A selected ion flow tube study of the reactions of small CmHn+ ions with O atoms

Article abstract of DOI:10.1063/1.481050

The reactions of thirteen C_mH_n⁺ ions (m ≤ 6) with atomic and molecular oxygen and nitric oxide have been measured using a selected ion flow tube operating at room temperature. Rate coefficients and product distributions are reported for each reaction. Most of the hydrocarbon ions studied exhibit relatively rapid reactions with O atoms and proceed at substantial fractions of the collision rate. All O atom reactions showed multiple product channels and the formation of a -C-O bond, either in the ion product or the neutral product of reaction.

Full text of DOI:10.1063/1.481050

A selected ion flow tube study of the reactions of small CmHn+ ions with O
atoms
Graham B. I. Scott, Daniel B. Milligan, David A. Fairley, Colin G. Freeman, and Murray J. McEwan
Citation: J. Chem. Phys. 112, 4959 (2000); doi: 10.1063/1.481050
View online: http://dx.doi.org/10.1063/1.481050
View Table of Contents: http://jcp.aip.org/resource/1/JCPSA6/v112/i11
Published by the American Institute of Physics.
Additional information on J. Chem. Phys.
Journal Homepage: http://jcp.aip.org/
Journal Information: http://jcp.aip.org/about/about_the_journal
Top downloads: http://jcp.aip.org/features/most_downloaded
Information for Authors: http://jcp.aip.org/authors
Downloaded 07 May 2013 to 193.1.100.108. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://jcp.aip.org/about/rights_and_permissions

JOURNAL OF CHEMICAL PHYSICS
VOLUME 112, NUMBER 11
15 MARCH 2000
A selected ion flow tube study of the reactions of small CmH¿n ions
with O atoms
Graham B. I. Scott, Daniel B. Milligan, David A. Fairley, Colin G. Freeman,
and Murray J. McEwana)
Department of Chemistry, University of Canterbury, Christchurch, New Zealand
͑Received 21 October 1999; accepted 23 December 1999͒
The reactions of thirteen CmHϩn ions (mр6) with atomic and molecular oxygen and nitric oxide
have been measured using a selected ion flow tube operating at room temperature. Rate coefficients
and product distributions are reported for each reaction. Most of the hydrocarbon ions studied
exhibit relatively rapid reactions with O atoms and proceed at substantial fractions of the collision
rate. All O atom reactions showed multiple product channels and the formation of a –C–O bond,
either in the ion product or the neutral product of reaction. © 2000 American Institute of Physics.
͓S0021-9606͑00͒00811-4͔
association ion, C10H8Oϩ, in the other. Finally, we reported
the results of a study of the first nitrile cation reactions with
O atoms together with the reactions of H2Cϩ, HCOϩ, and
HCOϩ2 .7
In the present study, we investigate the reactions of thir-
teen hydrocarbon ions with O atoms and discuss some of the
implications of these measurements to interstellar cloud
chemistry. In most reactions in this study, O atoms were
generated via reaction ͑2͒ in which NO and traces of O2 were
also present. In order to obtain accurate product distributions
from the O atom reactions, it was necessary to examine sepa-
rately the reactions of each ion with O2 and NO. The rate
coefficients and product ratios for these reagents are reported
as well.
I. INTRODUCTION
Reactions of ions with atomic oxygen pose a number of
difficulties for experimental investigators. Part of the diffi-
culty is linked to the generation of O atoms and part with
monitoring the atoms produced. Despite more than 30 years
having elapsed since the first ion–O atom reaction was
studied,1 there have not been many additional reports of
ion–O atom investigations. The impetus for the first study,
that of Nϩ2 ϩO, came from aeronomy and the role reaction
͑1͒ played in the E and F1 regions of the earth’s ionosphere,
Nϩ2 ϩO→NOϩϩN,
kϭ 2.5Ϯ1.0͒ϫ10Ϫ10 cm3 sϪ1
͑1͒
.
͑
II. EXPERIMENT
The O atoms in this study were generated from N atoms
formed in a microwave discharge of N2. The N atoms so
formed ͑ϳ1% of the N2 flow͒ were titrated with measured
flows of nitric oxide, thereby generating O(3P) atoms, viz.
Most of the reactions reported here were investigated
using our original selected ion flow tube ͑SIFT͒ apparatus
operating at room temperature (295Ϯ5) K. This apparatus
has been described elsewhere8 and will not be discussed
here. The NO titration procedure used to generate O atoms
from N atoms via reaction ͑2͒ has also been previously
described.4,7,9
A diagram of the neutral probe in which the titration
took place is shown as Fig. 1 in an earlier study.10 A small
flux of atomic nitrogen was generated from a microwave
discharge of pure nitrogen.7,10 Measured flow rates of nitric
oxide were admitted through the NO inlet of the probe
downstream from the discharge region. This probe was de-
signed to allow sufficient flow time for reaction ͑2͒ to pro-
ceed to completion before the gas stream entered the SIFT
reaction tube.
