Role of TiO2-based photocatalysts on the synthesis of the pharmaceutical precursor benzhydrol by UVA-LED radiation

Article abstract of DOI:10.1016/j.jphotochem.2019.112350

The influence of several reactional parameters was investigated on the photocatalytic reduction of benzophenone to the pharmaceutical precursor benzhydrol under UVA-LED radiation. The optical semiconductor titanium dioxide was loaded with 1.0 wt.% of nobl

Full text of DOI:10.1016/j.jphotochem.2019.112350

Journal Pre-proof
Role of TiO2-based photocatalysts on the synthesis of the
pharmaceutical precursor benzhydrol by UVA-LED radiation
Joana C. Lopes, Maria J. Sampaio, Bernardo Rosa, Maria J. Lima,
´
Joaquim L. Faria, Claudia G. Silva
PII:
S1010-6030(19)31371-1
DOI:
https://doi.org/10.1016/j.jphotochem.2019.112350
JPC 112350
Reference:
To appear in:
Journal of Photochemistry & Photobiology, A: Chemistry
Received Date:
Revised Date:
Accepted Date:
10 August 2019
16 November 2019
28 December 2019
Please cite this article as: Lopes JC, Sampaio MJ, Rosa B, Lima MJ, Faria JL, Silva CG, Role
of TiO2-based photocatalysts on the synthesis of the pharmaceutical precursor benzhydrol by
UVA-LED radiation, Journal of Photochemistry and amp; Photobiology, A: Chemistry (2019),
doi: https://doi.org/10.1016/j.jphotochem.2019.112350
This is a PDF file of an article that has undergone enhancements after acceptance, such as
the addition of a cover page and metadata, and formatting for readability, but it is not yet the
definitive version of record. This version will undergo additional copyediting, typesetting and
review before it is published in its final form, but we are providing this version to give early
visibility of the article. Please note that, during the production process, errors may be
discovered which could affect the content, and all legal disclaimers that apply to the journal
pertain.
© 2019 Published by Elsevier.

Role of TiO2-based photocatalysts on the synthesis of the
pharmaceutical precursor benzhydrol by UVA-LED radiation
Joana C. Lopes, Maria J. Sampaio*, Bernardo Rosa, Maria J. Lima, Joaquim L.
Faria, Cláudia G. Silva
Laboratory of Separation and Reaction Engineering – Laboratory of Catalysis and
Materials (LSRE-LCM), Departamento de Engenharia Química, Faculdade de
Engenharia, Universidade do Porto, Rua Dr. Roberto Frias s/n, 4200-465 Porto,
Portugal.
*Corresponding author: mjsampaio@fe.up.pt (Maria J. Sampaio)
Graphical abstract
Highlights
 Benzhydrol formation was evaluated.by photochemical and photocatalytic routes.
 TiO2-based materials were synthesized and characterized.
1

 The photocatalysts were studied under UVA-LED radiation.
 Efficient formation of benzhydrol was achieved using Pd/TiO2 as photocatalyst.
 Photoluminescence analyses suggest effective electron transfer between TiO2 and
Pd particles.
Abstract
The influence of several reactional parameters was investigated on the photocatalytic
reduction of benzophenone to the pharmaceutical precursor benzhydrol under UVA-LED
radiation. The optical semiconductor titanium dioxide was loaded with 1.0 wt.% of noble
metals (Pd, Pt and Au) by incipient wetness impregnation method. The TiO2-based
materials were characterized by various techniques. Results showed that the benzhydrol
selectivity is highly influenced by the nature of the solvent and the presence of KOH
(acting as nucleophilic agent) The most efficient system was obtained using 2-propanol,
5.0 mM of KOH, and 0.1 g L-1 of catalyst, achieving 100% selectivity of benzhydrol after
10 min reaction. Among the TiO2-based materials, the Pd/TiO2 showed high efficiency
for benzhydrol formation (in absence KOH) with an enhancing of c.a. 49% compared
with neat TiO2. The decrease of photoluminescence intensity observed for Pd/TiO2
material suggests an effective electron transfer from TiO2 to Pd nanoparticles.
Keywords: Photocatalysis; optical semiconductor; fine organic chemicals; Redox
reactions; noble metals.
1. Introduction
The high demand of fine organic chemicals including fragrances and flavours,
agrochemicals and pharmaceuticals are driving worldwide industry and researchers to
2

developing new processes affording high selectivity of these chemicals. New perspectives
on the development of alternative industrial processes that can be safe, environmentally
friendly, and profitable have been attempted [1-4]. Heterogeneous photocatalysis
appeared as a promising technology by providing a green alternative to conventional
thermal catalytic organic synthesis processes [5-9]. Advantages of photocatalytic
processes include the possibility of operating under mild conditions of temperature and
pressure, exempt of using hazardous reagents and the prospect of using low energy
consumption irradiation sources for reaction activation [10-12]. Benzhydrol
(diphenylmethanol, BH) is a precursor commonly used in the pharmaceutical industry for
the production of drugs such as antihistamine, antihypertensive and antiallergenic
agents[13]. In general, the synthesis of benzhydrol may occur through Grignard reaction
(e.g., using ethylmagnesium or phenylmagnesium bromide [14] or by catalytic reduction
of benzophenone (BP) using catalysts such as sodium or aluminum amalgams [15, 16].
In organic photochemistry, the reduction of benzophenone by UV irradiation, in the
presence of alcohols, is generally considered as a model ketone photoreduction reaction
[17, 18]. In such conditions, the formation of benzhydrol, acetone and other intermediates
(e.g., benzopinacol and benzopinacolone) occur by hydrogen abstraction from the
photoexcited benzophenone in the presence of an alcohol [18]. Schuster et al. [19],
described the photoreduction of BP in 2-propanol under UV irradiation, being
benzopinacol the main product and BH detected in very low concentrations. Bochet et al.
[20] reported the formation of BH by photoreduction of BP in 2-propanol, using
potassium hydroxide (KOH) as reaction promoter. The use of TiO2 as heterogeneous
photocatalyst for the reduction of benzophenone was also reported by Escobar et al. [21].
The novelty of this work relies on the application of heterogeneous photocatalysis
technology using noble metals-TiO2 as catalysts, avoiding in this way the addition of
3

