Towards Sustainable H2 Production: Rational Design of Hydrophobic Triphenylamine-based Dyes for Sensitized Ethanol Photoreforming

Article abstract of DOI:10.1002/cssc.201701707

Donor–acceptor dyes are a well-established class of photosensitizers, used to enhance visible-light harvesting in solar cells and in direct photocatalytic reactions, such as H₂ production by photoreforming of sacrificial electron donors (SEDs). Amines—typically triethanolamine (TEOA)—are commonly employed as SEDs in such reactions. Dye-sensitized photoreforming of more sustainable, biomass-derived alcohols, on the other hand, was only recently reported by using methanol as the electron donor. In this work, several rationally designed donor–acceptor dyes were used as sensitizers in H₂ photocatalytic production, comparing the efficiency of TEOA and EtOH as SEDs. In particular, the effect of hydrophobic chains in the spacer and/or the donor unit of the dyes was systematically studied. The H₂ production rates were higher when TEOA was used as SED, whereas the activity trends depended on the SED used. The best performance was obtained with TEOA by using a sensitizer with just one bulky hydrophobic moiety, propylenedioxythiophene, placed on the spacer unit. In the case of EtOH, the best-performing sensitizers were the ones featuring a thiazolo[5,4-d]thiazole internal unit, needed for enhancing light harvesting, and carrying alkyl chains on both the donor part and the spacer unit. The results are discussed in terms of reaction mechanism, interaction with the SED, and structural/electrochemical properties of the sensitizers.

Full text of DOI:10.1002/cssc.201701707

Accepted Article
Title: Towards Sustainable H2 Production: Rational Design of
Hydrophobic Triphenylamine-based Dyes for Sensitized Ethanol
Photoreforming
Authors: Alessio Dessì, Matteo Monai, Matteo Bessi, Tiziano Montini,
Massimo Calamante, Alessandro Mordini, Gianna Reginato,
Cosimo Trono, Paolo Fornasiero, and Lorenzo Zani
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemSusChem 10.1002/cssc.201701707
Link to VoR: http://dx.doi.org/10.1002/cssc.201701707
A Journal of
www.chemsuschem.org

10.1002/cssc.201701707
ChemSusChem
FULL PAPER
Towards Sustainable H2 Production: Rational Design of
Hydrophobic Triphenylamine-based Dyes for Sensitized Ethanol
Photoreforming
Alessio Dessì,[a],§ Matteo Monai,[b],§ Matteo Bessi,[a,c],§ Tiziano Montini,[b] Massimo Calamante,[a]
Alessandro Mordini,[a] Gianna Reginato,[a] Cosimo Trono,[d] Paolo Fornasiero,[b],* Lorenzo Zani[a],*
Abstract: Donor-acceptor dyes are a well-established class of
photosensitizers, used to enhance visible light harvesting in solar cells
and in direct photocatalytic reactions, such as H2 production by
photoreforming of sacrificial electron donors (SEDs). Amines -
typically triethanolamine (TEOA) - are commonly employed as SEDs
in such reactions. Dye-sensitized photoreforming of more sustainable,
biomass-derived alcohols, on the other hand, was only recently
reported using methanol as the electron donor. In this work, several
rationally designed donor-acceptor dyes were used as sensitizers in
H2 photocatalytic production, comparing the efficiency of TEOA and
EtOH as SEDs. In particular, the effect of hydrophobic chains in the
spacer and/or the donor unit of the dyes was systematically studied.
The H2 production rates were higher when TEOA was used as SED,
while the activity trends were dependent on the SED used. With TEOA,
the best performance was obtained using a sensitizer (AD418)
Introduction
Hydrogen is an essential chemical for nowadays most important
industrial and manufacturing processes, ranging from ammonia
and methanol synthesis to fuel cells power generation and food
processing. However, despite the great efforts made in the last
decades, sustainable H2 production from low-carbon or
renewable energies remains challenging and will require large-
scale changes to our energy systems.[1]
While photocatalytic water splitting is the dream reaction for clean
H2 production, its low efficiency limits its feasibility.
Photoreforming of renewable feedstocks, in which an organic
compound acts as a sacrificial electron donor (SED), bypassing
the water oxidation reaction, is a valid alternative. In this field,
titanium dioxide, TiO2, is the most investigated material, because
it is cheap, nontoxic, chemically and biologically inert, and photo-
stable. In particular, when coupled with appropriate metal co-
catalysts such as, among others, Pt, Pd, Au or Ni, TiO2 performs
very well in many photocatalytic reactions,[2] and has been
demonstrated to efficiently mediate photoreforming of biomass-
derived SEDs (glucose, wood hydrolysate, simple alcohols).[3-10]
However, TiO2 does not absorb visible light, because of its large
band gap (3.2 eV, depending on polymorph and nanostructure),
so that only a narrow, UV (Ultraviolet) portion of the solar
spectrum can contribute to electron excitation from its valence
band (VB) to the conduction band (CB). Among the many
strategies to enhance light harvesting, coupling TiO2 with visible
light-absorbing moieties such as colored dyes is a well-
established approach, which is largely employed in the field of
solar cells (i.e. dye-sensitized solar cells (DSSCs)).[11] The dye
acts as a photosensitizer, or antenna, whose main purposes are
to efficiently absorb Vis light and trigger the remaining steps of the
hydrogen production process by electron injection into TiO2 CB.
Obviously, the photosensitizer must also be stable in the working
conditions, i.e. under continuous irradiation over long periods.
Figure 1 illustrates the working mechanism of a dye-sensitized
Pt/TiO2 photocatalyst for H2 production, summarized in Equations
(1-6).[12] First, the dye absorbs a photon and enters an excited
state (1). Electron injection into the CB of TiO2 results in charge
separation (2). The regeneration of the oxidized dye occurs by
oxidation of the SED agent (3). Notably, no holes are generated
in the semiconductor since TiO2 cannot generate photoelectrons
under visible light illumination, so the electron transfer must occur
between the SED and the excited dye for the photocatalytic cycle
to be completed. Electrons are then transferred from TiO2 to Pt(0)
nanoparticles adsorbed on the TiO2 surface, on which protons are
reduced to molecular hydrogen (4). This ideal process can be
endowed
with
just
one
bulky
hydrophobic
moiety,
propylenedioxythiophene (ProDOT), placed on the spacer unit. In the
case of EtOH, the best performing sensitizers (TTZ4-5) were the ones
featuring a thiazolo[5,4-d]thiazole internal unit, needed for enhancing
light harvesting, and carrying alkyl chains on both the donor part and
the spacer unit. The results are discussed in terms of reaction
mechanism, interaction with SED and structural/ electrochemical
properties of the sensitizers.
[a]
[b]
[c]
Dr. A. Dessì, M. Bessi, Dr. M. Calamante, Dr. A. Mordini, Dr. G.
Reginato, Dr. L. Zani
Institute for the Chemistry of Organometallic Compounds (ICCOM-
CNR)
Via Madonna del Piano 10, 50019 Sesto Fiorentino, Italy
E-mail: lorenzo.zani@iccom.cnr.it
Dr. M. Monai, Dr. T. Montini, Prof. P. Fornasiero
Department of Chemical and Pharmaceutical Sciences, ICCOM-
CNR Trieste Research Unit and INSTM Research Unit,
University of Trieste
via L. Giorgieri 1, 34127 Trieste, Italy
E-mail: pfornasiero@units.it
M. Bessi
Department of Biotechnology, Chemistry and Pharmacy
University of Siena
Via A. Moro 2, 53100 Siena, Italy
[d]
Dr. C. Trono
Institute of Applied Physics “Nello Carrara” (IFAC-CNR)
Via Madonna del Piano 10, 50019 Sesto Fiorentino, Italy
§
These authors contributed equally to this work.
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

