Synthesis of an Acceptor-Donor-Acceptor Multichromophore Consisting of Terrylene and Perylene Diimides for Multistep Energy Transfer Studies

Article abstract of DOI:10.1021/acs.chemmater.5b04602

Motivated by the results obtained from the investigation of singlet-singlet annihilation in a linear multichromophore comprising terrylene diimides (TDI) and perylene diimide (PDI) in 2010, we report the detailed process toward the successful synthesis of a TDI-PDI-TDI dyad. Ineffective synthetic pathways, which were necessary for the understanding of the step-by-step construction of the complex multichromophore, are described, leading toward a universal synthetic plan for multicomponent systems containing rylene diimides separated by rigid oligophenylene spacers.

Full text of DOI:10.1021/acs.chemmater.5b04602

Article
pubs.acs.org/cm
Synthesis of an Acceptor−Donor−Acceptor Multichromophore
Consisting of Terrylene and Perylene Diimides for Multistep Energy
Transfer Studies
Sebastian Stappert, Chen Li, and Klaus Mullen*
̈
Max-Planck Institute for Polymer Research Ackermannweg 10, 55128 Mainz, Germany
Thomas Basche
́
Institute of Physical Chemistry, Johannes Gutenberg University, Mainz Duesbergweg 10-14, 55128 Mainz, Germany
S
* Supporting Information
ABSTRACT: Motivated by the results obtained from the
investigation of singlet−singlet annihilation in a linear multi-
chromophore comprising terrylene diimides (TDI) and
perylene diimide (PDI) in 2010, we report the detailed process
toward the successful synthesis of a TDI−PDI−TDI dyad.
Ineffective synthetic pathways, which were necessary for the
understanding of the step-by-step construction of the complex
multichromophore, are described, leading toward a universal synthetic plan for multicomponent systems containing rylene
diimides separated by rigid oligophenylene spacers.
Unfortunately, further investigation of 1 was limited due to the
complex, low-yielding multistep synthesis with difficult
purification giving only milligram quantities, which prompted
us to explore efficient and facile synthetic routes to such
multichromophores. As a consequence of the envisioned
applications in single molecule energy transfer studies, even
minor variations in the design of each chromophore subunit are
relevant, and especially the connection between donor and
acceptor needs to be investigated thoroughly with regard to
energy transfer efficiency as well as synthetic accessibility. Here
we outline the reasons for redesigning the molecular structure
of the formerly published dyad 1 together with a new synthetic
strategy refined toward greater reproducibility and target
structural versatility. One can imagine a variety of different
strategies to prepare the well-defined TDI-bridge-PDI-bridge-
TDI oligomer, but each with its own special challenges. For
instance, (1) while the symmetric PDI core-units are readily
available, more complicated syntheses are required to prepare
the requisite lower-symmetry TDI building blocks. (2)
Preparation and purification of any such extended rigid
oligomer of rylene diimides will be hindered to varying degrees
by solubility/aggregation. Thus, equipping the dyes with
solubilizing groups and diminishing π-stacking are of pivotal
importance but require more elaborate synthetic work. We
could not turn to tetra-phenoxy bay-functionalized PDI and
TDI building blocks for this purpose, because the phenoxy
INTRODUCTION
■
Light-harvesting systems in nature generate a sustainable
energy source to support the life of organisms. Complexes of
chromophores and proteins have the outstanding ability of
efficient energy transfer and conversion after absorption of
sunlight, especially in the process of photosynthesis. Until now,
the different steps involved in photosynthesis are still not
completely understood. Thus, synthetic multichromophores
have been used to investigate the fundamentals of excitation
energy transfer down to the single-molecule level.1,2 In this
regard, we have investigated several multichromophoric
systems in which a perylene diimide (PDI) donor moiety and
terrylene diimide (TDI) acceptor units were linked by rigid
para-oligophenylene bridges (Figure 1).3−6 PDI and TDI had
been selected with the consideration of their large absorption
cross-section, high fluorescence quantum yields, and out-
standing photostability with low photobleaching yields.7−9 In
2010, we reported a TDI−PDI dyad 1, which had a PDI at the
center position, two TDIs as end-caps, and two rigid bridges
linking the PDI and TDIs (Figure 1).10 In a first set of
experiments, electronic coupling between the two TDI moieties
was studied by their selective excitation at 635 nm. In this case,
the central PDI moiety solely acted as part of a long rigid spacer
and neither contributed to absorption/emission or energy
transfer pathways. A photon coincidence setup was imple-
mented to quantitatively reveal the competition between
singlet−singlet annihilation (SSA) and simultaneous emission
once both TDI chromophores were excited. It turned out that
the SSA process between the two TDI molecules in dyad 1 was
on average three-times faster than the competing fluorescence.
Received: November 26, 2015
Revised: January 11, 2016
© XXXX American Chemical Society
A
DOI: 10.1021/acs.chemmater.5b04602
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
Figure 1. TDI−PDI−TDI dyad synthesized for the investigation of SSA of weakly interacting chromophores.
Figure 2. Types of reaction involved in the synthesis of the target structure of TDI-bridge-PDI-bridge-TDI.
