Synthesis and photo physical properties of 9,10-bis(hydroxyphenyl) anthracene derivatives

Article abstract of DOI:10.1007/s00706-007-0625-2

Among the most promising fluorescent materials for sensor applications, anthracene and its derivatives have been widely studied. In this contribution, the synthesis and characterization of a series of isomers of bis(hydroxyphenyl)anthracene derivatives ar

Full text of DOI:10.1007/s00706-007-0625-2

Monatshefte fu¨r Chemie 138, 453–464 (2007)
DOI 10.1007/s00706-007-0625-2
Printed in The Netherlands
Synthesis and Photo Physical Properties of 9,10-
Bis(hydroxyphenyl)anthracene Derivatives
Mischa Zelzer1, Stefan Kappaun1, Egbert Zojer2, and Christian Slugovc1;ꢀ
1
Institute of Chemistry and Technology of Organic Materials, Graz University of Technology, Graz, Austria
2
Institute of Solid State Physics, Graz University of Technology, Graz, Austria
Received November 7, 2006; accepted (revised) November 19, 2006; published online April 20, 2007
# Springer-Verlag 2007
Summary. Among the most promising fluorescent materi-
als for sensor applications, anthracene and its derivatives
have been widely studied. In this contribution, the synthesis
and characterization of a series of isomers of bis(hydroxy-
[1c]. Further, the development of advanced analyti-
cal techniques (e.g. multiphoton microscopy) requires
fluorophores meeting even more specific criteria [3]
and, thus, boosts scientific efforts to identify suitable
phenyl)anthracene derivatives are described. The preparation
classes of materials with adequate material properties.
Among the most promising fluorescent materials
for sensor applications, anthracene and its derivatives
have been intensely studied because of their excellent
photoluminescence (PL) characteristics and planar
conjugated backbone [4]. Utilizing these desirable
properties, a series of fluorescence sensors especially
for metal ions and biological molecules has been re-
alized [1a, 4a, 5]. Nevertheless, investigations on hy-
droxy functionalized anthracene derivatives and the
influence of deprotonation on their photo physical
properties are rare [6].
of the corresponding derivatives via Suzuki cross-coupling re-
actions is presented with particular emphasis on finding ap-
propriate protecting groups and studying the influence of the
position of the hydroxy group on the outcome of the reaction.
Moreover, the effects of protonation and deprotonation on
absorption and emission spectra are reported and correlated
to semi-empirical quantum mechanical calculations.
Keywords. Sensors; Semiempirical calculations; Protona-
tion; Fluorescence spectroscopy; Absorption spectra.
Introduction
The design, synthesis, and characterization of fluo-
rescent materials with photo physical properties
altered by external stimuli are a field of intense re-
search [1]. Due to the high sensitivity and low-cost
instrumentation of fluorescence spectroscopy, numer-
ous fluorophores are already used as reporter mole-
cules for labeling and sensing purposes in a variety
of areas such as medicine or environmental monitor-
ing [2]. Although extensive work has been devoted
to the design and characterization of fluorescent ma-
terials with a sensing function, only few classes of
molecules fulfill the criteria for practical applications
As a further advance in the design of anthracene
based sensor materials, in this contribution we re-
port on the preparation and characterization of
9,10-bis(hydroxyphenyl)anthracene derivatives. Three
different derivatives, namely 9,10-bis(4-hydroxyphe-
nyl)anthracene (6), 9,10-bis(3-hydroxyphenyl)anthra-
cene (10), and 9,10-bis(2-hydroxyphenyl)anthracene
(11) are discussed. We present procedures for their
synthesis, and demonstrate the impact of deprotona-
tion on absorption and emission spectra of the cor-
responding materials. The latter aims at establishing
a correlation between pH-dependent luminescence
properties and molecular structure. To provide deep-
er insight into the effects of deprotonation on the
ꢀ
Corresponding author. E-mail: slugovc@tugraz.at

454
M. Zelzer et al.
microscopic electronic structure of these 9,10-bis(hy-
droxyphenyl)anthracenes, semi-empirical quantum
mechanical calculations are enclosed and allow a
conclusive interpretation of the experimental data.
derivative gave in our hands the corresponding
4-benzyloxyphenylboronic acid (3) without the for-
mation of undesired by-products in a yield of 80%
(cf. Scheme 1) [9].
Besides the choice of the protecting group, the
reaction conditions applied for the Suzuki cross-
coupling reaction are crucial for the outcome of the
reaction. Cross-coupling reactions of phenylboronic
acid and its derivatives with 9,10-dibromoanthracene
(4) have hitherto only been carried out with Na2CO3
as base and [Pd(PPh3)4] as the catalyst in either
toluene or a mixture of toluene and THF with good
yields of 80 to 95% [8]. However, using this classical
catalyst in THF and Na2CO3 as base for the coupling
of 3 and 4 resulted in difficult removal of by-prod-
ucts such as phosphines and other decomposition
products of the catalyst leading to lower yields of
approx. 50%. To avoid this cumbersome purification
step, the recently described, highly active catalyst
bis(dicyclohexylamine)palladium acetate (DAPCy)
was applied in ethanol with KOH as base [10]. This
catalyst indeed proved to be a promising alternative
producing higher yields of 79% of 9,10-bis(4-benzyl-
oxyphenyl)anthracene (5) combined with a facili-
tated work up.
Results and Discussion
The presented strategy for the synthesis of 9,10-
bis(hydroxyphenyl)anthracenes makes use of boronic
acids obtained from the corresponding bromophe-
nole 1 coupled with 9,10-dibromoanthracene (4) via
a Suzuki cross-coupling reaction (cf. Scheme 1).
While this approach offers the clear advantage of high
functional group tolerance concerning the Suzuki
cross-coupling reaction [7], more attention has to be
paid to the preparation of the boronic acids. Due to
the harsh conditions applied during their synthesis, a
careful choice of protecting groups for the hydroxy
functionality is essential. Thus, for the preparation of
the corresponding 9,10-bis(hydroxyphenyl)anthracene
derivatives a de-protection step subsequent to the
Suzuki cross-coupling reaction is necessary, as re-
ported by Chung et al. for the preparation of 9,10-
bis(4-hydroxyphenyl)anthracene (6) [8]. Although
protection of the hydroxy-function could be ac-
complished with several protecting groups in good
to excellent yields (e.g. utilizing acetyl, pyran or
benzyloxy groups), only the benzyloxy protected
To obtain the desired product 6 cleavage of the
benzyloxy group was attempted with several meth-
ods reported in literature such as the reduction with
Scheme 1

