Using a photoacid generator to switch the direction of electronic energy transfer in a molecular triad

Article abstract of DOI:10.1002/anie.201102065

A switch in time: A sequence of highly efficient, intramolecular electronic energy transfer steps follows from selective illumination of the fluorescent center (DPP) present in a new class of molecular triads (see picture). The direction of energy flow de

Full text of DOI:10.1002/anie.201102065

DOI: 10.1002/anie.201102065
Energy Transfer
Using a Photoacid Generator to Switch the Direction of Electronic
Energy Transfer in a Molecular Triad**
Delphine Hablot, Anthony Harriman,* and Raymond Ziessel*
Over the past decade there has been a renaissance of interest
in the processes by which electronic energy can be trans-
ported around an organized molecular array with minimal
loss.[1] The motivation for this research stems from a desire to
apply, at a molecular level, the lessons acquired from our
ever-deepening understanding of the natural light-harvesting
machinery[2] that powers bacterial and green-plant photosyn-
thesis and other biological processes.[3] A key requirement of
all such functional units is the need to move the photonic
energy to a site where chemical reactions are initiated.[4]
Paramount to the successful design of artificial prototypes
able to operate in this way is the logical positioning of
individual units in a way that favors vectorial electronic
energy transfer (EET) along the molecular axis[5,6] or by way
of some other preferred pathway.[7,8] An obvious, and indeed
enviable, extension for these materials is to devise a simple
means by which the EET flow can be reversed, while
maintaining very high efficiency. Such switching protocols
might involve a change in solvent polarity[9] or temperature,[10]
coordination of substrates,[11] light-induced conformational
exchange,[12] or modulation of the excitation wavelength.[13]
Herein, we introduce a new concept for alternating the EET
direction. Our approach, which works in both liquid and solid
states, uses a photoacid generator (PAG) to trigger the switch.
The target molecule was constructed with a central 1,4-
oxo-3,6-diphenylpyrrolo[3,4-c]pyrrole unit, DPP, connecting
two disparate Bodipy dyes; namely, a blue dye absorbing
strongly at 650 nm and a green dye with prominent absorption
centered at 695 nm. Note the green dye is readily protonated
at the amine sites to give a blue dye that exhibits an
absorption maximum at 630 nm. As such, protonation of the
green dye is involved in the switching mechanism; this
observation in itself is not a new idea but follows from original
work by Armaroli et al.[14] A key feature of our design
principle requires that the DPP core absorbs and emits at
slightly higher energy than either of the termini. An
appropriate prototypic compound is B(DPP)G, as shown in
Scheme 1. Preparation of this triad involves cross-coupling
the bromo function of the DPP unit with the dedicated
terminal alkynes B and G in the presence of Pd0.[15] The
corresponding symmetrical controls, B(DPP)B and G-
(DPP)G, were prepared using a slight excess of the corre-
Scheme 1. Molecular formulae of the molecular triads used herein and
their respective abbreviations.
sponding starting dye B or G, respectively. The mixed dye,
B(DPP)G, was prepared by first cross-coupling the blue dye
with the central platform and then attaching the green dye in
a separate step (Scheme 1). The boron substituents import
polarity to assist purification by column chromatography. The
peripheral units attached to the Bodipy dyes were selected in
order to modulate the optical properties and, in the case of
the green dye, to provide the protonation venue. These new
dyes are soluble in common organic solvents and give well-
defined proton NMR spectra in keeping with the lack of
aggregation in solution (see the Supporting Information).
The isolated DPP unit is highly fluorescent (lFLU
=
538 nm) in solution (see the Supporting Information for a
summary of the photophysical properties of the isolated
dyes); for example in dioxane at 208C, the emission quantum
yield (FF) is 0.87 and the excited-singlet-state lifetime (tS) is
4.9 ns. On selective excitation at 470 nm of the DPP unit
present in B(DPP)B, FF falls to 0.001 while tS is decreased to
5.7 ps. There is concomitant appearance of strong emission
from the blue dye and, on the basis of comparing the
excitation and absorption spectra, it is concluded that intra-
molecular EET is highly efficient (i.e., PEET = 99.8%) in this
system (see the Supporting Information for full details); the
corresponding rate constant, kEET, is 17.5 ꢀ 1010 sÀ1. Indeed,
EET is promoted by a moderately high spectral overlap
integral (JDA = 0.00060 cm) and thermodynamic driving force,
although the center-to-center separation distance (R = 15 ꢁ)
is reasonably high. Similar behavior holds for G(DPP)G in
dioxane solution, although the spectral overlap integral is
[*] D. Hablot, Dr. R. Ziessel
LCOSA, Ecole Europꢀenne de Chimie, Polymꢁres et Matꢀriaux
CNRS, 25 rue Becquerel, 67087 Strasbourg Cedex 02 (France)
E-mail: ziessel@unistra.fr
Homepage: http://www-lmspc.u-strasbg.fr/lcosa
Prof. Dr. A. Harriman
Molecular Photonics Laboratory, School of Chemistry, Bedson Bldg
Newcastle University, Newcastle upon Tyne, NE1 7RU (UK)
E-mail: anthony.harriman@ncl.ac.uk
[**] This work was supported by the CNRS (UMR 7515), EPSRC, and by
Newcastle University.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201102065.
