Sustainable photocatalytic synthesis of benzimidazoles

Article abstract of DOI:10.1016/j.ica.2021.120289

Among the 17 Sustainable Development Goals presented by the United Nations in 2015, great attention is devoted to the production of goods and chemicals by use of renewable raw materials, by recycling of products and by extensive use of renewable energy sources. In this context, photocatalysis attracted great attention for the possibility to exploit Solar light to promote the desired chemical reactions. Besides its use in degradation of pollutants and in the production of fuels, some efforts have been devoted in the development of photocatalytic processes for the synthesis of fine chemicals with high added-value. In this work, we investigated the sustainable photocatalytic synthesis of benzimidazole derivatives through a one-pot, tandem process starting from a nitro compound and ethanol. By a photocatalytic approach, ethanol is dehydrogenated producing the hydrogen required for reduction of nitro groups and the aldehyde required for cyclization and production of the benzimidazole unit. Co-doping of TiO₂ with B and N is beneficial to increase the photocatalytic activity in H₂ production from ethanol. The effect of various metal co-catalysts (Pt, Pd Ag, Cu) have been evaluated on H₂ production rate and on selectivity in the synthesis of substituted benzimidazoles: Pt showed the highest selectivity in the desired products while Pd demonstrated a great activity for hydrodehalogenation, with potential interest for degradation of persistent pollutants.

Full text of DOI:10.1016/j.ica.2021.120289

Inorganica Chimica Acta 520 (2021) 120289
Contents lists available at ScienceDirect
Inorganica Chimica Acta
journal homepage: www.elsevier.com/locate/ica
Research paper
Sustainable photocatalytic synthesis of benzimidazoles
,
b
,
,
Tiziano Montinia *, Valentina Gombaca b, Juan J. Delgadoc, Anna Maria Veneziad,
Gianpiero Adamia, Paolo Fornasieroa
,
b
a Department of Chemical and Pharmaceutical Sciences, INSTM Trieste Research Unit, University of Trieste, via L. Giorgieri 1, 34217 Trieste, Italy
b ICCOM-CNR Trieste, University of Trieste, via L. Giorgieri 1, 34217 Trieste, Italy
c
´
´
Departamento de Ciencia de los Materiales e Ingeniería Metalúrgica y Química Inorganica, Facultad de Ciencias, Universidad de Cadiz, Campus Río San Pedro, 11510,
´
Puerto Real, Cadiz, Spain
d ISMN-CNR, Via Ugo La Malfa 153, 90146 Palermo, Italy
A R T I C L E I N F O
A B S T R A C T
Among the 17 Sustainable Development Goals presented by the United Nations in 2015, great attention is
devoted to the production of goods and chemicals by use of renewable raw materials, by recycling of products
and by extensive use of renewable energy sources. In this context, photocatalysis attracted great attention for the
possibility to exploit Solar light to promote the desired chemical reactions. Besides its use in degradation of
pollutants and in the production of fuels, some efforts have been devoted in the development of photocatalytic
processes for the synthesis of fine chemicals with high added-value. In this work, we investigated the sustainable
photocatalytic synthesis of benzimidazole derivatives through a one-pot, tandem process starting from a nitro
compound and ethanol. By a photocatalytic approach, ethanol is dehydrogenated producing the hydrogen
required for reduction of nitro groups and the aldehyde required for cyclization and production of the benz-
imidazole unit. Co-doping of TiO2 with B and N is beneficial to increase the photocatalytic activity in H2 pro-
duction from ethanol. The effect of various metal co-catalysts (Pt, Pd Ag, Cu) have been evaluated on H2
production rate and on selectivity in the synthesis of substituted benzimidazoles: Pt showed the highest selec-
tivity in the desired products while Pd demonstrated a great activity for hydrodehalogenation, with potential
interest for degradation of persistent pollutants.
Dedicated to Dr. M. Peruzzini.
Keywords:
Photocatalysis
Benzimidazole
One-pot synthesis
Sustainability
Ethanol
1. Introduction
represents an intriguing task for research in chemical sciences. In this
context, photocatalysis is a key technology to promote chemical re-
The demand for energy, food and goods of our modern society has
been progressively increased in the last decades due to both the growth
of world population and the improvements in life quality. The 12th of
the 17 Sustainable Development Goals (SDG) adopted by the General
Assembly of United Nation with the 2030 Agenda for Sustainable
Development in January 2015 claims to “Ensure sustainable consump-
tion and production patterns” [1]. The actions within this goal intend to
favour sustainable exploiting of the natural resources, with extended
recycling of materials and with favouring the transition from the use of
fossil fuels to renewable raw materials.
actions using, in the best cases, solar light, a primary and renewable
energy source, which is potentially well spread and largely available to
everyone. Photocatalysis has been employed in the synthesis of organic
compounds [2], pharmaceuticals [3] or of natural products [4]. Notably,
in the majority of the studies, the photoactive component is a metal
complex (based on Ru, Ir, Cu etc.) or an organic molecule (i.e. Eosin Y)
[3,4]. These systems are usually working under visible light but suffer of
the consequent problem of degradation of the molecular photoactive
compound and of separation/recovery difficulties. Heterogeneous pho-
tocatalysts offers undoubtedly advantages of robustness and easy sepa-
ration/recovery. Beside their well-known application in pollutants
abatements from air or wastewaters, heterogeneous photocatalysis is
obtaining promising performances in the production of solar fuels by
valorisation of biomasses [5–8] and of liquid fuels from forestal/
The main beliefs of the 12th SDG and the principle of Green chem-
istry should be applied also in the production of fine chemicals with
high-added value. In particular, the search of innovative production
processes taking advantage of renewable reagents and energy sources
* Corresponding author at: Department of Chemical and Pharmaceutical Sciences, INSTM Trieste Research Unit, University of Trieste, via L. Giorgieri 1, 34217
Trieste, Italy.
E-mail address: tmontini@units.it (T. Montini).
https://doi.org/10.1016/j.ica.2021.120289
Received 30 November 2020; Received in revised form 26 January 2021; Accepted 3 February 2021
Available online 12 February 2021
0020-1693/© 2021 Elsevier B.V. All rights reserved.

T. Montini et al.
Inorganica Chimica Acta 520 (2021) 120289
agricultural wastes [9]. In the chemical synthesis, in which high selec-
tivity to the organic product is required, heterogeneous photocatalysts
have been applied in reactions such as oxidation of benzylic alcohols
[10–15], reduction of nitro compounds [14,16,17], functionalization of
photocatalysts have been studied for the one-pot synthesis of 2-MBI. The
experimental conditions (nitro compounds used as precursor, solvent)
have been optimized using 2-nitroaniline. Finally, the effect of the metal
co-catalyst in the product selectivity has been evaluated for the synthesis
of substituted 2-MBI, in order to obtain products that can be further
functionalised on the aromatic ring and, therefore, interesting for syn-
thesis of more complex molecules.
–
C
H bonds [15,18], Suzuki coupling [19,20] and others [2,21]. The
weakness of this approach is its low selectivity. In fact, high selectivity to
the desired product is typically obtained only when the molecular
structure of the substrate is well defined.
The sustainability of chemical production will be further improved if
heterogeneous photocatalysts will be applied in tandem processes, in
which the same catalyst is employed to promote cascade reactions using
solar light as primary energy source without the need to isolate the in-
termediate reaction products. A relevant example of this situation is the
synthesis of benzimidazolic products from nitro compound. Benzimid-
azole and its derivative are fundamental building blocks in the synthesis
of natural products and pharmaceuticals [22], showing biological ac-
tivity mainly as bactericides [23], anticarcinogens [24] and peptic
ulcera agents [25]. Moreover, benzimidazoles are used as ligands in
metal complexes [26–28], that are employed in catalysts and sensors.