The end point of the titration reaction is readily apparent
as the intersection between the two linear decays in a semi-
logarithmic plot shown in Fig. 1 for the reaction between O
atoms and c-C6Hϩ6 and in Fig. 2 for the reaction between
ac-C3Hϩ3 and O. The form of these plots is dependent on the
relative rates of reaction for the ion of interest with N atoms,
O atoms, and NO. Thus in Fig. 1, N atoms and O atoms react
3
NϩNO→N ϩO P͒.
͑2͒
͑
2
Several ion–O atom reactions were investigated with a
motivation other than aeronomy, in that the potential appli-
cation of these systems to the synthetic chemistry occurring
in interstellar clouds was explored. Fehsenfeld2 examined the
reactions of Hϩ3 and CH3ϩ , Bohme, Mackay, and Schiff3 ex-
amined the reactions of N2Hϩ and CH5ϩ and Viggiano et al.4
reported the results of O atom reactions with CHϩ, CH5ϩ ,
Cϩ2 , C2Hϩ, C2Hϩ2 , Dϩ, and H2Oϩ. Eight years later in
1988, Viggiano et al.5 investigated reactions of the isoto-
pomers of CHϩ3 , ͑i.e., CDnH3ϩϪn), with O atoms. More re-
cently, Le Page et al.6 reported their investigation of the pro-
totypical polycyclic aromatic hydrocarbon ions C10Hϩ6 ,
C10Hϩ7 , and C10H8ϩ with H, N, and O atoms. Of these ions,
only the C10Hϩ8 species was found to react with atomic oxy-
gen, forming C9Hϩ8 ϩCO in one reaction channel and the
a͒
Electronic mail: m.mcewan@chem.canterbury.ac.nz
0021-9606/2000/112(11)/4959/7/$17.00
4959
© 2000 American Institute of Physics
Downloaded 07 May 2013 to 193.1.100.108. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://jcp.aip.org/about/rights_and_permissions

4960
J. Chem. Phys., Vol. 112, No. 11, 15 March 2000
Scott et al.
FIG. 1. Data obtained for the reaction
between c-C6Hϩ6 with N, O, and NO.
The end point of the NO titration at
the intersection of the two lines corre-
sponds to the complete conversion of
N atoms to O atoms. The point marked
by a triangle represents the experimen-
tal ion count without N atoms present.
The difference between this point and
that denoted by an open circle at zero
NO flow is the result of the reaction
between c-C6Hϩ6 and N atoms.
at similar rates with c-C6Hϩ6 giving an almost horizontal line
prior to the ‘‘end point’’ as N atoms are converted to O
atoms by NO. On the other hand in Fig. 2, N atoms do not
react with ac-C3Hϩ3 while O atoms do, with the conse-
quence that the plot prior to the ‘‘end point’’ resembles a
standard semilogarithmic decay. At the point of intersection
atom flux is zero, the O atom flux is unchanged from FNO
,
and the rate of decay is then a consequence of the CmHnϩ
ϩNO reaction.
Generally, identification of the products of the ion ϩN
atoms and ion ϩO atom reactions was not confusing as the
N atom products were identified in the absence of O atoms.
Further, the product ions themselves were usually of a mass
that made ready unambiguous identification possible. Rate
coefficients and product distribution data for the reactions of
these thirteen cations with atomic nitrogen have been re-
ported in an earlier publication.11
between the two lines in Figs. 1 and 2, the flow of NO, FNO
,
equals the original flux of N atoms ͑i.e., the atomic nitrogen
flow before any NO was introduced͒. Precisely at this same
intersection point, all the N atoms have been converted to O
atoms. Thus at this end point, FNOϭFO , that is the flow of O
atoms and NO are equivalent, and by similar reasoning for
any flow of NO less than FNO , the NO flow rate equals the
O atom flow rate. This statement is only approximately cor-
rect however as some recombination of O atoms occurs be-
fore entry into the SIFT reaction tube and as a result the rate
coefficients for O atom reactions reported here may be low
by as much as 20%. At NO flow rates larger than FNO , the N
III. RESULTS
Rate coefficients and product distributions for the reac-
tions studied here are summarized in Tables I–III. The reac-
tions of CmHϩn with atomic oxygen are listed in Table I, with
FIG. 2. Data obtained for the reaction
between ac,c-C3Hϩ3 with N, O, and
NO. The end point of the NO titration
of the intersection point corresponds
to the complete conversion of N atoms
to O atoms. The slope of the left line
corresponds to the reaction of
ac-C3Hϩ3 with O atoms and the slope
of the right-hand line corresponds to a
nonreaction with NO.
Downloaded 07 May 2013 to 193.1.100.108. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://jcp.aip.org/about/rights_and_permissions

4961
J. Chem. Phys., Vol. 112, No. 11, 15 March 2000
TABLE I. Reactions of the given CmHϩn ion with O.
