Mesoporous silica-supported Pt catalysts in enantioselective hydrogenation of ethyl pyruvate

Article abstract of DOI:10.1016/j.cattod.2020.06.010

Catalytic properties of Pt catalysts supported on mesoporous silica (Pt/m-SiO₂) have been studied in enantioselective hydrogenation of ethyl pyruvate. The influences of pore structure of mesoporous silica (m-SiO₂), type of chiral modifier, and H₂ pressure on the catalytic performance have been investigated by using various m-SiO₂ supports and cinchona alkaloids and by varying H₂ pressure. The use of MCM-41, SBA-15, KIT-6, and MCF reveals that characteristic pore structure and size of m-SiO₂ supports significantly affect both activity and enantioselectivity. A facile diffusion of chiral modifier through large mesopores of MCF support enables Pt/MCF to exhibit excellent performance. A comparison of the efficiency of cinchona alkaloids-modified Pt catalysts shows that QN and QD lead to higher performance than CD and CN at ambient H₂ pressure. The influence of cinchona alkaloids on enantioselectivity noticeably depends on H₂ pressure. Cinchona alkaloid-modified Pt/m-SiO₂ exhibit superior enantioselectivity to the corresponding Pt/Al₂O₃ under various H₂ pressures. These results imply that m-SiO₂ is a promising support and that fine control of pore structure can further improve catalytic performance of Pt/m-SiO₂ in heterogeneous enantioselective hydrogenation.

Full text of DOI:10.1016/j.cattod.2020.06.010

CatalysisTodayxxx(xxxx)xxx–xxx
Contents lists available at ScienceDirect
Catalysis Today
journal homepage: www.elsevier.com/locate/cattod
Mesoporous silica-supported Pt catalysts in enantioselective hydrogenation
of ethyl pyruvate
Byeongju Song1, Jeongmyeong Kim1, Iljun Chung, Yongju Yun*
Department of Chemical Engineering, Pohang University of Science and Technology (POSTECH), Pohang, Gyeongbuk, 37673, Republic of Korea
A R T I C L E I N F O
A B S T R A C T
Keywords:
Catalytic properties of Pt catalysts supported on mesoporous silica (Pt/m-SiO2) have been studied in en-
antioselective hydrogenation of ethyl pyruvate. The influences of pore structure of mesoporous silica (m-SiO2),
type of chiral modifier, and H2 pressure on the catalytic performance have been investigated by using various m-
SiO2 supports and cinchona alkaloids and by varying H2 pressure. The use of MCM-41, SBA-15, KIT-6, and MCF
reveals that characteristic pore structure and size of m-SiO2 supports significantly affect both activity and en-
antioselectivity. A facile diffusion of chiral modifier through large mesopores of MCF support enables Pt/MCF to
exhibit excellent performance. A comparison of the efficiency of cinchona alkaloids-modified Pt catalysts shows
that QN and QD lead to higher performance than CD and CN at ambient H2 pressure. The influence of cinchona
alkaloids on enantioselectivity noticeably depends on H2 pressure. Cinchona alkaloid-modified Pt/m-SiO2 ex-
hibit superior enantioselectivity to the corresponding Pt/Al2O3 under various H2 pressures. These results imply
that m-SiO2 is a promising support and that fine control of pore structure can further improve catalytic per-
formance of Pt/m-SiO2 in heterogeneous enantioselective hydrogenation.
Enantioselective
Heterogeneous
Catalysis
Hydrogenation
Platinum
Mesoporous silica
1. Introduction
hydrogenation [13–17]. Among them, only 5 wt% Pt/CNT showed
higher catalytic performance than 5 wt% Pt/Al2O3. Pt nanoparticles
Chiral compounds have been widely used as key intermediates for
the production of pharmaceuticals, fragrances, and agrochemicals
[1–4]. Two enantiomers of a chiral compound can have different phy-
siological impacts because of the homochirality of life [3,4]. Hence, the
production of enantiomerically pure chemicals via catalytic reactions
has attracted considerable attention over the past decades. Significant
efforts have been devoted to the development of heterogeneous cata-
lysts for enantioselective reactions because of their advantages, such as
easy separation from products and high reusability [5,6]. The Orito
reaction, asymmetric hydrogenation of α-keto esters, is the most suc-
cessful example of heterogeneous enantioselective catalysis [7]. The
enantioselectivity of heterogeneous catalysts can be achieved by mod-
ifying metal surface through the adsorption of a chiral organic mole-
cule.
Since the discovery of the Orito reaction, 5 wt% Pt/Al2O3 catalyst
has been extensively studied in asymmetric hydrogenation of α-keto
esters due to its high enantioselectivity [8–12]. Recently, high-surface-
area materials such as zeolite, CMK-3, carbon nanotube (CNT), metal
organic framework (MOF), and resol have been used as alternatives to
Al2O3, one of the most effective supports in enantioselective
deposited inside CNT channel led to higher reaction rate and en-
antioselectivity than those outside the channel. The improvement was
attributed to confinement effect of chiral modifier and reactant mole-
cules inside the CNT channel of ∼20 nm in diameter. Thus, this ob-
servation suggests that the pore size and structure of supports can sig-
nificantly affect catalytic efficiency in heterogeneous enantioselective
hydrogenation.
Mesoporous silica (m-SiO2) is a widely used support material in
heterogeneous catalysis because of its high surface area, superior
thermal stability, and facile preparation [18–22]. Furthermore, fine
control of pore size and structure of silica in nanoscale enables sys-
tematic study of the effect of support materials on catalytic perfor-
mance. It has been also employed as a support for Pt nanoparticles in
heterogeneous enantioselective hydrogenation. In most cases Pt cata-
lysts supported on m-SiO2 (Pt/m-SiO2) showed inferior enantioselec-
tivity, revealing that m-SiO2 is a less efficient support than Al2O3
[12,23,24]. On the other hand, comparably high enantioselectivity was
also observed for 5 wt% Pt/m-SiO2 in enantioselective hydrogenation of
ethyl pyruvate (EtPy) [25]. Recently, we have shown that despite 5
times less metal content, 1 wt% Pt supported on mesocellular silica
⁎ Corresponding author.
E-mail address: yjyun@postech.ac.kr (Y. Yun).
1 Byeongju Song and Jeongmyeong Kim equally contributed to this work.
https://doi.org/10.1016/j.cattod.2020.06.010
Received 1 March 2020; Received in revised form 22 May 2020; Accepted 3 June 2020
0920-5861/©2020ElsevierB.V.Allrightsreserved.
