Tuning the Reactivity of Cofacial Porphyrin Prisms for Oxygen Reduction Using Modular Building Blocks

Article abstract of DOI:10.1021/jacs.0c11895

We assembled eight cofacial porphyrin prisms using MTPyP (M = Co(II) or Zn(II), TPyP = 4-tetrapyridylporphyrin) and functionalized ruthenium-based "molecular clips"using coordination-driven self-assembly. Our approach allows for the rapid synthesis of these architectures in isolated yields as high as 98% for the assembly step. Structural and reactivity studies provided a deeper understanding of the role of the building blocks on the oxygen reduction reaction (ORR). Catalytic efficacy was probed by using cyclic and hydrodynamic voltammetry on heterogeneous catalyst inks in aqueous media. The reported prisms showed outstanding selectivity (>98%) for the kinetically hindered 4e-/4H+ reduction of O2 to H2O over the kinetically more accessible 2e-/2H+ reduction to H2O2. Furthermore, we have demonstrated significant cofacial enhancement in the observed catalytic rate constant ks (μ5 orders of magnitude) over the mononuclear analogue. We conclude that the steric bulk of the clip plays an important role in the structural dynamics of these prisms, which in turn modulates the ORR reactivity with respect to selectivity and kinetics.

Full text of DOI:10.1021/jacs.0c11895

pubs.acs.org/JACS
Article
Tuning the Reactivity of Cofacial Porphyrin Prisms for Oxygen
Reduction Using Modular Building Blocks
Matthew R. Crawley, Daoyang Zhang, Amanda N. Oldacre, Christine M. Beavers, Alan E. Friedman,
and Timothy R. Cook*
Cite This: J. Am. Chem. Soc. 2021, 143, 1098−1106
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: We assembled eight cofacial porphyrin prisms using MTPyP (M
= Co(II) or Zn(II), TPyP = 4-tetrapyridylporphyrin) and functionalized
ruthenium-based “molecular clips” using coordination-driven self-assembly. Our
approach allows for the rapid synthesis of these architectures in isolated yields as
high as 98% for the assembly step. Structural and reactivity studies provided a
deeper understanding of the role of the building blocks on the oxygen reduction
reaction (ORR). Catalytic efficacy was probed by using cyclic and hydrodynamic
voltammetry on heterogeneous catalyst inks in aqueous media. The reported
prisms showed outstanding selectivity (>98%) for the kinetically hindered 4e−/
4H+ reduction of O2 to H2O over the kinetically more accessible 2e−/2H+
reduction to H2O2. Furthermore, we have demonstrated significant cofacial
enhancement in the observed catalytic rate constant ks (∼5 orders of magnitude)
over the mononuclear analogue. We conclude that the steric bulk of the clip
plays an important role in the structural dynamics of these prisms, which in turn modulates the ORR reactivity with respect to
selectivity and kinetics.
face”, orientation. This allows for multiple metal centers to
interact with one another and substrates.12−17 Initial interest in
these cofacial systems stemmed from the structural similarities
to the active site in cytochrome c oxidase and was therefore
interrogated for O2 reduction chemistry.14,15,18 More recently,
there have been reports of cofacial porphyrin systems carrying
out other small molecule activations such as CO2 reduc-
tion.19−21 These dinuclear catalysts have shown considerable
“cofacial enhancements” on both reaction kinetics and
selectivity.15,22 Despite their impressive reactivity, cofacial
porphyrin prisms have, to date, been predominantly linked
through organic tethers (i.e., dibenzofuran, biphenylene,
anthracene, etc.) via covalent bonds (C−C or amide
bonds).23−25 This poses a significant synthetic challenge as
the porphyrin rings must be either synthesized directly off of
the “clipping” group or specially designed porphyrins
synthesized through complex, and oftentimes low yielding,
syntheses. As a result, it is difficult to generate a library of
catalysts, and consequently, there have been few systematic
studies of these cofacial systems where key parameters such as
INTRODUCTION
■
Growing global energy consumption, in tandem with
increasing initiatives to shed dependence on carbon-based
fuels, has motivated the study of small-molecule activations
and other energy-relevant transformations.1 Our aim is to
develop an understanding of the oxygen reduction reaction
(ORR) catalyzed by polynuclear molecules with an emphasis
on the factors that govern parameters such as Faradaic
efficiencies, observed standard rate constants (ks), and
overpotentials (η). Much of our efforts are inspired by natural
systems that necessarily use self-assembly to enforce exacting
control over physical and electronic structure to influence
these factors.
Many metalloenzymes feature polynuclear active sites such
as the Ni−Fe sulfur cluster in carbon monoxide dehydrogenase
(CODH),2 MoFe nitrogenases,3 and the Fe-heme/Cu-His
active site of cytochrome c oxidase.4 Furthermore, there is an
ever-growing number of synthetic polynuclear catalysts for a
variety of transformations.5−10 Multiple metal centers allow for
the spread of coordination and redox demand. This is of
particular importance for small-molecule (O2, N2, CO2, etc.)
transformations which are often multiproton, multielectron
processes. When multiple redox pathways are possible, careful
control over electron and proton inventories is necessary to
achieve selectivity.11
Received: November 12, 2020
Published: December 30, 2020
Over the past four decades, there has been interest in
tethering multiple porphyrin groups in a cofacial, or “face-to-
https://dx.doi.org/10.1021/jacs.0c11895
J. Am. Chem. Soc. 2021, 143, 1098−1106
© 2020 American Chemical Society
1098

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
metal−metal separation and/or electronic structure have been
varied. Our lab has overcome this challenge through the use of
coordination-driven self-assembly26−35 wherein two metal-
loporphyrins decorated with Lewis basic donors may combine
with four so-called “molecular clips” that serve as spacers.