The synthesis of benzimidazoles is classically performed by coupling
1,2-arylenediammine with aldehydes, carboxylic acids or their de-
rivatives [29–33], requiring medium – high temperatures (100–200 ◦C)
and often strong oxidants. To overcome these harsh conditions, Shiraishi
et al. reported an one-pot method in which 1,2-phenylenediammines are
coupled with aldehydes produced by photocatalytic dehydrogenation of
alcohols using Pt/TiO2 under UV–vis irradiation (λ > 300 nm, Xe lamp,
2 kW) [34]. The photocatalyst is involved in both the dehydrogenation
of alcohol to produce the required aldehyde and in the dehydrogenation
of the intermediates, necessary to obtain the final benzimidazole prod-
ucts. Notably, the most important by-products reported in this paper are
N-substituted-2-alkyl-1H-benzimidazoles. Although this method allows
the preparation of 2-substitured 1H-benzimidazoles under very mild
conditions, it does not overcome the important safety problem related
with the employment of highly toxic compounds, such as 1,2-arylene-
diammine. In a subsequent study, Selvam and Swaminathan [35] re-
ported the first example of an one-pot synthesis of benzimidazoles from
nitrocompounds in alcoholic solvent, using the dehydrogenation of the
alcohol to reduce the –NO2 to –NH2 and produce the aldehyde required
for the synthesis of the cyclic compound. Anyway, this work shows
important limitations. First, the photocatalyst employed (Pt/TiO2) is
irradiated with a monochromatic UV light at 365 nm. Second, the
nitrocompounds employed are N-substituted 2-nitro arenes, in which
the amino group is already bearing a substituent. This was done in order
to reduce the number of possible by-products obtained in the reaction.
Despite these limitations, the work by Selvan and Swaminathan [35]
represents the elegant attempt to combine more reaction steps into a
single one-pot photocatalytic process.
2. Experimental section
2.1. Catalyst preparation
TiO2 and TiO2-B,N supports were prepared following a sol–gel
approach. Briefly, two solutions were rapidly mixed: the first solution
was made by tert-butyl orthotitanate dissolved in absolute ethanol (1.7
M) while the second contains H2O and nitric acid (as catalyst). In the
case of TiO2-B,N, H3BO3 (in 9:100 M ratio with respect to Ti) and urea
(in 5:100 M ratio with respect to Ti) were added simultaneously in the
second (aqueous) solution. After dropping the ethanolic solution con-
taining the Ti precursor into the aqueous solution under gentle stirring, a
gel was obtained in a few minutes. The gel was aged for 24 h at room
temperature, dried at 120 ◦C for 12 h and finally calcined in a static oven
at 450 ◦C for 6 h.
Metal co-catalyst (Pd, Pt, Ag, or Cu) was deposited on the surface of
the supports by a simple photodeposition technique starting from a
water/methanol solution of correspondent nitrate salt. Briefly, 500 mg
of the support was suspended into 120 mL of water/methanol solution
containing an adequate amount of the metal precursor in order to reach
the desired metal loading (0.5 wt% for Pd, Pt, and Ag and 1.0 wt% for
Cu). After equilibration in the dark for 30 min, the suspension was
irradiated with UV–vis light (125 W Hg medium pressure lamp, Helios)
for 2 h. Finally, the solid was recovered by filtration and dried in an oven
at 80 ◦C for 12 h, without any further treatment.
2.2. Sample characterization
Textural properties of the samples were characterized by N2 phys-
isorption at the liquid nitrogen temperature using a Micromeritics ASAP
2020 porosimeter. Before the analysis, the samples have been degassed
at 120 ◦C overnight.
Powder X-ray diffraction patterns were collected using a Philips
X’Pert diffractometer, using Cu Kα radiation. Cu Kα2 was removed by
computer-processing of the data. Phase analysis was performed using
the Rietveld method through the PowderCell 2.0 program while mean
crystallite size was calculated applying the Scherrer’s equation to the
main reflection of each phase.
UV–vis spectra of solutions were collected using a Shimadzu Shi-
madzu UV-2450, acquiring absorbance in the range 300 – 800 nm.
ICP-OES analysis has been performed using an Optima 8000 Spec-
trometer (PerkinElmer) equipped with a S10 autosampler. The total
copper concentration was evaluated using calibration curves obtained
by dilution of SpectrascanTM copper standard solutions for ICP-OES
analysis (Teknolab A/S, Norway). All standards (range: 0.1–50.0 mg/
L) were prepared using a EtOH 96% solution to compensate the matrix
effect. The used emission wavelength was 324.754 nm, the limit of
detection 0.03 mg/L and the repeatability of measurements expressed as
relative standard deviations (RSD%) and calculated on 6 replicates of
various samples was lower than 4%. Calibration curves obtained by
means of 5 standard solutions had correlation coefficients higher than
0.998.
A key step in the tandem synthesis of benzimidazoles is the efficient
hydrogen abstraction from alcohols. Dehydrogenation of alcohols to
produce H2 is a well know process [5,7,36] and, in our research group,
great efforts have been dedicated to the development of TiO2-based
photocatalysts able to harvest solar light extracting H2 from renewable
raw materials like ethanol [37–45] and glycerol [38,39,42–44,46,47].
Activity of TiO2 materials can be further improved by doping, in order to
modify the electronic properties of the semiconductor. Doping results
usually in increasing on solar light harvesting and/or in increasing the
lifetime of photogenerated e-/h + couples [49].
In this work, we investigated the tandem, one-pot synthesis of 1H-
benzimidazoles starting from nitro compounds (1,2-dinitrobenzene or
substituted 2-nitroaniline) by a photocatalytic approach, in which
ethanol acts as source for the hydrogen, required in the reduction of
nitro groups, and for the aldehyde, required in the cyclization process.
The main product of this process is a 2-methyl-1H-benzimidazole (2-
MBI). In this research work, we investigated the photocatalytic perfor-
mances in ethanol dehydrogenation of TiO2 co-doped with B and N and
of the nature of various metal co-catalysts. The most promising
The samples were characterized by means of High Resolution Elec-
tron Microscopy (HREM) on a JEOL2010 field emission gun microscope
operated at 200 kV and with a spatial resolution of 0.19 nm at Scherzer
defocus conditions. In order to obtain the particle size distribution and
chemical information of the sample, High Angle Annular Dark Field-
Scanning Transmission Electron Microscopy (HAADF-STEM)
2

T. Montini et al.
Inorganica Chimica Acta 520 (2021) 120289
technique, in combination with X-Ray Energy Dispersive Spectroscopy
(X-EDS, Oxford Instrument), was used. These experiments were carried
out at the same microscope using an electron probe of 0.5 nm of
diameter and a diffraction camera length of 10 cm. The sample was
deposited on a Cu grid covered with a thin layer of holey carbon (Ni
grind in the case of Cu/TiO2-B,N).
were performed in the same conditions in ethanol 96% without sam-
pling. At the end of the synthesis, the catalyst was removed by filtration
and solvent has been evaporated under vacuum. The crude mixture
containing the benzimidazole products were dissolved in 15 mL of HCl
5% and extracted with CH2Cl2 (3 × 10 mL) to remove organic con-
taminants. Then, the aqueous solution was neutralized with NaOH and
extracted with CH2Cl2 (3 × 10 mL) to extract the benzimidazole prod-
ucts. After drying on Na2SO4 and evaporation to dryness under vacuum,
the product was dissolved in 2 mL of methanol for GC/MS analysis.
The X-ray photoelectron spectroscopy (XPS) analyses of the powder
samples were performed with a VG Microtech ESCA 3000 Multilab,
using Al Kα radiation (1486.6 eV) from a dual Mg/Al anode. The con-
stant charging of the samples was corrected by referencing all the
binding energies to the C 1 s peak energy set at 285.1 eV, arising fom
adventitious carbon. Qualitative and quantitative analyses of the peaks
were performed with the CasaXPS software.
3. Results and discussion
3.1. Characterization of the catalysts
UV–vis diffuse reflectance spectra were measured on a Varian Cary
4000 UV–vis spectrophotometer equipped with an integrating sphere.
Spectra were collected in the range of 200 – 800 nm. MgO was used as a
white standard material. Band gap energies were determined by con-
structing Tauc plots for an indirect semiconductor from the calculated
Kubelka-Munk functions of the respective diffuse reflectance spectra.
Produced H2 has been quantified using a gas-chromatograph Agilent
7890 mounting a TCD detector connected with a Supelco Carboxen 1010
column (30 m, 0.52 mm ID, 0.5 µm film) with Ar as gas carrier. Volatile
products (by-products from ethanol dehydrogenation and various
benzimidazoles) were analysed GC/MS Agilent 7890 GC equipped with
an Agilent 5975C mass spectrometer and mounting a J&D DB-225 m
column (60 m, 0.25 mm ID, 0.25 µm film) and using He as carrier.
After preparation, the TiO2 based materials have been characterized
to gain information on phase composition and textural properties.
