An artificial light-harvesting array constructed from multiple bodipy dyes

Article abstract of DOI:10.1021/ja4049306

An artificial light-harvesting array, comprising 21 discrete chromophores arranged in a rational manner, has been synthesized and characterized fully. The design strategy follows a convergent approach that leads to a molecular-scale funnel, having an effective chromophore concentration of 0.6 M condensed into ca. 55 nm³, able to direct the excitation energy to a focal point. A cascade of electronic energy-transfer steps occurs from the rim to the focal point, with the rate slowing down as the exciton moves toward its ultimate target. Situated midway along each branch of the V-shaped array, two chromophoric relays differ only slightly in terms of their excitation energies, and this situation facilitates reverse energy transfer. Thus, the excitation energy becomes spread around the array, a situation reminiscent of a giant holding pattern for the photon that can sample many different chromophores before being trapped by the terminal acceptor. At high photon flux under conditions of relatively slow off-load to a device, such as a solar cell, electronic energy transfer encounters one or more barriers that hinder forward progress of the exciton and thereby delays arrival of the second photon. Preliminary studies have addressed the ability of the array to function as a sensitizer for amorphous silicon solar cells.

Full text of DOI:10.1021/ja4049306

Subscriber access provided by University of South Dakota
Article
An Artificial Light-Harvesting Array Constructed from Multiple Bodipy Dyes
Raymond Ziessel, Gilles Ulrich, Alexandre Haefele, and Anthony Harriman
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/ja4049306 • Publication Date (Web): 03 Jul 2013
Downloaded from http://pubs.acs.org on July 10, 2013
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of the American Chemical Society is published by the American Chemical
Society. 1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 17
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
An Artificial Light-Harvesting Array Constructed from Multiple
Bodipy Dyes
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Raymond Ziessel,*,a Gilles Ulrich,a Alexandre Haefele,a and Anthony Harriman*,b
aLaboratoire de Chimie Organique et Spectroscopies Avancées (LCOSA), UMR 7515 au CNRS, Ecole Européenne
de Chimie, Polymères et Matériaux, Université de Strasbourg, 25 rue Becquerel, 67087 Strasbourg Cedex 02,
France
bMolecular Photonics Laboratory, School of Chemistry, Bedson Building, Newcastle University, Newcastle upon
Tyne, NE1 7RU, United Kingdom
KEYWORDS: Bodipy dyes, molecular photophysics, fluorescence, electronic energy transfer, artificial photosyn-
thesis.
Supporting Information
ABSTRACT: An artificial light-harvesting array, comprising 21 discrete chromophores arranged in a rational manner,
has been synthesized and characterized fully. The design strategy follows a convergent approach that leads to a molecu-
lar-scale funnel, having an effective chromophore concentration of 0.6 M condensed into ca. 55 nm3, able to direct the
excitation energy to a focal point. A cascade of electronic energy-transfer steps occurs from the rim to the focal point,
with the rate slowing down as the exciton moves towards its ultimate target. Situated midway along each branch of the
V-shaped array, two chromophoric relays differ only slightly in terms of their excitation energies and this situation facil-
itates reverse energy transfer. Thus, the excitation energy becomes spread around the array, a situation reminiscent of a
giant holding pattern for the photon that can sample many different chromophores before being trapped by the terminal
acceptor. At high photon flux under conditions of relatively slow off-load to a device such as a solar cell, electronic ener-
gy transfer encounters one or more barriers that hinder forward progress of the exciton and thereby delays arrival of the
second photon. Preliminary studies have addressed the ability of the array to function as a sensitizer for amorphous sili-
con solar cells.
the light-harvesting machinery usually involves some
kind of energy-dependent quenching,9 although the de-
tails are still under active debate.10 Incorporation of self-
protection, however primitive this might be, is seen as an
essential feature of any robust artificial system developed
for solar energy conversion.11 Otherwise, the array will
not survive the harsh conditions associated with pro-
longed exposure of a sophisticated, multi-component
organic molecule to sunlight. Here, we describe an at-
tempt to construct an artificial light-harvesting array
able to dissipate excess photonic energy through a self-
regulating mechanism. At the molecular level, the array
is intended to collect photons over most of the visible
range, terminating in a threshold wavelength appropriate
for the sensitization of an amorphous silicon solar cell,12
and channel the energy to an acceptor that can be an-
chored to a solid surface.13 The capability to re-distribute
excess excitation energy is built into the system at an
early stage. Further engineering of next-generation pro-
totypes should ensure improved levels of photostability.
Introduction
The complex machinery of the photosynthetic appa-
ratus found in higher plants serves two essential purpos-
es; namely, to harvest sunlight and direct the resultant
excitation energy to a reaction center1 but also to provide
protection against photochemical damage.2 At first
glance, these features appear to be mutually opposing!
The collection of sunlight by artificial arrays is well doc-
umented3-5 and there are numerous descriptions of ra-
tionally designed molecular systems displaying effica-
cious intramolecular electronic energy transfer (EET).
Additional attention has been given to the construction
of plastic solar concentrators6 and to the formulation of
light-harvesting sensitizers7 for use with specific types of
solar cell. Much less attention, however, has been given
to the problem of equipping such arrays with some form
of self-protection against deleterious photochemical
damage. In the natural system,8 various mechanisms
have evolved to ensure plant survival on those inevitable
occasions when highly reactive intermediate species arise
from exposure to a high photon flux. Such regulation of
In designing the molecular array, we have relied
heavily on prior work14-16 using the boron dipyrrome-
1
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 17
thene (Bodipy) family of highly fluorescent dyes. Such wards the terminal acceptor. Energy transfer to the latter
1
2
3
4
5
6
7
8
materials, possessing triplet quantum yields well below
5% in the absence of spin orbit perturbation, have been
exploited for numerous photonic devices, including cas-
cade energy-transfer processes,17 flow cytometry,18 pro-
tein labeling,19 cell internalization,20 bio-labeling,21
chemical probes22 and two-photon imaging.23 The facile
chemical modification of the generic Bodipy framework24
provides the impetus to construct elaborate molecular
architectures whereby selected dyes can be arranged in a
logical sequence and where the optical properties of indi-
vidual dyes can be fine-tuned25 for particular purposes. A
key discovery26 in this field relates to the replacement of
the conventional B-F bonds with B-C analogues since
this permits the attachment of a multitude of disparate
functionalities and diversification of the dimensionality
of the construct. The “chemistry at boron” approach27 has
been applied here to covalently attach a total of 20 ancil-
lary chromophores (including 6 tolane linkers) to the
terminal acceptor, thereby producing a cascade-type mo-
lecular array capable of directed EET (Chart I). The gra-
dient of excitation energies is intended to facilitate EET
from the outermost pyrene (PY) to the expanded Bodipy
(EXBod), with the intermediate dyes functioning as en-
ergy relays. The mechanism for each EET step, that is the
distinction between through-bond or through-space in-
teractions, is not important in the present context but the
overall EET efficiency is critical.
might be restricted if it is already in an excited state (i.e.,
the rate of EET depends on the state of excitation of the
terminal acceptor). As such, a key feature of the molecu-
lar topology is to provide diverse routes for the exciton
that will ensure disparate arrival times. It might be ar-
gued that such behavior is best examined at the single-
molecule level where light intensities can be varied dra-
matically.30 There have, in fact, been several studies of
the exciton blockade carried out with single-molecule
dendrimers31 but the impact of conformational heteroge-
neity prevents unambiguous recognition of the effect.
Structural distortions that fluctuate on the nanosecond
timescale32 are always likely to interfere with data analy-
sis in elaborate multi-component molecular arrays33 and
must be considered during data interpretation.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Perhaps it is instructive to draw attention to other
molecular arrays that display cascade-type EET, neglect-
ing linear dyads and triads. The most relevant architec-
tures are dendrimers34 and there are numerous such en-
tities able to direct excitons from the periphery to the
core by way of successive EET steps.35 These dendritic
frameworks can be flexible36 or semi-rigid37 and, because
of their topology, EET does not need to follow a single
branch. Indeed, energy hopping between branches could
serve to distribute the excitation energy around the den-
drimer before being localized at the central unit.35 Conju-
gated polymers also provide facile means for both inter-
and intra-strand EET38 that leads to widespread redistri-
bution of energy throughout the matrix. Advantages of
the cluster described herein are that the terminal energy
acceptor is readily available for anchoring to a surface
where the exciton can be off-loaded, easy control of the
rates of EET via slight modification of the bridging units,
and simple tuning of the optical properties so as to meet
any desired threshold wavelength (e.g., sensitization of a
particular semiconductor). Other strategies, such as dye-
decorated platforms39 or self-assembly of hetero-dyes40
should also be considered. All such architectures, where
multiple dyes are brought into close proximity, are likely
to exhibit nonlinear optical behavior in respect of inci-
dent photon densities.41 Again, it should be stressed that
single-molecule approaches42 offer obvious advantages
for attaching several photons to a solitary array.
Chart 1. Molecular formula of the target array
comprising 8 pyrene units (grey), attached in
pairs to 4 tetramethyl-Bodipy dyes (green) that
are themselves bundled in pairs to 2 fully alkyl-
ated Bodipy dyes (orange) that are linked to the
terminal expanded Bodipy dye (blue). Note the
use of tolane spacer units to separate the various
Bodipy dyes. There is an excitation energy gradi-
ent running from pyrene to the expanded Bodipy
dye. See Supporting Information (Figure S1) for
a 3D representation of the cluster as generated
by molecular dynamics simulations.
Results and Discussion
Design and synthesis
The target molecular entity is intended to serve two func-
tions; namely, (i) the collection of photons over a wide
spectral window, with subsequent channeling of the exci-
tation energy to a single site, and (ii) the dissipation of
excess photonic energy by way of harmless processes.
The first requirement leads to the design of a molecular
funnel comprising a series of closely spaced chromo-
phores graded so as to favor a cascade of EET steps lead-
ing to a stable fluorescent species. Many such systems
are known43 and we have opted to use Bodipy building
blocks for most of the construction work. The second
requirement, being essentially unknown in artificial sys-
tems but being an integral part of all natural photosys-
tems,8 is much more demanding. Our approach to this
The design principle is based on the exciton blockade
approach introduced some years ago,28 this being a de-
rivative of the better known coulomb blockade recog-
nized in low temperature physics.29 The general idea is
that, under intense illumination, several excitons will be
added to the light-harvesting array and will migrate to-
2
ACS Paragon Plus Environment

Page 3 of 17
Journal of the American Chemical Society
problem involves using a pair of chromophores for which
tions for isolation of the target dye in acceptable yield
(Scheme 1).