10.1002/cssc.201701707
ChemSusChem
FULL PAPER
hindered by some unwanted events, such as: the relaxation of the
dye to its ground state before electron injection into the CB of the
semiconductor (5) and hole–electron recombination between
TiO2 and dye (6), which competes with the H2 generation (4) and
dye regeneration (3) steps.
molecules, the semiconductor, and the SED.[17] These
interactions can be tuned by a rational molecular engineering of
the organic sensitizer, leading to an enhancement of the
photocatalytic performance with a particular SED.
For instance, it has been recently shown[18,19] that hydrophilic
substituents placed on the terminal portion of organic dyes can
enhance their affinity towards the aqueous environment used in
the experiments, thus improving catalytic performances by
favoring the interaction with reactants in solution. On the other
hand, positioning hydrophobic side chains on dye molecules
could also have a positive impact on H2 production, since they
should prevent adsorption of water on TiO2 surface, reducing the
possibility of dye deactivation / desorption from the semiconductor.
Furthermore, in analogy to what happens in dye-sensitized solar
cells,[11] hydrophobic side chains should also be able to suppress
undesirable dye aggregation and increase the stability of oxidized
dyes by preventing fast charge recombination between them and
electrons in the conduction band of TiO2, as well as protecting
them from the attack of nucleophilic species present in the
reaction environment.
ℎ휈
푃푡/푇ꢀ푂2/퐷푦푒 → 푃푡/푇ꢀ푂2/퐷푦푒∗
(1)
푃푡/푇ꢀ푂2/퐷푦푒∗ →푃푡/푇ꢀ푂2(푒−)/퐷푦푒+ ∙
(2)
푃푡/푇ꢀ푂2(푒−)/퐷푦푒+ ∙ +푆퐸퐷 →푃푡/푇ꢀ푂2(푒−)/퐷푦푒 +푆퐸퐷+
(3)
1
푃푡/푇ꢀ푂2(푒−)/퐷푦푒 +퐻+ → 푃푡/푇ꢀ푂2/퐷푦푒 + 퐻2
2
(4)
푃푡/푇ꢀ푂2/퐷푦푒∗→푃푡/푇ꢀ푂2/퐷푦푒 +ℎ휈
푃푡/푇ꢀ푂2(푒−)/퐷푦푒+ ∙ →푃푡/푇ꢀ푂2/퐷푦푒
(5)
(6)
Some studies concerning the use of dyes endowed with
hydrophobic substituents have already been published,[20-23] and
in a few cases it was actually shown that elongation of the alkyl
chains had a positive influence on hydrogen generation.[24,25]
Nevertheless, a specific investigation on the effect that a different
positioning of the lipophilic side chains could have on dye
properties, and therefore on photocatalytic activity, was not
reported so far. In this work, we decided to test a series of D-π-A
dyes having hydrophobic side chains in different positions of their
molecular scaffold, presuming that such design could influence
both the reduction of dye desorption and the anti-aggregation /
recombination effect, and, at the same time, still allow SED
molecules dissolved in the aqueous environment to interact with
the oxidized dye. To do that, alkyl chains were placed on the
internal, conjugated section of the sensitizers and/or on their
donor part. The organic dyes were tested as sensitizers in H2
photocatalytic using either TEOA or EtOH as SEDs, in order to
Figure 1. Graphical representation of the working mechanism for H2 production
over dye-sensitized Pt/TiO2 photocatalysts.
In most cases, the organic sensitizers employed are
characterized by a donor-π-acceptor (D-π-A) architecture,[12] in
which the HOMO is located on the donor core and the LUMO on
the acceptor end-group. Upon irradiation, an electron is promoted
from the HOMO to the LUMO of the dye, corresponding to an
intramolecular charge transfer (ICT) transition. Once in its excited
state, electron transfer from the LUMO of the sensitizer to the
conduction band of TiO2 takes place, thus triggering the
photocatalytic cycle.
assess the possibility to produce hydrogen employing
renewable and nontoxic reducing agent.
a
Results and Discussion
A series of D-π-A organic dyes having a triphenylamine donor unit
(Figure 2) was synthesized and tested in the photo-sensitization
of Pt/TiO2 for the photocatalytic production of H2. The simple
organic dye D5 (Figure 2) was used as a reference and
systematically modified in order to get insight in structure-activity
and stability relations. Compound D5 is well-known for its use in
DSSCs[26] and has also been applied to sensitize Pt-TiO2
photocatalysts.[21] With the aim of evaluating the influence of
introducing hydrophobic groups in different positions of the
molecule, the performances of D5 were compared with those of
two derivate dyes: the previously reported sensitizer DF15,
having two hexyloxy chains on its terminal triarylamine group,[27]
The nature of the SED plays an important role in photocatalytic
reactions kinetics as well. The most commonly used electron
donors in dye-sensitized photocatalysis are TEOA and EDTA for
–
H2 production, while redox couples such as IO3 /I– and Fe2+/Fe3+
are used for water splitting.[13] On the other hand, among alcohols,
a very well-known class of SEDs in photocatalysis,[14] only MeOH
has been used in dye-sensitized systems.[12,15] In particular, to the
best of our knowledge, ethanol was never reported as SED for
dye-sensitized photocatalysts, despite its relevance as
sustainable feedstock, which can be cost-efficiently produced
from renewable biomasses.[16] Finally, it must be taken into
account that dye-sensitized systems are sensitive not only to the
relative energy levels of the dye and SED, but also to the electron
transfer kinetics that are related to the interactions among the dye
This article is protected by copyright. All rights reserved.

10.1002/cssc.201701707
ChemSusChem
FULL PAPER
and the newly synthesized dye MB25, having the intermediate
thiophene ring replaced by a propylenedioxythiophene (ProDOT)
moiety, bearing two n-pentyl chains (Figure 2). Such substitution
was aimed at favoring dye disaggregation and enhancing the
shielding effect of the TiO2 surface, as the alkyl chains on the
ProDOT group should stick out of the molecular plane due to the
tetrahedral geometry of its quaternary carbon atom.[28]
Considering that D5 has a relatively large HOMO-LUMO-gap with
a low-lying HOMO (Table 1), we anticipated that light-harvesting
properties could also be improved by introducing the alkyl chains
as a part of electron-donating alkoxy groups. Finally, in order to
study the effect of the spacer moiety, dye AD418, in which the
double bond was replaced by a thiophene ring, was also prepared
and tested together with known, more elaborated dyes TTZ3-5.
TTZ dyes have an electron withdrawing thiazolo[5,4-d]thiazole
moiety in the -scaffold and are characterized by long term
stability and by very high molar extinction coefficients, which has
made them especially suitable for the fabrication of thin-layer dye-
sensitized solar cells (Figure 2).[28,29] TTZ4 and TTZ5 carry
hydrophobic chains both on the central scaffold and on the donor
moiety, while TTZ3 only has chains on its central scaffold.
The synthesis of compound MB25 is outlined in Scheme 1. We
anticipated that its conjugated scaffold could be generated by a
diastereoselective Horner-Wadsworth-Emmons (HWE) reaction,
whose application required access to diethylphosphonate 4. To
this end, ProDOT aldehyde 1[30] was first reduced to the
corresponding primary alcohol 2 using sodium borohydride in a
methanol / dichloromethane mixture. Compound 2 was then
converted to the corresponding chloride 3 by reaction with SOCl2
in the presence of triethylamine and catalytic DMAP, and the
Figure 2. Structure of the dyes investigated in this study.
Scheme 1. Synthesis of compound MB25 starting from ProDOT aldehyde 1.
This article is protected by copyright. All rights reserved.

10.1002/cssc.201701707
ChemSusChem
FULL PAPER
Scheme 2. Synthesis of compound AD418 starting from ProDOT aldehyde 1 and substituted thiophene 8.
chloride was then reacted with triethylphosphite, under typical
Michaelis-Arbuzov conditions, to give phosphonate 4, however in
low yield (21%, over two steps). Since this result was mostly due
to the difficult isolation of compound 3, which underwent extensive
decomposition during chromatographic purification, we decided to
improve the process generating the halide in a single-pot
procedure. Thus, a THF solution of alcohol 2 was treated with
PBr3 at 0 °C for 30 min to obtain the corresponding bromide, which
was then reacted in situ with P(OEt)3 at reflux overnight. In these
conditions compound 4 was more conveniently prepared in 82%
catalyzed Stille-Migita coupling with bromide 10, obtained by
treatment of aldehyde with N-bromosuccinimide and
1
immediately employed without the need for further purification. In
the final step of the synthesis, aldehyde 11 underwent
Knoevenagel condensation with cyanoacetic acid in the presence
of an AcOH / AcONH4 mixture, to provide dye AD418 in a
satisfactory yield.
The optical properties of the complete series of dyes investigated
herein are reported in Figure 3 and Table 1. DF15, MB25 and
AD418 showed a significantly red-shifted absorption compared to
D5 in CH2Cl2 solution (Figure 3a), due to the presence of
electrondonating alkoxy substituents on the molecular backbone,
which enhance the donor-acceptor character of the compounds.
This effect was more pronounced for DF15, which bears the
alkoxy substituents on the donor part of the molecule. On the
other hand, TTZ dyes display a broader and stronger absorption
of visible light, with a maximum at 510-520 nm (Figure 3c). The
molar extinction coefficients for D5, DF15, MB25 and AD418 dyes
were comprised in the 2.7-3.5 × 104 M–1 cm–1 range, while TTZ
dyes have much higher molar extinction coefficients, above 8.0 ×
104 M–1 cm–1 in THF solution.
Diffuse reflectance spectra (DRS) of the sensitized Pt/TiO2
photocatalysts used for the H2 production tests (see below) were
recorded (Figure 3b, 3d) to investigate the optical properties of
the dyes in the solid state. All dyes exhibited a wide absorption
band encompassing the visible region in the 400-700 nm range,
confirming that they were successfully loaded onto the powder.
Similar to what observed in solution, DF15 and MB25 displayed
red-shifted spectra relative to D5, although in this case the
difference was somewhat smaller; AD418, on the other hand,
gave rise to a DRS spectrum very similar to that obtained with D5.
Concerning the TTZ dye series, no remarkable differences were
observed among the curves registered with the different dyes,
suggesting that their anchoring mode and adsorption density on
Pt/TiO2 was similar in all cases.
overall yield. Phosphonate
4 was then reacted with 4-
(diphenylamino)benzaldehyde (5) in the presence of a base, to
obtain disubstituted olefin 6. While employment of t-BuONa and
NaH allowed recovery of the product only in moderate yields (43-
50%, respectively), use of 1.7 eq. of NaHMDS proved more
efficient, leading to selective formation of olefin (E)-6 in 76% yield
(as demonstrated by analysis of the 1H-NMR coupling constants).
Formylation of compound (E)-6 was then easily accomplished by
metalation with n-BuLi in THF at –78 °C and subsequent quench
with N,N-DMF. However, 1H-NMR analysis of the crude reaction
mixture indicated formation of desired aldehyde 7 together with a
side-product in a 10:1 molar ratio, as shown by the presence of
two close aldehyde peaks at 9.88 – 9.86 ppm and two sets of
ProDOT CH2 signals at 3.9-4.1 ppm (Figure S1). Fortunately, we
determined that the crude 10:1 mixture could still be successfully
used for the final Knoevenagel reaction. Indeed, by treatment with
cyanoacetic acid in the presence of ammonium acetate / glacial
acetic acid, compound MB25 was predominantly formed, and
only a small amount of byproduct was identified by NMR analysis
of the crude. Finally, dye MB25 was obtained in pure form by
chromatographic purification and repeated washings with
pentane and methanol.
To prepare sensitizer AD418 (Scheme 2) compound 8 was
transformed into intermediate 9, according to
a reported
stannylation procedure.[31] The latter was used in a second Pd-
This article is protected by copyright. All rights reserved.

