Solvent-Driven Conformational Exchange for Amide-Linked Bichromophoric BODIPY Derivatives

Article abstract of DOI:10.1002/chem.201602354

The fluorescence lifetime and quantum yield are seen to depend in an unexpected manner on the nature of the solvent for a pair of tripartite molecules composed of two identical boron dipyrromethene (BODIPY) residues attached to a 1,10-phenanthroline core. A key feature of these molecular architectures concerns the presence of an amide linkage that connects the BODIPY dye to the heterocyclic platform. The secondary amide derivative is more sensitive to environmental change than is the corresponding tertiary amide. In general, increasing solvent polarity, as measured by the static dielectric constant, above a critical threshold tends to reduce fluorescence but certain hydrogen bond accepting solvents exhibit anomolous behaviour. Fluorescence quenching is believed to arise from light-induced charge transfer between the two BODIPY dyes, but thermodynamic arguments alone do not explain the experimental findings. Molecular modelling is used to argue that the conformation changes in strongly polar media in such a way as to facilitate improved rates of light-induced charge transfer. These solvent-induced changes, however, differ remarkably for the two types of amide.

Full text of DOI:10.1002/chem.201602354

DOI: 10.1002/chem.201602354
Full Paper
&
Dyes/Pigments
Solvent-Driven Conformational Exchange for Amide-Linked
Bichromophoric BODIPY Derivatives
Shrikant Thakare,[b] Patrycja Stachelek,[a] Soumyaditya Mula,[c] Ankush B. More,[b]
Subrata Chattopadhyay,[c] Alok K. Ray,[d] Nagaiyan Sekar,[b] Raymond Ziessel,[e] and
Anthony Harriman*[a]
Abstract: The fluorescence lifetime and quantum yield are
seen to depend in an unexpected manner on the nature of
the solvent for a pair of tripartite molecules composed of
two identical boron dipyrromethene (BODIPY) residues at-
tached to a 1,10-phenanthroline core. A key feature of these
molecular architectures concerns the presence of an amide
linkage that connects the BODIPY dye to the heterocyclic
platform. The secondary amide derivative is more sensitive
to environmental change than is the corresponding tertiary
amide. In general, increasing solvent polarity, as measured
by the static dielectric constant, above a critical threshold
tends to reduce fluorescence but certain hydrogen bond ac-
cepting solvents exhibit anomolous behaviour. Fluorescence
quenching is believed to arise from light-induced charge
transfer between the two BODIPY dyes, but thermodynamic
arguments alone do not explain the experimental findings.
Molecular modelling is used to argue that the conformation
changes in strongly polar media in such a way as to facilitate
improved rates of light-induced charge transfer. These sol-
vent-induced changes, however, differ remarkably for the
two types of amide.
Introduction
zwitterionic resonance structures (Scheme 1). In each case,
both trans and cis forms are possible. The partial double-bond
character imposed on the central CÀN bond helps position the
atoms around the amide bond such that they reside in the
same plane and provides a high (i.e., 10–25 kcalmolÀ1) barrier
for internal rotation.[4d] In most proteins, the amide-trans tauto-
mer (Scheme 1) is the more stable species, but the correspond-
ing amide-cis form has been clearly recognised by X-ray crys-
tallography. Indeed, the amide-cis tautomer is often found in
simple amides, such as formamide and N-methylacetamide.[5]
Much less information is available regarding the possible oc-
currence of amide–iminol tautomerism for the amide bond
(Scheme 1), although this situation is interesting from the the-
oretical viewpoint.[6] Such transformations, which require
a large structural rearrangement, might be more easily accom-
plished in solution, especially in the presence of hydrogen-
bonding solvents. It is also significant that the amide oxygen
atom is a good hydrogen-bond acceptor, while the nitrogen
atom is an effective hydrogen-bond donor. This aspect of
amide chemistry is of crucial importance in terms of establish-
ing the secondary structures of proteins, enzymes, bacteria, vi-
ruses and so forth and has been studied in great detail.[7]
The amide bond plays a special role in biochemistry—it pro-
vides the key structural features needed to assemble helical
peptides and folded proteins.[1,2] Indeed, the amide carbonyl
function is less electrophilic than in ketones, aldehydes and
carboxylic acids, while sp2 hybridisation renders the nitrogen
atom non-basic.[3] This last point is well illustrated by the fact
that protonation of amides in acidic media occurs at the
oxygen atom. In general, the amide group persists in two tau-
tomeric forms; namely, the amide and iminol structures.[4]
However, the amide form is generally considered as a set of
[a] P. Stachelek, Prof. Dr. A. Harriman
Molecular Photonics Laboratory, School of Chemistry
Newcastle University, Bedson Building, Newcastle upon Tyne, NE1 7RU (UK)
E-mail: anthony.harriman@ncl.ac.uk
[b] S. Thakare, A. B. More, Prof. N. Sekar
Department of Dyestuff Technology, Institute of Chemical Technology
Mumbai 400019 (India)
[c] Dr. S. Mula, Prof. S. Chattopadhyay
Bio-Organic Division, Bhabha Atomic Research Centre
Mumbai 400085 (India)
[d] Prof. A. K. Ray
The amide bond, although relatively small, might be expect-
ed to help dictate the global structures of large organic mole-
cules and even supramolecular entities.[8] This realisation takes
on additional importance when allowance is made for the fact
that the amide linkage can be sensitive to changes in the local
environment. We now report bichromophoric molecules,
BAB(H) and BAB(ET), wherein two identical boron dipyrrome-
thene (BODIPY) dyes are appended at the 2,9-positions of
Laser and Plasma Technology Division, Bhabha Atomic Research Centre
Mumbai 400085 (India)
[e] Dr. R. Ziessel
Laboratoire de Chimie Organique et Spectroscopies Avancꢀes (LCOSA)
Ecole Europꢀenne de Chimie, Polymꢁres et Matꢀriaux
Universitꢀ de Strasbourg, 25 rue Becquerel
67087 Strasbourg Cedex 02 (France)
Supporting information for this article can be found under http://
dx.doi.org/10.1002/chem.201602354.
Chem. Eur. J. 2016, 22, 1 – 12
1
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
Our strategy is based on the realisation that the BODIPY
dyes are likely to be strongly fluorescent when embedded in
a solvent reservoir.[11] Rotations around the connecting units
should ensure that the two BODIPY dyes remain mutually inde-
pendent and it might be noted that NMR spectra are fully con-
sistent with this notion. In the event that the amide connector
responds to changes in the nature of the surrounding solvent,
the molecular conformation might be expected to change in
such a way that the two BODIPY fluorophores are brought into
closer proximity. In turn, this effect should be visualised by
a change in the overall fluorescence output. The idea behind
using 1,10-phenanthroline is to make use of its well-known
ability[12] to complex cations from solution, including protons.
It is also apparent that repulsion between lone-pairs on the
carbonyl and aza-nitrogen atoms should help establish the
molecular conformation. If successful, this approach could be
adapted to generate related structural changes brought about
by external or internal triggering so that the kinetics for the
folding and/or unfolding steps might be monitored. Such in-
formation could be helpful in terms of better understanding
the critical issues of protein denaturation[13] and unfolding.[14]
Scheme 1. Major tautomeric and resonance forms of the amide bond; the
positive charge assigned to the nitrogen atom should not be taken literally,
since the overall electronic charge at this site is likely to remain negative.
a 1,10-phenanthroline residue by way of amide connectors
(Figure 1). Such BODIPY dyes are highly fluorescent markers[9]
that, at least in the case of monomeric dyes, are insensitive to
the nature of the surrounding medium, temperature or pres-
sure.[10] The specific intention of this work is to explore wheth- Results and Discussion
er the amide bond, which constitutes less than 3% of the total
Synthesis and characterisation
solvent-accessible surface area of the supermolecule, can exert
control over the molecular geometry. It might be noted that
there is a complete library of BODIPY derivatives bearing aryl
hydrocarbons attached directly to the dipyrrin backbone or
through the pseudo-meso-position,[9] but none of these struc-
tures use an amide spacer as the means by which to intercon-
nect the terminals.
