Ligand self-assembling through complementary hydrogen-bonding in the coordination sphere of a transition metal center: The 6- diphenylphosphanylpyridin-2(1H)-one system

Article abstract of DOI:10.1021/ja108639e

Motivated by previous findings which had shown that transition metal catalysts based on the 6-diphenylphosphanylpyridone ligand (6-DPPon, 2) display properties as a self-assembling bidentate ligand-metal complex, we have performed a thorough study on the bonding situation of this ligand, alone and in the coordination sphere of a late transition metal. Thus, combining a number of spectroscopic methods (UV-vis, IR, NMR, X-ray), we gained insights into the unique structural characteristics of 2. These experimental studies were corroborated by DFT calculations, which were in all cases in good agreement with the experimental results. The free ligand 2 prefers to exist as the pyridone tautomer 2A and dimerizes to the pyridone-pyridone dimer 4A in solution as well as in the crystal state. The corresponding hydroxypyridine tautomer 2B is energetically slightly disfavored (ca. 0.9 kcal/mol within the up-conformer relevant for metal coordination); hence, hydrogen bond formation within the complex may easily compensate this small energy penalty. Coordination properties of 2 were studied in the coordination sphere of a platinum(II) center. As a model complex, [Cl₂Pt(6-DPPon)₂] (II) was prepared and investigated. All experimental and theoretical methods used prove the existence of a hydrogenbonding interligand network in solution as well as in the crystal state of 11 between one 6-DPPon ligand existing as the pyridone tautomer 2A and the other ligand occupying the complementary hydroxypyridine form 2B. Dynamic proton NMR allowed to determine the barrier for interligand hydrogen bond breaking and, in combination with theory, enabled us to determine the enthalpic stabilization through hydrogen-bonding to contribute 14-15 kcal/mol.

Full text of DOI:10.1021/ja108639e

ARTICLE
pubs.acs.org/JACS
Ligand Self-Assembling through Complementary Hydrogen-Bonding
in the Coordination Sphere of a Transition Metal Center:
The 6-Diphenylphosphanylpyridin-2(1H)-one System
Urs Gellrich,† Jing Huang,§ Wolfgang Seiche,† Manfred Keller,† Markus Meuwly,*,§ and Bernhard Breit*,†,‡
†Institut f€ur Organische Chemie und Biochemie and ‡Freiburg Institute for Advanced Studies, Albert-Ludwigs-Unversit€at,
79104 Freiburg, Germany
§Department of Chemistry, University of Basel, Klingelbergstrasse 80, CH-4056 Basel, Switzerland
S Supporting Information
b
ABSTRACT: Motivated by previous findings which had
shown that transition metal catalysts based on the 6-di-
phenylphosphanylpyridone ligand (6-DPPon, 2) display
properties as a self-assembling bidentate ligand-metal
complex, we have performed a thorough study on the
bonding situation of this ligand, alone and in the coordina-
tion sphere of a late transition metal. Thus, combining a
number of spectroscopic methods (UV-vis, IR, NMR,
X-ray), we gained insights into the unique structural char-
acteristics of 2. These experimental studies were corrobo-
rated by DFT calculations, which were in all cases in good agreement with the experimental results. The free ligand 2 prefers to exist
as the pyridone tautomer 2A and dimerizes to the pyridone-pyridone dimer 4A in solution as well as in the crystal state. The
corresponding hydroxypyridine tautomer 2B is energetically slightly disfavored (ca. 0.9 kcal/mol within the up-conformer relevant
for metal coordination); hence, hydrogen bond formation within the complex may easily compensate this small energy penalty.
Coordination properties of 2 were studied in the coordination sphere of a platinum(II) center. As a model complex, [Cl2Pt(6-
DPPon)2] (11) was prepared and investigated. All experimental and theoretical methods used prove the existence of a hydrogen-
bonding interligand network in solution as well as in the crystal state of 11 between one 6-DPPon ligand existing as the pyridone
tautomer 2A and the other ligand occupying the complementary hydroxypyridine form 2B. Dynamic proton NMR allowed to
determine the barrier for interligand hydrogen bond breaking and, in combination with theory, enabled us to determine the
enthalpic stabilization through hydrogen-bonding to contribute 14-15 kcal/mol.
’ INTRODUCTION
While these bidentate ligands provide unique selectivities,
their synthesis can be troublesome and may render the ligands
more expensive than the noble metal source. Thus, as an
alternative to classical bidentate ligands in which the two donor
sites are connected by a covalent connection, we4a-d and others5a-h
recently introduced the concept of monodentate-to-bidentate
ligand self-assembly. Thus, mixing monodentate ligands
equipped with functional groups capable of complementary
hydrogen bonding in the presence of a transition metal salt leads
to defined bidentate ligand-metal complexes. The parent sys-
tem we selected for this approach was the 2-pyridone (1A)/2-
hydroxypyridine (1B) tautomer system. It is known to dimerize
in aprotic solvents to form predominantly the symmetrical pyridone
dimer 3A through hydrogen bonding (Scheme 2).6-9 If D would
be a donor atom capable of binding a metal center, a bidentate
binding mode would be impossible for dimer 4A for geometric
reasons. Alternatively, formation of the mixed dimer between the
Selectivity control in homogeneous metal complex catalysis is
most frequently achieved by crafting the microenvironment of
the catalytically active metal center upon binding of appropriate
ligand architectures to the metal center. Among the many ligands
used, bidentate ligands occupy an important position. Chelation
induces a higher binding constant for the metal center and thus
provides a higher stability of the resulting metal complexes. Further-
more, the bidentate nature of the ligand reduces the degrees of
freedom in the ligand backbone, which in turn decreases the
number of competing transition states, resulting in higher
reaction selectivity. A prominent example for enantioselectivity
control provides β-ketoester reduction with ruthenium(II)-
BINAP catalysts (Scheme 1).1 Another example is the control
of regioselectivity for the industrially important rhodium-
catalyzed hydroformylation of terminal alkenes, which requires
tailor-made bidentate ligands.2a-c Among the few ligands known
to achieve efficient regiocontrol is the wide bite angle diphosphine
XANTPHOS (Scheme 1).3
Received: September 24, 2010
Published: December 9, 2010
r
2010 American Chemical Society
964
dx.doi.org/10.1021/ja108639e J. Am. Chem. Soc. 2011, 133, 964–975
|

