Ultraviolet photochemistry of diacetylene: Reactions with benzene and toluene

Article abstract of DOI:10.1021/jp001427v

The reactions of metastable diacetylene with benzene and toluene are explored using a molecular beam pump-probe time-of-flight mass spectrometer. Diacetylene is laser-excited to the 2¹⁰6¹₀ band of the ¹Δ_u←X¹Σ⁺_g transition, whereupon rapid intersystem crossing occurs to the lowest triplet states. The triplet state diacetylene then reacts with either benzene or toluene as the gas mixture traverses a short reaction tube (approximately 20 μs). The reactions are quenched as the gas mixture expands into the ion source region of a time-of-flight mass spectrometer where the primary photoproducts are detected using vacuum ultraviolet (VUV) photoionization or resonant two-photon ionization (R2PI). The major products from the reaction of diacetylene and benzene have molecular formulas C₈H₆ and C₁₀H₆, and are identified as phenylacetylene and phenyldiacetylene using R2PI spectroscopy. The major products from metastable diacetylene's reaction with toluene are C₉H₈ and C₁₁H₈. The C₉H₈ product is confirmed as a mixture of o-, m-, and p-ethynyltoluene, with the ortho product dominating. Mechanisms for the formation of the above products are proposed based on deuterium substitution studies of the reactions. The potential importance of these reactions is discussed as they relate to hydrocarbon growth in sooting flames.

Full text of DOI:10.1021/jp001427v

10312
J. Phys. Chem. A 2000, 104, 10312-10320
Ultraviolet Photochemistry of Diacetylene: Reactions with Benzene and Toluene†
Allison G. Robinson, Paul R. Winter, Christopher Ramos, and Timothy S. Zwier*
Department of Chemistry, Purdue UniVersity, West Lafayette, Indiana 47907-1393
ReceiVed: April 14, 2000; In Final Form: June 12, 2000
The reactions of metastable diacetylene with benzene and toluene are explored using a molecular beam pump-
1
1
1
1 +
probe time-of-flight mass spectrometer. Diacetylene is laser-excited to the 2
0
6
0
band of the ∆
u
r X Σ
g
transition, whereupon rapid intersystem crossing occurs to the lowest triplet states. The triplet state diacetylene
then reacts with either benzene or toluene as the gas mixture traverses a short reaction tube (∼20 µs). The
reactions are quenched as the gas mixture expands into the ion source region of a time-of-flight mass
spectrometer where the primary photoproducts are detected using vacuum ultraviolet (VUV) photoionization
or resonant two-photon ionization (R2PI). The major products from the reaction of diacetylene and benzene
have molecular formulas C
R2PI spectroscopy. The major products from metastable diacetylene’s reaction with toluene are C
. The C product is confirmed as a mixture of o-, m-, and p-ethynyltoluene, with the ortho product
8
H
6
and C10
H
6
, and are identified as phenylacetylene and phenyldiacetylene using
9 8
H
and
C
11
H
8
9 8
H
dominating. Mechanisms for the formation of the above products are proposed based on deuterium substitution
studies of the reactions. The potential importance of these reactions is discussed as they relate to hydrocarbon
growth in sooting flames.
I. Introduction
photochemistry of diacetylene.18 Among the important results
of their work was the determination that following ultraviolet
Diacetylene (C4H2, HsCtC-CtCsH) is an important
molecule in combustion chemistry. It is one of the most
abundant C4 species found in sooting flames, particularly
acetylene flames.1 Diacetylene is also thought to play a role
in the formation of aromatic molecules in flames, which lead
to soot formation.5 There are two main pathways in current
models for benzene formation in flames. First, the radical/radical
recombination of two propargyl radicals ( C˙ H2sCtCsH)
produces 1,5 hexadiyne (HCtC(CH2)2CtCH) which can
subsequently isomerize to benzene.4 Second, acetylene can
react with an isomer of either C4H3 or C4H5 to produce C6H5
or C6H7, which can add or lose an H atom to form benzene.
excitation, diacetylene reacts out of metastable triplet states and
not via radical reactions. This molecular reactivity is a natural
consequence of the strong bonds in diacetylene, which prevent
radical formation following photoexcitation throughout much
-4
18
of the ultraviolet. However, rapid polymerization in their bulb
studies prevented the determination of primary products. We
have studied the primary products of the reactions of metastable
,6
19-23
diacetylene with several small hydrocarbons,
but the present
work marks the first study of its reactions with an aromatic
molecule.
,7
In seeking to understand the reactivity and importance of the
triplet states of C4H2, further spectroscopic characterization of
the triplet states themselves is needed. The electron energy loss
spectra (EELS) of Allan identified two low-lying triplet states,
Both C4H3 and C4H5 are closely coupled with reactions either
forming or depleting diacetylene.8
-10
We recently demonstrated
the importance of the second pathway in forming benzene in
3
+
3
1
1
T1( Σu ) and T2( ∆u), which are likely candidates for the
methane flames doped with diacetylene and vinylacetylene.
metastable state(s) responsible for the observed photochemis-
After the formation of the first aromatic ring, further growth
to soot requires production of larger polycyclic aromatic
hydrocarbons (PAHs). One of the proposed routes to formation
of the second aromatic ring is the addition of acetylene to
2
4
try. We have recently recorded a singlet-triplet absorption
spectrum of diacetylene in the 370 nm region using cavity
ringdown spectroscopy, probing the transition assigned to the
3
T2 upper state, which is of ∆u symmetry in the linear
benzene through the hydrogen abstraction acetylene addition
2
5
configuration.
,8,12
(
HACA) mechanism.5
Alternatively, recombination of cy-
In this paper, we focus instead on the chemistry of the triplet
state(s) of diacetylene. Recently, we demonstrated that the
reaction of metastable diacetylene with 1,3-butadiene formed
benzene and phenylacetylene, raising the prospect that such
reactions could play a role in aromatic formation in flames.23
Accordingly, we are interested in exploring whether diacetylene
can play a role in forming larger fused aromatic rings. The
present paper, therefore, analyzes the reactions of metastable
diacetylene with benzene and toluene using the same laser
clopentadienyl radicals (C5H5) is thought to be an efficient route
to naphthalene.8,13,14
Several other pathways have been proposed,
including one in which diacetylene reacts with phenyl radical
to form the naphthyl radical,15 which could then abstract a
hydrogen to give naphthalene. Further elucidation of the possible
pathways to the formation of substituted aromatics and subse-
quently naphthalene or substituted naphthalenes and higher order
1
6,17
ring structures
is important in determining suitable mech-
anisms for growth of PAHs in flames.
Diacetylene has a rich history of photochemical studies.
Glicker and Okabe carried out one of the early studies on the
1
9-23,26
pump-probe scheme used in earlier studies.
II. Experimental Section
The experiments described in this paper were carried out in
a molecular beam time-of-flight (TOF) mass spectrometer
designed to study the primary products of photoinitiated
†
Part of the special issue “C. Bradley Moore Festschrift”.
