Synthesis and reactivity of nickel and palladium fluoride complexes with PCP pincer ligands. NMR-based assessment of electron-donating properties of fluoride and other monoanionic ligands

Article abstract of DOI:10.1021/om2009793

A series of complexes of the type [(^iPrPCP)M-L] (M = Ni, Pd), containing the 2,6-bis(diisopropylphosphinomethyl)phenyl (^iPrPCP) pincer ligand and simple monoanionic ligands L (F, Cl, Br, H, Me, Ph, NO ₃, and OTf), have bee

Full text of DOI:10.1021/om2009793

Article
pubs.acs.org/Organometallics
Synthesis and Reactivity of Nickel and Palladium Fluoride Complexes
with PCP Pincer Ligands. NMR-Based Assessment of Electron-
Donating Properties of Fluoride and Other Monoanionic Ligands
†
†
‡
†
,†
́
́
Luis Miguel Martınez-Prieto, Cristobal Melero, Diego del Rıo, Pilar Palma, Juan Campora,*
́
́
†
́
and Eleuterio Alvarez
†
́
Instituto de Investigaciones Quımicas, CSIC-Universidad de Sevilla, c/Amer
́
ico Vespucio, 49, 41092, Sevilla, Spain
‡SRI International, 333 Ravenswood Avenue, Menlo Park, California 94025, United States
S
* Supporting Information
ABSTRACT: A series of complexes of the type [(iPrPCP)M-
L] (M
= Ni, Pd), containing the 2, 6-bis-
(diisopropylphosphinomethyl)phenyl (iPrPCP) pincer ligand
and simple monoanionic ligands L (F, Cl, Br, H, Me, Ph, NO3,
and OTf), have been synthesized and characterized. The
fluoride derivatives [(iPrPCP)M-F] were prepared from the
halides [(iPrPCP)M-X] (M = Ni, X = Br; M = Pd, X = I) by
exchange reactions with AgF or, alternatively, by protonolysis
of the methyl complexes [(iPrPCP)M-Me] with Et3N·3HF
(TREAT-HF). A survey of the 13C NMR data for the new
complexes and previously reported ones with L = OH, OMe,
and NH2 revealed significant trends that can be directly related
to the electronic properties of the anionic ligands L coordinated to Ni(II) and Pd(II) centers. The chemical shift of the 13C
resonance of the metal-bound ipso carbon is very sensitive to the σ-donating capacity of L, which increases in the order L = OTf
> NO3 > F > OAc > OMe > Cl > OH ≈ Br > I > NH2 > Ph > Me > H. In addition, the reactivity of [(iPrPCP)M-F] complexes
was investigated. These readily undergo fluoride exchange with LiOMe, affording the corresponding methoxides [(iPrPCP)M-
OMe] and LiF. They also react with n-C12H13I to cleanly yield [(iPrPCP)M-I] and n-C12H13F. Comparison of the rates of the
latter reactions and analogous ones involving bromide or chloride complexes showed that the fluorides are significantly more
nucleophilic, particularly the palladium derivative. This property was applied in a palladium-catalyzed reaction for the synthesis of
alkyl fluorides using AgF as the fluoride source.