Porphyrin-based systems are particularly amenable to coordi-
nation-driven self-assembly due to their rigid structure and
well-defined angularity.9,10,14,36−40
We hypothesized that the electron-withdrawing/donating
nature of the molecular clips used in self-assembly may alter
overpotentials and electron transfer rates, resulting in differ-
ences to reactivity and selectivity for ORR. Coordination-
driven self-assembly is a unique and powerful way to modify
distal sites since the modifications we have made to the clips
do not interfere with the assembly process. In this work we
report the synthesis, characterization, and study of catalytic
reactivity of eight new functionalized cofacial porphyrin prisms
(Figure 1). Functional groups were appended to the molecular
include an investigation of the physical structure based on X-
ray diffraction and related discussions of whether molecular
clips play a steric role in addition to tuning electronic structure.
We emphasize that this distal tuning is enabled by
coordination-driven self-assembly where we avoid the need
to redesign the synthesis of each new prisms based on a
conventional stepwise approach.
RESULTS AND DISCUSSION
■
The molecular clips used for assembly reactions were
synthesized by using an adapted literature procedure as
shown in ref 41. Metalation of the clips used [Ru(η6-p-
cymene)μ2-Cl]2Cl2. The dinuclear complexes were readily
isolated via recrystallization with the exception of R = −Me, in
which case column chromatography was required.
Clip formation was confirmed, and purity assessed, by using
1
a combination of H NMR, mass spectrometry, and FT-IR
spectroscopy (when discussing molecular clips, the following
naming convention will be used: Ru-R group clip, e.g., Ru-CF3
clip). In the case the Ru-CF3 clip, single-crystal X-ray
diffraction was used to determine the solid-state molecular
structure (Figure 2). Further details of the syntheses,
isolations, and characterizations of the molecular clips may
be found in the Supporting Information.
Figure 1. Modular nature of self-assembly allows for more
streamlined synthesis of functionalized catalysts.
Figure 2. Solid-state structure of Ru-CF3 clip determined by single-
crystal X-ray diffraction. Thermal ellipsoids are set at the 50%
probability level. Primed labels are used to denote symmetry-
equivalent atoms. Hydrogen atoms and chloroform solvent molecules
are omitted for clarity.
a
Scheme 1. Synthesis of Molecular Clips
Coordination-driven self-assembly of the prisms was
performed via a salt metathesis of the ruthenium molecular
clip with a slight excess of silver triflate (AgOTf) and
subsequent treatment with the desired porphyrin (either
ZnTPyP or CoTPyP) at reflux in methanol. Although all of
the prisms formed with high yields (lowest, 81%) for some
combinations, the assembly process was near quantitative, with
isolated yields as high as 98%. Reaction times varied48 h for
assemblies using ZnTPyP and 96 h for those using CoTPyP.
This disparity is a result of the solubility differences between
the two porphyrins; ZnTPyP exhibits higher solubilities in the
organic solvents used as assembly media (CH2Cl2, acetonitrile,
and methanol) than that of its cobalt counterpart. The
following naming convention is used for the prisms: porphyrin
metal−molecular clip−R group prism, e.g., Zn-Tolyl prism).
Isolation of the green (Zn) or red (Co) self-assembled prisms
was straightforward: precipitation and collection by centrifu-
gation followed by drying in vacuo (see the Supporting
Information for more details). Zn-based assemblies were
a
[Ru] = [Ru(η6-p-cymene)μ2-Cl]2Cl2, reflux for 24 h.
clips (Scheme 1) which were used to tether two tetra-4-
pyridylporphyrin (TPyP) groups together to form the cofacial
prisms. Specifically, the 4-position of N,N-phenyloxamido-
bridged Ru-based clips was modified with −Me, −H, −Cl, and
−CF3 groups. To the extent that these distal changes do not
significantly affect substrate access, it becomes possible to
isolate the role of the clip on catalysis. Toward that end, we
1099
https://dx.doi.org/10.1021/jacs.0c11895
J. Am. Chem. Soc. 2021, 143, 1098−1106

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
synthesized to act as diamagnetic, catalytically inactive
analogues to the Co-based assemblies for NMR and electro-
chemical experiments (vide infra).
One-dimensional NMR (1H and 19F{1H} in the case of the
Zn/Co-CF3 prisms, see Figures S11−S19) was used for
solution-state structure conformation, and diffusion ordered
spectroscopy (DOSY) NMR experiments were used to rule out
the formation of coordination polymers and/or smaller
1
oligomeric species in solution. The peaks in the H NMR
spectra of the Co-porphyrin prisms were broad and showed
shifts consistent with the presence of paramagnetic d7 Co(II)
centers. For the diamagnetic Zn analogues, the NMR spectra
are more standard and exhibit three major regions: the first
being from ∼10 to 7 ppm. Pyridyl-, pyrolic-, and aromatic
oxamido-proton resonances comprise this region. Second, from
∼6.5 to 5 ppm, are the resonances from the protons of the p-
cymene group. Finally, the aliphatic region (<3 ppm) contains
the resonances of the p-cymene isopropyl and methyl groups as
well as the methyl group of the Ru-Tolyl clip. When the
spectra were integrated, the relative number of protons for
each region was consistent with assemblies containing twice
the number of molecular clips as porphyrins. This ratio can be
narrowed down to specifically a 4 + 2 assembly by mass
spectrometry, explained later. Splitting of the α- and β-protons
of the pyridyl group into complex multiplets rather than sharp
doublets as in the free porphyrin suggests helical chirality
caused by tilting of the molecular clips, as observed by
Therrien and co-workers.37 The complex splitting pattern can
be attributed to the diastereomers that may exist for these
prisms, as further discussed in the crystallographic section.