To whom correspondence should be addressed. Fax: (765) 494-0239.
*
E-mail: zwier@purdue.edu.
1
0.1021/jp001427v CCC: $19.00 © 2000 American Chemical Society
Published on Web 07/29/2000

Ultraviolet Photochemistry of Diacetylene
J. Phys. Chem. A, Vol. 104, No. 45, 2000 10313
III. Results
reactions.20 The gas handling system was modified in order to
minimize the contact of the diacetylene with liquid benzene or
toluene because we found that diacetylene reacts with liquid
benzene to form polymeric products, while the two gas phase
reactants do not react on the time scale of our experiment.
Therefore, the reaction mixture was prepared by flowing helium
at approximately 1 sccm over liquid benzene (toluene) in a
stainless steel reservoir to pick up the sample at its vapor
pressure (100 and 33 Torr, respectively). The concentration of
benzene (toluene) in helium was approximately 5% (2%) at 20
psig backing pressure. The diacetylene (4-7% in helium) was
introduced to the pulsed valve through a separate line, again at
approximately 1 sccm flow. The two gas lines were combined
just prior to the pulsed valve (R.M. Jordan Co.) and the mixture
was pulsed into the chamber at 10 Hz.
A. C H * + C H VUV Photoionization Studies. Figure
4
2
6
6
1a shows the difference mass spectrum from the reaction of a
1:1.3 mixture of diacetylene and benzene, as determined from
the peak areas of the primary ions in the VUV mass spectra.
The reported spectrum is the difference between a mass
spectrum taken with the photoexcitation laser on and one taken
with the photoexcitation laser off. The difference mass spectrum
highlights the photochemical products of the reaction of
metastable diacetylene with benzene. The molecular formulas
in the figure were determined from the mass-to-charge ratios
of the respective peaks.
The two reactant peaks, C4H2 (m/z ) 50) and C6H6 (m/z )
8), are not shown in the figure. They are about 100-200 times
7
larger than the photoproduct peaks, consistent with the small
amount of diacetylene excited by the ultraviolet laser. A high
voltage pulse (+800 V) is applied to a deflection plate, located
early in the TOF tube, to deflect the reactant ions away from
the microsphere plate so the photoproduct signals can be
observed. Consequently, photoproducts with masses in the 45-
In these experiments, it was important to avoid the photo-
polymerization reactions that diacetylene experiences under bulb
conditions. The reactants were expanded into a short reaction
tube (1 cm long, 2 mm i.d.) affixed to the pulsed valve. While
the reactant mixture was in the tube, photochemistry was
initiated using the doubled output (0.7 mJ/pulse) of a Nd:YAG
pumped KTP/BBO optical parametric converter (LaserVision).
8
0 amu range are not detected, although no such products are
22
anticipated. Previous studies of polyyne, polyenyne, cumulene,
and aromatic molecules have shown no evidence of fragmen-
23
1
1
1
The photoexcitation laser was tuned to the 2 06 0 of the ∆u r
tation upon 118 nm ionization, consistent with other studies of
1
+
19
X Σ g transition in diacetylene (231.5 nm). From the known
31
a range of hydrocarbons. Therefore, we assume throughout
-
1
-1 18
absorption cross section of diacetylene (80 cm atm ), we
estimate that 2-3% of the diacetylene was photoexcited by the
ultraviolet laser. Rapid intersystem crossing occurs from the
excited singlet state, either directly or mediated by the lower
this paper that the species that appear in the difference mass
spectra represent the nascent, neutral products formed in the
gas phase reaction.
The two major photochemical product peaks observed for
the C H * + C H reaction (Figure 1a) are m/z ) 102 (C H )
1
+
energy Σu state, producing high vibrational levels of the low-
4
2
6
6
8
6
3
3
+ 18,20
lying triplet states ( ∆u and/or Σu ).
Reactions then occur
and m/z ) 126 (C H ) with m/z ) 127 (C H ) as a minor
1
0
6
10
7
3
2
in parallel with vibrational deactivation in the triplet manifold.
The 20 µs traversal of the gas mixture in the tube allows for
sufficient collisions to initiate primary C4H2* + C6H6/C7H8
chemistry, after which free expansion into the vacuum chamber
quenches further reactions. The absence of larger photoproducts
provides strong evidence that secondary reactions are of minor
importance using this scheme.
product. The percent product yields are given in Table 1. The
C8H2 and C8H3 products from the C4H2* + C4H2 reaction also
appear in Figure 1a, as this reaction occurs in competition with
the reaction of interest. The relevant reactions are as follows:
C H * + C H f C H + (C H )
k
k
k
(1a)
(1b)
(1c)
4
2
4
2
6
2
2
2
C H
6
2
2
3
f C H + (H or 2H)
8
2
2
C H
The gas mixture was ionized approximately 7 cm downstream
from the exit of the reaction tube in the extraction region of the
TOF mass spectrometer using either vacuum ultraviolet (VUV)
ionization or resonant two-photon ionization (R2PI). VUV
photoionization achieves general mass analysis with minimal
fragmentation using 118 nm (10.5 eV) light produced by tripling
the third harmonic of a Nd:YAG laser in a xenon/argon gas
mixture.2 Alternatively, R2PI produces ultraviolet spectra of
the products, which can be used in structural analysis. Scans
over the wavelength ranges depicted in this work (570-480
nm) were taken using the doubled output of a Nd:YAG pumped
dye laser in three dye regions (Coumarin 540A, 503, 480).
8
f C H + (H)
8
3
C H
8
C H * + C H f C H + (C H )
k
(2a)
(2b)
(2c)
4
2
6
6
8
6
2
2
C H
8
6
f C H + (H or 2H)
kC10H6
kC10H7
1
0
6
2
0,27
f C H + (H)
1
0
7
The products listed in parentheses have ionization potentials
greater than 10.5 eV, so they are not observable using 118 nm
VUV photoionization. The C6H2 and C8H2 products from C4H2*
Benzene, deuterated benzene, toluene, and deuterated toluene
were used as supplied (Fisher, 99.9%; Cambridge Isotope
Laboratories, 99.6%; Fisher, 99.8%; Aldrich, 99%; respectively).
Diacetylene and deuterated diacetylene were synthesized in our
laboratory using previously described methods.
yldiacetylene was synthesized using the method of Brandsma.
Due to the difficulty in isolating pure phenyldiacetylene, the
synthesis was verified by comparison of the ultraviolet spectrum
of the reaction mixture in methanol with a literature ultraviolet
spectrum of phenyldiacetylene.29 The o-ethynyltoluene was
synthesized at Hampford Research, Inc.,30 while the m- and
p-ethynyltoluenes are available commercially (Lancaster, 99%
and Aldrich, 97%).
+
C4H2 have been shown to be triacetylene and tetracetylene,
26
respectively in an earlier work. The C6H2 product is below
the mass of benzene and therefore is pulsed away by the high
voltage pulse.
