Cross-Linked "poisonous" Polymer: Thermochemically Stable Catalyst Support for Tuning Chemoselectivity

Article abstract of DOI:10.1021/acscatal.5b02613

Designed catalyst poisons can be deliberately added in various reactions for tuning chemoselectivity. In general, the poisons are "transient" selectivity modifiers that are readily leached out during reactions and thus should be continuously fed to maintain the selectivity. In this work, we supported Pd catalysts on a thermochemically stable cross-linked polymer containing diphenyl sulfide linkages, which can simultaneously act as a catalyst support and a "permanent" selectivity modifier. The entire surfaces of the Pd clusters were ligated (or poisoned) by sulfide groups of the polymer support. The sulfide groups capping the Pd surface behaved like a "molecular gate" that enabled exceptionally discriminative adsorption of alkynes over alkenes. H₂/D₂ isotope exchange revealed that the capped Pd surface alone is inactive for H₂ (or D₂) dissociation, but in the presence of coflowing acetylene (alkyne), it becomes active for H₂ dissociation as well as acetylene hydrogenation. The results indicated that acetylene adsorbs on the Pd surface and enables cooperative adsorption of H₂. In contrast, ethylene (alkene) did not facilitate H₂-D₂ exchange, and hydrogenation of ethylene was not observed. The results indicated that alkynes can induce decapping of the sulfide groups from the Pd surface, while alkenes with weaker adsorption strength cannot. The discriminative adsorption of alkynes over alkenes led to highly chemoselective hydrogenation of various alkynes to alkenes with minimal overhydrogenation and the conversion of side functional groups. The catalytic functions can be retained over a long reaction period due to the high thermochemical stability of the polymer.

Full text of DOI:10.1021/acscatal.5b02613

Research Article
pubs.acs.org/acscatalysis
Cross-Linked “Poisonous” Polymer: Thermochemically Stable
Catalyst Support for Tuning Chemoselectivity
Seongho Yun,†,# Songhyun Lee,†,# Sunwoo Yook,† Hasmukh A. Patel,‡ Cafer T. Yavuz,‡
and Minkee Choi*,†
†Department of Chemical and Biomolecular Engineering, Korea Advanced Institute of Science and Technology, Daejeon 305-701,
Republic of Korea
‡Graduate School of Energy, Environment, Water, and Sustainability (EEWS), Korea Advanced Institute of Science and Technology,
Daejeon 305-701, Republic of Korea
S
* Supporting Information
ABSTRACT: Designed catalyst poisons can be deliberately
added in various reactions for tuning chemoselectivity. In
general, the poisons are “transient” selectivity modifiers that
are readily leached out during reactions and thus should be
continuously fed to maintain the selectivity. In this work, we
supported Pd catalysts on a thermochemically stable cross-
linked polymer containing diphenyl sulfide linkages, which can
simultaneously act as a catalyst support and a “permanent”
selectivity modifier. The entire surfaces of the Pd clusters were
ligated (or poisoned) by sulfide groups of the polymer
support. The sulfide groups capping the Pd surface behaved like a “molecular gate” that enabled exceptionally discriminative
adsorption of alkynes over alkenes. H2/D2 isotope exchange revealed that the capped Pd surface alone is inactive for H2 (or D2)
dissociation, but in the presence of coflowing acetylene (alkyne), it becomes active for H2 dissociation as well as acetylene
hydrogenation. The results indicated that acetylene adsorbs on the Pd surface and enables cooperative adsorption of H2. In
contrast, ethylene (alkene) did not facilitate H2−D2 exchange, and hydrogenation of ethylene was not observed. The results
indicated that alkynes can induce decapping of the sulfide groups from the Pd surface, while alkenes with weaker adsorption
strength cannot. The discriminative adsorption of alkynes over alkenes led to highly chemoselective hydrogenation of various
alkynes to alkenes with minimal overhydrogenation and the conversion of side functional groups. The catalytic functions can be
retained over a long reaction period due to the high thermochemical stability of the polymer.
KEYWORDS: hydrogenation, chemoselectivity, polymers, poison, modifier, palladium
either during the catalyst preparation24−30 or to the reactant
INTRODUCTION
■
feeds during the reaction.22,23 These modifiers can be adsorbed
Partial hydrogenation of alkynes to alkenes is an important
on the catalyst surface in moderate strength, which modulates
model reaction for fundamental investigation of chemo-
the adsorption/desorption behaviors of reactants/intermedi-
selectivity in heterogeneous catalysis and also a key trans-
ates25 or preferably poison the catalytic active sites for
formation in fine chemical synthesis.1,2 Despite various efforts,
development of a catalyst system simultaneously showing high
unselective reactions.28 The beauty of this strategy is that, in
principle, it can provide a route for precise tuning of catalytic
selectivity and long catalyst lifetime remains a challenge. It has
properties, because a wide variety of modifiers having diverse
been demonstrated that Pd-based catalysts are the most
effective choice for selective hydrogenation, although the
unmodified Pd catalyst is not selective and readily forms an
overhydrogenated product, alkanes. Conventional approaches
molecular structures and adsorption properties can be designed
via organic synthesis. On the other hand, the molecular
modifiers also have a significant limitation compared to
inorganic promoters. In general, the modifiers should be
continuously supplied to the reaction feeds to maintain
selectivity,31 because they are readily decomposed27 or leached
out during reactions.32,33 In other words, the molecular
modifiers are not a “permanent” selectivity modifier but rather
for enhancing the catalytic selectivity of Pd to alkenes include
alloying with secondary metals (e.g., Pb,3−5 Cu,5−8 Ag,5,9−11
Au,5,12,13 and Ga5,14,15), surface decoration with various metal
oxide promoters,16 and interstitial modification with C15,17−19
and B.20 Recently, it was also reported that single-site metal
catalysts are highly selective to alkene production.6,7,11,21
Another important strategy is to add various modifiers (or
catalyst poisons) such as carbon monoxide22,23 and heter-
oatom-containing (N,24−26 S,27,28 P29,30) organic molecules
Received: November 19, 2015
Revised: February 3, 2016
© XXXX American Chemical Society
2435
DOI: 10.1021/acscatal.5b02613
ACS Catal. 2016, 6, 2435−2442

ACS Catalysis
Research Article
a “transient” modifier.28 This requires continuous consumption
of modifiers and additional separation steps for removing the
modifiers from products.
