N-monoalkylation of amines with alcohols by tandem photocatalytic and catalytic reactions on TiO2 loaded with Pd nanoparticles

Article abstract of DOI:10.1021/cs300756f

TiO₂ loaded with Pd particles (Pd/TiO₂), when photoirradiated at λ > 300 nm in alcohol containing primary amine, efficiently promotes N-monoalkylation of amine with alcohol, producing the corresponding secondary amine with almost quantitative yields. This occurs via tandem photocatalytic and catalytic reactions: (i) Pd-assisted alcohol oxidation on the photoactivated TiO₂, (ii) condensation of the formed aldehyde with amine on the TiO₂ surface, and (iii) hydrogenation of the formed imine by the surface H atoms on the Pd particles. The rate-determining step is the imine hydrogenation, and the reaction depends strongly on the size of the Pd particles. The catalyst with 0.3 wt % Pd, containing 2-2.5 nm Pd particles, shows the highest activity for imine hydrogenation, and smaller or larger Pd particles are inefficient. Calculations of the number of surface Pd atoms based on the cuboctahedron particle model revealed that the Pd atoms on the triangle site of Pd particles are the active site for hydrogenation. Larger Pd particles contain a larger number of these Pd atoms and are effective for imine hydrogenation. Alcohols, however, are strongly adsorbed onto the larger triangle site and suppress imine hydrogenation. As a result of this, the catalyst with 2-2.5 nm Pd particles, which contains a relatively larger number of Pd atoms on the triangle site and does not promote strong alcohol adsorption, shows the highest activity for imine hydrogenation and promotes efficient N-monoalkylation of amine with alcohol under photoirradiation.

Full text of DOI:10.1021/cs300756f

Research Article
pubs.acs.org/acscatalysis
N‑Monoalkylation of Amines with Alcohols by Tandem
Photocatalytic and Catalytic Reactions on TiO2 Loaded with Pd
Nanoparticles
Yasuhiro Shiraishi,*,† Keisuke Fujiwara,† Yoshitsune Sugano,† Satoshi Ichikawa,‡ and Takayuki Hirai†
†Research Center for Solar Energy Chemistry, and Division of Chemical Engineering, Graduate School of Engineering Science, Osaka
University, Toyonaka 560-8531, Japan
‡Institute for NanoScience Design, Osaka University, Toyonaka 560-8531, Japan
ABSTRACT: TiO2 loaded with Pd particles (Pd/TiO2), when
photoirradiated at λ > 300 nm in alcohol containing primary
amine, efficiently promotes N-monoalkylation of amine with
alcohol, producing the corresponding secondary amine with
almost quantitative yields. This occurs via tandem photo-
catalytic and catalytic reactions: (i) Pd-assisted alcohol
oxidation on the photoactivated TiO2, (ii) condensation of
the formed aldehyde with amine on the TiO2 surface, and (iii)
hydrogenation of the formed imine by the surface H atoms on
the Pd particles. The rate-determining step is the imine
hydrogenation, and the reaction depends strongly on the size of the Pd particles. The catalyst with 0.3 wt % Pd, containing 2−2.5
nm Pd particles, shows the highest activity for imine hydrogenation, and smaller or larger Pd particles are inefficient. Calculations
of the number of surface Pd atoms based on the cuboctahedron particle model revealed that the Pd atoms on the triangle site of
Pd particles are the active site for hydrogenation. Larger Pd particles contain a larger number of these Pd atoms and are effective
for imine hydrogenation. Alcohols, however, are strongly adsorbed onto the larger triangle site and suppress imine
hydrogenation. As a result of this, the catalyst with 2−2.5 nm Pd particles, which contains a relatively larger number of Pd atoms
on the triangle site and does not promote strong alcohol adsorption, shows the highest activity for imine hydrogenation and
promotes efficient N-monoalkylation of amine with alcohol under photoirradiation.
KEYWORDS: photocatalysis, titanium dioxide, nanoparticles, palladium, N-alkylation
catalytic steps in one pot: (i) dehydrogenation of an alcohol
initially proceeds, producing aldehyde and H atoms on the
metal; (ii) catalytic condensation of the formed aldehyde with
primary amine produces imine; and (iii) the imine is
hydrogenated by the H atoms, giving the secondary amine.
Although many homogeneous catalysts, such as Pt, Ru, and Ir
complexes, have been proposed so far,19−24 these methods have
shortcomings in the recovery and reuse of expensive catalysts,
the indispensable use of cocatalysts such as bases and stabilizing
ligands, or both. The design of easily recyclable heterogeneous
catalytic systems is therefore desirable.
Several heterogeneous systems have also been proposed for
one-pot secondary amine synthesis, such as Pd oxides
supported on Fe2O3,25 Pd particles supported on boehmite
nanofibers,26 Ag clusters supported on Al2O3 (in the presence
of FeCl3·6H2O as a homogeneous Lewis acid),27 Ru or Cu
hydroxide supported on Al2O3,28−31 Au particles supported on
TiO2,32 and Pd particles supported on MgO.33 All of these
INTRODUCTION
■
Tandem catalysis that enables multistep reactions in one pot
has attracted a great deal of attention because it avoids the
isolation of unstable intermediates and reduces the production
of wastes.1−3 A variety of one-pot synthetic procedures have
been proposed, but many of these employ homogeneous
catalysts, which generally suffer from product contamination
and limited recyclability.4−6 Development of heterogeneous
catalytic systems that promote efficient one-pot synthetic
reactions is currently the focus of attention.7−11
Secondary amines are one of the most important classes of
chemicals that are widely used for synthesis of pharmaceuticals
and agricultural chemicals.12 Traditionally, these compounds
are synthesized by N-monoalkylation of primary amines with
alkyl halides.12−14 This method, however, requires stoichio-
metric or excess amounts of inorganic bases, with a
concomitant formation of large amounts of inorganic salts as
waste.
