Alkyl nitrate, hydroxyalkyl nitrate, and hydroxycarbonyl formation from the NOx-air photooxidations of C5-C8 n-alkanes

Article abstract of DOI:10.1021/jp003292z

In the troposphere, alkanes present in the gas phase react mainly with the OH radical to form an alkyl radical. Products of the gas-phase reactions of OH radicals with the n-alkanes, n-pentane through n-octane at 298 ± 2 K and 1 atm of air were studied using GC with flame ionization detection, combined GC-MS, and in situ atmospheric pressure ionization tandem MS. The alkyl nitrate, hydroxyalkyl nitrate, and hydroxycarbonyl formation yields combined with the published carbonyl yields arising from alkoxy radical decomposition and reaction with O₂, account for ～ 95, ～ 80, ～ 70, and ～ 55% of n-pentane, n-hexane, n-heptane, and n-octane (with NO), respectively. Hydroxyalkyl nitrate formation accounts for a few percent of the overall reaction products, and was consistently ～ 25% of the corresponding alkyl nitrate yields. Significant fractions of the atmospheric reaction products of ≥ C₅ n-alkanes (and other alkanes whose intermediate alkoxy radicals can isomerize) were composed of hydroxycarbonyls and hydroxyalkyl nitrates, compound classes which are difficult to analyze under atmospheric conditions.

Full text of DOI:10.1021/jp003292z

1020
J. Phys. Chem. A 2001, 105, 1020-1027
Alkyl Nitrate, Hydroxyalkyl Nitrate, and Hydroxycarbonyl Formation from the NOx-Air
Photooxidations of C5-C8 n-Alkanes
Janet Arey,*,† Sara M. Aschmann, Eric S. C. Kwok,‡ and Roger Atkinson*†,§
Air Pollution Research Center, UniVersity of California, RiVerside, California 92521
ReceiVed: September 15, 2000; In Final Form: NoVember 29, 2000
Products of the gas-phase reactions of OH radicals with the n-alkanes n-pentane through n-octane at 298 (
2 K and atmospheric pressure of air have been investigated using gas chromatography with flame ionization
detection (GC-FID), combined gas chromatography-mass spectrometry (GC-MS), and in situ atmospheric
pressure ionization tandem mass spectrometry. The formation yields of alkyl nitrates from n-hexane, n-heptane,
and n-octane were measured by GC-FID, with the sum of the isomeric alkyl nitrates being 0.141 ( 0.020,
0.178 ( 0.024, and 0.226 ( 0.032, respectively. These alkyl nitrate yields are ∼35% lower than previous
data reported from this laboratory in the early 1980s. Using negative ion atmospheric pressure chemical
ionization with the addition of pentafluorobenzyl alcohol to study the n-pentane through n-octane reactions
and those of the fully deuterated n-alkanes, hydroxyalkyl nitrate products were identified from the n-pentane,
n-heptane, and n-octane reactions for the first time and the presence of hydroxycarbonyl products was confirmed.
Adding NO2 to the chamber reaction mixture postreaction to form [NO2‚M]- adducts of the hydroxycarbonyls
and hydroxynitrates, together with the use of 5-hydroxy-2-pentanone and 2-nitrooxy-3-butanol as internal
standards for the hydroxycarbonyls and hydroxynitrates, respectively, enabled the yields of the hydroxycarbonyl
and hydroxynitrate reaction products to be estimated.
Introduction
SCHEME 1
In the troposphere, alkanes present in the gas phase react
mainly with the OH radical1 to form an alkyl radical.
OH + RH f H2O + R˙
(1)
Alkyl radicals react rapidly with O2 in the troposphere to form
alkyl peroxy (RO˙ 2) radicals, which then react with NO, NO2,
HO2 radicals or with other organic peroxy (RÄ O˙ 2) radicals.1
R˙ + O2 f RO˙ 2
RO˙ 2 + NO f RONO2
(2)
(3a)
(3b)
(4)
RO˙ 2 + NO f RO˙ + NO2
RO˙ 2 + NO2 a ROONO2
RO˙ 2 + HO2 f ROOH + O2
(5)
RO˙ 2 + RO˙ 2 f R carbonyl + (1 - R) alcohol +
RÄ O˙ 2 products + O2 (6a)
RO˙ 2 + RÄ O˙ 2 f RO˙ + RÄ O˙ + O2
(6b)
In the presence of NO concentrations g10-30 parts-per-trillion,
peroxy radical reactions with NO [reaction 3] will dominate2
and the products of the alkyl peroxy radical reactions with NO
are the corresponding alkoxy radical plus NO2 or the alkyl
nitrate, RONO2,1 as illustrated in Scheme 1 for the 2-hexyl
radical formed from n-hexane.
The alkoxy radicals formed in reactions 3b and 6b then react
with O2, decompose by C-C bond scission, or isomerize
through a six-membered ring transition state,1 as shown in
Scheme 1. The isomerization reaction becomes important for
gC4 alkanes and isomerization followed by O2 addition to the
1,4-hydroxyalkyl radical results in the formation of a 1,4-
hydroxyalkyl peroxy radical (Scheme 1). 1,4-Hydroxyalkyl
peroxy radicals then react with NO, as shown in reactions 3a
* Authors to whom correspondence should be addressed.
† Also Interdepartmental Graduate Program in Environmental Toxicology
and Department of Environmental Sciences, University of California,
Riverside.
‡ Present address: California Air Resources Board, 2020 L Street, P.O.
Box 2815, Sacramento, CA 95812.
§ Also Department of Chemistry, University of California, Riverside.
10.1021/jp003292z CCC: $20.00 © 2001 American Chemical Society
Published on Web 01/18/2001

NOx-Air Photooxidations of C5-C8 n-Alkanes
J. Phys. Chem. A, Vol. 105, No. 6, 2001 1021
and 3b and Scheme 1, leading to the formation of a 1,4-
hydroxyalkoxy radical plus NO2 or a 1,4-hydroxyalkyl nitrate
(note that the products analyzed in this work are denoted by
boxes in Scheme 1). The 1,4-hydroxyalkoxy radical is expected
to undergo a second (rapid) isomerization, leading to the
formation of a 1,4-hydroxycarbonyl.1 In addition to the forma-
tion of alkyl nitrates from reactions of the peroxy radicals with
NO (as shown in reaction 3a and Scheme 1), reaction of the
alkoxy radicals with NO2 can also lead to alkyl nitrate formation.