Reactant ion Products Branching ratio
Reactions of CmHϩn ions with O atoms
ka
b
c
d
kprev
kcoll
Ϫ⌬H°/(kJ molϪ1
)
CHϩ3
HCOϩϩH2
517
380
156g
HOCϩϩH2
4.1
2.0
4.4e
1.7f
7.5
6.6
C2Hϩ2
HCOϩ ͑or HOCϩ͒ϩCH
0.5
HC2OϩϩH
H2CCOϩϩH
C2H3Oϩ
0.5
0.85
0.10
0.05
0.45
0.35
0.10
ϳ0.05
ϳ0.05
263
264h
521i
379
344g
177
1
C2Hϩ3
C2Hϩ4
1.0
6.6
CHϩ3 ϩCO
CHϩ3 ϩHCO
HCOϩϩCH3
HC2OϩϩH2ϩH
CH3COϩϩH
H2CCOϩϩH2
No reaction
C3H2OϩϩH
C2Hϩ3 ϩCO
2.4
Ͻ0.25
1.5
6.6
6.5
6.2
256j
436h
C2Hϩ5
ac-C3Hϩ3
0.30
0.30
236j
427k
C2Hϩ2 ϩHCO
HC3OϩϩH2
No reaction
No reaction
C4HOϩϩH
C3Hϩ2 ϩCO
0.25
0.15
57k
459k
c-C3Hϩ3
ac-C3Hϩ5
C4Hϩ2
Ͻ0.10
Ͻ0.25
6.2
6.2
l
0.50
0.40
45m
2.7
6.0
C3HOϩϩCH
C4H2Oϩ
ϳ0.05
ϳ0.05
105n
C4Hϩ3
No reaction
C5Hϩ5 ϩCO
Ͻ0.5
1.0
6.0
5.7
5.7
5.7
c-C6Hϩ5
0.6
432o
205p
378q
175r
C3Hϩ3 ϩC3H2O
No reaction
C5Hϩ6 ϩCO
0.4
ac-C6Hϩ5
c-C6Hϩ6
Ͻ0.25
1.4
0.90
0.10
C4H4OϩϩC2H2
aObserved rate coefficient in units of 10Ϫ10 cm3 sϪ1
.
bRate coefficients measured in other laboratories in units of 10Ϫ10 cm3 sϪ1
.
cLangevin collision rate in units of 10Ϫ10 cm3 sϪ1
dReference 12.
.
eReferences 2 and 5.
fReference 4.
gThermochemistry for the HCOϩ cation.
hThermochemistry for the H2CCOϩ cation.
iThermochemistry for the oxiranyl radical cation.
jThermochemistry for H2CwC–C–Oϩ.
kThermochemistry for HwC–CCvCϩ.
lAlthough the structure of this ion was not determined, subsequent experiments in Ref. 13 have characterized it
as having the allyl structure, H2CCHCHϩ2 .
mThermochemistry for the HCwCCHϩ cation.
nThermochemistry for the HCwC–CvOϩ cation.
oThermochemistry for the cyclopentadienyl cation.
pThermochemistry for the cyclopropenyl cation and propadienone, H2OvCvCvO.
qThermochemistry for the cyclopentadiene cation.
rThermochemistry for the furan cation.
IV. DISCUSSION OF RESULTS
A. CH¿3
molecular oxygen in Table II, and with nitric oxide in Table
III. All data were measured at room temperature, 295
Ϯ5 K, and at flow tube pressures between 0.30 and 0.35
Torr of helium. The reaction rate coefficients with O atoms
are considered accurate to Ϯ40% due to the difficulties in-
herent in accurately determining O atom fluxes. Rate coeffi-
cients for stable molecular species are more precisely defined
and are considered accurate to Ϯ15%.
The methyl cation was prepared by electron impact on
methyl bromide and undergoes a rapid reaction with atomic
oxygen. The rate coefficient measured here (kϭ4.1
ϫ10Ϫ10 cm3 sϪ1) is in excellent agreement with the two val-
ues reported previously for this reaction2,5 but the product
Downloaded 07 May 2013 to 193.1.100.108. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://jcp.aip.org/about/rights_and_permissions

4962
J. Chem. Phys., Vol. 112, No. 11, 15 March 2000
Scott et al.
TABLE II. Reactions of the given CmHϩn ion with O2.
b
c
d
Ϫ⌬H°/͑kJ molϪ1
)
Reactant ion
Products
CH3 .Oϩ2
Branching ratio
ka
kprev
kcoll
f,g
CHϩ3
1.0
0.7
0.25
0.05
0.002e
9.2
C2Hϩ2
HCOϩϩHCO
C2H2OϩϩO
C2Hϩ2 •O2
458
45i
567
0.014
Ͻ0.005
0.54
Ͻ0.1h
7.8
7.0
6.2
(c, ac)-C3Hϩ3
c-C6Hϩ5
No reactionj
C4Hϩ5 ϩ2CO
j
0.90
0.10
356k
C5H5OϩϩCO
aObserved rate coefficient in units of 10Ϫ10 cm3 sϪ1
.
bRate coefficients measured in other laboratories in units of 10Ϫ10 cm3 sϪ1
.
cLangevin collision rate in units of 10Ϫ10 cm3 sϪ1
dReference 12.
.
ePseudobimolecular reaction at 0.32 Torr. This value corresponds to a three-body association rate coefficient of
kϳ1.9ϫ10Ϫ29 cm6 sϪ1
.
fReference 14 lists the three-body rate coefficient at 300 K as kϭ5.3ϫ10Ϫ29 cm6 sϪ1
.
