C-H and O2Activation at a Pt(II) Center enabled by a novel sulfonated CNN pincer ligand

Article abstract of DOI:10.1021/acs.organomet.6b00613

A novel sulfonated CNN pincer ligand has been designed to support CH and O2 activation at a Pt(II) center. The derived cycloplatinated aqua complex 7 was found to be one of the most active reported homogeneous Pt catalysts for H/D exchange between studied arenes (benzene, benzene-d_6, toluene-d_8, p-xylene, and mesitylene) and 2,2,2-trifluoroethanol (TFE) or 2,2,2- trifluoroethanol-d; the TON for C₆D₆ as a substrate is >250 after 48 h at 80 °C. The reaction is very selective; no benzylic CH bond activation was observed. The per-CH-bond reactivity diminishes in the series benzene (19) > toluene (p-CH:m-CH:o-CH = 1:0.9:0.2) > xylene (2.9) > mesitylene (1.1). The complex 7 reacts slowly in TFE solutions under ambient light but not in the dark with O₂ to selectively produce a Pt(IV) trifluoroethoxo derivative. The H/D exchange reaction kinetics and results of the DFT study suggest that complex 7, and not its TFE derivatives, is the major species responsible for the arene CH bond activation. The reaction deuterium kinetic isotope effect, kH/kD = 1.7, the reaction selectivity, and reaction kinetics modeling suggest that the CH bond cleavage step is rate-determining.

Full text of DOI:10.1021/acs.organomet.6b00613

This is an open access article published under an ACS AuthorChoice License, which permits
copying and redistribution of the article or any adaptations for non-commercial purposes.
Article
pubs.acs.org/Organometallics
C−H and O Activation at a Pt(II) Center Enabled by a Novel
2
Sulfonated CNN Pincer Ligand
David Watts, Daoyong Wang, Mackenzie Adelberg, Peter Y. Zavalij, and Andrei N. Vedernikov*
Department of Chemistry and Biochemistry, University of Maryland, College Park, Maryland 20742, United States
*
S Supporting Information
ABSTRACT: A novel sulfonated CNN pincer ligand has been
designed to support CH and O activation at a Pt(II) center. The
2
derived cycloplatinated aqua complex 7 was found to be one of the
most active reported homogeneous Pt catalysts for H/D exchange
between studied arenes (benzene, benzene-d , toluene-d , p-xylene,
6
8
and mesitylene) and 2,2,2-trifluoroethanol (TFE) or 2,2,2-
trifluoroethanol-d; the TON for C D as a substrate is >250 after
6
6
4
8 h at 80 °C. The reaction is very selective; no benzylic CH bond
activation was observed. The per-CH-bond reactivity diminishes in
the series benzene (19) > toluene (p-CH:m-CH:o-CH = 1:0.9:0.2)
>
xylene (2.9) > mesitylene (1.1). The complex 7 reacts slowly in TFE solutions under ambient light but not in the dark with O2
to selectively produce a Pt(IV) trifluoroethoxo derivative. The H/D exchange reaction kinetics and results of the DFT study
suggest that complex 7, and not its TFE derivatives, is the major species responsible for the arene CH bond activation. The
reaction deuterium kinetic isotope effect, k /k = 1.7, the reaction selectivity, and reaction kinetics modeling suggest that the CH
H
D
bond cleavage step is rate-determining.
1
. INTRODUCTION
In spite of the demonstration of this important reactivity, the
catalytic aerobic C−H functionalization of alkanes or arenes by
The homogeneous C−H functionalization of hydrocarbons
mediated by Pt complexes has an impressive history spanning
almost 50 years from its discovery to subsequent mechanistic
studies
II
+
II
the aqua complex 4, [(dpms)Pt (OH ) ] (Scheme 1), could
2 2
1
not be developed. One of the reasons for that is a bimolecular
1
b,2,3
4−7
decomposition of 4 leading to the robust dinuclear complex
and up to current developments.
While some
13
(
dpms) Pt (μ-OH) 5. Fortunately, the bimolecular decom-
2 2 2
substantial progress in Pt-catalyzed “non-oxidative” C−H
7
position pathway can be shut down via a covalent
immobilization of 4. Using this strategy we recently were able
to achieve catalytic aerobic epoxidation of some olefins with an
functionalization has been achieved, the oxidative C−H
functionalization remains limited to the use of more or less
1
5
“impractical” oxidizing agents such as H PtCl , CuCl , and
14
2
6
2
impressive TON exceeding 40000.
6
SO (oleum). The use of a more economically attractive
oxidant, O , has only been reported for mediator-containing Pt
systems such as O /heteropolyacids and O /CuCl , among
others, but these systems suffer from their low selectivity. The
use of O as an inexpensive oxidant for selective oxidative C−H
functionalization mediated by Pt compounds remains
desirable and challenging. Potentially, such processes could be
realized in mediator-free systems where both C−H and O2
activation occur at the same metal center. Toward this goal,
previously we have demonstrated a facile mediator-free Pt −
C(sp ) bond functionalization in water or alcohols using O
from air. Such transformations include aerobic Pt (Alk) to
Pt (Alk) oxidation (e.g., 1 to 2 in Scheme 1) and subsequent
C(sp )−O elimination from Pt (Alk) intermediates such as 3,
all enabled by the tripod facially chelating ligand bis(2-
pyridyl)methanesulfonate (dpms). The pendant sulfonate
group of the dpms ligand serves as a surprisingly good donor
facilitating the formation of Pt intermediates such as 2.
The sulfonate also acts as a good leaving group, thus making
3
Another important factor contributing to the lack of the C−
H bond activation reactivity of 4 toward alkanes and arenes
may be related to the difficulty of coordinating alkane and arene
substrates to the Pt(II) center, which is a step required for
subsequent aerobic functionalization of the substrate in the
II
2
8
,9
4
2
2
2
2
II
II
10
(
dpms)Pt systems. In general, coordination of olefins is
thermodynamically more favorable and kinetically more facile
in comparison to alkanes and arenes, thus accounting for the
success of the aerobic epoxidation of olefins by immobilized
II
14
complex 4. In particular, olefins but not arenes can displace
3
2
the DMSO ligand in the DMSO complex (dpms)Pt(OH)-
1
0
II
(
DMSO) (6) and be engaged in CH activation by this complex
IV
15
even in aqueous media.
3
IV
II
To enhance the reactivity of (dpms)Pt (OH )- type
2
complexes in reactions requiring displacement of the aqua
ligand, one can attempt diminishing the aqua ligand
IV
11,12
Special Issue: Hydrocarbon Chemistry: Activation and Beyond
Received: July 31, 2016
3
IV
the subsequent C(sp )−O elimination from the Pt center
facile.
1
3
©
XXXX American Chemical Society
A
DOI: 10.1021/acs.organomet.6b00613
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Scheme 1. Aerobic Pt −C(sp ) Functionalization in Water
Article
II
3
II
Scheme 2. New Sulfonated CNN Pincer Preligand 8 and Derived Pt Complexes 7 and 9−14
dissociation energy. At this point, on the basis of the results of a
bimolecular decomposition pathway leading to the formation of
bis-μ-hydroxo complexes such as 5.
molecular modeling of 1 and 4, we hypothesized that a strong
−
II
intramolecular hydrogen bond, SO ···H O−Pt , may be the
In this work we have tested the hypotheses above. The novel
3
2
II
factor contributing to the high energy of the aqua ligand
dissociation. Expectedly, the hydrogen bonding can be
diminished if the sulfonate group and the Pt center are
sulfonated CNN pincer preligand 8 and the derived Pt
complexes K[(C H -dpms)PtCl] (9) and 7 have been prepared
6
4
II
and fully characterized. In contrast to the aqua complex 4,
solutions of 7 in 2,2,2-trifluoroethanol (TFE) allow for catalytic
H/D exchange between benzene as well as some (poly)-
methylbenzenes and TFE solvent at 21−80 °C and exhibit the
highest reactivity in this reaction, in comparison to other
slightly pulled apart, as in the more rigid pincer complex
(
C H -dpms)Pt(OH ) (7), supported by a novel sulfonated
6 4 2
CNN pincer ligand derived from the preligand 8 (Scheme 2).
The presence of a close-by pendant sulfonate group in 7 would
II
IV
10
II
allow for facile aerobic Pt to Pt oxidation, whereas the
reported homogeneous Pt -based systems, some of which
1
6−19
presence of a rigid CNN pincer scaffold would also allow
require a temperature of 150−180 °C to operate.
weakening of the Pt−OH bonding/facilitation of hydrocarbon
Interestingly, our data indicate that the aqua complex 7 and
not its TFE derivatives, such 10 and 12, is responsible for the
2
substrate coordination to the metal and would shut down the
B
DOI: 10.1021/acs.organomet.6b00613
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Scheme 3. Aerobic Oxidation of 7 in TFE Solutions
Scheme 4. Preparation of Complexes 9 and 7
Figure 1. ORTEP drawings (50% probability ellipsoids) of (a) 7 and (b) 9 (cation is not shown; hydrogen atoms are omitted for clarity). The
cocrystallized solvent is not shown.
CH activation in our systems. Finally, compound 7 also allows
(Scheme 4). Reaction of 9 with ethanolic KOH followed by
solvent removal and extraction of the solid product with
ethanol afforded K[(C H -dpms)Pt(OEt)] (16). The latter was
for O activation to selectively produce the corresponding
2
IV
trifluorethoxo Pt derivative 15, presumably via the inter-
6
4
mediacy of the TFE derivative 14 (Scheme 3). The oxidation
reaction occurs only under light, representing a simple tool to
dissolved in water and converted to the water-insoluble aqua
complex 7, which crystallized upon acidification of the solution
with H SO . The target aqua complex was isolated as the
control this type of reactivity. One can expect that the transient
2
4
II
(
L)Pt R species resulting from hydrocarbon RH activation are
tetrahydrate 7·4H O in analytically pure form in 51% yield.
