PtII-catalyzed hydrophenylation of α-olefins: Variation of linear/branched products as a function of ligand donor ability

Article abstract of DOI:10.1021/cs400988w

The Pt^II complexes [(^xbpy)Pt(Ph)(THF)]⁺ (^xbpy = 4,4′-X₂-2,2′-bipyridyl; x = OMe (1a), ^tBu (1b), H (1c), Br (1d), CO₂Et (1e) and NO₂ (1f)] catalyze the formation of n-propylbenzene and cumene from benzene and propene. The catalysts are selective for branched products, and the cumene/n-propylbenzene ratio decreases with increasing donor ability of the ^xbpy ligand. DFT(D) calculations predict more favorable activation barriers for 1,2-insertion into the Pt-Ph bond to give branched products. The calculations indicate that 1,2-insertion of propene should be faster than 2,1-insertion for all Pt(II) catalysts studied, but they also indicate that cumene/n-propylbenzene selectivity is under Curtin-Hammett control.

Full text of DOI:10.1021/cs400988w

Research Article
pubs.acs.org/acscatalysis
PtII-Catalyzed Hydrophenylation of α‑Olefins: Variation of Linear/
Branched Products as a Function of Ligand Donor Ability
Bradley A. McKeown,† Bruce M. Prince,‡ Zoraida Ramiro,† T. Brent Gunnoe,*,†
and Thomas R. Cundari*,‡
†Department of Chemistry, University of Virginia, Charlottesville, Virginia 22904, United States
‡Center for Advanced Scientific Computing and Modeling (CASCaM), Department of Chemistry, University of North Texas,
Denton, Texas 76203, United States
S
* Supporting Information
ABSTRACT: The PtII complexes [(xbpy)Pt(Ph)(THF)]+ (xbpy = 4,4′-X2-2,2′-
t
bipyridyl; x = OMe (1a), Bu (1b), H (1c), Br (1d), CO2Et (1e) and NO2 (1f)]
catalyze the formation of n-propylbenzene and cumene from benzene and propene.
The catalysts are selective for branched products, and the cumene/n-propylbenzene
ratio decreases with increasing donor ability of the xbpy ligand. DFT(D) calculations
predict more favorable activation barriers for 1,2-insertion into the Pt−Ph bond to
give branched products. The calculations indicate that 1,2-insertion of propene
should be faster than 2,1-insertion for all Pt(II) catalysts studied, but they also
indicate that cumene/n-propylbenzene selectivity is under Curtin−Hammett
control.
KEYWORDS: platinum, hydroarylation, α-olefins, C−H activation, Curtin−Hammett principle
favor the formation of the anti-Markovnikov over Markovnikov
addition products with moderate selectivity (∼1.5:1 linear-to-
branched ratio for both catalysts).1a,3
INTRODUCTION
■
Transition-metal-catalyzed hydroarylation of substituted olefins
by a nonacidic pathway (i.e., non-Friedel−Crafts catalysis)
offers the opportunity to control the regioselectivity of olefin
insertion into the M−aryl bond to obtain either Markovnikov
or anti-Markovnikov products.1 For example, catalysts that can
bias 2,1- over 1,2-insertion with α-olefins could selectively
produce linear alkyl arenes that are not accessible with acid-
based methodologies (Scheme 1).2 The hydrophenylation of α-
olefins using Ru or Ir catalyst precursors has been reported to
Unlike the Ir and Ru catalysts, product selectivity for PtII-
catalyzed hydrophenylation of propylene favors Markovnikov
addition products. Goldberg and co-workers reported that a
(pyridyl)pyrrolide ligated PtII catalyst precursor catalyzes the
formation of cumene and n-propylbenzene from benzene and
propylene in an ∼6:1 ratio.4 In addition, we have reported that
t
[(tbpy)Pt(Ph)(THF)][BAr′4] (1b; bpy = 4,4′-di-tert-butyl-
2,2′-bipyridine; Ar′ = 3, 5-(CF3)2−C6H3) catalyzes the
formation of cumene and n-propylbenzene in an ∼3:1 ratio
(Chart 1).5 Vitagliano and Tesauro studied the stoichiometric
C−C bond formation from cationic PtII(aryl)(olefin) (aryl =
C6H5, 4-C6H4R, or 3-C6H4R; R = OMe or Me) complexes
containing N,N bidentate ligands.6 These complexes effectively
mediate migratory insertion of H2CCHR (R = H, Me, or Ph)
into the Pt−Caryl bond. Consistent with the results from PtII-
catalyzed propene hydrophenylation, 1,2-insertion and sub-
sequent protonolysis yields the branched product.
Scheme 1. Product Selectivity from Transition Metal
Mediated Hydrophenylation of α-Olefins Can Be Dictated
by the Selectivity of Insertion into the M−Ph Bond
The proposed mechanism for transition-metal-catalyzed
olefin hydrophenylation involves two principal steps: olefin
insertion into the M−Ph bond and benzene C−H activation
(Scheme 2).1,3,4,7 The PtII complex 1b has been demonstrated
Received: October 29, 2013
Revised: April 7, 2014
Published: April 22, 2014
© 2014 American Chemical Society
1607
dx.doi.org/10.1021/cs400988w | ACS Catal. 2014, 4, 1607−1615

ACS Catalysis
Research Article
Chart 1. Catalysts Studied for Hydroarylation of Substituted
Olefins
Table 1. Hydrophenylation of Olefins with
a
[(tbpy)Pt(Ph)(THF)][BAr′4] (1b).