DOSY NMR supports the formation of uniform-sized species
in solution rather than a mixture of oligomers/fragments of
assembly. All peaks that were attributed to the prism were
associated with the same diffusion coefficient, and each of the
prisms exhibited similar diffusion coefficients. For the Zn-CF3
prism, 19F{1H} NMR further affirmed assembly. The −CF3
resonance shifted from the free clip (−61.98 ppm) to the
assembly (−63.70 ppm). Furthermore, the peak became
significantly broader, suggesting the formation of a larger
molecule with slowed tumbling. The relative integrations of the
triflate singlet (−79.22 ppm) to the −CF3 confirms the
expected 1:1 ratio.
Self-assembly stoichiometry and charge state were confirmed
by electrospray ionization Fourier transform ion cyclotron
resonance mass spectrometry (ESI-FT-ICR MS). Figure 3
shows the mass spectrum of the Zn-Tolyl prism as an example
of the type of data obtained for these assemblies. For most
prisms, the parent ion peak was either a 3+ or 4+ peak,
corresponding to the intact prism which was ionized by the
loss of three or four outer-sphere triflate counterions,
respectively. Furthermore, in some cases the 2+ and 5+ intact
prism peaks were also observed. Stoichiometry and charge
state confirmation were made on the basis of the excellent
agreement between the experimental mass spectrum and the
simulated pattern, in terms of both isotopic intensities and
charge spacings. This affirms that the desired assemblies were
present in the isolated powders and persist in solution. Lower
mass-to-charge ratio peaks were consistent with fragmentation
of the desired assembly, that is, the [A3D2 + 2OTf]4+ peak (see
Figure 3) which corresponds to the loss of a single molecular
clip from the assembly. In general, the same peak patterns are
observed for both Zn- and Co-metalated prisms. Furthermore,
Figure 3. ESI-FT-ICR mass spectrum of the Zn-Tolyl prism. The
bold peaks correspond to the intact prism. Top: full mass spectrum.
Bottom: zoom in on the 4+ peak of the intact prism.
the m/z values are consistent with a Co(II) species
(Supporting Information).
The vibrational spectra of the prisms are dominated by
signals due to the triflate counterions at ca. 1255, 1025, and
635 cm−1; these correspond to the C−F and S−O stretches
and −SO3 group deformation, respectively.42 This is consistent
with other reported Ru-clipped porphyrin prisms.36 Addition-
ally, there is no evidence of the N−H stretch associated with
the starting oxamide ligand. Comparison of the Co prism to its
Zn-analogue reveals near identical IR signatures, further
1100
https://dx.doi.org/10.1021/jacs.0c11895
J. Am. Chem. Soc. 2021, 143, 1098−1106

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Figure 4. Crystal structure of Zn-CF3 prism. Hydrogen atoms, axial bound water, and OTf− counterions have been omitted for clarity.
affirming the structural similarities between the two. Electronic
absorption spectroscopy reveals the expected Soret band and
Q-bands, indicative of a metalated porphyrin. The Co prism
Soret bands appeared around 415−420 nm, consistent with
other Co(II)-porphyrin complexes in the literature (Co(III)
porphyrins typical display Soret bands at ∼435−450 nm).43,44
Examining the Q-bands, the Zn prisms exhibit two distinct,
well-resolved Q-bands, while the Co prisms seem to show a
single Q-band (Supporting Information). This agrees with
other cobalt porphyrins in the literature and suggests the
geometry at the metal center is near ideal D4h point symmetry
for cobalt-containing prisms and more distorted from square
planar for the Zn prisms.44
X-ray diffraction experiments were used to confirm the
molecular geometry of both the Ru-CF3 clip and the Zn-CF3
prism. The Ru-CF3 clip presented the expected geometry; the
two inner-sphere chloro ligands in an anti-configuration
suggest that in solution an equilibrium between syn- and
anti-geometries exists. The packing reveals a layered structure
with layers of the Ru-CF3 clip separated by cocrystallized
chloroform. Crystals of the Zn-CF3 prism were small and
poorly diffracting on lab source instruments; data for the prism
were collected at Berkeley National Laboratories by using the
Advanced Light Source synchrotron. Diffraction beyond dmin of
1.00 Å was not achievable; however, the data were more than
sufficient for atomic connectivity. The structure of Zn-CF3 was
easily recognized in the density map and was quite well
behaved given the data quality. The p-cymene ligands were
difficult to model, and thus their geometries were refined
against coordinates produced from DFT structure optimization
of free p-cymene (now available in Ilia’s Fragment Library).45
Careful examination of the structures reveals the stereo-
chemical complexity associated with these architectures. First,
the tilting of the molecular clip persists in the solid state. This
gives rise to M and P helicity; the M enantiomer is shown in
Figure 5. However, Zn-CF3 crystallizes in the centrosymmetric
space group C2/c, and therefore both enantiomers must exist
in the crystalline phase. Once a part of the assembly, the
molecular clips possess idealized C2 point symmetry; this is
purely rotational and therefore chiral. As a result, there are two
orientations the clips can bind to the porphyrin (see Figure 5).
Specifically, it is the relative orientations of the oxamido phenyl
rings which result in these diastereomers. This leads to four
Figure 5. Symmetry analysis of conformers of functionalized cofacial
porphyrin prisms. (A) The two possible orientations the molecular
clip can bind to the TPyP. (B) The four possible diastereomers due to
the orientation of the phenyl groups and their associated Schoenflies
symbols.
possible diastereomers illustrated in Figure 5, along with the
idealized symmetry of each prism. The ideal C2 symmetric
prism was crystallographically observed; there are several
important consequences of this observation: First, the mixture
of diastereomers explains the very complex splitting in the
NMR spectra; different diastereomers will have different
chemical shifts. The complexity of the observed NMR spectra
suggests that all four configurations are present in solution.
