Experimental investigation of the low temperature oxidation of the five isomers of hexane

Article abstract of DOI:10.1021/jp503772h

The low-temperature oxidation of the five hexane isomers (n-hexane, 2-methyl-pentane, 3-methyl-pentane, 2,2-dimethylbutane, and 2,3-dimethylbutane) was studied in a jet-stirred reactor (JSR) at atmospheric pressure under stoichiometric conditions between 550 and 1000 K. The evolution of reactant and product mole fraction profiles were recorded as a function of the temperature using two analytical methods: gas chromatography and synchrotron vacuum ultraviolet photoionization mass spectrometry (SVUV-PIMS). Experimental data obtained with both methods were in good agreement for the five fuels. These data were used to compare the reactivity and the nature of the reaction products and their distribution. At low temperature (below 800 K), n-hexane was the most reactive isomer. The two methyl-pentane isomers have about the same reactivity, which was lower than that of n-hexane. 2,2-Dimethylbutane was less reactive than the two methyl-pentane isomers, and 2,3-dimethylbutane was the least reactive isomer. These observations are in good agreement with research octane numbers given in the literature. Cyclic ethers with rings including 3, 4, 5, and 6 atoms have been identified and quantified for the five fuels. While the cyclic ether distribution was notably more detailed than in other literature of JSR studies of branched alkane oxidation, some oxiranes were missing among the cyclic ethers expected from methyl-pentanes. Using SVUV-PIMS, the formation of C ₂-C₃ monocarboxylic acids, ketohydroperoxides, and species with two carbonyl groups have also been observed, supporting their possible formation from branched reactants. This is in line with what was previously experimentally demonstrated from linear fuels. Possible structures and ways of decomposition of the most probable ketohydroperoxides were discussed. Above 800 K, all five isomers have about the same reactivity, with a larger formation from branched alkanes of some unsaturated species, such as allene and propyne, which are known to be soot precursors.

Full text of DOI:10.1021/jp503772h

Article
Experimental Investigation of the Low Temperature
Oxidation of the Five Isomers of Hexane
Zhandong Wang, Olivier Herbinet, Zhanjun Cheng, Benoit
Husson, René Fournet, Fei Qi, and Frédérique Battin-Leclerc
J. Phys. Chem. A, Just Accepted Manuscript • DOI: 10.1021/jp503772h • Publication Date (Web): 09 Jul 2014
Downloaded from http://pubs.acs.org on July 14, 2014
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
The Journal of Physical Chemistry A is published by the American Chemical Society.
1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 44
The Journal of Physical Chemistry
Experimental Investigation of the Low Temperature Oxidation
of the Five Isomers of Hexane
1
2
3
4
5
6
7
8
9
Zhandong WANG1,2, Olivier HERBINET1,*, Zhanjun Cheng2, Benoit HUSSON1, René Fournet1, Fei QI2, Frédérique
BATTIN-LECLERC1
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1Laboratoire Réactions et Génie des Procédés, Nancy Université, CNRS UPR 3349,
BP 20451, 1 rue Grandville, 54000 Nancy, France
2National Synchrotron Radiation Laboratory, University of Science and Technology of China,
Hefei, Anhui 230029, P. R. China.
Abstract
The low-temperature oxidation of the five hexane isomers (n-hexane, 2-methyl-pentane, 3-methyl-pentane,
2,2-dimethyl-butane and 2,3-dimethyl-butane) was studied in a jet-stirred reactor (JSR) at atmospheric pressure
under stoichiometric conditions between 550 and 1000 K. The evolution of reactant and product mole fraction
profiles were recorded as a function of the temperature using two analytical methods: gas chromatography and
synchrotron vacuum ultraviolet photoionization mass spectrometry (SVUV-PIMS). Experimental data obtained with
both methods were in good agreement for the five fuels. These data were used to compare the reactivity, the nature
of the reaction products and their distribution. At low temperature (below 800 K), n-hexane was the most reactive
isomer. The two methyl-pentane isomers have about the same reactivity, which was lower than that of n-hexane.
2,2-dimethyl-butane was less reactive than the two methyl-pentane isomers, and 2,3-dimethyl-butane was the least
reactive isomer. These observations are in good agreement with research octane numbers given in literature. Cyclic
ethers with rings including 3, 4, 5 and 6 atoms have been identified and quantified for the five fuels. While the cyclic
ether distribution was notably more detailed than in other literature JSR studies of branched alkane oxidation, some
expected oxiranes were missing amongst the cyclic ethers expected from methyl-pentanes. Using SVUV-PIMS, the
formation of C2-C3 monocarboxylic acids, ketohydroperoxides, and species with two carbonyl groups has also been
observed, supporting their possible formation from branched reactants. This is in line with what was previously
experimentally demonstrated from linear fuels. Possible structures and ways of decomposition of the most probable
ketohydroperoxides were discussed. Above 800 K, all the five isomers have about the same reactivity, with a larger
formation from branched alkanes of some unsaturated species, such as allene and propyne, which are known to be
soot precursors.
1
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 2 of 44
*Corresponding author information:
1
2
Olivier HERBINET
3
4
5
6
7
8
9
Laboratoire Réactions et Génie des Procédés
Ecole Nationale Supérieure des Industries Chimiques
BP 20451
1 rue Grandville
54000 Nancy, France
Tel: +33 (0)3 83 17 53 60
Fax: +33 (0)3 83 17 81 20
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
E-mail: olivier.herbinet@univ-lorraine.fr
Keywords: oxidation, jet-stirred reactor; synchrotron VUV mass spectrometry; gasoline; hexane
2
ACS Paragon Plus Environment