Scheme 2. Generic Catalytic Cycle for Transition-Metal-
Catalyzed Olefin Hydroarylation Proposed for Ru-, Ir- and
Pt-based Catalyst Precursors
a
0.01 mol % catalyst (1 mM) dissolved in C6H6 with hexamethyl-
b
benzene as an internal standard at 100 °C. Ratio of branched to linear
c
d
isomer after 4 h. 0.1 MPa C3H6; see ref 10. 150 equiv relative to Pt.
e
f
Turnovers after 4 h as determined by GC/MS. Numbers in
parentheses are turnovers after 16 h.
for a nonacid-catalyzed mechanism because acid-based catalysts
are highly selective for Markovnikov addition.2
Increasing the length of the carbon chain upon substitution
of 1-pentene for propylene has a minimal influence on catalyst
efficiency and selectivity. Using 150 equiv of 1-pentene (relative
to 1b), 40 TO of 2-phenylpentane and n-pentylbenzene were
observed after 4 h at 100 °C in a ∼3:1 ratio with no further
activity observed after this time point. The formation of
approximately 1 equiv (relative to catalyst) of 3-phenylpentane
during the hydrophenylation of 1-pentene using complex 1b
indicates that olefin isomerization occurs but is not significantly
competitive on the time scale of catalysis. The catalyst is also
compatible with disubstituted olefins, as the reaction with
cyclohexene results in ∼20 TO of phenylcyclohexane after 16 h
at 100 °C.
to insert ethylene into the Pt−Ph bond in the presence of
excess olefin to yield [(tbpy)Pt(CH2CH2Ph)(η2-C2H4)]+,
which is the proposed resting state of catalytic ethylene
hydrophenylation.7h,8 We recently reported that expansion of
the chelate ring size using [(dpm)Pt(Ph)(THF)][BAr′4] (2;
dpm = di(2-pyridyl)methane) results in an increase in the
catalytic efficiency for ethylene hydrophenylation.8a
The efficiency of catalytic hydrophenylation of substituted
olefins using complex 2 is reduced compared to complex 1b.9
Using 2, propylene hydrophenylation results in 11.8 TO of
cumene and n-propylbenzene after 4 h in a 4.4:1.0 ratio.
Catalysis with 1-pentene and cyclohexene using 2 also results in
a reduction in TO compared to the 2,2′-bipyridyl complex 1b
(Schemes 3 and 4). Although 2 is a more effective catalyst for
ethylene hydrophenylation,8a its efficacy for hydrophenylation
of α-olefins is reduced relative to 1b. We believe that the
reduced activity of 2 relative to 1b is a result of the increased
As previously reported,8b the donor ability of bipyridyl
ligands for PtII catalyzed ethylene hydrophenylation influences
the selectivity for ethylbenzene vs styrene, and we sought to
determine the extent to which ancillary ligand donor ability
could be used to alter selectivity for hydrophenylation of α-
olefins. In this report, catalytic activity is evaluated using longer-
chain and internal olefins as well as Michael acceptors for
complexes 1b and 2. The impact of 2,2′-bipyridyl ligand donor
ability on PtII-catalyzed hydroarylation of α-olefins is examined
using various 2,2′-bipyridyl ligands with different 4,4′
substituents (Chart 1). A clear trend for the influence of bpy
donor ability has been established.
t
steric bulk of the dpm ligand compared to bpy.
Scheme 3. Comparison of Catalytic Hydrophenylation of 1-
Pentene (150 equiv relative to Pt) Using Complexes 1b and
2 (0.01 mol % relative to benzene)
RESULTS AND DISCUSSION
■
The hydroarylation of substituted olefins was studied using
complexes 1b and 2 as catalyst precursors. Under conditions of
0.01 mol % catalyst relative to benzene (1 mM Pt and 11 M
benzene) and 0.1 MPa of propylene at 100 °C, catalysis with
complex 1b results in 33.5 (4 h) and 39.8 (16 h) turnovers
(TO) of cumene and n-propylbenzene in an approximate 3:1
ratio (Table 1). Traces of isomers of di- and tripropylbenzenes
were also detected. Although the formation of the Markovnikov
addition product cumene is favored, the observation of
significant quantities of the anti-Markovnikov addition product
n-propylbenzene (∼10 TO after 16 h) provides further support
1608
dx.doi.org/10.1021/cs400988w | ACS Catal. 2014, 4, 1607−1615

ACS Catalysis
Research Article
phenylpropanoate and 4.9 TO of methyl cinnamate (11.7 total
TO). This corresponds to a 1.4:1 ratio of saturated to
unsaturated products. No further catalytic activity was observed
with continued reaction.
Scheme 4. Comparison of Catalytic Hydrophenylation of
Cyclohexene (150 equiv relative to Pt) Using Complexes 1b
and 2 (0.01 mol %, 1 mM, relative to benzene)
The same reaction was performed using methyl methacrylate
to determine if the presence of an α-methyl group would
inhibit β-hydride elimination after insertion into the Pt−Ph
bond and increase selectivity for the saturated addition product.
Indeed, the dominant product from catalysis was 2-methyl-3-
phenylpropanoate, which was identified by GC/MS and
1
comparison after catalysis of the H NMR spectrum of the
reaction mixture to reported NMR data.11 Only trace amounts
of 2-methyl-3-phenyl-2-propenoate were observed. Because of a
lack of analytically pure material, linear regressions for
quantification were not performed, but using the ratio of
product to internal standard peak areas, ∼7 TO of 2-methyl-3-
phenylpropanoate is estimated from catalysis with 1b and
methyl methacrylate in benzene. Catalysis with methyl vinyl
ketone (MVK) or cyclohexenone is less efficient than the
analogous reactions with the ester derivatives. For MVK,
increased temperature (120 °C) was required to observe the
addition product 4-phenylbutan-2-one, but only in near
stoichiometric amounts. Approximately 2 TO of 3-phenyl-
cyclohexanone is observed after 16 h at 100 °C for the
hydrophenylation of cyclohexenone.
