Diketopyrrolopyrrole-porphyrin conjugates as broadly absorbing sensitizers for dye-sensitized solar cells

Article abstract of DOI:10.1002/cssc.201100764

A series of four new push-pull zinc porphyrin-based dyes was synthesised for hybrid photovoltaic solar cells with a view to enhancing the light-harvesting efficiency at approximately 550 nm with a diketopyrrolopyrrole (DPP) unit. The strength of the donor side of the push-pull porphyrin was tuned by affixing the electron-rich 4,4'-dimethoxydiphenylamine group at the meso position of the macrocycle, and the influence of the distance between the semiconductor surface and the porphyrin chromophore was assessed by introducing different π-conjugated spacers. Charge-transfer transitions over great distances were characterised by electronic absorption spectroscopy and DFT calculations. The absorption and photoactivity spectra of the new bichromophoric dyes spans the whole visible spectrum to the red, implying a better light-harvesting efficiency than regular porphyrin as the absorption spectra of DPP and porphyrin complement one another. Photovoltaic conversion efficiencies accordingly increase from 2.40 to 5.19%. Interestingly, the best overall efficiency was reached with dye 3, which lacks the powerful donating group in the meso position of the porphyrin core. Optical and electrochemical measurements coupled to time dependent (TD)-DFT calculations give insight into the deleterious effect of the 4,4'-dimethoxydiphenylamine unit on the photovoltaic performances, paving the way towards the design of efficient push-pull porphyrin-based sensitizers. Copyright

Full text of DOI:10.1002/cssc.201100764

DOI: 10.1002/cssc.201100764
Diketopyrrolopyrrole–Porphyrin Conjugates as Broadly
Absorbing Sensitizers for Dye-Sensitized Solar Cells
Julien Warnan, Ludovic Favereau, Frꢀdꢀric Meslin, Marjorie Severac, Errol Blart,
Yann Pellegrin, Denis Jacquemin,* and Fabrice Odobel*[a]
A series of four new push–pull zinc porphyrin-based dyes was
synthesised for hybrid photovoltaic solar cells with a view to
enhancing the light-harvesting efficiency at approximately
550 nm with a diketopyrrolopyrrole (DPP) unit. The strength of
the donor side of the push–pull porphyrin was tuned by affix-
ing the electron-rich 4,4’-dimethoxydiphenylamine group at
the meso position of the macrocycle, and the influence of the
distance between the semiconductor surface and the porphy-
rin chromophore was assessed by introducing different p-con-
jugated spacers. Charge-transfer transitions over great distan-
ces were characterised by electronic absorption spectroscopy
and DFT calculations. The absorption and photoactivity spectra
of the new bichromophoric dyes spans the whole visible spec-
trum to the red, implying a better light-harvesting efficiency
than regular porphyrin as the absorption spectra of DPP and
porphyrin complement one another. Photovoltaic conversion
efficiencies accordingly increase from 2.40 to 5.19%. Interest-
ingly, the best overall efficiency was reached with dye 3, which
lacks the powerful donating group in the meso position of the
porphyrin core. Optical and electrochemical measurements
coupled to time dependent (TD)-DFT calculations give insight
into the deleterious effect of the 4,4’-dimethoxydiphenylamine
unit on the photovoltaic performances, paving the way to-
wards the design of efficient push–pull porphyrin-based
sensitizers.
Introduction
Nanocrystalline dye-sensitised solar cells (DSSCs) represent
promising low-cost alternatives to conventional inorganic pho-
tovoltaic cells.[1] During the last decade, there has been in-
creasing interest in the development of ruthenium-free sensi-
tizers for DSSCs.[2] However, organic sensitizers that approach
the performances of the best ruthenium complexes are still
rare.[3–7] Porphyrin derivatives are the archetypal light collectors
of photosynthetic organisms and they are naturally becoming
the main focus of the design of new organic sensitizers for
DSSCs. Indeed, there have recently been many studies in this
context.[6–27] More specifically, Diau and co-workers developed
very efficient push–pull porphyrin dyes, affording a record effi-
ties: very good photostability, direct synthetic accessibility, and
a hallmark intense absorption band at around 550 nm, where
porphyrins weakly absorb. Therefore, DPPs present well-suited
properties to be incorporated as a complementary chromogen
along with a porphyrin sensitizer; an association that has not
been reported previously, even in the framework of other
applications.
This idea was at the origin of the development of the bi-
chromophoric dyes 3 and 4 shown in Figure 1. Two additional
dyes, 1 and 2, were also prepared to investigate the structure–
property relationship, to assess the role of the DPP moiety, and
to draw molecular design rules. These four new porphyrin
dyes are all composed of a zinc porphyrin and contain the
same cyanoacrylic acid anchoring group, but they differ in the
nature of the spacer between the porphyrin and the anchor:
a direct bond in 1, a phenylethynyl spacer in 2, or a DPP unit
in 3 and 4. The two DPP–porphyrin conjugates only differ by
the meso substituent at the extremity of the system; 3 bears
a simple phenyl group, whereas the other porphyrins are
ꢀ
ciency of 10.17% with an I3 /Iꢀ-based electrolyte,[5–7] and more
recently a 11.9% efficiency for a cobalt electrolyte,[7] which are
the best performing sensitizers reported so far. However, it is
observed that porphyrins exhibit a weak absorbance at around
550 nm, between the Soret and Q-bands, which limits the pho-
tocurrent density in a region of intense solar flux. Therefore, it
is unsurprising that a record efficiency of 12.3%[7] was achieved
by co-sensitisation when a porphyrin is associated with
a push–pull dye, the maximum absorption of which was spe-
cifically positioned at 550 nm. The golden design principle of
organic dyes is the construction of push–pull systems, which
intrinsically exhibit charge-transfer absorption bands in the
visible region and facilitate charge injection. On the other
hand, diketopyrrolopyrrole (DPP) is a well-known chromogen
that has received surprisingly little attention as a building
block for the development of organic sensitizers in DSSCs.[28–30]
However, DPPs could be targets as they possess several quali-
[a] J. Warnan, L. Favereau, Dr. F. Meslin, Dr. M. Severac, Dr. E. Blart,
Dr. Y. Pellegrin, Prof. D. Jacquemin, Dr. F. Odobel
Universitꢀ de Nantes, CNRS
Chimie et Interdisciplinaritꢀ: Synthꢁse, Analyse, Modꢀlisation (CEISAM)
UMR CNRS no 6230
2, rue de la Houssiniꢁre - BP 92208–44322 Nantes Cedex 3 (France)
Fax: (+33)02-51-12-54-02
E-mail: fabrice.odobel@univ-nantes.fr
Supporting Information for this article is available on the WWW under
http://dx.doi.org/10.1002/cssc.201100764.