1
2
3
4
5
6
7
8
the individual energy levels are sufficiently similar for
the respective excited states to establish a thermal equi-
librium.44 The final design links 21 chromophores in an
ordered array; individual units being 8 ethynylpyrene
(PY) residues, 4 tetramethyl-Bodipy (TMBod) dyes, 2
tetramethyldiethyl-Bodipy (MEBod) dyes, 6 tolane func-
tions and an expanded Bodipy (EXBod) dye as the ter-
minal acceptor. The most logical strategy by which to
construct such an edifice is to use a convergent protocol45
whereby the peripheral chromophores are introduced
early in the synthesis. A critical feature of this approach
concerns the need to provide for adequate solubility and
characterization at each stage of evolution so that the
stoichiometry can be assured.
All intermediates and target compounds were ana-
lyzed by proton NMR spectroscopy and a selection of
relevant spectra is given in Figure 1. For compound 3b,
for example, the presence of a singlet for both β-pyrrolic
protons at 6.2 ppm, which integrates for two protons,
compared to the doublet at 8.8 ppm (integration for two
protons) belonging to the pyrene units allows us to con-
clude that double substitution has occurred at the boron
center (Figure 1). Interestingly, the spectrum of com-
pound 7b resembles a linear combination of spectra for
compounds 3b and 6 without notable overlapping be-
tween respective peaks. In particular, integration of the
ethyl peaks compared to the methyl peaks of dyes 6 and
3b confirms that double substitution of both iodo-groups
has taken place. This situation was confirmed subse-
quently by the Maldi-TOF MS spectrum that exhibits a
molecular peak at 2087.0/2086.0 amu, with an isotopic
profile in agreement with the calculated spectrum.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Isolation of the two conventional Bodipy-based
synthons involved similar but not identical protocols.
Thus, the tetramethyl-Bodipy analogue 3 was prepared
in three steps, starting with selective substitution at the
B-F bonds using a Grignard reagent obtained from 1-
ethynylpyrene and leading to isolation of compound 2.
This was followed by functionalization of the iodo-group
using trimethylsilylacetylene and deprotection with a
mineral base, thereby affording compound 3 in an over-
all yield of 58% (left hand side of Scheme 1). Also
straightforward is the preparation of synthon 6 by a se-
quence of reactions starting by cross coupling with tri-
ethylsilylacetylene. This latter reaction was promoted by
Pd(0) complexes and led to isolation of compound 5.
This was followed by reaction with the Grignard reagent
formed from 1-ethynyl-4-iodobenzene, thereby affording
derivative 6 in an overall yield of 70% (right hand side of
Scheme 1). The final step involves cross-coupling two
equivalents of 3 with one equivalent of 6 using a Pd(0)
complex in the absence of copper salts and molecular
oxygen. The pivotal intermediate 7 was isolated as a pure
sample in excellent yield by standard separation tech-
niques (see Supporting Information).
The well-defined NMR spectrum found for the final
compound 9 is also quite informative (Figure 2). The
methoxy signal for this terminal blue dye indicates that
both iodine moieties present in 8 have been replaced
with the intended modules. In particular, integration of
this methoxy signal (singlet at 3.8 ppm with integration
for 6 protons) perfectly matches the 8-proton integration
of the β-pyrrolic signal belonging to the incoming four-
component module 7b. Further confirmation is provided
by the 4-proton integration of the doublet at 7.05 ppm
belonging to the AB quartet of the anisole fragment of
the blue dye which matches the expected 8 protons (lo-
calized at the ortho position of the alkyne) for 8 pyrene
residues at 8.8 ppm (Figure 2). Confirmation of the mo-
lecular composition of the final dye is provided by the
Maldi-TOF MS spectrum that exhibits a molecular peak
at 4836.8/4837.8 amu, with an isotopic profile in agree-
ment with the calculated spectrum. Treating the edifice
as a cone of height 48 Å and radius 33 Å containing 21
chromophores leads to an estimate for the effective pig-
ment concentration as being ca. 0.6 M. The threshold
wavelength for absorbing sunlight is ca. 670 nm, this
being appropriate for sensitization of amorphous silicon
solar cells, and with the target cluster dispersed in a plas-
tic matrix approximately 30% of the solar spectrum can
be harvested without self-absorption or concentration
quenching. This is a marked improvement on conven-
tional plastic solar concentrators.6
The final, and indeed critical, step consists of cross
coupling two equivalents of module 7b with the terminal
acceptor 8. The major difficulty here is to circumvent the
strong tendency of 7b to form the homo-coupled com-
pound (i.e., the butadiyne-bridged derivative). This type
of reaction is usually promoted by copper salts,46 but can
be catalyzed also with palladium salts in specific cases.47
To find ways around this undesirable side reaction, many
different experimental conditions were tried; these in-
cluded changes in solvent, base, catalyst, concentration,
and temperature. On the basis of these studies, it appears
that the use of a low concentration of catalyst and the
strict avoidance of copper salts provides the best condi-
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 17
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 1. Key: (i) 1-ethynylpyrene, EtMgBr, THF, 60 °C; (ii) trimethylsilylacetylene, [Pd(PPh3)2Cl2] (6 mol%), CuI (10
mol%), diisopropylamine, THF, rt; (iii) KOH, MeOH, H2O, rt; (iv) triethylsilylacetylene, [Pd(PPh3)2Cl2] (6 mol%), CuI
(10 mol%), di-isopropylamine, THF, rt; (v) 1-ethynyl-4-iodophenyl, EtMgBr, THF, 60 °C; (vi) compound 3 (2 equiv.),
compound 6 (1 equiv.) [Pd(PPh3)4] (6 mol%), triethylamine, benzene, 60 °C; (vii) [Pd(PPh3)4] (2 mol%), triethylamine,
benzene, 60 °C.
4
ACS Paragon Plus Environment

Page 5 of 17
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
Figure 1. Proton NMR spectra of the key model compound 3b, 6 and 7b from top to bottom, recorded in CDCl3 at 300 MHz.
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2. Proton NMR spectra of the key model compound 9 recorded at room temperature in CDCl3 at 400 MHz.
5
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 17
Photophysical investigations
1
The starting point for this investigation lies with explor-
ing EET from 1-ethynylpyrene (PY) to the appended tet-
ramethyl-Bodipy (TMBod) derivative, taking note that
two PY units are attached to the dye by way of the boron
center. Similar molecular dyads have been studied be-
fore,48 having various aryl polycycles attached to the bo-
ron atom (or meso-carbon), and it has been established
that EET from polycycle to Bodipy is rapid and essential-
ly quantitative. The same situation is found for 3b (here-
after abbreviated as PMe for simplicity) in 2-
methyltetrahydrofuran (MTHF) at room temperature.
Thus, the absorption spectrum (Figure 3) indicates that
the TMBod chromophore exhibits pronounced absorp-
tion bands with a maximum at 501 nm and a vibronic
progression stretching toward higher energies. Absorp-
tion transitions localized on PY are seen as a series of
well-resolved bands in the wavelength range 300 to 390
nm. These latter transitions overlap with higher-energy
absorption bands associated49 with TMBod. Excitation
into the PY unit at around 365 nm leads to the appear-
ance of strong fluorescence with a maximum at 518 nm
that is characteristic of TMBod (Figure 3); the same
spectral profile is observed following illumination at 470
nm where only TMBod absorbs. The excitation spectrum
matches well with the absorption spectrum over the en-
tire spectral window and it is clear that photons collected
by PY are transferred quantitatively to TMBod under
these conditions. Both the fluorescence quantum yield
(ΦF = 0.78 ± 0.03) and the excited-state lifetime (τS = 4.6
± 0.1 ns) recorded for TMBod remain independent of
excitation wavelength. In addition, the time-resolved
fluorescence decay profiles obtained after excitation at
310, 470 or 490 nm are mono-exponential. These ΦF and
τS values are unexceptional50 for a conventional Bodipy
dye.
For PMe in MTHF it was not possible to resolve fluo-
2
3
4
5
6
7
8
9
rescence characteristic of PY from the baseline, even in a
glassy matrix at 77K. Emission from the pyrene unit is
expected over the region 430-500 nm and will overlap
with absorption by the Bodipy chromophore (Figure S2).
Unlike the situation with other pyrene-Bodipy dyads,25c
the absorption spectral profile for PMe favors EET to the
first-excited singlet state resident on TMBod. Compari-
son to a 2:1 molar mixture of 1-ethynylpyrene and 1 in
MTHF led to the conclusion that the level of quenching
of the excited-singlet state of the PY unit present in PMe
exceeds 99.5% while, on the basis of time-resolved emis-
sion studies, the PY fluorescence lifetime is less than 10
ps (cf τS of 3.0 ns for 1-ethynylpyrene50). The center-to-
center separation distance is ca. 8.5 Å while the connec-
tion at the pyrene terminal is likely to favor through-
bond EET.51 Consequently fluorescence quenching is
attributed to intramolecular EET, as has been reported
for many related dyads25 and used15 to extend the Stokes’
shift for Bodipy-based fluorescent labels.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Turning attention now to 7b (hereafter abbreviated
PMEt for simplicity) where the additional Bodipy residue
has been added, we note that in structural terms the new
fluorophore differs from TMBod only by the presence of
two ethyl groups on the dipyrrin nucleus; we will refer to
this new dye as MEBod to signify the presence of both
methyl and ethyl functions. These ethyl groups push the
absorption maximum from 501 nm to 522 nm in MTHF
at room temperature.14 The individual Bodipy-based
transitions can be resolved in the absorption spectrum,
although the stoichiometry has to be taken into account
when considering the relative intensities (Figure 3). No-
tice should also be given to the arrangement of these
dyes and, in particular, their proximity to the PY unit
which will serve as the principal excitation point for EET
studies. Thus, following illumination at 365 nm, the fluo-
rescence spectrum shows no indication for emission
from the PY unit but exhibits two overlapping fluores-
cence profiles with maxima at 518 and 540 nm (Figure
3). This dual emission can be attributed to fluorescence
from both TMBod and MEBod, although the ratio is
clearly in favor of the latter when due allowance is made
for the relative ΦF values.52 Excitation spectra recorded
for emission at 600 nm confirm that photons absorbed
by both PY and TMBod units lead to emission from the
MEBod fluorophore (Figure S3). Excitation into TMBod
at 475 nm gives rise to the same dual emission profile as
observed on illumination into PY. Consequently, we can
conclude that intramolecular EET occurs from TMBod to
MEBod in the array. Deconvoluting the spectral profile
into components associated with TMBod and MEBod,
spectra being available for the reference compounds,
indicates the former accounts for only 7% of the total
fluorescence signal (Figure S4).
