Hydrogenation and dehydrogenation reactions of the phenalenyl radical/1H-phenalene system at low temperatures

Article abstract of DOI:10.1016/j.cplett.2020.137183

Phenalenyl radical was generated in situ in inert matrices by laser UV photolysis of 1H-phenalene, and was identified by recording laser induced fluorescence and IR spectra. Theoretical computations predict that the H atom addition to phenalenyl radical is barrierless, while the H atom abstraction from 1H-phenalene has only a small barrier. Upon annealing the Xe matrix after the photolysis, the radical and the H atom recombine, while H atom abstraction from 1H-phenalene is a possible explanation for additional spectral changes. These results show that the phenalenyl radical/1H-phenalene system can be a very effective catalyst in the formation of interstellar H₂.

Full text of DOI:10.1016/j.cplett.2020.137183

Journal Pre-proofs
Research paper
Hydrogenation and dehydrogenation reactions of the phenalenyl radical/1H-
phenalene system at low temperatures
Anita Schneiker, István Pál Csonka, György Tarczay
PII:
S0009-2614(20)30098-1
DOI:
https://doi.org/10.1016/j.cplett.2020.137183
CPLETT 137183
Reference:
To appear in:
Chemical Physics Letters
Received Date:
Revised Date:
Accepted Date:
25 November 2019
3 February 2020
4 February 2020
Please cite this article as: A. Schneiker, I. Pál Csonka, G. Tarczay, Hydrogenation and dehydrogenation reactions
of the phenalenyl radical/1H-phenalene system at low temperatures, Chemical Physics Letters (2020), doi:
https://doi.org/10.1016/j.cplett.2020.137183
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover
page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version
will undergo additional copyediting, typesetting and review before it is published in its final form, but we are
providing this version to give early visibility of the article. Please note that, during the production process, errors
may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
© 2020 Published by Elsevier B.V.

Hydrogenation and dehydrogenation reactions of the
phenalenyl radical/1H-phenalene system at low
temperatures
Anita Schneiker,a,b István Pál Csonka,a György Tarczaya,b,*
aELTE-MTA Lendület Laboratory Astrochemistry Research Group,
PO Box 32, H-1518, Budapest 112, Hungary
bLaboratory of Molecular Spectroscopy, Institute of Chemistry, ELTE Eötvös University,
PO Box 32, H-1518, Budapest 112, Hungary
Abstract
Phenalenyl radical was generated in situ in inert matrices by laser UV photolysis of 1H-
phenalene, and was identified by recording laser induced fluorescence and IR spectra.
Theoretical computations predict that the H atom addition to phenalenyl radical is barrierless,
while the H atom abstraction from 1H-phenalene has only a small barrier. Upon annealing the
Xe matrix after the photolysis, the radical and the H atom recombine, while H atom abstraction
from 1H-phenalene is a possible explanation for additional spectral changes. These results show
that the phenalenyl radical/1H-phenalene system can be a very effective catalyst in the
formation of interstellar H2.
Keywords: polyaromatic hydrocarbons, photochemistry, matrix isolation, astrochemistry,
interstellar H2 formation
*Corresponding author. e-mail: tarczay@chem.elte.hu, phone: +36-1-372-2500/6587
1

1. Introduction
The phenalenyl radical (C13H9) is a polyaromatic hydrocarbon (PAH), which can be
considered as the smallest model system of graphene. It is a neutral 13- electron odd alternant
aromatic species, possessing D3h symmetry in its 2A1″ ground electronic state [1]. Due to strong
resonance stabilization in the phenalenyl radical (see Fig. 1a), it could be prepared even in
solution and was characterized by ESR spectroscopy [2,3]. The first excited state of phenalenyl
has 2E″ symmetry, therefore in this state the radical shows Jahn-Teller activity [4,5]. This was
2
proven experimentally by the analysis of 2E″→ A ″ fluorescence emission spectrum,
퐴
푋
1
recorded in polycrystalline n-pentane between 20 and 72 K by Cofino and coworkers [4]. They
have prepared the radical by in situ broad-band UV photolysis of cyclopropanoacenaphthylene.
More recently, O’Connor et al. have obtained the jet-cooled REMPI and the jet-cooled
dispersed fluorescence spectrum of the phenalenyl radical [5]. In that case the radical was
generated by DC discharge from 1H-phenalene. They have analyzed the spectra by the help of
high-level (EOM-CCSD) quantum chemical computations, and they have shown that the 2E″
퐴
and the 2E″ states are strongly coupled by pseudo-Jahn-Teller mechanism.
퐵
Both the 12- electron phenalenyl cation [6] and the 14- electron anionic system were
prepared [7]. It was shown that both of these ionic species, as well as the neutral radical, are
aromatic systems, as they sustain well‐defined diatropic perimeter ring currents [8]. Due to this
behavior the phenalenyl motif was described as a magnetic chameleon [9]. The hydrogen
addition to the neutral radical, or the protonation of the anion can lead to the phenalene isomers
(see Fig. 1b). Among the phenalenes, only the 1H-phenalene was synthetized so far [10]. The
structure and the relative stability of the four possible phenalene isomers in their lowest energy
singlet and triplet states were investigated by Zoellner and Zoellner by quantum chemical
computations at the HF/6-31G* and B3LYP/6-31G* levels of theory [11]. They have concluded
that the singlet 1H-phenalene is the most stable form, and the second most stable isomer, the
triplet 2H-phenalene, might also be synthesized.
PAHs have been considered as catalysts of H2 formation in the interstellar medium,
especially in the photon-dominated regions (PDRs) [12–15]. In the proposed catalytic
mechanism, first H atom is (or several H atoms are) chemisorbed by a PAH molecule. In the
second step, another H atom can abstract a chemisorbed H-atom, generating back the original
PAH molecule and a H2 molecule. Several studies have pointed out that the second process is
usually barrierless, while the first process has an activation energy [16–19]. Therefore, the first
2