Synthesis of the target bichromophoric molecules BAB(ET) and
BAB(H), which differ only in terms of the substitution pattern
at the amide sites, was aided by earlier work[15] which explored
the catalytic hydrogenation of the corresponding mononuclear
BODIPY derivative, 1, bearing a meso-4-nitrophenyl group.
Thus, the Pd-catalysed hydrogenation of 1 in a mixture of di-
chloromethane and ethanol at 258C gave the expected amino-
BODIPY 3 as the only product. However, when the reaction
was carried out in a mixture of acetonitrile and ethanol at
708C, dye 2 was the major product along with a small amount
of 3 (2:3=95:5) (Scheme 2). Apparently, the initially formed 3
is readily transformed to 2 by an in situ reaction with
CH3CN.[16] Compound 3 was acetylated with acetyl chloride in
CH2Cl2 to obtain the control compound, BOD, in 81% yield
(Figure 1 and Scheme 2). The target compounds BAB(ET) and
BAB(H) were synthesised by reacting 1,10-phenanthroline-2,9-
diacyl chloride[17] with 2 and 3, respectively. After workup and
purification by flash chromatography, the desired compounds
were isolated in 42% and 52% yield, respectively (Scheme 2).
The target compounds were characterised fully by NMR
spectroscopy, giving well-defined spectra. For example, in the
case of dye BAB(H), aromatic protons of the 1,10-phenanthro-
line unit appear as two different sets of signals. Protons of the
central aromatic ring resonate as a singlet at 8.04 ppm, where-
as the other aromatic protons exhibit two doublets at 8.57 and
8.74 ppm. The methyl groups of the BODIPY moieties resonate
as singlets at 1.35 and 2.50 ppm, while the ethyl groups reso-
nate as quartets (2.22 ppm) and triplets (0.91 ppm). The aro-
matic phenyl protons of the two BODIPY units are found as
doublets at 7.33 and 8.11 ppm with the expected 4-proton in-
tegration for each signal. A diagnostic signal is provided by
Figure 1. Molecular formulae of the new BODIPY-based bichromophores,
BAB(H) and BAB(ET), and of the control compound, BOD.
&
&
Chem. Eur. J. 2016, 22, 1 – 12
www.chemeurj.org
2
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
Table 1. Electrochemical data recorded for the target BODIPY-based dyes
in CH2Cl2.[a]
Compound
E0 [V] (DE [mV])
E0 [V] (DE [mV])
ox
red
BOD
BAB(ET)
BAB(H)
+0.57 (32)
+0.61 (66)
+0.60 (82)
À1.77 (74)
À1.69 (70)
À1.69 (62); À1.98 (78)
[a] Half-wave potentials determined by cyclic voltammetry in deoxygenat-
ed CH2Cl2, containing 0.1m TBAPF6, at a solute concentration of 1.5 mm
at 258C. Potentials were standardised using ferrocene (Fc) as an internal
reference and the half-wave potentials are reported relative to the Fc/Fc+
couple. Error in half-wave potentials is Æ15 mV.
Scheme 2. Synthetic scheme for preparation of dyes 2, 3, BAB(ET), BAB(H)
and BOD. Reagents and conditions: a) H2, 10% Pd/C (10%), H2O, EtOH/
CH3CN (1:1), 708C (for 2) and H2, 10% Pd/C (10%), H2O, CH2Cl2/EtOH (1:1),
258C (for 3); b) SOCl2, reflux, 24 h; c) 2 (for BAB(ET))/3 (for BAB(H)), Et3N,
CH2Cl2, 258C, 24 h; d) CH3COCl, CH2Cl2, 258C, 12 h.
the NÀH proton which is found as a singlet at 10.88 ppm. The
13C NMR spectra present the expected 4 and 15 signals for the
alkyl and aromatic carbon atoms, respectively. The amide car-
bonyl signal is found at 162.1 ppm in CDCl3.
Electrochemistry
Cyclic voltammetry studies were carried out for the target
compounds, BOD, BAB(ET) and BAB(H), in both CH2Cl2 and
CH3CN with tetra-N-butylammonium hexafluorophosphate as
the supporting electrolyte and the main results are summar-
ised in Table 1. The control compound, BOD, exhibits quasi-re-
versible one-electron oxidation and reduction steps with half-
wave potentials of +0.95 V and À1.40 V versus SCE, respective-
ly, that can be assigned to formation of the corresponding p-
radical cation and anion, respectively (Figure 2).[18] The amide
function has no effect on the reduction potentials and does
not make a direct contribution to the cyclic voltammograms.
The two solvents give comparable results. For BAB(ET), similar
oxidation and reduction processes are observed, with much
the same half-wave potentials to those noted for the control
compound, and there are no undue effects associated with the
bridging 1,10-phenanthroline unit. With the bichromophore,
however, both processes involve the quasi-reversible transfer
of two electrons. That the two electrons are transferred at the
same potential is taken as testimony to the fact that the
BODIPY units do not interact in an electronic sense.
Figure 2. Typical cyclic voltammograms recorded for the target compounds
(ca. 2 mm) in de-aerated CH2Cl2 containing background electrolyte. Note Fc
refers to ferrocene used as internal standard. The lower axis corresponds to
mV versus SCE.
even in polar solvents. The electrochemistry also indicates that
there is no electronic interaction between the appended
BODIPY residues.
Photophysical properties
The absorption and fluorescence spectra recorded for the con-
trol compound, BOD, in tetrahydrofuran (THF) at room temper-
ature are entirely in accord with what might be expected for
a conventional BODIPY dye.[9] The absorption maximum (lmax
)
Although BAB(H) displays similar behaviour to that noted for
BAB(ET), there is indication for the one-electron reduction of
the 1,10-phenanthroline-based bridge at about À1.6 V versus
SCE (Figure 2). The appearance of this wave is attributed to in-
ternal hydrogen bonding between the aza-N atom and the
amide proton. In fact, this situation was confirmed by IR spec-
troscopy. The significance of these results is that light-induced
charge transfer between the BODIPY units and the bridging
1,10-phenanthroline unit is thermodynamically unfavourable,
lies at 523 nm while the emission maximum (lfl) is found at
536 nm. Both values are unremarkable when considered in
terms of related BODIPY derivatives.[19] Moreover, both the
fluorescence quantum yield (FF =0.80) and excited-singlet-
state lifetime (tS =5.9 ns) are in line with values recorded for
structurally related dyes.[9,19] There is excellent agreement be-
tween absorption and excitation spectra and the time-resolved
emission decay profiles are well described by mono-exponen-
tial fits. Changes in solvent polarity have little effect on these
Chem. Eur. J. 2016, 22, 1 – 12
www.chemeurj.org
3
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
various parameters, although the absorption and emission
maxima do respond slightly to changes in solvent polarisabili-
ty.[20] Spectral properties recorded for the bichromophores
under identical conditions are closely comparable to those re-
corded for BOD (Table 2, Figure 3). Thus, we can conclude that
there are no undue electronic effects associated with either
the amide linker or the 1,10-phenanthroline spacer. In THF, the
quantum yields and excited-state lifetimes recorded for
BAB(ET) and BAB(H) are comparable to those found for BOD,
although fluorescence is decreased slightly, especially for
BAB(H) (Table 2). For the two bichromophores, excitation and
absorption spectra are in close agreement and time-resolved
decay curves remain mono-exponential over at least three
half-lives. The presence of molecular oxygen does not affect
the derived properties, while changes in solvent polarity have
only minimal effect on the absorption and emission spectral
maxima and on the band half-widths.
found for the quantum yield. However, in polar solvents (es >
10) the quality of the statistical fit to a mono-exponential
decay falls below the satisfactory level.[21] This fit can be
judged best in terms of the randomness of the weighted resid-
uals,[22] but is also evident in an increased chi-squared parame-
ter (c2) and in the Durbin–Watson term (note: c2 is given in
Table 3). Inclusion of a short-lived (i.e., <1 ns) or long-lived
(i.e., >10 ns) component did not improve the quality of the fit.