2
0,21
The phen-
In order to confirm that the photoproducts of the C4H2* +
C6H6 reaction were coming exclusively from photoexcited
diacetylene, action spectra of the three photoproducts were
28
1
1
recorded by tuning the photoexcitation laser through the 2 06 0
transition of diacetylene while monitoring the relevant mass
channels. All three products are formed following excitation of
the diacetylene, producing action spectra (not shown) that mirror
the absorption spectrum of diacetylene. Benzene has a small
1
5
1
1
absorption (6 01 0) near the diacetylene 2 06 0 transition, so

10314 J. Phys. Chem. A, Vol. 104, No. 45, 2000
Robinson et al.
out. Figure 1b and c, shows the difference mass spectra for the
C4D2* + C6H6 and the C4H2* + C6D6 reactions, respectively.
Percent product yields for these reactions are given in Table 1.
In the former reaction (Figure 1b), the mass of the two major
photoproducts (C8H6 and C10H6) and the minor C10H7 product
all increased by only one mass unit, consistent with one
deuterium atom from C4D2 being retained in each photoproduct
(C8H5D, C10H5D, and C10H6D).
C D * + C H f C H D + (C HD)
(3a)
(3b)
(3c)
4
2
6
6
8
5
2
f C H D + (HD or H + D)
1
0
5
f C H D + (D)
1
0
6
Figure 1. VUV photoionization difference mass spectra highlighting
the photoproducts for the reactions (a) C * + C , (b) C * +
, and (c) C * + C . The reactant mass peaks are 100-200
times larger than the photoproduct peaks and are not shown. Signal
from the fully non-deuterated reaction is present in the two deuterated
cases, due to residual non-deuterated sample in the gas handling system.
4
H
2
6
H
6
4
D
2
The exclusive formation of C10H6D in reaction 3c shows that it
is a diacetylenic hydrogen that is preferentially lost from the
reaction complex.
The reaction of C4H2* + C6D6 (Figure 1c) produces isotopic
products consistent with those from C4D2* + C6H6.
C
6
H
6
4
H
2
6 6
D
difference mass spectra were also taken at two other diacetylene
C H * + C D f C D H + (C HD)
(4a)
(4b)
(4c)
2
1
1
33
4
2
6
6
8
5
2
transitions, the 2 06 0 (220.7 nm) and the 6 0 (243.1 nm), where
there are no benzene absorptions. The data at the other two
photochemistry wavelengths reproduced the products and
f C D H + (HD or H + D)
1
0
5
1
1
intensity ratios shown in Figure 1a. The 2 06 0 transition was
used as the primary photoexcitation wavelength because it is
the most intense diacetylene transition in the wavelength region
well below the lowest dissociation threshold of C4H2.
f C D H + (H)
10 6
As before, the C8D5H and C10D5H products indicate that the
deuterated benzene is contributing five deuterium atoms and
the diacetylene one hydrogen to each product. The minor
C10D6H photoproduct also confirms that all of benzene’s
deuterium atoms are retained in this product.
B. C4H2* + C6H6 Product R2PI Studies. The mass spectral
data provides the molecular formulas of the photoproducts, but
it cannot distinguish among the various possible isomers that
could be produced. Therefore, R2PI spectra were taken of the
C8H6 and C10H6 products. The R2PI spectrum of the C8H6
product is shown in Figure 2a, compared with the R2PI spectrum
in the region of the S1 r S0 origin of a known sample of
One potential complication of the present study is the
possibility that C4H2* could undergo triplet-triplet energy
transfer to benzene, followed by the reaction of C6H6* with
C4H2. While not a focus of the present study, we also observed
C8H6 and C10H6 photoproducts from the C6H6* + C4H2 reaction
1
when benzene was excited via its 6 0 transition (259.15 nm).
-
1
Under the laser power (∼0.7 mJ) and bandwidth (∼2 cm )
conditions of this experiment, we estimate that the photoexci-
tation laser excites no more than a few percent of the benzene
molecules in the reaction tube. Following photoexcitation at the
1
0 transition, the benzene that does not fluoresce (18%)34
23
6
phenylacetylene taken under similar conditions (Figure 2b).
predominantly intersystem crosses to the triplet manifold
(
The match between the two spectra positively identifies the C8H6
photoproduct as phenylacetylene.
63%).35 The difference mass spectra recorded following excita-
tion of C6H6 had a qualitatively different intensity ratio for the
two primary products (∼2.5:1 C10H6:C8H6 in the benzene
excitation case, as opposed to ∼1:1 C10H6:C8H6 in the case of
diacetylene excitation). Furthermore, there was no evidence of
C4H2* + C4H2 reaction products when we excited the benzene
rather than diacetylene. These findings suggest that triplet-
triplet energy transfer in either direction does not compete
effectively with direct reactions of the triplet state molecule.
While the C6H6* chemistry is interesting in its own right, it is
The second major product, C10H6, could be any one of several
possible isomers, including o-, m-, and p-diethynylbenzene and
phenyldiacetylene.
The R2PI spectrum of the C H photoproduct is shown in
1
0
6
not pursued further here.
Figure 3a. The spectrum shows only broad features of about
350 cm fwhm, with hints of substructure beneath it, presenting
As in earlier studies,2
0-23
reaction time scans were carried
-1
out to ensure that the observed photoproducts have arrival times
consistent with primary, gas phase photoproducts uncompro-
mised by wall reactions. In these scans, the VUV photoioniza-
tion laser is fixed in time relative to the gas pulse while the
timing of the photoexcitation laser is scanned. The products all
appear in a symmetric, Gaussian-like distribution in time, with
a full width at half-maximum (fwhm) of about 20 µs. They are
peaked at an arrival time corresponding to their traversal from
the reaction tube to the ion source region. The form of this
distribution was the same for all of the C4H2* + C6H6 and C4H2*
a sharp contrast to the phenylacetylene spectrum in Figure 2.
A gas phase UV-visible absorption spectrum was taken of the
vapor above a known sample of phenyldiacetylene (Figure 3b).
The broad features of the room-temperature ultraviolet absorp-
tion spectrum of the vapor above the phenyldiacetylene match
the transitions in the C H photoproduct R2PI spectrum,
1
0
6
providing confirmation that the C H photoproduct is phenyl-
1
0
6
diacetylene.
Attempts to record an R2PI spectrum from the known
phenyldiacetylene were unsuccessful. The difficulties in obtain-
ing this spectrum were likely the combined consequence of the
problems with handling this compound and the modest vapor
pressure of phenyldiacetylene in our impure sample. The
+
C7H8 photoproducts.
