A deep-purple-grey thiophene-benzothiadiazole-thiophene BODIPY dye for solution-processed solar cells

Article abstract of DOI:10.1039/c4nj00294f

In this work we explore the synthesis of an extended 4,4-difluoro-4-bora- 3a,4a-diaza-s-indacene dye (BODIPY) engineered from thiophene-benzothiadiazole- thiophene modules linked in the 3,5-substitution positions. We found that this highly soluble dye absorbs up to 800 nm in solution and up to 900 nm in thin films. An effective charge transfer absorption band was found at around 479 nm. The hybrid dye emits at 778 nm with a quantum yield of about 6%. Similar electrochemical and optical gaps were determined about 1.36 eV. When deposited in thin films the dye exhibits an ambipolar nature with well-balanced hole and electron mobilities. Bulk heterojunction solar cells based upon this dye blended with [6,6]phenylC_61or71butyricacid methylester (PC₆₁BM or PC₇₁BM) provide a power conversion efficiency of about 1.26% upon a mild thermal annealing.

Full text of DOI:10.1039/c4nj00294f

NJC
View Article Online
View Journal
PAPER
A deep-purple-grey thiophene–benzothiadiazole–
thiophene BODIPY dye for solution-processed
solar cells†
Cite this: New J. Chem., 2014,
38, 3644
Antoine Mirloup,a Nicolas Leclerc,b Sandra Rihn,a Thomas Bura,a Rony Bechara,c
Anne Hebraud,b Patrick Leveque,c Thomas Heiserc and Raymond Ziessel*a
´
´ ˆ
In this work we explore the synthesis of an extended 4,4-difluoro-4-bora-3a,4a-diaza-s-indacene dye
(BODIPY) engineered from thiophene–benzothiadiazole–thiophene modules linked in the 3,5-
substitution positions. We found that this highly soluble dye absorbs up to 800 nm in solution and up to
900 nm in thin films. An effective charge transfer absorption band was found at around 479 nm. The
hybrid dye emits at 778 nm with a quantum yield of about 6%. Similar electrochemical and optical gaps
were determined about 1.36 eV. When deposited in thin films the dye exhibits an ambipolar nature with
well-balanced hole and electron mobilities. Bulk heterojunction solar cells based upon this dye blended
Received (in Montpellier, France)
27th February 2014,
Accepted 15th May 2014
DOI: 10.1039/c4nj00294f
with [6,6]phenylC61or71butyricacid methylester (PC61BM or PC71BM) provide
efficiency of about 1.26% upon a mild thermal annealing.
a power conversion
www.rsc.org/njc
absorbing dyes (labs 4 700 nm) display attractive optical and
interesting charge transporting properties and were highly effi-
1. Introduction
The engineering of photoactive dyes for application in organic cient when blended with PC61BM (electron acceptor) in bulk
(opto)-electronics and photovoltaic devices is a multi-parameter heterojunction (BHJ) solar cells.4 Our interest in this specific
challenge. The most innovative materials involve both the design relies on the fact that large-scale synthesis of purified
optimization of the functional units responsible for the colour compounds could be achieved in a minimum of synthetic steps.
and for the hole transport and the optimization of its combination
In the course of such work it appears important to decipher the
with an electron acceptor moiety in a molecular (or supramole- key elements to understand the system, and find new synthetic
cular) bulk heterojunction (BHJ) suitable for the targeted applica- options for the design of even more efficient and sustainable
tion.1,2 Because of significant issues related to the availability of materials. Along this line of using small-molecules as organic
innovative dyes, the photostability, the charge carrier mobility, the donors several types of dyes have been scrutinized and classified
light harvesting properties, boron dipyrromethene dyes (BODIPY’s) as oligothiophenes,5 diketopyrrolopyrrole (DPP),6 squaraine,7
are generally regarded as highly promising but still poorly inves- hexabenzo-coronenes,8 merocyanine,9 donor–acceptor oxoindane,10
tigated candidates for photovoltaic applications.3 Very recently, we and thiadiazolo-bithienyl dyes.11 Boron dipyrromethene (BODIPY)
became interested in the construction of BODIPY dyes containing dyes3,12 are characterized by outstanding chemical and photo-
a bis-thiophene heterocyclic ring system linked in the 3,5- chemical stabilities, redox activity and optical features that can
substitution positions via unsaturated linkers. These green easily be tailored by chemical transformation, allowing them to
be used in BHJ solar cells13,14 and dye sensitized solar cells.15–17
A logical progression in the design of novel dyes usable in
a
´ ´
´
Institut de Chimie et Procedes pour l’Energie, l’Environnement et la Sante
BHJ solar cells is to use the molecular items present in low-
band-gap polymer systems such as thieno[3,2-b]thiophene,18
N-alkyldithienopyrrole,19 thiazolothiazole-thiophene,20 dithieno-
silole,21 benzodithiophene,22 benzothiadiazole,23 bithiophene-
imide.24 Not only should this unit contribute to the hole
transporting properties, but it should also guarantee the avail-
ability of highly coloured materials when linked in the 3,5-
positions of the BODIPY.
´
´
(ICPEES), Laboratoire de Chimie Moleculaire et Spectroscopies Avancees LCOSA,
´
`
´
Ecole Europeenne de Chimie, Polymeres et Materiaux, ICPEES-LCOSA, UMR 7515
´
associe au CNRS, 25 rue Becquerel, 67087 Strasbourg Cedex 02, France.
E-mail: ziessel@unistra.fr
´ ´
b
c
´
Institut de Chimie et Procedes pour l’Energie, l’Environnement et la Sante
´
´
(ICPEES), Universite de Strasbourg, Ecole Europeenne de Chimie,
`
´
Polymeres et Materiaux, 25 rue Becquerel, 67087 Strasbourg, France
´
Laboratoire ICube, Universite de Strasbourg, CNRS, 23 rue du Loess,
67037 Strasbourg, France
We have now designed dye 6 (Scheme 1) based on bisthio-
phene–benzothiadiazole modules according to the push–pull–push
† Electronic supplementary information (ESI) available. See DOI: 10.1039/
c4nj00294f
3644 | New J. Chem., 2014, 38, 3644--3653
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2014

View Article Online
Paper
NJC
Scheme 1 General synthetic sketch for the preparation of dye 6.