100 cm3 volume were collected from the chamber onto Tenax-
TA solid adsorbent, with subsequent thermal desorption at ∼225
°C onto a DB-1701 megabore column in a Hewlett-Packard (HP)
5710 GC, initially held at 0 °C and then temperature pro-
grammed to 200 °C at 8 °C min-1. In addition, gas samples
were collected onto Tenax-TA solid adsorbent for GC-MS
analyses, with thermal desorption onto a 60 m HP-5 fused silica
capillary column in an HP 5890 GC interfaced to a HP 5970
mass selective detector operating in the scanning mode. GC-
FID response factors were determined as described previously.18
The alkyl nitrate standards contained the other secondary alkyl
nitrates as impurities. The GC-FID calibrations, therefore, were
based on summing the areas of the alkyl nitrate isomers and
assuming identical GC-FID response factors for each alkyl
nitrate. The molar GC-FID responses of the hexyl, heptyl, and
octyl nitrates were consistent with their expected Equivalent
Carbon Numbers,19 verifying that wall losses in these calibra-
tions were minor or negligible, in contrast to what is now
suspected for our previous study.5 The unique MS fragmentation
patterns of the alkyl nitrates allowed the various isomers to be
distinguished from one another. NO and initial NO2 concentra-
tions were measured using a Thermo Environmental Instru-
ments, Inc., Model 42 NO-NO2-NOx chemiluminescence
analyzer.
RO˙ + NO2 f RONO2
(7)
The formation of alkyl nitrates from the OH radical-initiated
reactions of C2-C8 n-alkanes and from a number of branched
and cyclic alkanes has been studied previously.3-13 For the
n-alkanes, the alkyl nitrate yields were observed to increase
monotonically with increasing carbon number in the n-alkane,5
with molar yields ranging from e0.014 for ethane to 0.33 for
n-octane at room temperature and atmospheric pressure of air.5
The alkyl nitrate yields also increase with increasing pressure
and with decreasing temperature.6,8,9,14 To date, the formation
of hydroxyalkyl nitrates from the OH radical-initiated reactions
of alkanes, arising after alkoxy radical isomerization (Scheme
1), has been reported only from n-hexane.15
Teflon Chamber with Analysis by API-MS. In the experi-
ments with API-MS analyses, the chamber contents were
sampled through a 25 mm diameter × 75 cm length Pyrex tube
at ∼20 L min-1 directly into the API mass spectrometer source.
The operation of the API-MS in the MS (scanning) and MS/
MS [with collision-activated dissociation (CAD)] modes has
been described elsewhere.11,16 Use of the MS/MS mode with
CAD allows the “product ion” or “precursor ion” spectrum of
a given ion peak observed in the MS scanning mode to be
obtained.11
In this work, we have extended our recent study of the OH
radical-initiated reaction of n-pentane,10 using sampling, analy-
sis, and calibration procedures which are significantly improved
over those employed in our early 1980s studies,5,6 to reinves-
tigate alkyl nitrate formation from the reactions of the OH
radical with n-hexane through n-octane. In addition, we have
used atmospheric pressure chemical ionization tandem mass
spectrometry in the negative ion mode to investigate products
of the n-pentane through n-octane and n-pentane-d12 through
n-octane-d18 reactions, with a goal of observing the formation
of the hydroxyalkyl nitrates predicted to be formed after alkoxy
radical isomerization1,15 and quantifying hydroxyalkyl nitrate
and hydroxycarbonyl15,16 products.
The negative ion mode was used in these API-MS and API-
MS/MS analyses, with negative ions being generated by the
negative corona around the discharge needle. The superoxide
ion (O2-), its hydrates [O2-(H2O)n], and O2 clusters [O2-(O2)n]
are the major reagent negative ions in the chamber diluent air.20
Experimental Methods
-
Other reagent ions, for example, NO2 and NO3-, are then
Experiments were carried out at 298 ( 2 K and 740 Torr
total pressure of air in a 7900 L Teflon chamber with analysis
by gas chromatography with flame ionization detection (GC-
FID) and combined gas chromatography-mass spectrometry
(GC-MS), with irradiation provided by two parallel banks of
blacklamps; and in a 7300 L Teflon chamber interfaced to a
PE SCIEX API III MS/MS direct air sampling, atmospheric
pressure ionization tandem mass spectrometer (API-MS), again
with irradiation provided by two parallel banks of blacklamps.
Both chambers are fitted with Teflon-coated fans to ensure rapid
mixing of reactants during their introduction into the chamber.
Hydroxyl radicals were generated by the photolysis of methyl
nitrite in air at wavelengths >300 nm,17 and NO was added to
the reactant mixtures to suppress the formation of O3 and hence
of NO3 radicals.17
Teflon Chamber with Analysis by GC-FID. For the OH
radical reactions carried out in the 7900 L Teflon chamber (at
∼5% relative humidity), the initial reactant concentrations (in
molecule cm-3 units) were CH3ONO, (2.2-2.5) × 1014; NO,
(2.2-2.3) × 1014; and n-alkane, (2.34-2.51) × 1013. Irradiations
were carried out at 20% of the maximum light intensity for
10-45 min, resulting in up to 37%, 43%, and 49% reaction of
the initially present n-hexane, n-heptane, and n-octane, respec-
tively. The concentrations of the n-alkane and alkyl nitrates were
measured during the experiments by GC-FID. Gas samples of
formed through reactions between the primary reagent ions and
neutral molecules such as NO2, and instrument tuning and
operation were designed to induce cluster formation.11,16 In one
series of experiments, pentafluorobenzyl alcohol (PFBOH) was
added by passing the sampled air stream from the chamber over
a heated vial containing PFBOH (with PFBOH effusing through
a pinhole in the cap of the vial).
Under the experimental conditions used, analytes were
detected as adducts ions formed from reaction of the neutral
analyte (M) and reagent ions such as O2-, NO2-, [PFBOH‚O2]-,
and [PFBOH‚NO2]- generated in the ion source region. For
example,
O2- + M f [O2‚M]-
NO2- + M f [NO2‚M]-
[PFBOH‚O2]- + M f [PFBOH‚O2‚M]-
[PFBOH‚NO2]- + M f [PFBOH‚NO2‚M]-
where M can include PFBOH. Because of the presence of
fluorine in the PFBOH-containing reagent ions, the PFBOH-
containing dimers are detected with high sensitivity under

1022 J. Phys. Chem. A, Vol. 105, No. 6, 2001
Arey et al.
negative ionization conditions and their higher masses allowed
them to be readily distinguished from fragment ions. Ions are
drawn by an electric potential from the ion source through the
sampling orifice into the mass-analyzing first quadrupole or third
quadrupole. Neutral molecules and particles are prevented from
entering the orifice by a flow of high-purity nitrogen (“curtain”
gas). The initial concentrations of CH3ONO, NO, and n-alkane
(or deuterated n-alkane) were ∼(2.4-24) × 1012 molecule cm-3
each, and irradiations were carried out at 100% of the maximum
light intensity for 0.5-7.5 min or at 20% of the maximum light
intensity for 2.5-10 min.
