A ruthenium and palladium bimetallic system superior to a rhodium co-catalyst for TiO2-photocatalyzed ring hydrogenation of aniline to cyclohexylamine

Article abstract of DOI:10.1016/j.jcat.2020.05.035

Cyclohexylamine (CHA) is used as an important raw material in chemical industries. Since the production of CHA requires a high temperature and excess amounts of hydrogen gas (H₂) and ammonia, the development of a simple reaction operated under milder conditions using a safe compound is indispensable for achieving sustainable chemistry. A highly selective ring hydrogenation of aniline (AN) to CHA over a titanium(IV) oxide photocatalyst having a co-catalyst in water at room temperature, in which oxalic acid was used as the hydrogen source in place of H₂, is described in this paper. Among TiO₂ photocatalysts having co-catalysts, Rh-TiO₂ showed high activity for CHA production, while Pd-TiO₂ and Ru-TiO₂ showed negligible activity. When a bimetallic photocatalyst, Ru/Pd-TiO₂, was used, the yield was larger than that when Rh-TiO₂ was used. Metal particles loaded on TiO₂ were observed by using TEM, and electronic states of Ru and Pd were evaluated by using XPS. Interactions among metal, AN and CHA were investigated through adsorption experiments. Based on the results, active sites for AN hydrogenation are discussed.

Full text of DOI:10.1016/j.jcat.2020.05.035

Journal Pre-proofs
A ruthenium and palladium bimetallic system superior to a rhodium co-cata-
lyst for TiO2-photocatalyzed ring hydrogenation of aniline to cyclohexyla-
mine
Sakae Araki, Kousuke Nakanishi, Atsuhiro Tanaka, Hiroshi Kominami
PII:
S0021-9517(20)30216-5
https://doi.org/10.1016/j.jcat.2020.05.035
YJCAT 13765
DOI:
Reference:
To appear in:
Journal of Catalysis
Received Date:
Revised Date:
Accepted Date:
7 March 2020
22 May 2020
22 May 2020
Please cite this article as: S. Araki, K. Nakanishi, A. Tanaka, H. Kominami, A ruthenium and palladium
bimetallic system superior to a rhodium co-catalyst for TiO2-photocatalyzed ring hydrogenation of aniline to
cyclohexylamine, Journal of Catalysis (2020), doi: https://doi.org/10.1016/j.jcat.2020.05.035
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.
© 2020 Elsevier Inc. All rights reserved.

1/31
A ruthenium and palladium bimetallic system superior to a rhodium co-
catalyst for TiO2-photocatalyzed ring hydrogenation of aniline to
cyclohexylamine
Sakae Araki a, Kousuke Nakanishi a, Atsuhiro Tanaka b, c, Hiroshi Kominami b,*
a Molecular and Material Engineering, Interdisciplinary Graduate School of Science and
Engineering, Kindai University, 3-4-1, Kowakae, Higashiosaka, Osaka 577-8502, Japan
b
Department of Applied Chemistry, Faculty of Science and Engineering, Kindai
University, 3-4-1 Kowakae, Higashiosaka, Osaka 577-8502, Japan.
c Precursory Research for Embryonic Science and Technology (PRESTO), Japan
Science and Technology Agency (JST), 4-1-8 Honcho, Kawaguchi 332-0012, Japan
E-mail: hiro@apch.kindai.ac.jp (H. Kominami)
Abstract
Cyclohexylamine (CHA) is used as an important raw material in chemical
industries.
Since the production of CHA requires a high temperature and excess
amounts of hydrogen gas (H2) and ammonia, the development of a simple reaction
operated under milder conditions using a safe compound is indispensable for achieving
sustainable chemistry.
A highly selective ring hydrogenation of aniline (AN) to CHA
over a titanium(IV) oxide photocatalyst having a co-catalyst in water at room temperature,
in which oxalic acid was used as the hydrogen source in place of H2, is described in this
paper. Among TiO2 photocatalysts having co-catalysts, Rh-TiO2 showed high activity
for CHA production, while Pd-TiO2 and Ru-TiO2 showed negligible activity.
When a
bimetallic photocatalyst, Ru/Pd-TiO2, was used, the yield was larger than that when Rh-

