Heterolytic Scission of Hydrogen within a Crystalline Frustrated Lewis Pair

Article abstract of DOI:10.1021/acs.inorgchem.0c02290

We report the heterolysis of molecular hydrogen under ambient conditions by the crystalline frustrated Lewis pair (FLP) 1-{2-[bis(pentafluorophenyl)boryl]phenyl}-2,2,6,6-tetramethylpiperidine (KCAT). The gas-solid reaction provides an approach to prepare the solvent-free, polycrystalline ion pair KCATH2 through a single crystal to single crystal transformation. The crystal lattice of KCATH2 increases in size relative to the parent KCAT by approximately 2%. Microscopy was used to follow the transformation of the highly colored red/orange KCAT to the colorless KCATH2 over a period of 2 h at 300 K under a flow of H2 gas. There is no evidence of crystal decrepitation during hydrogen uptake. Inelastic neutron scattering employed over a temperature range from 4-200 K did not provide evidence for the formation of polarized H2 in a precursor complex within the crystal at low temperatures and high pressures. However, at 300 K, the INS spectrum of KCAT transformed to the INS spectrum of KCATH2. Calculations suggest that the driving force is more favorable in the solid state compared to the solution or gas phase, but the addition of H2 into the KCAT crystal is unfavorable. Ab Initio methods were used to calculate the INS spectra of KCAT, KCATH2, and a possible precursor complex of H2 in the pocket between the B and N of crystalline KCAT. Ex-situ NMR showed that the transformation from KCAT to KCATH2 is quantitative and our results suggest that the hydrogen heterolysis process occurs via H2 diffusion into the FLP crystal with a rate-limiting movement of H2 from inactive positions to reactive sites.

Full text of DOI:10.1021/acs.inorgchem.0c02290

pubs.acs.org/IC
Article
Heterolytic Scission of Hydrogen Within a Crystalline Frustrated
Lewis Pair
Mark E. Bowden,* Bojana Ginovska,* Martin Owen Jones, Abhijeet J. Karkamkar,
Anibal J. Ramirez-Cuesta, Luke L. Daemen, Gregory K. Schenter, Seth A. Miller, Timo Repo,
Konstantin Chernichenko, Noemi Leick, Madison B. Martinez, and Tom Autrey
Cite This: https://dx.doi.org/10.1021/acs.inorgchem.0c02290
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: We report the heterolysis of molecular hydrogen
under ambient conditions by the crystalline frustrated Lewis pair
(FLP) 1-{2-[bis(pentafluorophenyl)boryl]phenyl}-2,2,6,6-tetrame-
thylpiperidine (KCAT). The gas−solid reaction provides an
approach to prepare the solvent-free, polycrystalline ion pair
KCATH2 through a single crystal to single crystal transformation.
The crystal lattice of KCATH2 increases in size relative to the
parent KCAT by approximately 2%. Microscopy was used to
follow the transformation of the highly colored red/orange KCAT
to the colorless KCATH2 over a period of 2 h at 300 K under a
flow of H2 gas. There is no evidence of crystal decrepitation during
hydrogen uptake. Inelastic neutron scattering employed over a
temperature range from 4−200 K did not provide evidence for the formation of polarized H2 in a precursor complex within the
crystal at low temperatures and high pressures. However, at 300 K, the INS spectrum of KCAT transformed to the INS spectrum of
KCATH2. Calculations suggest that the driving force is more favorable in the solid state compared to the solution or gas phase, but
the addition of H2 into the KCAT crystal is unfavorable. Ab Initio methods were used to calculate the INS spectra of KCAT,
KCATH2, and a possible precursor complex of H2 in the pocket between the B and N of crystalline KCAT. Ex-situ NMR showed
that the transformation from KCAT to KCATH2 is quantitative and our results suggest that the hydrogen heterolysis process occurs
via H2 diffusion into the FLP crystal with a rate-limiting movement of H2 from inactive positions to reactive sites.
activation by solid state FLPs that has followed since. Qu et al.8
reviewed FLP systems which have separate acidic and basic
molecules with at least one of these in the solid state during
activation. Computational studies have suggested solid state H2
INTRODUCTION
■
Catalytic activation of molecular hydrogen is critical for the
efficient upgrading of energy carriers ranging from conven-
tional hydrocarbons to renewables such as biomass. While
activation by Lewis acid−base pairs incorporated into MOF
structures;9−11 however, experimental observations have been
limited. In one recent example, the use of solid physical
mixtures of FLPs has been shown to be an effective strategy for
achieving frustration for molecular pairs that have insufficient
steric hindrance to prevent adduct formation in solution.12
These solid mixtures have been reported to react with
hydrogen to form the corresponding ion pairs in what was
considered to be a partially molten environment at the
interface with sufficient molecular flexibility to activate H2 that
had diffused there. Ozin et al. have demonstrated hydrogen
platinum group metals1 are recognized as efficient catalysts for
H2 activation, there has been significant progress in the
development of nonprecious metal catalysts for hydrogen
activation.2−5 In parallel, there has been an interest in the
ability of nonmetal catalysts to activate H2 in catalytic reactions.