2
much more reactive than 7 thanks to the presence of the
second hydrocarbyl group at the metal center resulting from a
CH bond cleavage event, so that they can be engaged in the
Slow crystallization of 9 from water−methanol mixtures
afforded 9·H O·MeOH. In both cases the crystalline products
2
so obtained were suitable for single-crystal X-ray character-
ization (Figure 1). According to the XRD characterization, in
reaction with O more readily. The analysis of such reactivity is
2
3
currently under investigation in our laboratory. In this work we
report the benzene and some (poly)methylbenzene CH
activation chemistry with TFE solutions of 7, including
mechanistic analysis of benzene C−H activation based on
experimental and computational (DFT) data, as well as some
their solid state both 9 and 7 feature the κ -C,N,N coordination
II
mode of the sulfonated pincer ligand bound to the Pt center.
In both complexes the phenylpyridine unit is almost planar.
The least-squares mean plane of this fragment forms a 35−40°
angle with the plane of the second pyridine ring. In the
details of photochemical oxidation of 7 with O2.
crystalline tetrahydrate 7·4H O the aqua ligand is not engaged
2
in H bonding with the sulfonate group of the same unit of 7 but
instead is involved in hydrogen bonding with the sulfonate
group of the adjacent unit of 7 indirectly, via a chain comprised
of two H-bonded water molecules.
2
. RESULTS AND DISCUSSION
.1. Preparation and Characterization of Novel
Sulfonated CNN-Pincer Complexes (C H -dpms)Pt(OH )
2
6
4
2
(
7) and K[(C H -dpms)PtCl] (9) (Scheme 4). The preligand
2.1.1. Transformations of 7 Involving the TFE for H O
2
6
4
8
was prepared from 6-phenylbis(2-pyridyl)methane by
Substitution and the Change of the Pincer Ligand
Coordination Mode from CNN to CNO. While in the solid
state 7·4H O and 9·H O·MeOH have the pincer ligand
20
sulfonation of the derived carbanion with Me NSO . Reaction
of 8 with 1 equiv of K PtCl in aqueous acetic acid led to its
quantitative conversion to the chloroplatinate(II) complex 9,
3
3
2
4
2
2
3
coordinated to the metal in a κ -C,N,N fashion, in solutions
which was isolated in 77% yield in analytically pure form
these complexes may coexist in equilibrium with their
C
DOI: 10.1021/acs.organomet.6b00613
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
3
respective κ -C,N,O-coordinated isomers: e.g., 7 vs 11 (Scheme
to 13 and the [TFE]:[H O] ratio (Figure 2). The linear
2
2). The presence of C,N,O-coordinated Pt(II) complexes in
relationship observed suggests that the Pt-containing compo-
TFE solutions of 7 is indirectly supported (vide infra) by the
fact that aerobic oxidation of these solutions produces 15
(Scheme 3), featuring a mer-CNO arrangement of the ligand
donor atoms. In addition to the coordination isomerism, both 7
and 11 may be engaged in a ligand exchange with the TFE
solvent, resulting in the formation of TFE adducts 10 and 12,
respectively. Finally, the CNO-coordinated complexes 11 and
1
2 having a basic pendant pyridyl group may undergo an
intramolecular proton transfer from a relatively acidic
coordinated solvent ligand to the pyridine nitrogen atom to
produce zwitterionic species 13 and 14, respectively. The
pyridinium proton, in turn, may be involved in a weak
II
hydrogen bonding with the Pt center, as was observed earlier
2
1
in similar systems.
Figure 2. Linear correlation between the integral intensity of broad
signals associated with LPt(TFE) (14 and its isomers): integral
Dissolution of yellow crystalline 7·4H O in TFE produces
2
yellow solutions which rapidly change their color to dark
intensity of sharp signals associated with 13 and the [TFE]:[H O]
2
II
orange, thus suggesting that the identity of the Pt complex
ratio at 80 °C. The [H O] value ranges from 30 mM to 1.0 M, whereas
2
changes upon its dissolution. The removal of the solvent under
vacuum or rapid precipitation of all of the solute upon addition
[TFE] = 12.2−12.4 M.
of diethyl ether to “aged” TFE solutions of 7 leads to a product
nents of the solution are likely to be related by a ligand
exchange equilibrium (eq 1):
1
that, on the basis of its H NMR spectra in DMSO-d , contains
6
II
two to three water molecules per LPt unit and 0.2 equiv of
LPt(H O) + TFE → LPt(TFE) + H O
(1)
2
2
TFE. The electrospray ionization mass spectrometry of aged
F
TFE solutions of 7 with additives of NaBAr shows a strong
4
Here LPt(H O) denotes 13 and LPt(TFE) designates some
2
+
signal corresponding to a 13·Na adduct (or its isomers) and a
mixture of TFE complexes that are rapidly equilibrating: e.g.,
+
weak signal corresponding to a 14·Na adduct (or its isomers).
1
0, 12, and 14.
The corresponding equilibrium constant KTFE value
+
No signal assignable to 13·H is observed in the positive mode;
the only signal seen in the negative mode mass spectrum is
+
[LPt(TFE)][H O]
2
assigned to the anion (14-H) . Our data presented below
KTFE
=
suggest that 13 is the major Pt-containing species present in
[LPt(H O)][TFE]
2
TFE solutions of 7·4H O at 21 °C (e.g., the fraction of 13 is
2
was found to be (1.32 ± 0.06) × 10−2 at 80 °C, with ΔG (1)
rxn
8
5% at 21 °C and [7·4H O] = 13 mM), whereas the minor
2
=
3.0 ± 0.1 kcal/mol.
component is, predominantly, a mixture of TFE complexes
LPt(TFE) existing in a fast (on the NMR time scale)
equilibrium. The fraction of 13 changes reversibly as a function
Consistent with the analysis above, our VT 1H NMR
experiments with 13 mM TFE solutions of partially hydrated 7·
xH O, x = 2, show that, as the temperature increases from 21 to
2
of the [TFE]:[H O] ratio and temperature, as described below.
2
8
0 °C, the fraction of 13 decreases from 85% (the equilibrium
1
Two sets of signals are observed in H NMR spectra of aged
−4
−2
constant value KTFE = 4.4 × 10 ) to 15% (K = 1.4 × 10 ).
From the van’t Hoff plot corresponding to the interconversion
TFE
solutions of 7 in CF CH OH. In the first set of signals assigned
3
2
to 13 all but one of its signals remain sharp in the temperature
range of 21−80 °C. The second component of “aged” solutions
of LPt(H O) (13) and LPt(TFE) in reaction 1 the reaction
2
thermodynamic parameters are as follows: ΔH = 11.9 ± 0.1
kcal/mol, ΔS = 25.2 ± 0.4 cal/K mol, and ΔG = 4.5 ± 0.2 kcal/
mol at 25 °C and 3.0 ± 0.2 kcal/mol at 80 °C (Figure 3). The
latter value matches the reaction 1 Gibbs energy of 3.0 kcal/
mol found using the plot in Figure 2.
1
of 7 exhibits broad H NMR signals at lower temperatures of
this range which become somewhat sharper at 80 °C. Notably,
the subspectrum associated with 13 features a slightly
broadened singlet at 7.61 ppm integrating as 1H and assigned
+
to the NH group proton. This signal disappears upon addition
of TFE-d to the above solutions due to an H/D exchange with
the additive. Notably, 14, the trifluoroethoxo analogue of 13,
+
also has an acidic NH fragment in it. Using solutions with
different [TFE]:[H O] ratios turned out to be an efficient way
2
to distinguish between 13 and 14 as the source of the 7.61 ppm
signal. The use of water-poorer samples 7·xH O where x = 0.6,
2
0
leads to a progressively lower fraction of 13 (sharp signals) in
such solutions at 21 °C: 66% and 55%, respectively. In turn,
addition of water to the above solutions leads to an almost
complete disappearance of broad signals (97% fraction of 13 at
1
[
H O] = 1 M) in their H NMR spectra.
2
Our experiments with varying concentrations of water
performed at 80 °C, when the integral intensities of both
sharp and broad signals can be comparable, allowed us to find a
linear correlation between the ratio of intensity of broad signals
assigned to LPt(TFE) to the intensity of sharp signals assigned
Figure 3. van’t Hoff plot for reaction 1 in the temperature range of
21−80 °C at [LPt(solv)] = 13 mM and [TFE]/[H O] = 415.
2
D
DOI: 10.1021/acs.organomet.6b00613
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Finally, our observations of aged TFE solutions of 7 showed
that, on exposure to a prolonged (several hours) heating at 80
C, they slowly produce a new derived compound, the dark
precipitate 17. This solid dissolves in TFE in the presence of
HBF4 or KCl. Accordingly, the formation of 17 can be
precluded in the presence of acid additives: e.g., HBF4.
Compound 17 can also be dissolved in DMSO to form
TON of 100 could be achieved after 8 h of reaction at 80 °C.
An analysis of forward and reverse (C H −CF CH OD
system) reaction 2 kinetics allowed us to determine the
deuterium kinetic isotope effect for (2). The observed reaction
2 rate constants were estimated as described below.
6
6
3
2
°
2.2.1. C−H Activation of Benzene. 2.2.1.1. Modeling the
Reaction Kinetics. To calculate the observed rate constants for
22
(
C H -dpms)Pt(DMSO), which could be prepared independ-
6 4
reaction 2, a numerical integration method was used along
with the simple model presented in Scheme 6. An example of
fitting experimental TONs is given in Figure 4.