However, analysis of the decay curves as dual-exponential fits
with lifetimes in the range of 5–7 ns and 1–3 ns gave superior
(i.e., more random) residuals and c2 parameters closer to unity.
Furthermore, the fractional contribution (A1) of the shorter-
lived component (t1) was found to increase in significance
with increasing solvent polarity (see Supporting Information).
Although the two lifetimes are too close for unique solutions
to be extracted from these fits, the dual-exponential behaviour
appears to better represent the situation in polar solution.
Even so, certain hydrogen-bond acceptor solvents, specifically
N,N-diethylformamide, N,N-diethylacetamide and dimethylsulf-
oxide, appear to behave anomalously.
Table 2. Comparison of the photophysical properties determined for the
target compounds in THF at room temperature.
At relatively low levels of precision, the dual-exponential or
stretched-exponential fits give an adequate representation of
the time-resolved emission decay profiles. At higher levels of
precision, these fits become less satisfactory in terms of the
randomness of the residuals and it is necessary to add a third
exponential term. Even so, unique solutions could not be re-
covered from data collected over different time scales or count
rates. An additional problem is that the derived lifetimes are
too similar for accurate analysis. Under such conditions, it is
impossible to justify the use of two or three discrete lifetimes
Compound lmax [nm] lfl [nm] FF tS [ns] SS [cmÀ1
]
krad [ꢁ108 sÀ1
]
[a]
[b]
BOD
BAB(H)
BAB(ET)
523
522
524
536
540
538
0.80 5.9
0.73 5.4
0.75 5.5
465
640
500
1.38
1.40
1.40
[a] SS refers to the Stokes’ shift for normalised spectra. [b] Radiative rate
constant calculated from the Strickler–Berg expression.
Table 3. Effect of solvent dielectric constant on the photophysical prop-
erties of BAB(ET) as recorded at room temperature.
Solvent
es
FF
tS [ns][a]
krad [ꢁ108 sÀ1
]
c2[b]
toluene
2.43
3.18
4.33
4.89
6.02
7.47
7.58
9.02
13.0
20.0
21.4
24.3
24.6
27.3
28.9
29.0
30.0
33.6
37.5
38.3
46.7
0.78
0.83
0.86
0.85
0.82
0.79
0.75
0.78
0.68
0.58
0.67
0.48
0.55
0.46
0.44
0.66
0.51
0.43
0.35
0.66
0.64
5.74
6.85
6.35
6.45
6.60
6.60
5.85
6.17
6.10
6.00
6.00
5.60
5.90
5.55
5.60
6.05
4.85
4.15
3.35
5.65
5.74
1.36
1.21
1.35
1.32
1.24
1.20
1.28
1.26
1.12
0.97
1.12
0.86
0.93
0.83
0.79
1.09
1.05
1.04
1.05
1.17
1.11
0.95
0.92
1.05
1.07
1.06
0.99
0.96
1.12
1.30
1.43
1.12
1.39
1.52
1.87
1.51
1.09
1.73
1.87
2.20
1.32
1.28
dibutyl ether
diethyl ether
CHCl3
ethyl acetate
MTHF
THF
CH2Cl2
heptyl cyanide
valeronitrile
acetone
Figure 3. Normalised absorption (black curve) and fluorescence (grey curve)
spectra recorded for BAB(H) in THF at room temperature.
ethanol
butyronitrile
nitropropane
propionitrile
DEF[c]
The photophysical properties of the tertiary amide deriva-
tive, BAB(ET), were recorded in a range of organic solvents of
differing dielectric constant, es, and the results are summarised
in Table 3. The variation in FF is small across the range of sol-
vents, despite es changing from 2 to 47, and there is no notice-
able influence of hydrogen-bond donor solvents. On close
scrutiny, it seems that there is a gradual decrease in FF with in-
creasing solvent polarity, but the effect is shallow and certain
solvents do not follow the generic pattern. The excited-state
lifetime measured in the same series of solvents decreases
with increasing solvent polarity in much the same manner as
chloroacetonitrile
methanol
acetonitrile
DEA[d]
DMSO[e]
[a] Lifetime of the excited-singlet state based on a single-exponential fit.
[b] Reduced chi-squared parameter associated with the single-exponen-
tial fit. [c] DEF=N,N-Diethylformamide. [d] DEA=N,N-Diethylacetamide.
[e] DMSO=Dimethylsulfoxide.
&
&
Chem. Eur. J. 2016, 22, 1 – 12
www.chemeurj.org
4
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
as opposed to a continuous distribution of lifetimes. Certainly,
the molecular structure is not suggestive of several discrete
conformations because of what might be considered facile ro-
tations. This situation is not uncommon in fluorescence spec-
troscopy, especially with regards to biologically relevant mate-
rials, and has led to the introduction of the maximum entropy
method[23] (MEM). This analytical approach, modified to in-
crease reliability about the zero-time shift,[24] allows recovery of
the shape of distributions of lifetimes. The MEM method
allows determination of the coefficients of an exponential
series of pre-set lifetimes isolated from correlation effects and
instrumental oscillations.
An increased level of solvent sensitivity is displayed by
BAB(H) across the same series of solvents (Table 4). In general,
increasing es tends to decrease FF, although the overall behav-
iour is non-linear and again there are a few anomalous sol-
vents, which were identified as being acetone (es =21.4), N,N-
diethylformamide (es =29), N,N-diethylacetamide (es =38.3) and
dimethylsulfoxide (es =46.7) and are hydrogen-bond acceptors.
As such, these solvents might be expected to form a hydrogen
bond with the amide proton and thereby perturb the molecu-
lar conformation. Otherwise, protic and aprotic solvents follow
a common trend and there is no indication that solvent attach-
ment to the aza-N atoms is important in the fluorescence
quenching process. In mixtures of THF (es =7.6) and acetoni-
trile (es =37.5), FF decreases progressively with increasing
mole fraction of acetonitrile, while that for BOD, used as the
control, shows no such effect (Figure 5).
Using a variety of time ranges for each sample, it proved
possible to recover reproducible lifetime distributions for
BAB(ET) in polar and weakly polar solvents. In each case, the
longer-lived species gave a lifetime of around 4–6 ns, while the
shorter-lived component was in the region of 1–2 ns (Fig-
ure 4a). There was no evident correlation between mean life-
time and solvent polarity, but the total contribution of the
shorter-lived species, integrated as a Gaussian profile, in-
creased progressively with increasing es (see Supporting Infor-
mation). Thus, based on the MEM analysis, there appears to be
a wide variety of slowly interconverting conformers that fall
into two loose families, which differ in terms of the relative ra-
diative probability of the BODIPY fluorophores.
Table 4. Effect of solvent dielectric constant on the photophysical prop-
erties of BAB(H) as recorded at room temperature.