step is the rate limiting process. However, in contrast to graphene or graphite, computations
revealed that the activation barrier of the H atom addition to benzene [20] and pyrene [21] is
permeable by tunneling, and the reaction rate is non-negligible even at 40 K. From the
experimental side, Menella and coworkers have exposed coronene to D atoms and observed the
formation of HD and D2 molecules [22]. With this experiment they proved for the first time the
catalytic role of neutral PAH molecules in H2 formation. In a very recent paper Barrales-
Martínez and Gutiérrez-Oliva have shown by computations that for some N- and Si-doped
coronenes the H addition process has no barrier, therefore the H atom chemisorption rates for
these species are higher than that for coronene [23].
Being a radical-radical reaction, the H atom addition to phenalenyl radical is expected to be
barrierless, while the radical itself is relatively stable due to resonance stabilization. Therefore,
phenalenyl-type systems could be efficient interstellar H2 catalysts. Besides the astrochemical
importance, similar graphene/graphene fragments may also play a role in H-storage [1]. The
goals of the present work were to generate and spectroscopically identify the phenalenyl radical
in an inert matrix, and investigate the reactions of the phenalenyl radical and 1H-phenalene with
H atoms at low temperatures upon annealing the matrix. Besides Ar, Xe was used as matrix
host, because H atoms can be mobilized by warming up the Xe matrix to 30–50 K [24,25].
2. Methods
2.1. Experimental details
1H-Phenalene was prepared from 1-hydroxy-2,3-dihydro-1H-phenalene (prepared in a past
project in our Institute) by water elimination according to a modified recipe of Boekelheide and
Larrabee [26]. 1-hydroxy-2,3-dihydro-1H-phenalene was dissolved in absolute ethanol and
mixed with absolute ethanol saturated with HCl. The solution was refluxed in a N2 atmosphere.
After cooling down to room temperature, the solution was poured onto water, and the product
was extracted by CH2Cl2. After evaporation of CH2Cl2, the product was purified by vacuum
sublimation. The purity was checked by 1H-NMR spectroscopy and GC-MS.
1H-Phenalene was evaporated into the vacuum chamber using a home-built Knudsen
effusion cell. The evaporated sample was mixed with argon (Messer, 99.9999%) or Xe (Sigma-
Aldrich, 99.995%) before deposition. The gas flow was kept at ~0.05 mmol min1.
3

