Effect of porphyrin loading on performance of dye sensitized solar cells based on iodide/tri-iodide and cobalt electrolytes

Article abstract of DOI:10.1039/c3ta12955a

Two zinc-porphyrin sensitizers, 1a and 1b, bearing triphenylamine donor groups, were synthesized and their efficiencies measured in nanocrystalline TiO₂ dye sensitized solar cells employing iodide/tri-iodide and tris(1,10-phenanthroline) cobalt electrolytes. Optimized sensitization time for the TiO₂ photoanode was found to depend on the electrolyte employed: devices based on iodide/tri-iodide showed better efficiencies with shorter sensitization times (1.5 hours) whereas those based on tris(1,10-phenanthroline) cobalt showed better efficiencies with longer sensitization times (6 hours). From UV-Vis absorption spectra it is estimated that there is roughly twice as much dye loaded onto the TiO₂ film sensitized for 6 hours compared to the 1.5 hour film. Interfacial processes were probed using transient photovoltage, transient absorption and fluorescence lifetime measurements. The results indicate that sensitization time does not affect either dye regeneration or interfacial recombination processes in the presence of either electrolyte. However, sensitization time does have a considerable impact on device photocurrent, and moreover, the effect is different for the two electrolytes studied. This work demonstrates how device preparation must be tailored carefully depending on the electrolyte red/ox couple used. The Royal Society of Chemistry 2013.

Full text of DOI:10.1039/c3ta12955a

Journal of
Materials Chemistry A
View Article Online
View Journal
PAPER
Effect of porphyrin loading on performance of dye
sensitized solar cells based on iodide/tri-iodide and
cobalt electrolytes†
Cite this: DOI: 10.1039/c3ta12955a
a
b
b
b
a
*
`
Ana Aljarilla, John N. Clifford, Laia Pelleja, Antonio Moncho, Susana Arrechea,
Pilar de la Cruz, Fernando Langa and Emilio Palomaresbc
a
a
*
Two zinc–porphyrin sensitizers, 1a and 1b, bearing triphenylamine donor groups, were synthesized and
their efficiencies measured in nanocrystalline TiO2 dye sensitized solar cells employing iodide/tri-iodide
and tris(1,10-phenanthroline) cobalt electrolytes. Optimized sensitization time for the TiO2 photoanode
was found to depend on the electrolyte employed: devices based on iodide/tri-iodide showed better
efficiencies with shorter sensitization times (1.5 hours) whereas those based on tris(1,10-phenanthroline)
cobalt showed better efficiencies with longer sensitization times (6 hours). From UV-Vis absorption
spectra it is estimated that there is roughly twice as much dye loaded onto the TiO2 film sensitized for
6 hours compared to the 1.5 hour film. Interfacial processes were probed using transient photovoltage,
transient absorption and fluorescence lifetime measurements. The results indicate that sensitization time
does not affect either dye regeneration or interfacial recombination processes in the presence of either
electrolyte. However, sensitization time does have a considerable impact on device photocurrent, and
moreover, the effect is different for the two electrolytes studied. This work demonstrates how device
preparation must be tailored carefully depending on the electrolyte red/ox couple used.
Received 29th July 2013
Accepted 24th September 2013
DOI: 10.1039/c3ta12955a
www.rsc.org/MaterialsA
absorption bands and high photostability.15–17 However, their
properties in DSCs still need to be fully investigated and
Introduction
Research into Dye Sensitized Solar Cells (DSCs) continues to be
an extremely active area.1–4 Aer considerable time in which
DSCs based on Ru(II)polypyridyl sensitizers were clearly the
most efficient, DSCs based on organic sensitizers have recently
made huge strides forward in development.5,6 In particular,
there are many recent examples of DSCs employing iodide/tri-
iodide (Iꢀ/I3ꢀ) with >10% device efficiency based on D–p–A
porphyrins designed by Diau and co-workers.7–9 Indeed, with
recently developed novel cobalt complex based electrolytes10–13
porphyrins have shown their promise with the current highest
efficiency recorded for a DSC device of 12.05% being based on a
cell containing a porphyrin sensitizer and a tris(bipyridyl)
cobalt electrolyte.14 Regardless of the red/ox couple, porphyrins
offer several distinct advantages over Ru(II)polypyridyl sensi-
tizers due to their high molar extinction coefficients, sharp
molecular structure-device function rules need to be outlined to
fully understand performance for devices based on these dyes.
In this article we investigate the performance of two
porphyrin sensitizers, 1a and 1b (Scheme 1), in DSC devices
ꢀ
employing Iꢀ/I3
electrolytes.
and tris(1,10-phenanthroline) cobalt
Interfacial processes were investigated using a number of
spectroscopic techniques. In particular, these processes were
investigated as a function of TiO2 dye loading. It was found that
device efficiency was very sensitive to dye loading. Thisꢀstudy
underlines how device rules for DSCs based on the Iꢀ/I3 elec-
trolyte are not necessary applicable to DSCs based on cobalt
aInstitute of Nanoscience, Nanotechnology and Molecular Materials (INAMOL),
´
Campus de la Fabrica de Armas (UCLM), Toledo, Spain. E-mail: Fernando.Langa@
uclm.es; Fax: +34 9252 68840; Tel: +34 9252 68843
b
¨
Institute of Chemical Research of Catalonia (ICIQ), Avda. Paısos Catalans 16,
Tarragona E-43007, Spain. E-mail: jnclifford@iciq.es; Fax: +34 9779 20224; Tel:
+34 9779 20200
c
´
ICREA, Avda. Lluıs Companys 28, Barcelona E-08030, Spain
† Electronic supplementary information (ESI) available: 1H NMR, 13C NMR, FT-IR,
MS, UV-Visible and emission spectra. CV and OSWV plots. Thermogravimetric
analysis diagrams. See DOI: 10.1039/c3ta12955a
Scheme 1 Structures of dyes 1a and 1b.
This journal is ª The Royal Society of Chemistry 2013
J. Mater. Chem. A

View Article Online
Journal of Materials Chemistry A
Paper
electrolytes. This study is pertinent and timely given the
increasing interest in DSCs based on cobalt electrolytes.
