Dimerization of the Benzyl Radical in a High-Temperature Pyrolysis Reactor Investigated by IR/UV Ion Dip Spectroscopy

Article abstract of DOI:10.1002/chem.201800852

We investigate the self-reaction of benzyl, C₇H₇, in a high-temperature pyrolysis reactor. The work is motivated by the observation that resonance-stabilized benzyl radicals can accumulate in reactive environments and contribute to the formation of polycyclic aromatic hydrocarbons (PAHs) and soot. Reaction products are detected by IR/UV ion dip spectroscopy, using infrared radiation from the free electron laser FELIX, and are identified by comparison with computed spectra. Among the reaction products identified by their IR absorption are several PAHs linked to toluene combustion such as bibenzyl, phenanthrene, diphenylmethane, and fluorene. The identification of 9,10-dihydrophenanthrene provides evidence for a mechanism of phenanthrene formation from bibenzyl that proceeds by initial cyclization rather than an initial hydrogen loss to stilbene.

Full text of DOI:10.1002/chem.201800852

DOI: 10.1002/chem.201800852
Full Paper
&
High-Temperature Pyrolysis |Hot Paper|
Dimerization of the Benzyl Radical in a High-Temperature
Pyrolysis Reactor Investigated by IR/UV Ion Dip Spectroscopy
Florian Hirsch,[a] Philipp Constantinidis,[a] Ingo Fischer,*[a] Sjors Bakels,[b] and Anouk M. Rijs*[b]
Abstract: We investigate the self-reaction of benzyl, C7H7, in
a high-temperature pyrolysis reactor. The work is motivated
by the observation that resonance-stabilized benzyl radicals
can accumulate in reactive environments and contribute to
the formation of polycyclic aromatic hydrocarbons (PAHs)
and soot. Reaction products are detected by IR/UV ion dip
spectroscopy, using infrared radiation from the free electron
laser FELIX, and are identified by comparison with computed
spectra. Among the reaction products identified by their IR
absorption are several PAHs linked to toluene combustion
such as bibenzyl, phenanthrene, diphenylmethane, and fluo-
rene. The identification of 9,10-dihydrophenanthrene pro-
vides evidence for a mechanism of phenanthrene formation
from bibenzyl that proceeds by initial cyclization rather than
an initial hydrogen loss to stilbene.
Introduction
Benzyl, C7H7, is the prototype of a resonance-stabilized radical
with a p-electron system.[1] Owing to its low reactivity, a com-
paratively long lifetime in a reactive environment can be ex-
pected and knowledge about the high-temperature reactions
of benzyl is considered relevant for understanding its role in
combustion chemistry. In combustion, the benzyl radical is
seen as the primary unimolecular reaction product of toluene,
a common fuel additive. Indeed, recent flame experiments sug-
gested benzyl to be a major precursor of various aromatic hy-
drocarbons.[2]
Figure 1. Thermal decomposition of the precursor 1 by pyrolysis is em-
ployed to generate 2.
sition of 2.[10] They observed C7H6, C7H5, C5H5, and acetylene as
unimolecular decomposition products.