9,10-Bis(hydroxyphenyl)anthracene Derivatives
455
ports on this very specific topic are scarce and so far
most systematic studies focused on the influence of
substituents on the aryl bromide component used in
the Suzuki cross-coupling reaction [14]. A to some
extent comparable study has been reported by Prieto
et al. who coupled stereochemically different phenyl-
boronic acids and phenylboronates bearing various
substituents to indol derivatives [15]. Although they
established some trends related to the nature of the
substituent, their work compared only ortho- and
para-substituted phenylboronic acids and does, there-
fore, not allow the establishment of a correlation to
the effect observed here. From these publications it
can be concluded that the yield of the cross-coupling
reaction rises if the boronic acid bears substituents
that push electrons to the aryl moiety. Thus, the nu-
cleophilic character is increased making it a better
reactant for the transmetalation step in the Suzuki
reaction. Although steric effects on the differences
in the reactivity of 7–9 must not be neglected, a
strong influence of the electron donating character
and the position of the hydroxy group can be anti-
cipated. It has to be kept in mind that the reaction
is performed under alkaline conditions using KOH
as base. Consequently, the proton of the phenolic
OH-function is withdrawn by this strong base, result-
ing in an anionic species as reactive agent. As pro-
posed in Scheme 3 for the mesomeric forms of the
boronates 7–9, this anionic species is capable of de-
localizing its negative charge over the whole aromatic
ring, rendering the ortho- and para-position even more
Scheme 2
1,4-cyclohexadien and Pd=C (10%) [11], with H2
and Pd=C (10%) [8d], with elemental sodium in
butanol [12], the dealkylation with BBr3 in CH2Cl2
[8d, 11], and with KI in BF3 ꢁ Et2O [13]. Among
these cleavage procedures, only the reaction with
BBr3 in CH2Cl2 gave a satisfying conversion yield-
ing 47% of 6.
Although the synthesis procedure described above
was successful, the approach of Chung et al. [8a]
making use of a methoxy protecting group and our
approach utilizing the benzyloxy group for OH-
protection suffer from the drawback of difficulties in
removing the corresponding protecting group. There-
fore, a synthesis route avoiding this yield-limiting
step is desirable. As depicted in Scheme 2, our further
attempts to prepare the 9,10-bis(hydroxyphenyl)an-
thracenes in a one-pot synthesis were based on the
commercially available but much more expensive
hydroxyphenyl pinacolboronates 7–9. Indeed, we
succeeded in the preparation of 9,10-bis(3-hydroxy-
phenyl)anthracene (10) in excellent yields of 95%
using the DAPCy catalyst in EtOH and KOH, even
though the corresponding ortho- and para-substituted
derivatives 6 and 11 could not be obtained by this
approach. It is worth noting that modifications of the
synthesis procedure such as the use of the catalyst
[Pd(PPh3)4] instead of DAPCy or utilization of dif-
ferent solvents and solvent mixtures (toluene, THF,
or ethanol=THF) did not improve the outcome of the
reaction. However, the meta-derivative 10 was not
only synthesized in high yields, but it was also easily
purified by mere washing of the product with CH2Cl2
and acetone.
For a possible explanation of this somewhat un-
expected behavior of the three different hydroxyphe-
nyl pinacolboronates 7–9, stereochemical effects of
electron rich and electron deficient substituents on
boronic acids have to be considered. However, re-
Scheme 3

456
M. Zelzer et al.
Scheme 4
nucleophilic [16]. However, the negative charge can
also be withdrawn by the boron atom, thus giving
rise to a less nucleophilic species, which also ex-
plains the significant number of undefined products
observed during the Suzuki cross-coupling reaction
of 4 with 7 and 9. This proposed explanation is con-
sistent with the high reactivity of 8 as it can be ex-
pected from the electron donating mesomeric effect
of the phenolate group in the meta-position.
Since our attempts to obtain 11 directly from the
boronate 9 failed, another pathway for its prepa-
ration had to be established. In order to give a gen-
eral overview of possibilities to synthesize different
9,10-bis(hydroxyphenyl)anthracene derivatives, the
aptness of a synthetic protocol coupling lithium-
aryles to anthraquinone is herein described as shown
in Scheme 4 [17]. Due to the strong alkaline media
used for this reaction, the OH group was protected
by the comparatively small methoxy group to avoid
any possible sterical hindrance. Starting from 2-
bromoanisol, the lithiated species was reacted with
anthraquinone (14) to obtain 15 after work up with
HI. Cleaving of the protecting group was again ac-
complished by BBr3 in CH2Cl2 [8d, 11] giving 11 in
pure form in a yield of 28%. Although a similar
reaction was published by Kwon et al. giving consid-
erable better yields [17], this procedure gave in our
hands a number of unwanted by-products, which were
hard to separate from 15. Subsequent cleavage of the
protecting group of crudely purified 15 turned out to
facilitate the purification of the unprotected com-
pound 11 due to its low solubility in apolar solvents.
To determine the photo physical properties of the
compounds under investigation, UV-Vis absorption
and photoluminescence (PL) measurements were per-
formed in diluted solutions of methanol at room tem-
Fig. 1. Absorption and emission spectra of 6 in the proto-
&
nated and deprotonated form recorded in MeOH. Absorp-
tion of the protonated species, ꢂ absorption upon addition of
&
NaOH, emission of the protonated species (ꢀexc ¼ 375 nm)
and ꢃ emission upon addition of NaOH. The inset shows the
absorption spectra in the high energy region. Deprotonation
was achieved by addition of 0.1 cm3 NaOH solution (0.03 g
in 1.5 cm3 MeOH) to the diluted sample solutions. Further
addition of base did not change the corresponding absorp-
tion and emission spectra
perature under ambient conditions. Because it was
assumed that deprotonation of the hydroxy group
leads to significant changes in the electronic states of
the molecules, pH-dependent absorption and emission
spectra of the protonated and deprotonated species
were investigated in neutral and alkaline environ-
ments (cf. Fig. 1 and Table 1). From literature it is
known that anthracene displays a structured absorp-
tion spectrum that is mirrored by the structure of the
fluorescence spectrum [18]. Further it has been re-
ported that phenyl substituents on the position 9 and
10 of the anthracene core do not change the absorp-