Chemicals. The chemicals used, and their stated purities, were
as follows: n-heptane, Mallinckrodt; n-heptane-d16 (98%),
n-hexane (99+%), 1-hydroxy-2-butanone (95%), 4-hydroxy-4-
methyl-2-pentanone (99%), n-octane (99+%), pentafluorobenzyl
alcohol (98%), and n-pentane (99+%), Aldrich Chemical Co.;
n-hexane-d14 (99%), n-octane-d18 (99%), and n-pentane-d12
(98%), Cambridge Isotope Laboratories; 3-heptyl nitrate, 2-hexyl
nitrate, and 3-octyl nitrate, Fluorochem, Inc.; 5-hydroxy-2-
pentanone (96%), TCI America; and NO (g99.0% and cis-2-
butene (g 95%), Matheson Gas Products. Methyl nitrite was
prepared and stored as described previously.17
Figure 1. Plots of the amounts of 2-, 3-, and 4-heptyl nitrate formed,
corrected for reaction with the OH radical (see text), against the amounts
of n-heptane reacted with the OH radical in the presence of NO.
Results and Discussion
molecule cm-3, and with the expectation that ([NO] + [NO2])
∼ constant throughout the irradiations23 then the maximum NO2
concentrations were <1.3 × 1014 molecule cm-3, corresponding
to a reaction rate of alkoxy radicals due to reaction with NO2
of <5 × 103 s-1. All of the alkoxy radicals formed from
n-hexane, n-heptane, and n-octane can isomerize through a six-
membered transition state,1 with estimated isomerization rate
constants at 298 K of ∼2 × 105 s-1 for the 3-hexoxy radical,
∼4 × 105 s-1 for the 4-heptoxy radical, and g2 × 106 s-1 for
the 2-hexoxy, 2- and 3-heptoxy, and 2-, 3-, and 4-octoxy
radicals.1 Based on these estimated rate constants for the
competing isomerization reactions, formation of alkyl nitrates
from reaction 7 could lead to a maximum formation yield of
3-hexyl nitrate and 4-heptyl nitrate of 0.025 and 0.012 (per
3-hexoxy or 4-heptoxy radical formed), respectively, and of the
other alkyl nitrates measured of e0.0025 (per alkoxy radical
formed). These alkyl nitrate formation yields from reaction 7
are much lower than the measured yields (per alkyl peroxy
radical) of g0.14 (see Table 2) and are within the experimental
uncertainties.
Our present alkyl nitrate yields (together with the pentyl
nitrate yields from n-pentane10) are compared to our previous
early 1980s data5 in Table 1, showing that while our newer
pentyl nitrates yield10 is in agreement with the earlier data of
Atkinson et al.,5 our present alkyl nitrate yields from n-hexane,
n-heptane, and n-octane are significantly lower than our previous
data.5 The previous measurements5 were carried out in an ∼60
L Teflon reaction vessel with GC-FID analyses on a packed
column, with 100 cm3 samples being collected in a gastight
all-glass syringe and concentrated by freezing out at liquid argon
temperature prior to introduction onto the column via a gas
sampling valve. Moreover, calibrations for the alkyl nitrates were
carried out by introducing measured quantities of the liquids
into a 46.5 L volume borosilicate glass vessel with replicate
sampling and analysis. We now believe that this earlier study5
had the potential for losses of the larger (gC6) alkyl nitrates
during the experiments, including (especially) during the alkyl
nitrate calibrations, and that the data obtained5 may have been
subject to significant (and then unrecognized) uncertainties. Our
present total heptyl nitrates yield of 0.178 ( 0.024 at 298 ( 2
K and 740 Torr total pressure of air is in good agreement with
GC-FID and GC-MS Analyses of Alkyl Nitrates. GC-MS
analyses of irradiated CH3ONO/NO/n-alkane/air mixtures showed
the formation of 2- and 3-hexyl nitrate from n-hexane, 2-, 3-,
and 4-heptyl nitrate from n-heptane, and 2-, 3-, and 4-octyl
nitrate from n-octane, and the n-alkane reactants and the alkyl
nitrate products were quantified by GC-FID. Because the alkyl
nitrates also react with the OH radical, these secondary reactions
were taken into account as described previously5 using rate
constants for reactions of the OH radical (in units of 10-12 cm3
molecule-1 s-1) of n-hexane, 5.45; n-heptane, 7.01; n-octane,
8.71; 2-hexyl nitrate, 3.06; 3-hexyl nitrate, 2.60; 2-hepyl nitrate,
4.67; 3-heptyl nitrate, 3.75; 4-heptyl nitrate, 3.56; 2-octyl nitrate,
6.03; 3-octyl nitrate, 5.11; and 4-octyl nitrate, 4.92. The rate
constants for the n-alkanes at 298 K were those recommended
by Atkinson,1 those for 2- and 3-hexyl nitrate were those cited
in Atkinson21 and reevaluated using the recently recommended
rate constant for the reference organic cyclohexane,1 and those
for the heptyl and octyl nitrates were calculated22 using the group
rate constants and group substituent factors given by Atkinson22
and decreased by 3.6% to account for the recent reference
organic recommendations of Atkinson.1 The multiplicative
correction factors F to take into account the secondary reactions
of OH radicals with the alkyl nitrates increase with the rate
constant ratio k(OH + alkyl nitrate)/k(OH + n-alkane) and with
the extent of reaction,5 and were <1.15 for the hexyl nitrates,
<1.22 for the heptyl nitrates, and <1.28 for the octyl nitrates.
Figure 1 shows plots of the amounts of the 2-, 3-, and 4-heptyl
nitrates, corrected for reaction with the OH radical, against the
amounts of n-heptane reacted. Similar linear plots were observed
for the n-hexane and n-octane reactions, and the alkyl nitrate
formation yields (for the individual isomers as well as their sum)
obtained by least-squares analyses of the data are given in Table
1, together with analogous formation yields from our study of
the n-pentane reaction10 using similar analysis procedures.