An alternative environmentally friendly way for secondary
amine synthesis is the N-alkylation of primary amines with
alcohols as the alkylating reagents in the presence of transition
metal catalysts, the so-called “borrowing hydrogen (H)
strategy”.15−18 The reaction proceeds via three consecutive
Received: November 23, 2012
Revised: January 8, 2013
Published: January 15, 2013
© 2013 American Chemical Society
312
dx.doi.org/10.1021/cs300756f | ACS Catal. 2013, 3, 312−320

ACS Catalysis
Research Article
Table 1. Catalyst Properties and the Results for N-Alkylation of Aniline with Benzyl Alcohol on Various Catalysts under
Photoirradiation
bc
,
b
yields (%)
amt product formed (μmol)
a
b
entry
catalyst
TiO2
dpd (nm)
aniline conv (%)
1
2
benzaldehyde
toluene
H2
1
2
3
4
5
6
7
8
11
4
10
4
0
<1
16
4
0
0
<1
15
3
Au0.3/TiO2
Ag0.3/TiO2
Pt0.3/TiO2
Pd0.1/TiO2
Pd0.3/TiO2
Pd0.5/TiO2
Pd1.0/TiO2
Pd0.3/TiO2
0
2
3
9
6
0
56
65
>99
91
89
>99
53
8
84
96
90
62
56
0
98
93
79
18
6
1.6
2.3
2.6
4.1
49
92
81
64
92
0.9
14
25
42
6
9
19
d
9
7
a
b
c
Average diameter of Pd particles determined by TEM observations (Figure 1). Determined by GC. Yield = [product formed]/[initial amount of
d
aniline] × 100. The result obtained by the reuse of catalyst (entry 6) after simple washing with ethanol.
systems produce secondary amines selectively, but they require
relatively high reaction temperatures (>363 K).
loadings were carried out by the deposition−precipitation
method for Au41 and by the impregnation−reduction method34
for Pt, Ag, and Pd. These catalysts were used for N-alkylation of
aniline with benzyl alcohol. The reactions were carried out by
photoirradiation (λ > 300 nm, 6 h) of a benzyl alcohol solution
(5 mL) containing aniline (50 μmol) with the respective
catalyst (10 mg) at room temperature under N2 atmosphere (1
atm). The conversion of aniline and the yields of N-
benzylidenaniline (1) and N-benzylphenylamine (2) (=
[product formed]/[initial amount of aniline] × 100) are
summarized in Table 1. With bare TiO2 (entry 1), the aniline
conversion is only 11%, and the imine (1) is formed as the
main product, and the secondary amine (2) is scarcely
produced. As shown in entries 2 and 3, Au0.3/TiO2 and
Ag0.3/TiO2 catalysts are also ineffective; their aniline con-
versions are <10%. In these cases, the amount of benzaldehyde
formed is significantly low (<20 μmol). This implies that
photocatalytic oxidation of alcohol does not occur efficiently on
these catalysts and does not provide a large enough amount of
aldehyde for condensation with aniline. As shown in entry 4,
Pt0.3/TiO2 shows a relatively high aniline conversion (56%).
The yield of 2, however, is only 3%, and imine (1) is mainly
produced. This indicates that hydrogenation of imine does not
occur efficiently on the Pt0.3/TiO2 catalyst. In contrast, Pd0.3/
TiO2 (entry 6) produces 2 with almost quantitative yield
(92%). The results clearly indicate that Pd/TiO2 catalyst is
highly active for N-monoalkylation.
The purpose of the present work is to design heterogeneous
catalytic systems that promote N-monoalkylation of primary
amines with alcohols at room temperature. It is well-known that
photoexcitation of semiconductor TiO2 loaded with noble
metal particles, such as Pt,34 Ag,35 Au,36 and Pd,37 under inert
gas atmosphere successfully promotes dehydrogenation of
alcohols and produces aldehydes at room temperature. The
removed H+ are reduced by the photoformed electrons on the
metal particles and are transformed to H atoms (H−metal
species), which are finally removed from the metal surface by
the formation of H2.38 This indicates that TiO2 loaded with
metal particles, when photoactivated in alcohol, produces
aldehyde and H atoms at room temperature. The formed
aldehyde may react with primary amine on the Lewis acid site
on the TiO2 surface and produce imine.34,39 The imine may
then be hydrogenated by the H atoms formed on the metal
surface40 and converted to secondary amine.
On the basis of the above scenario, we studied the N-
alkylation of primary amines with alcohols in the presence of
metal-loaded TiO2 under photoirradiation (λ > 300 nm). Here,
we report that TiO2 loaded with Pd particles (Pd/TiO2)
successfully promotes N-monoalkylation of primary amines at
room temperature via three consecutive steps involving (i)
dehydrogenation of alcohols on the photoactivated TiO2
surface, (ii) catalytic condensation of the formed aldehyde
and amine, and (iii) hydrogenation of the formed imine on the
Pd surface. We found that the reaction efficiency strongly
depends on the size of Pd particles. The catalyst loaded with 0.3
wt % Pd, containing 2−2.5 nm Pd particles, shows the highest
activity.
Catalytic Activity of Pd/TiO2. Pdx/TiO2 catalysts with
different Pd loadings (x = 0.1, 0.3, 0.5, and 1.0 wt %) were
prepared to clarify their activity. Figure 1 shows the typical
transmission electron microscopy (TEM) images of catalysts.
