Organometallic Complexes Anchored to Conductive Carbon for Electrocatalytic Oxidation of Methane at Low Temperature

Article abstract of DOI:10.1021/jacs.5b06392

Low-temperature direct methane fuel cells (DMEFCs) offer the opportunity to substantially improve the efficiency of energy production from natural gas. This study focuses on the development of well-defined platinum organometallic complexes covalently anchored to ordered mesoporous carbon (OMC) for electrochemical oxidation of methane in a proton exchange membrane fuel cell at 80 °C. A maximum normalized power of 403 μW/mg Pt was obtained, which was 5 times higher than the power obtained from a modern commercial catalyst and 2 orders of magnitude greater than that from a Pt black catalyst. The observed differences in catalytic activities for oxidation of methane are linked to the chemistry of the tethered catalysts, determined by X-ray photoelectron spectroscopy. The chemistry/activity relationships demonstrate a tangible path for the design of electrocatalytic systems for C-H bond activation that afford superior performance in DMEFC for potential commercial applications.

Full text of DOI:10.1021/jacs.5b06392

Article
pubs.acs.org/JACS
Organometallic Complexes Anchored to Conductive Carbon for
Electrocatalytic Oxidation of Methane at Low Temperature
Madhura Joglekar,†,⊥ Vinh Nguyen, § Svitlana Pylypenko, Chilan Ngo, Quanning Li,
‡,⊥
†
∥
†
‡
§
,§
,‡
Matthew E. O’Reilly, Tristan S. Gray, William A. Hubbard, T. Brent Gunnoe,* Andrew M. Herring,*
,†
and Brian G. Trewyn*
†
Department of Chemistry and Geochemistry, Colorado School of Mines, Golden, Colorado 80401, United States
Department of Chemical and Biological Engineering, Colorado School of Mines, Golden, Colorado 80401, United States
Department of Chemistry, University of Virginia, Charlottesville, Virginia 22904, United States
‡
§
∥
Department of Physics, University of CaliforniaLos Angeles, Los Angeles, California 90095, United States
*
S Supporting Information
ABSTRACT: Low-temperature direct methane fuel cells (DMEFCs) offer the
opportunity to substantially improve the efficiency of energy production from
natural gas. This study focuses on the development of well-defined platinum
organometallic complexes covalently anchored to ordered mesoporous carbon
(
OMC) for electrochemical oxidation of methane in a proton exchange membrane
fuel cell at 80 °C. A maximum normalized power of 403 μW/mg Pt was obtained,
which was 5 times higher than the power obtained from a modern commercial
catalyst and 2 orders of magnitude greater than that from a Pt black catalyst. The
observed differences in catalytic activities for oxidation of methane are linked to the
chemistry of the tethered catalysts, determined by X-ray photoelectron spectros-
copy. The chemistry/activity relationships demonstrate a tangible path for the
design of electrocatalytic systems for C−H bond activation that afford superior
performance in DMEFC for potential commercial applications.
INTRODUCTION
using commercially available Pt ELAT and Pt−Ru ELAT gas-
■
8
diffusion anode electrodes in a PEMFC. However, negligible
Methane, a primary constituent of natural gas, has attracted
widespread attention as a fuel due to its higher energy content
per mass unit (55.7 kJ g ) compared to other hydrocarbons. In
the past decade, advances in drilling technologies have expanded
the access and reduced the expense of natural gas. Currently,
the predominant use of methane in the energy sector is via
combustion. An alternative approach is to utilize methane in fuel
cells. Fuel cells are dramatically more efficient than heat engines
at lower temperatures and their exhaust is composed of only CO2
and water making them attractive clean energy conversion
devices. With the use of current technology, methane has been
directly electrochemically oxidized only in solid oxide fuel cells
current densities were achieved in both studies, and in general,
−1
the cell could not be stabilized. Considering the very high C−H
−1
bond dissociation energy (435 kJ mol ) for methane, one of the
major challenges is the development of catalysts that provide
sufficient rates of C−H activation at the operating temperatures
of PEMFC (60−100 °C). Despite the fact that some molecular
transition complexes have been demonstrated to activate methane
1
C−H bonds in homogeneous environments at temperatures more
9−18
relevant for PEMFCs (≤200 °C),
to our knowledge there
have been no attempts to incorporate these molecular systems
into electrochemical environments for hydrocarbon oxidation in
fuel cells.
(
SOFCs) with most research focused on the development of
new anode materials for this conversion.
2
−6
Joglekar et al. have recently demonstrated a method to covalently
anchor molecular complexes onto a conductive mesoporous
However, the high
operating temperatures (650−1100 °C) and substantial capital
expenses for SOFCs keep this technology from being cost-
effective and necessitates the need to develop alternatives.
Proton exchange membrane fuel cells (PEMFCs) have the
advantage of much higher power densities, faster start up and
shut down, good cyclability, and the potential for scalability from
micro to large-scale distributed power generation; however, their
lower temperature of operation makes the activation of methane
extremely challenging under the operating conditions of the fuel
cell. In 1962, Niedrach made the first attempt to demonstrate
19
carbon support. This has been achieved through a lithiation
strategy to selectively deprotonate defect sites in the graphitic
structure of mesoporous carbons, thereby allowing covalent
functionalization of the surface at the defect sites. A limited
number of reports have indicated the use of molecular complexes
adsorbed onto carbon surface as efficient catalysts for the oxygen
20,21
reduction reaction (ORR) in methanol or hydrogen PEMFCs.
Received: June 19, 2015
7
a DMEFC. In 2012, Ferrell et al. reported methane activation
©
XXXX American Chemical Society
A
DOI: 10.1021/jacs.5b06392
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Recently, an ethanol fuel cell was developed using Rh-based
organometallic complexes adsorbed onto conductive Vulcan
support, which generated electric power from the oxidation of
the larger aggregates. The mixture was kept under vigorous stirring and
2
.5 mL of n-butyllithium (2.5 M in hexane, Sigma) was added dropwise
at −78 °C. Following the addition, the mixture was stirred at room
temperature for 4 h. In the meantime, 1.06 mmol of 6-bromo-2,2′-
bipyridine purchased from Sigma was dried under vacuum. After 4 h, it
was added to the Schlenk flask, and the mixture was further stirred for 2 h
at 35 °C. To obtain surface functionalized OMC powder, the mixture
was centrifuged and the supernatant was removed. During this step and
all the further steps, no efforts were taken to exclude air. The product
was washed with methanol 5 times and subsequently centrifuged. Finally,
it was suspended in methanol overnight to remove any unreacted
bipyridine and Li impurities from the pores and surface of OMC. It was
then centrifuged and dried at 100 °C for 4 h. This product was termed as
OMC-6Bp. For the synthesis of OMC-4Bp and OMC-phen, 1.06 mmol
of 4,4′-dibromo-2,2′-bipyridine (purchased from Carbosynth) and
22,23
ethanol to the acetate product.