Frustrated Lewis pairs (FLPs),6 composed of Lewis acidic
borane and Lewis basic amine or phosphine, activate hydrogen
efficiently and have been demonstrated to reduce a wide range
of substrates in the solution phase. FLPs are of interest as they
provide an approach to heterolytically activate H2 catalytically
at moderate and often ambient temperatures and pressures,
without recourse to expensive transition metals and complex
nanoscale composites.
Received: August 3, 2020
FLP catalysts, almost exclusively, activate H2 as solvated
molecular systems and thus have been investigated predom-
inately in homogeneous catalytic processes. The work of
O’Hare et al.7 in 2016, demonstrating H2 activation by silica
supported BCF as a Lewis base, initiated the activity in H2
https://dx.doi.org/10.1021/acs.inorgchem.0c02290
© XXXX American Chemical Society
Inorg. Chem. XXXX, XXX, XXX−XXX
A

Inorganic Chemistry
pubs.acs.org/IC
Article
activation at the surface of rhombohedral In2O3‑x(OH)y, and
used this to catalyze various hydrogenation reactions.13,14
Their work showed how surface analogs of FLPs are generated
where terminally bonded surface hydroxide groups act as Lewis
bases and the coordinately unsaturated surface indium site acts
as a Lewis acid. A research interest of ours is to understand
how H2 is activated in the “pocket” of an FLP. We have
measured the activation barrier in the solution phase and
found the enthalpic barrier to be quite low and apparently
controlled by diffusion of H2 through solution into the
activated pocket.15,16 Our hypothesis was that in the solid
phase, the heterolysis of H2 would be slower assuming a higher
barrier for rearrangement of bonding in the solid state to
optimize the transition state structure for heterolysis of H2.
The goal of the work was to observe polarization of the H2
molecule in a precursor complex, the “pocket” of an FLP, at
low temperature by inelastic neutron scattering (INS)
spectroscopy, then warm the crystalline material to room
temperature, and follow the heterolysis reaction. The
heterolysis of H2 by the acid−base pair is subtly different
than the ligand exchange reactions on organometallic
compounds reported by Brookhart and Weller. Brookhart
and coauthors17 demonstrated that the nonporous iridium
molecular complex, {C6H3-2,6-[OP-(C6H2(CF3)3-2,4,6)2]2}Ir-
(N2), underwent a single crystal to single crystal (SC-SC)
transformation to form a dihydrogen complex on exposure to
H2 and provided a scaffold for the catalytic transformation of
CH2=CH2 to ethane. This followed work by Bianchini et al.18
who used an Ir tripodal polyphosphine complex in the solid
state to catalytically hydrogenate ethene. Similarly, Weller et
al.19 showed that solid rhodium bis-phosphine complexes
would undergo solid−gas ligand exchange reactions to become
active catalysts. These workers showed how SC-SC chemistry
can be used to synthesize σ-alkane complexes,20 which are
typically difficult to prepare otherwise because the weak M···
H−C interactions are easily replaced by solvent coordination.
They were also able to demonstrate how C−H bonds are
activated according to their metal interactions, leading to
selective deuteration21 of exo- hydrogens in coordinated
norbornane by exposing crystals to D2 gas at room temper-
ature.
EXPERIMENTAL AND COMPUTATIONAL
METHODS
■
We employed an extensive range of spectroscopic and diffraction
experiments described in the introduction as well as DFT and
molecular dynamics calculations. These followed common laboratory
practice, and details of the methods used are given in the Supporting
Information. All sample manipulations were carried out using glove
boxes or other protective atmosphere techniques to avoid reaction
with water or other atmospheric gases.
RESULTS
■
NMR Spectroscopy. The quantitative transformation of
the solid-state KCAT to KCATH2 was undertaken by filling
the headspace above ca. 20 mg of KCAT in an NMR tube with
H2 gas at ambient pressure. The KCAT sample was a coarse
crystalline powder comprising particles approximately 50−200
μm in size. Over the course of 2 h, the orange/red crystals of
KCAT in the NMR tube were observed to lose their color,
indicating a transformation to the colorless compound
KCATH2 (Figure 1a). The reaction was confirmed by 11B
Here, we report heterolytic hydrogen scission by crystalline
1-{2-[bis(pentafluorophenyl)boryl]phenyl}-2,2,6,6-tetrame-
thylpiperidine (KCAT) at ambient temperatures and pressures
to quantitatively yield the corresponding zwitterionic pair. The
single crystal to single crystal (SC-SC) transformation provides
the first demonstration of a nonmetal example of H2
heterolysis by a unimolecular FLP in the solid state. Single
crystal and powder X-ray diffraction were used to determine
the structure and lattice expansion on hydrogen uptake to form
the ion pair (KCATH2). We use a combination of temper-
ature-programmed desorption (TPD) and inelastic neutron
scattering (INS) to demonstrate that there is little molecular
hydrogen solubility within the solid FLP prior to heterolytic
dissociation and no evidence for a precursor, polarized-H2,
intermediate phase. Optical and NMR measurements were
used to confirm the quantitative conversion of KCAT to
KCATH2 at ambient temperatures. First principle calculations
were used to determine the free energy changes for
incorporation of H2 into the crystal lattice, for the reaction
to the zwitterion, and to model INS spectra for comparison
with experiment.
Figure 1. (a) KCAT and KCATH2 structures; (b) 11B NMR (in
bromobenzene) of KCAT after reaction with H2 gas.