Highly dispersed Pd nanoparticles were observed for all Pdx/
TiO2 catalysts. As shown in Figure 1, the size of the Pd particles
(dPd) determined by the TEM observations increases with an
increase in the Pd loadings; the average diameters for x = 0.1,
0.3, 0.5, and 1.0 catalysts are 1.6, 2.3, 2.6, and 4.1 nm,
respectively. In addition, high-resolution TEM image of
catalysts (Figure 2) revealed that the Pd particles can be
indexed as fcc in structure, as well as bulk Pd metal (JCPDS 46-
1043). Figure 3 shows the diffuse reflectance UV−vis spectra of
the respective catalysts. The catalysts with higher Pd loadings
RESULTS AND DISCUSSION
■
Catalytic Activity of Metal-Loaded TiO2. The catalytic
activity of metal-loaded TiO2, Mx/TiO2 [x (wt %) = M/(M +
TiO2) × 100], was studied for N-alkylation of primary amine
with alcohol. The catalysts loaded with 0.3 wt % metal particles
such as Au, Pt, Ag, and Pd were prepared with JRC-TIO-4 TiO2
particles supplied from the Catalyst Society of Japan
(equivalent to Degussa P25; anatase/rutile = ∼80/20; average
particle size, 24 nm; BET surface area, 59 m2 g−1). Metal
313
dx.doi.org/10.1021/cs300756f | ACS Catal. 2013, 3, 312−320

ACS Catalysis
Research Article
Figure 2. High-resolution TEM images of Pdx/TiO2 catalysts.
Figure 1. Typical TEM image of Pdx/TiO2 catalysts and the size
distributions of Pd particles on the respective catalysts.
show increased absorbance at λ > 300 nm due to the light
scattering by the Pd particles.
Table 1 (entries 5−8) summarizes the results for N-
alkylation of aniline with benzyl alcohol in the presence of
Pdx/TiO2 catalysts. Among the catalysts, Pd0.3/TiO2 (entry 6)
shows the highest aniline conversion (>99%) and the highest 2
yield (92%). The catalysts with lower or higher Pd loadings
show decreased activity. These data suggest that Pd0.3/TiO2
shows the best catalytic activity, and the activity strongly
depends on the amount of Pd loaded. It is noted that the Pd0.3/
TiO2 catalyst is reusable for further reaction. As shown in Table
1 (entry 9), Pd0.3/TiO2, when reused for reaction after simple
washing with ethanol, shows aniline conversions and a yield of
2 that are similar to that obtained with the virgin catalyst (entry
6). This indicates that the catalyst is reusable without the loss of
activity and selectivity. In addition, Pd0.3/TiO2 is applicable for
synthesis of several kinds of secondary amines. As shown in
Table 2, photoirradiation of Pd0.3/TiO2 in alkyl or benzyl
Figure 3. Diffuse reflectance UV−vis spectra of Pdx/TiO2 catalysts.
alcohols containing several kinds of primary amines produces
the corresponding secondary amines with very high yields
(>82%).
Mechanism for N-Alkylation. The Pd/TiO2 catalysts
promote N-alkylation of amine with alcohol via tandem
photocatalytic and catalytic reactions. The reactions are
314
dx.doi.org/10.1021/cs300756f | ACS Catal. 2013, 3, 312−320

ACS Catalysis
Research Article
a
Table 2. N-Alkylation of Various Amines with Alcohols on the Pd0.3/TiO2 Catalyst under Photoirradiation
a
b
Reaction conditions: alcohol, 5 mL; amine, 50 μmol; catalyst, 10 mg; temperature, 298 K; N2, 1 atm; λ > 300 nm; Determined by GC.
Figure 4. Time-dependent change in the amounts of substrate and products during photoreaction of benzyl alcohol and aniline with (a) Pd0.1/TiO2,
(b) Pd0.3/TiO2, and (c) Pd1.0/TiO2 catalysts. Reaction conditions are identical to those in Table 1.
initiated by photoexcitation of TiO2, producing the electron
(e−) and positive hole (h+) pairs.
Condensation of the formed aldehyde with primary amine by
the Lewis acid site on the TiO2 surface produces the imine. The
reaction occurs reversibly as follows:34,39
TiO2 + hν → h+ + e−
(1)
R1−CHO + R2−NH2 ⇄ R1−CHN−R2 + H2O
(5)
The h+ oxidizes alcohol and produces aldehyde and H+.
The imine is hydrogenated by the H−Pd species and is
transformed to secondary amine.40
R1−CH2OH + 2h+ → R1−CHO + 2H+
(2)
R1−CHN−R2 + 2H−Pd → R1−CH2−NH−R2 + 2Pd
(6)
H+ is reduced on the surface of Pd particles by the e−
transferred from the TiO2 conduction band, and transformed
to the surface H atom (H−Pd species).
Figure 4 shows the time-dependent change in the amounts of
substrate and products during the photocatalytic reaction of
aniline with benzyl alcohol in the presence of Pd0.1/TiO2, Pd0.3/
TiO2, and Pd1.0/TiO2 catalysts, respectively. As shown in Figure
4b, photoirradiation of Pd0.3/TiO2 produces benzaldehyde and
H2 at the initial stage, then the amount of aniline decreases, and
the imine 1 and secondary amine 2 appear. The amount of 1
stays very low, and the amount of 2 increases with time; 6 h of
photoirradiation leads to almost quantitative transformation of
H+ + e− + Pd → H−Pd
(3)
Parts of the H atoms are removed from the surface of Pd
particles as H2 gas by coalescence.
H−Pd + H−Pd → H2↑+2Pd
(4)
315
dx.doi.org/10.1021/cs300756f | ACS Catal. 2013, 3, 312−320

ACS Catalysis
Research Article
aniline to 2. These substrate and product profiles suggest that
the above reaction sequence (eqs 1−6) proceeds efficiently on
the Pd0.3/TiO2 catalyst. In contrast, with Pd0.1/TiO2 and Pd1.0/
TiO2 (Figure 4a and c), the formation of 2 is much slower than
that obtained with Pd0.3/TiO2; quantitative formation of 2
requires more than 10 h. These data suggest that Pd0.3/TiO2
shows the highest activity for N-alkylation, and the amount of
Pd loaded strongly affects the catalytic activity.