alkoxo, and amido complexes of Ni and Pd containing the
strongly donating and relatively bulky 2,6-bis-
((diisopropylphosphino)methyl)phenyl ligand (iPrPCP), which
imparts a balance between stability and reactivity to such
complexes.5
Late-transition-metal fluoride complexes have been recently
the object of much attention, particularly in the context of C−F
bond activation6 and formation,1b,7 as well as in relation with
their use as sources of the hypernucleophilic (and elusive)
“naked fluoride” anion.8 In addition, fluoride complexes can be
valuable starting materials in organometallic synthesis. For
example, we have shown before that the fluoride ligand is
cleanly displaced from the precursor complex [(dippe)Ni(Me)-
(F)] by lithium alkoxides and amides, providing a versatile
route to monomeric methyl alkoxo species and methyl amides
that would have been difficult to prepare by a different
INTRODUCTION
■
In the last two decades, late-transition-metal complexes
containing covalently bound hard monoanionic ligands (L)
such as amides, alkoxides, and fluorides have been the subject
of sustained interest.1,2 This class of compounds is involved in
many catalytic reactions, such as hydrogen transfer hydro-
genations, aerobic oxidations, and carbon−heteroatom cou-
plings.3 The ability to catalyze these processes is connected to
the unique reactivity of the M−L linkage, which combines high
nucleophilicity and basicity and the ability to undergo many
typical organometallic reactions: e.g., insertion reactions, β-
hydrogen elimination, and C−heteroatom reductive couplings.4
Although the chemistry of late-transition-metal complexes
containing terminal (nonbridging) M−N, M−O, or M−F
bonds has lately experienced a considerable development, their
synthesis is still challenging. This is due to a number of reasons,
such as the tendency of terminal ligands to bridge other metal
centers, thermal instability, and hydrolytic sensitivity. A useful
strategy for minimizing such problems is using rigid PCP pincer
chelates to stabilize the metal-containing moiety. We have
previously prepared and studied several monomeric hydroxo,
Special Issue: Fluorine in Organometallic Chemistry
Received: October 14, 2011
Published: January 17, 2012
© 2012 American Chemical Society
1425
dx.doi.org/10.1021/om2009793 | Organometallics 2012, 31, 1425−1438

Organometallics
Article
Scheme 1
method.9 Although a number of monomeric nickel and
palladium fluorides stabilized by monodentate or bidentate
ligands have been reported in the literature, only a few
examples contain pincer-type ligands.10 These would be
interesting candidates to use in the generation of new C−F
bonds, since many pincer complexes have found interesting
applications as catalysts for coupling reactions.11 Herein we
report the synthesis of new Ni(II) and Pd(II) fluoride
complexes: the pincer derivatives [(iPrPCP)Ni-F] and
[(iPrPCP)Pd-F]. We also investigate the use of Ni and Pd
pincers as starting materials for the synthesis of organometallic
alkoxides and their ability to catalyze fluoride exchange between
AgF and alkyl halides.
One of the most striking questions regarding the bonding
and reactivity of late-transition-metal complexes with strongly
π-electron-donating ligands such as OR, NR2, and F is to what
extent these metal−ligand linkages are destabilized by non-
bonding dπ−pπ interactions between the filled d orbitals in the
metal and the lone electron pairs in the heteroatom.12
Answering this question would improve the understanding of
the reactivity of compounds containing strong π-donor ligands
such as fluoride. Spectroscopic data have been of critical
importance to support the concept of repulsive dπ−pπ
interactions.13,14 The availability of spectroscopic data for
many pincer complexes of the type [(iPrPCP)M-L] containing
widely different anionic ligands L provides an unique
opportunity for investigating their correlation to the σ/π
donor capacity of L. With the aim of extending this comparison,
we have prepared some new derivatives with L = triflate, nitrate,
acetate, methyl, phenyl, hydride.
We originally synthesized the nickel precursor [(iPrPCP)Ni-
Br] by oxidative addition of 1-bromo-2,6-bis-
(diisopropylphosphinomethyl)benzene (iPrPCBrP) to Ni-
(cod)2. A similar reaction was used to prepare the Pd analogue
[(iPr(PCP)Pd-Br], using [Pd(η5-Cp)(η3-C3H5)], generated in
situ from [Pd(η3-C3H5)(μ-Cl)]2 and NaCp, as the Pd(0)
source (see the Experimental Section). For the synthesis of
[(iPrPCP)Pd-Cl], we relied on literature procedures for
involving direct palladation of the diphosphine 2,6-bis-
((diisopropylphosphino)methyl)benzene (iPrPCHP) by Pd(II)
sources.15 Several pincer derivatives of nickel have also been
prepared by direct metalation,16 including the chloro derivative
[(iPrPCP)Ni-Cl].17 This method can be regarded as being more
practical than those based on oxidative addition methods,
because it does not require the previous synthesis of
organometallic precursors. Hence, we developed a new
synthesis of [(iPrPCP)Ni-Br] involving direct reaction of
(iPrPCP)H and NiBr2(dme), using triethylamine to neutralize
the resulting HBr. This method leads to the desired nickel
complex in nearly the same yield as achieved by oxidative
addition (ca. 80%).