10.1002/cssc.201701707
ChemSusChem
FULL PAPER
Figure 3. UV-Vis absorption spectra of: (a) compounds D5, DF15, MB25 and AD418 in dichloromethane solution, (c) TTZ3-5 in THF solution {reproduced from ref.
20}. (b, d) Normalized DRS spectra of the compounds adsorbed on the Pt/TiO2 catalyst used for hydrogen production tests.
Table 1. Optical and electrochemical properties of the sensitizers employed in this work.
ε [M–1 cm–1
]
λmax Pt/TiO2 [nm]
Dye
λmax [nm]
501[e]
E0-0 [eV][a]
2.19
Eox [V][b]
+1.09[f]
+0.94[g]
+1.03
EHOMO [eV][c]
–5.69
ELUMO [eV][c,d]
–3.50
D5
27300
478
498
490
475
484
491
497
DF15
MB25
AD418
TTZ3
TTZ4
TTZ5
530[e,g]
515[e]
35040[g]
27925
2.07
–5.54
–3.47
2.14
–5.63
–3.49
512[e]
30400
2.16
+1.05
–5.65
–3.49
510[h,i]
518[h,i]
510[h,i]
81400[i]
86600[i]
94100[i]
2.16i
2.12i
2.16i
+1.02[i]
+0.77[i]
+0.91[i]
–5.62
–3.46
–5.37
–3.25
–5.51
–3.35
[a] Estimated from the corresponding Tauc plot; [b] Potentials vs. NHE; [c] Values relative to a vacuum, using a potential value of
4.6 ± 0.2 eV for NHE vs. vacuum[34]; [d] Values calculated using the following equation: ELUMO = EHOMO – E0-0; [e] Measured in
dichloromethane solution; [f] Value taken from ref. 35; [g] Values taken from ref. 27; [h] Measured in THF solution; [i] Values
reproduced from ref. 29.
The optical bandgaps (E0-0) of the dyes were estimated based on
the corresponding Tauc plot (Figure S2),[32,33] while their
electrochemical properties were determined by means of cyclic
voltammetry (Table 1, Figure S3). As expected, the presence of
electrondonating groups on the backbone of DF15, MB25 and
AD418 caused a negative shift in the ground-state oxidation
potential (Eox) compared to D5, in agreement with what observed
in the optical measurements. Their LUMO energy, on the other
This article is protected by copyright. All rights reserved.

10.1002/cssc.201701707
ChemSusChem
FULL PAPER
hand, was similar to that of D5 and in all cases appeared much
higher than that of the conduction band of TiO2 (–4.0 V vs.
vacuum), thus allowing electron injection from the excited
compounds to the semiconductor. Concerning TTZ4 and TTZ5,
they clearly showed higher HOMO orbitals compared to the other
dyes, which could influence the efficiency of dye regeneration
evaluate the fraction of light energy stored in the form of H2.[38,39]
LFE values after 3 h (LFE03) were almost the same for the two
compounds, but after 20 h (LFE20) the value relative to D5 was
almost 25% lower than that of DF15 (Table 2). Such enhanced
stability of DF15 is consistent with the protecting effect of the long
alkyl chains, preventing approach of potential quenchers present
processes, but at the same time presented higher LUMO energies, in the aqueous environment, in agreement with previous
so that a larger driving force for electron injection in the CB of TiO2
can be expected for them.
observations.[24] As a result, the overall quantity of produced
hydrogen and the TON recorded with DF15 were slightly superior
to those obtained with D5 (Table 2). MB25-sensitized catalysts
showed a much higher initial activity, as can be seen by LFE03,
but after 4-5 h a sharp decrease in H2 production was observed,
leading to LFE20 values similar to those of D5 and DF15 at the
end of the experiment. Nevertheless, thanks to the excellent
performances registered at the beginning of the reaction, TON
To perform dye-sensitized H2 photocatalytic production
experiments, the investigated dyes were adsorbed over
benchmark Pt/TiO2 nanocomposite catalyst, prepared by
photodeposition of Pt on TiO2 P25, as previously reported.[36]
Briefly, the dye was dissolved in THF and was loaded onto the
catalyst by mixing its solution with the nanoparticle suspension in
EtOH for 12 h in the dark. After filtration, the remaining solution
was colorless, indicating quantitative adsorption of the dye.
Experiments were first conducted using TEOA as SED at pH =
7.0, under irradiation with visible light (λ > 420 nm, to exclude
contribution from TiO2). After optimization of the experimental
conditions,[37] the optimal dye loading was set at 10 μmol g−1. The
observed hydrogen production rates with dyes D5, DF15, MB25
and AD418 are shown in Figure 4 and in Table 2.
Independently from the dye employed as sensitizer, the H2
production rates observed using the sensitized photocatalysts
showed an initial increase in the first 1─2 h of illumination, as
already observed for similar systems, due to chromatographic
effects and to the activation of passivated Pt under irradiation.[36]
D5 and DF15 showed similar and moderate hydrogen production
rates after stabilization (Figure 4), possibly due to a trade-off
between the better light-harvesting capability of DF15, as
demonstrated by its more intense and red-shifted UV-Vis
spectrum, and the more positive redox potential of D5, which
should facilitate its regeneration (for a detailed discussion, see
below). In any case, for dye D5 a rate decrease was observed
over time, which was less pronounced in the case of DF15. Such
phenomenon was evident by looking at the light-to-fuel efficiency
(LFE) of the two dyes, which was introduced to quantitatively
Table 2. Photocatalytic performances of the dye/Pt/TiO2 catalysts in H2
production with TEOA as SED under irradiation with visible light (λ > 420
nm) for the dyes tested in this study.
H2 amount
[c]
[d]
Dye
TON[b]
LFE03
LFE20
[μmol g–1 [a]
]
D5
1884
2371
2846
4359
1424
1550
3432
397
474
569
872
285
310
686
0.065%
0.066%
0.127%
0.117%
0.044%
0.043%
0.093%
0.038%
0.050%
0.041%
0.121%
0.043%
0.044%
0.094%
DF15
MB25
AD418
TTZ3
TTZ4
TTZ5
[a] Overall H2 amount produced after 20 h of irradiation per gram of catalyst;
[b] TON = (2 × H2 total amount after 20h of irradiation)/(dye loading); [c] Light-
to-fuel efficiency calculated after 3 h of irradiation; [d] Light-to-fuel efficiency
calculated after 20 h of irradiation.
Figure 4. H2 production from TEOA 10% v/v aqueous solution at pH = 7.0 under irradiation with Vis light (λ > 420 nm) from 0 to 20 h over Pt/TiO2 catalysts sensitized
with: A) D5; B) DF15; C) MB25; D) AD418.
This article is protected by copyright. All rights reserved.

10.1002/cssc.201701707
ChemSusChem
FULL PAPER
Figure 5. H2 production from TEOA 10% v/v aqueous solution at pH = 7.0 under irradiation with Vis light (λ > 420 nm) from 0 to 20 h over Pt/TiO2 catalysts sensitized
with: A) TTZ3; B) TTZ4; C) TTZ5.
value for MB25 was superior to those of the other dyes. The high
initial rate of hydrogen production for MB25 compared to DF15
was encouraging, as it appeared to confirm our hypothesis that
moving the bulky and hydrophobic alkyl chains from the donor to
the intermediate part of the molecular scaffold could have a
beneficial effect on dye regeneration (by facilitating approach of
the SED to the sensitizer) while slowing down undesired
recombination events (by better coverage of the TiO2 surface).
Nevertheless, to better understand the reasons of the observed
progressive deactivation of MB25, the compounds anchored to
TiO2 at the end of the photocatalysis were characterized with the
aim of identifying potential decomposition side-products (See SI
for detailed information). UV-Vis and ESI-MS spectroscopy
(Figure S4-S6) suggest that a reaction of the unsaturated
backbone of MB25 takes place,[40,41] possibly by protonation
followed by nucleophilic addition of water or of a species derived
from TEOA (see Figure S7 for a possible pathway involving H2O).
Such process could be favored by the presence of a double bond
in the vicinity of the ProDOT unit, which could facilitate protonation
thanks to its electrondonating ability.
Such hypothesis is supported by the fact that the photocatalyst
obtained by sensitization with AD418, an analog of MB25 in which
the double bond was replaced with an extra thiophene moiety
(see Figure 2 and Scheme 2), was very active and stable until the
end of the experiment, as evident from the hydrogen production
rate curve in Figure 4 and the comparison of LFE03 and LFE20
values (Table 2). Clearly, removing the double bond from the
structure of MB25, while keeping all other properties as close as
possible to the original dye, provided an equally efficient but much
more stable sensitizer. As a consequence, the TON value
obtained with AD418 was the highest among the tested dyes.[42]
Table 2), relatively close to AD418, suggesting that alkyl sulfide
substituents on the donor have a beneficial effect in the
photocatalytic process. Notably, TTZ5 was also the best
performing sensitizer among TTZ dyes in recent DSSC studies
carried out by some of us.[28,29]
The dyes employed in the present study have ground-state
oxidation potentials ranging from +0.77 V to +1.09 V vs. NHE
(Table 1), and therefore very close to the reported TEOA redox
potential (+0.82 to +1.07 V vs. NHE, depending on the
conditions).[43,44] Despite that, H2 production was operative in all
cases, clearly demonstrating that dye regeneration was not
hampered even in the presence of
a seemingly small
overpotential. A closer inspection of the data did not reveal any
obvious trend linking Eox values with H2 production rates (Table 1
and 2). Clearly, such observation reflects the fact that H2 evolution
activity depends on the delicate balance of several different
factors, among which the properties of the water/dye/TiO2
interface, influenced by the nature and position of the hydrophobic
alkyl chains placed on the sensitizers, play an important role.
In view of the promising results obtained with TEOA, we were
interested in evaluating the catalytic performances of the dye
sensitized Pt/TiO2 photocatalysts in combination with a different
sacrificial reagent. In particular, ethanol was chosen as SED since
it is an environmentally friendly and non-toxic feedstock that can
be produced on a large scale from renewable biomass.[16] Ethanol
photoreforming has already been investigated for the production
of H2 using several metal/TiO2 photocatalytic systems,[5,6,9,14] but,
to the best of our knowledge, its use as a sacrificial reagent in
combination with a dye sensitized photocatalyst has never been
reported.
The activity trend emerging from these experiments was very
different from that observed in the presence of TEOA. Indeed,
when dyes D5, DF15 or MB25 were used as photosensitizers for
the Pt/TiO2 catalyst in an aqueous EtOH solution, no hydrogen
production could be detected. On the contrary, AD418 and TTZ3-
5 sensitized photocatalysts were active and quite stable in H2
photocatalytic production, although their TONs were lower than
Thiazolothiazole
dyes-sensitized
photocatalysts
showed
intermediate hydrogen production rates compared to the other
compounds when TEOA was used as a SED, and were also
remarkably stable (Figure 5 and Table 2). TTZ4 displayed only
slightly higher TON values than TTZ3, while TTZ5 showed a much
higher H2 production activity (slightly higher than 3400 μmol g–1,
This article is protected by copyright. All rights reserved.