These types of systems,
wherein the molecular catalysts are adsorbed on the surface of
carbon through noncovalent interactions, tend to fail in more
complicated C−H activation systems, such as DMEFCs, due to
the weak interactive forces between the molecular catalyst and the
conductive carbon support. To address this, we have developed
a new selective surface functionalization strategy for covalently
tethering molecular complexes on the surface of ordered meso-
porous carbon (OMC) materials. This enables us to introduce
unique molecular systems into the fuel cell configurations.
Previously, such materials were difficult to prepare due to the
limitations of carbon functionalization techniques. The new
OMC-support Pt catalysts afford the direct oxidation of methane
without poisoning by carbon monoxide adsorption at low
temperatures (<150 °C), leading to the design, fabrication and
testing of new low temperature DMEFCs.
5
-bromo-1,10-phenanthroline synthesized via a previously reported
25
procedure were vacuum-dried and added after 4 h following the
addition of n-butyllithium in the above procedure.
Synthesis of [PtPh (μ-SEt )] . The Pt dimeric complex was synthe-
2
2 2
2
6
sized using a modification of a previously reported procedure. In brief,
440 mg of commercially available yellow crystalline cis-PtCl (μ-SEt )
2 2
2
In contrast to the current technology for the direct conversion
of methane using SOFCs that operate exclusively at very
high temperatures (650−1100 °C), we report the first successful
demonstration of electrooxidation of methane at 80 °C utilizing
these novel OMC-bound C−H activating molecular catalysts.
was dispersed in 10 mL of diethyl ether in a Schlenk flask. The synthesis
was done under inert atmosphere using standard Schlenk line
techniques. To the above suspension, 4 mL of phenyl lithium solution
(
2
1.8 M in dibutylether, Sigma) was added dropwise over a course of
0 min at −78 °C. This suspension was allowed to stir for 2 h at 0 °C.
After 2 h, 8 mL of distilled water was added dropwise over a course of
0 min under a constant nitrogen flow. A thick and nonsettling
1
EXPERIMENTAL SECTION
suspension was formed, which was filtered through 545 Celite. At this
and all the subsequent points, no efforts were made to exclude air. The
filtrate was poured into a separatory funnel and dichloromethane was
added to it. The dichloromethane layer was washed three times with
water, dried with magnesium sulfate, filtered through activated charcoal,
and reduced to dryness to obtain a white solid. This was dissolved in
15 mL of dichloromethane and gently heated (40 °C) to ensure complete
dissolution of the white solid. The mixture was filtered through a medium
porosity frit into a flask, and the filtrate was concentrated to approxi-
mately 6 mL in vacuo. A white crystalline solid precipitated, which was
dissolved in 50 mL of pentane. The crystallization was left at −20 °C
overnight. White crystalline solid (75% yield) was obtained following
filtration.
■
(
Synthesis of Large Pore Mesoporous Silica Nanoparticles
l-MSN). l-MSN template was synthesized according to a previously
2
4
reported literature procedure. A nonionic surfactant Pluronic
P104 (7.0 g, BASF) was added to 1.6 M HCl (273.0 g) in a 500 mL
Erlenmeyer flask. After the mixture stirred at 55 °C for 1 h,
tetramethylorthosilicate (TMOS, 10.64 g, Sigma) was added at once
and the reaction mixture was further stirred at 55 °C for 24 h. It is crucial
to maintain a constant reaction temperature for 24 h in order to obtain
l-MSN with uniform morphology and pore size. The mixture was then
hydrothermally treated at 150 °C for 24 h in a Teflon lined autoclave. In
the final step, the mixture was cooled, filtered, and washed with water
and copious amounts of methanol to obtain a white powder. It was then
lyophilized overnight, followed by calcination at 550 °C for 6 h at a ramp
Synthesis of OMC-4Bp-Pt-Ph , OMC-6Bp-Pt-Ph , and OMC-
2
2
−1
phen-Pt-Ph . Platinum was coordinated to the bipyridyl ligand on the
rate of 1.5 °C min to remove the nonionic surfactant P104.
Synthesis of Ordered Mesoporous Carbon (OMC). OMC with
uniform morphology was synthesized according to our recently
2
surface of OMC using the following procedure. OMC-4Bp, OMC-6Bp,
or OMC-phen (0.1 g) and 0.03 mmol Pt-dimer were added in a Schlenk
flask in anhydrous diethyl ether as a solvent. The reaction mixture was
stirred at room temperature for 24 h under argon. Finally, the product
was centrifuged and the supernatant was discarded. It was washed
six times with ether to remove any unreacted, excess Pt-dimer and
lyophilized overnight.
19
published report. Typically, 1 g of l-MSN having a pore volume of
3
−1
1
.12 cm g was impregnated with 1.20 g of sucrose and 7 g of water
in a centrifuge tube and sonicated until all the particles were evenly
dispersed. After this solution was transferred to a crucible, 0.13 g of
concentrated sulfuric acid was added. The mixture was stirred to break
any large chunks of l-MSN or sucrose and then heated at 100 and 160 °C
for 6 h each. This process of addition and partial carbonization of
sucrose, water, and concentrated sulfuric acid was repeated until the
pore volume of the l-MSN template was reduced to approximately zero.
The pore volume of the silica−carbon composite was determined after
each step using the nitrogen sorption analysis. Also, the amounts of
sucrose and sulfuric acid required for each step were determined based
on the pore volume of the silica−carbon composite. The complete
pyrolysis of carbon was carried out under nitrogen atmosphere at 900 °C
for 5 h in a tube furnace. In the final step, the l-MSN template was etched
with 10% HF overnight in centrifuge tubes. The resulting OMC was
washed with copious amounts of water until the pH was neutral.
Covalent Attachment of Bipyridine or Phenanthroline to the
Surface of OMC. All the reactions were performed under inert
atmosphere using standard Schlenk line techniques. For the covalent
attachment of 2,2′-bipyridine, the surface of OMC was first lithiated
using n-butyllithium. In a typical procedure, a Schlenk flask was charged
with 0.25 g of OMC and dried overnight under vacuum at 100 °C to
remove all the moisture. The OMC was then suspended in 25 mL of
diethyl ether and sonicated for 15 min to disperse the particles and break
Synthesis of OMC-4Bp-Pt-Cl , OMC-6Bp-Pt-Cl , and OMC-
2
2
phen-Pt-Cl . The chloro versions of the molecular complexes were
2
synthesized using a procedure similar to the synthesis of homogeneous
2
7
complexes and is as follows. OMC-4Bp, OMC-6Bp, or OMC-phen
0.1 g) was added to a round-bottom flask containing 250 mL of
distilled water and 0.2 mL of concentrated hydrochloric acid. A platinum
precursor, K PtCl (100 mg), was added to the same flask, and the
(
2
4
reaction mixture was heated to reflux for 24 h. Finally, the powder
was centrifuged, washed with copious amounts of distilled water, and
lyophilized overnight.