Journal of the American Chemical Society
ARTICLE
Scheme 1. Enantio- and Regioselectivity Control with Selected Metal Catalysts Derived from Bidentate Ligands
Scheme 2. Self-Assembling of the 6-DPPon Ligand in the Presence of a Transition Metal Center
2-pyridone and the 2-hydroxypyridine tautomer would generate
a ligand architecture suited for a bidentate binding mode 5.
Although this nonsymmetrical dimerization mode is unusual,
recent calculations in the gas phase have shown that it is just 4.8
kcal/mol less favorable (D = H) compared to the commonly
observed 2-pyridone dimer.8 We speculated that if D would be an
appropriate donor atom such as a phosphine, the chelation effect
exhibited through coordinative binding to a metal center might
overcome this energy penalty to provide the bidentate binding
mode 5 as shown in Scheme 2. Thus, choosing for D a
diphenylphosphanyl function led us to the 6-diphenylphospha-
nylpyridin-2(1H)-one (6-DPPon) system (2).4c,10
Scheme 3. Rhodium-Catalyzed Hydroformylation of 1-Oc-
tene: Complementary Behavior of PPh3 and 2a
a Reaction conditions: Rh:L:1-octene (1:10:7000), c(1-octene) = 1.4 M,
4 h, toluene.
In previous studies we showed that 6-DPPon (2) indeed
behaves as a bidentate ligand in homogeneous catalysis. We
selected the hydroformylation of terminal olefins as a test case
since the hydroformylation displays strong chelating effects and
only chelating ligands with appropriate large bite angles such as
the XANTPHOS system allow for high regioselectivities. Indeed,
a rhodium/6-DPPon system furnished a highly active hydro-
formylation catalyst which displayed high regioselectivity in favor
of the linear aldehyde. This catalytic transformation was selected
since its regioselectivity is highly sensitive to chelation effects.
High regioselectivities in favor of the linear aldehyde were
observed with this catalyst in unpolar solvents such as
toluene.4c However, changing the solvent from toluene to
methanol led to significantly reduced levels of regioselectivities
which were more typical for a monodentate phosphine ligand
(Scheme 3).11 Hence, these experiments strongly suggest that
6-DPPon forms a bidentate ligand based on complementary
hydrogen bonding during the course of the hydroformylation
reaction.
In order to get a deeper insight into the nature and properties
of the self-assembling 6-DPPon system, we performed a detailed
study on the synthesis of the 6-DPPon ligand (2) and its tautomer
equilibrium in solution, as well as its coordination behavior in the
presence of a platinum(II) center as a model complex for a late
transition metal catalyst bearing 2. Employing UV-vis, IR, and
NMR spectroscopy, X-ray crystallography, and DFT methods
allowed us to get a detailed insight into the hydrogen-bonding
situation and its properties with regard to monodentate to
bidentate ligand self-assembly.
’ RESULTS AND DISCUSSION
Synthesis of the 6-DPPon Ligand. The synthesis of 2
commenced from either 2,6-dichloropyridine 6A or alternatively
the dibromo compound 6B, both commercially available (Scheme 4).
One of the halogens was replaced in a nucleophilic aromatic
substitution with potassium tert-butoxide to furnish the corresponding
965
dx.doi.org/10.1021/ja108639e |J. Am. Chem. Soc. 2011, 133, 964–975

Journal of the American Chemical Society
ARTICLE
Scheme 4. Alternative Preparation of 6-DPPon (2) via a Nucleophilic (Upper Pathway) or Electrophilic Aromatic
(Lower Pathway) Displacement
Scheme 5. Preparation of Methyl-Fixed Tautomers 10A
(6-DPMePon) and 10B (2-MeODPP)
Figure 1. UV-vis spectra of the fixed tautomers 10A (red) and 10B
(blue) and the free ligand 2 (black) in different solvents. All spectra were
recorded at room temperature and a concentration of 0.14 mM.
butoxy ethers 7A and 7B. Treatment of 7A with sodium
diphenylphosphide, generated under Birch-type conditions from
triphenylphosphine, gave the phosphine 8. This reaction can be
run at large scale (50 g). Alternatively, 8 can be obtained from the
bromopyridine 7B upon treatment with n-butyllithium at 0 °C in
diethyl ether to initiate a bromine-lithium exchange reaction,
followed by electrophilic trapping with chlorodiphenylphos-
phine. Cleavage of the tert-butyl ether occurred upon treatment
of 8 with formic acid to liberate the 6-DPPon ligand 2, which was
obtained as a colorless solid after recrystallization.
UV-Vis Spectroscopy of 2. Our first goal was to have a
deeper look into the equilibrium between the two tautomers of 2.
Hence, we utilized UV-vis spectroscopy, which is a frequently
applied method to investigate such tautomeric equilibria. To
obtain experimental data for each of the individual tautomers,
covalently fixed model substrates 10A and 10B were prepared
first (Scheme 5). N-Methylated and fixed lactam tautomer 10A
was obtained from 6-DPPon (2) upon deprotonation with sodium
hydride, followed by methylation with methyl iodide. The O-
methylated system 10B was obtained starting from bromopyridine
9. Halogen lithium exchange with n-butyllithium and subsequent
trapping of the resulting organolithium compound with chloro-
diphenylphosphine furnished 10B.
λ = 290 nm. Interestingly, the 6-DPPon (2) ligand itself showed
an absorption maximum at λ = 330 nm irrespective of the solvent
polarity, which shows that the pyridone form 2A is the prevalent
tautomer and thermodynamically favored.
IR Spectroscopy of 6-DPPon (2) in CCl4 Solution. Typi-
cally, NH and OH bonds involved in hydrogen-bonding display
stretching vibrations which are shifted to lower wavenumbers
compared to those not involved in hydrogen-bonding, and the
corresponding absorption bands are significantly broadened.12
Since the parent 2-pyridone system 1 is known to form symmetric
pyridone dimers in aprotic solvents, the appearance of NH
O
3 3 3
hydrogen bonds in the IR spectrum of a solution of 6-DPPon (2)
would suggest the existence of such a dimer 4A in a similar
manner. The IR spectrum of 2 in CCl4 at c = 10-1 M was
recorded and is depicted in Figure 2. Additionally, the theoretical
IR spectrum for the dimer 4A and the most stable confomers of
both tautomeric forms of the monomer (2A and 2B) were
calculated employing DFT methods. The results of experimen-
tally found and calculated absorptions are depicted in Table 1.
According to the calculations, the sharp peak at 1644 cm-1
corresponds to the CO vibration of the pyridone form which, in
agreement with the results obtained by UV-vis spectroscopy,
represents the more stable tautomer of 2. On the other hand, the
extremely broad peak between 2200 and 3200 cm-1 corresponds
The UV-vis spectra of 6-DPMePon (10A) and 2-MeODPP
(10B), as well as that of 6-DPPon (2), were recorded at a
concentration c = 0.14 mM in solvents of different polarity
(Figure 1). Spectra recorded in dichloromethane showed an
absorption maximum for 6-DPMePon (9A) at λ = 330 nm, while
the absorption maximum for 2-MeODPP (10B) was detected at
to the N-H O vibration involved in hydrogen bonding. The
3 3 3
origin for the exceptionally broad shape of this peak will be
discussed later (vide infra). The presence of the N-H
O
3 3 3
966
dx.doi.org/10.1021/ja108639e |J. Am. Chem. Soc. 2011, 133, 964–975

Journal of the American Chemical Society
ARTICLE
Table 1. Comparison of Characteristic Vibrations (cm-1) Obtained from Experiment and by DFT Calculation of 6-DPPon (2)
B3LYP/6-31G(d,p)a
expt
2A-down
2B-up
4A
NH bending vibration
OH bending vibration
CO stretching vibration
CN stretching vibration
1595/1578
1518 (1580)
1578 (1642)/ 1546 (1609)
1580 (1644)
1567 (1631)
1644
1715 (1785)
1671 (1738)
3029 (3152)
N-H
O
2200-3200
3 3 3
a The vibrations obtained by DFT calculation are scaled by a factor of 0.9611, as proposed by Kacker et al.13 Unscaled vibrations obtained by means of
DFT calculations are shown in parentheses.
Figure 2. IR spectrum of 2 in CCl4 (c = 10-1 M).
Figure 3. X-ray structure of the symmetric dimer 4A of 6-DPPon (2).
Scheme 6. Association of the Monomeric 6-DPPon Ligand
2A to the Pyridone-Pyridone Dimer 4A
The results obtained from NMR titration as well as those
obtained from DFT calculations are in excellent agreement and
suggest that dimer formation is thermodynamically favored and,
hence, that 4A is the prevalent species in solution. This result
correlates well with former calculations on the pyridone parent
system.8
Structure of the Symmetric 6-DPPon Dimer. From a
saturated solution of 2 in CH2Cl2, single crystals could be grown
which were suitable for X-ray diffraction analysis. The structure
of 2 in the solid state is depicted in Figure 3 and shows the
formation of the expected symmetrical pyridone-pyridone
dimer 4A. The formation of the theoretically accessible hydroxy-
pyridine dimer can be excluded on the basis of a comparison of
the calculated geometrical data for dimers 4A and 4B with those
found in the X-ray crystal structure analysis (see Table 3).
The bond lengths show an average difference of 0.05 Å, and
Table 2. Association Constant and ΔG for the Dimerization
of 2 Observed by NMR Dilution Experiments and by DFT
Calculations
the N-H O angle differs by 2.6°. The only major deviation
3 3 3
NMR expt
B3LYP/6-31G(d,p)
between theory and experiment is the O H distance, which
3 3 3
K
ass (M-1
ΔG (kcal/mol)
)
(2.31 ( 0.29) ꢀ 104
-5.95 ( 0.075
2.69 ꢀ 104
-6.05
differs by 0.17 Å. Here it seems important to point out that the
position of the hydrogen atoms in a structure determined by
X-ray cannot be detected directly because of the low scattering
power of the hydrogen atoms and strongly depends on the
mathematical algorithms used to refine the structure. Therefore,
the hydrogen positions from the computations are probably
more realistic.
vibration strongly suggests that a symmetric dimer 4A of 2 is
prevalent in solution under these conditions. Notably, the
calculations for 4A suggest a shift for the CO stretching vibration
to lower wavenumbers due to its involvement in hydrogen bonding.
Determination of Kass for the Formation of the Symmetric
6-DPPon Dimer. In order to estimate Kass for the dimerization
process of the 6-DPPon monomer (Scheme 6), we performed 1H
NMR dilution eperiments.14 Additionally, we estimated Kass by
means of DFT calculations (Table 2).
Theoretical Analysis of the Conformers and Tautomers
of 2. As a next step, we wanted to support our spectroscopic
results concerning the tautomerism of 2 by DFT calculations.
However, not only is this equilibrium essential for the formation
of the proposed hydrogen bonding in the self-assembling process
at a metal center, but the heterocycle also has to be in a specific
967
dx.doi.org/10.1021/ja108639e |J. Am. Chem. Soc. 2011, 133, 964–975