(a)
(b)
300
400
500
600
700
The control compound for MEBod, namely 4, shows
strong fluorescence in MTHF at room temperature. Here,
ΦF = 0.75 ± 0.03 while τS = 4.8 ± 0.2 ns; these values
being normal14 for a conventional Bodipy dye. Time-
resolved fluorescence studies indicate that τS for the ME-
Bod residue present in PMEt is reduced slightly to 4.5 ±
λ / nm
Figure 3. Normalized absorption (black curve) and fluores-
cence (grey curve) spectra recorded for (a) PMe (3b) and (b)
PMEt (7b) in MTHF at room temperature. The excitation
wavelength was 365 nm.
6
0.2 ns; fluorescence quantum yields for MEBod and
ACS Paragon Plus Environment

Page 7 of 17
Journal of the American Chemical Society
More reliable results were obtained by time-
PMEt (λEX = 515 nm) are within ±8%. There is no sugges-
1
2
3
4
5
6
7
8
correlated, single photon counting with excitation at 340
nm delivered from a picosecond laser diode (average
power 1 mW at 40 MHz) and with detection via a cooled
microchannel plate PMT. The instrument response func-
tion recorded under optimized conditions was <70 ps.
Analysis with iterative re-convolution of emission from
the TMBod unit indicated that decay occurs via dual-
exponential kinetics. The shorter-lived component,53
accounting for ca. 90% of the total signal, gave a con-
sistent estimate for τS of 35 ± 6 ps (Figure 5). The slower
component required analysis on much longer timescales
but was found to correspond to a lifetime of 4.0 ± 0.6 ns.
Monitoring at longer wavelength showed that emission
associated with MEBod grows-in with a time constant of
39 ± 5 ps but barely decays over several hundred ps.
Global analysis across the 500-600 nm window was con-
sistent with a reaction scheme whereby the excited-
singlet state of TMBod decays with a lifetime of ca. 40 ps
to populate the corresponding excited-singlet state resi-
tion that the slight shortening of the lifetime is indicative
of (e.g., electron transfer) quenching in the array and
instead we attribute this minor effect to some kind of
geometric distortion introduced at the S1 level by impos-
ing bulky substituents at the boron atom.17f Somewhat
surprisingly, emission from the TMBod unit in PMEt also
decays with this same lifetime, no matter how carefully
this measurement is made with respect to excluding fluo-
rescence from MEBod. Close spectral overlap between
the two fluorescence profiles, together with a small
amount of hot fluorescence if excitation is made in the
near-UV region, means it is difficult to completely ex-
clude emission from MEBod when examining the region
associated with TMBod; the best interrogation wave-
length is 510 nm (Figure S5).
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In this regard it is important to note that the fraction-
al contribution made by TMBod emission increases
slightly with increasing temperature over the range 255K
to 385K (Figure 4), thereby indicating that the two emit-
ting species are in thermal equilibrium44 over this region.
Assuming the ratio of the fluorescence yields is set by the
Boltzmann distribution law, the energy gap between the
two states is estimated to be 700 cm-1 (i.e., 8.4 kJ/mol).
This can be compared with the spectroscopic energy gap
of 795 cm-1 derived from absorption and emission spec-
tra of the reference compounds. In turn, the calculated
energy gap can be used to estimate the equilibrium con-
stant, K = 0.04, at room temperature.
30000
25000
20000
15000
10000
5000
0
1.00
0.75
0.50
0.25
0.00
0
50
100
150
200
250
300
350
400
Time / ps
dent on MEBod with the same time constant (Figure 6).
Figure 5. Time-resolved emission profiles recorded for
PMEt in MTHF at RT following excitation at 340 nm. The
IRF is shown as a grey curve while fluorescence from
TMBod at 510 nm is shown as a red curve with the fit (χ2 =
1.09) is superimposed as a black curve. Growth of fluores-
cence from MEBod at 590 nm is shown as a blue curve with
the fit (χ2 = 1.24) superimposed as a black curve. See Figure
S6 for a plot of the weighted residuals.
520
560
600
640
λ / nm
Figure 4. Effect of increasing temperature on the fluores-
cence spectrum recorded for PMEt (7) in MTHF; the insert
shows an expansion of the wavelength range from 500 to
520 nm.
Measurement of the rate of intramolecular EET from
TMBod to MEBod was not straightforward because of the
significant overlap between fluorescence from the two
Bodipy residues (Figure S5). Thus, it proved difficult to
isolate emission from either of these dyes without con-
tamination from the second dye. In principle, the re-
quired level of resolution might be expected from fre-
quency-domain measurements made with phase modu-
lation techniques but this was not so. Here, the problem
was poor sensitivity, although the lifetime of the MEBod
(τS = 4.5 ± 0.3 ns) could be recovered on excitation into
either PY at 310 nm or one of the Bodipy dyes at 490 nm.
It should be recalled that the temperature depend-
ence noted above indicates that these two excited-singlet
states are in thermal equilibrium under ambient condi-
tions. Effectively, this situation corresponds to E-type
delayed fluorescence.54 Now, in the simplest case, the
inverse of τS can be equated to the sum of the forward
and reverse EET rate constants while the equilibrium
constant represents the ratio of these two rate constants.
This simplified scheme allows derivation of the rate con-
stants for forward (TMBod* → MEBod) and reverse
(MEBod* → TMBod) EET steps as being 2.4 x 1010 and
1.0 x 109 s-1, respectively, at 295K. The thermally equili-
brated mixture of excited-singlet states decays with a
common lifetime of ca. 4 ns.
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 17
1
2
3
4
5
6
7
8
1.0
0.8
0.6
0.4
0.2
0.0
1.5
1.0
0.5
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
0.0
300
400
500
600
700
510
530
550
570
λ / nm
λ / nm
Figure 7. Absorption (black curve) and fluorescence (red
curve) spectra recorded for PMEX (9) in CH2Cl2 at room
temperature. The excitation wavelength was 365 nm. The
insert shows an expansion of the emission spectrum over
the range 500-600 nm.
Figure 6. Time-resolved fluorescence spectra recorded for
PMEt in MTHF at room temperature following excitation at
340 nm. Spectra were compiled at 25 (blue), 40 (red), 65
(lilac), 145 (light blue) and 225 ps (plum).
Because of the molecular topology, reverse EET in
PMEt could populate the S1 state associated with either of
the two TMBod units. In this way, the photonic energy
migrates around the array and samples all three Bodipy-
based chromophores. Through-space (i.e., direct) EET
between the two TMBod units is unlikely because of the
extended separation distance, estimated from computer-
generated molecular models55 to be ca. 29 Å, and limited
spectral overlap integral.56 On the assumption that this
particular EET step occurs exclusively via Förster dipole-
dipole coupling,57 the critical distance (RCD) for EET be-
tween TMBod molecules held in random orientation can
be computed from spectroscopic data to be 20.5 Å. As
such, we estimate the rate constant for through-space
EET between TMB units as being ca. 3.0 x 107 s-1. This
can be compared to the rate constant (kEET = 2.4 x 1010 s-
1) for EET from TMBod to MEBod, where the separation
distance is 18.0 Å and where the mechanism is most like-
ly dominated by through-bond interactions.27 At best,
energy migration between the two TMBod units, there-
fore, will account for about 1% of the initial excitation
energy but this will be supplemented by reverse EET
from MEBod: see Figure S7 for a reconstruction of the
steady-state fluorescence spectrum based on the derived
photophysics. It is important to note that the rate con-
stant (kREET) for reverse energy transfer from the excited-
singlet state of MEBod to TMBod has a value of 1.0 x 109
s-1 and is therefore competitive with inherent deactiva-
tion of the exciton (see Figure S8 for a distribution pro-
file based on solving the appropriate ordinary differential
equations).
Continuing along the projected EET axis, the photo-
physics of 9 (hereafter abbreviated as PMEX for simplici-
ty) were studied in CH2Cl2; this compound being poorly
soluble in MTHF at ambient temperature. The absorp-
tion spectrum (Figure 7) shows the presence of a rich
variety of transitions which are easily explained in terms
of the earlier description: Thus, the terminal acceptor,
EXBod, which is a class of expanded Bodipy dyes,58 has
an absorption maximum located at 645 nm and is clearly
visible on the low-energy side of the spectrum. This opti-
cal transition is broader than those found for conven-
tional Bodipy dyes but comparable to spectra reported59
for related dyes in solution. The familiar patterns for the
TMBod and MEBod chromophores are recognizable in
the window from 450 to 540 nm while the pyrene ab-
sorption transitions dominate the near-UV region. Exci-
tation into the PY chromophore at 365 nm leads to fluo-
rescence from all three Bodipy-based emitters. Most of
the observed fluorescence is associated with the EXBod
unit, which appears at 675 nm, but this is complemented
by weaker fluorescence from both MEBod and TMBod.
Selective excitation into the EXBod chromophore at 595
nm gives the anticipated emission spectral profile and
allows determination of ΦF and τS as being 0.92 ± 0.04
and 6.0 ± 0.1 ns, respectively. Following excitation of
PMEX at 310 or 505 nm, τS for the terminal acceptor
EXBod is 6.1 ± 0.1 ns.
Fluorescence spectra recorded for MEBod and
TMBod are well resolved from those recorded for EXBod
(Figure S5). Following excitation of PMEX in CH2Cl2 at
370 nm it was not possible to resolve emission from PY
and we assume that rapid EET occurs to one of the ap-
pended TMB units. Using time-correlated, single photon
counting methodology with excitation at 340 nm it was
possible to detect fluorescence from TMB at 520 nm. The
assertion of efficient EET from PY to TMB was fully sup-
ported by excitation spectra (Figure S9). Emission from
TMB was found to decay with an average lifetime of 33 ±
7 ps, although this is close to the temporal resolution of
our set-up following numerical re-convolution of the da-
ta and the quality of the fit was quite poor.60 There was
ACS Paragon Plus Environment

Page 9 of 17
Journal of the American Chemical Society
data points are shown as open circles. Non-linear least-
also a minor (i.e., 8%) contribution from a longer-lived
1
2
3
4
5
6
7
8
squares fits to the growth and decay are shown as red
(TMBod at 515 nm) and blue (MEBod at 590 nm) lines su-
perimposed on the experimental data. For MEBod, analysis
corresponds to a growth of 30 ps and a decay of 148 ps, the
weighted residuals (χ2 = 1.32) being given in the upper pan-
el. For TMBod, analysis corresponds to a dual-exponential
decay of 30 ps (92%) and 190 ps (8%), with the weighted
residuals (χ2 = 1.51) being shown on the central panel.
component having a lifetime of ca. 190 ps (Figure 8).
Fluorescence associated with MEBod grows-in after the
excitation pulse with a time constant of 30 ± 5 ps before
decaying via exponential kinetics. Indeed, the fluores-
cence signal at 555 nm decayed with an average lifetime
of 148 ± 9 ps (Figure 8).