As noted above, in addition to alkyl nitrate formation from
peroxy radical reactions with NO [reaction 3a], alkyl nitrates
can also be formed from the combination reaction of alkoxy
radicals with NO2 [reaction 7]. The rate constant for reaction 7
is ∼3.8 × 10-11 cm3 molecule-1 s-1 at room temperature.1 The
final NO concentrations in the experiments were >1.0 × 1014

NOx-Air Photooxidations of C5-C8 n-Alkanes
J. Phys. Chem. A, Vol. 105, No. 6, 2001 1023
TABLE 1: Formation Yields of Alkyl Nitrates from the OH Radical-Initiated Reactions of n-Hexane, n-Heptane, and n-Octane
in the Presence of NO, at 298 ( 2 K and Atmospheric Pressure of Aira
total nitratesb
alkane
ethane
propane
n-butane
n-pentane
n-hexane
n-heptane
n-octane
2-nitratea
3-nitratea
4-nitratea
this worka
previous studyc
<0.014
0.029 ( 0.002c
0.073 ( 0.008c
0.060 ( 0.010d
0.059 ( 0.007
0.059 ( 0.006
0.062 ( 0.006
0.036 ( 0.005
0.077 ( 0.009
0.117 ( 0.013
0.208 ( 0.027
0.293 ( 0.042
0.318 ( 0.027
0.044 ( 0.006d
0.082 ( 0.010
0.083 ( 0.009
0.081 ( 0.008
0.105 ( 0.014d
0.141 ( 0.020
0.178 ( 0.024
0.226 ( 0.032
0.036 ( 0.005
0.083 ( 0.012
a Indicated errors are two least-squares standard deviations combined with estimated overall uncertainties in the GC-FID response factors for the
alkanes and alkyl nitrates of (5% each. b From least-squares analyses of the sum of the nitrates. c From Atkinson et al.,5 with the indicated errors
being two least-squares standard deviations. d From Atkinson et al.;10 see text.
TABLE 2: Formation Yields of Secondary Alkyl Nitrates from the Corresponding Secondary Alkyl Peroxy Radical in the OH
Radical-Initiated Reactions of n-Hexane, n-Heptane, and n-Octane in the Presence of NO, at 298 ( 2 K and Atmospheric
Pressure of Air
alkane
2-nitratea
3-nitratea
4-nitratea
sec nitratesc
3-nitrate/2-nitrateb
4-nitrate/2-nitrateb
propane
0.039
0.084
0.106
0.140
0.177
0.224
0.039
0.084
0.115 ( 0.013
0.150 ( 0.019
0.187 ( 0.023
0.235 ( 0.031
n-butane
n-pentaned
n-hexane
n-heptane
n-octane
0.126
0.158
0.202
0.238
1.41 ( 0.08
1.40 ( 0.06
1.44 ( 0.11
1.29 ( 0.08
0.175
0.243
1.26 ( 0.08
1.32 ( 0.14
a Using the -CH3 and -CH2- group rate constants and group substituent factors of Kwok and Atkinson24 to calculate the fractions of the
overall OH radical reaction proceeding by C-H abstraction at the individual carbon atoms. b From least-squares analyses of the 3-alkyl (or 4-alkyl)
nitrate concentrations against those of the 2-alkyl nitrates (corrected for reactions with the OH radical). Indicated errors are two least-squares
standard deviations. c From least-squares analyses of the sum of the nitrates, using the method of Kwok and Atkinson24 to calculate the fraction of
the overall OH radical reaction proceeding by C-H abstraction from secondary -CH2- groups. d From Atkinson et al.10 (see text).
the total yield of heptyl nitrates measured by Harris and Kerr9
at 730 Torr total pressure of air of 0.201 ( 0.041 at 298 K and
0.180 ( 0.015 at 299 K, although the heptyl nitrate isomer
distribution reported by Harris and Kerr9 differed significantly
from our present and previous5,6 heptyl nitrate isomer distribu-
tions. It should be noted that Harris and Kerr9 quantified the
heptyl nitrates by calibrating for 2-heptyl nitrate and assuming
that the GC-electron capture detector responses of the isomeric
heptyl nitrates were identical, while in the present work it is
assumed that the GC-FID responses are identical.
that while the calculated overall rate constants for the reaction
of the OH radical with the C2-C8 n-alkanes at 298 K agree
with the recommended values1 to within 13%, the calculated
partial rate constants and hence fractions of the overall reaction
proceeding at the various -CH3 and -CH2- groups may be
less reliable.10
The secondary alkyl nitrate formation yields from the
corresponding RO˙ 2 radicals given in Table 2 show that the
nitrate yield increases with the carbon number, as observed
previously.5 Table 2 also shows that the nitrate yields from the
various secondary RO˙ 2 radicals, and hence values of k3a/(k3a
+
Table 1 also contains the alkyl nitrate formation yields
reported by Atkinson et al.5 at room temperature and atmo-
spheric pressure of air from the reactions of Cl atoms with
ethane and the OH radical with propane and n-butane (all in
the presence of NO). Combining these earlier data with the
present work, the observed overall alkyl nitrate yields increase
linearly with the carbon number of the n-alkane, with essentially
zero ethyl nitrate formation from ethane and with an increase
in the alkyl nitrate yield of 0.0354 ( 0.0026 per additional
carbon atom in the n-alkane. As evident from the data presented
in Table 1, only alkyl nitrates formed from the reactions of
secondary alkyl peroxy radicals with NO were observed in our
present study of n-hexane through n-octane, consistent with our
previous data.5,6 In the absence of direct experimental data, the
estimation method of Kwok and Atkinson24 for calculation of
OH radical reaction rate constants can be used to calculate the
rate constants for H-atom abstraction by the OH radical from
the various -CH3 and -CH2- groups in the n-alkanes, and
hence the fractions of the overall OH radical reaction proceeding
at the various -CH2- groups. Table 2 gives the formation yields
of the 2-, 3-, and 4-alkyl nitrates from their precursor secondary
RO˙ 2 radicals (i.e., rate constant ratios k3a/(k3a + k3b) derived
from our present and previous5,10 data and the calculated
fractions of the overall OH radical reaction proceeding at the
2-, 3-, and 4-position -CH2- groups.24 It needs to be recognized
k3b), for a given n-alkane are similar, although the 2-alkyl nitrate
yields are consistently lower than those of the 3-alkyl nitrates.
The measured 3-nitrate/2-nitrate and 4-nitrate/2-nitrate formation
yield ratios are all similar, with an average value of 1.35 (
0.15 (two standard deviations), and this ratio can be compared
with that of 1.23 obtained from the estimation method of Kwok
and Atkinson24 assuming that for a given n-alkane the secondary
alkyl nitrate yield per RO˙ 2 precursor radical is identical. Given
the likely uncertainties in the substituent group factors in the
estimation method,24,25 our data are consistent with the second-
ary alkyl nitrate yield per RO˙ 2 precursor radical being identical
for a given n-alkane.