10.1002/cssc.201701707
ChemSusChem
FULL PAPER
Figure 6. H2 production rates from EtOH 20% v/v aqueous solution under irradiation with Vis light (λ > 420 nm) from 0 to 20 h over Pt/TiO2 catalysts sensitized with:
A) AD418; B) TTZ3; C) TTZ4; D) TTZ5. Please mind the different scale with respect to Figures 4 and 5.
Table 3. Photocatalytic performances of the dye/Pt/TiO2 catalysts in H2
production with EtOH as SED under irradiation with visible light (λ > 420 nm)
for dyes AD418 and TTZ3-5.
potentials for the conversion of ethanol to acetaldehyde (see
below).
2퐷푦푒+ ∙ +퐶퐻3퐶퐻2푂퐻 → 2퐷푦푒 +퐶퐻3퐶퐻푂 +2퐻+
(7)
(8)
H2 amount
[μmol g–1 [a]
[c]
[d]
Dye
TON[b]
LFE03
LFE20
]
1
푃푡/푇ꢀ푂2(푒−) +퐻+ →푃푡/푇ꢀ푂2/퐷푦푒 + 퐻2
2
AD418
TTZ3
TTZ4
TTZ5
403
81
0.010%
0.014%
0.033%
0.033%
0.009%
0.012%
0.027%
0.030%
With the aim to understand the reason of the different behavior of
the various dyes in the presence of EtOH, we re-examined their
structural and optoelectronic properties. Compared to dyes
AD418 and TTZ3, both compounds TTZ4-5 have electron
donating chains placed on the terminal triarylamine unit and
display red-shifted absorption spectra when anchored on TiO2,
superior molar extinction coefficients and higher LUMO energies
(see the corresponding values in Table 1). The latter observation
suggests that a larger driving force for electron injection is
especially beneficial to H2 production efficiency when using EtOH
as a SED. We hypothesized that this could be due to a negative
shift of TiO2 CB[45] caused by the adsorption on its surface of
negatively-charged ethoxide species formed by the dissociative
adsorption of EtOH on TiO2,[46,47] in turn making electron injection
from the LUMO of the excited sensitizers more difficult. To
determine if that was the case, we performed selected
experiments on AD418-sensitized photocatalysts (those showing
the largest performance difference when switching from TEOA to
EtOH) by running the reaction with TEOA both in the absence and
in the presence of EtOH (20% v/v). The results of such
experiments suggested that electron injection was not
significantly hindered by the alleged ethoxide adsorption, since
the H2 production activity observed using TEOA was only slightly
(and positively) affected by EtOH addition, probably due to its
capability to act as a “supplementary” SED itself (Figure S8). On
the other hand, additional species present in solution (such as, for
example, TEOAH+ and Cl–, coming from neutralization with aq.
HCl, see Experimental Section) could also have an influence on
the position of TiO2 CB, making it difficult to draw a definitive
conclusion. However, the resulting effect on hydrogen production
rate seemed not very large (Figures 4-6), and therefore CB band
493
99
1102
1105
220
221
[a] Overall H2 amount produced after 20 h of irradiation per gram of catalyst;
[b] TON = (2 × H2 total amount after 20h of irradiation)/(dye loading); [c] Light-
to-fuel efficiency calculated after 3 h of irradiation; [d] Light-to-fuel efficiency
calculated after 20 h of irradiation.
those observed when TEOA was used as SED, ranging from 80
to 220 (Figure 6, Table 3). More in detail, the catalytic
performance obtained with dye TTZ3 was slightly superior to that
registered with dye AD418, although the TON improvement was
inferior to 20% (Table 3). On the other hand, employment of dyes
TTZ4-5 was more successful, resulting in an increase of more
than 2.5 times in hydrogen production rates and TONs.
Interestingly, TTZ4 and TTZ5 exhibited comparable performance,
and were also sufficiently stable over the 20 h of the experiment,
as evidenced by the comparison between the light-to-fuel
efficiency values calculated at the beginning and at the end of the
test. Notably, acetaldehyde was the only product observed in the
liquid phase, in stoichiometric amounts compared to H2 in the gas
phase. This is consistent with the oxidation of ethanol into
acetaldehyde by the oxidized dye and reduction of protons to H2
by the photoelectrons injected in the TiO2 VB, according to
reactions (7) and (8). Conversely to un-sensitized photocatalysts,
in which photogenerated holes are involved in the process,[3-10] in
this case the dyes act as oxidating agents, having suitable
This article is protected by copyright. All rights reserved.

10.1002/cssc.201701707
ChemSusChem
FULL PAPER
shift is likely not the main reason of the lower performances in the Conclusions
presence of the sole EtOH compared to TEOA
Fluorescence lifetime measurements of dyes D5, DF15, AD418
and TTZ5 were then carried out using the time-correlated single
photon counting (TCSPC) technique to get further insights in the
reason for the different trends observed in the hydrogen
production experiments. Samples of the dyes were dissolved in
methylene chloride, irradiated with a green led (λ = 499 nm), and
the corresponding fluorescence decay curves were recorded
(Figure S9). After deconvolution from the instrument response
function (IRF), all decay profiles could be fitted with a single
exponential function, giving similar excited state lifetimes (τ)
comprised in the 1.4 – 2.5 ns range, which did not change
significantly when using different solvents and did not seem to
correlate with the results of the photocatalytic experiments. Upon
absorption of the dyes on TiO2, a rapid fluorescence quenching
was observed in all cases, which confirmed the dyes capability to
inject electrons in the semiconductor CB. These results suggest
that the different performances of the dyes do not depend on
significant differences in the corresponding excited state lifetimes.
Clearly, although thiazolothiazole-bearing dyes were active and
stable in H2 photocatalytic production, ethanol appeared a much
In this work, we presented the results obtained in the
photocatalytic hydrogen production using dye-sensitized Pt/TiO2
catalysts in combination with two different sacrificial reagents
(TEOA and EtOH). All sensitizers were metal-free organic
compounds characterized by a D-π-A structure, and some of
them (compounds TTZ3-5) featured a thiazolo[5,4-d]thiazole
internal unit introduced to modulate their light-absorption and
electronic properties. The sensitizers were endowed with
hydrophobic alkyl chains on different parts of their molecular
scaffold, whose position and number was systematically altered
to evaluate their influence on catalytic performances.
When employing TEOA as SED, the best result was obtained with
sensitizer AD418, carrying a propylenedioxythiophene (ProDOT)
moiety on the intermediate part of the structure. Comparison of its
performance with that yielded by compound DF15 (bearing linear
alkoxy substituents on its donor group) supported the hypothesis
that placing alkyl chains on the conjugated molecular scaffold
rather than on the donor could have a beneficial effect on dye
regeneration while slowing down undesired recombination events.
Thiazolothiazole-bearing dyes, in particular TTZ4 and TTZ5
provided the best results when using EtOH as SED. Such
behavior is especially interesting as use of EtOH as a sacrificial
reagent in combination with a dye sensitized photocatalyst had
never been described so far. Analysis of the molecular structure
as well as their electronic properties of the dyes suggested that in
this case dye regeneration is slower, so charge recombination
between injected electrons in TiO2 and the oxidized dye can
less efficient SED compared to TEOA, despite being
a
thermodynamically competent reducing agent. Indeed, its one-
electron oxidation potential in aqueous solution has been
measured several times by means of cyclic voltammetry in a
range of different conditions,[48-51] and was found to shift from
approx. 0.0 V vs. NHE in strongly basic medium[52] to approx. +0.9
V vs. NHE in strongly acidic solution,[53] values generally
compatible with the ground-state oxidation potentials of all dyes
used in this study (+0.77 - +1.09 V vs. NHE, Table 1). Thus, the
observed results suggest that the efficiency of the selected SED
is not only dictated by its thermodynamic ability to reduce the
oxidized photosensitizer, but also by the kinetic barrier for the
electron transfer process. In this regard, it has been already
observed that TEOA is a more efficient hole-scavenger than
methanol or ethanol in photocatalytic H2 generation processes
using inorganic[54,55] or carbon-based catalysts,[56] and this
difference has been ascribed to the fact that TEOA can use the
lone pair of nitrogen for the first electron transfer,[57] in agreement
with earlier reports indicating that only the deprotonated form of
the amine is an active reducing agent.[43] Based on relevant
hinder the photocatalytic process. Thus,
a stronger light
harvesting capability coupled with a larger driving force for
electron injection can help increasing the electron density on TiO2,
thereby enhancing hydrogen production rates.
This study indicates that the molecular design of sensitizers
should be carefully adjusted depending on the particular SED
(and set of conditions) used in the experiments in order to yield
the best results in hydrogen production experiments. Finely tuned
sensitizers may lead to a breakthrough in dye-sensitized
photocatalytic H2 production sustainability, extending the range of
efficient SEDs to alcohols and possibly other biomass-derived
feedstocks.
literature
references
on
TEOA[58]
and
alcohol
photoreforming,[3,5,10,14] a mechanistic proposal for the two
different regeneration processes is presented in Figure S10,
bearing in mind that in this instance no photogenerated holes are
present on TiO2 due to the filtering of simulated solar light.
In our case, the slower dye regeneration when ethanol is used
could compete with charge recombination between injected
electrons and the oxidized dye, hampering the H2 production
process. In this hypothesis, the excellent spectroscopic properties
and the large driving force for electron injection displayed by dyes
TTZ4-5 should enhance the efficiency of the elementary light-
harvesting and charge injection steps, increasing the electron
density on TiO2 and ultimately improving hydrogen production
rates.
Experimental Section
General Experimental Remarks. All air-sensitive reactions were
performed under inert atmosphere in a flame- or oven-dried apparatus
using Schlenk techniques. Solvents used in metal-catalyzed reactions
were degassed by means of the “freeze-pump-thaw” method.
Tetrahydrofuran (THF) was distilled over metallic sodium with
benzophenone as an indicator, methanol was distilled from metallic
magnesium in the presence of iodine (cat.), CH2Cl2 was distilled over CaH2.
Toluene was dried on a resin exchange Solvent Purification System
(MBraun).
carbaldehyde
3,3-Dipentyl-3,4-dihydro-2H-thieno[3,4-b][1,4]dioxepine-6-
(1),[30]
N,N-diphenyl-4-(5-(tributylstannyl)thiophen-2-
yl)aniline (9),[31] and dyes DF15[27] and TTZ3-5[28,29] were prepared as
previously reported. 4-(diphenylamino)-benzaldehyde (5) and all other
chemicals employed were commercially available and used as received.
Thin-layer chromatography was carried out on aluminum-supported plates
This article is protected by copyright. All rights reserved.