The relative contribution of each diastereomer is at this time
unknown because of the overlap of the peaks that makes
integrations nonspecific, and the factors influencing the
1101
https://dx.doi.org/10.1021/jacs.0c11895
J. Am. Chem. Soc. 2021, 143, 1098−1106

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
distribution are a point of ongoing research in our lab. Variable
temperature NMR spectra at 50 °C do not show substantial
changes relative to those collected at room temperature (see
the Supporting Information). The lack of change implies
different configurations (e.g., D4, C2, C1, and D2) are not
interconvertible on the NMR time scale under conditions
probed, indicating that the splitting is due to the different
configurations, not helical chirality. The mixture of diaster-
eomers can also be used to rationalize the small, poorly
diffracting crystals. Assuming that one diastereomer does not
dominate, the effective concentration of any one will be low
during crystallization, and similar diastereomers may crystallize
together, leading to defects and disorder, which dramatically
reduce periodicity. Nonetheless, we can extract a metal−metal
separation; in this case, the Zn−Zn distance was determined to
be ca. 4.5 Å. This is slightly shorter than previous self-
assembled prisms but in line with the Fe−Cu separation in
cytochrome c oxidase.10,46 Both zinc centers were found to
have an axially bound water molecule; this supports the
distorted symmetry of the Zn center predicted by the Q-bands
in the UV−vis spectrum. The homogeneous electrochemical
behavior of the cofacial prisms was assessed by using cyclic
voltammetry in acetonitrile. Under a nitrogen atmosphere, Zn-
and Co-metalated prisms exhibited similar cyclic voltammo-
grams (CVs); however, the cobalt prisms in general showed
two reduction events (ca. −1 V vs Fc+/0) before a third current
response at around −1.4 V. These two events have been
attributed to the two Co(II/I) couples for each of the metal
centers, in agreement with other Co(II/I) couples.47 Two
distinct Co(II/I) couples imply some degree of electronic
communication between the two metal centers. The remaining
reductive events at ca. −1.4 V and potentials more negative are
attributed to clip centered reductions (see Figure 6). The most
electron-rich clip (-tolyl) shows the most negative reduction
potential while the most electron-deficient clip shows the most
positive.
chemistry. Sparging the solution with O2 did not lead to any
appreciable change in current response within the reduced
electrochemical window (reversible superoxide formation at
the glassy carbon electrode limited negative potential scans).
With the addition of trifluoroacetic acid (TFA) to the
electrochemical cell, a large catalytic wave was observed (see
Figure 7). To rule out proton reduction, a solution was sparged
Figure 7. CV of Co-Chloro prism under variable conditions. Blank
(purple, N2 atmosphere), catalyst under inert conditions (blue, 0.1
mM Co-Chloro, N2 atmosphere), catalyst with O2 (red, 0.1 mM Co-
Chloro, O2 atmosphere), and catalyst under catalytic conditions
(green, 0.1 mM Co-Chloro, O2 atmosphere, 100 mM TFA). In all
cases, dry acetonitrile with 100 mM TBAPF6 was used.
with N2 and TFA was added; no current response was
observed at the same potential with this absence of oxygen. To
confirm that catalysis occurs at the Co centers and not at the
Ru clips, the same CV experiments were conducted on the Zn-
analogues with no observable catalytic response. All four of the
Co-containing prisms showed similar current responses at
similar overpotentials (Supporting Information). To compare
overpotentials, the Ecat/2 method of Appel was used.48 The
Ecat/2 values showed very little dependence on the identity of
the functionalized-clip R-group, all showing a Ecat/2 of ca. −0.3
V vs Fc+/0. Comparing the Ecat/2 values against the Hammett
substituent constants does not show any apparent trend (R2 of
0.38). Given the absence of a linear free energy relationship,
we can conclude that the electronic nature of the clip does not
appreciably tune the overpotentials of these prisms, though we
do note that the kinetics and selectivity of these different
prisms do vary significantly. To interrogate these differences
without the possibility of complications due to ill-defined film
formation due to adsorption, heterogeneous catalyst inks were
used.
Scanning oxidizing reveals two reversible oxidation events
followed by a large irreversible wave (see the Supporting
Information). These are found in both the Co- and Zn-
containing prisms suggestive of ring- or clip-centered redox
Catalyst inks of the cofacial prisms were formed by using
reported methods.10,49 Inks were then drop-cast onto a glassy
carbon electrode, and the electrode was immersed in a 0.5 M
H2SO2 aqueous solution. CV were then acquired under an
inert atmosphere, which showed no current response. Upon
sparging with O2, a catalytic current response was observed
(Figure 8).
Comparing the Ecat/2 values under homogeneous and
heterogeneous conditions, a similar trend is observed with a
trend of (positive) tolyl > chloro > phenyl ∼ CF3- (negative).
We interpret this preserved trend as evidence that the
electronic structure and conformation of the prisms are not
significantly altered under heterogeneous conditions relative to
homogeneous conditions.
To gather further insight into the parameters governing
ORR, hydrodynamic voltammetric techniques were performed
by using a rotating ring disk electrode (RRDE). Faradaic
efficiencies for each prism were determined by comparing disk
Figure 6. CV of Co prisms illustrating reductive current responses. All
CVs were acquired at room temperature under a N2 atmosphere with
rigorously dry solvent. The prism concentration was 0.1 mM with 100
mM TBAPF6 added as supporting electrolyte. Scan rate: 100 mV/s.
Vertical lines indicate the E1/2 value of the clip-centered reduction.
1102
https://dx.doi.org/10.1021/jacs.0c11895
J. Am. Chem. Soc. 2021, 143, 1098−1106

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Table 1. Summary of Electrochemically Determined
Catalyst Benchmarks of Co-Prism Containing Catalyst Inks
Ecat/2 (V vs
Ag/AgCl)
Tafel slope
(mV/decade)
prism
% H2O2
ks (M−1 s−1
)
Co-Tolyl
Co-Phenyl
Co-Chloro
Co-CF3
0.38(1)
0.35(2)
0.37(2)
0.37(1)
0.16
1.6(1)
3.5(3)
10.7(6)
7.5(4)
50
−130(7)
−101(2)
−128(2)
−120(3)
−126
2.2(4) × 105
3.4(5) × 105
2.8(4) × 105
9(2) × 104
9.5 × 100
CoTPyP10
electron, four-proton chemistry. Under the same conditions
normalized by Co concentration, CoTPyP shows 50% H2O
production. Tafel slopes for each of the prisms were all ca.
