EXAFS Characterization of Pd Catalysts for Enantioselective Hydrogenation of α-Phenylcinnamic Acid: Pretreatment Effects and Thiol Adsorption

Article abstract of DOI:10.1007/s10562-016-1859-6

Abstract: The effects of the pretreatment and Pd-distribution on the enantioselective hydrogenation of α-phenylcinnamic acid over cinchonidine-modified Pd/C and Pd/Al₂O₃ were studied by use of EXAFS techniques combined with thiol ads

Full text of DOI:10.1007/s10562-016-1859-6

Catal Lett (2016) 146:2430–2440
DOI 10.1007/s10562-016-1859-6
EXAFS Characterization of Pd Catalysts for Enantioselective
Hydrogenation of α-Phenylcinnamic Acid: Pretreatment Effects
and Thiol Adsorption
1
2
2
2
1
Takeshi Kubota · Haruka Sato · Takayuki Uchida · Tae Yeon Kim · Koji Omata ·
2
2
2
Tomonori Misaki · Yasuaki Okamoto · Takashi Sugimura
Received: 12 June 2016 / Accepted: 4 September 2016 / Published online: 22 October 2016
Springer Science+Business Media New York 2016
©
Abstract The effects of the pretreatment and Pd-
distribution on the enantioselective hydrogenation of
α-phenylcinnamic acid over cinchonidine-modified Pd/C
and Pd/Al O were studied by use of EXAFS techniques
Keywords EXAFS · Pd catalysts ·
Enantioselective hydrogenation · α-Phenylcinnamic acid ·
Thiol adsorption
2
3
combined with thiol adsorption. It is suggested that the
enhanced performance by the H 2-pretreatment at 353
K
1 Introduction
is ascribed to the elimination of contaminants and/or H 2O
from Pd/C. The Pd–S CN is related to the Pd dispersion and
the reactivity and diffusivity of thiols used.
Enantioselective hydrogenation of prochiral ketones or olefins
over heterogeneous metal catalysts modified with adsorbed
chiral molecules has been extensively studied because of
advantages for industrial applications of solid catalysts over
Graphical Abstract
the corresponding homogeneous catalyst systems [
1–5].
Nevertheless, heterogeneous enantioselective hydrogenation
needs further improvements in the selectivity and reaction
rate to meet industrial requirements. One of feasible hetero -
geneous catalyst systems is cinchonidine (CD)-modified Pd
catalysts for enantioselective hydrogenation of C =C double
bonds of α,β-unsaturated carboxylic acids represented by
α-phenylcinnamic acid (PCA) (Scheme 1) [ 6–8]. A variety
of supported Pd catalysts have been used for the enantio-
differentiating hydrogenation of PCA. It was reported that a
R-SH
Pd Distribuꢀon Effects
Pretreatment Effects
CD-modified Pd/C
R-H
CD-modified 5% Pd/TiO catalyst prepared by a precipita -
2
tion-deposition method was more active and selective (79%ee
under optimized reaction conditions) for the hydrogenation of
PCA (298 K) than SiO -, Al O - or ZrO -supported Pd cata-
Electronic supplementary material The online version of this
article (doi:10.1007/s10562-016-1859-6) contains supplementary
material, which is available to authorized users.
2
2
3
2
lysts, suggesting that strong Pd metal-support interactions are
unfavorable for the enantioselective hydrogenation [6, 9–11].
Pd/C has been widely used as a typical Pd metal catalyst
supported on an almost inactive support. However, Nitta
et al. [ 9] showed that commercial and home-made Pd/C
catalysts modified with CD were unexpectedly less active
and less selective for the enantioselective hydrogenation
of PCA (10–21%ee) than Pd catalysts supported on SiO2,
Al O or TiO . More recently, it was found that the selec -
ꢀ
ꢀ
Yasuaki Okamoto
yokamoto@sci.u-hyogo.ac.jp
Takashi Sugimura
sugimura@sci.u-hyogo.ac.jp
1
2
Interdisciplinary Graduate School of Science and
Engineering, Shimane University, Matsue 690-8504, Japan
Graduate School of Material Science, University of Hyogo,
2
3
2
3
-2-1 Kohto, Kamigori, Ako-Gun, Hyogo 678-1297, Japan
tivity was increased up to moderate one, 49%ee, when an
1
23

EXAFS Characterization of Pd Catalysts for Enantioselective Hydrogenation of α-Phenylcinnamic Acid:…
2431
Scheme 1 Enantioselective hydrogenation of α-phenylcinnamic acid over cinchonidine-modified Pd/C
eggshell-type Pd/C catalyst was employed for the hydroge-
nation of PCA [12]. The relatively low enantioselectivity of
Pd/C catalysts was, however, overcome by figuring out a
catalyst pretreatment. Later on, we demonstrated that only
when the catalyst was modified with CD after a pretreatment
systems, the effects of Pd distribution on the selectivity may
be explained in terms of diffusion of the substrate/modifier/
additive or in terms of the formation and/or configuration of
PCA-CD surface interaction species. Thus, it seems difficult
to determine the factors controlling the selectivity in these
complicated reaction systems. In the present study, we tried
at 353 K in a solvent under an atmospheric pressure of H, a
2
5
% Pd/C commercial catalyst exhibited a very high selectiv-
to directly monitor the diffusion effects of the organic mol
-
ity of 79%ee for the hydrogenation of PCA in the presence
of benzylamine (BA) as an additive, in contrast to 43%ee
without the pretreatment [ 13]. The pretreatment was not
very effective for Pd/TiO and Pd/Al O . On the other hand,
ecules (PCA and CD) with a rather large molecular size dur -
ing the reaction by use of several organic sulfur compounds
(thiols) with a different molecular size as probe molecules.
It is surmised that thiols are easily decomposed on Pd metal
surface to form adsorbed sulfur species, the amount of which
is easily detected by EXAFS as Pd–S contributions [ 23]. In
addition, the decomposition of sulfur compounds on Pd metal
surface itself is an important subject in connection with sul -
fur poisoning, e.g., during hydrogenation of olefins [24, 25].
We previously observed sulfur poisoning of Pd/C during the
hydrogenation of 5-alkylidene-2,4-thiazolidiones [ 26]. We
also compared, in the present study, the sulfur-susceptibility of
Pd metal catalysts supported on carbon and Al O to get some
2
2
3
we later found that the enantioselectivity of PCA hydroge -
nation over CD-modified Pd/C catalysts was considerably
improved by a series of pretreatments of evacuation-solvent
injection-evacuation (co-evaporation pretreatment) before
CD-modification [14]. In spite of extensive increases in the
enantioselectivity and hydrogenation rate by these pretreat-
ments of Pd/C catalysts, it has not been disclosed yet what
causes these improvements of the catalyst performances. In
the present study, we tried to clarify the reasons from the
structural viewpoints of Pd/C by use of an extended X-ray
absorption fine structure (EXAFS) technique.
2
3
insights into the effects of the support on sulfur poisoning.
A variety of Pd/C catalysts are commercially available,
each of which is specially designed and prepared for cer
tain reactions under specific reaction conditions. We previ-
ously proposed to employ commercial Pd/C catalysts for
-
2 Experimental
2.1 Catalysts
a high-throughput screening research on the enantioselec
-
tive hydrogenation of α,β-unsaturated carboxylic acids [15].