BODIPY format associated to the building of highly coloured dyes. and excitation spectra were obtained by using a HORIBA JOBIN
This design principle popularized earlier is attractive due to its YVON FLUOROMAX 4. All fluorescence spectra were corrected. The
modular nature. In the present design the hole transporting fluorescence quantum yield (Fexp) was calculated using eqn (1).
subunits were directly attached to the BODIPY core avoiding
unsaturated (alkyne or vinyl) linkers. The novel dye 6 was prepared
as shown in Scheme 1.
I ODRef Z2
IRef OD ZRef
Fexp ¼ FRef
(1)
2
Here, I denotes the integral of the corrected emission spectrum, OD
is the optical density at the excitation wavelength and Z is the
refractive index of the medium. The reference systems used were
2. Experimental part
2.1. General methods
rhodamine 6G (F
= 0.78) in air equilibrated water and
tetramethoxydiisoindomethene-difluoroborate (F = 0.51). Lumines-
All reactions were performed under an atmosphere of dried cence lifetimes were measured on an Edinburgh Instrument
argon using standard Schlenk tube techniques. All chemicals spectrofluorometer equipped with a R928 photomultiplier and
were used as received from commercial sources unless stated a PicoQuant PDL 800-D pulsed diode connected to a GwInstect
otherwise. THF was distilled from sodium and benzophenone GFG-8015G delay generator. No filter was used for the excitation.
1
under an Ar atmosphere. H NMR (400.1 MHz) and 13C NMR Emission wavelengths were selected using a monochromator. Life-
(100.5 MHz) spectra were recorded at room temperature (rt) on times were deconvoluted with the FS-900 software using a light-
a Bruker Advance 400 MHz spectrometer, 1H NMR (300.1 MHz) scattering solution (LUDOX) for instrument response.
and 13C NMR (75.5 MHz) on a Bruker Advance 300 MHz
Potentials were determined by cyclic votammetry in deoxy-
spectrometer, H NMR (200.1 MHz) and 13C NMR (50.5 MHz) genated CH2Cl2 solutions, containing 0.1 M TBAPF6, at a solute
on a Bruker Advance 200 MHz spectrometer using perdeu- concentration range of ca. 1 mM and at rt. Potentials are given
teriated solvents as internal standards. Chromatographic versus the saturated calomel electrode (SCE) and standardized vs.
purifications were performed using silica gel (40–63 mm). TLC ferrocene (Fc) as the internal reference assuming that E1/2 (Fc/Fc+) =
was performed on silica gel plates coated with a fluorescent +0.38 V (DEp = 60 mV) vs. SCE. The error in half-wave potential is
1
indicator.
ꢀ10 mV. Where the redox processes are irreversible, the peak
Absorption spectra were recorded on a Schimadzu UV-3000 potentials (Eap or Ecp) are quoted. All reversible redox steps result
absorption spectrometer. The steady-state fluorescence emission from one-electron processes unless otherwise specified.
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2014
New J. Chem., 2014, 38, 3644--3653 | 3645

View Article Online
NJC
Paper
2.2. Preparative work
dichloromethane–pentane to obtain a purple powder (102 mg,
83%). H NMR (CDCl3, 400 MHz): d 2.48 (s, 3H), 3.41 (s, 6H),
1
Materials. Compounds 1,25 24 and 526 were synthesized
according to the indicated literature procedure.
3.59–3.61 (m, 4H), 3.73–3.75 (m, 4H), 3.89–3.91 (m, 4H),
3
3
4.30–4.32 (m, 4H), 6.34 (d, J = 4.1 Hz, 2H), 6.82 (d, J = 4.2 Hz,
2H), 6.95 (d, 3J = 4.3 Hz, 2H), 7.32 (d, 3J = 7.7 Hz, 2H), 7.43 (d, 3J =
7.8 Hz, 2H), 7.63 (d, 3J = 7.5 Hz, 2H), 7.83 (d, 3J = 4.0 Hz, 2H), 7.87
(d, 3J = 7.9 Hz, 2H), 8.13 (d, 3J = 4.1 Hz, 2H), 8.38 (d, 3J = 4.2 Hz,
2H). 13C NMR (CDCl3, 75 MHz): d 21.6, 29.8, 59.3, 69.6, 71.0, 72.1,
73.2, 106.7, 121.0, 123.9, 124.2, 126.1, 126.4, 126.8, 127.2, 129.2,
130.0, 130.8, 131.9, 135.5, 140.4, 152.6, 152.8, 167.3. EIMS, m/z
(%): 1114.1 ([M], 100). Anal. calcd for C54H45BF2N6O6S6 (Mr =
1115.17): C, 58.16; H, 4.07; N, 7.54; found: C, 57.94; H, 3.82;
N, 7.38.
Compound 3. Compound 1 (0.9 g, 3.03 mmol), compound 2
(3.3 g, 12.8 mmol) and THF (25 mL) were placed in a 100 mL
flask. The solution was degassed with argon for 20 minutes and
then [Pd(PPh3)4] (0.35 g, 0.3 mmol) was added. The solution
was stirred and heated to reflux for one day. After cooling down,
the mixture was washed with water and extracted with ethyl
acetate. Organic extracts were dried over anhydrous magne-
sium sulfate or absorbent cotton, and the solvent was removed
on a rotator evaporator. The crude product was purified by
column chromatography (silica gel, solvent: 80/20 cyclohexane–
AcOEt). The product was then recrystallized in acetone to obtain
1
a red powder (570 mg, 43%). H NMR (acetone d6, 300 MHz):
2.3. Device preparation
3
3
d 3.32 (s, 3H), 3.54 (t, J = 5.2 Hz, 2H), 3.68 (t, J = 5.2 Hz, 2H),
Bulk heterojunction devices were elaborated using dye 6 as the
electron donor and PC61BM or PC71BM as the electron acceptor.
Chloroform and chlorobenzene (CB) were used as solvents with
typical dye 6 concentrations ranging from 3 to 10 mg mLꢁ1. For
all CB mixtures, DIO with a volumic concentration ranging
from 0.3 to 0.5% were tested. The standard device structure was
as follows: ITO/PEDOT:PSS(B40 nm)/active layer/Al(B120 nm).