Scheme 1. Reaction Scheme of the Cyclodehydrogenation Pathway
groups induce additional radiationless relaxation pathways
leading to a decrease in energy transfer efficiency between
the chromophores and to stronger electron−phonon coupling
in low-temperature experiments. Special attention was therefore
focused on the rigid linker between the chromophores.
Although the dimension of the bridging unit is of less
importance, reactivity, connectivity, and contribution to
solubility play major roles. As will be outlined here, not every
building block can be combined in standard coupling reactions,
revealing the complexity and nontriviality of the dyad. Limited
combinatory possibilities of donor, bridge, and acceptor
prompted many strategies to fail in the final stage; hence, the
most important issue was to find the best balance between
solubility, reactivity, and ease of purification. Benefiting from
progress in the synthesis of rylene dyes over the past decades,
we herein describe the experimental process toward the desired
A−D−A structure culminating in the successful synthesis of
dyad 26 in high yields.
B
DOI: 10.1021/acs.chemmater.5b04602
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
Scheme 2. Reaction Scheme of the Sonogashira−Hagihara Pathway
Scheme 3. Reaction Scheme of the Stille Pathway versus Suzuki−Miyaura Pathway
sterically hindered or less reactive amines or anilines. Aniline 3
was prepared to provide a linker with sterically hindered,
twisted terphenyl linkage and solubilizing alkyl chains. By
imidization of 2 with bromo-functionalized 4′-bromo-2′,5′-
dioctyl-[1,1′-biphenyl]-4-amine (3), a symmetric, solubilized,
and reactive PDI building block 4 was obtained (Scheme 1).
Coupling of TDI 5 to PDI 4 was unsatisfactory, mainly giving
monocoupled intermediates which partially precipitated from
the reaction mixture. We attribute the poor yields to the poor
solubility of the intermediate in combination with low reactivity
of the TDI-boronic acid ester 5. Suzuki coupling of the same
PDI 4 with the unfused TDI precursor 7 alleviates some of the
solubility problem and gives better yields but then requires a
subsequent cyclodehydrogenation step. During the cyclo-
dehydrogenation reaction, fusion and saponification occurred
simultaneously, leading mostly to unwanted reaction products.
Attempts to improve this by reducing reaction time or
temperature failed as saponification of at least one of the six
imide linkages dominated over cyclization.
RESULTS/DISCUSSION
■
The symmetry of the target molecule gave rise to several
possible synthetic pathways, each of them bearing specific
challenges. In general, three key reactions have to be considered
for building the scaffold (Figure 2), which are (1) cyclo-
dehydrogenation of unfused PDI or TDI precursors; (2)
imidization to functionalize the PDI unit; and (3) aryl C−C
coupling to connect the PDI and TDI moieties. Because the
C−C coupling is aiming to connect two different units, the
most common Pd-mediated reactions are used such as Suzuki,
Stille, and Sonogashira coupling reactions. The primary
obstacle then is to determine the most effective sequence of
these steps. As it turns out, in most cases, the final-step reaction
is critical. Thus, three overall synthetic strategies can be defined
by the type of the final-step reaction: Cyclodehydrogenation,
Sonogashira, and Suzuki/Stille pathways.
Cyclodehydrogenation Pathway. Initially, we concen-
trated on readily accessible building blocks, that is, moderately
soluble PDIs and TDIs, for example, 4 and 5, with major
consideration on reducing the number of synthetic steps
(Scheme 1). Imidization of commercially available
perylenetetracarboxylic dianhydride 2 proceeds readily with
various anilines but requires greater synthetic effort with
Sonogashira Pathway. Because of unexpected saponifica-
tion when cyclodehydrogenation was employed as the final
step, we turned to strategies with aryl C−C coupling as the final
reaction. The first fruitful attempts yielding the desired A−D−
C
DOI: 10.1021/acs.chemmater.5b04602
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
Scheme 4. Left: Failed C−C Cross-Coupling Reactions Targeting Disubstituted Rylenediimides. Ryleneimide Substructure
Indicates PDI, TDI, or Unfused TDI. Right: Multi-Suzuki−Miyaura Pathway Successfully Giving the Cross-Coupling Product in
High Yields
Scheme 5. Reaction Scheme for the Multi-Suzuki−Miyaura Pathway
A structure 1 made use of Sonogashira coupling of an ethynyl
functionalized PDI-moiety 10 with the halo-functionalized TDI
12. The synthetic strategy was geared to generate soluble
building blocks with highly reactive and accessible coupling
functionality. Studies have shown that coupling of one TDI-
unit to one PDI-unit can be achieved in some cases,4 but
connection of two TDI units to a central PDI-moiety failed
under those conditions. As a result, more solubilizing groups
were introduced, and the Sonogashira−Hagihara reaction was
chosen as the final C−C-coupling step (Scheme 2). Both
chromophores, PDI 9 and TDI 11, were equipped with 2,6-
diisopropylphenyl substituents in the imide position to impart
steric hindrance to aggregation as well as resistance to
saponification. Starting from well-known dibromo-diisopropyl-
phenyl PDI 9 and monobromo diisopropylphenyl TDI 11, the
key reaction of the following multistep synthesis was the
desymmetrizing monolithiation of the precursor 1,4-dibromo-
2,5-dioctylbenzene (13) and subsequent silylation. Unfortu-
nately, the final Sonogashira−Hagihara coupling of PDI 10 and
TDI 12 suffered from poor reproducibility and yield, which is
unacceptable given the immense effort in this 20-step synthesis.