100−130 mV/decade. Similar slopes imply that there is no
major shift in the operative mechanism due to different
molecular clips. Rate constants for the prisms were all within
an order of magnitude around 10000 times greater than that of
CoTPyP, further illustrating the value of cofacial enhance-
ments to electrocatalysts. Furthermore, the Ecat/2 values show
that these catalysts operate at, or below, the overpotential
measured for 10% Pt/C.10 These cofacial prisms are among the
most active ORR catalyst reported to date.
Figure 8. CV of Co-prism catalyst inks. All CV were acquired at room
temperature, after the 0.5 M H2SO4 solution had been sparged with
O2. Scan rate: 100 mV/s.
and ring currents to determine the amount of hydrogen
peroxide generated via eq 1.10,15,50
2iring
N
%H2O2 =
× 100
iring
N
idisk
+
(1)
where iring is the current measured at the platinum ring, idisk is
the current response at the modified glassy carbon disk (see
Figure 9), and N is the collection efficiency of the platinum
Examining all the data in concert, some fundamental
conclusions regarding self-assembled cofacial catalysts may
now be made. Although drastic shifts in mechanism and/or
reactivity were not observed, there was modulation of said
reactivity in terms of both selectivity and kinetics. These
differences in reactivity were prompted by molecular changes
made ∼1.1 nm away from the “active sites”. We were skeptical
that changing the electronic structure so far away from the
metals would influence catalysis, and in fact we did not observe
a linear free energy relationship between Hammett parameters
and overpotential under homogeneous conditions or ks under
heterogeneous conditions (Supporting Information). Because
our Tafel slopes do not indicate a mechanistic shift, these
observations together rule out that the changes in selectivities
and rates constants are purely based on electronic effects.
Instead, we surmise that variations to the clips, although
occurring spatially far from the metals, impose structural
ramifications on the two Co centers. We find evidence for
these structural changes from our crystal data: the clip
functional groups point outward, perpendicular to the Ru−
pyridyl bond vector (see Figure 4). As a result, they point
directly at one another and, spatially, are much closer than one
may predict based on an oversimplified 2D drawing. In 3D
(see Figure 5) it can be seen that the clips tilt to minimize the
steric interaction between adjacent clips. The Co-Phenyl prism
showed the highest rate constant followed by the Co-Tolyl and
Co-Chloro prisms, which have similar sized substituents, and
finally the Co-CF3 showed the lowest rate constant, following a
clean smallest-to-largest trend. It is clear that the prisms
significantly break the idealized D4 symmetric one may expect
for vertical clips spanning two offset porphyrins. As the
porphyrins twist, the molecular clips necessarily tilt, but
because of their fixed Ru−Ru separation, the porphyrins
themselves must move closer together, as illustrated in Figure
10.
Figure 9. Hydrodynamic voltammogram of the Co-Tolyl prism using
RRDE. Lower traces represent modified-disk current where catalysis is
occurring (scan rate 20 mV/s); upper traces are the current response
of the Pt-ring electrode held at a constant +1.0 V. Very little current
response was observed at the ring electrode, suggesting high
selectivity for O2 reduction to water.
ring. The collection efficiency was experimentally determined
(0.362(2)) by using the FeIII/II couple of potassium
ferricyanide (Supporting Information).51
Hydrodynamic voltammetry allows for the minimization of
mass-transport effects on the observed kinetics of electron
transfer, enabling measurements of faradaic efficiency along
with rate constants associated with electron transfer using
́
Koutecky−Levich analysis. Tafel plots were generated to
examine whether modifications to the clip lead to changes in
́
the operative mechanism, and Koutecky−Levich analysis
allowed for the determination of the standard rate constant
(ks). These results are summarized in Table 1.
There are several striking features to the data. First, the
remarkable selectivity of the Co-Tolyl prisms, forming 1.6%
H2O2 over the kinetically more challenging product, H2O, with
the remaining prisms ranging from ∼3−10%. The excellent
selectivity of this entire family of catalysts, ≥90% H2O
formation, is a testament to the cofacial design to favor four-
Whereas our solid-state structures are a snapshot of one
conformation, this twisting may be dynamic in solution, and
thus the bulky R groups of the clip play a role in Co−Co
separation, which explains how changes to the prisms over 1
nm away can actually impact selectivity and kinetics.
1103
https://dx.doi.org/10.1021/jacs.0c11895
J. Am. Chem. Soc. 2021, 143, 1098−1106

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
14260, United States; orcid.org/0000-0002-7668-8089;
Email: trcook@buffalo.edu
Authors
Matthew R. Crawley − Department of Chemistry, University
at Buffalo, The State University of New York, Buffalo, New
York 14260, United States; orcid.org/0000-0002-2555-
9543
Daoyang Zhang − Department of Chemistry, University at
Buffalo, The State University of New York, Buffalo, New York
14260, United States; orcid.org/0000-0003-3722-1336
Amanda N. Oldacre − Department of Chemistry, University at
Buffalo, The State University of New York, Buffalo, New York
14260, United States; orcid.org/0000-0001-8873-7186
Christine M. Beavers − Advanced Light Source, Lawrence
Berkeley National Laboratory, Berkeley, California 94720,
United States; orcid.org/0000-0001-8653-5513
Alan E. Friedman − Department of Materials, Design, and
Innovation, University at Buffalo, The State University of
New York, Buffalo, New York 14260, United States;
orcid.org/0000-0002-4764-8168
Figure 10. Modulating metal−metal separation through the canting
of molecular clips.