Characterization. Small-angle X-ray scattering (SAXS) data was
collected using a Rigaku S-Max 3000 High Brilliance three-pinhole
SAXS system outfitted with a MicroMax-007HFM rotating anode
(Cu Kα), Confocal Max-Flux Optic, Gabriel multiwire area detector,
and a Linkam thermal stage. Exposure times for samples were typically
on the order of 1200 s. Nitrogen sorption analyses were done on a
Micromeritics Tristar 3000 surface area and porosity analyzer using
Brunauer−Emmett−Teller (BET) equation to calculate the surface area
and pore volume and the Barrett−Joyner−Halenda (BJH) method
to calculate pore size distribution. Raman spectroscopy was done using
B
DOI: 10.1021/jacs.5b06392
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
WITEC Alpha 300 confocal microscope. The morphology of the
samples was analyzed using JEOL JSM-7000F field emission scanning
electron microscope (FESEM). The samples were dispersed on a con-
ductive carbon tape and analyzed using an accelerating voltage of 7 kV.
High-resolution micrographs and X-ray energy dispersive spectroscopy
during all fuel cells testing. The anode exit gas line was also connected
with a CO trap that contained 1 M NaOH solution. At the end of the
2
testing procedure, the bicarbonate solution collected from this trap was
analyzed by 13C NMR spectroscopy to probe for the formation of CO2.
To test consistency and reproducibility of the MEA, each experiment
was done in triplicate and demonstrated consistent results.
(
EDS) spot analysis were obtained using Philips CM200 transmission
electron microscope operated at 200 kV and FEI Titan S/TEM operated
at 300 kV. The samples were dispersed in 2 mL of ethanol and drop-
casted onto lacey carbon TEM grids for observation. Thermogravi-
metric analysis was done in a SETSYS Evo system with B-type DTA
A Gamry Instruments potentiostat was used to perform polarization
and electrochemical impedance spectroscopy (EIS) experiments as
2
9
previously reported. The polarization curves were obtained by first
remaining at open circuit for 20 min. The potential was then stepped
down from the open circuit potential until the limiting current was
reached. For anode polarization, nitrogen instead of oxygen was fed to
the cathode, which became a pseudoreference electrode. The potential
was then scanned from open circuit to more positive potentials until a
limiting current was achieved.
−1
measurement head and a temperature ramp rate of 10 °C min in
air for quantification of metal. Helium gas was used to maintain inert
atmosphere in order to study the decomposition behavior of the
tethered catalyst. XPS analysis was performed on a Kratos Nova X-ray
photoelectron spectrometer equipped with a monochromatic Al Kα
source operating at 300 W. Survey and high-resolution C 1s, O 1s, N 1s,
Si 2p, Cl 2p, and Pt 4f spectra were acquired at 160 and 20 eV,
respectively, providing charge compensation using low energy electrons.
Three areas per sample were analyzed. Data analysis was performed
using CasaXPS software. A linear background was applied to C 1s, O 1s,
N 1s, Si 2p, and Cl 2p regions, and a Shirley background was applied to
Pt 4f region. Quantification was performed using sensitivity factors
supplied by manufacturer. Analysis included charge referencing to the
internal aromatic carbon signal at 284.4 eV.
Measurement of Electrical Conductivity of the OMC Support
and the OMC Bound Catalyst. A standard procedure was used
30
to measure the bulk electrical conductivity of the OMC. The
results showed a decrease in the conductivity of OMC from 3.96 to
−1
0.609 S cm after functionalization measured at 127 psig. To boost the
electrical conductivity of the catalyst, we added 20 wt % unfunction-
alized OMC in with the OMC-4Bp-Pt-Cl
catalytic activity. Our measured conductivity of 80 wt% OMC-4Bp-
which had the highest
2
−
1
Pt-Cl and 20 wt % OMC was 2.14 S cm , demonstrating a clear
2
Materials for Methane Fuel Cell Testing. Methane cylinder
purchased from General Air Corporation was used as received. Nafion-
improvement in conductivity by mixing functionalized and unfunction-
alized OMC.
1
110 membranes purchased from Ion Power were first cleaned and
protonated by refluxing in 3% H O for 1 h, followed by 1 h refluxing in
2
2
RESULTS AND DISCUSSION
■
DI water, 1 h refluxing in 0.5 N H SO , and finally 1 h refluxing in DI
2
4
water. After conditionings, the membranes were stored in DI water at
room temperature in the dark before use. The anode was fabricated with
six different molecular catalysts prepared in our lab as previously
described coating onto hydrophobic gas diffusion layers GDL LT 1400-W
carbon cloth (E-tek). Detailed fabrication is illustrated in the following
Membrane Electrode Assembly (MEA) Fabrication section. For the
cathode, all fuel cells used Electrode Los Alamos Type (ELAT) prepared
with ELAT LT 1400 gas diffusion layers (Nuvant Systems, Inc.) and
Synthesis and Characterization of OMC-Tethered
Single-Site Catalysts. Pt(II) complexes with chelating bis-
nitrogen ligands have been studied for C−H activation of
13,14,31−33
hydrocarbons, including alkanes.
The use of [(bpy)-
Pt(Ph)(THF)]+ (bpy = various 2,2′-bipyridyl ligands) and
closely related complexes has been recently reported to catalyze
34−36
the functionalization of benzene,
which has a C−H bond
2
dissociation energy that is ∼29.3 kJ/mol stronger than that
of methane. The demonstrated ability of these Pt(II) complexes
to activate C−H bonds combined with the availability (both
commercially and synthetically) of a variety of substituted
,2′-bipyridyl compounds motivated us to start with (bpy)PtX
X = Cl, phenyl) complexes as the molecular component for the
containing 0.5 mg Pt/cm (20% Pt on Vulcan XC-72 carbon). The
3M perfluorosulfonic acid ionomer with EW = 733 (structure given in
the Supporting Information) as an aqueous dispersion was used in the
28
catalyst layer.
Membrane Electrode Assembly (MEA) Fabrication. Catalyst
2
(
2
inks were first prepared as previously described by combining the
8
desired catalyst, water, 2-propanol, and 3M ionomer dispersion. The
new OMC-supported catalyst. OMCs have attracted significant
attention as a conductive support for application in fuel cells
due to their unique properties such as high surface area, uniform
pore size distribution, interconnected mesopores and high con-
ionomer dispersion was added such that the 3M solid was 25% of the
total mass of the catalyst and 3M solid in the ink. Water was added in an
amount that was 10 times the mass of the catalyst in the ink. 2-Propanol
was finally added in the ink to make water/2-propanol (3:2). The inks
were sonicated in an ultrasonic bath for 5 min, followed by mixing with
Vortex Mixer for 2 min. The sonicating and mixing process was repeated
three times. The inks were finally airbrushed onto GDLs to make gas
diffusion electrodes. Following airbrushing, the electrodes were placed
under an IR 250W heat lamp to evaporate the water/2-propanol solvent
in the catalyst layer. Both anode and cathode were prepared with an area
37−39
ductivity.
In this study, a high surface area OMC support,
used for anchoring the platinum-based catalysts, was prepared
according to our recent report using large pore mesoporous silica
19
nanoparticles (l-MSN) as a hard template. As previously noted,
this OMC support has a number of structural defect sites that
can be leveraged for covalently anchoring catalysts on its surface.