Suzuki versus Stille Pathway. Our earlier unsuccessful
attempts with cyclodehydrogenation as final step relied on C−
C coupling of halo-functionalized PDI central units with TDI-
boronate coupling partners. As described here, we then
evaluated a few routes with the reactivity “inverted” according
to Scheme 3. Coupling of the unfused brominated TDI-
precursor 16 to boronated PDI 15a via Suzuki coupling
(Scheme 3, left) was superior to the coupling of the respective
fused TDI moiety 11, but yields were still unsatisfactory. We
then turned to a better-solubilized TDI, (Scheme 3, right) but
surprisingly, no reaction was observed at all with 15a.
Electrostatic repulsion of the shielding octyl-chains with the
neighboring functionalities could explain the unsatisfyingly low
reaction yields (<5%) but does not explain the high amounts of
monocoupled intermediates, whereas two-fold coupling was
rarely observed (Scheme 1). Since the poor outcomes could be
related to the increasingly hydrophobic environments immedi-
ately surrounding the reactive sites under biphasic Suzuki
coupling conditions, we attempted Stille coupling. However, in
accordance with the analogous Suzuki-reaction of 15a and 12,
no reaction between 15b and 12 could be observed, proving
the above suggestions wrong. These results point toward steric
problems in the C−C-coupling step at the octyl-chain bearing
phenylene within the para-terphenylene spacer, regardless of
which reaction type was utilized. As will be shown later,
D
DOI: 10.1021/acs.chemmater.5b04602
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
Scheme 6. Reaction Scheme for the Synthesis of the Bridging Dioctyl-Substituted Terphenylene
Scheme 7. Reaction Scheme of the Synthesis of the Donor Unit PDI 9. Upper Part Showing the Reported Four-Step Procedure,
Lower Part Displaying Our Novel Two-Step Procedure
sterically unhindered and electronically activated moieties do
indeed react with the dioctyl-substituted phenylene system in
very high yields (Scheme 6).
SYNTHESIS
■
Our synthetic procedure includes the synthesis of the bis-
functionalized PDI 9, attachment of a solubilizing linker via 24,
and final coupling to the asymmetric TDI-units 11 (Scheme 5).
The bis-functionalized PDI 9 was synthesized by a new,
palladium-free method in only two steps via the tetraester of
perylenetetracarboxylic acid in high overall yields of 73%. For
reasons of solubility, a terphenylene-linker with n-octyl chains
was introduced. These linear substituents not only had a
positive effect on solubility, but also were needed to increase
the interphenylene twisting angle. According to the literature,
ortho-unsubstituted poly(para-phenylenes) show a twist of 23°
between consecutive repeating units along the phenylene
backbone, which allow for non-negligible π-overlap and are
greatly decreased upon ortho-substitution, thus reducing the
degree of conjugation of the bridging unit.11,12 The syntheses of
the individual building blocks are described below.
Spacer. Oligo para-phenylenes are ideal building blocks as
rigid, nonconjugated spacer moieties up to a spacer-length of
approximately 30 phenylene-units.13 The intrinsic twist
between the phenylene units was enhanced by introduction
of two n-octyl chains, thus optimizing solubility and minimizing
conjugation. By using symmetric 1,4-dibromo-2,5-diocytlben-
zene (13), the para-difunctionalized terphenylene unit was
synthesized by palladium catalyzed coupling reaction with p-
(trimethylsilyl)phenylboronic acid (Scheme 6). Subsequent
exchange of the TMS group gave the corresponding para-
diiodo terphenylene 28 in excellent yields, which was further
converted into the boronic acid pinacolester 24 via Miyaura-
borylation reaction. Conversion of TMS into the pinacolester
Multi-Suzuki Pathway. After the unsuccessful Stille and
Suzuki pathways (Scheme 3), we reconsidered the synthetic
strategy and tried to solve the reactivity problem of the aryl C−
C coupling pairs (e.g., 4 and 5, 11 and 15a, 12 and 15).
Regardless of the substitution pattern at the dioctylphenylene-
spacer and the diisopropylphenyl-group in the imide-position,
yields of direct aryl−aryl coupling of any rylene diimide 19 to
any dioctyl substituted oligo para-phenylene bridge 20 were
very low (Scheme 4, left). In our final synthetic procedure
described herein, the coupling reaction center was separated by
one unhindered phenylene spacer from the sterically hindered
bridging parts, which proved to be of crucial importance
(Scheme 4, right). Shifting the center of reaction away from the
dioctylphenyl-moiety enabled two-fold cross-coupling of the
bridge 22 with the respective rylene diimide, even in the case of
the less soluble fused TDI 11. With this finding, we identify the
right balance of solubility and reactivity as essential feature of
the final synthesis.
On the basis of this result, a convergent synthesis was
developed using two symmetric and one unsymmetric unit,
yielding the final dyad in decent amounts. All three final
building blocks, that is, dibromo diisopropylphenyl PDI 9,
monobromo-diisopropylphenyl TDI 11, and dioctyl-substituted
terphenyl diboronic acid pinacolester 24, can be synthesized in
up to a gram scale (Scheme 5).
E
DOI: 10.1021/acs.chemmater.5b04602
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
Scheme 8. Reaction Scheme for the Synthesis of the Acceptor Unit TDI 11
via the aryl iodide was found to be superior to the direct
conversion of TMS into the boronic acid for reasons of
purification.
yields due to shorter reaction times, thus minimizing the risk of
saponification or debromination, which can cause serious losses
in isolated yield.
PDI (Donor). The central element of the dyad is the bis-
functionalized perylenediimide 9. For reasons of stability and
solubility during all following steps, we chose the diisopropyl-
phenyl imide. Therefore, a quick and easy synthetic procedure
was introduced to yield dibromo-PDI 9 in only two steps
making use of readily available compounds (Scheme 7). PDI 9
has previously been prepared by a four-step procedure via
esterification of 4-chloro naphthalic anhydride (29), nickel
catalyzed homocoupling followed by imidization with aniline
33, and final cyclodehydrogenation.14,15 We instead converted
3,4,9,10-perylenetetracarboxylic dianhydride (PDA, 2) in two
easy steps to PDI 9. Since low reactivity of 4-bromo-2,6-
diisopropylaniline in combination with the insolubility of PDA
2 limits their direct conversion to 9, the esterification and
accompanying increase of solubility was of crucial importance.16
By starting with the hydrolysis of PDA 2 and base induced ester
formation, a highly soluble perylene tetracarboxylic-tetrabuty-
lester 34 was obtained, which only needed to be precipitated
from its chloroform solution to yield clean compound 34 in
high yields (89%). Imidization of the tetraester 34 with the
respective aniline 33 and stoichiometric amounts of ZnCl2 in
imidazole gave target dibromo PDI 9 on a multigram scale.
TDI (Acceptor). The acceptor-unit employed is the well-
established diisopropylphenyl-TDI 11.17 Synthetic procedures
(Scheme 8) and purification methods have been developed and
optimized by our group. Bromination of N-diisopropylphenyl-
perylene-3,4-dicarboxylic imide (PMI, 35) gave isomerically
pure monobrominated 36 in almost quantitative yields. After
borylation of 36 under Miyaura-reaction conditions and
subsequent Suzuki−Miyaura coupling with 4-bromo-1,8-naph-
thalic anhydride, the precursor 38 for the synthesis of the TDI-
core was obtained. Because of the comparably good solubility
of the unfused TDI precursor 38 and the higher reactivity of
the naphthalene monoimide-subunit compared to the former
case of PDA 2, the imidization reaction of 38 with p-bromo-2,6-
diisopropylaniline (33) could be executed in propionic acid at
elevated temperature. Reaction temperature and time both
were regulated to minimize the debrominated sideproduct.
Oxidative cyclodehydrogenation of 16 was conducted in
monoethanolamine with solid potassium carbonate as base.
Owing to the poor solubility of the starting compound 16 in
monoethanolamine, finely powdered substance led to improved
Dyad (A−D−A). The final coupling of the three building
blocks, dibromo PDI 9, boronic acid ester functionalized
terphenylene 24, and monobromo TDI 11, was performed
from the inside out (Scheme 5). By using a slight excess of 1.5
equiv of the terphenylene 24 for each bromo-functionality, the
terphenylene-elongated PDI 25 could be obtained without any
higher oligomers. Purification of 25 was done by column
chromatography on silica as well as on GPC-gel. Finally, both
TDI-groups were attached by standard Suzuki−Miyaura
coupling of PDI 25 with a slight excess of the TDI-moiety
11. By displacing the reaction center of the final C−C-coupling
by one phenylene group from the dioctylphenyl-moiety, the
final dyad 26 could be obtained in comparatively high yields of
32% after purification.
Absorption and Emission Spectra. Photophysical
investigation of target compound 26 proves that a donor−
acceptor system containing PDI and TDI has been established.
Absorption and emission spectra (Figure 3) were recorded in
toluene at concentrations ranging from 10−6 M to 10−8 M. The
emission spectrum of PDI (Figure 3, left) shows very good
overlap with the absorption of TDI, which is necessary for
Figure 3. Absorption and emission spectra of PDI 9 and TDI 11 in
toluene. Red line, absorption of PDI 9; green line, emission of PDI 9;
blue line, absorption of TDI 11. Shaded area displaying the spectral
overlaps of PDI-emission and TDI-absorption for Forster energy
̈
transfer.
F
DOI: 10.1021/acs.chemmater.5b04602
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
efficient Forster energy transfer. The absorption spectrum of
dyad 26 (Figure 4) appears as the superposition of the
acceptors are involved in the energy transfer process and how
they communicate after collecting energy from PDI. A very
promising surplus of the low temperature experiments relates
to the fact that the sharp zero-phonon transitions of the TDIs
in excitation and emission can be spectrally separated.
Consequently, because of the reduced spectral overlaps
compared to room temperature,19 it might be possible to
modify the energy flow from PDI to one of the TDIs by
selective saturation of the optical transition of this particular
TDI.