CONCLUSIONS
■
We have demonstrated that coordination-driven self-assembly
allows for the rapid population of a library of zinc- and cobalt-
containing cofacial porphyrin prisms in three high-yielding
steps. The modularity of self-assembly permitted tuning of
electronic structure and steric bulk of a family of Co-based
ORR electrocatalysts. Spectroscopic and solid-state character-
ization confirmed the anticipated cofacial porphyrin prisms
were indeed synthesized assembly yields of 81−98%. Electro-
chemical techniques were used to probe ORR reactivity with
an emphasis on correlating electronic structure to redox
parameters. Kinetics and selectivities were assessed by using
hydrodynamic voltammetry. Highly competitive standard rate
Complete contact information is available at:
https://pubs.acs.org/10.1021/jacs.0c11895
Author Contributions
M.R.C. and D.Z. contributed equally to this work.
Notes
The authors declare no competing financial interest.
́
constants were obtained by using Koutecky−Levich analysis.
Tafel slopes analysis ruled out fundamental mechanistic shift
between the Co prisms. Faradaic efficiencies revealed that the
cofacial prisms strongly favor the four-proton, four-electron
reduction to water over the kinetically more accessible two-
proton, two-electron reduction to hydrogen peroxide. Sub-
stantial cofacial enhancements can been seen in all metrics of
catalysis. The reported prisms even show reactivity that is
comparable and, in one case, superior to Pt/C as a catalyst for
ORR. Because our self-assembly approach to catalyst formation
is high-yielding, rapid, and modular, we were able to make
systematic changes to our prisms and establish that distal
modifications can have a pronounced role in selectivity and
activity. These changes arise from steric effects on the twisting
of the two porphyrin faces that results in a tuning of the Co−
Co distance. Efforts are underway to interrogate the solution-
phase dynamics of these prisms now that we have established
its importance in catalytic activity.
ACKNOWLEDGMENTS
■
SC-XRD data for Zn-CF3 were collected on the LBNL
Beamline 12.2.1. This research used resources of the Advanced
Light Source, which is a DOE Office of Science User Facility
under Contract DE-AC02-05CH11231. The authors thank
SUNY Fredonia for the use of their X-ray diffractometer. This
work was supported by NSF Career Award #1847950 (T.R.C.)
and a UB Silbert Fellowship (M.R.C.).
REFERENCES
■
(1) Lewis, N. S.; Nocera, D. G. Powering the planet: chemical
challenges in solar energy utilization. Proc. Natl. Acad. Sci. U. S. A.
2006, 103, 15729−15735.
(2) Ragsdale, S. W. Metals and Their Scaffolds To Promote Difficult
Enzymatic Reactions. Chem. Rev. 2006, 106, 3317−3337.
(3) Bjornsson, R.; Delgado-Jaime, M. U.; Lima, F. A.; Sippel, D.;
Schlesier, J.; Weyhermuller, T.; Einsle, O.; Neese, F.; DeBeer, S.
̈
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/jacs.0c11895.
Molybdenum L-Edge XAS Spectra of MoFe Nitrogenase. Z. Anorg.
Allg. Chem. 2015, 641, 65−71.
■
sı
(4) Yoshikawa, S.; Shimada, A. Reaction Mechanism of Cytochrome
c Oxidase. Chem. Rev. 2015, 115, 1936−1989.
(5) Hetterscheid, D. G. H.; Chikkali, S. H.; de Bruin, B.; Reek, J. N.
H. Binuclear Cooperative Catalysts for the Hydrogenation and
Hydroformylation of Olefins. ChemCatChem 2013, 5, 2785−2793.
(6) Tanaka, H.; Nishibayashi, Y.; Yoshizawa, K. Interplay between
Theory and Experiment for Ammonia Synthesis Catalyzed by
Transition Metal Complexes. Acc. Chem. Res. 2016, 49, 987−995.
(7) Trehoux, A.; Mahy, J.-P.; Avenier, F. A growing family of O2
activating dinuclear iron enzymes with key catalytic diiron(III)-peroxo
intermediates: Biological systems and chemical models. Coord. Chem.
Rev. 2016, 322, 142−158.
Synthetic procedure and spectroscopic characterization
data (HR-MS, NMR, IR, UV/vis), crystallographic
details, electrochemical characterization, and a detailed
discussion of catalyst performance (PDF)
Crystallographic data of Ru-CF3 (CIF)
Crystallographic data of Zn-CF3 (CIF)
AUTHOR INFORMATION
Corresponding Author
Timothy R. Cook − Department of Chemistry, University at
Buffalo, The State University of New York, Buffalo, New York
■
(8) Rej, S.; Tsurugi, H.; Mashima, K. Multiply-bonded dinuclear
complexes of early-transition metals as minimum entities of metal
cluster catalysts. Coord. Chem. Rev. 2018, 355, 223−239.
1104
https://dx.doi.org/10.1021/jacs.0c11895
J. Am. Chem. Soc. 2021, 143, 1098−1106

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
(9) Oldacre, A. N.; Friedman, A. E.; Cook, T. R. A Self-Assembled
Cofacial Cobalt Porphyrin Prism for Oxygen Reduction Catalysis. J.
Am. Chem. Soc. 2017, 139, 1424−1427.
(10) Oldacre, A. N.; Crawley, M. R.; Friedman, A. E.; Cook, T. R.
Tuning the Activity of Heterogeneous Cofacial Cobalt Porphyrins for
Oxygen Reduction Electrocatalysis through Self-Assembly. Chem. -
Eur. J. 2018, 24, 10984−10987.
(11) Zhang, R.; Warren, J. J. Controlling the Oxygen Reduction
Selectivity of Asymmetric Cobalt Porphyrins by Using Local
Electrostatic Interactions. J. Am. Chem. Soc. 2020, 142, 13426−13434.