Results and discussion
Synthesis of 1a and 1b
Scheme 2 illustrates the synthetic route to dyes 1a and 1b. We
prepared the free base porphyrins 2a,b in 14% and 11% yield
respectively, according to the Lindsey method.18 This procedure
improves signicantly the yield with respect to previous proce-
dures described for similar porphyrins (4–6%).19 2a,b were
reacted with zinc acetate in chloroform giving the metallated
porphyrins 3a,b in 91% and 95% yield respectively aer puri-
cation by column chromatography. Finally, the trimethylsylil
group was quantitatively removed by TBAF and in situ reacted
with p-iodobenzoic acid under Pd-catalyzed Sonogashira
coupling conditions affording the target dyes 1a,b in 70% and
63% yield respectively. The intermediate and nal compounds
Fig. 1 Normalized absorption spectra of dyes 1a (---) and 1b (—) in dichloro-
methane solution (10ꢀ5 M).
Table 1 Absorption, emission and electrochemical data of dyes 1a and 1b
E1red
E
ox
EHOMO
ELUMO
a
b
1 b
c
d
were fully characterized by means of UV-Vis, FT-IR, 1H and 13
C
lem
/
lmaxa/nm (log 3)
nm
(V)
(V)
(eV)
(eV)
NMR spectroscopies; the structures of all compounds were
conrmed by MALDI-TOF mass spectrometry (see the Experi-
mental section and ESI†). The thermal stabilities of compounds
1a and 1b were evaluated by thermogravimetric analysis (TGA)
under nitrogen, with a heating rate of 10 ꢁC minꢀ1. The
decomposition temperatures (Td) were estimated from the TGA
plot as the temperature of the intercept of the leading edge of
the weight loss curve. Under these conditions, compounds 1a
and 1b display excellent thermal stability up to 220 ꢁC (Fig. S33
and S34, ESI†) which is in principle satisfactory for their
application in photovoltaic devices.
1a 630 (4.18), 572 (4.02), 652
450 (5.08), 306 (4.71)
1b 637 (4.39), 579 (4.07), 669
449 (5.12), 305 (4.77)
ꢀ0.52 0.33 ꢀ5.43
ꢀ0.58 0.28 ꢀ5.38
ꢀ3.50
ꢀ3.49
a
b
10ꢀ5 M, CH2Cl2. [10ꢀ3 M] in THF versus Fc/Fc+, glassy carbon,ꢀP1t
counter electrode, 20 ꢁC, 0.1 M Bu4NClO4, scan rate ¼ 100 mV s
.
c
Calculated using equation EHOMO (vs. vacuum) ¼ ꢀ5.1 ꢀ E1ox (vs.
Fc/Fc+) in eV.20 ELUMO was calculated using ELUMO ¼ EHOMO + E0–0
,
d
where E0–0 is the intersection of the absorption and emission spectra.
bandwidth is observed. For the uorescence spectra in CH2Cl2
the trend for the variation of the emission wavelength is similar
to that of the absorption bands. The emission bands at 652 nm
(1a, lexc ¼ 450 nm) and 669 nm (1b, lexc ¼ 449 nm) (Fig. S26 and
S27 in ESI†) were totally quenched aer adsorption onto TiO2
indicating efficient photoinduced electron transfer from the
dyes to the TiO2 nanoparticles.
Absorption, emission, electrochemistry and computational
calculations
The absorption spectra of dyes 1a,b in dichloromethane
(CH2Cl2) solution are shown in Fig. 1. Both dyes exhibit the
typical features of zinc porphyrins, with an intense Soret band
between 400 and 500 nm and less intense Q bands in the range
550 to 700 nm (see Table 1). As expected, the absorption bands
are not very sensitive to the nature of the substituents, pointing
to weak electronic interactions between the porphyrin and the
attached moiety in the ground state. Due to the presence of the
hexyloxy groups in dye 1b, a bathochromic shi is observed in
the Q band, while in the Soret band an enhancement in
The redox properties of 1a,b were investigated by cyclic vol-
tammetry and square wave voltammetry in tetrahydrofuran
(THF), (Table 1, Fig. S30–S32, ESI†). In the cathodic side,
compounds 1a,b show the rst reversible oxidation peaks at
0.33 and 0.28 V respectively; the presence of the electron-
donating alkoxy groups signicantly reduces the oxidation
potential of 1b compared to 1a. On the reduction side, both
compounds show rst reduction potentials at ꢀ0.52 V and
ꢀ0.57 V as irreversible waves, showing that the electron-
donating alkoxy groups in 1b increase its reduction potential
with respect to 1a, as one would expect. The EHOMO values of 1a
to 1b vary only by 0.05 eV and were determined as ꢀ5.43 eV (1a)
and ꢀ5.38 eV (1b), indicating regeneration is energetically
ꢀ
feasible by both Iꢀ/I3 (Eredox ¼ ꢀ4.75 eV) and Co(II)(phen)3/
Co(III)(phen)3 (Eredox ¼ ꢀ5.06 eV) red/ox couples. The ELUMO
values also indicate that efficient electron injection into the
TiO2 conduction band (ETiO ¼ ꢀ4.00 eV) is energetically
2
possible.
In order to gain insight into the geometries and electronic
properties of dyes 1a and 1b, computational studies were
Scheme 2 Synthetic route to dyes 1a and 1b.
J. Mater. Chem. A
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
Journal of Materials Chemistry A
performed using density functional theory (DFT) at the B3LYP/
6-31G level. Regarding geometry, for both dyes the dihedral
angles between the phenyl ring and the porphyrin macrocycle
(4) are similar (4 ꢂ ꢀ65ꢁ) and are in agreement with those
calculated for similar systems (Fig. 2).21 The HOMOs were found
to be delocalized through both the porphyrin macrocycle and
TPA for both 1a and 1b, whereas the LUMOs are extended along
the porphyrin system, the linker and the acid group, indicating
electronic coupling with the TiO2 nanocrystals. The HOMO–
LUMO gap is similar for both dyes although the LUMO level of
dye 1b is higher (ꢀ2.45 eV) than that of 1a (ꢀ2.61 eV) due to the
electronic coupling of the alkoxy groups of the TPA moieties.