̈
EXPERIMENTAL SECTION
■
Perylenetetracaboxylic tetrabutylester,16 1,4-dibromo-2,5-dioctylben-
zene,20 and N-(4-bromo-2,6-diisopropylphenyl)-N′-(2,6-
diisopropylphenyl)terrylene-3,4:11,12-tetracarboxdiimide17 were pre-
pared as described in the literature.
N,N′-Bis(4-bromo-2,6-diisopropylphenyl)perylene-3,4:9,10-
tetracarboxylic bisimide (9). Perylenetetracaboxylic tetrabutylester
(34) (5.0 g, 7.66 mmol), ZnCl2 (2.09 g, 15.32 mmol), and 4-bromo-
2,6-diisopropylaniline (29.43 g, 114.89 mmol) were stirred in molten
imidazole (150 g) at 180 °C under argon atmosphere overnight. After
the mixture was cooled down to 100 °C, the slurry was poured into
150 mL of methanol. The precipitate was collected by filtration and
dried under vacuum. Purification by column chromatography on silica-
gel (dichloromethane) yielded compound 9 as a dark red solid (5.46 g,
82%). HR MALDI-TOF MS (dithranol): m/z = 866.1371, calc. for
Figure 4. Absorption and emission spectra of dyad 26 in toluene.
Black line, absorption of dyad 26; green line, emission of dyad 26; red
line, absorption of PDI 9; blue line, absorption of TDI 11.
absorptions of PDI and TDI in very good agreement with the
absorption spectra of the individual chromophores (red and
blue lines in Figure 4). The emission spectra were independent
of the excitation wavelength, giving rise to the same spectral
characteristics when excited at the PDI site (λexc. = 460 nm) or
the TDI site (λexc. = 600 nm). Upon excitation of PDI, fast and
efficient energy transfer from PDI to TDI occurs since the
1
C48H41Br2N2O4: 866.1355. H NMR (300 MHz, CD2Cl2) δ = 8.79
(m, 8H), 7.49 (s, 4H), 2.74 (p, J = 6.8 Hz, 4H), 1.16 (s, 12H), 1.13 (s,
12H). 13C NMR (75 MHz, CD2Cl2) δ = 149.2, 135.8, 132.6, 130.83,
130.76, 128.1, 127.4, 124.3, 124.2, 123.8, 29.9, 24.1. MP: > 400 °C.
distance between PDI and TDI is well within the Forster
̈
radius.3
̃
FT-IR: ν = 748, 812, 841, 862, 957, 970, 1176, 1198, 1250, 1344, 1356,
1404, 1452, 1459, 1468, 1577, 1593, 1662, 1674, 1705, 2850, 2868,
2924, 2956 cm−1.
CONCLUSION
■
2′,5′-Dioctyl-4,4″-bis(trimethylsilane)-1,1′:4′,1″-terpheny-
lene (27). The 1,4-dibromo-2,5-dioctylbenzene (2.32 g, 5.03 mmol),
4-trimethylsilylphenylboronic acid (2.15 g, 11.08 mmol), and K2CO3
(13.92 g, 100.69 mmol) were dissolved in a mixture of solvents of
toluene (50 mL), water (10 mL), and ethanol (2 mL). The mixture
was degassed by argon bubbling for 30 min before Pd(PPh3)4 (582
mg, 0.50 mmol) was added. The reaction was run at 95 °C under
argon atmosphere overnight. The aqueous phase was extracted with
toluene twice, and the combined organic solutions were dried over
Mg2SO4. The solution was concentrated under reduced pressure and
then filtered over a pad of silica. The solvent was removed by rotary
evaporation to give a yellow oil, which was treated with cold ethanol.
After shaking of the two phases, a precipitate formed, which was
filtered from the remaining solvent. The solid was washed with cold
methanol to give the title compound as a colorless solid (2.80 g, 93%).
1H NMR (300 MHz, CD2Cl2) δ = 7.59 (d, J = 8.1 Hz, 4H), 7.35 (d, J
After pursuing different strategies of PDI- and TDI-based A−
D−A synthesis, we presented a successful procedure for the
formation of a photophysically important compound consisting
of two TDI chromophores connected to a central PDI moiety
via a rigid terphenylene linker. A balance between solubility,
reactivity, and complexity of the synthesis was found, leading to
a reproducible strategy toward acceptor−donor systems
containing rylene diimides. By designing the chromophores
and linkers separately and combining them in the final stage,
our method of constructing A−D−A is not limited to the
presented structure but enables the synthesis of various related
compounds for photophysical purposes. Investigation of basic
photophysical behavior of the dyad presented herein gives
proof for the identity of the product. The negligible emission
from PDI in the recorded fluorescence spectra suggests highly
efficient energy transfer between donor and acceptor, which can
be explained by the close proximity of the chromophores within
a rigid scaffold. Comparison of absorption and emission spectra
of dyad 1 and dyad 26 displays the equivalence of both
structures for the given purposes, thus justifying the change
from bisphenylethyne to a terphenylene spacer in design for
reasons of synthetic feasibility.