The samplerare gas mixture was deposited onto a cold (8–10 K or 14 K for Ar, 20 K for
Xe) CsI window, attached to a Janis CCS-350R cold head cooled by a CTI Cryogenics 22
closed-cycle refrigerator unit. The temperature of the cold window was controlled by a Lake
Shore 321 thermostat equipped with a silicon diode thermometer. The cold window was set at
45° to the optical path of the spectrometer, and the irradiating laser beam was perpendicular to
the optical path. The IR spectra were recorded by a Bruker IFS 55 spectrometer, using a Globar
source and an MCT detector. The spectra were recorded at 1 cm1 instrumental resolution.
Laser induced fluorescence (LIF) spectra were recorded by an Ocean Optics HR2000 fiber
optic UV-Vis spectrometer. The instrumental resolution was ~11 cm1.
For UV photolysis and for LIF spectroscopy the frequency doubled signal of an optical
parametric oscillator (VersaScan MB 240 OPO, GWU/Spectra Physics) was used. The OPO
was pumped with the third harmonic (355 nm) of a pulsed (10 Hz, 2–3 ns) Quanta Ray Lab 150
Nd:YAG laser (Spectra Physics). The pulse energy was ~2 and 4.5 mJ at 220 nm and 250 nm,
respectively. The laser beam was unfocused; its diameter was about 0.8 cm. An LPW 3860 low
pass filter was placed between the sample and the detector of the IR spectrometer to make
possible IR spectroscopic measurements during UV irradiation.
2.2. Computational details
Equilibrium structures, harmonic vibrational frequencies and intensities of 1H-phenalene
and the phenalenyl radical were computed at the B3LYP/cc-pVTZ level of theory. The
optimized geometries, the computed vibrational frequencies and intensities are summarized in
Tables S1 and S2. Scaling factors for the harmonic wavenumbers were determined by fitting
the computed wavenumbers to the experimental wavenumbers of 1H-phenalene. This scaling
factor was applied for the phenalenyl radical too. The transition state structure of the 1H-
phenalene + H → phenalenyl radical + H2 reaction was located and barrier heights were
determined at the B3LYP/cc-pVTZ level of theory using the Synchronous Transit-Guided
Quasi-Newton (STQN) method [27,28]. Intrinsic reaction path (IRC) computations were
carried out to check that the transition structure corresponds to the right reaction. The potential
energy profile along the phenalenyl radical + H → 1H-phenalene reaction path was computed
by relaxed potential energy scan. Single-point energy computations were carried out at the
CCSD(T)/cc-pVDZ//B3LYP/cc-pVTZ and CCSD(T)/cc-pVTZ//B3LYP/cc-pVTZ levels of
theory. All computations were performed by the GAUSSIAN suite of programs [29].
4

3. Results and Discussion
The IR spectra (in the 1550–1010 cm–1 and 880–700 cm–1 regions) of 1H-phenalene
deposited in an 8–10 K Ar and in a 20 K Xe matrix are shown in Fig. 2a and 2c, respectively.
The spectra in the 3500–650 cm–1 range are shown in Fig. S1 of the Supplementary Material.
The spectra are dominated by three strong out-of-plane aromatic vibrations at 720, 777,
and 827 cm–1 in Ar, and 718, 778, and 826 cm–1 in Xe matrix. Table 1 summarizes the
experimentally observed bands assigned to computed vibrational transitions. (The complete list
of computed harmonic wavenumbers and intensities are given in Table S2 of the Supplementary
Material.) The computations show very good agreement with the observations regarding the
wavenumbers. The intensities of the C‒H stretching modes are somewhat overestimated by the
harmonic computations, while the experimental intensities of these modes are reduced by site
splitting.
The matrix isolated 1H-phenalene was exposed to laser UV radiation at wavelengths 335,
322, 250 and 220 nm. These photon energies are slightly higher than the excitation energies
from the ground electronic state to the first (339 nm, experimental), to the second (323 nm,
experimental) and third (261 nm, computed) singlet excited states of 1H-phenalene [1]. In the
case of irradiations at 335 and 322 nm no change was observed in the IR spectrum. In contrast
to this, both the 220 and the 250 nm irradiation bleached the bands of 1H-phenalene, and new
bands were observed at 754, 757 and 839 cm–1 in Ar matrix (deposited at 8–10 K), and 754,
758 and 837 cm–1 in Xe matrix. These bands and some further weak features could be assigned
in a very good agreement to the computed vibrational wavenumbers of phenalenyl radical (see
Table 2; the complete list of computed harmonic wavenumbers and intensities are given in
Table S2 of the Supplementary Material). The computed intensities of the vibrational transitions
are also in a good correspondence with the observations. (Note that the band at 1181 cm–1
coincides with a band of XeH2, a possible product of the photolysis. However, the more intense
1166 cm–1 band of XeH2 [30,31] was not observed.) The observation that 1H-phenalene is
dissociative in the third singlet excited state explains why O’Connor et al. were not able to
observe this state of 1H-phenalene [1].
The identification of phenalenyl radical was also supported by recording its fluorescence
spectra, as shown in Fig. 3. In both Ar and Xe matrices broad fluorescence bands were observed,
5