.
gReference 15 lists the three-body rate coefficient at 300 K as kϭ7.3ϫ10Ϫ30 cm6 sϪ1
hReference 16.
iThermochemistry based on the HCwCOHϩ cation.
jThe ions C2Hϩ3 , C2H4ϩ ,C2H5ϩ,ac-C3Hϩ5 ,C4Hϩ2 ,C4Hϩ3 ,ac-C6Hϩ5 , and c-C6Hϩ6 also did not exhibit any reaction
with O2 :kϽ5ϫ10Ϫ13 cm3 sϪ1
.
kThermochemistry for the methyl cyclopropenyl cation.
distribution has not previously been identified. Although
three viable reaction channels are possible, we were able to
exclude channel ͑3a͒ from subsequent experiments utilizing
a helium/O2 mixture in the microwave discharge port con-
nected to our new FA/SIFDT apparatus,20
In this experiment we found no evidence for any Hϩ3 produc-
tion. The ratio of HOCϩ/HCOϩ was not measured in this
study, but both products are apparently spin forbidden, a
characteristic that is not uncommon in ion atom reactions.21
The reaction between CHϩ3 and molecular oxygen has
been previously measured by Smith and Adams14,15 and the
current result agrees well with this data.
CHϩ3 1A͒ϩO P͒
3
͑
͑
ϩ
g
yHϩ3 1AЈ͒ϩCO ⌺ ͒ϩ345 kJ molϪ1
͑3a͒
͑3b͒
͑3c͒
1
͑
͑
1
B. C2H¿2
→HCOϩ 1⌺ ͒ϩH 1⌺ϩg ͒ϩ517 kJ molϪ1
͑
͑
g
2
This ion, produced by electron impact on acetylene in a
low pressure ion source, exhibited a moderately fast reaction
→HOCϩ 1⌺ϩ͒ϩH 1⌺ϩg ͒ϩ380 kJ molϪ1
.
͑
͑
2
TABLE III. Reactions of the given CmHϩn ion with NO.
ka
b
c
d
kprev
kcoll
Ϫ⌬H°/(kJ molϪ1
)
Reactant ion
Products
Branching ratio
1.0
1.0
10.0
1.0,
Ͻ0.005
4.7
Ͻ0.005
0.014
Ͻ0.05
3.8
Ͻ0.15
1.9
10.0e, 9.4f, 11.0g 10.2
54
207
CHϩ3
NOϩϩCH3
NOϩϩC2H2
No reaction
NOϩϩC2H4
No reaction
C3Hϩ3 .NO
No reaction
NOϩϩC4H2
No reaction
c-C6Hϩ5 .NO
No reaction
NOϩϩC6H6
c-C6Hϩ6 .NO
C2Hϩ2
1.1h, 1.2e
3.6g
8.6
8.5
8.5
8.4
7.8
7.7
7.4
7.4
6.9
6.9
6.9
C2Hϩ3
C2Hϩ4
1.0
1.0
1.0
1.0
120
129e
90
C2Hϩ5
c, ac-C3Hϩ3
ac-C3Hϩ5
C4Hϩ2
C4Hϩ3
c-C6Hϩ5
ac-C6Hϩ5
c-C6Hϩ6
235i
Ͻ0.005
1.4
0.25
0.75
Ϫ0.4
193j
aObserved rate coefficient in units of 10Ϫ10 cm3 sϪ1
.
bRate coefficients measured in other laboratories in units of 10Ϫ10 cm3 sϪ1
.
cLangevin collision rate in units of 10Ϫ10 cm3 sϪ1
dReference 12.
.
eReference 4.
fReference 17.
gReference 18.
hReference 19 at 15 K.
iThermochemistry based on c-C3Hϩ3 and the isoxazole cation.
jThermochemistry based on the protonated nitrosobenzene cation.
Downloaded 07 May 2013 to 193.1.100.108. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://jcp.aip.org/about/rights_and_permissions

Reactions of CmHϩn ions with O atoms
4963
J. Chem. Phys., Vol. 112, No. 11, 15 March 2000
0.10
with O atoms. The measured rate coefficient of kϭ2.0
ϫ10Ϫ10 cm3 sϪ1, is in good agreement with the one previ-
ously reported measurement of this reaction,4
ϩ
g
2
→ HC2Oϩ 1A ͒ϩH 1⌺ ͒ϩH S͒ϩ1 kJ molϪ1 ͑6c͒
͑
͑
͑
Ј
2
ϳ0.05
→ CH3COϩ 1A ͒ϩH S͒ϩ256 kJ molϪ1
͑6d͒
͑6e͒
2
͑
͑
1
C2Hϩ 2⌺͒ϩO P͒
3
͑
͑
2
ϳ0.05
→ H2CCOϩ 2B ͒ϩH 1⌺ϩg ͒ϩ436 kJ molϪ1
.
͑
͑
2
1
0.5
→HCOϩ 1⌺ ͒ϩCH ⌸͒ϩ156 kJ molϪ1
͑4a͒
͑4b͒
2
The two major channels ͓͑6a͒ and ͑6b͔͒ proceed by fragmen-
tation of either HCO or HCOϩ from the complex and ac-
count for 80% of the total product channels. The remaining
three channels require more extensive rearrangement. Reac-
tion ͑6c͒ is interesting in that it is almost thermoneutral. Spin
is conserved in all product channels.