10.1002/cssc.201701707
ChemSusChem
FULL PAPER
coated with a fluorescent indicator; detection was carried out using UV light
(λ = 254 and 365 nm) and permanganate or molybdophosphoric acid
solutions followed by heating. Flash column chromatography was
performed using Merck Kieselgel 60 (300-400 mesh) as the stationary
phase. 1H-NMR spectra were recorded at 300 or 400 MHz, 13C-NMR
spectra were recorded at 75.5 or 100.6 MHz, respectively, and 31P-NMR
spectra were recorded at 121 MHz. Chemical shifts were referenced to the
residual solvent peak (CDCl3, δ 7.26 ppm for 1H-NMR, δ 77.16 ppm for
13C-NMR; THF-d8, δ 1.72 and 3.58 ppm for 1H-NMR, δ 25.31 and 67.21
ppm for 13C-NMR). FT-IR spectra were recorded in the range 4000-400
cm–1 with a 2 cm–1 resolution. ESI-MS spectra were obtained by direct
injection of the sample and are reported in the form m/z (intensity relative
to base = 100).
aqueous phase was washed with fresh ethyl acetate (3  10 mL), then the
combined organic phases were washed with brine (20 mL) and dried with
Na2SO4. After filtration and removal of the solvent under vacuum, the
reaction crude was purified by flash column chromatography (SiO2,
petroleum ether/AcOEt 30:1). Desired product 6 (83 mg, 0.15 mmol, 76%
yield) was isolated as a dense yellow oil. 1H–NMR (300 MHz, CDCl3): δ =
7.30–7.37 (m, 2H), 7.21–7.30 (m, 4H), 7.07–7.15 (m, 5H), 6.98–7.06 (m,
4H), 6.73 (d, J = 16.2 Hz, 1H), 6.28 (s, 1H), 3.92 (s, 2H), 3.86 (s, 2H),
1.26–1.39 (m, 16H), 0.91 (t, J = 7.0 Hz, 6H) ppm. 13C–NMR (75 MHz,
CDCl3): δ = 150.1, 147.7, 147.1, 146.7, 131.8, 129.4, 127.2, 125.6, 124.5,
123.8, 123.1, 122.0, 117.4, 101.8, 77.9, 77.7, 43.9, 32.8 (2), 32.0 (2),
22.7 (2), 22.6 (2), 14.2 (2) ppm. IR (KBr): 휈̃ = 3026, 2927, 1591, 1483,
1375, 1277, 1030 cm–1. ESI–MS: m/z = 565 [M]+.
(3,3-dipentyl-3,4-dihydro-2H-thieno[3,4-b][1,4]dioxepin-6-
(E)-8-(4-(diphenylamino)styryl)-3,3-dipentyl-3,4-dihydro-2H-
yl)methanol (2). 3,3-dipentyl-3,4-dihydro-2H-thieno[3,4-b][1,4]dioxepine-
6-carbaldehyde (1, 1.02 g, 3.15 mmol, 1.0 eq) was dissolved in CH2Cl2 (15
mL) and methanol (20 mL), then, after cooling to 0 °C, sodium borohydride
(0.202 g, 5.34 mmol, 1.7 eq.) was added into the reaction vessel. The
resulting mixture was stirred at 0 °C for 30 minutes, then at room
temperature for other 30 minutes. A saturated aqueous solution of NH4Cl
(45 mL) was added into the flask and the mixture was stirred until hydrogen
production ceased, then the two phases were separated. The aqueous
phase was extracted with CH2Cl2 (2  15 mL), then the combined organic
phases were washed with brine and dried with Na2SO4. After filtration,
removal of the solvent afforded alcohol 2 (1.01 g, 3.09 mmol, 98% yield)
as an orange oil which was used for the next step without further
purification. 1H–NMR (300 MHz, CDCl3): δ = 6.36 (s, 1H), 4.61 (s, 2H),
3.84 (s, 2H), 3.82 (s, 2H), 2.41 (bs, 1H), 1.26–1.34 (m, 16H), 0.89 (t, J =
7.0 Hz, 6H) ppm. 13C–NMR (75 MHz, CDCl3): δ = 149.5, 146.8, 121.2,
thieno[3,4-b][1,4]dioxepine-6-carbaldehyde (7). Compound 6 (0.85 g,
1.53 mmol, 1.0 eq.) was dissolved in THF (15 mL), then, after cooling to –
78 °C, n-BuLi (1.6 M solution in pentane, 1.40 mL, 2.25 mmol, 1.5 eq) was
added in the reaction vessel. After one hour under stirring at – 78 °C, N,N-
DMF (0.22 g, 3.00 mmol, 0.23 mL, 2.0 eq.) was added, then the reaction
mixture was slowly warmed to room temperature and stirred for 16 h. After
this time, a saturated aqueous solution of NH4Cl (12 mL) was added, then
THF was evaporated under vacuum and the aqueous phase was extracted
with ethyl acetate (3  30 mL). The combined organic phases were washed
with brine (2  70 mL) and dried with Na2SO4. After filtration and removal
of the solvent under vacuum, a dense orange oil corresponding to a 10:1
mixture of aldehyde 7 and an unidentified side product (0.84 g, 90% approx.
yield of aldehyde 7) was obtained and used as such for the following
reaction. Characterization data for aldehyde 7: 1H–NMR (300 MHz,
CDCl3): δ = 9.88 (s, 1H), 6.95–7.37 (m, 16H), 4.07 (s, 2H), 3.98 (s, 2H),
1.25–1.42 (m, 16H), 0.91 (t, J = 6.8 Hz, 6H) ppm. 13C–NMR (75 MHz,
CDCl3): δ = 180.6, 156.7, 148.3, 147.4, 145.8, 132.8, 130.7, 130.1, 129.5,
128.0, 125.0, 123.6, 123.0, 118.2, 116.0, 78.2, 78.0, 43.9, 32.7 (×2), 32.2
103.2, 77.2, 76.7, 56.7, 43.9, 32.7, 31.8, 22.6, 22.5, 14.1 ppm. IR (KBr): 휈
̃
= 3410, 3112, 2930, 2861, 1459, 1412, 1376, 1206, 1164, 1016 cm–1. ESI–
MS: m/z = 349 [M+Na]+, 309 [M–OH]+.
(×2), 22.7 (×4), 14.2 (×2) ppm. IR (KBr): 휈̃ = 2924, 1642, 1589, 1485, 1424,
1278, 1049 cm–1. ESI–MS: m/z = 608 [M+Me]+ (peak derived from the
formation of the emiacetal of aldehyde 7 with MeOH).
Diethyl
((3,3-dipentyl-3,4-dihydro-2H-thieno[3,4-b][1,4]dioxepin-6-
yl)methyl)phosphonate (4). Alcohol 2 (1.39 g, 4.25 mmol, 1.0 eq.) was
dissolved in THF (30 mL), then the solution was cooled to –5 °C and PBr3
(1.17 g, 4.33 mmol, 0.41 mL, 3.0 eq.) was added into the reaction vessel.
The resulting dark mixture was stirred at 0 °C for 30 minutes, then triethyl
phosphite (2.12 g, 12.8 mmol, 2.17 mL, 3.0 eq.) was added dropwise at
room temperature. The resulting mixture was stirred at reflux for 7 h and
then at 60 °C for 10 h. After cooling to room temperature, the solvent was
removed under vacuum. Purification by flash column chromatography
(SiO2, petroleum ether/AcOEt from 3:1 to 1:1) afforded desired product 4
(1.56 g, 3.49 mmol, 82% yield) as a yellow oil. 1H–NMR (300 MHz, CDCl3):
δ = 6.31 (d, J = 2.7 Hz, 1H), 4.03–4.13 (m, 4H), 3.82 (s, 2H), 3.80 (s, 2H),
3.20 (d, J = 20.3, 2H), 1.33–1.25 (m, 22H), 0.88 (t, J = 7.0 Hz, 6H) ppm.
13C–NMR (75 MHz, CDCl3): δ = 149.3 (d, J = 3.6 Hz), 147.4 (d, J = 9.7 Hz),
111.5 (d, J = 11.3 Hz), 102.5 (d, J = 4.2 Hz), 77.9, 77.8, 62.4 (d, J = 6.5
Hz), 44.0, 32.8 (2), 31.7 (2), 24.6 (d, J = 144.3 Hz), 22.7 (2), 22.6 (2),
16.5 (d, J = 6.0 Hz), 14.2 (2) ppm. 31P–NMR (121 MHz, CDCl3): δ = 24.5
(E)-2-cyano-3-(8-((E)-4-(diphenylamino)styryl)-3,3-dipentyl-3,4-
dihydro-2H-thieno[3,4-b][1,4]dioxepin-6-yl)acrylic
acid
(MB25).
Aldehyde 7 (0.48 g, 0.82 mmol, 1.0 eq.), as obtained in the previous step,
was dissolved in toluene (45 mL), then cyanoacetic acid (0.69 g, 8.15
mmol, 10 eq.), ammonium acetate (0.25 g, 3.28 mmol, 4.0 eq.) and acetic
acid (65 mL) were added to the reaction vessel. The reaction mixture was
heated under stirring at 120° C for 4 h, then cooled to room temperature
and diluted with AcOEt (200 mL). The organic phase was washed with aq.
NH4Cl 1M (230 mL + 4  70 mL) and brine (150 mL), then dried with
Na2SO4. After filtration and removal of the solvent under vacuum, the
reaction crude was purified by flash column chromatography (SiO2, CH2Cl2,
then CH2Cl2/AcOH 98:2). Desired product MB25 (0.35 g, 0.53 mmol, 65%
yield) was isolated as dark red solid. 1H–NMR (300 MHz, CDCl3): δ = 8.41
(s, 1H), 7.33–7.40 (m, 2H), 7.24–7.32 (m, 4H), 7.04–7.16 (m, 8H), 6.97–
7.04 (m, 2H), 4.10 (s, 2H), 3.98 (s, 2H), 1.31–1.39 (m, 16H), 0.91 (t, J =
6.9 Hz, 6H) ppm. 13C–NMR (75 MHz, CDCl3): δ = 169.1, 156.8, 148.7,
147.2, 145.7, 143.5, 135.3, 132.5, 129.7, 129.5, 128.3, 125.2, 123.8, 122.5,
116.8, 115.3, 113.2, 91.5, 78.3, 77.4, 43.7, 32.7 (×2), 32.2 (×2), 22.7 (×4),
(s) ppm. IR (KBr): 휈̃ = 3112, 2930, 1724, 1648, 1497, 1377, 1258, 1029
cm–1. ESI–MS: m/z = 915 [2M+Na]+, 447 [M+H]+.
(E)-4-(2-(3,3-dipentyl-3,4-dihydro-2H-thieno[3,4-b][1,4]dioxepin-6-
yl)vinyl)-N,N-diphenylaniline (6). 4-(diphenylamino)-benzaldehyde (5,
53 mg, 0.19 mmol, 1.0 eq.) was dissolved in THF (0.3 mL), then a solution
of phosphonate 4 (100 mg, 0.22 mmol, 1.2 eq.) in THF (1.0 mL) was added
to the reaction vessel. After cooling to 0 °C, NaHMDS (70 mg, 0.38 mmol,
2.0 eq.) was added. The reaction mixture was stirred at 0 °C for 15 minutes,
then at room temperature for other 15 minutes and, finally, at reflux
temperature for 16 h. After cooling to room temperature, water (4 mL) was
added and the resulting mixture was concentrated under vacuum. After
dilution with ethyl acetate (5 mL), the two phases were separated. The
14.2 (×2) ppm. IR (KBr): 휈̃ = 3423, 2930, 2857, 2213, 1681, 1564, 1480,
1242, 1049 cm–1. ESI-MS: m/z = 661 [M+1]+.
8-(5-(4-(diphenylamino)phenyl)thiophen-2-yl)-3,3-dipentyl-3,4-
dihydro-2H-thieno[3,4-b][1,4]dioxepine-6-carbaldehyde (11). 3,3-
dipentyl-3,4-dihydro-2H-thieno[3,4-b][1,4]dioxepine-6-carbaldehyde
(1,
0.35 g, 1.08 mmol, 1.0 eq) was dissolved in chloroform (8 mL) and acetic
acid (4 mL), then N-bromosuccinimide (0.21 g, 1.19 mmol, 1.1 eq.) was
added to the reaction vessel. The reaction mixture was stirred at room
This article is protected by copyright. All rights reserved.