Pleasecitethisarticleas:ByeongjuSong,etal.,CatalysisToday,https://doi.org/10.1016/j.cattod.2020.06.010

B. Song, et al.
CatalysisTodayxxx(xxxx)xxx–xxx
foam (MCF) achieves excellent catalytic performance [26]. The high
activity and enantioselectivity are attributed to the large mesopores of
MCFs that facilitate intraparticle diffusion of bulky chiral modifier
molecules. Therefore, these conflicting results of the catalytic perfor-
mance of Pt/m-SiO2 in enantioselective hydrogenation require a com-
parative study on the performance and the effect of structures of m-
SiO2.
Along with a support material of Pt catalysts, chiral modifier and H2
pressure are important factors determining catalytic performance in
heterogeneous enantioselective hydrogenation. Cinchona alkaloids,
including cinchonidine (CD), cinchonine (CN), quinine (QN), and qui-
nidine (QD), are known as the most effective chiral modifiers in hy-
drogenation of α-keto esters over Pt/Al2O3 [8]. For cinchona alkaloid-
modified Pt catalysts, high H2 pressure has a positive effect on their
activity because dissociated hydrogen acts as one of reactants [27].
However, this is not the case for the enantioselectivity of Pt catalysts.
While Pt/Al2O3 modified with CD exhibited high enantioselectivity
even at high H2 pressure, one modified with synthetic chiral modifier
showed a noticeable decrease [28–31]. Although several studies re-
ported the effect of chiral modifier and H2 pressure on catalytic per-
formance, there is no study on Pt/m-SiO2. Hence, a systematic in-
vestigation of efficiency of Pt/m-SiO2 requires exploring the influences
of chiral modifiers and H2 pressure.
M.W. = 5800, BASF), Hexadecyltrimethylammonium bromide (CTAB,
98 %, Alfa Aesar), 1-butanol (> 99.4 %), 1,3,5-trimethylbenzene (TMB,
98 %, Alfa Aesar), ammonium fluoride (NH4F, 98 %, Alfa Aesar), hy-
drochloric acid (HCl, 35–37 %, Samchun Pure Chemical), ammonia
solution (28–30 %, Samchun Pure Chemical) and ethanol (99.9 %,
Samchun Pure Chemical). As reactant and solvent in the en-
antioselective hydrogenation, ethyl pyruvate (EtPy, 98 %, Alfa Aesar)
and acetic acid (99.7 %, Samchun Pure Chemical) were used, respec-
tively. As chiral modifiers, cinchonidine (CD, 96 %, Sigma-Aldrich),
cinchonine (CN, 98 %, Alfa Aesar), quinine (QN, 98 %, Sigma-Aldrich)
and quinidine (QD, 98 %, Alfa Aesar) were purchased and used as re-
ceived.
The 1 wt% Pt/m-SiO2 were prepared via wet impregnation method
with platinum(II) acetylacetonate (Pt(acac)2, 98 %, Acros) as a pre-
cursor. For the preparation of a reference 1 wt% Pt/SiO2 catalyst,
commercial fumed silica (AEROSIL®380, Evonik Industries) was used as
a support material. Commercial 1 wt% Pt/Al2O3 (Alfa Aesar) and 5 wt%
Pt/Al2O3 (Strem) were purchased and used as reference Pt/Al2O3 cat-
alysts.
2.2. Preparation of mesoporous silica supports and Pt catalysts
Four different m-SiO2 supports were synthesized by previously re-
ported methods [32–34]. Briefly, MCM-41 was synthesized as follows:
CTAB (2.4 g) was dissolved in 120 mL deionized water. Ammonia so-
lution (8 mL) was added into the solution and stirred for 5 min. Then,
TEOS (10 mL) was added slowly into the solution and stirred at room
temperature for 12 h. The mixture was aged for 2 days at 373 K under
static conditions. SBA-15 was prepared as follows: P123 (4 g) was dis-
solved in 1.6 M HCl solution (150 mL) and TEOS (9.1 mL) was added
dropwise into the solution. The mixture was stirred at 313 K for 1 day,
then aged for 2 days at 373 K. KIT-6 was synthesized as follows: P123
(6 g) was dissolved in 216 mL of deionized water and 9.9 mL of HCl.
Then, 1-butanol (6 g) was added into the solution under vigorous stir-
ring at 308 K. After 1 h stirring, TEOS (13.8 mL) was added slowly into
the solution at 308 K and subsequently heated at 373 K for 1 day under
static conditions. MCF was prepared as follows: P123 (4 g) was dis-
solved in 1.6 M HCl solution (75 mL) and then TMB (5 g) was added
into the solution. The solution was stirred vigorously at 313 K for 2 h,
then TEOS (9.2 mL) was added to the mixture and it was aged under
static conditions for 20 h. After NH4F (46 mg) was added, the mixture
was transferred to an autoclave for hydrothermal treatment at 413 K for
24 h. All precipitates obtained after the hydrothermal treatment, were
In this work, we report catalytic performance of 1 wt% Pt/m-SiO2
and reference catalysts, Pt/SiO2 and Pt/Al2O3, in enantioselective hy-
drogenation of EtPy (Scheme 1). The use of different types of m-SiO2
supports, such as MCM-41, SBA-15, KIT-6, and MCF, reveals that
characteristic pore structure and size of m-SiO2 supports significantly
affect both activity and enantioselectivity. A comparison of the effi-
ciency of cinchona alkaloids-modified Pt catalysts shows that QN and
QD result in higher performance than CD and CN at ambient H2 pres-
sure. Depending on the cinchona alkaloid, excess H2 pressure can lead
to a decrease in enantioselectivity. Cinchona alkaloid-modified Pt/m-
SiO2 exhibit superior enantioselectivity to the corresponding Pt/Al2O3
under various H2 pressures.
2. Experimental section
2.1. Chemicals and materials
For synthesis of m-SiO2 supports including MCM-41, SBA-15, KIT-6,
and MCF, the following chemicals were used as received: tetraethyl
orthosilicate (TEOS, Sigma-Aldrich), Pluronic P123 (EO20PO70EO20
,
Scheme 1. Enantioselective hydrogenation of EtPy over cinchona alkaloid-modified Pt catalysts.
2

B. Song, et al.
CatalysisTodayxxx(xxxx)xxx–xxx
Fig. 1. Nitrogen physisorption results for m-SiO2 supports. (a) Nitrogen adsorption-desorption isotherms at 77 K and (b) pore size distributions of m-SiO2 supports.
filtered, dried and finally calcined at 823 K for 4 h or 873 K for 6 h in air
to remove surfactant.
ee did not exceed 3% and 1%, respectively.
