Off the back or on the side: Comparison of meso and 2-substituted donor-acceptor difluoroborondipyrromethene (bodipy) dyads

Article abstract of DOI:10.1002/ejoc.201000215

The preparation of several difluoroborondipyrromethene (Bodipy) dyads is described Incorporating covalently attached hydroqumone/qumone groups at the 2-position (BDSHQ, BD-SQ, BD-SPHQ, BD-SPQ). The compounds, currently under investigation as chemical sensors for reactive oxygen species, show various levels of fluorescence depending on the oxidation state of the appended group. The ¹⁹F NMR spectrum for BD-SHQ in CDCl³ at room temperature reveals the two fluorines are inequivalent on the NMR timescale. In contrast, the ¹⁹F NMR spectrum for the counterpart quinone compound, BD-SQ, is consistent with two equivalent: fluorine atoms. The two results are interpreted as the quinone is free to rotate around the connector bond, whereas this motion is restricted for the hydroquinone group and makes the fluorines chemically inequivalent. Cyclic voltammograms recorded for all derivatives in CH₂Cl₂ electrolyte solution are consistent: with typical Bodipy-based redox chemistry; the potentials of which depend on factors such as presence of the phenylene spacer and oxidation state of the appended group. A comparison of the electrochemical behaviour with the counterpart meso derivatives reveals some interesting trends which are associated with the location of the HOMO/LUMOs. The absorption profiles for the compounds in CH ₃CN are again consistent with Bodipy-based derivatives, though there are some subtle differences in the band-shapes of the closely-coupled systems. In particular, the absorption spectra for the dyad, BD-SQ, in a wide range of solvents are appreciably broader than for BD-SHQ. Femtosecond transient absorption spectroscopy performed on the hydroquinone derivatives, BD-SHQ and its meso analogue is interpreted as electron transfer occurs from, the hydroquinone unit to the first-excited singlet (S₁) state of the Bodipy center, followed by ultrafast charge recombination to reinstate the ground state. The coupling of OH vibrations to the return electron transfer process is invoked to explain the lack of clear identification of the charge-separated state in the transient records.

Full text of DOI:10.1002/ejoc.201000215

FULL PAPER
DOI: 10.1002/ejoc.201000215
Off the Back or on the Side: Comparison of meso and 2-Substituted
Donor-Acceptor Difluoroborondipyrromethene (Bodipy) Dyads
Andrew C. Benniston,*[a] Graeme Copley,[a] Helge Lemmetyinen,*[b] and
Nikolai V. Tkachenko[b]
Keywords: Dyes / Luminescence / Fluorescence / Electrochemistry / Photochemistry / Electron transfer
The preparation of several difluoroborondipyrromethene
(Bodipy) dyads is described incorporating covalently at-
tached hydroquinone/quinone groups at the 2-position (BD-
SHQ, BD-SQ, BD-SPHQ, BD-SPQ). The compounds, cur-
rently under investigation as chemical sensors for reactive
oxygen species, show various levels of fluorescence de-
pending on the oxidation state of the appended group. The
19F NMR spectrum for BD-SHQ in CDCl3 at room tempera-
ture reveals the two fluorines are inequivalent on the NMR
appended group. A comparison of the electrochemical be-
haviour with the counterpart meso derivatives reveals some
interesting trends which are associated with the location of
the HOMO/LUMOs. The absorption profiles for the com-
pounds in CH3CN are again consistent with Bodipy-based
derivatives, though there are some subtle differences in the
band-shapes of the closely-coupled systems. In particular,
the absorption spectra for the dyad, BD-SQ, in a wide range
of solvents are appreciably broader than for BD-SHQ. Femto-
timescale. In contrast, the 19F NMR spectrum for the counter- second transient absorption spectroscopy performed on the
part quinone compound, BD-SQ, is consistent with two
equivalent fluorine atoms. The two results are interpreted as
the quinone is free to rotate around the connector bond,
whereas this motion is restricted for the hydroquinone group
and makes the fluorines chemically inequivalent. Cyclic vol-
tammograms recorded for all derivatives in CH2Cl2 electro-
lyte solution are consistent with typical Bodipy-based redox
chemistry; the potentials of which depend on factors such as
presence of the phenylene spacer and oxidation state of the
hydroquinone derivatives, BD-SHQ and its meso analogue is
interpreted as electron transfer occurs from the hydroqui-
none unit to the first-excited singlet (S1) state of the Bodipy
center, followed by ultrafast charge recombination to re-
instate the ground state. The coupling of OH vibrations to
the return electron transfer process is invoked to explain the
lack of clear identification of the charge-separated state in
the transient records.
how this can be used in the development of molecular
probes[3] that respond to outside influences such as pH,[4]
metal ions,[5] gases,[6] and radicals.[7] The detection of reac-
tive oxygen species (ROS) in biological specimens is one
area that current resources are being put into considering
the extreme importance of ROS in many biologically impor-
tant processes.[8] Previously we described the response of
the meso hydroquione derivatives, BD-MPHQ (Figure 1),
to hydrogen peroxide in a membrane mimic.[9] The deriva-
tive BD-MPHQ behaves as an excellent fluorescence-by-eye
detector of hydrogen peroxide, and has the added benefit of
being reversible since the quinone derivative (BD-MPQ)
can be reduced back in situ to the active probe. In seeking
new probes we developed the analogue series in which the
ROS sensitive group is appended in the 2-position. One in-
tention was to see if the on/off discrimination level could
be improved by a change in substitution pattern. Here we
discuss the comparison of the two series highlighting in par-
ticular differences in their electrochemical and photophysi-
cal behaviour.
Introduction
In the search for new and more robust fluorescent dyes
one type of molecular system is certainly emerging as a
serious frontrunner: the half-porphyrin dye, difluoroboron-
dipyrromethene (Bodipy).[1] The rise in popularity in the
dye can be certainly traced to its highly beneficial traits
such as high fluorescence yield, high molar absorption coef-
ficient, good photostability, readily tunable absorption pro-
file and excellent excited-state redox behaviour.[2] Like many
other researches we have focussed on the latter point and
[a] Molecular Photonics Laboratory, School of Chemistry,
Newcastle University,
Newcastle upon Tyne, NE1 7RU, UK
Fax: +44-191-222-6929
E-mail: a.c.benniston@ncl.ac.uk
[b] The Department of Chemistry and Bioengineering, Tampere
University of Technology,
P. O. Box 541, 33101, Finland
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201000215.
Eur. J. Org. Chem. 2010, 2867–2877
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2867

A. C. Benniston, G. Copley, H. Lemmetyinen, N. V. Tkachenko
FULL PAPER
(Boc and Bn) and having the protected quinone in place
from the start of the synthesis.[11] Other protecting groups
for the hydroxy groups (e.g., TIPS) were tried but failed to
survive the full synthetic procedure. The starting building
block is compound 1, which is readily prepared in three
steps using reported literature procedures.[12] Compound 2a
is an easily prepared precursor starting from 2-bromobenz-
ene-1,4-diol in two steps. To prepare the analogue com-
pound, 2b, a literature procedure[13] was slightly adapted in
which 4-bromoaniline is firstly converted to 4Ј-bromobi-
phenyl-2,5-diol, then protected with the benzyl groups and
finally converted into the boronic acid. Cross coupling of
2a/2b with 1 using standard Suzuki conditions[14] produced
the substituted pyrrole derivatives 3a/3b in good yields. The
next step and formation of the Bodipy core followed the
standard literature procedure[1a] of condensation of 3a/3b
with 3-ethyl-2,4-dimethylpyrrole in the presence of base,
coupled with oxidation and chelation to the difluoroboron
group. Both the protected compounds, BD-SBn and BD-
SPBn, were purified extensively by column chromatography
to afford dark violet solids. The first attempted method for
removal of the benzyl protecting group under one atmo-
sphere of H2 in the presence of Pd/C failed to yield the
desired compounds. Instead, both the compounds decom-
posed to dipyrromethane side products due to hydrogena-
tion of the unmasked meso double bond.[1b] Use of the
milder condition of 1,4-cyclohexadiene and Pd(OH)2/C,
however, was more successful and gave the hydroquinone
compounds, BD-SHQ and BD-SPHQ, in excellent yields.
Oxidation of the hydroquinone group to the quinone moi-
Figure 1. The Bodipy compounds discussed in the text and their
basic nomenclature.
Results and Discussion
Synthesis
In devising a synthetic procedure towards the 2-substi- ety performed using DDQ worked extremely well to afford
tuted Bodipy analogues our main aim was to prepare BD-SQ and BD-SPQ in good yields after careful
mono-substituted derivatives only, even though symmetri- chromatography. For all spectroscopic work the quinone
cal versions may have been more easier to produce since a compounds were also purified by preparative tlc (silica gel,
literature procedure was at hand for preparing 2,6-dibromo/ CH2Cl2) to remove trace amounts of the hydroquinone
iodo Bodipy derivatives.[10] The outlined method shown in compounds. All compounds were fully characterized by nu-
Scheme 1 towards the target molecules represents the most merous analytical techniques including 1H, 13C, 11B and 19
F
successful approach, utilizing optimized protecting groups NMR spectroscopy, mass spectrometry and elemental
Scheme 1. Reagents and conditions. (i) DMF, Na2CO3, Pd(PPh3)4, reflux. (ii) CH2Cl2, TFA, N,N-diisopropylethylamine, BF3·Et2O, room
temperature, (iii) EtOH, 1,4-cyclohexadiene, Pd(OH)2/C, reflux (iv) THF, DDQ.