Solvent
es
FF
tS [ns][a]
krad [ꢁ108 sÀ1
]
c2[b]
toluene
2.43
3.18
4.33
4.89
6.02
7.47
7.58
9.02
13.0
20.0
21.4
24.3
24.6
27.3
28.9
29.0
30.0
33.6
37.5
38.3
46.7
0.60
0.65
0.71
0.73
0.74
0.73
0.73
0.63
0.66
0.56
0.50
0.46
0.44
0.39
0.23
0.48
0.21
0.19
0.07
0.42
0.12
4.19
4.50
5.15
5.52
5.23
4.88
5.36
5.40
5.20
5.14
5.18
4.60
4.55
3.57
1.96
4.85
1.91
1.00
0.74
3.92
0.90
1.43
1.44
1.38
1.32
1.41
1.50
1.36
1.17
1.27
1.10
0.97
1.00
0.97
1.11
1.12
1.00
1.10
1.90
0.95
1.07
1.33
1.02
0.96
0.97
1.10
1.08
1.18
1.14
1.38
1.53
1.82
1.48
1.94
2.11
1.82
2.35
1.60
2.28
2.35
3.10
2.23
1.93
dibutyl ether
diethyl ether
CHCl3
ethyl acetate
MTHF
THF
CH2Cl2
heptyl cyanide
valeronitrile
acetone
It might be important to stress that es is not the only means
for expressing solvent polarity[25] and, in fact, several alterna-
tives are available. These include Reichardt’s empirical ET(30)
parameter,[26] the Kirkwood factor,[27] and the Catalan SPP[28]
and SB[29] factors. It is not the purpose of the present investiga-
tion to critically compare these solvent descriptors, but it
should be emphasised that similar behaviour is noted in all
cases with regards to the polarity effect on fluorescence proba-
bility.
ethanol
butyronitrile
nitropropane
propionitrile
DEF[c]
chloroacetonitrile
methanol
acetonitrile
DEA[d]
DMSO[e]
[a] Lifetime for the excited-singlet state as recovered from single-expo-
nential fits. [b] Reduced chi-squared parameter associated with the sin-
glet-exponential fits. [c] DEF=N,N-diethylformamide. [d] DEA=N,N-dieth-
ylacetamide. [e] DMSO=dimethylsulfoxide.
As above, the fluorescence decay curves recorded for
BAB(H) could be analysed satisfactorily in terms of mono-expo-
nential fits in non-polar (i.e., es <8) solvents. However, with in-
creasing solvent polarity the quality of the exponential fit was
seen to worsen; this judgement was based[21] on the reduced
chi-squared parameters (c2 in Table 4) and the randomness[22]
of the weighted residuals. Furthermore, the radiative rate con-
stant (krad =FF/tS) adopts a strong dependence on solvent po-
larity that does not seem justified.[30] The quality of the fit im-
proved on inclusion of a second component, corresponding to
a species with a shorter lifetime, but analysis of different decay
profiles recorded for the same solvent on differing time bases
Figure 4. Maximum entropy method plots derived by fitting the deconvolut-
ed emission decay curves for a) BAB(ET) and b) BAB(H) in butyronitrile (grey
curve) and acetonitrile (black curve). See the text and Table 5 for the derived
parameters.
Chem. Eur. J. 2016, 22, 1 – 12
www.chemeurj.org
5
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
Fluorescence quenching
In consideration of the fluorescence quenching mechanism,
we draw attention to the similar qualitative behaviour of the
two compounds, but note that fluorescence quenching is
much more pronounced for BAB(H) than for BAB(ET). The two
bichromophores exhibit comparable optical spectroscopy and
electrochemistry such that the disparate quenching level dis-
played in any given solvent cannot be ascribed to thermody-
namic effects. The only chemical difference between the com-
pounds relates to the substitution pattern around the amide
linker. These linkers are not in electronic communication with
the excited state localised on the BODIPY chromophore, sug-
gesting that their role in the quenching event is to perturb the
molecular conformation. We can also eliminate light-induced
charge transfer between BODIPY and 1,10-phenanthroline as
playing an important role, since such processes are thermody-
namically unfavourable, unless the latter unit is protonated.
Consideration of the cyclic voltammograms and taking due al-
lowance for the excitation energy of the BODIPY chromophore,
as derived from the intersection of normalised absorption and
fluorescence spectra, indicates that light-induced charge trans-
fer between the two BODIPY units is weakly exergonic in
CH2Cl2. Indeed, the thermodynamic driving force (ÀDGCS) is
50 meV in the absence of electrostatic effects. This mechanism
remains the most likely cause of the observed emission
quenching in the target bichromophores and it is well estab-
lished[31] that the thermodynamics for light-induced charge
transfer are sensitive to the nature of the solvent. It might be
mentioned that other molecules containing two BODIPY-based
chromophores have been observed to show excimer emis-
sion,[32] dimerisation[33] and light-induced charge transfer[34] in
fluid solution.
Based on existing theoretical principles,[31,35,36] the effects of
changes in solvent polarity and mutual separation distance be-
tween the reactants on DGCS were simulated. The results are
compiled in the Supporting Information in terms of driving
force, solvent re-organisation energy and overall activation
energy. At any given separation distance, DGCS decreases
sharply with increasing es, but the general effect tends to satu-
rate at moderate dielectric constant (i.e., es >10). Similarly, at
a given es, DGCS decreases with decreasing separation distance,
but tends towards a plateau as RCC exceeds about 10 ꢂ. Similar
effects are found for the activation energy, which requires
knowledge of the re-organisation energy. Most of the change
is expected to occur for weakly polar solvents at quite short
separations.[36,37] Although charge-transfer quenching will
become more significant in polar solvents, the calculated
changes in thermodynamic effects cannot explain the solvent-
induced variations in fluorescence seen for these bichromo-
phores. There are no differences in DGCS between BAB(ET) and
BAB(H), but the level of fluorescence quenching is quite dispa-
rate. Taking into account the MEM-derived results, we can con-
clude that there must be an accompanying solvent-induced
structural change[38] and we now examine this possibility using
molecular modelling.
Figure 5. Effect of solvent composition on the fluorescence intensity record-
ed for BAB(H). The arrow indicates the direction of increasing mole fraction
of acetonitrile in the mixture with THF; individual spectra refer to acetoni-
trile mole fractions of 0, 0.44, 0.61, 0.76 and 1.0.
did not return unique lifetime values. Again, data collected at
high precision required analysis in terms of three discrete life-
times.
Using the MEM analytical approach, it was possible to realise
reproducible fits to two broad distributions of lifetimes in
polar and weakly polar solvents (Figure 4b). The shorter-lived
distribution had a mean lifetime of about 0.5 ns, while the
second series correspond to a mean lifetime of about 3 ns
(Table 5). The total contribution of the shorter component in-
creased in more polar solvents but the mean lifetime did not
correlate with es. We are led to the conclusion that the re-
duced quantum yield arises from increased population of
a family of bichromophores more susceptible to intramolecular
fluorescence quenching. The nature of the solvent determines
the extent of this population but there is no relationship be-
tween the level of quenching inherent to that family and the
solvent polarity.
Table 5. Summary of the parameters derived from the MEM-based analy-
sis of the time-resolved emission data collected for BAB(H) in polar sol-
vents at room temperature.
Solvent
es
tA [ns][a]
tB [ns]
A1
c2[b]
CH2Cl2
9.02
13.0
20.0
21.4
24.3
24.6
27.3
28.9
30.0
33.6
37.5
46.7
1.80
0.84
0.30
1.45
1.00
0.55
0.23
0.33
0.27
1.10
0.34
0.80
5.1
5.5
4.85
3.9
3.65
4.3
3.75
3.65
2.5
0.65
2.3
0.14
0.14
0.16
0.09
0.10
0.28
0.25
0.45
0.48
0.80
0.91
0.95
1.51
1.47
1.90
1.15
1.44
1.09
1.68
1.47
1.36
1.90
1.15
1.33
heptyl cyanide
valeronitrile
acetone
ethanol
butyronitrile
nitropropane
propionitrile
chloroacetonitrile
methanol
acetonitrile
DMSO[c]
2.9
[a] Mean lifetime for the shorter-lived component derived from fitting the
distribution to a Gaussian profile. [b] Reduced chi-squared parameter as-
sociated with the fit between simulated and experimental decay curves.