In order to gain further insight into the structure of the
photoproducts, deuterium isotope labeling studies were carried

Ultraviolet Photochemistry of Diacetylene
J. Phys. Chem. A, Vol. 104, No. 45, 2000 10315
TABLE 1: Percent Product Yieldsa,b for Metastable Diacetylene Reacting with Benzene and Toluene and Their Corresponding
Isotopic Reactions
C
4
2
H * + C
H
6 6
C
4
2
D * + C
H
6 6
C
4 2
H * + C
6
D
6
C
4 2
H * + C
7
H
8
C
4
D
2
* + C
H
7 8
C
H
4 2
7 8
* + C D
C
C
C
8
H
6
39%
48%
13%
C
C
C
8
H
5
D
40%
44%
16%
C
C
C
8
D
5
H
38%
31%
31%
C
C
C
C
C
8
H
H
10
11
11
6
4%
42%
5%
38%
11%
C
C
C
C
C
8
9
H
H
5
D
D
4%c
45%
6%
34%
11%
C
C
C
C
C
8
D
D
5
H
H
4%c
25%
8%
25%
38%
10
H
H
6
10
H
H
5
D
D
10
D
D
5
H
H
9
8
H
H
H
7
9
7
10
7
10
6
10
6
6
8
9
10
11
11
H
H
H
5
7
8
D
10
D
11
D
11
D
5
7
8
H
D
D
H
H
a
Percent product yields were determined from the integrated peak areas in the difference mass spectrum. b Estimated error on the percent yields
c
is (3%. This yield was not available due to interference from other features in the mass spectrum. The yield is assumed it to be the same as in
the non-deuterated case.
1
02), C10H6 (m/z ) 126), and C11H9 (m/z ) 141). The major
photoproducts differ from the corresponding products in the
C4H2* + C6H6 reaction by 14 mass units, as would be
anticipated if each product is a methyl-substituted version of
its analogue in the benzene reaction. The relevant reactions are
detailed below, listed in order of percent photoproduct yield
(see Table 1).
C H * + C H f C H + (C H )
k
C H
9 8
(5a)
4
2
7
8
9
8
2
2
f C H + (H or 2H)
kC11H8 (5b)
kC11H9 (5c)
1
1
8
2
f C H + (H)
1
1
9
Figure 2. (a) One-color R2PI of the C
. (b) R2PI spectrum of the S
phenylacetylene seeded in helium under similar conditions to those used
for the photochemistry, identifying the C photoproduct from C H *
as phenylacetylene (taken from ref 23).
8
H
1
6
photoproduct from C
4 2
H *
f C H + (CH ) or CH + (H) k
(5d)
(5e)
+
C
6
H
6
r S origin region of
0
10
6
4
3
C10H6
f C H + C H
4
k
8
H
6
4
2
8
6
3
C H
8
6
+
6 6
C H
Once again, the C4H2* + C4H2 reaction occurs in parallel
with the C4H2* + C7H8 reaction, but these products are outside
the mass range shown in Figure 4. As with the benzene reaction,
action spectra and reaction time profiles were taken to confirm
that the photoproducts were all coming from the excitation of
gas phase diacetylene and that wall effects were negligible.
The reactions of C4D2* + C7H8 and C4H2* + C7D8 were
also carried out (not shown). The percent product yields are
included in Table 1. The results were consistent with what was
observed in the deuterated benzene experiments, giving the
following suite of reactions:
C D * + C H f C H D + (C HD)
(6a)
(6b)
(6c)
(6d)
(6e)
4
2
7
8
9
7
2
Figure 3. (a) One-color R2PI of the C10
H
6
photoproduct from C
4
H
2
*
f C H D + (HD or H + D)
11 7
+
6 6
C H
, smoothed using a third-order boxcar function. (b) Ultraviolet
absorption spectrum of gas phase phenyldiacetylene at room temper-
ature.
f C H D + (D)
11 8
f C H D + (CH D) or CH + (D)
1
0
5
3
3
inherent breadth of phenyldiacetylene’s spectrum in the ultra-
violet region suggests that fast nonradiative processes occur in
its excited state. Similar fast, nonradiative processes are present
in diacetylene’s S2 r S0 spectrum,33 suggesting that the
diacetylenic moiety is responsible for the fast, nonradiative
processes in phenyldiacetylene. The long Franck-Condon
progression that appears in the spectrum of phenyldiacetylene
is due to the CtC stretch (2053 cm ), and is similar to the
corresponding progression in diacetylene’s gas phase ultraviolet
spectrum, but shifted about 30 nm to the red.
C. C4H2* + C7H8 VUV Photoionization Studies. In order
to probe the effect that methyl substitution on an aromatic ring
has on the diacetylene photochemistry, a similar set of experi-
ments was carried out with toluene. Figure 4 shows a difference
TOF mass spectrum of the photoproducts of the C4H2* + C7H8
reaction from a mixture containing a 1:1.1 ratio of diacetylene
to toluene. The two major products are C9H8 (m/z ) 116) and
C11H8 (m/z ) 140) with minor contributions from C8H6 (m/z )
f C H D + C H D
8
5
3
3
C H * + C D f C D H + (C HD)
(7a)
(7b)
(7c)
(7d)
(7e)
4
2
7
8
9
7
2
f C D H + (HD or H + D)
11
7
f C D H + (H)
-
1
11
8
f C D H + (CD H) or CD + (H)
1
0
5
3
3
f C
D
H + C
D
H
8
5
3
3
The deuterated toluene reaction results confirmed that the C8H6
and C10H6 products are indeed minor products of the toluene
reaction and not simply products from a reaction of diacetylene
with residual benzene in the system.
D. C4H2* + C7H8 Product R2PI Studies. On the basis of a
comparison with the C4H2* + C6H6 reaction, one anticipates

10316 J. Phys. Chem. A, Vol. 104, No. 45, 2000
Robinson et al.
Figure 6. Blow up of the one-color R2PI of the (a) C
9 8
H photoproduct
Figure 4. VUV photoionization difference mass spectrum highlighting
the photoproducts for the reaction C * + C . The reactant mass
from C , (b) m-ethynyltoluene, and (c) o-ethynyltoluene
4
H
2
7 8
* + C H
4
H
2
7 8
H
over the origin region.
peaks are 100-200 times larger than the photoproduct peaks and are
therefore not shown. The minor product C
interference effects from an impurity.
8 6
H is also not shown due to
The R2PI spectrum of the photochemical product in the 268-
85 nm region is shown in Figure 5a. Indene was removed as
2
a candidate based on a comparison with a literature R2PI
36
spectrum. The R2PI spectrum of 3-phenyl-1-propyne (Aldrich,
7%) also was not a match with the observed photochemical
9
product. The R2PI spectra of the three ethynyltoluene isomers
are shown in Figure 5b-d. The C9H8 photochemical product
spectrum (Figure 5a) contains transitions readily assignable to
both o- and p-ethynyltoluene isomers. Figure 6a shows a blow
up of the photoproduct spectrum around the origin region of
the ortho and meta isomers, while Figure 6b and c shows the
corresponding m- and o-ethynyltoluene spectra over the same
region. The one-to-one correspondence between several of the
weak features in the photoproduct spectrum with the m-
ethynyltoluene transitions identifies the meta isomer as also
present in the photoproduct spectrum. In order to quantify the
amount of o- and p-ethynyltoluene in the photoproduct, a
spectrum was recorded of a known mixture of the ortho and
para isomers. Ratios were taken of peak areas from that spectrum
to their corresponding peaks in the photoproduct spectrum, and
it was determined that the photoproduct represents a 4.9((1.8):1
mixture of o- to p-ethynyltoluene. Due to the weak intensity of
the meta transitions in the photoproduct, no attempt was made
to quantify its yield relative to the other isomers.