Page 3 of 44
1. Introduction
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
n-Alkanes and iso-alkanes are two important families of hydrocarbons which are present in conventional
fuels derived from petroleum1 and Fischer-Tropsch fuels2. The oxidation of n-alkanes3–16 and iso-alkanes3–8,17–20 has
been the subject of numerous experimental studies. But there are only a few experimental studies with the aim to
show the difference of reactivity at low temperature according to the structure of the different isomers of alkanes3–
5,7,8
.
9
Several studies compared the ignition delay times of n- and iso-butanes3–6. In 2007, Ogura et al.3 measured
ignition delay times of mixtures of n-butane and iso-butane (1% of fuel, equivalence ratio of 0.72 with dilution in
argon) behind reflected shock waves by monitoring time histories of UV emission and pressure in the temperature
range 1200−1600 K. This experimental study showed that the igniꢀon delay ꢀme of iso-butane was longer than that
of n-butane. In 2010, Healy et al.4,5 performed rapid compression machine and shock-tube ignition experiments for
fuel/air mixtures for n- and iso-butanes at equivalence ratios of 0.3, 0.5, 1, and 2. The wide range of studied
experimental conditions included temperatures from 590 to 1567 K and pressures of approximately 1, 10, 20, 30 and
45 atm. It was also found that iso-butane was less reactive than n-butane and that both species exhibit a slight
negative temperature coefficient (NTC) behavior. Gersen et al.6 also measured ignition delay times of n- and
iso-butanes in a rapid compression machine in the temperature range 660–1010 K, at pressures from 14 to 36 bar
and at equivalence ratios of 1.0 and 0.5. They obtained conclusions similar to that of Healy et al.4,5.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
C5 and C7 alkane isomers7,8 are the only other species for which ignition delay times were compared in an
extensive way. In 1996, Minetti et al.7 measured ignition delay times of n-pentane, iso-pentane and neo-pentane in a
rapid compression machine over the temperature range 600-950 K, the pressure range 6-11 bar, at an equivalence
ratio of 1, and in air. This study again showed that the linear isomer was the most reactive species and that neo-
pentane, the most branched isomer, was the least reactive, with the reactivity of iso-pentane being in between. In
2005, Silke et al.8 measured ignition delay times for the nine isomers of heptane (n-heptane, 2-methyl-hexane,
3-methyl-hexane, 2,2-dimethyl-pentane, 2,3-dimethyl-pentane, 2,4-dimethyl-pentane, 3,3-dimethyl-pentane,
3-ethyl-pentane, and 2,2,3-trimethyl-butane) using a rapid compression machine. Stoichiometric fuel-air mixtures
were studied at compressed gas pressures of 10, 15, and 20 atm for the n-heptane study, and at 15 atm for all other
isomers, in the compressed gas temperature range of 640–960 K. NTC behavior was observed for each of the
isomers, with the most branched isomer exhibiting a lower reactivity than other isomers. Besides, three branched
isomers of octane (2-methyl-heptane19, 3-methyl-heptane20 and 2,5-dimethyl-hexane21) have been studied in a
jet-stirred reactor operated at 10 atm under the same conditions (initial mole fraction of 0.001). The results indicate
about the same reactivity for the three compounds.
All these studies showed that linear alkanes are more reactive than branched species and that the position
of methyl groups can also have an impact on the reactivity and the product selectivity. This observation can be
related to the evolution of octane numbers for the anti-knock rating of gasoline fuels. As an example, the research
octane number (RON) of n-butane is 94 whereas the RON of iso-butane is 102, meaning that iso-butane ignites more
difficultly than n-butane under the engine test conditions. The ability of n-butane to ignite easily is related to the
3
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 4 of 44
higher reactivity of linear alkanes at low temperature. The linearity of the molecule favors the formation of
branching agents providing the multiplication of radicals and the heat release22,23
1
2
.
3
4
5
6
7
8
9
Eskola et al. studied the reactions of n- and iso-butyl radicals with oxygen using chorine laser-photolysis to
initiate the low-temperature oxidation of n-butane24 or iso-butane25 in a flow tube using molecular beam
photoionization mass spectrometry. They observed the formation of ketohydroperoxides from the linear reactant,
but not from the branched one.
The goal of this paper is to present new experimental results from the study of the low-temperature
oxidation of the five isomers of hexane (see structures in Figure 1) with an analysis of a maximum of reaction
products, including ketohydroperoxides which have not yet been detected from branched reactants. According to
the research octane numbers26 given in Table 1, n-hexane is the most reactive isomer, 2- and 3-methyl-pentanes
have about the same reactivity, 2,2-dimethyl-butane is much less reactive than the two methyl-pentane isomers and
2,3-dimethyl-butane is the least reactive isomer. In spark ignition engines, due to gas compression, fuels with low
octane numbers (e.g., n-heptane) start to react and self-ignite before the spark plug ignition occurs. This is the cause
of a sharp metallic sound, called knocking. Research octane numbers, which characterize the tendency of fuels to
resist to auto-ignition, are measured in a cooperative fuel research engine. The fuel reactivity during the
compression is directly related to low-temperature oxidation chemistry which is responsible for the cool flame
occurring prior to the self-ignition26.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Experiments carried out in this work were performed using a jet-stirred reactor. This type of reactor is well
adapted to the kinetic studies of fuel low-temperature oxidation chemistry, although the conditions are different
from those observed in an engine (the present reactor is an isothermal continuous flow reactor working at
atmospheric pressure, whereas an engine is a closed vessel with variable volume working at significantly higher
pressures, up to 30 bars). The advantage of using this type of reactor is the possibility of comparing the reactivity of
the different hexane isomers, but also of highlighting the differences in the nature and the concentration profiles of
the reaction products under well defined conditions. These data are useful for a better understanding of the low-
temperature oxidation chemistry of alkanes and for further validations of detailed kinetic models of the oxidation of
hexane isomers. Experiments were performed using two complementary analytical methods: gas chromatography
analyses in Nancy (France) and synchrotron vacuum ultraviolet photoionization mass spectrometry (SVUV-PIMS)
analyses in Hefei (China). The same reactor and analytical methods have been recently used to investigate the
low-temperature oxidation of the three isomers of hexene27.
4
ACS Paragon Plus Environment

Page 5 of 44
The Journal of Physical Chemistry
1
2
n-hexane
3
4
5
6
7
8
9
2-methyl-pentane
3-methyl-pentane
10
11
12
13
14
15
16
17
18
19
20
21
22
23
2,2-dimethyl-butane
2,3-dimethyl-butane
Figure 1: Structures of the five isomers of hexane.
Table 1: Research octane numbers (RON) of the five hexane isomers 26
.
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Species
RON
24.8
73.4
74.5
91.8
103.5
n-hexane
2-methyl-pentane
3-methyl-pentane
2,2-dimethyl-butane
2,3-dimethyl-butane
2. Experimental method
The oxidation of the isomers of hexane was studied using a jet-stirred reactor (JSR), a type of reactor
described in detail in 28. This type of reactor has often been used by our research team in Nancy for gas phase kinetic
studies of the oxidation of hydrocarbon and oxygenated molecule under isothermal conditions29–32. Experiments
were performed at a constant pressure of 1.06 bar, at a residence time of 2±0.1 s, at temperatures ranging from 550
to 1000 K, under stoichiometric conditions and the fuel was diluted in an inert gas (inlet mole fraction of 0.02 for the
five isomers and 0.04 for 2,2- and 2,3-dimethyl-butanes). Inlet mole fraction values were chosen for the least
reactive compound to show enough reactivity at low temperature. However, the chosen values are relatively high
and the range of investigated temperatures had to be reduced for the most reactive species to avoid a too large
exothermicity in the reactor (as an example, the oxidation of n-hexane was studied in the temperature range 550-
850 K). The inert gas was helium in Nancy and Argon in Hefei. Argon was required in Hefei because its signal at a
photo-ionization energy of 16.6 eV is used for the normalization of the signals of other peaks.
5
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 6 of 44
The reactor, made of fused silica, consists of a quartz sphere (volume ≈ 90 cm3) into which the diluted
reactant enters through an injection cross located at its center. A lateral sampling cone was inserted in the reactor
used in Hefei for the connection with the mass spectrometer. The reactor is operated at constant temperature and
pressure and it is preceded by an annular preheating zone in which the temperature of the gases is increased up to
the reaction temperature before entering into the reactor. Gas mixture residence time inside the annular preheater
is very short compared to the residence time inside the reactor (a few percent). In Nancy, both the spherical reactor
and the annular preheating zone are heated by Thermocoax heating resistances rolled up around their walls. In
Hefei, an oven having the same shape as the reactor with its lateral cone was used to provide the heating of the
reactor part. The uncertainty in the reaction temperature measurement was estimated to ± 5% with both devices
(resistances in Nancy and oven in Hefei). Gas flow rates for the feed of the reactor were controlled by mass flow
controllers provided by Bronkhorst for the experiments in Nancy and by mass flow controllers provided by MKS and
Bronkhorst in Hefei. According to manufacturers, errors in mass flow measurements were ± 0.5%. Liquid flow rates
were controlled by Coriolis flow controllers provided by Bronkhorst. The uncertainty in fuel flow rates was ± 1%.
Fuels used in China were provided by Aladdin Reagent (n-hexane assay ≥ 99% and 2,3-dimethyl-butane
assay ≥ 98%) and by Alfa Aesar (methyl-pentane isomers assay ≥ 99% and 2,2-dimethyl-butane assay ≥ 98.5%).
Fuels used in France were provided by Sigma-Aldrich (assay ≥ 97% for n-hexane, ≥ 98% for 2,3-dimethyl-butane,
and ≥ 99% for other isomers).
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2.1. Gas chromatography analyses (Nancy)
Analyses in Nancy were performed using online gas chromatography. The online analysis of products which
are liquid under standard conditions was performed by a heated transfer line between the reactor outlet and the
chromatograph sampling gate which was also heated. The temperature of the transfer line was set at 423 K. This
temperature was high enough to keep all the reaction products in the gas phase. Three gas chromatographs were
used for the quantification of the different species.
The first gas chromatograph, equipped with a Carbosphere packed column, a thermal conductivity detector
(TCD), and a flame ionization detector (FID), was used for the quantification of O2, CO, CO2, methane, ethylene,
acetylene and ethane. The second one was fitted with a PlotQ capillary column and a FID and was used for the
quantification of molecules from methane to reaction products with up to 5 carbon atoms and 1 to 2 oxygen atoms
maximum. The third one was fitted with a HP-1 capillary column and a FID. It was used for the quantification of
heavier species with more than five heavy atoms.
Note that C2-C4 carboxylic acids were observed by GC analyses, but with much larger mole fractions than
those measured by SVUV-PIMS (about a factor of 10). According to previous studies 32, larger mole fractions of these
species measured by GC are likely due to the decomposition of unstable species (e.g., ketohydroperoxides) in the
heated transfer line between the outlet of the reactor and the GC. These results will then not be presented in this
paper.
The identification of reaction products which were not calibrated before was performed using a gas
chromatograph equipped with a PlotQ or HP-1 capillary column and a mass spectrometer. The mass spectra of most
6
ACS Paragon Plus Environment