2/31
TiO2 was used.
Metal particles loaded on TiO2 were observed by using TEM, and
electronic states of Ru and Pd were evaluated by using XPS.
Interactions among metal,
Based on the results,
AN and CHA were investigated through adsorption experiments.
active sites for AN hydrogenation are discussed.
Keywords: Photocatalyst, TiO2, Hydrogenation, Aniline, Co-catalyst, Palladium,
Ruthenium
1. Introduction
Cyclohexylamine (CHA) is used as a raw material for metal corrosion inhibitors,
rubber vulcanization accelerators, synthetic dyes, artificial sweeteners, plasticizers,
surfactants and agricultural chemicals [1,2].
by catalytic hydrogenation of aniline (AN) or amine alkylation of cyclohexanol with
ammonia (NH3) [3-5]. However, these production methods require (1) high
Cyclohexylamine is industrially produced
temperature and high pressure conditions, (2) excess amounts of hydrogen (H2) gas and
(3) a large amount of NH3 to prevent side reactions. From the viewpoint of sustainable
chemistry, it is necessary to develop a reaction system for hydrogenation of AN to CHA
under mild conditions.
A photocatalytic reaction can use solar light as an energy source for driving the
reaction, which is consistent with the concept of sustainable chemistry.
oxide (TiO2) is one of the most widely used photocatalysts. When TiO2 is irradiated
with ultraviolet light, electrons (e-) are excited from the valence band (VB) to the
conduction band (CB) and holes (h+) are generated in the VB.
These cause reduction
and oxidation reactions, respectively. When metal nanoparticles are supported on TiO2,
they function as a co-catalyst, not only suppressing recombination of e- and h+ but also
exhibiting specific reduction activity for various compounds. We have reported
various photocatalytic reduction reactions over TiO2 modified with metals such as silver,
Titanium(IV)

3/31
copper, platinum, rhodium and palladium [6-10].
Among the various reduction
reactions, ring hydrogenation of aromatic compounds is attractive research field in recent
years from the viewpoint of hydrogen storage as well as utilization for organic synthesis
materials [11-13].
It is known that a bimetal catalyst having two different elements exhibits an
activity in a (thermo)catalytic reaction that is not achieved in a single metal system [14-
16].
photocatalytic reaction system.
In this study, we examined the effect of ruthenium (Ru) as a second element on
the catalytic performance of metal-loaded TiO2 (M-TiO2) in ring hydrogenation of AN to
CHA. Here we report 1) preparation and characterization of M-TiO2 and Ru/M-TiO2,
Therefore, it is expected that the same effect can also be obtained in a
2) photocatalytic performance of M-TiO2 and Ru/M-TiO2, 3) effect of the reaction
conditions on ring hydrogenation and 4) results of physicochemical studies on ring
hydrogenation.
2. Experimental
2.1 Preparation of M-TiO2
Loading various metals (Au, Ag, Cu, Pt, Rh, Pd and Ru) as co-catalysts on TiO2
(P25, Degussa) was performed by the photodeposition method.
TiO2 powder was
suspended in an aqueous methanol (MeOH) solution (50 vol%) containing a X mol% of
the metal source (HAuCl4, AgNO3, CuCl2, H2PtCl6, PdCl2, RhCl3 and RuCl3) in a test
tube.
photoirradiated for 2 h at λ > 300 nm by a 400 W high-pressure mercury arc (Eiko-sha,
Osaka, Japan) with magnetic stirring in a water bath continuously kept at 293 K. The
The test tube was sealed with a rubber septum under argon (Ar) and then
resulting powder was washed repeatedly with distilled water and dried for 2 h in vacuo.
The obtained sample is hereafter designated as M(X)-TiO2.

4/31
2.2 Introduction of Ru to M-TiO2 (Ru/M-TiO2)
A two-step photodeposition method [17-20] was used for the preparation of
Ru/M-TiO2.
The prepared M(X)/TiO2 powder was suspended in an aqueous isopropyl
alcohol (IPA) solution (50 vol%) containing a Y mol% of RuCl3 and hydrochloric acid
(HCl) in a test tube.
photoirradiated for 6 h under the same condition.
the aqueous IPA solution was smaller than that of photodeposition in the aqueous
methanol solution, time of photoirradiation was prolonged to 6 h. The small
photodeposition rate may be the reason for success in the selective formation of Ru layer
on M-TiO2. The resulting powder was washed repeatedly with distilled water and
The test tube was sealed with a rubber septum under Ar and then
Since the rate of photodeposition in
dried for 2 h in vacuo.
The obtained bimetallic sample is hereafter designated as
Ru(Y)/M(X)-TiO2.
2.3 Characterization of Pd-TiO2, Ru-TiO2 and Ru/Pd-TiO2
Morphology of the photocatalysts was observed with a transmission electron
microscope (TEM, JEM-2100F, JEOL, Tokyo) operated at 200 kV in the Joint Research
Center at Kindai University (JRC).
X-ray photoelectron spectroscopy (XPS) was
The spectrum
acquired using an AXIS-NOVA ESCA (Shimadzu, Kyoto) in the JRC.
was calibrated by the peak that can be assigned to C 1s and was normalized.
2.4 Photocatalytic ring hydrogenation of AN
In a typical run, M-TiO2 (or Ru/M-TiO2, 50 mg) was suspended in 5 cm3 of water
containing AN (ca. 50 μmol) and oxalic acid (OA) used as a hole scavenger in a test tube,
which was sealed with a rubber septum under Ar and then photoirradiated at λ > 300 nm
with a 400 W high-pressure mercury arc in a water bath with stirring at a temperature of

