A Short and Efficient Synthesis of the [3]Triangulene Ring System

Article abstract of DOI:10.1002/anie.201907226

Triangulenes are of current interest for potential applications in molecular electronics. We describe here a three step synthesis of the 4,8,12-trihydro[3]triangulenium cation by cascade cyclization of a tetra-benzyl alcohol precursor in triflic acid solu

Full text of DOI:10.1002/anie.201907226

Angewandte
Communications
Chemie
International Edition: DOI: 10.1002/anie.201907226
German Edition: DOI: 10.1002/ange.201907226
Polycyclic Hydrocarbons
A Short and Efficient Synthesis of the [3]Triangulene Ring System
Carter J. Holt, Katelyn J. Wentworth, and Richard P. Johnson*
Abstract: Triangulenes are of current interest for potential
applications in molecular electronics. We describe here a three
step synthesis of the 4,8,12-trihydro[3]triangulenium cation by
cascade cyclization of a tetra-benzyl alcohol precursor in triflic
acid solution. This stable carbocation is easily observed by
NMR and optical spectroscopy and is highly fluorescent.
Quenching of the cation into basic solutions or by hydride
transfer from triethylsilane provides access to stable dihydro
and tetrahydro[3]triangulenes. These neutral species intercon-
vert with cations in a complex series of proton and hydride
transfers. This route provides several important [3]triangulene
precursors. Preliminary experiments designed to generate
[3]triangulene in the solution phase provide evidence for its
formation and rapid oligomerization.
effort in this field, the parent substance 2, commonly known
as triangulene or Clarꢀs hydrocarbon, was only recently
prepared and observed as single molecules by scanning probe
microscopy.[31] Similar experiments describing the molecular-
scale preparation of higher homologues 3 and 4 by surface-
assisted dehydrogenation of suitable precursors have been
reported very recently.[32,33]
Further development in the evolving triangulene field
requires efficient syntheses. Earlier synthetic routes to the all-
carbon ring system of 2 are relatively lengthy;[31] improve-
ments in the Clar route to triangulene precursors have been
reported as this work was under review.[34] We report here
a very short and efficient synthesis of the [3]triangulene ring
system by cascade cyclization that provides multiple precur-
sors to 2 and potential derivatives. Preliminary efforts to
generate 2 in the solution phase are also described.
Scheme 1 summarizes our route to the [3]triangulene ring
system. Taking advantage of the 3-fold symmetry of this
structure, condensation of three equivalents of protected
benzyl alcohol 5 with diethyl carbonate afforded 6 in a single
synthetic step and high yield. Palladium catalyzed deprotec-
tion yielded tetraol 7. Sequential ring closing strategies have
been reported earlier but we envisioned a cascade cyclization
of benzyl cations. It is noteworthy that the 4,8,12-trioxytrian-
gulenium cation was prepared in 1964 by Martin and Smith
through a structurally similar cascade cyclization forming new
aryl ether linkages.[23]
T
riangular-shaped aromatic hydrocarbons (Figure 1) com-
prise a homologous series of open shell structures, beginning
with the well-studied phenalenyl radical (1).[1–3] Higher
homologs 2, 3 and 4 are expected to exist as di-, tri- and
tetraradicals, respectively; all have been predicted to possess
high spin ground states.[4–9] Synthetic attempts to make
[3]triangulene (2) were first described by Clar in 1953[10–12]
although the first example of a triangulene ring system had
been reported much earlier by Weiss.[13,14]
Figure 1. The [n]triangulene series (n=rings on a side).
Among milestones in triangulene chemistry, the aromatic
[3]triangulene dianion has been prepared from the Clar
route.[15] Synthesis and ESR spectroscopy of 2,6,10-tri-tert-
butyl triangulene confirmed the existence of a kinetically
unstable ground state triplet.[16] 4,8,12-Trioxotriangulenes are
well known, including the stable radical.[17,18] There is a grow-
ing literature predicting the chemistry of 2[19–22] and a variety
of heterocyclic versions are known.[23–30] In spite of much
Scheme 1. Synthesis of the 4,8,12-trihydro[3]triangulenium ion.