Intramolecular olefin hydroarylation was attempted with
allylbenzene to form indane (eq 1). Heating complex 1b or 2 in
neat allylbenzene at 100 °C resulted in trace amounts of indane
(observed by GC/MS). The small amount of cyclized product
is not surprising because successful catalysis requires activation
of an ortho C−H bond of allylbenzene, and product
distributions from ethylene hydroarylation using substituted
benzenes demonstrate that ortho C−H activation is not favored
relative to meta or para activation.7h In addition, cationic PtII
complexes containing phenyl ligands have been reported to be
active for the isomerization of allylbenzene to the internal olefin
trans-β-methyl-styrene.10 Analysis of the mixture from reaction
Use of 4,4′-Substituted Bipyridyl Ligands. The series of
complexes [(xbpy)Pt(Ph)(THF)][BAr′4] [xbpy = 4,4′-X2-2,2′-
t
bipyridyl; X = OMe (1a), Bu (1b), H (1c), Br (1d), CO2Et
(1e), and NO2 (1f)] was screened for propylene hydro-
phenylation to evaluate the influence of ligand donor ability on
product selectivity (Table 3). Catalysis with complexes 1a−1c
demonstrates a slight enhancement in catalytic efficiency
compared with catalysis with 1d−1f, which is the same trend
in catalyst efficiency that has been observed for ethylene
hydrophenylation.8b Catalyst precursors 1d−1f exhibit a greater
predilection for the formation of unsaturated products (i.e.,
t
of both the bpy complex 1b and the dpm complex 2 with
allylbenzene reveals that both of these Pt complexes catalyze
olefin isomerization. Monitoring the reaction of 1b (20 mol %
1
relative to substrate) and allylbenzene in CD2Cl2 by H NMR
spectroscopy shows complete conversion of allylbenzene to
trans-β-methyl-styrene in ∼1 h at room temperature (eq 1).
Functionalized Olefins. Hydrophenylation of α,β-unsatu-
rated carbonyl compounds by complex 1b was performed to
probe catalyst tolerance toward heteroatomic functionality
(Table 2). In all cases, catalysis was selective for the anticipated
Michael addition product. Catalysis in benzene for 4 h with
0.01 mol % 1b and 150 equiv of methacrylate (relative to 1b) at
100 °C results in the formation of 6.8 TO of methyl 3-
Table 3. Comparison of Catalytic Propylene
a
Hydrophenylation Using Complexes 1a−1f.
Table 2. Hydrophenylation of Michael Acceptors Using
a
[(tbpy)Pt(Ph)(THF)][BAr′4] (1b)
a
0.01 mol % catalyst (1 mM) dissolved in C6H6 with 150 equiv of
olefin (relative to Pt loading) and hexamethylbenzene as an internal
b
a
standard at 100 °C. Turnovers after 4 h as determined by GC/MS.
0.01 mol % catalyst dissolved in C6H6 with hexamethylbenzene as an
c
d
b
Numbers in parentheses are turnovers after 16 h. Turnovers
internal standard at 100 °C with 0.1 MPa C3H6. Ratio of cumene to
c
estimated from GC peak area ratios between product and internal
standard. Catalysis performed at 120 °C.
n-propylbenzene after 4 h. Turnovers after 4 h as determined by GC/
e
d
MS. Numbers in parentheses are turnovers after 16 h.
1609
dx.doi.org/10.1021/cs400988w | ACS Catal. 2014, 4, 1607−1615

ACS Catalysis
Research Article
isomers of (propenyl)benzene), as observed by GC/MS,
relative to 1a−1c. After 16 h at 100 °C, the nitro-substituted
complex 1f gives 8.1 TO of propylbenzenes with a 4.6:1.0
branched/linear ratio. In contrast, ethylene hydrophenylation
using 1f gives an approximately stoichiometric yield of
ethylbenzene and styrene.8b The 1,2-insertion of propylene
Scheme 5. Comparison of Branched to Linear Ratios of
Propylene (0.1 MPa) Hydroarylation Using Substituted
Benzenes
into the Pt−Ph bond forms the intermediate [(NO bpy)Pt-
2
(CH2CH(Me)Ph)]+, and the presence of the β-methyl group
may result in a difference in activation barrier for β-hydride
elimination relative to elimination of styrene from [(NO bpy)-
2
Pt(CH2CH2Ph)]+.12
The ratio of cumene/n-propylbenzene is influenced by the
donor ability of the bipyridyl ligand. For example, catalysis
using complex 1a (OMe, σp = −0.27) and 0.1 MPa of
propylene (100 °C) results in a cumene/n-propylbenzene ratio
of 2.9 (after 4 h) compared with 4.6 for complex 1f (NO2, σp =
0.78). A Hammett plot was constructed using product ratios (n-
propylbenzene vs cumene), and the Hammett parameter, σp
(Figure 1). Although the effects of substituted pyridyl ligands
Functional groups on benzene exert little influence on
selectivity for catalytic propylene hydroarylation; however, the
substitution of benzene for furan biases the reaction to favor the
Markovnikov addition product by a substantial amount
(Scheme 6). After 4 h at 100 °C, complex 1b catalyzes the
Scheme 6. Ratio of 2-Isopropylfuran and 2-n-Propylfuran
from Propylene (0.1 MPa) Hydroarylation with Furan and
[(tbpy)Pt(Ph)(THF)]+ at 100 °C after 4 h, as Determined
from Peak Area Ratios from Analysis by GC/MS
formation of 2-isopropylfuran and 2-n-propylfuran in an
approximate 44:1 ratio, as determined by comparison of peak
areas from GC/MS analysis.
Figure 1. Hammett plot using the ratios of cumene to n-
propylbenzene from [(xbpy)Pt(Ph)(THF)]+-catalyzed propylene
hydrophenylation after 4 h at 100 °C with 0.1 MPa of ethylene
(slope = 0.2, R2 = 0.94).
Computational Studies. Density functional theory (DFT-
(D)) was employed to probe hydroarylation of α-olefins
catalyzed by [(xbpy)Pt(Ph)]+ complexes (x = tBu, NO2 or Br).
The Gaussian 0914 package was used to perform all simulations.