2868
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 2867–2877

Comparison of Substituted Donor-Acceptor Bodipy Dyads
analysis. Single-crystal structural analysis was performed on the compound BD-SQ can be considered a redox molecular
several compounds, including BD-SBn, BD-SPBn and BD- brake,[18] since free rotation is retarded upon reduction to
SPQ, and confirmed their identity.[15]
the hydroquinone, and reversed upon oxidation back to the
quinone. This idea was not followed up further, but it was
noted that BD-SHQ (and any derivative thereof) is chiral
at the boron centre since the Bodipy core is asymmetrical
and the two fluorines are inequivalent. No attempt was
made to try and isolate the two chiral rotamers, but in prin-
ciple this may be possible by attaching chiral group(s) to
the hydroxys to produce diasteroisomers.[19]
19F NMR Spectroscopy
Previous findings have shown that 19F chemical shifts for
Bodipy derivatives can be highly sensitive to the local envi-
ronment.[9,16] For example, the two fluorines in BD-MQ are
inequivalent on the NMR time scale owing to restricted
rotation of the quinone group which is caused by the two
1,7-methyl groups. Inspection of a simple space-filling
model generated from the crystal structure determination
of BD-SBn reveals that the 1–3 dimethyl groups will restrict
rotation of the appended aromatic group. This notion was
in fact borne out in the room temperature 19F NMR spec-
trum for the compound which shows 14 lines. The spectrum
is adequately simulated as two chemically inequivalent flu-
orines (∆δ = 0.43 ppm) coupling (Fa–FbJ = 107 Hz) to the
quadropolar 11B nuclei (B–FJ = 35 and 34 Hz) (see Support-
ing Information). A somewhat similar looking 19F spectrum
(12 lines) is seen for BD-SHQ (Figure 2), but in this case
the difference in chemical shift between the two fluorines is
smaller (∆δ = 0.29 ppm) (see Supporting Information). To
account for these two observations the aromatic group must
undergo, at least on the NMR timescale, restricted rotation
so that each fluorine is in a different chemical environment.
In contrast to these two cases, the 19F NMR spectrum for
BD-SQ is very simple and contains a quartet with signs of
smaller secondary coupling. The spectrum is simulated as
two equivalent fluorines coupling to the 11B (I = 3/2,
80.42%) and 10B (I = 3, 19.58%) nuclei. It would appear
that the small change in molar volume (ca. 9 cm3 mol–1)[17]
in shifting from hydroquinone to quinone is enough to per-
mit rotation around the connector bond. At a basic level
Electrochemistry
The electrochemical behavior of the Bodipy derivatives
was measured by the standard cyclic voltammetry (CV)
technique in dry CH2Cl2 containing 0.2  N-tetrabutylam-
monium tetrafluoroborate (TBATFB) electrolyte at a plati-
num electrode. The redox half-wave potentials (E1/2) are
quoted relative to the ferrocenium/ferrocene couple which
is taken to be fully reversible, and peak separations (Epa
–
Epc) for other redox couples are compared to that observed
for ferrocene (60 mV as measured). Very clean looking CVs
were recorded in most of the cases for the Bodipy deriva-
tives, and there was no appreciable sign of electrode fouling
under the employed conditions. The quinone derivatives,
BD-SQ and BD-SPQ, display upon reductive scanning a
clear quasi-reversible one-electron wave at around –0.90 V
(peak separation ≈ 160 mV) that is assigned to the redox
processes associated with the quinone group (see Support-
ing Information). At a considerable more cathodic potential
the one electron reduction of the Bodipy moiety is ob-
served, which for BD-SPQ is completely irreversible. The
redox process at the Bodipy site is clearly perturbed by the
presence of the quinone radical anion. The oxidative por-
tion of the CV is dominated by a quasi-reversible wave at
around +0.75 V (peak separation ≈ 100 mV), that is readily
assigned, by comparison to literature examples,[20] to one-
electron oxidation of the Bodipy core.
The CVs. of the dihydroxy (BD-SHQ/BD-SPHQ) and
benzyl-protected derivatives (BD-SBn/BD-SPBn) again, in
the reductive segment of the CV, show the characteristic
wave associated with one-electron reduction of the Bodipy
unit. The oxidative portion contains two waves with one
being assigned to the one-electron oxidation of the Bodipy
unit and the second to oxidation of the hydroquinone or
benzyloxy group. Electrochemical data for all the com-
pounds, and that previously published for the meso deriva-
tives,[9] are collected in Table 1. In comparing the redox po-
tentials for the 2-substituted Bodipy compounds, as well as
the meso counterparts, several interesting trends are re-
vealed, including effects of adding the phenylene spacer,
change from hydroquinone to quinone and alteration in
substitution position. In attempting to rationalize the elec-
trochemical results several, often competing factors need to
be taken into account including, for example, effects from
change in the extent of π-conjugation (steric induced), alter-
ation in electron donor/acceptor ability of appended group,
Figure 2. The room temperature 19F NMR (470 MHz) spectrum
for BD-SHQ in CDCl3, and a basic molecular model showing the
proximity of the oxygen atoms to the fluorine atoms.
Eur. J. Org. Chem. 2010, 2867–2877
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
2869

A. C. Benniston, G. Copley, H. Lemmetyinen, N. V. Tkachenko
FULL PAPER
adjustment in charge distribution in the molecular frame- two methyl groups restrict any extended π-conjugation. It
work and change in spatial arrangement of the HOMO/ appears that the reduction potential for BD-MQ is the odd
LUMOs on the Bodipy core. In addition, considering that one out. The comparison of the oxidation potentials for the
quinone is a good electron acceptor a strongly-coupled po- hydroquinone group is difficult because of the irreversibility
larization in the ground-state molecule might be expected of the redox couple for the two phenylene spacer deriva-
along the line BDδ+-Qδ–, and would make oxidation more tives. However, oxidation of the hydroquinone group in BD-
difficult. This argument would also hold for the hydro- MHQ is slightly more difficult than for BD-SHQ. This is,
quinone which is more of an electron donor and the polari- as expected, the opposite of what is observed in the quinone
zation would be BDδ–-HQδ+ and make oxidation easier. reduction potentials. It is worth noting that the change
Electronic coupling would be enhanced by increased conju- from hydroquinone to quinone has very little affect on the
gation along the molecular framework. The aryl spacer Bodipy-based oxidation potentials (E1) for the 2-substituted
group could act to either decouple/couple the two subunits. Bodipy derivatives, and only affects slightly the reduction
potentials where the shift is anodic (+ 60 mV). In contrast,
alteration in oxidation potentials for the meso-substituted
Table 1. Electrochemical data for the Bodipy compounds discussed
Bodipy derivatives is significant and cathodic (–110 mV),
in the text.
and coupled to a concomitant shift to a more cathodic po-
Compd.
E1 /V[a] E2 /V[b] E3 /V[c] E4 /V[d] ∆E1 /V[e] ∆E2 /V[f]
tential (–230 mV) for the Bodipy-based reduction. These
findings suggest that the Bodipy-based LUMO and HOMO
for the meso derivatives are highly perturbed by a change
in oxidation state of the appended group, but any change
for the 2-substituted derivatives is much less pronounced.
A small shift to a lower potential (–40 mV) is observed
for the Bodipy-based oxidation when a comparison is made
between BD-SQ and BD-MQ. For the hydroquinone deriv-
atives, BD-SHQ and BD-MHQ, the change in substitution
pattern results in an anodic shift (+ 60 mV). Again, for the
directly linked derivatives, there is a noticeable cathodic
shift (–250 mV) in the Bodipy-based reduction potentials
by modification of the substitution pattern from side-on to
the meso position. This is less evident in the hydroquinone
case where in fact the shift (+ 40 mV) is to a higher re-
duction potential. What is apparent from Table 1 is that the
peak separation between redox potentials for the quinone
BD-SBn
0.61
0.64
0.76
0.67
0.75
0.66
0.82
0.76[g]
0.71
0.63
–1.70
–1.64
–1.62
–1.58
–1.56
–
–
–
–
1.10
0.97
0.53
0.58[g]
–
2.31
2.28
2.38
2.25
2.31
2.12
2.40
2.26
2.52
2.42
0.49
0.33
BD-SPBn
BD-SHQ
BD-SPHQ
BD-SQ
BD-SPQ
BD-MHQ
BD-MPHQ
BD-MQ
0.23
[h]
–0.91
0.65
–1.46[g] –0.87
–
[h]
–1.58
–1.50
–1.81
–1.79
–
–
0.61
0.56[g]
–
0.21
[h]
–0.79
–0.90
1.02
0.89
BD-MPQ
–
[a] E1/2 for one-electron oxidation of Bodipy. [b] E1/2 for one-elec-
tron reduction of Bodipy. [c] E1/2 for one-electron reduction of
quinone. [d] E1/2 for one-electron oxidation of hyroquinone/benz-
yloxy. [e] E1 – E2. [f] Modulus of E4 – E1 or E2 – E3. [g] No reverse
peak observed. [h] Not calculated because of irreversibility of redox
process.