9,10-Bis(hydroxyphenyl)anthracene Derivatives
457
Table 1. Absorption (ꢀabs) and emission wavelengths (ꢀem) in solution of the protonated and deprotonated species recorded
in methanol. The molecular coefficients of extinction for all compounds are in the same order of magnitude as shown for
6 in Fig. 1
Compound
Protonated
Deprotonated
ꢀabs=nm
ꢀem=nm
ꢀabs=nm
ꢀem=nm
Anthracenea
297, 309, 323
339, 356, 376
340, 357, 375, 395
337, 355, 373, 393
339, 355, 373, 394
357, 397
420, 446
423
421
421
–
–
6
10
11
360, 379, 397
341, 357, 375, 395
341, 357, 376, 396
ꢄ488
b
–
b
–
a
b
Data taken from Ref. [18]; 10 and 11 are not luminescent in their deprotonated forms
Table 2. INDO=SCI calculated transition energies (E), absorption maxima (ꢀabs,calc), and oscillator strengths (OS) for the most
relevant excited states in compounds 6, 10, and 11 in their ground-state equilibrium conformations
ꢅR1=ꢅR2
Compound
E=eV
ꢀabs,calc=nm
OS
Dominant
CI contribution
ꢅOH=ꢅOH
6
3.38
4.80
4.84
3.38
3.40
4.82
3.40
3.44
4.84
2.14
3.40
3.96
4.87
2.29
3.22
4.77
2.30
3.37
4.80
4.89
2.46
3.09
3.61
4.10
4.85
2.97
3.28
3.85
4.74
4.84
4.88
2.79
3.28
3.44
3.74
4.46
4.84
367
258
256
366
364
257
365
360
256
579
365
313
255
541
386
260
540
368
258
254
503
402
343
303
256
418
378
322
261
256
254
445
378
361
331
278
256
0.47
1.25
0.91
0.07
0.44
1.99
0.06
0.39
2.23
0.48
0.15
0.27
1.79
0.01
0.41
1.29
0.14
0.29
0.81
1.01
0.48
0.14
0.08
0.72
1.52
0.06
0.41
0.22
0.49
0.47
1.48
0.23
0.12
0.19
0.27
0.42
0.27
H ! L
H-3 ! L, H ! L þ 1
H ! L þ 3
H ! L þ 1
H ! L
10
11
6
H-3 ! L
H ! L þ 1
H ! L
H-3 ! L
ꢅOꢅ=ꢅOH
H ! L
H ! L þ 2, H-1 ! L
H ! L þ 5
H-4 ! L
10
11
H ! L
H-1 ! L
H-4 ! L
H ! L
H-1 ! L
H-4 ! L
H-4 ! L
ꢅOꢅ=ꢅOꢅ
6
H ! L
H ! L þ 1
H-2 ! L
H ! L þ 3
H-6 ! L
10
H ! L
H-2 ! L
H ! L þ 3
H-4 ! L
several
H-7 ! L
11
H ! L
H ! L þ 1
H-2 ! L
H-1 ! L þ 4, H ! L þ 3
H-1 ! L þ 5, H ! L þ 6
H-2 ! L þ 3

458
M. Zelzer et al.
tion and emission characteristics significantly, though
the absorption and emission bands are broadened [8].
Additionally, aromatic compounds bearing hydroxy
groups generally tend to shift their maxima to lon-
ger wavelengths combined with a loss of structural
features of the corresponding spectra [8] and a
Stoke’s shift dependent on the solvent polarity [19].
Comparing the absorption spectra of 6, 10, and 11
(cf. Table 1), only a slight broadening of the parental
9,10-diphenylanthracene structure [8] in the spec-
tra of 10 and 11 is obtained upon deprotonation,
whereas the three absorption peaks of 6 are notice-
ably broadened as shown in Fig. 1. Hence, the effect
of deprotonation seems to be most striking when the
hydroxy function is located para to the anthracene
core. For the fluorescence spectra of 6, 10, and 11
a dramatic decrease of fluorescence intensity under
alkaline conditions is observed for all compounds
rendering the deprotonated species almost non fluo-
rescent (cf. Fig. 1). Although upon deprotonation a
red-shifted, very weak and broad emission maximum
of 6 and 11 can be noticed, 10 does not show any
luminescence under these experimental conditions.
It is worth noting that a similar red-shifted emission
band as the one obtained, e.g., for 6 has been
observed by Chung et al. for 9,10-bis(4-hydroxyphe-
nyl)anthracene-ester derivatives [8a]. Photolumines-
cence quantum yields for all studied compounds
were found to be around 80% for the fully protonat-
ed forms (experimental error ꢆ7%) [18], while upon
deprotonation quantum yields smaller than 1% were
determined.
For a better understanding and explanation of
the photo physical properties of the bis(hydroxyphe-
nyl)anthracences, semi-empirical quantum mechan-
ical calculations were performed (cf. Experimental)
for the protonated as well as deprotonated species
of 6, 10, and 11. The transition energies, associated
wavelengths, and oscillator strengths are summar-
ized in Table 2 for compounds 6, 10, and 11 in their
ground state equilibrium geometries for various de-
grees of deprotonation. As the present calculations
only consider vertical transitions, the calculated tran-
sition energies correspond to the maximum of an
envelope to the experimentally observed vibronic
progressions.
The data for the fully protonated systems are con-
sistent with the experimental observations. In parti-
cular, the energy of the first absorption maximum is
largely unaffected by the position of the –OH sub-
Fig. 2. Top left: INDO-SCI calculated electron-hole two-particle wavefunction for the lowest excited state in the fully
protonated species of 10. The shading gives the probability for finding the hole at the site denoted by x, while the electron
is at the site denoted by y; top right: numbering of axes of the electron-hole two-particle wavefunction; bottom: illustrations of
the HOMO and LUMO orbitals