which can be explained most likely by the strong scattering of the matrix and by the interaction
between the noble gas and the phenalenyl radical. In the case of Xe, fluorescence was observed
when the excitation laser wavelength was set between 514 and 516 nm, while for the Ar matrix
the strongest fluorescence signal was observed on excitation at ~512.5 nm. This corresponds to
1
2
the strongest vibronic band ( ) of the 2E″( E″) ← A ″ transition, observed around 19600
휈0
퐴
퐵
푋
1
cm‒1 (510 nm) for the free phenalenyl radical [5]. In the case of matrices containing 1H-
phenalene as-deposited, without 220 or 250 nm photolysis, fluorescence was not observed upon
512.5 nm or ~515 nm excitation. The spectrum corresponding to Xe matrix is red-shifted by ca.
120 cm–1 compared to the fluorescence spectrum of phenalenyl radical in Ar. This can be
explained by van der Waals interaction between phenalenyl radical and the noble gas atoms
being stronger for Xe than for Ar. This interaction is stronger in the excited state than in the
electronic ground state. At first sight it is not obvious to compare the spectra shown in Fig. 3
and the fluorescence spectrum of phenalenyl radical isolated in polycrystalline n-pentane [4],
because of the different resolutions. The positions of the maxima of the most intense bands in
n-pentane are observed between the corresponding values measured in Ar and Xe matrices. The
differences in the relative intensities can be explained by the different resolution and by the
different environment and temperature.
It is interesting to note that other intense bands were observed in the fluorescence spectra
between 27300 and 29200 in Ar, and between 26900 and 28400 cm‒1 in Xe (see full scale
fluorescence spectra in Fig. S2 of the Supplementary Material). These bands are considerably
blue shifted compared to the wavelength of the laser excitation. This interesting observation
might be explained by the recombination phenalenyl radical and a nearby H atom promoted by
the laser excitation, resulting in 1H-phenalene in the S1 excited state. Therefore, these
1
fluorescence bands correspond to the 1A′ ← A′ transition of 1H-phenalene.
퐴
푋
It is interesting to note that either the phenalenyl radical could not be observed or bands
with intensity comparable with noise level were only seen when 1H-phenalene was deposited
in a 14 K Ar matrix. A small amount of the radical could be prepared when the precursor was
photolyzed in an Ar matrix deposited at 8–10 K. In contrast to this, the photolytic generation of
phenalenyl radical was very efficient in a Xe matrix deposited at 20 K. This can likely be
explained by the morphology of the matrix. At a low-temperature deposition a more amorphous
matrix with more defects could be prepared, while at higher deposition temperatures a more
crystalline, more compact matrix was made. In the latter case, due to the cage effect, the
phenalenyl radical and the H atom likely recombine, while in the case of amorphous matrix
6

some of the H atoms can diffuse and escape from the site. In case of larger Xe atoms, the H
atoms can more efficiently escape from the cages due to the larger interstitial holes.
The matrices were annealed after photolysis. Changes in the IR spectra upon warming were
observed both for Ar and Xe matrices, but the change was much more pronounced for Xe matrix
(see Fig. 4 and Fig. S3 of the Supplementary Material). Upon annealing several processes can
take place simultaneously, including matrix structure reorganization, site change, diffusion of
H atoms and the reaction of H atoms with 1H-phenalene and the phenalenyl radical. Therefore,
the interpretation of spectral changes upon annealing is not straightforward.
To analyze these processes, the integrated intensity as a function of temperature of the 31
band of phenalenyl radical at 837 cm–1 (trace a) and the site-split 51 band of 1H-phenalene at
823–827 cm–1 (trace b) are plotted in Fig. 5. The intensity change of the 51 band of 1H-
phenalene observed in a non-irradiated matrix is also shown in the Figure (trace c). The latter
shows an increase up to ~30 K, above this temperature it is constant. This increase can be
explained either by matrix structure reorganization or by changing the matrix site of 1H-
phenalene. Both of these processes can result in a small change of band positions and intensities.
Upon warming the Xe matrix from 20 K to ~40 K, the band of phenalenyl radical clearly
decreased (see Fig. 4b). At the same time, the higher-wavenumber side of the 51 band of 1H-
phenalene increased, while a decrease in intensity was observed at its lower-wavenumber side.
Because of the very similar structures of 1H-phenalene and phenalenyl radical we can assume
that the intensity changes due to site change processes of phenalenyl radical are also completed
by 30 K. This means, that the ~15% decrease in intensity of the 31 band of phenalenyl radical
between 30 K and 40 K is caused very likely by the recombination of phenalenyl radicals with
H atoms.
Between 35 K and 50 K a small, but monotonic decrease in the intensity of the 51 band of
1H-phenalene was observed (see trace b in Fig. 5). In this temperature region the lower
wavenumber side of this band of 1H-phenalene is decreasing much more than its higher
wavenumber side is increasing. This type of change was not observed for the band of 1H-
phenalene in a non-irradiated Xe matrix (See Fig. S3 of the Supplementary Material). A possible
explanation of this observation is that in this temperature range, H atoms can diffuse farther
from the site where they were generated, and can react with 1H-phenalene molecules.
According to our computations the H atom abstraction from 1H-phenalene by a free H atom,
7