HOMO and LUMO (Fig. 3) are overlapped, favouring the HOMO
to LUMO electronic transitions.
Device properties
Fig. 4 and Table 2 show the IV curves and device characteristics,
namely Jsc, Voc, ll factor (FF) and overall efficiency (h), for
Fig. 4 I–V curves and IPCE spectra of DSC devices composed of sensitizers 1a and
1b based on Iꢀ/I3ꢀ (a and b) and Co(II)(phen)3/Co(III)(phen)3 electrolytes (c and d).
ꢀ
sensitizers 1a and 1b in DSC devices based on Iꢀ/I3 and
Co(II)(phen)3/Co(III)(ꢀphen)3 electrolytes.
Employing Iꢀ/I3 electrolyte, the best device efficiencies
(Fig. 4(a)) were obtained aer 1.5 hours sensitization time with
maximum efficiencies of 5.14% and 4.15% for 1a and 1b
respectively (without mask). These values compare well with a
recent study by Liu et al. involving similar dyes.19 Upon the
longer sensitization time of 6 hours the device efficiency drops
to 3.93% and 3.60% for 1a and 1b respectively (without mask).
The principle reason for this is a loss in Jsc. When cobalt elec-
trolyte is used, however, the dependence of device efficiency on
sensitization time is reversed (Fig. 4(c)). The best efficiencies are
recorded for 6 hours sensitization (3.11% and 2.45% for 1a and
Table 2 Photovoltaic performance of cells recorded under AM 1.5G 1 sun
illumination
Dye
Time
Jsc (mA cmꢀ2
)
Voc (V) FF (%)
h
d (%)
1a
1a
1b
1b
1a
1a
1b
1.5 hoursa 10.09
0.63
0.62
0.62
0.61
0.75
0.76
0.77
0.75
0.76
67
62
70
69
75
74
66
72
72
4.19 (5.14)
3.27 (3.93)
3.40 (4.15)
2.54 (3.60)
2.06 (2.53)
2.48 (3.11)
1.61 (2.01)
2.05 (2.45)
8.42 (11.20)
6 ha
8.51
7.83
6.18
3.67
4.44
3.17
3.82
15.25
1.5 hoursa
6 ha
1.5 hoursb
6 hb
1.5 hoursb
6 hb
1b
YD2-o-C8c
a
Electrolyte: 0.6 M 1-butyl-3-methylimidazolium iodide (BMII), 0.1 M
lithium iodide, 0.05 M iodine and 0.5 M 4-tert-butylpyridine in a 1 : 1
b
mixture of acetonitrile–valeronitrile. Electrolyte: 0.2
M
tris(1,10-
tris(1,10-phenanthroline)
lithium perchlorate and 0.5 4-tert-
85 : 15 mixture of acetonitrile–valeronitrile.
phenanthroline)cobalt(II)(TFSI)2, 0.02
cobalt(III)(TFSI)3, 0.1
butylpyridine in
M
M
M
a
c
Electrolyte: 0.5 M 1-butyl-3-methylimidazolium iodide (BMII), 0.1 M
lithium iodide, 0.05 M iodine, and 0.5 M 4-tert-butylpyridine in a
d
85 : 15 mixture of acetonitrile–valeronitrile. Data in parenthesis
recorded without mask.
Fig. 2 Optimized structure for dye 1b (hexyl groups have been changed by
1b respectively). Again, the main difference arising from
sensitization time is accounted for by a difference in Jsc. The
increase in device Voc (approx. 140 mV) afforded by employing
Co(II)(phen)3/Co(III)(phen)3 with respect to Iꢀ/I3ꢀ is in line with
increases observed in other studies.22 Finally, under similar
conditions the YD2-o-C8 device gave an efficiency of 11.20%
(without mask) indicating testing conditions are indeed
comparable with studies demonstrating the best literature
values for porphyrin based DSCs.23
methyl for calculations).
Fig. 4(b) and (d) show the IPCE spectra for all of the devices
measured in this study. These spectra show the contributions to
device current from the Soret and Q bands centred at around
450 and 650 nm respectively. Integration of these spectra agrees
with the Jsc values for the same devices in Fig. 4(a) and (c).
Fig. 3 Frontier orbitals of dye 1b: HOMO (left) and LUMO (right).
This journal is ª The Royal Society of Chemistry 2013
J. Mater. Chem. A

View Article Online
Journal of Materials Chemistry A
Paper
In order to have an indication of dye loading in these devices,
The data in Fig. 6 demonstrate that sensitization time has
the absorption spectra of thin lms of transparent TiO2 lms little bearing either on the position of the TiO2 conduction band
(6 mm) were measured aer 1.5 hours and 6 hour sensitization or on device electron lifetime for 1a devices based on the two
in 0.2 mM solutions of la and lb in chlorobenzene (Fig. 5). These different electrolytes. This is not surprising if we consider that
data indicate that there is roughly twice as much dye loading on the Voc for devices does not change with sensitization time.
the lm sensitized for 6 hours compared to the equivalent lm Despite the estimated doubling in dye loading for 6 hours
of 1.5 hours.
sensitization compared to 1.5 hours, Voc, TiO2 electron density
and electron lifetime do not change. This may be because the
TiO2 lm is entirely covered by 1a following 1.5 hours and any
subsequent dye loading does not come into direct contact with
TiO2 or improve the dye blocking effect increasing TiO2 electron
lifetime.
Transient absorption spectroscopy was then used to monitor
dye regeneration in these devices and these kinetics are shown
in Fig. 7.
In the absence of electrolyte (black decays) the kinetics are
slow and dispersive, which become faster when electrolyte is
added. The t50% for the device prepared with 1.5 hour sensiti-
zation (0.5 ms) is slightly longer than that prepared with 6 hour
sensitization (0.1 ms), probably due to the fewer charge sepa-
rated species generated in this device due to lower dye loading.
On the other hand, sensitization time appears to have little
effect on t50% in the presence of electrolyte (approx. 20 ms
(Iꢀ/I3ꢀ) and 90 ms (Co(II)(phen)3/Co(III)(phen)3))ꢀ. It is noted that
kinetics recorded in the presence of the Iꢀ/I3 electrolyte are
faster due to the potential of this red/ox couple affording more
Charge extraction, transient photovoltage and transient
absorption spectroscopy measurements
Device electron densities and electron lifetimes were probed
using charge extraction and transient photovoltage measure-
ments respectively (Fig. 6). To simplify the discussion, and
because trends in DSC performance based on 1a and 1b are the
same, data relating to DSCs fabricated using sensitizer 1a are
presented only.