͑
͑
g
0.5
→HC2Oϩ 1A ͒ϩH S͒ϩ263 kJ molϪ1
.
2
͑
͑
Ј
Spin is conserved in each product channel. Sufficient energy
is available in reaction ͑4a͒ to form either HOCϩ or HCOϩ
but O atom attack at the–CwC–triple bond would favor the
more stable HCOϩ product. Reaction ͑4b͒ is the result of O
atom–H atom interchange.
C2Hϩ4 is unreactive with O2 but exhibits a fast charge
transfer reaction with NO.
The reaction with molecular oxygen is slow, yielding
two bimolecular channels (HCOϩϩHCO and C2H2OϩϩO)
plus a small ϳ5% association product, C2Hϩ2 •O2. Charge
transfer occurring at ϳ12% of the collision rate was the only
channel observed in the reaction with NO.
E. C2H¿5
Electron impact on ethyl bromide generated the C2H5ϩ
cation. This species is unreactive with O atoms, O2, and NO.
C. C2H¿3
F. c,ac-C3H¿3
C2Hϩ3 was formed by collisional dissociation of C2H5ϩ
during the injection process into the SIFT reaction tube. The
precursor ion, C2Hϩ5 , was generated by electron impact on
ethyl bromide. A moderately fast reaction with atomic oxy-
gen ensued (kϭ1.0ϫ10Ϫ10 cm3 sϪ1) having multiple reac-
tion channels,
These ions designated c,ac-C3Hϩ3 refer to the mixture of
ion structures formed when ethene was subjected to electron
impact in a high pressure ion source. Previous work in our
laboratory has indicated that this mode of generation forms a
mixture that is ϳ40% c-C3Hϩ3 and ϳ60% ac-C3H3ϩ. Al-
though the ratio was not measured in this study, a mixture of
the cyclo-propenylium, c-C3Hϩ3 , and propargyl HCCCHϩ2
ions in approximately a 40/60 ratio was undoubtedly present
in these studies. Subsequent work using our new FA/SIFDT
apparatus and generating only c-C3Hϩ3 ,22 enabled the O
atom c-C3Hϩ3 reaction to be confirmed as nonreactive. The O
atoms in these latter experiments were produced in a micro-
wave discharge of a helium/oxygen mixture. A moderately
fast reaction was observed with atomic oxygen ͑Fig. 2͒, k
ϭ1.5ϫ10Ϫ10 cm3 s1, having multiple reaction channels,
C2Hϩ 1A͒ϩO P͒
3
͑
͑
3
0.85
→ H2CCOϩ 2B ͒ϩH S͒ϩ264 kJ molϪ1
͑5a͒
͑5b͒
͑5c͒
2
͑
͑
1
0.10
→ CH3COϩ 1A1͒ϩ521 kJ molϪ1
͑
ϳ0.05
→ CHϩ3 1A͒ϩCO ⌺ϩ͒ϩ379 kJ molϪ1
.
1
͑
͑
Spin is conserved in the major channel but not in the minor
channels. The structure of the main ion product, H2CCOϩ,
was not determined but likely possibilities are the ketone
structure or the HCwCOHϩ structure. The latter structure is
less likely in view of the greater rearrangement required.
Sufficient energy is available in the reactants to form either
product ion in the absence of large barriers on the potential
surface.
HCCCHϩ2 1A͒ϩO P͒
3
͑
͑
0.30
→ C3H2Oϩ 2A ͒ϩH S͒ϩ236 kJ molϪ1
͑7a͒
͑7b͒
͑7c͒
͑7d͒
2
͑
͑
Ј
0.30
→ C2Hϩ 1A͒ϩCO ⌺ϩ͒ϩ427 kJ molϪ1
1
͑
͑
3
0.25
No reaction of C2H3ϩ with O2 or NO was observed.
→ C2Hϩ 2⌺͒ϩHCO A ͒ϩ57 kJ molϪ1
2
͑
͑
Ј
2
0.15
→ C3OHϩ 3A ͒ϩH 1⌺ϩg ͒ϩ459 kJ molϪ1
.
D. C2H¿4
͑
͑
Ј
2
C2Hϩ4 was formed directly by electron impact on ethene,
Reaction ͑7a͒ proceeds via spin allowed O, H exchange but
the structure of the C3H2Oϩ production is not known with
possible candidates being HCwCCHOϩ, H2CvCvCvOϩ,
and 2-cyclopropen-1-one. All three structures are energeti-
cally accessible from HCCCHϩ2 . Reaction ͑7b͒ proceeds via
C atom transfer and does not conserve spin whereas in reac-
tions ͑7c͒ and ͑7d͒ spin is conserved.
and undergoes
a
relatively fast reaction (kϭ2.4
ϫ10Ϫ10 cm3 sϪ1) with O atoms in multiple product channels,
i.e.,
C2Hϩ 2A͒ϩO P͒
3
͑
͑
4
0.45
→ CHϩ3 1A͒ϩHCO A ͒ϩ344 kJ molϪ1
͑6a͒
͑6b͒
2
Neither c-C3Hϩ3 nor HCCCHϩ2 are reactive with molecu-
lar oxygen but they do exhibit a slow termolecular associa-
tion reaction with NO (kϳ1.2ϫ10Ϫ28 cm6 sϪ1).