[c] DMSO=dimethylsulfoxide.
&
&
Chem. Eur. J. 2016, 22, 1 – 12
www.chemeurj.org
6
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
Computational studies made with BAB(ET)
group, the first barrier to be encountered relates to the ethyl
group moving into space occupied by the meso-phenyl ring of
the second BODIPY unit. This barrier is slightly less than 20 kcal
molÀ1, but can be reduced by concerted motion of the oppos-
ing phenyl ring. A more substantial barrier is raised by close
approach of the ethyl-CH2 group to the H3 atom of the aza-ar-
omatic unit, followed by a less severe clash between the ethyl-
CH3 group and the same proton; these barriers are essentially
insurmountable at ambient temperature and restrict the direc-
tion through which the carbonyl group can oscillate between
opposite and adjacent sites (Figure 6a).
Molecular modelling studies (DFT, B3LPY, 6-31G(d,p)) were un-
dertaken with a view to better understand the mechanism of
light-induced electron transfer in the two bichromophores.
Starting firstly with BAB(ET), model calculations indicate that
the carbonyl groups are hindered from adopting a planar ori-
entation with the 1,10-phenanthroline unit because of steric
crowding. Instead, the lowest energy configuration has the
two carbonyl groups sitting almost perpendicular to the plane
of the 1,10-phenanthroline nucleus. This gives rise to two inter-
convertible species; namely, the opposite structure having
oxygen atoms pointing away from each other and the adjacent
structure having the oxygen atoms pointing in the same direc-
tion (Scheme 3). For structures generated in vacuo, each amide
group adopts the cis geometry, with the opposite form provid-
ing the more stable species. The same situation holds for struc-
tures calculated in a solvent reservoir of es =20. However, in
weakly polar solvents, there is increased likelihood that one of
the carbonyl groups adopts the trans geometry. In water, the
lowest energy conformation has both carbonyl groups existing
as the trans species, although the mixed trans/cis geometry is
only marginally less stable in polar media. This matrix of com-
puted structures is illustrated in Scheme 3 and it might be
stressed that it is not possible to generate a reasonable geom-
etry for the hypothetical cis/cis species in the adjacent form
because of severe steric crowding.
Calculations made for es =20 suggest the presence of sever-
al barriers for rotation of the carbonyl unit around the plane of
the 1,10-phenanthroline residue. Thus, starting from the adja-
cent trans/trans geometry and rotating a single carbonyl
Figure 6. Rotational barriers computed for progressive twisting of one of the
carbonyl groups present in BAB(ET) in a solvent reservoir with dielectric con-
stant equal to 20. Panel a) refers to the all-trans form with zero degrees rep-
resenting the opposite form. Panel b) refers to the cis-trans species with ro-
tation around the cis-amide. Here, the energy-minimised geometry corre-
sponds to a dihedral angle of 808.
For the mixed cis/trans species, the carbonyl groups show
only a minor preference for the opposite configuration, the dif-
ference in energy between opposite and adjacent species
being only a few kcalmolÀ1 at es =20. Full rotation around the
cis-amide is not possible because of severe steric clashes. The
energy-minimised geometry has the carbonyl group on the
cis-amide lying 858 to the plane of the 1,10-phenanthroline res-
idue. Rotation in one direction causes the BODIPY unit to fold
back onto the 1,10-phenathroline nucleus, while rotation in
the other direction brings the phenyl ring on the cis-amide
into contact with the 1,10-phenanthroline H3 atom (Figure 6b).
With the carbonyl groups in the adjacent configuration, rota-
tion brings the BODIPY unit into contact with the ethyl group
on the other amide. Interconversion between the opposite
and adjacent configurations is possible only with concerted
movement of the other BODIPY-based arm.
Scheme 3. Effect of solvent polarity on the computed energies for the vari-
ous conformers predicted for BAB(ET). The upper panel shows energies in
kcalmolÀ1 and BÀB separation distances for opposite and adjacent species;
the values given in parenthesis are the solvent dielectric constants. The
lower panel shows the anticipated structural evolution as the dielectric con-
stant increases.
Similar calculations were performed for isomerisation of the
amide group, starting with the trans/trans species with the ad-
jacent configuration. Allowing rotation in one direction, an
energy profile for isomerisation around the CÀN bond[39]
Chem. Eur. J. 2016, 22, 1 – 12
www.chemeurj.org
7
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
shows the maximum barrier of about 20 kcalmolÀ1 occurring
at 908, as might be expected.[40] As the solvent polarity increas-
es, the barrier height is raised[41] for trans-to-cis isomerization,
but decreases for the reverse step by a few kcalmolÀ1 for es =
40. Rotation in the opposite direction imposes an insurmount-
able barrier, in excess of 100 kcalmolÀ1, which arises from
steric clashes between the two BODIPY units (Figure 7). The
corresponding cis isomer cannot be reached by this route. A
similar barrier was observed for isomerisation from the oppo-
site configuration, although here the cis isomer is the more
favoured product in moderately polar solvents.
in the trans configuration[45] for all solvent polarities. However,
the energy difference between the all-trans and cis/trans spe-
cies becomes negligible at high dielectric constant. As such, in
highly polar solvents we would expect to attain almost equal
populations of the all-trans and cis/trans species.
An energy profile[46] for rotation of the carbonyl group
around the connection with the 1,10-phenanthroline nucleus,
while retaining the trans geometry, shows a modest barrier
due to repulsion between the lone-pairs on carbonyl group
and aza-N atom (Figure 8a). The height of this barrier decreas-
Figure 7. Rotational barrier computed for isomerisation of an amide group
for BAB(ET) in a solvent reservoir with dielectric constant equal to 20. Zero
degrees corresponds to the cis-geometry.
Molecular dynamics simulations made for the all-trans tauto-
mer in a bath of water molecules (50ꢁ50ꢁ50 ꢂ3)[42] using the
AMBER-03 force field[43] indicate that the main structural mo-
tions relate to fluctuation of the carbonyl group around its
connection to the 1,10-phenanthroline nucleus. At 250 K, there
is no suggestion for isomerisation of the amide bond on the
time scale of the simulations. Likewise, the carbonyl group
does not alternate between adjacent and opposite configura-
tions. The all-cis species oscillates around the mean geometry
without switching to the trans form, at least on short simula-
tions. As such, the two BODIPY units sample many different
mutual orientations but remain at crude B–B separations of
about 14–20 ꢂ, regardless of starting geometry. None of these
migratory pathways brings the BODIPY units into direct con-
tact.
Figure 8. Rotational barriers calculated for motion of the carbonyl group
around the 1,10-phenathroline nucleus for a) the all-trans and b) all-cis geo-
metries.
es in polar solvents and amounts to about 4 kcalmolÀ1 for es =
20. Clearly, this dihedral angle will cover a wide variance under
ambient conditions. There are small differences in the calculat-
ed energies of the adjacent and opposite configurations, with
increased solvent polarity moving the equilibrium position in
favour of the former. This has a small effect on the average dis-
tance between the two BODIPY residues, but these units
remain well separated in all cases for the trans-geometry.
The corresponding cis-amide[47] in the mixed cis/trans species
positions the carbonyl group orthogonal to the heterocycle.
Rotation in one direction is blocked by steric clashes between
the two BODIPY residues (Figure 8b). In the other direction, ro-
tation is blocked by clashes between the phenyl ring and the
1,10-phenanthroline H3 atom. As such, the cis species has the
propensity to bring the two BODIPY units into unusually close
proximity.
Computational studies made with BAB(H)
Attention now turns to the bichromophore built around secon-
dary amide linkages. In order to avoid repulsion between lone
pairs localised on O and aza-N atoms, the preferred dihedral
angle for the carbonyl group and the heterocycle is about 908.