For the most promising results, the cathode was replaced by a
Ca(20 nm)/Al(120 nm) bilayer. Indium tin oxide coated glass
with a surface resistance lower than 20 O sqꢁ1 was used as a
transparent substrate. Substrates were cleaned sequentially by
ultrasonic treatment in acetone, isopropyl alcohol, and deio-
nized water. After an additional cleaning for 30 minutes under
an ultra-violet generated ozone, a highly conductive polyethy-
lene dioxythiophene:polystyrene-sulphonate PEDOT:PSS was
spin coated (1500 rpm: 40 nm) from an aqueous solution and
dried for 30 minutes at 120 1C under vacuum before being
transferred to the nitrogen filled glove box. The chloroform or
chlorobenzene molecule/PCBM solutions were stirred for at
least 24 hours at 50 1C before spin-coating. The relative dye
6/PCBM weight ratio was varied from 1/1 to 1/2. The active layer
spin coating conditions (one fast and one slow program) were
also varied to change the active layer thickness. Both spin
coating programs included two steps with the following para-
meters: step 1 (2000 rpm, 120 seconds and 600 rpm sꢁ1) and
step 2 (2500 rpm, 60 seconds and 600 rpm sꢁ1) for the ‘‘fast’’
3
3
3
3.87 (t, J = 4.6 Hz, 2H), 4.34 (t, J = 4.6 Hz, 2H), 6.46 (d, J =
4.2 Hz, 1H), 7.25 (dd, 3J = 5.1 Hz, 3J = 3.7 Hz, 1H), 7.63 (dd,
3J = 5.1 Hz, 4J = 1.0 Hz, 1H), 7.84 (d, 3J = 7.7 Hz, 1H), 7.82 (d, 3J =
4.2 Hz, 1H), 7.99 (d, 3J = 7.7 Hz, 1H), 8.17 (dd, 3J = 3.7 Hz,
4J = 1.0 Hz, 1H). 13C NMR (acetone d6, 75 MHz): d 59.6, 70.7, 72.0,
73.4, 74.7, 105.5, 125.3, 126.1, 126.9, 127.5, 127.7, 128.0, 128.5,
128.8, 129.4, 140.8, 153.8, 154.0, 168.9. EIMS, m/z (%): 418.0
([M], 100). Anal. calcd for C19H18N2O3S3 (Mr = 418.55): C, 54.52;
H, 4.33; N, 6.69; found: C, 54.33; H, 4.47; N, 6.42.
Compound 4. 0.426 mL of a solution of n-BuLi at 2.5 M in
hexane were added dropwise in a solution of diisopropylamine
(0.164 mL, 1.17 mmol) in 10 mL of distilled THF at ꢁ78 1C. The
solution was warmed up to ꢁ40 1C for 30 min and then cooled
again at ꢁ78 1C. A solution of compound 3 (0.41 g, 0.97 mmol)
in 5 mL of THF was then added at ꢁ78 1C. The mixture was
stirred at room temperature for 12 h. The reaction was
quenched with a saturated solution of NH4Cl. The mixture
was washed with water and extracted with ethyl acetate.
Organic extracts were dried over anhydrous magnesium sulfate
or absorbent cotton, and the solvent was removed on a rotator
evaporator. No further purification is possible for this stannic
compound and the conversion yield was estimated by 1H NMR
around 35%. 1H NMR (CDCl3, 300 MHz): d 0.44 (s, 9H), 3.41
(s, 3H), 3.61 (t, 3J = 4.9 Hz, 2H), 3.74 (t, 3J = 4.5 Hz, 2H), 3.89 (t, 3J =
4.9 Hz, 2H), 4.31 (t, 3J = 4.6 Hz, 2H), 6.33 (d, 3J = 4.2 Hz, 2H), 7.28
3
3
program and step 1 (1200 rpm, 60 seconds and 200 rpm sꢁ1
)
(d, J = 3.5 Hz, 1H), 7.63 (d, J = 7.6 Hz, 1H), 7.81 (m, 2H), 8.14
3
and step 2 (2000 rpm, 120 seconds and 200 rpm sꢁ1) for the
‘‘slow’’ one. Finally, a 120 nm thick aluminum layer with or
without a pre-20 nm thick calcium layer was thermally evapo-
rated and used as the cathode. The device active area was
12 mm2, while each sample included four independent diodes.
Current versus voltage (I–V) characteristics were measured
using a source measurement unit Keithley 2400 under dark
conditions and under AM1.5G (100 mW cmꢁ2) illumination.
The standard illumination was provided by a Sun 3000 (ABET
Technologies) solar simulator and the illumination power was
set using a calibrated silicon solar cell. After IPEDOT deposition
procedure made under air, the photovoltaic device elaboration
and their characterizations were performed in glove boxes.
(d, J = 3.5 Hz, 1H).
Compound 6. Compound 4 (160 mg, 0.275 mmol), com-
pound 5 (48 mg, 0.110 mmol), distilled toluene (10 mL) and
tri(o-tolyl)phosphine (7 mg, 21 mmol) were added to a Schlenck
tube. The solution was degassed with argon for 45 min.
[Pd2(dba)3] (5 mg, 5.5 mmol) was then added and the mixture
was heated at 110 1C for 3 h. After cooling down, the mixture
was washed with water and extracted with dichloromethane.
Organic extracts were dried over anhydrous magnesium sulfate
or absorbent cotton, and the solvent was removed on a rotator
evaporator. The crude product was purified by column chro-
matography (silica gel, solvent: 90/10 dichloromethane–
ethyl acetate). The obtained product was then recrystallised in
3646 | New J. Chem., 2014, 38, 3644--3653
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2014

View Article Online
Paper
NJC
3. Results and discussion
3.1. Synthesis and characterisation
Molecules 1,25 24 and 526 were prepared according to proce-
dures found in the literature. The first step consists of a Stille
coupling reaction between 1 and 2. This reaction is carried out
under standard conditions using palladium(0) catalysts and
leads to trimer 3 with 43% yield. In a second step the stannic
group is introduced by the formation of the lithiated thiophene
in the last free a-position of thiophenes thanks to LDA, which
then reacts with trimethyltin chloride leading to compound 4.
The obtained crude product is not purified, as stannic com-
pounds are not quite stable on silica gel, and the conversion
is estimated by 1H NMR spectroscopy (around 35%). The
resulting mixture is then engaged in another Stille coupling
reaction with the dibromo-BODIPY 5. The fraction of 3 that did
not react to form the stannic compound 4 is recovered and can
be recycled. The target compound 6 is obtained in 83% yield.
The proton NMR is a diagnostic tool for the desired molecule.