As shown in Scheme 1, complexes [(iPr(PCP)Ni-Br] and
[(iPrPCP)Pd-Cl] are starting materials for the synthesis of other
compounds. We previously reported that the palladium nitrate
complex is readily obtained by reacting the corresponding
chloro precursor with silver nitrate.5a Similarly, the nickel
analogue is also prepared from the corresponding bromide
precursor and AgNO3, but in this case special care should be
taken to adjust 1/1 Ni/Ag stoichiometry. This is because an
excess of the silver reagent causes the oxidation of the product
to an stable Ni(III) complex, [(iPrPCP)Ni(NO3)2], which will
be described elsewhere.18 The acetate complexes [(iPrPCP)M-
OAc] are obtained in an analogous fashion from the halide
precursors and an stoichiometric amount of silver acetate.
Transmetalation reactions of halide precursors with methyl-
or phenyllithium also proceed easily, affording the correspond-
ing derivatives [(iPrPCP)M-R] (R = Me, Ph) in good yields.
The Pd phenyl derivative was mentioned in the literature
before.15a We characterized the nickel methyl and phenyl
complexes by X-ray diffraction. Figures 1 and 2 represent one
RESULTS AND DISCUSSION
■
Synthesis of New Pincer Derivatives of Ni and Pd. We
have described previously the hydroxo, alkoxo, and amido
derivatives of Ni and Pd stabilized by the iPrPCP pincer ligand.5
In this section we detail the synthesis of new complexes
[(iPrPCP)M-L] containing anionic ligands L with widely
different electron-donor strengths, including different halides,
triflates, nitrates, alkyls, and hydrides.
1426
dx.doi.org/10.1021/om2009793 | Organometallics 2012, 31, 1425−1438

Organometallics
Article
Figure 3. Crystal structure of 1-Me-2,6-(iPr2P(O)CH2)2C6H3·H2O.
Selected distances (Å) and angles (deg): C1−C21, 1.509(3); P1−O1,
1.4978(14); P2−O2, 1.4867(13); O1−H3BO, 1.884(17); O3−H3BO,
0.881(15); O1−O3, 2.763(3); O2−H3AO, 1.93(2); O3−H3AO,
0.921(15); O2−O3, 2.842(3); O1−H3BO−O3, 175(2); O2−H3AO−
O3, 174(2); H3BO−O3−H3AO, 104(2).
Figure 1. Crystal structure of [(iPrPCP)NiMe] (one of two
independent molecules). Selected bond distances (Å): Ni1−C1,
1.954(2); Ni1−C21, 1.991(2); Ni1−P1, 2.1418(15); Ni1−P2,
2.1445(15). Angle between mean planes (deg): ring C1−C6 and
NiP2C2, 9.1.
strong hydrogen bridges (oxygen−oxygen distances 2.763(3)
and 2.842(3) Å).
Both the nickel bromide and the palladium chloride react
smoothly with MeI at 60 °C, affording the iodides [iPr(PCP)M-
I]15a,16d in high yield. This appears to be a very general reaction
for pincer complexes,19 since other iPrPCP derivatives of Ni or
Pd we tried do react in a similar fashion with MeI. This reaction
provides a simple method for recycling mixtures of these
compounds, since the iodide complexes can easily be recovered
by column chromatography on silica.