A plot of the formation yields of the sum of the secondary
alkyl nitrates for a given alkane from the corresponding
secondary alkyl peroxy radicals for propane through n-octane
(data from Atkinson et al.5,10 and this work) and cyclohexane
(data from Aschmann et al.11) against carbon number is shown
in Figure 2. The cyclohexyl nitrate yield of Aschmann et al.11
of 0.165 ( 0.021 at 298 ( 2 K and 740 Torr pressure of air is
in good agreement with the yields reported by Platz et al.12 (0.16
( 0.04 at 296 ( 2 K and 700 Torr total pressure) and Orlando
et al.13 (0.15 ( 0.04 at 296 K and 700 Torr total pressure), but
is higher than the value of 0.090 ( 0.044 at atmospheric pressure
and room temperature measured by Takagi et al.4 The nitrate

1024 J. Phys. Chem. A, Vol. 105, No. 6, 2001
Arey et al.
Figure 3. API-MS negative ion spectrum (300-500 u) of an irradiated
CH3ONO/NO/n-hexane/air mixture, with addition of pentafluorobenzyl
alcohol (PFBOH) to the sampled air stream. The ion peaks marked
with (*) are attributed to the following: 428 u, [(PFBOH)2‚O2]-; 442
u, [(PFBOH)2‚NO2]-; 446 u, [(PFBOH)2‚O2‚H2O]-; 458 u, [(PFBOH)2‚
O2‚NO]-; and 474 u, [(PFBOH)2‚O2‚NO2]-. The 307 u ion peak is
always present when the PFBOH is added and may be due to a proton-
transfer reaction.
Figure 2. Plot of the yields of secondary alkyl nitrates from their
precursor secondary alkyl peroxy radicals, k3a/(k3a + k3b), for the C3-
C8 n-alkanes (O) and cyclohexane (0) against carbon number of the
alkane. The data for the propane and n-butane reactions are from
Atkinson et al.;5 for the n-pentane reaction are from Atkinson et al.;10
and for the cyclohexane reaction are from Aschmann et al.11
yield per secondary RO˙ 2 radical from cyclohexane fits very well
with the data from the n-alkanes, and the alkyl nitrate formation
yield from secondary alkyl peroxy radicals is linear with the
carbon number of the alkane. From the data shown in Figure 2
the ratio k3a/(k3a + k3b) for secondary RO˙ 2 radicals at 298 K
and atmospheric pressure of air is
3-pentyl nitrate formation yields from the n-pentane reaction
measured by Atkinson et al.6 is correct (with all of the measured
yields being consistently slightly high due to calibration
problems). The secondary alkyl nitrate yield data can then be
reasonably well fit with the following values of the parameters
in eq I: R ) 2 × 10-22 cm3 molecule-1, â ) 1.0, Y∞
)
298
k3a/(k3a + k3b) ) (0.0381 ( 0.0031)n - (0.073 ( 0.0178)
0.43, F ) 0.41, mo ) 0, and m∞ ) 8.0. The major difference
between these revised parameters and those given in Carter and
Atkinson14 is in the value of Y∞298, with Carter and Atkinson14
deriving Y∞298 ) 0.87 (0.826 at 300 K), and our revised formula
leads to alkyl nitrate yields which are increasingly lower than
the previous14 prediction for n > 5.
where n is the number of carbon atoms in the alkane and the
indicated errors are two least-squares standard deviations. The
rate constant ratio k3a/k3b for primary, secondary, and tertiary
alkyl groups is needed for extrapolation to temperatures and
pressures outside of the range for which experimental data are
available and for >C8 alkanes.14 Our present data can be used
with previous literature data for alkyl nitrate formation from
propane,5 n-butane,5 n-pentane,10 2-methylbutane,7,8 2-methyl-
pentane,7 3-methylpentane,7,8 cyclohexane,11-13 and n-heptane9
to revise the formula proposed by Carter and Atkinson.14
API-MS Analyses. A series of CH3ONO/NO/n-alkane/air
irradiations were carried out with analysis using in situ
atmospheric pressure ionization tandem mass spectrometry (API-
MS). In these analyses, the API-MS was operated in the negative
ion mode, and either PFBOH was added to the gas stream
sampled from the chamber or NO2 was introduced into the
chamber after the reaction was completed.
o
k3a
k3b
Yo298[M](T/298)-m
Yo298[M](T/298)-m
)
Fz
(I)
PFBOH Addition. The formation of dimers in the API source
has proved to be an effective means of product identification
in a number of complex reaction systems.11,26,27 PFBOH has a
strong response in the negative ion mode and readily forms
o
1 +
[
]
Y∞298(T/298)-m
∞
-
dimers as well as adduct ions with O2 and NO2-, and its
where
and
addition to the sample stream was utilized to identify the
hydroxycarbonyl and hydroxynitrate products from the OH
radical-initiated reactions of n-pentane through n-octane and of
n-pentane-d12 through n-octane-d18. All of these reactions gave
analogous results, as illustrated using the n-hexane reaction as
an example. Figure 3 shows the API-MS spectrum, in the adduct
ion region above 300 u, of an irradiated CH3ONO/NO/n-hexane/
air mixture. The majority of the ion peaks (those marked with
(*) are present when PFBOH is added to the pre-reaction
chamber sample, i.e., to the n-alkane/CH3ONO/NO/air mixture
prior to irradiation) can be attributed to dimers of PFBOH
forming adducts with O2, NO2, NO, and H2O. The NO and NO2
adducts are present from the CH3ONO/NO/air mixture used to
generate OH radicals and the NO concentration decreases, and
Yo298[M](T/298)-m
Y∞298(T/298)-m
2
-1
o
z ) 1 + log
∞
(
(
) )
]
[
Yo298 ) Reân
298
in which Yo298[M] and Y∞ are the limiting low-pressure and
high-pressure rate constant ratios k3a/k3b at 298 K, respectively,
[M] is the concentration of diluent air, F, mo, m∞, R, and â are
parameters, and n is the number of carbon atoms in the alkane.