Powder XRD analysis (Fig. 1) showed that the obtained materials are
multiphasic, comprising a mixture of anatase, brookite and rutile. The
phase composition has been determined by Rietveld analysis of the XRD
patterns presented in Fig. 1 and the quantitative results (summarized in
Table 1) show that the relative amounts of the three phases are not
affected by the introduction of B and N as co-dopants for TiO2. The cell
parameters of the 3 phases are only marginally affected by the intro-
duction of the co-dopants (see Table S1), in agreement with the low
loading of B and N in the TiO2-B,N material. On the other hand, a sig-
nificant reduction of average crystallite size has been observed for all the
phases, in particular for the rutile polymorph. This effect has been
already observed for TiO2 material doped with non-metal elements
[50,51].
2.3. Photocatalytic tests
As a result of the reduction of crystallite size, the textural properties
of the materials are strongly affected by the addition of dopants. Both
the materials present a Type IV isotherm (Fig. S1a), typical of meso-
porous material [52]. The large hysteresis of Type H2 (Fig. S1a) is
indicative of the presence of bottled-neck pores [52]. Despite these
general similarities, the specific surface area and the cumulative pore
volume of TiO2-B,N are much larger than that of the corresponding
undoped TiO2 (Table 1 and Fig. S1b). At the same time, the maximum of
the pore size distributions (from the BJH analysis of the desorption
branch of the isotherms, Fig. S1c) is shifted to slightly smaller diameter
in the case of the doped material. All the modification in textural
properties are clearly related with the different aggregation of crystal-
lites depending on their sizes.
Photocatalytic activity experiments were performed using a Solar
Simulator (LOT-Oriel) equipped with a 150 W Xe lamp and an Atmo-
spheric filter (cut-off at 300 nm), in order to simulate the radiation that
reaches the Earth surface. The radiation intensity reaching the reactor,
as measured with a radiometer (Delta Ohm HD 2302.0 LightMeter), was
25 mW/cm2 in the UV-A range (315–400 nm) and 180 mW/cm2 in the
vis-NIR range (400–1080 nm). This situation corresponds to a powder
density around 2 Suns, typical of a small solar concentrator.
A semi-batch reactor, equipped with a glass window, was used for
photocatalytic experiments. In H2 production studies, 150 mg of the
photocatalysts were suspended into 60 mL of ethanol 96% (commercial
azeotropic mixture) and the reactor temperature was controlled at 30 ◦C
using a thermostatic water bath. After purging the reactor with Ar (15
mL/min) for 30 min, the lamp was switched on and on-line analysis of
the H2 produced and transferred by the Ar flow (15 mL/min) to the GC.
Apparent quantum yields (AQY) were measured using the same
apparatus. After 3 h for activation of the photocatalysts and to reach
steady-state conditions, a 10 nm bandpass filter centred at 350 nm was
inserted obtaining an irradiation power of 2.78 mW/cm2. H2 production
was monitored for at least other 12 h, in order to reach a stable value.
Considering that 2 electrons are required to produce one H2 molecule,
AQY was calculated following the formula:
These important differences in structural and textural properties of
TiO2 and TiO2-B,N could be easily related with the presence of the
dopants. The detection of B and N into TiO2 matrix is not a simple task,
considering that both are light elements. Moreover, it is know that
calcination treatments in air are able to remove the most part of them
[51]. In our case, XPS has been employed to reveal the presence of B and
N in the TiO2-B,N material (spectra reported in Fig. S2). Despite the fact
that XPS is a surface-sensitive technique, the results clearly reveal that B
is actually present in the sample: the position of B 1 s signal (192.70 eV)
is typical of species of B interacting with O, like H3BO3 or B2O3 [53]. Ti
2p signal shows the typical bands on Ti4+ in TiO2 while O 1 s can be
deconvoluted into 2 peaks, related with O in TiO2 (530.01 eV) and in
B2O3 (531.80 eV) [53]. The atomic contents of B and Ti revealed by XPS
are 7.03 at% and 29.02 at%, which correspond to a B/Ti ratio of 0.242
that is much higher that the nominal B/Ti ratio of 0.09 employed during
the synthesis. This suggests that B is mainly segregated in the outermost
layers of TiO2 nanocrystals and can justify the different sintering
behaviour of the TiO2-B,N with respect to undoped TiO2, finally leading
to higher surface area for the doped material. Notably, no signals can be
distinguished from the background in the region of binding energy of the
N 1 s (396 – 410 eV). This result suggests that the most part of N has been
removed during the calcination treatment at the end of sol–gel synthesis
or, if present, it is located mainly in the bulk of the material.
2 x num ber of H2 m olucules
num ber of incident photons at 350 nm
AQY =
Synthesis of benzimidazoles was performed suspending 150 mg of
catalyst into 60 mL of a 4 mM solution of precursor in ethanol–water
solution (with different composition). After purging with Ar for 30 min,
lamp was switched on and small portions of the suspension were
collected every hour to check the progress of the reaction by UV–vis
analysis and/or GC/MS analysis (after filtration of the catalyst). During
irradiation, Ar effluent was periodically analyzed to check the presence
of gaseous products by GC: experiments were stopped when H2 was
detected in the gas phase. Various experiments have been performed to
evaluate the effect of metal co-catalyst and water amount in the solvent
on yield and selectivity on desired benzimidazole.
For the selectivity tests with substituted 2-nitroanilines, experiments
The optical properties of TiO2 and TiO2-B,N has been characterized
3

T. Montini et al.
Inorganica Chimica Acta 520 (2021) 120289
Fig. 1. Powder XRD patterns of (a) TiO2 and (b) TiO2-B,N supports.
Table 1
Structural and textural characteristics of TiO2 and TiO2-B,N photoactive
materials.
Composition
Crystallite size
SSA
CPV
dM
(wt%)
(nm)
(m2/g)
(mL/g)
(nm)
A
B a
R
A
B a
R a
a
a
a
TiO2
TiO2-B,
N
64
69
28
27
8
4
11
8
11
7
32
12
80
0.149
0.276
6.8
6.3
138
aA = anatase; B = brookite; R = rutile.
bSSA = Specific surface Area; CPV = Cumulative Pore Volume; dM = maximum
of the pore size distribution obtained from BJH analysis of the desorption branch
of the N2 physisorption isotherm.
by diffuse reflectance UV–vis spectroscopy. As expected, the two pho-
toactive materials show photon absorption only in the UV range, below
400 nm (Fig. S3a). By extrapolation of the linear parts of Tauc plots
(Fig. S3b), indirect band gaps of 2.96 and 2.97 eV were calculated for
TiO2 and TiO2-B,N, respectively. These results are in good agreement
with experimental [51] and theoretical [49,54] studies regarding the
effect of B doping on the electronic structure of TiO2: with its low nu-
clear charge, the electronic 2p states of B are located at energy higher
than the lower edge of conduction band of TiO2 (made essentially by 3d
states of Ti) [49,54]. As a consequence, band-gap band gap is marginally
affected by B doping. On the other hand, a decrease in band gap was
expected by doping with N [49,51,54], since N introduces filled 2p states
just above the top of the valence band (made essentially by O 2p states).
This effect was not observed in TiO2-B,N since the amount of N dopant is
negligible (if present).
Fig. 2. Representative HAADF-STEM micrographs of the photocatalysts based
on TiO2-B,N after photodeposition of the metal cocatalysts: (a) Pd 0.5 wt%, (b)
Pt 0.5 wt%, (c) Ag 0.5 wt% and (d) Cu 1.0 wt%.
other hand, Pd and Ag show very broad size distributions, with the
presence of few very large metal nanoparticles (up to ~ 40 nm for Ag). A
rare image of this situation is presented in Fig. S4, in which a very large
Ag particle is present together with many small nanoparticles. In these
cases, it is evident that growth rate becomes competitive with nucleation
rate, even if similar amounts of photogenerated electrons are expected
for all the samples supported on TiO2-B,N. These results are in good
agreement with the recent study of Puga et al. [55]. As a comparison,
Pd/TiO2 has also been analysed by HAADF-STEM (Fig. S5). The broad
distribution of Pd nanoparticle sizes obtained also in this case suggests
that, when Pd is photodeposited, the growth process is competitive with
the nucleation process without an evident dependence from the doping
of the TiO2 support.
The morphology of the photocatalysts has been investigated by
means of HAADF-STEM. The results are summarized by the represen-
tative images presented in Fig. 2 while the size distribution of the metal
nanoparticles are compared in Fig. 3. TiO2-B,N is composed by an
agglomeration of crystallites of TiO2 with an irregular shape and crys-
tallite sizes in the range 5 – 20 nm, in agreement with the results ob-
tained from the powder XRD.