In recent years, metal catalysts supported on organic
polymers have attracted increasing attention in various catalytic
reactions as a means of achieving enhanced activity, selectivity,
and recyclability, compared with metal catalysts on conven-
tional inorganic supports.34−36 In the present work, we
supported Pd catalysts on a thermochemically stable covalent
organic polymer (COP) containing diphenyl sulfide linkages
which can act as a permanent modifier (Scheme 1). Pd clusters
were formed on the surface of the polymer matrix, where the
sulfide groups of the polymer framework can cover the entire
Pd surface. The sulfide functional groups capping the Pd
surface behave like a “molecular gate” that enables exceptionally
discriminative adsorption of alkynes over alkenes. This leads to
highly selective partial hydrogenation of various alkynes to
alkenes with minimal conversion of other functional groups.
Due to the high thermochemical stability and suppressed coke
formation, the catalyst retains its catalytic functions during
acetylene hydrogenation at 393 K over a long period (≫15
days). The result indicates that the COP simultaneously acts as
a stable catalyst support and a permanent selectivity modifier.
Scheme 1. Synthesis of Pd Catalyst Supported on a Covalent
Organic Polymer (COP) and Its Selective Alkyne-to-Alkene
Hydrogenation Mechanism
RESULTS AND DISCUSSION
■
The COP was synthesized by a single-step nucleophilic
substitution of cyanuric chloride with 4,4′-thiobisbenzenethiol
(Scheme 1), following a modified version of a procedure
reported elsewhere.37 Namely, we modified the polymer
synthesis conditions to maximize the surface area of the
essentially nonporous polymer (Figure S1),37 thereby making
the material more suitable as a catalyst support. It appears that
the polymer morphology is highly sensitive to solvent polarity,
and a polymer having a nanoneedle morphology (Figure 1a)
can be synthesized by using cyclohexanone as a solvent. The
COP showed several well-distinguished XRD peaks (Figure S2)
that indicates crystallinity, although earlier work reported a
completely amorphous nature for this polymer.37 The result
implies that solvent polarity significantly affects the rate of
polymerization and self-assembling rate of growing polymer
chains which determine the crystallinity of the resultant
polymer framework. The unique nanoneedle morphology
appears to have originated from the crystalline framework
nature. According to N2 adsorption−desorption at 77 K
(Figure S1), the resultant polymer exhibited a BET surface area
of 55 m2 g−1 and only the presence of interparticular
macroporosity (pore volume: 0.53 mL g−1). Elemental analysis
revealed that the polymer contains 9.1 wt % N and 31 wt % S,
Figure 1. (a) TEM image and (b) solid-state CP/MAS 13C NMR spectrum of COP. (c) Fourier transforms of k3-weighted Pd K-edge EXAFS of Pd/
γ-Al2O3, Pd/COP, PdO, PdS, and Pd foil. (d) HAADF-STEM image of Pd/COP. The inset shows a histogram of particle size distribution for Pd.
Surface-area-weighted cluster diameter, dTEM, was calculated from dTEM = Σnidi3/Σnidi2. (e) XPS-Pd (3d) spectra of Pd/COP and Pd/γ-Al2O3. (f)
Weight loss and H2S evolution of Pd/COP during temperature ramping under H2, which were measured with TGA-MS.
2436
DOI: 10.1021/acscatal.5b02613
ACS Catal. 2016, 6, 2435−2442

ACS Catalysis
Research Article
Table 1. Hydrogenation of Various Alkyne Compounds over Pd/COP and Pd/γ-Al2O3
a
Catalytic turnover rate (TOR) was calculated on the basis of the total number of Pd rather than that of the accessible surface Pd determined by H2
or CO chemisorption. This is because Pd/COP showed almost zero H/Pd and CO/Pd values in chemisorptions, which led to a scientifically
meaningless TOR value of “infinity” for the latter definition.
which are very close to the theoretically predicted values (9.3
wt % N and 32 wt % S). CP/MAS 13C NMR spectrum showed
three peaks, which can be assigned to triazine ring carbons (179
ppm) and aromatic benzene ring carbons (134 and 128 ppm),
respectively (Figure 1b).
reported to be 2.05 Å38). We believe that the triazine ring is
mainly functioning as a building block for cross-linked polymer,
whereas the sulfur groups of 4,4′-thiobisbenzenethiol are
predominantly interacting with the Pd surface, as shown in
Scheme 1. On the other hand, a conventional Pd/γ-Al2O3
catalyst prepared for comparison predominantly showed Pd−
Pd coordination (CN: 10.4) (Figure 1c).
XPS for Pd (3d) (Figure 1e) were deconvoluted by assuming
that a signal for an individual species comprises 3d3/2 and 3d5/2
peaks with an intensity ratio of 2:3, which are separated by an
energy difference of 5.26 eV.39 In the case of Pd/COP, the
3d5/2 peak at a binding energy of 335.3 eV indicates the
presence of metallic Pd (Pd0).39 The more intense peak at
337.1 eV represents Pd2+−S species.40 On the other hand, Pd/
γ-Al2O3 showed a major peak at a binding energy of 335.3 eV,
which represents Pd0 species.