When operated under suitable conditions (i.e., backing gas
pressure, temperature, precursor concentration), a pyrolysis re-
actor also provides an environment to study bimolecular reac-
tions of radicals at high temperature and to detect polycyclic
aromatic hydrocarbons (PAHs) that are efficiently formed. Such
reactions have initially been observed in photoionization mass
spectrometry (PIMS) experiments with tunable synchrotron ra-
diation, where PAH molecules are identified through their ioni-
zation energy (IE). For example, mixtures of phenyl[11] and
naphthyl[12] radicals with acetylene were studied with the aim
to check the validity of the HACA (hydrogen abstraction C2 ad-
dition) mechanism for PAH and soot formation. A better iden-
tification of products isomers is achieved, when ions and pho-
toelectrons are detected in coincidence.[13] Alternatively, we re-
cently showed that IR/UV ion dip spectroscopy[14] in a super-
sonically cooled free jet is also a successful tool to study high-
temperature reactions in a flow reactor, because it combines
the structural sensitivity of infrared spectroscopy with mass in-
formation. Although not applicable to small molecules without
a UV chromophore and thus less general than photoionization,
isomer-selective identification of aromatic molecules and PAH
is straightforward. The ground-state IR spectra are obtained by
monitoring the depletion in the resonance-enhance ionization
signal (REMPI) that occurs upon vibronic excitation, so the
technique results in mass-selected IR spectra. However, to
High-temperature (pyrolysis) flow microreactors are an effi-
cient tool to generate hydrocarbon radicals and study them
under collision-free conditions in a free jet by laser spectrosco-
py.[3] Pyrolytically generated benzyl radicals 2 (Figure 1) have
been studied by time-resolved spectroscopy,[4] threshold pho-
toelectron spectroscopy,[5] and photofragment spectroscopy.[6]
Ellison and co-workers demonstrated that pyrolysis reactors
are also a suitable tool to study the high-temperature chemis-
try of radicals and analyzed the flow dynamics in such a reac-
tor in detail.[7] They focused on the decomposition of organic
molecules, like furan[8] and anisole[9] at high temperature inside
the reactor and recently extended their work to the decompo-
[a] F. Hirsch, P. Constantinidis, Prof. Dr. I. Fischer
Institute of Physical and Theoretical Chemistry, University of Wꢀrzburg
Am Hubland Sꢀd, 97074 Wꢀrzburg (Germany)
E-mail: ingo.fischer@uni-wuerzburg.de
[b] S. Bakels, Dr. A. M. Rijs
Radboud University, Institute for Molecules and Materials
FELIX Laboratory, Toernooiveld 7-c, 6525 ED Nijmegen (The Netherlands)
E-mail: a.rijs@science.ru.nl
Supporting information and the ORCID identification number for the
author of this article can be found under:
https://doi.org/10.1002/chem.201800852.
Chem. Eur. J. 2018, 24, 1 – 7
1
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
excite molecules in a free jet in the mid-IR fingerprint region,
the high photon flux of a free electron laser (FEL) is required.
In previous experiments, we investigated bimolecular reactions
of propargyl,[15] phenyl,[16] and phenylpropargyl radicals[17] and
observed formation of PAHs with up to three aromatic rings.
Here, we extend our work to the benzyl radical, employing
phenylethyl nitrite 1 as the precursor, see Figure 1. Note that
an IR/UV ion dip spectrum of benzyl itself has already been ob-
served.[18]
mass spectrum at 265 nm, which is the basis of the IR/UV ion
dip experiment. Because of the high sensitivity for aromatic
molecules with a UV chromophore, additional mass signals are
visible, in particular strong signals at m/z=166 and 168. The
peak at m/z=168 corresponds to the addition of a phenyl unit
to benzyl; additional H2 loss leads to m/z=166. A similar reac-
tion path can be observed for the benzyl dimer. Close to the
mass of the dimerization product (m/z=182), additional peaks
are visible two mass units apart at 178 and 180, suggesting
consecutive H2 loss. Interestingly, m/z=106 cannot be ob-
served in the REMPI spectrum. This further supports the assign-
ment to benzaldehyde with IE=9.47 eV, which is too high to
be ionized with two photons at 265 nm (IE=9.36 eV).[19]
Results and Discussion
Mass spectra
The photoionization mass spectra of the pyrolysis products of
1 (m/z=151) are depicted in Figure 2. The upper trace shows
a one-photon ionization mass spectrum at 118 nm. Although
IR spectra of benzyl self-reaction products
In the following section, the IR/UV ion dip spectra will be dis-
cussed and compared with calculated IR spectra. The IR/UV
spectrum of the benzyl radical itself has already been mea-
sured by Satink et al.[18] with the same methodology at
!
305 nm, utilizing the D3 D0 transition. In our study, the UV
wavelength (265 nm) excites the red edge of an intense UV
band, which is dominated by excitation into several short-lived
electronic states.[4b] The excitation wavelength was selected to
efficiently ionize the most important reaction products. In all
spectra presented below, the solid line corresponds to the ex-
perimental spectrum, whereas the dotted line represents a
computed spectrum. In addition to the molecules discussed
here, m/z=104 has been identified as styrene. As it is possibly
also a decomposition product of the alcohol contamination
(m/z=122), it will also not be further discussed.
The peaks at m/z=182–178
The IR action spectra recorded for the peaks at m/z=182, 180,
and 178 are depicted in Figure 3. The largest detectable mole-
cule in our pyrolysis experiment is the dimerization product at
m/z=182. It is the only condensation product with a sufficient
concentration to be detected in the 118 nm mass spectrum.