9,10-Bis(hydroxyphenyl)anthracene Derivatives
459
stituent. This can be explained by the nature of the
lowest optically allowed state of the fully protonated
molecules. Its configuration interaction (CI) descrip-
tion is largely dominated by a HOMO ! LUMO ex-
citation with both orbitals localized strongly on the
anthracene unit (cf. Fig. 2 for 10 as a typical exam-
ple). Also the associated electron-hole two-particle
wavefunction [20] shown in Fig. 2 confirms the lo-
calization of the exciton. There, the shading is pro-
portional to the correlated probability for finding the
hole at site x while the electron is at site y includ-
ing the full CI description of the excited state. At
higher energies, the calculations also well reproduce
the strong peak(s) around 4.8 eV and the shoulder
between 4.0 and 4.3 eV (the latter is related to two
states not included in Table 2; they are described by
excitations from the HOMO to higher-lying unoccu-
pied orbitals).
Fig. 3. Top: INDO-SCI calculated electron-hole two-particle wavefunction for the lowest excited state in the singly (left) and
doubly (right) deprotonated species of 10. The shading gives the probability for finding the hole at the site denoted by x, while
the electron is at the site denoted by y (the numbering of the axes is the same as depicted in Fig. 2); center: electron-hole two
particle wavefunction for first significantly allowed excited state; bottom: illustrations of the HOMO and LUMO orbitals

460
M. Zelzer et al.
For the mono and fully deprotonated molecules, a
The vanishing overlap between electron and hole is
also responsible for the vanishing oscillator strength
and resulting strongly increased radiative lifetime as-
sociated with the lowest excited state. The first opti-
cally allowed state, S2 for the mono (S3 for the fully)
deprotonated molecule, is described by an excitation
from the HOMO-1 (HOMO-2) to the LUMO. The
respective orbitals are similar to the HOMO of the
fully protonated system, the main difference being
that they somewhat extend onto the deprotonated
ring(s). The conclusions drawn from the orbital pic-
tures are again confirmed by the electron-hole wave-
functions. Consequently, this state is closely related
to the first optically allowed state in the fully proto-
nated molecule. The overall picture developed above,
provides a consistent explanation for the experimen-
tal observation that the emission from 10 is fully
quenched upon deprotonation, while the optical ab-
sorption spectra are only slightly modified.
Considering the experimental data, a similar behav-
ior as for 10 should also be expected for 6 and 11.
Indeed, also there the lowest excited state possesses
a significant charge transfer character. The main dif-
ference compared to the above described situation is
the relatively high oscillator strength of that state,
which should render the deprotonated forms of 6
and 11 emissive. The reason is that in 6 and 11 the
degree of localization of the HOMO on the deproto-
nated ring(s) is less pronounced than for 10, which is
presumably related to the –Oꢅ groups in the ortho-
and para-positions. This results in a stronger overlap
between the electron and hole (in the simple orbital
picture between LUMO and HOMO) and conse-
quently a larger transition dipole. To understand the
origin of this behavior, we have calculated the elec-
tronic structure for the ground and excited states in
the fully deprotonated form of 6 as a function of the
twist angle between the anthracene and the two phe-
nyl rings. The corresponding transition energies, os-
cillator strengths, and orbitals are shown in Fig. 4.
From these plots it becomes obvious that a decreased
coupling between the anthracene core and the phenyl
rings (achieved here by increasing the twist) would
yield the desired results by localizing the orbitals
more strongly, significantly decreasing the oscillator
strength of the lowest excited state (rendering the ma-
terial essentially non-emissive) and recovering the
oscillator strength of the state in the region of the
first absorption peak of the fully protonated molecule.
The overestimation of the coupling in our calcula-
strong red-shift of the lowest excited state is observed.
With the exception of 10, a significant oscillator
strength is associated with the strongly red-shifted
state(s), which is not consistent with the experimen-
tal observations. This cannot be directly related to
the applied semi-empirical methodology, as when
instead using density functional based methods for
geometry optimization and excitation energy calcu-
lation the deviations are even larger. Thus, in the
following, we will adopt a dual strategy: first, the
results for 10, with the –OH groups in meta position
will be discussed and, subsequently, based on these
results, the presumable origin of the deviations for 6
and 11 will be elucidated.
In the mono (fully) deprotonated form of 10, the
lowest-lying excited state at 2.29 (2.97) eV is char-
acterized by a vanishing oscillator strength and the
first allowed state at 3.22 (3.28) eV is only slightly
red-shifted compared to the fully protonated mole-
cule. This small shift of the absorption maximum is
in excellent agreement with the experimental ob-
servation. The low-lying optically forbidden state,
on the other hand, is responsible for the molecules
becoming non-emissive in the deprotonated form.
To more accurately describe the emission process,
we have also calculated the equilibrium geometries
of the excited molecules. For the fully protonated
form of 10, we find the lowest excited state shifted
to 2.99 eV (414 nm; OS ¼ 0.56), while for the mono
and fully deprotonated molecules optically forbidden
states at 1.99 eV (622 nm; OS ¼ 0.00) and 2.68 eV
(462 nm; OS ¼ 0.03) are lowest in energy. Based on
this good correlation between experimental and the-
oretical results for 10, a more in depth analysis of the
nature of the various states in the deprotonated mole-
cule was performed. As shown in Fig. 3, the LUMO
in the deprotonated molecule is strongly reminiscent
of that of the protonated one. The HOMO, however
is localized on the ring(s) bearing the –Oꢅ group(s).
As a consequence, the lowest excited state, which is
dominantly described by a HOMO to LUMO ex-
citation (cf. Table 2) possesses a pronounced charge
transfer character. This is also confirmed by the elec-
tron-hole two-particle wavefunctions (which con-
tain the full CI-based information of the excited
state going beyond the simple orbital picture) in
the top part of Fig. 3. They clearly show that the
hole is confined to the deprotonated ring(s), while
the electron is localized on the anthracene backbone.