In particular, the proton NMR spectra display ten proton
patterns in the aromatic window as would be expected from
a first order spectrum and prove that the molecule has a
symmetry axis bisecting the tolyl and the BF2 fragments. The
short polyoxoethylene chains also display four characteristic
patterns in the 3.5 to 4.5 ppm range. All other routine analysis
(13C, EI-MS and elemental analysis) are in keeping with the
molecular assignment.
Fig. 1 Absorption (orange trace), emission (green trace) and excitation
(dashed line) spectra of 3 in THF.
The rationale to use polyoxoethylene short chains is based
on previous observations: (i) a highly pure material could be
prepared due to the polarity imported by the oxygen atoms;
(ii) high solubility in most common solvents and (iii) good
filmability and increase of dye stability.
Fig. 2 Absorption (orange trace), emission (green trace) and excitation
(dashed line) spectra of 6 in THF.
3.2. Spectroscopic characterization
The absorption, emission and excitation spectra of the trimer
thiophene–benzothiadiazole–thiophene (T–Bz–T) 3 and the
The absorption of the final compound 6 (Fig. 2) shows
final compound 6 are shown in Fig. 1 and 2. The absorption three peaks. We can notice the presence of the structured
spectrum of 3 shows two intense peaks. The first absorption at peak at around 320 nm due to the presence of thiophene
around 320 nm corresponds to a p–p* transition with an fragments, overlapping with (S0 - S2) of the BODIPY.27 The
extinction coefficient e of about 28 000 Mꢁ1 cmꢁ1. The second peak corresponding to the CT absorption band is now found
absorption peak, at around 479 nm, is broader and is likely a at 537 nm and its e value is approximately twice as high as the
charge transfer band (CT) due to the presence of a strong donor one found for 3 (e about 50 000 Mꢁ1 cmꢁ1). The absorption
(the thiophene subunit) and a strong acceptor (the benzothia- spectrum also shows an intense peak (e around 90 000 Mꢁ1 cmꢁ1
diazole moiety) in the trimer. The e value of the CT band is at 737 nm corresponding to the p–p* transition (S0 - S1) of
weaker than the p–p* transition one by about 15 000 Mꢁ1 cmꢁ1
the BODIPY unit in keeping with previous spectral analysis.
)
.
Excitation at 450 nm gave the emission spectrum with a The weak Stokes shift (715 cmꢁ1) and the short lifetime
maximum at 631 nm, this emission being independent of the (around 1 ns) confirm that the excited state is likely a singlet
excitation wavelength (Fig. 1). The excitation spectrum matches with little reorganization in the excited state. The quantum
the absorption spectrum, a result excluding aggregation or yield is this time around 6%. Once more, there was no
the presence of impurities. The large Stokes shift (about influence of the solvent (THF, dichloromethane, toluene) on
5000 cmꢁ1) and a long lifetime (about 10 ns) confirm the CT the fluorescence characteristics and the excitation spectrum
nature of this transition. However, there was no influence of the matches fairly well with the absorption spectrum in all cases.
solvent (THF, dichloromethane, toluene) on the spectroscopic Notice that, relative to compound 5, the non-radiative rate
characteristics of 3 and its quantum yield was around 62%. The constant is larger than the radiative rate constant, a result in
radiative rate constant is larger than the non-radiative rate keeping with the energy gap law for dyes absorbing at low
constant confirming the efficiency of the process (Table 1).
excitation energy.28
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2014
New J. Chem., 2014, 38, 3644--3653 | 3647

View Article Online
NJC
Paper
Table 1 Spectroscopic data for 3 and 6 in THF
e (Mꢁ1 cmꢁ1
)
lem (nm) FF (lex, nm) t (ns) kra (107 sꢁ1
)
knr (107 sꢁ1
b
)
DSS (cmꢁ1
)
FWHMd (cmꢁ1
)
c
Compd labs (nm)
3
6
320 (479 for CT band) 15 500
631
778
0.62 (@450) 10.5
5.90
4.55
3.62
71.8
5030
720
4200
1900
737
93 800
0.06 (@680)e
1.3
kr
1 ꢁ FF
.
a
b
c
d
e
kr was calculated from FF
¼
¼ krt. knr was calculated from knr
¼
Stokes shifts. Full width at half maximum. Using
kr þ knr
kr
tetramethoxydiisoindomethene-difluoroborate (F = 0.51) as the reference.
Fig. 4 Cyclic voltammetry spectra of 5 (black trace), 3 (orange trace) and
6 (green trace).
Fig. 3 Absorption spectra of 6 in solution in THF (orange trace) and in a
thin film (black trace).
Table 2 Experimentally determined energetic levels for compound 6
The thin film absorption spectrum of 6 (Fig. 3) shows a broad
peak at 798 nm with a bathochromic shift of about 60 nm
compared to the spectrum in solution and suggests the existence
of strong p–p stacking interactions in the solid state.29 The low
energy absorption edge leads to an optical band gap of 1.37 eV,
which is low in comparison to most reported organic low band-
gap materials and matches the optimal value predicted by
Queisser et al.30 for homojunction solar cells.
HOMO level
LUMO level
Electrochemical gap
1.35 eV
Optical gap
1.37 eV
ꢁ5.43 eV
ꢁ4.08 eV
constructed with bis-thienylhexyl side arms (Fig. 5).4 Compar-
ison with TB2 is motivated by: (i) the similar molecular struc-
ture; (ii) its high efficiency in photon to electricity conversion in
solar cells; (iii) its ambipolar charge mobility in thin films.
The tolyl group is less twisted in the case of dye 6 (551 for 6
and 901 for TB2) relative to the BODIPY core due to the absence
of methyl groups in positions 1 and 7 of the BODIPY core. Both
T–Bz–T segments (dye 6) and T–T segments (TB2) show a high
degree of planarity but are slightly more twisted relative to the
BODIPY core for dye 6 (201) than for TB2 (101). This small twist
angle difference is most likely due to the different links
between the BODIPY core and the thiophene-based segments
used for both dyes and has a negligible effect on the conjuga-
tion. The direct linking in compound 6 induces more steric
crowding with the proton in the 2,6-positions relative to the
double bond case (TB2) where the steric constraint is partially
released. The most pronounced conformational difference
between the two dyes is the orientation angle between the
two thiophene-based segments that is as high as 1201 for dye
6 while it is more than twice smaller for TB2. This difference is
again mainly due to the single C–C bond between T–Bz–T and
3.3. Redox properties
Cyclic voltammetry data obtained for 3, 5, and 6 in solution are
shown in Fig. 4. The spectrum of 5 shows only a reversible
reduction at ꢁ0.79 V and no oxidation in the accessible redox
window (up to 1.7 V), while the trimer 3 shows one reversible
reduction at ꢁ1.37 V and one irreversible oxidation at +0.95 V.