Journal of the American Chemical Society
ARTICLE
Table 3. Equilibrium Structure of the Dimer of 2 Measured
by X-ray and Computed by DFT
Table 4. Relative Energies and Gibbs Free Enthalpy at 298.15
K of the Confomers and Tautomers Shown in Figure 4
E þ ZPE (kcal/mol)a
ΔG (kcal/mol)
B3LYP/6-31G(d,p)
2A-down
2A-up
0.0
0.6
2.3
1.5
0.0
0.2
2.2
1.5
X-ray
4A
4B
N-H O (Å)
3 3 3
2.777(1)
1.89(2)
2.772
1.72
2.737
1.01
2B-down
O
H (Å)
3 3 3
2B-up
a E þ ZPE = electronic energy þ zero point energy.
C2-N (Å)
1.377(2)
1.250(2)
177.3(2)
1.392
1.248
179.7
1.341
1.326
173.4
C2-O (Å)
N-H O (deg)
3 3 3
Scheme 7. Preparation of cis-[Cl2Pt(6-DPPon)2] Complex 11
energy differences are small enough to allow the existence of
the disfavored hydroxypyridine form if this energetic penalty
could be compensated for by the bond enthalpy of the proposed
hydrogen bonds.
Coordination Properties of 6-DPPon: Formation of and
Bonding Situation in cis-[Cl2Pt(6-DPPon)2] (11). Square pla-
nar coordination geometries are the most common coordination
geometries (among others, including trigonal bipyramidal, octa-
hedral, etc.) passed in catalytic cycles of late transition metal
catalysis, and as such it is an ideal model system to explore the
coordination behavior of the 6-DPPon ligands. For this reason a
platinum(II) salt was selected, since these are known to form
square planar coordination geometries upon coordination of two
donor ligands, which may exist in either a cis or trans isomeric
form. Reaction of [Cl2Pt(COD)] with 2 equiv of 6-DPPon
ligand 2 furnished quantitatively cis-[Cl2Pt(6-DPPon)2] (11)
via displacement of 1,5-cyclooctadiene with the two phosphine
donor ligands (Scheme 7).
NMR Spectroscopy of 11. 31P NMR spectroscopy can be
used for structural characterization of square planar platinum
complexes with two phosphorus ligands, as 30% of the platinum
exists as the NMR-active 195Pt isotope (I = 1/2), coupling with
1
the 31P atoms. JP-Pt coupling constants greater than 3000 Hz
Figure 4. Possible conformers and tautomers of 2 and their DFT
optimized structures.
are typical for a cisoid arrangement of the two phosphane ligands
at the platinum center.15
The 31P NMR of [Cl2Pt(6-DPPon)2] (11) is depicted in
Figure 5. A 1JP,Pt coupling constant of 3591 Hz was monitored
which clearly indicates the two 6-DPPon ligands to be in a cisoid
arrangement.16 The cis geometry should enable the formation of
hydrogen bonds between the two ligands, which could be
monitored by 1H NMR spectroscopy (Figure 6).
geometrical orientation relative to the PPh2 group in order to
allow for self-assembly to occur. Figure 4 shows the optimized
structures of the relevant conformers and tautomers of 2.
Thus, from the selected conformers, only the “up” conformers
allow 2 to form hydrogen bonds to a neighbor ligand system
while the phosphorus is bound to the metal center. Furthermore,
one ligand must exist in the lactim form 2B to render hydrogen
bonding possible. The energy differences between all conformers
and tautomers were determined by DFT calculations. Table 4
shows the energy and Gibbs free enthalpy differences between
the conformers and tautomers of 2.
Indeed, the 1H NMR shows two distinct signals in the
chemical shift region typically expected for OH or NH bonds
involved in hydrogen bonding (δ = 11-13, see Figure 6).
Considering the electronegativity of the heteroatoms involved
in the hydrogen bonds, the signal at δ = 12.90 could be assigned
The calculations clearly show, in agreement with the results
obtained by UV-vis spectroscopy, that the pyridone form
represents the more stable tautomer. On the other hand, the
to the O-H O proton, whereas the signal at δ = 12.43
3 3 3
corresponds to the absorption of the N-H N proton. Taking
3 3 3
a closer look at this resonance, a poorly resolved doublet can be
968
dx.doi.org/10.1021/ja108639e |J. Am. Chem. Soc. 2011, 133, 964–975

Journal of the American Chemical Society
ARTICLE
Figure 7. 1H NMR spectrum of 11 as a function of the temperature.
With a frequency splitting of Δν = 213.6 Hz at 210 K, a rate
constant kc = 514 s-1 was calculated. In the context of Eyring
theory, the rate constant is related to the free activation enthalpy.
Figure 5. Coupling in complexes of the general type [Cl2PtP2] (top)
and 31P NMR spectrum for [Cl2Pt(6-DPPon)2] (bottom).
kbT
h
TS=RT
kc ¼
e- ΔG
ð2Þ
In eq 2, kB represents the Boltzmann constant, h the Planck
constant, and R the ideal gas constant. Using this equation, a free
activation energy of ΔGTS = 15.1 kcal/mol can be estimated.
The dynamic process which renders the two hydrogen atoms
involved in hydrogen bonding equal on the NMR time scale must
involve a tautomerization process. However, since within the
hydrogen-bonded pseudochelate complex the hydrogen atoms
involved in H-bonding are in a fixed orientation, tautomerization
is impossible. Thus, in order to tautomerize, the hydrogen-
bonding network has to break first, which is achieved by rotating
the pyridone and hydroxypyridine nuclei out of plane (step a
in Scheme 8). Once the H-bonds are broken, the known
pyridone/hydroxypyridine tautomerization can occur (step b
in Scheme 8). Renewed hydrogen-bonding (step c in Scheme 8)
renders the protons involved in hydrogen bonding magnetically
undistinguishable.
For a more quantitative assesssment, the strength of the
double-H-bond motif was determined, and the tautomerization
reaction was considered in more detail. To computationally
characterize the energy required to break the H-bond motif, the
two aromatic rings were simultaneously rotated in a disrotatory
manner (fixing the two Pt-P-C-H dihedral angles), and the
remaining internal degrees of freedom were optimized. The
computed energy barrier is 14.6 kcal/mol, which is close to the
experimentally determined value of 15.1 kcal/mol. However, it
should be pointed out that, experimentally, a free energy (ΔG) is
derived from kc, whereas the present computational approach
only reports on the enthalpy change. The N-H and O-H bond
lengths are closer to those in the separate 6-DPPon system
(1.014 and 0.966 Å) at φ = 80°, showing that both hydrogen
bonds have been broken (Figure 8).
Figure 6. 1H NMR (CDCl3, 500 MHz, 210 K) spectrum of 11 and a
proposed coupling for the signal located at δ = 12.43.
3
deduced, which results from a JH-P coupling with the phos-
phorus atom in proximity. Dynamic proton NMR spectra show a
distinct temperature dependence for the resonances of these
protons involved in hydrogen bonding. Thus, coalescence of
signals was observed at 325 K, which suggests a highly dynamic
and fast tautomerization of the two ligands, rendering the protons
on the NMR time scale magnetically equivalent (Figure 7).
Without an explicit line shape analysis, the rate constant was
calculated with eq 1.17
πΔν
pffiffi
kc ¼
¼ 2:22Δν
ð1Þ
2
969
dx.doi.org/10.1021/ja108639e |J. Am. Chem. Soc. 2011, 133, 964–975

