Semi-fused hexaphenyl hexa-peri-hexabenzocoronene: a novel fluorophore from an intramolecular Scholl reaction

Article abstract of DOI:10.1016/j.tetlet.2009.04.103

A semi-fused hexaphenyl hexabenzocoronene derivative (hexaphenyl-1/2HBC) was discovered and isolated from an intramolecular Scholl reaction of an oligo(ethyleneglycol) substituted hexabiphenylbenzene. This new derivative exhibits blue-shifted λ_em

Full text of DOI:10.1016/j.tetlet.2009.04.103

Tetrahedron Letters 50 (2009) 4071–4077
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Semi-fused hexaphenyl hexa-peri-hexabenzocoronene: a novel fluorophore
from an intramolecular Scholl reaction
*
Yunyi Lu, Jeffrey S. Moore
Roger Adams Laboratory, Departments of Chemistry and Materials Science & Engineering, University of Illinois at Urbana-Champaign, Urbana, IL 61801, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
A semi-fused hexaphenyl hexabenzocoronene derivative (hexaphenyl-1/2HBC) was discovered and iso-
lated from an intramolecular Scholl reaction of an oligo(ethyleneglycol) substituted hexabiphenylben-
zene. This new derivative exhibits blue-shifted kem and significantly enhanced fluorescence in
dichloromethane compared to the completely cyclized hexaphenyl-HBC derivative.
Ó 2009 Elsevier Ltd. All rights reserved.
Received 22 February 2009
Revised 13 April 2009
Accepted 24 April 2009
Available online 3 May 2009
Keywords:
Scholl condensation
Dehydrogenation
Fluorophore
Polycyclic aromatic hydrocarbons
Hexa-peri-hexabenzocoronenes (HBCs) are an interesting class
of polycyclic aromatic hydrocarbon molecules that can be regarded
as disc-like, small graphene sections. Prompted by their interesting
structures and properties, HBC derivatives have aroused intense
interest in synthesis and applications in organic electronics. Their
conjugated ‘superbenzene’ surfaces tend to strongly aggregate into
columnar nanostructures with enhanced charge carrier mobility
compared to other discotic liquid crystalline materials.1
tion intermediates 5–6 (Scheme 1b).12 Steric hindrance and the
stability of the radical cation were suggested to render the prefer-
ential formation of 5 and 6, respectively. A recent report on the
preparation of nitrogen-functionalized HBCs involving pyrimidines
showed that the metal catalyst might interact with the substrate to
terminate the Scholl reaction halfway: condensation of 1,2-dipy-
rimidyl-3,4,5,6-tetra(4-tert-butylphenyl)benzene 7 in FeCl3/MeNO2
provided fused-ring species 8 which cannot be further converted
to HBC analog 9 (Scheme 1c).13 Further, several reports indicated
that only properly designed hexaphenylbenzene precursors can
undergo Scholl condensation to give alkoxy-substituted HBCs.14–17
Due to the lack of generality of Scholl reaction conditions to pro-
duce different HBCs and elusiveness of the reaction mechanisms,
reaction intermediates are of great significance to advance our
understanding and to guide the synthetic formulation of HBC
molecules and HBC-based conjugated polymers. In this Letter, we
report on the synthesis and photophysical properties of the unprec-
edented HBC intermediate 11 from Scholl condensation of an all-
carbon hexabiphenylbenzene precursor 10 (Scheme 2).
Recently, during our investigation of Scholl reaction to synthe-
size amphiphilic hexaphenyl-HBC 12 using FeCl3/MeNO2, we found
that a compound 11 with molecular weight corresponding to three
newly formed carbon–carbon bonds (hexaphenyl-1/2HBC) was al-
ways observed by MALDI-MS. This interesting finding prompted us
to isolate 11 and study the reaction to further investigate its nature
and unusual product formation. The oxidative cyclodehydrogena-
tion of precursor 10 (50 mg) was conducted under concentrated
(2.6 mM), medium (0.6 mM), and dilute (0.4 mM) conditions, using
45 equiv of FeCl3. When the highest concentration (2.6 mM) was
applied, after 45 min, the starting material 10 was found to
Harsh reaction conditions2 were used to prepare HBC molecules
until Müllen and co-workers demonstrated that mild Scholl con-
densation3 conditions, either CuCl2/AlCl3/CS2 or FeCl3/CH3NO2,
can be used to prepare HBCs conveniently from hexaphenylben-
zene precursors.4,5 This protocol allowed their subsequent break-
through syntheses of extended 2D graphene nanosheets.6,7 The
reaction mechanism, however, was not clarified. In the chemical
literature, two possible reaction mechanisms were considered for
the Scholl reaction, including the arenium cation mechanism and
radical cation mechanism.8 Aside from that, only few computa-
tional studies of the two mechanisms have been reported.9,10 The
aryl–aryl dehydrogenative coupling reaction was concluded to oc-
cur intramolecularly in separate steps, as evidenced experimen-
tally from isolation of reaction intermediate 2 with four new
carbon–carbon bonds in 10% yield (Scheme 1a).5 So far, this has
been the only report of isolated intermediate from Scholl conden-
sation of an all-carbon hexaphenylbenzene precursor 1.11 In an-
other case, Scholl condensation of programmed all-carbon
oligophenylene precursors 3–4 generated two partially fused reac-
* Corresponding author. Tel.: +1 217 244 4024; fax: +1 217 244 8024.
E-mail address: jsmoore@illinois.edu (J.S. Moore).
0040-4039/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2009.04.103