strong bases (such as KOH). The present work explored systematic aspects of the
photocatalytic synthesis of the pharmaceutical precursor benzhydrol from benzophenone
reduction under an energy-efficient and low-cost light UVA-LEDs as irradiation source.
2. Experimental
2.1 Chemical and methods
Benzophenone (BP, (C6H5)2CO  99%) and 2-propanol ((CH3)2CHOH, 99.5%) were
purchased from Sigma-Aldrich. Potassium hydroxide (KOH,  86%) and benzhydrol
(BH, (C6H5)2CHOH, 99%) were supplied from Fluka. Benzopinacol (BPC,
(C6H5)2C(OH)C(C6H5)2OH, 99%) was purchased from Alfa Aesar. Ethanol
(CH3CH2OH, ≥99.5%) was obtained from Panreac. Methanol (CH3OH, 99%) was
supplied from VWR. Ultrapure water was produced in a Direct-Q Millipore system.
TiO2 Aeroxide® P25 (P25, 80% anatase and 20% rutile) powder was obtained from
Evonik Degussa Corporation (Essen, Germany). Gold (III) chloride trihydrate
(HAuCl4.3H2O,  49%), palladium (II) chloride (PdCl2,  99%) and chloroplatinic acid
solution (H2PtCl6, 8 wt.% in H2O) were supplied from Sigma-Aldrich.
2.2 Preparation and characterization of the catalysts
Noble metals were loaded on TiO2 by incipient wetness impregnation method. Briefly,
the TiO2 was placed in a flask, and was sonicated for 30 min. Then, an aqueous solution
with the noble metal precursor (1.0 wt.%) was added drop by drop into the flask
containing the TiO2 under sonication for 90 min. The final catalysts were calcined under
N2 flow at 200 ºC for 1h followed by a H2 flow at the same temperature for 3 h. The
catalysts were labeled as M/TiO2, where M corresponds to Au, Pt or Pd.
The specific surface area of the resulting materials was monitored by multipoint analysis
of N2 adsorption isotherms at -196 °C using the Brunauer-Emmett-Teller (BET) method
4

in a Quantachrome NOVA 4200e apparatus. Transmission electron microscopy (TEM)
images were obtained using a LEO 906E instrument operating at 120 kV, equipped with
a 4 M pixel 28x28 CCD camera from TRS.
X-ray diffraction (XRD) analysis was conducted in a PANalytical X’Pert MPD apparatus
equipped with an X’Celerator detector and secondary monochromator (Cu Kα λ = 0.154
nm, 50 kV, 40 mA; data recorded at a 0.017° step size, 100 s/step). Rietveld refinement
with a PowderCell software was applied for identification of the crystallographic phases.
X-ray photoelectron spectroscopy (XPS) analysis was performed in a Kratos AXIS Ultra
HAS apparatus, using the VISION software for data acquisition and CASAXPS software
for data analysis. The analyses were carried out with a monochromatic Al K X-ray
source (1486.7 eV), operating at 15kV (90 W), in FAT (Fixed Analyser Transmission)
mode, with a pass energy of 40 eV for regions ROI and 80 eV for survey.
Diffuse reflectance UV–Vis spectra (DRUV-Vis) were conducted in a JASCO V-560
spectrophotometer equipped with an integrating sphere attachment (JASCO ISV-469).
The spectra were recorded in diffuse reflectance mode and transformed by the JASCO
software to equivalent absorption Kubelka–Munk units.
Photoluminescence (PL) measurements were carried out at room temperature on a
JASCO (FP-82000) fluorescence spectrometer with a 150W Xenon lamp as light source.
2.3 Photocatalytic setup
In a typical photocatalytic experiment, a glass-immersion reactor was filled with 100 mL
of a 2-propanol solution containing a certain concentration of benzophenone as subtract
for benzhydrol synthesis. The solution was magnetically stirred and continuously purged
5

with argon. The photocatalytic reactions were performed in the absence of KOH and
using 1 g L-1 of catalyst. For comparison purposes, photochemical experiments under
homogeneous conditions were also performed, using KOH as a promoter to hydrogen
transfer. The reactions were performed under UV light irradiation using four 10 W LEDs
with an emission line at 385 nm located axially at a 3.5 cm distance from the reactor
walls. The irradiation power of each LED reaching the reactor was c.a. 400 W m-2, which
was determined using a UV-Vis spectroradiometer (USB2000+, OceanOptics, USA).
Before turning the illumination on, the benzophenone solution was saturated with argon
and the first sample was taken after 1h. Then, the solution was irradiated for 1 h and
samples were withdrawn regularly from the reactor and analyzed by High Performance
Liquid Chromatography (HPLC) using a Hitachi Elite LaChrom instrument equipped
with a Diode Array Detector (L-2450), a Purospher Star RP-18 column (250 mm x 4.6
mm, 5 mm particles), and a solvent delivery pump (L-2130) at a fixed flow rate of 1 mL
min-1. The method starts with an equilibrated mixture of water (A):methanol (B) (20:80)
followed by a linear gradient step to A:B (10:90) in 6 min; finally the initial conditions
were re-established in a 9 min gradient step and the A:B (20:80) mixture was isocratically
eluted for 4 min. The production of benzhydrol and benzophenone consumption during
the photocatalytic reactions was followed as function of conversion (X), selectivity (S)
and the reaction yield (Y).
3. Results and discussion
3.1 Catalyst characterization
The specific surface area (SBET) of the catalysts was determined by N2 adsorption
at -196 ºC. As already reported in the literature the SBET of the commercial Aeroxide®
6