Effects of Pd Amount on the Reaction Steps. The
effects of the Pd amount on the respective reaction steps (eqs
2−6) for N-alkylation were studied to clarify the reason for the
high activity of Pd0.3/TiO2. The oxidation of alcohol by the
photoformed h+ (eq 2) as the first step for reaction was studied
by photoirradiation of the respective catalysts (10 mg) in
benzyl alcohol (5 mL) without aniline. Figure 5 (white) shows
Table 3. Condensation of Benzaldehyde with Aniline on the
Respective Catalysts in the Dark.
a
a
entry
catalyst
none
1 yield (%)
1
2
3
4
5
6
29
90
92
94
92
95
TiO2
Pd0.1/TiO2
Pd0.3/TiO2
Pd0.5/TiO2
Pd1.0/TiO2
a
Determined by GC.
the respective catalysts for 3 h at 298 K in the dark. The
absence of catalyst (entry 1) produces 1 with only 29% yield. In
contrast, addition of bare TiO2 (entry 2) produces 1 with >90%
yield. The yield of 1 scarcely changes when using the catalysts
with different Pd loadings (entries 3−6). These data clearly
suggest that the imine formation (eq 5) is also not the rate-
determining step for N-alkylation.
As a result of the above, the rate-determining step for N-
alkylation is the hydrogenation of imine by the H atom formed
on the surface of Pd particles (eq 6) as the final step for
reaction sequence. The imine hydrogenation is strongly affected
by the size of the Pd particles loaded. This is confirmed by
hydrogenation of imine 1 with molecular hydrogen (H2) as a
hydrogen source. The reaction was performed in MeCN
containing 1 with H2 (1 atm) in the presence of respective Pdx/
TiO2 catalysts at a constant Pd amount (0.28 μmol) for 2 h in
the dark. Figure 6 (black) shows the relationship between the
diameter of Pd particles (dPd) and the yield of secondary amine
(2) formed. Smaller Pd particles have a larger surface area and
possess a larger number of surface Pd atoms. The 2 yields,
however, increase with an increase in the size of the Pd
Figure 5. Amount of products formed during photoreaction of benzyl
alcohol with respective catalysts in the absence of amine.
the amount of benzaldehyde formed by 6 h of photoirradiation.
Pd0.1/TiO2 produces the highest amount of aldehyde, and the
amount decreases with the Pd loadings. This suggests that
lower Pd loading catalysts are active for photocatalytic
oxidation of alcohol. It is well-known that a metal/semi-
conductor heterojunction creates a Schottky barrier.42 Photo-
formed conduction band e− of a semiconductor overcomes this
barrier and is trapped by the metal particles, resulting in an
enhanced charge separation between h+ and e−. The charge
separation efficiency therefore strongly depends on the height
of the Schottky barrier. The increased metal loadings onto the
semiconductor lead to an increase in the height of the Schottky
barrier as a result of the electron transfer from the
semiconductor to the metal.43 The increased barrier height
by larger Pd loadings probably suppresses the migration of the
conduction band e− to the Pd particles. This may decrease the
charge separation efficiency and result in decreased photo-
catalytic aldehyde formation (Figure 5). However, as shown in
Table 1 (entries 5−8), all of the Pdx/TiO2 catalysts produce
benzaldehyde with amounts larger than that of aniline (50
μmol). This suggests that an amount of aldehyde that is
sufficient for condensation with aniline is produced on all of the
catalysts, and the photocatalytic alcohol oxidation is not the
rate-determining step for N-alkylation.
Condensation of amine with aldehyde (eq 5) is the next step
for N-alkylation. The reaction is catalyzed by the Lewis acid site
on the TiO2 surface34,39 and is not affected by the amount of
Pd loaded. Table 3 summarizes the results for condensation
between aniline and 2 equiv of benzaldehyde in the presence of
Figure 6. The yields of secondary amine 2 obtained during
hydrogenation of imine 1 in MeCN or benzylalcohol with respective
Pdx/TiO2 catalysts (Pd: 0.28 μmol) under H2 in the dark. The
amounts of catalysts used are 30 (Pd0.1), 10 (Pd0.3), 6 (Pd0.5), and 3
mg (Pd1), respectively.
316
dx.doi.org/10.1021/cs300756f | ACS Catal. 2013, 3, 312−320

ACS Catalysis
Research Article
particles. This indicates that not all of the surface Pd atoms are
the active site, and the Pd atoms on the specific surface site
behave as the active site for imine hydrogenation.
particle is expressed as 2m−1. The lattice constant of Pd is
0.389 nm (Figure 7c),52 and the atomic diameter of Pd is 0.274
nm.53 The diameter of the Pd particle (dPd) is therefore
expressed as follows:
Active Site for Imine Hydrogenation. The specific active
site for imine hydrogenation on the Pd particles must be
identified. Morphology and size of metal particles strongly
affect the catalytic activity because of the electronic and
geometric effects.44−47 Metal particles contain different types of
surface atoms located at the vertex, edge, square, and triangle
sites. The number of these atoms changes substantially with the
size of the particles. As shown in Figure 2, high-resolution TEM
images of catalysts revealed that the shape of the Pd particle is
part of cuboctahedron, which is surrounded by [111] and [100]
surfaces. The Pd particles on the TiO2 surface can therefore
simply be modeled as a fcc cuboctahedron, as shown in Figure
7a, which is often used for related nanoparticle systems.48−50
0.389
dPd(nm) =
{(2m − 1) − 1} + 0.274
(13)
2
The number of Pd particles per unit weight of Pd (nparticle),
and the number of Pd atoms on the specific surface site (vertex,
edge, square, triangle) per unit weight of Pd (Nspecific) can
therefore be expressed using the molecular weight of Pd [Mw
(= 106.42 g mol−1)], as follows:
1
nparticle (mol g−1) =
*
MW × Ntotal
(14)
(15)
−1
*
(mol g ) = Nspecific × nparticle
N
specific
As shown in Figure 1, the average diameters of Pd particles
(dPd) for catalysts were determined by TEM observations to be
1.6 (Pd0.1/TiO2), 2.3 (Pd0.3/TiO2), 2.6 (Pd0.5/TiO2), and 4.1
nm (Pd1.0/TiO2), respectively. The shell numbers (m) for Pd
particles on the respective catalysts can therefore be calculated
using eq 13. The numbers of Pd atoms on the specific surface
site per unit weight of Pd (Nspecific) can then be determined
using eqs 9−12, 14, and 15. Figure 8 shows the relationship
Figure 7. (a) The fcc cuboctahedron model for Pd particles. (b) The
largest cross-sectional (111) hexagonal facet of the Pd particle. (c) The
unit lattice of Pd particle.