Attempts to record an R2PI spectrum of the C11H8 photo-
product produced a weak signal with insufficient intensity to
make any clear structural deductions. However, the resonantly
enhanced signal occurs in the same wavelength region as
phenyldiacetylene, suggesting that this product is the analogous
diethynyltoluene. On the basis of a comparison with the C9H8
photoproduct, one might anticipate that C11H8 is a mixture of
the ortho, meta, and para isomers. However, given the difficul-
ties encountered in preparing the phenyldiacetylene sample, no
efforts were made to synthesize any of the diethnyltoluene
isomers, and no positive structural identification was made. In
addition, no attempts were made to record R2PI spectra of the
minor photoproducts.
Figure 5. (a) One-color R2PI of the C
. (b) R2PI spectrum of the S
o-ethynyltoluene seeded in helium under similar conditions to those
used for the photochemistry. (c) R2PI spectrum of the S r S origin
region of p-ethynyltoluene seeded in helium under similar conditions
to those used for the photochemistry. (d) R2PI spectrum of the S
origin region of meta-ethynyltoluene seeded in helium under similar
9
H
1
8
photoproduct from C
4 2
H *
+
C
7
H
8
r S origin region of
0
1
0
1
r
S
0
conditions to those used for the photochemistry. Spectra in this figure
are not power normalized.
the C9H8 photochemical product to be one or more of the three
ethynyltoluene isomers (ortho, meta, and para). Indene and
3
-phenyl-1-propyne are other possibilities consistent with the
IV. Discussion
molecular formula.
A. Percent Product Yields. The percent product yields
extracted from the difference TOF mass spectra are given in
Table 1; analogous sets of products dominate in all circum-
stances (C8H6-xDx and C10H6-xDx for the benzene reaction,
C9H8-xDx and C11H8-xDx for the toluene reaction). In fact, the
distributions of products both between the two reactions and
between those formed from non-deuterated and deuterated
reactants are consistent throughout. The only exceptions are the
This range of potential products highlights the importance of
having spectroscopic information to aid in positive identification
of the structure(s) of the products.

Ultraviolet Photochemistry of Diacetylene
J. Phys. Chem. A, Vol. 104, No. 45, 2000 10317
Figure 8. Thermodynamic energetics for the reactions C
and C * + C . See Table 3 for more detail.
H
4 2
6 6
* + C H
H
4 2
7 8
H
TABLE 2: Heats of Formation for the Various
Photoproducts
Figure 7. (a) Concentration study for the reaction C
4
H
2
* + C
products]/I[C
products] versus the ratio of reactant ion intensities I[C ]/I[C
The slope of the line fit to these points gives the relative rate constant
for the reaction, kC6H6/kC4H ) 1.67. (b) Concentration study for the
6
H
6
.
2
Plotted are the ratio of product ion intensities I[C
H
6 6
4
H
species
∆H
f
(kcal/mol) at 298 °C
ref
40
6
H
6
4 2
H ].
H
H
52.103 ( 0.001
0
2
2
reaction C
I[C products]/I[C
intensities I[C ]/I[C
4
H
2
* + C
7
H
8
. Plotted are the ratio of product ion intensities
C H
135.1 ( 0.7
54.35 ( 0.19
35.0 ( 0.1
-17.8 ( 0.1
44.32 ( 0.21
111 ( 2
19.7 ( 0.2
12.0 ( 0.1
73.27 ( 0.41
71 ( 3
133 ( 3
126 ( 3
40
41
40
42
43
44
40
45
46
47
47
47
2
7
H
8
4
H
2
products] versus the ratio of reactant ion
]. The slope of the line fit to these points gives
C H
2
2
H
7 8
4
H
2
CH3
CH4
the relative rate constant for the reaction kC7H8/kC4H2 ) 1.79.
C
3
C
4
C
6
C
7
C
8
H
H
H
H
H
4
2
6
8
6
(propyne)
(diacetylene)
(benzene)
(toluene)
(phenylacetylene)
minor C10D6H and C11D8H products from the C4H2* + C6D6
and C4H2* + C7D8 reactions, respectively. These two products
are increased in yield significantly relative to their non-
deuterated analogues. We will return to this point following a
consideration of the energetics of the various product channels
in section C.
C H (o-ethynyltoluene)
9
8
H
H
C
C
10
6
(phenyldiacetylene)
11
8
(o-diethynyltoluene)
In order to extract quantitatively accurate quantum yields for
the products from the data in Table 1, one would need to correct
the observed ion signal for differences in absolute photoion-
ization efficiencies at 118 nm. However, such data are not
available, so we instead rely on the fact that the observed
photoproducts are chemically similar in structure (e.g., pheny-
lacetylene and phenyldiacetylene) and therefore are expected
to have similar photoionization cross sections. On this basis,
we assume that the photoion signals in Table 1 faithfully reflect
the neutral product yields, recognizing the need for a quantitative
determination in the future.
B. Relative Rate Constants. The metastable diacetylene
reactions are carried out under early-time conditions in which
secondary reactions are negligible. As we have shown in
previous work, under similar conditions the results of a
concentration study can be used to extract an effective rate
constant for the reaction of interest relative to that for C4H2* +
C4H2, which occurs in parallel.22 Within this pseudo-first-order
kinetic model, which assumes similar photoionization cross
sections for all species, a plot of the integrated intensity ratio
of the product ion intensities from the C4H2* + C6H6 (C7H8)
reaction to those from the C4H2* + C4H2 reaction versus the
integrated intensity ratio of the benzene (toluene) to diacetylene
primary ion signals should produce a linear plot with a slope
equal to the rate constant ratio.
TABLE 3: Heats of Reaction for the Various Photoproduct
Reactions
∆
H
rxn (kcal/mol)
at 298 °C
reactiona,b
C
4
C
4
C
4
C
4
C
4
C
4
H
2
H
2
H
2
H
2
H
2
H
2
+ hν(231.5 nm) f C
4
H
2
*
123.5
+ C
+ C
+ C
+ C
+ C
6
6
6
6
6
H
6
H
6
H
6
H
6
H
6
f C
f C
8
H
H
6
+ (C
+ (C
2
H + H)
2
130 ( 2
8
6
H
2
)
-3 ( 2
2 ( 6
f C10
f C10
f C10
H
6
H
6
H
7
+ (H
+ (2H)
+ (H)
2
)
106 ( 6
54 ( 6
C
C
C
C
4
4
4
4
H
H
H
H
2
2
2
2
+ C
+ C
+ C
+ C
7
7
7
7
H
H
H
H
8
8
8
8
f C
f C
f C
8
9
9
H
6
+ C
3
H
4
-5 ( 2
135 ( 7
2 ( 6
97 ( 6
-8 ( 6
3 ( 6
107 ( 6
55 ( 6
H
H
8
+ (C
+ (C
2
H + H)
2
8
2
H )
f C10
H
6
+ CH
3
+ (H)
4
C H + C H f C H + (CH )
4
4
4
4
2
2
2
2
7
7
7
7
8
8
8
8
10
6
8
8
9
C
C
C
H
H
H
+ C
+ C
+ C
H
H
H
f C11
f C11
f C11
H
H
H
2
+ (H )
+ (2H)
+ (H)
a
The products listed in parentheses are not directly detected in the
present work. b The product structures are those indicated in Table 2.
quite similar. These rate constants are also appreciably larger
than those previously obtained in studies of C4H2* reactions
20-23
with small, aliphatic hydrocarbons.