10.1002/cssc.201701707
ChemSusChem
FULL PAPER
temperature for 1.5 h, then, after dilution with chloroform (30 mL), the
organic phase was washed with a saturated aqueous solution of NaHCO3
(40 mL) and brine (40 mL) and dried with Na2SO4. After filtration, the
solvent was removed under vacuum. The reaction crude was dissolved in
N,N-DMF (10 mL), then N,N-diphenyl-4-(5-(tributylstannyl)thiophen-2-
yl)aniline (9, 0.727 g, 1.18 mmol, 1.1 eq.) and Pd(PPh3)2Cl2 (0.038 g, 0.054
mmol, 5.0 mol%) were added to the reaction vessel. The mixture was
stirred at 90 °C for 2 h, then, after cooling, to room temperature, filtered
over Celite®. The filtrate was diluted with ethyl acetate (50 mL) and the
organic phase was washed with water (4  50 mL) and brine (2  50 mL).
After dehydration with Na2SO4, filtration and removal of the solvent under
vacuum, flash column chromatography (SiO2, petroleum ether/ CH2Cl2
gradient from 3:1 to 1:1) afforded desired product 11 (0.47 g, 0.72 mmol,
67% yield) as a dark red solid. 1H–NMR (400 MHz, CDCl3): δ = 9.90 (s,
1H), 7.48 (d, J = 8.6 Hz, 2H), 7.36 (d, J = 3.9 Hz, 1H), 7.24–7.30 (m, 4H),
7.16 (d, J = 3.9 Hz, 1H), 7.12 (d, J = 7.7 Hz, 4H), 7.02–7.08 (m, 4H), 4.11
(s, 2H), 4.06 (s, 2H), 1.42–1.49 (m, 4H), 1.25–1.38 (m, 12H), 0.92 (t, J =
6.1 Hz, 6H). 13C–NMR (100 MHz, CDCl3): δ = 180.5, 156.7, 147.9, 147.4,
146.2, 144.1, 132.2, 129.5, 127.7, 126.9, 126.7, 124.8, 123.4, 122.4, 117.7,
78.3, 78.2, 44.0, 32.7, 32.2, 22.7, 14.2. (Two carbon signals are missing,
likely covered by other signals). ESI-MS: m/z = 649 [M]+; 695 [M+,
dimethylacetal of aldehyde 11]+. IR (KBr): 3070, 2956, 2928, 2857, 1647,
power of 33 W, peak power of 947 W, pulse energy of 0.83 pJ, spectral
width of 50 nm (FWHM) and time pulse duration of 947 ps (FWHM). The
acquisition system allowed to acquire 4096 time channels with a minimum
time resolution of 34 ps. The sample was directly illuminated by the
collimated LED light (excitation light). The fluorescence signal was
collected at an angle of 90 degrees with respect to the excitation direction
by means of a 200 m core diameter multimode fiber coupled with a 1 mm
diameter GRIN lens, so that only collimated photons could be collected.
The other end of the fiber was coupled with a 1 mm GRIN lens and then
faced to the PM detector. The instrument response function (IRF) was
collected using an aqueous suspension of polystyrene latex beads. Data
were analyzed by means of Picoquant FluoFit 4.4 fluorescence decay data
analysis software.
Preparation of Pt/TiO2 nanopowder. Pt was photodeposited on TiO2
Degussa P25 following a previously reported procedure.[18a,60] Briefly, 2 g
of TiO2 Degussa P25 were suspended in 400 mL Pt(NO3)2 aqueous
solution (EtOH 50 % v/v), in order to reach a final Pt loading of 1.0 wt%.
After stirring for 1 h in the dark, the suspension was irradiated with a 450
W medium pressure Hg lamp for 4 h. Nanopowders were recovered
through centrifugation, washed with EtOH 3 times, and dried under
vacuum at 50 °C overnight.
1593, 1485, 1426, 1281, 1051 cm–1
.
Dyes adsorption on Pt/TiO2. 200 mg of Pt/TiO2 nanopowder was
suspended in 20 mL of dye solution (0.1 mM in ethanol) for 24 h in the
dark. Then, the nanopowder was separated through centrifugation,
washed twice with ethanol, and dried under vacuum at room temperature
overnight. After staining, the concentration of the dyes in the solution was
measured by UV-vis spectroscopy. More than 95% of the dye was
adsorbed on the Pt/TiO2 material in all cases.
(E)-2-cyano-3-(8-(5-(4-(diphenylamino)phenyl)thiophen-2-yl)-3,3-
dipentyl-3,4-dihydro-2H-thieno[3,4-b][1,4]dioxepin-6-yl)acrylic acid
(AD418). Aldehyde 11 (0.075 g, 0.115 mmol, 1.0 eq) was dissolved in
toluene (2 mL), then a solution of cyanoacetic acid (0.098 g, 1.15 mmol,
10 eq.) and ammonium acetate (0.035 g, 0.46 mmol, 4.0 eq.) in acetic acid
(2 mL) was added in the reaction vessel. The mixture was stirred at 100 °C
for 4 h, then, after cooling to room temperature, the solvent was removed
and replaced with DCM (30 mL). The organic phase was washed with aq.
HCl 0.03 M (30 mL) and brine (30 mL), and dried with Na2SO4. After
filtration and removal of the solvent, washings of the crude with small
portions of pentane and methanol afforded the desired product AD418
(0.076 g, 0.106 mmol, 92% yield) as a dark purple solid. 1H–NMR (400
MHz, THF-d8): δ = 8.33 (s, 1H), 7.55 (d, J = 8.2 Hz, 2H), 7.48 (d, J = 3.3
Hz, 1H), 7.32 (d, J = 3.5 Hz, 1H), 7.24–7.28 (m, 4H), 7.09 (d, J = 7.8 Hz,
4H), 7.00–7.06 (m, 4H), 4.17 (s, 2H), 4.12 (s, 2H), 1.46–1.52 (m, 4H),
1.25–1.42 (m, 12H), 0.92 (t, J = 5.9 Hz, 6H). 13C–NMR (100 MHz, THF-
d8): δ = 164.4, 156.3, 148.7, 148.2, 147.1, 145.0, 141.3, 132.7, 130.0,
128.6, 127.8, 127.2, 127.0, 125.3, 124.1, 123.9, 123.4, 116.9, 113.2, 95.8,
79.0, 78.8, 44.4, 33.4, 32.7, 23.23, 23.17, 14.2. ESI-MS: m/z = 716.48 [M]+.
IR (KBr): 3447, 3061, 2928, 2860, 2214, 1654, 1576, 1542, 1419, 1251
cm–1. HRMS (ESI) for C43H44O4N2S2 [M+]: calcd 716.2737, found 716.2741.
Hydrogen production through photoreforming. The dye-sensitized
Pt/TiO2 nanomaterials have been tested for H2 production using TEOA or
EtOH as sacrificial electron donors, following or slightly modifying a
previously described procedure.[36] 60 mg of the dye-sensitized Pt/TiO2
catalyst was suspended into 60 mL of either 10 % v/v aqueous solution of
TEOA previously neutralized with HCl, or 20 % v/v aqueous solution of
EtOH. After purging with Ar (15 mL min−1) for 30 min, the suspension was
irradiated using a 150 W Xe lamp with a cut-off filter at 420 nm. Irradiance
was ~ 6 x 10–3 W m–2 in the UV-A range and ~ 1080 W m–2 in the visible
and near-IR range (400 – 1000 nm). The concentration of H2 in gas stream
coming from the reactor has been quantified using a Agilent 7890
gaschromatograph equipped with
a TCD detector, connected to a
Carboxen 1010 column (Supelco, 30 m x 0.53 mm ID, 30 μm film) using
Ar as carrier. Blank experiments on Pt/TiO2 in the absence of dyes showed
no H2 evolution under any of the experimental conditions used in this work.
The performances of the sensitized photocatalysts have been reported in
terms of H2 production rate and overall H2 productivity. Turn-Over
Numbers (TON) were calculated from the total amount of H2 produced in
20h of irradiation as:
Optical and electrochemical measurements. UV-Vis absorption spectra
were recorded on diluted solutions of the compounds (approx. 10–5 M in
the solvents specified in the text). DRS spectra were recorded on
sensitized Pt/TiO2 catalyst powders in the reflectance mode using an
integrating sphere (with Ba2SO4 as a reference standard), and were
converted to the corresponding absorption spectra using the Kubelka-
Munk equation. Cyclic voltammetry experiments were conducted in
dichloromethane solution employing a three-electrode cell having a glassy
carbon working electrode, a platinum counter-electrode and an aqueous
Ag/AgCl (sat. KCl) reference electrode. The supporting electrolyte was
electrochemical-grade [N(Bu)4]PF6. Under these experimental conditions,
the one-electron oxidation of ferrocene occurs at E°′ = 0.62 V.
The technique used for the measurement of the fluorophores lifetime is the
Time Correlated Single Photon Counting (TCSPC).[59] The apparatus
consisted of a PDL 808 “Sepia” laser driver system, a TimeHarp 200 PCI
acquisition board, a PMA182-P-M photomultiplier (PM) detector and a PLS
500 collimated LED source, all from PicoQuant (Germany). The chosen
LED source had peak emission wavelength at 499 nm, average maximum
2푥 표푣푒푟푎푙푙 퐻2 푎푚표푢푛푡 (휇푚표푙 ꢁ−1
)
푇푂푁 =
2푥 푑푦푒 푙표푎푑ꢀ푛ꢁ (휇푚표푙 ꢁ−1
)
Light-to-Fuel Efficiency (LFE) was calculated as:
퐹퐻 ∆퐻0퐻
2
2
퐿퐹퐸 =
푆 퐴ꢂ푟푟
where FH2 is the flow of H2 produced (expressed in mol s–1), ΔH0H2 is the
enthalpy associated with H2 combustion (285.8 kJ mol–1), S is the total
incident light irradiance, as measured by adequate radiometers in 400 –
1000 nm range (expressed in W cm–2) and Airr is the irradiated area
This article is protected by copyright. All rights reserved.