The synthesis of the palladium triflate [(iPrPCP)PdOTf]
from the chloride complex and silver triflate has been described
previously in the literature.15b We found that the readily
available Ni and Pd methyl derivatives react with triflic acid,
affording the corresponding triflates in good yields (eq 1). This
Figure 2. Crystal structure of [(iPrPCP)NiPh] (one of two
independent molecules). Selected bond distances (Å): Ni1−C1,
1.940(2); Ni1−C21, 1.925(2); Ni1−P1, 2.1470(6); Ni1−P2,
2.1484(6). Angles between mean planes (deg): Ph and NiP2C2,
79.3; ring C1−C6 and NiP2C2, 11.1; Ph and ring C1−C6, 88.1.
is a versatile route to complexes containing weakly coordinating
anions that avoids the use of expensive and potentially reactive
silver salts.
of the two crystallographically independent molecules found in
the unit cells of these complexes. The methyl complex has a
typical square-planar configuration without remarkable features.
Figure 2 shows that the rings of the phenyl and the pincer
ligands are approximately perpendicular. Although π-electron
back-donation from the Ni center to each of the aromatic
systems is expected to be different, due to their different
orientations with regard to the metal coordination plane, the
Ni−C bond distances are very similar (Ni1−C1, 1.940(2);
Ni1−C21, 1.925(2) Å).
The methyl and phenyl complexes are thermally stable, and
they resist exposure to air for some time in the solid state.
However, during one of the attempts to crystallize the nickel
methyl complex we isolated colorless crystals which we
subsequently identified as the diphosphine oxide 1-Me-2,6-
(iPr2P(O)CH2)2C6H3·H2O (Figure 3). Interestingly, a methyl
group saturates the ipso position originally bound to nickel,
suggesting that the reaction with oxygen induces the reductive
coupling of the methyl and the pincer ligand. Although the
molecule has no crystallographically imposed symmetry, it
adopts a symmetric arrangement with both phosphine oxide
functionalities associated to a single water molecule through
Nickel hydride complexes containing RPCP ligands (R = iPr,
tBu, cyclohexyl) have been synthesized by the reaction of the
corresponding halide precursors and different hydride
sources.17 In our hands, the reaction of [(iPrPCP)Ni-Br] with
LiBHEt3 (Super-Hydride) was clean and essentially complete,
but the purification of the hydride from the boron byproducts
was problematic, leading to erratic isolated yields. Alternatively,
this compound can be prepared by reacting [(iPrPCP)Ni-OMe]
with triethoxysilane. As shown in Scheme 2, the Ni methoxide
is prepared by the previously reported method,5b from the
amide [(iPrPCP)Ni-NH2] and methanol. It is not necessary to
isolate the amido and methoxide intermediates, because all
three steps of this synthesis can be performed in a sequential
manner. Since most reagents and byproducts (methanol,
HSi(OEt)3, ammonia, and Si(OMe)(OEt)3) are volatile, this
method minimizes side reactions and simplifies product
isolation. The palladium hydride [(iPrPCP)Pd-H] has also
been prepared in the same way, in spite of the thermal
instability of the corresponding amido complex [(iPrPCP)Pd-
NH2].5c The identity of the latter intermediate has been
1427
dx.doi.org/10.1021/om2009793 | Organometallics 2012, 31, 1425−1438

Organometallics
Article
Scheme 2
confirmed spectroscopically, and full NMR data (1H, 13C, and
31P) are supplied in the Experimental Section.
Scheme 3
The crystal structure of the nickel hydride is shown in Figure
4. The crystals of this compound contain three crystallo-
reactions are driven by the insolubility of AgI.20 Hence,
[(iPrPCP)Pd−F] was satisfactorily obtained from the corre-
sponding palladium iodide [(iPrPCP)Pd-I] and AgF. We found
that an excess of ≥2 equiv of AgF was necessary to complete
the formation of the nickel and palladium fluorides (Scheme 3).
Although this could be attributed to the transformation of AgF
into a less reactive mixed halide compound, analysis of the salt
residue by powder X-ray diffraction gave no indication of
inorganic phases different from AgF and AgBr or AgI.