4072
Y. Lu, J. S. Moore / Tetrahedron Letters 50 (2009) 4071–4077
(a)
CuCl2/AlCl3
CS2
CuCl2/AlCl3
CS2
1
2
X
(b)
FeCl3/MeNO2
CH2Cl2
FeCl3/MeNO2
CH2Cl2
-Bu
-
-Bu
-Bu
-Bu
-Bu
t
t
t Bu
t
t
t
3
5
I
I
I
FeCl3/MeNO2
CH2Cl2
FeCl3/MeNO2
CH2Cl2
X
I
I
I
I
I
I
6
4
(c)
t-Bu
t-Bu
t-Bu
t-Bu
t-Bu
t-Bu
t-Bu
t-Bu
t-Bu
t-Bu
t-Bu
t-Bu
t-Bu
CuCl2/AlCl3
CS2
FeCl3/MeNO2
CH2Cl2
N
N
N
t-Bu
N
N
t-Bu
N
N
N
N
N
N
N
t-Bu
t-Bu
t-Bu
9
8
7
X
Scheme 1. HBC intermediates in the Scholl condensation reactions.
OR
OR
OR
RO
OR
RO
OR
RO
OR
FeCl3/MeNO2
°
CH2Cl2, 25
C
RO
OR
RO
OR
RO
OR
OR
10
OR
11
OR
12
R = -(CH2CH2O)3CH3
Scheme 2. Syntheses of hexaphenyl-1/2HBC 11 and hexaphenyl HBC 12.
transform primarily into completely cyclized hexaphenyl HBC 12 (
m/z = 1952) along with oligomer products from intermolecular
condensation as well as a minor amount of incompletely cyclized
mixture product (m/z = 1958 amu) being the only clearly identified

Y. Lu, J. S. Moore / Tetrahedron Letters 50 (2009) 4071–4077
4073
a
9.0
8.5
8.0
7.5
7.0
ppm
4
10
9
8
7
6
5
3
ppm
4.4
4.2
4.0
3.8
3.6
3.4
ppm
b
10
9
8
7
6
5
4
3
ppm
Figure 1. (a) 1H NMR spectrum of hexaphenyl-1/2HBC 11 (CD2Cl2, 500 MHz, 20 °C) after purification; (b) 1H NMR spectrum of hexaphenyl-HBC 12 (CDCl3, 500 MHz, 20 °C).
species above the detection limit of MALDI. We found that the
transformation into oligomer products can be eliminated by
decreasing the substrate concentration during the reaction. It
was possible to separate the incompletely cyclized product and
precursor residues from 12 by silica gel column chromatography
using a more polar solvent (CH2Cl2/CH3OH, 49/1, v/v), while by
Figure 2. Isotopic envelop of (a) hexaphenyl-1/2HBC 11 and (b) hexaphenyl-HBC 12 by MALDI-MS.