Accordingly, the defect sites on the surface of OMCs were initially
deprotonated using a strong base, n-butyllithium, followed by
the addition of a brominated ligand as shown in Figure 1.
This product was isolated, washed, lyophilized overnight and
then coordinated with a platinum precursor such as K PtCl or
2
of 5.48 cm . The electrodes were hot pressed on cleaned Nafion-1110
membranes using a digital combo multipurpose press, DC14 (GEO
Knight & Co., Inc.), at 80 °C and 60 psig for 90 s. There was a small
variation in the platinum loading on the anode ranging from 0.055 to
19,40
−2
0
.073 mg cm , assuming homogeneous distribution. The loading of Pt
on OMC for various complexes was in the range of 1.5−2.1 wt %.
Fuel Cell Testing. A single-cell hardware with an area of 5.48 cm2
and single serpentine flow fields (Fuel Cell Technologies, Inc.) was used
for this study. Humidified methane and oxygen were fed to the anode
and cathode, respectively, at the same flow rate of 0.3 L/min. Both gases
were flowed and humidified through sparging bottles with modular gas
handling and gas metering system (Lynntech Industry, Inc.). Methane
and oxygen were humidified at 60 and 80 °C, respectively. The cell,
however, was maintained at 80 °C. The effluent from the fuel cell sweeps
through the backpressure regulators and condenses liquid water in trap
bottles. For this study, the backpressure was always kept at 30 psig
2
4
[PtPh (μ-SEt )] to give the corresponding chlorine or phenyl
2
2
2
versions of the molecular complexes (Figure 1). The OMC
support and all the as-synthesized OMC-based catalysts were
characterized using a range of spectroscopy and microscopy
techniques. As seen in Figure S1 (Supporting Information), the
nitrogen sorption isotherms of l-MSN, OMC and the OMC with
covalently attached Pt indicate a type IV isotherm characteristic of
41
mesoporous nanomaterials. The pore size distribution is narrow
for both the unfunctionalized OMC support and the surface
C
DOI: 10.1021/jacs.5b06392
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
arrangement of pores indicated by the 100, 110, and 200 peaks
(
Figure S2, Supporting Information), while the as-synthesized
OMC support has a disordered arrangement of pores along
the 110 and 200 planes as indicated by the absence of peaks
(
Figure S2, Supporting Information). The OMC particles are
discrete with uniform morphology as indicated by the high-
resolution transmission and scanning electron micrographs
(Figure 2). The pore channels and the structure are also clearly
visible in these micrographs (Figure 2D). High resolution
scanning transmission electron micrographs (STEM) of the
catalyst sample (Figures 2D and S14) OMC-4Bp-Pt-Cl indicated
2
negligible Pt nanoparticles on the OMC surface. Hence, the
catalytic activity can be attributed to the tethered Pt (II) complex.
EDX analysis revealed that C, O, N and Si were the major
components (Figures S13 and S15, Supporting Information) in
the catalyst samples. The Si content was attributed to the
remnants of the silica template, in the amount less than 0.25 wt %.
A small change in the I /I ratio (0.96 for OMC to 1.06 for
D
G
OMC-4Bp-Pt-Cl ) in the Raman analysis of the OMC-4Bp-Pt-
2
Cl sample revealed that the OMC structure was retained even
2
after chemical modification of the surface (Figure S3c, Supporting
Information). This observation was also consistent for the other
OMC-based samples modified with bipyridine ligands tethered at
the 6-position and phenanthroline ligands (Figure S3, Supporting
Information).
Application of OMC-Tethered Single-Site Catalysts in
−1
DMEFC. Despite good gravimetric energy density (55.6 MJ kg ),
the volumetric energy density of methane is very low
Figure 1. Schematic representation of the general synthetic procedure
for OMC-tethered single-site catalysts.
−
1
(
0.0378 MJ L ) and difficult to utilize at 1 bar since the fuel
is not easily compressed. In addition, the electrooxidation
of methane in acidic environments at low temperature remains
modified ligand anchored OMC. Small angle X-ray scattering
8
(
SAXS) measurements were obtained to determine the pore
exceedingly challenging. A fuel cell running on methane requires
arrangement of the original hard template and the synthesized
water as an additional reactant at the anode (Figure 3A) and
OMC support. The l-MSN hard template has a 2D hexagonal
produces CO at the anode as an oxidation product.
2
Figure 2. Characterization of the as-synthesized OMC support and the tethered molecular catalyst using electron microscopy. (A) Scanning electron
micrograph of l-MSN hard template particles. Scale bar = 500 nm. (B and C) Transmission electron micrographs of a single OMC particle revealing the
pore structure. Scale bars = 100 and 50 nm, respectively. (D) Scanning transmission electron micrograph of several overlapping OMC-4Bp-Pt-Cl2
particles. Scale bar = 200 nm. (E) Bright-field TEM image corresponding to the STEM image in (D). Scale bar = 200 nm. (F) Higher magnification image
of a thinner area on the edge of one particle, indicated by the blue box in (D) and showing the pore structure viewed from the side. Scale bar = 20 nm.
D
DOI: 10.1021/jacs.5b06392
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 3. Polarization curves with power density performance of fuel cells with different molecular catalysts on the anode and the same commercial GDE
−2
containing 0.5 mg Pt cm on the cathode. (A) Schematic representation and equation of the direct methane PEM fuel cell. (B) V−I curves of all
catalysts tested in DMEFC. (C) P−I curves of all catalysts tested in DMEFC. (D) V−I curves of OMC-6Bp-Pt-Ph (light blue and red) and OMC-4Bp-
2
Pt-Cl (dark blue and purple) on the anode before and after adding 20 wt % OMC. (E) P−I curves of OMC-6Bp-Pt-Ph (light blue and red) and OMC-
2
2
4
Bp-Pt-Cl (dark blue and purple) on the anode before and after adding 20 wt % OMC.
2
For the first time, using a series of OMC-tethered molecular
catalysts, we demonstrate power densities ranging from 100 to
00 μW/mg Pt (Table 1), as opposed to previous technology
The polarization curves and the power density performance
for all six OMC-bound molecular catalysts, measured using
optimized fuel cell conditions, are shown in Figure 3B−E. All
OMC-tethered single-site catalysts gave substantially higher OCV
than previous DMEFC studies. Specifically, OMC-4Bp-Pt-Cl2
reached a voltage as high as 0.53 V, whereas its control, OMC-
4Bp, which has no platinum complex bound on the surface, had an
OCV of 0 V (Figure S7). Also, controls using Vulcanas the carbon
4
that showed very low, unsteady open circuit voltage (OCV) at
less than 105 mV and produced no measurable current from
8
the fuel cell. Additionally, this work demonstrates substantial
improvement in power density over the initial publication from7
the General Electric Corporation demonstrating 2.2 μW/mg Pt.