NMR spectroscopy after dissolving the product in bromo-
1
benzene (Figure 1b) which shows the H-coupled doublet
resonance at −22.5 ppm characteristic of the ion pair. The
broad resonance at +62 ppm displayed by KCAT (published in
earlier work,22 SI) was no longer detected.
Single Crystal and Powder X-ray Diffraction. Powder
X-ray diffraction (Figure S1) of KCAT and KCATH2
indicated that both the reactant and product were crystalline.
The reactant pattern matched that calculated from the crystal
structure of KCAT,22 but the previously published structure
for KCATH2 has entrained solvent and is therefore not
B
https://dx.doi.org/10.1021/acs.inorgchem.0c02290
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
pubs.acs.org/IC
Article
suitable for comparison. However, our solid−gas reaction
resulted in crystals of KCATH2 free of solvent which were
suitable for structure solution by single crystal diffraction. This
suggests the reaction is a single crystal−single crystal
transformation. The structure we obtained (Figure 2) has a
microscope slide affixed to one face. After locating a suitable
mass of crystals, H2 gas was flowed through the cell at 1 bar
pressure, and images were taken every 30 s, with typical images
presented in Figure 3. These images demonstrated that the
external features of the crystal are unaffected by the
transformation to the ion pair, with no evidence for
decrepitation of hydrogenated material from the host crystal.
The most significant change observed was the color change,
which occurred slowly throughout most of the crystal.
Although the color change was not perfectly homogeneous,
we did not observe a sharp reaction front progressing from the
outside surface as seen by Weller.19 Some portions of the
crystal reacted faster than others, but the boundaries between
these regions were diffuse and therefore there did not appear
to be a clear preference for surface or interior sites.
The volume expansion predicted by the crystal structures
was accommodated without any obvious changes to the size
and shape of the crystals. The crystal under observation moved
once between successive frames (between Figure 3a,b), and
the field of view was adjusted. However, the retention of the
overall integrity of the crystal is surprising given the structural
changes which result from the formation of the ion pair.
Inelastic Neutron Scattering. Investigation of the
reaction was undertaken using INS to provide insight into
the chemical environment of H2 prior to formation of the ion
pair. We were especially interested to see whether any weakly
bound H2 between the boron and nitrogen centers could be
observed, as suggested by theoretical studies of related
systems.23−25 Although multiple INS experiments were
performed at a variety of temperatures and pressures (see
Supporting Information) no unambiguous evidence for
dihydrogen, polarized or otherwise, was obtained. For example,
the INS spectra measured at 10 K for KCAT before and after
exposure to ca. 1 equiv H2 at up to 200 K were
indistinguishable (Figure 4). However, exposure at 300 K
resulted in transformation to the ion pair and a concomitant
change in the INS spectrum.
Computational Modeling. DFT calculations were done
for the periodic system, with a cell containing 4 FLP molecules,
replicated in XYZ directions for a bulk model. The models
were built from the experimentally determined crystal
structures reported here. All calculations were done with the
CP2K program,26 using PBE functional27 with Grimme’s D2
dispersion correction.28,29 DZVP basis set and Goedeker-
Hutter pseudopotentials30 consistent with the PBE functional
were employed. Details of the methods and the approach are
provided in the SI (page 10). The energy difference between
the FLP and ion pair revealed a strong driving force for the
heterolysis with ΔE = −135 kJ mol−1H2 and a free energy
change at 300 K of ΔA = −125 kJ mol−1 H2. However, the
presence of molecular H2 within the KCAT lattice was
Figure 2. Structure of solvent-free KCATH2 obtained from single
crystal XRD. C, gray; F, green; N, blue; B, pink; H, white.
molecular geometry similar to the solvent-containing crystal.
Selected bond distances and angles are tabulated (Table S2) to
show how the environments around N and B change upon
reaction. There is a small (ca. 3.5%) increase in N − C bond
distances, but the angles around N do not change significantly
because of the more rigid piperidine residue and the presence
of the lone pair in KCAT. The increase in B − C bond
distances upon heterolysis were slightly larger (ca. 4−6%) and
the geometry around B changed from approximately planar
triangular to tetrahedral as expected. This changes the
orientations of the pentafluorophenyl rings and consequently
some of the intermolecular packing interactions. It is
remarkable that the compound remains as a single crystal
during this transformation. Further discussion of the structural
differences between KCATH2 with and without solvent is
given in the SI.
The lattice parameters show that KCATH2 has a unit cell
volume 2.1% larger than KCAT. Single crystal measurements
of KCAT were also undertaken at temperatures between 100
and 300 K to detect any phase transition. The structures
obtained were the same as the published structure22 without
any phase transition or unusual thermal expansion (Figure S3).
Optical Microscopy. Optical microscopy was used to
dynamically observe the hydrogen uptake in KCAT. Large
crystals of KCAT were grown by slow evaporation from
heptane and placed under N2 in a custom gastight cell with a
Figure 3. Optical micrographs of KCAT crystals exposed to H2 gas for (a) 0 min, (b) 55 min, (c) 85 min, (d) 112 min, (e) 150 min.