ChemSusChem 0000, 00, 1 – 11
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
1
&
ÞÞ
These are not the final page numbers!

D. Jacquemin, F. Odobel, et al.
venagel reaction, followed by
zinc(II) insertion into the macro-
cyclic cavity to achieve the syn-
thesis of 1.
The synthesis of 2 starts with
the halogenation of 6 by an
electrophilic addition of bromide
on the meso position, which led
to 8 (Scheme 2). To avoid any
copper coordination by the por-
phyrin macrocycle during the
next step (Sonogashira cou-
pling), 8 was first metalated by
zinc acetate. The addition of the
4-ethynylbenzaldehyde moiety
Figure 1. Structures of the porphyrin sensitizers investigated herein.
was performed using palladium
chemistry with Sonogashira’s
capped with a dianisylamine group. The synthesis and photo-
voltaic investigations of this series of four new sensitizers dem-
onstrate the great potential of the DPP–porphyrin mix and
provide some guidelines for the design of new porphyrin-
based sensitizers.
classical conditions for porphyrins[37] to yield 9. This reaction
furnished only 13% yield after 15 h, and the starting material
was principally recovered during the purification step. This
poor conversion is certainly a result of the sluggish oxidative
addition of palladium in the first step of the catalytic cycle
owing to the electron-rich dianisylamine that significantly in-
creases electron density on the meso position. Despite this low
yield, the synthesis of 2 was achieved by a Knoevenagel con-
densation using cyanoacetic acid and piperidine.
The two bichromophoric sensitizers 3 and 4 were prepared
by using a convergent strategy. The porphyrin part of 3 was
obtained in one step through a Suzuki–Miyaura cross-coupling
between 5 and phenylboronic acid (Scheme 3). The reaction
furnished the trisarylporphyrin, which was immediately halo-
genated in the presence of N-bromosuccinimide (NBS). This
double reaction permitted the easy purification of bromopor-
phyrin 10 without significantly affecting the overall yield (ca.
73%). The additional DPP subunit, featuring a terminal alkyne
and the anchoring group 15, was synthesised according to
a previously described route and was subsequently alkylated
with 2-ethylbromohexane to enhance its solubility in organic
solvents to yield 11 (Scheme 3).[38,39] A Suzuki–Miyaura reaction
was then performed with one equivalent of 4-formylthiophe-
nylboronic acid to yield the DPP derivative 12. Statistical condi-
tions afforded 12 in 34% yield as well as the starting material
and the biscoupled product. A Sonogashira cross-coupling
with ethynyltrimethylsilane followed by tetrabutylammonium
fluoride (TBAF) deprotection furnished the terminal alkyne 14
Results and Discussion
Synthesis of the sensitizers
The synthesis of trans-A2B2-porphyrin is well described in the
literature as it represents the basic framework for push–pull
molecules used in several applications.[31,32] The initial halogen-
ation of one of the electron-rich meso positions furnishes an
open door to versatile functionalisation.[33] Indeed, the synthe-
ses of all four dyes 1–4 start with 5-bromo-10,20-dimesitylpor-
phyrin (5), which was engaged in different palladium coupling
reactions depending on the targeted product. Compound 5
was engaged in a Buchwald–Hartwig cross-coupling with the
electron-rich dianisylamine in the presence of palladium(II) ace-
tate, dry caesium carbonate, and bis-(2-diphenylphosphino-
phenyl) ether (DPEphos) to afford the aminoporphyrin 6 in
a good yield of 85% (Scheme 1).[34,35] An aldehyde group was
then introduced on the remaining electron-rich meso position
by using a Vielsmeier–Hack reaction performed on the cop-
per(II) porphyrin.[36] The copper was then removed under
strongly acidic conditions to give porphyrin 7 in 70% yield.
Compound 7 was then reacted with cyanoacetic acid in a Knoe-
Scheme 1. Synthesis of 1. Reagents and conditions: i) dianisylamine, Pd(OAc)2, Cs2CO3, DPEphos, THF, 85%; ii) 1) Cu(OAc)2·2H2O, CH3OH/CH2Cl2, 100%; 2) DMF,
POCl3, CH2Cl2, 80%; 3) H2SO4, CH2Cl2, 87%; iii) 1) cyanoacetic acid, piperidine, CH3CN, 62%; 2) Zn(OAc)2·2H2O, CH3OH/CH2Cl2, 100%.
&2
&
www.chemsuschem.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 0000, 00, 1 – 11
ÝÝ
These are not the final page numbers!

Diketopyrrolopyrrole–Porphyrin Conjugates for Solar Cells
Scheme 2. Synthesis of 2. Reagents and conditions: i) NBS, CH2Cl2, 97%; ii) 1) Zn(OAc)2·2H2O, CH3OH/CH2Cl2, 100%; 2) 4-ethynylbenzaldehyde, [Pd2(dba)3]
(dba=dibenzylideneacetone, PPh3, CuI, Et3N, THF, 13%; iii) cyanoacetic acid, piperidine, CH3CN, 31%.
in good yield. Initially, we considered the introduction of the
anchoring group after the chromophore coupling reaction to
limit the number of difficult purification steps of acidic com-
pounds; unfortunately, we did not succeed in performing
a Knoevenagel reaction with the porphyrin–DPP–aldehyde
adduct. Consequently, we chose to introduce the carboxylic
anchor earlier in the synthesis on the DPP derivative 14, which
led to 15 in 50% yield (Scheme 3).
Finally, porphyrin 10 and the
DPP unit 15 were linked togeth-
er by a copper-free Heck-type
alkynylation
reaction
(see
Scheme 4).[37] Zinc(II) was ulti-
mately introduced into 16 to
yield 3.