leading to phenalenyl radical, has only a barrier of 0.3, 11.8 and 10.6 kJ mol–1 at the B3LYP/cc-
pVTZ, CCSD(T)/cc-pVDZ//B3LYP/cc-pVTZ and CCSD(T)/cc-pVTZ//B3LYP/cc-pVTZ
levels of theory, respectively. Former computational and experimental results showed that,
although the classical reaction channels are closed, a barrier with this height can be permeable
to H atoms even at this low temperature by tunneling [20‒23]. Therefore, according to the
computational results, the H abstraction reaction from 1H-phenalene is possible under our
experimental conditions.
The back formation of phenalenyl radical (see Fig. 6) can be a possible explanation for the
observation that the intensity change of the 31 band of phenalenyl radical is constant above
40 K. Assuming the mechanism depicted in Fig. 6, one would expect if the intensity of the 51
band of 1H-phenalene is decreasing then the intensity 31 band of phenalenyl radical should not
only be constant, but it should increase. This discrepancy might be explained by other (e.g.,
addition) reactions between 1H-phenalene and H atoms. It has to be also mentioned, that in
principle the reaction of a H atom with phenalenyl radical could also lead to other phenalene
isomers. However, these reactions are energetically unfavorable compared to the above
discussed processes. No new bands remarkably larger than the noise level were observed upon
annealing, therefore we cannot identify any of these species.
4. Conclusions
In the present study the IR spectrum of phenalenyl radical isolated in Ar and Xe matrices
was observed for the first time. The radicals were generated by 220 or 250 nm photolysis of
1H-phenalene. The identification of the radical was based both on comparison of computed and
experimental IR spectra, and on the comparison of fluorescence spectra to the spectrum
recorded in low-temperature polycrystalline n-pentane [4]. The annealing of the matrices
containing phenalenyl radicals and H atoms revealed that phenalenyl radical can recombine
with a H atom at low temperatures. Furthermore, the H atom abstraction by a free H atom from
1H-phenalene, leading to phenalenyl radical can be a possible explanation for the spectral
changes observed between 40 K and 50 K. This H atom abstraction/addition mechanism (Fig.
6) is supported by quantum chemical computations, which predict no barriers for the H addition
reaction of phenalenyl radical, and a very small barrier, permeable for H atoms by tunneling,
for the H abstraction reaction for 1H-phenalene. This suggests that the 1H-
phenalene/phenalenyl system might be a very effective catalyst, more effective than other
8

PAHs, in interstellar H2 formation. In order to prove further that H atom abstraction reaction
from 1H-phenalene is effective at low temperatures, we plan to carry out reactions with H atoms
in a para-H2 matrix. In this medium H atoms can be produced very effectively and the diffusion
of H atoms is fast even at 2–4 K [32]. Therefore, no annealing is needed, and the interpretation
of these experiments is much more straightforward. This method was used very recently to
prove that small organic molecules, e.g. formamide, can also catalyze the interstellar H2
formation by hydrogen abstraction/addition cycles [33].
Acknowledgements
The authors thank Attila Nagy (retired from Eötvös University) for providing the
1-hydroxy-2,3-dihydro-1H-phenalene sample, and Dr Sándor Góbi (currently at University of
Coimbra) for preliminary measurements. The support of the Lendület program of the Hungarian
Academy of Sciences is acknowledged. This work was completed within the framework of the
ELTE Excellence Program (1783-3/2018/FEKUTSTRAT) supported by the Hungarian
Ministry of Human Capacities (EMMI). The work of A.S. was also supported by the ÚNKP-
19-2 New National Excellence Program of the Ministry for Innovation and Technology.
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in the online version, at
http://dx.doi.org/
9