In order to analyze the surface accessibility of the supported
Pd catalysts, H2 and CO chemisorption experiments were
carried out. The results showed that Pd/COP cannot adsorb
H2 or CO on the surface of Pd clusters (H/Pd ≪0.01; CO/Pd
≪0.01). In contrast, Pd/γ-Al2O3 catalyst showed appreciable
chemisorption amounts for both gas titrants (H/Pd = 0.53;
CO/Pd = 0.58). The results reveal that the entire surface of the
Pd clusters in Pd/COP is ligated (or poisoned) with the
diphenyl sulfide-like moieties of the polymer matrix, and thus,
no clean Pd surface is accessible to H2 and CO. Considering
the fact that Pd was supported on the surface of premade
Pd catalysts were supported on the COP by wet
impregnation of Pd(acac)2 in THF solution, followed by H2
reduction at 453 K. ICP-MS analysis revealed that Pd content is
1.2 wt %. High-angle annular dark field scanning transmission
electron microscopy (HAADF-STEM) revealed that Pd clusters
having an average diameter of ca. 2.0 nm were dispersed on the
polymer support (Figure 1d and S3). No characteristic peak for
Pd0 was observed in the XRD pattern of Pd/COP (Figure S2),
which also supports the predominant presence of highly
dispersed small Pd clusters. The radial distribution functions
obtained from the k3-weighted Pd K-edge extended X-ray
absorption fine structure (EXAFS) showed that the catalyst
possesses substantial Pd−S coordination (CN: 2.67) at 2.33 Å
along with Pd−Pd coordination (CN: 7.01) at 2.78 Å (Figure
1c and Table S1 in the Supporting Information). The
substantial Pd−S coordination indicated that the surface of
the Pd catalyst is ligated with the sulfur moieties of the polymer
framework. Since double-shell EXAFS fitting with Pd−Pd and
Pd−S scattering shells already provided a good fitting result (R
factor: 0.004), we could not find any strong evidence proving
the interaction between Pd and the nitrogen groups of the
triazine rings of COP in EXAFS (Pd−N bond length is
2437
DOI: 10.1021/acscatal.5b02613
ACS Catal. 2016, 6, 2435−2442

ACS Catalysis
Research Article
nonporous COP particles, it is remarkable that the entire Pd
surface is decorated with a sulfur-containing polymer frame-
work, which resulted in zero chemisorption of H2 and CO. We
believe that the organic COP framework has at least some
flexibility, and thus, the polymer framework nearby supported
Pd clusters can cover the Pd surface to maximize the strong
interaction between sulfur moieties and Pd during the catalyst
preparation. This is conceptually analogous to the “strong
metal−support interaction (SMSI)” often found in metal
catalysts supported on inorganic metal oxide supports.41−43
Conventional organic polymers generally show very low
thermochemical stability and thus have been considered to be
unsuitable as catalytic materials for reactions at elevated
temperatures (>373 K). In this sense, it is remarkable that
the present COP possesses extra-high thermal stability up to
673 K under an inert atmosphere due to its highly cross-linked
aromatic structure.37 Nevertheless, thermochemical stability
under catalytically relevant conditions (i.e., in the presence of
embedded Pd catalysts and a H2 atmosphere) may be
significantly lower than the one under inert gas, because Pd
under H2 can catalyze hydrodesulfurization and hydrogenolysis
of the polymer framework. With this regard, we carried out a
thermogravimetric analyzer-mass spectrometer (TGA-MS)
analysis of Pd/COP under H2 (Figure 1f). The result shows
that polymer decomposition starts at ∼503 K with concomitant
generation of H2S, indicating that polymer degradation starts
via hydrodesulfurization. Although the decomposition temper-
ature is lower than the one under inert gas, it is still high
enough for carrying out various chemoselective hydrogenations
described in this work.
Gas-phase hydrogenation of alkynes with various functional
groups was carried out with Pd/COP and Pd/γ-Al2O3 catalysts
in a plug-flow reactor (Table 1). Pd/COP showed generally
lower catalytic turnover rates (TOR) than Pd/γ-Al2O3 due to
the surface poisoning with the sulfur moieties. As shown in
reaction 1 with an internal alkyne, a major product on Pd/COP
is cis-alkene. This indicates that the hydrogenation takes place
via syn-addition of H atoms.4 In contrast, Pd/γ-Al2O3 showed
the production of various isomers having different double bond
positions and cis/trans geometries (Table 1), indicating that the
alkenes can be further isomerized on the unpromoted Pd
surface.44,45 Pd/γ-Al2O3 also produced a substantial amount of
an overhydrogenated product, octane (15%). Similarly, in
reactions 2−3 with the alkynes having CC and CO side
functional groups, Pd/COP exhibited superior selectivities
(85−89%) to corresponding alkenes with minimal hydro-
genation of the side functional groups. In contrast, Pd/γ-Al2O3
produced substantial amounts of overhydrogenated products
(Table 1). In reactions 4−6 with alkynes having cyclopropane,
−OH and −NH2 functional groups, Pd/COP produced
predominantly the corresponding alkenes. In contrast, Pd/γ-
Al2O3 showed more pronounced production of overhydro-
genated products and hydrocarbons without cyclopropane ring
structure and −OH functional groups (Table 1) due to
hydrogenolysis of these functional groups. We also carried out
the liquid-phase phenylacetylene (aromatic alkyne) hydro-
genation using Pd/COP and Pd/γ-Al2O3. Pd/COP showed a
significantly higher selectivity (90.5%) for the corresponding
alkene (styrene) than Pd/γ-Al2O3 (26.2%), that produced a
large amount of overhydrogenated product, ethylbenzene
(Figure 2). Pd/COP showed excellent recyclability up to five
cycles without any substantial loss of activity or selectivity. ICP-
MS analysis also confirmed that Pd leaching was negligible
Figure 2. Liquid-phase phenylacetylene hydrogenation with Pd/γ-
Al2O3 (1 cycle) and Pd/COP (5 cycles) (reaction condition: 1 MPa
H2, 303 K, 5 h reaction).
during the reaction cycles (Pd content was maintained at 1.2 wt
%). On the basis of the results, it can be concluded that Pd/
COP exhibits superior chemoselectivities in converting alkynes
to corresponding alkenes with minimal overhydrogenation,
isomerization, and hydrogenolysis of various side functional
groups.