TiO2 P25 is c.a. 53 ± 3 m2 g-1. Some reports have also been showed that the deposition of
noble metals may improve the adsorption ability of optical semiconductors by promote
an enhancing on their specific surface area [22-24]. Yet, in most cases the surface area of
the metal-loaded materials is increase by change the textural properties of the
semiconductors (e.g., by thermal treatments). In the present study the thermal treatment
used for the reduction of the noble metal phase was maintained at 200 ºC, which did not
interfere with the textural properties of the TiO2, i.e., it wasn’t notice any change on the
SBET of the resulted photocatalysts. This observation could be also explain due to the low
loading of the noble metal added during the photocatalysts preparation.
Fig. 1 shows representative TEM images of the TiO2 based photocatalysts used in this
study (~ 1 wt.% of each noble metal on TiO2 P25) and the particles size distribution of
the noble metal used in this study. The mean of the each noble metal was investigated by
TEM analysis and was based on a count of more than 200 individual nanoparticles. The
characteristic structure of TiO2 P25 particles is shown (Fig. 1a), revealing the presence of
spheroidal TiO2 nanoparticles with an average size near to that of metal particles
(30-50 nm). Regarding the results TiO2 based materials was noticed that the mean of
particle size of the noble metals loaded on TiO2 present a narrow size when compared
with the TiO2 particles size, which difficult their detection. Yet, some spots ranging 20-
25 nm are observed on the TiO2 based materials, which seems to indicate the presence of
the noble metals particles on the surface of the photocatalysts. Additionally, elemental
mapping obtained by EDXS analysis (Fig. SD1) revealed a good dispersion of the noble
metals on the TiO2 surface.
The X-ray powder diffractograms of neat TiO2 P25 and noble metals/TiO2 materials are
shown in Figure 2. The presence of crystalline phases (anatase and rutile) of the TiO2 was
7

confirmed by XRD measurements, in which the typical diffraction lines such as 25.4º and
27.5º, corresponding to the lattice planes of (101) and (110), respectively can be observed.
In addition, comparing the patterns of the neat TiO2 with the synthesized noble
metals/TiO2 (Fig. 2) no diffraction peaks corresponding to the respective noble metal
were detected. This could be attributed to the low amount of noble metals loaded on the
TiO2 surface, which could be beyond the detection limit of the diffractometer.
The surface chemical composition of the noble metal/TiO2 materials was investigated by
deconvolution of the XPS profiles. As displayed in Fig. 3a, the Au/TiO2 material shows
two peaks at binding energy (BE) 87.96 eV and 84.26 eV corresponding to Au 4f5/2 and
Au 4f7/2, respectively, with spin-orbital splitting photoelectrons at 3.70 eV, which is in
agreement with the literature [25, 26]. Regarding the Pt/TiO2 catalyst (Fig. 3b), it was
observed the presence of two peaks in the region of Pt 4f7/2 at 71.56 eV and 72.90 eV,
corresponding to reduced Pt nanoparticles (Pt0) and to Pt2+ ions, respectively [26-28]. The
same evidence was verified at the Pt 4f5/2 region, in which two peaks at 74.81 eV and
75.89 eV were also detected (Pt0 and Pt2+, respectively).
The Pd/TiO2 sample was analyzed before and after (fresh and used samples) the
photocatalytic reduction of BP to BH (Fig. 3c and 3d, respectively). Concerning the fresh
sample, the XPS deconvolution profiles revealed three Pd species in the Pd 3d5/2 region
located at 335.30, 336.08 and 338.18 eV, which are generally assigned to Pd0, palladium
oxide (PdO) and Pd2+, respectively [26, 29]. Moreover, in the Pd 3d3/2 region the peaks
corresponding to the same Pd species (Pd0, PdO and Pd2+) were also observed with BE
of 340.51, 341.30 and 343.19 eV, respectively.
Similar peaks were observed in both sub regions (Pd 3d5/2 and Pd 3d3/2) in the sample
collected after the reutilization studies (Fig. 3d), however with different proportions of
8

the Pd species. From the XPS analyses it was found that the fresh sample only contains
28% of Pd in its reduced form, whereas after the photocatalytic reuse reactions a high
amount of Pd0 species was detected (~68%). This increase in the metallic Pd content may
be ascribed to the photoreduction of palladium species during the photocatalytic
experiments [30]. In fact, it has been reported that Pd2+ species may suffer reduction by
solvated electrons and 2-propanol when exposed to a light source [31], which corroborate
the XPS results obtained for the used photocatalyst sample (Fig. 3d).
Additionally, from DRUV-Vis (Fig. 4a) can be noticed the characteristic plasmon
resonance band spectra of the noble metals such as Pt and Pd, corroborate with the TEM
images.Moreover, the strong absorption band with onset around 400 nm is observed for
TiO2 being attributed to the bandgap transition of the semiconductor. The Pd/TiO2 and
Au/TiO2 catalysts a second absorption band is observed in the visible range at 470 nm
and 576 nm, respectively, which indicates the capacity of these materials to harvest the
light from UV to the visible range. Moreover, in the case of the Pd/TiO2 catalyst the
maximum peak at 470 nm is commonly attributed to d–d transition of Pd2+ [32], which is
in agreement with XPS analysis (Fig. 3c), where these Pd species are well noticed.
The wavelength position of surface plasmon resonance band is commonly related with
the size and shape of noble metals particles, however was noticed from TEM images that
the size and shape of the particles is maintain independently of the noble metal deposited
in TiO2 surface. Additionally, from Tauc plot obtained by diffuse reflectance UV-Vis
spectra (Fig. 4 inset) was observed that a presence of noble metals on TiO2 particles
induce a slight decrease in the bandgap energy of the resulted materials comparing with
the bare TiO2 (3.23 eV). A similar bandgaps were obtained for the TiO2-based catalysts
Au/TiO2 (3.10 eV), Pt/TiO2 (3.08 eV) and Pd/TiO2 (3.05 eV). These results corroborate
9