Furthermore, combining the Pd3d/C1s X-ray photoelectron
spectroscopy (XPS) intensity ratio and enantioselectivity for
a series of commercial Pd/C catalysts, we showed that the
The catalysts employed in the present study were wet-form
5% Pd/C commercial catalysts (STD, AER, and BNA),
some of which were already used in our previous study
[15, 16, 27]. A commercial catalyst, 5% Pd/Al O (a dry
2
3
distribution of Pd metal inside the catalyst particles primar
ily determined the selectivity of the enantio-differentiating
hydrogenation of PCA [ 16]. We suggested that Pd/C having
a homogeneous distribution or an egg-shell type distribution
of Pd was more suitable for the hydrogenation of PCA than
that with an egg-white- or egg-yolk-type distribution. In the
enantioselective hydrogenation, it is generally accepted that
-
form), was also used for comparison. All the catalysts were
donated by N.E. Chemcat.
2.2 Hydrogenation Reaction Procedures
The substrate, α-phenylcinnamic acid (PCA, Aldrich; 98%),
was purified by recrystallization from an acetone solution.
Dioxane (Wako Pure Chemicals; special grade), cinchonidine
(CD, Wako Pure Chemicals; 99%), and benzylamine (BA,
Wako Pure Chemicals, 98%) were used as received [13–15].
The enantioselective hydrogenation of PCA was conducted
at 296 K as reported in our previous study after one of the
following pretreatments employed previously; Pretreatment
(A) [13]: wet 5% Pd/C (23 mg as a dry form, 0.01 mmol of
the substrate, PCA, interacts with the chiral modifier, CD,
adsorbed on the Pd metal surface to generate enantioselec
tivity [1–4, 6–8]. In addition, auxiliary achiral amine, BA in
the present study, added in the catalyst system improves the
enantioselectivity as well as the hydrogenation rate, although
the roles of the added amine are still under debate [6, 17–22].
Taking into consideration these rather complicated reaction
-
123

2
432
T. Kubota et al.
Pd) was pretreated at 353 K in 5 mL of wet dioxane (diox -
ane containing 2.5% water) under an atmospheric pressure of
for 10 min. The sample was then heated to 313 K in flowing
He and then the He stream was replaced with a stream of H2
for 10 min, followed by flushing with He for 10 min. CO
pulses (4.1 µmol/pulse) were repeatedly injected at 313 K
to the carrier gas (He) until no adsorption was detected (3–6
pulses). Pd/Al O was also dried at 373 K prior to the CO
H for 30 min. After the catalyst was cooled to 296 K, CD
2
(0.02 mmol) in wet dioxane (1 mL) was added to the catalyst.
The resulting suspension was then stirred for 30 min, while
the H -atmosphere was maintained. Next, the substrate (PCA,
2
2
3
0
.5 mmol) and then BA (0.3mmol) were successively injected
adsorption measurements. The Pd dispersion was calculated
from the amounts of chemisorbed CO and Pd, assuming a
ratio of 1:2 for chemisorbed CO/surface Pd atoms.
into the reactor through a septum port. Pretreatment (B): Pd/C
in the reactor was evacuated at room temperature (rt) with a
diaphragm pump at 13 kPa for 10 min, which was sufficient
time to reach a stationary state. After quickly flashing with H2,
dioxane (1 mL) was added to Pd/C, followed by evacuation
In situ Pd K-edge X-Ray absorption fine structure (XAFS)
spectra for the supported Pd catalysts and reference com
-
pounds were measured in a transmission mode at BL14B in
SPring-8 with 8.0 GeV ring energy and 100 mA stored cur-
rent [15] (proposal No. 2008A1848). The synchrotron radi-
ation was monochromatized by a Si (311) double-crystal
monochromator. The higher harmonics were eliminated by
two Rh coated mirrors. The incidence and transmitted X-ray
were monitored by ion chambers with flowing appropriate
mixed gases. The sample (Pd/C: 300 mg and Pd/Al O :
for 10 min and subsequent quick flashing with H (the co-
2
evaporation treatment [ 14]). Wet dioxane (5 mL) was added
prior to the CD modification. A solution of CD (0.02 mmol) in
wet dioxane (1 mL) was added to the pretreated catalyst, and
stirred for 30 min under an atmospheric pressure of H . PCA
2
(
0.5 mmol) in wet dioxane (4 mL) and then BA (0.5 mmol)
were added to the reactor via the septum port. Pretreatment
C): as employed in our earlier study [15, 28], the catalyst and
2
3
(
150 mg in 1 mL of wet dioxane) was reduced at rt [Pretreat-
ment (C)] for 30 min and then transferred to an EXAFS cell
CD were pretreated at 296K in 5 mL of the solvent, while stir-
ring for 30 min under H -atmosphere prior to the injection of
in the presence of an atmospheric pressure of H . A solution
2
2
PCA (0.5mmol) in wet dioxane (4 mL), followed by 0.5mmol
of BA addition. After one of the above pretreatments, the rate
containing a sulfur compound (0.1 mmol in 3 mL of wet
dioxane) was injected to the catalyst at rt. The sulfur com -
pounds used in the present study were t-butylthiol (t-BuSH),
thio-p-cresol (MePhSH), 1-hexadecanethiol (n-C H SH),
of H -uptake was measured to monitor the progress of the
2
reaction, while the reaction solution was maintained under an
1
6
33
atmospheric pressure of H . Hydrogenation rates were deter -
mined at a 20–25% conversion of the substrate.
The product enantioselectivity expressed by %ee
was determined via HPLC (Daicel Chiralpak OJ-H;
and 2-naphtothiol (2-NapSH). They were used as received.
After keeping for 30min at rt, the EXAFS data of the catalyst
were collected using a quick scan mode. Each accumulation
time of the spectral data was set at 5 min. The adsorption
temperature was raised stepwise (from rt to 333K, and then
to 353 K) and kept for 30min at each temperature, followed
by EXAFS measurements at the temperature.
2
2
1
5 cm×4.6 mm; Hex/IPA/TFA=95/5/0.1, retention times:
5.4 min (R) and 24.4 min (S) at a flow rate of 1 mL/min),
and the final reaction conversions were determined via
1
H-NMR (600 MHz) spectroscopy [29]. After washing each
The EXAFS data were analyzed assuming a spheri
cal wave approximation and a single scattering model
-
portion with an aqueous solution of 1 N HCl, the reaction
mixture was found to contain only the substrate and the
hydrogenation products. The enantiomeric excess (%ee)
was calculated according to the following equation.
(REX2000, Rigaku Co. Ltd., Japan). The degree of error bars
for the each parameter was estimated using the diagonal ele-
ments of the covariance matrix. The EXAFS data were Fou-
−1
rier-transformed from k-space (30–135 nm ) to R-space.
The empirical backscattering amplitudes and phase shifts for
Pd–Pd, Pd–O, and Pd–S pairs were extracted from EXAFS
data for Pd foil, PdO, and PdS, respectively. Table1 presents
the EXAFS parameters for the reference compounds.
%
ee =100×
(
[
S
]
−
[
R
]
)
/
(
[
S
]
+
[
R ,
]
)
where [ S] and [ R] represent the concentrations of S- and
R-enantiomers, respectively.