The voltammogram of 6 corresponds to a combination of both
moieties: it shows three reversible reductions at ꢁ0.80 V,
ꢁ1.32 V and ꢁ1.40 V and one irreversible oxidation at
+0.71 V. From the splitting of the reduction wave at ꢁ1.37 V
we may infer that the two T–Bz–T side arms are not equivalent
toward reduction processes. The HOMO level is estimated to be
ꢁ5.43 eV whereas the LUMO is around ꢁ4.08 eV. The electro-
chemical band-gap is thus equal to 1.35 eV, a value in keeping
with the optical gap determined above (1.37 eV in Table 2).
3.4. DFT calculations
Density functional theory (DFT) calculations have been per- the BODIPY core (dye 6). We expect that this larger angle has a
formed using SPARTAN 10 at the B3LYP/6-311+G* level of strong impact on the volume occupied by the dyes.
theory in vacuum on dye 6 and are compared below those
The calculated HOMO and LUMO orbitals are more deloca-
obtained previously on another BODIPY dye named TB2 lized for dye 6 than for TB2 (Fig. 6 and 7). Even though the
3648 | New J. Chem., 2014, 38, 3644--3653
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2014

View Article Online
Paper
NJC
Fig. 7 HOMO (left) and LUMO (right) orbitals for TB2 as calculated by DFT
in vacuum.
voltammetry. The DFT LUMO level for dye 6 is found to be
0.3 eV above the one of TB2 while cyclic voltammetry gave
LUMO values of ꢁ4.08 eV and ꢁ3.86 eV for dye 6 and TB2,
respectively. Also, the DFT HOMO level for dye 6 is found to be
0.1 eV below that of TB2, which is again close to the measured
values (ꢁ5.43 eV and ꢁ5.32 eV for dye 6 and TB2, respectively).
In a previous study,4 the LUMO level of TB2 and PC61BM
were measured by cyclic voltammetry at ꢁ3.86 and ꢁ4.09 eV,
respectively. Based on the present DFT calculations and cyclic
voltammetry measurements, the energy offset between the
LUMO level of dye 6 and the LUMO level of PC61BM may hinder
exciton dissociation if 6 is to be used as an electron donor
material in donor–PC61BM blends. Yet, the relatively large span
in reported PC61BM LUMO level values may reflect the influ-
ence of the degree of fullerene aggregation on the LUMO value,
which varies with blend composition and processing condi-
tions and is difficult to anticipate. Also, since a deep HOMO
level is a prerequisite for air stable devices and a high open
circuit voltage, it remains interesting to investigate the photo-
voltaic properties of 6–PC61BM blends and to evaluate the
impact of the exciton dissociation bottleneck as well as its
potential in chemically stable high performance solar cells
(see below).31
Fig. 5 Conformation of dye TB2 (top) and 6 (bottom) as calculated by
DFT in vacuum.
3.5. AFM measurements
The morphology of the dye 6–PCBM films was investigated by
AFM. Two representative topography images of a film including a
1/2 ratio of dye 6–PC71BM (see ESI†) before and after annealing
show no clear morphology differences indicating that the thermal
Fig. 6 HOMO (top) and LUMO (bottom) orbitals for dye 6 as calculated by
DFT in vacuum.
absolute values of the frontier energy levels obtained by DFT annealing step does not deeply change the microphase separa-
(given in the ESI†) may not be accurate, energy level shifts tion. The low roughness of the films (maximum height value
induced by changes in the molecular structure are more reli- around 1.4 nm) is worth noting, which does not change after
able. For dye 6 and TB2, the differences in frontier orbital thermal annealing confirming that the BODIPY derivative does
energy levels are close to those found experimentally by cyclic not crystallize during the thermal annealing.
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2014
New J. Chem., 2014, 38, 3644--3653 | 3649

View Article Online
NJC
Paper
3.6. Mobility measurements
Charge transport is another important issue that needs to be
addressed when designing new molecules for photovoltaic appli-
cations. We therefore used dye 6 as a semiconducting layer in thin
film transistors and extracted the field-effect charge carrier mobi-
lities from the transfer characteristics in the saturation regime on
pristine films (the device characteristics are given in the ESI†).
Although field-effect transistors probe charge transport under very
specific conditions (parallel to the dielectric interface and at high
charge carrier densities), the results can give insight into the
correlation between the molecular structure and charge transport
properties.32 In particular, the field-effect mobilities can be com-
pared to those of TB2, which we consider as a reference.
Interestingly, compound 6 exhibits a rather uncommon
ambipolar nature with well-balanced hole and electron mobi-
lities (Fig. 8 and Fig. S2, ESI†). This behavior is not surprising
given the deep LUMO level of this dye and considering previous
observations done on TB2, which exhibited ambipolar charge
transport.4 The electron and hole mobilities of 6 are in the
lower range of 10ꢁ4 cm2 Vꢁ1 sꢁ1 which is significantly lower
than those observed for TB2, but does not necessarily impede
its utilization as a donor material in a BHJ solar cell.
Fig. 9 Current density versus voltage in the dark (open symbols) and
under standard AM1.5G (100 mW cmꢁ2) illumination (closed symbols) for
the last conditions of Table 3.
the experimental conditions can be found in the ESI†). Typical
current–voltage curves are displayed in Fig. 9. Surprisingly, the
elaboration process has only a minor impact on the measured
power conversion efficiency (PCE), which remains close to 1.2%
independent of the solvent nature and the fullerene derivatives
used. The major performance limiting factors are a low fill factor
(close to 33%) and rather low short-circuit current density. The
latter is much below expectation if we consider the rather broad
absorption band and high extinction coefficient of 6. The fill factor
in turn is linked to the low parallel resistance (Rshunt) measured
under illumination (inverse slope at short-circuit current). Increas-
ing the active layer thickness by using more concentrated solutions
to enhance light absorption did lead only to a weak increase in
device performances (similar fill factors but slightly higher short-
circuit currents were obtained) (see Table S2, ESI†).