Page 7 of 44
The Journal of Physical Chemistry
reaction products were included in the NIST 08 mass spectral library. When mass spectra were not available (e.g., for
species like cyclic ethers formed from the low-temperature oxidation of branched alkanes), the deciphering of the
experimental mass spectra was carried out. The mass spectra of all the cyclic ethers which are not available in the
NIST 08 mass spectral library are given in Supplementary Material.
1
2
3
4
5
6
7
8
9
The calibration was performed by injecting known amounts of the pure substances when available,
otherwise the method of the effective carbon number was used (species having the same number of carbon atoms
and the same functional groups were assumed to have the same response in the FID). The detection threshold was
about 100 ppb for the heaviest species (FID) and about 100 ppm for carbon monoxide, carbon dioxide and oxygen
(TCD). Based on the repeatability of the experiments, uncertainty estimates on obtained mole fractions were about ±
5% (measurement reading) for species which were calibrated using gaseous standards (oxygen, fuels, CO, CO2, C1-C4
hydrocarbons) and ±10% for those calibrated using the effective carbon number method33.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2.2. SVUV-PIMS analysis (Hefei)
In Hefei, species from the reactor were analyzed by a reflectron time-of-flight mass spectrometer (RTOF MS)
with photo-ionization by synchrotron radiation. This apparatus was already described in detail in 34. The reactor was
coupled to the low pressure photo-ionization chamber through a lateral fused silica cone-like nozzle which was
inserted in the spherical part. The tip of the cone was pierced by a 50 µm orifice. A nickel skimmer with a 1.25 mm
diameter aperture was located 15 mm downstream from the sampling nozzle. The sampled gases formed a
molecular beam, which passed horizontally through the 10 mm gap between the repeller and extractor plates of
RTOF MS. The molecular beam intersected perpendicularly with synchrotron vacuum ultraviolet light. The ion signal
was then detected with the RTOF MS, which was installed in the photo-ionization chamber vertically. Details about
the tunable synchrotron light are available in 35. The evolution of ion signals was recorded as a temperature function
(from 500 to 700 K) over the mass range 2-172 Da and at several photon energies (9.5-10-10.5-11-13-16.6 eV). Scans
in energy were also performed at the temperature of 625 K for the identification of species according to their
ionization energies. The ion signal of argon measured at 16.6 eV was used to normalize ion signals obtained for other
masses.
The quantification of reactants (fuels and oxygen) was carried out by measuring the signal under unreactive
conditions. Reaction products were quantified from equation (1) using another species as reference (a reactant or a
product, whose mole fraction and cross section are already known).
ꢂ ꢄ
ꢂ
ꢄ
ꢊꢅꢆꢇ
ꢉ
ꢋꢅꢆꢇ
ꢋꢁ
ꢈꢁ ꢃ, ꢉ
ꢂ ꢄ
ꢂ ꢄ
(1)
ꢀꢁ ꢃ = ꢀꢅꢆꢇ ꢃ ×
×
×
ꢂ
ꢄ
ꢂ ꢄ
ꢊꢁ ꢉ
ꢈꢅꢆꢇ ꢃ, ꢉ
ꢂ ꢄ
ꢂ
ꢄ
where: ꢀꢁ ꢃ and ꢈꢁ ꢃ, ꢉ are the mole fraction and the signal of species i at the temperature ꢃ and at the photon
energy ꢉ, respectively.
ꢂ ꢄ
ꢊꢁ ꢉ is the photo-ionization cross-section of species i at the photon energy ꢉ.
ꢋꢁ is the mass discrimination factor of species i.
7
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 8 of 44
Carbon oxides, water and methane mole fractions were calculated using equation (1) and the signal of m/z
1
2
3
4
40 (argon) at 16.6 eV as reference. Cross sections at 16.6 eV used for the calculation were 32.1 36, 19.42 37, 34.88 38
15.1 39 and 44.38 Mb 40 for argon, carbon monoxide, carbon dioxide, water and methane, respectively.
,
Other products were quantified using the related hydrocarbon fuels as the reference. Cross sections were
only available in the literature for n-hexane41. Cross sections of the four other hexane isomers were measured, as
well as that of n-hexane, in the present study. The data are presented in Figure 2 as well as that of n-hexane
measured in41. Formaldehyde, methanol, acetic acid, propanoic acid and ethylene were quantified using the signals
obtained at a photon energy of 11 eV; acetaldehyde, propene, 1-butene and iso-butene were quantified using the
signals at 10.5 eV. Cross sections at 11 eV were 9.0242, 3.6343, 6.8232, 11.9632 and 7.79 Mb43 for formaldehyde,
methanol, acetic acid, propanoic acid and ethylene, respectively. Cross sections at 10.5 eV were 4.444, 11.0945, 9.9145
and 10.08 Mb46 for acetaldehyde, propene, 1-butene and iso-butene, respectively.
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Uncertainties were estimated to ±10% for species which were directly calibrated using standards when
available (hexane fuels, oxygen and argon) and to ±25% for species quantified by the mean of equation (1) with
known photoionization cross sections.
20
n-hexane
2-methyl-pentane
3-methyl-pentane
2,3-dimethyl-butane
2,2-dimethyl-butane
15
10
5
0
9.0
9.5
10.0
10.5
11.0
11.5
Photon Energy (eV)
Figure 2: Evolution of the cross sections of the five hexane isomers as a function of the photon energy (data were
obtained in the present work except for n-hexane41).
3. Experimental results and discussion
Experiments performed in Hefei were repeated in Nancy using the same conditions. In the first part of this
section, both sets of experimental results obtained for 3-methyl-pentane are compared in order to show the
agreement between both analytical methods. The comparison for other hexane isomers is given as Supplementary
Material. In the second part, the reactivity and the concentration profiles of the common reaction products are
compared for the five isomers of hexane, using the data obtained by gas chromatography or by SVUV-PIMS for
species that could not be analyzed using gas chromatography. In the third part, the structures and the concentration
profiles of cyclic ethers and large alkenes are discussed. Finally the fourth part presents the signals obtained for
ketohydroperoxides and species including two carbonyl groups (named from here diones).
8
ACS Paragon Plus Environment