2
.3 Catalyst Characterizations
Table 1 EXAFS parameters for the reference compounds
Five percent Pd/C and Pd/Al O catalysts were character -
−3
2
3
Reference
Shell
CN
R (nm)
∆E (eV)
σ (10 nm)
ized by use of CO adsorption and EXAFS techniques. The
dispersion of Pd metal particles on the catalyst was evalu -
ated on the basis of CO adsorption capacity using a pulse
technique [16, 27]. A wet Pd/C catalyst was dried in air at
Pd foil
PdO
Pd–Pd
Pd–O
Pd–S
12
4
0.2770
0.2018
0.2333
0
0
0
6.0
6.0
6.0
PdS
4
3
5
73 K for 24 h. A certain amount of the dried Pd/C (20–
0 mg) was placed in a U-shaped tube and purged with He
CN coordination number, R interatomic distance, σ Debye–Waller
factor, ∆E inner potential
1
23

EXAFS Characterization of Pd Catalysts for Enantioselective Hydrogenation of α-Phenylcinnamic Acid:…
2433
3
Results and Discussion
(phase shift: uncorrected), assigned to Pd–O and Pd–Pd
metal) atomic pairs, respectively, by comparing with the
(
3
.1 Characterization of Pd Catalysts by CO Adsorption
and EXAFS
FT features of PdO and Pd metal. Tables 3 and 4 summa-
rize the structural parameters as derived from the EXAFS
analysis for Pd/C (STD) and Pd/Al O , respectively. With
2
3
The dispersion of Pd was evaluated by the CO adsorption
capacity. Table 2 summarizes the dispersion of Pd metal on
Pd/C and Pd/Al O . Pd/C (STD) shows the highest disper -
as-received Pd/C (STD), the presence of Pd–O shows that
oxidized Pd species, which are probably assigned to PdO-
like species from the Pd–O bond distance of 0.200–0.20 2n m
2
3
sion, 0.60, which is consistent with the previous results (0.57 compared with 0.202 nm for PdO in Table 1, are present
and 0.59) [16, 27], while Pd/Al O the lowest Pd dispersion,
as well as metallic Pd. Previous XPS characterization of
the catalyst also demonstrated the coexistence of metallic
(40%) and PdO-like species (60%) on Pd/C (STD) [27].
However, Table 3 shows that when Pd/C (STD) has been
2
3
0
.41. It is noted that the catalysts used here have high dis -
persions of Pd metal. The Pd metal particle sizes estimated
assuming a spherical particle are shown in Table2. With the
distribution of Pd inside the catalyst particles, Pd/C (STD)
is a commercial catalyst nominally having a uniform distri-
bution of Pd, whereas Pd/C (AER) is characterized by an
eggshell type distribution. These Pd distributions were sub-
stantiated by XPS [16, 27].
pretreated with H at rt for 30 min under rigorous stirring,
2
PdO-like species are completely removed and only metallic
Pd particles are formed, indicating that the reduction of Pd
is completed even by the H -treatment at rt [Pretreatment
2
(C)]. When Pd/C (STD) has been pretreated at 353 K with
3
Figure 1 shows representative k -weighted Pd K-edge
H , the oxidized Pd species are completely reduced to form
2
EXAFS oscillations χ(k) and Fourier transforms (FTs) for
the Pd catalysts as received or after the pretreatments. Two
characteristic peaks are observed around 0.16 and 0.25 nm
metallic Pd [Pretreatment (A)]. As shown in Table 3, the
Pd–Pd coordination number (CN) for Pd metal particles in
Pd/C (STD) increases from 8.6 to 10 as the H 2-treatment
temperature increases from rt to 353 K, suggesting a slight
sintering of Pd metal particles at the higher temperature. No
extra peaks due to Pd–C nor Pd–N bonding were observed
by the addition of CD to the reduced catalyst (Table3). This
is due to only a small fraction of surface Pd atoms bonded
with adsorbed CD, possible disorder of Pd–C or Pd–N
bonds, and/or high mobility of adsorbed CD.
With Pretreatment (B), we examined the effect of the
first step of the procedure, evacuation. Table 3 shows that
PdO-like species on Pd/C (STD) remain intact after the
evacuation, followed by the solvent-injection. Similarly, the
subsequent solvent injection and evacuation procedures in
Table 2 Dispersion of Pd metal particles on 5% Pd catalysts as esti-
mated by CO adsorption
Catalyst
Dispersion of
Pd metal
Pd metal particle
size (nm)
a
Pd/C (STD)
Pd/C (AER)
Pd/C (BNA)
0.60
0.49
0.47
0.41
1.9
2.3
2.4
2.8
Pd/Al O3
2
aCalculated from the dispersion of Pd metal assuming spherical par-
ticles
3
Fig. 1 k -Weighted Pd K-edge
EXAFS oscillations χ(k) A and
3
k χ(k) Fourier transforms B for
Pd/C and Pd/Al O . (a) Pd/C
2
3
(
(
(
STD) as received, (b) Pd/C
STD) after H -treatment at rt,
2
c) Pd/C (STD) after H -treat-
2
ment at 353 K, (d) Pd/C (STD)
after evacuation/dioxane/evacu-
ation treatment, (e) Pd/Al O
2
3
after H -treatment at rt, and (f)
2
Pd/Al O after H -treatment at
2
3
2
3
53 K. The Fourier transforms
of PdO and Pd metal foil are
shown for comparison. Dotted
lines present the best fitting
curves in the EXAFS analysis
123

2
434
T. Kubota et al.
Pretreatment (B) cause no noticeable change in the chemi -
cal states of Pd in Pd/C (STD) despite brief flashings with
H twice. It is considered that a brief contact with H under
static conditions does not lead to reduction of PdO-like spe-
cies. The effect of the solvent, dioxane or toluene, used as
Pd–Pd bond distance (0.280 ± 0.001 nm) is slightly longer
than that for Pd metal (0.277 nm), while after reduced at
rt, it is almost identical with. It is well established that the
metal–metal bond distance becomes shorter for nm-sized
metal particles [ 30–32]. Accordingly, the enlarged Pd–Pd
bond distance is attributed to the dissolution of H-atoms into
the metal phase to partially form subsurface Pd-hydride [3 3,
2
2
the “co-evaporation” medium does not affect the Pd spe
-
cies in the catalyst at all. Thus it is concluded that Pretreat -
ment (B) does not modify the Pd species in Pd/C (STD) and 34]. Comparisons of the Pd K-edge X-ray absorption near-
that the catalyst is reduced only when H is introduced just
edge structures (XANES) for Pd/C (STD) and Pd/Al 2O3
2
before the reaction as in the case of Pretreatment (C).
H -pretreated at rt and 353 K (Supplementary materials,
2
5
% Pd/Al O as received shows the presence of oxidized
Figure S-1) showed no notable changes, suggesting that the
amount of dissolved hydrogen into Pd metal particles was
2
3
and metallic Pd species, as presented in Table 4. Taking into
consideration the relative CNs of Pd–O and Pd–Pd shells, the
fraction of oxidized Pd species, assigned to PdO-like species
from the Pd–O atomic distance, is considered to be smaller
in Pd/Al O compared with Pd/C (STD). But, the proportion
small because of the low H -pressure applied (1 atm). It is
2
considered from Table 3 that the Pd-hydride phase is hardly
affected by the adsorption of CD on Pd/C (STD).