Pt/m-SiO2 and Pt/SiO2 catalysts were obtained by facile wet im-
pregnation with an acetone solution of Pt(acac)2. After evaporation of
the solvent at 313 K overnight, the powder was dried at 353 K for 12 h
in an oven. Then it was calcined at 623 K for 3 h in a flow of Ar.
3. Results and discussion
3.1. Characterization of synthesized mesoporous silica supports and Pt
catalysts
2.3. Characterization of mesoporous silica supports and Pt catalysts
Four different types of m-SiO2 supports, MCM-41, SBA-15, KIT-6,
and MCF, were synthesized to study the effect of pore structure of
supports on enantioselective hydrogenation of EtPy. Each m-SiO2 sup-
port has its characteristic pore structure and size. Both MCM-41 and
SBA-15 have long 1D channels while KIT-6 has a mesostructure, which
consist of a pair of interpenetrating bicontinuous networks. Compared
to other m-SiO2 supports, MCF has significantly larger pores consisting
of large spherical cell pores interconnected with small window pores.
The properties of synthesized m-SiO2 supports were characterized
by low-angle X-ray diffraction (XRD), nitrogen physisorption, and
transmission electron microscopy (TEM). The low-angle XRD patterns
show the degree of structural ordering of the synthesized m-SiO2 sup-
ports (Fig. S1). The synthesized MCM-41, SBA-15, and KIT-6 showed
three well-resolved diffraction peaks, indicating ordered mesos-
tructured array of these materials [34,36–40]. In contrast to other m-
SiO2 supports, MCF exhibited no characteristic peak in the low-angle
XRD pattern, showing absence of long range structural ordering
[41–44].
The nitrogen adsorption-desorption isotherms of the m-SiO2 sup-
ports are shown in Fig. 1a. These samples displayed type IV curves. The
MCM-41 exhibited reversible isotherm and other m-SiO2 supports re-
vealed isotherms with a H1 type hysteresis loop due to capillary con-
densation. The pore size distributions shown in Fig. 1b confirmed that
the pore diameters of these supports were well-controlled in a narrow
range. The order of average pore size of the m-SiO2 supports was as
follows: MCF > SBA-15 = KIT-6 > MCM-41. A detailed information
on their textural properties is listed in Table 1. All samples showed high
surface areas of 331–928 m2 g−1 and pore volumes of
Nitrogen physisorption was performed at 77 K using a Micromeritics
ASAP 2010 system. Brunauer-Emmett-Teller (BET) surface areas were
calculated using data obtained over the relative pressure range of 0.05
≤ p/p0 ≤ 0.2 [35]. The cell and window size distributions were cal-
culated by the Barrett-Joyner-Halenda (BJH) method from the ad-
sorption and desorption branches of the isotherm, respectively. Trans-
mission electron microscopy (TEM) images were obtained with a JEOL
JEM-2100 F TEM at an acceleration voltage of 200 kV at Korea Basic
Science Institute (KBSI), Daegu center. Low-angle X-ray diffraction
(XRD) patterns were performed on a PANanalytical Empyrean multi-
purpose diffractometer with Cu-Kα radiation over the 2θ range from
0.2° to 5° at Korea Basic Science Institute (KBSI), Daegu center. Wide-
angle X-ray powder diffraction (XRD) patterns were collected on a Ri-
gaku D/MAX-2500 powder diffraction system using Cu-Kα radiation
over the 2θ range from 5° to 90°. The amount of Pt loading was de-
termined by inductively coupled plasma-optical emission spectroscopy
(ICP-OES) on a Thermo Scientific model iCAP 6300 spectrometer. The
metal dispersion of the Pt catalysts was determined by the CO pulse
chemisorption technique using MicrotracBEL BELCAT-Ⅱ. The disper-
sion of Pt and average particle size were calculated from the amount of
chemisorbed CO on Pt by assuming an adsorption stoichiometry of CO/
Pt = 1. UV–vis spectroscopic measurements were performed on Agilent
Cary 8454 spectrophotometer using quartz cells.
2.4. Catalytic performance test
The catalytic tests were performed in a 50 mL glass reactor for the
reactions at 0.1 MPa H2 pressure. A stainless steel autoclave equipped
with a 100 mL glass liner was used for the reactions at 0.5–5.0 MPa H2
pressure. Before the catalytic tests, all Pt catalysts were pretreated in a
flow of 5% H2/Ar at 673 K for 90 min. The enantioselective hydro-
genation was performed with Pt catalyst (50 mg), chiral modifier
(2–6 mg), and substrate (0.07 mL of EtPy) in solvent (10 mL of acetic
acid) at 293 K. The products were analyzed by a gas chromatograph
(YL6500) equipped with a Chirasil-Dex CB column (CP7502, Agilent).
The enantiomeric excess (ee) was calculated as below:
Table 1
Textural properties of synthesized m-SiO2 supports determined by nitrogen
physisorption.
Sample
SBET (m2 g−1
)
Dwindow (nm)a
Dcell (nm)b
Vpore (cm3 g−1
)
MCM-41
SBA-15
KIT-6
928
883
712
331
2.5
6.4
6.4
22.7
0.82
1.24
1.03
2.38
MCF
41.5
a
Window pore diameter (Dwindow
) determined by the Barrett-Joyner-
ee(%) = |[R]–[S]|/([R] + [S]) × 100%
(1)
Halenda (BJH) method from the desorption branches of the isotherms.
b
Where [R] and [S] are the concentrations of (R)- and (S)-ethyl lactate
Cell pore diameter (Dcell) determined by the BJH method from the ad-
(EtLac), respectively. The estimated absolute errors of conversion and
sorption branches of the isotherms.
3

B. Song, et al.
CatalysisTodayxxx(xxxx)xxx–xxx
Fig. 2. TEM images of (a) MCM-41, (b) SBA-15, (c) KIT-6, and (d) MCF. The synthesized m-SiO2 supports show well-developed mesopores.
0.82–2.38 cm3 g−1. Their pore diameter was estimated from the deso-
rption branches of the isotherms varied from 2.5 nm to 22.7 nm. The
pore structure of MCF consists of large spherical cell pores inter-
connected with small window pores. In the case of cell pores of MCF,
the average diameter was estimated to be 41.5 nm from adsorption
branch of the isotherm. The representative TEM images of the synthe-
sized silica supports confirmed that they had well-developed mesopores
as shown in Fig. 2.