Therefore, it is reason-
able to assume that the increased collisional cross section of
the larger species is more likely to be responsible for the
increased rate rather than any reactant specific dynamic effect.
Figure 7a and b shows such plots for the C4H2* + C6H6 and
C4H2* + C7H8 reactions. The linearity of both plots is consistent
with the simple kinetic model used. The rate constant ratios
obtained are kC H /kC H ) 1.67 ( 0.07 and kC H /kC H ) 1.79
C. Energetics of the Reactions. Figure 8 is a schematic of
the energetics for the diacetylene + benzene and diacetylene
+ toluene reactions. The heats of formation for the various
photoproducts are listed in Table 2, while Table 3 gives the
numerical values for the heats of reaction relative to the two
ground state reactant molecules. Diacetylene is excited to the
6
6
4
2
7
8
4 2
(
0.04, where the error bars represent 1 standard deviation of
the mean. As expected, the relative rate constants for the reaction
of metastable diacetylene with the two aromatics studied are

10318 J. Phys. Chem. A, Vol. 104, No. 45, 2000
Robinson et al.
Figure 9. Proposed mechanism for the formation of phenylacetylene, phenyldiacetylene, and C10
shows the details for the bond rearrangement of 3 to give the major reaction products. The triplet/singlet conversion may happen during the elimination,
H
7
radical species from C
4
H
2
6 6
* + C H . The inset
or triplet products may be formed (see text). The present experiment cannot distinguish at what point in the reaction the crossing occurs.
1
0
1
0
1
2
6 level of the ∆u state providing 123.5 kcal/mol of energy
In spite of the degree of exothermicity of the reaction, the
R2PI spectra of the photoproducts show very little internal
energy. The spectra, particularly of the phenylacetylene and the
o-ethynyltoluene, mimic the jet-cooled known sample spectra.
It appears that the cooling which accompanies the expansion
of the reaction mixture as it leaves the reaction tube is sufficient
to remove most of the internal energy in the nascent products.
This cooling simplifies the R2PI spectra of the products, but
eliminates the possibility of learning anything about the reaction
mechanism from the internal excitation observed in the products.
Previously, the rotational energy of the benzene photoproduct
in the C4H2* + C4H6 reaction was estimated from the rotational
band contour to be in the 50-100 K temperature range.23
D. Reaction Mechanisms. There are three key experimental
findings that a mechanism must justify to explain the reaction
of metastable diacetylene plus benzene or toluene within a
similar framework. First, the major photoproducts in both the
benzene reaction (C8H6 and C10H6) and in the toluene reaction
(C9H8 and C11H8) were formed with similar propensities,
suggesting that they arise from a common intermediate. Second,
the loss of C2H2 or C2HD must happen in a concerted fashion
from the reaction complex. An analogous concerted loss of H2
or HD seems plausible as well. The deuterium labeling
experiments prove that one hydrogen must come from the
diacetylene and the other from the benzene or toluene molecule.
Finally, the mechanism must account for the observed distribu-
tion of o-, m-, and p-ethynyltoluene isomers formed in the C4H2*
+ C7H8 reaction.
above the ground state reactants, but the precise energy available
for reaction depends on the degree of vibrational deactivation
in the triplet manifold.
Due to the large C-H bond strength of C2H2, the C8H6 and
C9H8 products can only be formed in conjunction with intact
C2H2. On the other hand, C10H6 and C11H8 products can be
formed with loss of either H2 of 2H, depending on the degree
of vibrational deactivation of C4H2*. On the basis of the
energetics of the two product channels, one would anticipate
that loss of H2 is preferred, but the present experiment cannot
distinguish between those two pathways.
Since the minor products C10H7 and C11H9 are observed, some
of the reaction complexes do lose a single hydrogen atom under
the conditions of our experiment. If the C10H7 or C11H9 products
have enough internal energy to fragment on a time scale fast
by comparison to its collisional stabilization, they could then
lose a second hydrogen to form the C10H6 or C11H8 product. In
a previous study of the C4H2* + C4H2 reaction, the ratio of
intensities of the C8H2 to C8H3 products (involving loss of H2/
2
H or H, respectively) shifted toward C8H3 when the buffer
2
0
gas used in the reaction was nitrogen as opposed to helium.
The nitrogen buffer serves as a more effective collisional
deactivator than helium. If a similar competition prevails in the
present case, it would provide a potential explanation for the
anomalous increases in C10D6H and C11D8H products from the
C4H2* + C6D6 and C4H2* + C7D8 reactions noted previously.
This increase could indicate that the time scale for further
fragmentation is slower when a deuterium atom on the aromatic
is lost from the reaction complex, and therefore deactivation
occurs before the second atom is ejected.
Finally, it should be noted that two of the minor channels
from the diacetylene plus toluene reaction, namely the C8H6 +
C3H4 and C10H6 + CH4 product channels, have similar heats
of reaction to the C9H8 + C2H2 and C11H8 + H2 product
channels, yet they clearly are not formed in comparable
quantities. We ascribe this difference to mechanistic effects as
will be discussed in the next section.
A mechanism that meets the above criteria is shown for the
two reactions in Figures 9 and 10. The mechanisms presented
use C4D2* rather than C4H2* as the reactant in order to
distinguish the benzene and diacetylene hydrogens, and the
triplet state diacetylene is represented as a cumulene diradical,
3
7
38
as the calculations of Karpfen and Lischka and Vila et al.
have suggested. For the C4D2* + C6H6 reaction, initial attack
of the cumulene diradical on a double bond of benzene produces
1. Subsequently, the ortho radical center attacks the double bond
in the carbon chain, resulting in ring closure and the triplet,

Ultraviolet Photochemistry of Diacetylene
J. Phys. Chem. A, Vol. 104, No. 45, 2000 10319
Figure 10. Proposed mechanism for the formation of o-ethynyltoluene, o-diethynyltoluene, phenylacetylene, phenyldiacetylene, and C11
9
H radical
species from C * + C
4
H
2
7 8
H .
bicyclic carbene 2. A 1,2-deuterium shift then produces the key
triplet diradical intermediate 3. This intermediate can undergo
a six-electron elimination to generate phenylacetylene (4) or
phenyldiacetylene (5) with equal likelihood, or it can eliminate
a single hydrogen to produce the C10H7 radical (6). We favor a
mechanism that proceeds on a triplet surface through to
intermediate 3 for two reasons. First, the 1,2-deuterium shift
may proceed more readily in the triplet carbene than on a singlet
manifold. Second, the bent triplet state of 3 brings the ring
hydrogen close to the end of the carbon chain for the elimination
step. The inset of Figure 9 shows in detail the ene rearrangement
that 3 undergoes to generate the major products, where R1 and
R2 can represent either a hydrogen or an acetylene unit. It should
be noted that some triplet state products could be formed in the
elimination step, particularly triplet phenylacetylene. However,
the detection of ground state phenylacetylene in R2PI spectros-
copy indicates either that a spin crossing occurs during the
elimination step or that the product undergoes intersystem
crossing during the expansion.