5/31
293 K.
The amounts of AN and products (CHA and cyclohexanone (CHON) were
only produced under the conditions.) were determined with an FID-type gas
chromatograph (GC-2025, Shimadzu) equipped with a DB-1 column. Toluene was
used as an internal standard sample. The reaction solution (2 cm3) was added to diethyl
ether (2 cm3) and aqueous sodium hydroxide (NaOH) solution (1 cm3).
After the
mixture had been stirred for 10 min, AN, CHA and CHON in the ether layer were
analyzed. The amounts of AN, CHA and CHON were determined from the ratios of
their peak areas to the peak area of toluene. The amounts of H2 as the reduction product
of protons (H+) and carbon dioxide (CO2) as the oxidation product of OA were determined
using two TCD-type gas chromatographs (GC-8A, Shimadzu) equipped with MS-5A and
Porapac Q-S columns, respectively.
2.5 Adsorption experiment
M-TiO2 (or Ru/M-TiO2, 100 mg) was suspended in 5 cm3 of water containing
AN or CHA (ca. 50 μmol) and OA (ca. 900 μmol) in a test tube, which was sealed with
a rubber septum under Ar and then the mixture was stirred at a temperature of 293 K for
20 hours in a dark place.
The amounts of AN and CHA were determined in the same
way as that for the photocatalytic reaction (section 2.4).
3. Results and Discussion
3.1 Effect of metal co-catalyst on ring hydrogenation of AN
Figure 1 shows the results of photocatalytic reactions of AN in aqueous
suspensions of M(3.5)-TiO2 for 30 min. When bare TiO2 was used as the photocatalyst,
CHA and CHON were not formed and the color of TiO2 changed from white to blue just
after photoirradiation, indicating that the tetravalent titanium in TiO2 was reduced to
trivalent titanium by photogenerated electrons.
When Au, Ag, Cu, and Ru were used

6/31
as co-catalysts, CHA and CHON were not formed probably due to their large activation
energies for the reactions.
small amount of CHA.
On the other hand, Pd(3.5)-TiO2 produced CHON with a
This result means that hydrogenation of the benzene ring and
elimination of ammonia occurred over the Pd co-catalyst.
will be discussed later. In contrast to Pd(3.5)-TiO2, Rh(3.5)-TiO2 and Pt(3.5)-TiO2
produced a large amount of CHA with suppressed formation of CHON. These results
The reaction mechanism
indicate that selectivity in the photocatalytic reaction of AN is dependent of the type of
co-catalyst.
Figure 2 shows the effect of introduction of Ru (1.0 mol%) as the second element
to M(3.5)-TiO2 on yields of CHA and CHON in the photocatalytic reaction of AN.
The
yields of CHA and CHON over Ru(1.0)/Pt(3.5)-TiO2 and Ru(1.0)/Rh(3.5)-TiO2
decreased, indicating that Ru had a negative effect on Rh(3.5)-TiO2 and Pt(3.5)-TiO2.
The decrease in the catalytic performance suggests that Pt and Rh particles were covered
with Ru.
CHA yield and suppressed the CHON yield, even though Ru-TiO2 and Pd-TiO2 showed
negligible activity for CHA production. Figures 1 and 2 show that Ru/Pd-TiO2 was
Very interestingly, introduction of Ru to Pd(3.5)-TiO2 greatly increased the
the most selective photocatalyst in ring hydrogenation of AN to CHA.
3.2 Photocatalytic ring hydrogenation of AN over Ru/Pd-TiO2
Figure 3 shows the effects of Pd and Ru loadings on the yields of CHA and
CHON in the photocatalytic reaction of AN.
volcano-like tendency against both Ru and Pd contents, and the highest yield of CHA was
obtained at 3.5 mol% Pd and 1.0 mol% Ru. Figure 4(a) shows the time course of
photocatalytic ring hydrogenation of AN over Ru(1.0)/Pd(3.5)-TiO2. Along with
Ru/Pd-TiO2 photocatalysts showed a
phothoirradiation, CHA was predominantly produced with a small amount of CHON.
When AN was consumed almost completely after photoirradiation for 90 min, CHA was