The synthesized m-SiO2 samples were used as supports for 1 wt% Pt
catalysts. The textural properties of Pt/m-SiO2 and other reference
catalysts are summarized in Table 2. All Pt/m-SiO2 still exhibited high
surface areas and pore volumes. More importantly, they revealed nearly
the same pore sizes as the synthesized m-SiO2 supports. The nitrogen
physisorption isotherms shown in Fig. 3a confirmed that there was no
noticeable change in pore structures after deposition of Pt nano-
particles. As shown in pore size distributions in Fig. 3b, the pore size of
the Pt/m-SiO2 was in following order: Pt/MCF > Pt/SBA-15 = Pt/KIT-
6 > Pt/MCM-41.
Most 1 wt% Pt catalysts, including Pt/SBA-15, Pt/KIT-6, Pt/SiO2,
and Pt/Al2O3, revealed similar Pt dispersion of 40.0–46.4 % and mean
Pt particle size of 2.4–2.8 nm. However, 1 wt% Pt/MCM-41 and Pt/MCF
showed slightly lower Pt dispersion and larger mean Pt particle size of
∼4.6 nm despite the same preparation procedure. The larger Pt size of
the catalysts seems to be attributed to aggregation of Pt nanoparticles
during the preparation. TEM images of the 1 wt% Pt/m-SiO2 revealed
that Pt nanoparticles were highly dispersed on the mesopores with
narrow size distribution as displayed in Fig. 4. Moreover, the wide-
angle XRD patterns of these catalysts showed no Pt diffraction peak,
indicating that the Pt particles were well-dispersed on the silica sup-
ports (Fig. S2a). Similarly, the wide-angle XRD patterns and the TEM
images of reference catalysts, Pt/SiO2 and Pt/Al2O3, revealed that Pt
nanoparticles were well-dispersed on the supports (Fig. S2b and Fig.
S3).
3.2. Enantioselective hydrogenation of ethyl pyruvate over Pt catalysts
supported on mesoporous silica
Catalytic performance of 1 wt% Pt/m-SiO2 was evaluated in hy-
drogenation of EtPy at 0.1 MPa H2 pressure. It was compared with that
of reference 1 wt% Pt catalysts, Pt/SiO2 and Pt/Al2O3. All Pt catalysts
were chirally modified by CD to achieve enantioselectivity. The con-
version and enantioselectivity of the Pt catalysts are summarized in
Table 3. The Pt/m-SiO2 and reference Pt/SiO2 exhibited conversion in
the range of 18–29 % after 5 min and enantiomeric excess (ee) in the
range of 73–90 % at full conversion. A reference Pt/Al2O3 catalyst re-
vealed conversion of 17 % and ee of 80 %. There have been several
studies reporting inferior efficiency of silica as a supporting material for
Pt catalysts in enantioselective hydrogenation of α-keto esters
[12,23,24]. However, our results clearly show that with the same 1 wt%
Pt loading, catalytic performance of Pt/m-SiO2 and reference Pt/SiO2 is
comparable or superior to that of reference Pt/Al2O3. This also indicates
that silica can be used as highly efficient support materials for Pt cat-
alysts in enantioselective hydrogenation.
Enantioselectivity of the silica-supported Pt catalysts was in fol-
lowing order: Pt/MCF (90 % ee) > Pt/SBA-15 (82 % ee) > Pt/MCM-41
(80 % ee) > Pt/SiO2 (76 % ee) > Pt/KIT-6 (73 % ee). The different
enantioselectivity is attributed to the pore structure and size of the si-
lica supports. The enantioselectivity of Pt/SiO2 reveals that mesopores
of m-SiO2 supports can have a positive or a negative effect depending on
their pore structures. It is interesting that Pt/SBA-15 with 1D channel
structure showed higher ee than Pt/KIT-6 with complex 3D channel
structure despite similar pore size and surface area. This suggests that
channel structure with higher pore volume is more effective in en-
antioselective hydrogenation. The fact that Pt/KIT-6 exhibited lower
enantioselectivity than Pt/MCM-41 implies that its pore structure and
lower surface area have a negative effect on enantioselectivity. The Pt/
m-SiO2, in general, exhibited higher ee value as the pore size increases,
Pt/MCF (Dwindow =22.5 nm) > Pt/SBA-15 (Dwindow =6.4 nm) > Pt/
MCM-41 (Dwindow =2.7 nm). These observations show that control of
pore structure of support materials can improve catalytic performance
of Pt catalysts in enantioselective hydrogenation.
Table 2
Physicochemical properties of Pt catalysts determined by nitrogen physisorption.
Sample
Pt content (wt%)a
SBET (m2 g−1
)
Dwindow (nm)b
Dcell (nm)c
Vpore (cm3/g)
Pt dispersion (%)d
dPt (nm)d
Pt/MCM-41
Pt/SBA-15
Pt/KIT-6
Pt/MCF
1.1
1.2
1.1
1.0
1.2
1.0
930
742
733
321
372
152
2.7
6.4
6.6
22.5
–
0.81
1.22
0.88
2.27
1.64
0.44
24.2
40.0
44.5
24.5
44.6
46.4
4.7
2.8
2.5
4.6
2.5
2.4
38.0
Pt/SiO2
e
Pt/Al2O3
9.1
a
Pt loadings determined by ICP-OES.
b
c
Window pore diameter (Dwindow) determined by the BJH method from the desorption branches of the isotherms.
Cell pore diameter (Dcell) determined by the BJH method from the adsorption branches of the isotherms.
Pt dispersions and particle sizes (dPt) determined by CO chemisorption.
d
e
Commercial 1 wt% Pt/Al2O3 catalyst purchased from Alfa Aesar.
4

B. Song, et al.
CatalysisTodayxxx(xxxx)xxx–xxx
Fig. 3. Nitrogen physisorption results for Pt/m-SiO2. (a) Nitrogen adsorption-desorption isotherms at 77 K, and (b) pore size distributions of Pt/m-SiO2.