The affect of the phenylene group is clearly seen by com- reduction and Bodipy reduction (∆E2) is much smaller for
paring oxidation potentials and, to a lesser extent, the re- BD-SQ when compared to BD-MQ. The values for ∆E2
duction potentials for the hydroquinone/quinone deriva- when comparing the oxidation potentials associated with
tives for both the series. Ignoring redox potentials obtained the hydroquinone oxidation and Bodipy oxidation are very
from CVs that are not fully reversible, then phenylene inser- similar and small. The magnitude of ∆E2 affords direct in-
tion leads to a definite shift to a less positive potential (ca. formation on the spatial location of charge on the Bodipy
–90 mV) for oxidation of the Bodipy core, and arguably a core and the quinone/hydroquinone groups. A large ∆E2
small anodic shift in the Bodipy reduction potential. It is indicates a close confinement of charge as observed for BD-
noticeable that ∆E1 (E1 – E2) which represents the HOMO– MQ, which is consistent with the reduction in ∆E2 by phen-
LUMO gap is reduced by insertion of the aryl spacer. An ylene insertion (cf. BD-MPQ).
increase in π-conjugation (raising of the HOMO) is consis-
The smaller the separation the more remote are the op-
tent with the first observation but not the second since the posite charges as seen in the case of BD-SQ. The location
LUMO would be lowered in energy and make the Bodipy of the LUMO on the distal pyrrol ring in BD-SQ is in fit-
center harder to reduce. Presumably a secondary electronic ting with this idea, so as to maximize its separation from
effect perturbs the LUMO. The quinone-based redox po- the singly-occupied molecular orbital (SOMO) on the qui-
tentials for BD-SQ, BD-SPQ and BD-MPQ are very sim- none radical anion. The asymmetry of BD-SQ facilitates
ilar and appreciably more cathodic than the redox potential this process especially if the SOMO can mix with orbitals
for BD-MQ. It is clear that no real ground-state polariza- on the proximal pyrrol ring to bring the quinone radical
tion dominates since reduction would be expected to be the anion into conjugation. In the symmetrical, BD-MQ, the
more difficult for the directly-linked quinone derivatives. SOMO cannot come into extended conjugation because of
Additionally, any effect from extended conjugation of the steric constraints. Previously published molecular orbital
quinone with the Bodipy core seems inconsistent with the calculations performed on Bodipy derivatives[21] reveal that
lack of difference between BD-SQ and BD-SPQ. The sim- the HOMO state is mainly localized on the pyrrol units,
ilar redox potentials for BD-SQ and BD-MPQ are also in with the LUMO spread over the backbone but with con-
line with this argument, since in the latter derivative the siderable electron density at the central meso atom. This
2870
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 2867–2877

Comparison of Substituted Donor-Acceptor Bodipy Dyads
latter finding is consistent with the fact that addition of an Supporting Information). It has been suggested that in-
electron to the LUMO is harder for BD-MQ because of the ternal conversion in Bodipy dyes is affected by the flexibil-
proximity of the SOMO on the quinone radical anion.
ity, and extent of distortion from planarity, of the dipyrro-
methene backbone.[24] This is possibly one contribution
coupled to the reduction of kNR as predicted by the energy-
gap law.[25]
Absorption and Fluorescence Spectroscopy
The removal of the protecting groups to produce the hy-
Absorption and fluorescence spectra were recorded for droquinone derivatives BD-SHQ and BD-SPHQ does not
the 2-substituted Bodipy compounds in dilute CH3CN bring about any major alterations in the absorption spectra
solution at room temperature. Data are collected in Table for the two compounds. Fluorescence is still readily ob-
S2 (see Supporting Information) for all the compounds served from BD-SPHQ with an emission profile that is very
along with that for the meso derivatives. To aid in under- similar in appearance to BD-SPBn; the magnitude of the
standing the basic photophysical properties of the series, Stokes’ shift is also very similar. The values for φflu and τs
the starting materials BD-SBn and BD-SPBn were studied, are, however, greatly reduced indicating the introduction of
and the findings discussed firstly for comparison purposes. an additional first-excited singlet state deactivation process.
The absorption spectrum for the benzyl protected com- Fluorescence quenching of the first-excited singlet state for
pound BD-SBn is dominated by the typical Bodipy-like ab- BD-SHQ is more pronounced, and τs as measured by up-
sorption profile at λabs = 525 nm, corresponding to the S0– conversion spectroscopy is in the sub-picosecond region. It
S1 electronic transition.[1] The less intense band seen at is very noticeable that τs for BD-SPHQ is reduced signifi-
around 382 nm is associated with the S0–S2 transition. The cantly (≈ 25 fold) when compared to its complement meso
longer wavelength absorption profile for BD-SBn is some- derivative BD-MPHQ.
what broader and does not show the pronounced higher-
The most interesting findings relate to the major change
energy shoulder observed for the counterpart meso-substi- in the absorption profile for the quinone derivative BD-SQ
tuted derivative. This finding is in fitting with the overall (Figure 3). The absorption profile in CH3CN is noticeably
reduction in symmetry of the molecule by the 2-substitution broad (λABS = 519 nm) with a very obvious lower-energy
which perturbs the S0 and S1 states. A similar finding has tail stretching off to around 700 nm. A Beer–Lambert plot
been reported by Burgess et al.[22] for 2-substituted Bodipy- (see Supporting Information) shows no deviation from
anthracene cassettes. Fluorescence is readily detected for linearity over the concentration range 2ϫ10–5
 to
BD-SBn and is centered at λflu = 549 nm, corresponding to 6.0ϫ10–6 , which rules out broadening caused by dye ag-
a measured quantum yield of fluorescence (φflu) of 0.045. gregation or intermolecular exciton coupling.[26] Absorp-
The fluorescence lifetime (τs) as determined by the single- tion spectra recorded in a range of solvents show several
photon-counting method is 300 ps. The radiative rate con- interesting effects. In toluene, for example, the λABS is red
stant (kRAD = φflu/τs) of 1.5ϫ108 s–1 is typical for Bodipy- shifted by ca. 6 nm (cf. CH3CN), and the absorption profile
based dyes and similar to the value calculated using the is considerably broader. Changing the solvent from toluene
Strickler–Berg expression.[23] The Stokes’ shift of 833 cm–1 to MeTHF and finally formamide results in a blue-shift in
is somewhat larger than the value found for the meso deriv- λABS and a narrowing of the absorption profile (Figure 3).
ative (499 cm–1), which implies a more substantial alteration This blue-shift is also observed for many other Bodipy de-
in molecular geometry takes place following the S0–S1 elec- rivatives[27] and in specific cases can be related to the sol-
tronic transition and relaxation. It is more noticeable that vent polarizability function [f(n2) = (n2–1)/(2n2+1)]. To see
non-radiative deactivation of the BD-SBn S1 state is far if this was the case here, absorption spectra were recorded
more efficient than in BD-MBn, since the non-radiative rate for BD-SQ in solvents of different polarizability for which
constant (kNR) is two orders of magnitude greater. The ba- it was soluble. Unfortunately, there is no clear connection
sic photophysical properties of the Bodipy group are notice- between λABS and f(n2) for the limited range of solvents
ably affected by insertion of the phenylene group (cf. BD- used. The assignment of the lower-energy tail in the absorp-
SPBn). The absorption profile for BD-SPBn is similar in tion spectrum to an intramolecular Bodipy-to-quinone op-
overall appearance, but slightly broadened and red-shifted, tical charge-transfer (CT) band can be likely ruled out on
when compared to BD-SBn. A slight broadening of the the grounds that in polar solvents (i.e., CH3CN), the band
fluorescence profile is also observed, and more importantly would be expected to be more pronounced at lower energy,
is accompanied by a significant increase in the Stokes’ shift. and shift to higher energy as the polarity of the solvent
These findings support increased conjugation of the phenyl- decreases; this is opposite to what is actually observed. In-
ene group with the Bodipy unit in the S1 excited state. Both sertion of the phenylene group between the Bodipy and quin-
the φflu and τs values for BD-SPBn are appreciably in- one removes the severe absorption band broadening, and
creased compared to BD-SBn. Clearly, the non-radiative de- in fact the spectrum for BD-SPQ (Figure 3) is very similar
cay process (Table S2, see Supporting Information) is cur- to that for BD-SPHQ. Severe perturbation of the electronic
tailed by the insertion of the phenylene group. Reductive transitions within the 2-substituted Bodipy derivatives is
electron transfer quenching of the first-excited singlet state seen only for the quinone group attached directly at the
by the dibenzyloxyphenylene group in BD-SBn and BD- 2-position. This is corroborated by spectroelectrochemical
SPBn can be ruled out on thermodynamic grounds (see experiments performed on a solution of BD-SQ in CH3CN
Eur. J. Org. Chem. 2010, 2867–2877
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
2871

A. C. Benniston, G. Copley, H. Lemmetyinen, N. V. Tkachenko
FULL PAPER
containing 0.2  TBATFB in a standard OTTLE cell. Con-
trolled coulombic reduction of the solution gives way to
appearance of an absorption band that matches almost ex-
actly with the hydroquinone derivative. Furthermore, the
process is fully reversible since controlled oxidation of the
solution affords the original spectrum. That the quinone
unit in BD-SQ can rotate freely may have some bearing on
the absorption spectrum profile. Unhindered rotation
means that a wider envelope of geometries can be sampled,
especially those which bring the Bodipy and quinone into
partial conjugation. This is not so possible in the hydro-
quinone/benzyloxy versions because of steric constraints as
discussed previously.
Figure 4. Differential transient absorption records measured at dif-
ferent time delays (1.2 ps, 1.6 ps, 7.8 ps, 19.7 ps) following exci-
tation of BD-SHQ in CH3CN with a 70 fs laser pulse delivered
at 420 nm. Inserts show decay kinetics measured at two different
wavelengths and the least-squares fit (solid line) to single ex-
ponential decays.
Figure 3. Normalized absorption spectra for BD-SQ in CH3CN
(–), toluene (– – ), Me/THF (···), and formamide (–·–) and spectrum
for BD-SPQ in CH3CN (–··).