7/31
obtained with a high selectivity of 90%.
Figure 4(a) also shows the amount of CHON
and material balance (MB) calculated from the amounts of CHA, CHON and AN by using
Equation (1),
n(CHA) + n(CHON) + n(AN)
.
(1)
MB =
(
)
n0 AN
MB decreased in an early stage before 30 min due to the formation of a small amount
CHON and was almost unchanged after that, indicating that no other intermediate(s) and
by-product(s) were formed.
H2) simultaneously occurred as the electron-consuming reaction.
reaction, CO2 was produced by oxidation of OA ((COOH)2 + 2(e-—h+) → 2CO2 + 2e- +
2H+).
The pressure inside the test tube reached 1.48 atm at 90 min due to the
production of 450 µmol of CO2. According to the Henry’s law, the total amount of
CO2 produced was roughly estimated to be 523 µmol and the value means that 529 µmol
of electrons were provided from OA. Using this value, redox balance was calculated
to be 0.99. For comparison, the time course of photocatalytic ring hydrogenation of
AN over Ru-free Pd(3.5)-TiO2 is shown in Figure S1. In contrast to the result of
As shown in Figure 4(b), evolution of H2 (2H+ + 2e- →
As the hole-trapping
Ru(1.0)/Pd(3.5)-TiO2, a large amount of CHON was produced over Pd(3.5)-TiO2,
resulting in the small value of CHA/H2.
3.3 Characterization of Pd-TiO2 and Ru/Pd-TiO2
A TEM image of Pd(3.5)/TiO2 is shown in Figure 5(a), indicating that Pd
nanoparticles were highly dispersed on TiO2.
(Figure 5(b)), the average diameter of the Pd nanoparticles was determined to be 5.0 nm.
A TEM image of Ru(1.0)/Pd(3.5)-TiO2 is shown in Figure 5(c). A large change was
not observed in the image; however, the determined average diameter of the particles (6.6
nm) (Figure 5(d)) was slightly larger than that of the original Pd nanoparticles. Since
introduction of Ru to Pd-TiO2 by the photodeposition method was performed at room
From the particle size distribution

8/31
temperature, the possibility of an increase in the size of the original Pd particles can be
ruled out.
Ru(1.0)/Pd(3.5)-TiO2 suggests that Pd nanoparticles were covered with an Ru layer with
a thickness of 0.8 nm (Figure S2). However, there is a possibility that a part of the Ru
Therefore, the increased average diameter of the particles in
element was not deposited on the Pd particles but on the TiO2 surface.
Figure 6(a) shows Pd3d5/2 and Pd3d3/2 XPS spectra of Pd(3.5)-TiO2 and
Ru(1.0)/Pd(3.5)-TiO2.
Metallic Pd (Pd0) and electron-deficient Pd species (Pd2+ and
Pd4+) were observed in Pd(3.5)-TiO2, indicating that Pd atoms at the outer surface of the
particles that had formed on TiO2 by using the photodeposition method were partially re-
oxidized by exposure to air.
decreased, suggesting that the surface of Pd particles was covered with Ru introduced to
Pd(3.5)-TiO2. Fitting parameters of the Pd3d5/2 spectra are summarized in Table 1.
By introduction of Ru to Pd(3.5)-TiO2, the proportion of metallic Pd (Pd0) decreased and
the proportions of electron-deficient Pd species (Pd2+ and Pd4+) increased.
This result
suggests that Ru and Pd interact with each other and that Pd at the interface is in an
oxidized state. Figure 6(b) shows an Ru3d5/2 XPS spectrum of Ru(1.0)/Pd(3.5)-TiO2
The intensity of the peaks of Ru(1.0)/Pd(3.5)-TiO2
together with a spectrum of Ru(1.0)/TiO2 prepared by the photodeposition method.
Most of the Ru species in Ru(1.0)/TiO2 was in an electron-deficient state (Ru4+),
indicating that Ru particles that had formed on TiO2 by using the photodeposition method
were easily re-oxidized by exposure to air.
Interestingly, a part of Ru in
Ru(1.0)/Pd(3.5)-TiO2 remained in the metallic state (Ru0), although Ru(1.0)/Pd(3.5)-TiO2
was stored under air as well as Ru(1.0)/TiO2.
As shown in Table 1, ca. 30% of Ru was
These results indicate that an inner
in the metallic state (Ru0) in the presence of Pd.
part of the Ru species in Ru(1.0)/Pd(3.5)-TiO2 was stabilized in the metallic state due to
the presence of Pd (Figure S2). From these results, it can be concluded that the metallic
Ru species work as active sites for ring hydrogenation reaction from AN to CHA over an