Journal of the American Chemical Society
ARTICLE
Scheme 8. Process That Renders the Hydrogen Bonds Indistinguishable: (a) Rotation along the CR-P Bond, (b) Intramolecular
Tautomerism, and (c) Rotation Back in Plane and Formation of the Hydrogen Bonds
most likely the rotatory motion of the pyridone and hydroxypyr-
idine rings and not intramolecular H-transfer. However, to firmly
establish this, more elaborate calculations including entropic
effects, H-bond dynamics, and explicit solvent will be necessary.
For this, molecular dynamics (MD) simulations together with
dedicated force fields, such as VALBOND-TRANS,19 combined
with provisions for a realistic treatment of the hydrogen dy-
namics, such as molecular mechanics with proton transfer, are
required.20 In summary, experimental results from dynamic
proton NMR in combination with theoretical DFT calculations
show that the hydrogen bonding between the two 6-DPPon
ligands within [Cl2Pt(6-DPPon)2] (11) results in a thermo-
dynamic stabilization of ca. 15 kcal/mol.
Ab initio calculations of the 1H NMR signals of the hydrogen
atoms in the N-H-N/O-H-O motifs have also been carried
out. In the experimental spectrum, the H signal is found at 12.9/
12.43 ppm (T = 210 K) and disappears from the 11-15 ppm
range at 320 K. This compares with computed 1H NMR signals
of the hydrogen atoms in the N-H N/O-H O motifs at
3 3 3 3 3 3
13.33/13.04 ppm (for the equilibrium structure, φ = 0°), where
the two H-bonds are formed, and signals at 5.12/8.60 ppm for a
structure with broken hydrogen bonds (φ = 80°). Thus, the
computational results are in qualitative agreement with the
experimental data. However, a more accurate assessment of the
temperature-dependent 1H NMR spectrum requires conforma-
tional sampling by means of finite-temperature molecular dy-
namics (MD) simulations together with dedicated force fields
(see above).
Figure 8. Rotation of pyridone and hydroxypyridine rings from the
equilibrium structure. (Top) Change of total energy. (Bottom) Change
of N-H and O-H bond lengths.
UV-Vis Spectroscopy. In addition to NMR spectroscopy,
UV-vis spectroscopy should be capable to provide further
evidence for the formation of the hydrogen-bonding interligand
network within [Cl2Pt(6-DPPon)2] (11). Since we showed that
the isolated 6-DPPon ligand 2 alone energetically prefers to exist
as the pyridone tautomer 2A, detection of the hydroxypyridine
tautomer 2B in the UV-vis spectrum of [Cl2Pt(6-DPPon)2]
(11) would be further evidence for the existence of the inter-
ligand hydrogen bonding. Since the coordination of the ligands
to the platinum center might change the UV spectroscopy of the
pyridone/hydroxypyridine chromophores, we prepared the pla-
tinum complexes with the “fixed tautomer” models 10A and 10B
first (Scheme 9), following the same protocol as for the prepara-
tion of the 6-DPPon complex 11.
Tautomerization after hydrogen bond breaking (step b in
Scheme 8) for compound 11 was also investigated by using DFT
calculations. For this, the barrier for intramolecular OH-N
H-transfer for the open conformation (both H-bonds broken)
was determined and yields an isomerization barrier of 34 kcal/
mol, close to calculations on free 2-pyridone, which found 32
kcal/mol for this process.18 The work on isolated pyridones has
further established that, with varying numbers of water molecules
as proton relays, the barrier for enol-keto tautomerization is
appreciably affected. It decreases to about 12 and 6 kcal/mol with
one and two additional water molecules, respectively. Similar
effects are expected for [Cl2Pt(6-DPPon)2] (11) in contact with
water, which cannot be excluded to be present in catalytic
amounts in an NMR experiment.
While 12B was formed as the cis isomer only, 12A was obtained
as a cis/trans mixture of isomers in the ratio of 45/55, as deter-
The experimental ΔGTS = 15.1 kcal/mol agrees favorably with
the DFT energies for pyridone/hydroxypyridine rotation and
concomitant deletion of the double H-bond motif in (11), but
differs considerably from water-free tautomerization (>30 kcal/
mol). The only competitive process is water-assisted tautomer-
ization (around 12 kcal/mol for free 2-pyridone), which, how-
ever, requires deletion of the double H-bond motif to occur
initially. Thus, the process captured by dynamic proton NMR is
1
mined by 31P NMR spectroscopy and analysis of the JP-Pt
coupling constants. Figure 9 displays the UV-vis spectra of the
complexes 12B, 12A, and 11 in CH2Cl2 solution (c = 2.5 μM).
Indeed, the UV-vis spectrum of 11 appears as the super-
position of the spectra obtained for fixed tautomer complexes
12A and 12B. Hence, the fact that both tautomeric forms 2A and
2B of the 6-DPPon ligand are present in the complex gives
970
dx.doi.org/10.1021/ja108639e |J. Am. Chem. Soc. 2011, 133, 964–975