The corresponding IR spectrum is depicted in the top trace of
Figure 3. Owing to mesomeric stabilization of the radical
center in the benzyl, a large number of product isomers are
possible; however, a comparison with computations (dotted
line) shows that bibenzyl (3, diphenylethane) is the dominant
reaction product and dimerization proceeds preferentially in
the benzylic position. For comparison IR spectra of several
other isomers of m/z=182 were computed that would be
formed by ortho- or para-attack, see the Supporting Informa-
tion. Most of them can be ruled out immediately. Only 2-ben-
zyltoluene shows peaks at the measured energies, but the
computed intensities do not agree with the experimental
ones.
Figure 2. Photoionization mass spectra of 1 with pyrolysis at 118 nm (upper
trace) and 265 nm (lower trace).
ionization cross-sections vary for the different molecules ob-
served in the experiment, the spectrum gives an approximate
measure of the relative concentration of molecules in the jet.
The spectrum shows a complete conversion of the precursor
(m/z=151) and the generation of 2 as the predominant mass
peak at m/z=91. Furthermore, dimerization products of 2 are
observed at m/z=182. The lower masses are side products of
the pyrolysis such as NO or, less likely owing to its high IE,
formaldehyde (m/z=30) and unimolecular decomposition
products such as benzene (m/z=78) and cyclopentadienyl
(m/z=65). The observation of benzene is expected as it is one
of the main decomposition products in the combustion of tol-
uene, whereas cyclopentadienyl has been observed before by
Buckingham et al.[10] Mass 122 corresponds to phenyl ethanol,
one of the reactants used in the precursor synthesis. Therefore,
masses 122 and 106, which correspond to benzaldehyde or
ethylbenzene (possible pyrolysis products of phenyl ethanol),
will not be further considered in the discussion of the benzyl
reaction products. The lower trace of Figure 2 shows the [1+1]
Product 3 has two different conformational isomers, anti and
gauche. Identification of the lowest energy isomer turned out
to be inconclusive previously because the energy difference
between both structures seems to be negligibly small.[20]
Owing to the different symmetries of these rotamers (C2h for
&
&
Chem. Eur. J. 2018, 24, 1 – 7
www.chemeurj.org
2
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
out-of-plane vibrational mode at 734 cmꢀ1 as well as two fur-
ther bands between 1400 cmꢀ1 and 1500 cmꢀ1. Finally, the IR
spectrum of m/z=178 is depicted in the bottom trace of
Figure 3. Although the signal/noise ratio is low, the carrier can
be identified as phenanthrene 5 by the two distinctive out-of-
plane vibrations at 724 cmꢀ1 and 797 cmꢀ1, as well as two
smaller enclosing signals at 628 cmꢀ1 and 867 cmꢀ1.
The dimerization of benzyl itself is well investigated. Sinha
and Raj[21] calculated the reaction to be barrierless and exother-
mic [Eq. (1)].
kJ
mol
2 C7H7 ! C14H14 DH0R ¼ ꢀ283:2
ð1Þ
The rate constants of this reaction show a low pressure but
high temperature dependence.[22] Whereas at lower tempera-
ture the dimerization of benzyl dominates, the back reaction
to benzyl gains importance with increasing temperature.[19,23]
Our spectra indicate sequential H2 loss as a reaction pathway
that leads from bibenzyl to phenanthrene formation, according
to Figure 4. The reaction is endothermic by 66.6 kJmolꢀ1 and
thus entropy-driven.[30] This is in agreement with recent work
in toluene flames that suggested 3 to be a phenanthrene pre-
cursor simply based on the ratio of mole fractions of 3 and 5
under varying conditions.[2,19] Shukla et al.[24] also concluded in
their toluene pyrolysis study that the formation of 5 is pre-
dominantly due to consecutive H2 elimination of the bibenzyl
dimer product. It can alternatively be formed in HACA-like
pathways, but suitable precursors such as acetylene and bi-
phenyl[19,25] have not been added in the experiments.