This allows rough determination of the number of Pd atoms on
the respective surface sites. Considering the full shell close
packing cuboctahedron for the Pd particle where one Pd atom
is surrounded by 12 others, the number of total Pd atoms per
particle (Ntotal*) and the number of surface Pd atoms per
particle (N*surface) can be expressed by the following equations
using the shell number (m).
Figure 8. Relationship between the diameter of Pd particles on the
respective Pdx/TiO2 catalyst and the number of Pd atoms on the
specific surface site of the Pd particle per unit weight of Pd.
between the diameter of Pd particles and their respective
Nspecific values. The Nvertex and Nedge values decrease with an
increase in the size of Pd particle, and the Nsquare values scarcely
change with the Pd particle size. These profiles are completely
different from the catalytic activity of Pd particles for imine
hydrogenation (Figure 6, black). In contrast, the Ntiangle value
increases with an increase in the Pd particle size, and the profile
is very similar to the imine hydrogenation profile. This
calculation result implies that Pd atoms on the triangle site
are active for imine hydrogenation.
1
3
(2m − 1)(5m2 − 5m + 3)
*
Ntotal
=
(7)
Nsurface = 10m2 − 20m + 12
*
*
*
*
*
= Nvertex + Nedge + Nsquare + Ntriangle
(8)
The numbers of Pd atoms on the specific surface sites
(Nspecific*) such as vertex, edge, square, and triangle sites per Pd
particles are therefore expressed by the following equations.51
This is further confirmed by the turnover number for imine
hydrogenation per number of Pd atoms on the specific surface
site (TONspecific). The amount of secondary amine 2 (mol)
formed during hydrogenation of imine 1 by H2 with respective
Pdx/TiO2 catalysts in the dark (Figure 6) was divided by the Pd
amount (0.28 μmol) and the respective Nspecific values, using the
following equation:
*
Nvertex = 12
(9)
(10)
(11)
(12)
*
Nedge = 24(m − 2)
Nsquare = 6(m − 2)2
*
*
Ntriangle = 4(m − 3)(m − 2)
[amount of product formed]
TONspecific
=
As shown in Figure 7b, the number of Pd atoms on the
diagonal line of the largest cross-sectional (111) facet of Pd
N
× (0.28 × 10−6)
specific
(16)
317
dx.doi.org/10.1021/cs300756f | ACS Catal. 2013, 3, 312−320

ACS Catalysis
Research Article
Figure 9 shows the relationship between the size of Pd
particle on the catalysts and respective TONspecific values. The
of smaller Pd particles is similar to that obtained in MeCN
(black), but the activity of larger Pd particles is decreased
significantly. These data clearly indicate that the imine
hydrogenation on larger Pd particles is suppressed by alcohol.
In these cases, GC analysis detected the formation of toluene.
This indicates that, as reported,56 Pd particles promote
hydrogenation of alcohol by the H−Pd species, leading to
hydrogenolysis.
Figure 10a shows the amount of toluene formed during
hydrogenation of benzyl alcohol with H2 in the dark with the
Figure 9. Relationship between the Pd particle size and the turnover
number for hydrogenation of imine 1 per number of Pd atoms on the
specific surface site (TONspecific). The TONspecific value was calculated
with the results for imine hydrogenation (Figure 6, black) using eq 16.
TONvertex, TONedge, and TONsquare values change with the Pd
particle size. In contrast, the TONtriangle values for the respective
catalysts are similar and are independent of the Pd particle size.
This result clearly indicates that the Pd atoms on the triangle
site are the active site for imine hydrogenation.
The H atoms formed on the Pd particles diffuse around the
particle surface,54 meaning that the H atoms exist randomly on
all parts of the Pd particle surface. The strong dependence of
the imine hydrogenation activity on the number of Pd atoms on
the triangle site is probably due to the adsorption of imine onto
the triangle site. The Pd atoms on the triangle site have a higher
coordination number (9) than those on the other surface sites,
such as vertex (5), edge (7), and square (8), and are charged
more positively. The imine is therefore preferentially adsorbed
onto the positively charged triangle site via π electronic
interaction with the aromatic ring and CN bond. This
adsorption may promote efficient hydrogenation by the H−Pd
species and result in a clear relationship between the imine
hydrogenation activity and the number of Pd atoms on the
triangle site. In the hydrogenation of an alkyne and a diene,
such as 2-methyl-3-butyn-2-ol44 and 1,3-butadiene,55 on Pd
particle catalysts with H2, the catalytic activity also depends on
the number of Pd atoms on the triangle site, and the
dependence is also considered to be due to the adsorption of
these olefins onto the positively charged triangle site. These
reports support the above mechanism for imine hydrogenation.
Effect of Alcohol on the Imine Hydrogenation. The
above results suggest that the triangle site of Pd particles is the
active site for imine hydrogenation. The number of these Pd
atoms increases with an increase in the Pd particle size, and the
Pdx/TiO2 catalyst with higher Pd loadings, which contain larger
Pd particles, efficiently promotes imine hydrogenation.
However, as shown in Figure 4b and c, photocatalytic reaction
of benzyl alcohol and aniline with Pd1.0/TiO2 shows much
lower activity for imine hydrogenation than Pd0.3/TiO2. This is
inconsistent with the results for imine hydrogenation with H2
(Figure 6, black).