Fig. 5 Absorption spectra of transparent TiO2 films (6 mm) following 1.5 and 6
hour sensitization in 0.2 mM solutions of la and lb in chlorobenzene.
Fig. 7 Transient absorption kinetics for DSC devices of 1a prepared following (a)
1.5 hours and (b) 6 hours sensitization. The black, red and blue decays correspond
to kinetics recorded in the presence of a blank electrolyte, Iꢀ/I3 electrolyte and
ꢀ
Fig. 6 Charge density as a function of voltage and electron lifetime as a function
of charge density of DSC devices composed of sensitizer 1a based on Iꢀ/I3ꢀ (a and
b) and Co(II)(phen)3/Co(III)(phen)3 electrolytes (c and d).
Co(II)(phen)3/Co(III)(phen)3 electrolyte respectively. Kinetics were recorded at
620 nm following excitation at 650 nm.
J. Mater. Chem. A
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
Journal of Materials Chemistry A
sensitization times were found tꢀo depend on the electrolyte
used with devices based on Iꢀ/I3 showing better efficiencies
with shorter sensitization times (1.5 hours) while those based
on Co(II)(phen)3/Co(III)(phen)3 showing better efficiencies with
longer sensitization times (6 hours). UV-Vis absorption spectra
indicate that there is roughly twice as much dye loaded onto the
TiO2 lm sensitized for 6 hours. It was found that dye loading
had little inuence over device Voc, electron density or electron
lifetime. However, at shorter sensitization times and lower dye
loading, electron injection efficiency and photocurrent were
higher in devices based on Iꢀ/I3ꢀ. The opposite was true for
devices based on Co(II)(phen)3/Co(III)(phen)3 with higher elec-
tron injection efficiencies and photocurrents found for longer
sensitization times and higher dye loadings. This work
demonstrates how device preparation must be tailored carefully
depending on the electrolyte red/ox couple used.
Experimental
Materials
Fig. 8 Emission lifetime decays of 1a on transparent Al2O3 and TiO2 films
following 1.5 and 6 hour sensitization in the presence of Iꢀ/I3ꢀ and Co(II)(phen)3/
Co(III)(phen)3 electrolytes. In all cases emission was recorded at 660 nm following
excitation at 405 nm.
Synthetic procedures were carried out under inert argon atmo-
sphere, in dry solvent unless otherwise noted. All reagents and
solvents were reagent grade and were used without further
purication. Chromatographic purications were performed
using silica gel 60 SDS (particle size 0.040–0.063 mm).
free energy for regeneration compared to Co(II)(phen)3/
Co(III)(phen)3.
Fluorescence lifetime measurements
Instruments
To investigate the difference in the Jsc values for la devices,
uorescence emission lifetimes were measured using time-
correlated single photon counting (Fig. 8).
Analytical thin-layer chromatography was performed using
Merck TLC silica gel 60 F254. 1H NMR spectra were obtained on
Bruker TopSpin AV-400 (400 MHz) spectrometer. Chemical
shis are reported in parts per million (ppm) relative to the
solvent residual peak (CDCl3, 7.27 ppm). 13C NMR chemical
shis are reported relative to the solvent residual peak (CDCl3,
77.00 ppm). UV-Vis measurements were carried out on a Shi-
madzu UV 3600 spectrophotometer. For extinction coefficient
determination, solutions of different concentration were
prepared in CH2Cl2, HPLC grade, with absorption between 0.1
and 1 of absorbance using a 1 cm UV cuvette. The emission
measurements were carried out on Cary Eclipse uorescence
spectrophotometer. Mass spectra (MALDI-TOF) were recorded
on a VOYAGER DEꢀ STR mass spectrometer using dithranol as
matrix. The thermal stability was evaluated by TGA on Mettler
Toledo TGA/DSC Starte System under nitrogen, with a heating
rate of 10 ꢁC minꢀ1. Heating of crystalline samples leads to
melting of the solids, but no recrystallization was observed.
Dye loading of 1a onto the above lms was controlled so that
the absorbance of Al2O3 and TiO2 lms in each set of samples
was the same. Electron injection yields, F, were estimated by
integrating the areas under each emission decay, as done
previously by Koops et al.24 The results show that as the sensi-
tization time increases from 1.5 to 6 hours, F drops from 77 to
62%, respectively, in the presence of Iꢀ/I3ꢀ. On the other hand,
in the presence of Co(II)(phen)3/Co(III)(phen)3, F increases from
27 to 40% on increasing sensitization time from 1.5 to 6 hours.
Clearly, F values correlate well with Jsc for the differently
prepared devices.
Dye aggregation has been shown to quench dye excited states
ꢀ
in Iꢀ/I3 based devices limiting device photocurrent.25 This
would explain the drop in Jsc in 1a,b DSC devices at high dye
loading in Iꢀ/I3ꢀ. However, the higher F and Jsc of Co(II)(phen)3/
Co(III)(phen)3 devices at high dye loading indicate that this
electrolyte has some inuence over, and is able to control,
aggregation. This can be explained by the bulky nature of the
cobalt complexes allowing them to break-up dye aggregates of
the planar porphyrin systems, improving Jsc and ultimately
device performance.