1
ently from complex 7, as judged by H NMR spectroscopy
Scheme 5). These observations are consistent with the
(
Scheme 5. “Dimerization” Reactivity of TFE Solutions of 7
Scheme 6. Simple Kinetics Model Used in This Work To
Account for the H/D Exchange Kinetics between Benzene-d6
and TFE and the Catalyst Decomposition
hypothesis that 17 is a product of a reversible “oligomerization”
of 7 that may occur by displacement of solvento ligands in
LPt(solv) species with a pendant donor group belonging to
another such species. ESI-MS spectra of aged TFE solutions of
F
The reaction model in Scheme 6 assumes a slow formation of
PtIV aryl hydrido intermediates which are involved in a fast
hydride ligand H atom exchange with the TFE solvent
following a deprotonation−protonation reaction sequence. A
slow decomposition of LPt(solv) leading to 17 was taken into
account as follows. The best least-squares fit to the plot of the
total concentration of all LPt(solv) species vs time allowed us
7
containing NaBAr show the presence of a weak signal
4
16
+
II
attributable to a Na adduct with L Pt , one of the possible
2
2
components of 17.
2
.2. Aromatic C−H Bond Activation in Benzene and
Methylarenes by 7 and 9 in 2,2,2-Trifluoroethanol. As
was mentioned earlier, the (dpms)Pt aqua complex 4 (Scheme
II
1
) does not react with alkanes or arenes in water or TFE
13
to find the reaction rate law, −d[LPt(solv)]/dt = k [LPt-
solvent. Hence, it was exciting for us to discover that TFE
solutions of the Pt aqua complex 7 can catalyze the H/D
2
4
II
(
solv)] − k , and the rate constants for the forward and
−2
−3 −3
reverse reactions, k = 1.1 ± 0.2 M
s
and k = 0. The
−2
exchange between benzene-d and CF CH OH (eq 2):
2
6
3
2
reaction order 4 may reflect the complexity of the process of
coordination polymerization of LPt(solv). When the catalyst
loss due to formation of 17 was accounted for, the reaction 2
profile could be fully modeled as shown in Figure 4 using the
rate law in eq 3
C D + nCFCH OH ⇌ C D H + nCFCH OD
(2)
6
6
3
2
6
6−n
n
3
2
1
The reaction progress could be monitored by means of H
NMR spectroscopy. The H/D exchange occurs at a noticeable
rate already at 21 °C and is reasonably fast at 80 °C (Figure 4).
No H/D exchange was noticed in the absence of the catalyst. A
−
d[CD]
dt
[ROH]
=
k1obs,D[Pt(solv)][CD]
[ROD] + [ROH]
[
ROD]
ROD] + [ROH]
3)
−
k 1obs,H[Pt(solv)][CH]
−
[
(
where [CD] and [CH] are the molar concentrations of the
substrate CD and CH bonds in solution and [ROH] and
[
ROD] are the concentrations of water and/or TFE, serving as
20
a source of “acidic” H and D, respectively.
Using the reaction model in Scheme 6 and the rate law (3),
we were able to calculate the observed per-bond second-order
rate constant corresponding to the C−D bond cleavage in
−
6
−1
−1
benzene-d , k
, equal to (7.2 ± 0.2) × 10
s
M
(initial
value is
6
1obs,D
−
5
−1
TOF 4.9 × 10 s ) at 21 °C. At 80 °C the k
Figure 4. Reaction profile for the H/D exchange between benzene-d6
1obs,D
(
1.28 ± 0.05) × 10−
3
s
−1
M
−1
(initial TOF 8.7 × 10 s ). The
−3 −1
(
10 vol %)−CF CH OH catalyzed by 13 mM 7·4H O at 80 °C: black
3
2
2
circles denote experimental data; the red line corresponds to the
k1obs,D values can be compared with those reported in the
TONs predicted by the reaction model in Scheme 6.
literature for some other Pt-based systems used for
E
DOI: 10.1021/acs.organomet.6b00613
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Scheme 7. Reactivity of Previously Reported Systems for Arene CH Bond Activation at a Pt Center
Article
II
1
6
stoichiometric benzene CH activation or catalytic H/D
Scheme 8. More Detailed Reaction Model Used for the
Analysis of the Initial Reaction Rates Involving LPt(TFE)
exchange involving benzene−TFE-d or benzene−CF COOD
3
17−19
and LPt(H O) Complexes
systems (Scheme 7).
The highest rate constants for
benzene CH bond cleavage at a Pt center have been reported
by Bercaw and Labinger for a series of cationic complexes
N,N′-diaryldiimine)PtMe(OH2) such as 18. In TFE
solution at 20 °C the estimated second-order rate constant
2
II
+
16
(
−4
−1
−1
reported is 6.9 × 10
s
M . Neglecting the reaction 2
kinetic isotope effect at this point (vide infra), this value is
almost 100 times greater than k1obs,D for 7, which may be
II
attributed to the presence of a methyl ligand at the Pt center
and, therefore, an electron-richer metal complex, a factor
16
recognized as very important in this type of reaction. The
d[CD]
dt
II
−
next most active series of complexes, (N,N′-diaryldiimine)Pt
species such as 19, catalyze the H/D exchange between C H
6
6
k1 [LPt(H O)] + k1,TFE[LPt(TFE)]
,H O
2
2
and CF CH OD (TFE-d) with comparable rate constants of
3
2
−
= (k
[PhD]
3
−1
/k )[H O] + (k
/k )[TFE] + 1
(
1−8) × 10
s
but only at the much higher temperature of
−1,H O
2
2
−1,TFE
2
2
1
7
150 °C. One might expect that TFE solutions of 7 would
(4)
demonstrate a much higher reactivity toward benzene under
comparable conditions. Two more examples of Pt -catalyzed
Using the least-squares fitting of the experimentally found
II
initial reaction rates, the reaction parameters k1,TFE and k
as
1,H
20
2
O
H/D exchange with C H (catalysts 20 and 21) involve
6
6
well as k−1,TFE/k and k
/k ratios could be found. The
2
2
‑1,H
2
O
trifluoroacetic acid and also exhibit comparable (20) or slower
rates of catalysis (21).
first two parameters characterize the reactivity of LPt(TFE) and
LPt(H O) species, respectively, in the ligand substitution
2.2.1.2. Effect of Water Additives on the Rate of CH
2
reaction with benzene, whereas the last two ratios can be used
to elucidate the nature of the reaction rate-determining step: if
Activation. A more detailed mechanistic analysis of reaction 2
could be carried out using the model presented in Scheme 8.
This model considers concurrent participation in the reaction
with benzene of two solvento species of different reactivity,
for a given solvento ligand, solv = H O, TFE, the ratio k‑1,solv/k2
2
>
1, the reaction rate-determining step is the substrate CH
bond cleavage.
LPt(TFE) and LPt(H O). For simplicity, only the initial
2
Our experiments were performed in aqueous TFE−
reaction period (<5% substrate conversion) is considered when
the contribution of the reverse reaction can be neglected. The
benzene−complex 7 mixtures with a 0.040−1.16 M total
1
concentration of water at 80 °C. Using H NMR spectroscopy
20
1
reaction rate law used along with Scheme 8 is given by eq 4:
the concentrations of LPt(TFE) (broader H NMR signals)
F
DOI: 10.1021/acs.organomet.6b00613
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
1
and LPt(H O) (sharper H NMR signals) could be determined
2
experimentally. As the water concentration increases, the
[
LPt(TFE)]/[LPt(H O)] ratio and the observed initial H/D
2
exchange rates decrease, as shown in Table 1.
Table 1. Effect of Water Additives on the Initial Rates of
reaction 3 Measured in the System H O−CF CH OH−
2
3
2
Benzene-d (10 vol %)−Complex 7 (0.00515 M) at 80 °C
6
initial rate, M h−
1
[
H O], M
[LPt(TFE)]/[LPt(H O)]
2
2
0.038
0.110
0.275
0.477
0.765
1.16
3.71
0.178 ± 0.003
0.174 ± 0.003
0.132 ± 0.001
0.097 ± 0.001
0.073 ± 0.001
0.060 ± 0.002
1.77
0.454
0.171
0.065
0.029
Figure 5. Eyring plot for the second-order rate constant k1obs,D for the
H/D exchange in the system benzene-d (10 vol %)−CF CH OH−
6
3
2
complex 7.
6
. The effective reaction Gibbs activation energy is 25.4 kcal/
mol at 80 °C and 24.2 kcal/mol at 25 °C.
2
0
The experimental data fitting led us to the value of k‑1,TFE
/
A more detailed analysis of the CH bond cleavage reactivity
of LPt(solv) species is given in section 2.4.
−5
k = 3.8 × 10 , thus suggesting that the LPt(TFE) species are
virtually unreactive. A repeated data-fitting procedure taking
this conclusion into account leads to the reaction parameter
2
Finally, the reactivity of the chloro complex 9 in TFE
solutions in the H/D exchange in a 1/9 C D /TFE mixture was
6
6
−2
−1 −1
values k1,H2O = 8.67 × 10
M
s
(the ligand exchange
also analyzed. Due to its activity being lower than that of 7, a
higher temperature of 120 °C was used along with a 15 mM
⧧
activation energy ΔG
= 22.5 kcal/mol) and k k1,H2O/k‑1,H2O
3
3
53
2
(3.0 ± 0.2) × 10 M−1 s , corresponding to the overall
−
−1
HBF additive to prevent formation of 17. The second-order
4
=
⧧
reaction rate constant k1obs,D estimated using the simple
reaction model in Scheme 6 was found to be (3.5 ± 0.1) ×
reaction 2 activation energy ΔG
= 24.8 kcal/mol and
3
53
k
‑
/k = 28.7. The conclusion that the LPt(TFE) species are
−5 −1
−1
1,H
2
O
2
1
0
s
M . The much lower reactivity of 9 in comparison to
less reactive in CH bond activation in comparison to their
LPt(H O) analogues is important. It means that the effect of
the TFE as a cosolvent present in aqueous solutions is just to
serve the role of an “inert” diluting agent, diminishing the high
water concentration and the associated inhibiting effect of water
on the CH activation. In turn, the k‑1,H2 /k value well above 1
suggests realization of the rate-determining CH activation step
7
is, most likely, due to the relatively high energy cost required
16
2
for the chloride ligand substitution with benzene (eq 5):
K[LPtCl] + PhH → KCl + LPt(PhH)
(5)
Indeed, 9 is much more robust than 7; mixing TFE solutions
of 7 with LiCl produces 9 virtually instantaneously and
quantitatively.