Metal nanoparticles can be easily observed by HAADF-STEM on the
surface of TiO2-B,N, even if with different contrast (Fig. 2). The size
distributions of metal nanoparticles strongly depends on the type of
metal. The mean size of the Pt and Cu nanoparticles is around 2 nm, with
a narrow distribution. This is in agreement with previous observation on
similar materials [38,44]: the small metal nanoparticles have been
formed by reduction of metal ions through the electrons excited in the
conduction band. In this way, many nucleation centres are formed and
nucleation results faster than growth of metal nanoparticles. On the
3.2. Photocatalytic H2 production
With the aim to use H2 produced in situ for the subsequent reduction
of nitro groups, the prepared catalysts have been preliminary tested for
the photocatalytic H2 production from aqueous solutions containing
4

T. Montini et al.
Inorganica Chimica Acta 520 (2021) 120289
of the photoactive material and the nature of the co-catalyst strongly
affect the activity. In particular, the H2 production is enhanced by co-
doping of TiO2 with B and N (Fig. 4a for Pd-based photocatalysts), as
a result of more efficient light harvesting, increased life-time of photo-
generated h+/eꢀ pairs and increased surface area [49–51,67,68]. The
same trend has been observed for Cu-based photocatalysts (Fig. S7).
Normalizing the H2 production rate with respect to the surface area
(Fig. S8), the photocatalysts supported on undoped TiO2 show a higher
efficiency in H2 evolution with respect to the materials prepared using
TiO2-B,N. It must be underlined here that the experimental conditions
employed in the H2 production tests are optimized for the forthcoming
benzimidazole synthesis: the optimal amount of TiO2 photocatalyst,
accordingly to [69], is around 60 mg [38,44] instead of the 150 mg
employed here. In this work, the amount of H2 produced and available
for nitro groups reduction has been estimated and the most part of the
catalyst will be required for the “dark” steps involved in bezimidazole
synthesis (see below).
More relevant is the effect of metal employed as co-catalyst (Fig. 4b).
Among the studied metals, the best performances was observed by Pt
and Pd followed by Cu that showed higher H2 production rate than Ag.
This trend is in agreement with previous reports on the effect of work
function of the metal co-catalyst on H2 production rate in photocatalytic
experiments [70–72]. It was demonstrated that higher is the work
function, higher is the Schottky barrier formed at the interface between
metal nanoparticles and semiconductor. As a result, eꢀ /h+ pairs
recombination is strongly inhibited, favouring their redox processes
taking places on the surface of the photocatalyst, like proton reduction
to H2 and oxidation of the adsorbed ethoxy groups to acetaldehyde [7].
Notably, Cu-based photocatalysts suffer of some metal leaching
during photocatalytic experiments, even under reducing conditions like
in the case of H2 production from alcohol dehydrogenation [38,39]. In
the present case, Cu concentration in the liquid phase at the end of
photocatalytic H2 production experiments has been measured by ICP-
OES. The amount of Cu detected in solution corresponds to 16.1 wt%
of the nominal Cu introduced into the reactor for Cu/TiO2 while this
quantity decreases to 5.9% for Cu/TiO2-B,N. This significant reduction is
very important in view of a practical application of this type of photo-
catalysts. The reduction of Cu leaching could be due to the combination
of different factors, including an increase of the surface area, the pres-
ence of different basic sites due to the presence of dopants that favour
the adsorption of Cu(II) ions and/or the modification of valence band
edge that hinder re-oxidation of metal nanoparticles. Reasonably, Cu
leaching could be further reduced employing a light source with a larger
fraction of UV photons [39].
Fig. 3. Size distribution of the metal nanoparticles in TiO2-B,N samples with
photodeposited metal co-catalysts. Approximatively 400 particles were counted
for each distribution.
EtOH as sacrificial agent. The tests have been performed to confirm the
positive role of the co-doping and the effect of the type of metal on the
rate of H2 production.
Under the adopted experimental conditions, the photocatalytic H2
production takes place mainly by dehydrogenation of ethanol to acet-
aldehyde [7,38,44,45,57]. The GC/MS analysis of the liquids collected
at the end of each experiment always show the presence of acetaldehyde
and 1,1-diethoxyethane, the acetal formed by acetaldehyde and excess
ethanol (Fig. S6). Moreover, no detectable amounts of CO, CO2 or CH4
have been observed in the gas effluents from the photoreactor. Typi-
cally, H2 production starts immediately but a progressive increase is
observed in the first hours due to the chromatographic effect within the
reactor. As the only exceptions, Cu-based photocatalysts show an acti-
vation period (Fig. 4 and S7), with H2 production becoming significant
only after 1.5 h of irradiation. The delay in H2 production at the
beginning of the experiments is due mainly to 2 effects: stabilization of
H2 concentration in the dead volume of photoreactor and activation of
the photocatalyst. In the case of Pt, Pd and Ag co-catalysts, metal
nanoparticles (deposited in the zero-valent state during preliminary
photodeposition) maintain their conditions, being immediately avail-
able to promote H2 production. Therefore, in these cases, initial delay is
due only to H2 diffusion in the dead volume of photoreactor. In the case
of Cu/TiO2-B,N, it is demonstrated that Cu nanoparticles produced by
photodeposition are rapidly oxidized by exposure to air [38,55]. In this
case, a longer time is needed to reach a stable H2 production since
photogenerated electrons are consumed to reduce CuOx species to Cu
nanoparticles before H2 production begins [38,55,58–61]. Various re-
ports in the literature assess that photocatalysts comprising nano-
composite made by different TiO2 polymorphs are more active than the
single phase materials both in oxidation [62,63] and in reduction
[64–66] processes. This is due by the formation of heterojunctions be-
tween the crystallites of the different phases and the capability of pho-
togenerated electrons and holes to transfer from phase to phase.
Considering this, electron/hole recombination is highly inhibited in
multiphasic TiO2, leading to a higher number of sites for nucleation of
metal nanoparticles, essential to have a high rate of H+ reduction to H2,
and for oxidation of the molecules of the sacrificial agent.
Apparent quantum yields for H2 production under the present con-
ditions have been estimated irradiating the samples based on TiO2-B,N
with UV light around 350 nm, by the use of a 10 nm bandpass filter. The
results (Fig. S9) clearly show that, under “monochromatic” irradiation
in the UV range, the H2 production rates are obviously lower because of
the lower total number of photons but follow the same trend presented
in Fig. 4b. Apparent quantum yields are in the range 1.7 – 6.5%, in
agreement with previous reports on the use of TiO2-based photocalysts
for H2 production from alcohols [74,75].
On the bases of the discussed positive textural characteristics (larger
surface area) and promising photocatalytic H2 production of TiO2-B,N
the subsequent studies on the photocatalytic synthesis of benzimidazolic
compounds were performed on those materials only.
3.3. Photocatalytic synthesis of benzimidazole: influence of reaction
conditions
The experimental conditions for the synthesis of benzimidazolic
compounds have been first optimized using Pt/TiO2-B,N photocatalyst.
Notably, the amount of photocatalyst employed is much larger than that
optimized for photocatalytic H2 production from ethanol solution, since
TiO2-B,N must provide also sufficient acid sites to catalyse the
The obtained experimental data indicate that both the composition
5

T. Montini et al.
Inorganica Chimica Acta 520 (2021) 120289
Fig. 4. H2 production rate during photocatalytic dehydrogenation of ethanol: (a) effect of doping on TiO2 and (b) effect of metal co-catalyst.
condensation and cyclization steps.