Using time-correlated, single photon counting with
detection across the 680-720 nm window, fluorescence
associated with the EXBod acceptor grows-in after the
excitation pulse (Figure 9); in contrast, direct excitation
into EXBod at 635 nm gives a comparable decay time of
6 ns but without the slow growth (Figure S10). The grow-
in for EXBod emission following excitation at 340 nm
corresponds to an average time constant of 150 ± 15 ps,
followed by decay with a lifetime of 6.0 ± 0.2 ns, alt-
hough it has to be stressed that the initial part of the fit is
quite poor. We attribute this slow appearance of acceptor
fluorescence to intramolecular EET from MEBod, in
agreement with the excitation spectra (Figure S9).61 Be-
cause of the significantly red shifted absorption transi-
tion of the expanded Bodipy residue, spectral overlap
between emission from MEBod and absorption by
EXBod is kept modest56 (JDA = 0.061 cm) and this serves
to restrict kEET. The distance between the centers of the
two reactants is ca. 21.2 Å and, on the basis of previous
findings with related dyads,27 it is likely that the domi-
nant EET mechanism involves through-bond interac-
tions. The spectroscopic energy gap between excited-
singlet states resident on the reactants (ꢀESS = 3,650 cm-
1) is sufficient for ETT to be unidirectional. The poor fit
for the appearance of EXBod emission is attributed to
secondary EET events that follow from distribution of
the exciton around the array. Indeed, the best analytical
fit to the data after deconvolution corresponds to a decay
of 6 ns but with two growth signals of 140 ps(91%) and
700 ps (9%) respectively.
Interestingly, the fluorescence spectrum shows resid-
ual emission from TMBod, although the yield is rather
small (Figure 7). This finding is indicative of the thermal
equilibrium between excited states resident on TMBod
and MEBod being set up across the full array (consult
Figure S11 for a distribution profile based on computer
modeling). This situation is further testament to the rela-
tively slow rate of EET from MEBod to EXBod in this
system. Also contingent on this slow EET to the terminal
acceptor is the possibility for through-space EET be-
tween the appended MEBod units and, using the same
approach as illustrated for PMEt, the rate constant for
this step62 was estimated to be 1.5 x 108 s-1. Although inef-
ficient, this process helps to distribute the photon density
around the molecular array, as illustrated in Figure S12.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The average lifetime derived for the S1 state of MEBod
in the fully array PMEX is 155 ± 20 ps, compared to the
inherent lifetime of 4.6 ± 0.1 ns in CH2Cl2. In addition to
radiative (kRAD = 1.6 x 108 s-1) and nonradiative (kNR = 5.4
x 107 s-1) decay to the ground state, deactivation of this
excited state can be partitioned into contributions due to
energy migration to an ancillary MEBod unit (k = 1.5 x
108 s-1, 2%), reverse EET to one of the attached TMBod
units (k = 1.0 x 109 s-1, 15%) and EET to the terminal
EXBod unit (k = 5.4 x 109 s-1, 83%). Photons directed
away from the terminal acceptor will arrive at EXBod
after a short delay and it is this branching process that
accounts for the poor fit for the growth process. In fact,
the best fit gives two growth steps with lifetimes of 160 ps
(91%) and 700 ps (9%).
6
3
0
9
6
3
0
-3
-6
-3
-6
4
2
0
4
-2
-4
2
0
100000
10000
1000
-2
100000
10000
1000
100
100
0
100
200
300
400
Time / ps
0
100 200 300 400 500 600 700 800
Time / ps
Figure 8. Time-resolved fluorescence decay profiles rec-
orded for PMEX in CH2Cl2 at room temperature following
excitation at 340 nm. The IRF is shown as a grey curve while
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 17
Figure 9. Time-resolved fluorescence decay profiles rec-
Preliminary sensitization studies made with PMEX
1
2
3
4
5
6
7
8
orded for PMEX in CH2Cl2 at room temperature following
excitation at 370 nm. The IRF is shown as a grey curve while
data points are shown as open circles. Non-linear least-
squares fits to the growth and decay are shown as red (ME-
Bod at 590 nm) and blue (EXBod at 720 nm) lines superim-
posed on the experimental data. For MEBod, analysis corre-
sponds to a growth of 34 ps and a decay of 160 ps, the
weighted residuals (χ2 = 1.39) being given in the upper pan-
el. For EXBod, analysis corresponds to a dual-exponential
decay of 6 ns and a growth of 175 ps, with the weighted re-
siduals (χ2 = 1.75) being shown on the central panel.
There has been considerable interest64,65 of late in the
development of artificial light-harvesting arrays formed
by mixtures of organic dyes as potential sensitizers for
solar cells. We,27 and others,66 have shown that Bodipy
dyes can be used for such purposes while related Bodipy-
based dyes have shown exceptional promise in organic
devices for solar-to-electricity conversion.67 Cascade-type
arrays such as PMEX collect around 30% of the solar
spectrum (Figure 10) and emit in the region where
amorphous silicon and GaInP exhibit absorption thresh-
olds. The most attractive feature of PMEX relative to a
cocktail of unattached dyes is that it does not suffer from
problems of self-absorption, this being the limiting factor
for most organic dyes used as solar concentrators in plas-
tic films.68 In the solid state, it was also found that PMEX
does not undergo self-quenching, again this is a major
problem for simple organic dyes, probably because of its
complex 3D topology.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
With minor modification of these EET rate constants,
the steady-state emission spectrum can be recovered by
computer simulation63 assuming a static geometry and
allowing for mono-photonic excitation of the array (Fig-
ure S12). With this increased confidence, we can summa-
rize the experimental results in the form of Scheme 2.
Here, photons enter the array by way of PY, TMB, ME-
Bod or EXBod. We consider EET from PY to TMB to be
both unidirectional and quantitative while that from
TMB to MEBod is too fast for significant sideways trans-
fer to a second TMB unit. Because onwards EET from
MEBod to EXBod is relatively slow, reverse EET can oc-
cur to repopulate the S1 state of TMB. Since the system
has no memory, this process will activate either of the
two available TMB units. Backwards EET sets up the
equilibrium between S1 states on these three chromo-
phores such that the exciton samples all three within its
lifetime of 4 ns. There is now the possibility that the exci-
ton migrates across to the MEBod unit on the opposite
branch and from there to one of the appended TMB
units. This leads to widespread distribution of the exciton
and causes further delay in its (almost) inevitable arrival
at EXBod.
The following experiments were made in order to
demonstrate the ability of PMEX to sensitize current
generation from amorphous silicon: The compound (1
mg/25 mg, w/w) was dispersed in molten sucrose octa-
acetate (SOA) and used to construct a disk (1 cm diame-
ter and 4 mm thickness). The disk was adhered to a di-
chroic
mirror,
via
a
thin
layer
of
poly(methylmethacrylate), that reflects λ >650 nm and
the edges coated with reflective paint. The disk was
mounted at the focal plane of a small integrating sphere
and illuminated via an optical fiber with panchromatic
light (λ <650 nm). Fluorescence from the disk, isolated
with a 650 nm cut-off filter, was directed through a se-
cond optical fiber and beam expander to an amorphous
Si solar cell (10 x 10 mm). The excitation light was modu-
lated at 1 kHz and the photocurrent detected at the same
frequency with a lock-in amplifier. The light intensity
was adjusted to give a steady output of 150 lux, resulting
in an average current of 5.7 ꢁA. Comparison was made
with a conventional LED emitting at 660 ± 25 nm and
collimated with an identical beam expander. Restricting
light output to equal that from the SOA disk at 150 lux
illumination gave a similar average photocurrent of 5.3
ꢁA. This experiment serves to demonstrate the basic
principle of sensitization.
During 100 hours of continuous illumination with the
LED, the current decreased by 11% and was not recov-
ered by a dark period. The same experiment made with
an SOA disk impregnated with PMEX gave a current loss
of 24% over 100 hours. A similar experiment made with
EXBod only incorporated into the SOA disk stopped pro-
ducing current after 36 hours continuous illumination.
Increasing the light intensity by a factor of 4-fold,
achieved by removal of a neutral density filter, caused
complete bleaching of EXBod in less than 24 hours illu-
mination. The high photon flux was much less harmful to
the PMEX disk, although the photocurrent dropped by
almost 35% during 100 hours illumination. This can be
compared with a current loss of 18% with the LED. Inter-
estingly, the initial current increased a factor of almost 4-
fold for the LED at the higher photon flux but the in-
crease for PMEX was only from 5.7 to 14.1 ꢁA. This might
Scheme 2. Pictorial representation of EET flow in the
PMEX array, allowing for excitation into any of the main
chromophores.
ACS Paragon Plus Environment

Page 11 of 17
Journal of the American Chemical Society
be the onset of photon management due to the anticipat- with shorter pulse durations, attention was turned to-
1
2
3
4
5
6
7
8
ed photo-regulation. Although far from definitive, these
initial results hold promise for the design of photostable
arrays. It should also be noted that PMEX has a built-in
capability to dissipate UV photons via rapid internal con-
version and EET. This stabilizes the dye against UV dam-
age, which is harmful to the Si solar cell. Thus, direct
illumination of the Si solar cell with UV light at 290 <λ
<340 nm causes rapid loss of current but this does not
occur when the SOA/PMEX disk is in place. In passing,
we note that output from PMEX is of the optimum color
for stimulated plant growth using far red light.69 This
might be a further application70 for such arrays, especial-
ly if the photo-regulatory effect can be perfected.
wards transient absorption spectroscopy.
Transient absorption spectroscopy with PMEt
The transient differential absorption spectrum recorded
for 4 in MTHF following excitation at 420 nm with a 20-
ps laser pulse is shown in Figure S13. The negative signal
observed across the region from 500 to 570 nm corre-
sponds to a combination of ground-state bleaching and
stimulated fluorescence. These contributions are super-
imposed onto a broad, relatively weak absorption band
that must correspond to the S1-Sn transition. The latter
appears on either side of the bleaching signal. By sub-
tracting combined spectra for the emission and ground-
state bleach, using an iterative routine to optimize the
bandshape,72 the actual absorption spectrum for the Bod-
ipy-based excited-singlet state can be revealed (Figure
S13). This latter band is diffuse but centered at about 480
nm. It overlaps with fluorescence from the TMBod
chromophore, although the optical absorption transition
is quite weak relative to bleaching of the ground state so
the molar absorption coefficient must be small through-
out the overlap region. Under these conditions, the signal
decays with a lifetime in the ns range and is formed with-
in the excitation pulse. Experiments carried out with
TMBod show remarkably similar behavior.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
100
80
60
40
20
0
Using the above results to assist interpretation, tran-
sient absorption spectral studies were made with PMEt
in MTHF following laser excitation into the PY chromo-
phore at 355 nm with a 9-ps laser pulse (Figure 11). The
pulse width is comparable to the EET timescale, and
longer than vibrational relaxation,49 but nonetheless the
transient spectral records can be used to confirm EET
from TMBod to MEBod. The rate constant for the latter
process, as derived by global analysis72 across the wave-
length range from 500 to 540 nm, is 2.7 x 1010 s-1 and
remains in reasonable agreement with the time-resolved
emission data. The main spectral changes observed at
low laser intensity concern a gradual shift of the bleach-
ing maximum from ca. 500 to ca. 520 nm as EET pro-
ceeds. The final signal hardly decays over a few hundred
ps or so and corresponds to the thermally established
equilibrium of singlet-excited states associated with the
two Bodipy-based chromophores; clearly the final signal
is dominated by the exciton localized on MEBod. Repre-
sentative kinetic traces are shown in Figure 12 and con-
firm the EET pathway derived from the emission studies.
300
400
500
600
700
800
900 1000
λ / nm
Figure 10. Crude comparison of the average solar output at
AM1 (light gray curve) with the action spectrum recorded
for the amorphous Si solar cell (dark gray curve) and the
absorption spectrum of PMEX recorded in solution (red
curve).
The illumination levels used for these experiments
were kept quite low and it is questionable if, under these
conditions, more than one photon is present on the array
at any given time. Certainly, the time between absorption
of successive photons is likely to be much longer than the
EET timescales.71 As such, the improved photostability
found for PMEX relative to EXBod cannot be ascribed
simply to redistribution of photons around the array. To
further explore this situation, the following experiment
was performed: A thick (ca. 2 mm) film of PMEX in
poly(methylmethacrylate) was illuminated with a train of
5-ns laser pulses at 355 nm. The laser beam was passed
through a pinhole (diameter 0.25 cm) before the sample
and the intensity was adjusted to be within the range 0-
25 mJ per pulse at a repetition rate of 10 Hz. The fluores-
cence spectrum was recorded with a spectrograph over
the wavelength range 500 to 700 nm, which allows esti-
mation of the ratio of emission bands for EXBod and
MEBod. The ratio found at low laser intensity (3 mJ per
pulse), adjusted with metal screen filters, was roughly in
agreement with that seen from steady-state emission
spectra but it did not change significantly with increasing
laser power. Under these conditions, the time between
incoming photons is very long relative to the EET events
and, on simple statistical grounds, the likelihood of a
single molecule of PMEX absorbing two photons is less
than 20%. Since higher photon densities are attainable
0.05
0.00
-0.05
-0.10
-0.15
-0.20
400
450
500
550
600
650
700
λ / nm
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 12 of 17
Figure 11. Transient differential absorption spectra record-
1
2
3
4
ed after laser excitation of PMEt in MTHF at 355 nm. Spec-
tra were recorded across the time window between 0 and
200 ps, with the arrow indicating the course of the spectral
changes.
1.0
0.8
0.6
0.4
0.2
0.0
5
6
7
8
1.00
0.75
0.50
0.25
0.00
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
0
100
200
300
400
Time / ps
Figure 13. Representative kinetic traces recorded for
bleaching of TMBod at 500 nm and MEBod at 535 nm fol-
lowing excitation of PMEt at 355 nm with a high intensity
pulse. The Gaussian IRF inferred from autocorrelation of
the 1064 nm pulse is shown as a grey curve. Experimental
data are shown as open circles and the fit to the data is given
as a red line. Data points are normalized at the peak channel
and inverted for easier presentation.
0
50 100 150 200 250 300 350 400
Time / ps
Figure 12. Representative kinetic traces recorded for
bleaching of TMBod at 500 nm and MEBod at 535 nm fol-
lowing excitation of PMEt at 355 nm with a low intensity
pulse. The Gaussian IRF inferred from autocorrelation of
the 1064 nm pulse is shown as a grey curve. Experimental
data are shown as open circles and the fit to the data is given
as a red line. Data points are normalized at the peak channel
and inverted for easier presentation.
At low laser intensity, the transient absorption signal
from TMBod grows in within the Gaussian-shaped pulse
and decays with a time constant of 32 ps. There is a mi-
nor component amounting to ca. 8% that decays very
slowly and is attributed to the equilibrium mixture of
excited states. Bleaching of MEBod can be accommodat-
ed with first-order kinetics, having a lifetime of 35 ps,
while decay is on the ns timescale (Figure 12). The situa-
tion at the highest laser intensity is similar, except that
recovery of the TMBod signal occurs on a slower time-
scale, corresponding to a first-order lifetime of 46.5 ps.
Again, the bleaching occurs with a time constant of 10 ps
and there is a minor contribution from a very long-lived
component. At 535 nm, bleaching of the MEBod chro-
mophore occurs with a time constant of 49.6 ps before
recovery on the ns timescale (Figure 13). The quality of
the experimental data, together with the complexity of
the reaction scheme, precludes more protracted data
analysis. In fact, we emphasize that other kinetic models,
such as a stretched exponential, will also fit the data73,74
One possible explanation for this lengthening of the
mean lifetime involves relatively slow S1-S1 annihilation
between excitons localized on MEBod and TMBod
(Scheme 3). We presume that the resultant doubly-
excited state of whichever chromophore undergoes rapid
internal relaxation to form the S1 state, with subsequent
loss of the excess photonic energy as heat.
For the above experiment, the laser intensity was kept
to a low level, being equivalent to 2 x 1016 photons/cm2 s,
so as to minimize the possibility to excite multiple PY
units on a given molecule. Using a slight modification of
the protocol introduced by Melnikov et al.31b to allow for
the 3D nature of the sample, calculations for the 10 MHz
repetition rate indicate that the probability of adding
one, two or three photons to PMEt is 0.16, 0.012 and 0,
respectively. At the highest accessible laser intensity (i.e.,
2 x 1017 photons/cm2 s), the same probabilities are calcu-
lated to be 0.37, 0.53 and 0.05. These are the optimized
conditions for attaching two photons onto the array, at
least with our available equipment. As a convenient
means by which to monitor the course of EET, we restrict
attention to two fixed wavelengths corresponding to de-
activation of the S1 state of TMBod and appearance of the
corresponding S1 state associated with MEBod. The ini-
tial part of the bleaching signal is eliminated from the
analysis on the basis that it could be contaminated with
contributions from vibrational cooling and signals due to
direct excitation by MEBod.
Two obvious possibilities co-exist to explain this situ-
ation: The first case requires slow EET from TMBod* to
MEBod*, which could be explained in terms of the small-
er spectral overlap integral (JDA). Thus, for EET from the
first-excited singlet state (S1) of TMBod to the ground
state (S0) of MEBod JDA is calculated56 from the respec-
tive emission and absorption spectra to be 0.59 cm. For
the corresponding EET step from TMBod* to MEBod*,
JDA is calculated to be 0.19 cm. The rate of ETT is ex-
pected75 to vary in a linear manner with JDA unless other
factors, notably the coupling matrix element, change
ACS Paragon Plus Environment

Page 13 of 17
Journal of the American Chemical Society
significantly. A case could be made, therefore, for this molecular conformations will be especially favorable for
1
2
3
4
5
6
7
8
type of exciton blockade. In fact, the relevant energy gaps
also change for the two processes; the energies of the S1
states localized on donor and acceptor, taken as the over-
lap between normalized absorption and emission spec-
tra, are 19,630 and 18,840 cm-1, respectively, for TMBod
and MEBod while the S2 state associated with MEBod
lies at 24,390 cm-1. As such, there is little excess energy
(i.e., 790 cm-1) to be dissipated during EET to the ground
state but EET to the upper-lying S2 state requires loss of
some 17,870 cm-1 by way of vibrational cooling. This will
likely contribute towards a decreased rate of through-
bond EET to the S2 state.76
fast EET between TMBod molecules since their transi-
tion dipole moment vectors are perfectly aligned for
through-space interactions. It is also recognized80 that
Förster theory tends to underestimate the rate of EET at
short separations; the closest edge-to-edge approach
between the two reactants is ca. 18 Å. In the extreme
case, inter-branch EET within this family of conformers
could reach 5 x 108 s-1. The net result is that any factor
that slows the rate of onwards EET could promote side-
ways EET between TMBod sites. This, in turn, would
delay arrival of the second photon at the acceptor.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Transient absorption spectroscopy with PMEX
Singlet-singlet exciton annihilation is well known in
dendrimers,81 conducting polymers,82 and other molecu-
lar clusters.83 However, the behavior recorded for PMEt
is unusual in that the rate of annihilation appears to be
slow relative to EET to the ground-state acceptor; it has
to be recognized that, in many clusters, the rate of anni-
hilation reflects the molecular topology such that com-
parisons between different systems is not straightfor-
ward. Although the time constants for successive EET at
high photon flux cannot be vastly different, the molecule
could be deemed to possess the capacity to “hold” the
extra photon. It is interesting to note in this context that
there have been several attempts to develop protocols for
mapping excitonic flow in complex molecular architec-
tures.84 Extending the work to include EXBod is made
difficult by the realization that the excitation conditions
for attaching only two photons onto the array cannot be
identified. The best conditions, at least according to our
models,31b allow for a probability distribution of the type
0.32, 0.38, 0.20 and 0.02, respectively, for adding one,
two, three or four photons to the network.
1.0
Scheme 3. Illustration of the delayed EET step expected for
PMEt at high excitation densities.
0.8
1.0
The second case involves (reverse) EET from MEBod*
to TMBod*, such that the surviving exciton is associated
with the higher-energy S1 localized on TMBod. This sit-
uation is in accord with the energy-gap law,77 but still
demands dissipation of excess energy as heat. The con-
sequence of such annihilation is that the surviving exci-
ton will be transferred to the MEBod acceptor. There will
be a delay in the average arrival time corresponding to
the mean time taken for the annihilation step. In the ab-
sence of further information, it is assumed that S1-S1 an-
nihilation shows little preference for which chromophore
retains the exciton.
0.6
0.4
0.2
0.0
0.5
0.0
0
500
1000
0
200
400
600
800
1000
Time / ps
Figure 14. Representative kinetic traces recorded for
bleaching of EXBod at 647 nm and MEBod at 535 nm fol-
lowing excitation of PMEX at 355 nm with a low intensity
pulse. The Gaussian IRF inferred from autocorrelation of
the 1064 nm pulse is shown as a grey curve. Experimental
data are shown as open circles and the fit to the data is given
as a red line. Data points are normalized at the peak channel
and inverted for easier presentation. The insert shows the
corresponding data for TMBod.