This configuration might be stabilised by hydrogen bonding
between the carbonyl O atom and an ortho-phenyl-H atom.[44]
Again, there exists the possibility for opposite and adjacent
forms, but these possess similar heats of formation and give
rise to comparable distances between the appended BODIPY
residues. The energy-minimised geometry has the amide bond
The barrier for isomerisation of the amide bond is calculat-
ed[48] to be about 15 kcalmolÀ1 for es =20, but steric crowding
between the two BODIPY residues prevents formation of the
all-cis species. Calculations made for BAB(H) in vacuo give
a barrier for isomerisation of about 32 kcalmolÀ1, but this falls
to slightly less than 15 kcalmolÀ1 in polar solvent. In fact, the
&
&
Chem. Eur. J. 2016, 22, 1 – 12
www.chemeurj.org
8
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
barrier height[49] becomes almost independent of solvent po-
larity for es >10. We might expect that isomerisation will be
too slow to be competitive with decay of the excited singlet
state, but can take place for the ground state. Furthermore, in
non-polar solvents, the all-trans species will dominate, but
a mixture of all-trans and cis/trans species should abound in
solvents of higher polarity. The trans species can sample
a wide variety of configurations through rotation of the car-
bonyl group, but the cis species experiences more restricted
rotation. This situation seems to be in good accord with the
MEM analysis in terms of there being two distinct groups of
conformers.[50]
moment of excitation. In turn, this would lead to the type of
distributed lifetimes consistent with the MEM analysis.
Quite unexpectedly, there are major structural differences
between BAB(ET) and BAB(H) caused by internal steric crowd-
ing and/or electronic effects. Fluorescence quenching becomes
more significant for BAB(H), although the thermodynamic driv-
ing force is the same as that for BAB(ET) and there is a similar
sensitivity towards solvent polarity. Calculations made for
BAB(H) predict that the trans geometry is favoured in all sol-
vents, this being the opposite situation to that found for
BAB(ET), but polar solvent promotes transformation to the cis
isomer. In non-polar solvents, we would expect to see only the
all-trans species. Increasing the solvent polarity raises the pos-
sibility for finding the trans/cis species.
By analogy to BAB(ET), increasing the contribution of the cis
species might be expected to extinguish through-bond charge
separation. This would restore fluorescence. However, for
BAB(H) the cis/trans species has the two BODIPY units close to-
gether, contact being possible in the extreme case, while rota-
tion around one of the carbonyl groups further reduces the
edge-to-edge separation (Figure 9). As such, the cis isomer can
be expected to promote through-space, light-induced charge
separation[57] between the two BODIPY units. This situation
would introduce a short-lived component into the decay re-
cords in polar solvents. Since each isomer samples a variety of
conformations, the lifetime distributions provided by the MEM
analysis appear to mirror the solvent-induced conformational
change.
Relationship between fluorescence quenching and molecu-
lar structure
We raise the hypothesis that the only viable mechanism able
to account for the observed solvent effect on the emission
properties of the BODIPY unit in these bichromophores is
light-induced charge transfer[31,34,51] between the terminal dyes.
As such, it is instructive to enquire if the proposed changes in
molecular conformation can explain the experimental observa-
tions. For BAB(ET), the two BODIPY units are held apart under
all reasonable conditions to such an extent that through-space
charge transfer[52] is unlikely to compete effectively with the in-
herent radiative and nonradiative decay routes. Fluorescence
quenching is ineffectual for this compound, except in strongly
polar media. Polar solvents promote conversion of the cis spe-
cies to the corresponding trans tautomer. Our modelling stud-
ies would suggest that, in strongly polar media, the bichromo-
phore should persist as a mixture of all-trans and cis/trans iso-
mers. Combining this result with the fluorescence behaviour,
we can speculate that the trans geometry provides a better
conduit[53] for through-bond charge transfer. The long path-
way,[54] together with the modest thermodynamic driving
force, means emission quenching will be kept at a minimum,
as is observed. The mean difference between the two sets of
emission lifetimes, taken together with the lifetime of the con-
trol compound, translates to a ratio of rate constants for
charge-separation of twofold in favour of the trans-amide.
There is, in fact, ample evidence to indicate that the trans
geometry provides a better pathway for super-exchange inter-
actions in many different types of molecular bridge.[55–58] This is
attributed to improved electronic coupling and nicely explains
the observations made with BAB(ET). The computational stud-
ies suggest that the all-cis species will predominate in non-
polar solvents and also in weakly polar media. Strongly polar
solvents, however, trigger the switch to the trans geometry
and we would expect to see increased population of the trans/
cis species in the more polar solvents. Isomerisation is unlikely
to be competitive with inherent deactivation of the excited
state. Instead, illumination of the ground-state equilibrium will
produce a distribution of geometries that does not intercon-
vert on the existing time scale. This situation would equate to
two families of conformers, each displaying a range of non-ra-
diative rate constants representing the mean geometry at the
It might be noted that transient absorption spectral studies
did not indicate the formation of an intermediate species with
a lifetime longer than that of the excited-singlet state. Thus,
laser excitation of BAB(ET) in deaerated CH3CN at 400 nm indi-
cated the presence of the S1 state immediately after the pulse.
This species is recognised by strong bleaching of the lowest-
energy absorption transition centred at around 525 nm, to-
gether with weak absorption bands at higher and lower
energy. There is an accompanying contribution from stimulat-
ed fluorescence. The signal decays with an approximate life-
time of 3.0Æ0.7 ns to restore the pre-pulse baseline. Although
the decay kinetics are not strictly mono-exponential, global
analysis of the transient spectra recorded at different time
delays showed only the S1 state to be present. The same con-
clusion was reached for BAB(H) in CH3CN, for which the recov-
ery of the transient bleaching signal at 525 nm could be ana-
lysed as the sum of two exponential terms, with lifetimes of
0.4Æ0.1 (80%) and 2.1Æ0.5 ns (20%). Again, no transient spe-
cies could be observed. This behaviour is not unusual and indi-
cates that subsequent charge recombination occurs on a faster
time scale than does light-induced charge transfer.[59] The
former process will be assisted by electrostatic attractive forces
that should minimise separation between the two BODIPY
units.
Conclusions
This work has shown that the amide linkage provides for multi-
ple molecular conformations that differ in terms of their pro-
Chem. Eur. J. 2016, 22, 1 – 12
www.chemeurj.org
9
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
ometry while BAB(H) takes up the trans-geometry. These vari-
ous conformations influence the quantum yield for fluores-
cence from the BODIPY appendages, but quenching is kept to
a minimum by limited thermodynamics. It seems likely that
these effects could be greatly amplified by appropriate choice
of the terminals. Thus, the amide linkage might offer some un-
usual opportunities to switch between emissive and dark
states under suitable stimulation. This situation is made possi-
ble because the absolute energies of the various tautomers are
quite comparable, while rotational barriers will impose kinetic
limitations at the excited-state level.
Experimental Section
Experimental details for the synthesis of the target compounds are
provided as part of the Supporting Information. Solvents used for
the spectroscopic studies were obtained from commercial sources
and were checked for fluorescent impurities before use. Absorption
spectra were recorded with a Hitachi U-3310 spectrophotometer
while emission spectra were recorded with a Hitachi F-4500 spec-
trophotometer. Fluorescence spectra were recorded for optically
dilute solution, the absorbance being less than 0.1 at the excitation
wavelength, and were fully corrected for any instrumental imper-
fections. Fluorescence quantum yields were recorded relative to
monomeric BODIPY derivatives already reported in the literature.
Emission lifetimes were recorded via time-correlated, single photon
counting methodologies using a short duration laser diode emit-
ting at 505 nm for excitation. Quantum chemical calculations were
made with TURBOMOLE and are described in more detail in the
Supporting Information.