= 8.1 Hz, 4H), 7.12 (s, 2H), 2.62−2.52 (m, 4H), 1.55−1.43 (m, 4H),
1.32−1.11 (m, 20H), 0.86 (t, J = 6.9 Hz, 6H), 0.32 (s, 18H). 13C
NMR (75 MHz, CD2Cl2) δ = 143.7, 142.1, 139.9, 138.8, 134.4, 132.3,
130.0, 33.8, 33.2, 32.9, 30.8, 30.61, 30.5, 24.0, 15.3, 0.00. HRMS (ESI
+): calc. for C40H63Si2 [M + H]+ 599.4461, found: 599.4468. MP: 78
̃
°C. FT-IR: ν = 627, 690, 725, 754, 825, 837, 1113, 1248, 1261, 1379,
1458, 1483, 1599, 2854, 2923, 2954 cm−1.
4,4″-Diiodo-2′,5′-dioctyl-1,1′:4′,1″-terphenylene (28). Com-
pound 27 (2.80 g, 4.67 mmol) was dissolved in dry dichloromethane
(80 mL). While being stirred in an ice bath, iodine monochloride (1
molar solution in CH2Cl2, 9.79 mL, 9.79 mmol) was added dropwise
under exclusion of light. After 2 h at 0 °C, the excess of iodine
monochloride was quenched by addition of saturated sodium sulfite
solution. The aqueous phase was extracted with dichloromethane, the
combined organic extracts dried over MgSO4, and finally the solvent
was removed under reduced pressure. The obtained dark colored oil
was dissolved in hot ethanol before the same amount of n-hexane was
added. Upon slowly cooling the mixture in a refrigerator, the product
precipitated as a fine colorless powder, which was collected by suction
In future experiments, a multistep sequence of energy
transfer events can be induced in 26 by selective excitation of
the PDI chromophore. In this scenario, the excited PDI first
transfers efficiently the excitation energy to any of both TDI
acceptors. This process can be studied in great detail by single-
molecule spectroscopy at liquid helium temperatures, which
not only allows one to deduce the energy transfer rate constant
from PDI to TDI from the zero-phonon line width of the PDI
donor,18 but also will reveal to which extent both TDI
G
DOI: 10.1021/acs.chemmater.5b04602
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
filtration and dried under vacuum. The final product was obtained as a
colorless solid (2.73 g, 83%). FD-MS: 706.20 (100%). HR MALDI-
TOF MS (dithranol): m/z = 706.1536, calc. for C34H44I2: 706.1532.
1H NMR (300 MHz, CD2Cl2) δ = 7.76 (d, J = 8.3 Hz, 4H), 7.10 (d, J
= 8.3 Hz, 4H), 7.07 (s, 2H), 2.57−2.49 (m, 4H), 1.50−1.36 (m, 4H),
1.34−1.09 (m, 20H), 0.86 (t, J = 6.8 Hz, 6H). 13C NMR (75 MHz,
CD2Cl2) δ = 142.0, 140.5, 138.1, 137.7, 131.9, 131.3, 92.8, 33.0, 32.4,
recorded. MP: > 400 °C. FT-IR: ν
̃
= 501, 740, 806, 841, 1014, 1039,
1072, 1095, 1120, 1261, 1358, 1379, 1462, 1585, 1666, 1705, 1727,
2848, 2870, 2916, 2957 cm−1.
ASSOCIATED CONTENT
■
S
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.chemma-
ter.5b04602.
̃
32.0, 30.0, 29.8, 29.7, 23.2, 14.5. MP: 70 °C. FT-IR: ν = 513, 534, 719,
741, 825, 1005, 1065, 1379, 1473, 2852, 2922, 2952 cm−1.
2′,5′-Dioctyl-4,4″-bis(4,4,5,5-tetramethyl-1,3,2-dioxaboro-
lane)-1,1′:4′,1″-terphenylene (24). The 4,4″-diiodo-2′,5′-dioctyl-
1,1′:4′,1″-terphenyl (1.41 g; 2.00 mmol), KOAc (1.41 g, 14.37 mmol),
and bis(pinacolato)diboron (1.52 g, 6.00 mmol) were suspended in
100 mL of dry DMF. After 30 min of argon bubbling, Pd(dppf)Cl2
(158 mg, 0.20 mmol) was added, and the mixture was stirred under
argon atmosphere at 85 °C for 2 h. After completion, the solvent was
removed under reduced pressure, and the black residue was purified by
column chromatography on silica-gel (petroleum ether/dichloro-
methane, 1:1) yielding 1.02 g (73%) of the title compound as a
colorless solid. HRMS (ESI+): calc. for C46H69B2O4 [M + H]+
All NMR spectra (PDF)
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: muellen@mpip-mainz.mpg.de.
Notes
The authors declare no competing financial interest.
1
707.5382, found: 707.5378. H NMR (300 MHz, CD2Cl2) δ = 7.82
(d, J = 8.1 Hz, 4H), 7.37 (d, J = 8.0 Hz, 4H), 7.12 (s, 2H), 2.67−2.44
(m, 4H), 1.54−1.41 (m, 4H), 1.37 (s, 24H), 1.32−1.11 (m, 20H), 0.86
(t, J = 6.8 Hz, 6H). 13C NMR (75 MHz, CD2Cl2) δ = 145.5, 141.4,
138.1, 135.0, 131.4, 129.3, 84.4, 33.2, 32.4, 32.1, 30.1, 29.8, 29.7, 25.3,
ACKNOWLEDGMENTS
■
We are grateful for financial support by the VW foundation and
BASF SE. Additionally, we thank Prof. Mark Watson for fruitful
discussions.