O
2
in reaction 2 involving the aqua complex LPt(H O). On the
2.3. CH Activation of Toluene, p-Xylene, and
Mesitylene. The selectivity of CH activation of hydrocarbon
substrates possessing CH bonds of different type is of prime
2
basis of this value, the difference in the Gibbs energy of the
transition states corresponding to the benzene displacement by
H O in LPt(H O) and CH bond cleavage in the resulting
16b
interest for practical applications of CH activation. Hence,
we analyzed the selectivity of CH bond activation in some
(poly)methylbenzenes using TFE solutions of 7 at 80 °C.
Using toluene-d instead of benzene-d as a substrate in our
2
2
LPt(PhD) is −RT ln(28.7) = −2.4 kcal/mol.
.2.1.3. Deuterium Kinetic Isotope Effect. The deuterium
kinetic isotope effect is a standard reaction parameter used to
2
8
6
2
3
analyze the mechanism of CH activation by transition metals
and, in particular, arenes at a Pt center.
H/D exchange experiments, we calculated the following TONs
II
16,17
2
To estimate the
after 18 h of reaction at 80 °C: 120 for the meta C(sp )−D
bonds with the initial TOF of 1.8 × 10−
3
s
−1
and 85 for the
reaction deuterium kinetic isotope effect (KIE) in our system,
in addition to C D −CF CH OH systems, C H −CF CH OD
ortho/para C−D bonds altogether with the initial TOF of 1.31
6
6
3
2
6
6
3
2
−3
−1
(
TFE-d ) mixtures were analyzed at 80 °C. The KIE value
× 10
s
(Table 2). Notably, when TOF values normalized
1
found for 7 as a catalyst was derived from the initial H/D
per number of equivalent C−D bonds involved in reaction 2
exchange rate ratio, rate(H)/rate(D) = 1.7, and falls into the
are compared, the value for the toluene-d meta C−D bonds,
8
II
−3 −1
range of 1.6−2.2 typical for benzene CH activation at Pt
0.98 × 10 s , is lower in comparison to that for benzene-d ,
6
1.28 × 10− s . Remarkably, the benzylic C(sp )−D bonds of
3
−1
3
center where the substrate C−H bond cleavage is rate-
1
6,17
determining.
toluene were not detectably involved in the reaction.
The reaction 2 activation parameters for the system benzene-
To estimate the initial reaction selectivity more accurately by
resolving the H NMR signals corresponding to the toluene
1
d (10 vol %)−CF CH OH−complex 7·4H O in the temper-
6
3
2
2
ature range of 21−80 °C were also determined. The resulting
linear dependence of −ln(k h/k T) vs 1/T (Eyring plot
para and ortho C−H bonds with close values of chemical shifts,
the H/D exchange between toluene-d and TFE was carried out
1
B
8
coordinates) is shown in Figure 5. The high quality of the
to achieve only near-stoichiometric (TON ≈ 1) yield of the
protio isotopologues of the substrate so that each of the signals
would appear as a virtual singlet originating, predominantly,
from isomeric monoprotio species CD C D H (Figure 6).
linear correlation suggests the predominant involvement of one
II
active Pt species, LPt(H O), according to our analysis above,
2
in reaction 2 in this temperature range. The reaction activation
3
6
4
⧧
⧧
parameters, ΔH = 16.8 ± 0.5 kcal/mol and ΔS = −25 ± 2
cal/(K mol), are consistent with a stepwise bimolecular
transformation corresponding to the steps k and k in Scheme
The difference in the integral intensity of various CH bond
signals in Figure 6 before and after a short-term heating of the
mixture allowed us to find the relative selectivity per bond for
1
2
G
DOI: 10.1021/acs.organomet.6b00613
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Table 2. Observed Aromatic CH Bond Activation Rate Constants k (Per-Bond Values) and Initial TOF for the Reaction
1
a
between 7 and Aromatic Substrates (10 vol %) in TFE Solution
k1, M− s−1
1
initial TOF, s−1
T, °C
method
substrate
(7.2 ± 0.2) × 10−
6
(4.9 ± 0.1) × 10
(8.7 ± 0.3) × 10
−5
21
80
80
80
80
80
80
80
eq 3
C D
6
6
(1.28 ± 0.05) × 10−
3
3
3
3
3
−3
eq 3
C D
6
6
−
−2
−3
−3
C H
6
(1.89 ± 0.05) × 10
(1.27 ± 0.03) × 10
(1.31 ± 0.10) × 10
(1.85 ± 0.10) × 10
eq 3
6
CD C D , o-CD
(0.19 ± 0.02) × 10−
(1.01 ± 0.05) × 10−
init rates
init rates
init rates
init rates
init rates
3
6
5
CD C D , p-CD
3
6
5
−
CD C D , m-CD
(0.98 ± 0.05) × 10
3
6
5
(3.5 ± 0.2) × 10−
4
(1.14 ± 0.06) × 10
−3
1
1
,4-(CH ) C H
4
3
2
6
(1.53 ± 0.05) × 10−
4
(3.3 ± 0.1) × 10
−4
,3,5-(CH ) C H
3
3
3
6
a
The rate constants were found using the simple model in Scheme 6 and the rate law (3) or initial reaction rates.
1.14 ± 0.06) × 10− and (3.3 ± 0.1) × 10−4 s−1 for p-xylene
3
(
and mesitylene, respectively, at 80 °C. If per-bond TOF values
−3
are compared, (0.29 ± 0.02) × 10 (p-xylene), (0.11 ± 0.1) ×
−
3
−3
−1
1
0
(mesitylene), and 2.1 × 10
s
(C H ), the
6 6
methylarenes are noticeably less reactive than benzene,
presumably, for both electronic and steric reasons. A DFT
computational analysis of the CH bond activation of benzene
with TFE solutions of complex 7 is presented next.
2
.4. Mechanism of Benzene C−H Bond Activation in
TFE Solutions of 7: DFT Analysis (Schemes 9 and 10). In
the Supporting Information we present some results of the
II
DFT analysis of the distribution of major Pt -containing
species in the system 7−TFE. The gas-phase calculations for
this system predict the predominance of 13−TFE, a hydrogen-
bond-stabilized adduct of 13 and TFE, and a low population for
the TFE complexes 10, 12, and 14 that all have approximately
same Gibbs energy. In turn, for TFE solutions the solvation
model used in this work fails to predict the predominance of 13
in this solvent with a slightly positive Gibbs energy of 1.0 kcal/
mol for the reaction leading from TFE and 7 to the 13−TFE
adduct. As such, the “ground state” species we consider in
Schemes 9 and 10 is the aqua complex 7. The error appears to
be of the magnitude of a few kcal/mol, which is not uncommon
for such calculations. The uncertainty in determining the most
stable species in TFE solutions of 7 implies an underestimation
by a few kcal/mol of the reaction 2 effective activation energy
discussed below.
1
Figure 6. H NMR spectrum of a toluene-d (10 vol %)−CF CH OH
8
3
2
(
90 vol %)−complex 7 reaction mixture after a short heating at 80 °C
corresponding to TON ≈ 1, showing well-resolved individual aromatic
C−H bond signals (left). The toluene methyl CHD signal intensity
2
remained virtually unchanged.
ortho, meta, and para C−D bonds as 0.19:0.92:1, respectively.
2
The observed much higher preference of C(sp )−H vs
3
C(sp )−H bonds is very similar to that found earlier using
II
+
cationic (N,N′-diaryldiimine)Pt Me(H O) complexes with a
2
reduced diimine ligand steric bulk (e.g., aryl = 3,5-bis-
24
(
trifluoromethyl)phenyl). The reduced steric bulk, in turn,
In the following analysis (Schemes 9 and 10) the Gibbs
energies were calculated for TFE solutions under standard
conditions, at 1 M concentration of each component except for
the solvent: [TFE] = 13.8 M. To adjust the calculated ΔG
values to our typical working conditions corresponding to the
favors a lower arene for solvento ligand exchange activation
barrier and the C−H bond cleavage being rate determining, as
follows from the notable deuterium kinetic isotope effect k /k
H
D
of 2.2 in the reaction of complex 18 (Scheme 7) with
1
6a
benzene. In contrast, bulkier ligands at the metal (aryl = 2,6-
24
dimethylphenyl) favor a 21% selective benzylic CH activation
and exhibit a low deuterium kinetic isotope effect k /k of 1.1
use of about 15 mM 7·4H O and the associated [H O] = 60
2
2
mM, one needs to subtract 1.7 kcal/mol from any reaction
energy value in Schemes 9 and 10 that corresponds to
H
D
II
for the reaction of this Pt complex with benzene. The arene
for solvento ligand exchange step was proposed to be rate-
determining in this system. On the basis of this comparison,
processes leading to the formation of H O.