4074
Y. Lu, J. S. Moore / Tetrahedron Letters 50 (2009) 4071–4077
OR
C2
OR
RO
a
1
6 3
4
5
10
6
9
5
7
b
c
b
2
a
1
8
c
7
8
2
OR
4
RO
6.50
7.00
7.50
8.00
8.50
3
9
10
OR
R = -(CH2CH2O)3CH3
9.00
ppm (t1)
9.00
8.50
8.00
7.50
7.00
6.50
ppm (t2)
Figure 3. 1H–1H COSY NMR spectrum (CD2Cl2, 500 MHz, 20 °C) of hexaphenyl-1/2HBC 11.
switching to a carefully performed chromatography using a less
polar solvent (CH2Cl2/acetone, 3/1, v/v) it was possible to separate
the m/z = 1958 amu product from the precursor residues. An
incompletely cyclized product with m/z of 1958 amu was also ob-
tained from medium and dilute concentration reactions in 17–20%
yield. Medium concentration also provided completely cyclized 12
in 44–60% yield.
MALDI-MS (Fig. 2), and 1H–1H COSY (Fig. 3) experiments. The sin-
gle isotopic envelop of 11 at m/z 1958.00 (M)+ and 1981.00
(M+Na)+ compared to that of hexaphenyl-HBC 12 at m/z 1951.62
(M)+ and 1974.58 (M+Na)+ revealed a mass difference of 6, which
is in agreement with dehydrogenation of 6 protons due to three
carbon-carbon bond formation. The covalent connectivity of com-
pound 11 was unambiguously determined by 1H–1H COSY NMR.
Figure 3 shows the enlarged aromatic areas of the COSY NMR spec-
trum. There are only three aromatic singlets (d 9.09, d 8.98, and d
The incompletely cyclized product was a pure substance whose
structure was determined to be 11 using 1H NMR (Fig. 1), 13C NMR,
OR
OR
OR
RO
RO
OR
OR
OR
RO
RO
OR
OR
RO
RO
OR
OR
OR
OR
11a
11b
11c
R = -(CH2CH2O)3CH3
D3h
Ch
C2v
Figure 4. Space-filling models of three possible constitutional isomers of 11 where methoxy oligophenylenes were used for calculation simplicity. The conformations shown
are energy-minimized structures.18

Y. Lu, J. S. Moore / Tetrahedron Letters 50 (2009) 4071–4077
4075
tional isomers of 11 in Figure 4, the only qualified structure is
semi-fused isomer 11(a) that possesses a C2 symmetry axis. These
three singlets integrate to six hydrogens, and the remaining ten
aromatic doublet signals integrate to 36 hydrogens in total. The
resonance correlated doublets at d 7.97 and d 7.35 each integrate
to two hydrogens and differ from the other eight doublets each
integrating to four hydrogens. These two doublets correspond to
the two sets of neighboring hydrogens of the fused ring, while
the eight doublets were assigned to the hydrogens of the eight
uncyclized, free-rotating phenyl rings. The NOESY1D, CH COSY
(HMBC-HMQC), and 135o DEPT NMR experiments made it possible
to completely assign the aromatic protons of 11 based on the pro-
posed structure. Key correlations of singlets a, b, c and doublets 1,
3–6 are shown in Figure 5. The complete assignments of all aro-
matic protons were summarized in Figure 3. The proposed C2v
symmetry of 11 was also confirmed by the observation of three
sets of alkoxy signals for the oligo(ethylene glycol)ether chains as
shown in Figure 1.
c
c
b
a
1
b
6
9.5
9.0
8.5
8.0
7.5
7.0
6.5
PPM
3
5
4
Figure 5. NOESY1D NMR spectrum (CDCl3, 500 MHz, 20 °C) of hexaphenyl-1/2HBC
11 (the colored resonance regions were irradiated selectively).
To better understand the formation of semi-fused structure 11,
modeling studies were performed to calculate the relative stabili-
ties of three constitutional isomers 11(a)–(c) (Table 1, Fig. 4). Re-
sults indicate that isomer 11(a) has the lowest energy compared
to isomers 11(b) and 11(c). Therefore, the observed product is also
the thermodynamically most favored product among three
isomers.
Table 1
Approximate energy of three constitutional isomers of 11 based on MM2
calculations18
Isomer
À
122.8
130.1
140.5
DH (kcal/mol)
11(a)
11(b)
11(c)
The unexpected discovery of hexaphenyl-1/2HBC 11 also
prompted us to further investigate whether 11 is a kinetically fa-
vored precursor of hexaphenyl-HBC 12 generated during the step-
wise cyclodehydrogenation reaction by treatment with FeCl3, or
whether 11 is a thermodynamically stable product formed inde-
8.82) deshielded to the lowest field, corresponding to the isolated
ortho-hydrogens of the fused ring. Among three possible constitu-
(a) 70 min
10
(b) 80 min
9
8
7
6
5
4
ppm
10
(c) 90 min
9
8
7
6
5
4
ppm
10
9
8
7
6
5
4
ppm
Figure 6. Comparison of 1H NMR spectra of crude condensation reaction mixtures upon MeOH quenching for cyclodehydrogenation of 10 after 70 min, 80 min, and 90 min,
respectively (CDCl3, 500 MHz, 20 °C). Condensation conditions: 50 mg 10, 1.13 mmol FeCl3, 4.8 mL MeNO2, 40 mL CH2Cl2, 25 °C; quenching with 40 mL MeOH.