The synthesis of the push–pull
porphyrin 4 was inspired by pre-
viously published routes.[40,41]
A protected alkyne was first in-
troduced on the porphyrin core
of 5, followed by an iodination
with diiodine in the presence of
hypervalent iodine (Scheme 5).
The dianisylamine was subse-
quently introduced by a Buch-
wald–Hartwig coupling reaction
Scheme 3. Synthesis of precursors of 3. Reagents and conditions: i) 1) phenylboronic acid, [Pd(PPh3)4], K3PO4,THF;
2) NBS, pyridine, CHCl3, 73%; ii) 4-formylthiophenylboronic acid, [Pd(PPh3)4], Na2CO3, THF/H2O, 34%; iii) ethynyltri-
metylsilane, [Pd(PPh3)4], CuI, Et3N, toluene, 97%; iv) TBAF, THF, 95%; v) cyanoacetic acid, piperidine, CH3CN/CHCl3,
reflux, 50%.
Scheme 4. Synthesis of 3. Reagents and conditions: i) [Pd2(dba)3], AsPh3, Et3N, THF/MeOH, 38%; ii) Zn(OAc)2, CH2Cl2/MeOH, 98%.
ChemSusChem 0000, 00, 1 – 11
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemsuschem.org
&
3
&
ÞÞ
These are not the final page numbers!

D. Jacquemin, F. Odobel, et al.
Scheme 5. Synthesis of 4. Reagents and conditions: i) 1) Zn(OAc)2·2H2O, CH2Cl2/MeOH; 2) ethynyltriisopropylsilane, [Pd(PPh3)4], CuI, Et3N, THF; 3) HCl (2n),
93%; ii) I2, PhI(CF3CO2)2, pyridine, CH2Cl2, 80%; iii) dianysylamine, Pd(OAc)2, DPEPhos, Cs2CO3, THF, 57%; iv) 1) TBAF, THF, 2) Compound 12, [Pd2(dba)3], AsPh3,
Et3N, THF, 19%; v) cyanoacetic acid, piperidine, THF, 20%; vi) Zn(OAc)2·2H2O, CH2Cl2/MeOH, 95%.
in a satisfactory 57% yield. The advantage of this higher-yield-
ing synthetic pathway, compared to the route used for 2,
stems from the fact that the palladium cross-coupling reaction
was performed earlier, which avoided the sluggish oxidative
addition generally encountered with electron-rich substrates.
As in the case of 3, the freshly deprotected porphyrin 19 was
linked to the bromo-susbtituted DPP 12 through a Heck-type
alkynylation reaction. The synthesis of 4 ended with the con-
densation of cyanoacetic acid and the zinc(II) metalation of the
porphyrin.
Figure 3. Normalised absorption spectra of the porphyrin sensitizers immo-
bilised on TiO2 electrodes in the absence of chenodeoxycholic acid (spectra
of the porphyrin sensitizers immobilised on TiO2 in the presence of cheno-
deoxycholic acid are given in the SI).
Electronic absorption spectra in solution and on TiO2
electrodes
The absorption spectra of the sensitizers, recorded in dichloro-
methane solution and on TiO2 electrodes, are shown in Fig-
ures 2 and 3, respectively. The absorption spectra of the four
sensitizers exhibit the classical intense Soret absorption band
at around 420 nm and the Q-bands at around 650 nm, which
contain significant charge-transfer character corresponding to
a charge shift from the porphyrin moiety towards the anchor-
ing group (see quantum mechanical calculations below). How-
ever, the replacement of the dianisylamine group by a phenyl
group (from 4 to 3) induces a blueshift and a decrease of the
intensity of this Q-band, owing to a larger delocalisation of the
HOMO levels (Figures 2 and 3 and calculations below). The
presence of the DPP unit reinforces the absorbance at around
500 nm; this is particularly perceptible in the absorption spec-
tra of 3 on TiO2, which exhibits a broad absorption band with
a continuous plateau from 380–680 nm (Figure 3). For the pho-
tovoltaic measurements, the dyes were chemisorbed onto TiO2
films in the presence and absence of a co-adsorbate (cheno-
deoxycholic acid, CDCA) to limit their propensity to aggregate.
From Figure 3, we can observe that CDCA does not significant-
ly modify the absorption spectra as the shape and full-width-
at-half-maximum of the bands are constant upon CDCA co-ad-
sorption. This indicates a modest degree of aggregation of
these dyes upon chemisorption on TiO2, probably thanks to
the long alkyl chains on the DPP unit and the mesityl groups
on the meso positions of the porphyrins, which prevent the
close packing of the dyes. This is also consistent with the
slight (if any) improvement of the photovoltaic performances
of the dyes in the presence of CDCA.
Except for 3, the porphyrins studied here are not lumines-
cent, therefore the zero–zero excitation energy was calculated
Figure 2. Normalised absorption spectra of 1–4 recorded in dichlorome-
thane.
&4
&
www.chemsuschem.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 0000, 00, 1 – 11
ÝÝ
These are not the final page numbers!

Diketopyrrolopyrrole–Porphyrin Conjugates for Solar Cells
from the wavelength at the edge of the lowest energy Q-band
(Table 1). This indicates that when a zinc porphyrin is substitut-
ed in the meso position with both a strong electron-releasing
group, such as dianisylamine, and a strong electron-withdraw-
ing group, such as ethynylphenylcyanoacrylic acid, the nonra-
diative deactivation pathways are particularly efficient, because
ZnP remains fluorescent in the presence of only one of these
substituents.[7,14,42,43] This can be a result of the stronger
to that of a regular DPP unit, and in agreement with the quan-
tum chemical calculations, which show that the HOMO level is
distributed on the zinc porphyrin with a strong contribution
on the amino group when it is present. In these systems, the
diarylamino group contains methoxy substituents that signifi-
cantly increase the electron-donating ability and, therefore,
cathodically shift the redox potential compared to similar por-
phyrins reported by Diau and co-workers that contain alkyl
groups instead (EOx ꢁ0.7 V vs. SCE).[41] Conversely, the
first oxidation potential of 3, lacking the diarylamino
substituent, is anodically shifted compared to 1, 2,
Table 1. Absorption and emission characteristics, redox potentials, and Gibbs free en-
ergies of 1, 2, 3, and 4 (SCE=saturated calomel electrode).