To further explore the selective alkyne hydrogenation
properties, we investigated acetylene partial hydrogenation in
an ethylene-rich feedstock (ethylene/acetylene = 82). The
reaction is highly important in industry for removing acetylene
impurity from ethylene produced in naphtha crackers.1,2 For
industrial applications, high catalytic selectivity to ethylene over
ethane at a very high acetylene conversion level (>99%)46 is
required for minimizing the loss of ethylene. As shown in
Figure 3a, Pd/γ-Al2O3 showed low selectivity to ethylene (29−
56%), and substantial formation of an overhydrogenated
Figure 3. Acetylene conversion and product selectivities as a function
of time-on-stream during acetylene hydrogenation in an ethylene-rich
stream (reaction condition: 0.6 kPa C2H2, 0.6 kPa C3H8, 49.3 kPa
C2H4, 0.9 kPa H2, and 48.6 kPa N2 at 393 K). (a) Pd/γ-Al2O3, (b) Pd/
γ-Al2O3 modified with diphenyl sulfide, and (c) Pd/COP. To obtain
complete conversion level during the initial reaction period, space
velocities of 0.3, 0.04, and 0.02 molC H molPd−1 s−1 were used for Pd/
2
2
γ-Al2O3, Pd/γ-Al2O3 modified with diphenyl sulfide, and Pd/COP
catalyst, respectively.
2438
DOI: 10.1021/acscatal.5b02613
ACS Catal. 2016, 6, 2435−2442

ACS Catalysis
Research Article
product, ethane, was detected. Furthermore, the catalytic
conversion and ethylene selectivity rapidly decrease with
time-on-stream, which is consistent with earlier reports.1 It
was proposed that coke deposits on the catalyst surface can act
as an active site for unselective hydrogenation of acetylene to
ethane via hydrogen spillover.1,47 Indeed, after 1 day of
reaction, the deactivated catalyst showed substantial coke
formation (28 wt % by TGA). In clear contrast, Pd/COP
(Figure 3c) showed 70−80% ethylene selectivity even at a
complete acetylene conversion level (>99%). Most notably, the
catalytic activity and selectivity did not significantly deteriorate
over 15 days. Elemental analysis results revealed that Pd/COP
contains only 8.4 wt % coke even after 15 days reaction, and the
original S/N elemental mass ratio (3.3) did not change
significantly (S/N of the fresh catalyst: 3.5). The results
strongly support that, despite the organic nature of the COP,
the catalyst is thermochemically stable over a long reaction
period.
To compare the thermochemical stabilities of the covalently
bonded and physically adsorbed modifiers, we impregnated the
presynthesized Pd/γ-Al2O3 with 10 wt % diphenyl sulfide as a
molecular modifier. The modified Pd/γ-Al2O3 was also tested
for acetylene hydrogenation in an ethylene-rich stream (Figure
3b). The catalyst initially showed improved selectivity to
ethylene (∼70%) compared with a pristine Pd/γ-Al2O3.
However, the catalyst showed a gradual decrease of catalytic
activity and selectivity with time-on-stream. The rates of
decrease in the catalytic activity and selectivity are much faster
than those of Pd/COP. The results strongly support that the
physically adsorbed diphenyl sulfide is not stable under the
reaction condition. Indeed, TGA analysis confirmed that the
physically adsorbed diphenyl sulfide started being leached or
decomposed even at 323 K under H2 (Figure S4), which is
substantially lower than the decomposition temperature of Pd/
COP (503 K). The results indicate that the modifiers covalently
bound to a support framework have a significant advantage in
terms of thermochemical stability, compared with physically
adsorbed modifiers.
From a mechanistic point of view, it is interesting that Pd/
COP does not adsorb H2 according to the aforementioned H2
chemisorption experiment (H/Pd ≪ 0.01). This result
indicates that Pd/COP lacks H−H dissociation ability because
the Pd surface is fully “poisoned” by the sulfur moieties. This is
contradictory to the observed alkyne hydrogenation activities,
because hydrogenation is generally impossible without H2
activation. A plausible explanation for the reaction mechanism
could be obtained from the H2-D2 exchange experiments
summarized in Figure 4. Consistent with the chemisorption
experiment, H2-D2 exchange does not occur on Pd/COP in the
absence of acetylene (Figure 4a). However, in the presence of
coflowing acetylene, substantial formation of HD was detected
along with ethylene (either nondeuterated or deuterated). The
results strongly support that acetylene adsorbs on the Pd
surface and enables “cooperative” adsorption of H2 (or D2). In
other words, H2 dissociative adsorption is possible only in the
presence of acetylene adsorbed on the Pd surface. Notably,
coflowing ethylene with H2/D2 did not facilitate H2-D2
exchange, and moreover, ethane formation was negligible
(Figure 4b). The results indicated that acetylene, which has a
strong interaction with the Pd surface (acetylene binding
energy on Pd (111): −136 kJ mol−1),48 can induce the
decapping of the sulfide groups from the Pd surface, while
ethylene with weaker adsorption strength (−82 kJ mol−1)48
Figure 4. H2−D2 isotope exchange reactions in the presence of
coflowing (a) acetylene and (b) ethylene. Temperature-programmed
desorption of C2H2 and C2H4 for (c) Pd/COP and (d) Pd/γ-Al2O3
catalysts. The desorption profiles were detected by the mass
spectrometer signal at m/z = 15 amu (fragmented C1 species). (e)
Schematic explanation of the H2−D2 exchange mechanism on Pd/
COP.
cannot. We believe that the binding energy between the sulfur-
moieties of COP and Pd is likely to be positioned between
those of alkyne and alkene on Pd surface, and thus, only the
alkynes with stronger binding energy can be adsorbed on the
Pd surface through competitive adsorption. A similar
mechanistic role of molecular modifiers was reported by
Kwon et al.25 Temperature-programmed desorption (TPD)
experiments with acetylene and ethylene also confirmed that
Pd/COP can adsorb acetylene but not ethylene (Figure 4c). In
contrast, Pd/γ-Al2O3 showed strong adsorptions of both
acetylene and ethylene (Figure 4d). These results infer that
the sulfur moieties of the COP ligating the Pd surface act as a
“molecular gate” that selectively enables the adsorption of
acetylene over ethylene (Figure 4e). It is reasonably expected
that the discriminative adsorption behavior would be generally
applied for other alkyne/alkene pairs.