TABLES AND FIGURES
Table 1. Tentative assignments of the experimentally observed vibrational transitions of
1H-phenalene.
Experimental
Computed
B3LYP/cc-pVTZ
ṽ /cm−1 a I / km mol−1
Assignment
(symmetry)
Ar matrix
ṽ /cm−1
720
748
750
777
802
818
827
941
Xe matrix
I
ṽ /cm−1
I
m
w
vw
s
vw
vvw
s
vw
vw
vw
vvw
vw
w
vw
w
vw
vw
vw
w
vs
m
w
718
747
750
720.5
746.8
757.4
777.5
794.7
803.9
830.7
927.7
942.3
23.4
4.9
3.9
52.9
2.1
0.4
33.7
8.9
1.5
1.0
0.2
2.1
4.5
1.7
2.7
0.9
1.1
1.3
4.1
5.2
6.1
11.3
3.1
3.3
28.3
11.8
3.5
7.9
47.3
ν54(a″)
ν53(a″)
ν36(a′)
ν52(a″)
ν35(a′)
ν34(a′)
ν51(a″)
ν33(a′)
ν48(a″)
ν46(a″)
ν45(a″)
ν32(a′)
ν31(a′)
ν29(a′)
ν26(a′)
ν44(a″)
ν24(a′)
ν22(a′)
ν19(a′)
ν17(a′)
ν14(a′)
ν12(a′)
ν11(a′)
ν10(a′)
ν9(a′)
vvs
w
778
802
vw
vvs
w
817
826
939
945
960
985
w
945
w
959
963.1
979.3
vvw
w
m
w
m
w
w
w
w
m
m
s
985
1025
1049
1104
1172
1180
1212
1309
1379
1421
1513
1593
1608
1624
2874b
2882b
3041b
3059b
3105b
a Scaled by 0.9718.
1024
1049
1104
1171
1179
1212
1310
1379
1416
1514
1593
1609
1623
2863b
2873b
3038b
3060b
3105b
1019.2
1038.9
1093.0
1172.5
1190.2
1222.3
1300.2
1382.6
1418.8
1500.6
1588.3
1601.8
1658.8
2901.3
2906.7
3060.4
3063.9
3081.4
w
w
m
vw
vvw
w
w
m
w
w
m
m
m
m
w
ν43(a″)
ν8(a′)
ν7(a′)
w
vw
ν3(a′)
b C‒H stretching modes are site split; only the most intense bands are listed.
10

Table 2. Assignments of the experimentally observed vibrational transitions of phenalenyl
radical.
Experimental
Computed
Assignment
(symmetry)
Ar matrix
Xe matrixa
B3LYP/cc-pVTZ
ṽ /cm−1
I
s
s
ṽ /cm−1
I
s
s
vw
vw
w
vw
m
ṽ /cm−1 c
I / km mol−1
754; 757b
839
754; 758b
837
753.5
843.5
115.1
23.0
5.3
ν32 (a″)
ν31 (a″)
ν24 (e′)
ν23 (e′)
ν22 (e′)
ν20 (e′)
ν17 (e′)
1036
1029.5
1094.0
1171.9
1309.4
1546.0
1120, 1118b
1180
2.6
1181
1300
vw
vw
w
11.2
3.7
11.7
1300
1546; 1548b
1545; 1547b
a
Further weak bands observed in some experiments at 1251, 1330, 1328, 1334, 1336, 1345,
1377, 1425 cm−1. Since the intensity of these bands (except for 1376.7 cm−1) was weak and
varied from experiment to experiment, their assignment is uncertain.
b Site split bands.
c Scaled by 0.9718.
11

Figure 1. a) Resonance stabilization in phenalenyl radical. b) Structure of phenalene
isomers.
12

Figure 2. a) IR spectrum of 1H-phenalene recorded after deposition in an 8‒10 K Ar matrix.
b) Difference spectrum obtained by subtracting the spectrum a from the spectrum recorded after
90 min 250 nm irradiation. c) IR spectrum of 1H-phenalene recorded after deposition in a 20 K
Xe matrix. d) Difference spectrum obtained by subtracting the spectrum c from the spectrum
recorded after 90 min 220 nm irradiation. The absorbance scale is multiplied by 3 in the region
of 1550–1010 cm–1 compared that of in the 880–700 cm–1 region. The relative intensity of the
bands marked by asterisk in spectrum d varied from experiment to experiment, their assignment
is uncertain.
13

b
a
Xe matrix
Ar matrix
18500 18000 17500 17000 16500 16000
Wavenumber / cm-1
Figure 3. Fluorescence spectrum of the phenalenyl radical in a) Ar and b) Xe matrices. The
scattering background was subtracted; the band origins were not observed due to strong laser
scattering. The excitation wavelength was 512.5 nm and 516 nm for Ar and Xe matrices,
respectively.
14

A c
A b
x5
x5
Xe matrix
A
a
840
830
820
Wavenumber / cm-1
Figure 4. a) Part of the IR spectrum recorded after 90 min 250 nm photolysis of 1H-
phenalene deposited in a 20 K Xe matrix. b) Difference spectrum obtained by subtracting
spectrum a from the spectrum recorded after warming up the matrix to 41 K. c) Difference
spectrum obtained by subtracting the spectrum recorded at 41 K from the spectrum recorded
after warming up the matrix to 50 K. The dashed lines are at A = 0. The absorbance scale of
the b and c difference spectra are multiplied by 5 compared to the scale of spectrum a. The
feature around 826 cm-1 is assigned to 1H-phenalene, while at the one around 837 cm-1 is
assigned to phenalenyl radical, see text.
15