3.7. Photovoltaic device elaboration and characterization
Bulk heterojunction devices were elaborated with a standard ITO/
PEDOT:PSS/active layer/Al or Ca–Al structure. The active layer has
been deposited by spin coating from organic solutions of com-
pound 6 and PCBM derivatives. Several conditions, including
different solvents, different concentrations, the use of two different
PCBM derivatives (either PC61BM or PC71BM) and variations in the
donor/acceptor weight ratio, have been tested. Some representative
device properties are summarized in Table 3 (more details about
The low FF (or low parallel resistance) could originate from
either fast charge carrier recombination (geminate or non-
geminate) or field-assisted exciton dissociation. In the former
case, increasing the electric field by reverse biasing the device
would lower the sweep-out time, minimize the loss of charge
carriers by recombination and give rise to higher photocurrents
(defined as the difference between light and dark current at a
given voltage). The weak thickness dependence of the fill factor
and shunt resistance points out that charge extraction may not
be the major bottleneck, despite the relatively low charge
carrier mobilities. This conclusion is also in line with the weak
offset between the LUMO levels of dye 6 and PCBM, which
suggests the current D/A interface energetics to be unfavorable
to exciton dissociation. On the other hand, the open-circuit
Fig. 8 Transfer characteristics in the saturation regime obtained for
electrons (left) and holes (right) on a pristine film. ID is the drain current
and UG is the gate voltage. The drain voltage was 100 V and ꢁ100 V for
electrons and holes, respectively.
Table 3 Selected thin-film compositions and deposition conditions with associated OPV performances
Ratio
(6/PCBM)
BODIPY dye
Jsc
Acceptor
concentration (mg mLꢁ1
)
Solvent
Cathode
Voc (V)
(mA cmꢁ2
)
FF
PCE (%)
Rshunt (O)
PC61BM
PC61BM
PC71BM
1/1.5
1/2
1/2
5
3
10
CHCl3
Al
Ca/Al
Al
0.69
0.62
0.61
5.2
5.8
7.4
0.32
0.35
0.28
1.15
1700
1550
1250
CHCl3
1.26b
1.26c
CB + 0.3 vol% DIOa
a
b
c
CB denotes chlorobenzene and DIO denotes 1,8-diiodooctane. Annealed for 10 minutes at 60 1C prior to cathode deposition. Annealed for 10
minutes at 80 1C and at 100 1C for 3 minutes.
3650 | New J. Chem., 2014, 38, 3644--3653
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2014

View Article Online
Paper
NJC
Table 4 Selected OPV data for analogous BODIPY-based molecules and polymers depicted in Fig. 10
Compd
Device
Jsc (mA cmꢁ2
)
Voc (V)
FF
Z (%)
mh (V cmꢁ2 sꢁ1
)
me (V cmꢁ2 sꢁ1
)
6
ITO/PEDOT:PSS/6:PC61BM/Ca/Al
ITO/PEDOT:PSS/TB2:PC61BM/Ca/Al
ITO/Ca/A:PC71BM/MoO3/Ag
5.8
14.3
8.3
7.5
6.9
4.8
2.65
7.87
0.62
0.70
0.99
0.93
0.74
0.80
0.88
0.31
0.35
0.47
0.40
0.37
0.38
0.51
0.26
0.39
1.26
4.7
3.22
2.56
2.17
2.0
B10ꢁ4 a
B10ꢁ4 a
B10ꢁ3 a
2.3 ꢂ 10ꢁ4 b
6.0 ꢂ 10ꢁ5 b
—
TB24
A33
B33
C34
D35
E36
F37
B10ꢁ3 a
4.1 ꢂ 10ꢁ6 b
1.2 ꢂ 10ꢁ6 b
9.7 ꢂ 10ꢁ5
3.0 ꢂ 10ꢁ7
—
ITO/Ca/B:PC71BM/MoO3/Ag
ITO/PEDOT:PSS/C:PC61BM/Al
ITO/PEDOT:PSS/D:PC61BM/Al
ITO/PEDOT:PSS/E:PC61BM/Ca/Al
ITO/PEDOT:PSS/F:PC71BM/Al
—
—
0.62
0.95
1.5 ꢂ 10ꢁ5 c
7.7 ꢂ 10ꢁ5 c
a
b
c
Values obtained by OFET measurements on pure materials. Values obtained by SCLC measurements on blends with PC71BM. Values obtained
by TOF measurements on pure materials.
voltage Voc varies between 0.6 and 0.7 V, which is well below the been observed in both categories and the lead is held by TB2
value expected according to the frequently observed linear which also display the highest photon to current conversion.
dependence of Voc on the difference between the acceptor
LUMO and the donor HOMO level in BHJ devices (assuming
a PCBM LUMO level of ꢁ4 eV and an energy loss DE due to
recombination of B0.3 V leads to a value well above 1 V).29
High recombination rates or poor charge generation are possi-
ble reasons for this discrepancy. Distinguishing between both
processes mentioned above is not trivial and requires utiliza-
tion of advanced time-resolved measurements. These however
lie beyond the scope of the present article.
This behaviour is surely due to the deep LUMO level usually
measured in BODIPY based materials. However, a further
comparison on the basis of values found in the literature
(and reported in the above table) is hazardous as they are
provided by different techniques, including space charge
limited current (SCLC) on blends, time of flight (TOF) or OFET
on pure materials. In most cases the short-circuit current (Jsc) is
rather high reaching 14.3 mA cmꢁ2 in the best case but the fill
factor (FF) remains modest (Table 4). Further comparison
between those materials must be done with care because the
structures of devices (standard versus inverted cells), the metallic
3.8. Comparison with related BODIPY based devices
Upon comparison of BODIPY based small molecules and poly- electrodes, the interfacial layers and the device active areas are
meric materials it appears that ambipolar charge transport has significantly different.
Fig. 10 Chemical structures of related BODIPY frameworks studied in bulk heterojunction solar cells.