͑
͑
Ј
0.35
→ HCOϩ 1⌺ ͒ϩCH 2AЉ2͒ϩ117 kJ molϪ1
͑
͑
3
g
Downloaded 07 May 2013 to 193.1.100.108. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://jcp.aip.org/about/rights_and_permissions

4964
J. Chem. Phys., Vol. 112, No. 11, 15 March 2000
Scott et al.
G. C3H¿5
Due to the presence of two C6Hϩ5 species in these measure-
ments, the uncertainty in the rate coefficients is somewhat
larger ͑Ϯ50%͒ than in the other O atom reactions. Both
product ions, C5Hϩ5 and C3Hϩ3 , are relatively stable com-
pared with their (nϪ1) and (nϩ1) homologues. The C3H2O
neutral product in reaction ͑9b͒ may be propadienone, pro-
pynal, or cyclopropenone or a mixture of species as all three
isomers are accessible from the amount of energy available
in the complex. Spin is conserved in reaction ͑9a͒ but not in
reaction ͑9b͒.
C3Hϩ5 was produced by electron impact on ethene in a
high pressure source. We have shown that the C3Hϩ5 ion
made in the reaction C2Hϩ4 ϩC2H4 after injection into the
flow tube possesses the allyl structure exclusively,
H2CCHCHϩ2 .13 This ion was unreactive with O, O2, and NO.
H. C4H¿2 ,C4H¿3
C4Hϩ2 ,C4Hϩ3 were produced by electron impact on acety-
lene in a high pressure source. C4Hϩ2 was also made sepa-
rately from electron impact on diacetylene, C4H2, and within
the experimental uncertainty there was no difference in the
behavior of C4Hϩ2 formed from either precursor.
A moderately fast reaction was found between C4Hϩ2 and
O atoms (kϭ2.7ϫ10Ϫ10 cm3 sϪ1) with multiple product
channels, including a small association reaction channel,
͑8d͒,
The c-C6Hϩ5 ion also undergoes a fast termolecular as-
sociation reaction with NO having a rate coefficient kϳ1.7
ϫ10Ϫ26 cm6 sϪ1 and also exhibits a slow bimolecular reac-
tion with O2:
0.90
c-C6Hϩ5 ϩO2→ C4H5ϩϩ2COϩ356 kJ molϪ1
,
͑10a͒
͑10b͒
0.10
→ C5H5OϩϩCO.
C4Hϩ 2⌸ ͒ϩO P͒
3
J. c-C6H¿6
͑
͑
g
2
0.50
c-C6Hϩ6 was formed by electron impact on benzene. A
moderately fast reaction was found with O atoms (kϭ1.4
ϫ10Ϫ10 cm3 sϪ1) that had two product channels each con-
serving spin viz.
2
→ C HOϩϩH S͒
͑8a͒
͑8b͒
͑8c͒
͑8d͒
͑
4
0.40
→ C3Hϩ 2A ͒ϩCO ⌺ϩ͒ϩ45 kJ molϪ1
1
͑
͑
Ј
2
ϳ0.05
c-C6Hϩ E1g ϩO P
͒ ͒
2
→ C3OHϩ 3A ͒ϩCH ⌸͒ϩ105 kJ molϪ1
2
3
͑
͑
Ј
͑ ͑
6
ϳ0.05
0.90
→ C4H2Oϩ.
→ C5Hϩ 2A ͒ϩCO ⌺ϩ͒ϩ378 kJ molϪ1
,
͑11a͒
͑11b͒
1
͑
͑
2
6
0.10
No thermochemical data are available for the product ions
C4HOϩ and C4H2Oϩ and their structures are unknown. The
two major channels are similar to the mechanisms in previ-
ous CmHϩn reactions in that the O atom inserts into the hy-
drocarbon cation followed by CO loss ͑8b͒ on H loss ͑8a͒.
No reaction of C4Hϩ2 with O2 was observed but a fast
charge transfer reaction with NO was found.
→ C4H4Oϩ 2A ͒ϩC H 1⌺ϩg ͒ϩ175 kJ molϪ1
,
͑
͑
2
2
2
In the major product channel, ͑11a͒, C atom transfer occurs
leaving a hydrocarbon ion, C5Hϩ6 , of unknown structure.
There was no reaction of c-C6Hϩ6 with molecular oxy-
gen but the slightly endothermic charge transfer reaction
with NO is sufficiently fast to compete with association
(ktotϭ1.4ϫ10Ϫ10 cm3 sϪ1). That the termolecular association
reaction is
C4Hϩ3 was unreactive with O, O2, and NO.
I. c-C6H¿5 ,ac-C6H¿5
c-C6Hϩ6 ϩNO
A mixture of these two ions is formed when bromoben-
zene is subjected to electron impact. A double exponential
semilogarithmic decay was found from this mixture of C6H5ϩ
ions corresponding to different rates of reaction for each iso-
mer with O, O2, and NO. According to the findings of Aus-
loos et al.23 the less reactive isomer is ac-C6Hϩ5 and the
more reactive isomer is c-C6Hϩ5 . Fitting the data to a sum of
two exponentials8 enabled the mixture composition to be de-
fined as ϳ70% c-C6Hϩ5 and ϳ30% ac-C6Hϩ5 .