All stationary points were obtained using the B3LYP15
functional with a pseudopotential scheme that is valence
double-ζ-plus-polarization (CEP-31G(d)) on the main group
elements and triple-ζ on the transition metals. Reported Gibbs
free energies were determined at 298.15 K and 1 atm and utilize
unscaled B3LYP/CEP-31G(d)16 vibrational frequencies. The
D3(BJ) dispersion correction17 with Becke−Johnson damp-
ing18 was utilized. The modeled system is cationic; thus, the
CPCM19 formalism was used with THF (ε = 7.43) as the
solvent. Stationary points were differentiated as minima or
transition states (TSs) via computation of the energy Hessian
and observation of the correct number of imaginary
are rarely amenable to Hammett correlations,13 a good linear
correlation is observed (R2 = 0.94). The plot demonstrates that
less-donating 4,4′ substituents result in an increase in the ratio
of cumene/n-propylbenzene, but the magnitude of the slope (ρ
= 0.2) suggests that the ligand donor ability has a small, but
clear and quantifiable, influence on the selectivity for propylene
insertion into the Pt−Ph bond and overall product selectivity.
Functionalized Arenes. Next, we sought to determine the
electronic effect of arene functionality on linear/branched
selectivity. The isomers of propylene hydroarylation using
anisole, benzene, and chlorobenzene were resolved using GC/
MS. Only para-substituted isomers were used for comparison,
but meta- and ortho-substituted benzenes were also observed.
Comparing the peak area ratios of 4-iPr-1-X-C6H4 and 4-nPr-1-
X-C6H4 (X = OMe, H and Cl) after 4 h of heating at 100 °C
under 0.1 MPa of propylene with 0.2 mol % 1b demonstrates
no correlation between the Hammett σ parameter of the
benzene functionality and Markovnikov/anti-Markovnikov
product selectivity for the resulting alkyl arenes (Scheme 5).
Catalysis with both anisole (OMe, σp = −0.27) and
chlorobenzene (Cl, σp = 0.23) provides a greater predilection
for formation of the isopropyl functionalized arene compared
with that of benzene.
frequencies, zero (0) and one (1), respectively. Intrinsic
20
reaction coordinates
from the TS geometries were
computed. All geometry optimizations were performed without
symmetry or coordinate restraint, and all complexes were
singlets for which restricted Kohn−Sham theory21 was
employed.
Product selectivity for the hydrophenylation of propylene
(i.e., linear/branched ratios) could be determined by the
regioselectivity of olefin insertion or by the relative rates of
subsequent reactions (i.e., Curtin−Hammett conditions).
Provided that olefin insertion into the Pt−Ph bond is reversible,
1610
dx.doi.org/10.1021/cs400988w | ACS Catal. 2014, 4, 1607−1615

ACS Catalysis
Research Article
as depicted in Scheme 1, the formation of cumene could be
preferential because of more favorable barriers for benzene
coordination or C−H activation. Thus, we calculated the
energetics of all steps, starting from propylene coordination to
the PtII complexes. For the computational modeling, we
on the bipyridyl ligand. For all three catalyst models, the
calculated ΔG‡rot is ∼10−11 kcal/mol. There is good agreement
between experiment and theory for olefin rotation. For
[(tbpy)Pt(CH2CH2Ph)(η2-C2H4)]+, the experimentally deter-
mined and ΔG‡rot for ethylene rotation is 11.7 kcal/mol at
−33.4 °C, and the calculated ΔG‡rot is 12.4 kcal/mol.7h For all
three modeled complexes, the dissociation of propylene from
[(xbpy)Pt(Ph)(η2-C3H6)]+ is calculated to be endergonic by
>27 kcal/mol, which is higher by at least 17 kcal/mol in
comparison to the propylene rotational transition states
(Scheme 7). This suggests that interconversion of the two
propylene adduct conformers will involve rotation of the olefin
rather than dissociation and recoordination of propylene. When
x = NO2 or Br, propylene binding to the Pt complex is
calculated to be more exergonic by 1.3 or 0.8 kcal/mol,
respectively, as compared with [(tbpy)Pt(Ph)]+ (Scheme 7),
which can likely be attributed to the electron-withdrawing nitro
or bromo groups on the bpy, which enhances olefin
coordination by providing a more electrophilic Pt center.
Selectivity for Branched vs Linear Insertion. Two
isomeric pathways for propylene insertion into the Pt−phenyl
bond were investigated from [(xbpy)Pt(Ph)(η2-C3H6)]+. The
two products of propylene insertion yield cumene or n-
propylbenzene after benzene C−H activation (Scheme 8). For
the tbpy complex, computations reveal a ΔΔGi‡ns = 1.6 kcal/mol
between transition states for propylene insertion with the lower
energy transition state leading to the branched product
(Scheme 8). At the same level of theory, relative free energy
focused on three Pt complexes: [(tbpy)Pt(Ph)]+ (1b),
+
[(Brbpy)Pt(Ph)]+ (1d), and [(NO bpy)Pt(Ph)] (1f)].
2
The energetics of the coordinated propylene complexes
[(xbpy)Pt(Ph)(η3-C3H6)]+ were studied to assess the preferred
olefin binding mode. Calculated stationary points relevant to
propylene binding are depicted in Scheme 7 for x = tBu (black),
Scheme 7. B3LYP-D3(BJ)/CEP-31G(d)/THF-Calculated
Propylene Binding and Rotational Transition States for
a
[(xbpy)Pt(Ph)(η2-C3H6)]+
barriers (ΔΔGins) for propylene insertion into the (NO bpy)Pt−
‡
2
Ph and (Brbpy)Pt−Ph bonds were 2.0 and 1.6 kcal/mol. Similar
to the tbpy complex, the ΔG‡ins leading to the branched product
NO2
is lower than for the linear product in
bpy and Brbpy cases.