In other systems where an exciton blockade has been
considered28 it has been necessary to draw attention to
alternative possibilities, such as bleaching of the accep-
tor,78 conformational exchange31 or triplet formation.79
In the case of PMEt, a potential escape path for the exci-
ton involves inter-branch energy migration between
TMBod units. This route was ignored earlier on the basis
that the average center-to-center separation distance of
29 Å is too long for competing EET given the relative
efficacy of onwards EET to MEBod. However, certain
Working within this constraint, laser flash photolysis
studies were made for PMEX in CH2Cl2 with excitation at
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 14 of 17
355 nm and using the conditions optimized to give the of either efficiency or self-protection of the array. At low
1
2
3
4
5
6
7
8
above excitation density (Figure S14). First, experiments
were made at an excitation intensity 10-fold lower than
the optimum value; this is intended to serve as a low in-
tensity control. These conditions give a projected proba-
bility distribution that favors mono-excitation (i.e., 0.16,
0.013 and 0 for addition of one, two and three photons to
the cluster). Bleaching of the EXBod chromophore fol-
lows first-order kinetics with an average lifetime of 153 ±
8 ps. Again, the initial part of the bleaching curve was
omitted from the analysis. Recovery of the ground-state
signal at 647 nm also follows first-order kinetics and cor-
responds to a lifetime of ca. 2.8 ± 0.3 ns. The data relat-
ing to bleaching and recovery of TMBod is very noisy but
gives a crude estimate of the lifetime for the recovery step
as being 30 ± 8 ps. Meanwhile, bleaching and recovery of
MEBod gives analyzed lifetimes of 26 ± 4 ps and 139 ± 9
ps, respectively (Figure 14). These values seem to be in
surprisingly good accord with data obtained by time-
resolved emission spectroscopy, although the latter are
believed to be far more precise.
light levels under conditions where there is fast discharge
of photons from EXBod to a solar cell the distributive
mechanism is redundant and the limiting step is EET
from MEBod to EXBod. On average, an absorbed photon
reaches the terminal acceptor within a few hundred pico-
seconds and overall EET to the terminal should exceed
90% despite the slow rate for the final step. Increasing
the fraction of sunlight harvested by the array and opti-
mizing the threshold wavelength for the terminal accep-
tor are well within the synthetic capability. Thus, within
this constraint, PMEX serves as a useful artificial leaf.
Future development of this type of system will involve
accreting numerous such units39 into a cooperative net-
work where EET can occur between PMEX units. In this
respect, the use of supramolecular interactions40 is con-
sidered mandatory.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
At high photon densities, PMEX possesses three sep-
arate junctions where photon annihilation might take
place. Although such reactions lose excitation energy as
low-grade heat, there is little propensity for triplet for-
mation25a and the molecule is quite stable toward photo-
degradation when dispersed in SOA, especially under UV
illumination. A crude estimate for the photon turnover
frequency is ca. 0.1 Hz such that each chromophore has
an average turnover number in excess of 10,000 during
sensitization. As such, it appears that the system pro-
vides a relatively effective means for photon manage-
ment. As the rate of discharge becomes slower, or in cas-
es where the light intensity fluctuates, there could be es-
calating significance for a re-distributive system. Clearly,
the kinetic balance becomes critical and the overall sys-
tem seems to be an ideal vehicle for sophisticated molec-
ular simulations.84 On-going research aims to increase
ΦF for the terminal acceptor and to add additional relays.
Interrogation of the system at the higher laser power
shows that, in this case, transient bleaching of EXBod
occurs on a slightly slower timescale. Thus, the time con-
stant for transient bleaching of the terminal acceptor is
lengthened to ca. 170 ± 8 ps while that for recovery re-
mains at 2.8 ± 0.3 ns. In addition, the recovery of ME-
Bod bleaching is somewhat slower than found at low
light intensity, the time constant increasing from 139 to
150 ps. Thus, kinetic indicators for some type of exciton
blockage, or energy pooling,71 are less evident for the full
array. This might be expected from the respective spec-
tral overlap integrals; JDA for conventional (i.e., MEBod*
to EXBod) EET being 0.061 cm while that for the second
step (i.e., MEBod* to EXBod*) is 0.225 cm. The resultant
doubly-excited state of EXBod (i.e., EXBod**) is ex-
pected to decay rapidly to form the corresponding S1
state. Backwards EET (from EXBod* to MEBod* to give
MEBod**) should be promoted by the energy-gap law77
ASSOCIATED CONTENT
Supporting Information: Supporting Information covers
full synthetic details, experimental methods, additional ex-
amples of experimental output, and examples of NMR spec-
tra. This material is available free of charge via the Internet
at http://pubs.acs.org.
and by the modest spectral overlap integral (JDA
=
0.0052 cm) to such an extent that it might be competi-
tive with forward annihilation. As such, this remains the
most likely route for the observed retardation of the
bleaching kinetics. There is no obvious indication for
charge transfer between the two dyes, as happens in cer-
tain singlet-singlet annihilation processes, but this can-
not be ruled out entirely.
AUTHOR INFORMATION
Corresponding Author
*e-mail: Anthony.harriman@ncl.ac.uk. Tel and fax: +44
191222 8660.
Concluding Remarks
Author Contributions
The molecular topology inherent to PMEX funnels pho-
tons absorbed by the various subunits to the terminal
acceptor with small losses at each stage. It is also inter-
esting to note the stepwise fall in the rate of EET on mov-
ing along the energy gradient. Reversible EET and
through-space couplings allow exciton distribution over
much of the array at low photon flux. This behavior is not
unique and sequential EET has been widely reported in
dendrimers37 and conducting polymers.38 In part, this
situation arises from the logical positioning of chromo-
phores in PMEX that directs photons along convergent
pathways. Although unusual, we have to question the
advantage offered by such photon redistribution in terms
All authors have given approval to the final version of the
manuscript.
ACKNOWLEDGMENT
We thank CNRS, EPSRC, Université Louis Pasteur de Stras-
bourg and Newcastle University for financial support of this
work.
REFERENCES
1) (a) Lewis, N. S. Nature 2001, 414, 589-590. (b) Blanken-
ship R. E. “Molecular Mechanisms of Photosynthesis” Black-
well Science, Oxford, 2002. (c) Zhu, X.-G.; Long, S. P.; Ort, D.
ACS Paragon Plus Environment

Page 15 of 17
Journal of the American Chemical Society
R. Ann. Rev. Plant Biol. 2010, 61, 235-261. (d) Ziessel, R.; Har-
riman, A. Chem. Commun., 2011, 47, 611-631.
(2) Quina, F. H.; Moreira, P. F.; Vautier-Giongo, Jr., C.;
Rettori, D.; Rodrigues, R. F.; Freitas, A. A.; Silva, P. F.;
Maçanita, A. L. Pure Appl. Chem. 2009, 81, 1687 –1694 and
references cited therein.
(3) (a) Holten, D.; Bocian, D. F.; Lindsey J. S. Acc. Chem.
Res. 2002, 35, 57-69; (b) Harriman, A.; Ziessel R. in “Carbon-
Rich Compounds” Eds. Haley, M. M.; Tykwinski, R. R. Wiley-
VCH, 2006, Weinheim, pp 26-89.
(4) (a) Nakamura, Y.; Aratani, N.; Osuka A. Chem. Soc. Rev.
2007, 36, 831-845. (b) Armaroli, N.; Balzani V. Angew. Chem.
Int. Ed., 2007, 46, 52-66.
(5) (a) Serin, J. M.; Brousmiche, D. W.; Fréchet J. M. J.
Chem. Commun. 2002, 2605-2607. (b) Weil, T.; Reuther, E.;
Müllen K. Angew. Chem. Int. Ed. 2002, 41, 1900-1904.
(6) (a) Earp, A. A.; Smith, G. B.; Franklin, J.; Swift, P. Sol.
Energy Mater. Sol. Cells, 2004, 84, 411-421. (b) Brühwiler, D.;
Calzaferri, G.; Torres, T.; Ramm, J. H.; Gartmann, N.; Dieu, L.-
Q.; Lopez-Duarte, I.; Martinez-Diaz, M. V. J. Mater. Chem.
2009, 19, 8040-8067. (c) Knuesel, R. J.; Jacobs, H. O. Adv.
Mater. 2011, 23, 2727-2733.
(7) (a) Dienel, T.; Bauer, C.; Dolamic, I.; Brühwiler, D. Solar
Energy, 2010, 84, 1366-1369. (b) Ziessel, R.; Goze, C.; Ulrich,
G.; Cesario, M.; Retailleau, P.; Harriman, A.; Rostron, J. P.
Chem. Eur. J. 2005, 11, 7366-7378.
(8) (a) Silva, P. F.; Paulo, L.; Barbafina, A.; Eisei, F.; Quina,
F. H.; Maçanita, A. L. Chem. Eur. J. 2012, 18, 3736–3744. (b)
Bora, D. K.; Rozhkova, E. A.; Schrantz, K.; Wyss, P. P.; Braun,
A.; Graule, T.; Constable E. C. Adv. Funct. Mater. 2012, 22,
490–502.
(9) Ahn, T. K.; Avenson, T. J.; Ballottari, M.; Cheng, Y.-C.;
Niyogi, K. K.; Bassi, R.; Fleming, G. R. Science 2008, 320, 794-
797.
(10) (a) Purcell, E. B.; Crosson, S. Curr. Opinion Microbiol.
2008, 11, 168-178. (b) Matsuoka, D.; Iwata, T.; Zikihara, K.;
Kandori, H.; Tokutomi, S. Photochem. Photobiol. 2007, 83,
122-130. (c) Gust, D.; Moore, T. A.; Moore, A. L. Faraday Disc.
2012, 155, 9-26.
(11) Boghossian, A. A.; Ham, M.-H.; Choic, J. H.; Strano, M.
S. Energy Environ. Sci., 2011, 4, 3834-3843.
(12) Wagner, S.; Bauer, S. MRS Bulletin 2012, 37, 207-217
and articles therein.
(13) (a) Zhang, J.; Hoeben, F. J. M.; Pouderoijen, M. J.;
Schenning, A. P. H. J.; Meijer, E. W.; De Schryver, F. C.; De
Feyter, S. Chem. Eur. J. 2006, 12, 9046-9055. (b) Werwie, M.;
Xu, X. X.; Haase, M.; Basche, T.; Paulsen, H. Langmuir 2012,
28, 5810-5818.
(14) (a) Loudet, A.; Burgess, K. Chem. Rev. 2007, 107, 4891-
4932. (b) Ulrich, G.; Ziessel, R.; Harriman, A. Angew. Chem.,
Int. Ed. 2008, 47, 1184-1201.