and 4, which is in agreement with the neutral elec-
tronic effect of the phenyl group. The calculated
Gibbs free energies indicate that the electron injec-
tion and the dye regeneration reactions are both
thermodynamically allowed processes for 1–4. How-
ever, 3 differs from the others because it is the single
sensitizer for which the regeneration reaction is sig-
nificantly exergonic (DG=ꢀ0.62 eV), whereas the
other dyes exhibit relatively weak driving forces
Dye
EOx(S+/S)
[V vs. SCE]
labs
E00(S*)
[eV]
EOx(S+/S*)
[V vs. SCE][a]
DGinj
[eV][b]
DGreg
[eV][c]
[nm]
1
2
3
4
0.51
0.54
0.80
0.50
424, 568, 656
439, 574, 668
422, 520, 625
427, 654
1.63[d]
1.70[d]
1.90[e]
1.67[d]
ꢀ1.12
ꢀ1.16
ꢀ1.10
ꢀ1.17
ꢀ0.38
ꢀ0.42
ꢀ0.36
ꢀ0.43
ꢀ0.33
ꢀ0.36
ꢀ0.62
ꢀ0.32
[a] Calculated according to the equation: EOx(S+/S*)=EOx(S+/S)ꢀE00. [b] Calculated ac-
cording to the equation: DGinj =0.74+EOx(S+/S*). [c] Calculated according to the equa-
tion: DGreg =0.18ꢀEOx(S+/S). [d] Estimated from the wavelength at the tail of the
lowest energy Q-band. [e] Calculated with the wavelength at the intersection (linter) of
(DGreg ꢁꢀ0.3 eV) for
a bimolecular reaction and
multi-electronic process known for their high overpo-
tential.[48,49] Furthermore, one should bear in mind
normalised absorption and emission spectra with the equation E00 =1240/linter
.
ꢀ
that the reduction of I3 is a non-elemental, polyelec-
tronic process, involving much higher potentials than
E(I3ꢀ/Iꢀ).[48] Experimental evidence supports the for-
·ꢀ
charge-transfer character of the lowest excited-state, which
may interconnect with the ground state as the energy gap is
decreased. Furthermore, the presence of a double bond direct-
ly connected to the porphyrin core can also quench the por-
phyrin excited state by possible cis–trans isomerism, which has
been reported for other porphyrins directly substituted with
double bonds.[44,45]
mation of I2 from Iꢀ during the regeneration of oxidised
N719;[50,51] the relevant potential to estimate the dye regenera-
tion thermodynamics is then E(I2·ꢀ/Iꢀ), which is higher than
E(I3ꢀ/Iꢀ), and therefore even a small difference in E(Dye+/Dye)
may account for significant changes in the regeneration of the
photo-oxidised dye.[48]
This indicates that the presence of both dianisylamine and
electron-withdrawing groups (cyanoacrylic anchor for 1 and 2,
and DPP–bridge–cyanoacetic anchor for 4) probably decreases
the luminescence emission quantum yield because of the
stronger charge-transfer character of the lowest excited state.
This is in agreement with other push–pull porphyrins pub-
lished recently.[41,42]
Quantum mechanical calculations
Time-dependent (TD)-DFT calculations have been performed
for the four dyes (see Experimental Section in the Supporting
Information) to probe the nature of the electronic transitions
implied in the different absorption bands. The main results are
collated in Tables 2 and 3, which displays the nature of the
frontier orbitals. The optimised structure of the four com-
pounds present the expected trends: the core is almost flat,
but the DPP moiety forms a dihedral angle of approximately
308 with its side phenyl rings, whereas the electron-rich diani-
sylamine group is arranged in a propeller-helix form. Neverthe-
less, the absence of a spacer between the porphyrin and the
anchoring group in 1 induces a twist of 428 between these
two moieties. The dipole moments computed are relatively
uniform for all structures: 8–9 D. The theoretical wavelengths
listed in Table 2 are in reasonable agreement with the experi-
mental values (see above): i) the Q-band is located at around
650 nm for all systems but 1 (608 nm) and is also less intense
in 1; ii) the introduction of a DPP unit causes the emergence of
an extra band at approximately 510 nm (but its relative intensi-
ty seems to be overestimated by TD-DFT); iii) the most intense
Soret transitions appear at around 400 nm and, notably, 2
presents the most redshifted Soret peak, which fits Figure 2.
Electrochemical properties
The oxidation potentials of the dyes were determined by dif-
ferential pulse voltammetry recorded on TiO2 electrodes to
assess the thermodynamic feasibility of the different electron
transfer processes. Using the zero–zero energy of the singlet
excited state (E00) and the well-accepted values of the conduc-
tion band of TiO2 (ꢀ0.74 V vs. SCE)[46] and that of the redox
ꢀ
couple I3 /Iꢀ (0.18 V vs. SCE),[47] the electron injection Gibbs
free energy (DGinj) and the dye regeneration Gibbs free energy
(DGreg) have been calculated (Table 1). The first oxidation pro-
cess corresponds to the removal of an electron from the zinc
porphyrin with a significant contribution of the dianisylamino
substituent (see DFT calculation section) as they represent the
most electron-rich fragments of 1, 2, and 4. This is consistent
with the easier oxidation of this type of porphyrin, compared
ChemSusChem 0000, 00, 1 – 11
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemsuschem.org
&
5
&
ÞÞ
These are not the final page numbers!

D. Jacquemin, F. Odobel, et al.
V) traces and incident photon-
to-current conversion efficiency
(IPCE) spectra are depicted in
Figures 4 and 5, respectively, and
the photovoltaic data are collect-
ed in Table 4. First, we can
notice that the open circuit pho-
tovoltage Voc values are relatively
low as they did not exceed
605 mV for all dyes. Interestingly,
the highest Voc value was ob-
tained for 4, which exhibits one
of the lowest overall photocon-
version efficiencies within this
series owing to a particularly
poor integral photocurrent den-
sity value (Jsc, Table 4). Low Voc
values, which are quite usual
with organic sensitizers, most
probably stem from the high
charge recombination reactions
(CR) between the injected elec-
tron with the hole on the sensi-
tizer or with the liquid electro-
lyte (dark current).[53] Sensitizer 4
Table 2. Summary of the theoretical results obtained at the PCM-TD-CAM-B3LYP/6-31+G(d) level. The dipole-
allowed vertical absorption wavelengths (l) are listed with their oscillator strengths (f) and the three largest or-
bital contributions in decreasing order. Only transitions with l>350 nm and f>0.4 are reported. For each tran-
sition, charge-transfer parameters (see Experimental Section) have been computed: the distance (dCT) and the
transferred charge (qCT). The energies of the HOMO and LUMO levels as well as the ground-state dipole
moment are also listed.