9/31
Ru(1.0)/Pd(3.5)-TiO2 photocatalyst.
However, since the increase in the proportion of
metallic Ru cannot be explained by the reduction potentials of RuO2 and PdO (0.68 V vs
SHE and 0.79 V vs SHE, respectively), some special interaction between Ru and Pd at
the interface may stabilize metallic Ru atoms, although we have no clear information on
this at present.
3.4 Effects of reaction conditions on ring hydrogenation of AN
Table 2 shows the effects of various reaction conditions on the ring
hydrogenation of AN in aqueous suspensions of metal-loaded TiO2 photocatalysts.
Among monometallic photocatalysts, Pt(3.5)/TiO2 and Rh(3.5)/TiO2 showed high CHA
selectivities (51% and 80%, respectively) (Entries 1 and 2).
Ru(1.0)/Pd(3.5)-TiO2 exhibited a higher selectivity (90%) than those of Pt(3.5)/TiO2 and
Rh(3.5)/TiO2 (Entry 3). The suppressed CHON production over Ru(1.0)/Pd(3.5)-TiO2
As shown previously,
was in contrast to the production over Pt(3.5)/TiO2 and Rh(3.5)/TiO2.
When the amount of OA was reduced from 900 μmol to 300 μmol, the CHA
selectivity decreased to 80% and the CHON selectivity increased to 14% (Entry 4, Figure
S3).
were investigated (Figure S4).
acidic (ca. pH 1) during the entire reaction (Figure S4(a)) because OA still remained in
the suspension. On the other hand, the suspension in the presence of 300 μmol changed
to neutral at about 90 min due to the consumption of OA (Figure S4(b)). These results
To understand the difference in selectivity, the values of pH of the reaction media
The suspension in the presence of 900 μmol of OA was
suggest that an acidic condition is important for ring hydrogenation of AN to CHA.
When the reaction in the presence of 300 μmol of OA was carried out in 0.1 mol dm-3
HCl solution, the CHA selectivity was improved together with suppression of CHON
selectivity (Entry 5).
It was clarified that the reaction system in which the reaction
suspension was kept acidic (Entries 3 and 5) showed higher CHA selectivity than that in

10/31
the system in which the reaction solution was changed to neutral (Entry 4).
Under an
acidic condition, AN exists as anilinium ions having a lower electron density of aromatic
rings and being more susceptible to ring hydrogenation during the reaction.
mechanisms for the formation of CHA and CHON will be discussed later.
Reaction
Three blank tests, 1) in the presence of Ru/Pd-TiO2 in the dark (thermocatalytic
reaction over Ru/Pd-TiO2, Entry 6), 2) in the absence of Ru/Pd-TiO2 (photochemical
reaction between AN and OA, Entry 7) and 3) in the absence of Ru/Pd-TiO2 in the dark
(chemical reaction between AN and OA, Entry 8), were examined.
indicated that Ru/Pd-TiO2 and photoirradiation are necessary for this reaction, in other
words, ring hydrogenation is a photocatalytic reaction. Thermocatalytic reactions in
the presence of H2 gas in the dark were also examined (Entries 9 and 10). Ring
The results
hydrogenation of AN in the presence of ca. 220 µmol H2 was negligible, and even under
1 atm H2 (ca. 1300 μmol), the AN yield was less than half of the yield of the photocatalytic
reaction (Entry 4).
These results show that the photocatalytic process using Ru/Pd-
TiO2 is an efficient method for ring hydrogenation of AN.
Ring hydrogenation was performed in alcohols, with the alcohols working as
hole scavengers as well as solvents for the photocatalytic reactions (Entries 11-13).
However, the reaction hardly proceeded in any of the alcohols for the following three
reasons: (1) a decrease in the amount of AN adsorbed, (2) the absence of anilinium ions
under the OA-free condition, and (3) the reaction between AN (or CHA) and carbonyl
compounds produced from alcohols by hole scavenging.
Photocatalytic reaction of
This result provides
CHA in water in the presence of OA gave no product (Entry 14).
important information on the reaction mechanism, i.e., CHON was not produced from
CHA but from AN.
3.5 Effect of reaction temperature