0.25
→ NOϩϩc-C6H6Ϫ0.4 kJ molϪ1
,
͑12a͒
͑12b͒
0.75
→ c-C6Hϩ6 •NOϩ193 kJ molϪ1
,
relatively fast (kϳ1ϫ10Ϫ26 cm6 sϪ1) implies that a strong
bond is formed between NO and c-C6Hϩ6 and the thermo-
chemistry for ͑12b͒ was calculated with the assumption that
the product ion is protonated nitrosobenzene. Arnold et al.24
have very recently reported a study of the reverse process
͑i.e., the reaction of NOϩ and c-C6H6). Under our flow tube
conditions they report that the total rate coefficient for the
reverse reaction is 1.5ϫ10Ϫ9 cm3 sϪ1 with branching ratios
of around 0.77 for charge transfer and 0.23 for association.
ac-C6Hϩ5 is unreactive with all three neutrals examined
in this study: O, O2, and NO.
c-C6Hϩ5 reacts at
a
moderate rate (kϭ1.0
ϫ10Ϫ10 cm3 sϪ1) with atomic oxygen via two product chan-
nels:
c-C6Hϩ 1A ͒ϩO P͒
3
V. CONCLUSIONS
͑
͑
1
5
0.60
Of the thirteen hydrocarbon ions investigated in this
work, eight exhibited relatively rapid reactions with atomic
oxygen whereas only one, c-C6Hϩ5 , showed a significant re-
action with molecular oxygen. A characteristic of hydrocar-
→ C5Hϩ 3AЈ͒ϩCO ⌺ϩ͒ϩ432 kJ molϪ1
,
͑9a͒
͑9b͒
1
͑
͑
5
2
0.40
→ C3Hϩ 1A͒ϩC H O A ͒ϩ205 kJ molϪ1
,
1
͑
͑
2
Ј
3
3
Downloaded 07 May 2013 to 193.1.100.108. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://jcp.aip.org/about/rights_and_permissions

Reactions of CmHϩn ions with O atoms
4965
J. Chem. Phys., Vol. 112, No. 11, 15 March 2000
the reactions of those CmHnϩ ions that are unreactive with H2
but reactive with O need to be re-examined in the models.
Included in this category are the reactions with C2Hϩ4 , C3H3ϩ,
C4Hϩ2 , c-C6Hϩ5 , and c-C6Hϩ6 .
Finally we note that in diffuse interstellar clouds, atom
densities are enhanced29 and ion atom reactions may have a
different role than in dense clouds.
bon ion reactions with atomic oxygen is the presence of mul-
tiple reaction channels that were found in all reactive
encounters. In a similar study of seventeen hydrocarbons
with atomic nitrogen,11 seven had reactions with rate coeffi-
cients kϾ1ϫ10Ϫ10 cm3 sϪ1, and therefore the fraction of re-
active encounters was slightly greater for atomic oxygen
with eight out of thirteen having kу1ϫ10Ϫ10 cm3 sϪ1. Fur-
ther, the reaction complexity was greater for reactions with O
atoms. Generally, ions unreactive with N atoms were also
unreactive with O atoms ͑e.g., C2Hϩ5 , c-C3H3ϩ, ac-C3H5ϩ,
C4Hϩ3 , and ac-C6Hϩ5 ). Another aspect of the comparison is
that whereas CmHϩn ion N atom reactions had a propensity to
form HCN ion or neutral products, O atoms favored HCO
ion products for mϽ3, but CO neutral products for mу3.
Other characteristics common to both investigations were the
number of reactions that apparently violate spin conserva-
tion. This observation has been commented on previously by
Ferguson and colleagues.21,25
ACKNOWLEDGMENT
We thank the Marsden fund for financial support of this
work.
1 E. E. Ferguson, F. C. Fehsenfeld, P. D. Golden, A. L. Schmeltekopf, and
H. I. Schiff, Planet. Space Sci. 13, 823 ͑1965͒.
2 F. C. Fehsenfeld, Astrophys. J. 209, 638 ͑1976͒.
3 D. K. Bohme, G. I. Mackay, and H. I. Schiff, J. Chem. Phys. 73, 4976
͑1980͒.
4 A. A. Viggiano, F. Howorka, D. L. Albritton, F. C. Fehsenfeld, N. G.
Adams, and D. Smith, Astrophys. J. 236, 492 ͑1980͒.
5 A. A. Viggiano, R. A. Morris, J. F. Paulson, and E. E. Ferguson, Chem.
Phys. Lett. 148, 296 ͑1988͒.
The absence of any reaction with O atoms by four hy-
drocarbon ions was not because of a lack of exothermic
product channels. For example, although C2Hϩ5 did not react
with atomic oxygen, the product channel to HCOϩϩCH4 is
6 V. Le Page, Y. Keheyan, V. M. Bierbaum, and T. P. Snow, J. Am. Chem.
Soc. 119, 8373 ͑1997͒.