Monitoring the halide exchange reactions by 31P{1H} NMR
shows that, after 12 h of sonication, the initial resonances of the
nickel bromide or palladium iodide precursors fully disappear
and are replaced by broader signals of the products. Analytically
pure samples of the yellow (M = Ni) or colorless (M = Pd)
fluoride complexes [(iPrPCP)M-F] were obtained after
recrystallization from hexane. The 31P{1H} signals of the pure
materials appear split into doublets by coupling with 19F. These
are significantly narrower and appear a few ppm below those
observed for the crude reaction mixtures. The differences of
line shape and chemical shifts of the 31P spectra of the reaction
mixtures could be due to the presence of small amounts of
water, which is known to promote rapid intermolecular fluoride
exchange.21 However, since efforts were made to maintain
rigorously anhydrous conditions, it cannot be ruled out that the
presence of small amounts of AgF could induce effects similar
to those of water in the NMR spectra of the crude solutions.
The reaction of hydroxo complexes with TREAT-HF has
proven to be an useful and versatile route for the synthesis of
fluoride derivatives.20 In addition, we showed that the
methylnickel complex [NiMe2(dippe)] also reacts readily with
TREAT-HF, affording the methyl fluoride complex [NiMeF-
(dippe)].9 Since methyl derivatives [(iPrPCP)M-Me] are robust
and readily available, we decided to investigate their reaction
with TREAT-HF. This reaction proceeds rapidly for M = Ni,
affording [(iPrPCP)Ni-F] in high yield. As with the AgF
method, the 31P{1H} spectrum of the crude product is
Figure 4. ORTEP view of [(iPrPCP)Ni-H] (one of three independent
molecules in the crystal). Selected bond distances (Å): Ni−H,
1.597(15); Ni1−C1, 1.946(13); Ni1−P1, 2.112(3); Ni1−P2,
2.121(3). Angle between mean planes (deg): ring C1−C6 and
NiP2C2, 14.1.
graphically independent molecules and are isomorphous with
those of the nickel fluoride derivative. Both complexes share an
unusual crystal-packing model, which is described in detail
below for the fluoride complex. Although the quality of the X-
ray data is not excellent, the hydride ligand was located ca. 1.60
Å from the Ni center. The lengths of the N−C bonds in the
three independent molecules are slightly different (1.943, 1.911,
and 1.893 Å). The average length of the Ni−C bond found in
the unit cell is 1.916 Å, somewhat shorter than in the methyl
and phenyl complexes (1.94−1.95 Å) and very similar to those
in the hydroxo and amido derivatives (1.917 and 1.929 Å), in
spite of the widely different strengths of the trans influence of
such ligands. This suggests that these M−C bonds can be
significantly biased by the rigidity of the pincer scaffold and are
not good indicators of the trans influence exerted by the
anionic ligand.
Synthesis of Ni and Pd Fluoride Complexes. In order to
synthesize these compounds, we explored two general methods
commonly used for the preparation of organometallic
transition-metal fluoride complexes: halide exchange with AgF
and protonation of suitable precursors with Et3N·3HF
(TREAT-HF) (Scheme 3).1a We investigated the reaction of
the halide precursors [(iPrPCP)Ni-Br] and [(iPrPCP)Pd-Cl]
with an excess of AgF in toluene, assisted with ultrasound.
Under these conditions, the Ni precursor is transformed into
the corresponding fluoride complex [(iPrPCP)Ni-F] in good
yield, but the Pd chloro complex only reacts very sluggishly.
Grushin observed that palladium iodide complexes are better
precursors for this purpose, probably because such exchange
1428
dx.doi.org/10.1021/om2009793 | Organometallics 2012, 31, 1425−1438

Organometallics
Article
somewhat broad, but in this case the line broadening persists
after recrystallization. This suggests the presence of small
amounts of HF, probably retained in the form of fluxional
bifluoride species that cannot be easily removed by
recrystallization.21 However, we found that THF solutions of
the crude material are effectively cleaned by stirring with CaH2
for 24 h, affording [iPr(PCP)Ni-F] with well-resolved spectra.
Presumably, CaH2 reacts with the HF but not with the nickel
fluoride complex, forming highly insoluble CaF2.
At room temperature, the reaction of [(iPrPCP)Pd-Me] with
an equivalent amount of TREAT-HF is slow and incomplete.