11/31
To investigate the effect of reaction temperature, photocatalytic ring
hydrogenation of AN over Ru/Pd-TiO2 was performed in a water bath kept at various
temperatures under irradiation of light from a UV-LED (UVL-365A, San-ei Electric, λmax
= 366 nm, 30 mW cm-2).
temperature (298-328 K) with photoirradiation time, indicating zero-order kinetics
(Figure S5). Figure 7 shows Arrhenius plots (logarithm of k vs reciprocal of T) for ring
hydrogenation of AN to CHA and CHON over Ru/Pd-TiO2. A linear correlation was
obtained in each reaction. The values of apparent activation energy (Ea) and pre-
exponential factor (A) calculated from the slope and intercept of the Arrhenius plot are
summarized in Table 3. The value of Ea for ring hydrogenation from AN to CHA was
determined to be 17 kJ mol-1, which was much smaller than the value of Ea for ring
hydrogenation from AN to CHON (44 kJ mol-1).
The value of A for ring hydrogenation
reaction from AN to CHA (2.4 × 102 μmol min-1) was much smaller than the value of A
for ring hydrogenation from AN to CHON (8.0 × 106 μmol min-1).
The production of
The amounts of CHA and CHON increased linearly at each
CHON having a large value of A was greatly suppressed at a lower temperature due to
the large value of Ea, resulting in highly selective formation of CHA in ring hydrogenation
of AN.
3.6 Adsorption experiments
Figure 8 shows the amounts of AN and CHA adsorbed onto TiO2, Pd(3.5)-TiO2,
Ru(1.0)-TiO2 and Ru(1.0)/Pd(3.5)-TiO2.
predominant compared with adsorption of AN.
TiO2 and Ru(1.0)-TiO2 revealed that Pd strongly interacts with AN, while Ru does not
show a strong interaction with AN. The adsorption property of Ru(1.0)/Pd(3.5)-TiO2
was completely different from that of Pd(3.5)-TiO2 and very similar to that of Ru(1.0)-
TiO2. These adsorption properties of Pd(3.5)-TiO2, Ru(1.0)-TiO2 and Ru(1.0)/Pd(3.5)-
In the case of TiO2, adsorption of CHA is
Adsorption experiments of Pd(3.5)-

12/31
TiO2 suggest that the surface of Pd particles is covered with Ru.
The smaller amount
of CHA adsorption than that of AN adsorption contributes to the smooth adsorption of
AN and desorption of CHA in the photocatalytic reaction, resulting in efficient
conversion of AN to CHA.
3.7 Expected reaction mechanism
Figure 9 shows an expected reaction mechanism of photocatalytic ring
hydrogenation of AN to CHA over Ru/Pd-TiO2.
TEM, XPS and AN adsorption support the idea that the surface of Pd particles is covered
with Ru and that metallic Ru works as active sites. Under UV light irradiation, OA is
oxidized by h+ in the VB of TiO2 and, as a result, H+ and CO2 are produced.
Over
Results of photocatalytic reaction,
Ru/Pd particles, reduction of H+ by e- occurs and active hydrogen species are formed.
The aromatic ring of AN is hydrogenated by six active hydrogen species, resulting in the
formation of CHA.
Two reactions from AN to CHA (main reaction) and from AN to CHON (side
reaction) are shown in Scheme 1.
protonated to anilinium ions (2), which are converted to a compound (3) by six active
hydrogen species. The compound (3) is de-protonated to CHA (4) under a basic
condition of the extraction operation. When free AN is hydrogenated with active
Under an acidic condition, most of the AN (1) is
hydrogen species, 1-cyclohexene-1-amine (5) probably forms as an intermediate, which
is converted to cyclohexanimine (6) by imine-enamine tautomerism.
Cyclohexanimine reacts with water and releases NH3, which results in the formation of
CHON (7).
To obtain CHA with a high selectivity, an acidic condition is necessary as
shown in Figure S4 and Table 2.
4. Conclusions

13/31
By using the photodeposition method, M-TiO2 and Ru/M-TiO2 were prepared
and used for photocatalytic ring hydrogenation of AN to CHA in an aqueous solution in
the presence of OA.
suppressed formation of CHON, while Pd-TiO2 produced CHON with a small amount of
CHA. Introduction of Ru to Pd-TiO2 greatly increased the CHA yield and suppressed
the CHON yield, and Ru/Pd-TiO2 showed the best performance (90% CHA selectivity at
>99% AN conversion) among the photocatalysts prepared in this study. In this reaction
system, OA works as a hole scavenger (electron and hydrogen sources) and as an acid in
order to keep the reaction solution acidic. Results of photocatalytic reaction, TEM,
XPS and AN adsorption support the idea that the surface of Pd particles is covered with
Ru and that metallic Ru works as active sites for ring hydrogenation of AN. In an
acidic condition, the amino group was protonated and preserved from elimination via
imine-enamine tautomerism. The side reaction (production of CHON) was also
Rh-TiO2 and Pt-TiO2 produced a large amount of CHA with
suppressed at a low temperature due to the large Ea over Ru/Pd-TiO2 (44 kJ mol-1)
compared with Ea (17 kJ mol-1) for the CHA formation.
Acknowledgements
This work was partially supported by Grant-in-Aid for Scientific Research (No.
17H03462 and 17H04967) from the Japan Society for the Promotion of Science (JSPS).
References
[1] M. Chatterjee, M. Sato, H. Kawanami, T. Ishizaka, T. Yokoyama, T. Suzuki., Appl.
Catal. A: Gen. 396 (2011) 186–193.
[2] F. Leng, I. C. Gerber, P. Lecante, S. Moldovan, M. Girleanu, M. R. Axet. P. Serp,
ACS Catal. 6 (2016) 6018–6024.
[3] H. L. Merten, G. R. Wilder, US Patent 4,384,142 (1983) to Monsanto Co. Ltd.