increased. A similar trend has been observed also analysing the UV–vis
spectra of the solutions collected during the synthesis (Fig. S10). After
switching on of the lamp, a band related to NA and centered around 406
nm appears, with intensity increasing in the first hours and then quickly
disappearing after 18 h. At the same time, the bands related to 2-MBI
(sharp bands in the range 250 – 300 nm) progressively increases. The
H2 evolution is very low during 18 h of irradiation but, after this period,
a sharp increase in the H2 production is observed, in concomitance with
the decrease in NA amount. The steeply increase in H2 production rate
corresponds to the complete reduction of nitro groups. This scenario
suggests that the pathway followed to produce the 2-MBI product starts
with the reduction of DNB to NA (Fig. 5). At this point, NA can be
reduced to 1,2-phenylenediamine, which quickly undergoes condensa-
tion with acetaldehyde (produced from ethanol dehydrogenation). The
formation of the mono-imine derivate is crucial for the following
cyclization/dehydrogenation to 2-MBI. Notably, no 1,2-phenylenedi-
amine has been detected by GC/MS analysis of the aliquots sampled
during the experiment indicating that, under the present conditions,
cyclization reaction is very fast with respect to reduction of the nitro
groups and all the equilibria involved in imine formation are shifted to
the products. Together with 2-MBI, other two compounds have been
detected in GC/MS analysis as by-products: 1-N-(1-ethoxyethyl)-2-
methyl-1H-benzimidazole (N-EtOEt-2-MBI, 5.3 mol%) and 1-N-ethyl-2-
methyl-1H-benzimidazole (N-Et-2-MBI, 3.1 mol%). Their formation is
clearly related with the accumulation of acetaldehyde from ethanol
dehydrogenation, which favours the formation of the di-imine inter-
mediate that, after various steps involving cyclization, dehydrogenation
and reaction with the solvent, produces the N-substituted 2-MBI. A
plausible pathway for the obtainment of the N-substituted 2-MBI is
presented in Fig. S11 and involves also the catalysis by acid sited of TiO2
The proposed reaction scheme is presented in Fig. 5. It starts with the
production of H2 and aldehyde by alcohol photoassisted dehydrogena-
tion, followed by in situ reduction of a nitro compounds to an inter-
mediate phenylenediamine, that is finally condensed with the aldehyde
forming an imine compounds. This intermediate undergoes rapid
cyclization and the desired product is obtained by oxidation/dehydro-
genation. Even if often carboxylic acids or their derivatives have been
employed in the synthesis of benzimidazoles, in this mechanism the
oxidation of acetaldehyde has been excluded considering that:
- Oxidation of acetaldehyde has never been observed under the
experimental conditions adopted in this study [38,44];
- Formation of benzimidazoles from aldehydes requires very mild
conditions, like those employed in this study [33].
The initial tests has been performed using 2-nitroalinine (NA) or 1,2-
dinitrobenzene (DNB) as starting compounds and ethanol as alcohol,
being 2-methyl-1H-benzimidazole (2-MBI) the expected product of the
reaction.
The results obtained during irradiation of DNB in EtOH 96% in the
presence of Pt/TiO2-B,N photocatalyst are presented in Fig. 6a. The
synthesis of 2-MBI from DNB in EtOH 96% showed that the precursor
starts to be consumed immediately after beginning of irradiation of the
photocatalyst. The molar fraction of DNB progressively decreased with
the complete conversion achieved around 17 h. During this period, the
molar fraction of NA increased rapidly in the initial 2 h up to 30 – 40%,
maintaining in this range until DNB is totally consumed; after that, NA
amount decreased, being reduced and converted into the benzimidazolic
product. During all this period, the molar fraction of 2-MBI progressively
Fig. 5. Possible reaction pathway for the photocatalytic production of the 2-MBI from DNB and/or NA coupled with photocatalytic dehydrogenation of ethanol. The
intermediates within square brackets have net been detected by GC/MS analysis (see text).
6

T. Montini et al.
Inorganica Chimica Acta 520 (2021) 120289
Fig. 6. Photocatalytic synthesis of 2-MBI using DNB (a) or NA (b) as starting molecule. Experimental condition: 150 mg Pt/TiO2-B,N, 60 mL solution of precursor 4
mM in EtOH 96%, 30 ◦C, 30 mL/min Ar flow, simulated sunlight irradiation.
and the ability of Pt nanoparticles to promote hydrogenation/
dehydrogenation.
also using Pt/TiO2. Accordingly to the lower H2 production rates
observed for the undoped TiO2 photocatalysts, 15 h were required to
completely convert NA in 2-MBI. Moreover, GC/MS analysis of the
procuts mixture revealed that, besides 2-MBI, N-Et-2-MBI and N-EtOEt-
2-MBI were present (3.54 mol% and 1.32 mol%, respectively). This
result indicates that the use of TiO2-B,N affects both the H2 production
rate and the following cyclization and dehydrogenation steps required
for the synthesis if 2-MBI.
Efforts to reduce the formation of these by-products is mandatory.
Therefore, we studied the influence of the presence of H2O in the solvent
on the imine formation equilibrium. In fact, an optimized amount of
water could reduce and even avoid the formation of the di-imine in-
termediate. The results obtained varying the H2O amount from 0 to 20
vol% in the solvent are presented in Fig. S12. In the absence of H2O, the
main by-product is N-EtOEt-2-MBI (19.6 mol%) followed by N-Et-2-MBI
(3.3 mol%). Adding a large amount of H2O (20 vol%), the main by-
product observed is 1,2- phenylenediamine (10.4 mol%), suggesting
the incomplete conversion into benzimidazoles as an effect of shifting
the equilibrium reaction of imine formation to the reagents. As
observed, the optimal amount of H2O for benzimidazole formation is
between 4 and 10 vol%. The by-products cannot be totally eliminated
since the acetaldehyde concentration increases due to the high amount
of H2 required for the reduction of 2 nitro groups. Notably, it is also
known that a certain amount of H2O strongly enhances the H2 produc-
tion by photocatalytic dehydrogenation of EtOH [76] allowing, in this
case, to reduce the irradiation time required for the complete reduction
of the nitro compound.
3.4. Photocatalytic synthesis of substituted benzimidazole: influence of
metal co-catalyst on selectivity
To extend the photocatalytic approach in the synthesis of more
relevant molecules, the synthesis of benzimidazole unit has been per-
formed also using substituted 2-nitro anilines as starting materials. The
scope of this investigation was the evaluation of selectivity in benz-
imidazole production of the various metal co-catalyst employed, with
the attempt to preserve the substituents on the aromatic ring for further
functionalization of the product. As representative examples, we
selected a bromo- and a dichloro-substituted 2-nitroanilines and another
one with an ester group. In the first 2 cases, the expected products can be
further functionalized by the well-known coupling reactions [77–79]
while the ester group in the last one can be easily converted into many
other organic functionalities. The experiments have been conducted
irradiating the solution on the substituted NA in EtOH 96% at 30 ◦C in
the presence of the TiO2-B,N photocatalyst loaded with the various
metal co-catalysts. Irradiation has been performed until a stable and
significant H2 evolution is observed and UV–vis analysis confirms the
complete reduction of the nitro group.
To avoid a large excess of acetaldehyde during benzimidazole syn-
thesis, the molecular precursor has been changed to 2-nitro aniline (NA),
a compound with a much lower toxicity with respect to 1,2-phenylenedi-
amine and producible on a large scale in chemical industry. The results
obtained during irradiation of NA in EtOH 96% in the presence of Pt/
TiO2-B,N photocatalyst are presented in Fig. 6b. NA conversion and 2-
MBI production start immediately after the beginning of irradiation of
the photocatalyst and the reaction proceeds regularly with reduction of
nitro group, condensation with acetaldehyde and cyclization/dehydro-
genation to 2-MBI. In agreement with GC/MS analysis, UV–vis spectra
recorded on the solution at various irradiation times (Fig. S13) show the
progressive decrease of the intensity of the band centred around 406 nm
related to NA and the concomitant increase on the sharp bands in the
range 250–300 nm related to 2-MBI. No H2 is observed in the gas phase
up to 8 h, indicating that hydrogen atoms extracted from ethanol are
completely employed for reduction of NA. After 8 h, when NA conver-
sion is around 90%, H2 begins to be detected in gas phase, with a sharp
increase in its concentration. After approximatively 10 h of irradiation,
NA is completely converted with a 2-MBI yield close to 100%. Obvi-
ously, the time required to reach the complete conversion of NA is
almost half of that required in the case on DNB, since NA contains only
one nitro group. Only a very minor amount of N-Et-2-MBI (0.84 mol%) is
detected as by-product. Considering the very high selectivity in 2-MBI
synthesis from NA, substituted NA have been employed for the further
investigations in the preparation of substituted benzimidazoles.
To highlight the importance of the use of TiO2-B,N as photoactive
support in benzimidazole synthesis, the reaction has been performed
The results obtained with the different co-catalyst in the synthesis of
substituted 2-MBI starting from 4-bromo-2-nitro-aniline are presented in
Fig. 7, with the GC/MS analysis of the product mixture (after purifica-
tion from eventual by-products from aldol condensation of acetaldehyde
by extraction) are reported in Fig. S14. It’s clearly possible to observe
that the metal co-catalyst strongly affects the selectivity of the synthesis.