However, using an excess of the acidic reagent to force the
reaction leads to ill-defined products that could not be purified
by recrystallization. Treating this mixture with CaH2 was
unsuccessful, because it leads to the formation of [(iPrPCP)Pd-
H], identified by its characteristic low-field 31P signal at 71.7
ppm. Monitoring the reaction of TREAT-HF with [(iPrPCP)-
Pd-Me] by 31P{1H} shows that full neutralization requires
gentle warming at 50 °C for 1−2 h. Clean samples of
[(iPrPCP)Pd-F] are obtained when a somewhat less than the
stoichiometric amount of TREAT-HF is reacted with
[(iPrPCP)Pd-Me] at 50 °C, since all the HF contained in the
acidic reagent is consumed under these conditions. Therefore,
careful adjustment of the TREAT-HF and the reaction
temperature is essential for the successful preparation of
[(iPrPCP)Pd-F]. This can be routinely done by a two-stage
addition method, as described in the Experimental Section.
The room-temperature 31P{1H} spectra of pure fluoride
complexes [(iPrPCP)M-F] in C6D6 show signals at 54.3 ppm
(M = Ni) and 57.5 ppm (M = Pd), split into doublets by
coupling to 19F. The coupling constant 2JPF is larger for M = Ni
(36.7 Hz) than for M = Pd (5.9 Hz). The trans 19F−13C
coupling is transmitted more efficiently in the case of the
palladium complex, which gives a broad doublet at 155.2 ppm
in its 13C{1H} spectrum for C(ipso) with 2JCF = 48.8 Hz, while
the analogous signal of the Ni complex appears unresolved at
152.1 ppm. This signal splits into a doublet of triplets with 2JCF
Figure 5. Crystal structure of [(iPrPCP)Ni-F] (one of three
independent molecules). Selected bond distances (Å): Ni1−F1,
1.844(2); Ni1−C1, 1.901(4); Ni1−P1, 2.1606(13); Ni1−P2,
2.1608(12).
nar contacts involving the fluoride or hydride and the C−H
bonds of neighboring molecules. Although in the case of the
fluoride complex it would be tempting to attribute these
contacts to nonclassical hydrogen bridges, this appears less
likely for the hydride, since the Ni−H bond is much less polar
than the Ni−F bond. No signs of other intermolecular
interactions, e.g., π stacking, seem to stabilize the columnar
arrangement. Thus, it seems likely that the peculiar crystal
packing of these compounds stems from the shape of their
molecules, rather than from the existence of strong
intermolecular forces.
Assessment of the σ- and π-Bonding Strength of
Anionic Ligands L− on the Basis of the 13C NMR Spectra
of [(iPrPCP)M-L] Complexes. The special reactivity of late-
transition-metal complexes containing strongly π-basic ligands
such as fluoride, alkoxo, and amido has been explained in light
of different theories.1,22 One of the most appealing ideas is the
existence of nonbonding dπ−pπ interactions between the filled
d orbitals in the metal and the ligand lone pairs.12 These would
destabilize the metal−ligand bond, increasing the reactivity of
the complexes. Unfortunately, quantification of such effects has
proven to be a difficult task. Precise measurements of the
relative metal−alkoxide and metal−amido bond strengths in
Ni(II), Pt(II), and Ru(II) complexes have been interpreted in
terms of electrostatic and covalent contributions to essentially σ
bonds, leading to the conclusion that dπ−pπ interactions are
weak or nonexistent.23 However, analysis of spectroscopic data
for late-transition-metal complexes containing π-donor ligands
has led to different conclusions. A study by Caulton and
Eisenstein on the Ru(II) carbonyl complexes [RuHL(CO)-
(PR3)2] showed that the frequency of the ν(CO) band depends
on the relative π-donor strength of the anionic ligands L−
present in the molecule, decreasing in the order I < Br < Cl < F
< OEt.14 Since the metal d orbitals of π symmetry are all
occupied, the observed decrease of ν(CO) was assumed to be
induced by a push−pull effect: the metal d orbitals become
destabilized by the nonbonding M−L pπ−dπ interactions,
which increases the π-back-bonding capacity of the Ru(II)
center to the carbonyl. Theoretical calculations supported this
interpretation and reproduced the experimental trend of the
ν(CO) frequency. This suggested that ν(CO) provides an
indication of the intensity of the filled−filled π interactions.