̃
23.3, 14.5. MP: 141 °C. FT-IR: ν = 511, 660, 741, 845, 860, 962, 1018,
1093, 1144, 1269, 1323, 1360, 1398, 1608, 2853, 2924, 2956, 2976
cm−1.
REFERENCES
■
Terphenylene−PDI−Terphenylene (25). PDI 9 (410 mg, 0.471
mmol) and terphenylene 24 (1.00 g, 1.42 mmol) were dissolved in
toluene (350 mL) before a solution of K2CO3 (1.96 g, 14.15 mmol) in
water, and some drops of ethanol were added. The mixture was
degassed by argon bubbling for 30 min before Pd(PPh3)4 (55 mg,
0.047 mmol) was added, and then the mixture was stirred at 85 °C
overnight. After consumption of the starting compound 9, the two
phases were separated, the organic solvent removed by rotary
evaporation, and the crude mixture purified by column chromatog-
raphy on silica with CH2Cl2 as eluent. Final purification was done by
GPC in toluene, and the title compound was isolated as orange-red
(1) Ensslen, P.; Wagenknecht, H.-A. One-Dimensional Multi-
chromophor Arrays Based on DNA: From Self-Assembly to Light-
Harvesting. Acc. Chem. Res. 2015, 48, 2724−2733.
(2) Kozaki, M.; Suzuki, S.; Okada, K. Dendritic Light-harvesting
Antennas with Excitation Energy Gradients. Chem. Lett. 2013, 42,
1112−1118.
́ ́
(3) Metivier, R.; Nolde, F.; Mullen, K.; Basche, T. Electronic
̈
Excitation Energy Transfer Between two Single Molecules Embedded
in a Polymer Host. Phys. Rev. Lett. 2007, 98, 047802.
(4) Kim, H. N.; Puhl, L.; Nolde, F.; Li, C.; Chen, L.; Basche,
Mullen, K. Energy Transfer at the Single-Molecule Level: Synthesis of
a Donor-Acceptor Dyad from Perylene and Terrylene Diimides. Chem.
- Eur. J. 2013, 19, 9160−9166.
(5) Bell, T. D. M.; Jacob, J.; Angeles-Izquierdo, M.; Fron, E.; Nolde,
́
T.;
1
̈
solid (480 mg, 54%). FD-MS: 1867.35 (100%). H NMR (300 MHz,
CD2Cl2) δ = 8.83 (s, 8H), 7.84 (d, J = 8.0 Hz, 4H), 7.80 (d, J = 8.2 Hz,
4H), 7.65 (s, 4H), 7.52 (d, J = 8.2 Hz, 4H), 7.40 (d, J = 8.0 Hz, 4H),
7.22 (s, 2H), 7.16 (s, 2H), 2.86 (p, J = 6.8 Hz, 4H), 2.72−2.54 (m,
8H), 1.62−1.41 (m, 8H), 1.37 (s, 24H), 1.32−1.17 (m, 64H), 0.91−
0.81 (m, 12H). 13C NMR (75 MHz, CD2Cl2) δ = 164.3, 147.1, 145.5,
142.8, 141.9, 141.4, 141.1, 140.2, 138.2, 138.1, 135.8, 135.0, 132.6,
131.6, 131.4, 131.1, 130.8, 130.4, 129.3, 127.7, 127.5, 124.2, 124.0,
123.7, 84.4, 33.19, 32.4, 32.13, 32.08, 30.3, 30.10, 30.08, 29.88, 29.85,
29.76, 29.74, 25.3, 24.4, 23.3, 14.5. EA calc. for C128H152B2N2O8, C:
82.29; H: 8.20; N: 1.50; found C: 82.13; H: 8.22; N: 1.50. MP: 278
F.; Hofkens, J.; Mullen, K.; De Schryver, F. C. Charge Transfer
̈
Enhanced Annihilation Leading to Deterministic Single Photon
Emission in Rigid Perylene End-Capped Polyphenylenes. Chem.
Commun. 2005, 39, 4973−4975.
(6) Fron, E.; Puhl, L.; Oesterling, I.; Li, C.; Mullen, K.; De Schryver,
̈
F. C.; Hofkens, J.; Vosch, T. Energy Transfer Pathways in a Rylene-
Based Triad. ChemPhysChem 2011, 12, 595−608.
(7) Zhan, X.; Facchetti, A.; Barlow, S.; Marks, T. J.; Ratner, M. A.;
Wasielewski, M. R.; Marder, S. R. Rylene and Related Diimides for
Organic Electronics. Adv. Mater. 2011, 23, 268−284.
̃
°C. FT-IR: ν = 513, 521, 660, 748, 804, 812, 841, 858, 960, 1016,
1090, 1144, 1176, 1196, 1254, 1269, 1321, 1344, 1356, 1404, 1468,
1579, 1595, 1668, 1707, 2854, 2869, 2924, 2958 cm−1.