Moreover, it must be immediately underlined that the irradiation time
required to complete the reaction (complete disappearing of the nitro
compound checked by UV–vis and significant and constant H2 produc-
tion detected in the gas stream from the reactor) also depends on the
nature of metal co-catalyst, in agreement with the H2 production rates
reported in Fig. 4b.
The shorter reaction time is observed for Pt (11.5 h) followed by Pd
(14.0 h). This is in agreement with the very high H2 production observed
for Pt/TiO2-B,N and Pd/TiO2-B,N. The reason why the reaction time
required by Pd/TiO2-B,N is significantly longer than that of Pt can be
explained considering the selectivity of the products. Pt/TiO2-B,N shows
a very high selectivity (93.5%) for 5-bromo-2-methyl-1H-benzimidazole
(Br-2-MBI), with only 2-MBI as by-product. In the case of Pd/TiO2-B,N,
7

T. Montini et al.
Inorganica Chimica Acta 520 (2021) 120289
the N-ethyl derivative formed by cyclization the diimine intermediate
(Fig. 5): they differ only for the position of the Br substituent on the
aromatic ring because they are formed by the nucleophilic attach of N
atom of one imino group to the first C atom on the other imino group.
Due to their very similar molecular structure, these two compounds
cannot be univocally distinguished from their MS spectra and they are
counted together in Fig. 7. In comparison with Pd, Ag and Cu show very
limited hydrodehalogenation activity but the yields in the N-ethyl de-
rivatives is much higher than that observed for Pt. The production of N-
ethyl derivatives could be related with the lower activity of Ag- and Cu-
based photocatalysts in the dehydrogenation process, a step that is
crucial in producing the desired benzimidazole after cyclization (Fig. 5).
In fact, if the intermediate obtained from cyclization is not dehydro-
genated/oxidized quickly, it can be subjected to re-opening of the ring
and formation of the di-imine intermediate, the precursor of N-ethyl
derivate.
Considering the overall performance of the 4 investigated photo-
catalysts, only Pt and Pd have been have been employed for the prose-
cution of the study considering that, even if Pd showed the total removal
of Br from the product, the N-ethyl by-products are absent. Therefore,
Pt/TiO2-B,N and Pd/TiO2-B,N have been tested in the synthesis of the
benzimidazoles starting from 4,5-dichloro-2-nitro-aniline and from
methyl 4-amino-3-nitro-benzoate. The selectivity in the various prod-
ucts are presented in Fig. 8.
In the synthesis of 5,6-dichloro-2-methyl-1H-benzimidazole (diCl-2-
MBI) from 4,5-dichloro-2-nitro-aniline, the behaviours of Pt and Pd-
based photocatalysts is again significantly different. When using Pt/
TiO2-B,N, the selectivity in the desired product (diCl-2-MBI) is still very
high (95.7%) and only a minor amount of 5-chloro-2-methyl-1H-benz-
imidazole (Cl-2-MBI) was detected. Using Pd/TiO2-B,N, the selectivity
of the desired product is much lower, with the production of large
amounts of the by-products Cl-2-MBI and 2-MBI as a result of the
hydrodehalogenation reaction. In agreement with this, the irradiation
Fig. 7. Effect of metal co-catalyst in the photocatalytic synthesis of 5-bromo-2-
methyl-1H-benzimidazole (Br-2-MBI) from 4-bromo-2-nitro-aniline. Experi-
mental condition: 150 mg Pt/TiO2-B,N, 60 mL solution of precursor 4 mM in
EtOH 96%, 30 ◦C, 30 mL/min Ar flow, simulated sunlight irradiation.
the product produced with 97.8% of selectivity is 2-MBI, indicating that
dehydrohalogenation is been performed with a very high efficiency on
the Pd nanoparticles. Pd-based heterogeneous catalysts are among the
most efficient for hydrodehalogenation of organic substrates following
many different strategies: using H2 gas [80,81], by H-transfer reaction
[82] or by electrocatalytic H generation [83]. Beside of synthetic pur-
poses, hydrodehalogenation with Pd-based catalysts have already been
reported for the remediation of contaminated soils [84] and abatement
of chlorofluorocarbons (CFC) [85]. In the present case, Pd/TiO2-B,N
demonstrated to be highly efficient in removing Br from the aromatic
ring during synthesis of benzimidazole, by H-transfer from EtOH. The
present results suggest the potentiality of the photocatalytic approach in
the hydrodehalogenation of pollutant compounds, using organic re-
newables (already present in contaminates soils) as sacrificial agents
and H donors. From our data, it is evident that Pt has a much lower
hydrodehalogenation activity, favouring the reduction of nitro groups
instead of breaking of the C-Br bond. These results, together with the
possibility to recycle Pd from spent catalysts in a sustainable way [47],
make Pd-based photocatalysts among the most promising in pollutant
removal.
The irradiation times required for Ag/TiO2-B,N and Cu/TiO2-B,N to
complete the benzimidazole synthesis (22.0 and 20.5 h, respectively) are
in agreement with the lower H2 production rate observed for these
photocatalysts (Fig. 4b). GC/MS analysis of the benzimidazole products
shows important differences with respect to the previous photocatalysts.
When Ag/TiO2-B,N is used, the selectivity in the desired product (Br-2-
MBI) is comparable with that of Pt/TiO2-B,N (94.4%) but all the by-
products listed in Fig. 7 are observed (Fig. S14). In the case of Cu/
TiO2-B,N, the selectivity for the main product is the lowest observed
(68.2%) and high amounts of the by-products are detected. In particular,
high amounts of the two isomers of bromo-1-N-ethyl-1-methyl-benz-
imidazole (Br-N-Et-2-MBI) are present. Notably, these compounds are
Fig. 8. Effect of metal co-catalyst in the photocatalytic synthesis of (a) 5,6-
dichloro-2-methyl-1H-benzimidazole (diCl-2-MBI) from 4,5-dichloro-2-nitro-
aniline and (b) 5-carboxymethyl-2-methyl-1H-benzimidazole (COOMe-2-MBI)
from methyl 4-amino-3-nitro-benzoate. Experimental condition: 150 mg Pt/
TiO2-B,N, 60 mL solution of precursor 4 mM in EtOH 96%, 30 ◦C, 30 mL/min Ar
flow, simulated sunlight irradiation.
8

T. Montini et al.
Inorganica Chimica Acta 520 (2021) 120289
time required for the complete conversion of the nitro aniline is much
longer using Pd/TiO2-B,N (13.0 h) than using Pt/TiO2-B,N (11.5 h). The
observed differences in the selectivity is again related with the ability of
Pt and Pd to promote the reduction of nitro groups over the hydro-
dehalogenation, in agreement with the results previously reported for
the selective catalytic hydrogenation of nitro groups in the presence of
activated aryl halides [86]. The GC/MS analysis of the product mixtures
obtained from 4,5-dichloro-2-nitro-aniline are presented in Fig. S15.
Finally, in the synthesis of 5-carboxymethyl-2-methyl-1H-benzimid-
azole from methyl 4-amino-3-nitro-benzoate both Pt/TiO2-B,N and Pd/
TiO2-B,N show very high selectivity to the desired product (98.5% and
94.7%, respectively). In the case of Pt/TiO2-B,N, GC/MS analysis
revealed the presence of only a very low amount of 2-MBI, indicating
that the ester group has been completely removed from the product. On
the other hand, using Pd/TiO2-B,N relevant amounts on the N-ethyl
derivatives has been detected (Fig. S1). Again, the two isomers cannot be
distinguished only on the basis of their MS spectra and are quantified
together.
Acknowledgements
TM, GC, GA, and PF acknowledge the University of Trieste (FRA
2018), Ministry for University and Research (PRIN 2017PBXPN4),
INSTM Consortium and ICCOM-CNR for financially supporting their
research activities. JJD thanks the financial support of the Ministry of
Science and Innovation of Spain/FEDER Program of the EU (ENE2017-
82451-C3-2-R).
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.ica.2021.120289.
References
[1] United Nation, 17 Sustain. Delevopment Goals (n.d.) https://sdgs.un.org/.
[2] G. Palmisano, V. Augugliaro, M. Pagliaro, L. Palmisano, Chem. Commun. (Camb).
(2007) 3425–3437.
[3] J.J. Douglas, M.J. Sevrin, C.R.J. Stephenson, Org. Process Res. Dev. 20 (2016)
1134–1147.