Angew. Chem. Int. Ed. 2011, 50, 7833 –7836
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
7833

Communications
decreased (JDA = 0.00018 cm) because of the red-shifted
absorption profile that is inherent to the green dye; here,
EET = 4.0 ꢀ 1010 sÀ1 and PEET = 99.5%. Spectroscopic meas-
appropriate photobase[19] but these were not pursued. In the
first instance, a solution of G(DPP)G in dioxane was
examined by fluorescence spectroscopy, with excitation at
525 nm, an isosbestic point for the neutral and dicationic
species. Emission occurs exclusively from the neutral species
and the intensity at 825 nm was measured. The photoacid
generator, PAG (Scheme 2), was added and the solution was
kept in the dark except for weak excitation at 525 nm. High
concentrations (> 5 mgmLÀ1) of PAG caused fluorescence
k
urements indicate that the quenching of DPP fluorescence is
entirely due to intramolecular EET. For the asymmetric triad
B(DPP)G, selective illumination into DPP results in EET to
both the blue (PEET = 75%) and green (PEET = 25%) termini
but is followed by fast (kEET = 1.0 ꢀ 1010 sÀ1; PEET = 98%) EET
from the blue dye to the green analogue. This latter step is
driven by a high spectral overlap integral (JDA = 0.00164 cm),
offset by the large spatial isolation (R = 34 ꢁ), and ensures
that the green dye is the recipient for essentially all photons
absorbed by B(DPP)G.
Protonation[16] of the green dye (lMAX = 695 nm) in
dioxane with gaseous HCl generates a blue dye (lMAX
=
630 nm) that is an excellent acceptor for photons absorbed
by the DPP unit. Thus, for GH+(DPP)GH+, the overlap
integral is high (JDA = 0.00098 cm) and intramolecular EET is
quantitative (PEET = 100%; kEET = 29.5 ꢀ 1010 sÀ1). The pro-
tonated dye shows modest fluorescence centered at 642 nm
Scheme 2. Molecular formulae of the photoacid generator (PAG) used
to switch the protonation state of the green Bodipy dye and of the
oxothioxanthene (SEN) sensitizer employed as UV activator. The
spectral absorption properties are given in each case and spectra are
shown in the Supporting Information.
(FF = 0.15; tS = 3.2 ns) that overlaps strongly (JDA
=
0.0025 cm) with absorption by the blue dye. Consequently,
selective excitation of the central DPP unit in B(DPP)GH+
leads to highly efficient intramolecular EET to the proton-
ated dye (PEET = 62%; kEET = 22 ꢀ 1010 sÀ1), with a contribu-
tion of direct EET to the blue terminus (PEET = 38%; kEET
=
quenching but this quenching was minimal at low loadings
(i.e., 400 mgmLÀ1). Under the latter conditions, the solution
was illuminated for short periods (typically 30 second bursts)
at (330 Æ 15) nm from a 250 W xenon lamp to activate the
photoacid. The liberated acid causes protonation of the green
dye as evidenced by the appearance of fluorescence at
645 nm. Indeed, monitoring the ratio of emission intensities
at 645 and 825 nm after each illumination period shows
complete conversion of G(DPP)G into the corresponding
diacid (see the Supporting Information). Addition of triethyl-
amine leads to partial recovery of the fluorescence character-
istic of G(DPP)G, although there is an overall loss of
approximately 10%. This decomposition route is most likely
due to free radical attack during liberation of the photoacid.
Nonetheless, the principle of fluorescence switching by
photolysis has been demonstrated.
14 ꢀ 1010 sÀ1), followed by long-range EET to the blue
terminus (PEET = 98%; kEET = 3.1 ꢀ 1010 sÀ1). The net result,
as illustrated in Figure 1, is that protonation of the green dye
reverses the EET flow without affecting the overall efficiency,
which remains extremely high in all cases (see the Supporting
Information for details).