Accordingly, fluoride and alkoxide are among the stronger π
donor ligands toward the Ru center. Grushin has applied similar
ideas to the analysis of the bonding in a series of palladium
complexes of the type trans-Pd(Ar)(X)(PPh3)2 (Ar = Ph, 4-
C6H4NO2; X = F, Cl, Br, I).24 In these compounds, the 13C
2
= 27.3 Hz and JCP = 16.0 Hz when the carbon spectrum is
recorded at −75 °C. The 19F signals, which appear at −295.0
and −245.3 ppm for the Ni and the Pd complexes, respectively,
are too broad to allow resolution of the 19F−31P splitting.
The nickel fluoride complex has been characterized by X-ray
diffraction. The unit cell of the crystal contains three
crystallographically independent molecules. These show no
significant differences in their general configuration or
intramolecular bond distances or angles. One of them is
shown in Figure 5. The Ni center is found in a square-planar
coordination environment, with nearly zero tetrahedral
distortion. Intramolecular bond distances and angles are
unexceptional. The Ni−F bond lengths are 1.844(2),
1.861(2), and 1.846(2) Å (1.850(2) Å in average), which are
comparable to those of other monomeric nickel fluoride
complexes. As mentioned previously, the nickel fluoride and
hydride complexes form isomorphous crystals characterized by
very similar cell parameters. The crystal packing of these
compounds is worth some comment. This is shown in Figure 6
for [(iPrPCP)Ni-F]. The molecules pile themselves, forming
columnar structures separated by narrow channels of about 1.6
Å wide. Each C1−Ni−L vector (L = H, F) is twisted ca. 120°
with respect to the next one, in such a way that the columns
resemble screw threads making one turn for each set of three
molecules. The Ni−L bonds are oriented pointing outward
from the columns. This arrangement allows short intercolum-
1429
dx.doi.org/10.1021/om2009793 | Organometallics 2012, 31, 1425−1438

Organometallics
Article
Figure 6. Crystal packing of [(iPrPCP)Ni-F]: (A) view along axis C, showing the columnar motif from the top; (B) detail of one of the columns
formed by three crystallographically independent molecules.
signal of the metal-bound carbon atom (C(ipso)) experiences a
significant shift of ca. 10 ppm, growing increasingly shielded on
going from iodide to fluoride. The trend is counterintuitive,
because the trans influence25 of the halide ligands, and
therefore their electron-donor capacity, increases in the
opposite order (as confirmed by the structural data gathered
for this particular set of complexes). As a rationale, it was
proposed that a π push−pull effect analogous to that suggested
by Caulton and Eisenstein actually increases the electron
density at the ipso carbon as the halogen varies from I to F,
shifting the 13C resonance to lower frequency. This is
schematically shown in Figure 7A.
structural data are of little use for this purpose, because the
influence of the anionic ligands on the rigid pincer fragment is
small. However, the structural invariance of the [(iPrPCP)M]
moiety is a positive factor for analyzing spectroscopic
properties, since any variations can be directly attributed to
electronic effects. Table 1 collects selected 13C NMR data for
[(iPrPCP)M-L] derivatives, analogous to the signals considered
by Grushin in the analysis of arylpalladium complexes. Most of
the 13C data come from spectra recorded in C6D6, but in some
cases the spectra were recorded in CD2Cl2 for solubility
reasons. Although solvent effects on chemical shifts are fairly
small, these have been minimized with the use of coordination
shifts, Δδ(13C): i.e., the differences between the chemical shifts
of each signal in the complex and those in the free ligand (in
the same solvent).
Table 1 shows that changes in the coordination shifts are
more pronounced for the ipso carbon than for the rest of the
13C resonances. Just as observed for the trans-[Pd(X)(Ar)-
(PPh3)2] complexes, this signal becomes increasingly de-
shielded on descending along the halide series, from F to I.