4. Conclusions
[4] T.P. Nicholls, D. Leonori, A.C. Bissember, Nat. Prod. Rep. 33 (2016) 1248–1254.
[5] A.V. Puga, Coord. Chem. Rev. 315 (2016) 1–66.
˜
´
[6] I. Romero Ocana, A. Beltram, J.J. Delgado Jaen, G. Adami, T. Montini,
P. Fornasiero, Inorganica Chim. Acta 431 (2015) 197–205.
[7] M. Cargnello, A. Gasparotto, V. Gombac, T. Montini, D. Barreca, P. Fornasiero, Eur.
J. Inorg. Chem. (2011) 4309–4323.
In this work, we investigated the role of co-doping of TiO2 with B and
N and of the employed co-catalyst on the photocatalytic activity of the
obtained materials in H2 production by dehydrogenation of ethanol and
its use for the production of 2-methylbenzimidazole by a one-pot process
starting from non-toxic nitroaniline compounds. Co-coping of TiO2 in-
creases surface area of TiO2 by significantly reducing the crystallite sizes
without affecting the phase composition of the materials. The H2 pro-
duction rate is favourable influenced by this increase of surface area
while the various metal co-catalysts (Pt, Pd, Cu, Ag) are affecting the
photocatalytic performance accordingly to their work function.
Photocatalytic 2-MBI synthesis has been optimized starting from
dinitrobenzene or 2-nitro aniline, showing that a low amount of H2O in
the solvent is convenient to avoid formation of highly toxic in-
termediates (such as phenylenediamine) or of un-desired N-substituted
benzimidazoles. The synthesis of 2-MBI proceeds through reduction of
the nitro groups by the H abstracted from ethanol by the photocatalyst,
reaction of amino groups with the acetaldehyde and rapid cyclization of
the imine intermediate, which produces the 2-MBI after oxidation/
dehydrogenation.
[8] M.F. Kuehnel, E. Reisner, Angew. Chemie Int. Ed. 57 (2018) 3290–3296.
[9] N. Luo, T. Montini, J. Zhang, P. Fornasiero, E. Fonda, T. Hou, W. Nie, J. Lu, J. Liu,
M. Heggen, L. Lin, C. Ma, M. Wang, F. Fan, S. Jin, F. Wang, Nat. Energy 4 (2019)
575–584.
[10] F. Su, S.C. Mathew, G. Lipner, X. Fu, M. Antonietti, S. Blechert, X. Wang, J. Am.
Chem. Soc. 132 (2010) 16299–16301.
[11] A. Tanaka, K. Hashimoto, H. Kominami, J. Am. Chem. Soc. 134 (2012)
14526–14533.
[12] S. Higashimoto, N. Kitao, N. Yoshida, T. Sakura, M. Azuma, H. Ohue, Y. Sakata,
J. Catal. 266 (2009) 279–285.
[13] J. Diniz, C.D. Nunes, O.C. Monteiro, Inorg. Chem. Commun. 119 (2020), 108099.
[14] X. Lin, S. Xu, Z.Q. Wei, S. Hou, Q.L. Mo, X.Y. Fu, F.X. Xiao, J. Mater. Chem. A 8
(2020) 20151–20161.
[15] Y. Wang, X. Wang, M. Antonietti, Angew. Chemie - Int. Ed. 51 (2012) 68–89.
[16] Q. Xiao, S. Sarina, E.R. Waclawik, J. Jia, J. Chang, J.D. Riches, H. Wu, Z. Zheng,
H. Zhu, ACS Catal. 6 (2016) 1744–1753.
[17] H. Wang, J. Yan, W. Chang, Z. Zhang, Catal. Commun. 10 (2009) 989–994.
[18] Z. Ding, X. Chen, M. Antonietti, X. Wang, ChemSusChem 4 (2010) 274–281.
[19] X.H. Li, M. Baar, S. Blechert, M. Antonietti, Sci. Rep. 3 (2013) 1–6.
[20] S. Zhang, C. Chang, Z. Huang, Y. Ma, W. Gao, J. Li, Y. Qu, ACS Catal. 5 (2015)
6481–6488.
¨
[21] A. Savateev, I. Ghosh, B. Konig, M. Antonietti, Angew. Chemie - Int. Ed. 57 (2018)
The metal loaded co-catalysts have been investigated also in the
synthesis of substituted benzimidazoles. The best performance of Pt/
TiO2-B,N in benzimidazole synthesis derive from a combination of two
factors. The first is the high activity in H2 production, which reduces the
time required for complete reduction of nitro groups. The second factor
is the low ability of Pt to hydrogenate the C-Cl and C-Br bonds in
15936–15947.
[22] M. Boiani, M. Gonzalez, Mini-Rev. Med. Chem. 5 (2005) 409–424.
[23] P.S. Charifson, A.L. Grillot, T.H. Grossman, J.D. Parsons, M. Badia, S. Bellon, D.
D. Deininger, J.E. Drumm, C.H. Gross, A. LeTiran, Y. Liao, N. Mani, D.P. Nicolau,
E. Perola, S. Ronkin, D. Shannon, L.L. Swenson, Q. Tang, P.R. Tessier, S.K. Tian,
M. Trudeau, T. Wang, Y. Wei, H. Zhang, D. Stamos, J. Med. Chem. 51 (2008)
5243–5263.
–
halogen-substituted precursors or the C O bond in the esther-
[24] A.W. White, R. Almassy, A.H. Calvert, N.J. Curtin, R.J. Griffin, Z. Hostomsky,
K. Maegley, D.R. Newell, S. Srinivasan, B.T. Golding, J. Med. Chem. 43 (2000)
4084–4097.
functionalized precursor. These aspects are now better summarized in
the conclusion section. Cu and Ag showed the production of large
amounts of N-ethyl benzimidazole products while Pd is able to effi-
ciently promote the hydrodehalogenation of the aromatic ring. Despite
this result could be useful for other applications (like degradation of
halogenated environmental pollutants), in this case halogens must be
preserved for further functionalization of the products. The study clearly
indicate the potentiality of metal loaded photocatalysts for organic
synthesis and of benzimidazole in particular. However, a fine tuning of
the nature of the metal co-catalyst, of the properties of the semi-
conductor, from its surface area, light absorption characteristic and
acidic nature, is necessary to guaranty promising activity and selectivity.
¨
[25] G. Rackur, M. Bickel, H.W. Fehlhaber, A. Herling, V. Hitzel, H.J. Lang, M. Rosner,
R. Weyer, Biochem. Biophys. Res. Commun. 128 (1985) 477–484.
[26] D. Luneau, P. Rey, Coord. Chem. Rev. 249 (2005) 2591–2611.
´
´
[27] F. Tellez, H. Lopez-Sandoval, S.E. Castillo-Blum, N. Barba-Behrens, Arkivoc 2008
(2008) 245–275.
[28] Y. Dong, H. Zhang, F. Lei, M. Liang, X. Qian, P. Shen, H. Xu, Z. Chen, J. Gao, J. Yao,
J. Solid State Chem. 245 (2017) 160–163.
[29] P.N. Preston, Chemical Reviews Synthesis, Reactions, and Spectroscopic Properties
of Benzimidazoles, 1974.
[30] J.B. Wright, Chem. Rev. 48 (1951) 397–541.
[31] S.I. Alaqeel, J. Saudi Chem. Soc. 21 (2017) 229–237.
[32] K. Bahrami, M.M. Khodaei, I. Kavianinia, Synthesis 2007 (2007) 547–550.
´
[33] P.L. Beaulieu, B. Hache, E. Von Moos, Synthesis 2003 (2003) 1683–1692.
[34] Y. Shiraishi, Y. Sugano, S. Tanaka, T. Hirai, Angew. Chemie - Int. Ed. 49 (2010)
1656–1660.
[35] K. Selvam, M. Swaminathan, Tetrahedron Lett. 52 (2011) 3386–3392.
˜
[36] T. Montini, M. Monai, A. Beltram, I. Romero-Ocana, P. Fornasiero, Mater. Sci.
Declaration of Competing Interest
Semicond. Process. 42 (2016) 122–130.
´
[37] K.C. Christoforidis, T. Montini, M. Fittipaldi, J.J.D. Jaen, P. Fornasiero,
ChemCatChem 11 (2019) 6408–6416.
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influence
the work reported in this paper.