Attempts at cyclization of 7 with catalysts and acidic
media known to promote benzyl cation chemistry afforded
incomplete cyclization or mixtures of the desired products
and oligomers, often with poor mass balances. As one
example, exposure of 7 to polyphosphoric acid/water at
1008C gave nearly quantitative conversion to 8. This result
supports initial ionization at the central hydroxyl group and
capture of the cation by one of the proximate benzyl alcohols.
An NMR experiment provided the key to complete
cyclization. When tetraol 7 was added to an NMR tube
[*] C. J. Holt, K. J. Wentworth, Prof. Dr. R. P. Johnson
Department of Chemistry, University of New Hampshire
23 Academic Way, Durham, NH 03824 (USA)
E-mail: Richard.johnson@unh.edu
Supporting information and the ORCID identification number(s) for
the author(s) of this article can be found under https://doi.org/10.
1002/anie.201907226.
Angew. Chem. Int. Ed. 2019, 58, 15793 –15796
ꢀ 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
15793

Angewandte
Communications
Chemie
containing 1m TfOD in DCE-d4 at ambient temperature, the
dark green solution immediately exhibited very simple 1H and
13C NMR spectra, displaying three unique hydrogens and six
carbon signals for the major species in solution. These data
support formation of the symmetrical 4,8,12-trihydro-
[3]triangulenium ion (9). DFT computations on 9 predict
a D3h symmetrical structure, with calculated NMR chemical
shifts in good agreement with experiment. NMR samples of
cation 9 were stable for > 24 h. As expected, exposure of the
partially cyclized structure 8 to the same acidic reaction
conditions led cleanly to 9.
Optical spectroscopy of 9 also supports this structure. The
absorption spectrum in TfOH/DCE (lmax = 404.5 and
463.5 nm), is similar but shifted to longer wavelength relative
to the triphenylmethyl cation (lmax = 411.5 and 433.5 nm),
consistent with improved conjugation in the planar structure.
TDDFT calculations correctly predict intensities for the two
longest wavelength absorption bands but computed transition
energies are too high, as previously seen for cations.[35] These
two absorption bands arise due to electronic excitation to the
LUMO from pairs of degenerate bonding orbitals. Cation 9
showed strong fluorescence, with a lifetime of ca. 5 ns, as
measured by single photon counting. By comparison, the
more flexible triphenylmethyl cation showed much weaker
fluorescence, as judged by photon intensity and a higher
signal/noise ratio.
characterized by its very simple 1H and 13C NMR spectra. The
central methine hydrogen in 10 appears as a quartet (J =
5.5 Hz) in the NMR spectrum because of long range coupling
to the three pseudoaxial methylene hydrogens. 5J Homoallylic
coupling has been described previously in 1,4-cyclohexa-
dienes.[37] The predicted structure of 10 from DFT computa-
tions is modestly bowl-shaped. Neutralization of 9 by addition
to saturated aqueous NaHCO3 afforded a mixture of 11 and
12, while slow addition to Et3N gave a 66% isolated yield of
pure 12.
Olympicene is known to undergo air oxidation at the
methylene carbon[36] and thus 11 is a logical precursor to
ketone 13, a substance isolated by Clar in his initial synthetic
study.[12] Consistent with this expectation, exposure of sol-
utions of 11 to air over 24 h (Scheme 3) led to partial
conversion to ketone 13. Air oxidation of 12 has been
reported previously[31] but the product was not identified. We
find that diketone 14 is the major product. This oxidation is
more efficiently carried out by reaction of 12 at ambient
temperature with excess TEMPO, giving 14 in 44% yield after
chromatography.
We next turned to larger scale reactions (Scheme 2,
Table 1), with the goal of preparing stable dihydro and
tetrahydro[3]triangulenes. After preparation of cation 9 by
the method described above, neutralization by hydride trans-
fer during slow addition to triethylsilane in DCM produced
a mixture composed primarily of hydrocarbons 10–12 in 96%
isolated yield. The three components were separable by flash
chromatography, providing pure samples of each hydrocar-
bon. Dihydrotriangulene 12 has been described previously as
a precursor to 2[31] and its dianion[15] but has only been
prepared according to the longer synthetic route of Clar.[11,12]
Our spectral data match those reported earlier. 1,2,3,8-
Tetrahydrotriangulene (11) is a new derivative of olympi-
cene,[36] possessing one additional saturated ring. Isomeric
tetrahydrotriangulene 10 is also a new substance, confidently
Scheme 3. Benzylic oxidation.