C
https://dx.doi.org/10.1021/acs.inorgchem.0c02290
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
pubs.acs.org/IC
Article
Table 1. Calculated Electronic Energy Changes and
Distances to Nearest B and N Atoms for Addition of
Molecular H2 in Different Sites within Crystalline KCAT
B···H
distance
(Å)
B···N
distance
(Å)
H−H
distance
(Å)
H2
ΔE
N···H
distance (Å)
site (kJ mol−1
)
1
2
3
4
26
32
39
36
5.10
2.39
2.52
6.67
4.72
2.15
2.35
5.25
3.07
3.14
3.05
3.05
0.75
0.76
0.75
0.75
detected during the temperature increase (Figure S7) and the
crystals retained their bright color at the end of the experiment.
DISCUSSION
■
Activation or heterolysis of molecular H2 to a proton and a
hydride, H+/H−, by solid phase KCAT at ambient temperature
and pressure is readily detected by the visual bleaching of the
brightly colored FLP. The heterolysis was confirmed by several
analytical methods and takes place as a single crystal to single
crystal transformation. This is apparent from visual observa-
tion; for example, Figure 3 shows a small number of crystals
whose external appearance is unchanged as the heterolysis
takes place within their interior. Further, a single crystal of
KCATH2 suitable for structure determination was readily
obtained from KCAT crystals exposed to H2 gas. The color
change suggests that formation of KCATH2 takes place in the
solid state rather than during dissolution in the NMR solvent
as an example. This was confirmed by the INS experiments,
where the ion pair was observed after reaction only of the solid
KCAT and gaseous hydrogen (Figure 4). The heterolysis is
more exergonic in the solid state (ΔAcalc = −125 kJ mol−1 H2)
than in solution (ΔGcalc = −50 kJ mol−1 H2).31 The reason for
this was not examined but is likely due to electrostatic
stabilization between the zwitterionic charges in the closely
packed crystalline KCATH2. Benchmarking computational
studies by Grimme et al.32 have also reported additional
stabilization in solid-state FLPs for H2 hydrolysis; however,
their observation was that the stabilization in intramolecular
FLPs is up to ∼20 kJ mol−1. We calculate a significantly larger
difference of ∼75 kJ mol−1: closer to the value Grimme
reported for intermolecular FLPs. We attribute this additional
stabilization to the different packing of KCAT and KCATH2
in contrast to the earlier study32 where only the structure of
the FLP is used. This difference suggests that in solid state
FLPs, in addition to the crystal field effects, packing can have
significant impact on the thermochemistry of the reaction. We
considered two potential mechanisms for the phase transition
of KCAT to KCATH2: (i) a diffusion mechanism where
gaseous H2 diffuses into crystalline KCAT followed by the
activated heterolysis of H2 and (ii) a decrepitation mechanism
where the outside of the solid phase particle reacts with
gaseous H2 and peels away layer by layer until complete
conversion of the FLP to the corresponding ion pair has
occurred. Optical microscopy of the FLP crystallites eliminated
the possibility of a decrepitation mechanism, and this
observation was supported by the small lattice expansion of
2.1 vol % on hydrogen uptake observed by diffraction. This
may be compared to classical decrepitation systems, such as
metal hydrides and Nd−Fe−B, which show expansions in the
range of 10−15 vol %.33,34 The lattice expansion is, however,
consistent with those reported for organometallic SC-SC
Figure 4. INS spectra at 10 K before (black) and after exposure to H2
gas at 200 (blue) and 300 K (green). The difference after exposure to
H2 at 200 K is shown at top.
energetically disfavored. Calculations after addition of four
different H2 molecules near each of the four B···N sites in a
symmetry-free KCAT unit gave a free energy change after
relaxation of ΔA = 50 kJ mol−1 H2 and an electronic energy
change of ΔE = 33 kJ mol−1H2. The electronic energy for
individual H2 molecules (Table 1) spanned a range of ΔE =
+26 to +39 kJ mol−1. The most stable structure was found to
have H2 in a position far from the active B···N site and
interacting with the π system of the pentafluorophenyl rings
(Figure S9, panel 1). A second H2 molecule (site 2, Table 1),
although 6 kJ mol−1 higher in energy, showed evidence of
interaction with nearby B and N atoms. Both the B···N and
H−H distances for this H2 were elongated compared to their
normal values toward distances calculated for the transition
state in solution heterolysis.31
INS spectra were calculated both from the harmonic modes
of the quantum mechanical optimized structure with and
without H2 as well as Ab Initio Molecular Dynamics (AIMD)
simulations that allow for anharmonic treatment of the
frequencies. Both approaches indicated that the presence of
H2 within the lattice would lead to significant changes in the
INS spectra (Figures S8 and S13). Our experimental results
(Figure 4), which did not detect changes, therefore lead us to
believe that the concentration of H2 in the lattice is much
lower than the stoichiometric quantity used in the calculations.
The MD simulations also indicated that H2 has significant
mobility within the KCAT lattice. Each of the four sites
traversed a large volume within the lattice during the 12 ps
simulation, up to 5 Å from the original starting location
(Figures S11 and S12). These diffuse volumes are consistent
with shallow minima that H2 can access near 300 K.
