Highly effective Ir-based catalysts for benzoic acid hydrogenation: Experiment- and theory-guided catalyst rational design

Article abstract of DOI:10.1039/c7gc00387k

On the way to exploring superior hydrogenation catalysts, Ir-based catalysts with a record catalytic activity (up to 40 h^-1) for the hydrogenation of benzoic acid to cyclohexanecarboxylic acid under mild reaction conditions (85 °C, 0.1 MPa H₂, in water) have been successfully developed. By excluding various factors, the experimental results showed that the main factor governing the activity discrepancy between the Ir-based catalysts is actually the dispersion stability of the supports (such as N-doped carbon, active carbon, SBA-15 and various metal oxides) in the reaction solution, rather than the interaction between the Ir active component and the supports. Combined with theoretical investigation from first principles, an activity volcano curve considering the competing adsorption between the reactants (H₂) and solvent (H₂O) for aqueous aromatic ring hydrogenation was presented for the first time. The high activity of Ir can be deduced by the proper discrepancy of dissociation energies or adsorption energies between H₂ and H₂O on the catalysts. This activity volcano curve provides guidance for further rational design of promising catalysts for benzoic acid or even aromatic ring hydrogenation under true reaction conditions for practical applications.

Full text of DOI:10.1039/c7gc00387k

View Article Online
View Journal
Green
Chemistry
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: M. Tang, S. Mao,
X. Li, C. Chen, M. li and Y. Wang, Green Chem., 2017, DOI: 10.1039/C7GC00387K.
Volume 18 Number
7
7
April 2016 Pages 1821–2242
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
Green
Chemistry
Cutting-edge research for a greener sustainable future
www.rsc.org/greenchem
Accepted Manuscripts are published online shortly after
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
author guidelines.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the ethical guidelines, outlined
in our author and reviewer resource centre, still apply. In no
event shall the Royal Society of Chemistry be held responsible
for any errors or omissions in this Accepted Manuscript or any
consequences arising from the use of any information it contains.
ISSN 1463-9262
CRITICAL REVIEW
G. Chatel et al.
Heterogeneous catalytic oxidation for lignin valorization into valuable
chemicals: what results? What limitations? What trends?
rsc.li/green-chem

Please do not adjust m aa rgins
Green Chemistry
Page 1 of 9
View Article Online
DOI: 10.1039/C7GC00387K
Journal Name
ARTICLE
Highly effective Ir-based catalysts for the benzoic acid
hydrogenation: experiment and theory guided catalysts rational
design
Received 00th January 20xx,
Accepted 00th January 20xx
‡
‡
DOI: 10.1039/x0xx00000x
Minghui Tang , Shanjun Mao , Xuefeng Li, Chunhong Chen, Mingming Li, and Yong Wang*
-1
www.rsc.org/
On the way to explore superior hydrogenation catalysts, Ir-based catalysts with a record catalytic activity (up to 40 h ) for
o
the hydrogenation of benzoic acid to cyclohexanecarboxylic under mild reaction conditions (85 C, 0.1 MPa H
2
, in water)
have been successfully developed. By excluding various factors, experimental results showed that the main factor
governing the activity discrepancy of the Ir-based catalysts is actually the dispersing stability of the supports (such as N-
doped carbon, active carbon, SBA-15 and various metal oxides) in the reaction solution, rather than the interaction
between the Ir active component and the supports. Combining with theoretical investigation from first principle, an
activity volcano curve considering the competing adsorption between reactants (H
aromatic ring hydrogenation was given out for the first time. The high activity of Ir can be deduced by the proper
discrepancy of dissociation energies or adsorption energies between H and H O on the catalysts. This activity volcano
curve provides guidance for further rational design of promising catalysts for benzoic acid or even aromatic ring
2 2
) and solvent (H O) for aqueous
2
2
hydrogenation
in
true
reaction
conditions
for
practical
applications.
conditions. Firstly, Pd NPs supported on
a
N-doped
Introduction
hydrothermal carbon (Pd/CN) catalyzed BA hydrogenation
-1
efficiently, exhibiting high activity with a TOF of 0.883 h ) at
o
relatively mild reaction conditions (85 C, 0.1 MPa H , in water)
The catalytic hydrogenation of aromatics is of great
significant in both the general synthetic and industrial fields of
2
1
6
1
,
2
. Then, considering the bimetallic effects and the advantage
chemistry
.
Among which, the chemoselective
of lowering the cost of the catalyst, we prepared a bimetallic
RuPd/CN catalyst, exhibiting a remarkably superior activity
hydrogenation of benzoic acid (BA) to cyclohexanecarboxylic
acid (CCA) is a typical reaction with practical application
significance, as CCA is a very important organic intermediate
used in the synthesis of hexanolactam, praziquantel,
-1
with a TOF up to 11.5 h under the same mild reaction
1
7
conditions as Pd/CN . However, it is notable that the catalytic
activity for BA hydrogenation of the Pd or RuPd based catalysts
strongly depends on the catalyst supports (Table S1, Table S2).