7 G. B. I. Scott, D. A. Fairley, D. B. Milligan, C. G. Freeman, and M. J.
McEwan, J. Phys. Chem. A 103, 7470 ͑1999͒.
exothermic by 399 kJ molϪ1
,
8 M. J. McEwan, in Advances in Gas Phase Ion Chemistry, edited by N. G.
Adams and L. M. Babcock ͑JAI, Greenwich, CT, 1992͒, Vol. 1, p. 1.
9 P. D. Goldan, A. L. Schmeltekopf, F. C. Fehsenfeld, H. I. Schiff, and E. E.
Ferguson, J. Chem. Phys. 44, 4095 ͑1966͒.
C2Hϩ5 ϩOyHCOϩϩCH4.
͑13͒
Presumably the barriers on the potential surface are too high
for the extensive rearrangement required. An alternative to a
bimolecular reaction is a termolecular association reactions,
i.e.,
10 G. B. I. Scott, D. A. Fairley, C. G. Freeman, and M. J. McEwan, J. Chem.
Phys. 109, 9010 ͑1998͒.
11 G. B. I. Scott, D. A. Fairley, C. G. Freeman, M. J. McEwan, and V. G.
Anicich, J. Phys. Chem. A 103, 1073 ͑1999͒.
12 Webbook, W. G. Mallard, Gen. Ed. ͑NIST, Gaithersburg, MD, 1998͒.
13 D. A. Fairley, D. B. Milligan, L. M. Wheadon, C. G. Freeman, R. G. A. R.
Maclagan, and M. J. McEwan, Int. J. Mass Spectrom. Ion Processes 185Õ
186Õ187, 253 ͑1999͒.
C2Hϩ5 ϩOϩMyC2H5OϩϩM.
͑14͒
Although reaction ͑14͒ is exothermic by 568 kJ molϪ1 for the
formation of CH3CHOHϩ, it is not observed. The lack of
association reactions for small CmHϩn –atom reactions is a
consequence of the short lifetime of the intermediate because
of the energy carried into the complex by the atom. O atom
association may be a more important outcome for larger
CmHϩn ions where the excess energy in the complex can be
accommodated by the large number of energy modes. In-
deed, a 40% association channel was observed by Le Page
et al.6 for the C10Hϩ8 ϩO reaction.
14 D. Smith and N. G. Adams, Chem. Phys. Lett. 54, 535 ͑1978͒.
15 N. G. Adams and D. Smith, Chem. Phys. Lett. 79, 563 ͑1981͒.
16 V. G. Anicich, J. Phys. Chem. Ref. Data 22, 1469 ͑1993͒.
17 W. T. Huntress, Astrophys. J., Suppl. 33, 495 ͑1977͒.
18 L. W. Sieck and J. H. Futrell, J. Chem. Phys. 48, 1409 ͑1968͒.
19 H. Hawley, T. L. Mazely, L. K. Randeniya, R. S. Smith, X. K. Zeng, and
M. A. Smith, Int. J. Mass Spectrom. Ion Processes 97, 55 ͑1990͒.
20 D. B. Milligan, D. A. Fairley, C. G. Freeman, and M. J. McEwan, Int. J.
Mass Spectrom. Ion Processes ͑submitted͒.
21 E. E. Ferguson, Chem. Phys. Lett. 99, 89 ͑1983͒.
22 M. J. McEwan, C. L. McConnell, C. G. Freeman, and V. G. Anicich, J.
Phys. Chem. 98, 5068 ͑1994͒.
Atomic oxygen can be a significant component of inter-
stellar clouds26 and in dark clouds the aggregate amount of
oxygen present is greater than the total amount of carbon.27
In specific discrete regions, the O atom density can be sig-
nificantly elevated above the ambient levels such as the
Orion compact ridge near the luminous infrared source
IRc2.28 Bettens and Brown27 have considered participation
by hydrogen ion–O atom reactions in interstellar cloud
chemistry and concluded that the influence of these reactions
on the chemistry of the clouds was small whereas neutral–O
atom chemistry may play a much more significant role. Since
their study, however, more data have become available and
23 P. Ausloos, S. G. Lias, T. J. Buckley, and E. E. Rogers, Int. J. Mass
Spectrom. Ion Processes 92, 65 ͑1989͒.
24 S. T. Arnold, S. Williams, I. Dotan, A. J. Midey, R. A. Morris, and A. A.
Viggiano, J. Phys. Chem. A 103, 8421 ͑1999͒.
25 W. Federer, H. Villinger, W. Lindinger, and E. E. Ferguson, Chem. Phys.
Lett. 123, 12 ͑1986͒.
26 H. H. Lee, R. P. A. Bettens, and E. Herbst, Astron. Astrophys., Suppl. Ser.
B119, 111 ͑1996͒.
27 R. P. A. Bettens and R. D. Brown, Mon. Not. R. Astron. Soc. 258, 347
͑1992͒.
28 T. J. Millar, E. Herbst, and S. B. Charnley, Astrophys. J. 369, 147 ͑1991͒.
29
˚
W. M. Irvine, P. F. Goldsmith, and A. Hjalmarson, in Interstellar Pro-
cesses, edited by D. J. Hollenbach and H. A. Thronson ͑Reidel, Dordrecht,
1987͒.
Downloaded 07 May 2013 to 193.1.100.108. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://jcp.aip.org/about/rights_and_permissions