Figure 10. (a) Amount of toluene formed during reaction of benzyl
alcohol with respective Pdx/TiO2 catalysts (Pd: 0.28 μmol) under H2
in the dark. The amounts of catalysts used are 30 (Pd0.1), 10 (Pd0.3), 6
(Pd0.5), and 3 mg (Pd1), respectively. (b) Relationship between the Pd
particle size and the turnover number for toluene formation per
number of Pd atoms on the specific surface site (TONspecific). The
TONspecific values were calculated using eq 16.
respective Pdx/TiO2 catalysts at the identical Pd amount (0.28
μmol). The hydrogenation activity increases with the Pd
particle size, and the profile is very similar to the imine
hydrogenation profile (Figure 6, black). Figure 10b shows the
turnover number for toluene formation per number of specific
surface sites (TONspecific) calculated using eq 16. The
TONtriangle values for all of the catalysts are similar, as is the
case for imine hydrogenation (Figure 9). These findings
indicate that the hydrogenation of alcohol is also promoted on
the triangle site of Pd particles, and this suppresses the imine
hydrogenation.
The strongly suppressed imine hydrogenation on larger Pd
particles is probably due to the stronger adsorption of alcohol
than imine onto larger triangle site by their different adsorption
modes. As shown in Figure 11, benzyl alcohol has a planar
structure and is adsorbed very strongly onto the Pd surface via
π interaction of the aromatic ring and the lone pair electron of
The lower imine hydrogenation activity of larger Pd particles
is because alcohols are adsorbed onto the larger triangle site
more strongly than imines and undergo hydrogenation. Figure
6 (white) shows the results for imine hydrogenation with H2
when carried out in benzyl alcohol. The hydrogenation activity
318
dx.doi.org/10.1021/cs300756f | ACS Catal. 2013, 3, 312−320

ACS Catalysis
Research Article
purification. Japan Reference Catalyst JRC-TIO-4 TiO2
particles were kindly supplied by the Catalysis Society of Japan.
Pdx/TiO2 [x (wt %) = 0.1, 0.3, 0.5, and 1.0] were prepared as
follows: TiO2 (1.0 g) and Pd(NO3)2 (2.2, 6.5, 10.9, or 21.9 mg)
were added to water (40 mL) and evaporated under vigorous
stirring at 353 K for 12 h. The obtained powders were calcined
at 673 K under air flow (0.5 L min−1) and then reduced at 673
K under H2 flow (0.2 L min−1). The heating rate and holding
time at 673 K for these treatments were 2 K min−1 and 2 h,
respectively. Ag0.3/TiO2 and Pt0.3/TiO2 were prepared in a
manner similar to those for Pdx/TiO2, using AgNO3 (4.7 mg),
or H2PtCl6·6H2O (8.0 mg) as precursor.
Figure 11. Schematic representation of different adsorption modes of
benzyl alcohol and imine 1 onto the surface of a Pd particle. Geometry
optimizations of the molecules were performed with B3LYP/6-31G*
basis set.
Au0.3/TiO2 was prepared by a deposition-precipitation
method as follows: HAuCl4·4H2O (6.3 mg) was added to
water (50 mL). The pH of the solution was adjusted to 7 by an
addition of 1 M NaOH. TiO2 (1.0 g) was added, and the
solution was stirred vigorously at 353 K for 3 h. The solids were
recovered by centrifugation and washed with water. The
obtained solids were dried at 353 K and calcined at 673 K for 2
h under air flow (0.5 L min−1).
Photoreaction. Each of the respective catalysts (10 mg)
was suspended in alcohol (5 mL) containing a required amount
of amine within a Pyrex glass tube (ϕ, 10 mm; capacity, 20
mL). The tube was sealed with a rubber septum cap. The
catalyst was dispersed by ultrasonication for 5 min, and N2 was
bubbled through the solution for 5 min. The tube was
photoirradiated with magnetic stirring at 298 K by a 2 kW Xe
lamp (λ > 300 nm; Ushio Inc.). The light intensity at 300−400
nm is 18.2 W m−2. After photoirradiation, the gas-phase
product was analyzed by GC/TCD (Shimadzu; GC-8A). The
resulting solution was recovered by centrifugation and analyzed
by GC/FID (Shimadzu; GC-1700), and the substrate and
product concentrations were determined with authentic
samples. Identification of the products was performed by a
Shimadzu GC/MS system (GCMS-QP5050A).
Hydrogenation by H2. Each of the respective catalysts was
suspended in a solution (5 mL) containing substrate within a
Schlenk tube. The catalyst was dispersed by ultrasonication for
5 min, and H2 was bubbled through the solution for 5 min. The
H2 pressure was maintained at 1 atm using a balloon. The tube
was placed on the digitally controlled water bath with magnetic
stirring at 298 K.
Analysis. Total Pd amounts of the catalysts were
determined by an X-ray fluorescence spectrometer (Seiko
Instruments, Inc.; SEA2110). Diffuse reflectance UV−vis
spectra were measured on an UV−vis spectrophotometer
(Jasco Corp.; V-550 with Integrated Sphere Apparatus ISV-
469) with BaSO4 as a reference.59,60 TEM observations were
carried out using an FEI Tecnai G2 20ST analytical electron
microscope operated at 200 kV.61 The cuboctahedron Pd
particles were created on the Crystal Studio Ver. 9.0 software
(CrystalSolf. Inc.) and used for calculation of the number of Pd
atoms.50 Ab initio calculations were carried out with the
Gaussian 03 program, and the geometry optimization was
performed with the density functional theory (DFT) using the
B3LYP function with the 6-31G* basis set.62,63
the oxygen atom.57 In contrast, the imine has a twisted
structure58 and may be adsorbed more weakly onto the Pd
surface. The interaction between alcohol and the Pd surface is
probably strengthened by an increase in the area of the triangle
site. This probably leads to enhanced alcohol adsorption onto
the larger Pd particles and suppresses imine hydrogenation
more strongly. As shown in Figures 4 and 5, a larger amount of
toluene is produced during the photocatalytic reaction of
alcohol in the presence of catalysts with a larger Pd particle.