E
DOI: 10.1021/jacs.5b06392
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Table 1. Power Densities Obtained from the Series of OMC-
Tethered Molecular Catalysts As Compared to Previously
Reported and Commercially Available Catalysts in a DMEFC
In this methane system, the voltage quickly dropped to zero after
a finite load resistance allowed electrons to go through the
external circuit. Since the steep decreasing voltage occurred at
very low current density, it could be associated with mass
transport limitation and slow sorption kinetics of methane at
entry
sample
power (μW/mg Pt)
7
1
2
3
4
5
6
7
8
9
OMC-6Bp-Pt-Ph2
83
122
101
103
127
278
403
109
87
the anode. Several of the lower performing polarization curves
OMC-6Bp-Pt-Ph + 20% OMC
2
in Figure 3B appear to reach the linear ohmic region at 0.1 V,
indicating that the poor kinetics of methane oxidation still
dominate the system.
OMC-4Bp-Pt-Ph2
OMC-phen-Pt-Ph2
OMC-6Bp-Pt-Cl2
OMC-4Bp-Pt-Cl2
To further elucidate the limiting performance of the fuel cell,
we investigated both the utilization of methane and the kinetics of
methane oxidation. With the use of the Randles-Sevcik equation,
the diffusion coefficient of methane (shown in Figure 4B) was
OMC-4Bp-Pt-Cl + 20% OMC
2
OMC-phen-Pt-Cl2
Commercial GDE Pt/C
−12
2 −1
estimated at 3.19 × 10 cm s (see Supporting information
for detailed explanation on the calculation). The Cottrell plot
produced from our measurements approaches the origin as has
10
General Electric (Niedrach, 1962) Pt
2
44
been previously described. This describes the two-phase
boundary reaction zone where the methane permeates, dissolves
into the electrolyte, and diffuses toward the electrode. For
reference, the diffusion coefficient of hydrogen at the anode
support for one of the molecular catalysts demonstratednegligible
activity (Figure S8). At open circuit, chemical reactions at the
electrodes are in equilibrium and the OCV directly measures the
difference in the chemical activity of methane at the anode and
the cathode. Although it took approximately 30 min to reach the
equilibrium state, the improved OCV indicates superior perform-
ance of the OMC-supported molecular catalysts in DMEFC
in hydrogen PEM fuel cell is only about 10− cm s , which
indicates a major limitation to be overcome is the transport of
methane to reach the reactive sites. As previously mentioned,
another challenge in this system is the slow sorption kinetics.
Since adsorption of methane is an initial step before the reaction
occurs and desorption is the final step to remove the products,
slow sorption would delay the reaction and lower the cell
performance. To separate the effects of anode electrode reaction
kinetics from mass transport, anodic polarization was used to
analyze the anodic half-cell. For this experiment, nitrogen was
fed to the cathode (rather than oxgyen), which became a
pseudoreference electrode, and a potentiostat was used to
polarize the anode to force methane to oxidize. From the anode
polarization, the Tafel slope was extracted (and its value could be
2
2
−1 45
7
,8
compared to the previous catalysts.
The convex-down
curvature of the polarization curves is attributed to methane
crossover that is oxidized minimally by the reduced Pt on the
cathode, as is expected by previously published work on methane
7,8
oxidation catalyzed by Pt metal.
In general, the chlorine variants of the molecular complexes
gave higher current and power densities than their phenyl
counterparts possibly due to the difference in the mechanism of
C−H activation. Although we have no direct evidence, the Pt−Cl
catalysts might access cationic Pt/chloride ion pairs. Related ion
pairs have been suggested to be involved in Pt(II) mediated
31,42,43
46
methane activation using different catalysts.
Accessing
related directly to the electrocatalytic activity). The smaller the
such species is less likely from the corresponding Pt−Ph species.
slope, the better the cell performance since higher current density
is obtained at a given voltage.
OMC-4Bp-Pt-Cl showed the greatest activity in methane fuel
2
−2
cells at approximately 80 μA cm at 0.2 V with a maximum
power density of 278 μW/mg Pt. These values are significantly
higher than previously reported values (Table 1, entry 10).
Higher power densities were observed when 20% unfunc-
tionalized OMC was mixed with the molecular catalyst tethered
OMC, possibly due to the higher conductivity of the unfunc-
tionalized OMC support. The results showed a significant
improvement in both the current and the power density, which
were nearly 50% higher than those of the catalyst alone for both
From the inset on Figure 4C, which shows the data in a
semilog format, the Tafel slope was extracted between 0.2 and
−1
0.5 V, and the estimated slope was 592 mV dec . At the same
temperature and pressure conditions, using Pt catalyst, typical
Tafel slopes for hydrogen and methanol are about 25 and
−1
47,48
161 mV dec , respectively.
The values of the Tafel slopes
indicate that the activation energy required for methane oxidation
at the fuel cell anode is almost 24 times higher than that for
hydrogen oxidation and 4 times higher than that for methanol
oxidation. This is not surprising considering the challenge in low
temperature activation of methane; however, these results are the
first demonstration that methane activation at these temperatures
is possible.
the lowest (OMC-6Bp-Pt-Ph ) (Table 1, entries 1 and 2) and the
2
highest (OMC-4Bp-Pt-Cl ) (Table 1, entries 6 and 7) performing
2
catalysts. A maximum normalized power of 403 μW/mg Pt was
obtained, which was 5 times higher than that for a modern
commercial catalyst and 2 orders of magnitude greater than that
for Pt black catalyst (Table 1, entry 7). To our knowledge, this is
the highest power output (μW/mg Pt) obtained from a direct
methane fuel cell operating at low temperatures (80 °C).
After fuel cell testing, the catalyst was characterized with FT-IR
to probe for the presence of Pt-carbonyl, as CO adsorption
would have a negative impact on the performance of the catalyst.
The absence of CO adsorption peaks at at 2186, 2087, and
1860 cm− in the FT-IR after methane oxidation indicated no
1
Interestingly, the polarization curves measured for OMC-phen-
49−52
Pt-Cl did not follow the pattern typically observed in PEMFC,
catalyst poisoning by formation of Pt-CO units (Figure S7).
2
where high current densities are observed (see Figure 4A for
This coincides with data we obtained from measuring CO
reactivity with the molecular catalyst (Figures S10 and S11).
Finally, chronoamperometry was performed by holding the fuel
cell at 0.2 V to generate current and allowing the effluent to
OMC-phen-Pt-Cl ). In PEMFC, the polarization appears with a
2
logarithmic drop at the initial decrease in voltage, which is referred
to as the activation region, followed by a linear region due to
the ohmic resistance of the cell. At low voltage, the mass transfer
region appears where the current reaches a limiting value because
the transport hindrances limit the supply of fuel to the active sites.
pass through a CO trap that contained 1 M NaOH solution to
2
1
3
determine if CO was produced over a 20 h period. C NMR was
2
used to characterize the bicarbonate solution collected in this
F
DOI: 10.1021/jacs.5b06392
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 4. Studies of mass transport and reaction kinetics during methane oxidation at the anode. (A) Polarization curves at different scan rates of fuel cell
−2
with OMC-phen-Pt-Cl on the anode and commercial GDE containing 0.5 mg Pt cm on the cathode. (B) The corresponding Cottrell plot for the
2
polarization curves with OMC-phen-Pt-Cl measured at zero potential. (C) Anode polarization of OMC-4Bp-Pt-Cl at 30 psig with Tafel plot in
2
2
the inset.
trap. The spectra revealed the presence of CO32− indicating that
voltage became negative indicating a switch potential of the
electrodes. It is most likely that the redox reactions continued to
generate electrons at the anode that migrated to the cathode
through the external circuit. However, the transport of the
protons, produced from methane oxidation at the interface
between the anode and the membrane, was hindered due to
currently available ionomer polymer technology under anhydrous
conditions (see Supporting Information for details on hydration
effect). Experiments were completed at various humidity levels
and optimized conditions were used for all reported experiments
(see Supporting Information for optimization studies). As a
result, charges built up at the anode and cathode generating a
negative potential. Also, CV was measured on one sample that
demonstrated current production and no redox peak for Pt was
observed (Figure S9).