The products in Table 1 suggest a complex series of proton
and hydride transfers, which pass through triangulenium ion 9
and other cations. Scheme 4 summarizes likely connections,
along with relative energies of isomeric neutral and cationic
species, as estimated by DFT computations.
To further understand these interconversions, we exposed
pure samples of 10, 11 and 12 to TfOD in DCE-d4, monitoring
reactions by NMR spectroscopy and using DFT computations
to predict the NMR spectra of likely cationic intermediates.
Results of initial reaction were dependent on TfOD concen-
tration. Exposure of 10 to TfOD/DCE-d4 led to cation 9, the
presumed result of central hydride abstraction. A similar
experiment with 12 gave a mixture of 9 (88%) and two other
cations believed to be 16 (9%) and 18 (3%). Cation 16 is the
likely product of hydride abstraction from phenylanthracene
15, a substance not observed here although it was described
earlier by Clar.[10–12] Structure 15 is the expected product of
initial deprotonation of 9, an easily reversible process. We
have shown earlier that anthracene is fully protonated in 1m
TfOH solution[38] and 15 should be more basic than anthra-
cene. Cation 16 is C4 protonated triangulene 2. Upon
standing for 72 h, the solution led to a mixture of 9 (92%)
Scheme 2. Cation conversion to dihydro- and tetrahydrotriangulenes.
Table 1: Synthesis of dihydro- and tetrahydro[3]triangulenes from the
4,8,12-trihydrotriangulenium cation (9).
Reduction
conditions
10
11
12
Total yield
Et3SiH/DCM
Sat. NaHCO3
Et3N/DCM
38%
–
–
9%
27%
–
49%
53%
66%
96%
80%
66%
15794 www.angewandte.org
ꢀ 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2019, 58, 15793 –15796

Angewandte
Communications
Chemie
Diverse heterocyclic analogues are known but the absence of
an efficient synthesis for the parent hydrocarbon has limited
exploration and applications of [3]triangulenes. Clarꢀs initial
quest for the parent hydrocarbon 2[12] was only fulfilled on
a molecular scale by Pavlicek and co-workers in 2017.[31] The
first molecular-scale preparations of higher homologues [4]-
and [5]triangulene have very recently been reported as
further milestones[32,33] in this rapidly evolving field. Di- and
tetrahydro[3]triangulenes are now more available through
this work and a parallel study reported recently by Jurícek
and co-workers.[34] Our preliminary efforts to generate
[3]triangulene in the solution phase provide evidence for
rapid oligomerization, as might be expected for this highly
reactive substance. We believe it should be possible to tame
the solution phase reactivity of 2 through appropriate
chemical traps.
Acknowledgements
Scheme 4. Interconnections of neutral and cationic species, with
relative free energies (IEFPCM(DCE)/B3LYP/6-31+G(d,p) for isomers.
We are grateful for support from National Science Founda-
tion award CHE-1362519, to Professor Christine Caputo for
assistance with the fluorescence experiments and to Dr.
Sergei Dikler (Bruker Daltonics, Billerica, MA) for measur-
ing the LDI-TOF spectra. This research used the Extreme
Science and Engineering Discovery Environment (XSEDE),
supported by NSF Grant No. OCI-1053575.
and 18 (8%). This same aromatic olympicenium ion[39] (18) is
formed cleanly by reaction of 11 with TfOD.
Observed chemical shifts for the cations are in good
agreement with computation. According to the DFT free
energies, aromatic cation 18 is the most stable C22H15+ isomer
in this series. Consistent with this prediction, an NMR sample
of this cation is stable for over 2 months. During the same
period, an NMR sample of higher energy isomer 9 has
converted ca. 66% to 18. Among neutral structures, 12 is
predicted to be the most stable dihydrotriangulene; this may
explain its selective formation from reaction of 9 with
triethylamine, likely by the sequence 9!15!16!12. Full
characterization of these fascinating interconversions will
require more detailed study.