4076
Y. Lu, J. S. Moore / Tetrahedron Letters 50 (2009) 4071–4077
0.8
0.6
0.4
0.2
0.0
1.2
1.0
10
11
12
10
11
12
0.8
0.6
0.4
0.2
0.0
200
300
400
500
600
700
800
Wavelength (nm)
200
250
300
350
400
450
500
550
Wavelength (nm)
Figure 8. Normalized fluorescence spectra of 10–12 in dichloromethane (5.1 lM,
25 °C). The excitation wavelengths were set to the kmax (abs) of each compound (10,
282 nm; 11, 341 nm; 12, 377 nm).
Figure 7. UV–vis absorption spectra of 10–12 in dichloromethane at 25 °C. The
spectra are normalized to 5.1 lM concentration.
two bands provide the characteristic vibronic spectral pattern
responsible for the chromophore nature and planarity of 11,
respectively.19,20 The absorption maxima around 272 nm is due
to the oligophenylene chromophore, and the absorption in the
other UV region around 342 nm is due to the planarization of the
oligophenylene core to generate the semi-fused ring with added
conjugation. Similarly, in the case of 12, the absorption in the UV
region (230–325 nm) represents the oligophenylene chromophore.
Table 2
Absorption maxima and extinction coefficients of 10–12
Compd
kmax (abs), nm
e
, MÀ1 cmÀ1
10
229.0
282.0
6.90 Â 104
1.41 Â 105
11
12
205.5
234.0
271.5
341.5
389.5
402.5
6.47 Â 104
8.61 Â 104
1.23 Â 105
1.47 Â 105
4.12 Â 104
4.12 Â 104
The next absorption maxima of 12 (377 nm, 1.24 Â 105 MÀ1 cmÀ1
)
is bathochromically shifted compared to that of 11 (342 nm,
1.47 Â 105 MÀ1 cmÀ1) with a slight decrease in the absorption
intensity upon extension of the conjugation. For comparison, the
absorption spectrum of the reference compound 10 is also shown
in Figure 7, which contains only one distinct UV band (240–
350 nm) corresponding to its nonplanar polyphenylene chromo-
phore as that in 11 and 12.
225.5
262.5
377.0
419.5
8.04 Â 104
1.39 Â 105
1.24 Â 105
3.31 Â 104
The fluorescence spectra of 11 and 12 generally lacked fine fea-
tures possibly due to the high concentration used for the excitation
experiments.21 As shown in Figure 8, the emission maxima of 11
and 12 exhibited bathochromic shifts compared to that of precur-
sor 10 in dichloromethane (10: kem = 370 nm, colorless; 11:
pendent of 12. 1H NMR was used to monitor the product distribu-
tion of three individual reactions under the medium concentration
condition with varied reaction times: 70 min, 80 min, and 90 min
(Fig. 6). The reaction time within this 20 min window had a dra-
matic effect on the product distribution. The majority of 11, which
appeared to be the main product in the 1H NMR spectrum at
70 min, converted to fully cyclized 12 at 90 min. However, cyclode-
hydrogenation reaction pathways are known to be complex:13 the
sensitivity of the reaction time, oxidant loading, and addition rate,
as well as reaction concentration all lead to the difficulty in achiev-
ing accurate control of the product distribution. However, the com-
parison data demonstrate that 11 is a kinetically favored reaction
intermediate instead of a thermodynamically stable product. To
support this statement further, purified compound 11 was sub-
jected to Scholl condensation reaction to confirm that 11 can be
converted to completely cyclized 12. 1H NMR and ESI-MS results
revealed that by using 22.6 equiv of FeCl3, the majority of 11 was
transformed into 12 after 40 min. Thus, we conclude that 11 is in-
deed a Scholl reaction intermediate. Although the generation of
semi-fused HBC structure had been recently reported in the form
of a nitrogen-containing HBC, such analog was shown to be a ther-
modynamically stable product that cannot undergo further trans-
formation into the fully cyclized structure.13
kem = 441 nm, UF = 0.28, yellow-green; 12: kem = 520 nm, UF
=
0.08, yellow). The fluorescence quantum yield of 11 is markedly
higher than that of 12. Interestingly, the solubility of 11 was found
to be lower than that of 12. The finding of the non-heteropolyaro-
matic, semi-fused 11 not only supports the stepwise Scholl con-
densation pathway leading to HBCs demonstrated earlier by
Müllen and co-workers,5 but also offers a new generation of
conjugated fluorescent molecules. We imagine that introduction
of such unique structures into polymer films would create new
opportunities for developing optical and electronic materials from
semi-fused HBCs.
Acknowledgments
This work was supported by the National Science Foundation
Science (CHE-0642413) and the NSF Science Technology Center
of Advanced Materials for the Purification of Water with Systems
(WaterCAMPWS) (Grant No. CTS-0120978). We thank Furong Sun
from the Mass Spectrometry Laboratory, SCS, University of Illinois
for valuable expertise in ESI and MALDI-MS characterization
experiments. Y.L. thanks Dr. Mitch Schultz for helpful discussions.
The semi-fused 11 provides an opportunity to compare it with
12 for investigating the potential interesting photophysical proper-
ties and applications of semi-HBC structures. The UV–vis spectra of
10–12 in CH2Cl2 are presented in Figure 7, and the absorption max-
Supplementary data
ima (kmax) and molar extinction coefficient (e) are collected in Ta-
ble 2. Based on the spectra, the absorption of nonplanar 11 has two
distinct bands in the UV region (245–305 nm, 305–370 nm). These
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.tetlet.2009.04.103.

