, Ber.
2
Bunsen-Ges. Phys. Chem., 1990, 94, 639.
very close to the one predicted by Atkinson.
23 F. Caralp, P. Devolder, Ch. Fittschen, N. Gomez, H. Hippler, R.
Mereau, M. T. Rayez, F. Striebel and B. Viskolcz, Phys. Chem.
Chem. Phys., 1999, 1, 2935.
4
Summary and conclusions
24 P. Devolder, Ch. Fittschen, A. Frenzel, H. Hippler, G. Poskreby-
shev, F. Striebel and B. Viskolcz, Phys. Chem. Chem. Phys., 1999,
1, 675.
25 Ch. Fittschen, H. Hippler and B. Viskolcz, Phys. Chem. Chem.
Phys., 2000, 2, 1677.
26 Ch. Lotz and R. Zellner, Phys. Chem. Chem. Phys., 2000, 2, 2353.
27 C. C. Carter, J. R. Atwell, S. Gopalakrishnan and T. A. Miller, J.
Phys. Chem. A, 2000, 104, 9165.
28 Ch. Wang, W. Deng, G. G. Shemesh, M. D. Lilien, D. R. Katz
and T. S. Dibble, J. Phys. Chem. A, 2000, 104, 10368.
29 W. Deng, Ch. Wang, D. R. Katz, G. R. Gawinski, A. J. Davis and
T. S. Dibble, Chem. Phys. L ett., 2000, 330, 541.
30 A. H. Blatt, Organic Syntheses, Collective Volume 2, Wiley, New
York, 1943.
31 J. Heicklen, Adv. Photochem., 1988, 14, 177.
32 H. R. Wendt and H. E. Hunziker, J. Chem. Phys., 1979, 71, 5202.
33 St. C. Foster, P. Misra, T. D. Lin, C. P. Damo, Ch. C. Carter and
T. A. Miller, J. Phys. Chem., 1988, 92, 5914.
34 X. Liu, C. P. Damo, T. D. Lin, St. C. Foster, P. Misra, L. Yu and
T. A. Miller, J. Phys. Chem., 1989, 93, 2266.
35 G. Inoue, H. Akimoto and M. Okuda, J. Chem. Phys., 1980, 72,
1769.
LIF studies of the 2-butoxyl radical were performed and a
Ñuorescence excitation spectrum was obtained following the
351 nm photolysis of 2-butylnitrite. Reaction rate constants of
the 2-butoxyl radical with NO and NO were measured as a
2
function of temperature and pressure. Despite slightly negative
dependencies on temperature in both cases no pressure depen-
dence could be observed in the range 6.5È104 mbar at
T \ 295 K. Under most ambient tropospheric conditions the
reactions of 2-butoxyl with NO and NO are of minor impor-
2
tance compared to unimolecular decomposition and reaction
with oxygen. The reactions with NO and NO become only
2
signiÐcant in highly polluted areas as well as in the interpreta-
tion of smog chamber experiments.
References
1
2
3
W. P. L. Carter and R. Atkinson, J. Atmos. Chem., 1985, 3, 377.
R. Atkinson, Int. J. Chem. Kinet., 1997, 29, 99.
M. Blitz, M. J. Pilling, S. H. Pilling and P. W. Seakins, Phys.
Chem. Chem. Phys., 1999, 1, 73.
36 P. Misra, X. Zhu, Ch.-Y. Hsueh and J. B. Halpern, Chem. Phys.,
4
5
6
7
8
9
Ch. Wang, L. G. Shemesh, W. Deng, M. D. Lilien and T. S.
Dibble, J. Phys. Chem. A, 1999, 103, 8207.
1993, 178, 377.
37 P. Misra, X. Zhu and A. H. Nur, Spectrosc. L ett., 1992, 25, 639.
38 K. Fuke, K. Ozawa and K. Kaya, Chem. Phys. L ett., 1986, 126,
119.
C. Fittschen, A. Frenzel, K. Imrik and P. Devolder, Int. J. Chem.
Kinet., 1999, 31, 1999.