CO was produced in the fuel cell effluent under these conditions
2
(Figure S8). Also, no current density, and hence no bicarbonate
formation, was observed in the absence of methane ruling out
the possibility of OMC oxidation. The OMC bound ligands that
were used as controls, OMC-4Bp, OMC-6Bp, and OMC-phen,
that did not contain Pt as shown XPS spectra (see next section for
details) showed negligible current densities in DMEFCs in the
13
presence of methane and zero production of CO by NMR.
2
Clearly, the observed carbonate was the product of methane
oxidation to CO . The OMC-4Bp-Pt-Cl catalyst sample was
2
2
analyzed after the fuel cell reaction using electron microscopy
Figure 5). It was observed that the molecular Pt(II) complex
(
decomposed to metallic Pt nanoparticles which led to the
deactivation of the methane oxidation catalyst. High resolution
STEM and EDX (Figure 5E−G) indicated the presence of
crystalline Pt nanoparticles.
Insights Using X-ray Photoelectron Spectroscopy
(XPS). It was observed that most of the activity values obtained
from the electrochemical measurements for each OMC-tethered
catalyst correlate directly with the observations recorded from
the XPS studies. Considerable insights regarding the relationship
between the structure, composition and activity can be gained
Chronopotentiometry was used to test the durability of
the direct methane PEMFC at an applied current density of
2
1
9.3 mA cm− at 80 °C. As Figure S12 shows, the voltage was
stable at 0.2 V for 6 h before it dropped to 0 V. After 10 h, the
G
DOI: 10.1021/jacs.5b06392
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 5. Bright field and scanning transmission electron micrographs of the catalyst OMC-4Bp-Pt-Cl after fuel cell testing, showing high contrast Pt
2
metal nanoparticles disbursed across the support material. (A) TEM image of the entire particle. Scale bar = 200 nm. (B) Higher magnification image of
the particle edge, indicated by the red box in (A), showing the pore structure of the carbon support from the side, as well as a distribution of larger
nanoparticles. Scale bar = 100 nm. (C) Higher magnification image of a thinner area of the particle edge, indicated by the blue box in (A), showing the
presence of many smaller nanoparticles. Scale bar = 20 nm. (D) High resolution image of the medium-sized nanoparticle indicated by the green box in
(
(
C), with corresponding fast Fourier transform inset showing single crystalline structure. Scale bar = 2 nm. (E and F) STEM images corresponding to
A) and (B). Scale bars = 200 and 50 nm, respectively. (G) EDX spectra collected in the regions indicated by the colored circles in (F), corresponding to
the OMC support (orange) and a single Pt nanoparticle (purple). The spectrum acquired in the purple circle is representative of spectra taken on
numerous other particles in this region. Elemental analysis shows that the indicated particle and other particles consist of Pt, while the HRTEM (D)
shows that the nanoparticles are single crystalline, and are thus reduced metal Pt.
Figure 6. High-resolution XPS spectra. (A) Pt 4f, phenyl derivatives; and (B) Pt 4f, chloro derivatives of the complexes. (C) N 1s, comparing chloro,
phenyl, and non-platinum versions. The peaks at 398.6 and 399.4 eV are due to uncoordinated bipyridine moieties; peak at 400.6 eV, clearly visible in the
purple and pink spectra corresponding to Pt-containing chloro and phenyl versions, is attributed to the N−Pt coordination. The control made without
platinum (blue) shows only negligible peak intensity at 400.6 eV.
through these measurements. As seen from Figure 6A, the
high-resolution Pt 4f spectra of all three phenyl versions of the
OMC supported single-site catalysts showed a predominant peak
corresponding to the Pt 4f7/2 component positioned at 73.0 eV,
indicating that Pt was in the +2 oxidation state. Another peak
positioned at 75.0 eV was also observed in these spectra, which
may be attributed to the excess Pt precursor ([PtPh (μ-SEt )] )
2 2 2
adsorbed onto the surface of OMC. Likewise, the high-resolution
Pt 4f spectra obtained from three chloro derivatives of the
OMC supported complexes revealed a major peak positioned at
71.4 eV (Figure 6B). This lower binding energy peak is distinctive
to PtCl moieties and has been attributed to a higher charge
2
H
DOI: 10.1021/jacs.5b06392
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
accumulation on the Pt in the PtCl as compared to PtPh ,
maintaining Pt in the +2 oxidation state, which is also consistent
catalysts are responsible for the methane oxidation as opposed to
Pt nanoparticles.
2
2
with the notion that heterolytic Pt−Cl cleavage is a viable pathway
53
for methane C−H activation (see above).
CONCLUSIONS
■
Furthermore, it was evident from the XPS data that the
activities of the OMC-based catalysts were dependent on several
factors such as the quantity of bipyridine moieties coordinated
to Pt(II), the atomic concentration of Pt and N and finally,
the atomic concentration of Cl in the catalyst sample. In the
deconvoluted high-resolution N 1s spectra of these catalyst
samples (Figure 6C), two main peaks were observed. The peaks
positioned at 398.6 and 399.4 eV were attributed to the
uncoordinated bipyridine moieties on the surface of mesoporous
carbon. The peak located at 400.6 eV was attributed to the
N−Pt bonds, confirming the formation of the complex on OMC
The strategy of pairing molecular catalysts with OMC is a viable
approach for low temperature electrochemical methane
oxidation. Hence, molecular Pt catalysts, which have been
shown to thermally activate C−H bonds for chemical trans-
34−36
formations,
can be tethered to conductive OMC support to
perform similar C−H activation in an electrochemical environ-
ment. Given this initial success, expanding the studies to new
molecular catalysts tethered to OMC is promising; improving
methane conversion from the conversion calculated from the
optimized catalyst (8 × 10 ) would be an initial starting point.
Still, engineering and chemical challenges need to be addressed
including the poor solubility of methane and sorption in the
fuel cell along with enhancing the stability of the supported Pt
catalysts. Additional studies are underway to improve GDE
conductivity, optimize humidity levels, and identify more active
catalysts.
−3
5
4−56
support.