Page 9 of 44
The Journal of Physical Chemistry
1
2
3.1. Comparison of both sets of experimental results obtained for 3-methyl-pentane
3
4
5
6
7
8
9
Experimental mole fractions obtained with both analytical methods have been compared when available for
the different fuels. In the text, only the comparison for 3-methyl-pentane is displayed. The comparison for other
fuels is given as Supplementary Material (except for 2,3-dimethyl-butane which was not reactive under the
conditions used in Hefei). Figures 3 to 5 show that the agreement between both sets of data for 3-methyl-pentane is
overall satisfactory for both the reactants (3-methyl-pentane and oxygen in Figure 3) and the reaction products
(oxygenated species in Figure 4 and hydrocarbons in Figure 5). The agreement is also satisfactory for other fuels.
Data are displayed in Supplementary Material (Figures S1 to S9).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
25x10-3
0.20
0.15
0.10
0.05
0.00
3-methyl-pentane
Oxygen
20
15
10
5
0
500 550 600 650 700 750 800
Temperature (K)
500 550 600 650 700 750 800
Temperature (K)
Figure 3: Comparison of mole fraction profiles obtained using gas chromatography (ꢀ) and SVUV-PIMS (×) for
3-methyl-pentane and oxygen (fuel inlet mole fraction of 0.02). The signals recorded at 11 eV for m/z 86 and at 16.6
eV for m/z 32 were used for the quantification of 3-methyl-pentane and oxygen, respectively.
30x10-3
6x10-3
Carbon Dioxide
(SVUV PIMS data were obtained
from the signal at 16.6 eV for m/z 44)
Carbon Monoxide
(SVUV PIMS data were obtained
from the signal at 16.6 eV for m/z 28)
25
20
15
10
5
5
4
3
2
1
0
0
500
600
700
800
500
600
700
800
Temperature (K)
Temperature (K)
8x10-3
8x10-3
Methanol
Acetaldehyde
(SVUV PIMS data were obtained
from the signal at 10.5 eV for m/z 44)
(SVUV PIMS data were obtained
from the signal at 11 eV for m/z 32)
6
4
2
0
6
4
2
0
500
600
700
800
500
600
700
800
Temperature (K)
Temperature (K)
Figure 4: Comparison of mole fraction profiles obtained using gas chromatography (ꢀ) and SVUV-PIMS (×) for
oxygenated reaction products (3-methyl-pentane inlet mole fraction of 0.02).
9
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 10 of 44
1.0x10-3
5x10-3
Methane
Ethylene
1
2
3
4
5
6
7
8
(SVUV PIMS data were obtained
(SVUV PIMS data were obtained
from the signal at 11 eV for m/z 28)
from the signal at 16.6 eV for m/z 16)
4
3
2
1
0
0.8
0.6
0.4
0.2
0.0
500
600
700
800
500
600
700
800
9
Temperature (K)
Temperature (K)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1.2x10-3
Propene
(SVUV PIMS data were obtained
from the signal at 10 eV for m/z 42)
1.0
0.8
0.6
0.4
0.2
0.0
500
600
700
800
Temperature (K)
Figure 5: Comparison of mole fraction profiles obtained using gas chromatography (ꢀ) and SVUV-PIMS (×) for small
hydrocarbon reaction products (3-methyl-pentane inlet mole fraction of 0.02).
However, it should be noted that an important deviation can be observed between both used analytical
methods for propene. This is particularly the case for 3-methyl-pentane. This is due to the formation of ketene which
was simultaneously detected with propene by RTOF MS given that both species have close molecular weight (m/z
42.0797 for propene and m/z 42.0367 for ketene)34. The separation of these two species was not available by the
used RTOF. Note that ketene was not detected by GC (this may be due to adsorption or decomposition phenomena
in the transfer line between the reactor and the GC, and also because of the low concentrations of this species). As
can be seen from Figure 6, the visible threshold at around 9.2 eV measured in the SVUV-PIMS experiment well
corresponds to the ionization energy of ketene (9.617 eV47). Note also that for light species, such as methane or
ethylene, the production of fragments from heavier species may perturb the SVUV-PIMS measurements.
10
ACS Paragon Plus Environment

Page 11 of 44
The Journal of Physical Chemistry
120
80
40
0
100
m/z 42
(2-methyl-pentane)
m/z 42
(n-hexane)
1
2
3
4
5
6
7
8
80
60
40
20
0
propene
9.73
ketene
9.617
8.5
9.0
9.5
10.0 10.5 11.0
8.5
9.0
9.5
PIE (eV)
10.0 10.5 11.0
9
PIE (eV)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
80
60
40
20
0
30
m/z 42
(3-methyl-pentane)
m/z 42
(2,2-dimethyl-butane)
20
10
0
8.5
9.0
9.5
10.0
10.5
11.0
8.5
9.0
9.5
10.0
10.5
11.0
PIE (eV)
PIE (eV)
Figure 6: Photo-ionization efficiency spectra of m/z 42 obtained during the oxidation of hexane isomers.
Temperatures in the reactor were 653, 620, 620 and 623 K for n-hexane, 2- and 3-methyl-pentane isomers and
2,2-dimethyl-butane, respectively.
3.2. Comparison of the reactivity of the five isomers of hexane and of the mole fractions of common reaction
products
Figure 7 displays the comparison of the experimental fuel mole fraction profiles that were obtained by GC
for the five hexane isomers. It can be seen that n-hexane is the most reactive isomer with a conversion of about 75%
at 650 K for an initial mole fraction of 0.02 (Figure 7a) and that it is still reactive in the NTC region. The two
methyl-pentane isomers have about the same reactivity (conversion of about 50 % at a temperature of 625 K), which
is lower than that of n-hexane. The position of the methyl groups does not seem to have an impact on the reactivity
of the two methyl-pentane isomers. This phenomenon is in accord with the observation by Karsenty et al. for 2- and
3-methyl-heptanes 20. On the contrary, 2,2- and 2,3-dimethyl-butanes do not have the same reactivity. For an initial
mole fraction of 0.02, 2,3-dimethyl-butane has no reactivity at low temperature whereas 2,2-dimethyl-butane has a
conversion of about 25% at 625 K (Figure 7a). An initial mole fraction of 0.04 is needed for a 20% maximum
conversion of 2,3-dimethyl-butane (Figure 7b), while the maximum conversion of 2,2-dimethyl-butane is then close
to 50%. Note that these experimental observations are in good agreement with what was expected from the analysis
of the research octane numbers given in Table 1. Note that the oxidation of the three most reactive isomers was
performed over a limited range of temperature (up to 850 K) to avoid auto-ignition phenomena due to the high fuel
inlet mole fractions used in this study.
11
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 12 of 44
25x10-3
n-hexane
1
2
3
4
5
6
7
8
3-methyl-pentane
2-methyl-pentane
2,2-dimethyl-butane
2,3-dimethyl-butane
20
15
10
5
τ = 2 s
P = 106 kPa
9
xfuel = 0.02
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
xO = 0.19 (φ = 1)
2
0
400
600
800
1000
Temperature (K)
(a)
50x10-3
40
τ = 2 s
P = 106 kPa
xfuel = 0.04
xO = 0.38 (φ = 1)
2
30
20
2,2-dimethyl-butane
2,3-dimethyl-butane
10
0
400
600
800
1000
Temperature (K)
(b)
Figure 7: Comparison of the reactivity of the five hexane isomers (fuel inlet mole fraction of 0.02 in fig. (a) and 0.04
in fig. (b)). Mole fractions were obtained using gas chromatography.
Figures 8 to 12 display the comparison of the mole fraction profiles of C1-C5 hydrocarbons and C1-C4
oxygenated reaction products obtained in the oxidation of the five hexane isomers using GC for the analysis. As
hexane isomers have different reactivities, selectivities have been calculated at 650 K in order to do comparisons.
The selectivity of a species is obtained by divided its mole fraction by the sum of mole fractions of all reaction
products. Selectivity graphs are displayed in Figure 13 for olefins and carbonyl species. As far as olefins are
concerned, it can be seen that the distribution clearly depends on the structure of the fuel. As an example, the
experimental propene mole fractions obtained with n-hexane and 2-methyl-pentane are larger than those obtained
with 3-methyl-pentane and 2,2-dimethyl-butane. These differences can be simply explained by the structure of the
fuels (Figure 1): propene cannot be directly formed by β-scission of fuel radicals obtained from 2,2-dimethyl-butane
and 3-methyl-pentane. Also the selectivity of iso-butene is very large in the case of 2,2-dimethyl-butane due to the
presence of the tert-butyl structure in the reactant. Note that selectivities of 2-methyl-1-butene and 2-methyl-2-
butene are also important in the case of 2,2-dimethyl-butane. 3-methyl-pentane rather leads to 1- and 2-butenes
and 2-methyl-1-butene due to the methyl-group located in the middle of the chains. The selectivities of propene and
iso-butene are large in the case of 2-methyl-pentane due to the presence of the iso-butyl structure.
12
ACS Paragon Plus Environment