2
3
of PdO-like species may depend on the duration and atmo -
sphere of storage of the catalyst after preparation. The oxi -
dized Pd species in Pd/Al O are reduced to metallic Pd even
3.2 Effects of the Catalyst Pretreatment on the
Performance in Enantioselective Hydrogenation
2
3
at rt for 30min, as observed for Pd/C (STD). When pretreated Table 5 summarizes the effects of the pretreatment on the
at 353 K under H [Pretreatment (B)], metallic Pd is formed, reaction rate and enantioselectivity over Pd/C and Pd/Al O
3
2
2
accompanying the increase in Pd–Pd CN from 11 to 12 due to catalysts. Before discussion, it should be noted that the reac-
sintering of Pd metal particles by the heat-treatment. The CNs tion results cited in Table 5 are from our group and, accord-
of the Pd–Pd shell for Pd/Al O are slightly larger at the both
ingly, can be directly compared each other and with the
present EXAFS results, as a function of the pretreatment
explored by our group. We first examined the effect of a
2
3
treatment temperatures than the corresponding CNs for Pd/C
STD), suggesting that the Pd metal dispersion in Pd/AlO is
(
2
3
lower than that in Pd/C (STD). This is consistent with the CO H -pretreatment at 353 K, as employed in Pretreatment (A),
2
adsorption capacity in Table 2.
on the hydrogenation activity of Pd/C (STD) without CD
addition but in the presence of BA. The activity decreases
from 120 mmol/gh to 96 mmol/gh by the pretreatment at
353 K compared to a H 2-pretreatment at rt [Pretreatment
There is a tendency in Tables 3 and 4 that the Pd–Pd
bond distance of the catalyst depends on the pretreatment
irrespective of the support: after H 2-treated at 353 K, the
Table 3 Pd K-edge EXAFS analysis of 5% Pd/C (STD): effects of the pretreatments
−
3
Pretreatment
As-received
Shell
CN
R (nm)
∆E (eV)
σ (10 nm)
R (%)
f
Pd–O
Pd–Pd
Pd–Pd
Pd–Pd
Pd–Pd
Pd–O
Pd–Pd
Pd–O
Pd–Pd
Pd–O
Pd–Pd
2.2± 0.3
0.2001± 0.0010
0.2792± 0.0006
0.2774± 0.0002
0.2795± 0.0002
0.2806± 0.0002
0.2005± 0.0009
0.2784± 0.0006
0.2000± 0.0008
0.2789± 0.0007
0.2017± 0.0008
0.2778± 0.0006
−5.2± 2.5
4.8± 1.5
−4.5± 0.5
5.6± 0.5
3.9± 0.6
−1.6± 2.3
8.3± 1.5
−1.6± 2.2
8.3± 1.6
0.96± 2.0
9.0± 1.7
6.3± 1.7
7.1± 1.0
7.0± 0.4
7.4± 0.4
7.9± 0.4
6.1± 1.7
7.1± 1.1
6.1± 2.1
6.8± 1.1
6.2± 1.6
7.0± 1.2
3.5
3.6± 0.5
8.6± 0.5
10.1± 0.5
9.9± 0.6
2.5± 0.3
3.5± 0.5
2.1± 0.3
3.4± 0.6
2.6± 0.3
3.1± 0.5
(
C) H at rt
0.27
1.3
1.1
3.7
2
(
A) H at 353 K
2
CD-adsorptiona
Evacuationb
(
B) Evacuation/dioxane/evacuationc
B) Evacuation/toluene/evacuationc
3.5
7.7
(
CN coordination number, R interatomic distance, σ Debye–Waller factor, ∆E inner potential, R R-factor
f
aCD was injected to Pd/C pretreated at 353 K under H
2
bEvacuated for 10 min, followed by injection of wet dioxane
c
Evacuated for 10 min, followed by quick H -flashing, injection of dioxane or toluene, and then quick H -flashing and finally evacuated again
2
2
before injection of wet dioxane
1
23

EXAFS Characterization of Pd Catalysts for Enantioselective Hydrogenation of α-Phenylcinnamic Acid:…
2435
(
C)]. The activity decrease is ascribable to sintering of
particles, with the highest selectivity being observed around
20–25% dispersion [10], Pd/C catalysts showed high enan -
tioselectivity over a wide range of Pd dispersion [27]. Thus,
Pd metal particles, as revealed by the EXAFS analysis in
Table 3.
The reaction rates are compared in Table5 for the hydro-
genation of PCA without additions of CD and BA after the
it is considered that the increase in the selectivity by Pre
-
treatment (A) is not attributed to a slight increase in the size
of Pd metal particles. Instead, we tentatively propose from
these findings that contaminants on Pd metal surface may
H -treatment at 353 K. The activity of Pd/C (AER) is con -
2
siderably higher than Pd/C (STD) or Pd/Al O , when the
2
3
dispersion of Pd is taken into account. It is surmised that the be removed by H -treatment at 353 K, providing “cleaner”
2
eggshell-type distribution of Pd in Pd/C (AER) is favorable
for the PCA hydrogenation under the present reaction cond-i
Pd metal surface for CD-modification, although XPS analy-
sis of Pd/C (STD) could not detect the presence of chlorine
tions probably due to a smaller extent of diffusion limitation. or sulfur on the catalyst surface ( <0.2%) (Supplementary
The BA addition apparently reduces the reaction rate for all
the catalysts. However, the ratio of the reaction rates, (with
BA addition)/(without BA addition), is strongly affected by
the support used: 0.3–0.4 for Pd/C vs. 0.6 for Pd/AlO . It is
materials, Figure S-2). The effects of possible removable of
water from mesopores at 353 K cannot be ruled out, though
[14]. It was demonstrated by ATR-IR that CD molecules are
adsorbed on Pd metal surface in a parallel geometry through
2
3
considered that Pd/Al O is more resistant than Pd/C against aromatic π-bonding of the quinoline moiety as well as tilted
2
3
the catalyst poisoning by BA due to difference in the acid -
adsorption configuration via the N-lone pair of the quino-
line ring [21, 37]. It is suggested that the parallel adsorption
of CD provides more favorable configurations for enantio-
differentiation in PCA-CD interaction intermediates than the
tilted configuration [1, 38, 39]. It is readily surmised that the
parallel adsorption of CD on Pd metal surface via π-bonding
of the quinoline ring needs a much larger ensemble of sur -
face Pd metal atoms and therefore the formation of such an
adsorption configuration of CD is very sensitive to a small
extent of surface contamination. As a consequence, cleaner
ity of the support; Al O has a weak acidic property, facili -
2
3
tating the adsorption of basic BA more preferentially than
activated carbon to protect Pd metal from poisoning by BA.
It is noted in Table 5 that Pretreatment (A) improves the
performance of Pd/C (STD) with respect to Pretreatment
(C), yielding much higher selectivity and reaction rate [ 13].
Scrutinizing the details, after Pretreatment (C) the reaction
rate is significantly decreased from 120 mmol/gh to 47 or
44 mmol/gh by the modification of Pd/C (STD) with CD for
the enantioselective hydrogenation of PCA, with enantiose -
lectivity being moderate (46 or 58%ee). On the other hand,
the hydrogenation rate is not changed (96 mmol/gh vs. 93
or 102 mmol/gh) by the CD-modification after Pretreatment
Pd metal surface, possibly produced at 353
K, provides
enhanced enantioselectivity and a larger ligand acceleration
effect. With Pd/C (AER), Pretreatment (A) is also very effec-
tive in the enantioselective hydrogenation of PCA: the reac-
tion rate and selectivity are significantly increased compared
with Pretreatment (C). We can apply the same discussions to
the effect of the pretreatment of Pd/C (AER) as Pd/C (STD).