Dye
l
f
Orbital contributions
dCT
[ꢂ]
qCT
[e]
HOMO LUMO Dipole moment
[eV] [eV] [Debye]
[nm]
1
644
436
411
651
444
431
375
608
509
416
405
388
380
646
513
428
416
385
378
0.41 H!L; Hꢀ1!L+1; Hꢀ2!L
0.78 H!L+1; Hꢀ1!L
3.19
3.84
3.87
3.69
7.40
4.42
3.58
1.61
5.22
9.47
2.15
5.94
0.41 ꢀ6.15
0.51
0.25
0.36 ꢀ6.01
0.46
0.47
0.63
0.61 ꢀ6.16
0.24
0.25
0.19
ꢀ2.33 8.47
1.39 Hꢀ1!L+1; Hꢀ2!L; H!L+2
0.73 H!L; Hꢀ1!L+1; H!L+2
2.03 Hꢀ1!L+1; H!L+2; H!L
0.88 H!L+1; Hꢀ1!L; Hꢀ1!L+2
0.41 Hꢀ2!L+1; Hꢀ1!L
2
3
ꢀ2.31 8.10
0.95 H!L+1; H!L; Hꢀ1!L+2
1.55 Hꢀ2!L; H!L; Hꢀ1!L+2
0.74 Hꢀ1!L+2; Hꢀ2!L; H!L
1.47 H!L+2; Hꢀ1!L+1; Hꢀ1!L
0.96 Hꢀ3!L; Hꢀ2!L+3; Hꢀ2!L+1
0.56 H!L+1; H!L; H!L+3
ꢀ2.35 8.80
0.37
12.26 0.56
4
0.91 H!L+1; H!L; Hꢀ1!L+2
1.54 Hꢀ2!L; Hꢀ3!L; H!L+3
0.88 H!L+2; Hꢀ1!L+1; Hꢀ1!L
0.50 Hꢀ1!L+2; Hꢀ3!L; H!L
1.64 Hꢀ4!L; Hꢀ1!L+2; Hꢀ4!L+1
3.56
6.11
4.26
11.86 0.33
7.27
0.32 ꢀ5.98
ꢀ2.36 8.76
0.27
0.47
0.27
0.62
0.46 Hꢀ2!L+2; Hꢀ3!L+2; Hꢀ3!L+1 3.26
is
a long rigid molecule, in
According to Table 3, the HOMO is centred on the porphyrin
and the electron-rich arylamine group, whereas the LUMO is
either a mix of the porphyrin and cyanoacrylic groups (1 and
2) or is located on both the DPP and the anchoring group (3
and 4). To quantify the associated charge transfer (CT) with all
transitions, we have used a recently designed procedure.[52] For
the 509 and 513 nm peaks of 3 and 4, TD-DFT predicts a pro-
motion from the DPP (HOMOꢀ2) to the anchoring group
(LUMO). These extra transitions, therefore, imply a significant
CT (distance>5 ꢂ) in a key domain of the visible spectra, con-
firming the improvements brought by DPP. From our calcula-
tions, there are no clear hints that the CT could be significantly
different for 3 and 4, if absorption at around 500 nm occurs.
The Q-band also implies a CT over a significant distance for all
systems except 3, illustrating the positive impact of the “push-
ing” amino group. For 2, the Soret peak induces a stronger CT
than that in 1, which is consistent with the orbitals shown in
Table 3. For 3 and 4, the Soret bands imply the LUMO+2,
which is like the LUMO+1 of 2 and 1 and also the LUMO,
therefore indicating an improved situation compared to DPP-
free structures, which is consistent with the larger computed
CT distances.
which the positive charge, formed upon electron injection,
must be primarily located on the amino substituent (see calcu-
lations above). As a result, there is a large distance between
the two charges, which explains the slower CR and, conse-
quently, the higher Voc value.[54] In addition, the J–V curves
show that the dark current for porphyrin 4 is significantly
lower than that for the other dyes, probably a result of closer
molecular packing on the TiO2 surface, which prevents the ap-
proach of triiodide (Figure 4).
The low photovoltaic performances of sensitizers 1 and 4
originate from their poor photocurrent. Inspection of the distri-
bution of the LUMO orbitals reveals that the electronic cou-
Photovoltaic measurements
The four new porphyrins were tested as sensitizers in classical
DSSC devices by using a liquid electrolyte composed of 0.6m
1,2-dimethyl-3-butylimidazolium iodide, 0.1m LiI and 0.05m I2
in acetonitrile, and 12 mm thick TiO2 electrodes. For compari-
son, a reference cell sensitised with the ruthenium complex
N719 was also prepared. The photocurrent density–voltage (J–
Figure 4. Current/voltage characteristics of the porphyrin sensitizers record-
ed under AM1.5 (straight line) and under the dark (dashed line).
&6
&
www.chemsuschem.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 0000, 00, 1 – 11
ÝÝ
These are not the final page numbers!

Diketopyrrolopyrrole–Porphyrin Conjugates for Solar Cells
Table 3. Optimised structures and the frontier molecular orbitals (from HOMOꢀ3 (H) to LUMO+3 (L)) of the porphyrin sensitizers. Branched alkyl chains
were replaced by methyl groups for faster calculations.
Orbital
Dye
1
2
3
4
L+3
L+2
L+1
L
H
Hꢀ1
Hꢀ2
Hꢀ3
ChemSusChem 0000, 00, 1 – 11
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemsuschem.org
&
7
&
ÞÞ
These are not the final page numbers!