10
5
c
b
0
-5
-10
-15
-20
a
20
25
30
35
40
45
50
Temperature / K
Figure 5. Change of the integrated absorbances as a function of temperature. a) Band of
phenalenyl radical at 837 cm–1 (integrated between 834 and 840 cm–1) observed for Xe matrix
after 250 nm photolysis. b) Band of 1H-phenalene at 826 cm–1 (integrated between 831 and 821
cm–1) observed for Xe matrix after 250 nm photolysis. c) Band of 1H-phenalene at 826 cm–1
(integrated between 831 and 821 cm–1) observed for Xe matrix without photolysis.
Figure 6. H abstraction/addition cycle between 1H-phenalene and phenalenyl radical. The
numbers above the arrows are the computed zero-point vibrational energy corrected barriers.
The number below the arrows are the reaction enthalpies at 0 K. All energies are kJ mol–1 as
computed at the B3LYP/cc-pVTZ (top), CCSD(T)/cc-pVDZ//B3LYP/cc-pVTZ (middle,
italics), and CCSD(T)/cc-pVTZ//B3LYP/cc-pVTZ levels of theory. All energies are corrected
by B3LYP/cc-pVTZ ZPVE.
16

References
[1] G.D. O’Connor, B. Chan, J.A. Sanelli, K.M. Cergol, V. Dryza, R.J. Payne, E.J. Bieske, L. Radom, T.W.
Schmidt, Hydrogen-adduction to open-shell graphene fragments: spectroscopy, thermochemistry and
astrochemistry, Chem. Sci. 8 (2017) 1186.
[2] P.B. Sogo, M. Nakazaki, M. Calvin, Free radical from perinaphthene, J. Chem. Phys., 26 (1957) 1343.
[3] R. Biehl, Ch. Hass, H. Kurreck, W. Lubitz, S. Oestreich, Esr and endor studies of partially deuterated and
chlorinated phenalenyls : New synthetic pathways, Tetrahedron, 34 (1978) 419.
[4] W.P. Cofino, S.M. van Dam, D.A. Kamminga, G.Ph. Hoornweg, C. Gooijer, C. MacLean, N.H. Velthorst,
Jahn-Teller effect in highly resolved optical spectra of the phenalenyl radical, Mol. Phys. 51 (1984) 537.
[5] G.D. O’Connor, T.P. Troy, D.A. Roberts, N. Chalyavi, B. Fückel, M.J. Crossley, K. Nauta, J.F. Stanton,
T.W. Schmidt, Spectroscopy of the Free Phenalenyl Radical, J. Am. Chem. Soc. 133 (2011) 14554.
[6] R. Pettit, The synthesis and properties of the perinapthenylium cation, J. Am. Chem. Soc. 82 (1960) 1972.
[7] D.H. Reid, Stable π-electron systems and new aromatic structures, tetrahedron, 3 (1958) 339.
[8] P. Lazzeretti, Assessment of Aromaticity via Molecular Response Properties, Phys. Chem. Chem. Phys. 6
(2004) 217.
[9] M.K. Cyrański, R.W.A. Havenith, M.A. Dobrowolski, B.R. Gray, T.M. Krygowski, P.W. Fowler, L.W.
Jenneskens, The phenalenyl motif – A magnetic chameleon, Chem. Eur. J., 13 (2007) 2201.
[10] G. Lock, G. Gergely, Über Perinaphthinden, Ber. Deutch. Chem. Ges. 77 (1944) 461.
[11] J.M. Zoellner, R.W. Zoellner, The isomers of phenalene and their singlet and triplet states: A Hartree–Fock
and density functional computational investigation, J. Mol. Struct. THEOCHEM 863 (2008) 50.
[12] E. Habart, F. Boulanger, L. Verstraete, G. Pineau des Forêts, E. Falgarone, A. Abergel, H2 infrared line
emission across the bright side of the ρ Ophiuchi main cloud, Astron. Astrophys. 397 (2003) 623.
[13] E. Habart, F. Boulanger, L. Verstraete, C.M. Walmsley, G. Pineau des Forêts, Some empirical estimates of
the H2 formation rate in photon-dominated regions, Astron. Astrophys. 414 (2004) 531.
[14] Le Page, T. P. Snow, V. Bierbaum, Molecular hydrogen formation catalyzed by polycyclic aromatic
hydrocarbons in the interstellar medium, Astrophys. J., 704 (2009) 274.
[15] A.L. Skov, J.D. Thrower, L. L. Hornekær, Polycyclic Aromatic Hydrocarbons – Catalysts for Molecular
Hydrogen Formation, Faraday Discuss., 168 (2014) 223.
[16] E. Rauls, L. Hornekær, Catalyzed routes to molecular hydrogen formation and hydrogen addition reactions
on neutral polycyclic aromatic hydrocarbons under interstellar conditions, Astrophys. J. 679 (2008) 531.
[17] M. Bonfanti, S. Casolo, G.F. Tantardini, R. Martinazzo, Surface models and reaction barrier in Eley–Rideal
formation of H2 on graphitic surfaces, Phys. Chem. Chem. Phys. 13 (2011) 16680.
[18] C. Barrales-Martínez, D. Cortés-Arriagada, S. Gutiérrez-Oliva, Molecular hydrogen formation in the
interstellar medium: the role of polycyclic aromatic hydrocarbons analysed by the reaction force and activation
strain model, Monthly Notices Roy. Astron. Soc. 481 (2018) 3052.
[19] S. Cazaux, A.G.G.M. Tielens, H2 formation on grain surfaces, Astrophys. J. 604 (2004) 222.
[20] T. P. M. Gourmans, J. Kästner, Hydrogen-atom tunneling could contribute to H2 formation to space,
Angew. Chem. Int Ed., 49 (2010) 7350.
[21] T.P.M. Goumans, Hydrogen chemisrption on polycyclic aromatic hydrocarbons via tunneling, Monthly
Notices Roy. Astron. Soc. 415 (2011) 3129.
[22] V. Menella, L. Hornekær, J. Thrower, M. Accolla, The catalytic role of coronene for molecular hydrogen
formation, Astrophys. J. Lett. 745 (212) L2.
[23] C. Barrales-Martínez, S. Gutiérrez-Oliva, The effect of heteroatoms in carbonaceous surfaces:
computational analysis of H chemisorption on to a PANH and Si-doped PAH, Monthly Notices Roy. Astron.
Soc. 490 (2019) 172.
[24] R. Pan, L.M Raff, Classical variational transition state theory study of hydrogen atom diffusion dynamics in
imperfect xeenon matrices, J. Phys. Chem. A, 101 (1997) 235.
[25] L. Khriachtchev, H. Tanskanen, M. Pettersson, M. Räsänen, V. Feldman, F. Sukhov, A. Orlov, A.F.
Shestakov, Isotopic effect on thermal mobility of atomic hydrogen in solid xenon, J. Chem. Phys. 116 (2002)
5708.
[26] V. Boekelheide, C. E. Larrabee, An investigation of the preparation and some properties of perinaphthene, J.
Am. Chem. Soc. 72 (1950) 1245.
[27] C.Y. Peng, P.Y. Ayala, H.B. Schlegel, M.J. Frisch, Using redundant internal coordinates to optimize
equilibrium geometries and transition states, J. Comp. Chem. 17 (1996) 49.
[28] C.Y. Peng, H.B. Schlegel, Combining synchronous transit and quasi-Newton methods for finding transition
states, Israel J. Chem., 33 (1993) 449.
[29] M.J. Frisch, et al., Gaussian 09 D.01, Gaussian, Inc., Wallingford CT, 2009.
17