2
1
6a
According to our experimental observations, the CNO-
coordinated complex 13 is the major Pt-containing species in
TFE solutions at 21 °C. As a consequence, we started our
analysis with the LPt(solv) complexes featuring a CNO
coordination of the sulfonated pincer ligand (Scheme 9). The
II
it appears that the Pt center in the LPt(H O) species
2
responsible for CH activation in our systems is sterically
uncrowded so that LPt(H O) can react with benzene via rate-
2
determining substrate C−H bond cleavage.
Interestingly, the use of an even bulkier and electron-richer
CNN to CNO isomerization and the TFE for H
2
O ligand
24
⧧
substrate, p-xylene, allowed the authors to invert the reaction
selectivity in favor of the benzylic C−H activation with 90%
selectivity. We also tested p-xylene and mesitylene in the H/D
exchange experiments using 10 vol % of the appropriate arene
exchange are facile (ΔG298 < 12 kcal/mol) for 13 and all the
20
LPt(solv) species considered in Scheme 2.
On the basis of our analysis of the reaction 2 kinetics in
TFE−H O mixtures, complex 11 is one of two aqua complexes
2
in its mixtures with TFE-d. The initial rate method was used
that may be responsible for CH arene activation. According to
the DFT, the associative benzene for aqua ligand substitution in
11 is predicted to be a relatively slow process with a Gibbs
2
(
Table 2). Only C(sp )−H bonds of the substrates were
detectably involved in the reaction, showing initial TOFs of
H
DOI: 10.1021/acs.organomet.6b00613
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
II
Scheme 9. DFT Calculated Gibbs Energy Profile for Benzene CH Bond Activation by LPt (solv) Species Supported by a
C,N,O-Coordinated Pincer Ligand L in Neat TFE at 25 °C
activation energy of 24.2 kcal/mol (TS11‑22). Even less
competitive is the similar benzene for TFE ligand substitution
in the CNO-coordinated TFE analogue 12 with the
corresponding transition state (TS12‑22) Gibbs energy of 30.6
Summarizing results of this analysis, we conclude that the
reaction sequences involving either 11 or 12 have the rate-
determining benzene coordination step, which is not consistent
with our experimental observations.
It is interesting to note here that there is another kinetically
competitive reaction sequence leading to the transient CH
benzene σ complex 23 and the CH bond cleavage transition
state TS23‑−24 (vide infra). This sequence begins with the CNN-
coordinated aqua complex 7 and involves a CNN to CNO
isomerization of the σ-CH benzene complex 26 via TS26‑23
2
kcal/mol. The resulting η -benzene π complex 22 can undergo
a low-barrier π to σ-complex isomerization (TS22‑23) to form
the CH σ-complex 23 with subsequent low-barrier CH bond
IV
oxidative addition (TS23‑24). The resulting Pt phenyl hydride
2
4 may be involved in an isomerization reaction leading to the
even more stable zwitterionic structure 25, a phenyl analogue of
the hydroxo and trifluoroethoxo species 13 and 14, respectively
(
Scheme 10, bottom).
Our analysis of reactivity of the CNN-coordinated LPt(solv)
species toward benzene is illustrated in Scheme 10.
(Scheme 2). Though it was not calculated, we presume that the
activation barrier for such isomerization should not exceed the
activation energy needed for the reverse reaction leading from
An associative ligand substitution in the CNN-coordinated
TFE complex 10 with benzene as an incoming ligand is a
kinetically viable process with a transition state energy of 22.8
kcal/mol (TS10‑26). The reaction between 10 and benzene leads
to 26, a benzene σ-CH complex. It is worth noting that the
corresponding π complex appears to be nonexistent for CNN-
coordinated LPt species, presumably because of excessive steric
interactions between the arene and the pincer ligand C−H
bonds near the Pt atom in the expected LPt(PhH) derivative.
The last reaction step in Scheme 10 following the formation
of 26 corresponds to the coordinated arene CH bond cleavage
2
4 back to 23.
While looking for alternative routes for the formation of the
zwitterionic phenyl complex 25, we have also been able to find
a direct route to it from the CH σ complex 23 employing a
cyclometalation−deprotonation mechanism (transition state
TS ; Scheme 9, bottom). A water molecule is needed to
CMD
shuttle the proton of the Pt-coordinated benzene CH bond to
the nitrogen atom of the ligand pendant pyridyl group. At the
same time, with the transition state TSCMD Gibbs energy of
IV
29.9 kcal/mol this pathway is virtually out of competition with
to form the Pt phenyl hydride 27.
the CH bond oxidative addition to Pt(II) center, with the
Importantly, the CNN-coordinated aqua complex 7 can form
transition state TS23‑24 Gibbs energy of only 22.3 kcal/mol.
26 via a lower energy path with an overall activation barrier of
I
DOI: 10.1021/acs.organomet.6b00613
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Scheme 10. DFT Calculated Gibbs Energy Profile for
Benzene CH Bond Activation by LPt (solv) Species
Supported by a C,N,N-Coordinated Pincer Ligand L in Neat
TFE at 25 °C
energy than TS26‑27 (Scheme 10), the lowest energy path for
benzene CH activation in our system appears to be by complex
7 via the transition states TS23‑24 and TS26‑23 (Scheme 10) and
TS23‑24 (Scheme 9) with an overall barrier of 22.3 kcal/mol for
C H and 23.1 kcal/mol for C D . The latter value is a close
II
6
6
6
6
enough match to the experimentally determined effective
⧧
ΔG298 = 24.2 kcal/mol for this substrate.
IV
The predicted reaction path suggests that the Pt hydride 24
II
and the zwitterionic Pt phenyl complex 25 are also expected to
appear as prominent transient products of benzene CH
activation in reaction 2 catalyzed by 7.
2
.5. Oxidation of 7 by O in 2,2,2-Trifluoroethanol
2
Solutions. Solutions of 7 in TFE are sufficiently stable under
air and show no signs of degradation for at least 10 h at 21 °C.
Under pure oxygen gas at a partial pressure of 2.4 bar and
vigorous stirring under ambient light the TFE solutions of 7
react with O with 41% conversion after 5 days to form a single
2
IV
new product which was identified as the Pt trifluoroethoxide
complex 15 (Scheme 3). No reaction was observed when an
O -pressurized reaction mixture was left in the dark for 5 days.
2
Our preliminary kinetics observations suggest that the oxidation
20
reaction is second order in concentration of 7. Single crystals
of 15·TFE suitable for X-ray diffraction analysis (Figure 7)
could be grown from TFE−ether mixtures. In this product the
sulfonated pincer ligand adopts a meridional C,N,O arrange-
ment of three donor atoms.
only 17.7 kcal/mol (TS7‑26), in comparison to the reaction of
the TFE analogue 10 (TS10‑26). Hence, DFT predicts the
predominant involvement of 7 in the benzene CH activation.
This conclusion is consistent with the results of our modeling
of the reaction 2 kinetics in the presence of water additives
(Scheme 8, eq 4). Notably, the reaction sequence involving 7
and TS7‑26 has a rate-limiting CH bond cleavage step (TS26‑27),
which is also consistent with our experimental observations.
The noticeably lower energy of the benzene for aqua ligand
substitution transition state TS7‑26, according to our analysis,
may be due to the participation of the pincer ligand sulfonate
group in two hydrogen-bonding interactions with two protons
of the “departing” aqua ligand, which are absent in the ground
state (complex 7) and which help “pull out” the aqua ligand
from the metal coordination sphere. This effect is unique for
TS7‑26, in comparison to other benzene for solvento ligand
exchange transition states shown in Schemes 9 and 10.
To complete the analysis of reactions shown in Scheme 10, it
is worth adding a few more comments on the facile
isomerization of the benzene σ-CH complex 26 having CNN
coordination of the pincer ligand (Scheme 10) to its CNO-
coordinated analogue 23 (Scheme 9) via the low-energy
transition state TS26‑23. This transformation leaves the ligand−
metal bonding of the σ-coordinated arene virtually intact as the
intramolecular sulfonate for pyridyl fragment substitution
occurs. The transition state TS26‑23 energy of 19.9 kcal/mol is
lower than that of either of the CH bond oxidative addition
transition states, TS23‑24 (Scheme 9) and TS26‑27 (Scheme 10).
Since the transition state TS23‑24 (Scheme 9) is slightly lower in
Figure 7. ORTEP drawing (50% probability ellipsoids) for 15.
Hydrogen atoms, except that of the OH group, are omitted for clarity;
the cocrystallized solvent is not shown.
1
1
1
The results of our H− H COSY and H NOE NMR
2
0
experiments are consistent with the retention in TFE
solutions of the structure found in the solid state.
Considering the possible mechanism of formation of 15, it is
worth noting that this complex has the same meridional
arrangement of C, N, and O donor atoms as in the complexes
13 and 14, which may be expected to be involved in the
oxidation reaction in Scheme 3. It is our experimental
10
observation and a result of a previous theoretical (DFT)
II
analysis of the mechanism of aerobic oxidation of (dpms)Pt R
11
II
complexes that reactions between O2 and (dpms)Pt R
species usually occur without altering the donor atom
II
arrangement in the Pt coordination plane. Therefore, 13
and/or 14 (Scheme 2) could be directly involved in the
photochemical oxidation. These complexes are also expected to
be more reactive toward O in comparison to 10, 11, or 12
2
since they bear a formal negative charge on the metal, as
J
DOI: 10.1021/acs.organomet.6b00613
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
opposed to the case for 10, 11, and 12, an important factor
As a continuation of this work, we expect that the novel
sulfonated pincer ligand and its analogues will play an
important role in the development of new Pt-based catalysts
for selective aerobic CH bond functionalization.