We assume that the temperature dependence of the 2- and

NOx-Air Photooxidations of C5-C8 n-Alkanes
J. Phys. Chem. A, Vol. 105, No. 6, 2001 1025
Figure 4. API-MS/MS CAD “precursor ion” negative ion spectrum
of the 230 u [PFBOH‚O2]- ion for the same experiment as in Figure
3. The ion peaks at 428, 446, and 458 u are attributed as noted in
Figure 3. The 346 and 393 u ion peaks indicate two products of the
n-hexane reaction.
the NO2 concentration increases, as the reaction proceeds. As
shown in Figure 4, an API-MS/MS “precursor ion” scan of 230
u (which corresponds to [PFBOH‚O2]-) resulted in two ion
peaks at 346 and 393 u, in addition to those identified in Figure
3 as being from the PFBOH. API-MS/MS “product ion” spectra
of these 346 and 393 u ion peaks are shown in Figure 5 parts
A and B, respectively. In general, API-MS/MS “product ion”
spectra of the adduct ions [PFBOH‚O2‚M]- show a prominent
fragment ion at 230 u [PFBOH‚O2]-, with the difference in mass
of the adduct ion and the 230 u fragment ion being the molecular
weight of the product M. Thus products of molecular weight
116 (an hydroxycarbonyl) and 163 (an hydroxynitrate) are
indicated from the API-MS/MS spectra in Figure 5, partsA and
B, respectively. These products are also evident from the less
abundant fragment ions at 148 u and 195 u, corresponding to
[M‚O2]- formed from loss of PFBOH from the parent ion, that
are also seen in Figure 5, parts A and B, respectively.
Figure 5. API-MS/MS CAD “product ion” negative ion spectra of
the ion peaks at 346 u and 393 u observed in the API-MS/MS CAD
“precursor ion” spectrum of the 230 u [PFBOH‚O2]- ion shown in
Figure 4. (A) The molecular weight 116 product (M1) is seen from the
loss of M1 from the parent adduct ion [PFBOH‚O2‚M1]-s [M1] ) 230
-
u and by the presence of a fragment ion that is the O2 adduct at 148
u. (B) The molecular weight 163 product (M2) is seen from the loss of
M2 from the parent adduct ion [PFBOH‚O2‚M2]-s[M2] ) 230 u and
by the presence of a fragment ion that is the O2- adduct at 195 u. The
230, 193, and 165 u ion fragments are characteristic of PFBOH.
TABLE 3: Product Species Observed from the Reactions of
the OH Radical with n-Alkanes in the Presence of NO, from
the Presence of [PFBOH‚O2‚M]- Ions
molecular weight of product M
alkane
n-alkane
n-alkane-d2n+2
Table 3 shows the molecular weights of the reaction products
observed from the n-alkanes and deuterated n-alkanes studied
using this procedure. Noting that species containing -OD
groups rapidly D/H exchange to form -OH groups under our
experimental conditions,10,11,16,28 the product species observed
from the n-alkanes are CnH2n(OH)(ONO2) and CnH2n-1(O)(OH)
species and those from the n-alkane-d2n+2 reactions are CnD2n-
(OH)(ONO2) and CnD2n-1(O)(OH) species. The CnH2n(OH)-
(ONO2) species (and their deuterated analogues) are attributed
to the 1,4-hydroxynitrates formed as shown in Scheme 1. The
CnH2n-1(O)(OH) species (and their deuterated analogues) are
attributed to the 1,4-hydroxycarbonyls formed as also shown
in Scheme 1 and previously observed by us10,11,16 using API-
MS and API-MS/MS in the positive ion mode, with [(H3O)‚
(H2O)n]+ clusters as the ionizing agent, and by Eberhard et al.15
using GC-MS analyses after derivatization by 2,4-dinitro-
phenylhydrazine.
n-pentane
102
149
116
163
130
177
144
191
111
159
127
175
143
191
159
207
n-hexane
n-heptane
n-octane
g1012 molecule cm-3. For example, in Figure 3 the ion peak at
407 u is attributed to the [PFBOH‚NO2]- adduct with the
hydroxynitrate. However, the corresponding adduct did not form
with the hydroxycarbonyl and as the NO2 increased during the
reaction, the signal for the [PFBOH‚O2‚hydroxycarbonyl]-
adduct at 346 u disappeared. Therefore, because NO2 is formed
from the photooxidation of methyl nitrite (and also from
reactions of HO2 and organic peroxy radicals with NO), this
limited the potential of using [PFBOH‚O2‚M]- ions for quan-
tification.
While the formation of [PFBOH‚O2‚M]- ions was useful for
hydroxycarbonyl and hydroxyalkyl nitrate identification, this
method did not appear to be suitable for quantification purposes.
NO2 Addition. It was observed that in the absence of PFBOH,
the hydroxycarbonyls and hydroxyalkyl nitrates formed adducts
-
-
-
It was observed that the reactions of NO2 ions with PFBOH
with O2 and with NO2 ions and that the majority of the
(to form [PFBOH‚NO2]-) competed with the corresponding
adducts at the end of the alkane photooxidation were NO2
adducts. Figure 6 shows an API-MS spectrum of an irradiated
-
-
reaction of O2 ions with PFBOH at NO2 concentrations of

1026 J. Phys. Chem. A, Vol. 105, No. 6, 2001
Arey et al.
TABLE 4: Formation Yields of Hydroxycarbonyls and
Hydroxyalkyl Nitrates from the Reaction of the OH Radical
with n-Alkanes in the Presence of NO, As Obtained from
-
Negative Ion API-MS Using NO2 as the Reagent Ion
n-pentane
Hydroxycarbonyls
1.33 ( 0.67 1.97 ( 0.61 1.69 ( 0.29
n-hexane
n-heptane n-octane
relative yielda
relative to
internal standardb
yieldc
1.00
0.24
0.60
0.49
0.29
0.36
0.53
0.46
0.27
Hydroxyalkyl nitrates
0.48 ( 0.12 0.85 ( 0.20 0.87 ( 0.05
relative yielda
relative to
internal standardd
yieldc
1.00
0.056
0.024
0.046
0.048
0.026
0.046
0.047
0.054
a From five experiments with all four n-alkanes in each irradiated
CH3ONO/NO/n-alkane/air mixture. Indicated errors are one standard
deviation. b Yield for each alkane from an irradiated CH3ONO/NO/n-
alkane/air mixture with that alkane only present and with a measured
amount of added 5-hydroxy-2-pentanone as the internal standard. c Yield
obtained from combining the relative yields with the relative yield based
on the internal standard for each n-alkane. Overall uncertainties are
estimated to be a factor of 2. d From two of the above five experiments
with CH3CH(OH)CH(ONO2)CH3 being formed in situ from the reaction
of the OH radical with cis-2-butene (see text). The relative hydroxyalkyl
nitrate yields in these two experiments were within 10% of the average
from all five experiments.