9,10-Bis(hydroxyphenyl)anthracene Derivatives
461
Fig. 4. Fully deprotonated molecule 6. Top left: INDO=SCI calculated change of excited state energies (filled symbols) and
oscillator strengths (open symbols) as a function of the twist angle ꢁ for the lowest excited state (squares) and the state
corresponding to the main peak in the fully protonated species (circles, details see text). The vertical line at 60.4ꢃ corresponds
to the twist in the fully relaxed molecule. The HOMOs and LUMOs for selected twist angles are also shown. Note that the
twists are applied synchronously
tions on the fully geometry optimized molecules of 6
and 11 is tentatively attributed to a combination of
factors including solvent effects and the interaction
with counter ions in solution.
Experimental
Unless otherwise noted, materials were obtained from com-
mercial sources (Aldrich, Fluka or Lancaster) and were used
without further purification. Solvents for reactions were
freshly distilled over appropriate drying agents prior to use.
Reactions were carried out under inert atmosphere of Ar using
Conclusion
1
standard Schlenk techniques. H NMR spectra were recorded
on a Varian INOVA 500 MHz Spectrometer at 500 MHz,
13C{1H} NMR spectra were recorded at 125 MHz. Assign-
ment of the peaks was done by DEPT and COSY NMR spec-
troscopy. Solvent residual peaks were used for referencing the
NMR spectra to the corresponding values given in literature
[21]. Elemental analyses (C and H) were conducted for the
novel compounds 5 and 10 with results that were found to
be in good agreement (ꢆ0.2%) with the calculated values.
UV-Visible absorption spectra were recorded on a Cary 50
Bio UV-Visible Spectrophotometer, fluorescence spectra on a
Perkin Elmer Luminescence Spectrometer LS50B. PL quan-
tum yields were measured using a Shimadzu RF-5301PC
In this contribution three different synthesis routes
for the preparation of 9,10-bis(4-hydroxyphenyl)an-
thracene (6), 9,10-bis(3-hydroxyphenyl)anthracene
(10), and 9,10-bis(2-hydroxyphenyl)anthracene (11)
were presented giving these materials in good to
acceptable yields of 47, 95, and 28%. Further, the
photo physical properties in solution and the effects
of deprotonation on absorption and emission spectra
were discussed and explained via quantum mechan-
ical calculations.