R. J. Balla, H. H. Nelson and J. R. McDonald, Chem. Phys., 1985,
99, 323.
39 D. E. Powers, M. B. Pushkarsky and T. A. Miller, J. Chem. Phys.,
1997, 106, 6863.
Ch. Mund, Ch. Fockenberg and R. Zellner, Ber. Bunsen-Ges.
Phys. Chem., 1998, 102, 709.
40 G. Inoue, M. Okuda and H. Akimoto, J. Chem. Phys., 1981, 75,
2060.
H. Hein, A. Ho†mann and R. Zellner, Ber. Bunsen-Ges. Phys.
Chem., 1998, 102, 1840.
41 X. Zhu, M. M. Kamal and P. Misra, Pure Appl. Opt., 1996, 5,
1021.
H. Hein, A. Ho†mann and R. Zellner, Phys. Chem. Chem. Phys.,
1999, 1, 3743.
42 X. Q. Tan, J. M. Williamson, St. C. Foster and T. A. Miller, J.
Phys. Chem., 1993, 97, 9311.
10 C. Fittschen, B. Decroix, N. Gomez and P. Devolder, J. Chim.
Phys., 1998, 95, 2129.
43 St. C. Foster, Y.-Ch. Hsu, C.-P. Damo, X. Liu, Ch.-Yi Kung and
T. A. Miller, J. Phys. Chem., 1986, 90, 6766.
11 P. Biggs, C. E. Canosa-Mas, J.-M. Fracheboud, A. Douglas Parr,
D. E. Shallcross, R. P. Wayne and F. Caralp, J. Chem. Soc.,
Faraday T rans., 1993, 89, 4163.
12 N. Sanders, J. E. Butler, L. R. Pasternack and J. R. McDonald,
Chem. Phys., 1980, 48, 203.
13 A. Aranda, V. Daele, G. Le Bras and G. Poulet, Int. J. Chem.
Kinet., 1998, 30, 249.
14 D. Rhasa and R. Zellner, Chem. Phys. L ett., 1986, 132, 474.
15 K. Lorenz, D. Rhasa, R. Zellner and B. Fritz, Ber. Bunsen-Ges.
Phys. Chem., 1985, 89, 341.
16 P. J. Wantuck, R. C. Oldenborg, S. L. Baughcum and K. R.
Winn, J. Phys. Chem., 1987, 91, 4653.
17 P. Biggs, C. E. Canosa-Mas, J.-M. Fracheboud, D. E. Shallcross
and R. P. Wayne, J. Chem. Soc., Faraday T rans., 1997, 93, 2481.
18 F. Caralp, M.-T. Rayez, W. Forst, N. Gomez, B. Delcroix, Ch.
Fittschen and P. Devolder, J. Chem. Soc., Faraday T rans., 1998,
94, 3321.
44 T. Ebata, H. Yanagishita, K. Obi and I. Tanaka, Chem. Phys.,
1982, 69, 27.
45 J. Bai, H. Okabe and M. K. Emadi-Babaki, J. Photochem. Photo-
biol. A, 1989, 50, 163.
46 G. Inoue, H. Akimoto and M. Okuda, Chem. Phys. L ett., 1979,
63, 213.
47 Ch. Lotz, H. Somnitz and R. Zellner, to be published.
48 R. Atkinson, J. Phys. Chem. Ref. Data, 1997, 26, 232.
49 M. J. Pilling and P. W. Seakins, Reaction Kinetics, Oxford Uni-
versity Press, 1995.
50 J. Troe, J. Chem. Soc., Faraday T rans., 1994, 90, 2303.
51 L. Batt, Int. Rev. Phys. Chem., 1987, 6, 53.
52 W. B. Moore, S. P. Sander, D. M. Golden, R. F. Hampson, M. J.
Kurylo, A. R. Ravishankara, C. E. Kolb and M. J. Molina,
Chemical Kinetics and Photochemical Data for Use in Strato-
spheric Modelling, Jet Propulsion Lab., Pasadena, CA, 1997, JPL
Pub. 97-4.
Phys. Chem. Chem. Phys., 2001, 3, 2607È2613
2613