Similar improvement has been reported for CD-modified Pt catalysts
in enantioselective hydrogenation of EtPy. A comparative study on Pt
confined within CNT channels and Pt deposited outside CNTs showed
that the enrichment of chiral modifier and reactant inside the CNT
channels of ∼20 nm in diameter leads to the enhancement of en-
antioselectivity [14]. The fact that chiral modification of Pt surface by
bulky cinchona alkaloids is required for enantiodifferentiation implies
that the diffusion of the chiral modifier through the pores of a support
material is a critical factor determining efficiency. It explains the lowest
ee of Pt/KIT-6 with a complex pore structure and the highest ee of Pt/
MCF with the largest pores in our results. It is also noteworthy that a
facile diffusion of CD by the use of MCF support enables Pt/MCF to
exhibit superior enantioselectivity to Pt/Al2O3, one of the most effec-
tive heterogeneous catalysts in this reaction.
Table 3
Catalytic performance of Pt/m-SiO2 and reference Pt catalysts in en-
antioselective hydrogenation of EtPya.
Sampleb
Conversion (%)c
ee (%)d
Pt/MCM-41
Pt/SBA-15
Pt/KIT-6
Pt/MCF
Pt/SiO2
Pt/Al2O3
20
22
18
29
20
17
80
82
73
90
76
80
a
Reaction conditions: 50 mg of catalyst, 2 mg of CD (6.8 × 10–3 mmol),
0.07 mL of EtPy, 10 mL of acetic acid, H2 pressure of 0.1 MPa, 293 K.
b
1 wt% Pt loading.
Conversion measured at 5 min after reaction.
ee (%) measured at 100 % conversion.
c
d
3.3. Comparison of different cinchona alkaloids as chiral modifiers for Pt
catalysts
(QN) and CN (QD) are diastereomers (Scheme 1); however, they have
opposite chirality at C8 and C9. The molecular structures of QN and QD
are different from those of CD and CN in that they have a methoxy
group in the quinoline ring.
As described in the previous section, Pt/MCF exhibited superior
catalytic performance due to the efficient mesoporous structure for
intraparticle diffusion of bulky chiral modifier molecules. Thus, Pt/MCF
was chosen as a representative sample of Pt/m-SiO2 to minimize the
limitation by textural properties and diffusion of bulky chiral modifiers.
The creation of chiral sites on a metal surface by the adsorption of a
chiral modifier is a prerequisite for heterogeneous enantioselective
catalysis. The catalytic performance of Pt catalysts often significantly
depends on the chiral modifier used in the reaction [45,46]. Therefore,
the effects of different cinchona alkaloids including CD, CN, QN, and
QD on the catalytic efficiency of Pt catalysts were further investigated
in enantioselective hydrogenation of EtPy. The chiral modifiers, CD
Fig. 4. TEM images and Pt size distributions of (a) Pt/MCM-41, (b) Pt/SBA-15, (c) Pt/KIT-6, and (d) Pt/MCF. The Pt nanoparticles are well-dispersed on m-SiO2
supports.
5

B. Song, et al.
CatalysisTodayxxx(xxxx)xxx–xxx
through the methoxy group enhances π-π interaction with the surface
and thus, makes the flat adsorption geometry more favorable. This
agrees with our results that QN- and QD-modified Pt catalysts had
higher activity and enantioselectivity than CD- and CN-modified cata-
lysts regardless of the type of support.
Efficiency comparison of Pt/MCF with other reference catalysts re-
vealed that Pt/MCF outperformed Pt/SiO2 and Pt/Al2O3 for all chiral
modifiers. The conversions of Pt/MCF were nearly 2–3 times higher and
the ee values were 7–26 % higher than those of the reference catalysts.
Among all chiral modifier-catalyst combinations, QN-modified Pt/MCF
yielded the highest ee of 96 % at 0.1 MPa H2 pressure. This indicates
that the intrinsic property of silica is not responsible for the improve-
ment of catalytic performance for Pt/MCF because the reference Pt/
SiO2 showed comparable or lower enantioselectivity than Pt/Al2O3.
Our results confirm that the large mesopores of MCF improve the dif-
fusion of not only CD, but also other bulky cinchona alkaloids, thereby
significantly enhance the performance.
Table 4
Conversion (%)a and ee (%)b for hydrogenation of EtPy over Pt catalysts
modified with different cinchona alkaloidsc. Letters in parentheses represent the
chirality of the major product.
Catalyst
Pt/MCF
Pt content (wt%)d
Chiral modifier
Conversion (%)
ee (%)
1.0
CD
CN
QN
QD
CD
CN
QN
QD
CD
CN
QN
QD
29
25
56
49
12
7
29
22
17
7
91 (R)
85 (S)
96 (R)
90 (S)
75 (R)
59 (S)
86 (R)
78 (S)
80 (R)
73 (S)
89 (R)
76 (S)
Pt/SiO2
1.2
1.0
Pt/Al2O3
18
22
a
Conversion (%) measured at 5 min after reaction.
ee (%) measured at 100 % conversion.
Reaction conditions: 50 mg of catalyst, 6 mg of chiral modifier, 0.07 mL of
b
c
3.4. The effect of hydrogen pressure on cinchona alkaloid-modified Pt
EtPy, 10 mL of acetic acid, 0.1 MPa H2 pressure, 293 K.
d
catalysts
Pt loadings determined by ICP-OES.
Towards
a detailed understanding of catalytic properties of
The CD- and QN-modified Pt/MCF predominantly yielded (R)-ethyl
lactate (EtLac) with 91–96 % ee while CN- and QD-modified Pt/MCF
yielded (S)-EtLac with 85–90 % ee as summarized in Table 4. Although
both CD- and QN-modified Pt/MCF favored the same chiral product, the
latter showed nearly two times higher conversion and higher en-
antioselectivity than those of the former. Similarly, the QD-modified
Pt/MCF catalyst revealed better performance than that of the CN-
modified catalyst. The effects of cinchona alkaloids on the performance
of reference Pt/SiO2 and Pt/Al2O3 catalysts were similar to those ob-
served for Pt/MCF. The QN-modified Pt/SiO2 showed 86 % ee, which is
11 % higher than that of CD-modified Pt/SiO2. The conversion of QN-
modified Pt/SiO2 was 2–3 times higher than that of CD-modified Pt/
SiO2. Similarly, QD modified Pt/SiO2 exhibited three times higher
conversion and 19 % higher ee than CN-modified Pt/SiO2. QN- and QD-
modified Pt/Al2O3 also exhibited higher conversion and enantioselec-
tivity than CD- and CN-modified Pt/Al2O3.
Cinchona alkaloids adsorb on Pt surface through the quinoline
moiety and have several adsorption geometries on the surface [47–50].
Previous studies have shown that a flat adsorption geometry, in parallel
with the Pt surface, effectively achieves enantiodifferentiation [50,51].