(15) (a) Ziessel, R.; Ulrich, G.; Harriman, A. New J. Chem.
2007, 31, 496-501. (b) Ziessel, R. Comptes Rendus Chim.
2007, 10, 622-629.
(16) Boens, N.; Leen, V.; Dehaen, W. Chem. Soc. Rev., 2012,
41, 1130-1172.
(20) Zheng, Q.; Xu, G.; Prasad P. N. Chem. Eur. J. 2008, 14,
5812-5819.
(21) Niu, S. L.; Massif, C.; Ulrich, G.; Ziessel, R.; Renard, P.-
Y.; Romieu, A. Org. Biomol. Chem. 2011, 9, 66-69.
1
2
3
4
5
6
7
8
(22) (a) The Molecular Probes Handbook.
A Guide to
Fluorescent Probes and Labelling Technologies; 11th Edition.
Johnson, I., Spence, M., Eds.; Life Technologies: Eugene, Oregon,
USA, 2010. (b) Bricks, J. L.; Kovalchuk, A.; Trieflinger, C.; Nofz,
M.; Büschel, M.; Tolmachev, A. I.; Daub, J.; Rurack, K. J. Am.
Chem. Soc. 2005, 127, 13522-13529.
(23) Didier, P.; Ulrich, G.; Mély, Y.; Ziessel R. Org. Biomol.
Chem. 2009, 7, 3639-3642.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(24) (a) Barin, G.; Yilmaz, M. D.; Akkaya, E. U. Tetrahedron
Lett. 2009, 50, 1738-1740. (b) Liu, J. Y.; Ermilov, E. A.; Roder,
B.; Ng, D. K. P. Chem. Commun. 2009, 1517-1519. (c) Camerel,
F.; Ulrich, G.; Retailleau, P.; Ziessel, R. Angew. Chem., Int. Ed.
2008, 46, 8876-8880. (d) Zhang, X. L.; Xiao, Y.; Qian, X. H.
Org. Lett. 2008, 10, 29-32. (e) Isaksson, M.; Hagglof, P.;
Hakansson, P.; Ny, T.; Johansson, L. B. Å. Phys. Chem. Chem.
Phys. 2007, 9, 3914-3922. (f) Coskun, A.; Akkaya, E. U. J. Am.
Chem. Soc. 2006, 128, 14474-14475. (g) Yilmaz, M. D.;
Bozdemir, O. A.; Akkaya, E. U. Org. Lett. 2006, 8, 2871-2873.
(h) Kalinin, S.; Johansson, L. B. Å. J. Phys. Chem. B 2004,
108, 3092-9097. (i) Keller, R. C. A.; Silvius, J. R.; de Kruijff, B.
Biochem. Biophys. Res. Commun. 1995, 207, 508-514.
(25) (a) Harriman, A.; Mallon, L. J.; Ziessel, R. Chem. Eur. J.
2008, 14, 11461-11473. (b) (a) Harriman, A.; Mallon, L. J.;
Goeb, S.; Ziessel, R. Phys. Chem. Chem. Phys. 2007, 9, 5199-
5201. (c) Harriman, A.; Mallon, L. J.; Goeb, S.; Ulrich G.; Zies-
sel, R. Chem. Eur. J., 2009, 15, 4553-4560.
(26) Goze, C.; Ulrich, G.; Ziessel, R. J. Org. Chem. 2007, 72,
313-322.
(27) Harriman, A.; Mallon, L. J.; Elliot, K. J.; Haefele, A.;
Ulrich, G.; Ziessel, R. J. Am. Chem. Soc. 2009, 131, 13375-
13386.
(28) Berglund, A. J.; Doherty, A. C.; Mabuchi, H. Phys. Rev.
Rev. 2002, 89, 068101.
(29) Birnbaum, K. M.; Boca, A.; Miller, R.; Boozer, A. D.;
Northup, T. E.; Kimble, H. J. Nature 2005, 436, 87-90.
(30) (a) Nirmal, M.; Babbousi, B. O.; Bawendi, M. G.;
Macklin, J. J.; Trautman, J. K.; Harris, T. D.; Brus, L. E. Nature
1996, 387, 802-804. (b) Jordens, S.; De Belder, G.; Lor, M.;
Schweitzer, G.; Van der Auweraer, M.; Weil, T.; Reuther, E.;
Müllen, K.; De Schryver, F. C. Photochem. Photobiol. Sci.
2003, 2, 177-186. (c) Neuwahl, F. V. R.; Righini, R.; Adronov,
A.; Malenfant, P. R. L.; Fréchet, J. M. J. J. Phys. Chem. B 2001,
105, 1307-1312.
(31) (a) Cotlet, M.; Gronheid, R.; Habuchi, S.; Stefan, A.;
Barbafina, A.; Müllen, K.; Hofkens, J.; De Schryver, F. C. J. Am.
Chem. Soc. 2003, 125, 13609-13617. (b) Melnikov, S. M.; Yeow,
E. K. L.; Uji-i, H.; Cotlet, M.; Müllen, K.; De Schryver, F. C.;
Enderlein, J.; Hofkens, J. J. Phys. Chem. B 2007, 111, 708-719.
(32) (a) Wozniak, A. K.; Schroeder, G. F.; Grubmueller, H.;
Seidel, C. A. M.; Oesterheit, F. Proc. Natl. Acad. Sci. USA
2008, 105, 18337-18342. (b) Alamiry, M. A. H.; Hagan, J. P.;
Harriman, A.; Bura, T.; Ziessel, R. Chem. Sci. 2012, 3, 1041-
1048.
(17) (a) Bozdemir, O. A.; Erbas-Cakmak, S.; Ekiz, O. O.;
Dana, A.; Akkaya, E. U. Angew. Chem., Int. Ed. 2011, 50,
10907-10912. (b) Haefele, A.; Ulrich, G.; Retailleau, P.; Ziessel,
R. Tetrahedron Lett. 2008, 49, 3716-3721. (c) Ziessel, R.;
Goze, C.; Ulrich, G. Synthesis 2007, 936-949. (d) Nastasi, F.;
Puntoriero, F.; Campagna, S.; Olivier, J.-H.; Ziessel, R.. Phys.
Chem. Chem. Phys. 2010, 12, 7392-7402. (e) Goeb, S.; Ziessel,
R. Tetrahedron Lett. 2008, 49, 2569-2574. (f) Goze, C.; Ulrich,
G.; Mallon, L. J.; Allen, B. D.; Harriman, A.; Ziessel, R. J. Am.
Chem. Soc. 2006, 128, 10231-10239.
(18) Hara, T.; Hirasawa, A.; Sun, Q.; Koshimizu, T.-A.;
Itsubo, C.; Sadakane, K.; Awaji T.; Tsujimoto G. Mol. Pharm.
2009, 75, 85-91.
(19) Ulrich, G.; Goze, C.; Guardigli, M.; Roda, A.; Ziessel, R.
Angew. Chem., Int. Ed. 2005, 44, 3694-3698.
(33) Hablot, D.; Ziessel, R.; Alamiry, M. A. H.; Bahraidah, E.;
Harriman, A. Chem. Sci. 2013, 4, 444-453.
(34) (a) Weil, T.; Reuther, E.; Beer, C.; Müllen, K. Chem. Eur.
J. 2004, 10, 1398-1414. (b) Schweitzer, G.; Gronheid, R.;
Jordens, S.; Lor, M.; De Belder, G.; Weil, T.; Reuther, E.;
Müllen, K.; De Schryver, F. C. J. Phys. Chem. A 2003, 107,
3199-3207.
(35) (a) Thomas, J. K. R.; Thompson, A. L.; Sivakumar, A. V.;
Bardeen, C. J. J. Am. Chem. Soc. 2005, 127, 373-383. (b)
Marchi, E.; Baroncini, M.; Bergamini, G.; Van Heyst, J.; Vögtle,
F.; Ceroni, P. J. Am. Chem. Soc. 2012, 134, 15277-15280. (c)
Cotlet, M.; Vosch, T.; Habuchi, S.; Weil, T.; Müllen, K.;
Hofkens, J.; De Schryver, F. C. J. Am. Chem. Soc. 2005, 127,
9760-9768. (d) Augulis, R.; Pugzlys, A.; Hurenkamp, J. H.;
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 16 of 17
Feringa, B. L.; van Esch, J. H.; Loosdrecht, P. H. M. J. Phys.
magnitude of the short component might not be as stated and it
Chem. A 2007, 111, 12944-12953.
is possible that some additional fast processes are obscured by
the IRF.
(54) Rothe, C.; Monkman, A. J. Chem. Phys. 2005, 123,
244904.
(55) (a) Isaksson, M.; Norlin, N.; Westlund, P. O.; Johans-
son, L. B.-Å. Phys. Chem. Chem. Phys. 2007, 9, 1941-1951. (b)
Abramavicius, D.; Valkunas, L.; van Grondelle, R. Phys. Chem.
Chem. Phys. 2004, 6, 3097-3105.
(56) Saini, S.; Singh, H.; Bagchi, B. J. Chem. Sci. 2006, 118,
23-35.
(57) Scholes, G. D. Annu. Rev. Phys. Chem. 2003, 54, 57-87.
(58) (a) Ulrich, G.; Goeb, S.; De Nicola, A.; Retailleau, P.;
Ziessel, R. Synlett 2007, 1517–1520. (b) Goeb, S; Ziessel, R.
Org. Lett. 2007, 9, 737–740.
(59) Ulrich, G.; Goeb, S.; De Nicola, N.; Retailleau, P.; Zies-
sel, R. J. Org. Chem. 2011, 76, 4489–4505.
(60) Bajzer, Z.; Sharp, J. C.; Sedarous, S. S.; Prendergast, F.
G. Eur. J. Biophys. 1990, 18, 101-115.
(61) Moore, A. L.; Dirks, G.; Gust, D.; Moore, T. A. Photo-
chem. Photobiol. 1980, 32, 691-695.
1
2
3
4
5
6
7
8
(36) (a) Meskers, S. C. J.; Bender, M.; Hubner, J.;
Romanovskii, Y. V.; Oestreich, M.; Schenning, A. P. H. J.;
Meijer, E. W.; Bassler, H. J. Phys. Chem. A 2001, 105, 10220-
10229. (c) Megow, J.; Roder, B.; Kulesza, A.; Bonacic-Koutecky,
V.; May, V. ChemPhysChem 2011, 12, 645-656.
(37) (a) Vosch, T.; Cotlet, M.; Hofkens, J.; Van der Biest, K.;
Lor, M.; Weston, K.; Tinnefeld, P.; Sauer, M.; Latterini, L.;
Müllen, K.; De Schryver, F. C. J. Phys. Chem. A 2003, 107,
6920-6931. (b) Maus, M.; Mitra, S.; Lor, M.; Hofkens, J.; Weil,
T.; Herrmann, A.; Müllen, K.; De Schryver, F. C. J. Phys. Chem.