Figure 6. API-MS negative ion spectrum of an irradiated CH3ONO/
NO/n-pentane/n-hexane/n-heptane/n-octane/air mixture, with NO2 added
to the chamber after the irradiation and an NO2 concentration of 3 ×
1013 molecule cm-3. The initial n-alkane concentrations were each ∼2.2
× 1012 molecule cm-3 and the extents of reaction ranged from 5.6%
for n-pentane to 13% for n-octane. The C5- through C8-hydroxycar-
-
bonyls are observed as their NO2 adducts at 148, 162, 176, and 190
u, respectively, and the C5- through C8-hydroxyalkyl nitrates are
observed as their NO2 adducts at 195, 209, 223, and 237 u,
respectively. The concentration of the C5-hydroxyalkyl nitrate (expected
n-pentane reaction and the only commercially available 1,4-
hydroxycarbonyl) as the internal standard for quantification of
the 1,4-hydroxycarbonyls formed from the n-alkanes. The
hydroxyalkyl nitrate 2-nitrooxy-3-butanol CH3CH(ONO2)CH-
(OH)CH3 was formed in situ in the chamber from the reaction
of the OH radical with cis-2-butene [using an average of the
literature formation yields measured by Muthuramu et al.29
(0.037 ( 0.009) and O’Brien et al.30 (0.034 ( 0.005)] as the
internal standard for quantification of the 1,4-hydroxyalkyl
nitrates formed from the n-alkanes.
After a series of preliminary experiments to investigate this
approach for quantification of hydroxycarbonyls and hydroxy-
alkyl nitrates, we carried out one experiment for each n-alkane
with 5-hydroxy-2-pentanone being added as the internal standard
after the irradiation (for n-pentane, this corresponds to adding
an additional known amount of 5-hydroxy-2-pentanone to that
formed in the reaction). In addition to the formation of the Cn-
hydroxycarbonyl from the Cn n-alkane, these experiments
showed the formation (numbers in parentheses are the percents
relative to the Cn-hydroxycarbonyl) of C4- (25%), C5- (15%),
C6- (11%), and C7-hydroxycarbonyls (21%) from the n-octane
reaction, C4- (14%), C5- (10%), and C6-hydroxycarbonyls (12%)
from the n-heptane reaction, and C4- (10%) and C5-hydroxy-
carbonyls (17%) from the n-hexane reaction. These observations
indicate that decomposition reactions, such as that illustrated
in Scheme 1 for the 2-hexoxy radical, must occur. Note that it
is possible that the ion peaks attributed to NO2- adducts of Cn-1
hydroxycarbonyls were in part O2- adducts of the Cn hydroxy-
carbonyls.
-
to be almost totally 4-nitrooxypentan-1-ol) is estimated to be 3.3 ×
109 molecule cm-3
.
CH3ONO/NO/n-pentane/n-hexane/n-heptane/n-octane/air mix-
ture, with the C5- through C8-hydroxycarbonyls being observed
as their NO2- adducts at 148, 162, 176, and 190 u, respectively,
and the C5- through C8-hydroxyalkyl nitrates being observed
as their NO2- adducts at 195, 209, 223, and 237 u, respectively.
API-MS spectra and API-MS/MS “precursor ion” spectra of
the 32 u O2- ion and the 46 u NO2- ion showed that reactions
of the NO2- ion to form the [NO2‚M]- adducts dominated over
the corresponding reactions of the O2- ion at NO2 concentrations
g(2-3) × 1013 molecule cm-3, and that under these NO2
concentration conditions the dominant ion peaks involving the
hydroxcarbonyls and hydroxyalkyl nitrates were the [NO2‚M]-
ions.
Therefore addition of NO2 after the reaction was employed
to attempt to quantify the hydroxycarbonyl and hydroxyalkyl
nitrates in these OH radical-initiated reactions, with an internal
standard of similar chemical structure as the reaction products
of interest also being introduced into the chamber. A series of
experiments were carried out in which measured amounts of
5-hydroxy-2-pentanone, 4-hydroxy-4-methyl-2-pentanone, and
1-hydroxy-2-butanone were introduced into the chamber in the
presence of (1-3) × 1013 molecule cm-3 of NO2 (the
concentrations of the hydroxycarbonyls in the chamber were
in the range (2.4-24) × 1011 molecule cm-3). The API-MS
spectra obtained from two independent experiments showed that
the intensities of the [NO2‚hydroxycarbonyl]- adducts depended
on the particular hydroxycarbonyl M, with the relative API-
MS ion intensities of the [NO2‚hydroxycarbonyl]- adducts for
equal concentrations of the hydroxycarbonyls being the follow-
ing: 1-hydroxy-2-butanone (a 1,2-hydroxycarbonyl), 1.0; 5-hy-
droxy-2-pentanone (a 1,4-hydroxycarbonyl), 0.5 ( 0.1; and
4-hydroxy-4-methyl-2-pentanone (a 1,3-hydroxycarbonyl), 1.9
( 0.3. Given this range of responses, we used 5-hydroxy-2-
pentanone (the dominant 1,4-hydroxycarbonyl formed from the
Five experiments with all four n-alkanes present were also
carried out to determine the relative amounts of the various
hydroxycarbonyls formed (with no addition of 5-hydroxy-2-
pentanone). The initial concentrations of the n-alkanes were
varied over the range (0.206-2.49) × 1013 molecule cm-3, and
the maximum extent of reaction was 12-14% of the initial
n-alkane. Correcting for secondary reactions of the hydroxy-
carbonyls and for formation from the decomposition pathway,
these experimental data lead to the hydroxycarbonyl formation
yields given in Table 4.

NOx-Air Photooxidations of C5-C8 n-Alkanes
J. Phys. Chem. A, Vol. 105, No. 6, 2001 1027
The above CH3ONO/NO/n-alkane/air irradiations with a
single n-alkane present also showed the formation (numbers in
parentheses are the percents relative to the Cn-hydroxyalkyl
nitrate) of C5- (3%), C6- (3%), and C7-hydroxyalkyl nitrates
(7%) from the n-octane reaction, C5- (1.5%) and C6-hydroxy-
alkyl nitrates (6.5%) from the n-heptane reaction, and C5-
hydroxyalkyl nitrates (9%) from the n-hexane reaction. In two
of the experiments with all four n-alkanes present, CH3CH-
(OH)CH(ONO2)CH3 was formed in situ from the reaction of
the OH radical with cis-2-butene and used as an internal standard
for hydroxyalkyl nitrate quantification, and in the three ad-
ditional experiments the relative yields of hydroxyalkyl nitrates
from the four n-alkanes were measured. The hydroxyalkyl nitrate
formation yields resulting from these experiments are also given
in Table 4. It is of interest to note that the API-MS is quite
sensitive to hydroxyalkyl nitrates (more so than to hydroxy-
carbonyls, as evident from Figure 6 and the formation yield
data shown in Table 4), with (Figure 6) detection limits of <50
part-per-trillion (ppt) mixing ratio (<1 × 109 molecule cm-3);
indeed, dilution of the chamber contents by a factor of 8
indicated that 10 ppt mixing ratios (2 × 108 molecule cm-3) of
these hydroxyalkyl nitrates could still be readily observed.