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2014
New J. Chem., 2014, 38, 3644--3653 | 3651

View Article Online
NJC
Paper
5 R. Fitzner, E. Reinold, A. Mishra, E. Mena-Osteritz,
4. Conclusion
¨
H. Zielke, C. Korner, K. Leo, M. Riede, M. Weil,
We have synthesized a new deep-purple-grey 4,4-difluoro-4-
bora-3a,4a-diaza-s-indacene dye (BODIPY) incorporating two
thiophene–benzothiadiazole–thiophene units bridging an
unsubstituted BODIPY core at the wings. In the electronic
absorption spectra, an effective charge transfer band appears
at 537 nm relative to 479 nm in the thiophene–benzothiadia-
zole–thiophene module due to interaction between the strong
‘‘push’’ thiophene units at the periphery and the central
benzothiadiazole ‘‘pull’’ unit. This is an interesting approach,
which also contributes to the increase of photon absorption
either in solution or in the thin film. This strong absorption
and redox activity of this dye enable the preparation of bulk
heterojunction with thin films blended with [6,6]phenylC61or71-
butyricacidmethylester PCnBM (n = 61 or 71). Promising effi-
ciencies around 1.25% were obtained, these reproducible
efficiencies are weakly dependent on the composition, thick-
ness of the films and nature of the electrodes. The synthetic
routes outlined here provide the scope to further tune the
performance of the solar cells by post-functionalization of the
2,6 or (b,b0-pyrrolic positions) with electron donating modules
suitable to increase the driving force for photoinduced electron
injection in the PCBM acceptor.
O. Tsaryova, A. Weiss, C. Uhrich, M. Pfeiffer and
¨
P. Bauerle, Adv. Funct. Mater., 2011, 21, 897–910.
6 (a) B. Walker, A. B. Tamayo, X.-D. Dang, P. Zalar, J. H. Seo,
A. Garcia, M. Tantiwiwat and T.-Q. Nguyen, Adv. Funct.
Mater., 2009, 19, 3063–3069; (b) T. Bura, N. Leclerc,
´ ˆ
R. Bechara, P. Leveque, T. Heiser and R. Ziessel, Adv. Energy
Mater., 2013, 3, 1118–1124.
7 (a) G. Wei, S. Wang, K. Sun, M. E. Thompson and
S. R. Forrest, Adv. Energy Mater., 2011, 2, 184–187;
(b) A. Ajayaghosh, Acc. Chem. Res., 2005, 38, 449–459.
8 W. W. H. Wong, T. B. Singh, D. Vak, W. Pisula, C. Yan,
X. Feng, E. L. Williams, K. L. Chan, Q. Mao, D. J. Jones,
C.-Q. Ma, K. Mu¨llen, P. Bau¨erle and A. B. Holmes,
Adv. Funct. Mater., 2010, 20, 927–938.
9 N. M. Kronenberg, M. Deppisch, F. Wu¨rthner,
H. W. A. Ladermann, K. Deing and K. Meerholz, Chem.
Commun., 2008, 6489–6491.
10 H. Bu¨rckstu¨mmer, E. V. Tulyakova, M. Deppisch,
¨
M. R. Lenze, N. M. Kronenberg, M. Gsanger, M. Stolte,
K. Meerholz and F. Wu¨rthner, Angew. Chem., Int. Ed.,
2011, 4506, 11628–11632.
11 Y. Sun, G. C. Welch, W. L. Leong, C. J. Takacs, G. C. Bazan
and A. J. Heeger, Nat. Mater., 2011, 11, 44–48.
12 R. Ziessel, C. R. Chim., 2007, 10, 622–629.
13 (a) T. Rousseau, A. Cravino, T. Bura, G. Ulrich, R. Ziessel and
J. Roncali, Chem. Commun., 2009, 1673–1675; (b) T. Bura,
Acknowledgements
We thank the Centre National de la Recherche Scientifique
(CNRS), the Engineering School of Chemistry (ECPM) and
Rhin-Solar supported by the European Fund for Regional Devel-
opment (FEDER) in the framework of the Programme INTERREG
IV Upper Rhine, Project nr C25 for financial support of this work.
N. Zimmermann from ICube in Strasbourg is kindly acknowl-
edged for his help in the photovoltaic device elaboration.
´ ˆ
N. Leclerc, S. Fall, P. Leveque, T. Heiser and R. Ziessel, Org.
Lett., 2011, 22, 6030–6033.
14 A. Sutter, P. Retailleau, W.-C. Huang, H.-W. Lin and
R. Ziessel, New J. Chem., 2014, 38, 1701–1710.
15 (a) S. Erten-Ela, M. D. Yilmaz, B. Icli, Y. Dede, S. Icli and
E. U. Akkaya, Org. Lett., 2008, 10, 3299–3302; (b) S. Kolemen,
Y. Cakmak, S. Erten-Ela, Y. Altay, J. Brendel, M. Thelakkat and
E. U. Akkaya, Org. Lett., 2010, 12, 3812–3815; (c) S. Kolemen,
O. A. Bozdemir, Y. Cakmak, G. Barin, S. Erten-Ela, M. Marszalek,
References
¨
J.-H. Yum, S. M. Zakeeruddin, M. K. Nazeeruddin, M. Gratzel
and E. U. Akkaya, Chem. Sci., 2011, 2, 949–954.
1 T. M. Figueira-Duarte and K. Mu¨llen, Chem. Rev., 2011,
111, 7260.
16 D. Kumaresan, R. P. Thummel, T. Bura, G. Ulrich and
R. Ziessel, Chem. – Eur. J., 2009, 15, 6335–6339.
17 C. Qin, A. Mirloup, N. Leclerc, A. Islam, A. El-Shafei, L. Han
and R. Ziessel, Adv. Energy Mater., 2014, DOI: 10.1002/
aenm.201400085.
18 I. Culloch, M. Heeney, C. Bailey, K. Genevicius,
I. MacDonald, M. Shkunov, D. Sparrowe, S. Tierney,
R. Wagner, W. Zhang, M. L. Chabinyc, R. J. Kline,
M. D. McGehee and M. F. Toney, Nat. Mater., 2006, 5, 328.
2 (a) G. Yu, J. Gao, J. C. Hummelen, F. Wudl and A. J. Heeger,
Science, 1995, 270, 1789; (b) T. S. van der Poll, J. A. Love,
T.-Q. Nguyen and G. C. Bazan, Adv. Mater., 2012, 24, 3646;
(c) Y. Sun, G. C. Welch, W. L. Leong, C. J. Takacs, G. C. Bazan
and A. J. Heeger, Nat. Mater., 2012, 11, 44; (d) Z. Li, G. He,
X. Wan, Y. Liu, J. Zhou, G. Long, Y. Zuo, M. Zhang and
Y. Chen, Adv. Energy Mater., 2012, 2, 74; (e) J. Y. Zhou,
X. J. Wan, Y. S. Liu, Y. Zuo, Z. Li, G. R. He, G. K. Long, W. Ni,
C. X. Li, X. C. Su and Y. S. Chen, J. Am. Chem. Soc., 2012,
134, 16345.