DFT(D) predicts a slightly enhanced kinetic selectivity for the
a
The Gibbs free energy values are given in kcal/mol. The 2,2′-
NO2
t
branched product for the
bpy complex relative to the bpy
bipyridyl ligand with 4,4′-substituents are color coded: x = tBu (black),
Br (green), NO2 (red). Propylene adduct isomers are differentiated by
whether the propylene methyl group is proximal or distal to the phenyl
ligand, but are nearly degenerate. Model propylene rotational
transition states feature the propylene methyl group proximal (TS-a,
left) or distal (TS-b, right) to the phenyl ligand.
complex, which is consistent with experimental branched/linear
ratios from catalytic hydrophenylation of propylene (see Table
3 and Figure 1). However, as discussed below, the relative rates
of 1,2- vs 2,1-propene insertion may not dictate the ultimate
selectivity for formation of cumene vs n-propylbenzene.
Analysis of atomic charges for the olefin insertion TSs did
not reveal any obvious trends to explain product selectivity for
1,2- vs 2,1-insertion as a function of the xbpy ligand, as
expected, given the similar inductive impact of H and methyl
substituents. Analysis of the geometries of the propylene
insertion TSs showed very modest changes (typically ≪0.05 Å)
in bond lengths as a function of the X group or the insertion
regiochemistry. In addition, differences in computed electronic
energies and enthalpies and free energies were nearly identical,
suggesting that preference for insertion mode is due to the
looseness or tightness of the TS is not at the root of the
observed and computed selectivity differences. The calculations
reveal that the 1,2-insertion products are more stable than the
2,1-insertion products by 1.6 (tbpy), 1.5 (Brbpy), and 1.5 kcal/
Br (green), and NO2 (red) supporting ligand models. In the
ground state of [(xbpy)Pt(Ph)(η3-C3H6)]+, the CC bond of
propylene is perpendicular to the platinum square plane, which
is common for square planar d8 transition metals.7h,8a,22
Isomers of propylene coordination are thus differentiated by
whether the propylene methyl group is proximal or distal to the
phenyl ligand. There is, however, negligible calculated free
energy difference (<0.1 kcal/mol) between these isomers for
[(xbpy)Pt(Ph)(η2-C3H6)]+, implying minimal steric contact
between the propylene methyl moiety and the remainder of the
complex. Computations revealed subtle distinctions in the
propylene rotational barriers and more so in the propylene
binding free energies, which are linked to the electronic
properties of the bipyridyl substituents (Scheme 7).
mol (NO bpy).
2
Catalytic Cycle. The catalytic cycle for hydrophenylation
was investigated using DFT(D). For the bpy catalyst, the
Rotation of propylene about the metal/CC centroid axis
leads to two isomeric rotational transition states (Scheme 7 and
Figure 2) in which the methyl group of the propylene is either
proximal (TS-a) or distal (TS-b) to the phenyl ring,
respectively. In the rotational transition states, the CC
bond is in the square plane of the Pt complex. For all
complexes modeled, TS-a in Scheme 7 was computed to be
lower by ∼4 kcal/mol than TS-b, presumably because of steric
hindrance between the methyl group on propylene and a 6-H
t
potential energy surface (PES) depicts the energy pathway
leading to branched and linear alkyl arene (Scheme 9). The
reaction coordinates for other catalyst models are similar to
Scheme 9 (see Supporting Information). The pertinent free
energies for all modeled catalysts are collected in Table 4.
There are two plausible mechanisms for arene C−H bond
activation: oxidative hydrogen migration (OHM) and oxidative
1611
dx.doi.org/10.1021/cs400988w | ACS Catal. 2014, 4, 1607−1615

ACS Catalysis
Research Article
Figure 2. B3LYP/CEP-31G(d)/THF-calculated transition state geometries (top) for propylene rotation and ground state geometries (bottom) for
2
+
t
the complex [(NO bpy)Pt(Ph)(η -C3H6)] . Bond lengths are in angstroms, and angles are in degrees. Related bpy and Brbpy geometries are in the
2
Supporting Information. Note that geometries differ in the orientation of the methyl substituent on propylene relative to the remainder of the
catalyst.
lated transition state leading to cumene is 22.5 kcal/mol. This is
lower in free energy by 1.6 kcal/mol in comparison to the TS
Scheme 8. Calculated Free Energies (kcal/mol) for
Propylene Insertion into the Pt−Ph Bond Starting from
[(xbpy)Pt(Ph)(η2-C3H6)]+
for pathway leading to n-propylbenzene (Scheme 9 and Table
NO2
4). For the
bpy model, the lowest energy benzene C−H
oxidative addition TS is 21.3 kcal/mol (Table 4), lower by 1.2
kcal/mol than for the tbpy derivative, but nearly identical to the
Brbpy catalyst model. Hence, simulations suggest commensurate
activity for the electron-withdrawing substituents, and each
marginally lower than the tbpy congener. In all cases, cumene is
predicted to be the preferred product of hydrophenylation of
propylene, as is observed experimentally.
Above, we used DFT(D) calculations to assess the kinetic
selectivity for 1,2- vs 2,1-insertion, but the calculated energetics
suggest that Curtin−Hammett conditions would apply to the
PtII catalyzed propylene hydrophenylation. Thus, the rate of
propylene deinsertion and reinsertion into the Pt−Ph bond is
predicted to be faster than the rate of benzene C−H activation
by either the 1,2- or 2,1-insertion product. Previously, we have
reported that the resting state for catalytic hydrophenylation of
ethylene using several Ru(II) or Pt(II) catalysts is the insertion
product with coordinated ethylene, M(CH2CH2Ph)(η2-ethyl-
ene).1,21,23 Although the spectra are complicated, monitoring
the hydrophenylation of propene by 1b using 1H NMR
spectroscopy reveals resonances consistent with the analogous
resting state. Table 4 shows calculated energetics for the
production of cumene and n-propylbenzene for all three Pt
complexes. Under Curtin−Hammet conditions, the predicted
branched/linear ratio for each catalyst can be estimated using
the relative activation barriers for rate-limiting benzene C−H
addition/reductive elimination (OA/RE), with the latter
calculated to be the preferred pathway. A typical benzene
oxidative addition TS, which defines the highest computed
point on the PES, leading to a branched product, is shown in
Figure 3; all other Cartesian coordinates are collected in the
Supporting Information.