However, with Pd/Al O , Pretreatment (A) shows only a
(A) at 353 K, although the selectivity becomes very high
81–82%ee. The decrease in the reaction rate on Pd/C (STD)
after Pretreatment (C) clearly shows that CD adsorbed on Pd
metal surface plays contradictory roles; generation of enanti-
oselectivity and poisoning of the hydrogenation. It is consid-
ered that Pretreatment (A) significantly enhances the former
role over the latter one, accompanying ligand acceleration
2
3
marginal increase in the enantioselectivity from 34%ee to
39%ee with respect to Pretreatment (C), accompanying a
decrease in the reaction rate due to sintering of Pd as sub
stantiated by EXAFS (Table 4). It is considered that the
-
[29, 35, 36] caused by CD-PCA interactions favorable for
highly enantioselective hydrogenation. The EXAFS analysis
in Table 3, however, shows a slight sintering of Pd metal
particles by Pretreatment (A) at 353 K with respect to Pre -
H -treatment temperature of 353 K is not high enough to
2
eliminate contaminants from Pd metal surface on Pd/Al O .
2
3
With Pd/TiO , which was prepared by a precipitation method
2
treatment (C). In contrast to Pd/TiO , which showed strong
using PdCl and Na CO , showed an optimum performance
2
2
2
3
dependency of the enantioselectivity on the size of Pd metal
when H -treated at 473 K [10].
2
Table 4 Pd K-edge EXAFS analysis of 5% Pd/Al O
2
3
−
3
Pretreatment
As received
Shell
CN
R (nm)
∆E (eV)
σ (10 nm)
R (%)
f
Pd–O
1.9± 0.3
5.8± 0.5
10.5± 0.5
12.0± 0.5
0.2010± 0.0013
0.2775± 0.0004
0.2793± 0.0002
0.2785± 0.0002
−5.6± 3.5
4.2± 0.9
0.45± 0.52
9.4± 0.5
7.1± 2.5
7.4± 0.6
7.3± 0.3
7.6± 0.3
2.9
Pd–Pd
Pd–Pd
Pd–Pd
(
C) H at rt
1.0
1.4
2
H at 353 K
2
CN coordination number, R interatomic distance, σ Debye–Waller factor, ∆E inner potential, R R-factor
f
123

2
436
T. Kubota et al.
Table 5 Reaction rate and
enantioselectivity of the hydro-
genation of PCA over
Pretreatment
5% Pd/C (STD)
5% Pd/C (AER)
5% Pd/Al O
2 3
r (mmol/gh)
ee (%)
r (mmol/gh)
ee (%)
r (mmol/gh)
ee (%)
CD-modified 5% Pd catalysts
H at rt (CD=0)
120
96 (252)a
47, 44b
–
–
–
–
2
–
105 (360)a
74 (124)a
12b
8b
–
H at 353 K (CD=0)
2
58, 46b
78 , 66
b
c
59 , 41
81d
b
c
34b
39b
(
C) H at rt
2
b
c
b
c
102d
H at 353 K
102 , 93
82 , 81
2
Evacuation
45e
40e
60e
72e
Evacuation/dioxane/
evacuation
51d
76d
aThe reaction rate in the parentheses was measured without addition of BA
bRef [13]
cRef [27]
dRef [15]
eRef [14]
As revealed by the EXAFS analysis in Table 3, the co-
evaporation treatments with quick H 2-flashings (Pretreat-
ment (B)) do not essentially modify the chemical state of
Pd in as-received Pd/C (STD). However, the selectivity is
considerably improved from 46 or 58%ee to 72%ee after
Pretreatment (B). Thus, it is estimated that Pretreatment (B)
summarized in Table 6. As for t-BuSH, the CN of the Pd–S
shell increases from 1.2 to 2.2 as the adsorption temperature
is increased from rt to 353 K, accompanying a decrease in
the CN of the Pd–Pd shell due to Pd metal from 8.6 to 3.2.
The decrease in Pd–Pd CN may be alleviated to some extent
by possible sintering of the metal particles by the tempera -
may partially eliminate weakly adsorbed contaminants from ture increase. As for the other thiols, essentially identical
Pd/C (STD). As suggested previously [ 14], it is also likely
that Pretreatment (B) increases the catalyst performance by
tendencies in the structural parameters are observed as pre -
sented in Table 6. The EXAFS analysis of Pd/C (AER) and
(BNA) catalysts exposed to the sulfur compounds showed
similar changes in structural parameters as the exposure or
adsorption temperature was increased from rt to 353K (Sup-
plementary materials, Tables S-1, S-2). Table 7 presents the
structural parameters for t-BuSH-Pd/Al O as derived from
removing H O from the micropores of Pd/C. Although no
2
EXAFS data are available, we may suggest the same causes
for the selectivity increase of Pd/C (AER) by Pretreatment
(B) as presented in Table 5.
2
3
3
.3 3–3 Adsorption of Thiols
EXAFS analysis. As observed for Pd/C catalysts, the Pd–S
CN increases as the adsorption temperature is increased up
to 353 K, accompanying a decrease in the Pd–Pd CN. How-
ever, the Pd–S CNs for Pd/Al O are much smaller than
those for Pd/C (STD). With the other thiols, similar tenden-
cies are observed as presented in Table 7.
As mentioned in the introduction, we tried to obtain some
insights into the effects of Pd distribution in the catalyst
particles on the enantioselectivity [16] from the viewpoints
of diffusion of the substrate and/or modifier by combining
EXAFS techniques and sulfur compounds as a probe mol -
ecule having a varied molecular size. It is surmised from
Table 5 that the eggshell-type distribution of Pd in Pd/C
2
3
The growth of the Pd–S shell suggests the formation of
a PdS phase and/or the adsorption of the sulfur compound
and/or adsorption of sulfur atoms on Pd metal surface. The
Pd–S bond distances in Tables 6 and 7 (and in Tables S- 1
and S-2) are 0.236 ± 0.001 nm irrespective of the catalyst
and the sulfur compound. The Pd–S bond distance is slightly
longer than that of PdS (0.233 nm in Table 1), negating a
predominant formation of a bulk PdS phase under the pres-
ent adsorption conditions. It seems difficult to differenti-
ate adsorbed thiols and atomic sulfur from the Pd–S bond
distance alone. We conducted qualitative chemical analy -
sis of the solutions by use of Gas chromatography-Mass
spectrometry techniques after the EXAFS measurements
at 353 K. With t-BuSH, 2-methyl propene was observed
to form as a product after the exposure to the Pd catalysts.