Subtle change in the adsorption geometry significantly affects en-
antioselectivity. Thus, superior performance of QN- and QD-modified Pt
catalysts to CD- and CN-modified catalysts is attributed to the difference
in quinoline moiety; QN and QD have a methoxy-quinoline ring
whereas CD and CN have a quinoline ring. An in-situ reflection-ab-
sorption infrared spectroscopy study has demonstrated that QN and QD
adsorb on Pt surface via both the quinoline ring and oxygen atoms in a
methoxy group [52,53]. The additional bonding to the Pt surface
cinchona alkaloid-modified Pt/m-SiO2, the effects of H2 pressure on the
enantioselectivity of Pt/MCF were investigated over 0.1–5.0 MPa H2
pressure. The ee of 1 wt% Pt/MCF was compared with that of 5 wt% Pt/
Al2O3, one of the most efficient catalysts in this reaction [5,54,55].
Fig. 5 shows the changes in enantioselectivity of the Pt catalysts by H2
pressure. The influence of H2 pressure on the ee of Pt/Al2O3 was sig-
nificantly different depending on the type of cinchona alkaloid as dis-
played in Fig. 5a. The CD-modified Pt/Al2O3 (◼) revealed an im-
provement in enantioselectivity at H2 pressure > 0.1 MPa, achieving
the maximum ee of 90 % at 1.0 MPa. However, other cinchona alkaloids
generally led to a decrease in ee. Both CN-modified Pt/Al2O3 (□) and
QN-modified Pt/Al2O3 ( ) exhibited a similar trend and their en-
antioselectivity decreased to ∼77 % ee at 5.0 MPa H2 pressure. Severe
decrease in enantioselectivity was observed for QD-modified Pt/Al2O3
( ). It yielded 89 % ee at 0.1 MPa but showed only 52 % ee at 5.0 MPa
H2 pressure, showing a significantly negative effect of H2 pressure.
Compared to 5 wt% Pt/Al2O3, 1 wt% Pt/MCF exhibited superior
enantioselectivity during the reaction, particularly at high H2 pressure,
as shown in Fig. 5b. Both CD-modified Pt/MCF (◼) and CN-modified
Pt/MCF (□) showed an increase in enantioselectivity, varying H2
pressure from 0.5 MPa to 5.0 MPa. They exhibited 95 % ee and 91 % ee
at 5.0 MPa H2 pressure, respectively. It is noteworthy that the CN-
modified Pt/MCF revealed much higher ee than the CN-modified Pt/
Al2O3 at 5.0 MPa H2 pressure. In contrast to CD- and CN-modified Pt/
MCF, both QN-modified Pt/MCF ( ) and QD-modified Pt/MCF ( )
showed a decrease in enantioselectivity at high H2 pressure. However,
Pt/MCF exhibited higher stability than Pt/Al2O3 at high H2 pressure.
The decline of ee started at 1.0 MPa for Pt/MCF whereas it started at
Fig. 5. Enantiomeric excess (ee) of cinchona-
alkaloid-modified (a) Pt/Al2O3 and (b) Pt/MCF
as a function of H2 pressure. QN- and QD-
modified Pt/Al2O3 show
a considerable de-
crease in ee above 3.0 MPa H2 pressure while
Pt/MCF maintain a relatively high ee.
Reaction conditions: 50 mg of catalyst, 6 mg of
chiral modifier, 0.07 mL of EtPy, 10 mL of
acetic acid, 293 K.
6

B. Song, et al.
CatalysisTodayxxx(xxxx)xxx–xxx
adsorption geometries of QD and QN by an increase in hydrogen cov-
erage on the Pt surface. Subtle changes in adsorption geometry of the
chiral modifier significantly affect the enantiodifferentiation ability on
the chirally-modified Pt catalysts [50,61]. Indeed, DFT calculations
showed that co-adsorbed hydrogen can destabilize adsorbed chiral
modifier and change its adsorption geometry [62]. Compared to CD and
CN with quinoline ring, QN and QD with methoxy-quinoline ring are
weakly adsorbed on Pt surface [8,60,63]. Thus, it is expected that co-
adsorbed hydrogen more significantly affects the adsorption geometries
of QN and QD than those of CD and CN, resulting in the dramatic ee
drop at high H2 pressure. Although further study is required to under-
stand the different catalytic behaviors arising from changes of adsorp-
tion geometry along hydrogen coverage, our results suggest that the
adsorption geometries of QN and QD are more negatively influenced by
H2 pressure.
Fig. 6. Normalized UV–vis absorbance of CD and QD with intact quinoline
moiety as a function of reaction time. Reaction conditions: 50 mg of catalyst,
6 mg of chiral modifier, 10 mL of acetic acid, 5.0 MPa H2, 293 K.
4. Conclusions
We have prepared m-SiO2 supports with different pore structures
and have tested Pt/m-SiO2 in enantioselective hydrogenation of EtPy.
The pore structure and size of m-SiO2 supports significantly affected
both activity and enantioselectivity. Pt/MCF with the largest mesopores
showed the highest catalytic efficiency. The performance of cinchona
alkaloid-modified Pt/m-SiO2 exhibited a noticeably dependence on
chiral modifier and H2 pressure. Compared to CD (CN), QN (QD) re-
sulted in higher enantioselectivity at ambient H2 pressure but led to
lower enantioselectivity at high H2 pressure. Comparison of perfor-
mance of supported Pt catalysts, Pt/m-SiO2 and Pt/Al2O3, suggested
that m-SiO2 is a promising support material and that fine control of m-
SiO2 can further improve efficiency of Pt/m-SiO2 in enantioselective
hydrogenation.
0.5 MPa for Pt/Al2O3. The magnitude of the decrease in ee was also
significantly less (90 % to 79 %) for QD-modified Pt/MCF than that for
QD-modified Pt/Al2O3 (89 % to 52 %). In general, high H2 pressure
increases the reaction rate of enantioselective hydrogenation over Pt
catalysts [27,31,56,57]. Our results indicate that 1 wt% Pt/MCF is more
efficient than 5 wt% Pt/Al2O3 for the production of enantiomerically-
pure compounds in that cinchona alkaloid-modified Pt/MCF can benefit
from both high enantioselectivity and enhanced reaction rate under
high H2 pressure. Indeed, at the same H2 pressure, each cinchona al-
kaloid-modified Pt/MCF showed higher ee than that obtained from the
corresponding Pt/Al2O3.