Temperature-Programmed Desorption. The presence
of molecular H2 in the crystal was also sought using
temperature-programmed desorption (TPD). One mg of
KCAT was exposed to 1 atm of H2 gas for 30 min at 298 K
and cooled rapidly to 77 K under the H2 atmosphere. Our
previous microscopy and INS experiments indicated that H2
enters the crystal at 298 K but not at 77 K. The sample was
pumped to ca. 1 × 10−8 Torr at 77 K before heating at 5 K/
min to 330 K and monitoring any gas liberated with a
quadrupole mass spectrometer. Despite the sensitivity of this
experiment to small (≥0.01 wt %) quantities of gas, no H2 was
D
https://dx.doi.org/10.1021/acs.inorgchem.0c02290
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
pubs.acs.org/IC
Article
transformations.35 As a consequence of these observations, we
assign a diffusion mechanism for the phase transformation in
crystalline KCAT. The rate of diffusion is faster than, or at least
comparable to, the rate of H2 heterolysis. If the rate of
heterolysis was significantly faster than diffusion, we would
expect to see a more clearly defined skin of white KCATH2 on
the exterior of the crystals, much like the CO complex
observed by Pike et al.19 on the surface of their Rh butadiene
complex. However, our optical observations revealed a more
widespread and diffuse color change without a sharply defined
boundary.
Little has been reported on the absolute rates of H2
activation by FLPs in solution, and there are no reports on
the kinetics of solid phase FLPs reacting with gaseous H2. We
assume the reaction involves an activated process that requires
near ambient temperature. This conclusion is based on
observations where at temperatures <200 K, the ion pair was
not formed even under a high pressure of H2 gas in our in situ
INS measurements.
free space within a crystal based on van der Waals radii. At 300
K, KCAT has a packing index of 67%, which is within the
range of 65−77% found in most crystals.38 Although this is
near the lower bound, it indicates there is no reason to expect
an unusually high concentration of H2 in KCAT compared to
other compounds.
The reaction path for heterolysis by molecular KCAT in
solution has been calculated by Repo et al.31 This work did not
identify a precursor complex but a transition state with an
activation barrier of 74 kJ mol−1 where the B···N distance is
elongated to 3.23 Å and the H−H distance to 0.76 Å. Our
calculations (site 2, Table 1) also showed a state with an
elongated H−H bond and with a more modest elongation of
the B···N distance to 3.14 Å, likely restricted by the packing of
the crystal. Although not formally identified as a transition
state, it is interesting to note that the geometry of the H2 and
nearby KCAT molecules in our calculations are nearly identical
to those in solution and that the energy for H2 addition in the
solid state is smaller than the activation barrier in solution, at
50 kJ mol−1 compared to the reactants.
The observation of multiple calculated minima where the H2
can reside in the FLP lattice indicates that H2 may get trapped
in unfavorable positions at low temperatures, inhibiting access
to the reactive pocket. Equally, such conclusions suggest that
although H2 diffuses relatively rapidly into the lattice, the rate
limiting step for heterolysis of molecular H2 within KCAT is
local movement to the reactive site rather than the molecular
rearrangement of the FLP in the solid state. In this light, an
interesting follow up could focus on studying the kinetics of
the reaction using solid-state (e.g., Avrami) models, including
the use of D2 to distinguish diffusion from activated rate-
limiting processes.
We hypothesized that a diffusion mechanism offers an
opportunity to observe the precursor complex of KCAT with
molecular hydrogen, {H2KCAT(s)}, where the H2 is
energetically stabilized within the cavity of the FLP to form
a polarized H2 molecule
H2(g) + KCAT(s) F {H2KCAT(s)} F KCATH2(s)
We sought spectroscopic evidence for such a precursor
complex with INS spectroscopy, encouraged by calculations
(Figures S8, S13) which indicated that discernible differences
in the spectra would arise from the presence of H2 within the
lattice. We considered that KCAT possessed several features
that might be advantageous for the formation of such a
complex: (i) a linked FLP with the Lewis acid and base on the
same framework, (ii) a geometry with the Lewis acidic B and
Lewis basic N fixed about 3 Å apart, (iii) a strong Lewis acid
and base pair that would provide a large electric field to
polarize H2,29,36 and (iv) a favorable equilibrium toward the
zwitterion suggesting a relatively low activation energy for a
precursor complex. However, despite these favorable factors,
no evidence for an intermediate phase was found by INS.
The INS results were supported by calculations of the
energetic penalty for molecular H2 to reside in the KCAT
crystal. The relative stabilities of H2 occupying different
positions in the crystal showed that the most stable site was
not productive for H2 activation, since H2 was more than 6 Å
away from the nearest B···N site. This is consistent with the H2
accessing the active site at higher temperatures but not at low
temperatures. This observation is reminiscent of work by
Weller et al.,37 on Rh based MOFs where Xe gas was found to
diffuse away from the active site and reside in hydrophobic
pockets, similar to enzymatic gas channels. Further, although
the formation of the ion pair clearly showed that H2 had
entered the solid at room temperature, our TPD measure-
ments did not detect H2 diffusing out of the lattice after
freezing the sample to 77 K and evacuating the surrounding
gas. INS demonstrated heterolysis does not occur at this low
temperature, and so we anticipated that H2 would be trapped
inside the crystal. Upon heating, the low pressure of H2 outside
the crystal would reverse diffusion as the temperature was
raised.
CONCLUSIONS
■
We have demonstrated the first heterolysis of hydrogen in an
intramolecular, metal-free FLP system in the solid state.
Hydrogen diffuses into the crystal lattice of KCAT at 298 K
but remains at concentrations too low for detection by inelastic
neutron scattering or temperature-programmed desorption.
Heterolysis occurs at 298 K to form the zwitterion in a single
crystal−single crystal transformation with a strong exergonic
free energy change. Observation of the color change associated
with this reaction shows that diffusion within the KCAT crystal
occurs more rapidly than heterolysis at a local site. This is
likely to be a consequence of H2 moving between and
occupying sites not closely associated with the frustrated B···N
to a greater extent than the sites proximate to B···N.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.inorgchem.0c02290.
■
sı
Experimental and computational details; powder XRD
results, crystallographic tables, and discussion of solvated
and solvent-free KCATH2; VISION INS spectra at
temperatures between 10 and 200 K, TOSCA INS
spectra, TPD traces; calculated INS spectra and images
of optimized and dynamic H2 positions within KCAT
lattice (PDF)
Crystallographic data for KCATH2 have been deposited
with the Cambridge Crystallographic Data Centre with
deposition number 2020667.
All these results indicated that the concentration of H2
within the KCAT crystal was small. The Kitaigorodsky packing
index, calculated with PLATON, provides a measure of the
E
https://dx.doi.org/10.1021/acs.inorgchem.0c02290
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
pubs.acs.org/IC
Article
Accession Codes
Teachers and Scientists under the Science Undergraduate
Laboratory Internships program. A portion of the research
including some computation was performed using EMSL
(grid.436923.9), a DOE Office of Science User Facility
sponsored by the Office of Biological and Environmental
Research. Further computing resources were provided by
PNNL Research Computing. This research benefited from the
use of the VISION beamline (IPTS- 12815) at the Spallation
Neutron Source, Oak Ridge National Laboratory (ORNL),
which is supported by the Scientific User Facilities Division,
Office of Basic Energy Sciences, US Department of Energy
(DOE). TOSCA experiments at the ISIS Neutron and Muon
Source were supported by a beamtime allocation RB1510079
from the Science and Technology Facilities Council. K.C. and
T.R. thank the Academy of Finland (project # 276586) for
financial support.
CCDC 2020667 contains the supplementary crystallographic
data for this paper. These data can be obtained free of charge
via www.ccdc.cam.ac.uk/data_request/cif, or by emailing
data_request@ccdc.cam.ac.uk, or by contacting The Cam-
bridge Crystallographic Data Centre, 12 Union Road,
Cambridge CB2 1EZ, UK; fax: +44 1223 336033.
AUTHOR INFORMATION
Corresponding Authors
■
Mark E. Bowden − Environmental Molecular Sciences
Laboratory, Pacific Northwest National Laboratory, Richland,
Washington 99352, United States; orcid.org/0000-0003-
3812-3340; Email: mark.bowden@pnnl.gov
Bojana Ginovska − Physical and Computational Sciences
Directorate, Pacific Northwest National Laboratory, Richland,
Washington 99352, United States; orcid.org/0000-0003-
0165-8202; Email: bojana.ginovska@pnnl.gov
REFERENCES
■
Authors
(1) Chen, Q. A.; Ye, Z. S.; Duan, Y.; Zhou, Y. G. Homogeneous
palladium-catalyzed asymmetric hydrogenation. Chem. Soc. Rev. 2013,
42 (2), 497−511.
Martin Owen Jones − ISIS Neutron and Muon Spallation
Facility, STFC, RAL, Didcot OX11 0QX, U.K.; St Andrews
University, Fife KY16 9AJ, Scotland, U.K.
Abhijeet J. Karkamkar − Physical and Computational Sciences
Directorate, Pacific Northwest National Laboratory, Richland,
Washington 99352, United States
Anibal J. Ramirez-Cuesta − Spallation Neutron Source, Oak
Ridge National Laboratory, Oak Ridge, Tennessee 37830,
United States
Luke L. Daemen − Spallation Neutron Source, Oak Ridge
National Laboratory, Oak Ridge, Tennessee 37830, United
States
Gregory K. Schenter − Physical and Computational Sciences
Directorate, Pacific Northwest National Laboratory, Richland,
Washington 99352, United States
Seth A. Miller − Physical and Computational Sciences
Directorate, Pacific Northwest National Laboratory, Richland,
Washington 99352, United States
Timo Repo − Department of Chemistry, University of Helsinki,
00014 Helsinki, Finland; orcid.org/0000-0002-3116-6199
Konstantin Chernichenko − Department of Chemistry,
University of Helsinki, 00014 Helsinki, Finland
Noemi Leick − National Renewable Energy Laboratory, Golden,
Colorado 80403, United States; orcid.org/0000-0002-
2014-6264
Madison B. Martinez − National Renewable Energy
Laboratory, Golden, Colorado 80403, United States
Tom Autrey − Physical and Computational Sciences Directorate,
Pacific Northwest National Laboratory, Richland, Washington
99352, United States; orcid.org/0000-0002-7983-3667
(2) Rakowski DuBois, M.; DuBois, D. L. The roles of the first and
second coordination spheres in the design of molecular catalysts for
H2 production and oxidation. Chem. Soc. Rev. 2009, 38 (1), 62−72.
(3) Alig, L.; Fritz, M.; Schneider, S. First-Row Transition Metal
(De)Hydrogenation Catalysis Based On Functional Pincer Ligands.