D. Jacquemin, F. Odobel, et al.
creases the photocurrent density by more than 50% owing to
the upward band bending and narrows the energy difference
between the LUMO of the dye and the edge of the TiO2 con-
duction band. The important decrease of the overall photocon-
version efficiency upon tert-butylpyridine addition in the elec-
trolyte is only observed with 1 and 3, the electron injection
Gibbs free energies of which are not too large. However, 3 ex-
hibits a broad IPCE trace spanning a large spectral window
from 380–700 nm (Figures 3 and 5). This highlights the benefi-
cial effect of the DPP moiety that fills the absorption gap of
the porphyrin at around 500 nm. This positive effect can also
be observed with 4, although less markedly because its IPCE
values are lower than those of 3 (Figure 5). Another important
difference between 3 and the other sensitizers is that it is the
only fluorescent dye. The electron injection rate is considered
very high as it was measured to be faster than 100 ps with
similar push–pull porphyrin sensitizers on TiO2.[7] However, it is
therefore possible that systems 1, 2, and 4 feature noncompe-
titive injection kinetics with respect to nonradiative decay
routes, leading to non-optimised injection quantum yields in
the cases of 1, 2, and 4. This is in agreement with other por-
phyrin-based sensitizers.[17]
Table 4. Open circuit potential (Voc), current density (Jsc), fill factor (ff) and
photo conversion efficiency (h) of the dyes recorded under AM1.5
(100 mWcmꢀ2 illumination).
Dye
Voc
[mV]
Jsc
ff
[%]
h
[mAcmꢀ2
]
[%]
1[a]
545
535
555
605
645
6.02
9.10
12.88
5.50
73.1
69.2
72.7
76.6
63.6
2.40
3.37
5.19
2.55
8.75
2[a]
3[a]
4[a]
N719[b]
21.31
[a] In the presence of CDCA. [b] Without CDCA (see the Experimental
Section).
pling with the TiO2 conduction band must be large and similar
within this series of dyes as all the LUMO orbitals significantly
extend to the anchoring group, which allows good mixing of
the wavefunction of the excited state with that of the accept-
ing levels of the semiconductor (Table 4). Conversely, sensitiz-
ers 1 and 4 display lower light-harvesting efficiencies (LHE)
compared to those of 2 and 3, which explains the lower pho-
tocurrents measured for 1 and 4 (Figure 3). Another important
difference of 1 and 4 compared to 2 and particularly 3 (the
best dye) is the weak driving force for the dye regeneration re-
action (Table 1). Therefore, 1 and 4 are certainly limited by
a weak electron collection efficiency owing to the sluggish re-
action of iodide with the oxidised sensitizer after electron in-
jection. This is reflected by the lower IPCE values recorded for
1 and 4 compared to those of the other dyes (Figure 5). The
The structure of 2 is related to the efficient porphyrin sensi-
tizer YD2 developed by Diau and co-workers, which has
a 8.8% photoconversion efficiency.[5] YD2 differs from 2 in the
meso-phenyl substituents on the porphyrin (a tert-butyl group
at the meta position in YD2, methyl groups at the ortho and
para positions in 2), in the substituents of the amino electron-
releasing unit (hexyl chains in YD2, methoxy groups in 2), in
the anchoring group (simple carboxylic acid in YD2, cyanoa-
crylic acid in 2), and in that YD2 is fluorescent. This study re-
veals that the determining parameter is most certainly the re-
placement of the methoxy groups by the hexyl chains on the
diarylamino pushing unit because these substituents control
both the efficiency of the regeneration reaction (by the oxida-
tion potential of the dye), probably the degree of aggregation
and, just as important, the surface protection against triiodide
approach (to prevent losses by dark current). Also, the fluores-
cent quenching in 1, 2, and 4, which prevents efficient electron
injection, is another reason for the low photovoltaic perform-
ances of these dyes.
Figure 5. Photoactivity spectra of the porphyrin sensitizers.
Despite all the appealing properties of DPP, its use in DSSCs
is rare.[28,30,55,56] When used as a chromophore, respectable
yields have already been obtained, with minimal synthetic ef-
forts.[56] Nevertheless, the insertion of the porphyrin chromo-
phore entailed a significant improvement of the external quan-
tum efficiency by 25%, mainly due to better LHE. Various
push–pull dyes that incorporate DPP as a linker between elec-
tron-rich and electron-poor moieties have also been reported,
differing mainly in the nature of the spacer that links the cya-
noacryic and/or the dianisylamine groups to the DPP
unit.[28,30,56] Properly tuning the extent of the electronic com-
munication between the different moieties of the dye ap-
peared to be a parameter of paramount importance, brutally
influencing the performances of the DSSC, which vary from
less than 1% to more than 6%. The output potential was par-
ticularly affected as a result of encouraged charge recombina-
calculated regeneration reaction driving forces may be overes-
timated owing to the complex mechanism of triiodide reduc-
·ꢀ
tion, which probably involves the initial formation of I2 rather
than the direct production of I3ꢀ.[48] The introduction of
a weaker electron-donating substituent than dianisylamine
would raise the oxidation potential and increase the driving
force of the limiting dye regeneration reaction.
The dye regeneration reaction for 2 is slightly more exergon-
ic, therefore this dye consistently exhibits a higher, albeit low,
photocurrent density compared to that of 1 and 4 (Tables 1
and 4). Although not optimised in terms of electron injection
Gibbs free energy, the highest photoconversion efficiency was
obtained with porphyrin 3. This is clearly confirmed by the ad-
dition of 0.5m tert-butylpyridine in the electrolyte, which de-
&8
&
www.chemsuschem.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 0000, 00, 1 – 11
ÝÝ
These are not the final page numbers!

Diketopyrrolopyrrole–Porphyrin Conjugates for Solar Cells
[9] M. V. Martꢄnez-Dꢄaz, G. de la Torre, T. Torres, Chem. Commun. 2010, 46,
7090–7108.
[10] H. Imahori, T. Umeyama, S. Ito, Acc. Chem. Res. 2009, 42, 1809–1818.
[11] C.-L. Wang, Y.-C. Chang, C.-M. Lan, C.-F. Lo, E. W.-G. Diau, C.-Y. Lin, Energy
Environ. Sci. 2011, 4, 1788–1795.
tion, a feature that seems to be irrelevant in our case as
decent Voc values were recorded in all cases.