Figure 3. IR/UV ion dip spectra (solid lines) of m/z=182, 180, and 178 in
comparison with computed IR spectra (dotted lines). Upper trace: IR/UV of
mass 182 in comparison with bibenzyl. Middle trace: IR/UV of mass 180 in
comparison with 9,10-dihydrophenanthrene. Lower trace: IR/UV of mass 178
in comparison with phenanthrene.
anti and C2 for gauche), they can be identified by their respec-
tive gas-phase IR spectra. Our data indicates that both isomers
are present in the jet, in agreement with previous findings.[20c]
The peak at 1334 cmꢀ1 has been shown to be a unique feature
of the IR spectrum of the gauche isomer, whereas the relative
intensity of the signal at 1488 cmꢀ1 compared with the enclos-
ing vibrations at 1445 cmꢀ1 and 1593 cmꢀ1 suggest a contribu-
tion of the anti isomer. The relative composition at room tem-
perature was predicted to be 70% gauche and 30% anti, a dis-
tribution that would qualitatively explain the intensities in our
spectrum.
The reaction from 3 to 5 can proceed via two pathways:
(I) cyclization to 4, associated with H2 loss in a first step, fol-
lowed by a subsequent second H2 loss (see Figure 4) or (II) ini-
tial loss of H2 to form stilbene and then cyclization concomi-
tant with another H2 loss in a second step. Interestingly, m/z=
180 has been observed in several studies and was usually iden-
tified as stilbene, based mainly on chemical intuition and
simple calculations.[23a,24,26] However, recent CBS-QB3 calcula-
tions[21] questioned the involvement of stilbene in phenan-
threne formation from 3 and rather suggested the pathway via
4 to be the more efficient above 1200 K reaction temperature.
The experimental IR/UV spectrum of m/z=180 is presented
in the center trace of Figure 3. Based on the computation, it is
assigned to 9,10-dihydrophenanthrene 4, showing the major
Figure 4. Identified condensation products in the benzyl radical self-reaction. Top: Dimer formation and consecutive H2 loss to 9,10-dihydrophenanthrene and
phenanthrene. Bottom: Addition of phenyl in the formation of diphenylmethane and cyclization to fluorene.
Chem. Eur. J. 2018, 24, 1 – 7
www.chemeurj.org
3
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
Although the reaction from 3 to 4 was computed to be
The heat of reaction was calculated from Ref. [28]. In con-
nearly isoenergetic with an activation barrier of 70.1 kJmolꢀ1, trast, the reaction of benzyl and benzene to diphenylmethane
the formation of E-stilbene requires an activation energy of + H would be endothermic by 93 kJmolꢀ1. The final observed
168.8 kJmolꢀ1 and the reaction is endergonic with product of benzyl condensation appears at m/z=166 and can
90.2 kJmolꢀ1 [21]
.
be identified as fluorene 7. The IR spectrum, depicted in
The IR spectrum (see Figure 3) shows no features resulting Figure 5, is dominated by the intense B1 out-of-plane vibration
from E-stilbene (see the Supporting Information), which should at 732 cmꢀ1.[29] We accredit the formation of 7 to the H2 elimi-
be dominated by two vibrational bands in the 600–800 cmꢀ1 nation and ring-closing reaction of 6, see Figure 4, as suggest-
region with similar intensity. Hence, the data confirm this ed previously [Eq. (3)].[24,27]
recent theoretical prediction that the reaction to 5 proceeds
via pathway (I) with initial cyclization to dihydrophenanthrene.
kJ
mol
C13H12 ! 7 þ H2 DH0R ¼ þ 13
ð3Þ
The heat of reaction was calculated based on literature
data.[28,30] The reaction is slightly endothermic, so at high tem-
peratures the entropy increase will be the driving force. The in-
crease of the fluorene signal with pyrolysis temperature ob-
served in the mass spectra confirms this picture. Flame experi-
ments also claimed 7 to appear as a product in premixed tolu-
ene laminar flames.[19] The formation of 7 was attributed to a
HACA reaction growth chain starting from benzyl (C7H7!
C9H8!C11H8!C13H10) owing to the high concentration of acet-
ylene in their experiments and the observation of the inter-
mediate PAHs. In contrast, we do not observe these intermedi-
ate molecules. Note that an IR spectrum of the most stable
C9H8 isomer indene has been recorded in previous experiments
on phenyl condensation.[16] Thus, it is in principle identifiable,
when present. Therefore, we conclude that 7 is formed by H2
loss from 6 rather than in a HACA process via indene, as
shown in Figure 4.