Clearly, the manual addition of acid or base to switch the
EET direction is unattractive and impractical but it is a
demonstration of the generic principle. A logical improve-
ment is to employ a photoacid (or photobase) to create the
required proton gradient[17] and, by moving to a polymer
matrix, to restrict the spatial distribution of the switching
effect. As proof-of-concept, studies were carried out with a
conventional photoacid[18] in both solution and solid states; in
principle, related studies could be carried out with an
A thin (ca. 5 mm) film of poly(methyl methacrylate)
(PMMA) was spin-coated onto quartz slides from CHCl3
containing various amounts of B(DPP)G (2–50 mgmLÀ1
initial solution)[20] and dried under vacuum at 408C. Fluores-
cence spectroscopy confirms that EET is highly efficacious in
the dried film (see the Supporting Information); fluorescence
from DPP is essentially absent, while emission from the blue
dye is barely detectable in contrast to strong fluorescence
from the green dye. Indeed, the small amount of residual
fluorescence from the blue dye becomes quenched at higher
loading (> 80 mgmLÀ1 initial solution) of dye because of an
intermolecular EET process that does not occur at moderate
concentrations in solution. The net result is that only the
green dye emits in the film, regardless of the excitation
wavelength. At high loading of B(DPP)G, energy migration
among molecules of the green dye will disperse the electronic
energy since the intermolecular Fçrster critical distance
computed[21] for random orientations of B(DPP)G is approx-
Figure 1. Representation of the chemically modified switching process,
in which conversion of G into GH+ causes a change in the direction of
electronic energy transfer (EET) from the central donor to the
peripheral acceptors.
7834
www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 7833 –7836

imately 35 ꢁ. On loading the B(DPP)G-doped film
(20 mgmLÀ1 initial solution) with PAG (1 mgmLÀ1 of solution
prior to spin-coating), the EET step remains unaffected but
fluorescence from the green dye is decreased by approx-
imately 40% relative to a film prepared in the absence of
PAG because of intermolecular electron transfer.[23]
Previous work has demonstrated that certain photoacid
generators can be activated by two-photon excitation,[24] most
notably using high-power lasers emitting in the far-red
region.[25] Advantages of this mode of excitation include
improved selectivity, depending on the two-photon absorp-
tion cross-section of the photoacid precursor,[22] and better
spatial resolution.[26] Most photoacid generators can also be
activated by sensitization[27] with an added UV absorber. 9-
Oxothioxanthones,[28] such as SEN (Scheme 2), are known to
function as effective sensitizers for many classes of photoacid
generators, including PAG. The main benefit of using a
UVabsorber to sensitize breakdown of the photoacid is that
significantly lower concentrations of PAG become possible,
which thereby curtail quenching of fluorescence from the
green dye.
Thus, steady-state illumination at 400 nm of a PMMA film
containing SEN (2 mgmLÀ1 of initial solution), PAG
(200 mgmLÀ1 of initial solution), and G(DPP)G (20 mgmLÀ1
of initial solution) causes progressive replacement of the
725 nm fluorescence with an emission band at 645 nm as
protonation of the green dye proceeds. Under such condi-
tions, SEN absorbs more than 95% of incident near-UV
photons. Interestingly, the fluorescence characteristic of SEN
is weak relative to that of a PMMA film prepared in the
absence of PAG or G because of a combination of electron
transfer to PAG and EET to the Bodipy dye (Figure 3). On
replacing G with B(DPP)G, the normal fluorescence from the
blue dye, which is initially absent, starts to appear as
photolysis proceeds (as reported in Figure 2). In this case,
there is almost no residual fluorescence from SEN; presum-
ably this is due to an increased rate of EET to the tripartite
dye.
Figure 3. Effect of UV illumination on the fluorescence spectral profile
recorded for the green dye (G(DPP)G) in a PMMA matrix loaded with
PAG and SEN. Illumination was carried out at 400 nm where SEN is
the dominant chromophore.
from the mixture curtailed the characteristic evolution of the
fluorescence maximum from 725 to 645 nm. In principle, it
should be possible to activate the PAG by two-photon
absorption using wavelengths around 700 nm but this could
not be realized in practice. An action spectrum recorded for
the full mixture, monitoring the ratio of fluorescence inten-
sities at 645 and 825 nm after exposure to a constant photon
flux, shows a broad profile with a peak at 800 nm (Figure 4).
This observation is consistent with that expected for two-
photon excitation of SEN, the absorption maximum of which
is at 397 nm.[28] The dimensions of the irradiation area can be
varied by changing the microscope objective or by moving to
confocal microscopy conditions.
Dried films loaded with B(DPP)G, PAG, and SEN as
described above could also be activated by two-photon
absorption of the sensitizer by excitation with high photon
densities delivered at 800 nm. Eliminating either PAG or SEN
Figure 4. Two-photon action spectrum recorded for activation of the
PAG in PMMA doped with B(DPP)G and SEN. The sensitivity is
expressed in terms of the ratio of fluorescence intensities measured at
825 nm (green dye) and 645 nm (blue dye) following exposure to a
fixed photon density delivered by a train of laser pulses.
This work has verified the concept of switching the
direction of intramolecular EET by protonation of one
chromophore in a multiple-dye package, as was reported
earlier for a molecular dyad.[14] The switch is reversible, at
least to a high degree, and can be activated by photochemical
means. In the case of B(DPP)G, the state of the switch can be
read by fluorescence and/or absorption spectroscopy and
activation occurs equally well in solution and in plastic films.
Various means of activation are possible and, by using
Figure 2. Effect of UV illumination on the fluorescence spectral profile
recorded for B(DPP)G in a PMMA film containing PAG (1 mgmLÀ1
initial solution). The excitation wavelength corresponds to an isosbes-
tic point for the blue and green chromophores.
Angew. Chem. Int. Ed. 2011, 50, 7833 –7836
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
7835

Communications
[2] a) M. Y. Okamura, G. Feher, N. Nelson in Photosynthesis:
Energy Conversion by Plants and Bacteria (Ed.: Govindjee)
Academic Press, New York, 1982, p. 799; b) K. Sugiura, Top.
Curr. Chem. 2003, 228, 65.
[3] R. Ziessel, A. Harriman, Chem. Commun. 2011, 47, 611.
[4] R. E. Blankenship, Molecular Mechanism of Photosynthesis,
Blackwell Science, Oxford, 2002.
[5] A. Harriman, L. J. Mallon, R. Ziessel, Chem. Eur. J. 2008, 14,
11461.
[6] A. Harriman, L. J. Mallon, S. Goeb, G. Ulrich, R. Ziessel, Chem.
Eur. J. 2009, 15, 4553.
[7] a) S. Speiser, Chem. Rev. 1996, 96, 1953; b) D. Kim, A. Osuka,
Acc. Chem. Res. 2004, 37, 735.
[8] A. Harriman, L. J. Mallon, K. J. Elliot, A. Haefele, G. Ulrich, R.
Ziessel, J. Am. Chem. Soc. 2009, 131, 13375.
[9] T. L. Easun, W. Z. Alsindi, N. Deppermann, M. Towrie, K. L.
Ronayne, X.-Z. Sun, M. D. Ward, M. W. George, Inorg. Chem.
2009, 48, 8759.
[10] A. C. Benniston, A. Harriman, P. Y. Li, C. A. Sams, J. Am.
Chem. Soc. 2005, 127, 2553.
confocal microscopy with two-photon excitation, switching
can be restricted to very small zones. In this case, EET might
be expected from the blue dye inside the zone to green dye at
the periphery of the acidified region. The Fçrster critical
distance computed[21] for such EETs is approximately 30 ꢁ
and; with state-of-the-art confocal microscopy, the diameter
of the zone could be decreased to approximately 100 nm. The
effect of bimolecular EET under such conditions would be a
small reduction in the apparent dimensions of the activated
zone when measured by fluorescence microscopy. Further
modifications of this approach could involve attaching the
PAG to the molecular triad so as to minimize the amount of
photoacid that needs to be generated. Such protocols could be
developed as a means for designing rapid switching systems.
We have also shown[16] that green Bodipy-based dyes can be
appended to porous microspheres in such a way that
reversible protonation still takes place. It is reasonable to
suppose that B(DPP)G could likewise be anchored to
polystyrene spheres preloaded with PAG. Such materials
could be studied at the single-bead level and could be
developed as advanced probes for chemical pollutants (e.g.,
phosgene) and/or acidic contaminants. It should also be
emphasized that, while this work has focussed on photoacids,
corresponding studies could be conducted with photobases.
[11] J. A. Faiz, R. M. Williams, M. J. J. Pereira Silva, L. De Cola, Z.
Pikramenou, J. Am. Chem. Soc. 2006, 128, 4520.
[12] J. L. Bahr, G. Kodis, L. De La Garza, S. Lin, A. L. Moore, T. A.
Moore, D. Gust, J. Am. Chem. Soc. 2001, 123, 7124.
[13] A. C. Benniston, A. Harriman, P. Y. Li, J. Am. Chem. Soc. 2010,
132, 26.
[14] N. Armaroli, J.-F. Eckert, J.-F. Nierengarten, Chem. Commun.
2000, 2105.
[15] D. Hablot, P. Retailleau, R. Ziessel, Chem. Eur. J. 2010, 16,
13346.
Experimental Section
[16] R. Ziessel, G. Ulrich, A. Harriman, M. A. H. Alamiry, B.
Stewart, P. Retailleau, Chem. Eur. J. 2009, 15, 1359.
[17] C. Coenjarts, O. Garcia, L. Llauger, J. Palfreyman, A. L. Vinette,
J. C. Scaiano, J. Am. Chem. Soc. 2003, 125, 620.
[18] E. Reichmanis, L. F. Thompson, Chem. Rev. 1989, 89, 1273.
[19] K. Suyama, M. Shirai, Prog. Polym. Sci. 2009, 34, 194.
[20] Over this range, the fluorescence intensity increased linearly
with increasing concentration of the dye, for excitation into the
green dye at 670 nm, thereby discounting self-aggregation.
[21] Y. R. Khan, T. E. Dykstra, G. D. Scholes, Chem. Phys. Lett. 2008,
461, 305.
Synthesis and characterization of the new compounds is described in
the Supporting Information. All solvents were purchased in spectro-
scopic grade from Aldrich Chemicals Co., were used as received, and
were found to be free of fluorescent impurities. Samples of SEN were
obtained from Ward-Blenkinsop Co as the chloride salt and were
subjected to ion exchange. Samples of PAG were prepared according
to a literature procedure.[26] Absorption spectra were measured with a
Hitachi U3310 spectrophotometer and fluorescence spectra were
measured with a fully corrected Jobin–Yvon Fluorolog tau-3 spec-
trometer for quantitative measurements and
a Hitachi F-4500
fluorescence spectrometer for routine studies. Steady-state irradia-
tion studies were carried out with the sample solution contained in a
quartz cell, after purging with N2, using a 250 W xenon arc lamp. The
light beam was passed through a high radiance monochromator to
isolate the required excitation wavelength. The course of reaction was
followed by fluorescence spectroscopy. Irradiation studies made with
PMMA films, after casting by spin-coating (see the Supporting
Information), utilized the same excitation source. For the two-photon
excitation studies, a train of 50 pulses delivered from a frequency-
doubled Nd–YAG laser (FWHM = 4 ns) was directed to a microscope
objective and focused onto the plastic film. The laser beam was passed
through an OPA in order to isolate a spectral region centered at
approximately 800 nm.
[22] PAG shows negligible absorption at wavelengths longer than
380 nm and does not interfere in diagnostic studies of the dyes.
Likewise, SEN does not absorb at wavelengths longer than
470 nm.
[23] Protonation does not take place for photolysis under an
atmosphere of ammonia or triethylamine. Exposure of a photo-
lyzed film to NH3 gas recovers the original emission from the
green dye, with approximately 5% loss. See the Supporting
Information for details.
[24] L. Steidl, S. J. Jhaveri, R. Ayothi, J. Sha, J. D. McMullen, S. Y. C.
Ng, W. R. Zipfel, R. Zentel, C. K. Ober, J. Mater. Chem. 2009, 19,
505.
[25] W. Zhou, S. M. Kuebler, K. L. Braun, T. Yu, J. K. Cammack,
C. K. Ober, J. W. Perry, S. R. Marder, Science 2002, 296, 1106.
[26] J.-P. Malval, S. Suzuki, F. Morlet-Savary, X. Allonas, J.-P.
Fouassier, S. Takahara, T. Yamaoka, Chem. Phys. Lett. 2007,
443, 323.
Received: March 23, 2011
Revised: May 30, 2011
Published online: July 4, 2011
[27] J.-P. Malval, S. Suzuki, F. Morlet-Savary, X. Allonas, J.-P.
Fouassier, S. Takahara, T. Yamaoka, J. Phys. Chem. A 2008,
112, 3879.
[28] J. Davila, A. Harriman, M.-C. Richoux, J. Chem. Soc. Faraday
Trans. 2 1988, 84, 287.
Keywords: dyes/pigments · energy transfer · fluorescence ·
photoacid generators · photophysics
.
[1] a) D. Gust, T. A. Moore, A. L. Moore, Acc. Chem. Res. 2001, 34,
40; b) D. Holten, D. F. Bocian, J. S. Lindsey, Acc. Chem. Res.
2002, 35, 57; c) M. R. Wasielewski, J. Org. Chem. 2006, 71, 5051.
7836
www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 7833 –7836