Y. Lu, J. S. Moore / Tetrahedron Letters 50 (2009) 4071–4077
4077
12. Feng, X.; Wu, J.; Enkelmann, V.; Müllen, K. Org. Lett. 2006, 8, 1145–1148.
13. Gregg, D. J.; Bothe, E.; Hofer, P.; Passaniti, P.; Draper, S. M. Inorg. Chem. 2005, 44,
5654–5660.
14. Feng, X.; Pisula, W.; Takase, M.; Dou, X.; Enkelmann, V.; Wagner, M.; Ding, N.;
Müllen, K. Chem. Mater. 2008, 20, 2872–2874.
15. Wadumethrige, S. H.; Rathore, R. Org. Lett. 2008, 10, 5139–5142.
16. Wang, Z.; Dotz, F.; Enkelmann, V.; Müllen, K. Angew. Chem., Int. Ed. 2005, 44,
1247–1250.
17. Zhang, Q.; Prins, P.; Jones, S. C.; Barlow, S.; Kondo, T.; An, Z.; Siebbeles, L. D. A.;
Marder, S. R. Org. Lett. 2005, 7, 5019–5022.
18. The estimation of heats of formation and energy minimizations were
performed by using Cache 6.1 at the MM2 level.
References and notes
1. van de Craats, A. M.; Warman, J. M. Adv. Mater. 2001, 13, 130–133.
2. Clar, E.; Ironside, C. T.; Zander, M. J. Chem. Soc. 1959, 142–148.
3. Scholl, R.; Mansfeld, J. Ber. Dtsch. Chem. Chem. Ges. 1910, 43, 1734–1746.
4. Ito, S.; Wehmeier, M.; Brand, J. D.; Kubel, C.; Epsch, R.; Rabe, J. P.; Mullen, M.
Chem. Eur. J. 2000, 6, 4327–4342.
5. Kübel, C.; Eckhardt, K.; Enkelmann, V.; Wegner, G.; Müllen, K. J. Mater. Chem.
2000, 10, 879–886.
6. Watson, M. D.; Fechtenkotter, A.; Mullen, K. Chem. Rev. 2001, 101, 1267–1300.
7. Simpson, C. D.; Brand, J. D.; Berresheim, A. J.; Przybilla, L.; Räder, H. J.; Müllen,
K. Chem. Eur. J. 2002, 8, 1424–1429.
8. Balaban, A. T.; Nenitzescu, C. D.. In Friedel-Crafts and Related Reactions; Olah, G.,
Ed.; Wiley: New York, 1964; Vol. 2, pp 979–1047.
9. Rempala, P.; Kroulík, J.; King, B. T. J. Org. Chem. 2006, 71, 5067–5081.
10. Stefano, M. D.; Negri, F.; Carbone, P.; Müllen, K. J. Chem. Phys. 2005, 314, 85–99.
11. Patent Halleux, A. L. US 3,000,984, 1961 claimed three other reaction
intermediates with one, two, three carbon–carbon bond formation, respectively.
19. Lee, M.; Kim, J.-W.; Peleshanko, S.; Larson, K.; Yoo, Y.-S.; Vaknin, D.; Markutsya,
S.; Tsukruk, V. V. J. Am. Chem. Soc. 2002, 124, 9121–9128.
20. Devadoss, C.; Bharathi, P.; Moore, J. S. J. Am. Chem. Soc. 1996, 118, 9635–9644.
21. Fleming, A.; Coleman, J. N.; Dalton, A. B.; Fechtenkötter, A.; Watson, M. D.;
Müllen, K.; Byrne, H. J.; Blau, W. J. J. Phys. Chem. 2003, 107, 37–43.