Chem. Rev. 2019, 119 (4), 2681−2751.
(4) Filonenko, G. A.; van Putten, R.; Hensen, E. J. M.; Pidko, E. A.
Catalytic (de)hydrogenation promoted by non-precious metals - Co,
Fe and Mn: recent advances in an emerging field. Chem. Soc. Rev.
2018, 47 (4), 1459−1483.
(5) Belkova, N. V.; Epstein, L. M.; Filippov, O. A.; Shubina, E. S.
Hydrogen and Dihydrogen Bonds in the Reactions of Metal Hydrides.
Chem. Rev. 2016, 116 (15), 8545−87.
(6) Stephan, D. W. Catalysis, FLPs, and Beyond. Chem 2020, 6 (7),
1520−1526.
(7) Xing, J. Y.; Buffet, J. C.; Rees, N. H.; Norby, P.; O’Hare, D.
Hydrogen cleavage by solid-phase frustrated Lewis pairs. Chem.
Commun. 2016, 52 (69), 10478−10481.
(8) Ma, Y.; Zhang, S.; Chang, C. R.; Huang, Z. Q.; Ho, J. C.; Qu, Y.
Semi-solid and solid frustrated Lewis pair catalysts. Chem. Soc. Rev.
2018, 47 (15), 5541−5553.
(9) Ye, J. Y.; Johnson, J. K. Design of Lewis Pair-Functionalized
Metal Organic Frameworks for CO2 Hydrogenation. ACS Catal.
2015, 5 (5), 2921−2928.
(10) Ye, J. Y.; Johnson, J. K. Catalytic hydrogenation of CO2 to
methanol in a Lewis pair functionalized MOF. Catal. Sci. Technol.
2016, 6 (24), 8392−8405.
(11) Heshmat, M. Alternative Pathway of CO2 Hydrogenation by
Lewis-Pair-Functionalized UiO-66 MOF Revealed by Metadynamics
Simulations. J. Phys. Chem. C 2020, 124 (20), 10951−10960.
(12) Wang, L.; Kehr, G.; Daniliuc, C. G.; Brinkkotter, M.; Wiegand,
T.; Wubker, A. L.; Eckert, H.; Liu, L.; Brandenburg, J. G.; Grimme, S.;
Erker, G. Solid state frustrated Lewis pair chemistry. Chem. Sci. 2018,
9 (21), 4859−4865.
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.inorgchem.0c02290
(13) Ghuman, K. K.; Hoch, L. B.; Wood, T. E.; Mims, C.; Singh, C.
V.; Ozin, G. A. Surface Analogues of Molecular Frustrated Lewis Pairs
in: Heterogeneous CO2 Hydrogenation Catalysis. ACS Catal. 2016, 6
(9), 5764−5770.
(14) Wang, L.; Yan, T.; Song, R.; Sun, W.; Dong, Y.; Guo, J.; Zhang,
Z.; Wang, X.; Ozin, G. A. Room-Temperature Activation of H2 by a
Surface Frustrated Lewis Pair. Angew. Chem., Int. Ed. 2019, 58 (28),
9501−9505.
(15) Houghton, A. Y.; Autrey, T. Calorimetric Study of the
Activation of Hydrogen by Tris(pentafluorophenyl)borane and
Trimesitylphosphine. J. Phys. Chem. A 2017, 121 (46), 8785−8790.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The authors (M.E.B., B.G., A.J.K., G.K.S., and T.A.) gratefully
acknowledge support from U.S. Department of Energy (DOE),
Office of Science, Office of Basic Energy Sciences, Division of
Chemical Sciences, Geosciences & Biosciences for providing
the impetus for this research. S.M. was supported by DOE,
Office of Science, Office of Workforce Development for
F
https://dx.doi.org/10.1021/acs.inorgchem.0c02290
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
pubs.acs.org/IC
Article
(16) Whittemore, S. M.; Autrey, T. Kinetic and Thermodynamic
Study of the Reduction of 1,1-Diphenylethylene by a Thermally
Frustrated Diethyl Ether-BCF Lewis Pair. Isr. J. Chem. 2015, 55 (2),
196−201.
(35) Pike, S. D.; Weller, A. S. Organometallic synthesis, reactivity
and catalysis in the solid state using well-defined single-site species.
Philos. Trans A Math Phys. Eng. Sci. 2015, 373 (2037), 2037.
(36) Rokob, T. A.; Bako, I.; Stirling, A.; Hamza, A.; Papai, I.
Reactivity Models of Hydrogen Activation by Frustrated Lewis Pairs:
Synergistic Electron Transfers or Polarization by Electric Field? J. Am.
Chem. Soc. 2013, 135 (11), 4425−4437.
(37) Martinez-Martinez, A. J.; Rees, N. H.; Weller, A. S. Reversible
Encapsulation of Xenon and CH2Cl2 in a Solid-State Molecular
Organometallic Framework (Guest@SMOM). Angew. Chem., Int. Ed.
2019, 58 (47), 16873−16877.
(38) Kitaigorodsky, A. I. Molecular Crystals and Molecules. Academic
Press: New York, 1973.
(17) Huang, Z.; White, P. S.; Brookhart, M. Ligand exchanges and
selective catalytic hydrogenation in molecular single crystals. Nature
2010, 465 (7298), 598−601.