Journal of the American Chemical Society
ARTICLE
Scheme 9. Synthesis of the Pt Complexes Bearing the Fixed Tautomers of 2
Figure 9. UV-vis spectra of 11 (black), 12A (red), and 12B (blue).
The spectra were recorded in CH2Cl2 at room temperature and a con-
centration of 2.5 μM.
Figure 10. IR spectrum of complex 11 in CCl4 at room temperature
and a concentration of 10-3 M.
Table 5. Comparison of Characteristic Vibrations (cm-1
Obtained from Experiment and by DFT Calculation of
[Cl2Pt(6-DPPon)2] (11)
)
another hint that both tautomeric forms are present in [Cl2Pt(6-
DPPon)2] (11). The only possible explanation for the existence
of the energetically unfavorable hydroxypyridine isomer is its
involvement in hydrogen bonding with the complementary
pyridone tautomer, which more than compensates the energetic
cost for tautomerization through hydrogen bonding.
IR Spectroscopy of 11. For further insight into the bonding
situation within [Cl2Pt(6-DPPon)2] (11), we chose IR spectros-
copy, which should allow us to detect the presence of hydrogen
bonding as well as whether one of the ligands is present in its
hydroxypyridine tautomeric form 2B. The IR spectrum of 11 was
recorded as a solution in CCl4 (c = 10-3 M) and is depicted in
Figure 10. Additionally, DFT calculations of 11 were carried out
to allow assignment of vibrational bands (see Table 5).
expt
B3LYP/6-31G(d,p)a
CO stretching vibration
OH rocking vibration
1653
1668 (1735)
1584/1440
1554 (1616)/1451 (1510)
2500-3500b
N-H N, O-H
3 3 3
O
2500-3500
3 3
a Both scaled and unscaled frequencies (in parentheses) are shown, as in
Table 1. b See Figure 10 and discussion.
and OH O) in the complex. The vibrational transitions
3 3 3
associated with hydrogen motion in such strong hydrogen bonds
are very sensitive to the chemical environment and the corre-
sponding infrared signatures usually exhibit a diffuse character.22a,b
We computed the IR spectrum for the equilibrium structure of
the [Cl2Pt(6-DPPon)] complex 11 and also for a series of
structures for which both, the N-H and O-H bonds were
simultaneously elongated by 0.01, 0.03, and 0.05 Å. The harmo-
The vibration at 1653 cm-1 (Figure 10) can be identifed as the
CO stretching vibration of 2A. Furthermore, the absorptions at
1584 cm-1 (Figure 10) and 1440 cm-1 could be identified as the
OH rocking vibrations of the hydroxypyridine tautomer (Figure 10).
Taking into account the situation for the free 6-DPPon ligand 2,
in which the pyridone form is thermodynamically more stable and
prevalent, the presence of a hydroxypyridine form within complex
11 must be due to the fact that hydrogen bonds are formed
between the two 6-DPPon ligands within complex 11 in solution.
Thus, the bond enthalpy of the hydrogen bonds enables the
transition of the pyridone into the energetically less favored
hydroxypyridine form.20 By comparing the spectrum of free 2 with
the recorded IR spectrum of 11 (Figure 11), it is obvious that the
vibrations at 1584 cm-1 and 1440 cm-1, are only present in the
metal complex, indicating the formation of a heterodimeric species.
The unusually broad peak between 2500 and 3500 cm-1
nic frequencies of the H-stretching modes in the NH N and
3 3 3
OH O motifs are highlighted by red lines in Figure 12, and
3 3 3
clearly link the observed broad band between 2500 cm-1 and
3500 cm-1 to motion along the hydrogen bond.
Experimental and Computed Structure for the [Cl2Pt(6-
DPPon)2]. A final proof for the existence of interligand hydro-
gen bonding within [Cl2Pt(6-DPPon)2] (11) came from an
X-ray crystal structure analysis of suitable single crystals obtained
from a saturated solution of 11 in dichloromethane. The
molecular structure of 11 in the crystal state is depicted in
Figure 13. It clearly shows two 6-DPPon ligands coordinated to
the platinum center in a cis fashion. The two phosphine ligands
shown in Figure 10 is assigned to the hydrogen bonds (NH
N
3 3 3
971
dx.doi.org/10.1021/ja108639e |J. Am. Chem. Soc. 2011, 133, 964–975

Journal of the American Chemical Society
ARTICLE
Table 6. Equilibrium Structure of the [Cl2Pt(6-DPPon)2]
Complex Measured by X-ray and Computed by DFT
X-ray
B3LYP/6-31G(d,p)/LANL2DZ
Pt-P1 (Å)
2.2563(6)
2.2364(6)
97.58(2)
0.84(3)
2.40(3)
3.071(3)
137(3)
2.313
2.306
106.0
1.033
1.990
2.966
156.3
1.003
1.632
2.631
173.3
Pt-P2 (Å)
P1-Pt-P2 (deg)
N1-H (Å)
N2 H (Å)
3 3 3
N1 N2 (Å)
3 3 3
N1
H
N2 (deg)
3 3 3 3 3 3
O2-H2 (Å)
0.86(4)
1.85(4)
2.683(3)
164(4)
O1 H2 (Å)
Figure 11. Direct comparison between the IR spectra of 2 and 11.
3 3 3
O1 O2 (Å)
3 3 3
O2 H-O1 (deg)
3 3 3
finally suggests the formation of hydrogen bonds between the
two tautomeric complementary 6-DPPon ligands based on the
bond distances between O1-O2 and N1-N2, respectively (see
Table 6). A comparison of the experimental data with those
obtained from a DFT calculation is in good agreement (Table 6).
For example, the Pt-P bond lengths differ by only 0.06 Å, and
the P-Pt-P natural bite angles differ by 9°. The distances
between the pyridone ring and the hydroxypyridine ring are also
comparable: the difference in N N distances is 0.1 Å, while
3 3 3
that of O O distances is 0.05 Å. However, in the DFT
3 3 3
structure, the N-H and O-H bond lengths are elongated to
1.033 and 1.003 Å, respectively, and the N
Figure 12. Computed IR spectrum for a) equilibrium structures b)
N-H and O-H bond lengths increased by 0.01 Å; c) N-H and O-H
bond lengths increased by 0.03 Å; d) N-H and O-H bond lengths
increased by 0.05 Å, and with all other degrees of freedom optimized.
The red lines indicate the frequencies of hydrogen bond stretching as
assigned from the normal modes. The N-H/O-H stretching frequen-
cies are found to be very sensitive to the geometry of the H-bonds.
H
N and
3 3 3 3 3 3
O
H
O motifs are almost linear. This is a strong indication
3 3 3 3 3 3
for hydrogen bonding between the 6-DPPon ligands, which is
not seen in the X-ray structure since the corresponding electron
density is asymmetrical and not centered at the position of the
nucleus. As mentioned before, in general, locating hydrogen
atoms in X-ray crystallography is difficult because of their low
scattering power and librational effect, and the refined structure
depends on the computational algorithms used.
’ CONCLUSIONS
Motivated by previous findings, which had shown that transi-
tion metal catalysts based on the 6-diphenylphosphanylpyridone
ligand (6-DPPon, 2) display unique properties and suggested
that 6-DPPon behaves as a monodentate-to-bidentate self-
assembling ligand, we herein performed a thorough study on
the bonding situation of this ligand, alone and in the coordination
sphere of a late transition metal. Thus, combining a number of
spectroscopic methods (UV-vis, IR, NMR, X-ray) corroborated
by DFT calculations, insights into the unique structural char-
acteristics of 6-DPPon (2) were gained. The free ligand 2 prefers
to exist as the pyridone tautomer 2A and dimerizes to the
pyridone-pyridone dimer 4A in solution as well as in the crystal
state. The corresponding hydroxypyridine tautomer 2B is slightly
disfavored energetically (ca. 0.9 kcal/mol within the up-confor-
mer relevant for metal coordination). Hence, hydrogen bond
formation within the complex is expected to easily compensate
this small energy penalty. This was confirmed experimentally by
studying the coordination properties of 2 within the model
complex [Cl2Pt(6-DPPon)2] (11). All experimental and theoret-
ical methods used are consistent with the existence of a hydro-
gen-bonding interligand network in solution as well as in the
crystal state of 11 between one 6-DPPon ligand existing as the
Figure 13. Crystal structure of [Cl2Pt(6-DPPon)2] complex 11. The
hydrogen bonds are represented as dashed lines.
are crystallographically distinct. One of them exists in the lactam
form 2A and the other in the lactim form 2B, which is obvious
from the different bond lengths within the corresponding
heterocycles (pyridone vs hydroxypyridine). The X-ray structure
972
dx.doi.org/10.1021/ja108639e |J. Am. Chem. Soc. 2011, 133, 964–975