[38] T. Montini, V. Gombac, L. Sordelli, J.J. Delgado, X. Chen, G. Adami, P. Fornasiero,
ChemCatChem 3 (2011) 574–577.
9

T. Montini et al.
Inorganica Chimica Acta 520 (2021) 120289
[39] V. Gombac, L. Sordelli, T. Montini, J.J. Delgado, A. Adamski, G. Adami,
M. Cargnello, S. Bernai, P. Fornasiero, J. Phys. Chem. A 114 (2010).
[40] M. Melchionna, A. Beltram, T. Montini, M. Monai, L. Nasi, P. Fornasiero, M. Prato,
Chem. Commun. 52 (2016) 764–767.
[61] M. Imizcoz, A.V. Puga, Catal. Sci. Technol. 9 (2019) 1098–1102.
[62] C.A. Emilio, M.I. Litter, M. Kunst, M. Bouchard, C. Colbeau-Justin, Langmuir 22
(2006) 3606–3613.
[63] J.T. Carneiro, T.J. Savenije, J.A. Moulijn, G. Mul, J. Phys. Chem. C 115 (2011)
2211–2217.
[41] A. Beltram, M. Melchionna, T. Montini, L. Nasi, P. Fornasiero, M. Prato, Green
Chem. 19 (2017) 2379–2389.
[64] Y. Liu, Z. Wang, W. Wang, W. Huang, J. Catal. 310 (2014) 16–23.
[65] C.-H. Lin, C.-H. Lee, J.-H. Chao, C.-Y. Kuo, Y.-C. Cheng, W.-N. Huang, H.-W. Chang,
Y.-M. Huang, M.-K. Shih, Catal. Lett. 98 (2004) 61–66.
´
[42] M. Cargnello, T. Montini, S.Y. Smolin, J.B. Priebe, J.J. Delgado Jaen, V.V.T. Doan-
Nguyen, I.S. McKay, J.A. Schwalbe, M.-M. Pohl, T.R. Gordon, Y. Lu, J.B. Baxter,
A. Brückner, P. Fornasiero, C.B. Murray, Proc. Natl. Acad. Sci. 113 (2016)
3966–3971.
[66] H.-L. Kuo, C.-Y. Kuo, C.-H. Liu, J.-H. Chao, C.-H. Lin, Catal. Letters 113 (2007)
7–12.
[43] A. Gallo, T. Montini, M. Marelli, A. Minguzzi, V. Gombac, R. Psaro, P. Fornasiero,
V. Dal Santo, ChemSusChem 5 (2012) 1800–1811.
[67] M.V. Dozzi, E. Selli, J. Photochem. Photobiol. C Photochem. Rev. 14 (2013) 13–28.
[68] Y. Li, G. Ma, S. Peng, G. Lu, S. Li, Appl. Surf. Sci. 254 (2008) 6831–6836.
[69] H. Kisch, D. Bahnemann, J. Phys. Chem. Lett. 6 (2015) 1907–1910.
[70] B. Gupta, A.A. Melvin, T. Matthews, S. Dash, A.K. Tyagi, Renew. Sustain. Energy
Rev. 58 (2016) 1366–1375.
˜
`
[44] A. Beltram, I. Romero-Ocana, J.J. Delgado Jaen, T. Montini, P. Fornasiero, J. Jose
Delgado Jaen, T. Montini, P. Fornasiero, Appl. Catal. A Gen. 518 (2016) 167–175.
[45] M. Murdoch, G.I.N. Waterhouse, M.A. Nadeem, J.B. Metson, M.A. Keane, R.
F. Howe, J. Llorca, H. Idriss, Nat. Chem. 3 (2011) 489–492.
[71] Y. Mizukoshi, Y. Makise, T. Shuto, J. Hu, A. Tominaga, S. Shironita, S. Tanabe,
Ultrason. Sonochem. 14 (2007) 387–392.
´
[46] A. Naldoni, T. Montini, F. Malara, M.M. Mroz, A. Beltram, T. Virgili, C.L. Boldrini,
˜
M. Marelli, I. Romero-Ocana, J.J. Delgado, V. Dal Santo, P. Fornasiero, ACS Catal.
[72] A. Naldoni, M. D’Arienzo, M. Altomare, M. Marelli, R. Scotti, F. Morazzoni, E. Selli,
V. Dal Santo, Appl. Catal. B Environ. 130–131 (2013) 239–248.
[74] M. Yasuda, T. Matsumoto, T. Yamashita, Renew. Sustain. Energy Rev. 81 (2018)
1627–1635.
7 (2017) 1270–1278.
[47] V. Gombac, T. Montini, A. Falqui, D. Loche, M. Prato, A. Genovese, M.L. Mercuri,
A. Serpe, P. Fornasiero, P. Deplano, Green Chem. 18 (2016) 2745–2752.
[49] C. Di Valentin, G. Pacchioni, Catal. Today 206 (2013) 12–18.
[50] G. Liu, Y. Zhao, C. Sun, F. Li, G.Q. Lu, H.-M. Cheng, Angew. Chem. Int. Ed. Engl. 47
(2008) 4516–4520.
[75] S. Escobedo Salas, B. Serrano Rosales, H. De Lasa, Appl. Catal. B Environ. 140–141
(2013) 523–536.
[76] N. Strataki, V. Bekiari, D.I. Kondarides, P. Lianos, Appl. Catal. B Environ. 77 (2007)
184–189.
[51] V. Gombac, L. De Rogatis, A. Gasparotto, G. Vicario, T. Montini, D. Barreca,
G. Balducci, P. Fornasiero, E. Tondello, M. Graziani, Chem. Phys. 339 (2007)
111–123.
[77] A. Suzuki, Angew. Chemie - Int. Ed. 50 (2011) 6722–6737.
[78] I.P. Beletskaya, F. Alonso, V. Tyurin, Coord. Chem. Rev. 385 (2019) 137–173.
[79] E.I. Negishi, Angew. Chemie - Int. Ed. 50 (2011) 6738–6764.
[52] K.S.W. Sing, D.H. Everett, R.A.W. Haul, L. Moscou, R.A. Pierotti, J. Rouquerol,
T. Siemieniewska, Pure Appl Chem 57 (1985) 603–619.
´
[80] C.-H. Pelisson, A. Denicourt-Nowicki, C. Meriadec, J.-M. Greneche, A. Roucoux,
[53] A. Zaleska, E. Grabowska, J.W. Sobczak, M. Gazda, J. Hupka, Appl. Catal. B
Environ. 89 (2009) 469–475.
ChemCatChem 7 (2015) 309–315.
[81] H. Min, S. Lee, M. Park, J. Hwang, H.M. Jung, S. Lee, J. Organomet. Chem. 755
(2014) 7–11.
[54] R. Asahi, T. Morikawa, H. Irie, T. Ohwaki, Chem. Rev. 114 (2014) 9824–9852.
[55] A. V. Puga, N. Barka, M. Imizcoz, ChemCatChem (2020) cctc.202001048.
[57] A.V. Puga, A. Forneli, H. García, A. Corma, Adv. Funct. Mater. 24 (2014) 241–248.
[58] A. Petala, E. Ioannidou, A. Georgaka, K. Bourikas, D.I. Kondarides, Appl. Catal. B
Environ. 178 (2015) 201–209.
[82] M. Muzzio, H. Lin, K. Wei, X. Guo, C. Yu, T. Yom, Z. Xi, Z. Yin, S. Sun, A.C.
S. Sustain, Chem. Eng. 8 (2020) 2814–2821.
[83] C. Cui, J. Wu, Y. Xin, Y. Han, Korean J. Chem. Eng. 32 (2015) 1069–1074.
[84] H.Y. Wee, J. Cunningham, Environ. Prog. Sustain Energy 30 (2011) 589–598.
[85] S.C. Shekhar, J.K. Murthy, P.K. Rao, K.S.R. Rao, Appl. Catal. A Gen. 271 (2004)
95–101.
[59] V. Polliotto, S. Livraghi, A. Krukowska, M.V. Dozzi, A. Zaleska-Medynska, E. Selli,
E. Giamello, A.C.S. Appl, Mater. Interfaces 10 (2018) 27745–27756.
[60] Z. Wang, K. Teramura, T. Shishido, T. Tanaka, in: Top. Catal., Kluwer Academic
Publishers, 2014, pp. 975–983.
[86] A.J. Kasparian, C. Savarin, A.M. Allgeier, S.D. Walker, J. Org. Chem. 76 (2011)
9841–9844.
10