Differential transient absorption profiles collected for
BD-MHQ in CH3CN, following excitation with a 70 fs laser
pulse (see Supporting Information), are overall similar in
appearance to those shown in Figure 4. A strong bleach is
Femtosecond Time-Resolved Spectroscopy
Up-conversion fluorescence lifetime measurements car- seen at around 525 nm after 1.2 ps, which is accompanied
ried out on BD-SHQ and BD-MHQ in CH3CN affords de- by an increase in bleach to the lower-energy region at
cay profiles that cannot be fitted adequately to a single around 565 nm over 400 fs; this change is very similar to
mono-exponential.[28] The fluorescence decay curve for BD- what is observed for BD-SHQ. Whereas the bleach at
SHQ is best fitted as a tri-exponential τ1 = 0.8 ps (32%), 525 nm continues to grow-in slightly over a further 20 ps or
τ2 = 3.4 ps (43%), τ3 = 11.6 ps (25%). In comparison, the so, the change in optical density at 565 nm remains invari-
fluorescence decay for BD-MHQ is suitably fitted to a bi- able over this time scale (see Supporting Information). Op-
exponential [τ1 = 34 ps (35%), τ2 = 273 ps (65%)], and there tical changes at 525 nm fit well to two exponentials corre-
is no improvement to the fit by using a tri-exponential func- sponding to a grown-in over 23 ps and a slower decay of
tion (see Supporting Information).
289 ps. There is no sign of a long-lived transient in the spec-
Differential transient absorption records collected at four tral records.
different time delays following excitation of BD-SHQ in
Based on thermodynamic considerations (see Supporting
CH3CN with a 70 fs laser pulse delivered at 420 nm are Information) electron transfer from the hydroquinone to
illustrated in Figure 4. The transient absorption profile af- the excited Bodipy S1 state for derivatives, BD-MHQ and
ter 1.2 ps shows a clear bleach in the 565 nm region, which BD-SHQ, is favourable. Despite this, the transient spectral
is associated with the ground-state S0–S1 electronic transi- records show no clear features at around 580 nm being at-
tion. A minor modification to the transient profile is seen tributable to the Bodipy-based radical anion.[30] From this
over some 400 fs which is assigned to stimulated emission it is inferred that either once formed the charge separated
and/or is associated with relaxation of the S1 state.[29] The state (CSS) collapses back to the ground state by ultrafast
transient absorption profiles over some 20 ps remain rela- charge recombination [despite the large exogenicity for the
tively constant in shape, and at long time delays (≈ 1 ns) process (∆GCR ≈ –2.2 eV)], or an additional excited S1 deac-
there is no indication in the absorption records of a long- tivation process associated with the hydroxy groups is intro-
lived transient. Decay kinetics measured in the region asso- duced and the CSS is bypassed. Up-conversion fluorescence
ciated with the strong bleach (526 nm) fit to a single ex- data for BD-SHQ is certainly consistent with fast processes
ponential corresponding to a lifetime of 13 ps. A similar fit evolving from the excited S1 state; the 800 fs contribution
at 565 nm affords a decay lifetime of 2.5 ps. For both cases is basically on a timescale in line with vibrational relax-
the decay curves return to the pre-pulse baseline, which is ation. The slower components (3.4 ps and 11.6 ps) from the
consistent with the lack of existence of a long-lived tran- up-conversion fluorescence experiments are similar to the
sient.
two lifetimes (2.5 ps and 13 ps) as measured by transient
2872
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 2867–2877

Comparison of Substituted Donor-Acceptor Bodipy Dyads
absorption spectroscopy. A basic model to explain these fin- from proton-coupled electron transfer,[40] but more struc-
dings is the first lifetime represents the timescale required ture-sensitive ultrafast time-resolved spectroscopy (e.g., IR,
to create a suitable geometry, via structural alteration of the Raman) is needed to really address this issue properly.
vibrationally cooled S1 state, from which to facilitate elec-
The excited state quenching of the Bodipy center for the
tron transfer and produce the CSS. The second lifetime quinone derivatives is subject to a separate publication[41]
component symbolizes the timescale for charge separation, since solvent plays a very significant role and a more de-
with charge recombination occurring with a rate constant tailed analysis is required. In polar solvents time-resolved
(kCR) greater than 8ϫ1010 s–1. That the structural change records are consistent with the formation of the Bodipy
required to assist electron transfer within BD-MHQ is no- radical cation following photoinduced electron transfer
ticeably slower (cf. BD-SHQ), is at least consistent with the from the first-excited singlet state to the adjacent quinone
two compounds geometrical differences. The hydroquinone group. This is followed by rapid charge recombination. The
group in BD-MHQ is almost orthogonal to the Bodipy behavior for the phenylene spacer derivatives in non-polar
plane because of steric constraints imposed by the two solvents is a little more complex requiring the need to in-
flanking methyl groups; this restraint is more relaxed in voke an exciplex as the quencher state.
BD-SHQ. A simple way to promote rapid electron transfer
is to bring into partial π-conjugation the hydroquinone unit
Conclusions
and the dipyrromethene of the Bodipy. For this to occur,
especially for BD-MHQ, the two methyl groups must dis-
tort out of the Bodipy plane to allow the hydroquinone
moiety to rotate more freely. High-level computational
studies on excited-state structure dynamics for the two de-
rivatives may be able to address this issue, but are beyond
our calculation capabilities presently. This type of “gated”
electron transfer[31] has been shown to operate in donor-
bridge-acceptor molecules comprising a tetracene donor,
pyromellitimide acceptor and oligo-p-phenylenevinylenes
bridges.[32] For such assemblies the charge separation pro-
cess is controlled by the torsional motion between the tetra-
cene and the first phenyl group of the bridge. Very recent
work by Vauthey et al. has also noted that charge separa-
tion dynamics in a Bodipy-dinitrotriptycene derivative are
controlled by the reorientational motion of the donor rela-
tive to the acceptor.[29]
The one question that remains unanswered if our analy-
sis is correct is why is charge recombination so fast? Gen-
erally, highly exogenic electron transfer as described above
is associated with the so-called Marcus inverted region
(–∆GCR Ͼ λ), and one upshot is that CSS formation is gen-
erally faster than its decay.[33] The full effect of the Marcus
inverted region can be obscured by secondary factors such
as triplet formation,[34] quantum mechanical phenome-
non[35] and nuclear tunneling,[36] however. Evidently, triplet
formation at the Bodipy center can be ruled out, despite its
known low-energy (ca. 12,800 cm–1),[37] since time-resolved
spectral records show no sign of a long-lived transient.
High frequency vibrational modes play a critical role in re-
turn electron transfer in the inverted region, as seen for
strongly interacting zinc(II)-porphyrin-imide dyads de-
scribed by Osuka and co-workers.[38] The observed rate of
charge recombination is in the order of 1011 s–1 for –∆GCR
≈ 2 eV, albeit in aprotic DMF, which is at least in the correct
time-scale required to rationalize our results. An ideal can-
didate for the high-energy vibration that couples to return
electron transfer is the OH stretching vibration (Rω
≈ 3500 cm–1, 0.43 eV) for the proximal hydroquinone group;
two quanta being enough to reduce significantly the exer-
gonicity for charge recombination.[39] Unfortunately, it is
A series of side-on Bodipy-based dyads have been suc-
cessfully prepared incorporating either hydroquinone or
quinone groups. The dyads represent the complement series
to the previously published compounds where the meso po-
sition was the attachment site.[9] There are some noticeable
differences in the properties of the compounds when the
two series are compared, especially in their redox chemistry
and photophysical properties. In particular, it is immedi-
ately apparent that the first-excited state deactivation by
electron transfer for BD-SHQ is much faster (ca. 20 fold)
when compared to the meso derivative BD-MHQ. Similar
differences have been observed in energy transfer involving
Bodipy-anthracene cassettes, where the substitution pattern
is found to be crucial.[22] The excited state lifetime of the
Bodipy is not dramatically improved by insertion of a phen-
ylene group between the Bodipy and hydroquinone center
for the 2-substituted derivative. In comparison, the change
in lifetime for the meso compound is far more dramatic by
addition of the phenylene unit. This work has demonstrated
that the substitution position on the Bodipy group is criti-
cal, especially in the design of molecular probes where an
on/off switching in fluorescence signal is paramount.
Experimental Section
Materials: Bulk chemicals were purchased as the highest purity
possible from Aldrich Chemical Co. and used as received unless
otherwise stated. Tetra-n-butylammonium tetrafluoroborate
(TBATFB) purchased from Fluka was recrystallized several times
from methanol and dried thoroughly under vacuum before being
stored in a desiccator. Standard solvents were dried by literature
methods before being distilled and stored under nitrogen over 4 Å
molecular sieves. Spectroscopic grade solvents were used in all fast
kinetic experiments and fluorescence/absorption spectroscopy mea-
surements.
Instrumentation: 1H and 13C NMR spectra were recorded with
either Bruker AVANCE 300 MHz, JEOL 400 MHz, or JEOL
Lambda 500 MHz spectrometers. 11B and 19F NMR spectra were
1
recorded using the 400 MHz spectrometer. Chemical shifts for H
and 13C NMR spectra are referenced relative to the residual proti-
not practical at this stage to rule out possible contributions ated solvent. The 11B NMR chemical shift is referenced relative to
Eur. J. Org. Chem. 2010, 2867–2877
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
2873

A. C. Benniston, G. Copley, H. Lemmetyinen, N. V. Tkachenko
FULL PAPER
BF3·Et2O (δ = 0 ppm), and the 19F NMR chemical shift is given
relative to CFCl3 (δ = 0 ppm). Routine mass spectra and elemental
analyses were obtained using in-house facilities. MALDI mass
spectra were recorded at the EPSRC-sponsored Mass Spectrometry
Service at Swansea. Absorption spectra were recorded using a Hita-
chi U3310 spectrophotometer and corrected fluorescence spectra
were recorded using a Lambda Advanced F 4500 spectrometer. Un-
corrected melting points were measured using a Stuart SMP11 ap-
paratus and typically carried out twice to check for consistency in
the readings.
monium chloride (30 mL), followed by removal or the solvent un-
der reduced pressure. Ethyl acetate was used to extract the product,
before washing with water (100 mL) and brine (100 mL). The or-
ganic layer was then dried (MgSO4) and the solvent was removed
under reduced pressure, yielding the crude product which was chro-
matographed on silica gel (petroleum ether/ethyl acetate, 3:2). The
final product was isolated as a white solid (1.3 g, 83%); m.p. 113–
115 °C. 1H NMR (CDCl3, 300 MHz): δ = 7.55 (d, J = 3.1 Hz, 1
H), 7.39 (m, 10 H), 7.05 (dd, J = 8.9, JЈ = 3.1 Hz, 1 H), 6.92 (d, J
= 8.9 Hz, 1 H), 6.49 (br. s, 2 H), 5.10 (s, 2 H), 5.06 (s, 2 H) ppm.