The peaks at m/z=168 and 166
At m/z=166 and 168, two strong peaks appear in the mass
spectrum recorded at 265 nm. From a comparison of the ex-
perimental and the computed IR spectrum given in Figure 5
mass 168 is unambiguously identified as diphenylmethane 6.
All major features present in the experimental spectrum are
well represented by the calculated IR spectrum of diphenylme-
thane.
Conclusions
The products of the self-reaction of benzyl radicals in a pyroly-
sis microreactor at high temperatures have been investigated
by IR/UV ion dip spectroscopy. The approach combines the
mass selectivity of REMPI-TOF spectroscopy with the isomer se-
lectivity of IR spectroscopy. This novel method has allowed us
to identify several PAH products originating from the benzyl
radical 2, the primary decomposition product of toluene in
combustion. We identified bibenzyl (diphenylethane), dihydro-
phenanthrene, phenanthrene, diphenylmethane, and fluorene
Figure 5. Comparison of the IR spectra of m/z=168 and 166 (solid line) with
the calculated IR spectrum of diphenylmethane and fluorene (dotted line).
A product at m/z=168 has been observed in previous stud- as bimolecular reaction products. All condensation products
ies on toluene pyrolysis that used mass spectrometric detec- are formed from the two main decomposition products of tol-
tion.[24,27] It was explained by a reaction of benzyl with ben- uene combustion: benzyl and phenyl, and were identified by
zene, the second important pyrolysis product of toluene. The their IR spectra. Particularly, the identification of a species at
118 nm mass spectrum (see Figure 2) shows indeed a signal at m/z=180 as 9,10-dihydrophenanthrene gives new insight into
m/z=78, so benzene and phenyl radicals are also formed in the formation of phenanthrene (see Figure 4) and shows that
the pyrolysis reactor and become available for secondary reac- the reaction proceeds by initial cyclization to 4 followed by a
tions. Note that the concentration of phenyl is probably too second H2 loss, in agreement with recent computations. An in-
small to be observed in the mass spectra owing to its high re- termediate with m/z=180 has been observed before, but was
activity. Hence, 6 is likely formed by the strongly exothermic often assigned to stilbene owing to a lack of isomer sensitivity
reaction of phenyl formed from activated benzene with benzyl in the experiments, leading to the assumption of an initial H2
[Eq. (2)].
loss to stilbene followed by cyclization in a second step. A fur-
ther pathway in the bimolecular high-temperature chemistry
of benzyl leads to diphenylmethane, probably through a reac-
tion with phenyl radicals. Subsequent cyclization associated
kJ
mol
C6H5 þ C7H7 ! C13H12 DH0R ¼ ꢀ382:3
ð2Þ
&
&
Chem. Eur. J. 2018, 24, 1 – 7
www.chemeurj.org
4
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
with H2 loss leads to 7. The experiments thus suggest a new
route to fluorene formation that will occur in the high-temper-
ature chemistry of toluene.
thermore we would like to thank Jana Hemberger for creating
the front cover illustration.
Conflict of interest
Experimental Section
The authors declare no conflict of interest.
Benzyl radicals 2 were generated by flash pyrolysis of 1 (see
Figure 1), synthesized from 2-phenyl ethanol and used without fur-
ther purification.[31] The precursor was heated to 85–958C and
seeded in 1.4 bar argon. A pulsed jet was produced by using a sol-
enoid valve (20 Hz, Parker General Valve) with a resistively heated
SiC tube mounted at the exit to cleave the precursor molecules.
The pyrolysis unit measures 38 mm in length with an inner diame-
ter of 1 mm. It was heated to approximately 11008C and served as
a chemical flow reactor. Note that the comparably large length of
the SiC tube promotes bimolecular processes. The experimental
conditions like sample temperature, pyrolysis power, gas pressure,
UV wavelength, etc. have been optimized with respect to dimer
formation when using 118 nm one-photon photoionization.
Keywords: gas-phase reactions · high-temperature chemistry ·
IR spectroscopy · pyrolysis · radical reactions
[1] T. W. Schmidt, Int. Rev. Phys. Chem. 2016, 35, 209–242.
[2] Y. Li, J. Cai, L. Zhang, T. Yuan, K. Zhang, F. Qi, Proc. Combust. Inst. 2011,
33, 593–600.