Device preparation and characterization
In the present work two different types of TiO2 lms are utilized
depending on the measurements being conducted. Highly
transparent thin lms of 4–8 mm were utilized for L-TAS
measurements and also for UV-Vis absorption studies. On the
other hand, for efficient DSCs, devices were made using either 4
Conclusions
mm or 8 mm thick lms consisting of 20 nm TiO2 nanoparticles
Two zinc–porphyrin sensitizers were synthesized and their (Dyesol paste) for Co(II)(phen)3/Co(III)(phen)3 and Iꢀ/I3 elec-
efficiencies measured in dye sensitized solar cells employing trolytes respectively. A scatter layer of 4 mm of 400 nm TiO2
ꢀ
Iꢀ/I3 and Co(II)(phen)3/Co(III)(phen)3 electrolytes. Optimized particles (Dyesol paste) was then deposited on these lms. Prior
ꢀ
This journal is ª The Royal Society of Chemistry 2013
J. Mater. Chem. A

View Article Online
Journal of Materials Chemistry A
Paper
to the deposition of the TiO2 paste the conducting glass
substrates were immersed in a solution of TiCl4 (40 mM) for 30
min and then dried. The TiO2 nanoparticle paste was deposited
onto a conducting glass substrate (NSG glass with 8 U cmꢀ2
resistance) using the screen printing technique. The TiO2 elec-
trodes were gradually heated under airow at 325 ꢁC for 5 min,
375 ꢁC for 5 min, 450 ꢁC for 15 min and 500 ꢁC for 15 min. The
heated TiO2 electrodes were immersed again in a solution of
TiCl4 (40 mM) at 70 ꢁC for 30 min and then washed with
ethanol. The electrodes were heated again at 500 ꢁC for 30 min
and cooled before sensitization. In order to reduce scattered
light from the edge of the glass electrodes of the dyed TiO2 layer,
a light shading mask was used on the DSCs, so the active area of
DSCs was xed to 0.16 cm2. The counter electrode was made by
spreading a 5 mM solution of H2PtCl6 in isopropyl alcohol onto
a conducting glass substrate (TEC15, Pilkington) with a small
hole to allow the introduction of the liquid electrolyte using
vacuum, followed by heating at 390 ꢁC for 15 min. Dye solutions
of 1a and 1b at concentrations of 0.2 mM in chlorobenzene were
prepared and lms immersed for different periods of time at r.t.
The sensitized electrodes were washed with chlorobenzene and
dried under air. Finally, the working and counter electrodes
were sandwiched together using a thin thermoplastic (Surlyn)
frame that melts at 100 ꢁC. The iodide/tri-iodide electrolyte used
consisted of 0.6 M 1-butyl-3-methylimidazolium iodide (BMII),
0.1 M lithium iodide, 0.05 M iodine and 0.5 M 4-tert-butylpyr-
idine in a 1 : 1 mixture of acetonitrile–valeronitrile. The cobalt
electrolyte consisted of 0.2 M tris(1,10-phenanthroline)cobal-
t(II)(TFSI)2 0.02 M tris(1,10-phenanthroline)cobalt(III)(TFSI)3,
0.1 M lithium perchlorate and 0.5 M 4-tert-butylpyridine in a
85 : 15 mixture of acetonitrile–valeronitrile.
Reference devices containing YD2-o-C8 dye were made by
sensitizing lms for approximately 90 minutes consisting of
13 mm transparent and 4 mm scatter TiO2 in 0.1 mM ethanol
solutions containing 0.5 mM chenodeoxycholic acid. The
electrolyte consisted of 0.5 M 1-butyl-3-methylimidazolium
iodide (BMII), 0.1 M lithium iodide, 0.05 M iodine and 0.5 M
4-tert-butylpyridine in 85 : 15 mixture of acetonitrile–
valeronitrile.
The IV characteristics of cells were measured using a Sun
2000 Solar Simulator (150 W, ABET Technologies). The illumi-
nation intensity was measured to be 100 mW mꢀ2 with a cali-
brated silicon photodiode. The appropriate lters were utilized
to faithfully simulate the AM 1.5G spectrum. The applied
potential and cell current were measured with a Keithley 2400
digital source meter. The IPCE (Incident Photon to Current
conversion Efficiency) was measured using a homemade set up
consisting of a 150 W Oriel Xenon lamp, a motorized mono-
chromator and a Keithley 2400 digital source meter.
Synthesis and characterization
General procedure for the synthesis of 2a,b. To a solution of
corresponding (diphenylamino)benzaldehyde (3 mmol), pyrrol
(4 mmol) and 3-(trimethylsilyl)propiolaldehyde (1 mmol) in
CHCl3 (100 mL mmolꢀ1), under argon and degassed, was added
BF3(OEt)2 (0.9 mmol). The mixture was allowed to stir under
argon for 3 hours. Then, DDQ (2.8 eq.) was added and stirring
for one hour; aer addition of Et3N (2 mL) for 30 min, the
solvent was removed by rotary evaporation and the product
was puried by column chromatography (silica gel, Hex–
CHCl3, 1 : 1).
Synthesis of [10,15,20-tri(N,N-diphenylaniline)-5-ethynyltrimethyl-
silane]porphyrin (2a). From 3.00 g of (diphenylamino)benzalde-
hyde (11 mmol), 1.02 mL of pyrrol (14.7 mmol) and 0.547 mL of 3-
(trimethylsilyl)propiolaldehyde (3.7 mmol), reacted according to
the general procedure, affording 600 mg of 5a as a green solid
(14% yield). 1H NMR (400 MHz, CDCl3) d/ppm: 9.69 (d, J ¼ 4.7 Hz,
2H), 9.05 (d, J ¼ 4.7 Hz, 2H), 8.96 (d, J ¼ 4.7 Hz, 2H), 8.93 (d, J ¼
4.7 Hz, 2H), 8.07 (dd, J ¼ 8.5, 6.4 Hz, 6H), 7.49 (d, J ¼ 8.5 Hz, 6H),
7.45–7.42 (t, J ¼ 7.0 Hz, 24H), 7.20–7.15 (m, 6H), 0.64 (s, 9H),
ꢀ2.31 (s, 2H). 13C NMR (100 MHz, CDCl3) d/ppm: 147.7, 147.6,
135.7, 135.6, 135.5, 135.3, 129.5, 129.2, 124.9, 123.3, 122.1, 121.3,
121.2, 120.9, 107.7, 101.6, 98.6, 0.35. UV-Vis (CH2Cl2) lmax/nm
(log 3): 670 (3.94), 580 (4.30), 538 (4.02), 435 (5.13), 303 (4.77). FT-
IR n/cmꢀ1: 3315, 3058, 3030, 2954, 2137, 1595, 1492, 1328, 1313,
1270, 848, 798, 749, 695. MS (m/z) (MALDI-TOF): calculated for
C79H61N7Si: 1135.48; found: 1135.5 (M+).