(8) Weil, T.; Vosch, T.; Hofkens, J.; Peneva, K.; Mullen, K. The
̈
TDI−PDI−TDI Dyad (26). Compound 25 (20 mg, 0.011 mmol)
and monobromo-TDI 11 (39 mg, 0.043 mmol) were dissolved in o-
xylene (7.5 mL), a solution of K2CO3 (59 mg, 0.428 mmol) in water
(2 mL), and ethanol (1.5 mL) was added. After degassing for 15 min
by argon bubbling, Pd(PPh3)4 (1 mg) was added, and the reaction
mixture was stirred at 95 °C under argon atmosphere overnight. After
completion, the organic phase was separated from the aqueous phase,
and the solvent was removed under reduced pressure. The crude
mixture was subjected to size-exclusion chromatography with toluene
as eluent. The product was obtained as a dark blue solid (11 mg, 32%).
HR MALDI-TOF MS (dithranol): m/z = 3279.6709, calc. for
C232H218N6O12: 3279.6633. 1H NMR (300 MHz, CD2Cl2 + CS2
7:1) δ = 8.88−8.68 (m, 32H), 7.77 (d, J = 7.9 Hz, 8H), 7.61 (d, J
= 2.2 Hz, 8H), 7.51 (d, J = 7.7 Hz, 8H), 7.35 (d, J = 7.7 Hz, 4H), 7.23
(s 4H), 2.91−2.66 (m, 20H), 1.65−1.48 (m, 8H, superimposed with
H2O), 1.38−1.25 (m, 90H), 1.19 (d, J = 6.8 Hz, 24H), 0.91 (m, 12H).
13C NMR: because of low solubility, no carbon NMR could be
Rylene Colorant Family-Tailored Nanoemitters for Photonics
Research and Applications. Angew. Chem., Int. Ed. 2010, 49, 9068−
9093.
(9) Herrmann, A.; Mullen, K. From Industrial Colorants to Single
̈
Photon Sources and Biolabels: The Fascination and Function of
Rylene Dyes. Chem. Lett. 2006, 35, 978−985.
́
(10) Fuckel, B.; Hinze, G.; Nolde, F.; Mullen, K.; Basche, T.
̈
̈
Quantification of the Singlet-Singlet Annihilation Times of Individual
Bichromophoric Molecules by Photon Coincidence Measurements. J.
Phys. Chem. A 2010, 114, 7671−7676.
(11) Curutchet, C.; Feist, F. A.; Van Averbeke, B.; Mennucci, B.;
́
Jacob, J.; Mullen, K.; Basche, T.; Beljonne, D. Superexchange-
̈
Mediated Electronic Energy Transfer in a Model Dyad. Phys. Chem.
Chem. Phys. 2010, 12, 7378−7385.
(12) Tour, J. M. Soluble Oligo- and Polyphenylenes. Adv. Mater.
1994, 6, 190−198.
H
DOI: 10.1021/acs.chemmater.5b04602
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
(13) Vanhee, S.; Rulkens, R.; Lehmann, U.; Rosenauer, C.; Schulze,
M.; Kohler, W.; Wegner, G. Synthesis and Characterization of Rigid
̈
Rod Poly (p-phenylenes). Macromolecules 1996, 29, 5136−5142.
(14) Kelber, J.; Achard, M.-F.; Durola, F.; Bock, H. Distorted Arene
Core Allows Room-Temperature Columnar Liquid-Crystal Glass with
Minimal Side Chains. Angew. Chem., Int. Ed. 2012, 51, 5200−5203.
(15) Frischmann, P. D.; Wurthner, F. Synthesis of a Non-aggregating
̈
Bay-unsubstituted Perylene Bisimide Dye with Latent Bromo Groups
for C-C Cross Coupling. Org. Lett. 2013, 15, 4674−4677.
(16) Mo, X.; Shi, M. M.; Huang, J. C.; Wang, M.; Chen, H. Z.
Synthesis, Aggregation and Photoconductive Properties of Alkox-
ycarbonyl Substituted Perylenes. Dyes Pigm. 2008, 76, 236−242.
(17) Nolde, F.; Qu, J.; Kohl, C.; Pschirer, N. G.; Reuther, E.; Mullen,
̈
K. Synthesis and Modification of Terrylenediimides as High-
Performance Fluorescent Dyes. Chem. - Eur. J. 2005, 11, 3959−3967.
́ ́
(18) Metivier, R.; Kulzer, F.; Weil, T.; Mullen, K.; Basche, T. Energy
̈
Transfer Rates and Pathways of Single Donor Chromophores in a
Multichromophoric Dendrimer Built around a Central Acceptor Core.
J. Am. Chem. Soc. 2004, 126, 14364−14365.
́ ́
(19) Hinze, G.; Metivier, R.; Nolde, F.; Mullen, K.; Basche, T.
̈
Intramolecular Electronic Excitation Energy Transfer in Donor/
Acceptor Dyads Studied by Time and Frequency Resolved Single
Molecule Spectroscopy. J. Chem. Phys. 2008, 128, 124516.
(20) Rehahn, M.; Schluter, A.-D.; Feast, W. J. A High-Yield Route to
̈
2,5-Di-n-alkyl-1,4-benzenedicarboxylic Acids. Synthesis 1988, 1988,
386−388.
I
DOI: 10.1021/acs.chemmater.5b04602
Chem. Mater. XXXX, XXX, XXX−XXX