[3] D. W. Kohn, H. Clauberg, P. Chen, Rev. Sci. Instrum. 1992, 63, 4003–
4005.
[4] a) M. Margraf, B. Noller, C. Schroter, T. Schultz, I. Fischer, J. Chem. Phys.
2010, 133, 074304; b) A. Rçder, A. Humeniuk, J. Giegerich, I. Fischer, L.
Poisson, R. Mitric, Phys. Chem. Chem. Phys. 2017, 19, 12365–12374.
[5] J. D. Savee, J. Zador, P. Hemberger, B. Sztaray, A. Bodi, D. L. Osborn, Mol.
Phys. 2015, 113, 2217–2227.
After expansion into the source chamber of
a differentially
pumped vacuum apparatus, the beam was skimmed and entered
the ionization region of a time-of-flight mass spectrometer (TOF-
MS).[32] Here, the free jet was crossed by the IR beam and UV laser
radiation to perform the IR/UV ion dip experiment. This methodol-
ogy has previously been successfully employed to study a variety
of different system, for example, biomolecules,[33] hydrated aromat-
ic molecules,[34] and radicals.[16,17] A frequency-doubled Nd:YAG
pumped dye laser ionized molecules in the jet via [1+1] REMPI.
The UV wavelength was fixed at 265 nm by using frequency-dou-
bled Coumarin 153 with 0.65 mJ of power and a repetition rate of
20 Hz. The ions were subsequently detected by TOF-MS. The IR ra-
diation was provided by the free electron laser for infrared experi-
ments FELIX (FELIX Laboratory, Radboud University, The Nether-
lands). It was scanned over the fingerprint region from 550–
1750 cmꢀ1 with a step width of 2.5 cmꢀ1, delivering an output
power of up to 100 mJ at a repetition rate of 10 Hz. The spectra
above are only displayed from 600 cmꢀ1 upwards owing to experi-
mental artifacts appearing below this value. The IR spectra depict-
ed above were obtained by dividing the signal without and with IR
excitation, taking the decadic logarithm, correcting for IR laser
power and averaging. Finally, the spectra were smoothed by using
digital Savitzky–Golay filtering.
[6] a) M. Shapero, N. C. Cole-Filipiak, C. Haibach-Morris, D. M. Neumark, J.
Phys. Chem. A 2015, 119, 12349–12356; b) Y. Song, X. F. Zheng, M.
Lucas, J. S. Zhang, Phys. Chem. Chem. Phys. 2011, 13, 8296–8305.
[7] Q. Guan, K. N. Urness, T. K. Ormond, D. E. David, G. B. Ellison, J. W. Daily,
Int. Rev. Phys. Chem. 2014, 33, 447–487.
[8] A. Vasiliou, M. R. Nimlos, J. W. Daily, G. B. Ellison, J. Phys. Chem. A 2009,
113, 8540–8547.
[9] M. Scheer, C. Mukarakate, D. J. Robichaud, G. B. Ellison, M. R. Nimlos, J.
Phys. Chem. A 2010, 114, 9043–9056.
[10] a) G. T. Buckingham, T. K. Ormond, J. P. Porterfield, P. Hemberger, O.
Kostko, M. Ahmed, D. J. Robichaud, M. R. Nimlos, J. W. Daily, G. B. Elli-
son, J. Chem. Phys. 2015, 142, 044307; b) G. T. Buckingham, J. P. Porter-
field, O. Kostko, T. P. Troy, M. Ahmed, D. J. Robichaud, M. R. Nimlos, J. W.
Daily, G. B. Ellison, J. Chem. Phys. 2016, 145, 014305.
[11] D. S. N. Parker, R. I. Kaiser, T. P. Troy, M. Ahmed, Angew. Chem. Int. Ed.
2014, 53, 7740–7744; Angew. Chem. 2014, 126, 7874–7878.
[12] D. S. N. Parker, R. I. Kaiser, B. Bandyopadhyay, O. Kostko, T. P. Troy, M.
Ahmed, Angew. Chem. Int. Ed. 2015, 54, 5421–5424; Angew. Chem.
2015, 127, 5511–5514.
[13] P. Hemberger, A. J. Trevitt, T. Gerber, E. Ross, G. da Silva, J. Phys. Chem. A
2014, 118, 3593–3604.