An analogous mechanism accounts for the observed toluene
products as well, with additional consideration of the possible
sites for the cumulene radical attack. The attack on the ortho
position is shown in Figure 10, as it produces the major fraction
of the products. Following initial attack, the C4H2* + C7H8
mechanism proceeds via the same route as the benzene
mechanism to the key intermediate 7, which can undergo
elimination via one of several pathways leading to o-ethynyl-
toluene (8), o-diethynyltoluene (9), phenylacetylene (10), phen-
yldiacetylene (11), or the C11H9 radical species (12), all of which
are observed experimentally.
for formation of either para- or meta-substituted products. Note
that para attack cannot lead to the formation of the C8H6 and
C10H6 products, as the methyl group is too far removed to be
involved in a concerted elimination reaction to form propyne
or methane.
The proposed mechanism thus accounts for the distribution
of o- to p-ethynyltoluene observed in the photoproduct spectrum.
There are three attack sites, two ortho and one ipso, all of which
produce ortho-substituted products, while there is only one para
attack site, which produces meta- and para-substituted isomers
in equal quantities. These statistics suggest that the photoprod-
ucts would be formed with a 6:1:1 ratio of ortho:para:meta
substitution. The 4.9((1.8):1 ratio of o- to p-ethynyltoluene
observed in the photoproduct spectrum is consistent with this
effect, though other factors may also contribute to the observed
product ratio. The mechanism predicts equal populations of para
and meta isomers in the photoproducts. However, the meta
origin is spread out over several torsional bands, resulting in
weak meta transitions in the photoproduct spectrum, while the
para isomer origin is a single band (see Figure 5).
The reaction mechanism in Figures 9 and 10 hinges on the
unique properties of reaction intermediates 3 and 7. Estimates
of the heats of formation of the triplet state intermediates 3 and
7 (∼50 kcal/mol above ground state reactants) place them at
accessible energies in the C4H2* + benzene (toluene) reactions.39
These intermediates provide a common framework for concerted
loss of C2HD and HD to form the two main products of each
of the reactions with their correct isotopic compositions. The
similar propensities of these products arises naturally from the
cyclopropyl bond broken in 3 or 7 during the elimination of
C2HD and HD to form products. The mechanism also explains
the observed ratio for the o- to p-ethynyltoluene products.
Finally, intermediates 3 and 7 can account in a similar way for
the formation of even minor products, particularly the formation
of the C8H6 and C10H6 products from the toluene reaction by
loss of CH4 and C3H4, respectively, from intermediate 7. These
species may show up only as minor products due to a greater
activation barrier for the elimination of a methyl, as opposed
to a hydrogen, from the reaction complex.
The position of attack determines the distribution of isomers
formed for both the m/z ) 116 and m/z ) 140 photoproducts.
Methyl substitution on the ring favors cumulene diradical attack
at ortho and para sites. An ipso attack would also produce a
resonantly stabilized radical intermediate. Both ortho and ipso
attacks would give intermediate 7 leading to ortho-substituted
products. A para attack would produce an intermediate with
the cyclopropane ring bridging the para and meta sites. One
would expect this intermediate to eliminate with equal proclivity

10320 J. Phys. Chem. A, Vol. 104, No. 45, 2000
Robinson et al.
E. Implications for Flame Chemistry. The importance of
(21) Frost, R. K.; Zavarin, G.; Zwier, T. S. J. Phys. Chem. 1995, 99,
408.
9
metastable C4H2 reactions in flame chemistry is not yet
established. Spectroscopic signatures for C4H2* are needed
before measurements of its concentration in flames can be
carried out. Furthermore, absolute rate constants for its electronic
quenching and chemical reaction are not yet available. Never-
theless, it is interesting at this early stage to note the types of
products that are formed in these reactions in the context of
flame chemistry. The reaction of metastable diacetylene with
benzene forms phenylacetylene and phenyldiacetylene products
that are similar in kind to those produced by the HACA
mechanism. Yet, in bypassing free-radical formation, they
constitute reaction pathways with unusually low energetic
thresholds, which thereby deserve further investigation. Ad-
ditionally, the preferential formation of o-ethnyltoluene in the
reaction of metastable diacetylene with toluene is interesting
in light of the heightened efficiency with which ortho-substituted
aromatic rings form naphthalene in flames.8
(
22) Frost, R. K.; Arrington, C. A.; Ramos, C.; Zwier, T. S. J. Am. Chem.
Soc. 1996, 118, 4451.
(23) Arrington, C. A.; Ramos, C.; Robinson, A. D.; Zwier, T. S. J. Phys.
Chem. A 1998, 102, 3315.
(
24) Allan, M. J. Chem. Phys. 1984, 80, 6020.
(25) Hagemeister, F. C.; Arrington, C. A.; Giles, B. J.; Quimpo, B.;
Zhang, L.; Zwier, T. S. Cavity Ringdown methods for studying intramo-
lecular and intermolecular dynamics. In CaVity-Ringdown Spectroscopys
A New Technique for Trace Absorption Measurements; Busch, K. J., Busch,
M. A., Eds.; American Chemical Society: Washington, DC, 1999.
(26) Arrington, C. A.; Ramos, C.; Robinson, A. D.; Zwier, T. S. J. Phys.
Chem. A 1999, 103, 1294.
(27) Mahon, R.; McIlrath, T. J.; Myerscough, V. P.; Koopman, D. W.
IEEE J. Quant. Electron. 1979, 6, 444.
(
28) Brandsma, L. PreparatiVe Acetylenic Chemistry; Elsevier Publishing
Co.: New York, 1988.
(29) Eastmond, R.; Walton, D. R. M. Tetrahedron 1972, 28, 4591.
(30) The synthetic procedure started with o-methylstyrene which was
brominated to give the dibromo compound. Base-catalyzed elimination of
Br2 yielded the desired o-ethynyltoluene. The procedure was verified by
the positive identification of p-ethynyltoluene synthesized from p-methyl-
styrene. The synthesis was performed by Stephen L. Finson at Hampford
Research, Inc.