The benzene C−H activation barrier has been previously
reported to be the rate-determining step for catalytic ethylene
hydrophenylation catalyzed by 1b.7h Relative to the most stable
propylene adduct, [(tbpy)Pt(Ph)(η2-propylene)]+, the calcu-
t
activation. For the bpy complex, a ΔΔG‡ of 1.6 kcal/mol
(favoring cumene production) would give a branched/linear
1612
dx.doi.org/10.1021/cs400988w | ACS Catal. 2014, 4, 1607−1615

ACS Catalysis
Research Article
Scheme 9. Calculated Ground and Transition States for Propylene Hydrophenylation To Yield Cumene (solid line) or n-
a
Propylbenzene (dashed line) Catalyzed by [(tbpy)Pt(Ph)(THF)][BAr′4] (1b)
a
Similar pathways are calculated for the Brbpy (1d) and NO bpy (1f) derivatives; see Table 4. The Gibbs free energy barriers are in kcal/mol. Ground
2
and transition state structures relevant to the formation of the preferred product, cumene, are shown.
a
Table 4. Relative Free Energies for Olefin Hydroarylation
b
x
with bpyPtII Complex
NO2
NO2
tbpy
tbpy
bpy
bpy
nPB
Brbpy
cumene
Brbpy
nPB
cumene nPB
cumene
[xbpyPtII(Ph)
0
0
0
0
0
0
(η2-C3H6)]+
insertion TS
benzene adduct
benzene OA TS
products
19.3
6.0
20.7
8.6
16.5
4.5
18.4
8.0
18.3
5.4
19.7
9.1
22.5
12.3
24.1
13.3
21.3
13.7
23.0
14.7
21.5
13.1
23.5
14.1
a
b
+
(B3LYP/CEP-31G(d), STP, kcal/mol, in THF) . The superscript
denotes cationic complexes. All species are d8 reactants and products
with formally d6-hydride intermediates. OA = oxidative addition.
ratio of ∼4:1 (at room temperature), which is similar to the
experimental value of 2.9:1 at 100 °C. The same calculations for
Figure 3. B3LYP/CEP-31G(d)/THF-calculated transition state
geometry for C−H bond activation via oxidative addition by complex
‡
Br
‡
the NO bpy (ΔΔG = 1.7 kcal/mol) and the bpy (ΔΔG = 2.0
kcal/mol) predicts branched/linear ratios of 4.4:1 and 5.8:1,
respectively. Although the calculations correctly predict that
2
NO2
[
bpyPt(CH2CH(Ph)CH3)]+. Bond lengths are in angstroms, and
angles are in degrees.
x
cumene is favored over n-propylbenzene for all ligands bpy,
the observed experimental trend for variation of branched/
linear ratio is not reproduced. These results are not surprising,
given the energy differences to accurately model the selectivity
trend are much smaller than expected standard errors in the
calculations. The most salient result from the calculations,
which is well within expected errors, is that the branched/linear
ratio is likely a function of Curtin−Hammett conditions and
not controlled by the kinetic selectivity for 1,2- vs 2,1-insertion
of propene into Pt−Ph bonds.
substituted olefins. The systematic variation of bipyridyl ligand
electronic properties through 4,4′-substitution revealed a
correlation between ligand donor ability and product selectivity
for propylene hydrophenylation. Increasing the electron-
withdrawing properties of the 4,4′ substituent results in an
increase in the cumene/n-propylbenzene ratio. Computational
studies predict that 1,2-insertion (leading to branched product)
is kinetically more favorable than 2,1-insertion of propene and
that reducing the donor ability of the bpy ligands should further
bias the insertion selectivity toward branched products. The
results from the calculations also indicate that benzene C−H
activation is rate-limiting for pathways leading to either alkyl
arene. Thus, Curtin−Hammett conditions are likely responsible
CONCLUSIONS
■
In summary, cationic PtII complexes supported by bipyridyl
ligands are effective catalysts for the hydroarylation of
1613
dx.doi.org/10.1021/cs400988w | ACS Catal. 2014, 4, 1607−1615

ACS Catalysis
Research Article
for product selectivity, and future catalyst design to target linear
products should seek to optimize the thermodynamic bias
toward 2,1-insertion and lower the activation barrier for
benzene C−H activation from the 2,1-insertion product relative
to the 1,2-insertion product. Although the magnitude of the
changes in branched/linear selectivity is modest for the current
set of catalysts, the results suggest that more dramatic changes
in ligand donor ability might have more pronounced changes in
selectivity.
procedures were used to quantify the production of n-
propylbenzene (1.19; 0.99).
Catalytic Olefin Hydrophenylation with Non-Gaseous
Olefinic Substrates. A representative catalytic reaction is
described. [(tbpy)Pt(Ph)(THF)][BAr′4] (1b) (0.005 g, 0.003
mmol) was dissolved in 3.0 mL of benzene containing 1-
pentene (55 μL, 150 equiv relative to Pt,) and 0.01 mol %
hexamethylbenzene relative to benzene as an internal standard.
The reaction mixture was placed in a glass pressure tube and
heated to 100 °C. After a given time period, the reaction
mixture was allowed to cool to room temperature and analyzed
by GC/MS. Peak areas of the products and the internal
standard were used to calculate product yields from linear
regression analysis of standard samples. A set of five known
standards consisting of 2:1, 4:1, 6:1, 8:1, and 10:1 molar ratios
of alkylbenzene to HMB in CH2Cl2 were prepared. A plot of
the peak area ratios vs molar ratios gave a regression line. For
the GC/MS system, the slope and correlation coefficient (R2)
for each alkylbenzene is as follows: n-pentylbenzene (1.10;
0.99), 2-phenylpentane (1.33; 0.99), phenylcyclohexane (0.33;
0.99), 3-phenylcyclohexanone (0.33; 0.99), 4-phenylbutan-2-
one (0.31; 0.99), methyl cinnamate (0.50; 0.99), and methyl 3-
phenylpropanoate (0.53; 0.99).