[30] M. Pettersson, J. Lundell, M. Räsänen, Neutral rare‐gas containing charge‐transfer molecules in solid
matrices. II. HXeH, HXeD, and DXeD in Xe, J. Chem. Phys., 103 (1195) 205.
[31] M. Pettersson, L. Khriachtchev, S. Jolkkonen, M. Räsänen,, Photochemistry of HNCO in solid Xe: channels
of UV photolysis and creation of H2NCO radicals, J. Phys. Chem. A, 103 (1999) 9154.
[32] M. Bahou, P. Das, Y.F. Lee, Y.J. Wu, Y.P. Lee, Infrared spectra of free radicals and protonated species
produced in parahydrogen matrices, Phys. Chem. Chem. Phys., 16 (2014) 2200.
[33] K.A. Haupa, G. Tarczay, Y.P. Lee, Hydrogen abstraction/addition tunneling reactions elucidate the
interstellar H2NCHO/HNCO ratio and H2 formation, J. Am. Chem. Soc. 141 (2019) 11614.
18

 IR spectrum of 1H-phenalene was recorded in Ar and Kr matrices.
 Phenalenyl radical was generated by laser UV photolysis from H-phenalene.
 Phenalenyl radical recombines with an H atom upon warming up the Xe matrix.
 Computations indicate that H abstraction reaction from 1H-phenalene is also feasible.
 The investigated molecules are effective catalysts in interstellar H2 formation.
19

CRediT author statement
Anita Schneiker: Sample Praparation, Measurements, Computations, Data Analysis,
Reviewing and Editing; István Pál Csonka: Sample Praparation, Measurements, Data
Analysis, Reviewing and Editing; György Tarczay: Measurements, Computations, Data
Analysis, Writing the paper, Reviewing and Editing, Supervision
Declaration of interests
☒ The authors declare that they have no known competing financial interests or personal
relationships that could have appeared to influence the work reported in this paper.
☐The authors declare the following financial interests/personal relationships which may be
considered as potential competing interests:
20