Page 13 of 44
The Journal of Physical Chemistry
14x10-3
Ethylene
8x10-3
Methane
1
2
3
4
5
6
7
8
12
10
8
6
4
2
0
n-hexane
3-methyl-pentane
2-methyl-pentane
2,2-dimethyl-butane
2,3-dimethyl-butane
6
4
2
0
9
400
600
800
1000
1000
1000
1000
400
600
800
1000
1000
1000
1000
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
Temperature (K)
Temperature (K)
350x10-6
300
250
200
150
100
50
120x10-6
Acetylene
Ethane
100
80
60
40
20
0
0
400
600
800
400
600
800
Temperature (K)
Temperature (K)
6x10-3
25x10-6
Propene
Propane
5
4
3
2
1
0
20
15
10
5
0
400
600
800
400
600
800
Temperature (K)
Temperature (K)
35x10-6
30
40x10-6
Propyne
Allene
30
20
10
0
25
20
15
10
5
0
400
600
800
400
600
800
Temperature (K)
Temperature (K)
Figure 8: Comparison of mole fraction profiles of the C1-C3 hydrocarbons that were observed in the oxidation of the
five isomers of hexane (fuel inlet mole fraction of 0.02). Mole fractions were obtained using gas chromatography.
49
50
51
52
53
54
55
56
57
58
59
60
13
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 14 of 44
1.4x10-3
300x10-6
250
200
150
100
50
1-Butene
1,3-Butadiene
n-hexane
1
2
3
4
5
6
7
8
3-methyl-pentane
2-methyl-pentane
2,2-dimethyl-butane
2,3-dimethyl-butane
1.2
1.0
0.8
0.6
0.4
0.2
0.0
0
9
400
600
800
1000
400
600
800
1000
1000
1000
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Temperature (K)
Temperature (K)
2.5x10-3
2.0
600x10-6
500
400
300
200
100
0
iso-Butene
2-Butene (Cis or Trans)
1.5
1.0
0.5
0.0
400
600
800
1000
400
600
800
Temperature (K)
Temperature (K)
1.0x10-3
0.8
600x10-6
500
400
300
200
100
0
2-Butene (Cis or Trans)
1-Pentene (n-hexane)
0.6
0.4
0.2
0.0
400
600
800
1000
400
600
800
Temperature (K)
Temperature (K)
600x10-6
500
400
300
200
100
0
2-Methyl-1-Butene
400
600
800
1000
Temperature (K)
Figure 9: Comparison of mole fraction profiles of the C4-C5 hydrocarbons that were observed in the oxidation of the
five isomers of hexane (fuel inlet mole fraction of 0.02). Mole fractions were obtained using gas chromatography.
Note that two isomers were detected for 2-butene but we were unable to distinguish the cis and the trans
configurations. Note also that the formation of 1-pentene was only observed in the oxidation of n-hexane whereas
the formation of 2-methyl-1-butene was observed only for the branched hexane isomers.
14
ACS Paragon Plus Environment

Page 15 of 44
The Journal of Physical Chemistry
70x10-3
60
70x10-3
CO2
CO
1
2
3
4
5
6
7
8
60
50
40
30
20
10
0
n-hexane
2-methyl-pentane
3-methyl-pentane
2,2-dimethyl-butane
2,3-dimethyl-butane
50
40
30
20
10
0
9
400
600
800
1000
400
600
800
1000
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
Temperature (K)
Temperature (K)
50x10-3
40
8x10-3
Water
Formaldehyde
(SVUV PIMS data obtained
at 11 eV for m/z 30)
(SVUV PIMS data obtained
at 16.6 eV for m/z 18)
6
4
2
0
30
20
10
0
500 550 600 650 700 750 800
Temperature (K)
500
550
600 650
Temperature (K)
700
750
800
7x10-3
600x10-6
500
400
300
200
100
0
Ethylene Oxide
Methanol
6
5
4
3
2
1
0
O
400
600
800
1000
400
600
800
1000
Temperature (K)
Temperature (K)
8x10-3
1.6x10-3
1.2
Acetaldehyde
Acetic Acid
O
(SVUV-PIMS data obtained
at 11 eV for m/z 60)
O
6
4
2
0
OH
0.8
0.4
0.0
400
600
800
1000
500 550 600 650 700 750 800
Temperature (K)
Temperature (K)
Figure 10: Comparison of mole fraction profiles of the C1-C2 oxygenated reaction products that were observed in the
oxidation of the five isomers of hexane (fuel inlet mole fraction of 0.02). Mole fractions were obtained using gas
chromatography except for water, formaldehyde and acetic acid.
47
48
49
50
51
52
53
54
55
56
57
58
59
60
15
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 16 of 44
500x10-6
400
700x10-6
Acrolein
Propanoic Acid
O
1
2
3
4
5
6
7
8
(SVUV-PIMS data obtained
at 11 eV for m/z 74)
O
600
500
400
300
200
100
0
OH
300
200
100
0
9
500 550 600 650 700 750 800
Temperature (K)
400
600
800
1000
1000
1000
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Temperature (K)
300x10-6
250
200
150
100
50
1.4x10-3
1.2
Propanal
2-Methyl-Oxirane
O
O
1.0
0.8
0.6
0.4
0.2
0
0.0
400
600
800
1000
400
600
800
Temperature (K)
Temperature (K)
4x10-3
400x10-6
300
Acetone
n-hexane
Methacrolein
2-methyl-pentane
3-methyl-pentane
2,2-dimethyl-butane
2,3-dimethyl-butane
O
3
2
1
0
O
200
100
0
400
600
800
1000
400
600
800
Temperature (K)
Temperature (K)
800x10-6
600
160x10-6
Methyl-Vinyl-Ketone
Ketene
O
H2C
C
120
80
40
0
O
400
200
0
400
600
800
1000
500 550 600 650 700 750 800
Temperature (K)
Temperature (K)
Figure 11: Comparison of mole fraction profiles of ketene and the C3-C4 oxygenated reaction products that were
observed in the oxidation of the five isomers of hexane (fuel inlet mole fraction of 0.02). Mole fractions were
obtained using gas chromatography.
16
ACS Paragon Plus Environment