Journal of the American Chemical Society
ARTICLE
pyridone tautomer 2A and the other ligand occupying the
complementary hydroxypyridine form 2B. Dynamic proton
NMR allowed us to determine the barrier for interligand hydro-
gen bond breaking and, in combination with theory, led to an
estimate for the enthalpic stabilization through hydrogen bond-
ing of 15 kcal/mol. Hence, a theoretical understanding of the
ground-state structural features of this self-assembling ligand 2 is
now available and ready to be used to further explore its unique
properties as a ligand in homogeneous catalysis.
times with 150 mL portions of diethyl ether, and dried over Na2SO4.
Evaporation of the solvent in vacuo and recrystallization of the residual
oil from methanol yielded 31.0 g (92.4 mmol, 79%) of 2-tert-butoxy-6-
diphenylphosphanylpyridine (8) as a white solid.
Alternatively, 2-tert-butoxy-6-bromopyridine (7B, 100 mg, 0.43
mmol, 1.00 equiv) was dissolved in Et2O (3 mL) and cooled to 0 °C.
After addition of n-BuLi (0.28 mL, 0.43 mmol, 1.56 M in hexane), the
solution was stirred for 60 min at 0 °C. Chlorodiphenylphosphine was
added, and the reaction mixture was stirred at room temperature for an
additional 60 min. Water (1 mL) was added and the solvent removed
under reduced pressure. Purification of the residue by column chroma-
tography over silica gel (CH2Cl2) yielded 125 mg (0.34 mmol, 77%) of
2-tert-butoxy-6-diphenylphosphanylpyridine (8) as a white solid.
mp = 77 °C. 1H NMR (400.1 MHz, CDCl3, ppm) = 7.51-7.44 (m,
4H), 7.11-7.04 (m, 6H), 6.91 (ddd, J = 8.3, 7.3, 2.8 Hz, 1H), 6.75 (dd,
J = 7.3, 2.8 Hz, 1H), 6.45 (d, J = 8.3 Hz, 1H), 1.41 (s, 9H). 13C NMR
(100.6 MHz, CDCl3, ppm) = 164.1, 160.5, 138.0, 137.6, 134.7,
128.9, 128.7, 121.4, 112.2, 79.6, 28.5. 31P NMR (121.5 MHz, CDCl3,
ppm) = -1.7. CHN analysis calcd: C, 75.21; H, 6.61; N, 4.18. Found: C,
74.98; H, 6.57; N, 4.09.
Synthesis of 6-Diphenylphosphanyl-1H-pyridin-2-one (6-DPPon,
2). 2-tert-Butoxy-6-diphenylphosphanylpyridine (8, 9.7 g, 28.9 mmol)
was dissolved in concentrated formic acid (100 mL) saturated with
argon. After stirring for 30 min at room temperature, the solution was
diluted with water (distilled, 120 mL). The precipitate was collected by
filtration, washed with 30 mL of aqueous formic acid (2:1, v/v), and
dried. 6-Diphenylphosphanyl-1H-pyridin-2-one (2) was obtained as a
white solid (5.6 g, 20.05 mmol, 69%). The combined aqueous formic
acid solutions were concentrated in vacuo, and the residue was recrys-
tallized from acetone. This yielded another 1.8 g (6.45 mmol, 22%) of
6-diphenylphosphanyl-1H-pyridin-2-one (2).
’ METHODS
Experimental Methods. All reagents were obtained commer-
cially unless otherwise noted. Reactions were performed using oven-
dried glassware under an atmosphere of argon. Air- and moisture-
sensitive liquids and solutions were transferred via syringe. Organic
solutions were concentrated under reduced pressure (ca. 20 mbar) by
rotary evaporation. Toluene was distilled from sodium, dichlormethane
and MeOH were distilled from CaH2, THF was distilled from potas-
sium, and diethyl ether was refluxed over a potassium-sodium alloy
using benzophenone ketyl radical as indicator and distilled prior to use.
All solvents were stored under an argon atmosphere. Chromatographic
purification of products was accomplished using flash chromatography
on Macherey-Nagel silica gel 60 (230-400 mesh). Nuclear magnetic
resonance spectra were acquired on a Varian Mercury spectrometer
(300, 121.5, and 75.5 MHz for 1H, 31P, and 13C, respectively) and on a
Bruker AMX 400 (400.1 and 100.6 MHz for 1H and 13C, respectively)
and are referenced internally according to residual protio solvent signals.
1
Data for H and 31P NMR are recorded as follows: chemical shift (δ,
ppm), multiplicity (s, singlet; bs, broad singlet; d, doublet; t, triplet; pt,
pseudotriplet; q, quartet; quint, quintet; m, multiplet), coupling con-
stant (Hz), integration. Data for 13C NMR are reported in terms of
chemical shift (δ, ppm). 9 was prepared from commercially available 6B
according to the literature known procedure.23
Synthesis of 2-tert-Butoxy-6-chloropyridine (7A). To a solution of
2,6-dichloropyridine (6A) (10.0 g, 67.6 mmol, 1.00 equiv) in toluene
(150 mL) was added potassium tert-butoxide (9.10 g, 81.1 mmol, 1.20
equiv). After heating for 6 h to 80 °C, the suspension was filtered over
Celite and the solvent removed under reduced pressure. Distillation
(bp = 200 °C, 10-2 mbar) yielded 11.98 g (64.53 mmol, 96%) of 2-tert-
butoxy-6-chloropyridine (7A) as a colorless liquid.
mp = 187 °C. 1H NMR (400.1 MHz, CDCl3, ppm) = 9.29 (br s, 1H),
7.45-7.25 (m, 11H), 6.51 (d, J = 9.0 Hz, 1H), 6.20 (t, J = 6.3 Hz, 1H).
13C NMR (100.6 MHz, CDCl3, ppm) = 163.8, 146.6, 140.5, 133.9,
132.6, 130.2, 129.3, 120.7, 114.0. 31P NMR (121.5 MHz, CDCl3,
ppm) = -8.5. CHN analysis calcd: C, 73.11; H, 5.05; N, 5.02. Found:
C, 72.97; H, 5.18; N, 4.73.
Synthesis of 6-Diphenylphosphanyl-1-methylpyridin-2-one (6-
DPMePon, 10A). To a solution of 6-diphenylphosphanylpyridine-
2(1H)-one (2, 500 mg, 1.79 mmol, 1.00 equiv) in DMF (8 mL) was
added sodium hydride (46.0 mg, 1.91 mmol, 1.07 equiv) in one portion.
After gas evolution ceased, the solution was cooled to 0 °C. Methyl
iodide (255 mg, 1.80 mmol, 1.00 equiv) was added, and the reaction
mixture was stirred for 30 min at room temperature. After addition of
water (20 mL), the organic layer was separated and washed with Et2O.
After drying with MgSO4, the solvent was removed under reduced
pressure, and the residue was purified by column chromatography over
silica gel (CH2Cl2:PE 1:1 to CH2Cl2 to CH2Cl2:EE 1:1), yielding 300
mg (1.02 mmol, 57%) of the desired product as a colorless liquid.
mp = 159 °C. 1H NMR (400.1 MHz, CDCl3, ppm) = 7.37-7.46 (m,
6H), 7.29-7.35 (m, 4H), 7.14 (ddd, J = 8.8 Hz, J = 7.1 Hz, J = 1.5 Hz,
1H), 6.54 (d, J = 9.0 Hz, 1H), 5.62 (d, J = 6.9 Hz, 1H), 3.58 (s, 3H).
13C NMR (100.6 MHz, CDCl3, ppm) = 164.2, 150.6, 137.8, 134.3,
132.9, 130.2, 129.2, 119.8, 113.3, 33.8. 31P NMR (121.5 MHz, CDCl3,
ppm) = -11.1. CHN analysis calcd: C, 73.71; H, 5.50; N, 4.78. Found:
C, 73.42; H, 5.73; N, 4.75.
Synthesis of 2-Methoxy-6-diphenylphosphinopyridine (2-MeODPP,
10B). To a solution of 2-methoxy-6-bromopyridine (8, 855 mg,
4.55 mmol, 1.00 equiv) in Et2O (25 mL) was added n-BuLi (3.25 mL,
4.55 mmol, 1.40 M in hexane, 1.00 equiv) dropwise at 0 °C. The solution
was stirred for 1 h at 0 °C, and chlorodiphenylphosphane (0.82 mL,
1.00 g, 4.55 mmol, 1.00 equiv) was added at this temperature. The
reaction mixture was allowed to warm to room temperature and stirred
for 5 h. The reaction mixture was diluted with CH2Cl2 (50 mL) and
1H NMR (400.1 MHz, CDCl3, ppm) = 7.43 (t, J = 7.7 Hz, 1H), 6.81
(d, J = 7.3 Hz, 1H), 6.53 (d, J = 8.2 Hz, 1H), 1.58 (s, 9H). 13C NMR
(100.6 MHz, CDCl3, ppm) = 163.3, 147.7, 140.2, 115.7, 111.3, 80.9,
28.6. CHN analysis calcd: C, 58.23; H, 6.52; N, 7.54. Found: C, 58.27; H,
6.65; N, 7.48.
Synthesis of 2-tert-Butoxy-6-bromopyridine (7B). To a solution of
2,6-dibromopyridine (6B) (9.48 g, 40.0 mmol, 1.00 equiv) in toluene
(150 mL) was added potassium tert-butoxide, and the suspension was
stirred for 4 h at 80 °C. The reaction mixture was filtered over Celite and
the solvent removed under reduced pressure. Distillation (bp = 120 °C,
oil pump vacuum) yielded 8.50 g (36.7 mmol, 92%) of the desired
product as a colorless liquid.
1H NMR (400.1 MHz, CDCl3, ppm) = 7.34 (m, 1H), 6.98 (d, J =
7.3 Hz, 1H), 6.57 (d, J = 8.2 Hz, 1H), 1.59 (s, 9H). 13C NMR (100.6 MHz,
CDCl3, ppm) = 163.1, 140.1, 137.7, 119.5, 111.5, 80.9, 28.5.
Synthesis of 2-tert-Butoxy-6-diphenylphosphanylpyridine (8). To
liquid ammonia (ca. 500 mL) at -78 °C was added sodium (5.40 g,
235 mmol, 2.03 equiv) over 10 min. The dark blue solution was treated
portionwise first with triphenylphosphine (30.4 g, 116 mmol, 1.00 equiv)
and then, after stirring for 2 h at -78 °C, with 2-tert-butoxy-6-chloro-
pyridine (7A, 21.5 g, 116 mmol, 1.00 equiv). After addition of 175 mL of
tetrahydrofuran, the ammonia was allowed to evaporate overnight. The
residue was quenched with 200 mL of water (distilled), extracted three
973
dx.doi.org/10.1021/ja108639e |J. Am. Chem. Soc. 2011, 133, 964–975