CONCLUSIONS
■
In summary, we demonstrated that Pd catalysts supported on a
thermochemically stable polymer containing diphenyl sulfide-
like moieties (Pd/COP) can be used as a highly chemoselective
and long-lived catalyst for partial hydrogenation of various
alkynes to alkenes. The exceptionally high alkene production
selectivity originated from the distinct adsorption behaviors of
the reactant/intermediate on the Pd catalyst surface in the
presence of “poisoning” sulfide moieties of the polymer matrix.
It is reasonably expected that thermochemically stable polymers
containing diverse poison-like functional groups can be used as
2439
DOI: 10.1021/acscatal.5b02613
ACS Catal. 2016, 6, 2435−2442

ACS Catalysis
Research Article
versatile catalyst supports to achieve distinct chemoselectivities
in various reactions other than the alkyne hydrogenation
demonstrated herein. We believe that an optimum compromise
between the catalytic activity and selectivity can be achieved in
various reactions by rational design of the polymer structure.
Pd/γ-Al2O3 in in situ XAFS cells equipped with a 0.05 mm
thick aluminum window. X-ray photoelectron spectroscopy
(XPS) was carried out using a Sigma Probe (Thermo VG
Scientific) equipped with a microfocused monochromator X-
ray source. The XPS-Pd3d peak was deconvoluted with 2:3
spin−orbit splitting ratios (Δ = 5.26 eV).39 The binding
energies used for the peak deconvolution of XPS-Pd spectra
(for 3d5/2) were 335.3 eV for metallic Pd (Pd0)39 and 337.1 eV
for Pd2+−S species,40 respectively. TGA-MS data were obtained
on a thermogravimetric analyzer (TGA N-1500, Scinco)
coupled with a quadrupole mass analyzer (Thermostar-GSD
320, Pfeiffer Vacuum). The weight change of the sample and
mass spectral data of hydrogen sulfide (m/z = 34 amu) were
monitored during a temperature rise from 323 to 773 K with a
5 K min−1 heating rate under a H2 flow. H2 chemisorption was
performed using an ASAP2020 (Micromeritics) volumetric
analyzer at 343 K to avoid the formation of the β-hydride
phase.49 CO chemisorption was performed at 323 K. Prior to
the measurements, the samples were rereduced at 453 K for
Pd/COP and 573 K for Pd/γ-Al2O3 under a H2 flow. The
amounts of chemisorbed hydrogen were determined by
extrapolation of the linear region (10−30 kPa) of the isotherm
to zero pressure.
Acetylene and ethylene temperature-programmed desorption
(TPD) and H2-D2 isotope exchange on Pd/COP catalyst were
carried out in a PYREX plug-flow reactor connected to an
online mass spectrometer (Thermostar-GSD 320/quadrupole
mass analyzer, Pfeiffer Vacuum). Prior to the analysis, 0.5 g of
catalyst was rereduced at 453 K under H2 flow. For acetylene
and ethylene TPD, rereduced sample was cooled to 303 K,
acetylene or ethylene was adsorbed on the sample at 303 K for
3 h, and weakly adsorbed species were removed by purging
with He for 1 h at the same temperature. The species that were
desorbed from the surface of the catalyst (fragmented C1
species, m/z = 15 amu) were monitored during a temperature
rise from 303 to 623 K with 5 K min−1 heating rate under He
flow. For H2-D2 isotope exchange, rereduced sample was
cooled to 393 K and then H2/D2 (1:1 volume ratio) was flowed
(60 mL min−1 g−1). After 5 min reaction, C2H2 or C2H4 was
flowed simultaneously with a H2/D2 mixture (total flow: 100
mL min−1 g−1; 40 kPa hydrocarbon, 30 kPa H2, and 30 kPa
D2).
EXPERIMENTAL SECTION
■
Catalyst Preparation. A sulfur-containing covalent organic
polymer (COP) was synthesized by nucleophilic substitution of
cyanuric chloride with 4,4′-thiobisbenzenethiol, following the
previously reported procedure after modification.37 In the
synthesis, 5.02 mL of N,N-diisopropylethylamine (DIPEA) was
added to 2.16 g of 4,4′-thiobisbenzenethiol dissolved in 220 mL
of cyclohexanone. A solution of 1 g of cyanuric chloride
dissolved in 40 mL of cyclohexanone was added dropwise into
the aforementioned solution with continuous stirring at 273 K
in a cooling bath. A white precipitate formed almost
immediately. The mixture was further reacted at 303 K for 4
h and at 423 K for 24 h. The solid product was filtered, washed
with 600 mL of cyclohexanone, and dried at 373 K for 12 h. Pd
(1 wt %) was supported on the COP by conventional wet
impregnation of Pd(acac)2 dissolved in tetrahydrofuran (THF).
The impregnated sample was dried at 373 K overnight, calcined
in dry air (200 mL min−1 g−1) at 453 K for 2 h, and reduced in
H2 at 453 K (200 mL min−1 g−1) for 2 h. A 1 wt % Pd/γ-Al2O3
sample was prepared by impregnating an aqueous solution of
PdCl2. PdCl2 (0.1 mmol) dissolved in 0.43 mL of 6.63 M HCl
solution was used for 1 g of γ-Al2O3. The impregnated sample
was dried at 373 K overnight, calcined in dry air at 573 K for 2
h, and reduced in H2 at 573 K for 2 h. A part of the Pd/γ-Al2O3
sample was modified with 10 wt % diphenyl sulfide by wet
impregnation of a hexane solution and drying at 323 K.