462
M. Zelzer et al.
Spectrofluorimeter (detector corrected) and Perkin Elmer
Lambda 9 UV=VIS=NIR Spectrophotometer using quinine
sulphate dihydrate in 0.1 N H2SO4 as standard.
Bn4), 127.6 (4C, Bn2,6), 127.0 (4C, Ant1,4,5,8), 124.9 (4C,
Ant2,3,6,7), 114.7 (4C, Ph3,5), 70.2 (2C, OCH2) ppm.
Quantum mechanical calculations consisted of the optimi-
zation of molecule geometries in their ground state using the
semi-empirical Hartree-Fock Austin Model 1 (AM1) method
[22] as implemented in Ampac6.55. For optimizing the geo-
metries of the lowest excited states, the AM1 Hamiltonian was
coupled to a single configuration interaction (SCI) scheme
[23] including 7 occupied and 7 unoccupied orbitals in the
CI active space (necessary for avoiding an artificial symmetry
breaking at all levels of deprotonation). Transition energies
and oscillator strengths were calculated on the basis of the
semi-empirical Hartree-Fock intermediate neglect of differ-
ential overlap method. Electron correlation effects were also
here included via SCI (26ꢇ26 active orbitals) [24]. These
calculations were performed on the one hand using a slightly
modified version of the original ZINDO code by Zerner and
coworkers as well as using Gaussian 98 (Revision A.9) [25].
Consistent results were obtained with both codes. Electron-
hole two-particle wavefunctions were calculated using Zoa
2.5. [26]. Gaussian 98 (Revision A.11) was used for the com-
parative density-functional theory based calculations.
9,10-Bis(4-hydroxyphenyl)anthracene (6)
Compound 5 (0.40 g, 0.7 mmol) was suspended in 10 cm3 dry
CH2Cl2 under Ar and treated with ultrasound for 1 min. To the
dirty-white suspension, 0.7 cm3 BBr3 (1M, 0.7mmol) were
added drop wise followed by stirring at room temperature
for 24h. After addition of 30 cm3 H2O, the reaction mixture
was acidified with 2 cm3 aqueous HCl solution (10%) and
stirred for further 24h. The product was extracted four times
with 70cm3 CH2Cl2 from H2O, dried (Na2SO4), and purified
by recrystallization from little CH2Cl2 several times yielding
0.13g (47%) of a weakly brown colored powder. Rf ¼ 0.14
1
(cy:ee¼ 4:1). H NMR (500MHz, CDCl3): ꢂ ¼ 8.66 (s, 2H,
OH), 7.74 (dd, J ¼ 6.8, 3.0 Hz, 4H, Ant1,4,5,8), 7.38 (dd,
J ¼ 6.8, 3.0 Hz, 4H, Ant2,3,6,7), 7.29 (d, J ¼ 8.3 Hz, 4H, Ph2,6),
7.13 (d, J ¼ 8.3 Hz, 4H, Ph3,5) ppm; 13C NMR (125 MHz,
CDCl3): ꢂ ¼ 158.6 (2C, Ph4), 138.5 (2C, Ant9,10), 133.8 (4C,
Ph2,6), 131.9 (4C, Ant4a,8a,9a,10a), 131.2 (2C, Ph1), 128.5 (4C,
Ant1,4,5,8), 126.5 (4C, Ant2,3,6,7), 117.0 (4C, Ph3,5) ppm;
UV-Vis (methanol): ꢀabs,protonated ¼ 340, 357, 375, 395 nm,
ꢀabs,deprotonated ¼ 360, 379, 397 nm; fluorescence (methanol):
ꢀem,protonated ¼ 423 nm, ꢀem,deprotonated ¼ around 488 nm (broad
and weak). Spectra were found to be in accordance with
Ref. [8].
4-Benzyloxybromobenzene (2)
As an alternative to the procedure of Kim et al. [9c], 10.4g
1 (60.2mmol) were dissolved in a mixture of 70 cm3 aqueous
KOH solution (2 M) and 140 cm3 ethanol. After heating to
90ꢃC, 16 cm3 benzyl bromide (134.5mmol) were added fol-
lowed by refluxing the reaction mixture for 24h at that
temperature. Cooling to room temperature and stripping of
the aqueous phase yielded a white solid which was dried
and further purified by column chromatography on silica
(cy:ee ¼ 5:1). Sampling the band at Rf ¼ 0.67 (cy:ee ¼ 4:1)
gave 14.0g (88%) 2. 1H and 13C NMR spectra were found to
be identical with the ones described in Ref. [9c].
9,10-Bis(3-hydroxyphenyl)anthracene (10, C26H18O2)
Compound 4 (0.30 g, 0.9mmol), 0.47 g 8 (2.1mmol), and
0.18g KOH (3.2mmol) were dissolved in 20cm3 ethanol
under Ar. For acceleration of the dissolving process, the mix-
ture was treated with ultrasound for 1 min. After the addition
of 0.005 g of the catalyst DAPCy (0.01 mmol), the reaction
was stirred at 40ꢃC for 24h. The reaction mixture was filtered
over Celite+ followed by elution of the product with methanol.
Evaporation of the solvent and treatment with CH2Cl2 and
methanol gave a suspension which was filtrated. The resi-
due was washed with acetone several times yielding 0.31 g
4-Benzyloxyphenylboronic acid (3)
Compound 3 was prepared according to Gimeno et al. with 1H
NMR spectra as reported in Ref. [9d].
1
(95%) 10 as white powder. Rf ¼ 0.20 (cy:ee¼ 4:1). H NMR
(500MHz, CD3OD): ꢂ ¼ 7.71 (dd, J ¼ 6.8, 3.4 Hz, 4H,
Ant1,4,5,8), 7.38 (t, J ¼ 7.8 Hz, 2H, Ph5), 7.29 (dd, J ¼ 6.8,
3.4 Hz, 4H, Ant2,3,6,7), 6.96 (dd, J ¼ 8.3, 2.4 Hz, 2H, Ph6),
6.84 (dd, J ¼ 2.4 Hz, 2H, Ph2), 6.80 (d, J ¼ 7.3 Hz, 2H,
Ph4), 4.85 (s, 2H, OH) ppm; 13C NMR (125MHz, CD3OD):
ꢂ ¼ 160.8 (2C, Ph3), 141.6 (2C, Ph1), 138.5 (2C, Ant9,10),
131.0 (4C, Ant4a,8a,9a,10a), 130.4 (2C, Ph5), 128.0 (4C,
Ant1,4,5,8), 125.8(4C, Ant2,3,6,7), 122.3(2C, Ph6), 120.1(2C, Ph4),
116.3 (2C, Ph2) ppm; UV-Vis (methanol): ꢀabs,protonated ¼ 337,
355, 373, 393 nm, ꢀabs,deprotonated ¼ 341, 357, 375, 395 nm;
fluorescence (methanol): ꢀem,protonated ¼ 421 nm.
9,10-Bis(4-benzyloxyphenyl)anthracene (5, C40H30O2)
Compound 4 (0.98 g, 2.9 mmol), 1.64 g 3 (7.2 mmol), and
0.50g KOH (8.9mmol) were dissolved in 55 cm3 ethanol
under Ar. For acceleration of the dissolving process, the mix-
ture was treated with ultrasound for 1 min. After addition of
0.01g of the catalyst DAPCy (0.02 mmol), the reaction mix-
ture was stirred at 40ꢃC for 20 h during which precipitation
of the product occurred. The precipitate was isolated by fil-
tration, washed with 5 cm3 methanol, and recrystallized
from CH2Cl2 several times yielding 1.25 g (79%). Rf ¼ 0.66
(cy:ee¼ 4:1). 1H NMR (500MHz, CDCl3): ꢂ ¼ 7.75 (dd,
J ¼ 6.8, 3.4Hz, 4H, Ant1,4,5,8), 7.56 (d, J ¼ 7.3 Hz, 4H, Bn2,6),
7.46 (t, J ¼ 7.3 Hz, 4H, Bn3,5), 7.44–7.32 (m, 2H, Bn4), 7.40
(d, J ¼ 8.8 Hz, 4H, Ph3,5), 7.33 (dd, J ¼ 6.8, 3.4 Hz, 4H,
Ant2,3,6,7), 7.22 (d, J ¼ 8.8 Hz, 4H, Ph2,6), 5.22 (s, 4H, OCH2)
ppm; 13C NMR (125MHz, CDCl3): ꢂ ¼ 158.3 (2C, Ph4),
137.0, 136.7 (4C, Ant9,10, Bn1), 132.4 (4C, Ph2,6), 131.4 (2C,
Ph1), 130.2 (4C, Ant4a,8a,9a,10a), 128.7 (4C, Bn3,5), 128.1 (2C,
9,10-Bis(2-methoxyphenyl)anthracene (15)
Compound 12 (0.56 g, 3.0 mmol) was dissolved in 20cm3
anhydrous THF under Ar and cooled to ꢅ80ꢃC. n-Butyl
lithium (2 cm3, 2.5 M in hexane, 5.0mmol) were added drop
wise followed by stirring the reaction mixture at ꢅ80ꢃC for
1 h. After addition of 0.26g 14 (1.2mmol), the reaction was
allowed to attain room temperature and stirred for further 20h.