14/31
[4] T. Kiyuma, I. Naramoto, US Patent 4,914,239 (1990) to New Japan Chemical Co.
Ltd.
[5] L. Wambach, M. Irgang, DE Patent 4,207,314 (1993) to BASF AG.
[6] H. Kominami, S. Yamamoto, K. Imamura, A. Tanaka, K. Hashimoto, Chem.
Commun. 50 (2014) 4558–4560.
[7] K. Nakanishi, A. Tanaka, K. Hashimoto, H. Kominami, Phys. Chem. Chem. Phys.
19 (2017) 20206–20212.
[8] K. Nakanishi, R. Yagi, K. Imamura, A. Tanaka, K. Hashimoto, H. Kominami, Catal.
Sci. Technol. 8 (2018) 139–146.
[9] H. Kominami, M. Shiba, A. Hashimoto, S. Imai, K. Nakanishi, A. Tanaka, K.
Hashimoto, K. Imamura, Phys. Chem. Chem. Phys. 20 (2018) 19321–19325.
[10] A. Kinoshita, K. Nakanishi, R. Yagi, A. Tanaka, K. Hashimoto, H. Kominami, Appl.
Catal. A: Gen. 578 (2019) 83–88.
[11] K. M. Eblagon, S. C. E. Tsang, Appl. Catal. B: Environ. 160–161 (2014) 22–34.
[12] K. Matsuoka, K. Miyoshi, Y. Sato, J. Power Sources 343 (2017) 156–160.
[13] L. Shi, S. Qi, J. Qu, T Che, C Yi, B Yang, Science Direct 44 (2019) 5345–5354.
[14] K. Kusada, H. Kobayashi, R. Ikeda, Y. Kubota, M. Takata, S. Toh, T. Yamamoto,
S. Matsumura, N. Sumi, K. Sato, K. Nagaoka, H. Kitagawa, J. Am. Chem. Soc. 136
(2014) 1864–1871.
[15] H. Miura, K. Endo, R. Ogawa, T. Shishido, ACS Catal. 7 (2017) 1543–1553.
[16] T. Mitsudome, K. Miyagawa, Z. Maeno, T. Mizugaki, K. Jitsukawa, J. Yamasaki,
Y. Kitagawa, K. Kaneda, Angew. Chem. Int. Ed. 56 (2017) 9381–9385.
[17] A. Tanaka, K. Hashimoto, H. Kominami, ChemCatChem 3 (2011) 1619–1623.
[18] A. Tanaka, S. Sakaguchi, K. Hashimoto, H. Kominami, Catal. Sci. Technol. 2 (2012)
907–909.
[19] A. Tanaka, K. Hashimoto, H. Kominami, J. Am. Chem. Soc. 134 (2012) 14526–

15/31
14533.
[20] A. Tanaka, K. Fuku, T. Nishi, K. Hashimoto, H. Kominami, J. Phys. Chem. C 117
(2013) 16983–16989.

16/31
Figure 1 Photocatalytic ring hydrogenation of AN in an aqueous suspension of M(3.5)-
TiO2 in the presence of OA (900 µmol) for 30 min.

17/31
Figure 2 Photocatalytic ring hydrogenation of AN in an aqueous suspension of
Ru(1.0)/M(3.5)-TiO2 in the presence of OA (900 µmol) for 30 min.

18/31
Figure 3 Effect of (a) Pd loading (X) and (b) Ru loading (Y) on ring hydrogenation of
AN in an aqueous suspension of Ru(Y)/Pd(X)-TiO2 in the presence of OA (300 μmol) for
30 min.

19/31
Figure 4 Time courses of the amounts of aniline (●), CHA (◆), cyclohexanone (■),
H2 (◢), CO2 (◣), material balance (▲), and CHA/H2 (▼) in an aqueous suspension of
Ru(1.0)/Pd(3.5)-TiO2 in the presence of OA (900 µmol).