10.1002/cssc.201701707
ChemSusChem
FULL PAPER
(expressed in cm2). UV-vis spectra of the aqueous solutions recovered at
the end of the photocatalytic runs highlighted that no desorption of the dyes
took place during the experiments.
[22] A. Kumari, I. Mondal, U. Pal, New J. Chem., 2015, 39, 713-720.
[23] K. Narayanaswamy, A. Tiwari, I. Mondal, U. Pal, S. Niveditha, K.
Bhanuprakash, S. P. Singh, Phys. Chem. Chem. Phys. 2015, 17, 13710-
13718.
[24] J. Lee, J. Kwak, K. C. Ko, J. H. Park, J. H. Ko, N. Park, E. Kim, D. H. Ryu,
T. K. Ahn, J. Y. Lee, S. U. Son, Chem. Commun. 2012, 48, 11431-11433.
[25] A. Tiwari, I. Mondal, U. Pal, RSC Adv. 2015, 5, 31415-31421.
[26] (a) D. P. Hagberg, T. Edvinsson, T. Marinado, G. Boschloo, A. Hagfeldt,
L. Sun, Chem. Commun. 2006, 2245-2247; (b) D. P. Hagberg, T.
Marinado, K. M. Karlsson, K. Nonomura, P. Qin, G. Boschloo, T. Brinck,
A. Hagfeldt, L. Sun, J. Org. Chem. 2007, 72, 9550-9556.
[27] D. Franchi, M. Calamante, G. Reginato, L. Zani, M. Peruzzini, M. Taddei,
F. Fabrizi de Biani, R. Basosi, A. Sinicropi, D. Colonna, A. Di Carlo, A.
Mordini, Tetrahedron 2014, 70, 6285-6295.
Acknowledgements
M.M., T.M., and P.F. acknowledge financial support from
University of Trieste through the FRA2015 project. All the other
authors thank the Cassa di Risparmio di Firenze Foundation for
funding through the ENERGYLAB project.
Keywords: Photocatalytic hydrogen production •
Photosensitizers • Donor-acceptor dyes • Heterocyclic chemistry
• Ethanol photoreforming.
[28] A. Dessì, M. Calamante, A. Mordini, M. Peruzzini, A. Sinicropi, R. Basosi,
F. Fabrizi de Biani, M. Taddei, D. Colonna, A. Di Carlo, G. Reginato, L.
Zani, Chem. Commun. 2014, 50, 13952-13955.
[29] A. Dessì, M. Calamante, A. Mordini, M. Peruzzini, A. Sinicropi, R. Basosi,
F. Fabrizi de Biani, M. Taddei, D. Colonna, A. di Carlo, G. Reginato, L.
Zani, RSC Adv. 2015, 5, 32657-32668.
[1]
R. C. Armstrong, C. Wolfram, K. P. de Jong, R. Gross, N. S. Lewis, B.
Boardman, A. J. Ragauskas, K. Ehrhardt-Martinez, G. Crabtree, M. V.
Ramana, Nat. Energy 2016, 1, 15020.
[30] B. C. Thompson, Y.-G. Kim, T. D. McCarley, J. R. Reynolds, J. Am.
Chem. Soc. 2006, 128, 12714-12725.
[2]
[3]
[4]
[5]
[6]
[7]
J. Schneider, M. Matsuoka, M. Takeuchi, J. Zhang, Y. Horiuchi, M. Anpo,
D. W. Bahnemann, Chem. Rev. 2014, 114, 9919-9986.
X. Fu, J. Long, X. Wang, D. Y. C. Leung, Z. Ding, L. Wu, Z. Zhang, Z. Li,
X. Fu, Int. J. Hydrogen En. 2008, 33, 6484-6491.
Q. Gu, J. Long, Y. Zhou, R. Yuan, H. Lin, X. Wang, J. Catal. 2012, 289,
88-99.
[31] A. Leliège, J. Grolleau, M. Allain, P. Blanchard, D. Demeter, T. Rousseau,
J. Roncali, Chem. Eur. J. 2013, 19, 9948–9960.
[32] J. Tauc, Mater. Res. Bull. 1968, 3, 37-46.
[33] C. Coluccini, N. Manfredi, M. M. Salamone, R. Ruffo, M. Grazia Lobello,
F. De Angelis, A. Abbotto, J. Org. Chem. 2012, 77, 7945-7956.
[34] J. O. M. Bockris, S. U. M. Khan, Surface Electrochemistry - A Molecular
Level Approach, Springer US, New York, 1993.
Q. Gu, J. Long, L. Fan, L. Chen, L. Zhao, H. Lin, X. Wang, J. Catal. 2013,
303, 141-155.
[35] H. Tian, X. Yang, J. Cong, R. Chen, C. Teng, J. Liu, Y. Hao, L. Wang, L.
Sun, Dyes Pigments 2010, 84, 62-68.
A. V. Puga, A. Forneli, H. García, A. Corma, Adv. Funct. Mater. 2014, 24,
241-248.
[36] B. Cecconi, N. Manfredi, R. Ruffo, T. Montini, I. Romero-Ocaña, P.
Fornasiero, A. Abbotto, ChemSusChem 2015, 8, 4216-4228.
[37] H. Kisch, D. Bahnemann, J. Phys. Chem. Lett. 2015, 6, 1907-1910
[38] S. Y. Reece, J. A. Hamel, K. Sung, T. D. Jarvi, A. J. Esswein, J. J. Pijpers,
D. G. Nocera, Science 2011, 334, 645-648.
J. Long, H. Chang, Q. Gu, J. Xu, L. Fan, S. Wang, Y. Zhou, W. Wei, L.
Huang, X. Wang, P. Liu, W. Huang, Energy Environ. Sci. 2014, 7, 973-
977.
[8]
[9]
L. Fan, J. Long, Q. Gu, H. Huang, H. Lin, X. Wang, J. Catal. 2014, 320,
147-159.
[39] G. Carraro, C. Maccato, A. Gasparotto, T. Montini, S. Turner, O. I.
Lebedev, V. Gombac, G. Adami, G. Van Tendeloo, D. Barreca, P.
Fornasiero, Adv. Funct. Mater. 2014, 24 372–378.
Q. Gu, J. Long, H. Zhuang, C. Zhang, Y. Zhou, X. Wang, Phys. Chem.
Chem. Phys. 2014, 16, 12521-12534.
[10] R. Jaswal, R. Shende, W. Nan, A. Shende, Int. J. Hydrogen. En. 2017,
42, 2839-2848.
[40] R. Abe, K. Hara, K. Sayama, K. Domen, H. Arakawa, J. Photochem.
Photobiol. A 2000, 137, 63-69.
[11] A. Hagfeldt, G. Boschloo, L. Sun, L. Kloo, H. Pettersson, Chem. Rev.
2010, 110, 6595-6663.
[41] Q. Li, Y. Che, H. Ji, C. Chen, H. Zhu, W. Ma, J. Zhao, Phys. Chem. Chem.
Phys. 2014, 16, 6550-6554.
[12] B. Cecconi, N. Manfredi, T. Montini, P. Fornasiero, A. Abbotto, Eur. J.
Org. Chem. 2016, 5194-5215.
[42] It should be noted that although compound AD418 is similar to the
photosensitizers already reported by Kang and co-workers (see ref. 10),
in those cases the effect of hydrophobic alkyl chains placed in the
intermediate section of the molecules was not investigated, and H2
production was performed under different conditions using another SED
(EDTA).
[13] X. Zhang, T. Peng, S. Song, J. Mater. Chem. A 2016, 4, 2365-2402.
[14]
K. C. Christoforidis, P. Fornasiero, ChemCatChem 2017, 9, 1523-1544.
[15] (a) X. Li, S. Cui, D. Wang, Y. Zhou, H. Zhou, Y. Hu, J.-g. Liu, Y. Long, W.
Wu, J. Hua, H. Tian, ChemSusChem 2014, 7, 2879-2888; (b) S.
Balasubramanian, P. Wang, R. D. Schaller, T. Rajh, E. A. Rozhkova,
Nano Lett. 2013, 13, 3365-3371.
[43] K. Kalyanasundaram, J. Kiwi, M. Grätzel, Helv. Chim. Acta 1978, 61,
2720-2730.
[16] S. Zinoviev, F. Müller-Langer, P. Das, N. Bertero, P. Fornasiero, M.
Kaltschmitt, G. Centi, S. Miertus, ChemSusChem. 2010, 3, 1106–1133.
[17] X. Zhang, B. Peng, S. Zhang, T. Peng, ACS Sust. Chem. Eng. 2015, 3,
1501–1509.
[44] A.-M. Manke, K. Geisel, A. Fetzer, P. Kurz, Phys. Chem. Chem. Phys.
2014, 16, 12029-12042.
[45] A. Listorti, B. O’Regan, J. R. Durrant, Chem. Mater. 2011, 23, 3381-3399.
[46] S. Kundu, A. B. Vidal, M. A. Nadeem, S. D. Senanayake, H. Idriss, P. Liu,
J. A. Rodriguez, D. Stacchiola, J. Phys. Chem. C 2013, 117, 11149-
11158.
[18] (a) S.-H. Lee, Y. Park, K.-R. Wee, H.-J. Son, D. W. Cho, C. Pac, W. Choi,
S. O. Kang, Org. Lett. 2010, 12, 460-463; (b) W.-S. Han, K.-R. Wee, H.-
Y. Kim, C. Pac, Y. Nabetani, D. Yamamoto, T. Shimada, H. Inoue, H.
Choi, K. Cho, S. O. Kang, Chem. Eur. J. 2012, 18, 15368-15381.
[19] N. Manfredi, B. Cecconi, V. Calabrese, A. Minotti, F. Peri, R. Ruffo, M.
Monai, I. Romero-Ocaña, T. Montini, P. Fornasiero, A. Abbotto, Chem.
Commun. 2016, 52, 6977-6980.
[47] S. Yang, H. Wang, H. Yu, S. Zhang, Y. Fang, S. Zhang, F. Peng, Int. J.
Hydrogen En. 2016, 41, 3446-3455.
[48] R. A. Rightmire, R. L. Rowland, D. L. Boos, D. L. Beals, J. Electrochem.
Soc. 1964, 111, 242–247
[49] S. C. S. Lai, M. T. M. Koper, Faraday Discuss. 2008, 140, 399-416.
[50] M. J. Giz, G. A. Camara, J. Electroanal. Chem. 2009, 625, 117-122.
[51] B. Pierozynski, Int. J. Electrochem. Sci., 2012, 7, 4261-4271.
[20] M. Watanabe, H. Hagiwara, A. Iribe, Y. Ogata, K. Shiomi, A. Staykov, S.
Ida, K. Tanaka, T. Ishihara, J. Mater. Chem. A 2014, 2, 12952-12961.
[21] A. Tiwari, U. Pal, Int. J. Hydrogen En. 2015, 40, 9069-9080.
This article is protected by copyright. All rights reserved.