A 2001, 105, 3961-3966.
(38) (a) Cornil, J.; Beljonne, D.; Calbert, J. P.; Brédas, J. L.
Adv. Mater. 2001, 13, 1053-1067. (b) Schwartz, B. J. Ann. Rev.
Phys. Chem. 2003, 54, 141-172. (c) Scholes, G. D.; Rumbles, G.
Nature Mater. 2006, 5, 683-696.
(39) Iehl, J.; Nierengarten, J. F.; Harriman, A.; Bura, T.;
Ziessel, R. J. Am. Chem. Soc. 2012, 134, 988-998.
(40) Olivier, J. H.; Barbera, J.; Bahaidarah. E.; Harriman, A.;
Ziessel, R. J. Am. Chem. Soc. 2012, 134, 6100-6103.
(41) (a) Katan, C.; Terenziani, F.; Mongin, O.; Werts, M. H.
V.; Porres, L.; Pons, T.; Metz, J.; Tretiak, S.; Blanchard-Desce,
M. J. Phys. Chem. A 2005, 109, 3024-3037. (b) Oliva, M. M.;
Casado, J.; Hennrich, G.; Navarette, J. T. L. J. Phys. Chem. B
2006, 110, 19198-19206.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(62) According to molecular dynamics simulations made
with PMEX, the average center-to-center separation between
MEBod units is ca. 28 Å while the mean orientation factor is
1.07. It appears that this branch is less flexible than the lower
branches connecting MEBod to TMBod and this is responsible
for the higher inter-strand kEET.
(63) Flomenbom, O.; Amir, R. J.; Shabat, D.; Klafter, J. J.
Lumin. 2005, 111, 315-325.
(42) Wrigge, G.; Gerhardt, I.; Hwang, J.; Zymofen, G.;
Sandoghdar, V. Nature Phys. 2008, 4, 60-66.
(43) (a) Mishra, A.; Behera, R. K.; Behera, P. K.; Mishra B.
K.; Behera, G. B. Chem. Rev. 2000, 100, 1973-2011. (b) Fabian,
M.; Nakazumi H.; Matsuka, M. Chem. Rev. 1992, 92, 1197-
1226. (c) Yuan, M.; Yin, X.; Zheng, H.; Ouyang, C.; Zuo, Z.; Liu
H.; Li, Y. Chem. Asian J. 2009, 4, 707-713.
(44) Harriman, A.; Heitz, V.; Ebersole, M.; Van Willigen, H.
J. Phys. Chem. 1994, 98, 4982-4989.
(45) Haefele, A.; Ulrich, G.; Retailleau, P.; Ziessel, R. Tetra-
hedron Lett. 2008, 49, 2774-2782.
(46) Liu, T.; Morisaki, Y.; Cunningham, N.; Tykwinski, R. R.
J. Org. Chem. 2007, 72, 9622-9629.
(47) Goze, C.; Ulrich, G.; Ziessel, R. Org. Lett. 2006, 8,
4445-4448.
(48) Harriman, A.; Izzet, G.; Ziessel, R. J. Am. Chem. Soc.
2006, 128, 10868-10875.
(49) (a) Hissler, M.; Harriman, A.; Khatyr, A.; Ziessel, R.
Chem. Eur. J. 1999, 11, 3366. (b) Although the various Bodipy-
based chromophores do absorb at 365 nm, the stoichiometry
and molar absorption coefficients indicate that direct excitation
of these units never exceeds about 7%. This was confirmed us-
ing stoichiometric mixtures of the isolated reagents.
(50) Ziessel, R.; Rihn, S.; Harriman, A. Chem. Eur. J. 2010,
16, 11942–11953.
(51) (a) Burghart, A.; Thoresen, L. H.; Che, J.; Burgess, K.;
Bergström, F.; Johansson, L. B-Å. Chem. Commun. 2000,
2203-2202; (b) Wan, C.-W.; Burghart, A.; Chen, J.; Bergström,
F.; Johansson, L. B.-A.; Wolford, M. F.; Kim, T. G.; Topp, M. R.;
Hochstrasser, R. M.; Burgess, K. Chem. Eur. J. 2003, 9, 4430-
4441. (c) The connections to the various Bodipy fragments do
not promote observable electronic interactions between bridge
and dye. The absorption and emission spectra of the latter are
not perturbed by these linkages and so comparison can be
made with the isolated reagents.
(64) (a) Chemisana, D. Renewable Sustainable Ener. Rev.
2011, 15, 603-611. (b) Currie, M. J.; Mapel, J. K.; Heidel, T. D.;
Goffri, S.; Baldo, M. A. Science 2008, 321, 226-228. (c)
Mulden, C. L.; Theogarajan, L.; Currie, M.; Mapel, J. K.; Baldo,
M. A.; Vaughn, M.; Willard, P.; Bruce, B. D.; Moss, M. W.;
McLain, C. E.; Morseman, J. P. Adv. Mater. 2009, 21, 3181-
3185.
(65) (a) Calzaferri, G. Topics Catal. 2010, 53, 130-140. (b)
Campbell, W. M.; Burrell, A. K.; Officer, D. L.; Jolley, K. W.
Coord. Chem. Rev. 2004, 248, 1363-1379. (c) Poirel, A.; De
Nicola, A.; Ziessel, R. Org. Lett. 2012, 14, 5696-5699.
(66) Bozdemir, C. A.; Yilmaz, M. D.; Buyukcakir, O.; Siemi-
arczuk, A.; Tutas, M.; Akkaya, E. U. New J. Chem. 2010, 34,
151-155.
(67) Bura, T.; Leclerc, N.; Fall, S. ; Leveque, P. ; Heiser, T. ;
Retailleau, P. ; Rihn, S. ; Mirloup, A. ; Ziessel, R. J. Am. Chem.
Soc. 2012, 134, 17404-17407.
(68) (a) Reisfeld, R. Opt. Mater. 2010, 32, 880-886. (b) Tal-
eb, A. M. Renewable Ener. 2002, 26, 137-142.
(69) Cerdan, P. D.; Chory, J. Nature 2003, 423, 881-885.
(70) Hoffmann, S. Gartenbauwiss. 1999, 64, 100-105.
(71) Bradshaw, D. S.; Andrews, D. L. Polymers 2011, 3,
2053-2077.
(72) Müllen, K. M.; van Stokkum, I. H. M. J. Stastist. Soft-
ware 2007, 18, 1-46.
(73) Phillips, J. C. Rep. Prog. Phys. 1996, 59, 1133-1207.
(74) Alvarez, F.; Alegria, A.; Colmenero, J. Phys. Rev. B
1991, 44, 7306-7312.
(75) Curutchet, C.; Scholes, G. D.; Mennucci, B.; Cammi, R.
J. Phys. Chem. B 2007, 111, 13253-13265.
(76) Ghiggino, K. P.; Yeow, E. K. L.; Haines, D. J.; Scholes, G.
D.; Smith, T. A. J. Photochem. Photobiol. A: Chem. 1996, 102,
81-86.
(51) Ziessel, R.; Rihn, S.; Harriman, A. Chem. Eur. J. 2010,
16, 11942–11953.
(77) Englman, R.; Jortner, J. Mol. Phys. 1970, 18, 145-172.
(78) Liu, D. J.; De Feyter, S.; Cotlet, M.; Stefan, A. ; Wiesler,
U. M. ; Herrmann, A. ; Grebel-Kochler, D.; Qu, J. Q.; Müllen,
K.; De Schryver, F. C. Macromol. 2003, 36, 5918.
(79) Kodriano, Y.; Poem, E.; Lindner, N. H.; Tradonsky, C.;
Gerardot, B. D.; Petroff, P. M.; Avron, J. E.; Gershoni, D. Phys.
Rev. B 2010, 82, 155329.
(52) Fluorescence quantum yields measured for the individ-
ual reference compounds are as follows: 1-ethynylpyrene ΦF =
0.80; 1 ΦF = 0.78; and 4 ΦF = 0.75. Experiments made with a
4:2:1 molar mixture of these three compounds in MTHF show
emission only from PY under illumination at 365 nm.
(53) Because the derived lifetime is close to the limit of the
temporal resolution of this set-up, we should treat the absolute
value with caution. The underlying trend is correct but the
(80) Jordanides, X. J.; Scholes, G. D.; Fleming, G. R. J. Phys.
Chem. B 2001, 105, 1652-1669.
ACS Paragon Plus Environment

Page 17 of 17
Journal of the American Chemical Society
(81) (a) Fron, E.; Schweitzer, G.; Jacob, J.; Van Vooren, A.;
Rana, D.; Gangopadhyay, G. J. Chem. Phys. 2003, 118, 434-
443. (e) Vlaming, S. M.; Heijs, D. J.; Knoester, J. J. Lumin.
2005, 111, 349-358.
Beljonne, D.; Müllen, K.; Hofkens, J.; Van der Auweraer, M.; De
Schryver, F. C. ChemPhysChem 2007, 8, 1386-1393. (b)
Larsen, J.; Bruggemann, B.; Polivka, T.; Sundstrom, V.; Akes-
son, E.; Sly, J.; Crossley, M. J. J. Phys. Chem. A 2005, 109,
10654-10662. (c) De Schryver, F. C.; Vosch, T.; Cotlet, M.; Van
der Auweraer, M.; Mullen, K. ; Hofkens, J. Acc. Chem. Res.
2005, 38, 514-522.
1
2
3
4
Table of Contents artwork
5
6
7
8
9
(82) (a) Nguyen, T. Q.; Martini, I. B.; Liu, J.; Schwatz, B. J. J.
Phys. Chem. B 2000, 104, 237-255. (b) Ruseckas, A.; Ribierre,
J. C.; Shaw, P. E.; Staton, S. V.; Burn, P. L.; Samuel, I. D. W.
Appl. Phys. Lett. 2009, 95, 183305.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(83) (a) Khairutdinov, R. F.; Serpone, N. J. Phys. Chem. B
1999, 103, 761-769. (b) Kelley, R. F.; Goldsmith, R. H.;
Wasielewski, M. R. J. Am. Chem. Soc. 2007, 129, 6384-6385.
(84) (a) Andrews, D. L.; Li, S.; Rodriguez, J.; Slota, J. J.
Chem. Phys. 2007, 127, 134902. (b) Andrews, D. L.; Li, S.
Chem. Phys. Lett. 2006, 433, 239-243. (c) Blumen, A.; Volta,
A.; Jurjiu, A.; Koslowskib, T. Physica A 2005, 356, 12-18. (d)
ACS Paragon Plus Environment