20/31
Figure 5 TEM images and particle size distributions of Pd(3.5)-TiO2 (a, b) and
Ru(1.0)/Pd(3.5)-TiO2 (c, d).

21/31
Figure 6 XPS spectra of Pd 3d of Pd(3.5)-TiO2 and Ru(1.0)/Pd(3.5)-TiO2 (a) and Ru
3d of Ru(1.0)-TiO2 and Ru(1.0)/Pd(3.5)-TiO2 (b).

22/31
Table 1 Fitting parameters of Ru 3d and Pd 3d spectra.
Intensity ratio
B.E.a (eV)
Species
Pd(3.5)-TiO2 Ru(1.0)-TiO2
Ru(1.0)/Pd(3.5)-TiO2
Ru0
280
―
―
4.5
30.8
69.2
Ru4+
280.8
335
95.5
Pd0
64.5
25.2
10.3
―
―
―
40.4
46.3
13.3
340.2
335.9
341.2
Pd2+
337.6
342.6
Pd4+
a Binding energy

23/31
Table 2 Results of ring hydrogenation of AN to CHA and CHON under various
conditions at 293 K.
Time /
min
Entry
Substrate
Hole scavenger
solvent
Co-catalyst
UV light
Gas phase
Yield / mol
M.B.
NH2
O
Type
Amount / mol
1
2
AN
AN
AN
AN
AN
AN
AN
AN
AN
AN
AN
AN
AN
CHA
OA
OA
900
900
900
300
300
300
300
300
300
300
―
H2O
H2O
Pd(3.5)
On
On
90
90
90
90
90
90
90
90
90
90
90
90
90
90
Ar
Ar
24.9
39.6
44.1
39.5
41.5
n. d.b
n. d.
n. d.
1.1
7.6
6.0
0.66
0.92
0.92
0.94
0.89
―
Rh(3.5)
3
OA
H2O
Ru(1.0)/Pd(3.5)
Ru(1.0)/Pd(3.5)
Ru(1.0)/Pd(3.5)
Ru(1.0)/Pd(3.5)
―
On
Ar
1.8
4
OA
H2O
0.1M HCl aq.a
On
Ar
6.1
5
OA
On
Ar
2.6
6
OA
H2O
(Dark)
On
Ar
n. d.
n. d.
n. d.
2.8
7
OA
H2O
Ar
―
8
OA
H2O
―
(Dark)
(Dark)
(Dark)
On
Ar
H2/Arc
―
9
OA
H2O
Ru(1.0)/Pd(3.5)
Ru(1.0)/Pd(3.5)
Ru(1.0)/Pd(3.5)
Ru(1.0)/Pd(3.5)
Ru(1.0)/Pd(3.5)
Ru(1.0)/Pd(3.5)
0.91
0.87
―
d
10
11
12
13
14
OA
H2O
H2
16.9
n. d.
n. d.
1.2
5.3
Methanol
Ethanol
2-Propanol
OA
Methanol
Ethanol
2-Propanol
H2O
Ar
Ar
Ar
Ar
n. d.
n. d.
n. d.
n. d.
―
On
―
―
On
0.54
0.93
300
On
―
a 0.1 mol dm-3 of a hydrochloric acid aqueous solution. b Not detected. c Amounts of
H2 : ca. 220 μmol. d Amounts of H2 : ca. 1300 μmol

24/31
Figure 7 Arrhenius plots (logarithm of k vs reciprocal of T) of formation of CHA (●)
and CHON (○) in photocatalytic reaction of AN in an aqueous suspension of
Ru(1.0)/Pd(3.5)-TiO2 in the presence of OA (300 μmol).

25/31
Table 3 Apparent activation energy (Ea) and pre-exponential factor (A) for formation
of CHA (●) and CHON (○) in photocatalytic reaction of AN in an aqueous suspension
of Ru(1.0)/Pd(3.5)-TiO2 in the presence of OA (300 μmol).
Product
CHA
Ea / kJ mol-1
A / µmol min-1
2.4 × 102
17
44
CHON
8.0 × 106

26/31
Figure 8 Amounts of AN or CHA adsorbed on various samples in an aqueous solution
in the presence of OA (900 μmol) for 20 h in the dark.

27/31
Figure 9 Expected reaction mechanism of photocatalytic ring hydrogenation of AN to
CHA over Ru/Pd-TiO2 in the presence of OA.

28/31
Scheme Expected reaction schemes of photocatalytic ring hydrogenation of AN to
CHA (main reaction) and CHON (side reaction) over Ru/Pd-TiO2.

29/31