Crossed-molecular-beam study on the formation of phenylacetylene from phenyl radicals and acetylene

Article abstract of DOI:10.1002/anie.200701890

(Figure Presented) A hot topic! Crossed-molecular-beam experiments showed that phenylacetylene molecules - precursors to polycyclic aromatic hydrocarbons in combustion flames and in carbon-rich planetary nebulae - can be formed through reactions of phenyl radicals with acetylene under single-collision conditions (see picture).

Full text of DOI:10.1002/anie.200701890

Communications
DOI: 10.1002/anie.200701890
Combustion Chemistry
Crossed-Molecular-Beam Study on the Formation of Phenylacetylene
from Phenyl Radicals and Acetylene**
Xibin Gu,* Fangtong Zhang,* Ying Guo,* and Ralf I. Kaiser*
During the last decades, polycyclic aromatic hydrocarbons
(PAHs)[1] and other related aromatic compounds, such as
ionized PAHs,[2] have received considerable attention from
astronomers,[3] astrobiologists,[4] environmentalists,[5] and the
combustion community.[6] In the interstellar medium, PAH-
like species account for up to 10% of the cosmic carbon
budget,[7] have been suggested as carriers ofboth the
unidentified infrared (UIR) emission bands[8] and the diffuse
interstellar bands (DIBs),[9] and are also associated with the
origin oflife. On Earth, however, PAHs are severe pollutants
and are considered as airborne toxic chemicals because of
their mutagenic and carcinogenic character.[10] Considering a
yearly emission rate of1.6 million tons in combustion
processes,[5] PAHs and soot also contribute significantly to
global warming.[11] A quantitative understanding ofthe
formation of PAHs is therefore essential to develop clean
combustion devices and understand the astrochemical and
astrobiological evolution ofthe interstellar medium.
phenyl–acetylene reaction as the trigger to PAH formation,
the theoretical investigations have never been verified
experimentally under single-collision conditions. Conse-
quently, the nature ofthe true reaction products ofthis
elementary reaction has remained elusive to date. To shed
light on this fundamental question, we conducted a crossed-
molecular-beam study ofthe gas-phase reaction ofthe phenyl
radical with acetylene to synthesize phenylacetylene plus
atomic hydrogen via [C8H7]* intermediate(s).
This system presents the prototype reaction ofan
aromatic radical[17] with ubiquitous acetylene molecules to
form phenylacetylene by means of a single radical–neutral
collision. Benzene, a likely precursor ofthe phenyl radical,
has been observed together with acetylene, for example,
toward the carbon-rich planetary nebula CRL 618,[18] where it
might be photodissociated to form the phenyl radical. On the
other hand, the phenyl radical might result from benzene in
combustion flames through hydrogen abstraction process-
es.[19] The reaction is also interesting from the viewpoint of a
physical–organic chemist as it represents a benchmark system
to unravel the chemistry ofradical-substitution reactions
initiated by polyatomic organic radicals.
Despite the crucial importance ofPAHs, even the
formation route of the very first building block of this type
ofcompounds (namely, the phenylacetylene molecule,
C6H5CCH) is unknown. Chemical reaction networks that
model the formation of PAH in combustion flames[12] and in
the interstellar medium[13] imply that the phenyl radical (that
is, C6H5) represents the most important transient species to
initiate the formation of the very first PAH.[14] The crucial
steps ofthese reaction models are hydrogen abstraction–
acetylene addition sequences. The chemical networks concur
that the reaction ofthe phenyl radical with acetylene initiates
the synthesis ofPAH through the addition ofphenyl to the
acetylene molecule. Because ofthe central role ofthe phenyl–
acetylene reaction, the kinetics ofthis system have been well-
established (covering the temperature range up to 1500 K).[15]
Theoretical studies ofthis benchmark system predicted
the initial formation of a [C8H7]* adduct which either
decomposed back to the reactants or fragmented into the
phenylacetylene molecule (C6H5CCH) plus atomic hydrogen
The reactive-scattering signal was recorded at mass-to-
+
charge ratios (m/z) between m/z 102 (C8H6 ) and m/z 98
+
(C8H2 ) (at distinct laboratory angles). The time-of-flight
(TOF) spectra monitored at m/z 101–98 were superimpos-
able—at all angles (and after scaling)—to those taken at m/z
98 (Figure 1). This represents a crucial finding and suggests
that the phenyl-versus-atomic-hydrogen exchange pathway to
form the C8H6 isomer(s) is open and that the signal at lower
m/z values originates from the dissociative ionization of the
reaction product in the electron-impact ionizer.