Acknowledgment. The authors gratefully acknowledge the
Department of Energy, Office of Basic Energy Sciences,
Division of Chemical Sciences, for their support of this research
under Grant No. DE-FG02-96ER14656. A.G.R. is grateful for
fellowships from Purdue Research Foundation and Lubrizol
Corp. The authors also acknowledge Professors John Grutzner
and Rudolph Winter for the many useful discussions on reaction
mechanisms. Finally, the authors are indebted to Stephen Finson
of Hampford Research, Inc. for providing us with the o-
ethynyltoluene sample, and to Professor Jillian Buriak and
Michael Stewart of Purdue University for synthesizing the
phenyldiacetylene.
(
31) Arps, J. H.; Chen, C. H.; McCann, M. P.; Datskou, I. Appl.
Spectrosc. 1989, 43, 1211.
32) Under certain experimental conditions, a photoproduct ion signal
(
is also observed around m/z ) 91 (C7H7, not shown in Figure 1). The
photoion signal remains with full intensity when xenon is removed from
the tripling cell, indicating that this product is ionized exclusively by the
3
55 nm light (∼10 mJ/pulse) that also traverses the ion source region along
with the 118 nm light. No ion signal was observed at the calibrated arrival
time for m/z ) 91 arising from VUV light, indicating that it is an
insignificant product. The mass 91 signal has a greater than first-order power
dependence with 355 nm light, suggesting that the product appears only by
virtue of its low two-photon ionization threshold. The C H species is likely
7
7
to be tropyl radical, which has an ionization potential (6.28 eV) [Koenig,
T.; Chang, J. C. J. Am. Chem. Soc. 1978, 100, 2240] accessible by two-
photon, 355 nm ionization (6.985 eV) , as opposed to benzyl radical (IP )
7
9
.2487 eV) [Berkowitz, J.; Ellison, G. B.; Gutman, D. J. Phys. Chem. 1994,
References and Notes
8, 2744].
(
1) Bastin, E.; Delfau, J.-L.; Reuillon, M.; Vovelle, C.; Warnatz, J.
Symp. (Int.) Combust. [Proc.], 22nd 1988.
2) Lam, F. W.; Howard, J. B.; Longwell, J. P. Symp. (Int.) Combust.
Proc.], 22nd 1988, 323.
3) Melius, C. F.; Miller, J. A.; Evleth, E. M. Symp. (Int.) Combust.
Proc.], 24th 1992, 621.
(
33) Bandy, R. E.; Lakshminarayan, C.; Zwier, T. S. J. Phys. Chem.
992, 96, 5337.
34) Noyes, W. A., Jr.; Mulac, W. A.; Harter, D. A. J. Chem. Phys.
966, 44, 2100.
1
1
(
(
[
[
(
(
(
39.
(
(
(
35) Cundall, R. B.; Davies, A. S. Trans. Faraday Soc. 1966, 62, 1151.
36) Kendler, S.; Zilberg, S.; Haas, Y. Chem. Phys. Lett. 1995, 242,
4) Miller, J. A.; Melius, C. F. Combust. Flame 1992, 91, 21.
1
37) Karpfen, A.; Lischka, H. Chem. Phys. 1986, 102, 91.
2
57.
38) Vila, F.; Borowski, P.; Jordan, K. D., submitted for publication
39) Benson, S. Thermochemical Kinetics, 2nd ed.; John Wiley &
(
(
(
6) McEnally, C. S.; Pfefferle, L. D. Combust. Flame 1998, 115, 81.
7) Leung, K. M.; Lindstedt, R. P. Combust. Flame 1995, 102, 129.
8) Marinov, N. M.; Pitz, W. J.; Westbrook, C. K.; Castaldi, M. J.;
Sons: New York, 1976. Heats of formation were estimated using Benson’s
method of group additivity for the singlet ground state species. A factor of
Senkan, S. M. Combust. Sci. Technol. 1996, 116, 211.
9) Westmoreland, P. R.; Dean, A. M.; Howard, J. B.; Longwell, J. P.
J. Phys. Chem. 1989, 92, 8171.
2
eV was then added to account for the singlet-triplet splitting.
40) Berkowitz, J.; Ellison, G. B.; Gutman, D. J. Phys. Chem. 1994,
8, 2744.
41) Gurvich, L. V.; Veyts, I. V.; Alcock, C. B.; Iorish, V. S.
(
(
9
10) Lindstedt, R. P.; Skevis, G. Combust. Sci. Technol. 1997, 125, 73.
(
Thermodynamic Properties of IndiVidual Substances, 4th ed.; Hemisphere:
New York, 1991; Vol. 2.
(42) Pittam, D. A.; Pilcher, G. J. Chem. Soc., Faraday Trans. 1 1972,
68, 2224.
(43) Wagman, D. D.; Kilpatrick, J. E.; Pitzer, K. S.; Rossini, F. D. J.
Combust. Flame, in press.
(
12) Wang, H.; Frenklach, M. J. Phys. Chem. 1994, 98, 11465.
(13) Melius, C. F.; Colvin, M. E.; Marinov, N. M.; Pitz, W. J.; Senkan,
S. M. Symp. (Int.) Combust. [Proc.], 26th 1996, 685.
14) Castaldi, M. J.; Marinov, N. M.; Melius, C. F.; Huang, J.; Senkan,
S. M.; Pitz, W. J.; Westbrook, C. K. Symp. (Int.) Combust. [Proc.], 26th
(
Res. Natl. Bur. Stand. 1945, 35, 467.
1
996, 693.
(44) Stein, S. E.; Fahr, A. J. Phys. Chem. 1985, 89, 3714.
(45) Pedley, J. B.; Naylor, R. D.; Kirby, S. P. Thermochemical Data of
Organic Compounds, 2nd ed.; Chapman & Hall, Ltd.: London, 1986.
(46) Davis, H. E.; Allinger, N. L.; Rogers, D. W. J. Org. Chem. 1985,
50, 3601.
(47) Benson, S. W. Thermochemical Kinetics, 2nd ed.; John Wiley &
Sons: New York, 1976. Heats of formation for compounds for which no
literature value could be found were estimated using Benson’s method of
group additivity. Benson provides no group values for Ct-(Ct) or C-(CB)-
(H)3, so the groups Ct-(C) and C-(C)(H)3 were used instead. The
uncertainties listed represent estimates.
(15) Hausmann, M.; Homann, K.-H. Ber. Bunsen-Ges. Phys. Chem.
1
997, 101, 651.
(16) Bapat, J. B.; Brown, R. F. C.; Bulmer, G. H.; Childs, T.; Coulston,
K. J.; Eastwood, F. W.; Taylor, D. K. Aust. J. Chem. 1997, 50, 1159.
(
(
(
17) Necula, A.; Scott, L. T. J. Am. Chem. Soc. 2000, 122, 1548.
18) Glicker, S.; Okabe, H. J. Phys. Chem. 1987, 91, 437.
19) Bandy, R. E.; Lakshminarayan, C.; Frost, R. K.; Zwier, T. S. Science
1
992, 258, 1630.
20) Bandy, R. E.; Lakshminarayan, C.; Frost, R. K.; Zwier, T. S. J.
Chem. Phys. 1993, 98, 5362.
(