Conclusions
[12] K. K. Pasunooti, J.-L. Song, H. Chai, P. Amaladass, W.-Q. Deng, X.-W. Liu,
J. Photochem. Photobiol. A 2011, 218, 219–225.
This work illustrates that the photoactivity of push–pull por-
phyrin dyes in DSSCs can be extended by filling their usual
transparent window at around 500 nm with a DPP moiety
placed, as a conjugated spacer, between the porphyrin core
and the anchoring group. The two main weak points of the
present dyes lie in their relatively high cathodic oxidation po-
tentials, which makes the regeneration reaction particularly in-
efficient and subsequently limits the charge-collection efficien-
cy and the short singlet excited-state, which may limit the elec-
tron injection quantum yield. The first drawback essentially
stems from the too strong electron-releasing unit of dianisyla-
mine. Although 4 is one of the less efficient dyes within the
series, it is probably one of the most promising systems be-
cause, if one raises its oxidation potential by diminishing the
electron-donating strength of the push substituent and one in-
creases its excited-state lifetime, it is certainly possible to sig-
nificantly increase the photoconversion efficiency. These struc-
ture–property relationships may be useful for the future molec-
ular design of better-performing porphyrin sensitizers for
DSSCs.
[13] J. A. Mikroyannidis, G. Charalambidis, A. G. Coutsolelos, P. Balraju, G. D.
Sharma, J. Power Sources 2011, 196, 6622–6628.
[14] S. Mathew, H. Iijima, Y. Toude, T. Umeyama, Y. Matano, S. Ito, N. V. Tka-
chenko, H. Lemmetyinen, H. Imahori, J. Phys. Chem. C 2011, 115,
14415–14424.
[15] Y. Liu, H. Lin, J. T. Dy, K. Tamaki, J. Nakazaki, D. Nakayama, S. Uchida, T.
Kubo, H. Segawa, Chem. Commun. 2011, 47, 4010–4012.
[16] M. J. Lee, K. D. Seo, H. M. Song, M. S. Kang, Y. K. Eom, H. S. Kang, H. K.
Kim, Tetrahedron Lett. 2011, 52, 3879–3882.
[17] H. Imahori, H. Iijima, H. Hayashi, Y. Toude, T. Umeyama, Y. Matano, S. Ito,
ChemSusChem 2011, 4, 797–805.
[18] H. Imahori, Key Eng. Mater. 2010, 451, 29–40.
[19] E. M. Barea, V. Gonzalez-Pedro, T. Ripolles-Sanchis, H.-P. Wu, L.-L. Li, C.-Y.
Yeh, E. W.-G. Diau, J. Bisquert, J. Phys. Chem. C 2011, 115, 10898–10902.
[20] N. Xiang, X. Huang, X. Feng, Y. Liu, B. Zhao, L. Deng, P. Shen, J. Fei, S.
Tan, Dyes Pigm. 2011, 88, 75–83.
[21] S.-L. Wu, H.-P. Lu, H.-T. Yu, S.-H. Chuang, C.-L. Chiu, C.-W. Lee, E. W.-G.
Diau, C.-Y. Yeh, Energy Environ. Sci. 2010, 3, 949–955.
[22] X.-F. Wang, H. Tamiaki, Energy Environ. Sci. 2010, 3, 94–106.
[23] C. Y. Lee, J. T. Hupp, Langmuir 2010, 26, 3760–3765.
[24] J. Warnan, F. Buchet, Y. Pellegrin, E. Blart, F. Odobel, Org. Lett. 2011, 13,
3944–3947.
[25] F. Odobel, E. Blart, M. Lagree, M. Villieras, H. Boujtita, N. El Murr, S. Cara-
mori, C. Alberto Bignozzi, J. Mater. Chem. 2003, 13, 502–510.
[26] K. D. Seo, M. J. Lee, H. M. Song, H. S. Kang, H. K. Kim, Dyes Pigm. 2012,
94, 143–149.
[27] C. Jiao, N. Zu, K.-W. Huang, P. Wang, J. Wu, Org. Lett. 2011, 13, 3652–
3655.
[28] F.-L. Guo, S.-Y. Qu, W.-J. Wu, J. Li, W.-J. Ying, J.-L. Hua, Synth. Met. 2010,
160, 1767–1773.
Acknowledgements
The authors wish to thank the ANR HABISOL (program Asyscol,
n8 ANR-08-HABISOL-002) for financial support. D.J. acknowledges
the European Research Council (ERC) and the Rꢀgion des Pays de
la Loire for financial support in the framework of a Starting
Grant (Marches - 278845) and a recrutement sur poste stratꢀgi-
que, respectively. This research used resources of the GENCI-
CINES/IDRIS (Grant c2011085117) and of the CCIPL (Centre de
Calcul Intensif des Pays de Loire).
[29] C. Kanimozhi, P. Balraju, G. D. Sharma, S. Patil, J. Phys. Chem. C 2010,
114, 3287–3291.
[30] S. Qu, W. Wu, J. Hua, C. Kong, Y. Long, H. Tian, J. Phys. Chem. C 2010,
114, 1343–1349.
[31] J. S. Lindsey, Acc. Chem. Res. 2010, 43, 300–311.
[32] B. J. Littler, Y. Ciringh, J. S. Lindsey, J. Org. Chem. 1999, 64, 2864–2872.
[33] D. P. Arnold, R. C. Bott, H. Eldridge, F. M. Elms, G. Smith, M. Zojaji, Aust.
J. Chem. 1997, 50, 495–504.
[34] J. F. Hartwig, Acc. Chem. Res. 2008, 41, 1534–1544.
[35] J. F. Hartwig, Acc. Chem. Res. 1998, 31, 852–860.
[36] R. Grigg, G. Shelton, A. Sweeney, A. W. Johnson, Perkin Trans. 1 1972,
1789–1799.
Keywords: macrocycles · nanoparticles · semiconductors ·
sensitizers · zinc
[37] R. W. Wagner, T. E. Johnson, F. Li, J. S. Lindsey, J. Org. Chem. 1995, 60,
5266–5273.
[38] E. Q. Guo, P. H. Ren, Y. L. Zhang, H. C. Zhang, W. J. Yang, Chem. Commun.
2009, 5859–5861.
[1] A. Hagfeldt, G. Boschloo, L. Sun, L. Kloo, H. Pettersson, Chem. Rev. 2010,
110, 6595–6663.