Characterization. C, H, N, and S elemental analysis was
carried out by using an elemental analyzer, FLASH2000
(Thermo Scientific). Solid-state cross-polarization magic angle
spinning (CP/MAS) NMR spectra were obtained at a Larmor
frequency, ν0, of 100.53 MHz for 13C using an Agilent DD2
400 NMR spectrometer equipped with a narrow bore 9.4 T
magnet. CP/MAS 13C NMR spectra were recorded with rf-field
strength of 30 kHz and recycling delay of 3 s. Chemical shifts
were reported in ppm relative to adamantine. N2 adsorption−
desorption isotherms were measured using a BEL-Sorp max
(BEL Japan) volumetric analyzer at the liquid N2 temperature
(77 K) after degassing samples at 423 K for 4 h. Bright-field
TEM images were collected by a JEM-2100F (JEOL) at 200
kV. High-angle annular dark-field scanning transmission
electron microscopy (HAADF-STEM) images were collected
by a Cs-corrected STEM microscope (JEM-ARM200F, JEOL)
operating at 200 kV. Particle size distributions were determined
by counting more than 300 crystallites. Surface-area-weighted
mean cluster diameter was determined using TEM analysis and
calculated from dTEM = ∑nidi3/∑nidi2, where ni is the number
of crystallites having a diameter di.49,50 Pd contents of catalysts
were determined by inductively coupled plasma mass
spectrometer (ICP-MS) using ICP-MS7700S (Agilent).
Powder X-ray diffraction (XRD) patterns were recorded
using a D/MAX-2500 (RIGAKU) equipped with Cu radiation
(40 kV, 300 mA).
Catalytic Measurements. All the gas phase reactions were
carried out at atmospheric pressure in a stainless steel plug-flow
reactor (10.9 mm diameter) connected to an online gas
chromatograph. Catalysts were pelletized, crushed, sieved
(150−200 mesh), and then physically mixed with quartz
diluents (150−200 mesh). Prior to the reaction, Pd/γ-Al2O3
was rereduced at 573 K and Pd/COP was rereduced at 453 K
under H2 flow for 2 h. In order to obtain high alkyne
conversion (88−98%), reaction conditions including the space
velocity, reaction temperature, and molar ratio of reactant to
hydrogen were chosen differently for each reactant. The
detailed reaction conditions for each reaction in Table 1 is
summarized as follows: (Reaction 1) 1 kPa 4-octyne, 99 kPa H2
at 393 K; space velocity: 0.2 and 3.0 mol4‑octyne molPd s−1 for
−1
Pd/COP and Pd/γ-Al2O3, respectively; (Reaction 2) 1 kPa 1-
ethynylcyclohexene, 9 kPa H2, 90 kPa He at 373 K; space
velocity: 0.2 and 1.0 mol1‑ethynylcyclohexene molPd s−1 for Pd/
−1
X-ray absorption spectra of the Pd K-edge were measured in
transmission mode at Pohang Accelerator Laboratory (8C-
Nano XAFS beamline). Before the measurements, 0.3 g of a
sample was pressurized into self-standing pellets (13 mm
diameter) and rereduced at 453 K for Pd/COP and 573 K for
COP and Pd/γ-Al2O3, respectively; (Reaction 3) 1 kPa 3-
butyn-2-one, 4 kPa H2, 95 kPa He at 413 K; space velocity: 0.2
and 1.0 mol3‑butyn‑2‑one molPd−1 s−1 for Pd/COP and Pd/γ-Al2O3,
respectively; (Reaction 4) 1 kPa cyclopropylacetylene, 9 kPa
2440
DOI: 10.1021/acscatal.5b02613
ACS Catal. 2016, 6, 2435−2442

ACS Catalysis
Research Article
H2, 90 kPa He at 363 K; space velocity: 0.2 and 1.0
C2H4 selectivity(%)
molcyclopropylacetylene molPd s−1 for Pd/COP and Pd/γ-Al2O3,
−1
⎡
⎤
⎥
(C2H6 − C2H6 ) + 2(C4Hx − C4Hx
)
out
in
out
in
⎢
respectively; (Reaction 5) 1 kPa propargyl alcohol, 99 kPa H2
=
1 −
× 100
−1
⎢
⎣
C2H2 − C2H2
⎥
⎦
at 393 K; space velocity: 0.2 and 1.0 molpropargyl alcohol molPd
in
out
s−1 for Pd/COP and Pd/γ-Al2O3, respectively; (Reaction 6) 1
kPa propargyl amine, 99 kPa H2 at 363 K; space velocity: 0.05
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acscatal.5b02613.
−1
and 1.0 molpropargyl amine molPd s−1 for Pd/COP and Pd/γ-
■
S
Al2O3, respectively. Product distributions of Pd/COP and Pd/
γ-Al2O3 were compared after 30 min in all reactions.
Liquid-phase hydrogenation of phenylacetylene was per-
formed in a 100 mL Teflon-lined stainless steel autoclave
equipped with a magnetic stirrer and a heating jacket. Prior to
the reactions, 0.1 g of catalysts was placed in the reactor and
rereduced at 373 K under H2 flow for 1 h. After cooling to 303
K, 4.39 mL (40 mmol) of phenylacetylene, 35 mL of toluene
solvent, and 0.65 mL (4 mmol) of octane were added into the
reactor. Octane was used as an internal standard. The reaction
mixture was stirred (900 rpm) at 303 K under 1 MPa of H2 for
5 h. After the reaction was completed, the used catalyst was
collected by filtration, washed with toluene, and dried at 373 K
overnight for recycle test. Recyclability of Pd/COP catalyst was
investigated by repetitive reactions up to 5 runs after the
prereduction of catalyst. The products were analyzed with a gas
chromatograph equipped with a flame ionization detector and a
DB-WAX capillary column.
N2 adsorption−desorption isotherm of COP, EXAFS
fitting results of the catalysts, XRD patterns of COP and
Pd/COP, HAADF-STEM images of Pd/COP, and
weight loss of Pd/γ-Al2O3 catalyst modified with
diphenyl sulfide during the temperature ramping under
H2 (PDF)
AUTHOR INFORMATION
Corresponding Author
*E-mail: mkchoi@kaist.ac.kr.
■
Author Contributions
#These authors contributed equally (S.Y. and S.L.).
Notes
The authors declare no competing financial interest.
For selective hydrogenation of acetylene in an ethylene-rich
stream, a reactant mixture gas containing 0.9% H2, 0.6%
acetylene, 0.6% propane, and 49.3% ethylene in N2 balance was
used. Propane was used as an internal standard. Prior to the
reaction, Pd/γ-Al2O3 was rereduced at 573 K. Pd/γ-Al2O3
modified with diphenyl sulfide and Pd/COP were rereduced
at 393 and 453 K, respectively, considering the decomposition
temperature of organic moieties analyzed by TGA-MS. In the
ACKNOWLEDGMENTS
■
This work was supported by the Basic Science Research
Program through the National Research Foundation of Korea
(NRF-2011-0011392). This research was also supported by the
Advanced Biomass R&D Center (ABC) of Korea Grant funded
by the Ministry of Science, ICT and Future Planning (ABC-
2015M3A6A2066121). The authors acknowledge the Pohang
Accelerator Laboratory (PAL) for beamline use.
−1
reaction, space velocities of 0.3, 0.04, and 0.02 molC H molPd
2
2
s−1 were used for Pd/γ-Al2O3, Pd/γ-Al2O3 modified with
diphenyl sulfide, and Pd/COP catalysts, respectively. The
reactant conversion and product selectivities were calculated
using the equations given below:51
REFERENCES
■
́
(1) Molnar
173, 185−221.
́
, A.; Sar
́
kan
́
y, A.; Varga, M. J. Mol. Catal. A: Chem. 2001,
(2) Nikolaev, S. A.; Zanaveskin, L. N.; Smirnov, V. V.; Averyanov, V.
A.; Zanaveskin, K. L. Russ. Chem. Rev. 2009, 78, 231−247.
(3) Lindlar, H. Helv. Chim. Acta 1952, 35, 446−450.
(4) García-Mota, M.; Gom
Fuentes, C.; Bellarosa, L.; Bridier, B.; Per
Theor. Chem. Acc. 2011, 128, 663−673.
(5) Lopez, N.; Vargas-Fuentes, C. Chem. Commun. 2012, 48, 1379−
1391.
C2H2 − C2H2
in
out
C2H2 conversion(%) =
× 100
C2H2
in
́
ez-Díaz, J.; Novell-Leruth, G.; Vargas-
́
ez-Ramírez, J.; Lopez, N.
́
where C2H2 is the acetylene concentration in the feed before
in
the reaction and C2H2 is the acetylene concentration in the
́
out
product stream.
(6) Kyriakou, G.; Boucher, M. B.; Jewell, A. D.; Lewis, E. A.; Lawton,
T. J.; Baber, A. E.; Tierney, H. L.; Flytzani-Stephanopoulos, M.; Sykes,
E. C. H. Science 2012, 335, 1209−1212.
C2H6 − C2H6
out
in
C2H6 selectivity(%) =
× 100
C2H2 − C2H2
(7) Boucher, M. B.; Zugic, B.; Cladaras, G.; Kammert, J.;
Marcinkowski, M. D.; Lawton, T. J.; Sykes, E. C. H.; Flytzani-
Stephanopoulos, M. Phys. Chem. Chem. Phys. 2013, 15, 12187−12196.
(8) McCue, A. J.; McRitchie, C. J.; Shepherd, A. M.; Anderson, J. A. J.
Catal. 2014, 319, 127−135.
in
out
where C2H6 is the ethane concentration in the feed before the
in
reaction and C2H6 is the ethane concentration in the product
out
stream.
(9) Zhang, Q.; Li, J.; Liu, X.; Zhu, Q. Appl. Catal., A 2000, 197, 221−
228.
(10) Sheth, P. A.; Neurock, M.; Smith, C. M. J. Phys. Chem. B 2005,
109, 12449−12466.
2(C4Hx − C4Hx
)
in
out
C4Hx selectivity(%) =
× 100
C2H2 − C2H2
(11) Pei, G. X.; Liu, X. Y.; Wang, A.; Lee, A. F.; Isaacs, M. A.; Li, L.;
Pan, X.; Yang, X.; Wang, X.; Tai, Z.; Wilson, K.; Zhang, T. ACS Catal.
2015, 5, 3717−3725.
in
out
where C4Hx is the C4 hydrocarbon concentration in the feed
in
(12) Sar
117−125.
(13) Sar
157−163.
́
kan
́ ́
y, A.; Horvath, A.; Beck, A. Appl. Catal., A 2002, 229,
before the reaction and C4Hx is the C4 hydrocarbon
out
concentration in the product stream. C4Hx is the sum of the
C4 hydrocarbons.
́
́
kany, A.; Geszti, O.; Saf
́
ran
́
, G. Appl. Catal., A 2008, 350,
2441
DOI: 10.1021/acscatal.5b02613
ACS Catal. 2016, 6, 2435−2442

ACS Catalysis
Research Article
(14) Osswald, J.; Kovnir, K.; Armbruster, M.; Giedigkeit, R.; Jentoft,
(43) Fu, Q.; Wagner, T. Surf. Sci. Rep. 2007, 62, 431−498.
̈
R. E.; Wild, U.; Grin, Y.; Schlogl, R. J. Catal. 2008, 258, 219−227.
(44) Anderson, J. A.; Mellor, J.; Wells, R. P. K. J. Catal. 2009, 261,
208−216.
̈
(15) Armbruster, M.; Behrens, M.; Cinquini, F.; Fottinger, K.; Grin,
̈
̈
́
(45) Bridier, B.; Perez-Ramírez, J. J. Catal. 2011, 284, 165−175.
Y.; Haghofer, A.; Klotzer, B.; Knop-Gericke, A.; Lorenz, H.; Ota, A.;
Penner, S.; Prinz, J.; Rameshan, C.; Rev
Rupprechter, G.; Sautet, P.; Schlogl, R.; Shao, L.; Szentmiklos
̈
(46) Jin, Y.; Datye, A. K.; Rightor, E.; Gulotty, R.; Waterman, W.;
Smith, M.; Holbrook, M.; Maj, J.; Blackson, J. J. Catal. 2001, 203,
292−306.
́
ay, Z.; Rosenthal, D.;
i, L.;
́
̈
Teschner, D.; Torres, D.; Wagner, R.; Widmer, R.; Wowsnick, G.
ChemCatChem 2012, 4, 1048−1063.
(47) Asplund, S. J. Catal. 1996, 158, 267−278.
(48) Mei, D.; Neurock, M.; Smith, C. M. J. Catal. 2009, 268, 181−
195.
(16) Kim, W.-J.; Moon, S. H. Catal. Today 2012, 185, 2−16.
(17) Teschner, D.; Vass, E.; Havecker, M.; Zafeiratos, S.; Schnorch,
̈
̈
(49) Choi, M.; Wu, Z.; Iglesia, E. J. Am. Chem. Soc. 2010, 132, 9129−
9137.
P.; Sauer, H.; Knop-Gericke, A.; Schlogl, R.; Chamam, M.; Wootsch,
A.; Canning, A. S.; Gamman, J. J.; Jackson, S. D.; McGregor, J.;
̈
(50) Schneider, M.; Duff, D. G.; Mallat, T.; Wildberger, M.; Baiker,
A. J. Catal. 1994, 147, 500−514.
Gladden, L. F. J. Catal. 2006, 242, 26−37.
(18) Teschner, D.; Borsodi, J.; Wootsch, A.; Rev
́
ay, Z.; Havecker, M.;
̈
(51) Pachulski, A.; Schodel, R.; Claus, P. Appl. Catal., A 2011, 400,
̈
Knop-Gericke, A.; Jackson, S. D.; Schlogl, R. Science 2008, 320, 86−89.
̈
14−24.
(19) Chan, C. W. A.; Tam, K. Y.; Cookson, J.; Bishop, P.; Tsang, S.
C. Catal. Sci. Technol. 2011, 1, 1584−1592.
(20) Chan, C. W. A.; Mahadi, A. H.; Li, M. M. -J.; Corbos, E. C.;
Tang, C.; Jones, G.; Kuo, W. C. H.; Cookson, J.; Brown, C. M.;
Bishop, P. T.; Tsang, S. C. E. Nat. Commun. 2014, 5, 5787.
(21) Vile,
Antonietti, M.; Lop
2015, 54, 11265−11269.
(22) Lopez, N.; Bridier, B.; Per
112, 9346−9350.
(23) García-Mota, M.; Bridier, B.; Per
Catal. 2010, 273, 92−102.
́
G.; Albani, D.; Nachtegaal, M.; Chen, Z.; Dontsova, D.;
́
ez, N.; Perez-Ramírez, J. Angew. Chem., Int. Ed.
́
́
́
ez-Ramírez, J. J. Phys. Chem. C 2008,
́
́
ez-Ramírez, J.; Lopez, N. J.
(24) Crespo-Quesada, M.; Dykeman, R. R.; Laurenczy, G.; Dyson, P.
J.; Kiwi-Minsker, L. J. Catal. 2011, 279, 66−74.
(25) Kwon, S. G.; Krylova, G.; Sumer, A.; Schwartz, M. M.; Bunel, E.
E.; Marshall, C. L.; Chattopadhyay, S.; Lee, B.; Jellinek, J.; Shevchenko,
E. V. Nano Lett. 2012, 12, 5382−5388.
(26) Long, W.; Brunelli, N. A.; Didas, S. A.; Ping, E. W.; Jones, C. W.
ACS Catal. 2013, 3, 1700−1708.
(27) McKenna, F.-M.; Anderson, J. A. J. Catal. 2011, 281, 231−240.
(28) McCue, A. J.; Anderson, J. A. Catal. Sci. Technol. 2014, 4, 272−
294.
́ ́ ́
(29) Vile, G.; Almora-Barrios, N.; Mitchell, S.; Lopez, N.; Perez-
Ramírez, J. Chem. - Eur. J. 2014, 20, 5926−5937.
(30) McCue, A. J.; McKenna, F.-M.; Anderson, J. A. Catal. Sci.
Technol. 2015, 5, 2449−2459.
(31) Trimm, D. L.; Liu, I. O. Y.; Cant, N. W. Appl. Catal., A 2010,
374, 58−64.
(32) Mori, A.; Mizusaki, T.; Miyakawa, Y.; Ohashi, E.; Haga, T.;
Maegawa, T.; Monguchi, Y.; Sajiki, H. Tetrahedron 2006, 62, 11925−
11932.
(33) Makosch, M.; Lin, W.-I.; Bumbal
́
ek, V.; Sa,
́
J.; Medlin, J. W.;
Hungerbuhler, K.; van Bokhoven, J. A. ACS Catal. 2012, 2, 2079−
̈
2081.
(34) Xing, R.; Liu, Y.; Wu, H.; Li, X.; He, M.; Wu, P. Chem. Commun.
2008, 6297−6299.
(35) Fujiwara, S.; Takanashi, N.; Nishiyabu, R.; Kubo, Y. Green Chem.
2014, 16, 3230−3236.
(36) Sun, Q.; Dai, Z.; Meng, X.; Wang, L.; Xiao, F.-S. ACS Catal.
2015, 5, 4556−4567.
(37) Patel, H. A.; Karadas, F.; Byun, J.; Park, J.; Deniz, E.; Canlier, A.;
Jung, Y.; Atilhan, M.; Yavuz, C. T. Adv. Funct. Mater. 2013, 23, 2270−
2276.
(38) Mojet, B. L.; Hoogenraad, M. S.; van Dillen, A. J.; Geus, J. W.;
Koningsberger, D. C. J. Chem. Soc., Faraday Trans. 1997, 93, 4371−
4375.
(39) Moulder, J. F., Stickle, W. F., Sobol, P. E., Bomben, K. D. In
Handbook of X-ray photoelectron spectroscopy; Chastain, J., Ed.; Physical
Electronics: Eden Prairie, MN, 1992; pp 118−119.
(40) Hyland, M. M.; Bancroft, G. M. Geochim. Cosmochim. Acta
1990, 54, 117−130.
(41) Tauster, S. J.; Fung, S. C.; Garten, R. L. J. Am. Chem. Soc. 1978,
100, 170−175.
(42) Tauster, S. J. Acc. Chem. Res. 1987, 20, 389−394.
2442
DOI: 10.1021/acscatal.5b02613
ACS Catal. 2016, 6, 2435−2442