Page 17 of 44
The Journal of Physical Chemistry
600x10-6
400
250x10-6
Large linear ketones (n-hexane)
Butanal
1
2
3
4
5
6
n-hexane
200
150
100
50
3-methyl-pentane
2-hexanone
2-pentanone
200
7
8
9
0
0
400 500 600 700 800 900
Temperature (K)
400 500 600 700 800 900
Temperature (K)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4x10-3
n-hexane
2-methyl-pentane
3-methyl-pentane
Butanone
3
2
1
0
400
500
600
700
800
900
Temperature (K)
Figure 12: Mole fractions of ketones and aldehydes obtained in the oxidation of hexane isomers (fuel inlet mole
fraction of 0.02). Mole fractions were obtained using gas chromatography.
17
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 18 of 44
1
2
3
4
2-methyl-2-butene
2-methyl-1-butene
1-pentene
n-hexane
2-methyl-pentane
3-methyl-pentane
2,2-dimethyl-butane
5
iso-butene
2-butene
6
7
8
9
1-butene
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
propene
ethylene
0
5
10
15
20x10-3
Selectivity
a)
2-hexanone
2-pentanone
pentanal
n-hexane
2-methyl-pentane
3-methyl-pentane
2,2-dimethyl-butane
methacrolein
2-methyl-propanal
methyl-vinyl-ketone
2-butanone
butanal
acetone
acrolein
propanal
acetaldehyde
formaldehyde
0.00
0.02
0.04
0.06
0.08
0.10
Selectivity
b)
Figure 13: Comparison of the selectivities of olefins and carbonyl species at 650 K.
As far as carbonyl compounds are concerned, the formation of acetone and 2-methyl-propanal is enhanced
in the case of 2,2-dimethyl-butane. Methacrolein is likely a product coming from the oxidation of iso-butene through
secondary reactions and the larger selectivity of this species could be related to the large formation of iso-butene.
Note that the formation of acetic and propanoic acids, species the formation of which is rarely reported in
oxidation studies likely because of the low sensitivity of flame ionization detectors for this type of species, were
observed in the oxidation of n-hexane and the two methyl-pentane isomers (Figures 10 and 11). Note that no acid
formation was observed for 2,2-dimethyl-butane. Significant amounts of ketene were detected with n-hexane and
3-methyl-pentane (Figure 11). Mole fractions (with a maximum around 10-3) were deduced from the signal recorded
18
ACS Paragon Plus Environment

Page 19 of 44
The Journal of Physical Chemistry
for m/z 42 at 10.5 eV by subtracting the contribution of propene (deduced from gas chromatography data). The
cross section used for the quantification is 24.83 Mb at 10.5 eV47.
1
2
3
4
5
6
7
8
9
It can also be seen that mole fraction profiles of some small hydrocarbons, such as methane, acetylene, and
ethane, are very similar for all isomers over the common whole range of temperature (Figure 8). Note that allene
and propyne, two unsaturated species which are known to play a role in the formation of soot, are formed in larger
amounts with the branched alkanes (Figure 8) above 800 K.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3.3. Evolution of the mole fractions of cyclic ethers and large alkenes
Cyclic ethers are species which are formed in relatively large quantities during the low-temperature
oxidation of alkanes, whereas their formation is more difficult from alkenes27. The ways of formation of these cyclic
oxygenated species are now well established22,23: they are produced from the reactions of addition of alkyl radicals
to O2 followed by isomerization and decomposition reactions. The numerous possibilities of isomerizations and
decompositions (into three, four, five or six membered ring cyclic ethers) are the cause for the numerous possible
isomers of cyclic ethers formed during the low-temperature oxidation of alkanes. As an example, the formation of
eight C6H12O cyclic ethers is expected for n-hexane, with some of them present twice because of the cis/trans
conformation of the side alkyl groups (Table 2).As an example, the sequence of reactions forming 2,5-dimethyl-
oxolane (also named 2,5-dimethyl-tetrahydrofuran) from n-hexane is displayed in Figure 14. Reaction (1) is an
H-atom abstraction forming an alkyl radical (2-hexyl radical in the example of Figure 14). This radical adds to O2 to
form a 2-peroxyhexyl radical (reaction (2)) which reacts by isomerization (3) to form an hydroperoxyhexyl radical
(the 2-hydroperoxy-hex-5-yl radical in the example of Figure 14). This last radical decomposes to yield a cyclic ether
and an OH radical.
19
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 20 of 44
Table 2: List of three to six membered ring cyclic ethers which can be formed in the low-temperature oxidation of
1
2
n-hexane.
3
4
5
Number of possible
isomers
Species detected in
this study
Name
Structure
6
7
O
2-butyl-oxirane
1
2
yes
8
9
O
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
only one of the two
isomers
2-propyl,3-methyl-oxirane
2,3-diethyl-oxirane
O
only one of the two
isomers
2
1
2
1
2-propyl-oxetane
yes
O
O
the two isomers
were detected
2-ethyl,4-methyl-oxetane
O
O
2-ethyl-oxolane (also named 2-
ethyl-tetrahydrofuran)
yes
2,5-dimethyl-oxolane (also
named 2,5-dimethyl-
tetrahydrofuran)
the two isomers
were detected
2
1
O
2-methyl-oxane (also named 2-
methyl-tetrahydropyran)
yes
20
ACS Paragon Plus Environment