This indicates that the decreased imine hydrogenation activity
of larger Pd particles is due to the strong adsorption of alcohols.
These results suggest that the Pd0.3/TiO2 catalyst with 2−2.5
nm Pd particles, which contains a relatively larger number of
triangle site Pd atoms but does not allow strong adsorption of
alcohols, shows the highest activity for imine hydrogenation
and promotes efficient N-monoalkylation of primary amine
with alcohol under photoirradiation.
CONCLUSION
■
We found that Pd/TiO2 catalyst promotes N-monoalkylation of
primary amine with alcohol under photoirradiation at room
temperature. Several kinds of secondary amines are successfully
produced with high yields. Tandem photocatalytic and catalytic
reactions promote three consecutive reactions, consisting of Pd-
assisted alcohol oxidation on the photoactivated TiO2, catalytic
condensation of the formed aldehydes with amines on the TiO2
surface, and hydrogenation of formed imine with H atoms on
the Pd particles. The rate-determining reaction is the imine
hydrogenation as the final step. The reaction is promoted on
the Pd atoms on the triangle site of the Pd particle via an
adsorption of imine. The imine adsorption onto the larger
triangle site is strongly suppressed by competitive adsorption of
alcohol. As a result of this, the catalyst with 2−2.5 nm Pd
particles, which contain a relatively larger number of triangular
Pd atoms and do not promote strong alcohol adsorption, shows
the highest activity for imine hydrogenation and promotes
efficient N-alkylation of amine with alcohol under photo-
irradiation. This tandem catalytic system offers significant
advantages: (i) no harmful byproduct forms, (ii) the reaction
proceeds at room temperature, and (iii) several secondary
amines are successfully produced. The tandem reactions
promoted by photocatalytic and catalytic actions, therefore,
have a potential to be a powerful method for one-pot synthesis
of organic compounds in an environmentally friendly way.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: shiraish@cheng.es.osaka-u.ac.jp.
EXPERIMENTAL SECTION
Materials. All reagents were purchased from Wako, Tokyo
Kasei, and Sigma-Aldrich and were used without further
■
Notes
The authors declare no competing financial interest.
319
dx.doi.org/10.1021/cs300756f | ACS Catal. 2013, 3, 312−320

ACS Catalysis
Research Article
(34) Shiraishi, Y.; Ikeda, M.; Tsukamoto, D.; Tanaka, S.; Hirai, T.
Chem. Commun. 2011, 47, 4811−4813.
ACKNOWLEDGMENTS
■
This work was supported by a Grant-in-Aid for Scientific
Research (No. 23360349) from the Ministry of Education,
Culture, Sports, Science and Technology, Japan (MEXT).
(35) Sclafani, A.; Mozzanega, M.-N.; Pichat, P. J. Photochem.
Photobiol., A 1991, 59, 181−189.
(36) Bamwenda, G. R.; Tsubota, S.; Nakamura, T.; Haruta, M. J.
Photochem. Photobiol., A 1995, 89, 177−189.
(37) Ohtani, B.; Kakimoto, M.; Nishimoto, S.; Kagiya, T. J.
Photochem. Photobiol., A 1993, 70, 265−272.
REFERENCES
■
(1) Lee, J. M.; Na, Y.; Han, H.; Chang, S. Chem. Soc. Rev. 2004, 33,
302−312.
(38) Shiraishi, Y.; Takeda, Y.; Sugano, Y.; Ichikawa, S.; Tanaka, S.;
Hirai, T. Chem. Commun. 2011, 47, 7863−7865.
(39) Jennings, W. B.; Lovely, C. J. Tetrahedron 1991, 47, 5561−5568.
(40) Krupka, J.; Patera, J. Appl. Catal., A 2007, 330, 96−107.
(41) Tsukamoto, D.; Shiraishi, Y.; Sugano, Y.; Ichikawa, S.; Tanaka,
S.; Hirai, T. J. Am. Chem. Soc. 2012, 134, 6309−6315.
(42) Schottky, W. Z. Phys. 1939, 113, 367−414.
(43) Uchihara, T.; Matsumura, M.; Yamamoto, A.; Tsubomura, H. J.
Phys. Chem. 1989, 93, 5870−5874.
(2) Kolb, H. C.; VanNieuwenhze, M. S.; Sharpless, K. B. Chem. Rev.
1994, 94, 2483−2547.
(3) Wasilke, J.-C.; Obrey, S. J.; Baker, R. T.; Bazan, G. C. Chem. Rev.
2005, 105, 1001−1020.
(4) Gunanathan, C.; Ben-David, Y.; Milstein, D. Science 2007, 317,
790−792.
(5) Cadierno, V.; Francos, J.; Gimeno, J.; Nebra, N. Chem. Commun.
2007, 2536−2538.
(44) Semagina, N.; Renken, A.; Laub, D.; Kiwi-Minsker, L. J. Catal.
2007, 246, 308−314.
(6) Zweifel, T.; Naubron, J.-V.; Grutzmacher, H. Angew. Chem., Int.
̈
Ed. 2009, 48, 559−563.
(45) Bars, J.; Le.; Specht, U.; Bradlay, J. S.; Blackmond, D. G.
Langmuir 1999, 15, 7621−7625.
(7) Yin, L.; Liebscher, J. Chem. Rev. 2007, 107, 133−173.
(8) Motokura, K.; Fujita, N.; Mori, K.; Mizugaki, T.; Ebitani, K.;
Kaneda, K. J. Am. Chem. Soc. 2005, 127, 9674−9675.
(9) Felpin, F.-X.; Fouquet, E. ChemSusChem 2008, 1, 718−724.
(10) Kim, J. W.; Yamaguchi, K.; Mizuno, N. Angew. Chem., Int. Ed.
2008, 47, 9249−9251.
(46) Coq, B.; Figueras, F. J. Mol. Catal. A: Chem. 2001, 173, 117−
134.
(47) Molnar
173, 185−221.