25
affecting this type of reactivity. When distinguishing between
3 and 14 as the most likely direct participants of the reaction
1
with O , it is worth mentioning that 14 is expected to be the
2
more acidic of these two and would produce reactive anionic
II
25
Pt OCH CF intermediates more readily. Hence, we
2
3
4. EXPERIMENTAL SECTION
hypothesize that complex 14 is directly involved in the
oxidation reaction, as shown in Scheme 3. An alternative
reaction sequence when 13 is oxidized first and subsequently
one of the resulting OH ligands of the anticipated Pt
dihydroxo analogue of the complex 15 is selectively displaced
4.1. General Considerations. All manipulations were carried out
under an argon atmosphere unless otherwise noted. All reagents for
which the synthesis is not given were purchased from Aldrich, Acros,
Alfa-Aesar, TCI America, or Pressure Chemicals and were used
without further purification except where stated. Deuterium-labeled
solvents were purchased from Cambridge Isotope Laboratories (C D ,
IV
by CF CH O seems unlikely. The kinetics of such ligand
3
2
6
6
IV
1
exchange reactions at a Pt center are usually very slow, and
the observed exceptionally selective formation of 15 would be
difficult to rationalize. The spin-forbidden character of this
transformation and its photochemical nature require additional
efforts in the mechanistic investigation of the transformation in
Scheme 3. An experimental investigation and theoretical
analysis of this and similar aerobic oxidation reactions is
ongoing in our laboratory.
CDCl
, CD OD, and DMSO-d ) or Sigma-Aldrich (CF CH OD). H
3 3 6 3 2
13
(400 MHz) and C NMR (125 MHz) spectra were recorded on a
Bruker AVANCE 400 or Bruker DRX-500 instrument. Chemical shifts
are reported in parts per million (ppm) (δ) and referenced to residual
solvent resonance peaks. Multiplicities are reported as follows: br
(
(
(
broad signal), s (singlet), d (doublet), t (triplet), q (quartet), quin
quintet), sex (sextet), m (multiplet), dd (doublet of doublets), ddd
doublet of doublets of doublets), qd (quartet of doublets). Coupling
constants (J) are reported in Hz. High-resolution mass spectrometry
HRMS) experiments were performed using a JEOL AccuTOF-CS
instrument.
(
3
. CONCLUSIONS
4
.2. Li(Ph-dpms) (8). 4.2.1. (6-Phenylpyrid-2-yl)(2-pyridyl)-
The design of the novel sulfonated CNN pincer ligand L and
derived Pt complexes was shown to be successful: a derived
aqua complex was demonstrated experimentally and computa-
tionally (DFT) to be a kinetically competent catalyst for H/D
exchange reactions between protio- and deuteriobenzene, some
methane. A solution of 2-phenyl-6-methylpyridine (1.69 g, 10
mmol) in 20 mL of dry THF was cooled to −78 °C in a 50 mL
Schlenk flask using a dry ice−acetone bath. n-BuLi (2.5 M, 4.0 mL, 10
mmol) was slowly added to the stirred solution, resulting in a color
change from light yellow to dark red. After the solution was stirred for
an additional 1 h, 2-fluoropyridine (0.58 g, 6 mmol) was added
dropwise via syringe. The dry ice bath was subsequently removed, and
the reaction mixture was warmed to room temperature and stirred
overnight. A 30 mL portion of water was added to quench the
reaction. The organic and aqueous layers of the resulting yellow
II
(
poly)methylarenes, and TFE-d or TFE. The new pincer ligand
II
IV
also supports a clean Pt to Pt aerobic oxidation chemistry
which is manifested by a slow photochemical oxidation of the
H/D exchange reaction catalyst. The latter fact may be
considered as a limitation of this first-generation sulfonated
pincer ligand based system, but the ability to turn off the
solution were separated, the aqueous layer was extracted with Et O (3
2
×
30 mL), and the combined organic phases were dried over MgSO4.
reaction in the dark may be beneficial for reaction control.
Interestingly, the most stable of the LPt (H O) derivatives
present in the catalyst TFE solutions appears to be a
zwitterionic (HL)Pt (OH) compound featuring a CNO
coordination to Pt of the sulfonated pincer ligand L and the
presence of a protonated pendant pyridine NH group on the
ligand. In spite of the relatively high abundance of the
zwitterionic derivative in the TFE solutions at 21 °C, the
species most active in the benzene CH bond cleavage is the
CNN-coordinated LPt(H O) complex, which is connected
through a series of low-barrier reaction steps to other LPt(solv)
and (HL)Pt (OH) species. The CH bond activation reaction
by LPt(H O) is characterized by a noticeable deuterium kinetic
isotope effect k /k of 1.7, which suggests the realization of a
rate-determining substrate CH bond cleavage and a low steric
crowding around the metal center in the reactive LPt(H O)
species. The latter conclusion also follows from the analysis of
the CH bond cleavage selectivity in toluene, p-xylene, and
mesitylene; in all of these cases exclusive aromatic CH bond
activation was observed. Computationally it was also shown
that the kinetically accessible reaction intermediates resulting
from the substrate CH bond cleavage are isomeric Pt phenyl
hydrides and a derived zwitterionic Pt phenyl complex
featuring a mer-CNO arrangement of the pincer ligand donor
atoms. The DFT analysis suggests that these products are
accessible through a benzene reaction with the LPt (H O)
having the CNN coordinated pincer ligand through the CNN
to CNO isomerization of the transient LPt(σ-benzene)
complexes and subsequent CH bond cleavage in the CNO
isomer via the lowest energy CH bond cleavage transition state.
After filtration, the solvents were removed under reduced pressure,
and the residual oil was purified by flash chromatography (silica,
hexanes/EtOAc/TEA (1/1/0.01)) to yield 0.70 g of the product (57%
II
2
1
II
yield based on 2-fluoropyridine used). H NMR (400 MHz, 22 °C,
3
4
II
CDCl ): δ 4.66 (s, 2H), 7.17 (ddd, J = 7.6, 4.8 Hz, J = 1.6 Hz,
3 HH HH
3
3
+
1H), 7.44 (d, J = 8.0 Hz, 1H), 7.52 (d, J = 8.0 Hz, 1H), 7.62−
3 3
HH
HH
7
.76 (m, 4H), 7.79 (d, J = 8.8 Hz, 1H), 7.91 (dd, J = 7.2 Hz,
HH HH
4
3
4
JHH = 2.0 Hz, 1H), 8.60 (ddd, J = 4.8, 2.0 Hz, J = 0.8 Hz, 1H),
HH
HH
3
4
13
9
.36 (dd, J = 8.0 Hz, J = 0.6 Hz, 1H). C NMR (125 MHz, 22
HH HH
°
C, DMSO-d ): δ 47.1, 118.3, 122.0, 122.6, 124.0, 127.0, 129.1, 129.4,
6
2
137.0, 138.1, 139.1, 149.5, 155.9, 159.6, 159.8. ESI-MS: solution in
+
+
methanol, [M + H] 247.137; calcd for [M + H] C H N , 247.124.
17 15 2
II
4.2.2. Li(Ph-dpms) (8). (6-Phenylpyrid-2-yl)(2-pyridyl)methane
(0.70 g, 2.8 mmol) was mixed with 3 mL of dry THF in a 25 mL
Schlenk flask under argon. The solution was cooled to −78 °C, and 2.5
M n-BuLi (1.1 mL, 2.8 mmol) was added dropwise over 10 min with
stirring. The resulting dark red solution was stirred for an additional 1/
2
H
D
2
2
h before being removed from the cold bath and brought into an
argon atmosphere glovebox. NMe ·SO complex (0.39 g, 2.8 mmol)
3
3
was added. The Schlenk flask was then closed with a Teflon seal, and
the mixture was heated in an oil bath for 24 h at 120 °C with rapid
stirring. The next day a large amount of precipitate had formed. The
mixture was then cooled to 5 °C, and the reaction was quenched with
10 mL of deionized water. The product was extracted in water and
washed with diethyl ether (3 × 10 mL). The water layer was
concentrated, and the residue was dried under vacuum. The crude
product was dissolved in dry trifluoroethanol and filtered through a
paper filter to remove inorganic impurities. The filtrate was
concentrated and dried. The resulting residue was purified by flash
chromatography (silica, EtOAc/CH Cl /MeOH/Et N (2/2/1/0.01)).
IV
II
II
2
2
2
3
The eluent was treated with K CO to convert any triethylammonium
2
3
salts to TEA and then filtered and dried to yield 0.56 g of the product
(61% yield). The formation of the disulfonated product can be
K
DOI: 10.1021/acs.organomet.6b00613
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
minimized by running the reaction at high concentrations (≥0.5 M),
126.9, 129.3, 130.7, 134.5, 140.5, 141.2, 140.0, 145.9, 152.6, 152.7,
−
which promotes precipitation of the desired monosulfonated product.
153.7. ESI-MS: dilute solution of 7 in H O, [2-H] , 536.028; calcd for
2
1
3
−
H NMR (400 MHz, 22 °C, MeOD): δ 5.80 (s, 1H), 7.34 (ddd, J
[7-H] , C H N O PtS, 536.025. Anal. Calcd for C H N O PtS·
4H O: C, 33.50; H, 3.64; N, 4.60. Found: C, 33.47; H, 3.79; N, 4.58.