(AER) is suitable for the PCA hydrogenation probably due
to a smaller extent of diffusion limitation. Figure 2 shows
the Pd K-edge k χ(k)-EXAFS oscillations and FTs for Pd/C
3
(
3
STD) exposed to t-BuSH at increasing temperature (rt,
33 K, and then 353 K) together with the FT for PdS as
a reference. With t-BuSH-Pd/C (STD), a new peak around
.19 nm (phase shift: uncorrected) grows as the tempera -
0
ture is increased, accompanying a gradual decrease in inten-
sity of the FT peak due to Pd metal. Comparing with the
FT peaks for PdS, the new FT peak is readily assigned to
a Pd–S shell. With the other thiols, MePhSH, n-C 16H SH,
or 2-NapSH, we observed similar changes in the FTs. The
33
structural parameters of Pd/C (STD) exposed to the thiols are With MePhSH, toluene was found to form as a product.
1
23

EXAFS Characterization of Pd Catalysts for Enantioselective Hydrogenation of α-Phenylcinnamic Acid:…
2437
3
Fig. 2 k -Weighted Pd K-edge
EXAFS oscillations χ(k) (A)
3
and k χ(k) Fourier transforms
(
B) for Pd/C and Pd/Al O
2 3
exposed to a thiol under atmo-
spheric pressure of H . (a) Pd/C
2
(STD) exposed to t-BuSH at
rt, (b) Pd/C (STD) subsequent
temperature increase to 333 K,
(c) Pd/C (STD) subsequent tem-
perature increase to 353 K, (d)
Pd/C (AER) exposed to t-BuSH
up to 353 K, (e) Pd/C (BNA)
exposed to t-BuSH up to 353 K,
and (f) Pd/Al O exposed to
2
3
t-BuSH up to 353 K. The Fou-
rier transform of PdS is shown
for comparison. Dotted lines
present the best fitting curves in
the EXAFS analysis
Table 6 Pd K-edge EXAFS analysis of 5% Pd/C (STD) exposed to a sulfur compound: effect of adsorption temperature
−
3
Sulfur compound
Adsorption
temp. (K)
Shell
CN
R (nm)
∆E (eV)
σ (10 nm)
Rf (%)
Before adsorption
t-BuSH
Pd–Pd
Pd–S
8.6± 0.5
1.2± 0.5
7.1± 0.6
1.6± 0.5
5.4± 0.6
2.2± 0.4
3.2± 0.7
1.7± 0.3
6.7± 0.7
1.9± 0.3
5.1± 0.5
2.1± 0.3
4.3± 0.6
1.4± 0.6
7.3± 0.6
1.6± 0.4
6.3± 0.6
1.8± 0.3
5.9± 0.6
1.5± 0.3
6.8± 0.6
1.8± 0.3
5.5± 0.6
1.9± 0.3
5.1± 0.7
0.2774± 0.0002
0.2350± 0.0039
0.2774± 0.0003
0.2352± 0.0024
0.2772± 0.0004
0.2345± 0.0018
0.2772± 0.0010
0.2355± 0.0021
0.2784± 0.0004
0.2371± 0.0018
0.2770± 0.0005
0.2345± 0.0016
0.2768± 0.0006
0.2357± 0.0033
0.2771± 0.0003
0.2370± 0.0020
0.2770± 0.0004
0.2368± 0.0018
0.2771± 0.0004
0.2352± 0.0022
0.2776± 0.0004
0.2369± 0.0020
0.2770± 0.0005
0.2368± 0.0019
0.2771± 0.0006
−4.5± 0.5
−3.8± 6.8
−4.3± 0.8
−3.3± 4.2
−4.3± 1.1
−8.9± 3.5
−9.8± 2.3
−4.5± 4.2
−4.2± 0.9
−3.7± 3.7
−6.8± 1.2
−8.7± 3.3
−7.9± 1.5
−4.4± 5.6
−4.0± 0.7
−3.9± 3.9
−6.3± 0.9
−4.3± 3.7
−6.6± 1.0
−5.1± 4.4
−3.9± 0.9
−4.0± 3.7
−6.5± 1.1
−4.4± 3.5
−7.1± 1.3
7.0± 0.4
7.6± 6.0
8.5± 0.6
7.6± 4.6
7.6± 0.8
7.4± 3.4
8.7± 1.7
7.9± 5.5
8.2± 0.7
7.2± 4.9
7.6± 0.9
7.1± 4.1
7.1± 1.1
8.1± 5.7
7.4± 0.6
6.9± 5.6
7.3± 0.7
6.9± 5.9
7.7± 0.8
7.1± 6.1
7.8± 0.7
6.8± 4.5
7.6± 0.9
6.8± 4.0
8.4± 1.0
0.27
0.45
rt
Pd–Pd
Pd–S
3
33
53
0.45
1.5
Pd–Pd
Pd–S
3
Pd–Pd
Pd–S
MePhSH
rt
0.59
0.76
1.0
Pd–Pd
Pd–S
3
33
Pd–Pd
Pd–S
3
53
Pd–Pd
Pd–S
n-C H SH
rt
0.43
0.61
0.58
0.47
0.55
0.55
16
33
Pd–Pd
Pd–S
3
33
Pd–Pd
Pd–S
3
53
Pd–Pd
Pd–S
2
-NapSH
rt
Pd–Pd
Pd–S
3
33
Pd–Pd
Pd–S
3
53
Pd–Pd
Adsorption temperature was increased stepwise and kept for 30 min at each temperature, followed by EXAFS measurements
CN coordination number, R interatomic distance, σ Debye–Waller factor, ∆E inner potential, R R-factor
f
123

2
438
T. Kubota et al.
Table 7 Pd K-edge EXAFS analysis of 5% Pd/Al O exposed to a sulfur compound: effect of adsorption temperature
2
3
−
3
Sulfur compound
Adsorption
temp./K
Shell
CN
R (nm)
∆E (eV)
σ (10 nm)
R (%)
f
Before adsorption
t-BuSH
Pd–Pd
Pd–S
10. ± 0.5
0.2793± 0.0002
0.2360± 0.0053
0.2835± 0.0003
0.2340± 0.0041
0.2845± 0.0004
0.2354± 0.0036
0.2841± 0.0005
0.2373± 0.0049
0.2775± 0.0003
0.2354± 0.0035
0.2777± 0.0003
0.2340± 0.0028
0.2776± 0.0004
0.2352± 0.0051
0.2802± 0.0004
0.2345± 0.0039
0.2827± 0.0004
0.2347± 0.0034
0.2788± 0.0004
0.2344± 0.0050
0.2812± 0.0004
0.2340± 0.0033
0.2833± 0.0004
0.2337± 0.0028
0.2829± 0.0005
0.45± 0.52
−4.4± 10
−4.5± 0.7
−8.7± 7.6
−4.5± 0.8
−5.6± 6.5
−4.6± 1.1
−3.4± 8.8
−4.7± 0.6
−5.4± 6.7
−4.7± 0.7
−8.1± 5.2
−5.8± 0.9
−8.1± 9.6
−5.2± 0.8
−7.5± 7.4
−4.5± 0.8
−7.0± 6.4
−5.5± 0.9
−8.5± 8.8
−5.5± 0.8
−8.6± 6.4
−4.9± 0.9
−8.3± 5.2
−4.9± 1.1
7.3± 0.3
7.9± 9.6
8.4± 0.5
7.6± 6.3
8.4± 0.6
7.8± 5.5
9.6± 0.8
6.8± 10
8.2± 0.5
6.4± 8.6
8.2± 0.6
6.3± 5.3
7.7± 0.7
6.3± 9.5
8.9± 0.6
6.3± 7.2
8.8± 0.6
6.6± 6.2
8.3± 0.6
8.0± 7.3
8.5± 0.6
7.8± 6.1
8.5± 0.6
7.7± 4.6
9.2± 0.8
1.0
5
rt
0.8± 0.4
8.6± 0.7
0.9± 0.4
7.8± 0.7
1.1± 0.5
7.9± 0.9
0.7± 0.5
9.3± 0.6
1.0± 0.4
8.3± 0.7
1.2± 0.4
6.1± 0.6
0.7± 0.4
9.3± 0.8
0.8± 0.4
9.0± 0.7
1.0± 0.4
7.5± 0.7
0.9± 0.5
8.3± 0.7
1.2± 0.4
7.5± 0.7
1.4± 0.4
7.0± 0.8
0.55
Pd–Pd
Pd–S
3
3
33
53
0.62
1.1
Pd–Pd
Pd–S
Pd–Pd
Pd–S
MePhSH
rt
0.28
0.42
0.40
0.52
0.76
0.48
0.51
0.67
0.94
Pd–Pd
Pd–S
3
3
33
Pd–Pd
Pd–S
53
Pd–Pd
Pd–S
n-C H SH
rt
16
33
Pd–Pd
Pd–S
3
33
Pd–Pd
Pd–S
3
53
Pd–Pd
Pd–S
2
-NapSH
rt
Pd–Pd
Pd–S
3
33
Pd–Pd
Pd–S
3
53
Pd–Pd
Adsorption temperature was increased stepwise and kept for 30 min at each temperature, followed by EXAFS measurements
CN coordination number, R interatomic distance, σ Debye–Waller factor, ∆E inner potential, R R-factor
f
n-C H SH reacted to form C 16H34 and 2-NapSH to give
catalysts during the exposure up to 353 K. Taking into con-
1
6
33
naphthalene. These products clearly suggest that desulfur -
ization of the sulfur compounds takes place over the Pd
sideration the presence of atmospheric pressure of H , it is
2
considered that following hydrodesulfurization reactions
occur even at 353 K during the adsorption of the thiols on
the Pd catalysts.