Conflict of interest
The authors declare no conflict of interest.
Keywords: carbocations ·cyclization ·polycyclic hydrocarbons ·
superacids ·triangulene
How to cite: Angew. Chem. Int. Ed. 2019, 58, 15793–15796
Angew. Chem. 2019, 131, 15940–15943
As a preliminary exploration of the solution-phase
chemistry of 2, the reaction of dihydrotriangulene 12 with p-
chloranil (2 equivalents) was monitored by 1H NMR and
optical spectroscopy. The UV-VIS spectrum showed a new
lmax at 628 nm, which decayed over 24 h. This is unlikely to be
2 which is predicted by TDDFT calculations (SI) to have
a longest wavelength triplet-triplet absorption between 400–
500 nm. NMR analysis showed only oxidation product 19
remaining in solution. LDI-TOF analysis of the solid reaction
product revealed a complex spectrum (SI Figure 2) with
masses stretching up to ca.1400 Da. We believe this result
supports the formation of higher oligomers of 2.
[1] T. Kubo, Chem. Rec. 2015, 15, 218 –232.
[2] I. Ratera, J. Veciana, Chem. Soc. Rev. 2012, 41, 303 –349.
[3] A. Mukherjee, S. C. Sau, S. K. Mandal, Acc. Chem. Res. 2017, 50,
1679 –1691.
[4] M. J. Bearpark, M. A. Robb, F. Bernardi, M. Olivucci, Chem.
Phys. Lett. 1994, 217, 513 –519.
[5] M. R. Philpott, F. Cimpoesu, Y. Kawazoe, Chem. Phys. 2008, 354,
1 –15.
[6] O. A. Gapurenko, A. G. Starikov, R. M. Minyaev, V. I. Minkin,
Russ. Chem. Bull. 2011, 60, 1517 –1524.
[7] A. Das, T. Muller, F. Plasser, H. Lischka, J. Phys. Chem. A 2016,
120, 1625 –1636.
[8] V. Sharma, N. Som, S. D. Dabhi, P. K. Jha, ChemistrySelect 2018,
3, 2390 –2397.
The history of the [3]triangulene ring system now spans
nearly a century, with the first example reported in 1925.[14]
[9] A. M. Toader, C. M. Buta, B. Frecus, A. Mischie, F. Cimpoesu, J.
Phys. Chem. C 2019, 123, 6869 –6880.
[10] E. Clar, Chimia 1964, 18, 375 –386.
[11] E. Clar, D. G. Stewart, J. Am. Chem. Soc. 1954, 76, 3504 –3507.
[12] E. Clar, D. G. Stewart, J. Am. Chem. Soc. 1953, 75, 2667 –2672.
[13] R. Weiss, F. Muller, Monatsh. Chem. 1934, 65, 129 –136.
[14] R. Weiß, J. Korczyn, Monatsh. Chem. 1925, 45, 207 –214.
[15] O. Hara, K. Tanaka, K. Yamamoto, T. Nakazawa, I. Murata,
Tetrahedron Lett. 1977, 18, 2435 –2436.
Scheme 5. Attempt to generate [3]triangulene in solution.
Angew. Chem. Int. Ed. 2019, 58, 15793 –15796
ꢀ 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 15795

Angewandte
Communications
Chemie
[16] J. Inoue, K. Fukui, T. Kubo, S. Nakazawa, K. Sato, D. Shiomi, Y.
Morita, K. Yamamoto, T. Takui, K. Nakasuji, J. Am. Chem. Soc.
2001, 123, 12702 –12703.
[17] G. Allinson, R. J. Bushby, J.-L. Paillaud, M. Thornton-Pett, J.
Chem. Soc. Perkin Trans. 1 1995, 385 –390.
[29] J. Bosson, J. Gouin, J. Lacour, Chem. Soc. Rev. 2014, 43, 2824 –
2840.
[30] J. E. Field, D. Venkataraman, Chem. Mater. 2002, 14, 962 –964.