Synthesis of [10,15,20-tri(N,N-(bis(4-hexyloxy)phenyl)aniline)-5-
ethynyltrimethylsilane]porphyrin (2b). From 3.00 g of 4-(N,N-di-4-
(hexyloxy)phenylamino)benzaldehyde27 (6.3 mmol), 0.6 mL of
pyrrol (8.5 mmol) and 0.31 mL of (trimethylsilyl)-propiolalde-
hyde (2.1 mmol) reacted according to the general procedure
affording 397 mg of 2b as yellow solid (11% yield). 1H NMR (400
MHz, CDCl3) d/ppm: 9.65 (d, J ¼ 4.7 Hz, 2H), 9.04 (d, J ¼ 4.7 Hz,
2H), 8.94 (d, J ¼ 4.7 Hz, 2H), 8.92 (d, J ¼ 4.7 Hz, 2H), 7.98 (dd, J ¼
8.6, 6.1 Hz, 6H), 7.39–7.34 (m, 12H), 7.30 (t, J ¼ 8.6 Hz, 6H),
7.00–6.95 (m, 12H), 4.00 (dd, J ¼ 12.6, 6.6 Hz, 12H), 1.87–1.78
(m, 12H), 1.54–1.46 (m, 12H), 1.41–1.35 (m, 24H), 0.96–0.92
(s, 18H), 0.63 (s, 9H), ꢀ2.27 (s, 2H). 13C NMR (100 MHz, CDCl3)
d/ppm: 155.7, 148.5, 140.6, 135.5, 133.55, 133.2, 127.1, 122.6,
117.95, 117.8, 115.4, 107.4, 101.3, 98.1, 68.25, 31.6, 29.3, 25.8,
22.6, 14.1, 0.4. UV-Vis (CH2Cl2) lmax/nm (log 3): 677 (3.72),
591 (4.15), 422 (5.06), 300 (4.68). FT-IR n/cmꢀ1
2951, 2927, 2857, 2137, 1599, 1504, 1469, 1239. MS (m/z)
(MALDI-TOF): calculated for C115H133N7O6Si: 1736.01; found:
1736.00 (M+).
General procedure for the synthesis of 3a and 3b. To a
solution of corresponding porphyrin 2a or 2b (1 mmol) in CHCl3
(84 mL mmolꢀ1) under argon was added a solution of
Zn(OAc)2$2H2O (5 mmol) in MeOH (2.5 mL mmolꢀ1). The
mixture was stirred at room temperature overnight. The reac-
tion was quenched with water and the mixture extracted with
CHCl3 (3 ꢃ 50 mL). The combined organic extracts were washed
with H2O and dried over anhydrous MgSO4. The solvent was
removed under reduced pressure and the product was puried
by column chromatography (silica gel, Hex–CHCl3, 1 : 1).
: 3314,
Transient photovoltage, charge extraction measurements
and transient absorption spectroscopy measurements were
carried out on systems described previously.26 Emission lifetime
studies were carried out on transparent 8–10 mm Al2O3 and
TiO2 lms with a Lifespec© picosecond uorescence lifetime
spectrometer from Edinburgh Instruments© with an instru-
ment response at half width at half maximum (HWHM) of
duration 420 ps.
J. Mater. Chem. A
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
Synthesis of [10,15,20-tri-(N,N-diphenylaniline)-5-ethynyltrimethyl-
Journal of Materials Chemistry A
132.0, 131.9, 131.9, 131.4, 131.1, 130.25, 130.1, 128.5, 128.4, 124.6,
121.4, 121.3. UV-Vis (CH2Cl2) lmax/nm (log 3): 630 (4.18), 572
(4.02), 450 (5.08), 306 (4.71). FT-IR n/cmꢀ1: 3436, 3061, 3031, 2956,
2924, 2186, 1738, 1589, 1405, 1330, 1316, 1278, 1173, 698. MS (m/
z) (MALDI-TOF): calculated for C83H55N7O2Zn: 1245.37; found:
1245.4 (M+).
silane]porphinato zinc(II) (3a). From 250 mg of porphyrin 2a (0.22
mmol), reacted according to the general procedure, affording 251
mg of 3a as a green solid (95% yield). 1H NMR (400 MHz, CDCl3) d/
ppm: 9.79 (d, J ¼ 4.7 Hz, 2H), 9.14 (d, J ¼ 4.7 Hz, 2H), 9.07 (d, J ¼
4.7 Hz, 2H), 9.05 (d, J ¼ 4.7 Hz, 2H), 8.10–8.05 (m, 6H), 7.49 (t, J ¼
8.4 Hz, 6H), 7.46–7.43 (t, 24H), 7.19–7.14 (m, 6H), 0.66 (s, 9H). 13
C
Synthesis of [10,15,20-tri(N,N-(bis(4-hexyloxy)phenyl)aniline)-5-
5-carboxy-phenylethynyl-porphyrinato] zinc(II) (1b). From 100 mg
of porphyrin 3b (0.13 mmol) reacted according to the general
NMR (100 MHz, CDCl3) d/ppm: 150.04, 149.94, 147.85, 147.82,
147.36, 136.32, 135.40, 135.32, 132.23, 131.79, 130.93, 129.48,
124.81, 123.21, 122.93, 121.82, 121.36, 121.26, 107.66, 101.12,
99.37, 0.41. UV-Vis (CH2Cl2) lmax/nm (log 3): 619 (4.25), 571 (4.19),
444 (5.31), 305 (4.81). FT-IR n/cmꢀ1: 3060, 3030, 2955, 1924, 2853,
2138, 1589, 1490, 1329, 1315, 1280, 996, 842, 699. MS (m/z)
(MALDI-TOF): calculated for C79H59N7SiZn: 1197.39; found (M +
H+): 1197.6, 1198.6, 1199.6, 1200.5, 1201.5, 1202.5.