Acknowledgements
We thank Newcastle University, BRNS (BARC) , and ECPM-Stras-
bourg for their financial support of this work. Dr. J. G. Knight
(Newcastle University) is acknowledged for many helpful dis-
cussions on the chemistry of the amide bond.
Keywords: BODIPY · boron · charge transfer · conformational
exchange · dyes/pigments · fluorescence · isomerisation
[1] D. Whitford, Proteins: Structure and Function, Wiley, Chicester, 2005.
[2] N. Seward, H.-D. Jakubke, Peptides: Chemistry and Biology, 2nd ed.,
Wiley-VCH, Weinheim, 2009.
[3] E. Iglesias, L. Montenegro, J. Chem. Soc. Faraday Trans. 1996, 92, 1205–
1212.
Figure 9. Snapshots of molecular conformations representing the important
distributions for all-trans (uppermost panel), mixed-cis/trans (central panels),
and all-cis (lower panel) geometries for BAB(H) in solution.
[4] a) D. B. Brown, M. B. Robin, R. D. Burbank, J. Am. Chem. Soc. 1968, 90,
5621–5622; b) D. P. Fairlie, T. C. Woon, W. A. Wickramasinghe, A. C.
Willis, Inorg. Chem. 1994, 33, 6425–6428; c) C. R. Kemnitz, M. J. Loewen,
J. Am. Chem. Soc. 2007, 129, 2521–2528; d) E. M. Duffy, D. L. Severance,
W. L. Jorgensen, J. Am. Chem. Soc. 1992, 114, 7535–7542; e) J. Zhang,
M. W. Germann, Biopolymers 2011, 95, 755–762.
[5] a) H. B. Schlegel, P. Gund, E. M. Fluder, J. Am. Chem. Soc. 1982, 104,
5347–5351; b) K. B. Wiberg, P. R. Rablen, D. J. Rush, T. A. Keith, J. Am.
Chem. Soc. 1995, 117, 4261–4270.
[6] K. Kamiya, M. Boero, K. Shiraishi, A. Oshiyama, J. Phys. Chem. B 2006,
110, 4443–4450.
[7] S. H. Gellman, G. P. Dado, G. B. Liang, B. R. Adams, J. Am. Chem. Soc.
1991, 113, 1164–1173.
pensity to promote either through-bond or through-space
charge transfer. In addition to isomerisation around the central
CÀN amide bond, structural complications arise from electronic
interactions between the carbonyl O atom and the aza-N atom
of the 1,10-phenanthroline unit. Indeed, the BODIPY-based
arms attached to the heterocyclic linker provide a crude cavity
wherein hydrogen-bonding and steric interactions further re-
strict conformational freedom to such an extent that the two
target compounds appear structurally distinct. This situation is
exemplified by the realisation that BAB(ET) adopts the cis-ge-
[8] P. J. Cragg, Supramolecular Chemistry: From Biological Inspiration to Bio-
medical Applications, Springer, Dordrecht, 2010.
&
&
Chem. Eur. J. 2016, 22, 1 – 12
www.chemeurj.org
10
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
[9] a) A. Loudet, K. Burgess, Chem. Rev. 2007, 107, 4891–4932; b) G. Ulrich,
R. Ziessel, A. Harriman, Angew. Chem. Int. Ed. 2008, 47, 1184–1201;
Angew. Chem. 2008, 120, 1202–1219; c) N. Boens, V. Leen, W. Dehaen,
Chem. Soc. Rev. 2012, 41, 1130–1172; d) A. Kamkaew, S. H. Lim, H. B.
Lee, L. V. Kiew, L. Y. Chung, K. Burgess, Chem. Soc. Rev. 2013, 42, 77–88;
e) H. Lu, J. Mack, Y. Yanga, Z. Shen, Chem. Soc. Rev. 2014, 43, 4778–
4823.
[10] a) M. A. H. Alamiry, E. Bahaidarah, A. Harriman, J. H. Olivier, R. Ziessel,
Pure Appl. Chem. 2013, 85, 1349–1365; b) A. Harriman, R. Ziessel, Photo-
chem. Photobiol. Sci. 2010, 9, 960–967.
[33] a) A. Harriman, L. J. Mallon, B. Stewart, G. Ulrich, R. Ziessel, Eur. J. Org.
Chem. 2007, 3191–3198; b) A. N. Kursunlu, Tetrahedron Lett. 2015, 56,
1873–1877; c) R. Misra, B. Dhokale, T. Jadhav, S. M. Mobin, New J. Chem.
2014, 38, 3579–3585; d) S. Choi, J. Bouffard, Y. Kim, Chem. Sci. 2014, 5,
751–755; e) A. B. Nepomnyashchii, M. Broring, J. Ahrens, A. J. Bard, J.
Am. Chem. Soc. 2011, 133, 8633–8645.
[34] a) M. T. Whited, N. M. Patel, S. T. Roberts, K. Allen, P. I. Djurovich, S. E.
Bradforth, M. E. Thompson, Chem. Commun. 2012, 48, 284–286; b) A. C.
Benniston, G. Copley, A. Harriman, D. Howgego, R. W. Harrington, W.
Clegg, J. Org. Chem. 2010, 75, 2018–2027.
[11] a) R. Ziessel, G, Ulrich, A. Harriman, New J. Chem. 2007, 31, 496–501;
b) A. C. Benniston, G. Copley, Phys. Chem. Chem. Phys. 2009, 11, 4124–
4130; c) M. Gupta, S. Mula, M. Tyagi, T. K. Ghanty, S. Murudkar, A. K. Ray,
S. Chattopadhyay, Chem. Eur. J. 2013, 19, 17766–17772.
[12] J. Rosenthal, A. B. Nepomnyashchii, J. Kozhukh, A. J. Bard, S. J. Lippard,
J. Phys. Chem. C 2011, 115, 17993–18001.
[13] V. MuÇoz, L. Serrano, Nature Struct. Biol. 1994, 1, 399–409.
[14] A. Ben-Naim, Adv. Biol. Chem. 2013, 3, 29–39 and references therein.
[15] a) S. Mula, G. Ulrich, R. Ziessel, Tetrahedron Lett. 2009, 50, 6383–6388;
b) S. Mula, K. Elliott, A. Harriman, R. Ziessel, J. Phys. Chem. A 2010, 114,
10515–10522.
[35] P. F. Barbara, T. J. Meyer, M. A. Ratner, J. Phys. Chem. 1996, 100, 13148–
13168.
[36] a) D. Rehm, A. H. Weller, Isr. J. Chem. 1970, 8, 259–271; b) D. Rehm,
A. H. Weller, Ber. Bunsen-Ges. 1969, 73, 834–839; c) A. H. Weller, Z. Phys.
Chem. 1982, 133, 93–98.
[37] G. King, A. Warshel, J. Chem. Phys. 1990, 93, 8682–8692.
[38] T. Otosu, K. Ishii, T. Tahara, Nat. Commun. 2015, 6, 7685.
[39] A. G. Martꢆnez, E. T. Vilar, A. G. Fraile, P. Martꢆnez-Ruiz, J. Phys. Chem. A
2002, 106, 4942–4950.
[40] S. Samdal, J. Mol. Struct. 1998, 440, 165–174.
[41] Y. A. Mantz, H. Gerard, R. Iftimie, G. J. Martyna, J. Phys. Chem. B 2006,
110, 13523–13538.
[42] V. Lounnas, S. K. Ludemann, R. C. Wade, Biophys. Chem. 1999, 78, 157–
182.
[16] a) H. Sajiki, T. Ikawa, K. Hirota, Org. Lett. 2004, 6, 4977–4980; b) R. Na-
cario, S. Kotakonda, D. M. D. Fouchard, L. M. V. Tillekeratne, R. A.