HH
17 13
2
4
17 14
2
4
4
3
=
7.2, 4.8 Hz, J = 0.8 Hz, 1H), 7.38−7.45 (m, 3H), 7.75 (dd, J
=
2
HH
HH
4
7
8
1
1
1
3
.6 Hz, J = 0.8 Hz, 1H), 7.82−7.92 (m, 3H), 8.00−8.02 (m, 2H),
4.5. VT NMR Experiments. An NMR Young tube equipped with a
sealable Teflon cap and a sealed capillary tube containing D O were
taken into an argon-filled glovebox and charged with 0.5 mL of a TFE
solution containing 7·2H O (13.1 mM), C (0 or 1.13 M), and
O (60 mM from the hydrate). The solution was monitored by H
NMR over 21−80 °C. The bridging methine CH singlets at 5.66 and
6.01 ppm were used to monitor the ratio between the two (C H -
6 4
HH
3
3
4
.26 (d, J = 8.0 Hz, 1H), 8.47 (dd, J = 4.0 Hz, J = 0.4 Hz,
2
HH
HH
HH
H). 13C NMR (125 MHz, 22 °C, MeOD): δ 73.6, 117.3, 121.1,
21.6, 124.0, 125.1, 126.7, 127.0, 135.0, 135.5, 137.8, 146.3, 154.0−,
54.7, 154.9. ESI-MS: solution of Li(Ph-dpms) in water, [Ph-dpms]
D
6 6
2
1
H
2
−
25.085; calcd for [Ph-dpms] , C H N O S, 325.065.
17
13
2
3
4
.3. K[(C H -dpms)PtCl] (9). Li(Ph-dpms) (100 mg, 0.3 mmol)
6
4
and K PtCl4 (125 mg, 0.3 mmol) were dissolved in 4.0 mL of
dpms)Pt(solv) species in solution at 40 °C and above (Figure S4 in
the Supporting Information). In the 21 °C experiments extreme
broadening of the 6.01 ppm signal was observed and the LPt(solv)
ratio was found by integrating the whole aromatic region of the
2
deionized H O and transferred to a 25 mL Schlenk tube. The red
2
solution was diluted with 8.0 mL of glacial acetic acid, and a stir bar
was added. The tube was sealed and brought to 100 °C for 18 h with
stirring. The resulting yellow solution was concentrated and dried
under vacuum, and the residue was extracted with warm
trifluoroethanol (8 mL). The product was obtained as a 1:1 mixture
of 9 and LiCl by removing trifluoroethanol (163 mg, 91%), appearing
as a yellow solid. This mixture could be used in the preparation of 7.
spectrum (assignable to both solv = TFE and H
multiplet standing alone at 5.66 ppm was assigned to 13. Virtually the
same LPt(TFE)/LPt(H O) ratios were found whether the temper-
O species); a sharp
2
2
ature was ramping up or down (demonstrating reversibility) and
whether or not benzene was present or absent in the mixtures. The
reported Keq values are given in Table S3 in the Supporting
Information.
To remove LiCl, the yellow solid was recrystallized from H O/EtOH.
2
The yellow crystalline powder was filtered off, washed with cold
EtOH, and then dried in a vacuum oven at 60 °C for 12 h (137 mg,
4.6. General Procedure for H/D Exchange Reactions. In a
7
7%). XRD-quality crystals of 9 were prepared by slow evaporation of
scintillation vial was placed 7·4H
was then brought into an argon-filled glovebox. Degassed TFE (0.45
mL) and C (0.05 mL) were then added. After about 10 min all had
2
O (4.0 mg, 6.7 μmol), and the vial
1
a saturated methanolic solution. H NMR (125 MHz, 22 °C, DMSO-
3
d ): δ 5.933 (s, 1H), 7.05 (m, 2H), 7.49−7.57 (m, 2H), 7.61 (dd, J
D
6 6
6
=
HH
4
3
7.1 Hz, J = 1.9 Hz, 1H), 7.81 (vd, J = 7.7 Hz, 1H), 7.93 (vd,
dissolved to form a reddish brown solution, which was then transferred
to an NMR Young tube equipped with a sealable Teflon cap. For
HH
HH
3
3
3
JHH = 7.8 Hz, 2H), 8.04 (vt, J = 7.98 Hz, 1H), 8.07 (ddd, J
=
HH
HH
4
3
4
locking and shimming purposes, a D O-containing capillary tube that
9
1
1
1
5
.5, 7.6 Hz, J = 1.9 Hz, 1H), 9.6 (dd, J = 6.0 Hz, J = 1.3 Hz,
2
HH
HH
HH
H). 13C NMR (400 MHz, 22 °C, DMSO-d ): δ 76.6, 117.8, 123.3,
had been flame-sealed on both ends was inserted. The NMR tube was
6
removed from the glovebox and charged with 15 psi of argon on a
23.7, 124.0, 126.7, 128.5, 129.6, 135.2, 138.0, 138.1, 143.0, 145.7−,
1
Schlenk line to prevent refluxing at 80 °C. After an initial H NMR
49.8, 152.3, 153.1, 166.8. ESI-MS: solution of 1 in water, [1-K]
−
spectrum was taken, the NMR tube was submerged in a preheated
53.973; calcd for [9-K] , C H ClN O PtS 553.990. Anal. Calcd for
17
12
2
3
1
bath at 80 °C and periodically removed for H NMR analysis. All
C H ClKN O PtS·2H O: C, 32.41; H, 2.56; N, 4.45. Found: C,
1
7
16
2
5
2
reactions were run at least twice. The TFE CH quartet at 3.9 ppm was
3
2.35; H, 2.84; N, 4.46.
.4. (C H -dpms)Pt(H O) (7). In a Schlenk flask was placed
2
used as an internal standard to monitor the growth of the benzene CH
singlet at 7.48 ppm (Figure S7 in the Supporting Information). To
track the disappearance of 13, either the ortho pyridyl CH doublet at
4
6
4
2
K[(C H -dpms)PtCl] (9; 200 mg, 0.337 mmol), along with 10 mL of
6
4
EtOH. Then KOH pellets (∼85% KOH, ∼15% H O, 115 mg, 1.75
2
8
.87 ppm or the bridging methine singlet at 5.74 ppm was monitored.
All reactions were run at least twice for at least 20 h. A control reaction
containing only 10% v/v C D in TFE showed no detectable H/D
mmol, ∼5 equiv) were added. The flask was sealed and heated to 60
°
C with vigorous stirring overnight. The resulting dark red solution
containing 16 was then filtered through Celite. When 9 contaminated
6
6
exchange.
.7. Oxidation of 7 in TFE Solutions To Form 15. Only the
reactions run under ambient fluorescent lighting showed oxidation of 7
with LiCl was used as the starting material, the solution was then
+
4
passed through a K cation exchange column and filtered again
followed by solvent removal in vacuo. The dark red residue containing
1
to 15. In a 25 mL Schlenk tube was placed 7·4H O (40 mg, 66 μmol)
crude 16 was characterized by H NMR and used without purification.
2
dissolved in 5 mL of TFE. A stir bar was added, followed by purging of
The residue was redissolved in 3 mL of EtOH, diluted to 9 mL with
DCM, and then placed in the refrigerator. After 1 h a light-colored
precipitate had formed, which was removed by filtration through
Celite. The filter cake was washed with 2 × 5 mL of a 2/1 DCM/
EtOH solution. The filtrate was stripped of solvent under vacuum ,and
the red residue was dispersed in 10 mL of deionized argon-sparged
the tube headspace with O and pressurization to 20 psi of O before
2
2
being sealed. The solution was stirred vigorously at room temperature
over the course of 5 days. The mixture initially turned red and then
slowly turned yellow. A 0.5 mL aliquot was removed and submitted to
1
a H NMR analysis in DMSO-d , which showed one new species,
6
identified as complex 15 (vide infra) in a 0.7:1 ratio with (C H -
H O to which a few drops of TFE were added until all had dissolved.
6
4
2
dpms)Pt(DMSO-d ) (41% NMR yield of 15). The TFE solution was
With vigorous stirring, the pH was adjusted to 2 using a 1 M H SO4
6
2
dried in vacuo to yield a light yellow residue, which was extracted with
solution. The mixture turned light yellow followed by the formation of
a large amount of fluffy precipitate. The flask was placed in the
refrigerator overnight. The next day small yellow crystals had formed
on the walls of the flask which were suitable for XRD along with a
yellow silty precipitate. The solids were filtered off, washed with
deionized H O, and dried under vacuum to yield 7·4H O (146 mg,
3
× 1 mL of MeOH. The MeOH was removed to yield 19 mg of an
off-white powder. The powder was dissolved in 0.5 mL of TFE,
layered with Et O, and placed in the freezer. The next day small thin
2
needle crystals suitable for XRD analysis had formed. The structure of
1
1
1
5 in DMSO-d solution was additionally confirmed by H− H COSY
6
2
2
1
and H NOE NMR experiments and was found to be consistent with
7
3%). Product prepared in this manner usually contained a 1−5%
the observed crystal structure. Samples of complex 15 are stable on
storage in the freezer but slowly decompose at room temperature over
impurity of 9; however, this does not affect the rate of CH activation
by 7. Analytically pure 7·4H O could be obtained by recrystallization
2
1
the course of a few days. H NMR (400 MHz, 22 °C, DMSO-d ): δ
from TFE/H O to yield yellow crystals (105 mg, 51%). Complex 7 is
6
2
2
3
1
.53 (s, 1H), 3.77 (dq, J = 12.7 Hz, J = 10.1 Hz, 1H), 4.01 (dq,
soluble in TFE and DMSO, slightly soluble in MeOH, and insoluble in
HH HF
2
3
JHH = 12.7 Hz, J = 10.1 Hz, 1H), 6.88 (s, 1H), 7.35 (m, 2H), 7.62
(m, 1H), 7.78 (ddd, JHH = 8.1, 5.8 Hz, JHH = 1.3 Hz, 1H), 7.86 (dd,
JHH = 7.2 Hz, JHH = 1.4 Hz, 1H), 7.90 (m, 1H), 8.01 (dd, JHH = 8.0
H O/DCM/THF/acetone. The NMR data below are for the soluble
HF
2
1
3
4
DMSO derivative LPt(DMSO). H NMR (400 MHz, 22 °C, DMSO-
3
4
3
d ): δ 6.25 (s, 1H), 7.24−7.27 (m, 2H), 7.66−7.72 (m, 2H), 7.76 (dd,
6
3
4
3
4
3
J
HH
= 6.0 Hz, J = 2.8 Hz, 1H), 7.88 (m, 1H), 8.00 (d, J = 7.6
Hz, J = 0.7 Hz, 1H), 8.23−8.32 (m, 3H), 8.67 (dd, J = 5.9 Hz,
4 19
HH
HH
HH
HH
3
13
Hz, 1H), 8.18−8.22 (m, 3H), 9.04 (d, J = 4.8 Hz, 1H). C NMR
125 MHz, 22 °C, DMSO-d ): δ 75.3, 119.4, 125.3, 125.7, 125.9,
JHH = 1.5 Hz, 1H). F NMR (376 MHz, 21 °C, DMSO-d ): δ
6
HH
3
−
(
−73.49 (bt, J = 10.1 Hz). ESI-MS: solution of 15 in TFE, [15-H] ,
6
HF
L
DOI: 10.1021/acs.organomet.6b00613
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
−
−
6
33.98; calcd for [15-H] , C H F N O PtS , 634.02. The oxidation
(h) Zhang, F.; Kirby, C. W.; Hairsine, D. W.; Jennings, M. C.;
Puddephatt, R. J. J. Am. Chem. Soc. 2005, 127, 14196−7. (i) West, N.