EXPERIMENTAL SECTION
■
General Methods. Unless otherwise noted, all synthetic
procedures were performed under anaerobic conditions in a
nitrogen filled glovebox or by using standard Schlenk
techniques. Glovebox purity was maintained by periodic
nitrogen purges and was monitored by an oxygen analyzer
(O2 < 15 ppm for all reactions). Diethyl ether was dried by
distillation over CaH2. Tetrahydrofuran and n-pentane were
distilled over sodium/benzophenone and P2O5, respectively.
Methylene chloride and benzene were purified by passage
through a column of activated alumina. Acetone-d6 and CD2Cl2
were used as received and stored under a N2 atmosphere over 4
Å molecular sieves. 1H NMR spectra were recorded on a Varian
Mercury 300 or Unity Innova 500 MHz or on a Bruker 800
MHz spectrometer. 13C NMR spectra were recorded using a
Varian Mercury 300, Unity Innova 500 MHz (operating
frequency 75 or 125 MHz, respectively) or using a Bruker 800
MHz spectrometer (operating frequency 201 MHz). All 1H and
13C NMR spectra are referenced against residual proton signals
(1H NMR) or the 13C resonances (13C NMR) of the
deuterated solvents. 19F NMR (282 MHz operating frequency)
spectra were obtained on a Varian 300 MHz spectrometer and
referenced against an external standard of hexafluorobenzene (δ
= −164.9 ppm). GC/MS was performed using a Shimadzu
GCMS-QP2010 Plus system with a 30 m × 0.25 mm SHRXI-
5MS column with 0.25 mm film thickness using electron
impact (EI) ionization or negative chemical ionization (NCI),
which also allows for simulated electron impact (SEI)
ionization. Ethylene (99.5%) and propylene (99.5%) were
purchased in a gas cylinder from GTS-Welco and used as
received. All other reagents were used as purchased from
commercial sources. The preparation, isolation, and character-
ization of 3-phenylcyclohexanone23 and [(N ∼ N)Pt(Ph)-
(THF)][BAr′4]7h,8 (1a−1f and 2) have been previously
reported.
ASSOCIATED CONTENT
* Supporting Information
■
S
Full computational study, including reaction coordinates for all
catalyst models. This material is available free of charge via the
Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Authors
*E-mail: tbg7h@virginia.edu.
*E-mail: t@unt.edu
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
T.B.G. and T.R.C. acknowledge The Office of Basic Energy
Sciences, U.S. Department of Energy for support of this work
(DE-SC0000776 and DE-FG02-03ER15387, respectively).
REFERENCES
■
(1) (a) Foley, N. A.; Lee, J. P.; Ke, Z.; Gunnoe, T. B.; Cundari, T. R.
Acc. Chem. Res. 2009, 42, 585−597. (b) Andreatta, J. R.; McKeown, B.
A.; Gunnoe, T. B. J. Organomet. Chem. 2011, 696, 305−315. (c) Bhalla,
G.; Bischof, S. M.; Ganesh, S. K.; Liu, X. Y.; Jones, C. J.; Borzenko, A.;
Tenn, W. J., III; Ess, D. H.; Hashiguchi, B. G.; Lokare, K. S.; Leung, C.
H.; Oxgaard, J.; Goddard, W. A., III; Periana, R. A. Green Chem. 2011,
13, 69−81.
Catalytic Olefin Hydrophenylation with Propylene. A
representative catalytic reaction is described. [(tbpy)Pt(Ph)-
(THF)][BAr′4] (1b) (0.019 g, 0.013 mmol) was dissolved in
12.0 mL of benzene containing 0.01 mol % hexamethylbenzene
(HMB) relative to benzene as an internal standard. The
reaction mixture was placed in a stainless steel pressure reactor,
charged with propylene (0.1 MPa), pressurized to a total of 0.8
MPa with N2, and heated to 100 °C. After a given time period,
the reaction mixture was allowed to cool to room temperature
and was analyzed by GC/MS. Peak areas of the products and
the internal standard were used to calculate product yields.
Cumene production was quantified using linear regression
analysis of gas chromatograms of standard samples. A set of five
known standards consisting of 2:1, 4:1, 6:1, 8:1, and 10:1 molar
ratios of cumene to HMB in benzene were prepared. A plot of
the peak area ratios vs molar ratios gave a regression line. For
the GC/MS system, the slope and correlation coefficients (R2)
for cumene were 0.89 and 0.99, respectively. Identical
(2) (a) Price, C. C. Chem. Rev. 1941, 29, 37−67. (b) Friedel−Crafts
and Related Reactions; John Wiley and Sons: New York, 1964; Vol. 2.
(c) Roberts, R. M.; Khalaf, A. A. Friedel−Crafts Alkylation Chemistry: A
Century of Discovery; Marcel Dekker, Inc.: New York, 1984.
(3) (a) Lail, M.; Bell, C. M.; Conner, D.; Cundari, T. R.; Gunnoe, T.
B.; Petersen, J. L. Organometallics 2004, 23, 5007−5020. (b) Matsu-
moto, T.; Taube, D. J.; Periana, R. A.; Taube, H.; Yoshida, H. J. Am.
Chem. Soc. 2000, 122, 7414−7415.
(4) Luedtke, A. T.; Goldberg, K. I. Angew. Chem., Int. Ed. 2008, 47,
7694−7696.
(5) McKeown, B. A.; Foley, N. A.; Lee, J. P.; Gunnoe, T. B.
Organometallics 2008, 27, 4031−4033.
(6) (a) De Felice, V.; De Renzi, A.; Tesauro, D.; Vitagliano, A.
Organometallics 1992, 11, 3669−3676. (b) De Felice, V.; Cucciolito,
1614
dx.doi.org/10.1021/cs400988w | ACS Catal. 2014, 4, 1607−1615

ACS Catalysis
Research Article
M. E.; De Renzi, A.; Ruffo, F.; Tesauro, D. J. Organomet. Chem. 1995,
493, 1−11. (c) Cucciolito, M. E.; De Renzi, A.; Orabona, I.; Ruffo, F.;
Tesauro, D. J. Chem. Soc., Dalton Trans. 1998, 1675−1678.