[2] A. Mishra, M. K. R. Fischer, P. Bꢃuerle, Angew. Chem. 2009, 121, 2510–
2536; Angew. Chem. Int. Ed. 2009, 48, 2474–2499.
[3] W. Zeng, Y. Cao, Y. Bai, Y. Wang, Y. Shi, M. Zhang, F. Wang, C. Pan, P.
Wang, Chem. Mater. 2010, 22, 1915–1925.
[4] S. Ito, H. Miura, S. Uchida, M. Takata, K. Sumioka, P. Liska, P. Comte, P.
Pechy, M. Grꢃtzel, Chem. Commun. 2008, 5194–5196.
[5] T. Bessho, S. M. Zakeeruddin, C.-Y. Yeh, E. W.-G. Diau, M. Grꢃtzel, Angew.
Chem. 2010, 122, 6796–6799; Angew. Chem. Int. Ed. 2010, 49, 6646–
6649.
[6] Y.-C. Chang, C.-L. Wang, T.-Y. Pan, S.-H. Hong, C.-M. Lan, H.-H. Kuo, C.-F.
Lo, H.-Y. Hsu, C.-Y. Lin, E. W.-G. Diau, Chem. Commun. 2011, 47, 8910–
8912.
[39] G. Zhang, K. Liu, Y. Li, M. Yang, Polym. Int. 2009, 58, 665–673.
[40] C.-W. Lee, H.-P. Lu, C.-M. Lan, Y.-L. Huang, Y.-R. Liang, W.-N. Yen, Y.-C. Liu,
Y.-S. Lin, E. W.-G. Diau, C.-Y. Yeh, Chem. Eur. J. 2009, 15, 1403–1412.
[41] C.-P. Hsieh, H.-P. Lu, C.-L. Chiu, C.-W. Lee, S.-H. Chuang, C.-L. Mai, W.-N.
Yen, S.-J. Hsu, E. W.-G. Diau, C.-Y. Yeh, J. Mater. Chem. 2010, 20, 1127–
1134.
[42] H. Imahori, Y. Matsubara, H. Iijima, T. Umeyama, Y. Matano, S. Ito, M.
Niemi, N. V. Tkachenko, H. Lemmetyinen, J. Phys. Chem. C 2010, 114,
10656–10665.
[43] C. Y. Lee, C. She, N. C. Jeong, J. T. Hupp, Chem. Commun. 2010, 46,
6090–6092.
[44] M. Chachisvilis, V. S. Chirvony, A. M. Shulga, B. Kꢃllebring, S. Larsson, V.
Sundstrçm, J. phys. Chem. 1996, 100, 13867–13873.
[45] E. Blart, F. Suzenet, J.-P. Quintard, F. Odobel, J. Porphyrins Phthalocya-
nines 2003, 7, 207–213.
[7] A. Yella, H.-W. Lee, H. N. Tsao, C. Yi, A. K. Chandiran, M. K. Nazeeruddin,
E. W.-G. Diau, C.-Y. Yeh, S. M. Zakeeruddin, M. Grꢃtzel, Science 2011, 334,
629–634.
[8] M. G. Walter, A. B. Rudine, C. C. Wamser, J. Porphyrins Phthalocyanines
2010, 14, 759–792.
[46] A. Hagfeldt, M. Grꢃtzel, Chem. Rev. 1995, 95, 49–68.
[47] M. Nazeeruddin, M. Grꢃtzel in Photofunctional Transition Metal Com-
plexes, Vol. 123, Springer, Berlin, 2007, pp. 113–175.
ChemSusChem 0000, 00, 1 – 11
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemsuschem.org
&
9
&
ÞÞ
These are not the final page numbers!

D. Jacquemin, F. Odobel, et al.
[48] G. Boschloo, A. Hagfeldt, Acc. Chem. Res. 2009, 42, 1819–1826.
[49] N. W. Duffy, L. M. Peter, R. M. G. Rajapakse, K. G. U. Wijayantha, J. Phys.
Chem. B 2000, 104, 8916–8919.
[54] R. Jose, Appl. Phys. Lett. 2008, 93, 023125.
[55] J. Warnan, L. Favereau, Y. Pellegrin, E. Blart, D. Jacquemin, F. Odobel, J.
Photochem. Photobiol. A 2011, 226, 9–15.
[50] L. M. Peter, J. Phys. Chem. C 2007, 111, 6601–6612.
[51] A. Y. Anderson, P. R. F. Barnes, J. R. Durrant, B. C. O’Regan, J. Phys. Chem.
C 2011, 115, 2439–2447.
[56] S. Qu, B. Wang, F. Guo, J. Li, W. Wu, C. Kong, Y. Long, J. Hua, Dyes Pigm.
2012, 92, 1384–1393.
[52] T. Le Bahers, C. Adamo, I. Ciofini, J. Chem. Theory Comput. 2011, 7,
2498–2506.
Received: November 25, 2011
Revised: January 27, 2012
[53] M. Miyashita, K. Sunahara, T. Nishikawa, Y. Uemura, N. Koumura, K. Hara,
A. Mori, T. Abe, E. Suzuki, S. Mori, J. Am. Chem. Soc. 2008, 130, 17874–
17881.
Published online on && &&, 0000
&10
&
www.chemsuschem.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 0000, 00, 1 – 11
ÝÝ
These are not the final page numbers!

FULL PAPERS
Panoptic porphyrin: We present the
synthesis, the electronic properties, the
time dependent DFT calculations and
the photovoltaic performances of new
porphyrin sensitizers in a TiO2 dye-sensi-
tized solar cell. Porphyrin dyes represent
the most efficient sensitizer in DSSC,
but they exhibit a weak absorbance
around 550 nm. The bichromophoric
system presented (see picture) is syn-
thesized by connecting a diketopyrrolo-
pyrrole with a push–pull porphyrin to
obtain a real panchromatic dye.
J. Warnan, L. Favereau, F. Meslin,
M. Severac, E. Blart, Y. Pellegrin,
D. Jacquemin,* F. Odobel*
&& – &&
Diketopyrrolopyrrole–Porphyrin
Conjugates as Broadly Absorbing
Sensitizers for Dye-Sensitized Solar
Cells
ChemSusChem 0000, 00, 1 – 11
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemsuschem.org
&
11
&
ÞÞ
These are not the final page numbers!