1
procedure giving 65 mg of 1b as a green solid (63% yield). H
NMR (400 MHz, CDCl3/d5-pyridine) d/ppm: 9.71 (d, J ¼ 4.5 Hz,
2H), 9.03 (d, J ¼ 4.5 Hz, 2H), 8.90 (d, J ¼ 4.5 Hz, 2H), 8.87 (d, J ¼
4.5 Hz, 2H), 8.27 (d, J ¼ 6.8 Hz, 2H), 8.03 (d, J ¼ 6.8 Hz, 2H), 7.90
(dd, J ¼ 8.5, 6.7 Hz, 6H), 7.30–7.19 (m, 24H), 6.91–6.86 (m, 6H),
3.93–3.88 (m, 12H), 1.72 (q, J ¼ 6.7 Hz, 12H), 1.45–1.37 (m, 12H),
1.31–1.24 (m, 24H), 0.83 (t, J ¼ 7.0 Hz, 18H). 13C NMR (100 MHz,
CDCl3/d5-pyridine) d/ppm: 207.0, 155.6, 125.2, 150.9, 149.9,
149.9, 148.0, 140.9, 140.8, 134.9, 134.8, 132.8, 131.9, 131.4,
131.0, 130.1, 126.9, 122.05, 118.0, 117.9, 115.4, 97.4, 95.1, 68.2,
31.6, 30.9, 29.3, 25.75, 22.6, 14.0. UV-Vis (CH2Cl2) lmax/nm
(log 3): 637 (4.39), 579 (4.07), 449 (5.12), 305 (4.77). FT-IR
n/cmꢀ1: 3424, 3120, 3039, 2852, 2927, 2859, 2185, 1799, 1688,
1603, 1503, 1467, 1314, 1278, 1240, 1167, 998, 828, 794. MS (m/z)
(MALDI-TOF): calculated for C119H127N7O8Zn: 1845.90; found:
1846.4 (M+).
Synthesis of [10,15,20-tri-(N,N-(bis(4-hexyloxy)phenyl)aniline)-
5-ethynyltrimethylsilane]porphinato zinc(II) (3b). From 230 mg of
porphyrin 2b (0.13 mmol), reacted according to the general
procedure, affording 217 mg of 3b as green solid (91% yield). 1H
NMR (400 MHz, CDCl3) d/ppm: 9.75 (d, J ¼ 4.7 Hz, 2H), 9.13 (d,
J ¼ 4.7 Hz, 2H), 9.05 (d, J ¼ 4.7 Hz, 2H), 9.03 (d, J ¼ 4.7 Hz, 2H),
7.98 (dd, J ¼ 8.6, 7.0 Hz, 6H), 7.36 (dd, J ¼ 9.0, 5.8 Hz, 12H), 7.30
(t, J ¼ 8.6 Hz, 6H), 6.95 (dd, J ¼ 9.0, 6.7 Hz, 12H), 3.97 (dd, J ¼
13.0, 6.5 Hz, 12H), 1.85–1.75 (m, 12H), 1.54–1.45 (m, 12H), 1.41–
1.35 (m, 24H), 0.95–0.91 (m, 18H), 0.63 (s, 9H). 13C NMR (100
MHz, CDCl3) d/ppm: 152.45, 150.9, 150.2, 150.1, 148.3, 140.8,
140.8, 135.3, 135.2, 134.1, 133.0, 132.2, 131.7, 130.7, 127.0,
123.4, 122.15, 118.0, 117.9, 115.4, 107.95, 100.9, 99.0, 68.3, 31.6,
29.3, 25.8, 22.6, 14.1, 0.4. UV-Vis (CH2Cl2) lmax/nm (log 3): 623
(4.22), 570 (4.10), 441 (5.09), 301 (4.79). FT-IR n/cmꢀ1: 2954,
2951, 2854, 2138, 1604, 1503, 1235. MS (m/z) (MALDI-TOF):
calculated for C115H131N7O6SiZn: 1797.92; found: 1798.5 (M+).
General procedure for the synthesis of 1a and 1b. To a
solution of the corresponding porphyrin 3a or 3b (1 mmol) in
CH2Cl2 (200 mL mmolꢀ1), TBAF (1.25 mmol, 1 M in THF) was
added under argon. The solution was stirred at room tempera-
ture for 1 hour. The mixture was quenched with H2O and
extracted with CH2Cl2 (3 ꢃ 50 mL). The combined organic layer
was dried over anhydrous MgSO4 and the solvent was removed
under reducer pressure. The residue and 4-iodobenzoic acid
(5 mmol) were dissolved in dry THF (200 mL mmolꢀ1) and Et3N
(120 mL mmolꢀ1). The solution was degassed with argon for 15
min, Pd2(dba)3 (0.3 mmol) and AsPh3 (2 mmol) were added to
the mixture and the solution was reuxed over night. The solvent
was removed under reduced pressure. The product was puried
by column chromatography (silica gel, CHCl3: MeOH 95 : 5).
Synthesis of [10,15,20-tri(N,N-diphenylaniline)-5-carboxyphenyl-
ethynyl-porphyrinato] zinc(II) (1a). From 100 mg of porphyrin 3a
(0.22 mmol) reacted according to the general procedure giving 73
mg of 1a as a green solid (70% yield). 1H NMR (400 MHz, CDCl3/
d5-pyridine) d/ppm: 9.80 (d, J ¼ 4.6 Hz, 2H), 9.09 (d, J ¼ 4.6 Hz,
2H), 8.97 (d, J ¼ 4.6 Hz, 2H), 8.94 (d, J ¼ 4.6 Hz, 2H), 8.34 (d, J ¼
8.3 Hz, 2H), 8.11 (d, J ¼ 8.3 Hz, 2H), 8.05 (t, J ¼ 8.3 Hz, 6H), 7.70–
7.64 (m, 4H), 7.57–7.52 (m, 2H), 7.49–7.38 (m, 24H), 7.16–7.11 (m,
6H). 13C NMR (100 MHz, CDCl3/d5-pyridine) d/ppm: 207.05, 152.8,
152.3, 150.7, 149.8, 147.9, 147.1, 137.0, 132.0, 132.7, 132.1, 132.1,
Acknowledgements
Financial support from the Ministry of Science and Innovation
of Spain, (CTQ2010-17498, PLE2009-0038 and Consolider-
Ingenio Projects HOPE CSD2007-00007) is gratefully acknowl-
edged. EP would also like to thank the EU for the ERCstg
PolyDot, and the Catalan government for the 2009 SGR-207
projects.