[14] A. M. Rijs, J. Oomens, in Topics in Current Chemistry, Vol. 364: Gas-Phase
IR Spectroscopy and Structure of Biological Molecules (Eds.: A. M. Rijs, J.
Oomens), 2015, Springer, Cham, pp. 1–42.
[15] P. Constantinidis, F. Hirsch, I. Fischer, A. Dey, A. M. Rijs, J. Phys. Chem. A
2017, 121, 181–191.
[16] P. Constantinidis, H. C. Schmitt, I. Fischer, B. Yan, A. M. Rijs, Phys. Chem.
Chem. Phys. 2015, 17, 29064–29071.
[17] K. H. Fischer, J. Herterich, I. Fischer, S. Jaeqx, A. M. Rijs, J. Phys. Chem. A
2012, 116, 8515–8522.
Isomers were identified with the aid of frequency calculations, con-
ducted at the DFT/B3LYP/6–311G(p,d) level of theory and carried
out by using the Gaussian09 computational chemistry software.[35]
The computed vibrational spectra were scaled by a factor of 0.975
and convolved with a Lorentzian (FWHM=8 cmꢀ1).
[18] R. G. Satink, G. Meijer, G. Von Helden, J. Am. Chem. Soc. 2003, 125,
15714–15715.
[19] Y. Li, L. Zhang, Z. Tian, T. Yuan, J. Wang, B. Yang, F. Qi, Energy Fuels
2009, 1473–1485.
Acknowledgments
[20] a) P. M. Ivanov, J. Mol. Struct. 1997, 415, 179–186; b) N. Kurita, P. M.
Ivanov, J. Mol. Struct. 2000, 554, 183–190; c) C. Latouche, V. Barone, J.
Chem. Theory Comput. 2014, 10, 5586–5592; d) A. Horn, P. Klaeboe, B.
Jordanov, C. J. Nielsen, V. Aleksa, J. Mol. Struct. 2004, 695–696, 77–94;
e) Q. Shen, J. Mol. Struct. 1998, 471, 57–61; f) M. A. Martin-Drumel, O.
Pirali, C. Falvo, P. Parneix, A. Gamboa, F. Calvo, P. Brꢁchignac, Phys.
Chem. Chem. Phys. 2014, 16, 22062–22072.
[21] S. Sinha, A. Raj, Phys. Chem. Chem. Phys. 2016, 18, 8120–8131.
[22] a) K. Luther, K. Oum, K. Sekiguchi, J. R. Troe, Phys. Chem. Chem. Phys.
2004, 6, 4133–4141; b) A. A. Boyd, B. Noziere, R. Lesclaux, J. Phys.
Chem. 1995, 99, 10815–10823; c) D. C. Astholz, J. Durant, J. Troe, Symp.
Int. Combust. Proc. 1981, 18, 885–892; d) A. Matsugi, A. Miyoshi, Chem.
Phys. Lett. 2012, 521, 26–30; e) R. D. Burkhart, J. Am. Chem. Soc. 1968,
This work was supported by the Deutsche Forschungsgemein-
schaft, contract FI 575/8-2. Furthermore, the research leading
to these results has received funding from the European Com-
munity’s Seventh Framework Programme (FP7/2007–2013)
under grant agreement no. 312284 and from LASERLAB-
EUROPE (grant agreement no. 654148, European Union’s Hori-
zon 2020 research and innovation programme). We gratefully
thank the FELIX staff for their experimental support and we ac-
knowledge the Nederlandse Organisatie voor Wetenschappelijk
Onderzoek (NWO) for the support of the FELIX Laboratory. Fur-
Chem. Eur. J. 2018, 24, 1 – 7
www.chemeurj.org
5
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
90, 273–277; f) R. D. Burkhart, J. Phys. Chem. 1969, 73, 2703–2706; g) H.
Hippler, J. Troe, J. Phys. Chem. 1990, 94, 3803–3806.
[23] a) J. Jones, G. B. Bacskay, J. C. Mackie, J. Phys. Chem. A 1997, 101, 7105–
7113; b) W. Mueller-Markgraf, J. Troe, J. Phys. Chem. 1988, 92, 4899–
4905.