(12) Chang, C. K. μ-Superoxodicobalt complex of a cofacial
diporphyrin. J. Chem. Soc., Chem. Commun. 1977, 800−801.
(13) Collman, J. P.; Elliott, C. M.; Halbert, T. R.; Tovrog, B. S.
Synthesis and characterization of “face-to-face” porphyrins. Proc. Natl.
Acad. Sci. U. S. A. 1977, 74, 18−22.
(14) Collman, J. P.; Denisevich, P.; Konai, Y.; Marrocco, M.; Koval,
C.; Anson, F. C. Electrode catalysis of the four-electron reduction of
oxygen to water by dicobalt face-to-face porphyrins. J. Am. Chem. Soc.
1980, 102, 6027−6036.
(15) Durand, R. R.; Bencosme, C. S.; Collman, J. P.; Anson, F. C.
Mechanistic aspects of the catalytic reduction of dioxygen by cofacial
metalloporphyrins. J. Am. Chem. Soc. 1983, 105, 2710−2718.
(16) Chang, C. J.; Deng, Y.; Heyduk, A. F.; Chang, C. K.; Nocera, D.
G. Xanthene-Bridged Cofacial Bisporphyrins. Inorg. Chem. 2000, 39,
959−966.
(17) Deng, Y.; Chang, C. J.; Nocera, D. G. Direct Observation of the
“Pac-Man” Effect from Dibenzofuran-Bridged Cofacial Bisporphyrins.
J. Am. Chem. Soc. 2000, 122, 410−411.
(18) Collman, J. P.; Anson, F. C.; Barnes, C. E.; Bencosme, C. S.;
Geiger, T.; Evitt, E. R.; Kreh, R. P.; Meier, K.; Pettman, R. B. Further
studies of the dimeric.beta.-linked “face-to-face four” porphyrin:
FTF4. J. Am. Chem. Soc. 1983, 105, 2694−2699.
(19) Mohamed, E. A.; Zahran, Z. N.; Naruta, Y. Efficient
electrocatalytic CO2 reduction with a molecular cofacial iron
porphyrin dimer. Chem. Commun. 2015, 51, 16900−16903.
(20) Mohamed, E. A.; Zahran, Z. N.; Naruta, Y. Efficient
Heterogeneous CO2 to CO Conversion with a Phosphonic Acid
Fabricated Cofacial Iron Porphyrin Dimer. Chem. Mater. 2017, 29,
7140−7150.
(21) Zahran, Z. N.; Mohamed, E. A.; Abdel Haleem, A.; Naruta, Y.
Efficient Solar-Assisted O2 Reduction Using a Cofacial Iron
Porphyrin Dimer Catalyst Integrated into a p-CuBi2O4 Photocathode.
Chem. - Eur. J. 2018, 24, 10606−10611.
(22) Chang, C. K.; Liu, H. Y.; Abdalmuhdi, I. Electroreduction of
oxygen by pillared cobalt(II) cofacial diporphyrin catalysts. J. Am.
Chem. Soc. 1984, 106, 2725−2726.
(23) Hiom, J.; Paine, J. B., III; Zapf, U.; Dolphin, D. The synthesis of
cofacial porphyrin dimers. Can. J. Chem. 1983, 61, 2220−3.
(24) Le Mest, Y.; L’Her, M.; Courtot-Coupez, J.; Collman, J. P.;
Evitt, E. R.; Bencosme, C. S. Electrochemical behaviour of a binary
cofacial dicobalt porphyrin in aprotic media under a nitrogen
atmosphere: Formation of mixed-valence compounds. J. Electroanal.
Chem. Interfacial Electrochem. 1985, 184, 331−346.
(25) Kim, K.; Collman, J. P.; Ibers, J. A. Structure of the dicobalt
“face-to-face” porphyrin with two four-atom amide bridges. J. Am.
Chem. Soc. 1988, 110, 4242−4246.
(26) Fujita, M. Metal-directed self-assembly of two- and three-
dimensional synthetic receptors. Chem. Soc. Rev. 1998, 27, 417−425.
(27) Inokuma, Y.; Arai, T.; Fujita, M. Networked molecular cages as
crystalline sponges for fullerenes and other guests. Nat. Chem. 2010,
2, 780−783.
(28) Stang, P. J.; Olenyuk, B.; Fan, J.; Arif, A. M. Combining
Ferrocenes and Molecular Squares: Self-Assembly of Heterobimetallic
Macrocyclic Squares Incorporating Mixed Transition Metal Systems
and a Main Group Element. Single-Crystal X-ray Structure of
[Pt(dppf)(H2O)2][OTf]2. Organometallics 1996, 15, 904−908.
(29) Yang, H. B.; Das, N.; Huang, F.; Hawkridge, A. M.; Muddiman,
D. C.; Stang, P. J. Molecular architecture via coordination: self-
assembly of nanoscale hexagonal metallodendrimers with designed
building blocks. J. Am. Chem. Soc. 2006, 128, 10014−10015.
(30) Northrop, B. H.; Yang, H. B.; Stang, P. J. Coordination-driven
self-assembly of functionalized supramolecular metallacycles. Chem.
Commun. 2008, 5896−908.
(31) Li, S.; Huang, J.; Cook, T. R.; Pollock, J. B.; Kim, H.; Chi, K.-
W.; Stang, P. J. Formation of [3]Catenanes from 10 Precursors via
Multicomponent Coordination-Driven Self-Assembly of Metallarec-
tangles. J. Am. Chem. Soc. 2013, 135, 2084−2087.
(32) Mal, P.; Schultz, D.; Beyeh, K.; Rissanen, K.; Nitschke, J. R. An
Unlockable−Relockable Iron Cage by Subcomponent Self-Assembly.