These data, along with the atomic concentration
of Pt and N, lend evidence for the role of molecular catalysts in
the activation of methane in a PEM fuel cell. For example, the
percent atomic concentrations of Pt and N in the samples OMC-
4
Bp-Pt-Cl were 4 and 3.2, respectively (Table S2, entry 5).
2
The highest catalytic activity in the DMEFC (278 μW/mg Pt)
was observed when the major component of the N 1s peak was
located at 400.6 eV, which was attributed to the coordinated
nitrogen moieties (Figure 6C). On the contrary, in the phenyl
ASSOCIATED CONTENT
Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/jacs.5b06392.
■
*
S
version of this catalyst, OMC-4Bp-Pt-Ph , the percent atomic
2
concentrations of Pt and N were observed to be 1.3 and 4.1
Additional experimental procedures, materials character-
ization data and fuel cell optimization details (PDF)
(Table S2, entry 2), respectively, implying a large excess of
nitrogen compared to platinum. This was also reflected in
the high-resolution N 1s spectra which, in addition to a peak at
AUTHOR INFORMATION
Corresponding Authors
tbg7h@eservices.virginia.edu
*aherring@mines.edu
*btrewyn@mines.edu
Author Contributions
M.J. and V.N. have contributed equally.
4
00.6 eV due to coordinated species, also showed a substantial
amount of uncoordinated species located at 398.8 and 399.4 eV
Figure 6C). Consequently, the electrocatalytic activity in the
■
(
*
DMEFC was significantly lower (101 μW/mg Pt), presumably
due to the small amount of complex on the surface of OMC. These
results further validate the role of OMC-tethered molecular
catalysts in the activation of methane in fuel cells.
Finally, it was observed that another contributing factor
toward the activity in DMEFC was the quantity of Cl in the
⊥
Notes
The authors declare no competing financial interest.
catalyst materials. For example, the OMC-4Bp-Pt-Cl catalyst
2
had the highest atomic concentration of Cl (Table S2, 1.7 at %)
and demonstrated the highest catalytic activity (278 μW/mg Pt).
Importantly, this sample also had the highest amount of nitrogen
coordinated with Pt(II). The N/Pt ratio was 0.8 and was the
highest among the chloro derivatives. In comparison, the OMC-
ACKNOWLEDGMENTS
■
All authors would like to acknowledge funding by the Center for
Catalytic Hydrocarbon Functionalization, an Energy Frontier
Research Center (EFRC) funded by the U.S. Department of
Energy (DOE), Office of Science, Office of Basic Energy
Sciences, under award DE-SC0001298. TSG acknowledged AES
for support through an AES Graduate Fellowship in Energy
Research. The authors acknowledge the use of instruments at the
Electron Imaging Center for NanoMachines supported by NIH
phen-Pt-Cl sample had significantly less Cl (Table S2, 0.9 at %)
2
and a much smaller N/Pt ratio (0.4), which implies a lower
quantity of phenanthroline coordinated Pt moieties. This material
demonstrated lower activity (109 μW/mg Pt) than other
catalysts in the OMC-Pt series. The sample showing the highest
atomic concentration of platinum did not demonstrate the
highest electrocatalytic activity (Table S2, OMC-phen-Pt-Cl2,
(
1S10RR23057 to ZHZ) and CNSI at UCLA. The authors
would like to thank BASF for the generous gift of the templating
agent and 3M for the generous gift of the ionomer.
5.1 atom %), nor did the sample containing the highest atomic
concentration of nitrogen (OMC-4Bp-Pt-Ph , 4.1 atom %, most
2
REFERENCES
of it was in the uncoordinated state due to the low platinum
atomic concentration), which underscores the importance of
specific identity of the nitrogen-ligated platinum for the C−H
activation of methane in the DMEFC. The OMC bound ligands
that were used as controls, OMC-4Bp, OMC-6Bp, and OMC-
phen, did not show the presence of Pt in their spectra and
showed negligible current densities in DMEFCs. In general,
XPS studies reveal that a near 1:1 ratio of Pt/N gave the highest
power output. These data supports the fact that molecular
■
(1) Kerr, R. A. Science (Washington, DC, U. S.) 2010, 328, 1624−1626.
2) Lee, J.-J.; Myung, J.-H.; Ko, H. J.; Hyun, S.-H. ECS Trans. 2009, 25,
143−1148.
3) Murray, E. P.; Tsai, T.; Barnett, S. A. Nature (London, U. K.) 1999,
00, 649−651.
4) Putna, E. S.; Stubenrauch, J.; Vohs, J. M.; Gorte, R. J. Langmuir
(
1
(
4
(
1
995, 11, 4832−7.
(5) Wang, Z.; Weng, W.; Cheng, K.; Du, P.; Shen, G.; Han, G. J. Power
Sources 2008, 179, 541−546.
I
DOI: 10.1021/jacs.5b06392
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
(
6) Shin, T. H.; Ida, S.; Ishihara, T. J. Am. Chem. Soc. 2011, 133,
(40) Tessonnier, J. P.; Villa, A.; Majoulet, O.; Su, D. S.; Schlogl, R.
Angew. Chem., Int. Ed. 2009, 48, 6543−6546.
1
(
9399−19407.
7) Niedrach, L. W. J. Electrochem. Soc. 1962, 109, 1092−96.
(41) Kruk, M.; Jaroniec, M. Chem. Mater. 2001, 13, 3169−3183.
(42) Zhu, H. J.; Ziegler, T. J. Organomet. Chem. 2006, 691, 4486−4497.
(43) Kua, J.; Xu, X.; Periana, R. A.; Goddard, W. A. Organometallics
2002, 21, 511−525.
(
8) Ferrell, J. R., III; Sachdeva, S.; Strobel, T. A.; Gopalakrishan, G.;
Koh, C. A.; Pez, G.; Cooper, A. C.; Herring, A. M. J. Electrochem. Soc.
012, 159, B371−B377.
9) Arndtsen, B. A.; Bergman, R. G. Science (Washington, DC, U. S.)
995, 270, 1970−3.
10) Holtcamp, M. W.; Labinger, J. A.; Bercaw, J. E. J. Am. Chem. Soc.
2
(
1
(
(44) Kamat, A.; Huth, A.; Klein, O.; Scholl, S. Fuel Cells 2010, 10, 983−
9
92.
(
45) O’Hayre, R., Cha, S., Colella, W., Prinz, F. Fuel Cell Fundamentals;
John Wiley & Sons, Inc.: Hoboken, NJ, 2006; pp 137−168.
46) Cooper, K. R.; Ramani, V.; Fenton, J. M.; Kunz, H. R.
1
(
(
(
997, 119, 848−849.
(
11) Jones, W. D.; Feher, F. J. Acc. Chem. Res. 1989, 22, 91−100.
12) Labinger, J. A. J. Mol. Catal. A: Chem. 2004, 220, 27−35.
13) Lersch, M.; Tilset, M. Chem. Rev. (Washington, DC, U. S.) 2005,
Experimental Methods and Data Analyses for Polymer Electrolyte Fuel
Cells, 1.2 ed.; Scribner Associates, Inc.: Southern Pines, NC, 2005.