[24] B. Shukla, A. Susa, A. Miyoshi, M. Koshi, J. Phys. Chem. A 2007, 111,
8308–8324.
[25] A. D’Anna, A. Violi, A. D’Alessio, Combust. Flame 2000, 121, 418–429.
[26] A. Matsugi, A. Miyoshi, Proc. Combust. Inst. 2013, 34, 269–277.
[27] B. Shukla, A. Susa, A. Miyoshi, M. Koshi, J. Phys. Chem. A 2008, 112,
2362–2369.
[28] M. V. Roux, M. Temprado, J. S. Chickos, Y. Nagano, J. Phys. Chem. Ref.
Data 2008, 37, 1855–1996.
[29] L. Cuff, M. Kertesz, J. Phys. Chem. 1994, 98, 12223–12231.
[30] NIST Chemistry WebBook, NIST Standard Reference Database Number 69,
National Institute of Standards and Technology, Gaithersburg, MD,
2005.
Phys. 2014, 16, 10770–10778; c) J. Mahꢁ, D. J. Bakker, S. Jaeqx, A. M.
Rijs, M.-P. Gaigeot, Phys. Chem. Chem. Phys. 2017, 19, 13778–13787.
[34] D. J. Bakker, A. Dey, D. P. Tabor, Q. Ong, J. Mahꢁ, M.-P. Gaigeot, E. L.
Sibert, A. M. Rijs, Phys. Chem. Chem. Phys. 2017, 19, 20343–20356.
[35] Gaussian 09, Revision A.02, G. W. T. M. J. Frisch, H. B. Schlegel, G. E. Scu-
seria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, G. A. Peters-
son, H. Nakatsuji, X. Li, M. Caricato, A. Marenich, J. Bloino, B. G. Janesko,
R. Gomperts, B. Mennucci, H. P. Hratchian, J. V. Ortiz, A. F. Izmaylov, J. L.
Sonnenberg, D. Williams-Young, F. Ding, F. Lipparini, F. Egidi, J. Goings,
B. Peng, A. Petrone, T. Henderson, D. Ranasinghe, V. G. Zakrzewski, J.
Gao, N. Rega, G. Zheng, W. Liang, M. Hada, M. Ehara, K. Toyota, R.
Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H.
Nakai, T. Vreven, K. Throssell, J. A. Montgomery, Jr., J. E. Peralta, F.
Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov,
T. Keith, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C.
Burant, S. S. Iyengar, J. Tomasi, M. Cossi, J. M. Millam, M. Klene, C.
Adamo, R. Cammi, J. W. Ochterski, R. L. Martin, K. Morokuma, O. Farkas,
J. B. Foresman, D. J. Fox, Gaussian, Inc., Wallingford, CT, 2016.
[31] M. Gasser, J. A. Frey, J. M. Hostettler, A. Bach, Chem. Commun. 2011, 47,
301–303.
[32] A. M. Rijs, J. Oomens, in Gas-Phase IR Spectroscopy and Structure of Bio-
logical Molecules, Springer International Publishing, Cham, 2015, pp. 1–
42.
[33] a) S. Jaeqx, J. Oomens, A. Cimas, M.-P. Gaigeot, A. M. Rijs, Angew. Chem.
Int. Ed. 2014, 53, 3663–3666; Angew. Chem. 2014, 126, 3737–3740;
b) B. Yan, S. Jaeqx, W. J. van der Zande, A. M. Rijs, Phys. Chem. Chem.
Manuscript received: February 19, 2018
Revised manuscript received: March 9, 2018
Accepted manuscript online: March 12, 2018
Version of record online: && &&, 0000
&
&
Chem. Eur. J. 2018, 24, 1 – 7
www.chemeurj.org
6
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
FULL PAPER
Radical self-reaction: The self-reaction
of benzyl radicals at high temperature,
studied with IR/UV ion dip spectroscopy,
leads to phenanthrene and fluorene as
polycyclic aromatic hydrocarbons.
&
High-Temperature Pyrolysis
F. Hirsch, P. Constantinidis, I. Fischer,*
S. Bakels, A. M. Rijs*
&& – &&
Dimerization of the Benzyl Radical in
a High-Temperature Pyrolysis Reactor
Investigated by IR/UV Ion Dip
Spectroscopy
Chem. Eur. J. 2018, 24, 1 – 7
www.chemeurj.org
7
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