́ ́ ́
, A.; Sarkany, A.; Varga, M. J. Mol. Catal. A: Chem. 2001,
(48) Wilson, O. M.; Knecht, M. R.; Garcia-Martinez, J. C.; Crooks, R.
M. J. Am. Chem. Soc. 2006, 128, 4510−4511.
(49) Arruda, T. M.; Shyam, B.; Zeigelbauer, J. M.; Mukerjee, S.;
Ramaker, D. E. J. Phys. Chem. C 2008, 112, 18087−18097.
(50) Shiraishi, Y.; Tsukamoto, D.; Sugano, Y.; Shiro, A.; Ichikawa, S.;
Tanaka, S.; Hirai, T. ACS Catal. 2012, 2, 1984−1992.
(51) Benfield, R. E. J. Chem. Soc. Faraday Trans. 1992, 88, 1107−
1110.
(11) Shiraishi, Y.; Sugano, Y.; Tanaka, S.; Hirai, T. Angew. Chem., Int.
Ed. 2010, 49, 1656−1660.
(12) Salvatore, R. N.; Yoon, C. H.; Jung, K. W. Tetrahedron 2001, 57,
7785−7811.
(13) Buchwald, S. L.; Mauger, C.; Mignani, G.; Scholz, U. Adv. Synth.
Catal. 2006, 348, 23−39.
(14) Navarro, O.; Marion, N.; Mei, J.; Nolan, S. P. Chem.Eur. J.
2006, 12, 5142−5148.
(52) Fornander, H.; Hultman, L.; Birch, J.; Sundgren, J.-E. J. Cryst.
(15) Hamid, M. H. S. A.; Slatford, P. A.; Williams, J. M. J. Adv. Synth.
Catal. 2007, 349, 1555−1575.
Growth 1998, 186, 189−202.
(53) Pauling, L. J. Am. Chem. Soc. 1947, 69, 542−553.
(54) Parambhath, V. B.; Nagar, R.; Sethupathi, K.; Ramaprabhu, S. J.
Phys. Chem. C 2011, 115, 15679−15685.
(16) Guillena, G.; Ramon
́
, D. J.; Yus, M. Chem. Rev. 2010, 110,
1611−1641.
(17) Guillena, G.; Ramon
́
, D. J.; Yus, M. Angew. Chem., Int. Ed. 2007,
(55) Silvestre-Albero, J.; Rupprechter, G.; Freund, H. J. Chem.
Commun. 2006, 80−82.
46, 2358−2364.
(18) Nixon, T. D.; Whittlesey, M. K.; Williams, J. M. J. Dalton Trans.
2009, 753−762.
(56) Keresszegi, C.; Ferri, D.; Mallat, T.; Baiker, A. J. Phys. Chem. B
2005, 109, 958−967.
(19) Tsuji, Y.; Takeuchi, R.; Ogawa, H.; Watanabe, Y. Chem. Lett.
1986, 15, 293−294.
(57) Souto, R. M.; Rodríguez, J. L.; Pastor, E. Langumuir 2000, 16,
8456−8462.
(20) Watanabe, Y.; Tsuji, Y.; Ohsugi, Y. Tetrahedron Lett. 1981, 22,
2667−2670.
(58) Traetteberg, M.; Hilmo, I. J. Mol. Struct. 1978, 48, 395−405.
(59) Shiraishi, Y.; Saito, N.; Hirai, T. J. Am. Chem. Soc. 2005, 127,
8304−8306.
(21) Naskar, S.; Bhattacharjee, M. Tetrahedron Lett. 2007, 48, 3367−
3370.
(60) Shiraishi, Y.; Saito, N.; Hirai, T. J. Am. Chem. Soc. 2005, 127,
12820−12822.
(22) Fujita, K.; Enoki, Y.; Yamaguchi, R. Tetrahedron 2008, 64,
1943−1954.
(61) Tsukamoto, D.; Shiro, A.; Shiraishi, Y.; Sugano, Y.; Ichikawa, S.;
Tanaka, S.; Hirai, T. ACS Catal. 2012, 2, 599−603.
(62) Shiraishi, Y.; Ichimura, C.; Sumiya, S.; Hirai, T. Chem.Eur. J.
2011, 17, 8324−8332.
(23) Blank, B.; Madalska, M.; Kempe, R. Adv. Synth. Catal. 2008, 350,
749−758.
(24) Michlik, S.; Kempe, R. Chem.Eur. J. 2010, 16, 13193−13198.
(25) Zhang, Y.; Qi, X.; Cui, X.; Shi, F.; Deng, Y. Tetrahedron Lett.
2011, 52, 1334−1338.
(63) Shiraishi, Y.; Sumiya, S.; Hirai, T. Chem. Commun. 2011, 47,
4953−4955.
(26) Kwon, M. S.; Kim, S.; Park, S.; Bosco, W.; Chidrala, R. K.; Park,
J. J. Org. Chem. 2009, 74, 2877−2879.
(27) Shimizu, K.; Nishimura, M.; Satsuma, A. ChemCatChem 2009, 1,
497−503.
(28) Yamaguchi, K.; Mizuno, N. Synlett 2010, 16, 2365−2382.
(29) Kim, J. W.; Yamaguchi, K.; Mizuno, N. J. Catal. 2009, 263, 205−
208.
(30) Yamaguchi, K.; He, J.; Oishi, T.; Mizuno, N. Chem.Eur. J.
2010, 16, 7199−7207.
(31) He, J.; Yamaguchi, K.; Mizuno, N. Chem. Lett. 2010, 39, 1182−
1183.
(32) He, L.; Lou, X.-B.; Ni, J.; Liu, Y.-M.; Cao, Y.; He, H.-Y.; Fan, K.-
N. Chem.Eur. J. 2010, 16, 13965−13969.
(33) Corma, A.; Rod
254−260.
́
enas, T.; Sabater, M. J. Chem.Eur. J. 2010, 16,
320
dx.doi.org/10.1021/cs300756f | ACS Catal. 2013, 3, 312−320