Table 8 Pd–S coordination numbers as derived from EXAFS analysis
and those estimated from the dispersion of Pd and the formation of a
S-monolayer
R −SH + H → RH + H S
2
2
Catalyst
Pd–S CNa
Pd–S
CN
b
Hydrogen sulfide thus formed by the reaction readily
decomposes to atomic sulfur and/or SH on Pd metal sur
t-BuSH MePhSH n-C H SH 2-NapSH
1
6
33
-
Pd/C (STD) 2.2
Pd/C (AER) 1.6
Pd/C (BNA) 1.8
2.1
1.8
1.8
1.2
1.8
1.4
1.6
1.0
1.9
–
1.9
1.6
1.5
1.3
face. It was suggested by DFT that the Pd–S bond distance
strongly depended on thiol and its adsorption mode: For
instance, 0.228 nm and 0.239nm for dissociative adsorption
–
of CH SH and thiophene-2-thiol, respectively [40]. In con-
Pd/Al O3
1.1
1.4
3
2
aPd–S coordination number as derived from EXAFS analysis of the
Pd catalyst after exposure to the sulfur compound at 353K for 30min
bPd–S coordination number estimated from the dispersion of Pd and
trast, the Pd–S bond distance is independent of the thiols in
the present study, as noted above. Thus, it is concluded that
the sulfur species on the Pd catalysts detected by EXAFS
are attributed to adsorbed atomic sulfur or SH species rather
than adsorbed thiol molecules.
a S-monolayer, assuming 83% of Pd(111) (Pd–S CN=3) and 17% of
Pd(100) (Pd–S CN=4)
1
23

EXAFS Characterization of Pd Catalysts for Enantioselective Hydrogenation of α-Phenylcinnamic Acid:…
2439
Table 8 compares the Pd–S CNs obtained after the expo-
sure of the Pd catalysts to the sulfur compounds at 353 K.
Obviously, the Pd–S CN varies with the catalyst and the su-l
fur compound used. Before discussing the results in Table8,
it is considered that the Pd–S CN is affected by (1) the size
of Pd metal particles or dispersion of Pd metal, (2) the reac-
tivity of the sulfur compounds toward Pd metal, (3) the dis-
molecular diffusion of the latter two thiols having rather
larger molecular sizes. The latter possibility may explain the
favorable eggshell-type Pd-distribution of Pd/C (AER) for
the enantioselective hydrogenation of PCA [16].
As for Pd/Al O in Table 8, the Pd–S CNs derived
2
3
from the EXAFS analysis are rather lower than or compa -
rable to the Pd–S CN estimated from the dispersion of Pd
tribution of Pd metal inside the catalyst particles, and (4) the metal even with t-BuSH and MePhSH. It is surmised that
nature of the support.
Pd/Al O is more resistant than Pd/C against sulfur poison-
2 3
First, we discuss the effect of the metal dispersion on the
Pd–S CN. The adsorption of sulfur atoms on Pd metal was
extensively studied until now [ 41, 42]. With Pd(111) sur -
face, it was reported that adsorbed sulfur atoms formed a
ing as observed for BA poisoning (Table 5). It is considered
that thiols are more preferentially adsorbed on Al O sur-
face to reduce the concentration of the thiols in the solution
2
3
and/or that Al O modifies the electronic properties of Pd
2
3
(√3 × √3) structure and occupied fcc hollow sites [43, 44].
metal particles.
The Pd–S CN is thus 3 for the sulfur adsorption on Pd(111)
at a monolayer coverage. With Pd(100), formation of a
c(2×2) sulfur structure was observed at a monolayer cover-
age [45, 46]. The Pd–S CN is calculated to be 4 for Pd(100).
Assuming the formation of cubo-octahedron particles, the
ratio of Pd(111) and Pd(100) facets is calculated to be about
4 Conclusions
In the present study, we first tried to get deeper insights
from the structural point of view by use of EXAFS tech
niques into the pretreatment effects on the enantioselec
-
-
5
for Pd metal particles having a size of >1 nm from the
structural model by Hardeveld and Hartog [ 47]. Thus the
average Pd–S CN is estimated to be 3.2 for surface Pd atoms Pd/Al O . It is suggested that the enhanced performance
tive hydrogenation of PCA over CD-modified Pd/C and
2
3
at a monolayer coverage for Pd metal particles with a size
of >1 nm. Assuming the monolayer formation of adsorbed
sulfur at 353 K, the Pd–S CNs were estimated from the dis-
persion of Pd metal particles in Table 2. They are presented
by the H -pretreatment at 353 K is ascribed to the elimina -
tion of contaminants and/or water from Pd/C. It is shown
2
that subsurface Pd hydride is formed when H -pretreated at
2
353 K in contrast to the pretreatment at rt. Pd/Al O shows
2
3
in Table8. The Pd CNs observed by EXAFS agree well with a higher resistance in the hydrogenation against BA addi -
those estimated above. As a consequence, it is concluded
that the great difference in the observed Pd–S CN among
the catalysts is primarily attributed to the difference in the
dispersion of the metal particles in the catalyst. It is also
concluded that surface sulfur structures close to the mono -
tion than Pd/C. Eggshell-type Pd/C is more effective for the
hydrogenation than Pd/C with a uniform distribution of Pd.