Therefore, we wonder whether this phenomenon may also
exist in other noble metal based catalysts. In this work, it is
found that the dependence of catalytic activity on catalytic
supports (such as CN, active carbon (AC), SBA-15 and various
metal oxides) has reduced in the present Ir-based catalytic
3
-6
ansatrienin, etc . Noble metal catalysts, including Pd/Carbon
7
,
8
9
, Ru/C , Rh/C
10, 11
12
materials
, RuPt/SiO
2
, bimetallic
14
1
nanoparticles (NPs: RuPd, RuSn) and Pt nanowire , have
3
been applied in this reaction. However, harsh conditions
o
100~250 C, 5~15 MPa H
(
2
, 24 h) and organic solvents
(methanol, acetic acid, tetrahydrofuran, isopropanol etc.) are
typically required. These rather extreme conditions lead to
some disadvantages in both technical and economic aspects,
which restrict their application on a commercial scale. Hence,
there is an urgent need for rational designed effective catalysts
capable of promoting hydrogenation of BA under moderate
conditions.
-1
system, all showing much higher activity (20~40 h ) under
o
mild reaction conditions (85 C, 0.1 MPa H , in water) than
2
previously reported work. Experimental results showed that
the main factor affecting the activity of the Ir-based catalysts is
the dispersing stability of the supports in the reaction solution,
a new insight was obtained.
Indeed, great endeavors have been devoted to this
respect, and our group also performed a series of research
work. Considering the advantages of the hotly investigated
nitrogen-doped porous carbon materials in catalytic reactions
Furthermore, the superior catalytic activity of the Ir
catalysts thus inspires us to think what determines the intrinsic
activity for BA hydrogenation and how to design a highly active
catalyst for this reaction. In a previous work, we already
showed that the competitive adsorption of BA and H2
influences the catalytic activity and that relatively larger
adsorption energy of BA on the catalyst is essential to achieve
1
5
,
a serious of nitrogen-doped porous carbon supported
catalytic system (Pd, Ru, RuPd et al.) have been developed,
showing excellent catalytic performance under mild reaction
1
7
18
good performance . According to the Sabatier Principle
,
the adsorption of the reactants on the catalysts should neither
be too strong nor too weak, which resulting a volcano curve of
the activities vs adsorption energies. However, this is generally
suitable for gas phase reaction. When reactions were
conducted in liquid phase, especially in the case of strong
adsorption solvent like H O, the adsorption of solvent should
2
also be considered, resulting in the competing adsorption
1
9, 20
between the reactants and solvent
. Here we found that
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 1
Please do not adjust margins

Pleas Ge rdeo e nn o Ct ah de j mu si ts mt r ya rgins
Page 2 of 9
View Article Online
DOI: 10.1039/C7GC00387K
Journal Name
ARTICLE
the activities of the catalysts can be predicted by the and then the intrinsic reason of the high activity for the Ir-
distinction of the adsorption strength between H and H O on based catalysts. In general, particle size is a possible factor that
various catalysts for the aqueous hydrogenation of BA. A may influence catalytic activity
2
2
2
1, 22
. To address the influence
volcano curve considering competing adsorption between of the Ir particle size, HRTEM characterizations of various Ir-
reactants and solvent for aqueous aromatic ring based catalysts were performed (Figure 1). Surprisingly, little
hydrogenation was given out for the first time to our best difference of average size and distributions for Ir nanoparticles
knowledge. The volcano curve points out that a splendid (NPs) among various supports was observed, that is, the Ir NPs
catalyst in aqueous hydrogenation should make a compromise are uniformly dispersed on the supports with the average size
between the adsorption of reactants and solvent. In addition, of approximately 1.0~2.2 nm.
BA hydrogenation reaction pathways were also investigated
Furthermore, the XRD patterns for all the samples showed
both on Ir and IrPd catalysts in first principle to rule out the no characteristic peaks of Ir (Figure 2), indicating that the Ir
influence of the hydrogenation process to the catalytic activity. NPs are less than 4 nm in size and well dispersed on the
2
3
supports, in well agreement with the HRTEM results . Hence,
the influence of Ir particle size on the catalytic activity is
negligible in this case, which can also be proven by the
correlation of Ir particle size with activity (Figure S1).
Results and discussion
The hydrogenation of BA to CCA was carried out over
various heterogeneous Ir-based catalysts under the given mild
reaction conditions (Table 1). As is observed, Ir/γ-Al
2 3
O is the
most efficient catalyst for BA hydrogenation, achieving the
−
1
highest turnover frequency (TOF) of 40 h (Table 1, entry 1).