the previous observations by TEM images, which was observed a good distribution and
similar particle size of noble metals on TiO2 surface.
The materials were characterized by photoluminescence (PL) spectroscopy in order to
explore the effect of noble metals particles on TiO2 particles. The PL spectra at room
temperature of neat TiO2 and TiO2-based materials under an excitation energy fixed a 280
nm (4.43 eV) are shown in Fig. 5.
The direct bandgap found for TiO2 (3.23 eV) was confirmed by PL analysis with the
presence of intense emission both in the UV and visible range. For the TiO2-based
materials the bandgap energy was slight decrease, nevertheless an intense quenching of
the characteristic bands of TiO2 was observed, being more pronounced for Pd/TiO2
material. The extensive quenching of luminescence intensity shows that the noble metals
may act as efficient scavengers of photogenerated electrons produced in TiO2 particles
under UV irradiation. Additionally, the PL spectra quenching indicate a decrease on the
recombination of photogenerated electron-hole pairs [33], which may lead to higher
efficiencies in the photocatalytic reduction process.
3.2 Photo-induced production of benzhydrol
In general, heterogeneous photocatalysis are widely dependent from the reactional
parameters, such as catalyst load, initial substrate concentration, solvent used, presence
of oxidant/reductant agents, irradiation source, among others. Concerning the photo-
induced reduction of benzophenone, the selection of the type of solvent is an important
step that need to be considered, due to the ability of the solvent works as a source of
hydrogen. Moreover, to achieve the desirable products the photo-reduction of
10

benzophenone requires the presence of KOH as a promoter of hydrogen transfer [20].
Thus, the effect of selected operation parameters was evaluated for optimization of the
benzhydrol production from the photocatalytic reduction of benzophenone. In this sense,
to evaluate the photochemical and the heterogeneous photocatalytic regimes for
benzhydrol formation a couple of reactions were performed in the absence of KOH.
During those experiments, 2-propanol was used as a solvent and the initial concentration
of the substrate (benzophenone) was fixed at 1.5 mM. During the heterogeneous
photocatalytic reaction, the amount of catalyst was set as 1.0 g L-1. The results in terms
of photochemical and photocatalytic regimes are displayed in Fig. 6.
Results show that BP is easily converted to produce mainly benzopinacol (BPC) under
photochemical (Fig. 6a, filled symbols). The presence of TiO2 during the photocatalytic
reaction tends to increase the selectivity to benzhydrol formation (Fig. 6b), nevertheless
BPC remains as a main product (Fig. 6a, open symbols). The photochemical reduction of
benzophenone in 2-propanol yielding in benzopinacol and acetone as stable products is
well discussed in literature [21, 34, 35]. Briefly, the mechanism start with the molecule
of benzophenone absorbing energy (photons) reaching the triplet state, which is the
responsible for the intermolecular hydrogen abstraction reaction of photoexcited
benzophenone from 2-propanol. In order to obtain high yields of benzhydrol refuting the
photochemical regime, a study of catalyst amount (0.05 g L-1 – 1.25 g L-1) and
concentration of substrate (0.5 mM – 1.75 mM) was performed (SD2a and SD2b,
respectively). It was found higher selectivity and yield for benzhydrol production when
the reaction were performed with a catalysts amount of 0.1 g L-1, and using an initial
concentration of BP of 1.5 mM, therefore the following experiments were performed
using these concentrations of catalyst and substrate concentration.
11

The effect of using 2-propanol (iPrOH), methanol (MeOH) and ethanol (EtOH) as
solvents under alkaline medium using a selected concentration of KOH (5.0 mM) as a
promoter of hydrogen transfer is depicted in Fig. 7.
The results show 100% of selectivity for BH production after 10 min reaction when
iPrOH was used as solvent (Fig. 7b) in the presence of KOH, which indicate that the
efficiency of the process is highly influenced by the solvent nature and from the presence
of KOH. On the other hand, when MeOH and EtOH were used as solvents the process
showed high selectivity for benzopinacol production, however in small concentrations
when compared with the results obtained from BP photochemical reduction in iPrOH in
the absence of KOH (Fig. 6a). Weiner et al. showed that 2-propanol holds special features
due to effective hydrogen bonding for photochemical reduction of benzophenone,
however the main product obtained from this reaction was benzopinacol. In a similar
study, Waldemar et al. [36] studied the effect of the same alcohol solvents for
benzophenone reduction during laser-jet photolysis. Analogous results were obtained, the
benzophenone photoreduction in 2-propanol was more selective for benzopinacol
production. The same authors describe that effective photoreduction of benzophenone
follows the order ability iPrOH > EtOH > MeOH due to the quenching rate constants of
triplet-excited benzophenone through intermolecular hydrogen transfer, which are
7.5106, 1.9106 and 0.21106 s-1 M-1, respectively [21].
These studies proved that the selection of the alcohol solvents for benzophenone
photochemical reduction is crucial. In addition, during this study it was found that the
presence of homogeneous (KOH) and heterogeneous (TiO2) catalysts led to an increase
of on yield BH production. These results seems to be ascribed at two main factors: i) the
addition of KOH acting as nucleophilic agent (rich in electron species OH¯), which will
12

reacts with the electrophilic species poor in electrons (C=O of the benzophenone), leading
to the protonation of the intermediate alkoxide to give an alcohol derivative (in this case
benzhydrol) and ketones derivatives; ii) the deoxygenated suspension of TiO2 in
benzophenone in iPrOH is followed by simultaneous oxidation and reduction of adsorbed
reactants (Scheme 1).
The photochemical and photocatalytic reduction of benzophenone in 2-propanol show
higher formation of benzopinacol (Fig. 6a), which corroborate with the earlier studies [17,
21]. Thus, the effect of KOH as a hydrogen promoter on the photocatalytic reduction of
benzophenone was studied by varying the KOH concentration from 1.0 – 10 mM (Fig.
8). No further addition of this agent was conducted during the photocatalytic experiment.
For comparison propose the results obtained in absence of KOH (Fig. 6) are also shown.
As expected, the introduction of KOH leads to an increase of the efficiency of the
photocatalytic reduction of BP into BH (Fig. 8). Even for the lowest concentration of
KOH (1.0 mM), a significant enhancement for BH production was observed with respect
to the reaction in the absence of this hydrogen promoter transfer. The optimal
concentration of KOH found was 5 mM, which corresponds to the highest selectivity for
benzhydrol production, and in this case was not detected the formation of the intermediate
benzopinacol.
The present work showed that the photocatalytic regime could be a good candidate for
benzophenone reduction to benzhydrol. Nevertheless, in order to obtain higher yield of
BH is always required the addition of a nucleophilic agent such as KOH.
13