[31] N. Pavlicek, A. Mistry, Z. Majzik, N. Moll, G. Meyer, D. J. Fox, L.
Gross, Nat. Nanotechnol. 2017, 12, 308 –311.
[18] Y. Morita, T. Murata, A. Ueda, C. Yamada, Y. Kanzaki, D.
Shiomi, K. Sato, T. Takui, Bull. Chem. Soc. Jpn. 2018, 91, 922 –
931.
[32] S. Mishra, D. Beyer, K. Eimre, J. Liu, R. Berger, O. Grçning,
C. A. Pignedoli, K. Müllen, R. Fasel, X. Feng, P. Ruffieux, J. Am.
Chem. Soc. 2019, 141, 10621 –10625.
[19] Q. Wang, J. Li, Y. Nie, F. Xu, Y. Yu, B. Wang, Phys. Chem. Chem.
Phys. 2018, 20, 15736 –15745.
[20] V. Sharma, S. D. Dabhi, S. Shinde, P. K. Jha, AIP Conf. Proc.
2018, 1961, 0300311 –0300316.
[21] Z. Mou, M. Kertesz, Chem. Eur. J. 2018, 24, 6140 –6147.
[22] H. Jin, J. Li, T. Wang, Y. Yu, Carbon 2018, 137, 1 –5.
[23] J. C. Martin, R. G. Smith, J. Am. Chem. Soc. 1964, 86, 2252 –
2256.
[24] M. Rosenberg, M. Santella, S. A. Bogh, A. Vinas Munoz,
H. O. B. Andersen, O. Hammerich, I. Bora, K. Lincke, B. W.
Laursen, J. Org. Chem. 2019, 84, 2556 –2567.
[33] J. Su, M. Telychko, P. Hu, G. Macam, P. Mutombo, H. Zhang, Y.
Bao, F. Cheng, Z.-Q. Huang, Z. Qiu, S. J. R. Tan, H. Lin, P.
Jelínek, F.-C. Chuang, J. Wu, J. Lu, Sci. Adv. 2019, 5, eaav7717.
[34] P. Ribar, T. Solomek, M. Jurícek, Org. Lett. 2019, https://doi.org/
10.1021/acs.orglett.9b02683.
[35] J. Dominikowska, M. Domagala, M. Palusiak, J. Chem. Phys.
2014, 140, 044324.
[36] A. Mistry, B. Moreton, B. Schuler, F. Mohn, G. Meyer, L. Gross,
A. Williams, P. Scott, G. Costantini, D. J. Fox, Chem. Eur. J. 2015,
21, 2011 –2018.
[37] L. J. Durham, J. Studebaker, M. J. Perkins, Chem. Commun.
1965, 456 –457.
[25] I. Hernandez Delgado, S. Pascal, C. Besnard, S. Voci, L.
Bouffier, N. Sojic, J. Lacour, Chem. Eur. J. 2018, 24, 10186 – [38] S. L. Skraba-Joiner, J. W. Brulet, M. K. Song, R. P. Johnson, J.
10195.
Org. Chem. 2017, 82, 13076 –13083.
[26] S. Nakatsuka, H. Gotoh, K. Kinoshita, N. Yasuda, T. Hata-
keyama, Angew. Chem. Int. Ed. 2017, 56, 5087 –5090; Angew.
Chem. 2017, 129, 5169 –5172.
[39] D. H. Reid, W. Bonthrone, J. Chem. Soc. 1965, 5920 –5926.
[27] S. A. Bogh, M. Simmermacher, M. Westberg, M. Bregnhoej, M.
Rosenberg, L. De Vico, M. Veiga, B. W. Laursen, P. R. Ogilby,
S. P. A. Sauer, T. J. Soerensen, ACS Omega 2017, 2, 193 –203.
[28] N. Hammer, T. A. Schaub, U. Meinhardt, M. Kivala, Chem. Rec.
2015, 15, 1119 –1131.
Manuscript received: June 10, 2019
Revised manuscript received: September 2, 2019
Accepted manuscript online: September 5, 2019
Version of record online: September 24, 2019
15796 www.angewandte.org
ꢀ 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2019, 58, 15793 –15796