Journal of the American Chemical Society
ARTICLE
filtered through silica gel. The solvent was removed under reduced
pressure, and the residue was purified by column chromatography over
silica gel (PE/CH2Cl2 2:1 to 1:1). Distillation (bp = 160 °C, 10-2 mbar)
yielded 1.10 g (3.60 mmol, 79%) of the desired product as a colorless
solid.
analysis and the GIAO gauge method,27,28 respectively. All electronic
structure calculations were carried out using the Gaussian03 suite of
programs29 with the grid=ultrafine option.
’ ASSOCIATED CONTENT
mp = 59 °C. 1H NMR (400.1 MHz, CDCl3, ppm) = 7.40-7.48 (m,
5H), 7.33-7.38 (m, 6H), 6.77 (dd, J = 7.2 Hz, J = 2.2 Hz, 1H), 6.63 (d,
J = 8.3 Hz, 1H), 3.83 (s, 3H). 13C NMR (100.6 MHz, CDCl3, ppm) =
163.7, 160.2, 138.1, 136.8, 134.3, 128.9, 128.4, 121.8, 110.0, 53.5. 31P
NMR (121.5 MHz, CDCl3, ppm) = -2.6. CHN analysis calcd: C, 73.71;
H, 5.50; N, 4.78. Found: C, 73.54; H, 5.50; N, 4.72.
S
Supporting Information. Energies and Cartesian coor-
b
dinates of optimized 2A, 2B, 4A, 4B, and 11; 31P NMR spectra of
12A and 12B; complete ref 29; details of the NMR dilution
experiment; X-ray crystallographic data, in CIF format, for 4A
and 11. This material is available free of charge via the Internet at
http://pubs.acs.org.
Synthesis of [Cl2Pt(6-diphenylphosphanyl-1H-pyridin-2-one)2]
(11). To a solution of cis-[Cl2Pt(COD)] (68.4 mg, 182 μmol, 1.00 equiv)
in CH2Cl2 (2.5 mL) was added 6-diphenylphosphanyl-1H-pyridin-2-
one (2, 102.0 mg, 366 μmol, 2.00 equiv) at room temperature. The
solution was diluted with CH2Cl2 (2.5 mL) and stirred for 2 h at room
temperature. The solvent was removed under reduced pressure and the
resulting solid washed two times with pentane (2 ꢀ 5 mL). The product
was obtained as a colorless solid (117 mg, 182 μmol, 100%).
1H NMR (400 MHz, CDCl3, ppm) = 12.90 (bs, 1H), 12.43 (bs, 1H),
7.50-7.60 (m, 8H), 7.30-7.38 (m, 6H), 7.22 (pt, J = 6.8 Hz, 8H), 6.73
(d, J = 8.8 Hz, 2H), 6.48 (pt, J = 6.4 Hz, 2H). 13C NMR (100.6 MHz,
CDCl3, ppm) = 164.6, 143.8, 140.2, 134.8, 131.5, 128.3, 128.1, 118.3,
117.8. 31P NMR (121.5 MHz, CDCl3, ppm) = 13.0 (1JP-Pt = 3591.4 Hz).
Suitable crystals for X-ray analysis were grown from a solution of
20 mg of 11 in CH2Cl2 (absolute, 1 mL) at room temperature.
UV-Vis Spectroscopy. Solvents used for UV-vis spectroscopy
were purified as described in the Experimental Methods section. The
spectra of the fixed tautomers 10A and 10B and the free ligand 2 were
recorded on a Perkin Elmer 15 spectrometer at room temperature and a
concentration of 0.14 mM. The spectra of the platinum complexes 11,
12A, and 12B were recorded using a Perkin Elmer Lambda 950 spectro-
meter at a concentration of 2.5 μM in CH2Cl2 at room temperature. The
complexes [Cl2Pt(6-diphenylphosphanyl-1-methyl-pyridine-2-one)2] (12A)
and [Cl2Pt(2-methoxy-6-diphenylphosphanylpyridine)2] (12B) were
prepared by stirring [Cl2Pt(COD)] (10.9 mg, 0.02 mmol, 1.00 equiv)
with 6-diphenylphosphino-1-methylpyridin-2-one (10A) and 2-methoxy-
6-diphenylphosphinopyridine (10B) (10.0 mg, 0.03 mmol, 2.00 equiv)
in CH2Cl2 (2.5 mL) at room temperature for 2 h, respectively. The
solvent was removed under reduced pressure, and the residue was
diluted to a concentration of 2.5 μM. An aliquot was taken from the
solution to confirm the formation of the desired complex by means of
31P NMR.
’ AUTHOR INFORMATION
Corresponding Author
bernhard.breit@chemie.uni-freiburg.de; m.meuwly@unibas.ch
’ ACKNOWLEDGMENT
Funding of both groups via the IRTG 1038 “Catalysts and
Catalytic Reaction for Organic Synthesis” from DFG and the
Swiss National Science Foundation is acknowledged. The work
in Freiburg was further supported by the Fonds of the Chemical
Industry (Ph.D. fellowship to U.G.) and the Alfred-Krupp Award
(to B.B.). The Freiburg group is indebted to the companies BASF
and Umicore for gifts of precious metal salts. The work in Basel is
supported by the Swiss National Science Foundation under
Grant No. 200021-117810 (to M.M.).
’ REFERENCES
(1) Miyashita, A.; Yasuda, A.; Takaya, H.; Toriumi, K.; Ito, T.;
Souchi, T.; Noyori, R. J. Am. Chem. Soc. 1980, 102, 7932.
(2) (a) Devon, T. J.; Philips, G. W.; Puckette, T. A.; Stavinoha, J. L.;
Vanderbilt, J. J. U.S. Patent 4 694 109,1987. (b) Billig, E.; Abatjoglou,
A. G.; Bryant, D. R. U.S. Patent 4 769 498, 1988. (c) Cunny, G. D.;
Buchwald, S. L. J. Am. Chem. Soc. 1993, 114, 5535.
(3) Kranenburg, M.; v. d. Burgt, Y. E. M.; Kamer, P. C. J.; v. Leeuwen,
P. W. N. M.; Goubitz, K.; Fraanje, J. Organometallics 1995, 14, 3081.
(4) (a) Breit, B. Angew. Chem. 2005, 117, 6976. (b) Angew. Chem.,
Int. Ed. 2005, 44, 6816. (c) Breit, B.; Seiche, W. J. Am. Chem. Soc. 2003,
125, 6608. (d) Pure Appl. Chem. 2006, 78, 249.
(5) (a) For a review on alternative approaches to self-assembled
ligand/catalyst libraries, see ref 4b. See also: (b) Takacs, J. M.; Reddy,
D. S.; Moteki, S. A.; Wu, D.; Palencia, H. J. Am. Chem. Soc. 2004, 126,
4494. (c) Slagt, V. F.; R€oder, M.; Kamer, P. C. J.; van Leeuwen,
P. W. N. M.; Reek, J. N. H. J. Am. Chem. Soc. 2004, 126, 4056.
(d) Slagt, V. F.; van Leeuwen, P. W. N. M.; Reek, J. N. H. Angew. Chem.,
Int. Ed. 2003, 42, 5619. (e) Slagt, V. F.; van Leeuwen, P. W. N. M.; Reek,
J. N. H. Chem. Commun. 2003, 2474. (f) Ding, K.; Du, H.; Yuan, Y.;
cis-[Cl2Pt(6-diphenylphosphanyl-1-methylpyridin-2-one)2] (12A):
31P NMR (121.5 MHz, CDCl3, ppm) = 12.8 (1JP-Pt = 3626.8 Hz).
trans-[Cl2Pt(6-diphenylphosphanyl-1-methylpyridin-2-one)2]: 31P NMR
(121.5 MHz, CDCl3, ppm) = 17.3 (1JP-Pt = 2686.4 Hz).
cis-[Cl2Pt(2-methoxy-6-diphenylphosphanylpyridine)2] (12B):31PNMR
(121.5 MHz, CDCl3, ppm) = 13.8 (1JP-Pt = 3655.4 Hz).
Long, J. Chem. Eur. J. 2004, 10, 2872. (g) Sandee, A. J.; v. d. Burg,
IR Spectroscopy. Solvents used for IR spectroscopy were purified
as described in the Experimental Methods section. All spectra were re-
corded using a ReactIR 45 m FT-IR spectrometer. In a typical experi-
ment the IR flask, equipped with a magnetic stirrer, was filled with CCl4
(4 mL) under an inert atmosphere of argon at room temperature. After
16 min, solvent spectra were recorded and subtracted, and a solution of
11 (2.1 mg, 0.0025 mmol) or 2 (140 mg, 0.5 mmol), dissolved in CCl4
(1 mL), was added. The solution was stirred for 1 h 45 min, and a
spectrum was recorded every 60 s.
Computational Methods. All structures are fully optimized by
using Density Functional Theory (DFT) with the B3LYP functional.24
All calculations were carried out with the all-electron 6-31G(d,p) basis
set25 for C, H, N, O, and P atoms and an LANL2DZ effective core
potential26 for Pt and Cl atoms. Infrared (IR) and NMR spectra are
calculated at the same level of theory from a harmonic vibrational
;
A. M.; Reek, J. N. H. Chem. Commun. 2007, 864. (h) Gulyꢀas, H.; Benet-
Buchholz, J.; Escudero-Adan, E. C.; Freixaa, Z.; van Leeuwen, P. W. N.
M. Chem. Eur. J. 2007, 13, 3424.
;
(6) Beak, P. Acc. Chem. Res. 1977, 10, 186.
(7) Fischer, C. B.; Polborn, K.; Steiniger, H.; Zipse, H. Z. Naturforsch.
2004, 59b, 1121.
(8) Chou, P.-T.; Wei, C.-Y.; Hung, F.-T. J. Phys. Chem. B 1997, 101,
9119.
(9) For a review on complementary hydrogen-bonding motifs for
the self-assembly of supramolecular architectures see: Lehn, J.-M.;
Krische, M. J. Struct. Bonding (Berlin) 2000, 94, 3.
(10) Akazome, M.; Suzuki, S.; Shimizu, Y.; Henmi, K.; Ogura, K.
J. Org. Chem. 2000, 65, 6917.
(11) Breit, B.; Krische, M. J. Metal catalyzed reductive C-C bond
formation; Springer: Berlin-Heidelberg, 2007.
974
dx.doi.org/10.1021/ja108639e |J. Am. Chem. Soc. 2011, 133, 964–975