As described previously, the photocatalytic activity of TiO2 can be enhanced by loading
of noble metals nanoparticles, seeking for charge separation efficiency. The introduction
of noble metal nanoparticles in metal oxide semiconductors is considered one of the most
efficient techniques for enhancing their photocatalytic activity. Depending of the
wavelength of irradiation and the type of metal, noble metal nanoparticles may improve
charge separation as a result of localized electromagnetic field, store electrons driving the
Fermi level to more negative potentials, promote electron transfer to adsorbed species
and/or increase the light absorption due to surface plasmon and light-trapping effects [37-
39]. Moreover, several studies have demonstrated that noble metal nanoparticles can be
highly chemoselective for the hydrogenation of molecules containing carbonyl group [15,
40, 41]. Therefore, the study was followed by the loading the Au, Pt and Pd particles into
TiO2. It has also been reported that in photocatalytic hydrogenation reactions using noble
metals are very influenced by the nature and concentration of alcohol, the solution pH
and reaction temperature [42]. Thus, in order to understand the effect of noble metals
loading on TiO2 the following results were conducted in absence of KOH, using iPrOH
as solvent, and with an amount of catalyst of 0.1 g L-1. This catalyst concentration was
selected based on the previously studies on the effect of catalyst amount for BH formation
(Fig. SD2a). In Fig. 9 are depicted the results using noble metals based TiO2
photocatalysts.
In general the results show that highest selectivity for BH production is obtained in the
presence of noble metals, with exception for Au/TiO2 material in which the yield is
similar to the obtained with neat TiO2 (Fig. 9b). Moreover, the results suggest the
existence of different activation mechanisms depending on the type of noble metal used.
14

It is generally accepted that semiconductor loading with noble metals may enhance the
amount and lifetime of photogenerated electron-hole pair avoiding charge recombination
[42, 43]. This effect was observed by photoluminescence analysis were Pd/TiO2 being
the material with higher quenching of the luminescence intensity.
Additionally, the plasmonic effect of noble metals nanoparticles is often as the main
contribute for enhanced photoactivity in semiconductors such TiO2 under visible light
irradiation [44], however this effect was not expected to occur since the irradiation source
used was UVA-LEDs with an emission line at 385 nm. The main reasonable
rationalization of photocatalytic mechanism under UV light assumes the direct
photoexcitation of TiO2, in which photo-excited hot electrons may be transferred via a
Schottky junction from the conduction of TiO2 to noble metal particles deposited on its
surface, while photogenerated holes remains on TiO2, being available for oxidation of
adsorbed reactants (Scheme 1). Moreover, the higher photoelectric work function of Pt
(5.7 eV) compared with Pd (5.2 eV) and Au (5.1 eV) [45] could suggest higher efficiency
for Pt/TiO2 catalyst by facilitating electron transfer from the conduction band of TiO2 to
the noble metals extending the lifetime of charge carriers. Thus, the photocatalytic
reduction of the BP can be ascribed to the presence of an excess amount of electrons at
the noble metal surface. Yet, more photogenerated holes will accumulate at the interface
thus increasing the formation of hydroxyl radicals, which may favor oxidation reactions.
Regarding the photocatalytic results was found that Pd/TiO2 material shows high
efficiency for BH production, being observed an increase of ~ 49% on the concentration
compared with neat TiO2 after 1 h reaction. Although TiO2 based materials show similar
bandgaps, the main driving force involved in the production of BH from BP
photocatalytic reduction may be attributed to the significant depression quenching of the
PL emission spectrum observed for Pd/TiO2 (Fig. 5), suggesting an effective electron
15

transfer from TiO2 and Pd nanoparticles favoring the reduction of BP at catalyst surface.
The benefits of loaded Pd nanoparticles on TiO2 was already demonstrated by Li et al.
[46] in this case for the photocatalytic reduction of CO2 into high-value-added products.
The authors found that palladium nanoparticles on TiO2 not only prevent the electron-
hole recombination but may also promote the adsorption and activation of the substrate
forming active sites for the desired products. Moreover, as described above the noble
metals/TiO2 materials were calcined under an H2 stream. In general, during the
calcination treatment the H2 molecules are firstly adsorbed and dissociated into hydrogen
atoms over the metal surface, and then migrate onto the TiO2 support [28, 47]. This
phenomenon is known as hydrogen spillover, in which the combination of noble metals
and TiO2 may enhance its reducibility, leading to the generation of oxygen vacancies in
the noble metal/TiO2 materials, and thus may promote an enhancement on the activity of
the resulting photocatalysts.
Additionally, the photostability of the Pd/TiO2 was investigated by reusing the catalyst in
consecutive experiments for BH production. Before each reaction cycle, the photocatalyst
was washed with deionized water and dried at 100 ºC. As shown in Fig. 10, the efficiency
of Pd/TiO2, in terms of selectivity for BH formation, decrease (from 28.5% to 20.0%)
after the first run, yet it remains constant after the next cycles. This slightly activity loss
can be ascribed to the decrease of Pd nanoparticles during the first recycling reaction. In
fact, as revealed by XPS analyses (Fig. 3) the relative amounts of palladium species (Pd0,
Pd2+ and PdO) present in the fresh sample and after several reuses changed. Since XPS
analysis is considered a surface quantitative spectroscopic technique, it is difficult to infer
about the amount of each species in the bulk catalyst. From XPS it was found that the
atomic percentage of Pd species before and after reuse decrease from 1.6 at.% to 0.6 at.%,
16