´
19 J. Liu, R. Zhang, G. Sauve, T. Kowalewski and
R. D. McCullough, J. Am. Chem. Soc., 2008, 130, 13167.
20 I. Saka, R. Zhang, J. Liu, D.-M. Smilgies, T. Kowalewski and
R. D. McCullough, Chem. Mater., 2010, 22, 4191.
21 J. Zhou, X. Wan, Y. Liu, G. Long, F. Wang, Z. Li, Y. Zuo, C. Li
and Y. Chen, Chem. Mater., 2011, 23, 4666–4668.
3 (a) G. Ulrich, R. Ziessel and A. Harriman, Angew. Chem., Int.
Ed., 2008, 47, 1184–1201; (b) A. Loudet and K. Burgess,
Chem. Rev., 2007, 107, 4891–4932; (c) R. Ziessel, G. Ulrich
and A. Harriman, New J. Chem., 2007, 31, 496–501.
´ ˆ
4 T. Bura, N. Leclerc, S. Fall, P. Leveque, T. Heiser,
22 Y. Liu, X. Wan, F. Wang, J. Zhou, G. Long, J. Tian and
Y. Chen, Adv. Mater., 2011, 23, 5387–5391.
P. Retailleau, S. Rihn, A. Mirloup and R. Ziessel, J. Am.
Chem. Soc., 2012, 134, 17404–17407.
3652 | New J. Chem., 2014, 38, 3644--3653
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2014

View Article Online
Paper
NJC
23 (a) M. Zhang, H. N. Tsao, W. Pisula, C. Yang, A. K. Mishra and 28 M. Bixon, J. Jortner, J. Cortes, H. Heitele and M. E. Michel-
K. Mullen, J. Am. Chem. Soc., 2007, 129, 3472–3473; Beyerle, J. Phys. Chem., 1994, 98, 7289–7299.
(b) H. N. Tsao, D. M. Cho, I. Park, M. R. Hansen, 29 J. Mizuguchi and G. Wooden, Ber. Bunsen-Ges. Phys. Chem.,
A. Mavrinskiy, D. Y. Yoon, R. Graf, W. Pisula, H. W. Spiess 1991, 95, 1264–1274 and references cited therein.
and K. Mu¨llen, J. Am. Chem. Soc., 2011, 133, 2605–2612; 30 W. Shockley and H. J. Queisser, J. Appl. Phys., 1961, 32,
(c) Z. Zhu, D. Waller, R. Gaudiana, M. Morana, 510–519.
D. Muhlbacher, M. Scharber and C. Brabec, Macromolecules, 31 G. Dennler, M. C. Scharber and C. J. Brabec, Adv. Mater.,
2007, 40, 1981–1986; (d) P. M. Beaujuge, H. N. Tsao, 2009, 21, 1323–1338.
M. R. Hansen, C. M. Amb, C. Risko, J. Subbiah, 32 S. Fall, L. Biniek, N. Leclerc, P. Leveque and T. Heiser, Appl.
K. R. Choudhury, A. Mavrinskiy, W. Pisula, J.-L. Bredas, F. So, Phys. Lett., 2012, 101, 123301–123303.
K. Mu¨llen and J. R. Reynolds, J. Am. Chem. Soc., 2012, 134, 33 H.-Y. Lin, W.-C. Huang, Y.-C. Chen, H.-H. Chou, C.-Y. Hsu,
´ ˆ
8944–8957; (e) R. C. Coffin, J. Peet, J. Rogers and G. C. Bazan,
Nat. Chem., 2009, 1, 657–661; ( f ) S. H. Park, A. Roy, S. Beaupre,
J. T. Lin and H.-W. Lin, Chem. Commun., 2012, 48,
8913–8915.
S. Cho, N. Coates, J. S. Moon, D. Moses, M. Leclerc, K. Lee and 34 T. Rousseau, A. Cravino, E. Ripaud, P. Leriche, S. Rihn,
A. J. Heeger, Nat. Photonics, 2009, 3, 297–303.
24 X. Guo, N. Zhou, S. J. Lou, J. W. Hennek, R. P. Ortiz,
A. De Nicola, R. Ziessel and J. Roncali, Chem. Commun.,
2010, 46, 5082–5084.
´
M. R. Butler, P.-L. T. Boudreault, J. Strzalka, P.-O. Morin, 35 B. Kim, B. Ma, V. R. Donuru, H. Liu and J. M. Frechet, Chem.
M. Lecherc, J. T. Lopez Navarrete, M. A. Ratner, L. X. Chen, Commun., 2010, 46, 4148–4150.
R. P. H. Chang, A. Facchetti and T. J. Marks, J. Am. Chem. 36 D. Cortizo-Lacalle, C. T. Howells, S. Gambino, F. Vilela,
Soc., 2012, 134, 18427–18439.
25 Y. Zhou, Q. He, Y. Yang, H. Zhong, C. He, G. Sang, W. Liu,
C. Yang, F. Bai and Y. Li, Adv. Funct. Mater., 2008, 18, 3299–3306.
Z. Vobecka, N. J. Findlay, A. R. Inigo, S. A. J. Thomson,
P. J. Skabara and I. D. W. Samuel, J. Mater. Chem., 2012, 22,
14119–14126.
26 M. Rajeswara Rao, S. M. Mobin and M. Ravikanth, Tetra- 37 S. P. Economopoulos, C. L. Chochos, H. A. Ioannidou,
hedron, 2010, 66(9), 1728–1734.
M. Neophytou, C. Charilaou, G. A. Zissimou, J. M. Frost,
T. Sachetan, M. Shahid, J. Nelson, M. Heeney,
D. D. C. Bradley, G. Itskos, P. A. Koutentis and
S. A. Choulis, RSC Adv., 2013, 3, 10221–10229.
27 A. Burghart, H. J. Kim, M. B. Welch, L. H. Thorensen,
J. Reibenspies, K. Burgess, F. Bergstrom and
L. B. A. Johansson, J. Org. Chem., 1999, 64, 7813–7819.
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2014
New J. Chem., 2014, 38, 3644--3653 | 3653