9,10-Bis(hydroxyphenyl)anthracene Derivatives
463
HI (1.8cm3, ꢄ55%, 13.0mmol) was added followed by stir-
ring for another 20h. Work up consisted of adding 70cm3
H2O, 3 cm3 Na2S2O3 (2%), 1 cm3 aqueous HCl solution
(10%), and the subsequent extraction with CH2Cl2. The sol-
vent was evaporated and the residue purified by column chro-
matography on silica (cy:ee¼ 30:1). Sampling the band at
Rf ¼ 0.75 (cy:ee¼ 4:1) gave 0.03 g (3%) 15 as a slightly yellow
3.4 Hz, 4H, Ant1,4,6,8), 7.58–7.51 (m, 2H, Ph6), 7.36 (dd,
J ¼ 7.3, 1.9 Hz, 2H, Ph4), 7.30 (dd, J ¼ 6.8, 3.4 Hz, 4H,
Ant2,3,6,7), 7.22–7.14 (m, 4H, Ph3,5), 3.66 (s, 6H, OCH3) ppm;
13C NMR (125MHz, CDCl3): ꢂ ¼ 158.1 (2C, Ph2), 133.6
(2C, Ant9,10), 130.1 (4C, Ant4a,8a,9a,10a), 129.2 (2C, Ph6),
128.8 (2C, Ph4), 127.7 (2C, Ph1), 126.9 (4C, Ant1,4,5,8),
124.7 (4C, Ant2,3,6,7), 120.7 (2C, Ph5), 111.2 (2C, Ph3), 55.7
(2C, OCH3) ppm. Spectra were found to be in accordance with
Ref. [17].
[4] a) Yang WJ, Kim DY, Jeong M-Y, Kim HM, Lee YK,
Fang X, Jeon S-J, Cho BR (2005) Chem Eur J 11: 4191;
b) Kim Y-H, Jeong H-C, Kim S-H, Yang K, Kwon S-K
(2005) Adv Funct Mater 15: 1799
[5] Desvergne J-P, Lahrahar N, Gotta M, Zimmermann Y,
Bouas-Laurent H (2005) J Mater Chem 15: 2873
[6] Basaric N, Wan P (2006) J Org Chem 71: 2677
[7] Suzuki A (2005) Chem Commun 4759
[8] a) Chung S-J, Kim K-K, Jin J-I (1999) Polymer 40:
1943; b) Kotha S, Ghosh AK, Deodhar KD (2004)
Synthesis 549; c) Kotha S, Ghosh AK (2002) Synlett
451; d) Wagner G, Herrmann R, Scherer W (1996) J Orga-
nomet Chem 516: 225; e) Hu D, Fuh RA, Li J, Corkan A,
Lindsey JS (1998) Photochem Photobiol 68: 141
[9] a) Sharma SK, Kanamathareddy S, Gutsche CD (1997)
Synthesis 11: 1268; b) Geng Y, Trajkovska A, Katsis D,
Ou JJ, Culligan SW, Chen SH (2002) J Am Chem Soc
124: 8337; c) Kim J, Kim YK, Park N, Hahn JH, Ahn KH
(2005) J Org Chem 70: 7087; d) Gimeno N, Ros MB,
Serrano JL, de la Fuente MR (2004) Angew Chem Int Ed
43: 5235
[10] a) Tao B, Boykin DW (2004) J Org Chem 69: 4330;
b) Kappaun S, Zelzer M, Bartl K, Saf R, Stelzer F,
Slugovc C (2006) J Polym Sci Part A Polym Chem
44: 2130; c) Kappaun S, Sovic T, Stelzer F, Pogantsch
A, Zojer E, Slugovc C (2006) Org Biomol Chem 4: 1503
[11] Kocienski PJ (1994) Protecting Groups, Georg Thieme
Verlag, Stuttgart
[12] Loev B, Dawson CR (1956) J Am Chem Soc 78: 6095
[13] Vankar YD, Rao CT (1985) J Chem Research (S): 232
[14] For examples see: a) Bandgar BP, Bettigeri SV, Phopase J
(2004) Org Lett 6: 2105; b) Edsall RJJ, Harris HA, Manas
ES, Mewshaw RE (2003) Bioorg Med Chem 11: 3457
[15] Prieto M, Zurita E, Rosa E, Munoz L, Lloyd-Williams P,
Giralt E (2004) J Org Chem 69: 6812
1
powder. H NMR (500 MHz, CDCl3): ꢂ ¼ 7.63 (dd, J ¼ 6.8,
9,10-Bis(2-hydroxyphenyl)anthracene (11)
Compound 15 (0.015 g, 0.04 mmol) was dissolved in 4 cm3
dry CH2Cl2 under Ar. BBr3 (0.5cm3, 1 M, 0.5 mmol) was
added followed by stirring the reaction mixture at room tem-
perature for 24h. After addition of 60cm3 H2O, the reaction
mixture was acidified with 2 cm3 aqueous HCl solution (10%)
and stirred for further 24h. The product was extracted four
times with 50 cm3 CH2Cl2 from H2O, dried (Na2SO4), and
purified by column chromatography on silica (cy:ee¼ 10:1).
Sampling the band at Rf ¼ 0.36 (cy:ee¼ 4:1) gave 0.004 g
(28%) 11. 1H NMR (500MHz, (CD3)2CO): ꢂ ¼ 8.03–8.00
(m, 2H, OH), 7.73–7.67 (m, 4H, Ant1,4,5,8), 7.50–7.43 (m,
2H, Ph6), 7.41–7.34 (m, 4H, Ant2,3,6,7), 7.30–7.23 (m, 2H,
Ph4), 7.22–7.16 (m, 2H, Ph3), 7.16–7.09 (m, 2H, Ph5) ppm;
UV-Vis (methanol): ꢀabs,protonated ¼ 339, 355, 373, 394nm,
ꢀabs,deprotonated ¼ 341, 357, 376, 396 nm; fluorescence (metha-
nol): ꢀem,protonated ¼ 421 nm. Spectra were found to be in ac-
cordance with Ref. [17].
[16] Beyer H, Walter W (1998) Lehrbuch der Organischen
Chemie, S. Hirzel Verlag, Stuttgart
[17] a) Kwon SK, Kim YH, Shin SC (2002) Bull Korean
Chem Soc 23: 17; b) Stoessel P, Breuning E, Vestweber
H, Heil H (2006) PCT Int Appl, WO 2006048268 A1
[18] a) Valeur B (2001) Molecular Fluorescence – Principles
and Applications, Wiley-VCH, Weinheim; b) Lakowicz
JR (1999) Principles of Fluorescence Spectroscopy,
Kluwer Academic=Plenum Publishers, New York; c)
Morris JV, Mahaney MA, Huber JR (1976) J Phys Chem
80: 969; d) Kim Y-H, Kwon S-K (2006) J Appl Polym
Sci 100: 2151
Acknowledgements
Financial support by the Austrian Science Fund in the frame-
work of the Austrian Nano Initiative RPC ISOTEC – RP 0701
and 0702 is gratefully acknowledged.
References
[19] Tautges A (2002) J Phys Chem Lab 4: 8
[20] Here we follow the definition given in Romaner L,
Heimel G, Wiesenhofer H, Scandiucci de Freitas P,
[1] For examples see: a) de Silva AP, Gunaratne HQN,
Gunnlaugsson T, Huxley AJM, McCoy CP, Rademacher
JT, Rice TE (1997) Chem Rev 97: 1515; b) Balzani V,
Credi A, Raymo FM, Stoddart JF (2000) Angew Chem
Int Ed 39: 3348; c) Charier S, Ruel O, Baudin J-B, Alcor
D, Allemand J-F, Meglio A, Jullien L, Valeur B (2006)
Chem Eur J 12: 1097
[2] a) Mancin F, Rampazzo E, Tecilla P, Tonellato U (2006)
Chem Eur J 12: 1844; b) Wang Z, Zheng G, Lu P (2005)
Org Lett 7: 3669
´
Scherf U, Bredas JL, Zojer E, List EJW (2004) Chem
Mater 16: 4667 rather than applying the symmetrized
version we used in previous works
[21] Gottlieb HE, Kotylar V, Nudelman A (1997) J Org Chem
62: 7512
[22] Dewar MJS, Zoebisch EG, Healy EF, Stewart JJP (1985)
J Am Chem Soc 107: 3902
[23] Ampac 5.0 User’s Manual (1994) Copyright Semichem,
7128 Summit, Shwanee, KS 66216
[3] Werts MHV, Gmouh S, Mongin O, Pons T, Blanchard-
Desce M (2004) J Am Chem Soc 126: 16294