respectively. These results may rationalize the slight decrease on the efficiency of the
Pd/TiO2 after the first run.
Overall, the present work shows that noble metals based TiO2 materials can be used for
the synthesis of benzhydrol by photocatalytic reduction of benzophenone without adding
KOH, however further improvements on the photocatalysts to increase the benzhydrol
yield has to be attempted. An example, is the preparation of bimetallic TiO2 based
catalysts using transition metals which are already been used as efficient materials for
photocatalytic hydrogenation reactions [48].
Conclusions
Heterogeneous photocatalysis was used as efficient technology for benzhydrol
production. The photoreduction of benzophenone for benzhydrol production is mainly
dependent of the solvent and of the presence of a promotor of hydrogen transfer such as
KOH. Using 2-propanol as solvent benzophenone was easily converted into benzopinacol
under UVA-LED irradiation. By using a certain concentration of KOH, benzhydrol was
the main product obtained, achieving 100% selectivity after 10 min reaction.
In general, noble metal deposition on TiO2 particles increased the efficiency of the
photocatalysts. Among all the photocatalysts, the Pd/TiO2 material showed the higher
efficiency for benzhydrol production. The reaction reuses showed slight decrease after
the first run however, the photostability remained in the following cycles. The enhance
activity was ascribed to the action of noble metals acting as electron sinks, thus increasing
the efficiency and lifetime of charge separation in the semiconductor.
The highest quenching observed for Pd/TiO2 material by PL measurements suggests an
effective electron transfer from TiO2 to Pd particles, preventing the recombination of
17

photogenerated electron-hole pairs, and thus improve the photoactivity for benzhydrol
formation.
Declaration of interests
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.
Acknowledgements
This work was financially supported by Project POCI-01-0145-FEDER-030674 funded
by ERDF through COMPETE2020 - Programa Operacional Competitividade e
Internacionalização (POCI) – and by national funds through FCT - Fundação para a
Ciência e a Tecnologia. We would also like to thank the scientific collaboration from the
Associate Laboratory LSRE-LCM - UID/EQU/50020/2019 - funded by national funds
through FCT/MCTES (PIDDAC). M.J.L. and J.L. acknowledge to the Project POCI-01-
0145-FEDER-031398 funded by ERDF through COMPETE2020. C.G.S. acknowledges
the FCT Investigator Programme (IF/00514/2014) with financing from the European
Social Fund (ESF) and the Human Potential Operational Programme.
18

References
[1] M. Lakshmi Kantam, R. Kishore, J. Yadav, S.K. Bhargava, L.A. Jones, A. Venugopal,
Chapter 10 - Hydrogenation for Fine Chemical Synthesis: Status and Future Perspectives,
in: S.S. Joshi, V.V. Ranade (Eds.) Industrial Catalytic Processes for Fine and Specialty
Chemicals, Elsevier, Amsterdam, 2016, pp. 427-462.
[2] Y. Zhang, X. Cui, F. Shi, Y. Deng, Nano-Gold Catalysis in Fine Chemical Synthesis,
Chemical Reviews, 112 (2012) 2467-2505.
[3] S.S. Joshi, A. Bhatnagar, V.V. Ranade, Chapter 8 - Catalysis for Fine and Specialty
Chemicals, in: S.S. Joshi, V.V. Ranade (Eds.) Industrial Catalytic Processes for Fine and
Specialty Chemicals, Elsevier, Amsterdam, 2016, pp. 317-392.
[4] C. Wang, A. Li, J. Xu, J. Wen, H. Zhang, L. Zhang, Preparation of WO3/CNT catalysts
in presence of ionic liquid [C16mim]Cl and catalytic efficiency in oxidative
desulfurization, Journal of Chemical Technology & Biotechnology, 94 (2019) 3403-
3412.
[5] J.J. Douglas, M.J. Sevrin, C.R.J. Stephenson, Visible Light Photocatalysis:
Applications and New Disconnections in the Synthesis of Pharmaceutical Agents,
Organic Process Research & Development, 20 (2016) 1134-1147.
[6] D. Friedmann, A. Hakki, H. Kim, W. Choi, D. Bahnemann, Heterogeneous
photocatalytic organic synthesis: state-of-the-art and future perspectives, Green
Chemistry, 18 (2016) 5391-5411.
[7] C. Michelin, N. Hoffmann, Photocatalysis applied to organic synthesis – A green
chemistry approach, Current Opinion in Green and Sustainable Chemistry, 10 (2018) 40-
45.
[8] B. König, Photocatalysis in Organic Synthesis – Past, Present, and Future, European
Journal of Organic Chemistry, 2017 (2017) 1979-1981.
19

[9] Y. Liu, X. Zeng, X. Hu, J. Hu, X. Zhang, Two-dimensional nanomaterials for
photocatalytic water disinfection: recent progress and future challenges, Journal of
Chemical Technology & Biotechnology, 94 (2019) 22-37.
[10] R.A. Fernandes, M.J. Sampaio, E.S. Da Silva, P. Serp, J.L. Faria, C.G. Silva,
Synthesis of selected aromatic aldehydes under UV-LED irradiation over a hybrid
photocatalyst of carbon nanofibers and zinc oxide, Catalysis Today, 328 (2019) 286-292.
[11] J.C. Lopes, M.J. Sampaio, R.A. Fernandes, M.J. Lima, J.L. Faria, C.G. Silva,
Outstanding response of carbon nitride photocatalysts for selective synthesis of aldehydes
under UV-LED irradiation, Catalysis Today, (2019).
[12] H.H. Mohamed, Biphasic TiO2 microspheres/reduced graphene oxide for effective
simultaneous photocatalytic reduction and oxidation processes, Applied Catalysis A:
General, 541 (2017) 25-34.
[13] F. Bellamy, D. Horton, J. Millet, F. Picart, S. Samreth, J.B. Chazan, Glycosylated
derivatives of benzophenone, benzhydrol and benzhydril as potential venous
antithrombotic agents, Journal of Medicinal Chemistry, 36 (1993) 898-903.
[14] T. Imamoto, N. Takiyama, K. Nakamura, T. Hatajima, Y. Kamiya, Reactions of
carbonyl compounds with Grignard reagents in the presence of cerium chloride, Journal
of the American Chemical Society, 111 (1989) 4392-4398.
[15] C.M. Cirtiu, A. Brisach-Wittmeyer, H. Ménard, Comparative study of catalytic and
electrocatalytic hydrogenation of benzophenone, Catalysis Communications, 8 (2007)
751-754.
[16] S.P. Bawane, S.B. Sawant, Kinetics of Liquid-Phase Catalytic Hydrogenation of
Benzophenone to Benzhydrol, Organic Process Research & Development, 7 (2003) 769-
773.
20

[17] D.I. Schuster, P.B. Karp, Photochemistry of ketones in solution LVIII: Mechanism
of photoreduction of benzophenone by benzhydrol, Journal of Photochemistry, 12 (1980)
333-344.
[18] M. Marazzi, M. Wibowo, H. Gattuso, E. Dumont, D. Roca-Sanjuán, A. Monari,
Hydrogen abstraction by photoexcited benzophenone: consequences for DNA
photosensitization, Physical Chemistry Chemical Physics, 18 (2016) 7829-7836.
[19] D.I. Schuster, M.D. Goldstein, P. Bane, Photochemistry of unsaturated ketones in
solution. 51. Photophysical and photochemical studies of Michler's ketone, Journal of the
American Chemical Society, 99 (1977) 187-193.
[20] C.G. Bochet, Photolabile protecting groups and linkers, Journal of the Chemical
Society, Perkin Transactions 1, (2002) 125-142.
[21] E.A. Escobar, M.Á.V. Zapata, S.A. Hernández, S.O.F. Valle, O.R. Berny, V.J.G.
Ángeles, I.C. Reyes, Photocatalytic Reduction of Benzophenone on TiO2: Effect of
Preparation Method and Reaction Condition, J. Mex. Chem. Soc., 54 (2010) 133-138.
[22] B. Tryba, M. Piszcz, A.W. Morawski, Photocatalytic and self-cleaning properties of
ag-doped tio2, Open Materials Science Journal, 4 (2010) 5-8.
[23] J.W. Fan, X.H. Liu, J. Zhang, The synthesis of TiO2 and TiO2–Pt and their
application in the removal of Cr (VI), Environmental Technology, 32 (2011) 427-437.
[24] H.O. Laguna, J.J. Murcia, H. Rojas, C. Jaramillo-Paez, A.J. Navío, C.M. Hidalgo,
Differences in the Catalytic Behavior of Au-Metalized TiO2 Systems During Phenol
Photo-Degradation and CO Oxidation, Catalysts, 9 (2019).
[25] K. Song, X. Wang, Q. Xiang, J. Xu, Weakened negative effect of Au/TiO2
photocatalytic activity by CdS quantum dots deposited under UV-vis light illumination
at different intensity ratios, Physical Chemistry Chemical Physics, 18 (2016) 29131-
29138.
21

[26] J.F. Moulder, J. Chastain, Handbook of X-ray Photoelectron Spectroscopy: A
Reference Book of Standard Spectra for Identification and Interpretation of XPS Data,
Physical Electronics Division, Perkin-Elmer Corporation, 1992.
[27] L. Yu, Y. Shao, D. Li, Direct combination of hydrogen evolution from water and
methane conversion in a photocatalytic system over Pt/TiO2, Applied Catalysis B:
Environmental, 204 (2017) 216-223.
[28] K. Batalović, N. Bundaleski, J. Radaković, N. Abazović, M. Mitrić, R.A. Silva, M.
Savić, J. Belošević-Čavor, Z. Rakočević, C.M. Rangel, Modification of N-doped TiO2
photocatalysts using noble metals (Pt, Pd) – a combined XPS and DFT study, Physical
Chemistry Chemical Physics, 19 (2017) 7062-7071.
[29] A.L. Luna, D. Dragoe, K. Wang, P. Beaunier, E. Kowalska, B. Ohtani, D. Bahena
Uribe, M.A. Valenzuela, H. Remita, C. Colbeau-Justin, Photocatalytic Hydrogen
Evolution Using Ni–Pd/TiO2: Correlation of Light Absorption, Charge-Carrier
Dynamics, and Quantum Efficiency, The Journal of Physical Chemistry C, 121 (2017)
14302-14311.
[30] F.R. Lemke, R.M. Granger, D.A. Morgenstern, C.P. Kubiak, Photoreduction of
palladium radical cations. Transient absorbance kinetics of electron transfers to
photogenerated [Pd(CNMe)3].cntdot.+ radicals, Journal of the American Chemical
Society, 112 (1990) 4052-4054.
[31] T. Redjala, G. Apostolecu, P. Beaunier, M. Mostafavi, A. Etcheberry, D. Uzio, C.
Thomazeau, H. Remita, Palladium nanostructures synthesized by radiolysis or by
photoreduction, New Journal of Chemistry, 32 (2008) 1403-1408.
[32] Z. Wu, Z. Sheng, H. Wang, Y. Liu, Relationship between Pd oxidation states on
TiO(2) and the photocatalytic oxidation behaviors of nitric oxide, Chemosphere, 77
(2009) 264-268.
22

[33] R. Katal, M.H. Davood Abadi Farahani, H. Jiangyong, Degradation of
acetaminophen in photocatalytic (batch and continuous system) and
a
photoelectrocatalytic process by application of faceted-TiO2, Separation and Purification
Technology, 230 (2020) 115859.
[34] N. Filipescu, F.L. Minn, Photoreduction of benzophenone in isopropyl alcohol,
Journal of the American Chemical Society, 90 (1968) 1544-1547.
[35] J.P. Da Silva, I. Ferreira Machado, J.P. Lourenço, L.F. Vieira Ferreira,
Photochemistry of benzophenone adsorbed on MCM-41 surface, Microporous and
Mesoporous Materials, 84 (2005) 1-10.
[36] W. Adam, U. Kliem, V. Lucchini, Preparative UV-VIS laser photochemistry:
Photocycloadditions of methylenelactones with benzophenone and p-benzoquinone.
Oxygen trapping of paterno-Büchi triplet 1,4-diradicals as model reactions for
quinghaosu-type 1,2,4-trioxanes, Liebigs Annalen der Chemie, 1988 (1988) 869-875.
[37] C. Gomes Silva, R. Juárez, T. Marino, R. Molinari, H. García, Influence of Excitation
Wavelength (UV or Visible Light) on the Photocatalytic Activity of Titania Containing
Gold Nanoparticles for the Generation of Hydrogen or Oxygen from Water, Journal of
the American Chemical Society, 133 (2011) 595-602.
[38] H. Choi, W.T. Chen, P.V. Kamat, Know Thy Nano Neighbor. Plasmonic versus
Electron Charging Effects of Metal Nanoparticles in Dye-Sensitized Solar Cells, ACS
Nano, 6 (2012) 4418-4427.
[39] R. Katal, S.M. Panah, M. Saeedikhani, M. Kosari, C.C. Sheng, O.S. Leong, G. Xiao,
H. Jiangyong, Pd-Decorated CuO Thin Film for Photodegradation of Acetaminophen and
Triclosan under Visible Light Irradiation, Advanced Materials Interfaces, 5 (2018)
1801440.
23

[40] F.-Z. Su, L. He, J. Ni, Y. Cao, H.-Y. He, K.-N. Fan, Efficient and chemoselective
reduction of carbonyl compounds with supported gold catalysts under transfer
hydrogenation conditions, Chemical Communications, (2008) 3531-3533.
[41] U. Rayhan, J.-H. Do, T. Arimura, T. Yamato, Reduction of carbonyl compounds by
Raney Ni–Al alloy and Al powder in the presence of noble metal catalysts in water,
Comptes Rendus Chimie, 18 (2015) 685-692.
[42] G. Palmisano, E. García-López, G. Marcì, V. Loddo, S. Yurdakal, V. Augugliaro, L.
Palmisano, Advances in selective conversions by heterogeneous photocatalysis,
Chemical Communications, 46 (2010) 7074-7089.
[43] S. Masudy-Panah, R. Katal, N.D. Khiavi, E. Shekarian, J. Hu, X. Gong, A high-
performance cupric oxide photocatalyst with palladium light trapping nanostructures and
a hole transporting layer for photoelectrochemical hydrogen evolution, Journal of
Materials Chemistry A, 7 (2019) 22332-22345.
[44] Z. Bian, T. Tachikawa, P. Zhang, M. Fujitsuka, T. Majima, Au/TiO2 Superstructure-
Based Plasmonic Photocatalysts Exhibiting Efficient Charge Separation and
Unprecedented Activity, Journal of the American Chemical Society, 136 (2014) 458-465.
[45] K. Shimura, H. Yoshida, Heterogeneous photocatalytic hydrogen production from
water and biomass derivatives, Energy & Environmental Science, 4 (2011) 2467-2481.
[46] N. Li, M. Liu, B. Yang, W. Shu, Q. Shen, M. Liu, J. Zhou, Enhanced Photocatalytic
Performance toward CO2 Hydrogenation over Nanosized TiO2-Loaded Pd under UV
Irradiation, The Journal of Physical Chemistry C, 121 (2017) 2923-2932.
[47] Y. Zhu, D. Liu, M. Meng, H2 spillover enhanced hydrogenation capability of TiO2
used for photocatalytic splitting of water: a traditional phenomenon for new applications,
Chemical Communications, 50 (2014) 6049-6051.
24

[48] Anna Zielińska-Jurek, Progress, Challenge, and Perspective of Bimetallic TiO2-
Based Photocatalysts, Journal of Nanomaterials, 2014 (2014) 17.
25

25
20
15
10
5
a
b
0
<10
>35
10-15 15-20 20-25 25-30 30-35
dp (nm)
250 nm
250 nm
35
30
25
20
15
10
5
25
20
15
10
5
c
d
0
0
<10
>35
10-15 15-20 20-25 25-30 30-35
<10
>40
10-1515-2020-2525-3030-3535-40
dp (nm)
dp (nm)
250 nm
250 nm
Figure 1 – TEM micrographs of bare TiO2 (a), Au/TiO2 (b), Pt/TiO2 (c) and Pd/TiO2 (d),
and metal nanoparticle size distribution (inset).
26

TiO2
Pd/TiO2
Pt/TiO2
Au/TiO2
20
30
40
50
60
70
2
Figure 2 – XRD diffractograms of neat TiO2 and noble metals/TiO2 materials; anatase
(A) and rutile (R).
27

Figure 3 – XPS patterns of fresh Au/TiO2 (a), Pt/TiO2 (b) and Pd/TiO2 (c) catalysts and
used Pd/TiO2 (d).
TiO2
Pd/TiO2
Au/TiO2
Pt/TiO2
3.0
3.1
3.2
3.3
3.4
3.5
Energy (eV)
300
400
500
600
700
800
Wavelength (nm)
Figure 4 – DRUV-Vis spectra and Tauc plot (inset) of TiO2 based photocatalysts.
TiO2
Au/TiO2
Pt/TiO2
Pd/TiO2
360
380
400
420
440
460
480
500
Wavelength (nm)
Figure 5 – Photoluminescence spectra of neat TiO2 and noble metals/TiO2 materials.
28

100
80
C
S
b)
Y
a)
BH
1.0
0.8
0.6
0.4
0.2
0.0
BPC
60
20
10
0
0
10
20
30
40
50
60
chemical
catalytic
Time (min)
Photoreaction
Figure 6 – (a) Concentration profiles of BH and BCP under photochemical and
photocatalytic regimes (filled and open symbols, respectively); and (b) BP conversion,
selectivity and yield towards BH production after 10 min of reaction.
100
80
60
40
20
0
1.6
1.2
0.8
0.4
0.0
a)
b)
C
S
Y
MeOH
EtOH
iPrOH
MeOH
EtOH
iPrOH
0
10
20
30
40
50
60
Solvent
Time (min)
Figure 7 – (a) Concentration profiles of BH (filled symbols) and BPC (open symbols)
varying the solvent nature; and (b) BP conversion, selectivity and yield towards BH
production after 10 min of reaction.
29