We used thiol adsorption combined with EXAFS technique
to clarify this point. It has been found that thiol is decom -
posed on Pd metal at 353 K in the presence of H 2, leav-
layer are formed after the exposure of the Pd catalysts to the ing sulfur atoms on the surface. The Pd–S CN as derived
sulfur compound up to 353K under atmospheric pressure of
from EXAFS analysis is consistent with the CN estimated
by use of Pd dispersion at a monolayer adsorption of sulfur.
A slight deviation of Pd–S CN from the estimation is related
to the difference in the reactivity and, possibly, diffusivity
of thiols used. It is suggested that Pd/AlO is more resistant
H and in wet dioxane.
2
The reactivity of the sulfur compound toward Pd metal
particles can be discussed with Pd/C (AER), in which Pd
metal particles form an eggshell-type distribution [
since the effect of diffusion of the sulfur compound in
16],
2
3
against thiol adsorption than Pd/C.
catalyst pore systems is minimized there. Table 8 suggests
that the Pd–S CNs for t-BuSH and MePhSH are slightly
higher than that estimated above, while the Pd–S CN for
n-C H SH agrees with the estimated one. It may be sug -
gested that even subsurface Pd sulfides are slightly formed
with the former thiols, in contrast to a monolayer formation
with the latter thiol. With Pd/C (STD) and (BNA) having a
homogeneous distribution of Pd within the catalyst particles
References
1
6
33
1
2
3
4
. Mallat T, Orglmeister E, Baiker A (2007) Chem Rev 107:4863
. E. Zhan, C. Chen, Y. Li, W. Shen (2015) Catal Sci Technol 5:650
. Studer M, Blaser HU, Exner C (2003) Adv Synth Catal 345:45
. D. Y. Murzin, P. Maki-Arvela, E. Toukoniitty, T. Salmi (2005)
Catal Rev Sci Eng 47:175
5
6
7
8
. Sugimura T (1999) Catal Surv Jpn 3:37
. Nitta Y (2000) Top Catal 13:179
. Tungler, E., Sipos, H. Hada (2006) Curr Org Chem 10:1569
. Sugimura T (2008) In: Ding K, Uozumi Y (eds) Handbook of
asymmetric heterogeneous catalysis. Wiley, Weinheim, p 357
. Nitta Y, Ueda Y, Imanaka T (1994) Chem Lett 23:1095
[
16], similar tendencies in the Pd–S CNs are observed in
Table 8. 2-NapSH also forms a sulfur monolayer on Pd/C
STD) under the present conditions. These findings may
(
be correlated to higher reactivity of t-BuSH and MePhSH
than n-C H SH or 2-NapSH and, possibly, to more limited
9
1
6
33
123

2
440
T. Kubota et al.
1
0. Nitta Y, Kubota T, Okamoto Y (2000) Bull Chem Soc Jpn 73:2635 29. Mameda T, Nakai K, Misaki T, Okamoto Y, Sugimura T (2015)
11. Nitta Y, Kobiro K, Okamoto Y (1997) In: Blaser HU, Baiker A,
Catal Today 245:129
Prins R (eds) Heterogeneous catalysis and fine chemicals IV.
Elsevier, Amsterdam, p 191
30. Frenkel AI, Hills CW, Nuzzo RG (2001) J Phys Chem B
105:12689
1
1
2. Nitta Y, Okamoto Y (1998) Chem Lett 27:1115
3. Nitta Y, Watanabe J, Okuyama T, Sugimura T (2005) J Catal
31. Nepijko SA, Klimenkov M, Adelt M, Kuhlenbeck H, Schlögel
SR, Freund H-J (1999) Langmuir 15:5309
32. Krüger S, Vent S, Nörtemann F, Staufer M, Rösch N (2001) J
Chem Phys 115:2082
33. Kubota T, Kitajima Y, Asakura K, Iwasawa Y (1999) Bull Chem
Soc Jpn 72:673
34. Bugaev AL, Guda AA, Lomachenko KA, Srabionyan VV, Bugaev
LA, Soldatov AV, Lamberti C, Dmitriev VP, van Bokhoven JA
(2014) J Phys Chem C 118:10416
35. Sugimura T, Ogawa H (2010) Chem Lett 39:232
36. Ogawa H, Mameda T, Misaki T, Okamoto Y, Sugimura T (2013)
Chem Lett 42:813
37. Ferri D, Bürgi T, Baiker A (2002) J Catal 210:160
38. Zaera F (2009) Acc Chem Res 42:1152
39. Nakatsuji M, Misaki T, Okamoto Y, Sugimura T (2016) Bull
Chem Soc Jpn 89:1187
40. Majumder C (2008) Langmuir 24:10838
41. Love JC, Wolfe DB, Haasch R, Chabinyc ML, Paul KE, White -
sides GM, Nuzzo RG (2003) J Am Chem Soc 125:2597
42. John NS, Thomas PS, Kulkarni GU (2003) J Phys Chem B
107:11376
2
36:164
4. Kim TY, Yokota M, Uchida T, Sugimura T (2009) Catal Lett
31:279
1
1
1
5. Kim TY, Uchida T, Ogawa H, Nitta Y, Okuyama T, Sugimura T,
Hirayama S, Honma T, Sugiura M, Kubota T, Okamoto Y (2010)
Top Catal 53:116
6. H. Sato, T. Mameda, K. Nakai, T. Misaki, Y. Haruyama, S.
Sonobe, T. Kubota, Y. Okamoto, T. Sugimura (2016) Res React
Interm 42:31
1
1
1
1
2
2
7. Nitta Y (1999) Chem Lett 28:635
8. T. Y. Kim, T. Sugimura (2010) J Mol Catal A 327:58
9. Szöllösi G, Herman B, Fulop F, Bartok M (2010) J Catal 276:259
0. Makra Z, Szöllösi G, Bartok M (2012) Catal Today 181:56
1. F. Meemken, N. Maeda, K. Hungerbuhler, A. Baiker (2012) ACS
Catal 2:464
2. Nakai K, Misaki T, Okamoto Y, Sugimura T (2015) Bull Chem
Soc Jpn 88:300
3. Pazmino JH, Bai C, Miller JT, Ribeiro FH, Delgass WN (2013)
Catal Lett 143:1098
2
2
2
2
2
4. Barbouth N, Salame M (1987) J Catal 104:240
5. Marecot P, Mahoungou JR, Barbier J (1993) Appl Catal A 101:143
6. Sugimura T, Oie K, Misaki T, Okamoto Y, Tanaka K, Mori H
43. Rauch T, Bomermann J, Bormann B, Speller S, Heiland W (1999)
Surf Sci 441:107
44. Speller S, Rauch T, Postnikov A, Heiland W (2000) Phys Rev B
6:17297
(2013) Catal Lett 143:495
2
2
7. Kubota T, Ogawa H, Okamoto Y, Misaki T, Sugimura T (2012)
Appl Catal A 437/438:18
8. Kubota T, Kubota H, Kubota T, Moriyasu E, Uchida T, Nitta Y,
Sugimura T, Okamoto Y (2009) Catal Lett 129:387
45. Berndt W, Hora R, Scheffer M (1982) Surf Sci 117:188
46. Kolthoff D, Jurgens D, Schweicke C, Pfnur H (1996) Surf Sci
365:374
47. van Hardeveld R, Hartog F (1969) Surf Sci 15:222
1
23