Angew. Chem., Int. Ed. 2008, 47, 8297−8301.
(33) Roberts, D. A.; Pilgrim, B. S.; Cooper, J. D.; Ronson, T. K.;
Zarra, S.; Nitschke, J. R. Post-assembly Modification of Tetrazine-
Edged FeII4L6 Tetrahedra. J. Am. Chem. Soc. 2015, 137, 10068−
10071.
(34) Caulder, D. L.; Bruckner, C.; Powers, R. E.; Konig, S.; Parac, T.
̈
̈
N.; Leary, J. A.; Raymond, K. N. Design, Formation and Properties of
Tetrahedral M4L4 and M4L6 Supramolecular Clusters1. J. Am. Chem.
Soc. 2001, 123, 8923−8938.
(35) Cook, T. R.; Stang, P. J. Recent Developments in the
Preparation and Chemistry of Metallacycles and Metallacages via
Coordination. Chem. Rev. 2015, 115, 7001−45.
(36) Barry, N. P. E.; Govindaswamy, P.; Furrer, J.; Suess-Fink, G.;
Therrien, B. Organometallic boxes built from 5,10,15,20-tetra(4-
pyridyl)porphyrin panels and hydroxyquinonato-bridged diruthenium
clips. Inorg. Chem. Commun. 2008, 11, 1300−1303.
(37) Barry, N. P. E.; Austeri, M.; Lacour, J.; Therrien, B. Highly
Efficient NMR Enantiodiscrimination of Chiral Octanuclear Metalla-
Boxes in Polar Solvent. Organometallics 2009, 28, 4894−4897.
(38) Han, Y.-F.; Lin, Y.-J.; Weng, L.-H.; Berke, H.; Jin, G.-X.
Stepwise formation of “organometallic boxes” with half-sandwich Ir,
Rh and Ru fragments. Chem. Commun. 2008, 350−352.
(39) Ward, B.; Wang, C.-B.; Chang, C. K. Nonbonding steric effect
on carbon monoxide and oxygen binding to hemes. Kinetics of ligand
binding in iron-copper cofacial diporphyrins and strapped hemes. J.
Am. Chem. Soc. 1981, 103, 5236−5238.
(40) Collman, J. P.; Bencosme, C. S.; Durand, R. R.; Kreh, R. P.;
Anson, F. C. Mixed-metal face-to-face porphyrin dimers. J. Am. Chem.
Soc. 1983, 105, 2699−2703.
(41) Casellato, U.; Guerriero, P.; Tamburini, S.; Vigato, P. A. Metal
complexes with disubstituted oxamidic ligands. Inorg. Chim. Acta
1997, 260, 1−9.
(42) Johnston, D. H.; Shriver, D. F. Vibrational study of the
trifluoromethanesulfonate anion: unambiguous assignment of the
asymmetric stretching modes. Inorg. Chem. 1993, 32, 1045−1047.
(43) Sun, H.; Smirnov, V. V.; DiMagno, S. G. Slow Electron
Transfer Rates for Fluorinated Cobalt Porphyrins: Electronic and
Conformational Factors Modulating Metalloporphyrin ET. Inorg.
Chem. 2003, 42, 6032−6040.
(44) Lopes, J. M. S.; Sampaio, R. N.; Ito, A. S.; Batista, A. A.;
Machado, A. E. H.; Araujo, P. T.; Neto, N. M. B. Evolution of
electronic and vibronic transitions in metal(II) meso-tetra(4-pyridyl)-
porphyrins. Spectrochim. Acta, Part A 2019, 215, 327−333.
(45) Guzei, I. An idealized molecular geometry library for
refinement of poorly behaved molecular fragments with constraints.
J. Appl. Crystallogr. 2014, 47, 806−809.
(46) Yoshikawa, S.; Shinzawa-Itoh, K.; Nakashima, R.; Yaono, R.;
Yamashita, E.; Inoue, N.; Yao, M.; Fei, M. J.; Libeu, C. P.; Mizushima,
T.; Yamaguchi, H.; Tomizaki, T.; Tsukihara, T. Redox-Coupled
Crystal Structural Changes in Bovine Heart Cytochrome c Oxidase.
Science 1998, 280, 1723−1729.
(47) Lacy, D. C.; McCrory, C. C. L.; Peters, J. C. Studies of Cobalt-
Mediated Electrocatalytic CO2 Reduction Using a Redox-Active
Ligand. Inorg. Chem. 2014, 53, 4980−4988.
(48) Appel, A. M.; Helm, M. L. Determining the Overpotential for a
Molecular Electrocatalyst. ACS Catal. 2014, 4, 630−633.
(49) Mihara, N.; Yamada, Y.; Takaya, H.; Kitagawa, Y.; Aoyama, S.;
Igawa, K.; Tomooka, K.; Tanaka, K. Oxygen Reduction to Water by a
1105
https://dx.doi.org/10.1021/jacs.0c11895
J. Am. Chem. Soc. 2021, 143, 1098−1106

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Cofacial Dimer of Iron(III)−Porphyrin and Iron(III)−Phthalocya-
nine Linked through a Highly Flexible Fourfold Rotaxane. Chem. -
Eur. J. 2017, 23, 7508−7514.
(50) Pegis, M. L.; Martin, D. J.; Wise, C. F.; Brezny, A. C.; Johnson,
S. I.; Johnson, L. E.; Kumar, N.; Raugei, S.; Mayer, J. M. Mechanism
of Catalytic O2 Reduction by Iron Tetraphenylporphyrin. J. Am.
Chem. Soc. 2019, 141, 8315−8326.
(51) Bard, A. J.; Faulkner, L. R. Electrochemical Methods:
Fundamentals and Applications; Wiley: 2000.
1106
https://dx.doi.org/10.1021/jacs.0c11895
J. Am. Chem. Soc. 2021, 143, 1098−1106