Other heterogeneous Ir-based catalysts including Ir/metal
oxide (Ir/CeO
2 2 2
, Ir/TiO , Ir/ZrO etc.), Ir/SBA-15 and Ir/carbon
(Ir/CN, Ir/AC) presented relatively lower hydrogenation
activities at the same reaction conditions. But, in the case of
Ir/MgO, neither substrate nor products could be detected by
GC-FID. This could be due to the reaction of strong basicity
MgO and the weak acidic BA, similar situation was also
1
6
appeared in our previous studies of Pd@MgO
.
a
Table 1. Catalytic results for different catalysts
Entry
Catalyst
Ir/γ-Al
Conv. (%) Sel. (%) TOF (h-1)
1
2
3
4
5
6
7
8
2
O
3
52
50
44
37
35
31
25
-
100
100
100
100
100
100
100
-
40
38
34
29
27
24
19
-
Ir/CeO
Ir/CN
2
Figure 2. XRD patterns of different catalysts.
Ir/TiO
2
As is known, the exposed crystal planes of metal particles
2
Ir/SBA-15
4, 25
may exert great influence on the catalytic performance
.
Ir/ZrO
Ir/AC
2
2 3
Taking two distinguished samples Ir/γ-Al O and Ir/CN as
typical examples, the crystallographic nature of the exposed Ir
NPs was examined by Cs-corrected HR-STEM to see weather
this could be the factor in affecting activity (Figure 3). As is
Ir/MgO
a
Reaction conditions: benzoic acid 0.5 mmol, catalyst (5 wt.% Ir) 50 mg, observed, the Ir NPs supported on the two supports have
o
solvent water 25 mL, 85 C, H
2
1 bar, 0.5 h.
similar morphologies, and both exhibit distinct crystal faces of
111) and (200) for Ir NPs. As a consequence, the results of
(
It is noteworthy that the activity of Ir-based catalysts has
significant advantage over the prior reported catalysts under
the mild reaction conditions (Table S1~S3). Even more
surprisingly, with just a change of the supported metal,
present Ir-based catalysts showed much better activity.
TEM analysis clearly showed that the size and the morphology
of the Ir NPs for various catalysts are almost identical.
Therefore, the difference in catalytic performance seems to
not be due to the morphology and particle size of Ir NPs on
different supports.
−1
Specifically, the TOF of Ir/γ-Al O (40 h ) was about 215 times
2
3
Previously reported work have shown that the valence
state of the exposed NPs is responsible for the resulted
26
−
higher than that of Pd@γ-Al O (0.186 h ), the TOF of Ir/CN
1
2
3
-
1
(
34 h ) was about 110 times, 20 times and 3 times higher than
-
catalytic performance , the valence state for the supported Ir
-1
-1
that of Ru/CN (0.31 h ), Pd/CN (1.7 h ) and RuPd/CN (11.5 h
was thus confirmed. Figure 4 shows the H -TPR profiles for Ir-
2
1
)
, respectively (Table S1~S3).
based samples obtained before reduction in hydrogen
atmosphere. All the catalysts present main hydrogen
From the above results we would like to firstly illustrate
the activity difference among the various Ir-based catalysts,
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 2
Please do not adjust margins

Page 3 of 9
Pleas Ge rdeo e nn o Ct ah de j mu si ts mt r ya rgins
View Article Online
DOI: 10.1039/C7GC00387K
Journal Name
ARTICLE
o
o
consumption peaks at about 140~300 C, it can be assigned to higher than 300 C can be attributed to iridium oxide species
the reduction of iridium oxide particles. The weaker peaks with higher interaction with
Figure 1. HRTEM images and particle size distribution of different catalysts.
percentage (68%), but the highest activity. So there should be
other factors such as the properties of the supports that
2
6
influence the reaction
.
2 3
Figure 3. Cs-corrected HR-STEM images of Ir nanoparticles on Ir/γ-Al O and
Ir/CN.
27
the support . Therefore, most of the iridium oxide species
deposited on all supports can be reduced to metallic state Ir at
o
00 C.
3
Previously reported work have shown that the valence
state of the exposed NPs is responsible for the resulted
2
6
catalytic performance , the valence state for the supported Ir
was thus confirmed. Figure 4 shows the H -TPR profiles for Ir-
2
based samples obtained before reduction in hydrogen
atmosphere. All the catalysts present main hydrogen
o
consumption peaks at about 140~300 C, it can be assigned to
Figure 4.
2
H -TPR patterns of various samples before reduction.
the reduction of iridium oxide particles. The weaker peaks
o
higher than 300 C can be attributed to iridium oxide species
2
7
with higher interaction with the support . Therefore, most of
the iridium oxide species deposited on all supports can be
o
reduced to metallic state Ir at 300 C.
Then, the Ir status of as-prepared Ir-based catalysts is
determined by XPS, as shown in Figure 5, and it is found that
0
the ratio of Ir in different catalysts various from 66 % to 75 %.
It is found that the corresponding conversion of BA was not
0
correlated well with the percentage of Ir in the corresponding
0
2 3
catalysts (Figure S2). For example, Ir/γ-Al O has moderate Ir
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 3
Please do not adjust margins

Pleas Ge rdeo e nn o Ct ah de j mu si ts mt r ya rgins
Page 4 of 9
View Article Online
DOI: 10.1039/C7GC00387K
Journal Name
ARTICLE
Figure 5. XPS spectra of Ir 4f in the different catalysts.
Firstly, the specific BET surface area (SBET) of supports were
firstly measured (Table S4). However, it is noted that SBET also
has no obvious connection with activity. This thus purged us to
consider other factors. Our previous observation has shown
that the hydrophilic property of the catalyst exerted great
2
8, 29
effect on the reaction activity
. Compared with the highly
aggregated solid catalysts, well-dispersed catalysts in the
reaction solution could improve the exposure of the catalyst to
substrates, thereby increasing the catalytic performance
significantly. This thus inspired us to investigate the dispersity
of the catalysts in the present catalytic conditions. As
expected, it is interesting to observe the difference between
Figure 6. Relationship between the BA conversion on various catalysts (the data
were taken from Table 1) and zeta potential of various supports.
Zeta potential, a key indicator of the stability of colloidal
dispersions was thus measured for all of the supports. Then
30
the reaction solutions of various catalysts after reaction under
8
the BA conversion on various catalysts versus the zeta
potential of the corresponding supports were correlated,
which was shown in Figure 6. As is shown in Figure 6 (A), the
zeta potentials are the actual measured data, the effect of zeta
potentials of supports towards the catalytic activities displays
an “inverted volcano” relationship. To be specific, a suspension
o
5 C, 1 bar H and 0.5 h (Figure S3). Specifically, all catalysts
2
can disperse well in water at the beginning of the reaction. But
after a while, except for Ir/CN, Ir/CeO and Ir/γ-Al O , other
2
2 3
catalysts gathered around the stirring bar with different
extent. This phenomenon thus inspired us to think the stability
difference of the catalysts in the present reaction condition
may have some intrinsic correlation with the final reaction
activity.
of particles with a higher absolute value of zeta potential is
more stable in comparison to suspensions with lower values
31
,
which was illustrated in Figure 6 (B). Interestingly, a linear
relation between the zeta potential absolute values and TOFs
was observed, that is to say, the TOF of the catalysts increases
steadily with the increase of the zeta potential absolute values.
In combination with the phenomenon of the different
dispersity for the involved catalysts in the reaction solution
(Figure S3), it can be speculated that the significant factor that
affects the catalytic activity of the Ir-based catalysts is the
dispersing stability of the supports in the reaction solution.
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 4
Please do not adjust margins

Page 5 of 9
Pleas Ge rdeo e nn o Ct ah de j mu si ts mt r ya rgins
View Article Online
DOI: 10.1039/C7GC00387K
Journal Name
ARTICLE
According to our previously reported work, CN material
was the best support for Pd and RuPd served as
heterogeneous catalysts for the chemoselective hydrogenation
of BA to CCA. Pd/CN exhibited much higher activity than other
Pd supported catalysts, including Pd/carbon materials (AC, HC,
1
6
2 3 2 2
CN0.132) and Pd/metal oxide (γ-Al O , CeO , TiO , ZrO etc.).
Likewise, among various RuPd/carbon materials (HC, AC, CNT),
RuPd/CN was found to give the most active system, meaning
the N atoms in CN are of crucial importance to the catalytic
1
7
activity. However, in the case of Ir-based catalysts, the CN is
not the most active support. It needs to be emphasized that all
the Ir-based catalysts are highly active for BA hydrogenation,
though the performance of the catalysts vary with the zeta
potentials of the supports, the effect of the supports is not as
obvious as the Pd-based cases. showing that Ir NPs is intrinsic
more active for BA hydrogenation.
So here come a couple of more profound questions. Why is
Ir NPs so active in nature for BA hydrogenation? What is the
best catalyst for this reaction in principle? What is the fatal
factor that determines the intrinsic activity of the catalyst in
this reaction? For heterogeneous catalysis, the activities of the
catalysts follow the Sabatier Principle, which states that the
interaction between the reactants and the catalysts should be
neither too strong nor too weak, giving a volcano curve
between the catalytic activities and the adsorption energies of
the reactants on different catalysts.
Firstly, the TOFs of BA hydrogenation are plotted with the
adsorption energies of BA on various noble metal and bi-metal
catalysts, as shown in Figure 7 (A), hoping to find a volcano
curve. However, the fact is not the case as expected. Then the
2
TOFs are plotted with the dissociation energies of H and
adsorption energies or dissociation energies of H O. Same
2
Figure 7. Catalytic activity plotted as a function of (A) the adsorption energies of
results are found yet (see Figure 7 (B, C)). Dissociation energies
of H O on Ru, RuPd and Rh are used since the dissociation of
2
BA (Eads(BA)), (B) the dissociation energies of H (Edis(H2)), (C) the adsorption
2
energies or dissociation energies of H O (Eads/dis(H2O)) on various metal/bi-metal
2
2
H O on these metal surfaces is exothermic and experiments
nanoparticles calculated from DFT.
32, 33
have shown that H O can dissociate on Ru spontaneously
.
The mismatch between the activity and the adsorption or
2
A volcano curve was established between the TOFs of BA
dissociation energy of merely a reactant or solvent molecule hydrogenation and the discrepancy of dissociation energies or
induces us to consider the competing adsorption between adsorption energies between H and H O after several tries
2
2
reactants and solvent.
(Figure 8). It can be seen that the optimum value of the
discrepancy of dissociation energies or adsorption energies
between H and H O is about -0.6 eV to achieve the highest
2
2
2
activity, which means that H has to bind 0.6 eV stronger than
H O. The dissociation of H on the catalysts of the left branch is
2
2
so strong that the chemisorption or dissociation of H
limited and therefore the solvation effect is weakened
leading to the decrease of the activity. On the other hand, the
2
O is
3
4
,
chemisorption or dissociation of H O on the catalysts of the
2
2
right branch is so strong that H dissociation is limited. To
further testify the validity of this volcano curve, we apply it to
predict more alloys such as IrRu (Ir as the main component)
and RuIr (Ru as the main component). As can be seen in Figure
8, these two alloys match well with the volcano curve
relationship. Hence the volcano curve presented here is rather
confidential. Regarding the activities in Figure 8, Ir, Pt and Rh
give the most promising performance among the studied
catalysts. However, according to the latest reports from USGS
(
United States Geological Survey), the price of Ir is much
cheaper than the other noble metal such as Pd, Pt and Rh
Table S5). Therefore, Ir catalyst is considered to be the best
(
candidate for BA or may be even aromatics hydrogenation
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 5
Please do not adjust margins

Pleas Ge rdeo e nn o Ct ah de j mu si ts mt r ya rgins
Page 6 of 9
View Article Online
DOI: 10.1039/C7GC00387K
Journal Name
ARTICLE
when considering the price and high-activity (Table S3, Table the activity of the catalyst. Therein, the reaction pathway on Ir
S5). Note that the catalysts on the right branch are generally
Ru-based, which is caused by the big dissociation energy of
and IrPd is investigated.
Since (111) is the main thermodynamically stable surface,
H
2
O as mention above (also see Figure 7(C)), developing
efficient Ru-based alloys in future by weakening the
dissociation of H O is also very intriguing owing to the much
the reaction pathways are discussed only on this surface. There
are mainly two reaction pathways for BA hydrogenation on
both Ir and IrPd catalysts as shown in Figure 9 and Figure 10,
respectively.
2
low price (Table S5).
Figure 8. Catalytic activity plotted as a function of the discrepancy of dissociation
energies or adsorption energies between H
metal nanoparticles calculated from DFT. For Ru, RuPd, RuIr and Rh, dissociation
energies of H O are used, and for other catalysts, adsorption energies of H O are
used. For H , dissociation energies are used. Reaction condition: catalyst (5 wt.%
2 2
and H O (ΔE) on various metal/bi-
2
2
2
o
loading on CN) 25 mg, solvent water 25 mL, 55 C, H
2
0.4 MPa, 1 h; TOF was
calculated based on the BA conversion about 30 %.
Note that the activity of Pd deviates from the volcano curve
to some extent. This deviation maybe caused by the fast
attainment of the low platform of H pressure before 0.4 MPa,
2
making an unfair activity evaluation (Figure 8). The reason why
the volcano curve is not established based on the discrepancy of
dissociation energies or adsorption energies between BA and
H
2
O or between BA and H
2
can be explained as follows.
Comparing the values of the adsorption or dissociation energies
for BA, H
2
and H
2
O, one can find that the values for BA are Figure 9. (A) Two possible reaction pathways of BA hydrogenation on Ir (111), (B)
The related transition states along the hydrogenation pathway beginning with C4
generally much bigger than H
and H O are closer. This means that the competing
adsorption between H and H O is much more serious than that
between BA and H or BA and H O. Therein the different
activity platforms vs pressures shown in Figure S4 reflect
more about the competing adsorption between H and H O in
the presence of solvent. However, it doesn’t mean that the
volcano curve and the different platforms of H pressure are
2 2
and H O, while the values for
site. The initial and final structure for each hydrogenation step along the
hydrogenation pathway beginning with C4 site and the related structures
beginning with C1 site are listed in Figure S5 and Figure S6.
H
2
2
2
2
2
2
H
2
2
2
2
irrelevant to the adsorption of BA. Actually the situation could
be very complex if the adsorption strength of BA is comparable
to that of H or H O. And the strong adsorption of BA is quite
2 2
essential. To sum up, the competing adsorption between the
reactants and solvent on the catalysts greatly affects the
performance of the catalysts for BA or even aromatic ring
aqueous hydrogenation. This phenomenon should be
emphasized for rational catalyst design in liquid catalysis.
In consideration of the reaction involves multistep
hydrogenation process, the reaction channel may also influence
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 6
Please do not adjust margins

Page 7 of 9
Pleas Ge rdeo e nn o Ct ah de j mu si ts mt r ya rgins
View Article Online
DOI: 10.1039/C7GC00387K
Journal Name
ARTICLE
Figure 10. Two possible reaction pathways of BA hydrogenation on IrPd (111).
2 6 2
the H IrCl •6H O precursor was dissolved in deionized water
and impregnated into the CN to produce a catalyst of 5 wt. % Ir
One is first hydrogenating at C1 site, the other is first
hydrogenating at C4 site (see TS1 in Figure 9(b)). For Ir catalyst,
the two reaction pathways are comparable since their highest
energy barriers (1.09 and 1.00 eV respectively) are similar. The
rate-determining steps changes from the second
hydrogenation step to the fifth and sixth steps respectively for
over CN. After the impregnation, the catalysts were dried at 70
o
C overnight and then calcined in a 500 mL/min H
2
flow at 300
o
o
C for 1 hour with a heating rate of 10 C/ min. The obtained
catalyst was denoted as Ir/CN.
Catalyst characterization. The BET specific surface areas of
the two reaction pathways when alloyed with Pd. However, various catalysts materials were measured on a surface area
the kinetics of the hydrogenation process undergoes a very and porosity analyzer (Micromeritics, ASAP 2020 HD88). High-
limited change, with the difference of less than 0.1 eV for the resolution transmission electron microscopy (HRTEM)
lowest energy barriers of the rate-determining steps on Ir and
IrPd (1.00 eV for Ir, 0.91 eV for IrPd).
From the above, it is reasonable to rule out the influence of
the hydrogenation process to the catalytic activity. Hence the
performance of the catalysts is guarded by the competing
adsorption between the reactants and solvent.
characterization was carried out on an FEI Tecnai G2 F20 S-
TWIN microscope. Spherical aberration corrected scanning
transmission electron microscope (Cs-corrected HR-STEM)
analyses were recorded on a Titan G280-200 w/ ChemiSTEM
instrument. The X-ray diffraction (XRD) data was collected at
room temperature using a wide angle X-ray diffraction (Ultima
TV) equipped with Cu Ka radiation (1.54 Å). The X-ray
photoelectron spectra (XPS) were obtained by an ESCALAB
MARK II spherical analyzer using an aluminum-magnesium
binode (Al 1486.6 eV, Mg 1253.6 eV) X-ray source. The XPS
spectrum was corrected with reference to the C1s peak at
Conclusions
Firstly, Ir-based catalysts were found to be superior active
-1
for BA hydrogenation with a record TOF (up to 40 h ) while
maintaining 100% selectivity under mild reaction conditions.
By excluding various factors, we speculated that the main true
factor affecting the activity of the Ir-based catalysts of
different supports is the dispersing stability of the supports in
the reaction solution, rather than the electronic effects of the
supports. Secondly, the nature of the high intrinsic activity of Ir
towards BA hydrogenation is revealed. The competing
adsorption between the reactants and solvent on the catalysts
greatly affects the performance of the catalysts for BA or even
aromatic ring aqueous hydrogenation and should be valued.
An activity volcano curve considering the competing
adsorption between reactants and solvent for aqueous
aromatic ring hydrogenation was given out for the first time.
And the high activity of Ir can be deduced by the proper
discrepancy of dissociation energies or adsorption energies
2
84.5 eV so as to correct the charging effect. Hydrogen
temperature-programmed reduction (H -TPR) was conducted
2
in FINESORB-3010 apparatus equipped with a thermal
conductivity detector (TCD). Before a TPR run, catalysts were
pretreated in Ar at 473 K for 2 hours. TPR was performed in a
2
flow of 10% H /Ar mixture gas at a flow rate of 10 sccm with a
o
temperature ramp of 10 C /min. Zeta potential was measured
with 2 mg of samples dispersed in 10 mL deionized water (pH
=
6.0) on a zeta potential analyzer Nano-Zetasizer (Malvern
Instruments, Worcestershire, UK).
Hydrogenation of BA. Hydrogenation of BA under atmospheric
pressure: In a typical process, 0.5 mmol BA and 50 mg Ir/γ-
2 3
Al O were put into a three-neck flask, and 25 mL deionized
water was employed as a green solvent. The reaction was
o
carried out at a temperature of 85 C with magnetic stirring at
2 2
between H and H O on the catalysts.
a speed of 1000 rpm. Before reaction, a balloon filled with
hydrogen was connected to the flask to replace the air. After
the reaction, the contents of products and substrate were
determined by GC-FID and the products were identified by GC-
MS.
Based on the comprehensive study, the combination of
experiment and theory in this work not only help to provide
explanations for the experimentally observed phenomena, but
also provide guidance for the rational design of promising
catalysts for BA or even aromatic ring hydrogenation in true
reaction conditions for practical applications.
Hydrogenation of BA under high pressure: The reaction was
performed in a 100 mL stainless steel autoclave equipped with
a magnetic stirrer. In a typical experiment, BA, catalyst and
H
2
O were introduced into the reactor, seal the reactor. After
Experimental Section
being purged by low H pressure three times to remove air, the
2
Materials.
RuCl •xH O (Ru, 40%), H
Rh, 27.5%), γ-Al , ZrO
received from Aladdin Chemistry Co., Ltd. Benzoic acid (AR,
99.5%) was purchased from Sinopharm Chemical Reagent
H
2
IrCl
6
•6H
2
O
(Ir, 35%), Na
PtCl •xH O (Pt, ≥37.5%), H
, CeO , TiO and MgO were used as
2
PdCl
4
(Pd, 40%),
reactor was pressurized with hydrogen at a desired pressure.
Then the reactor was heated to the reaction temperature.
After reaction, the contents of products and substrate were
determined by GC-FID and the products were identified by GC-
MS.
3
2
2
6
2
2
Cl Rh
6 3
N
(
2
O
3
2
2
2
>
Co., Ltd. SBA-15 was purchased from Nanjing XFNANO
Materials Tech Co., Ltd. AC was purchased from Jiangsu
Liangyou Environmental Technology Co., Ltd.
Computational setup. The calculations reported here are
performed by using periodic, spin-polarized DFT as
3
5, 36
implemented in Vienna ab initio program package (VASP)
.
Catalyst preparation. The N-doped carbon (CN) was
1
The electron-ion interactions are described by the projector
7
3
augmented wave (PAW) method proposed by Blöchl and
7
synthesized as described previously . All of the catalysts were
prepared by the impregnation method. In a typical process,
3
implemented by Kresse . RPBE functional is used as
8
39
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 7
Please do not adjust margins

Pleas Ge rdeo e nn o Ct ah de j mu si ts mt r ya rgins
Page 8 of 9
View Article Online
DOI: 10.1039/C7GC00387K
Journal Name
ARTICLE
exchange-correlation functional approximation for the 8. J. Anderson, F.-M. McKenna, A. Linares-Solano and R. K.
reaction pathway calculation and only gamma point is used for Wells, Catal Lett., 2007, 119, 16-20.
the Brillouin zone sampling. However, for the adsorption or 9. Z. Jiang, G. Lan, X. Liu, H. Tang and Y. Li, Catal. Sci. Technol.,
4
0
dissociation energy calculation, PBEsol functional with van
der Waals interaction involved and a (2×2×1) k-point grid are 10. H. Wang and F. Zhao, Int. J. Mol. Sci., 2007,
used to obtain more precise results. A plane wave basis set 11. T. Maegawa, A. Akashi, K. Yaguchi, Y. Iwasaki, M. Shigetsura,
2016, 6, 7259-7266.
8
, 628.
with an energy cutoff of 400 eV is used. A p (5×5) surpercell
containing a four-layer slab with 100 atoms was modeled. The
Y. Monguchi and H. Sajiki, Chem. Eur. J., 2009, 15, 6953-
6963.
periodic condition is employed along the x and y direction. The 12. J. M. Thomas, B. F. G. Johnson, R. Raja, G. Sankar and P. A.
vacuum space along the z direction was set to be 13 Å. For Midgley, Acc. Chem. Res., 2003, 36, 20-30.
alloy catalysts, taking RuPd catalyst as an example, Pd was 13. R. Raja, T. Khimyak, J. M. Thomas, S. Hermans and B. F. G.
alloyed to the sub-layer of Ru catalyst. The upper two layer Johnson, Angew. Chem. Int. Ed., 2001, 40, 4638-4642.
atoms in the cell are allowed to relax during the structure 14. Z. Guo, L. Hu, H.-h. Yu, X. Cao and H. Gu, RSC Adv., 2012,
optimization and the bottom two layer atoms are fixed. The 3477-3480.
relaxation is stopped when the force residue on the atom is 15. M. Li, F. Xu, H. Li and Y. Wang, Catal. Sci. Technol., 2016,
smaller than 0.02 eV/ Å. The transition states are calculated by 3670-3693.
using the climbing image nudged elastic band (CI-NEB) method 16. X. Xu, M. Tang, M. Li, H. Li and Y. Wang, ACS Catal., 2014,
2
6
4
,
,
,
4
1
.
3132-3135.
The adsorption and dissociation energy for molecule 17. M. Tang, S. Mao, M. Li, Z. Wei, F. Xu, H. Li and Y. Wang, ACS
chemisorption or dissociation are defined respectively as:
Eads/dis = Etot – Eslab – Emol
Catal., 2015, 5, 3100-3107.
18. R. A. Van Santen and M. Neurock, Catal. Rev., 1995, 37, 557-
Where Etot is the total energy after a molecule adsorption
698.
or dissociation the on catalysts; Eslab is the energy of the clean 19. C.-R. Chang, Z.-Q. Huang and J. Li, WIREs Comput. Mol. Sci.,
catalyst alone; Emol is the energy of the molecule in the gas
phase.
2016, 6, 679-693.
20. S. Siahrostami and A. Vojvodic, J. Phys. Chem. C., 2015, 119
,
The energy discrepancy is defined as:
ΔE = Eads/dis (1) - Eads/dis (2)
1032-1037.
21. X.-F. Yang, A. Wang, B. Qiao, J. Li, J. Liu and T. Zhang, Acc.
Chem. Res., 2013, 46, 1740-1748.
where (1) and (2) represent different molecules.
1
7
More details can refer to our previous work
.
22. M. Valden, X. Lai and D. W. Goodman, Science, 1998, 281
647-1650.
3. C. Wang, B. Li, H. Lin and Y. Yuan, J. Power Sources, 2012,
02, 200-208.
4. K. M. Bratlie, H. Lee, K. Komvopoulos, P. Yang and G. A.
Somorjai, Nano Lett., 2007, , 3097-3101.
,
1
2
2
2
Acknowledgements
2
Financial support from the National Natural Science
Foundation of China (21622308, 91534114, 21376208), the
MOST (2016YFA0202900), the Fundamental Research Funds
for the Central Universities, and the computing time supported
by Special Program for Applied Research on Super
Computation of the NSFC-Guangdong Joint Fund (the second
phase) are greatly appreciated.
7
5. E. Schmidt, A. Vargas, T. Mallat and A. Baiker, J. Am. Chem.
Soc., 2009, 131, 12358-12367.
2
2
6. X. Xu, H. Li and Y. Wang, ChemCatChem, 2014, 6, 3328-3332.
7. A. Aguilar-Tapia, R. Zanella, C. Calers, C. Louis and L.
Delannoy, Phys. Chem. Chem. Phys., 2015, 17, 28022-28032.
8. X. Xu, Y. Li, Y. Gong, P. Zhang, H. Li and Y. Wang, J. Am. Chem.
Soc., 2012, 134, 16987-16990.
2
References
29. M. Li, X. Xu, Y. Gong, Z. Wei, Z. Hou, H. Li and Y. Wang, Green
Chem., 2014, 16, 4371-4377.
1
2
3
4
.
.
.
.
M. Zahmakıran, Y. Tonbul and S. Özkar, J. Am. Chem. Soc.,
010, 132, 6541-6549.
3
0. R. W. O'Brien, B. R. Midmore, A. Lamb and R. J. Hunter,
Faraday Discuss. Chem. Soc., 1990, 90, 301-312.
1. D. Hanaor, M. Michelazzi, C. Leonelli and C. C. Sorrell, J. Eur.
Ceram. Soc., 2012, 32, 235-244.
2
R. R. Deshmukh, J. W. Lee, U. S. Shin, J. Y. Lee and C. E. Song,
Angew. Chem., Int. Ed., 2008, 47, 8615-8617.
3
J. A. Bigot, T. A. J. Meijerink and L. J. Revallier, Recl. Trav.
Chim. Pays-Bas., 1962, 81, 363-364.
3
3
2. P. J. Feibelman, Science, 2002, 295, 99-102.
3. S. Maier, I. Stass, J. I. Cerda and M. Salmeron, Phys. Rev.
Lett., 2014, 112, 126101.
B. S. Moore, H. Cho, R. Casati, E. Kennedy, K. A. Reynolds, U.
Mocek, J. M. Beale and H. G. Floss, J. Am. Chem. Soc., 1993,
3
3
4. C. Zhu, S. Kais, X. C. Zeng, J. S. Francisco and I. Gladich, J. Am.
Chem. Soc., 2017, 139, 27-30.
115, 5254-5266.
5
6
.
.
X. Zhang, B. Zong and M. Qiao, AIChE J, 2009, 55, 2382-2388.
H. Cao, H. Liu and A. Dömling, Chem. Eur. J., 2010, 16, 12296-
5. G. Kresse and J. Furthmüller, Comp. Mater. Sci., 1996,
0.
6, 15-
5
1
2298.
3
3
3
6. G. Kresse and J. Furthmüller, Phys. Rev. B, 1996, 54, 11169.
7. P. E. Blöchl, Phys. Rev. B, 1994, 50, 17953.
7
.
B. Zong, X. Zhang and M. Qiao, AIChE J., 2009, 55, 192-197.
8. G. Kresse and D. Joubert, Phys. Rev. B, 1999, 59, 1758.
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 8
Please do not adjust margins

Page 9 of 9
Pleas Ge rdeo e nn o Ct ah de j mu si ts mt r ya rgins
View Article Online
DOI: 10.1039/C7GC00387K
Journal Name
ARTICLE
3
4
9. B. Hammer, L. B. Hansen and J. K. Nørskov, Phys. Rev. B,
999, 59, 7413.
1
0. J. P. Perdew, A. Ruzsinszky, G. I. Csonka, O. A. Vydrov, G. E.
Scuseria, L. A. Constantin, X. Zhou and K. Burke, Phys. Rev.
Lett., 2008, 100, 136406.
4
1. G. Henkelman, B. P. Uberuaga and H. Jonsson, J. Chem.
Phys., 2000, 113, 9901-9904.
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 9
Please do not adjust margins