Hudson, Org. Lett. 2005, 7, 471–474.
[17] a) C. J. Chandler, L. W. Deady, J. A. Reiss, J. Heterocycl. Chem. 1981, 18,
599–601; b) D. Manna, S. Mula, A. Bhattacharyya, S. Chattopadhyay,
T. K. Ghanty, Dalton Trans. 2015, 44, 1332–1340.
[18] a) R. Ziessel, L. Bonardi, P. Retailleau, G. Ulrich, J. Org. Chem. 2006, 71,
3093–3102; b) T. Bura, R. Ziessel, Tetrahedron Lett. 2010, 51, 2875–
2879; c) N. Shivran, S. Mula, T. K. Ghanty, S. Chattopadhyay, Org. Lett.
2011, 13, 5870–5873.
[43] C. G. Ricci, A. S. de Andrade, M. Mottin, P. A. Netz, J. Phys. Chem. B 2010,
114, 9882–9893.
[44] D. A. Dixon, K. D. Dobbs, J. J. Valenti, J. Phys. Chem. 1994, 98, 13435–
13439.
[45] E. Beausoleil, W. D. Lubell, J. Am. Chem. Soc. 1996, 118, 12902–12908.
[46] M. A. Leiva, R. G. E. Morales, V. Vargas, J. Phys. Org. Chem. 1996, 9, 455–
458.
[19] a) J. Karolin, L. B. A. Johansson, L. Strandberg, T. Ny, J. Am. Chem. Soc.
1994, 116, 7801–7806; b) A. Burghart, H. J. Kim, M. B. Welch, L. H. Thor-
ensen, J. Reibenspies, K. Burgess, F. Bergstrom, L. B. A. Johansson, J.
Org. Chem. 1999, 64, 7813–7819; c) H. Sunahara, Y. Urano, H. Kojima, T.
Nagano, J. Am. Chem. Soc. 2007, 129, 5597–5604; d) K. K. Jagtap, N.
Shivran, S. Mula, D. B. Naik, S. K. Sarkar, T. Mukherjee, D. K. Maity, A. K.
Ray, Chem. Eur. J. 2013, 19, 702–708.
[20] W. W. Qin, M. Baruah, M. Van der Auweraer, F. C. De Schryver, J. Phys.
Chem. A 2005, 109, 7371–7384.
[21] D. V. O’Connor, D. Phillips, Time-Correlated Single Photon Counting, Aca-
demic Press, New York, 1979.
[22] K. Santra, J. C. Zhan, X.-Y. Song, E. A. Smith, N. Vaswani, J. W. Petrich, J.
Phys. Chem. B 2016, 120, 2484–2490.
[23] A. Siemiarczuk, B. D. Wagner, W. R. Ware, J. Phys. Chem. 1990, 94, 1661–
1666.
[24] R. Esposito, G. Mensitieri, S. de Nicola, Analyst 2015, 140, 8138–8147.
[25] A. R. Katritzky, D. C. Fara, H. Yang, K. Tamm, T. Tamm, M. Karelson, Chem.
Rev. 2004, 104, 175–198.
[47] G. Fischer, Chem. Soc. Rev. 2000, 29, 119–127.
[48] Y. K. Yang, H. S. Park, J. Mol. Struct. THEOCHEM 2004, 676, 171–176.
[49] A. Radzicka, L. Pedersen, R. Wolfenden, Biochemistry 1988, 27, 4538–
4541.
[50] F. B. Dias, M. Knaapila, A. P. Monkman, H. D. Burrows, Macromolecules
2006, 39, 1598–1606.
[51] J. M. Giaimo, A. V. Gusev, M. R. Wasielewski, J. Am. Chem. Soc. 2002, 124,
8530–8532.
[52] V. Bandi, H. B. Gobeze, P. A. Karr, F. D’Souza, J. Phys. Chem. C 2014, 118,
18969–18982.
[53] a) R. M. Williams, Photochem. Photobiol. Sci. 2010, 9, 1018–1026; b) B. P.
Paulson, L. A. Curtiss, B. Bal, G. L. Closs, J. R. Miller, J. Am. Chem. Soc.
1996, 118, 378–387; c) A. M. Oliver, D. C. Craig, M. N. Paddon-Row, J.
Kroon, J. W. Verhoeven, Chem. Phys. Lett. 1988, 150, 366–373.
[54] A. M. Brun, A. Harriman, V. Heitz, J.-P. Sauvage, J. Am. Chem. Soc. 1991,
113, 8657–8663.
[55] A. C. Benniston, A. Harriman, Chem. Soc. Rev. 2006, 35, 169–179.
[56] M. N. Paddon-Row, Aust. J. Chem. 2003, 56, 729–748.
[57] K. Pettersson, J. Wiberg, T. Ljungdahl, J. Martensson, B. Albinsson, J.
Phys. Chem. A 2006, 110, 319–326.
[58] A. Harriman, V. Heitz, J.-P. Sauvage, J. Phys. Chem. 1993, 97, 5940–5946.
[59] a) K. J. Elliott, A. Harriman, L. Le Pleux, Y. Pellegrin, E. Blart, C. R. Mayer, F.
Odobel, Phys. Chem. Chem. Phys. 2009, 11, 8767–8773; b) A. Harriman,
K. J. Elliott, M. A. H. Alamiry, L. Le Pleux, M. Severac, Y. Pellegrin, E. Blart,
C. Fosse, C. Cannizzo, C. R. Mayer, F. Odobel, J. Phys. Chem. C 2009, 113,
5834–5842.
[26] C. Reichardt, Chem. Rev. 1994, 94, 2319–2358.
[27] Y. Q. Zhou, G. Stell, J. Chem. Phys. 1992, 96, 1507–1515.
[28] J. Catalꢃn, V. Lꢄpez, P. Pꢅrez, R. Martin-Villamil, J.-G. Rodriguez, Liebigs
Ann. 1995, 241–252.
[29] J. Catalꢃn, C. Diaz, V. Lꢄpez, P. Pꢅrez, J. L. G. De Paz, J.-G. Rodriguez, Lie-
bigs Ann. 1996, 1785–1794.
[30] S. J. Strickler, R. A. Berg, J. Chem. Phys. 1962, 37, 814–822.
[31] a) M. R. Wasielewski, Chem. Rev. 1992, 92, 435–461; b) R. Ziessel, B. D.
Allen, D. B. Rewinska, A. Harriman, Chem. Eur. J. 2009, 15, 7382–7393.
[32] a) M. A. H. Alamiry, A. C. Benniston, G. Copley, A. Harriman, D. Howgego,
J. Phys. Chem. A 2011, 115, 12111–12119; b) N. Saki, T. Dinc, E. U.
Akkaya, Tetrahedron 2006, 62, 2721–2725; c) H. L. Qi, J. J. Teesdale, R. C.
Pupillo, J. Rosenthal, A. J. Bard, J. Am. Chem. Soc. 2013, 135, 13558–
13566.
Received: May 17, 2016
Published online on && &&, 0000
Chem. Eur. J. 2016, 22, 1 – 12
www.chemeurj.org
11
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
FULL PAPER
&
Dyes/Pigments
S. Thakare, P. Stachelek, S. Mula,
A. B. More, S. Chattopadhyay, A. K. Ray,
N. Sekar, R. Ziessel, A. Harriman*
&& – &&
Solvent has its finger on the trigger:
Polar solvent triggers a conformational
change from cis- to trans-geometries for
a tertiary amide-bridged molecular
dyad, but the exact opposite situation is
found for the corresponding secondary
amide-bridged bichromophore (see
scheme).
Solvent-Driven Conformational
Exchange for Amide-Linked
Bichromophoric BODIPY Derivatives
&
&
Chem. Eur. J. 2016, 22, 1 – 12
www.chemeurj.org
12
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!