(
47) Ferrell, J. R.; Kuo, M. C.; Turner, J. A.; Herring, A. M. Electrochim.
Acta 2008, 53, 4927−4933.
48) Song, C. J.; Tang, Y. H.; Zhang, J. L.; Zhang, J. J.; Wang, H. J.;
Shen, J.; McDermid, S.; Li, J.; Kozak, P. Electrochim. Acta 2007, 52,
552−2561.
49) Benvenutti, E. V.; Franken, L.; Moro, C. C.; Davanzo, C. U.
Langmuir 1999, 15, 8140−8146.
50) Hill, G. S.; Rendina, L. M.; Puddephatt, R. J. J. Chem. Soc., Dalton
Trans. 1996, 1809−1813.
51) Yamamoto, T.; Shido, T.; Inagaki, S.; Fukushima, Y.; Ichikawa, M.
J. Phys. Chem. B 1998, 102, 3866−3875.
52) McKeown, B. A.; Foley, N. A.; Lee, J. P.; Gunnoe, T. B.
Organometallics 2008, 27, 4031−4033.
53) Aw, B. H.; Looh, K. K.; Chan, H. S. O.; Tan, K. L.; Hor, T. S. A. J.
Chem. Soc., Dalton Trans. 1994, 3177−82.
54) Artyushkova, K.; Kiefer, B.; Halevi, B.; Knop-Gericke, A.; Schlogl,
R.; Atanassov, P. Chem. Commun. (Cambridge, U. K.) 2013, 49, 2539−
541.
55) Ferragina, C.; Massucci, M. A.; Mattogno, G. J. Inclusion Phenom.
Mol. Recognit. Chem. 1989, 7, 529−36.
56) Onoa, G. B.; Moreno, V.; Font-Bardia, M.; Solans, X.; Perez, J. M.;
Alonso, C. J. Inorg. Biochem. 1999, 75, 205−212.
1
(
05, 2471−2526.
14) Periana, R. A.; Taube, D. J.; Gamble, S.; Taube, H.; Satoh, T.;
(
Fujii, H. Science (Washington, DC, U. S.) 1998, 280, 560−564.
(
(
15) Shilov, A. E.; S, G. B. Russ. Chem. Rev. 1987, 56, 442−464.
16) Bar-Nahum, I.; Khenkin, A. M.; Neumann, R. J. Am. Chem. Soc.
2
(
2
(
004, 126, 10236−10237.
17) Groothaert, M. H.; Smeets, P. J.; Sels, B. F.; Jacobs, P. A.;
Schoonheydt, R. A. J. Am. Chem. Soc. 2005, 127, 1394−1395.
18) Labinger, J. A.; Herring, A. M.; Bercaw, J. E. J. Am. Chem. Soc.
990, 112, 5628−5629.
19) Joglekar, M.; Pylypenko, S.; Otting, M. M.; Valenstein, J. S.;
Trewyn, B. G. Chem. Mater. 2014, 26, 2873−2882.
20) Brushett, F. R.; Thorum, M. S.; Lioutas, N. S.; Naughton, M. S.;
Tornow, C.; Jhong, H.-R.; Gewirth, A. A.; Kenis, P. J. A. J. Am. Chem. Soc.
010, 132, 12185−12187.
21) Ward, A. L.; Elbaz, L.; Kerr, J. B.; Arnold, J. Inorg. Chem. 2012, 51,
694−4706.
22) Annen, S. P.; Bambagioni, V.; Bevilacqua, M.; Filippi, J.;
(
(
1
(
(
(
(
(
2
(
4
(
(
2
(
Marchionni, A.; Oberhauser, W.; Schonberg, H.; Vizza, F.; Bianchini,
C.; Gritzmacher, H. Angew. Chem., Int. Ed. 2010, 49, 7229−7233.
(
(
23) Bevilacqua, M.; Bianchini, C.; Marchionni, A.; Filippi, J.; Lavacchi,
A.; Miller, H.; Oberhauser, W.; Vizza, F.; Granozzi, G.; Artiglia, L.;
Annen, S. P.; Krumeich, F.; Gruetzmacher, H. Energy Environ. Sci. 2012,
5
, 8608−8620.
24) Valenstein, J. S.; Kandel, K.; Melcher, F.; Slowing, I. I.; Lin, V. S.
Y.; Trewyn, B. G. ACS Appl. Mater. Interfaces 2012, 4, 1003−1009.
25) Monro, S.; Scott, J.; Chouai, A.; Lincoln, R.; Zong, R.; Thummel,
R. P.; McFarland, S. A. Inorg. Chem. 2010, 49, 2889−2900.
26) Steele, B. R.; Vrieze, K. Transition Met. Chem. (Dordrecht, Neth.)
(
(
(
1
(
(
977, 2, 140−4.
27) Hodges, K. D.; Rund, J. V. Inorg. Chem. 1975, 14, 525−8.
28) Giffin, G. A.; Haugen, G. M.; Hamrock, S. J.; Di Noto, V. J. Am.
Chem. Soc. 2013, 135, 822−834.
29) Ferrell, J. R.; Kuo, M. C.; Herring, A. M. J. Power Sources 2010,
95, 39−45.
30) Sanchez-Gonzalez, J.; Macias-Garcia, A.; Alexandre-Franco, M. F.;
Gomez-Serrano, V. Carbon 2005, 43, 741−747.
31) Shilov, A. E.; Shul’pin, G. B. Chem. Rev. (Washington, DC, U. S.)
997, 97, 2879−2932.
32) Johansson, L.; Ryan, O. B.; Tilset, M. J. Am. Chem. Soc. 1999, 121,
974−1975.
33) Johansson, L.; Tilset, M.; Labinger, J. A.; Bercaw, J. E. J. Am. Chem.
Soc. 2000, 122, 10846−10855.
34) McKeown, B. A.; Gonzalez, H. E.; Friedfeld, M. R.; Gunnoe, T. B.;
Cundari, T. R.; Sabat, M. J. Am. Chem. Soc. 2011, 133, 19131−19152.
35) McKeown, B. A.; Gonzalez, H. E.; Gunnoe, T. B.; Cundari, T. R.;
Sabat, M. ACS Catal. 2013, 3, 1165−1171.
36) McKeown, B. A.; Prince, B. M.; Ramiro, Z.; Gunnoe, T. B.;
Cundari, T. R. ACS Catal. 2014, 4, 1607−1615.
37) Ji, X.; Lee, K. T.; Holden, R.; Zhang, L.; Zhang, J.; Botton, G. A.;
Couillard, M.; Nazar, L. F. Nat. Chem. 2010, 2, 286−293.
38) Joo, S. H.; Choi, S. J.; Oh, I.; Kwak, J.; Liu, Z.; Terasaki, O.; Ryoo,
R. Nature (London, U. K.) 2001, 414, 470.
39) Nam, J.-H.; Jang, Y.-Y.; Kwon, Y.-U.; Nam, J.-D. Electrochem.
Commun. 2004, 6, 737−741.
(
1
(
(
1
(
1
(
(
(
(
(
(
(
J
DOI: 10.1021/jacs.5b06392
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX