Bidentate N-based Ligands for Highly Reusable, Ligand-coordinated, Supported Pt Hydrosilylation Catalysts

Article abstract of DOI:10.1002/cctc.202000085

A significant challenge in designing supported metal-ligand catalysts for solution-phase reactions is the stabilization of the metal active sites against leaching into solution. Here, we examine alkene hydrosilylation reactions as model systems to improve the stability of highly dispersed Pt using a metal-ligand coordination strategy on high surface area oxide supports. By evaluating a series of bidentate N-based ligands, we demonstrate several design strategies to improve stability of the highly dispersed Pt²⁺ centers against leaching, while maintaining a high level of catalytic activity, selectivity, and recyclability for alkene hydrosilylation batch reactions under mild conditions. These involve a bi-functional approach to ligand design, which considers interaction to the support and a well-defined coordination environment for the metal active site. Three strategies are reported: modifying ligands for stronger interaction with oxide surfaces, mixing ligands, and pre-depositing an “anchoring ligand” to the support before loading the metal-ligand catalyst. Each of these is successful in enhancing Pt recyclability. Particularly, two Pt-phenanthroline catalysts exhibit excellent reusability for multiple batch reaction cycles, due to high stability of the active Pt species. Addressing the active site leaching problem significantly enhances the utility of ligand-coordinated supported metal catalysts as highly stable and selective catalysts for solution-phase reactions.

Full text of DOI:10.1002/cctc.202000085

The European Society Journal for Catalysis
Supported by
Accepted Article
Title: Bidentate N-based Ligands for Highly Reusable, Ligand-
coordinated, Supported Pt Hydrosilylation Catalysts
Authors: Linxiao Chen, Iyad S. Ali, and Steven L Tait
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemCatChem 10.1002/cctc.202000085
Link to VoR: https://doi.org/10.1002/cctc.202000085
01/2020

10.1002/cctc.202000085
ChemCatChem
ChemCatChem
Bidentate N-based Ligands for Highly Reusable, Ligand-coordinated, Supported Pt
Hydrosilylation Catalysts
Dr. Linxiao Chen,† Mr. Iyad S. Ali, and Prof. Steven L. Tait*
Dept. of Chemistry, Indiana University Bloomington, Bloomington, Indiana 47401, USA
* E-mail: tait@indiana.edu, Tel: +1-812-855-1302, Twitter: @StevenLTait
† Present address: Institute for Integrated Catalysis, Pacific Northwest National Laboratory,
Richland, Washington 99352, USA
This article is protected by copyright. All rights reserved.

10.1002/cctc.202000085
ChemCatChem
Abstract
A significant challenge in designing supported metal-ligand catalysts for solution-phase
reactions is the stabilization of the metal active sites against leaching into solution. Here, we
examine alkene hydrosilylation reactions as model systems to improve the stability of highly
dispersed Pt using a metal-ligand coordination strategy on high surface area oxide supports. By
evaluating a series of bidentate N-based ligands, we demonstrate several design strategies to
improve stability of the highly dispersed Pt2+ centers against leaching, while maintaining a high
level of catalytic activity, selectivity, and recyclability for alkene hydrosilylation batch reactions
under mild conditions. These involve a bi-functional approach to ligand design, which considers
interaction to the support and a well-defined coordination environment for the metal active site.
Three strategies are reported: modifying ligands for stronger interaction with oxide surfaces,
mixing ligands, and pre-depositing an “anchoring ligand” to the support before loading the
metal-ligand catalyst. Each of these is successful in enhancing Pt recyclability. Particularly, two
Pt-phenanthroline catalysts exhibit excellent reusability for multiple batch reaction cycles, due to
high stability of the active Pt species. Addressing the active site leaching problem significantly
enhances the utility of ligand-coordinated supported metal catalysts as highly stable and selective
catalysts for solution-phase reactions.
2
This article is protected by copyright. All rights reserved.

10.1002/cctc.202000085
ChemCatChem
Introduction
Hydrosilylation, the addition of a Si−H bond to a multiple bond (particularly C=C, Scheme
1), has been of significant importance in silicon chemistry since its first report in 1947.[1] It
serves as a versatile tool to produce various functionalized silicon-based polymers,[2] which have
observed wide applications as lubricant oils[3] and as coating,[4] preceramic,[5] adhesive,[6] and
sealing materials.[7] Hydrosilylation also offers solutions to silicone curing[4a, 8] and Si−C bond
building in fine chemical synthesis.[9] Industrial hydrosilylation catalysts have long been
dominated by homogeneous Pt complexes due to their superior activity over other metals.[10]
Speier catalyst, H2PtCl6/iPr−OH,[11] prevailed for decades before being replaced by the more
active and selective Karstedt catalyst, a vinyl-siloxane Pt0 complex developed in 1970s.[12] One
significant drawback of Karstedt catalyst is the formation of colloidal Pt, which deactivates the
catalyst and reduces product quality.[4a, 13] The problem is further underscored by the high cost,
volatile market, and uncertain future supply of Pt. Besides the Pt aggregation problem, side
reactions of C=C isomerization and hydrogenation (Scheme 1) are difficult to avoid, and some
important functional groups, such as epoxy,[14] are unstable under reaction conditions.
Consequently, there is ongoing research effort to develop Pt hydrosilylation catalysts with
improved selectivity and stability. Examples of recent progress include Pt-carbene complexes,[14-
15] trinuclear alkyne Pt0 complexes,[16] and anti-sulfur-poisoning Pt catalysts.[17] Heterogeneous
Pt catalysts, such as Pt nanoparticles,[13a-c] PtO2,[18] and supported Pt single-atoms,[19] have also
inspired interest because they can be easily separated from the reaction mixture and recycled.
3
This article is protected by copyright. All rights reserved.

10.1002/cctc.202000085
ChemCatChem
Scheme 1. General reaction scheme of hydrosilylation and common byproducts from alkene.
The quest has long existed for next-generation catalysts that combine the easy recovery of
heterogeneous catalysts with the high metal utilization efficiency and selectivity of homogeneous
catalysts. This has sparked growing interest in immobilized organometallic catalysts[20] and
single-atom catalysts (SACs).[21] Both groups of catalysts fulfill the purpose of isolating single
metal atoms on solid supports. The former offers higher metal loading and tunability, while the
latter provides direct metal-support interaction, which is often crucial in catalysis. To combine
desirable characters from both, we have developed a metal-ligand self-assembly strategy, first on
single crystal surfaces in UHV[22] and then adapted to various high-surface-area oxide supports
under ambient conditions[23] to create high-loading of ligand-stabilized and tunable noble metal
sites with direct contact with the support. We demonstrated that Pt stabilized by 3,6-di-2-pyridyl-
1,2,4,5-tetrazine (DPTZ, Figure 1a) on CeO2 and MgO powders are atomically dispersed as
Pt2+,[23a] and are much more active than supported Pt aggregates and existing Pt SACs for alkene
hydrosilylation.[19a, 19b, 23b] Compared with the Karstedt catalyst, they improve selectivity, reduce
Pt aggregation, and suppress the decomposition of epoxy groups.[23] Despite desired catalytic
properties, Pt active sites on these catalysts partially leach into the solution during catalysis,
limiting their recyclability and, hence, practical applications.
4
This article is protected by copyright. All rights reserved.

10.1002/cctc.202000085
ChemCatChem
Figure 1. Structures of bidentate N-based ligands used in this work: (a) 3,6-(2-pyridyl)-1,2,4,5-
tetrazine (DPTZ), (b) 1,10-phenanthroline-5,6-dione (PDO), (c) bathophenanthroline (BPhen),
(d) 4,4’-dinonyl-2,2′-dipyridyl (C9BP), and (e) 2,2′-bipyridine-4,4′-dicarboxylic acid (4,4’-
BPDCA).
Therefore, in this work, we introduce a series of bifunctional bidentate N-based ligands
(Figure 1b-e), aiming to create supported Pt-ligand hydrosilylation catalysts with improved
active site recyclability. Three approaches are discussed: replacing DPTZ with a different ligand,
mixing DPTZ with another ligand, and pre-depositing an additional ligand as the “anchoring
ligand.” The resulting Pt catalysts are evaluated with a model hydrosilylation reaction between 1-
octene and dimethoxymethyl silane. We demonstrate that all three approaches are promising to
mitigate the active site leaching problem, and present two catalysts with excellent reusability
through multiple reaction cycles. Post-reaction characterization sheds light on the nature of
active sites on these two catalysts. This work not only significantly improves the application
potential of supported Pt-ligand hydrosilylation catalysts, it also expands the tool box for the
synthesis of highly-dispersed noble metal catalysts on oxide supports.
5
This article is protected by copyright. All rights reserved.

10.1002/cctc.202000085
ChemCatChem
Experimental
Synthesis of supported Pt-ligand catalysts. The synthesis follows a metal-ligand self-
assembly strategy reported in our previous publications.[23] For Pt-DPTZ/CeO2: 0.0108g (0.046
mmol) DPTZ (Sigma Aldrich, 96%) were completely dissolved in 25 mL 1-butanol (Alfa Aesar,
99%) by stirring for 20 min at room temperature. 0.3 g CeO2 (BET surface area: 4.8 m2/g) were
added to the pink DPTZ solution and the mixture was then stirred for 2 h at room temperature.
0.0080 g H2PtCl6 · 6H2O (Alfa Aesar, 99.95% metal basis, 0.015 mmol, 1 wt % by Pt with
respect to total catalyst mass; 3 eq. DPTZ with Pt) were dissolved in 5 mL 1-butanol. The Pt salt
solution was then added to the CeO2/DPTZ/1-butanol mixture dropwise under stirring within 30
min. The mixture was covered and stirred for 24 h, then dried at room temperature under dry air
flow overnight. The dried catalyst was washed with water, then dichloromethane (DCM) until
powders did not show any pink color (all free DPTZ removed).
For Pt-PDO/CeO2, Pt-BPhen/CeO2, Pt-C9BP/CeO2, and Pt-4,4’-BPDCA/CeO2: the same
procedure was followed, except that DPTZ was replaced by PDO (Sigma Aldrich, 98%), BPhen
(Sigma Aldrich, 97%), C9BP (Alfa Aesar, 97%), or 4,4’-BPDCA (Alfa Aesar, 98%) and H2O
was used instead of 1-butanol as the solvent in the case of PDO and 4,4’-BPDCA due to ligand
solubility.
For Pt-BPhen+DPTZ/CeO2 catalysts: DPTZ was replaced by a BPhen+DPTZ mixture of the
same total mole quantity and various ratios.
For the catalysts with anchoring ligands, 0.07 mmol anchoring ligand was impregnated onto
0.5 g CeO2 or MgO powder (BET surface area: ~5 m2/g) in a solvent (H2O for PDO and DCM
for 4,4’-BPDCA),[24] then that ligand-modified support was used instead of pristine support in
6
This article is protected by copyright. All rights reserved.

10.1002/cctc.202000085
ChemCatChem
the procedure described above. All supported Pt-ligand catalysts were yellow or light-yellow
powders.
General procedures for alkene hydrosilylation reactions. For the 70 °C, 30 min reactions:
30 mg supported Pt catalysts were weighed and kept in an empty reaction tube with cap. 5 mmol
dimethoxymethylsilane (Alfa Aesar, > 97%) and 6 mmol 1-octene (Alfa Aesar, > 97%) were
weighed into another reaction tube, and 3 mL toluene (Macron, ACS grade) was added to the
same tube. For the 60 °C, 20 min reactions, the amount of all chemicals was reduced to half.
Both tubes were pre-heated in a water bath at the reaction temperature for 10 min before
reactants and solvent were added into the tube with Pt catalysts. The tube was capped during the
reaction to avoid evaporation of silane (low boiling point). After the reaction, the tube was
cooled down quickly with cold water flow, the solid catalysts were centrifuged out for reuse or
post-reaction characterization, and the liquid mixture was diluted to 25 mL for GC-MS
measurements with an Agilent 6890N Gas Chromatograph and 5973 Inert Mass Selective
Detector. Product yield was calculated from its response intensity at m/z = 203.2 with respect to
the response of internal standard decane (Sigma Aldrich, > 99%, ~ 0.15 g added to all standard
and post-reaction solutions) at m/z = 142.2 using calibration curves pre-made with standard
solutions. The product used to make standard solutions was purchased from Sigma Aldrich
(≥95.0%). The calibration curve exhibits an almost-perfect linear relationship between
normalized GC-MS response and concentration over the concentration range relevant to the
experiments reported here. For all samples we measured, the GC-MS response from decane is
within 20% of standard solutions used in the calibration curves. Epoxy-containing alkene
substrate hydrosilylation reactions occurred under 80 °C, 100 min conditions with 50 mg
supported Pt catalyst weighed and kept in an empty reaction tube. 2.5 mmol trimethoxysilane
7
This article is protected by copyright. All rights reserved.

10.1002/cctc.202000085
ChemCatChem
(Sigma Aldrich, 95%) and 3 mmol 4-vinyl-1-cyclohexene 1,2-epoxide isomers (Sigma Aldrich,
98%) were weighed into another reaction tube, and 1.5 mL toluene (Macron, ACS grade) was
added to the same tube.
Characterization of supported Pt-ligand catalysts. X-ray photoemission spectroscopy
(XPS) measurements were performed with a PHI Versaprobe II XP spectrometer using a
monochromated Al X-ray source. A small amount of each powder sample was fixed onto a
platen with double-sided tape. For CeO2-supported samples, XP spectra were collected for Pt 4f,
N 1s, C 1s, Cl 2p, Ce 3d, and O 1s. For MgO-supported samples, the Ce 3d region was replaced
by the Mg 2p region. An Ar+ neutralizer was used to alleviate surface charging. The binding
energy was corrected by calibration with the adventitious C 1s peak (284.8 eV). Inductively
coupled plasma mass spectrometry (ICP-MS) measurements were performed with an Agilent
7700 quadrupole ICP-MS instrument. Solid catalysts were treated with aqua regia to dissolve all
Pt before measurement. For solution samples, the solvent was evaporated first and then the
residue was treated with aqua regia.
Results and Discussion
1. Supported Pt-ligand catalysts with single bidentate N-based ligand
Previously we have reported oxide-supported Pt-DPTZ catalysts. Extensive characterization
showed that Pt exist predominantly as isolated atoms stabilized by favorable Pt-DPTZ
coordination, with direct contact with oxide supports. Therefore, these catalysts have been
categorized as single-atom catalysts (SACs).[23] The SACs exhibit improvements over the
commercial hydrosilylation catalyst (Karstedt catalyst) in selectivity, stability, and substrate
8
This article is protected by copyright. All rights reserved.

10.1002/cctc.202000085
ChemCatChem
scope. Nevertheless, their application is challenged by low reusability due to active site leaching
into solution during reaction. We have established that Pt-DPTZ coordination through the
bidentate N binding pockets (Figure 1a) is highly stable, but the binding between Pt-DPTZ
complexes and the oxide support is not.[23b] Consequently, we first replaced DPTZ with one of
the following bifunctional bidentate N-based ligands: 1,10-phenanthroline-5,6-dione (PDO),
bathophenanthroline (BPhen), 4,4’-dinonyl-2,2′-dipyridyl (C9BP), or 2,2′-bipyridine-4,4′-
dicarboxylic acid (4,4’-BPDCA) (structures shown as Figure 1b-1e). Besides a bidentate N
pocket, each ligand has additional functional groups for improved interaction with the oxide
supports. Our hypothesis is that the replacement enhances the interaction between the Pt-ligand
complex and the support to create active hydrosilylation catalysts with improved active site
recyclability.
We synthesized supported Pt-ligand catalysts using the one-step metal-ligand impregnation
method reported previously.[23a] In short, H2PtCl6 ∙ 6H2O solution was added dropwise into pre-
mixed ligand solution and CeO2, followed by stirring, solvent evaporation, and rinsing (see
experimental section for details). We chose CeO2 as the support because, in previous studies, we
found higher catalytic activity and reusability with CeO2 than with other oxide supports.[23]
Synthesis solvent was varied to ensure simultaneous dissolution of H2PtCl6 ∙ 6H2O and the
ligand (Table 1). 1 wt% Pt was added and the actual Pt loading, determined by inductively
coupled plasma mass spectrometry (ICP-MS), was found to be in the range of 0.16 – 0.41 wt%
(Table 1). On all fresh catalysts, the binding energy (BE) of the Pt 4f7/2 XPS peak is consistent
with Pt2+ (~72.8 eV, Table 1 and Figure 2). Figure 2 also shows that Pt 4f XP spectra on most
catalysts can be described well with a single Pt2+ component of similar FWHM (~2.1 eV), except
for Pt-BPhen/CeO2, on which the wide, asymmetric peak requires an additional Pt4+ component
9
This article is protected by copyright. All rights reserved.

10.1002/cctc.202000085
ChemCatChem
for appropriate fitting (see SI for fitting details). Therefore, most Pt exist as highly dispersed Pt2+
cations and metallic Pt nanoparticles are not present. This is expected because the bidentate N
pockets are suitable for Pt coordination and have sufficient oxidizing potential to stabilize
dispersed single-atom Pt2+ on CeO2 and on MgO, as we have demonstrated previously.[23] On all
catalysts, N (from ligands) and Cl (from H2PtCl6 ∙ 6H2O) are clearly identified by XPS (Figure
S1, Table S2). N:Pt and Cl:Pt ratios are calculated from XPS peak areas and summarized in
Table 1. The sum of the two ratios is at least 3, indicating Pt are highly coordinated with N and
Cl. Besides, Pt need to coordinate with surface O to stay on CeO2. The high coordination is
consistent with the argument that Pt aggregates are not formed to a significant degree. In
summary, XPS indicates that on these catalysts, Pt bind with ligands, Cl, and O from CeO2,
forming dispersed Pt2+ sites. We note that these XPS results are consistent with our previous
studies of Pt-DPTZ SACs,[23a] for which EXAFS and STEM analysis conclusively proved a
single-atom character of the metal sites, which implies that atomic dispersion is likely on the
other catalysts reported here. However, due to a lack of comprehensive characterization, we refer
to these as ligand-coordinated supported catalysts (LCSCs), rather than SACs.
10
This article is protected by copyright. All rights reserved.

10.1002/cctc.202000085
ChemCatChem
Figure 2. XP spectra of fresh (before reaction) CeO2-supported Pt-ligand LCSCs in the Pt 4f
region. All spectra are normalized so that peaks have similar intensity for better peak shape
comparison. All spectra except for Pt-BPhen/CeO2 show a single Pt2+ components (~72.8 eV).
For Pt-BPhen/CeO2, a small fraction of Pt4+ is also present.
11
This article is protected by copyright. All rights reserved.

10.1002/cctc.202000085
ChemCatChem
Table 1. Synthesis and characterization of fresh CeO2-supported Pt-ligand catalysts for a series
of ligands. Pt loading is calculated as wt % from ICP-MS. From XPS analysis, we measure Pt
4f7/2 binding energy (BE) and molar ratios of N:Pt and Cl:Pt, calculated from peak area ratios and
corrected for relative sensitivity factors.
ICP Pt loading
(wt %)
Pt 4f7/2 BE
Ligand
DPTZ[a]
N:Pt
2.8
2.1
2.5
1.4
0.8
4.6
Cl:Pt
1.0
1.3
1.5
2.8
2.2
1.1
Synthesis solvent
1-butanol
water
(eV)
0.35
0.47
0.53
0.29
0.16
0.41
72.8
72.8
72.7
72.8
72.6
72.9
PDO
Bphen + DPTZ[b]
BPhen
1-butanol
1-butanol
1-butanol
water[c]
C9BP
4,4’-BPDCA
[a] Data on Pt-DPTZ/CeO2 were previously published in reference 23b.
[b] The catalyst was synthesized using a mixture of BPhen and DPTZ (1:2 molar ratio).
[c] 4,4’-BPDCA was heated with water (to approximately 60 °C) to increase its solubility. The heat was turned
off before adding Pt solution.
We tested these Pt LCSCs for a model hydrosilylation reaction between 1-octene (1) and
dimethoxymethylsilane (2) (Scheme 2), under two conditions: 70 °C for 30 min and 60 °C for 20
min. The former is the “standard condition” to evaluate active site recyclability. The latter is for
TON calculation and activity comparison, as it keeps the conversion of 2 far below 100%.
Toluene was selected as the solvent for the reaction (Table S3). All catalysts, except Pt-4,4’-
BPDCA/CeO2, exhibit catalytic activity under both conditions, yielding the anti-Markovnikov
addition product 3. Similar with most hydrosilylation catalysts, alkene isomerization and
hydrogenation are unavoidable. Therefore, we performed all reactions with 1.2 eq. 1. The most
concentrated by-products are silane oligomers 4 and 5, the concentration of which is very low
12
This article is protected by copyright. All rights reserved.

10.1002/cctc.202000085
ChemCatChem
compared with 3 (see Figure S2 for a representative GC-MS spectrum). In addition, no
Markovnikov addition product is observed, suggesting our catalysts are highly selective. The
conversion of 2 is difficult to quantify precisely due to its volatility, so we instead used the yield
of 3 to quantify catalytic activity.
Activity of all fresh catalysts at 70 °C, TON at 60 °C in the first 20 min, as well as total Pt
recovery percentage (calculated from XPS Pt:Ce ratio, see SI for justification) after 1 reaction
cycle are reported in Table 2. Both activity and Pt recyclability vary significantly with the choice
of ligand, highlighting one unique advantage of the LCSCs: metal sites can be tuned by changing
the ligand, as in organometallic complexes in homogeneous catalysis.
Scheme 2. Hydrosilylation reaction between 1-octene (1) and dimethoxymethylsilane (2) used to
test Pt LCSCs, as well as structures of the two most important Si-containing by-products.
13
This article is protected by copyright. All rights reserved.

10.1002/cctc.202000085
ChemCatChem
Table 2. Activity and total Pt recovery (by XPS) of Pt LCSCs.
XPS Pt recovery[b]
Yield[a]
(%)
TON per Pt[c]
(%)
Ligand
(103)
70 °C, 30 min
60 °C, 20 min
1 cycle
62
3 cycles
DPTZ
PDO
90[d]
66
83
74
23
0
−
69
23
−
6.2
2.7
2.2
0.1
10.2
0
97
Bphen + DPTZ
Bphen
86
66
C9BP
27
−
4,4’-BPDCA
90
−
[a] Reaction condition: T = 70 °C, t = 30 min, 30 mg catalyst, 6 mmol 1, 5 mmol 2, and 3 mL toluene.
[b] XPS Pt recovery was calculated based on Pt:Ce ratio compared with the fresh catalyst. Pt 4f XP spectra of
post-reaction Pt-PDO/CeO2 and Pt-BPhen+DPTZ/CeO2 are included as a part of Figure 4, while others are
presented as Figure S3.
[c] Reaction condition: T = 60 °C, t = 20 min, 15 mg catalyst, 3 mmol 1, 2.5 mmol 2, and 1.5 mL toluene.
[d] Complete conversion of 2 was achieved in this reaction.
We examined the reusability of the catalyst with the highest Pt recovery percentage after 1
reaction cycle, Pt-PDO/CeO2, by recycling it for 4 cycles, and results are shown in Figure 3. For
all reactions in Figure 3, the conversion of 2 is below 100%, and selectivity remains similar with
the first cycle. Pt-PDO/CeO2 shows excellent reusability, as its activity does not drop through the
4 cycles (~60% yield). In comparison, we have shown previously that product yield decreases
quickly on Pt-DPTZ/CeO2 once the conversion of 2 drops below 100% (Figure S4), due to active
site leaching.[23b] Figure 3 strongly indicates that almost all active sites on Pt-PDO/CeO2 can be
recycled effectively, a drastic improvement over Pt-DPTZ/CeO2. Pt recovery percentage after the
first cycle is also much higher on Pt-PDO/CeO2 than Pt-DPTZ/CeO2 (97% compared with 62%,
Table 2). Minor Pt leaching does not deactivate Pt-PDO/CeO2, possibly because only relatively
14
This article is protected by copyright. All rights reserved.

10.1002/cctc.202000085
ChemCatChem
inactive Pt sites are leached (see next paragraph). Besides Pt-PDO/CeO2, Pt-4,4’-BPDCA/CeO2
exhibits enhanced Pt recyclability (90% after the first cycle) too, but it is not catalytically active.
Pt-Bphen/CeO2 and Pt-C9BP/CeO2 exhibit underwhelming Pt recyclability (only 66% and 27%),
possibly due to the low ligand loading (low N:Pt in Table 1) leading to relatively
undercoordinated Pt. We recognize that Pt-PDO/CeO2 is not as active as Pt-DPTZ/CeO2 (lower
yield and TON per Pt under identical reaction conditions), but the difference is within 3-fold. In
practice, one can compensate for lower activity by using more catalysts, but active site leaching
is more challenging, especially for expensive and environmental unfriendly noble metals.
Therefore, from a practical perspective, significantly enhanced reusability is more valuable. For
the hydrosilylation of epoxy-containing alkene shown in Scheme S1, Pt-PDO/CeO2 exhibits
similar selectivity (66%) with Pt-DPTZ/CeO2 (71%) at 100% silane conversion (80 °C, 100
min),[23b] demonstrating desired stronger tolerance towards unstable groups than Karstedt
catalyst (~ 50% selectivity).[14a, 23b]
Figure 3. Reusability test results of Pt-PDO/CeO2 (4 cycles) and Pt-BPhen+DPTZ/CeO2 (3
cycles). Reaction condition: T = 70 °C, t = 30 min, 30 mg catalyst, 6 mmol 1, 5 mmol 2, and 3
mL toluene. After each reaction, the solid catalyst was centrifuged out and recycled. Both
catalysts can be reused multiple times without significant loss of activity.
15
This article is protected by copyright. All rights reserved.

10.1002/cctc.202000085
ChemCatChem
The highly reusable Pt-PDO/CeO2 was characterized by XPS after the reaction to investigate
its active sites. We found that most but not all Pt sites are recyclable, with 97% Pt recovered after
1 cycle and 69% recovered after 3 cycles (Table 2). ICP on post-reaction solution from the first
cycle shows Pt equivalent with < 2% total Pt leaching, consistent with XPS. After the first cycle,
Pt 4f XPS peak widens without significant shift in the BE, indicating broader oxidation state
distribution centered around +2 (Figure 4a). Fitting (Figure 4a, Table 3) suggests two Pt
components are formed: Pt(2−δ)+ (~72.2 eV, red) and Pt(2+δ)+ (~73.7 eV, green). This implies that
at least two different Pt species exist on fresh Pt-PDO/CeO2. Although both appear to be Pt2+
before catalysis, they evolve differently during catalysis, and hence the post-reaction oxidation
state varies. Figure 4a and Table 3 also exhibit that after 3 cycles, the Pt(2+δ)+ component does not
change much, while the Pt(2-δ)+ component drops, indicating the Pt leaching (~30% of total Pt) is
mainly from the Pt(2-δ)+ species. However, the partial loss of Pt(2-δ)+ has negligible impacts on the
activity (Figure 3). The lack of correlation between activity and the number of Pt(2-δ)+ sites
implies that Pt(2+δ)+ sites are likely much more active than Pt(2-δ)+. We have hence demonstrated
that although Pt leaching still occurs on Pt-PDO/CeO2, the main active sites are effectively
recycled. We note that no Pt0 are detected after catalysis, implying Pt remain highly dispersed.
XPS also reveals that N:Pt ratio remains constant through 3 cycles (Table 3, Figure S5), and thus
the Pt-PDO coordination is highly stable. In contrast, Cl:Pt ratio drops. The leaving of Cl has
been previously reported on Pt-DPTZ SACs, which is associated with the catalyst activation.[23b]
Cl might play a similar role on Pt-PDO/CeO2: it acts as a leaving group to open more
coordination vacancies and activate Pt. In previous publications, we have established that with
Pt-DPTZ ligand-stabilized SACs, the active Pt species may be generated during the reaction by
temporary leaving the support, and then partially recovered by recombination with the support
16
This article is protected by copyright. All rights reserved.

10.1002/cctc.202000085
ChemCatChem
after the reaction.[23b] This could also be a plausible mechanism for Pt-PDO/CeO2. Therefore, the
good reusability does not necessarily prove that active Pt sites stay on CeO2 throughout the
reaction: it might result from the complete recovery of active Pt when the reaction is complete.
Figure 4. Fittings of Pt 4f XP spectra of (a) Pt-PDO/CeO2 and (b) Pt-BPhen+DPTZ/CeO2,
including their fresh forms (top), after 1 reaction cycle (mid), and after 3 cycles (bottom). All
spectra are normalized based on Ce 3d peak area so that the peak area reflects Pt concentration
(Pt:Ce ratio) on the catalyst. See SI for discussion of fit components.
17
This article is protected by copyright. All rights reserved.

10.1002/cctc.202000085
ChemCatChem
Table 3. Changes in relative Pt concentrations of various oxidation states, N:Pt ratio, and Cl:Pt
ratio on Pt-PDO/CeO2 and Pt-BPhen+DPTZ/CeO2.
Relative concentration[a]
Ligand
PDO
Catalyst status
N:Pt
Cl:Pt
Pt(2–δ)+
Pt2+
1
Pt(2+δ)+ Total Pt
Fresh
0
0.54
0.31
0
0
0.43
0.38
0
1
2.1
2
1.3
0.9
1
post-1rxn
post-3rxn
Fresh
0
0.97
0.69
1
0
2.2
2.5
2.7
3
1
1.5
1.1
0.8
Bphen+DPTZ
post-1rxn
post-3rxn
0
0.55
0.23
0.31
0
0.86
0.23
0
[a] Relative concentration of a Pt component on a sample was calculated based on the peak area from fittings
(normalized to Ce 3d area of the same sample). For each catalyst, the total normalized Pt 4f peak area of its
fresh form was defined as 1.
2. Enhancing recyclability of Pt-DPTZ/CeO2 by combination with another ligand
We also attempted to improve active site recyclability of Pt-DPTZ SACs by mixing DPTZ
with another ligand. The mixing ligand must be soluble in 1-butanol, the only solvent we
discovered that dissolves DPTZ and H2PtCl6 · 6H2O simultaneously. This excludes PDO and
4,4’-BPDCA. BPhen was chosen over C9BP for this study because when used alone, BPhen
stabilizes Pt better than C9BP according to Table 2. This is potentially due to the benzene rings
on BPhen offering stronger van der Waals interaction with CeO2 than alkyl chains on C9BP.
Besides, the phenanthroline ring on BPhen is more rigid than the bi-pyridyl ring on C9BP,
possibly providing a more stable pocket for Pt binding. The synthesis procedure was adapted
from the original Pt-DPTZ/CeO2 recipe,[23a] with a fraction of DPTZ replaced by BPhen. We
tested two BPhen:DPTZ molar ratios: 2:1 and 1:2. The former does not create active
18
This article is protected by copyright. All rights reserved.

10.1002/cctc.202000085
ChemCatChem
hydrosilylation catalyst, so here we focus on the latter, referred to herein as Pt-
BPhen+DPTZ/CeO2.
Pt-Bphen+DPTZ/CeO2 exhibits significantly improved reusability over Pt-DPTZ/CeO2, as it
only shows minimal activity drop in the first three cycles (from 81% yield to 73%, Figure 3). Pt
recovery percentage after the first cycle is improved as well (86% compared with 62%, Table 2).
The reusability of Pt-BPhen+DPTZ/CeO2 is close to the most promising catalyst from the single-
ligand strategy, Pt-PDO/CeO2. The two also exhibit similar activity at 60 °C (Table 2, TON =
2.2 and 2.7 × 103 respectively in the first 20 min), and selectivity with epoxy-containing alkene
(63% and 66% at 100% silane conversion, respectively, for the reaction shown in Scheme S1).
These results indicate that the mixed-ligand method offers another valid approach to alleviate the
active site leaching problem. This strategy also provides an extra layer of tunability to the Pt
sites, as catalyst properties can be modified by varying either ligand and the molar ratio between
the two ligands.
Post-reaction XPS shows that, on Pt-BPhen+DPTZ, only a small fraction of Pt are the active
sites. According to Table 3, Pt leaching is significant after three cycles (23% total Pt recovery).
Nevertheless, the activity remains almost constant in Figure 3. This implies that, like the Pt-
PDO/CeO2 catalyst, leached Pt from Pt-BPhen+DPTZ/CeO2 contribute little to the activity, and
the main active sites are retained. Figure 2 and 4b show that fresh Pt-BPhen+DPTZ/CeO2 also
has predominately Pt2+ species. After the first cycle, Pt 4f peak shifts to slightly higher BE and
the peak widens. Fitting reveals that some Pt2+ are converted into Pt(2+δ)+ (Figure 4b, Table 3).
After three cycles, both BE and FWHM of Pt 4f peak change back to values similar with the
fresh catalyst, and the peak can again be described with a single Pt2+ component. These results
indicate that similar with Pt-PDO/CeO2, the Pt2+ on fresh Pt-BPhen+DPTZ/CeO2 also represents
19
This article is protected by copyright. All rights reserved.

10.1002/cctc.202000085
ChemCatChem
multiple species of similar oxidation states. During 3 reaction cycles, some are converted into
Pt(2+δ)+ first and then leached away. Only a small fraction are highly stable, remaining on CeO2
with unchanged oxidation state. However, they are much more catalytically active than other
species, and hence the catalyst shows good reusability overall. Table 3 and Figure S5 show that
the decrease in Cl:Pt ratio after the reaction, which has been linked with Pt activation, is also
observed on Pt-BPhen+DPTZ/CeO2.[23b] N:Pt ratio does not decrease, again highlighting the
strong binding between Pt and the bidentate N pockets. We also discovered that treating Pt-
BPhen+DPTZ/CeO2 with only 2 in toluene (without 1) at 70 °C leads to almost complete loss of
Pt, N, Cl, (Figure S6) and the reaction activity. Therefore, despite the main active Pt2+ sites being
highly recyclable under reaction conditions, interaction with silane without alkene induces
significant leaching of active sites.
3. Using bidentate N-based ligands as anchoring ligands
The third strategy we explored is to use a bidentate N-based ligand to modify oxide supports
for Pt-DPTZ LCSCs. The synthesis procedure includes two steps: an “anchoring” ligand was
first deposited onto an oxide to form a ligand-modified oxide. Then, Pt-DPTZ LCSCs were
synthesized using our usual procedure,[23a] with the anchoring-ligand-modified oxide as the
support. We hypothesized that the ligand-modified supports can enhance catalyst reusability
because the anchoring ligand can offer stronger interactions with Pt than surface O from pristine
oxides. We tested the concept using PDO and 4,4’-BPDCA as anchoring ligands on both CeO2
and MgO (with similar surface area). The successful deposition of the anchoring ligand after the
first step is verified by XPS N 1s peak area analysis (Figure S7), with higher loading on CeO2
than on MgO. Pt 4f XPS of both modified-CeO2-supported Pt-DPTZ catalysts show
20
This article is protected by copyright. All rights reserved.

10.1002/cctc.202000085
ChemCatChem
predominantly Pt2+ (Figure 5), as on pristine CeO2.[23a] On both MgO-supported catalysts, the Pt
4f peak is wider and more asymmetric (low-binding-energy tail), compared to the CeO2 catalysts,
and fitting shows a small fraction of Pt0 (Figure 5), likely from small amount of Pt nanoparticles,
which is not observed with Pt-DPTZ on pristine MgO.[23a]
Figure 5. Fittings of Pt 4f XP spectra of Pt-DPTZ LCSCs supported on an oxide (CeO2 or MgO)
modified by an anchoring ligand (PDO or 4,4’-BPDCA). Each spectrum is labeled as “anchoring
ligand, support”, and BPDCA refers to 4,4’-BPDCA. All spectra are normalized so that peaks are
of similar intensity for better peak shape comparison. Pt exist predominantly as Pt2+ on modified
CeO2, while a small fraction of Pt0 is observed on modified MgO.
The activity and Pt recyclability of these catalysts are reported in Table 4. The enhancement
in Pt recyclability by the anchoring ligand is clearly observed on both supports, as Pt recovery
increases on ligand-modified supports compared with pristine supports. Meanwhile, activity
21
This article is protected by copyright. All rights reserved.

10.1002/cctc.202000085
ChemCatChem
drops, the extent of which varies. Despite the activity loss, Table 4 shows that when the proper
anchoring ligand and support are used (PDO and CeO2), this strategy can create Pt-DPTZ LCSCs
with significantly improved Pt recyclability (from 62% to 82%) while maintaining acceptable
activity (68% yield at 70 °C for 30 min, similar with Pt-PDO/CeO2 in Table 2). Therefore, the
anchoring ligand method is a promising approach. Pt recyclability is better on CeO2 than on
MgO, likely due to the higher anchoring ligand coverage on CeO2 (Figure S7).
Table 4. Activity and total Pt recovery (after 1 reaction cycle) of Pt-DPTZ LCSCs supported on
anchoring-ligand-modified oxides.
Yield
(%)
XPS Pt recovery
(%)
Support
Anchoring ligand
No anchoring ligand
PDO
90[c, d]
68
62
82
91
32
36
47
[a]
CeO2
4,4’-BPDCA
No anchoring ligand
PDO
0
95[c, d]
49
MgO[b]
4,4’-BPDCA
93[c]
[a] Reaction condition: T = 70 °C, t = 30 min, 30 mg catalyst, 6 mmol 1, 5 mmol 2, and 3 mL toluene.
[b] Reaction condition: T = 75 °C, t = 120 min, 30 mg catalyst, 6 mmol 1, 5 mmol 2, and 3 mL toluene.
[c] Complete conversion of 2 was achieved in these reactions.
[d] Data on unmodified Pt-DPTZ SACs were previously published in reference 23b.
We note that, in some cases, the activity loss with the anchoring ligand is too significant. For
example, 4,4’-BPDCA leads to complete deactivation on CeO2. 5,5’-BPDCA, another ligand
with the same −COOH groups but at different positions (Figure S8a), does the same. Control
experiments show that modifying CeO2 with trifluoroacetic acid (Figure S8b), a ligand with a
−COOH group but without the bidentate N pocket to anchor Pt-DPTZ, does not completely
22
This article is protected by copyright. All rights reserved.

10.1002/cctc.202000085
ChemCatChem
deactivate the catalyst. Consequently, the deactivation cannot be simply attributed to −COOH
groups occupying certain catalytically relevant sites on CeO2, such as oxygen defects. We
suspect that the deactivation might be related to strong interactions between −COOH groups and
CeO2 forcing BPDCA to stand up, pushing Pt away from the support. The loss of Pt-support
interaction may impact the electronic structure of Pt enough to deactivate the site. We have
demonstrated previously that surface O temporarily detach from Pt during catalysis, acting as
“reversible leaving groups,”[23b] so the lack of Pt-support binding may also reduce the number of
leaving groups, so that the Pt is over-coordinated. The sensitivity of the metal center to the local
coordination environment is a topic of ongoing study and interest.
Conclusions
In this work, we reported a series of ligand-coordinated supported Pt hydrosilylation catalysts
with various bidentate N-based ligands, designed to improve catalyst recyclability over previous
Pt-DPTZ SACs. These heterogeneous catalysts mostly contain highly dispersed Pt2+ sites, and
are active for alkene hydrosilylation under mild conditions with high selectivity. We
demonstrated three approaches to alleviate the active site leaching problem on the original Pt-
DPTZ SACs. First, by replacing DPTZ with a ligand containing additional functional groups that
offer strong interactions with oxide supports, we developed Pt-PDO/CeO2, a catalyst with steady
activity through four reaction cycles and much less Pt leaching. A Pt(2+δ)+ species converted from
Pt2+ is likely the active site. Second, we discovered that mixing DPTZ with BPhen, a ligand
providing additional van der Waals interaction with the support, leads to a highly reusable
catalyst Pt-BPhen+DPTZ/CeO2, which shows only 10% activity loss after three reaction cycles,
due to a Pt2+ species that is highly active and stable, despite accounting for only a small fraction
23
This article is protected by copyright. All rights reserved.

10.1002/cctc.202000085
ChemCatChem
of total Pt. Finally, we showed that modifying oxide supports with an anchoring ligand enhances
Pt recyclability on Pt-DPTZ SACs. All three strategies are promising to enhance the reusability
of supported Pt-ligand single-atom hydrosilylation catalysts, while maintaining high activity and
selectivity. Meanwhile, new development in synthesis methods reported here also expands future
opportunities to fine-tune metal centers in these catalysts towards desired properties.
Acknowledgement
This work was supported by the U. S. Department of Energy, Office of Basic Energy
Sciences, Chemical Sciences program, DE-SC0016367. XPS measurements were carried out at
Indiana University (IU) Nanoscale Characterization Facility with assistance from Dr. Yaroslav
Lasovyj, Dr. Xuemei Zhou, and Eman Wasim. GC-MS measurements were carried out at the IU
Mass Spectrometry Facility with assistance from Dr. Jonathan A. Karty. ICP-MS measurements
were performed at IU Department of Earth & Atmospheric Sciences with assistance from Dr.
Laura Wasylenki and Benjamin S. Underwood.
Conflicts of interest
There are no conflicts of interest to declare.
Keywords: Heterogeneous Catalysis, Hydrosilylation, Platinum, N Ligands, X-ray
Photoemission Spectroscopy
24
This article is protected by copyright. All rights reserved.

10.1002/cctc.202000085
ChemCatChem
References
[1]
[2]
L. Sommer, E. Pietrusza, F. Whitmore, J. Am. Chem. Soc. 1947, 69, 188-188.
a) O. R. Pierce, Y. K. Kim, Rubber Chem. Technol. 1971, 44, 1350-1362; b) B.
Marciniec, E. Walczuk, P. Blazejewska-Chadyniak, D. Chadyniak, M. Kujawa-Welten,
S. Krompiec, N. Auner, J. Weiss, Wiley VCH, Weinheim 2003; c) Y. Morita, S. Tajima,
H. Suzuki, H. Sugino, J. Appl. Polym. Sci. 2006, 100, 2010-2019; d) E. Beyou, P. Babin,
B. Bennetau, J. Dunogues, D. Teyssie, S. Boileau, J. Polym. Sci., Part A: Polym. Chem.
1994, 32, 1673-1681; e) C. Iojoiu, M. J. Abadie, V. Harabagiu, M. Pinteala, B. C.
Simionescu, Eur. Polym. J. 2000, 36, 2115-2123; f) Z. Li, J. Qin, Z. Yang, C. Ye, J.
Appl. Polym. Sci. 2004, 94, 769-774; g) A. Sellinger, R. M. Laine, V. Chu, C. Viney, J.
Polym. Sci., Part A: Polym. Chem. 1994, 32, 3069-3089; h) D. B. Drazkowski, A. Lee, T.
S. Haddad, D. J. Cookson, Macromolecules 2006, 39, 1854-1863; i) A. Tuchbreiter, H.
Werner, L. H. Gade, Dalton Trans. 2005, 1394-1402.
[3]
[4]
H. Maciejewski, A. Wawrzyńczak, M. Dutkiewicz, R. Fiedorow, J. Mol. Catal. A: Chem.
2006, 257, 141-148.
a) D. Troegel, J. Stohrer, Coord. Chem. Rev. 2011, 255, 1440-1459; b) B. Marciniec,
Hydrosilylation: a comprehensive review on recent advances, Vol. 1, Springer Science &
Business Media, 2008; c) T. Ganicz, T. Pakuła, W. A. Stańczyk, J. Organomet. Chem.
2006, 691, 5052-5055.
[5]
a) B. Boury, R. J. Corriu, D. Leclercq, P. H. Mutin, J. M. Planeix, A. Vioux,
Organometallics 1991, 10, 1457-1461; b) A. Mori, H. Sato, K. Mizuno, T. Hiyama, K.
Shintani, Y. Kawakami, Chem. Lett. 1996, 25, 517-518.
25
This article is protected by copyright. All rights reserved.

10.1002/cctc.202000085
ChemCatChem
[6]
[7]
a) M. J. O'brien, US Patent 6,531,540, 2003; b) C. Herzig, B. Deubzer, D. Huettner, US
Patent 5,241,034, 1993.
a) H. Jyono, H. Odaka, H. Ito, H. Iwakiri, US Patent 6,444,775, 2002; b) T. Watabe, T.
Matsumoto, T. Onoguchi, K. Tsuruoka, US Patent 6,207,766, 2001.
P. Jerschow, Silicone elastomers, Vol. 137, Smart Publications, 2001.
a) R. H. Morris, Chem. Soc. Rev. 2009, 38, 2282-2291; b) B. K. Langlotz, H. Wadepohl,
L. H. Gade, Angew. Chem. Int. Ed. 2008, 47, 4670-4674; c) S. C. Bart, E. Lobkovsky, P.
J. Chirik, J. Am. Chem. Soc. 2004, 126, 13794-13807; d) I. Vankelecom, P. Jacobs,
Catal. Today 2000, 56, 147-157.
[8]
[9]
[10] a) M. Xue, J. Li, J. Peng, Y. Bai, G. Zhang, W. Xiao, G. Lai, Appl. Organomet. Chem.
2014, 28, 120-126; b) M. Igarashi, T. Matsumoto, T. Kobayashi, K. Sato, W. Ando, S.
Shimada, M. Hara, H. Uchida, J. Organomet. Chem. 2014, 752, 141-146; c) H. Dong, Y.
Jiang, H. Berke, J. Organomet. Chem. 2014, 750, 17-22; d) J. Y. Wu, B. N. Stanzl, T.
Ritter, J. Am. Chem. Soc. 2010, 132, 13214-13216; e) P. B. Glaser, T. D. Tilley, J. Am.
Chem. Soc. 2003, 125, 13640-13641; f) S. Nozakura, S. Konotsune, Bull. Chem. Soc.
Jpn. 1956, 29, 326-331; g) L. Bareille, S. Becht, J. L. Cui, P. Le Gendre, C. Moïse,
Organometallics 2005, 24, 5802-5806; h) S. Harder, J. Brettar, Angew. Chem. Int. Ed.
2006, 45, 3474-3478; i) V. Leich, T. P. Spaniol, L. Maron, J. Okuda, Chem. Commun.
2014, 50, 2311-2314.
[11] J. L. Speier, J. A. Webster, G. H. Barnes, J. Am. Chem. Soc. 1957, 79, 974-979.
[12] B. Karstedt, US Patent 3,775,452, 1973.
[13] a) T. Galeandro-Diamant, M.-L. Zanota, R. Sayah, L. Veyre, C. Nikitine, C. de Bellefon,
S. Marrot, V. Meille, C. Thieuleux, Chem. Commun. 2015, 51, 16194-16196; b) B. P.
26
This article is protected by copyright. All rights reserved.

10.1002/cctc.202000085
ChemCatChem
Chauhan, J. S. Rathore, J. Am. Chem. Soc. 2005, 127, 5790-5791; c) Y. Bai, S. Zhang, Y.
Deng, J. Peng, J. Li, Y. Hu, X. Li, G. Lai, J. Colloid Interface Sci. 2013, 394, 428-433; d)
J. Stein, L. Lewis, Y. Gao, R. Scott, J. Am. Chem. Soc. 1999, 121, 3693-3703; e) T. K.
Meister, K. Riener, P. Gigler, J. r. Stohrer, W. A. Herrmann, F. E. Kühn, ACS Catal.
2016, 6, 1274-1284.
[14] a) I. E. Markó, S. Stérin, O. Buisine, G. Mignani, P. Branlard, B. Tinant, J.-P. Declercq,
Science 2002, 298, 204-206; b) I. E. Markó, S. Sterin, O. Buisine, G. Berthon, G.
Michaud, B. Tinant, J. P. Declercq, Adv. Synth. Catal. 2004, 346, 1429-1434.
[15] a) J. C. Bernhammer, H. V. Huynh, Organometallics 2013, 33, 172-180; b) J. J.
Dunsford, K. J. Cavell, B. Kariuki, J. Organomet. Chem. 2011, 696, 188-194; c) M. A.
Taige, S. Ahrens, T. Strassner, J. Organomet. Chem. 2011, 696, 2918-2927.
[16] B. Marciniec, K. Posała, I. Kownacki, M. Kubicki, R. Taylor, ChemCatChem 2012, 4,
1935-1937.
[17] C. M. Downing, H. H. Kung, Catal. Commun. 2011, 12, 1166-1169.
[18] N. Sabourault, G. Mignani, A. Wagner, C. Mioskowski, Org. Lett. 2002, 4, 2117-2119.
[19] a) Y. J. Chen, S. F. Ji, W. M. Sun, W. X. Chen, J. C. Dong, J. F. Wen, J. Zhang, Z. Li, L.
R. Zheng, C. Chen, Q. Peng, D. S. Wang, Y. D. Li, J. Am. Chem. Soc. 2018, 140, 7407-
7410; b) Y. Zhu, T. Cao, C. Cao, J. Luo, W. Chen, L. Zheng, J. Dong, J. Zhang, Y. Han,
Z. Li, C. Chen, Q. Peng, D. Wang, Y. Li, ACS Catal. 2018, 8, 10004-10011; c) X. Cui, K.
Junge, X. Dai, C. Kreyenschulte, M.-M. Pohl, S. Wohlrab, F. Shi, A. Brückner, M.
Beller, ACS Central Science 2017, 3, 580-585.
[20] a) K. M. Mantovani, J. F. Stival, F. Wypych, L. Bach, P. G. P. Zamora, M. L. Rocco, S.
Nakagaki, J. Catal. 2017, 352, 442-451; b) M. Rimoldi, D. Fodor, J. A. van Bokhoven,
27
This article is protected by copyright. All rights reserved.

10.1002/cctc.202000085
ChemCatChem
A. Mezzetti, Chem. Comm. 2013, 49, 11314-11316; c) W. Xu, Y. Li, B. Yu, J. Yang, Y.
Zhang, X. Chen, G. Zhang, Z. Gao, J. Solid State Chem. 2015, 221, 208-215; d) L. Chen,
S. Rangan, J. Li, H. Jiang, Y. Li, Green Chemistry 2014, 16, 3978-3985.
[21] a) Z. Huang, X. Gu, Q. Cao, P. Hu, J. Hao, J. Li, X. Tang, Angew. Chem. 2012, 124,
4274-4279; b) E. Fako, Z. Lodziana, N. Lopez, Catal. Sci. Technol. 2017, 7, 4285-4293;
c) Y. X. Chen, Z. W. Huang, Z. Ma, J. M. Chen, X. F. Tang, Catal. Sci. Technol. 2017, 7,
4250-4258; d) G. Vilé, D. Albani, M. Nachtegaal, Z. Chen, D. Dontsova, M. Antonietti,
N. López, J. Pérez‐Ramírez, Angew. Chem. Int. Ed. 2015, 54, 11265-11269; e) W. Xu, B.
Yu, Y. Zhang, X. Chen, G. Zhang, Z. Gao, Appl. Surf. Sci. 2015, 325, 227-234; f) P. Ji,
K. Manna, Z. Lin, A. Urban, F. X. Greene, G. Lan, W. Lin, J. Am. Chem. Soc. 2016, 138,
12234-12242; g) N. M. Schweitzer, B. Hu, U. Das, H. Kim, J. Greeley, L. A. Curtiss, P.
C. Stair, J. T. Miller, A. S. Hock, ACS Catal. 2014, 4, 1091-1098; h) H. Sohn, J.
Camacho-Bunquin, R. Langeslay, P. Ignacio-de Leon, J. Niklas, O. Poluektov, C. Liu, J.
Connell, D. Yang, J. Kropf, Chem. Commun. 2017; i) B. Qiao, A. Wang, X. Yang, L. F.
Allard, Z. Jiang, Y. Cui, J. Liu, J. Li, T. Zhang, Nature chemistry 2011, 3, 634-641; j) Z.
Li, S. Ji, Y. Liu, X. Cao, S. Tian, Y. Chen, Z. Niu, Y. Li, Chem. Rev. 2019; k) L. DeRita,
S. Dai, K. Lopez-Zepeda, N. Pham, G. W. Graham, X. Pan, P. Christopher, J. Am. Chem.
Soc. 2017, 139, 14150-14165; l) J. Liu, ACS Catal. 2017, 7, 34-59; m) X. Cui, W. Li, P.
Ryabchuk, K. Junge, M. Beller, Nat. Catal. 2018, 1, 385-397; n) L. Liu, D. M. Meira, R.
Arenal, P. Concepcion, A. V. Puga, A. Corma, ACS Catal. 2019, 10626-10639.
[22] a) D. Skomski, C. D. Tempas, B. J. Cook, A. V. Polezhaev, K. A. Smith, K. G. Caulton,
S. L. Tait, J. Am. Chem. Soc. 2015, 137, 7898-7902; b) D. Skomski, C. D. Tempas, G. S.
Bukowski, K. A. Smith, S. L. Tait, J. Chem. Phys. 2015, 142, 101913; c) D. Skomski, C.
28
This article is protected by copyright. All rights reserved.

10.1002/cctc.202000085
ChemCatChem
D. Tempas, K. A. Smith, S. L. Tait, J. Am. Chem. Soc. 2014, 136, 9862-9865; d) C. D.
Tempas, D. Skomski, B. J. Cook, D. Le, K. A. Smith, T. S. Rahman, K. G. Caulton, S. L.
Tait, Chem. Eur. J. 2018, 24, 15852-15858; e) C. G. Williams, M. Wang, D. Skomski, C.
D. Tempas, L. L. Kesmodel, S. L. Tait, J. Phys. Chem. C 2017; f) T. W. Morris, I. J.
Huerfano, M. Wang, D. L. Wisman, A. C. Cabelof, N. U. Din, C. D. Tempas, D. Le, A.
V. Polezhaev, T. S. Rahman, K. G. Caulton, S. L. Tait, Chem. Eur. J. 2019, 25, 5565-
5573.
[23] a) L. Chen, G. E. Sterbinsky, S. L. Tait, J. Catal. 2018, 365, 303-312; b) L. Chen, I. S.
Ali, G. E. Sterbinsky, J. T. L. Gamler, S. E. Skrabalak, S. L. Tait, ChemCatChem 2019,
11, 2843-2854.
[24] L. Chen, V. Agrawal, S. L. Tait, Catalysis Science & Technology 2019, 9, 1802-1815.
29
This article is protected by copyright. All rights reserved.