Page 21 of 44
The Journal of Physical Chemistry
1
2
(1)
+ R / - RH
3
4
5
6
(2) + O2
7
8
O
O
9
(3)
O
10
11
12
13
14
15
16
17
18
(4)
O
HO
+ HO
Figure 14: Route of formation of the cyclic ether 2,5-dimethyl-tetrahydrofuran from n-hexane ((1): H-atom
abstraction from the reactant; (2): addition to oxygen; (3): isomerization; (4): decomposition into cyclic ether and the
OH radical).
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Due to the relatively high concentration and reactivity of this fuel, almost all expected isomers were
identified (Table 2 and Figure 15). Four and five membered ring cyclic ethers (also called oxetanes and oxolanes,
respectively) are formed in larger quantities than three and six membered ring cyclic ethers (also called oxiranes and
oxanes, respectively). Note that the formation of three membered ring cyclic ethers is rarely reported in JSR
oxidation studies, likely due to too low inlet fuel mole fractions, e.g.48. The identification of cyclic ethers was
relatively easy because most of their electron impact (70 eV) mass spectra are available in the NIST08 database.
When mass spectra were not available in the NIST08 database, experimental ones were deciphered by looking for
characteristic fragments49 (see Figures S10 to S17).
21
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 22 of 44
800x10-6
600
800x10-6
2,5-dimethyl-oxolane (2 isomers)
2-ethyl-oxolane
2-methyl-oxane
2,4-methyl,ethyl-oxetane (2 isomers)
2-propyl-oxetane
1
2
3
4
5
6
7
8
600
400
200
0
400
200
0
9
400
500
600
700
800
900
400
500
600
700
800
900
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
a)
b)
Temperature (K)
200x10-6
Temperature (K)
2,3-diethyl-oxirane
2,3-methyl,propyl-oxirane
2-butyl-oxirane
150
100
50
0
400
500
600
700
800
900
c)
Temperature (K)
Figure 15: Mole fraction profiles of cyclic ethers obtained in the low-temperature oxidation of n-hexane (fuel inlet
mole fraction of 0.02). a) five and six membered ring cyclic ethers. b) four membered ring cyclic ethers. c) three
membered ring cyclic ethers. Mole fractions were obtained using gas chromatography.
Expected cyclic ethers for 2-methyl-pentane, 3-methyl-pentane, 2,2-dimethyl-butane and 2,3-dimethyl-
butane are shown in Tables 3 to 6, respectively. Mole fraction profiles are displayed in Figures 16 and 17. With n-
hexane, 2,2dimethyl-butane and 2,3-dimethyl-butane as fuels, all expected cyclic ethers were detected. However,
that was not the case for 2-methyl-pentane and 3-methyl-pentane. As an example, all the expected cyclic ethers,
except all the oxiranes and one oxetane, were detected with 3-methyl-pentane. There is no obvious reason which
explains the missing cyclic ethers in the case of 2-methyl-pentane and 3-methyl-pentane. This could be due to the
presence of competitive consumption routes such as the decomposition to a HO2 radical and an olefin which is
competitive with the decomposition into an oxirane and an OH radical. This competition would be favored in methyl
substituted reactants. However, no obvious evidence of this competition can be deduced from the mole fraction
profiles of C6 olefins shown in Figure 18.
22
ACS Paragon Plus Environment

Page 23 of 44
The Journal of Physical Chemistry
Table 3: List of three to six membered ring cyclic ethers which can be formed during the low-temperature oxidation
of 2-methyl-pentane.
1
2
3
4
5
Number of possible
isomers
Species detected in
this study
Name
Structure
6
7
8
9
O
2,2-methyl,propyl-oxirane
1
1
no
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
O
2,2-dimethyl,3-ethyl-oxirane
2-methyl,3-isopropyl-oxirane
yes
O
2
no
O
2-isobutyl-oxirane
3-propyl-oxetane
1
1
no
yes
O
2-ethyl,3-methyl-oxetane
2
no
O
2,2,4-trimethyl-oxetane
2-isopropyl-oxetane
1
1
yes
yes
O
O
2,2-dimethyl-oxolane (also
named 2,2-dimethyl-
tetrahydrofuran)
O
1
2
1
yes
O
2,4-dimethyl-oxolane (also
named 2,4-dimethyl-
tetrahydrofuran)
the two isomers
were detected
O
3-methyl-oxane (also named 3-
methyl-tetrahydropyran)
no
23
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 24 of 44
Table 4: List of three to six membered ring cyclic ethers which can be formed during the low-temperature oxidation
of 3-methyl-pentane.
1
2
3
4
5
Number of possible
isomers
Species detected in
this study
Name
Structure
6
7
8
O
(2-methyl),2-propyl-oxirane
1
2
no
no
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
O
2,2-methyl,ethyl,3-methyl-
oxirane
O
2,2-diethyl-oxirane
1
no
2,3-methyl,ethyl-oxetane
2
1
no
O
2,2-methyl,ethyl-oxetane
2,3,4-trimethyl-oxetane
yes
O
the three isomers
were detected
3
1
2
1
O
3-ethyl-oxolane (also named 3-
ethyl-tetrahydrofuran)
O
yes
O
2,3-dimethyl-oxolane (also
named 2,3-dimethyl-
tetrahydrofuran)
the two isomers
were detected
4-methyl-oxane (also named 4-
methyl-tetrahydropyran)
O
yes
24
ACS Paragon Plus Environment

Page 25 of 44
The Journal of Physical Chemistry
Table 5: List of three to five membered ring cyclic ethers which can be formed during the low-temperature oxidation
of 2,2-dimethyl-butane.
1
2
3
4
5
Number of possible
isomers
Species detected in
this study
Name
Structure
6
7
8
O
2-tertbutyl-oxirane
1
1
yes
yes
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3,3-methyl-ethyl-oxetane
2,3,3-trimethyl-oxetane
O
1
1
yes
yes
O
3,3-dimethyl-oxolane (also
named 3,3-dimethyl-
tetrahydrofuran)
O
Table 6: List of three to five membered ring cyclic ethers which can be formed during the low-temperature oxidation
of 2,3-dimethyl-butane.
Number of possible
isomers
Species detected in
this study
Name
Structure
2,2-methyl-isopropyl-oxirane
1
1
1
yes
yes
yes
O
2,2,3,3-tetramethyl-oxirane
2,2,3-trimethyl-oxetane
O
O
3-isopropyl-oxetane
1
2
yes
O
3,4-dimethyl-oxolane (also
named) 3,4-dimethyl-
tetrahydrofuran
O
the two isomers were
detected
25
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 26 of 44
250x10-6
200
150
100
50
600x10-6
Five and six membered ring cyclic
ethers from 3-Methyl-Pentane
2,3-dimethyl-oxolane (2 isomers)
3-ethyl-oxolane (X2)
4-methyl-oxane (X5)
Four membered ring cyclic ethers
from 3-Methyl-Pentane
1
2
3
4
5
6
7
8
500
400
300
200
100
0
2,2-methyl,ethyl-oxetane (X5)
2,3,4-trimethyl-oxetane (3 isomers)
0
9
500
600
700
800
900
500
600
700
800
900
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Temperature (K)
Temperature (K)
1.0x10-3
0.8
1.0x10-3
0.8
Four and three membered ring cyclic
Five membered ring cyclic
ethers from 2-Methyl-Pentane
ethers from 2-Methyl-Pentane
3-propyl-oxetane
2,2-dimethyl-oxolane
2-isopropyl-oxetane
2,2,4-trimethyl-oxetane
2,2-dimethyl,3-ethyl-oxirane
2,4-dimethyl-oxolane
0.6
0.6
0.4
0.4
0.2
0.2
0.0
0.0
500
600
700
800
900
500
600
700
800
900
Temperature (K)
Temperature (K)
Figure 16: Mole fraction profiles of cyclic ethers obtained in the low-temperature oxidation of 2-methyl-pentane and
3-methyl-pentane (fuel inlet mole fraction of 0.02). Mole fractions were obtained using gas chromatography. Mole
fractions of species formed in small amounts were multiplied by a factor
α (×α) to better see their profiles.
1.0x10-3
0.8
200x10-6
150
100
50
Cyclic Ethers from 2,2-Dimethyl-Butane
Cyclic ethers from 2,3-Dimethyl-Butane
2,2,3,3-tetramethyl-oxirane
2,2-methyl,isopropyl-oxirane (X2)
2,2,3-trimethyl-oxetane
2,3,3-Trimethyl-Oxetane
3,3-Dimethyl-Tetrahydrofuran
3,3-Methyl-Ethyl-Oxetane (X10)
2-Tertbutyl-Oxirane (X10)
3-isopropyl-oxetane
3,4-dimethyl-oxolane (2 isomers)
0.6
0.4
0.2
0.0
0
500
700
900
1100
500
700
900
1100
Temperature (K)
Temperature (K)
Figure 17: Mole fraction profiles of cyclic ethers obtained in the low-temperature oxidation of 2,2-dimethyl-butane
and 2,3-dimethyl-butane (fuel inlet mole fraction of 0.02). Mole fractions were obtained using gas chromatography.
Mole fractions of species formed in small amounts were multiplied by a factor
α (×α) to better see their profiles.
26
ACS Paragon Plus Environment