(7) (a) Matsumoto, T.; Periana, R. A.; Taube, D. J.; Yoshida, H. J.
Mol. Catal. A.: Chem. 2002, 180, 1−18. (b) Oxgaard, J.; Muller, R. P.;
Goddard, W. A., III; Periana, R. A. J. Am. Chem. Soc. 2004, 126, 352−
363. (c) Oxgaard, J.; Periana, R. A.; Goddard, W. A., III J. Am. Chem.
Soc. 2004, 126, 11658−11665. (d) Lail, M.; Arrowood, B. N.; Gunnoe,
T. B. J. Am. Chem. Soc. 2003, 125, 7506−7507. (e) Goj, L. A.; Gunnoe,
T. B. Curr. Org. Chem. 2005, 9, 671−685. (f) Foley, N. A.; Lail, M.;
Lee, J. P.; Gunnoe, T. B.; Cundari, T. R.; Petersen, J. L. J. Am. Chem.
Soc. 2007, 129, 6765−6781. (g) Joslin, E. E.; McMullin, C. L.;
Gunnoe, T. B.; Cundari, T. R.; Sabat, M.; Myers, W. H.
Organometallics 2012, 31, 6851−6860. (h) McKeown, B. A.;
Gonzalez, H. E.; Friedfeld, M. R.; Gunnoe, T. B.; Cundari, T. R.;
Sabat, M. J. Am. Chem. Soc. 2011, 133, 19131−19152.
(8) (a) McKeown, B. A.; Gonzalez, H. E.; Gunnoe, T. B.; Cundari, T.
R.; Sabat, M. ACS Catal. 2013, 3, 1165−1171. (b) McKeown, B. A.;
Gonzalez, H. E.; Friedfeld, M. R.; Brosnahan, A. M.; Gunnoe, T. B.;
Cundari, T. R.; Sabat, M. Organometallics 2013, 32, 2857−2865.
(9) McKeown, B. A.; Gonzalez, H. E.; Michaelos, T.; Gunnoe, T. B.;
Cundari, T. R.; Crabtree, R. H.; Sabat, M. Organometallics 2013, 32,
3903−3913.
(10) Scarso, A.; Colladon, M.; Sgarbossa, P.; Santo, C.; Michelin, R.
A.; Strukul, G. Organometallics 2010, 29, 1487−1497.
(11) Liu, Y.; Mao, D.; Xu, D.; Xu, Z.; Zhang, Y. Synth. Commun.
2007, 37, 4389−4397.
(12) Suzuki, Y.; Yasumoto, T.; Mashima, K.; Okuda, J. J. Am. Chem.
Soc. 2006, 128, 13017−13025.
(13) Tomasik, P.; Ratajewicz, Z. Pyridine-Metal Complexes; John
Wiley & Sons, Inc.: New York, 1985; Vol. 14.
(14) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A.,
Jr.; Stratmann, R. E.; Burant, J. C.; Dapprich, S.; Millam, J. M.; Daniels,
A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.; Barone, V.;
Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo, C.; Clifford,
S.; Ochterski, J.; Petersson, G. A.; Ayala, P. Y.; Cui, Q.; Morokuma, K.;
Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.;
Cioslowski, J.; Ortiz, J. V.; Baboul, A. G.; Stefanov, B. B.; Liu, G.;
Liashenko, A.; Piskorz, P.; Komaromi, I.; Gomperts, R.; Martin, R. L.;
Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.;
Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M.
W.; Andres, J. L.; Gonzalez, C.; Head-Gordon, M.; Replogle, E. S.;
Pople, J. A. Gaussian 03, Revision C.02; Gaussian, Inc.: Wallingford,
CT, 2004.
(15) (a) Becke, A. D. Phys. Rev. 1988, A38, 3098−3100. (b) Lee, C.;
Yang, W.; Parr, R. G. Phys. Rev., B 1988, 37, 785−789.
(16) Stevens, W. J.; Basch, H.; Krauss, M. J. Chem. Phys. 1984, 81,
6026−6033.
(17) Grimme, S.; Antony, J.; Ehrlich, S.; Krieg, H. J. Chem. Phys.
2010, 132, 154104/1-19.
(18) (a) Becke, A. D.; Johnson, E. R. J. Chem. Phys. 2005, 122,
154101/1-5. (b) Johnson, E. R.; Becke, A. D. J. Chem. Phys. 2005, 123,
24101/1-7. (c) Johnson, E. R.; Becke, A. D. J. Chem. Phys. 2006, 124,
174104/1-9.
(19) Barone, V.; Cossi, M. J. Phys. Chem. A 1998, 102, 1995−2001.
(20) Hratchian, H. P. J. Chem. Theory Comput. 2005, 1, 61−69.
(21) Kohn, W.; Sham, L. J. Phys. Rev. 1980, A140, 1133−1138.
(22) (a) Shiotsuki, M.; White, P. S.; Brookhart, M.; Templeton, J. L.
J. Am. Chem. Soc. 2007, 129, 4058−4067. (b) Fusto, M.; Giordano, F.;
Orabona, I.; Ruffo, F.; Panunzi, A. Organometallics 1997, 16, 5981−
5987. (c) Friedrich, A.; Ghosh, R.; Kolb, R.; Herdtweck, E.; Schneider,
S. Organometallics 2009, 28, 708−718. (d) Buhl, M.; Hakansson, M.;
̊
̈
̈
Mahmoudkhani, A. H.; Ohrstrom, L. Organometallics 2000, 19, 5589−
̈
5596.
(23) Hayashi, T.; Sasaki, K. Chem. Lett. 2008, 37, 842−843.
1615
dx.doi.org/10.1021/cs400988w | ACS Catal. 2014, 4, 1607−1615