M.; Templeton, J. L. Can. J. Chem. 2009, 87, 288−96. (j) Vedernikov,
A. N. Curr. Org. Chem. 2007, 11, 1401−16.
19
16
3
2
5
1
of 7 dissolved in TFE was monitored by H NMR spectroscopy.
Complex 7·4H O (4.0 mg, 6.7 μmol) was dissolved in 0.5 mL of TFE
and transferred to an NMR Young tube equipped with a sealable
Teflon cap. After an initial NMR spectrum was taken, the headspace
was flushed with dry O and then pressurized to 20 psi with additional
O (35 psi of O total). The NMR tube was then placed on a rotator
to ensure adequate O /solution mixing. The disappearance of the
starting material 7 was monitored using signals of the solvent as an
internal standard, resulting in the reaction profile shown below. The
rate was found to follow a second-order rate law (Figure S10 in the
Supporting Information).
.8. DFT Calculations. Theoretical calculations in this work have
been performed using the density functional theory (DFT) method,
specifically the functional PBE and LACVP relativistic basis set with
two polarization functions implemented in the Jaguar program
package. Full geometry optimization has been performed without
constraints on symmetry in 2,2,2-trifluoroethanol as a solvent using a
2
(3) For a review on Pt-mediated CH activation see: Lersch, M.;
Tilset, M. Chem. Rev. 2005, 105, 2471−526.
2
2
2
(4) Lin, M. R.; Shen, C. Y.; Garcia-Zayas, E. A.; Sen, A. J. Am. Chem.
Soc. 2001, 123, 1000−1.
2
(5) Dangel, B. D.; Johnson, J. A.; Sames, D. J. Am. Chem. Soc. 2001,
123, 8149−50.
(6) Periana, R. A.; Taube, D. J.; Gamble, S.; Taube, H.; Satoh, T.;
Fujii, H. Science 1998, 280, 560−564.
4
(7) Hartwig, J. F. J. Am. Chem. Soc. 2016, 138, 2−24.
(8) Geletii, Yu. V.; Shilov, A. E. Kinet. Catal. 1983, 24, 413−416.
(9) Bar-Nahum, I.; Khenkin, A. M.; Neumann, R. J. Am. Chem. Soc.
2004, 126, 10236−7.
2
6
27
2
8
(10) Vedernikov, A. N. Acc. Chem. Res. 2012, 45, 803−13.
(11) Liu, W.-G.; Sberegaeva, A. V.; Nielsen, R. J.; Goddard, W. A.;
2
8
Poisson−Boltzmann continuum solvation model (PBF). For all
species under investigation frequency analysis has been carried out. All
energy minima have been checked for the absence of imaginary
frequencies. All transition states possessed just one imaginary
frequency. Using the intrinsic reaction coordinate method, reactants,
products, and the corresponding transition states were proven to be
connected by a single minimum energy reaction path.
Vedernikov, A. N. J. Am. Chem. Soc. 2014, 136, 2335−41.
12) Sberegaeva, A. V.; Liu, W.-G.; Nielsen, R. J.; Goddard, W. A., III;
Vedernikov, A. N. J. Am. Chem. Soc. 2014, 136, 4761−8.
13) Vedernikov, A. N.; Binfield, S. A.; Zavalij, P. Y.; Khusnutdinova,
J. R. J. Am. Chem. Soc. 2006, 128, 82−3.
14) Munz, D.; Wang, D.; Moyer, M. M.; Webster-Gardiner, M.;
(
(
(
Kunal, P.; Watts, D.; Trewyn, B. G.; Vedernikov, A. N.; Gunnoe, T. B.
ACS Catal. 2016, 6, 4584−93.
ASSOCIATED CONTENT
Supporting Information
(15) Khusnutdinova, J. R.; Zavalij, P. Y.; Vedernikov, A. N.
Organometallics 2011, 30, 3392−9.
■
*
S
(16) (a) Zhong, H. A.; Labinger, J. a; Bercaw, J. E. J. Am. Chem. Soc.
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.organo-
met.6b00613.
2
002, 124, 1378−99. (b) Driver, T.; Day, M. W.; Labinger, J. A.;
Bercaw, J. E. Organometallics 2005, 24, 3644.
17) Hickman, A. J.; Cismesia, M. A.; Sanford, M. S. Organometallics
(
Experimental details regarding synthesis and character-
ization of the preligand 8, derived metal complexes,
reactivity studies, kinetics modeling, X-ray character-
ization data for 7, 9, and 15 and DFT calculations (PDF)
Crystallographic data (CIF)
2012, 31, 1761−6.
(18) Ziatdinov, V. R.; Oxgaard, J.; Mironov, O. A.; Young, K. J. H.;
Goddard, W. A.; Periana, R. A. J. Am. Chem. Soc. 2006, 128, 7404−5.
(19) Young, K.; Meier, S.; Gonzales, J.; Periana, R. A. Organometallics
2
(
006, 25, 4734−4737.
20) See the Supporting Information for details.
Cartesian coordinates for the calculated structures (XYZ)
(21) Vedernikov, A. N.; Pink, M.; Caulton, K. G. Inorg. Chem. 2004,
4
3, 3642−6.
AUTHOR INFORMATION
Corresponding Author
E-mail for A.N.V.: avederni@umd.edu.
(22) Khusnutdinova, J. R.; Zavalij, P. Y.; Vedernikov, A. N.
■
Organometallics 2007, 26, 3466−83.
(
(
23) Jones, W. D. Acc. Chem. Res. 2003, 36, 140−6.
24) Johansson, L.; Ryan, O. B.; Romming, C.; Tilset, M. J. Am.
*
Notes
Chem. Soc. 2001, 123, 6579−90.
The authors declare no competing financial interest.
(25) Khusnutdinova, J. R.; Zavalij, P. Y.; Vedernikov, A. N. Can. J.
Chem. 2009, 87, 110−120.
(26) Parr, R. G.; Yang, W. Density-functional Theory of Atoms and
ACKNOWLEDGMENTS
This work was supported by the National Science Foundation
CHE-1464772) and, in part, by the Center for Catalytic
■
(
Molecules; Oxford University Press: Oxford, U.K., 1989.
27) Perdew, J. P.; Burke, K.; Ernzerhof, M. Phys. Rev. Lett. 1996, 77,
865−8.
28) Jaguar, version 8.4; Schro
(
3
Hydrocarbon Functionalization, an Energy Frontier Research
Center Funded by the U.S. Department of Energy, Office of
Science, Office of Basic Energy Sciences, under Award Number
DE-SC0001298.
(
̈
dinger, LLC, New York, 2014.
REFERENCES
■
(
1) (a) Gol’dshleger, N. F.; Es’kova, V. V.; Shilov, A. E.; Shteinman,
A. A. Zh. Fiz. Khim. 1972, 46, 1353−4. (b) Shilov, A. E.; Shulpin, G. B.
Activation and Catalytic Reactions of Saturated Hydrocarbons in the
Presence of Metal Complexes; Kluwer: Boston, MA, 2000.
(
2) (a) Zamashchikov, V. V.; Rudakov, E. S.; Mitchenko, S. A.;
Pekhtereva, T. M. Koord. Khim. 1985, 11, 69−72. (b) Crabtree, R. H.
Chem. Rev. 1995, 95, 987−1007. (c) Arndtsen, B. A.; Bergman, R. G.;
Mobley, T. A.; Peterson, T. H. Acc. Chem. Res. 1995, 28, 154−62.
(
(
d) Shilov, A. E.; Shu'lpin, G. B. Chem. Rev. 1997, 97, 2879−2932.
e) Stahl, S.; Labinger, J. A.; Bercaw, J. E. Angew. Chem., Int. Ed. 1998,
3
1
7, 2180−92. (f) Labinger, J. A.; Bercaw, J. E. Nature 2002, 417, 507−
4. (g) Fekl, U.; Goldberg, K. I. Adv. Inorg. Chem. 2003, 54, 259−320.
M
DOI: 10.1021/acs.organomet.6b00613
Organometallics XXXX, XXX, XXX−XXX