We also detected a signal at m/z 103. Since the patterns of
the TOF spectra were identical to those recorded at m/z 102
and the intensity ofthe ion counts at m/z 103 was only about
10% ofthat observed at m/z 102, we concluded that the signal
at m/z 103 originates from the ionized 13C isotopomer of
C8H6, that is, from 13CC7H6, but not from the C8H7 adduct. The
TOF spectra and laboratory angular distributions ofthe
[Eq. (1)].[16] Nevertheless, despite the central role ofthe
X
C6H5ðX2A1Þ þ C2H2ðX1
Þ ! ½C8H7ꢀ
Þ
þ
*
g
ð1Þ
reactive-scattering signal at m/z 102 (namely, ofC H6) are
8
2
! C6H5CCHðX1A1Þ þ Hð S1=2
shown in Figure 1. To pin down whether the hydrogen atom is
lost from the phenyl group or from the acetylenic unit, we
conducted the crossed-beam reaction ofphenyl radicals with
fully deuterated acetylene. In this case, the loss of a hydrogen
atom from the phenyl group should result in a signal at m/z
[*] Dr. X. Gu, Dr. F. Zhang, Dr. Y. Guo, Prof. R. I. Kaiser
Department ofChemistry
University ofHawaii at Manoa
Honolulu, HI 96822 (USA)
Fax: (+1)808-956-5908
E-mail: ralfk@hawaii.edu
+
104 (C8H4D2 ), while the release ofatomic deuterium is
expected to be monitored at m/z 103 (C8H5D+). We observed
signals at m/z 103 and 104. However, both TOF spectra were
superimposable, and the ion counts at m/z 104 were only
about 10% ofthose at m/z 103. This result indicates that
Homepage: http://www.chem.hawaii.edu/Bil301/welcome.html
[**] This work was supported by the US Department ofEnergy, Basic
Energy Sciences (DE-FG02-03ER15411).
6866
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 6866 –6869

Angewandte
Chemie
Figure 1. Representative laboratory angular distribution ofthe C 8H6
product at m/z 102 and a collision energy of96 kJmol ꢁ1. Shown is the
intensity I plotted versus the laboratory angle V. Circles and 1s error
bars indicate experimental data, the solid lines represent the calculated
distributions, and C.M. is the center-of-mass angle. Also shown are
the corresponding TOF data at selected laboratory angles (inset); the
flight time t is plotted against the counts C. The circles represent the
experimental data, and the solid lines show the corresponding fit.
Figure 2. Center-of-mass angular-flux distributions (T, left) and transla-
tional-energy-flux distributions (P; right) for the reaction of phenyl
radicals with acetylene at nominal collision energies of71 (upper row)
and 96 kJmolꢁ1 (lower row); the hatched areas limit the fits obtained
within the error limits (q: center-of-mass angle; Etrans: translational
energy).
show maxima around 25–35 kJmolꢁ1, thus indicating a tight
exit transition state in the decomposition ofthe intermediate.
In other words, the reverse reaction (hydrogen-atom addition
to the C8H6 isomer) is expected to have an entrance barrier of
about 30 kJmolꢁ1. Most importantly, the high-energy cutoffs
ofthe translational-energy distributions resemble the sum of
the absolute value ofthe reaction exoergicity plus the relative
collision energy; in this limit, all ofthe available energy goes
into the translational degrees ofrfeedom ofthe products.