Contract No. 97-312. The statements and conclusions are those
of the authors and not necessarily those of the California Air
Resources Board.
References and Notes
(1) Atkinson, R. J. Phys. Chem. Ref. Data 1997, 26, 215.
(2) Logan, J. A. J. Geophys. Res. 1985, 90, 10463.
(3) Darnall, K. R.; Carter, W. P. L.; Winer, A. M.; Lloyd, A. C.; Pitts,
J. N., Jr. J. Phys. Chem. 1976, 80, 1948.
(4) Takagi, H.; Washida, N.; Bandow, H.; Akimoto, H.; Okuda, M. J.
Phys. Chem. 1981, 85, 2701.
(5) Atkinson, R.; Aschmann, S. M.; Carter, W. P. L.; Winer, A. M.;
Pitts, J. N., Jr. J. Phys. Chem. 1982, 86, 4563.
(6) Atkinson, R.; Carter, W. P. L.; Winer, A. M. J. Phys. Chem. 1983,
87, 2012.
(7) Atkinson, R.; Aschmann, S. M.; Carter, W. P. L.; Winer, A. M.;
Pitts, J. N., Jr. Int. J. Chem. Kinet. 1984, 16, 1085.
(8) Atkinson, R.; Aschmann, S. M.; Winer, A. M. J. Atmos. Chem.
1987, 5, 91.
(9) Harris, S. J.; Kerr, J. A. Int. J. Chem. Kinet. 1989, 21, 207.
(10) Atkinson, R.; Kwok, E. S. C.; Arey, J.; Aschmann, S. M. Faraday
Discuss. 1995, 100, 23.
(11) Aschmann, S. M.; Chew, A. A.; Arey, J.; Atkinson, R. J. Phys.
Chem. A 1997, 101, 8042.
(12) Platz, J.; Sehested, J.; Nielsen, O. J.; Wallington, T. J. J. Phys.
Chem. A 1999, 103, 2688.
(13) Orlando, J. J.; Iraci, L. T.; Tyndall, G. S. J. Phys. Chem. A 2000,
104, 5072.
Conclusions
(14) Carter, W. P. L.; Atkinson, R. J. Atmos. Chem. 1989, 8, 165.
(15) Eberhard, J.; Mu¨ller, C.; Stocker, D. W.; Kerr, J. A. EnViron. Sci.
Technol. 1995, 29, 232.
(16) Kwok, E. S. C.; Arey, J.; Atkinson, R. J. Phys. Chem. 1996, 100,
214.
(17) Atkinson, R.; Carter, W. P. L.; Winer, A. M.; Pitts, J. N., Jr. J. Air
Pollut. Control Assoc. 1981, 31, 1090.
(18) Atkinson, R.; Tuazon, E. C.; Aschmann, S. M. EnViron. Sci.
Technol. 1995, 29, 1674.
(19) Scanlon, J. T.; Willis, D. E. J. Chromatogr. Sci. 1985, 23, 333.
(20) French, J. B.; Thomson, B. A.; Davidson, W. R.; Reid, N. M.;
Buckley, J. A. Atmospheric Pressure Chemical Ionization Mass spectrom-
etry. In Mass Spectrometry in EnVironmental Sciences; Karasek, F. W.,
Hutzinger, O., Safe, S., Eds.; Plenum Press: New York, 1985; pp 101-
121.
(21) Atkinson, R. J. Phys. Chem. Ref. Data 1989, Monograph 1, 1.
(22) Atkinson, R. Atmospheric Oxidation. In Handbook of Property
Estimation Methods for Chemicals: EnVironmental and Health Sciences;
Boethling, R. S., Mackay, D., Eds.; Lewis Publishers: Boca Raton, 2000,
pp 335-354.
(23) Atkinson, R.; Aschmann, S. M.; Arey, J.; Carter, W. P. L. Int. J.
Chem. Kinet. 1989, 21, 801.
(24) Kwok, E. S. C.; Atkinson, R. Atmos. EnViron. 1995, 29, 1685.
(25) Atkinson, R. Int. J. Chem. Kinet. 1987, 19, 799.
(26) Aschmann, S. M.; Reissell, A.; Atkinson, R.; Arey, J. J. Geophys.
Res. 1998, 103, 25553.
Our alkyl nitrate, hydroxyalkyl nitrate, and hydroxycarbonyl
formation yields given in Table 4, combined with the literature
carbonyl yields arising from alkoxy radical decomposition and
reaction with O2, account for ∼95%, ∼80%, ∼70%, and ∼55%
of the products formed from the OH radical-initiated reactions
of n-pentane, n-hexane, n-heptane, and n-octane (in the presence
of NO), respectively, with uncertainties of approximately a factor
of 1.4 in each case. We can therefore now account for most of
the reaction products formed from these n-alkanes, and hy-
droxycarbonyl formation accounts for a significant fraction of
these products (in fact, for most of the products from n-hexane,
n-heptane, and n-octane). Hydroxyalkyl nitrate formation ac-
counts for a few percent of the overall reaction products, and is
consistently ∼25% of the corresponding alkyl nitrate yields.
Our hydroxyalkyl nitrate yield from the n-hexane reaction of
0.046 (with an uncertainty of a factor of 2) is reasonably
consistent with the approximate upper limit estimate of Eberhard
et al.15 for the formation of 5-nitrooxyhexan-2-ol of 0.03-0.04.
Thus significant fractions of the atmospheric reaction products
of gC5 n-alkanes (and other alkanes whose intermediate alkoxy
radicals can isomerize) are comprised of hydroxycarbonyls and
hydroxyalkyl nitrates, compound classes which are presently
difficult to analyze under atmospheric conditions.
(27) Tuazon, E. C.; Alvarado, A.; Aschmann, S. M.; Atkinson, R.; Arey,
J. EnViron. Sci. Technol. 1999, 33, 3586.
(28) Kwok, E. S. C.; Atkinson, R.; Arey, J. EnViron. Sci. Technol. 1995,
29, 2467.
(29) Muthuramu, K.; Shepson, P. B.; O’Brien, J. M. EnViron. Sci.
Technol. 1993, 27, 1117.
(30) O’Brien, J. M.; Czuba, E.; Hastie, D. R.; Francisco, J. S.; Shepson,
P. B. J. Phys. Chem. A 1998, 102, 8903.
Acknowledgment. The authors gratefully thank the Cali-
fornia Air Resources Board for supporting this research through