10.1002/cssc.201701707
ChemSusChem
FULL PAPER
[52] D. Wang, J. Liu, Z. Wu, J. Zhang, Y. Su, Z. Liu, C. Xu, Int. J. Electrochem.
Sci. 2009, 4, 1672-1678.
[57] W. Jones, D. J. Martin, A. Caravaca, A. M. Beale, M. Bowker, T.
Maschmeyer, G. Hartley, A. Masters, Appl. Catal. B 2017, in press, DOI:
10.1016/j.apcatb.2017.01.042.
[53] A. von Weber, E. T. Baxter, S. Proch, M. D. Kane, M. Rosenfelder, H. S.
White, S. L. Anderson, Phys. Chem. Chem. Phys. 2015, 17, 17601-
17610.
[58] Y. Pellegrin, F. Odobel, C. R. Chim. 2017, 20, 283-295, and references
therein.
[54] U. Pal, S. Ghosh, D. Chatterjee, Transition Met. Chem. 2012, 37, 93-96.
[55] T.-N. Ye, M. Xu, W. Fu, Y.-Y. Cai, X. Wei, K.-X. Wang, Y.-N. Zhao, X.-H.
Li, J.-S. Chen, Chem. Commun. 2014, 50, 3021-3023.
[56] S. M. Unni, L. George, S. N. Bhange, R. N. Devi S. Kurungot, RSC Adv.
2016, 6, 82103-82111.
[59] D.V. O’Connor, D. Phillips, Time Correlated Single Photon Counting,
Academic Press, London, U.K., 1984.
[60] I. Romero Ocaña, A. Beltram, J. J. Delgado Jaén, G. Adami, T. Montini,
P. Fornasiero, Inorg. Chim. Acta 2015, 431, 197–205.
This article is protected by copyright. All rights reserved.

10.1002/cssc.201701707
ChemSusChem
FULL PAPER
Entry for the Table of Contents (Please choose one layout)
FULL PAPER
Photocatalytic H2 production mediated
by organic dye-sensitized Pt/TiO2
nanopowders was investigated using
TEOA and, for the first time, EtOH as
sacrificial electron donors (SEDs).
Results indicate that good compatibility
between individual dye properties and
the particular SED used in the
experiment is vital to maximize
catalytic efficiency.
A. Dessì, M. Monai, M. Bessi, T. Montini,
M. Calamante, A. Mordini, G. Reginato,
C. Trono, P. Fornasiero,* L. Zani*
Page No. – Page No.
Towards Sustainable H2 Production:
Rational Design of Hydrophobic
Triphenylamine-based Dyes for
Sensitized Ethanol Photoreforming
This article is protected by copyright. All rights reserved.