(18) Bianchini, C.; Farnetti, E.; Graziani, M.; Kaspar, J.; Vizza, F.
Molecular Solid-State Organometallic Chemistry of Tripodal
(Polyphosphine)Metal Complexes - Catalytic-Hydrogenation of
Ethylene at Iridium. J. Am. Chem. Soc. 1993, 115 (5), 1753−1759.
(19) Pike, S. D.; Kramer, T.; Rees, N. H.; Macgregor, S. A.; Weller,
A. S. Stoichiometric and Catalytic Solid-Gas Reactivity of Rhodium
Bisphosphine Complexes. Organometallics 2015, 34 (8), 1487−1497.
(20) Pike, S. D.; Chadwick, F. M.; Rees, N. H.; Scott, M. P.; Weller,
A. S.; Kramer, T.; Macgregor, S. A. Solid-state synthesis and
characterization of sigma-alkane complexes, [Rh(L2)(eta(2), eta(2)-
C7H12)][BAr(F)4] (L2 = bidentate chelating phosphine). J. Am.
Chem. Soc. 2015, 137 (2), 820−33.
(21) Chadwick, F. M.; Kramer, T.; Gutmann, T.; Rees, N. H.;
Thompson, A. L.; Edwards, A. J.; Buntkowsky, G.; Macgregor, S. A.;
Weller, A. S. Selective C-H Activation at a Molecular Rhodium Sigma-
Alkane Complex by Solid/Gas Single-Crystal to Single-Crystal H/D
Exchange. J. Am. Chem. Soc. 2016, 138 (40), 13369−13378.
(22) Chernichenko, K.; Nieger, M.; Leskela, M.; Repo, T. Hydrogen
activation by 2-boryl-N, N-dialkylanilines: a revision of Piers’ ansa-
aminoborane. Dalton Trans. 2012, 41 (30), 9029−9032.
(23) Camaioni, D. M.; Ginovska-Pangovska, B.; Schenter, G. K.;
Kathmann, S. M.; Autrey, T. Analysis of the Activation and
Heterolytic Dissociation of H-2 by Frustrated Lewis Pairs: NH3/
BX3 (X = H, F, and Cl). J. Phys. Chem. A 2012, 116 (26), 7228−
7237.
(24) Pu, M. P.; Privalov, T., Ab initio dynamics trajectory study of
the heterolytic cleavage of H-2 by a Lewis acid [B(C6F5)(3)] and a
Lewis base [P(tBu)(3)]. J. Chem. Phys. 2013, 138 (15).
(25) Wu, D.; Jia, D.; Liu, L.; Zhang, L.; Guo, J. Reactivity of 2,6-
lutidine/BR(3) and pyridine/BR(3) Lewis pairs (R = F, Me,
C(6)F(5)): a density functional study. J. Phys. Chem. A 2010, 114
(43), 11738−45.
(26) VandeVondele, J.; Krack, M.; Mohamed, F.; Parrinello, M.;
Chassaing, T.; Hutter, J. QUICKSTEP: Fast and accurate density
functional calculations using a mixed Gaussian and plane waves
approach. Comput. Phys. Commun. 2005, 167 (2), 103−128.
(27) Perdew, J. P.; Burke, K.; Ernzerhof, M. Generalized gradient
approximation made simple. Phys. Rev. Lett. 1996, 77 (18), 3865−
3868.
(28) Grimme, S. Semiempirical GGA-type density functional
constructed with a long-range dispersion correction. J. Comput.
Chem. 2006, 27 (15), 1787−1799.
(29) Grimme, S.; Kruse, H.; Goerigk, L.; Erker, G. The Mechanism
of Dihydrogen Activation by Frustrated Lewis Pairs Revisited. Angew.
Chem., Int. Ed. 2010, 49 (8), 1402−1405.
(30) Goedecker, S.; Teter, M.; Hutter, J. Separable dual-space
Gaussian pseudopotentials. Phys. Rev. B: Condens. Matter Mater. Phys.
1996, 54 (3), 1703−1710.
(31) Chernichenko, K.; Kotai, B.; Nieger, M.; Heikkinen, S.; Papai,
I.; Repo, T. Replacing C6F5 groups with Cl and H atoms in frustrated
Lewis pairs: H2 additions and catalytic hydrogenations. Dalton Trans.
2017, 46 (7), 2263−2269.
(32) Liu, L.; Brandenburg, J. G.; Grimme, S. On the hydrogen
activation by frustrated Lewis pairs in the solid state: benchmark
studies and theoretical insights. Philos. Trans. R. Soc., A 2017, 375
(2101), 20170006.
(33) Joubert, J. M.; Latroche, M.; Cerny, R.; Percheron-Guegan, A.;
Yvon, K. Hydrogen cycling induced degradation in LaNi5-type
materials. J. Alloys Compd. 2002, 330, 208−214.
(34) Liszkowski, P.; Turek, K.; Figiel, H. The influence of
decrepitation on the diffusion kinetics of hydrogen in Nd-Fe-B. J.
Alloys Compd. 2000, 307, 297−303.
G
https://dx.doi.org/10.1021/acs.inorgchem.0c02290
Inorg. Chem. XXXX, XXX, XXX−XXX