13C NMR (CDCl3, 75 MHz): δ = 158.82, 153.89, 137.81, 136.73,
129.24, 128.83, 128.16, 128.05, 127.83, 122.95, 120.06, 113.33,
Cyclic voltammetry experiments were performed using a fully auto-
mated HCH Instruments Electrochemical Analyzer and a three-
electrode set-up consisting of a platinum working electrode, a plati-
num wire counter electrode and a silver wire reference electrode.
Ferrocene was used an internal standard. All studies were per-
formed in deoxygenated CH2Cl2 containing TBATFB (0.2 ) as
background electrolyte. The solute concentrations were typically
0.1 m. Redox potentials were reproducible to within Ϯ15 mV.
Spectroelectrochemistry experiments were performed in CH2Cl2
(TBATFB, 0.2 ) solutions and an Omini-cell specac OTTLE cell
placed inside a Perkin–Elmer Lambda 35 UV/Vis spectrometer. In
a typical experiment the working electrode was stepped to a fixed
potential for several minutes and the absorption spectrum re-
corded. The reversibility of an electrochemical process was checked
by carrying out the same procedure but with a potential set with
an opposite sign. Several cycles were performed to check for com-
pound degradation.
71.99, 71.26 ppm. IR (neat): ν = 3494, 3389 (O–H), 2875 (C–H)
˜
cm–1.
Preparation of 2,5-Bis(benzyloxy)-4Ј-bromobiphenyl: To a stirred
solution of 2-(4-bromophenyl)cyclohexa-2,5-diene-1,4-diol (1.55 g,
5.85 mmol, 1 equiv.), K2CO3 (1.13 g, 8.19 mmol, 1.4 equiv.) in ace-
tone (25 mL) was added benzyl bromide (1.67 mL, 14 mmol,
2.4 equiv.). The mixture was heated at reflux until TLC showed
complete consumption of the starting material. At this point, the
reaction was allowed to cool and the K2CO3 was removed via fil-
tration. The solvent was removed under reduced pressure, yielding
the crude product which was chromatographed on silica gel (petro-
leum ether/CH2Cl2, 3:1) affording a white solid (2.39 g, 92%); m.p.
1
96–98 °C. H NMR (CDCl3, 300 MHz): δ = 7.31 (m, 14 H), 6.882
(m, 2 H), 6.81 (dd, J = 8.8, JЈ = 2.9 Hz, 1 H), 4.96 (s, 2 H), 4.89
(s, 2 H) ppm. 13C NMR (CDCl3, 75 MHz): δ = 154.11, 150.58,
137.75, 137.70, 137.65, 132.07, 131.53, 131.42, 128.86, 128.72,
128.22, 128.02, 127.76, 127.49, 121.60, 118.14, 116.26, 115.39,
72.41, 71.37 ppm. EI-MS: m/z calcd. for C26H21BrO2 (445), found
445. C26H21BrO2 (445): calcd. C 70.12, H 4.75; found C 70.19, H
4.67.
All luminescence measurements were made using optically dilute
solutions and were corrected for spectral imperfections of the in-
strument by reference to a standard lamp. Luminescence quantum
yields were measured relative to a standard Bodipy in acetonitrile
using optically matched solutions and were measured twice for con-
sistency in calculated values. Radiative rate constants were calcu-
lated from absorption and normalized fluorescence spectra using
the Strickler–Berg equation[23] and compared to those calculated
from lifetime and quantum yield measurements.
Preparation of 2b: To a stirred solution of 2,5-bis(benzyloxy)-4Ј-
bromobiphenyl (2.1 g, 4.72 mmol, 1 equiv.) in THF (100 mL) at
–78 °C was added tBuLi 1.7  (3.88 mL, 6.6 mmol, 1.4 equiv.)
dropwise, immediately followed by the dropwise addition of trieth-
ylborate (2.41 mL, 14 mmol, 3 equiv.) at –78 °C. This temperature
was maintained for 10 min, after which the solution was warmed
to room temperature before being stirred for a further 2 h. The
reaction was cooled to 0 °C and quenched with a saturated solution
of ammonium chloride (30 mL), followed by removal of the reac-
tion solvent under reduced pressure. Ethyl acetate was used to ex-
tract the product. The organic phase was washed with water
(100 mL) and brine (100 mL). The organic layer was then dried
(MgSO4), and the solvent was removed under reduced pressure,
yielding the crude product. This latter material was chromato-
graphed on silica gel (petroleum ether/ethyl acetate, 1:1) to afford
the product as a white solid (1.64 g, 85%); m.p. 70–72 °C. 1H NMR
(CDCl3, 300 MHz): δ = 8.24 (d, J = 8 Hz, 2 H), 7.66 (d, J = 8 Hz,
2 H), 7.24 (m, 10 H), 7.01 (d, J = 2.9 Hz, 1 H), 6.92 (d, J = 8.9 Hz,
1 H), 6.85 (dd, J = 8.9, JЈ = 2.9 Hz, 1 H), 5.00 (s, 2 H), 4.94 (s, 2
H) ppm. 13C NMR (CDCl3, 75 MHz): δ = 154.14, 150.81, 143.26,
137.82, 137.78, 135.70, 133.19, 129.54, 128.87, 128.71, 128.20,
127.94, 127.80, 127.50, 118.37, 116.48, 116.45, 115.51 ppm (note:
two carbon resonances missing due to accidental equivalence). EI-
MS: m/z calcd. for C26H23BO4 (410), found 410. C26H23BO4 (410):
calcd. C 76.12, H 5.65; found C 76.18, H 5.59.
Synthesis: All preparations were carried out, unless otherwise
stated, under a dry N2 atmosphere in thoroughly oven-dried glass-
ware.
Preparation of 1: To a stirred solution of 4-bromo-3,5-dimethyl-
1H-pyrrole-2-carbaldehyde (1 g, 4.95 mmol, 1 equiv.) and 4-(di-
methylamino)pyridine (200 mg) in THF (50 mL) at room tempera-
ture was added di-tert-butyl dicarbonate (1.4 g, 6.43 mmol,
1.3 equiv.) in a single portion. The reaction was heated to reflux
until TLC showed complete consumption of the starting material.
The THF was removed, and the residue was taken up into diethyl
ether, washed with 1  aqueous NaHSO4 (3ϫ100 mL), brine
(3ϫ100 mL), and saturated NaHCO3 before being dried (Na2SO4).
The solvent was removed under reduced pressure, yielding the
crude product. The latter product was chromatographed on a silica
gel (CH2Cl2/petroleum ether, 1:3), yielding the product as a pale
brown solid (1.36 g, 91%); m.p. 82–84 °C. 1H NMR (CDCl3,
300 MHz): δ = 9.96 (s, 1 H), 2.45 (s, 3 H), 2.32 (s, 3 H), 1.61 (s, 9
H) ppm.
Preparation 2a: To a stirred solution of (2-bromo-1,4-phenylene)-
bis(oxy)bis(methylene)dibenzene (1.8 g, 4.87 mmol, 1 equiv.) in
THF (50 mL) at –78 °C was added dropwise tBuLi 1.7  (4.01 mL,
6.82 mmol, 1.4 equiv.), immediately followed by addition of triethyl
borate (2.48 mL, 14.6 mmol, 3 equiv.) dropwise at –78 °C. This
temperature was maintained for 10 min, after which the solution
was warmed to room temperature and stirred for 2 h. The reaction
was cooled to 0 °C and quenched with a saturated solution of am-
Preparation of 3a: Compound 1 (1 g, 3.3 mmol, 1 equiv.), and 2a
(2.2 g, 6.6 mmol, 2 equiv.) were dissolved in DMF (100 mL), and
the resultant solution was purged with nitrogen for 1 h. To this
solution was added a nitrogen-purged aqueous solution of 2 
Na2CO3 (6.6 mL, 13.2 mmol, 4 equiv.), followed by Pd(PPh3)4
2874
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 2867–2877

Comparison of Substituted Donor-Acceptor Bodipy Dyads
(0.38 g, 10 mol-%). The reaction mixture was refluxed overnight.
The resulting reaction mixture was allowed to cool to room tem-
perature, and the product was extracted into CH2Cl2 (3ϫ100 mL).
The organic layer was washed with brine (3ϫ50 mL) and dried
(MgSO4). The solvent was removed under reduced pressure, to
yield a dark brown residue. This material was chromatographed on
silica gel (ethyl acetate/petroleum ether, 1:1), yielding a light brown
solid (1.01 g, 75% yield); m.p. 148–150 °C. 1H NMR (300 MHz,
CDCl3): δ = 10.26 (br. s, 1 H), 9.43 (s, 1 H), 7.00 (m, 10 H), 6.87
(d, J = 8.8 Hz, 1 H), 6.80 (dd, J = 8.8, JЈ = 2.8 Hz, 1 H), 6.68 (d,
J = 2.8 Hz, 1 H), 4.94 (s, 2 H), 4.82 (s, 2 H), 2.08 (s, 6 H) ppm.
13C NMR (75 MHz, CDCl3): δ = 176.52, 153.75, 151.81, 137.83,
136.85, 128.88, 128.69, 128.19, 127.97, 127.75, 127.46, 125.82,
122.25, 119.62, 116.44, 115.32, 72.53, 71.31, 12.62, 9.75 ppm. EI-
MS: m/z calcd. for C27H25NO3 (411), found 411. C27H25NO3 (411):
C 78.81, H 6.12 (6.19), N 3.40 (3.30).
NMR (CDCl3, 300 MHz): δ = 7.23 (m, 10 H), 6.81 (m, 3 H), 6.67
(s, 1 H), 4.94 (s, 2 H), 4.84 (s, 2 H), 2.44 (s, 3 H), 2.30 (m, 5 H),
2.09 (s, 3 H), 1.98 (s, 3 H), 1.04 (t, J = 7.5 Hz, 3 H) ppm. 13C
NMR (CDCl3, 75 MHz): δ = 156.28, 155.35, 153.73, 151.67,
138.25, 137.77, 137.42, 133.43, 132.88, 132.54, 128.85, 128.74,
128.57, 128.17, 127.94, 127.73, 127.50, 127.52, 119.51, 119.16,
116.45, 115.75, 72.53, 71.33, 17.64, 14.66, 13.69, 12.90, 10.51,
9.59 ppm (note: one carbon resonance is missing due to accidental
equivalence). 11B NMR (CDCl3, 160 MHz): δ (ppm) = –0.00 (t, Jav
= 33.09 Hz). 19F NMR (CDCl3, 470 MHz): δ = –146.3 (m) ppm.
IR (neat): ν = 2960 (C–H), 1598, 1496 (C=C, C=N), 1186 (B–
˜
F) cm–1. EI-MS: m/z calcd. for C35H35BF2N2O2 (564), found 564.
C35H35BF2N2O2 (564): calcd. C 74.47, H 6.25, N 4.96; found C
74.28, H 6.22, N 4.92.
Preparation of BD-SHQ: To a thoroughly degassed solution of BD-
SBn (0.3 g, 0.51 mmol, 1 equiv.) and 1,4-cyclohexadiene (1.98 mL,
21 mmol, 40 equiv.) in ethanol was added Pd(OH)2/C (50 mg). The
reaction was refluxed until TLC showed complete consumption of
the starting material. The mixture was filtered through a pad of
celite and concentrated under vacuum. The residue was chromato-
graphed on silica gel (chloroform), yielding the final product as a
red solid (0.2 g, 93%); m.p. 134–136 °C. 1H NMR (CDCl3,
Preparation of 3b: The same procedure are described above was
used. Compound
1 (0.4 g, 1.3 mmol, 1 equiv.), 2b (1.08 g,
2.63 mmol, 2 equiv.), DMF (80 mL), 2  Na2CO3 (5.5 mL,
11 mmol, 4 equiv.) and Pd(PPh3)4 (0.15 g, 10 mol-%). Purification:
silica gel, ethyl acetate/petroleum ether (1:2). Brown solid (0.43 g,
67%); m.p. 172–174 °C. 1H NMR (300 MHz, CDCl3): δ = 9.61 (br.
s, 1 H), 9.52 (s, 1 H), 7.56 (d, J = 8.2 Hz, 2 H), 7.28 (m, 12 H), 300 MHz): δ = 7.06 (s, 1 H), 6.78 (d, J = 8.6 Hz, 1 H), 6.73 (dd, J
7.01 (d, J = 2.9 Hz, 1 H), 6.915 (d, J = 8.9 Hz, 1 H), 6.83 (dd, J =
8.9, JЈ = 2.9 Hz, 1 H), 4.99 (s, 2 H), 4.93 (s, 2 H), 2.29 (s, 3 H),
= 8.6, JЈ = 2.6 Hz, 1 H), 6.53 (d, J = 2.6 Hz, 1 H), 5.06 (br. s, 1
H), 4.76 (br. s, 1 H), 2.54 (s, 3 H), 2.44 (m, 5 H), 2.19 (s, 3 H), 2.09
2.28 (s, 3 H) ppm. 13C NMR (75 MHz, CDCl3): δ = 176.77, 154.21, (s, 3 H), 1.08 (t, J = 7.5 Hz, 3 H) ppm. 13C NMR (CDCl3,
150.86, 137.91, 137.13, 133.29, 133.17, 129.92, 129.72, 128.84,
128.61, 128.17, 127.90, 127.77, 127.47, 118.23, 116.50, 115.12,
75 MHz): δ = 159.16, 153.35, 149.54, 148.25, 138.95, 137.85,
134.41, 133.76, 132.48, 124.94, 121.03, 119.76, 117.98, 116.81,
72.55, 71.37, 12.57, 9.66 ppm. EI-MS: m/z calcd. for C33H29NO3 116.54, 17.61, 14.47, 13.12, 13.11, 10.18, 9.61 ppm. 11B NMR
(487), found 487. C33H29NO3 (487): calcd. C 81.29, H 5.99, N 2.87;
found C 81.17, H 6.01, N 2.80.
(CDCl3, 160 MHz): δ = –0.00 (t, Jav = 34.4 Hz) ppm. 19F NMR
(CDCl , 470 MHz): δ = –145.63 (m, 2 F) ppm. IR (neat): ν = 3497
˜
3
(OH), 2965 (C–H), 1598, 1462 (C=C, C=N), 1186 (B–F) cm–1. EI-
MS: m/z calcd. for C21H23BF2N2O2 (384), found 384.
C21H23BF2N2O2 (384): calcd. C 65.65, H 6.03, N 7.29; found C
65.59, H 6.14, N 7.17.
Preparation of BD-SPBn: To a stirred solution of 3-ethyl-2,4-di-
methylpyrrole (0.16 mL, 1.2 mmol, 1 equiv.) and 3b (0.5 g,
1.2 mmol, 1 equiv.) in CH2Cl2 (100 mL) was added TFA (2 drops).
The reaction was allowed to stir at room temperature until TLC
showed complete consumption of the aldehyde. N,N-Diisopropyl-
ethylamine (1.164 mL, 6.9 mmol, 5.7 equiv.) and BF3·Et2O
(1.24 mL, 9.7 mmol, 8 equiv.) were added and the reaction was left
to stir for 6 h at room temperature. The mixture was washed with
water (3ϫ100 mL) and brine (3ϫ100 mL). The separated organic
fractions were dried (MgSO4), filtered and evaporated to yield a
black/dark violet residue with a green tint. The residue was chro-
Preparation of BD-SPHQ: The same procedure to above was used.
BD-SPBn (0.36 g, 0.56 mmol, 1 equiv.), 1,4-cyclohexadiene
(2.11 mL, 22 mmol, 40 equiv.), ethanol (50 mL), Pd(OH)2/C
(50 mg). Purification: silica gel, petroleum ether/ethyl acetate (1:1)
eluant. Red solid (0.25 g, 98%); m.p. 150–152 °C. 1H NMR
(MeOD, 300 MHz): δ = 7.59 (d, J = 8.2 Hz, 2 H), 7.20 (m, 3 H),
6.73 (m, 3 H), 2.47 (s, 3 H), 2.46 (s, 3 H), 2.33 (q, J = 7.5 Hz, 2
H), 2.10 (s, 3 H), 2.08 (s, 3 H), 1.01 (t, J = 7.5 Hz, 3 H) ppm. 13C
matographed on silica gel (toluene), yielding the product as a bright
1
red solid (0.33 g, 42% yield); m.p. 199–201 °C. H NMR (CDCl3, NMR (MeOD, 75 MHz): δ = 158.08, 154.46, 152.03, 148.78,
300 MHz): δ = 7.55 (d, J = 8.2 Hz, 2 H), 7.26 (m, 12 H), 7.00 (m, 140.05, 139.32, 138.40, 135.20, 135.18, 134.24, 134.08, 133.98,
2 H), 6.90 (d, J = 8.9 Hz, 1 H), 6.81 (dd, J = 8.9, JЈ = 2.9 Hz, 1 132.72, 130.89, 130.80, 121.54, 118.65, 118.47, 116.73, 18.47, 15.15,
H), 4.98 (s, 2 H), 4.91 (s, 2 H), 2.49 (s, 3 H), 2.46 (s, 3 H), 2.32 (q,
J = 7.5 Hz, 2 H), 2.17 (s, 3 H), 2.11 (s, 3 H), 1.00 (t, J = 7.5 Hz, 3
H) ppm. 13C NMR (CDCl3, 75 MHz): δ = 157.18, 154.19, 153.98,
150.85, 137.89, 137.82, 137.41, 136.53, 133.80, 133.14, 133.09,
132.93, 132.78, 131.64, 129.94, 129.64, 128.87, 128.65, 128.20,
14.00, 13.30, 10.67, 9.83 ppm. 11B NMR (MeOD, 160 MHz): δ
(ppm) = –0.057 (t, Jav = 31.99 Hz). 19F NMR (MeOD, 470 MHz):
δ = –145.550 (q, Jav = 28.15 Hz) ppm. IR (neat): ν = 3525 (OH),
˜
2967 (C–H), 1539, 1455 (C=C, C=N), 1184 (B–F) cm–1. EI-MS:
m/z calcd. for C27H27BF2N2O2 (460), found 460. C27H27BF2N2O2
127.95, 127.80, 127.48, 119.62, 118.16, 116.48, 115.16, 72.54, 71.34, (460): calcd. C 70.45, H 5.91, N 6.09; found C 70.54, H 5.90, N
17.66, 14.65, 13.63, 12.99, 10.46, 9.65 ppm. IR (neat): ν = 2960 (C–
6.00.
˜
H), 1599, 1453 (C=C, C=N), 1189 (B–F) cm–1. EI-MS: m/z calcd.
for C41H39BF2N2O2 (640), found 640. C41H39BF2N2O2 (640):
calcd. C 76.88, H 6.14, N 4.37; found C 76.82, H 6.17, N 4.31.
Preparation of BD-SQ: To a stirred solution of the BD-SHQ (0.9 g,
2.34 mmol, 1 equiv.) in THF (50 mL) was added DDQ (0.79 g,
3.51 mmol, 1.5 equiv.). On addition of the DDQ, the reaction mix-
ture changed from red to dark purple. When TLC showed complete
consumption of the starting material, the solvent was removed and
the residue was taken into CH2Cl2 (100 mL) and washed with water
(3ϫ100 mL). The combined organic fractions were dried (MgSO4).
The residue was chromatographed on silica gel (petroleum ether/
chloroform, 1:1), yielding the crude product as a purple solid
Preparation of BD-SBn: The same procedure as described above
was used. 3-Ethyl-2,4-dimethylpyrrole (0.49 mL, 3.6 mmol,
1 equiv.), 3a (1.37 g, 3.3 mmol, 1 equiv.), CH2Cl2 (250 mL), TFA (2
drops), N,N-diisopropylethylamine (3.18 mL, 19 mmol, 5.7 equiv.)
and BF3·Et2O (3.78 mL, 26 mmol, 8 equiv.). Purification: silica gel,
toluene eluant. Bright red solid (0.82 g, 44%); m.p. 128–130 °C. 1H
Eur. J. Org. Chem. 2010, 2867–2877
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
2875

A. C. Benniston, G. Copley, H. Lemmetyinen, N. V. Tkachenko
FULL PAPER
1
10833–10841; f) R. Guliyez, O. Buyukcakir, F. Sozman, O. A.
Bozdemir, Tetrahedron Lett. 2009, 50, 5139–5141.
(0.89 g, 94%); m.p. 178–180 °C. H NMR (CDCl3, 300 MHz): δ =
7.06 (s, 1 H), 6.82 (m, 2 H), 6.62 (d, J = 1.9 Hz, 1 H), 2.53 (s, 3
H), 2.38 (m, 5 H), 2.18 (s, 3 H), 2.12 (s, 3 H), 1.07 (t, J = 7.5 Hz,
3 H) ppm. 13C NMR (CDCl3, 75 MHz): δ (ppm) = 187.49, 186.19,
160.60, 152.70, 142.00, 139.50, 137.26, 136.72, 135.02, 134.91,
134.41, 132.29, 122.28, 120.00, 17.61, 14.39, 13.74, 13.19, 10.79,
[3] A. C. Benniston, G. Copley, Phys. Chem. Chem. Phys. 2009,
11, 4124–4131.
[4] a) W. Qin, M. Baruah, W. M De Borggraeve, N. Boens, J. Pho-
tochem. Photobiol. 2006, 183, 190–197; b) C. N. Baki, E. U.
Akkaya, J. Org. Chem. 2005, 70, 4152–4157.
9.60. 11B NMR (CDCl3, 160 MHz): δ (ppm) = –0.10 (t, Jav
33.09 Hz). 19F NMR (CDCl3, 470 MHz): δ (ppm) = –145.83 (q,
=
[5] a) D. P. Kennedy, C. M. Kormos, S. C. Burdette, J. Am. Chem.
Soc. 2009, 131, 8578–8586; b) X. Zhang, H.-S. Zhang, Spec.
Chim. Acta 2005, 61, 1045–1049; c) T. Matsumoto, Y. Urano,
T. Shoda, H. Kojima, T. Nagano, Org. Lett. 2007, 9, 3375–
3377; d) X. Peng, J. Du, J. Fan, J. Wang, Y. Wu, J. Zhao, S.
Sun, T. J. Xu, J. Am. Chem. Soc. 2007, 129, 1500–1501; e) A.
Coskun, M. D. Yilmaz, E. U. Akkaya, Org. Lett. 2007, 9, 607–
609; f) M. Baruah, W. Qin, R. A. L. Vallée, D. Beljonne, T.
Rohand, W. Dehaen, N. Boens, Org. Lett. 2005, 7, 4377–4380;
g) J.-P. Malval, I. Leray, B. Valeur, New J. Chem. 2005, 29,
1089–1094.
[6] R. Ziessel, G. Ulrich, A. Harriman, M. A. H. Alamiry, B. Ste-
wart, P. Retailleau, Chem. Eur. J. 2009, 15, 1359–1369.
[7] a) Z.-N. Sun, F.-Q. Liu, Y. Chen, P. K. H. Tam, D. Yang, Org.
Lett. 2008, 10, 2171–2174; b) Y. Yamada, Y. Tomiyama, A.
Morita, M. Ikekita, S. Aoki, ChemBioChem 2008, 9, 853–856.
[8] A. Gomes, E. Fernades, J. L. F. C. Lima, J. Biochem. Biophys.
Methods 2005, 65, 45–80.
Jav = 32.73 Hz). IR (neat): ν = 2961 (C–H), 1661 (C=O), 1532,
˜
1444 (C=C, C=N), 1187 (B–F) cm–1. EI-MS: m/z calcd. for
C21H21BF2N2O2 (382), found 382. C21H21BF2N2O2 (382): calcd. C
65.99, H 5.54, N 7.33; found C 65.93, H 5.68, N 7.32.
Preparation of BD-SPQ: The same procedure to above was used.
BD-SPHQ (0.23 g, 0.49 mmol, 1 equiv.), THF (50 mL), DDQ
(0.17 g, 0.74 mmol, 1.5 equiv.). Purification: silica gel, petroleum
ether/chloroform (1:1) eluant. Purple solid (0.22 g, 97%); m.p. 217–
219 °C. 1H NMR (CDCl3, 300 MHz): δ = 7.53 (d, J = 8.2 Hz, 2
H), 7.31 (d, J = 8.2 Hz, 2 H), 7.06 (s, 1 H), 6.89 (d, J = 2.3 Hz, 1
H), 6.86 (d, J = 10.1 Hz, 1 H), 6.81 (dd, J = 10.1, JЈ = 2.3 Hz, 1
H), 2.52 (s, 6 H), 2.38 (q, J = 7.5 Hz, 2 H), 2.20 (s, 3 H), 2.17 (s,
3 H), 1.06 (t, J = 7.5 Hz, 3 H) ppm. 13C NMR (CDCl3, 75 MHz):
δ = 187.63, 187.84, 158.00, 152.48, 145.53, 138.34, 137.14, 136.38,
136.32, 135.92, 133.77, 133.12, 132.42, 132.09, 131.05, 129.89,
129.41, 17.39, 14.53, 13.41, 12.92, 10.36, 9.53 ppm. 11B NMR
(CDCl3, 160 MHz): δ (ppm) = –0.00 (t, Jav = 33.21 Hz). 19F NMR
(CDCl3, 470 MHz): δ = –146.04 (q, Jav = 32.81 Hz) ppm. IR (neat):
[9] A. C. Benniston, G. Copley, K. J. Elliott, R. W. Harrington, W.
Clegg, Eur. J. Org. Chem. 2008, 2705–2713.
[10] a) Z. Dost, S. Atilgan, E. U. Akkaya, Tetrahedron 2006, 62,
8484–8488; b) D. Zhang, Y. Wen, Y. Xiao, G. Yu, Y. Liu, X. X.
Qian, Chem. Commun. 2008, 4777–4779.
ν = 2970 (C–H), 1652 (C=O), 1535, 1456 (C=C, C=N), 1184 (B–
˜
[11]
a) J. L. Sessler, M. R. Johnson, S. E. Creager, J. C. Fettinger,
J. A. Ibers, J. Am. Chem. Soc. 1990, 112, 9310–9329; b) J. L.
Sessler, V. L. Capuano, A. Harriman, J. Am. Chem. Soc. 1993,
115, 4618–4628.
F) cm–1. EI-MS: m/z calcd. for C27H25BF2N2O2 (458), found 458.
C27H25BF2N2O2 (458): calcd. C 70.76, H 5.50, N 6.11; found C
70.81, H 5.44, N 6.09.
[12]
[13]
H. D. Tabb, K. M. Smith, J. Org. Chem. 1984, 49, 1870–1875.
G. B. Marini-Bettolo, O. A. Polla, J. H. Abril, Gazz. Chim. Ital.
1950, 80, 76–84.
N. Miyaura, K. Yamada, A. Suzuki, Tetrahedron Lett. 1979,
36, 3437–3440.
The molecular structures can be viewed at the website http://
www.ncl.ac.uk/mpl/, or alternatively they are presented in G.
Copley, PhD Thesis, Newcastle University, 2009.
Supporting Information (see also the footnote on the first page of
this article): Calculated thermodynamic parameters, Table of ab-
sorption/fluorescence data, simulated NMR spectra, electrochemis-
try, Beer–Lambert plot and additional fast kinetic data.
[14]
[15]
Acknowledgments
[16]
[17]
[18]
M. Bröring, R. Krüger, S. Link, C. Kleeberg, S. Köhler, X. Xie,
B. Ventura, L. Flamigni, Chem. Eur. J. 2008, 14, 2976–2983.
Molar volumes were calculated using data taken from: A.
Bondi, J. Phys. Chem. 1964, 68, 441–451.
T. R. Kelly, M. C. Bowyer, K. V. Bhaskar, D. Bebbington, A.
Garcia, F. Lang, M. H. Kim, M. P. Jette, J. Am. Chem. Soc.
1994, 116, 3657–3658.
In order to isolate the diastereoisomers no rotation of the ap-
pended group can take place in solution at ambient tempera-
ture. Very hindered rotation would be more probable using a
large bulky chiral hydroxy protecting group starting from, for
example, 1-bromomethyl-(+)-cuparene.
a) Y. Zhou, Y. Xiao, D. Li, M. Fu, X. Qian, J. Org. Chem.
2008, 73, 1571–1574; b) J. Chen, A. Burghart, A. Derecskei-
Kovacs, K. Burgess, J. Org. Chem. 2000, 65, 2900–2906.
a) J. Bañuelos, F. L. Arbeloa, V. M. Martínez, T. A. López,
I. L. Arbeloa, Phys. Chem. Chem. Phys. 2004, 6, 4247–4253; b)
J. B. Prieto, F. L. Arbeloa, V. M. Martínez, I. L. Arbeloa,
Chem. Phys. 2004, 296, 13–22.
C.-W. Wan, A. Burghart, J. Chen, F. Bergström, L. B.-Å.
Johansson, M. F. Wolford, T. G. Kim, M. R. Topp, R. M.
Hochstrasser, K. Burgess, Chem. Eur. J. 2003, 9, 4430–4441.
S. J. Strickler, R. A. Berg, J. Chem. Phys. 1962, 37, 814–822.
J. H. Boyer, A. M. Haag, G. Sathyamoorthi, M.-L. Soong, K.
Thangaraj, T. G. Pavlopoulos, Heteroat. Chem. 1993, 4, 39–49.
R. Englman, J. Jortner, Mol. Phys. 1970, 18, 145–164.
F. Bergström, I. Mikhalyov, R. Hägglöf, R. Wortmann, T. Ny,
L. B.-Å. Johansson, J. Am. Chem. Soc. 2002, 124, 196–204.
We thank the Engineering and Physical Sciences Research Council
(EPSRC) and Newcastle University for financial support and the
EPSRC sponsored National Mass Spectrometry Service at Swansea
for obtaining mass spectra. We also thank Dr Bruce Tattershall
and Prof William McFarlane for help in collecting NMR spectra
and the fitting of the 19F NMR spectroscopic data.
[19]
[1] For comprehensive reviews see: a) A. Loudet, K. Burgess,
Chem. Rev. 2007, 107, 4891–4932; b) T. E. Wood, A. Thomp-
son, Chem. Rev. 2007, 107, 1831–1861; c) G. Ulrich, R. Ziessel,
A. Harriman, Angew. Chem. Int. Ed. 2008, 47, 1184–1201; d)
R. Ziessel, G. Ulrich, A. Harriman, New J. Chem. 2007, 31,
496–501; e) K. Burgess, L. Wu, Chem. Commun. 2008, 4933–
4935; f) M. Shah, K. Thangraj, M. L. Soong, L. Wolford, J. H.
Boyer, I. R. Poliitzer, T. G. Pavlopoulos, Heteroat. Chem. 1990,
1, 389–399; g) A. Burghart, H. Kim, M. B. Welch, L. H. Tho-
resnsen, J. Reibenspies, K. Burgess, J. Org. Chem. 1999, 64,
7813–7819.
[2] For a selection of examples see: a) M. D. Yilmaz, A. Bozdemir,
E. U. Akkaya, Org. Lett. 2006, 8, 2871–2873; b) H. Y. Lee,
D. R. Bae, J. C. Park, H. Song, W. S. Han, J. H. Jung, Angew.
Chem. Int. Ed. 2009, 48, 1239–1243; c) K. Umezawa, A. Mat-
sui, Y. Nakmura, D. Citterio, K. Suzuki, Chem. Eur. J. 2009,
15, 1096–1106; d) K. Rurack, M. Kollmannsberger, J. Daub,
Angew. Chem. Int. Ed. 2001, 40, 385–387; e) M. M. Sartin, F.
Camerel, R. Ziessel, A. J. Bard, J. Phys. Chem. C 2008, 112,
[20]
[21]
[22]
[23]
[24]
[25]
[26]
2876
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 2867–2877

Comparison of Substituted Donor-Acceptor Bodipy Dyads
[27] a) T. Rohand, J. Lycoops, S. Smout, E. Braeken, M. Sliwa,
M. Van der Auweraer, W. Dehaen, W. M. De Borggraeve, N.
Boens, Photochem. Photobiol. Sci. 2007, 6, 1061–1066; b) W.
Qin, T. Rohand, M. Baruah, A. Stefan, M. Van der Auweraer,
W. Dehaen, N. Boens, Chem. Phys. Lett. 2006, 420, 562–568;
c) W. Qin, M. Baruah, M. Van der Auweraer, F. C.
De Schryver, N. Boens, J. Phys. Chem. A 2005, 109, 7371–7384;
d) W. Qin, M. Baruah, A. Stefan, M. Van der Auweraer, N.
Boens, ChemPhysChem 2005, 6, 2343–2351.
[28] A single mono-exponential decay is not always observed and
is often associated with other processes stemming from the ex-
cited-state such as vibrational cooling, proton transfer etc., see
for example: H. Wang, H. Zhang, O. K. Abou-Zied, C. Yu,
F. E. Romesberg, M. Glasbeek, Chem. Phys. Lett. 2003, 367,
599–608.
2823–2824; e) J. Hu, B. Xia, D. Bao, A. Ferreira, J. Wan, G.
Jones II, V. I. Vullev, J. Phys. Chem. A 2009, 113, 3096–3107.
[35]
[36]
a) M. Bixon, J. Jortner, Adv. Chem. Phys. 1999, 106, 35–208; b)
J. Kroon, H. Oevering, J. W. Verhoeven, J. M. Warman, A. M.
Oliver, M. N. Paddon-Row, J. Phys. Chem. 1993, 97, 5065–
5069; c) K. J. Smit, J. M. Warman, M. P. de Haas, M. N. Pad-
don-Row, A. M. Oliver, Chem. Phys. Lett. 1988, 152, 177–182;
d) N. Liang, J. R. Miller, G. L. Closs, J. Am. Chem. Soc. 1990,
112, 5353–5354.
a) S. Franzen, J. Martin, Annu. Rev. Phys. Chem. 1995, 46, 453–
487; b) A. C. Benniston, G. M. Chapman, A. Harriman, M.
Mehrabi, J. Phys. Chem. A 2004, 108, 9026–9036.
A. A. Gorman, I. Hamblett, T. A. King, M. D. Rahn, J. Pho-
tochem. Photobiol. A: Chem. 2000, 130, 127–132.
N. Yoshida, T. Ishizuka, K. Yofu, M. Murakami, H. Miyasaka,
T. Okada, Y. Nagata, A. Itaya, H. S. Cho, D. Kim, A. Osuka,
Chem. Eur. J. 2003, 9, 2854–2866.
[37]
[38]
[29] G. Duvanel, N. Banerji, E. Vauthey, J. Phys. Chem. A 2007,
111, 5361–5369.
[30] S. Hattori, K. Ohkubo, Y. Urano, H. Sunhara, T. Nagano, Y.
Wada, N. V. Tkachenko, H. Lemmetyinen, S. Fukuzumi, J.
Phys. Chem. B 2005, 109, 15368–15375.
[31] a) P. P. Lainé, F. Bedioui, F. Loiseau, C. Chiorboli, S. Cam-
pagna, J. Am. Chem. Soc. 2006, 128, 7510–7521; b) K. Nom-
oto, S. Kume, H. Nishihara, J. Am. Chem. Soc. 2009, 131,
3830–3831.
[39] It also possible to explain the increase in the rate constant for
charge recombination in terms of the energy gap law (see ref.
25). Generally, Rω is taken to be around 1500 cm–1 when con-
sidering return electron transfer in donor-acceptor systems; the
increase in Rω to ca. 3500 cm–1 will concomitantly increase kNR
assuming all other parameters remain constant. This argument
has been used to explain the rapid non-radiative deactivation
of a charge-transfer state in a N–H-containing Bodipy-based
tripartite array, see: M. A. H. Alamiry, A. Harriman, L. J. Mal-
lon, G. Ulrich, R. Ziessel, Eur. J. Org. Chem. 2008, 2774–2782.
[40] R. I. Cukier, D. G. Nocera, Annu. Rev. Chem. 1998, 49, 337–
369.
[32] W. B. Davis, M. A. Ratner, M. R. Wasielewski, J. Am. Chem.
Soc. 2001, 123, 7877–7886.
[33] R. A. Marcus, Discuss. Faraday Soc. 1960, 29, 21–31.
[34] a) P. L. Dutton, J. S. Leight, M. Seibert, Biochem. Biophys. Res.
Commun. 1972, 46, 406–413; b) M. C. Thuranauer, J. J. Katz,
J. R. Norris, Proc. Natl. Acad. Sci. USA 1975, 72, 3270–3274;
c) K. Hasharoni, H. Levanon, S. R. Greenfield, D. J. Gosztola,
W. A. Svec, M. R. Wasielewski, J. Am. Chem. Soc. 1995, 117,
8055–8056; d) M. R. Wasielewski, D. W. Minsek, M. P. Niem-
czyk, W. A. Svec, N. C. Yang, J. Am. Chem. Soc. 1990, 112,
[41] A. C. Benniston, G. Copley, H. Lemmetyinen, N. V. Tkach-
enko, ChemPhysChem 2010, in press.
Received: February 17, 2010
Published Online: April 14, 2010
Eur. J. Org. Chem. 2010, 2867–2877
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
2877