Under ambient pressure, chiral modification by QN and QD rather
than by CD and CN resulted in higher enantioselectivity for both Pt/
MCF and Pt/Al2O3 catalysts. However, the detrimental effect of high H2
pressure on enantioselectivity was more severe for QN- and QD-mod-
ified Pt catalysts. As mentioned earlier, CD and CN possess quinoline
ring as adsorbing moiety while QN and QD have a methoxy-quinoline
ring. It has been reported that competitive hydrogenation of adsorbing
moiety of chiral modifier occurs over Pt catalysts during en-
antioselective hydrogenation of α-keto ester [58,59]. Pt catalyst mod-
ified with partially or completely hydrogenated CD exhibited 8–70 %
lower ee than Pt catalyst modified with intact CD [47]. Thus, faster
hydrogenation of adsorbing moiety of QN and QD than that of CD and
CN could result in a sharp decrease in ee at high H2 pressure. Hence, CD
and QD was chosen for comparison of hydrogenation rate of quinoline
ring and methoxy-quinoline ring because CD- and QD-modified Pt/MCF
exhibited the largest difference in ee at high H2 pressure.
CRediT authorship contribution statement
Byeongju Song: Conceptualization, Investigation, Writing - original
draft. Jeongmyeong Kim: Conceptualization, Investigation, Writing -
original draft. Iljun Chung: Investigation. Yongju Yun:
Conceptualization, Supervision, Writing - original draft, Writing - re-
view & editing.
Declaration of Competing Interest
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influ-
ence the work reported in this paper.
Acknowledgements
The hydrogenation rate of quinoline moiety of CD and QD over Pt/
MCF at 5.0 MPa H2 pressure was measured by UV–vis analysis.
Absorbance at 315 nm and 337 nm in the UV–vis spectra is character-
istic of intact quinoline moiety of CD and QD, respectively (Fig. S4)
[10,60]. Therefore, concentrations of CD and QD with intact quinoline
moiety were measured by the corresponding absorbance values. Nor-
malized UV–vis absorbance is plotted as a function of hydrogenation
reaction time in Fig. 6. A decrease in the absorbance after a certain time
of reaction indicates the loss of intact quinoline moiety of the chiral
modifier by hydrogenation. Interestingly, there was no significant dif-
ference in the rate of decrease in absorbance. This shows that the hy-
drogenation rate of methoxy-quinoline ring of QD was similar to that of
quinoline ring of CD. Our results suggest that the saturation of quino-
line moiety has no influence on the decrease in ee of QN- and QD-
modified Pt/MCF at high H2 pressure. Indeed, the same ee was obtained
from QD-modified Pt/MCF when partially hydrogenated QD was used
as the modifier instead of fresh one (Table S1).
This work was supported by the National Research Foundation of
Korea (NRF) grant (NRF-2019R1C1C1002846) funded by the Ministry
of Science and ICT. This work was also supported by LG Chem.
Appendix A. Supplementary data
Supplementary material related to this article can be found, in the
online version, at doi:https://doi.org/10.1016/j.cattod.2020.06.010.
References
[1] W.J. Zhang, S.H. Ruan, W.Y. Shen, Z. Wang, D.L. An, Y.Y. Li, J.X. Gao, Catal.
Commun. 119 (2019) 153–158.
[2] F.D. Klingler, Acc. Chem. Res. 40 (2007) 1367–1376.
[3] S.W. Smith, Toxicol. Sci. 110 (2009) 4–30.
[4] D.S. Sholl, A.J. Gellman, AlChE J. 55 (2009) 2484–2490.
[5] Q. Li, X.Q. Zhang, M.T. Xiao, Y.J. Liu, Catal. Commun. 42 (2013) 68–72.
[6] X.R. Xue, P. Chen, P. Xu, Y.H. Wang, Catal. Commun. 110 (2018) 55–58.
[7] Y. Orito, S. Imai, S. Niwa, G.H. Nguyen, J. Synth. Org. Chem Jpn. 37 (1979)
One possible reason for the decrease in enantioselectivity of QN-
and QD- modified Pt catalysts at high H2 pressure is changes in
7

B. Song, et al.
CatalysisTodayxxx(xxxx)xxx–xxx
173–174.
130 (2010) 333–337.
[8] W.M. Huck, T. Mallat, A. Baiker, Adv. Synth. Catal. 345 (2003) 255–260.
[9] M. Bartok, B. Torok, K. Balazsik, T. Bartok, Catal. Lett. 73 (2001) 127–131.
[10] B. Torok, K. Balazsik, M. Torok, G. Szollosi, M. Bartok, Ultrason. Sonochem. 7
(2000) 151–155.
[11] X.H. Li, X. You, P.L. Ying, J.L. Xiao, C. Li, Top. Catal. 25 (2003) 63–70.
[12] X.H. Li, H.H. Wang, H.Y. Pan, Y.M. Wang, P. Wu, Appl. Catal. A 488 (2014) 48–57.
[13] B. Li, X.H. Li, H.N. Wang, P. Wu, J. Mol. Catal. A: Chem. 345 (2011) 81–89.
[14] Z. Chen, Z. Guan, M. Li, Q. Yang, C. Li, Angew. Chem. Int. Ed. 50 (2011)
4913–4917.
[15] T.J. Hall, J.E. Halder, G.J. Hutchings, R.L. Jenkins, P. Johnston, P. McMorn,
P.B. Wells, R.P.K. Wells, Top. Catal. 11 (2000) 351–357.
[16] U. Bohmer, K. Morgenschweis, W. Reschetilowski, Catal. Today 24 (1995) 195–199.
[17] H. Pan, X. Li, D. Zhang, Y. Guan, P. Wu, J. Mol. Catal. A: Chem. 377 (2013)
108–114.
[18] L. Li, J.L. Shi, J.N. Yan, X.G. Zhao, H.G. Chen, Appl. Catal. A 263 (2004) 213–217.
[19] L. Chmielarz, P. Kustrowski, R. Dziembaj, P. Cool, E.F. Vansant, Appl. Catal. B 62
(2006) 369–380.
[20] C.X. Xiao, R.V. Maligal-Ganesh, T. Li, Z.Y. Qi, Z.Y. Guo, K.T. Brashler, S. Goes,
X.L. Li, T.W. Goh, R.E. Winans, W.Y. Huang, ChemSusChem 6 (2013) 1915–1922.
[21] S.H. Joo, J.Y. Park, C.K. Tsung, Y. Yamada, P.D. Yang, G.A. Somorjai, Nat. Mater. 8
(2009) 126–131.