Notes and references
1 B. O'Regan and M. Gratzel, Nature, 1991, 353, 737–740.
2 A. Hagfeldt, G. Boschloo, L. Sun, L. Kloo and H. Pettersson,
Chem. Rev., 2010, 110, 6595–6663.
3 J. N. Clifford, E. Martinez-Ferrero and E. Palomares, J. Mater.
Chem., 2012, 22, 12415–12422.
4 N. Robertson, Angew. Chem., Int. Ed., 2008, 47, 1012–1014.
5 J. N. Clifford, M. Planells and E. Palomares, J. Mater. Chem.,
2012, 22, 24195–24201.
¨
6 A. Mishra, M. K. R. Fischer and P. Bauerle, Angew. Chem., Int.
Ed., 2009, 48, 2474–2499.
7 T. Bessho, S. M. Zakeeruddin, C.-Y. Yeh, E. W.-G. Diau and
M. Gratzel, Angew. Chem., Int. Ed., 2010, 49, 6646–6649.
8 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 and E. W.-G. Diau,
Chem. Commun., 2011, 47, 8910–8912.
9 C.-L. Wang, Y.-C. Chang, C.-M. Lan, C.-F. Lo, E. Wei-Guang
Diau and C.-Y. Lin, Energy Environ. Sci., 2011, 4, 1788–1795.
10 S. M. Feldt, E. A. Gibson, E. Gabrielsson, L. Sun, G. Boschloo
and A. Hagfeldt, J. Am. Chem. Soc., 2010, 132, 16714–16724.
¨
This journal is ª The Royal Society of Chemistry 2013
J. Mater. Chem. A

View Article Online
Journal of Materials Chemistry A
Paper
11 M. Wang, C. Gratzel, S. M. Zakeeruddin and M. Gratzel,
Energy Environ. Sci., 2012, 5, 9394–9405.
12 J. Cong, X. Yang, L. Kloo and L. Sun, Energy Environ. Sci.,
2012, 5, 9180–9194.
M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai,
M. Klene, X. Li, J. E. Knox, H. P. Hratchian, J. B. Cross,
V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts,
R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi,
C. Pomelli, J. W. Ochterski, P. Y. Ayala, K. Morokuma,
G. A. Voth, P. Salvador, J. J. Dannenberg, V. G. Zakrzewski,
S. Dapprich, A. D. Daniels, M. C. Strain, O. Farkas,
D. K. Malick, A. D. Rabuck, K. Raghavachari,
J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul, S. Clifford,
J. Cioslowski, B. B. Stefanov, G. Liu, A. Liashenko,
P. Piskorz, I. Komaromi, R. L. Martin, D. J. Fox, T. Keith,
M. A. Al-Laham, C. Y. Peng, A. Nanayakkara,
M. Challacombe, P. M. W. Gill, B. Johnson, W. Chen,
M. W. Wong, C. Gonzalez and J. A. Pople, R. D. Gaussian
03, Gaussian Inc., Wallingford CT, 2004.
13 E. Mosconi, J.-H. Yum, F. Kessler, C. J. Gomez-Garcia,
C. Zuccaccia, A. Cinti, M. K. Nazeeruddin, M. Graetzel and
F. De Angelis, J. Am. Chem. Soc., 2012, 134, 19438–19453.
14 A. Yella, H. W. Lee, H. N. Tsao, C. Y. Yi, A. K. Chandiran,
M. K. Nazeeruddin, E. W. G. Diau, C. Y. Yeh,
S. M. Zakeeruddin and M. Gratzel, Science, 2011, 334, 629–
634.
15 L. L. Li and E. W. G. Diau, Chem. Soc. Rev., 2013, 42, 291–304.
16 H. Imahori, S. Hayashi, H. Hayashi, A. Oguro, S. Eu,
T. Umeyama and Y. Matano, J. Phys. Chem. C, 2009, 113,
18406–18413.
17 M. V. Martinez-Diaz, G. de la Torre and T. Torres, Chem. 22 Y. Bai, J. Zhang, D. Zhou, Y. Wang, M. Zhang and P. Wang,
Commun., 2010, 46, 7090–7108. J. Am. Chem. Soc., 2011, 133, 11442–11445.
18 J. S. Lindsey, S. Prathapan, T. E. Johnson and R. W. Wagner, 23 H.-P. Wu, Z.-W. Ou, T.-Y. Pan, C.-M. Lan, W.-K. Huang,
Tetrahedron, 1994, 50, 8941–8968.
19 B. Liu, W. Zhu, Y. Wang, W. Wu, X. Li, B. Chen, Y.-T. Long
and Y. Xie, J. Mater. Chem., 2012, 22, 7434–7444.
20 C. M. Cardona, W. Li, A. E. Kaifer, D. Stockdale and
G. C. Bazan, Adv. Mater., 2011, 23, 2367–2371.
21 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, J. A. Montgomery Jr,
H.-W. Lee, N. M. Reddy, C.-T. Chen, W.-S. Chao, C.-Y. Yeh
and E. W.-G. Diau, Energy Environ. Sci., 2012, 5, 9843–9848.
24 S. E. Koops, P. R. F. Barnes, B. C. O'Regan and J. R. Durrant,
J. Phys. Chem. C, 2010, 114, 8054–8061.
25 S. Tatay, S. A. Haque, B. O'Regan, J. R. Durrant,
W. J. H. Verhees, J. M. Kroon, A. Vidal-Ferran, P. Gavina
and E. Palomares, J. Mater. Chem., 2007, 17, 3037–3044.
T. Vreven, K. N. Kudin, J. C. Burant, J. M. Millam, 26 T. Zewdu, J. N. Clifford, J. P. Hernandez and E. Palomares,
S. S. Iyengar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, Energy Environ. Sci., 2011, 4, 4633–4638.
G. Scalmani, N. Rega, G. A. Petersson, H. Nakatsuji, 27 H.-M. Nguyen, R. S. Mane, T. Ganesh, S.-H. Han and N. Kim,
M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa,
J. Phys. Chem. C, 2009, 113, 9206–9209.
J. Mater. Chem. A
This journal is ª The Royal Society of Chemistry 2013