464
M. Zelzer et al.: 9,10-Bis(hydroxyphenyl)anthracene Derivatives
[24] a) Pople JA, Beveridge DL, Dobosh PA (1967) J Chem
Phys 47: 2026; b) Ridley J, Zerner M (1973) Theoret
Chim Acta 32: 111
[25] Gaussian 98, Revision A.9, Frisch MJ, Trucks GW,
Schlegel HB, Scuseria GE, Robb MA, Cheeseman JR,
Zakrzewski VG, Montgomery JA, Stratmann RE, Burant
JC,DapprichS,MillamJM,DanielsAD,KudinKN,Strain
MC, Farkas O, Tomasi J, Barone V, Cossi M, Cammi R,
Mennucci B, PomelliC, Adamo C, CliffordS, OchterskiJ,
PeterssonGA,AyalaPY,CuiQ,MorokumaK,MalickDK,
Rabuck AD, Raghavachari K, Foresman JB, Cioslowski J,
Ortiz JV, Baboul AG, Stefanov BB, Liu G, Liashenko A,
Piskorz P, Komaromi I, Gomperts R, Martin RL, Fox DJ,
Keith T, Al-Laham MA, Peng CY, Nanayakkara A,
Challacombe M, Gill PMW, Johnson B, Chen W, Wong
MW, Andres JL, Gonzalez C, Head-Gordon M, Replogle
ES, Pople JA (1998) Gaussian Inc., Pittsburgh PA
´
[26] Zoa v2.0, Calbert JP, Service de Chimie des Materiaux
Nouveaux, Mons (Belgium), options 3D and not anti-
symmetrized