[22] G.H. Liu, M. Yao, F. Zhang, Y. Gao, H.X. Li, Chem. Commun. (2008) 347–349.
[23] H.N. Wang, X.H. Li, Y.M. Wang, P. Wu, ChemCatChem 2 (2010) 1303–1311.
[24] M.U. Azmat, Y. Guo, Y. Guo, Y.Q. Wang, G.Z. Lu, J. Mol. Catal. A: Chem. 336 (2011)
42–50.
[37] D. Lee, G.S. Jung, H.C. Lee, J.S. Lee, Catal. Today 111 (2006) 373–378.
[38] D.Y. Zhao, J.L. Feng, Q.S. Huo, N. Melosh, G.H. Fredrickson, B.F. Chmelka,
G.D. Stucky, Science 279 (1998) 548–552.
[39] Z.P. Qu, S.J. Shen, D. Chen, Y. Wang, J. Mol. Catal. A: Chem. 356 (2012) 171–177.
[40] F. Kleitz, F. Berube, R. Guillet-Nicolas, C.M. Yang, M. Thommes, J. Phys. Chem. C
114 (2010) 9344–9355.
[41] S.P. Hudson, R.F. Padera, R. Langer, D.S. Kohane, Biomaterials 29 (2008)
4045–4055.
[42] E. Vilarrasa-Garcia, J.A. Cecilia, S.M.L. Santos, C.L. Cavalcante, J. Jimenez-
Jimenez, D.C.S. Azevedo, E. Rodriguez-Castellon, Microporous Mesoporous Mater.
187 (2014) 125–134.
[43] M.Y. Xing, D.Y. Qi, J.L. Zhang, F. Chen, B.Z. Tian, S. Bagwas, M. Anpo, J. Catal. 294
(2012) 37–46.
[44] A. Wach, M. Drozdek, B. Dudek, P. Latka, P. Kustrowski, Microporous Mesoporous
Mater. 226 (2016) 433–443.
[45] M. Heitbaum, F. Glorius, I. Escher, Angew. Chem. Int. Ed. 45 (2006) 4732–4762.
[46] Z. Ma, F. Zaera, J. Phys. Chem. B 109 (2005) 406–414.
[47] H.U. Blaser, H.P. Jalett, W. Lottenbach, M. Studer, J. Am. Chem. Soc. 122 (2000)
12675–12682.
[48] D. Ferri, T. Burgi, J. Am. Chem. Soc. 123 (2001) 12074–12084.
[49] M. von Arx, M. Wahl, T.A. Jung, A. Baiker, Phys. Chem. Chem. Phys. 7 (2005)
273–277.
[50] J. Kubota, F. Zaera, J. Am. Chem. Soc. 123 (2001) 11115–11116.
[51] A.J. Gellman, W.T. Tysoe, F. Zaera, Catal. Lett. 145 (2015) 220–232.
[52] Z. Ma, I. Lee, F. Zaera, J. Am. Chem. Soc. 129 (2007) 16083–16090.
[53] L. Mink, Z. Ma, R.A. Olsen, J.N. James, D.S. Sholl, L.J. Mueller, F. Zaera, Top. Catal.
48 (2008) 120–127.
[25] S. Bottcher, C. Hoffmann, K. Rauchle, W. Reschetilowski, ChemCatChem 3 (2011)
741–748.
[26] J. Kim, B. Song, G. Hwang, Y. Bang, Y. Yun, J. Catal. 373 (2019) 306–313.
[27] H.U. Blaser, H.P. Jalett, M. Garland, M. Studer, H. Thies, A.W. Tijani, J. Catal. 173
(1998) 282–294.
[54] H.U. Blaser, H.P. Jalett, M. Muller, M. Studer, Catal. Today 37 (1997) 441–463.
[55] L.L. Lou, T. Yang, W.J. Yu, H.X. Qu, Y.J. Feng, H. Li, K. Yu, S.X. Liu, Catal. Today
298 (2017) 197–202.
[28] M. Bartok, M. Sutyinszki, I. Bucsi, K. Felfoldi, G. Szollosi, F. Bartha, T. Bartok, J.
Catal. 231 (2005) 33–40.
[29] B. Minder, M. Schurch, T. Mallat, A. Baiker, T. Heinz, A. Pfaltz, J. Catal. 160 (1996)
261–268.
[56] Y.L. Huang, Y.Z. Li, J.Y. Hu, P.M. Cheng, H. Chen, R.X. Li, X.J. Li, C.W. Yip,
A.S.C. Chan, J. Mol. Catal. A: Chem. 189 (2002) 219–224.
[57] G. Szollosi, S. Cserenyi, K. Balazsik, F. Fulop, M. Bartok, J. Mol. Catal. A: Chem. 305
(2009) 155–160.
[30] G. Szollosi, C. Somlai, P.T. Szabo, M. Bartok, J. Mol. Catal. A: Chem. 170 (2001)
165–173.
[31] A. Marinas, T. Mallat, A. Baiker, J. Catal. 223 (2004) 466-466.
[32] N.E. Tari, A. Tadjarodi, J. Tamnanloo, S. Fatemi, Microporous Mesoporous Mater.
231 (2016) 154–162.
[58] M. von Arx, T. Mallat, A. Baiker, Top. Catal. 19 (2002) 75–87.
[59] M. Bartok, G. Szollosi, K. Balazsik, T. Bartok, J. Mol. Catal. A: Chem. 177 (2002)
299–305.
[60] W.R. Huck, T. Burgi, T. Mallat, A. Baiker, J. Catal. 216 (2003) 276–287.
[61] S. Diezi, A. Szabo, T. Mallat, A. Baiker, Tetrahedron: Asymmetry 14 (2003)
2573–2577.
[33] R.M. Rioux, H. Song, J.D. Hoefelmeyer, P. Yang, G.A. Somorjai, J. Phys. Chem. B
109 (2005) 2192–2202.
[62] K.R. Hahn, A.P. Seitsonen, A. Baiker, Phys. Chem. Chem. Phys. 17 (2015)
27615–27629.
[63] D.M. Meier, T. Mallat, D. Ferri, A. Baiker, J. Catal. 244 (2006) 260–263.
[34] F. Kleitz, S.H. Choi, R. Ryoo, Chem. Commun. (2003) 2136–2137.
[35] S. Brunauer, P.H. Emmett, E. Teller, J. Am. Chem. Soc. 60 (1938) 309–319.
[36] J.X. Li, Z.Y. Xiong, L.H. Zhou, X. Han, H.L. Liu, Microporous Mesoporous Mater.
8