Page 27 of 44
The Journal of Physical Chemistry
600x10-6
500
400
300
200
100
0
600x10-6
1
2
3
4
5
6
7
8
hexene isomers detected in
the n-hexane oxidation
1-hexene
hexene isomers detected in
the 2-methyl-pentane oxidation
500
400
300
200
100
0
4-methyl-1-pentene & 2-methyl-1-pentene
2-methyl-2-pentene & 4-methyl-2-pentene
2-hexene
3-hexene
9
500
600
700
800
900
500
600
700
800
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
Temperature (K)
Temperature (K)
600x10-6
500
400
300
200
100
0
1.2x10-3
1.0
hexene isomers detected in
3,3-dimethyl-1-butene detected in
the 2,2-dimethyl-butane oxidation
the 3-methyl-pentane oxidation
2-ethyl-1-butene
3-methyl-1-pentene
0.8
3-methyl-2-pentene isomers
0.6
0.4
0.2
0.0
500
600
700
800
500
600
700
800
900
Temperature (K)
160x10-6
Temperature (K)
hexene isomers detected in
the 2,3-dimethyl-butane oxidation
2,3-dimethyl-1-butene
120
80
40
0
2,3-dimethyl-2-butene
500
600
700
800
900
Temperature (K)
Figure 18: Mole fractions of C6H12 isomers detected during the oxidation of hexane isomers (gas chromatography
data). Data were obtained for an initial fuel mole fraction of 0.02.
Figure 19a displays the evolution between 500 and 750 K of the signals recorded using SVUV-PIMS for m/z
100 which corresponds to C6H12O species (cyclic ethers, hexanal and hexanones). This signal is the sum of the
contributions of each C6H12O isomers and the extraction of mole fractions was not carried out due to their similar
ionization energies and a lack of data for cross sections for these species. Signal intensities are very similar for
n-hexane and methyl-pentane isomers, but much lower for 2,2-dimethylbutane. Figure 19b presents the sum of
C6H12O species obtained by GC analyses. This confirms that despite the much larger reactivity for n-hexane than for
other fuels (as shown in Figure 7), the sum of the mole fractions of C6H12O species (which are mostly cyclic ethers) at
650 K is the highest for 2-methyl-pentane and similar for n-hexane, 3-methyl-pentane, and 2,2-dimethyl-butane. This
clearly shows that the formation of cyclic ethers is significantly favored from branched alkanes below 650 K. This is in
agreement with theoretical work of Wijaya et al.50. These calculations showed that methyl substitution reduces the
barrier height for the formation of oxetanes and oxiranes, while with little effect on oxolanes, because of the release
of alkyl strain in the cyclic transition state. In the NTC area, the cyclic ether formation from n-hexane is the largest
because it is the only fuel isomer with a significant reactivity in this zone.
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
27
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 28 of 44
400
300
200
100
0
m/z 100 (10.5 eV)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
500
550
600
650
700
750
Temperature (K)
(a)
4x10-3
n-hexane
2-methyl-pentane
3-methyl-pentane
2,2-dimethyl-butane
2,3-dimethyl-butane
3
2
1
0
500 550 600 650 700 750
Temperature (K)
(b)
Figure 19: (a) Signals at m/z 100 (corresponding to C6H12O species (cyclic ethers, hexanal and hexanones)) obtained
using SVUV-PIMS and b) sum of the C6H12O species obtained by GC (fuel inlet mole fraction of 0.02).
The formation of cyclic ethers has already been reported by Dagaut et al. during the oxidation of n-heptane,
iso-octane and n-decane51–53. All tetrahydrofuran isomers were identified in these studies (cis and trans 2-methyl,5-
ethyl-tetrahydrofuran, 2-propyl-tetrahydrofuran and 2,2,4,4-tetramethyl-tetrahydrofuran in the n-heptane and iso-
octane studies51,52; cis and trans 2,5-dipropyl-tetrahydrofuran, cis and trans 2-ethyl,5-butyl-tetrahydrofuran, cis and
trans 2-methyl,5-pentyl-tetrahydrofuran and 2-hexyl-tetrahydrofuran in the n-decane study53). While not reporting
all the possible isomers, Dagaut et al. also observed the formation of some oxetanes and oxiranes during the
oxidation of n-heptane and iso-octane51,52 (cis and trans 2-methyl,4-propyl-oxetane, 2-ethyl,3-propyl-oxirane, cis 2-
methyl-3-butyl-oxirane, 2-iso-propyl,3,3-dimethyl-oxetane, 2-terbutyl,3-methyl-oxetane, 2,4-diethyl-oxetane and 2-
methyl,3-butyl-oxirane). In the studies about the oxidation of some C8H18 isomers, Sarathy et al.19,21 and Karsenty et
al.20 observed the formation of only a few five membered ring cyclic ethers (oxolanes) whereas all possible isomers
were detected in the present hexane isomer study. As an example, the only five membered ring cyclic ether
20
detected during the oxidation of 3-methyl-heptane by was 2,5-dimethyl,2-ethyl-tetrahydrofuran, whereas the
formation of five other isomers could be expected. According to the flow rate analysis presented in 20, non negligible
amount of 2-butyl-2-methyl-oxetane would be formed during this reaction but the detection of this species was not
reported during those experiments. Note that these studies were performed with lower initial fuel mole fractions
28
ACS Paragon Plus Environment

Page 29 of 44
The Journal of Physical Chemistry
than in the present work yielding lower reaction product concentrations and inducing then difficulties to detect
them.
1
2
3
4
5
6
7
8
9
Alkenes of the same size as the reactant are also important products in the low-temperature oxidation of
hydrocarbons. They can be formed by H-abstraction of O2 from alkyl radicals produced from the reactant or by HO2
elimination from RO2 radicals. The structures of the C6 alkenes detected during this study are listed in Table 7 and
the mole fraction profiles are given in Figure 18 for the five hexane isomers. Note that the formation of 2-hexene is
favored from n-hexane since it can be obtained from alkyl radicals formed by the abstraction of four secondary
H-atoms. All the hexene isomers could not be separated in the GC analysis for 2-methyl-pentane. 3,3-dimethyl-1-
butene is formed in very large amounts since it is the only alkene that can be formed from 2,2-dimethyl-butane.
Since 2,3-dimethyl-butane is unreactive at lower temperature, alkene formation is only observed from 750 K. The
production of alkenes from alkyl radicals obtained by the abstraction of the two tertiary H-atoms is much favored.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
29
ACS Paragon Plus Environment