After extraction of the latter and averaging over both
experiments, the data suggest that the reaction is exoergic
by (45 ꢂ 11) kJmolꢁ1. These data are in excellent agreement
with the computed values of(41 ꢂ 8)[20] and (38 ꢂ
5) kJmolꢁ1 [16] to form the phenylacetylene isomer plus
atomic hydrogen. Based on this result, we can extract the
fraction of available energy channeled into the translational
degrees ofrfeedom ofthe reactants to be about (45 ꢂ 5)%.
Finally, the best fits to the TOF data and laboratory angular
within the signal-to-noise limitations ofour experiments, only
the deuterium atom was lost and that the signal at higher m/z
ratio originated—similar to that at m/z 103 in the original
experiment—from an ionized 13C isotopomer (here,
13CC7H5D).
Finally, we investigated to what extent a hydrogen-
abstraction pathway, by which benzene (m/z 78) and ethynyl
are formed, is involved in the dynamics. When we monitored
the signal at m/z 78, we detected a strong interference from
elastically scattered 13CC5H5 radicals, which overlapped with
the expected reactively scattered benzene molecules. To
circumvent this interference, we monitored the deuterium-
abstraction pathway to form C6H5D from phenyl and C2D2.
At m/z 79, no background molecules interfered, but we were
unable to detect any reactive-scattering signal. In summary,
the laboratory data suggest that the phenyl ring is conserved
in the phenyl-radical-versus-atomic-hydrogen/deuterium
exchange reaction and that the sole reaction product(s) is
(are) the C8H6 isomer(s). Accounting for our signal-to-noise
ratio and the dynamics ofthe reaction, the hydrogen/
deuterium-abstraction pathway was found to contribute less
than 5% to the scattering signal. The identification of this
phenyl–hydrogen exchange under single-collision conditions
alone underlines the importance ofthis reaction to build up
C8H6 isomers in interstellar environments and in combustion
settings.
Next we tried to elucidate the reaction energies and gain
additional information on potential entrance and exit barriers
in the phenyl–acetylene reaction by untangling its dynamics.
Most importantly, the dynamics assist in the identification of
the nature ofthe product isomer(s) and the intermediate(s)
involved. The center-of-mass translational energy P(E) and
the angular distribution T(q) for both collision energies help
us to solve this matter; the functions are depicted in Figure 2.
The translational-energy distributions assist in collecting
information on the reaction dynamics. Both distributions
distributions were achieved by adjusting the threshold energy
ꢁ1
ofthe reaction to be in the range of5 to 30 kJmol
.
Also, both center-of-mass angular distributions exhibit
intensity over the complete range (from 0 to 1808). This
finding indicates that the reaction dynamics are indirect and
involve the formation of C8H7 intermediate(s) before the
latter decompose—via atomic-hydrogen elimination—into
the C8H6 product(s). Moreover, the T(q) values show a
pronounced forward scattering, that is, the flux at 08 is
enhanced compared to that at 1808 and, as the collision
energy increases, the distribution becomes more forward-
scattered. These results suggest that the lifetime(s) of the
decomposing reaction intermediate(s) is(are) shorter than the
rotational period. Most importantly, the center-of-mass
angular distributions show an enhanced forward peaking as
the collision energy increases. This energy dependence ofthe
T(q) values infers the existence of an osculating complex,[21]
that is, a C8H7 intermediate whose lifetime decreases as the
collision energy is enhanced.
Angew. Chem. Int. Ed. 2007, 46, 6866 –6869
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 6867

Communications
Our results suggest that the phenyl radical adds with its
formation of naphthalene. In the interstellar medium, how-
ever, only bimolecular reactions take place, and the inter-
mediate always fragments, for example, in planetary nebulae,
to phenylacetylene plus atomic hydrogen.
radical center to the carbon–carbon triple bond, thereby
forming a C8H7 doublet radical intermediate (namely, 2-
phenylvinyl, see Figure 3). The threshold energy indicates
Consequently,
future
infrared spectroscopic surveys
ofplanetary nebulae should
search for infrared features of
the phenylacetylene mole-
cule—a crucial building block
ofcomplex PAHs in the inter-
stellar medium. Since the
reaction between phenyl and
acetylene has an entrance bar-
rier, it should be noted that it
does not contribute to the
formation of PAHs in cold
Figure 3. Schematic representation ofthe reaction mechanism ofphenyl radicals with acetylene (letf)
leading to phenylacetylene plus atomic hydrogen (right) via a 2-phenylvinyl radical intermediate (center).