Journal of the American Chemical Society
ARTICLE
(12) G€unzler, H.; Gremlich, H.-U. IR-Spektroskopie; Wiley-VCH:
Weinheim, 2003.
(13) Irikura, K. K.; Johnson, R. D., III; Kacker, R. N. J. Phys. Chem. A
2005, 109, 8430.
(14) For details see the Supporting Information.
(15) (a) Pregosin, P. S.; Kunz, R. W. 31P and 13C NMR of Transition
Metal Phosphine Complexes in NMR Basic Principles and Progress; Springer:
Heidelberg, 1979. (b) Alt, H. G.; Baumgartner, R.; Brune, H. A. Chem.
Ber. 1986, 119, 1694.
(16) The fact that the two P atoms in the platinum complex display a
singlet in the 31P NMR spectrum means that they are magnetically
equivalent at the NMR time scale. This can be explained by our
calculations which identified a transition state for proton transfer by
DFT. The activation energy was calculated to be 7.9 kcal/mol, which
corresponds to a rate for tautomerization of 0.011 ns-1. This is much
faster than the microsecond (NMR) time scale, which renders the P
atom connected to the pyridone ring and the P atom connected to the
hydroxypyridine system indistinguishable in the 31P NMR experiment.
(17) Friebolin, H. Ein- und zweidimensionale NMR-Spektroskopie;
Wiley-VCH: Weinheim, 1999.
(18) Tsuchida, N.; Yamabe, S. J. Phys. Chem. A 2005, 109, 1974.
(19) Tubert-Brohman, I.; Schmid, M.; Meuwly, M. J. Chem. Theory
Comput. 2009, 5, 530.
(20) Lammers, S.; Lutz, S.; Meuwly, M. J. Comput. Chem. 2008, 29,
1048.
(21) Shuklov, I. A.; Dubrovina, N. V.; Barsch, E.; Ludwig, R.;
Michalik, D.; B€orner, A. Chem. Commun. 2009, 1535.
(22) (a) Meuwly, M.; M€uller, A.; Leutwyler, S. Phys. Chem. Chem.
Phys. 2003, 5, 2663. (b) Roscioli, J. R.; McCunn, L. R.; Johnson, M. A.
Science 2007, 316, 249.
(23) Zhang, W.; Luo, Z.; Chen, C. H.-T.; Curran, D. P. J. Am. Chem.
Soc. 2002, 124, 10443.
(24) Becke, A. D. J. Chem. Phys. 1993, 98, 5648.
(25) Hehre, W. J.; Ditchfield, R.; Pople, J. A. J. Chem. Phys. 1972, 56,
2257.
(26) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 270.
(27) Ditchfield, R. Mol. Phys. 1974, 27, 789.
(28) Wolinski, K.; Hilton, J. F.; Pulay, P. J. Am. Chem. Soc. 1990, 112,
8251.
(29) Frisch, M. J.; et al. Gaussian 03, Revision C.02; Gaussian, Inc.:
Wallingford, CT, 2004.
975
dx.doi.org/10.1021/ja108639e |J. Am. Chem. Soc. 2011, 133, 964–975