Carbon and hydrogen atoms are depicted in black and blue, respectively. Note that the 2-phenylvinyl
intermediate can exist in its cis and trans form.
that this process has an entrance barrier between 5 and
30 kJmolꢁ1. The shapes ofthe center-o-fmass angular distri-
butions are indicative ofcomplex-forming “indirect” scatter-
ing dynamics. At both collision energies used, the lifetime of
the decomposing intermediate was much shorter than the
rotational period. The rotational period ofthe 2-phenylvinyl
intermediate acts as a clock in the molecular-beam experi-
ment and can be utilized to estimate the lifetime t ofthe
decomposing complex. A quantitative inspection ofthe T(q)
values (Figure 2) suggests that the lifetime of the intermediate
is less than halfofits rotational period.
molecular clouds, where averaged translational temperatures
ofthe reactants ofonly 10K reside. However, reactants in
planetary nebulae close to the central star can have temper-
atures of up to a few 1000K (which are sufficiently high to
overcome the entrance barrier). We anticipate that this
generalized concept ofa phenyl-radical-versus-atomic-hydro-
gen exchange is a starting point in a systematic investigation
ofphenyl-radical reactions with unsaturated hydrocarbons
under single-collision conditions and a search for hitherto
unobserved reaction intermediates ofthese reactions in space
and in combustion flames.
The short timescale ofthis reaction was also veriifed by
the relatively large fraction of energy that channels into the
translational modes, that is, (45 ꢂ 5)% ofthe total available
energy. The initial collision complex decomposed by atomic-
hydrogen elimination through a tight exit transition state
located 25–35 kJmolꢁ1 above the separated reaction products
(namely, phenylacetylene and atomic hydrogen). This reac-
tion mechanism fully supports a previous theoretical study,[16]
in which the authors concluded that the phenyl radical added
with its unpaired electron to the acetylene molecule through a
barrier ofabout 16 kJmol ꢁ1, leading to an intermediate which
was stabilized by 170 kJmolꢁ1. This shallow potential-energy
well can nicely account for the relatively short timescale of
the reaction (as found experimentally). The fate of this
intermediate was dictated by an atomic-hydrogen loss upon
passing an exit transition state located 28 kJmolꢁ1 above the
products; note that this intermediate could also undergo a cis–
trans isomerization prior to its decomposition.[16] Finally, the
experiments verify the sole existence of the hydrogen-loss
pathway—and the absence ofany hydrogen-abstraction
pathways—to form benzene.
Experimental Section
The reaction was conducted under single-collision conditions, at
collision energies of71 and 96 kJmol ꢁ1, using a crossed-molecular-
beam machine.[22] The supersonic phenyl-radical beam was generated
by flash pyrolysis of helium-seeded nitrosobenzene (Aldrich) at
seeding fractions of less than 0.1%. This mixture was expanded, at a
stagnation pressure of920 Torr, through a resistively heated silicon
carbide tube held at temperatures of1200–1500K. The pulsed valve
was operated at 200 Hz, with pulses that were 150 ms wide (under
these conditions, the decomposition ofnitrosobenzene is quantita-
tive). A chopper wheel, mounted after the skimmer, selected a
component ofthe phenyl-radical beam beofre it crossed a pulsed
acetylene beam at a right angle in the interaction region.[23] To extract
the position ofthe hydrogen-atom loss, we also perofrmed experi-
ments with [D2]acetylene. TOF spectra and laboratory angular
distributions ofthe reactively scattered products were probed
utilizing a quadrupole mass spectrometer with an electron-impact
ionizer. Information on the reaction dynamics was gained by fitting
the laboratory data using a forward-convolution routine, thereby
yielding the angular-flux distribution T(q) and the translational-
energy-flux distribution P(E) in the center-of-mass system.[24]
In summary, we have identified the phenylacetylene
molecule as the sole product in the radical–neutral reaction
ofphenyl radicals with acetylene molecules under single-
collision conditions. The proposed chemical dynamics infer
the existence ofa reaction intermediate, which decomposed
in the crossed-beam experiments through the loss ofatomic
hydrogen. However, in low-temperature and high-pressure
combustion flames, this intermediate may either be stabilized
or react with another acetylene molecule ifthe time between
collisions is short enough; this can ultimately lead to the
Received: April 28, 2007
Published online: July 19, 2007
Keywords: combustion chemistry · polycycles · radicals ·
.
reaction mechanisms · reactive intermediates
[1] J. L. Weisman, A. Mittioda, T. J. Lee, D. M. Hudgins, L. J.
Allamandola, C. W. Bauschlicher, M. Head-Gordon, Phys.
Chem. Chem. Phys. 2005, 7, 109.
6868 www.angewandte.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 6866 –6869

Angewandte
Chemie
[2] D. M. Hudgins, C. W. Bauschlicher, L. J. Allamandola, J. C.
Fetzer, J. Phys. Chem. A 2000, 104, 3655.
[3] F. Salama, L. J. Allamandola, Nature 1992, 358, 42.
[4] P. Ehrenfreund, Science 1999, 283, 1123.
[5] J. H. Seinfeld, J. F. Pankow, Annu. Rev. Phys. Chem. 2003, 54,
121.
[6] A. Kazakov, M. Frenklach, Combust. Flame 1998, 112, 270.
[7] Y. J. Pendleton, L. J. Allamandola, Astrophys. J. Suppl. Ser. 2002,
138, 75.
[14] R. J. McMahon, M. C. McCarhy, C. A. Gottlieb, J. B. Dudek, J. F.
Stanton, P. Thaddeus, Astrophys. J. 2003, 590, L61.
[15] H. Richter, O. A. Mazyar, R. Sumathi, W. H. Green, J. B.
Howard, J. W. Bozzelli, J. Phys. Chem. A 2001, 105, 1561.
[16] I. V. Tokmakov, M. C. Lin, J. Am. Chem. Soc. 2003, 125, 11397.
[17] P. von R. Schleyer, Chem. Rev. 2001, 101, 1115.
[18] J. Cernicharo, Astrophys. J. 2001, 546, L123.
[19] H. Böhm, H. Jander, Phys. Chem. Chem. Phys. 1999, 1, 3775.
[20] http://webbook.nist.gov.
[8] E. Peeters, A. Mattioda, D. M. Hudgins, L. J. Allamandola,
Astrophys. J. 2004, 617, L65.
[21] W. B. Miller, S. A. Safron, D. R. Herschbach, Discuss. Faraday
Soc. 1967, 44, 108.
[9] W. W. Duley, Faraday Discuss. 2006, 133, 415.
[10] B. J. Finlayson-Pitts, J. N. Pitts, Science 1997 276, 1045; K.
Hylland, J. Toxicol. Environ. Health Part A 2006 69, 109.
[11] A. Violi, A. Venkatnathan, J. Chem. Phys. 2006, 125, 054302.
[12] J. Appel, H. Bockhron, M. Frenklach, Combust. Flame 2000, 121,
122; H. Richter, J. B. Howard, Phys. Chem. Chem. Phys. 2002, 4,
2038.
[22] X. Gu, Y. Guo, F. Zhang, A. M. Mebel, R. I. Kaiser, Faraday
Discuss. 2006, 133, 245.
[23] The peak velocities ofthe phenyl-radical, acetylene, and
[D2]acetylene beam were determined to be 2541 (low collision
energy), 3047 (high collision energy), 925, and 920 msꢁ1
.
[24] R. I. Kaiser, C. Ochsenfeld, D. Stranges, M. Head-Gordon, Y. T.
Lee, Faraday Discuss. 1998, 109, 183.
[13] M. Frenklach, E. D. Feigelson, Astrophys. J. 1989, 341, 372.
Angew. Chem. Int. Ed. 2007, 46, 6866 –6869
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 6869