The magnitude of the variation is also very similar in the two
systems (10−13 ppm from F to I). The influence of the anionic
ligand follows the same trend along the full series of nickel and
palladium pincer complexes. This is confirmed by a plot of
Δδ(13C(ipso)) of nickel complexes versus the same parameter
in their palladium analogues (Figure 8A), which shows a good
linear correlation (R = 0.993) with a slope close to 1. The
Figure 7. Mechanisms explaining the influence of the metal fragment
on 13C NMR shifts of the aryl ring: (A) ligand-promoted π-electron
back-donation (push−pull effect); (B) paramagnetic contribution to
the chemical shift, due the to the mixing of the ground and excited
states.
The availability of a series of iPrPCP pincer complexes of Ni
and Pd provides an opportunity to extend the comparison to
various monoanionic ligands in addition to halides. X-ray
a
Table 1. Coordination Shifts (Δδ) for Selected 13C Signals of [(iPrPCP)M-L] Complexes Relative to Free PrPCHP Ligand
Δδ(13C), M = Ni
Δδ(13C), M = Pd
L
i-C
o-C
m-C
p-C
ref
i-C
o-C
m-C
p-C
ref
OTf
ONO2
F
16.1
21.8
23.5
26.1
29.1
29.6
31.8
32
12.5
12.1
13
−4.1
−4.3
−4.8
−4.5
−5.3
−4.6
−5.3
−4.3
−4.3
−5.2
−5.6
−5.7
−6.0
−4.5
−5.0
−6.1
−5.5
−6.7
−5.5
−7.0
−5.1
−5.0
−6.9
−5.8
−6.2
−6.1
b, this work
c, this work
b, this work
c, this work
b, ref 5b
20.9
21.8
26.6
27.9
30.6
31.8
31.7
34.0
37.7
39.2
47.6
50.7
51.1
10.8
10.8
10.9
10.9
10.6
10.5
10.0
10.5
10.3
9.4
−3.5
−3.3
−4.2
−4.0
−4.5
−3.8
−4.9
−3.8
−4.0
−5.0
−5.4
−5.6
−5.6
−4.9
−4.8
−6.1
−4.8
−6.3
−5.4
−6.8
−5.3
−5.2
−6.8
−6.0
−6.2
−5.8
b, this work
c, this work
b, this work
c, this work
b, ref 5c
OAc
OMe
Cl
12.6
11.7
12.5
11.5
12.3
12
b, ref 17
c, this work
c, this work
b, ref 5a
Br
c, this work
b, ref 5a
OH
I
36.3
39.3
46.8
49.1
49.5
c, this work
b, ref 5b
c, this work
b, this work
c, this work
c, this work
NH2
Ph
10.9
11.6
9.3
c, this work
c, this work
c, this work
10.3
10.7
10.0
Me
H
8.2
c, ref 5c
b
a
NMR signals (ppm) for the free ligand (C6D6, CD2Cl2): i-C, 128.5, 128.4; o-C, 140.4, 140.5; m-C, 127.0, 126.7; p-C: 130.8, 130.6. Spectrum
recorded in C6D6. Spectrum recorded in CD2Cl2.
c
1430
dx.doi.org/10.1021/om2009793 | Organometallics 2012, 31, 1425−1438

Organometallics
Article
Figure 8. Plots of (A) coordination shifts of the [(iPrPCP)M-L] complex C(ipso) signals for M = Pd vs M = Ni and (B) coordination shifts for
C(ipso) of hydride and halide nickel and palladium complexes vs electronegativity of the anionic ligand.
points in the plot, each one corresponding to a single ligand,
spread regularly along the straight line with Δδ(13C(ipso))
increasing in the order OTf < NO3 < F < OAc < OMe < Cl <
OH ≈ Br < I < NH2 < Ph < Me < H. This sequence matches
the qualitative σ-donor strengths of the ligands and suggests
that Δδ(13C(ipso)) has little relation to the π-donor capacities
of the anionic ligands. For example, extreme values of
Δδ(13C(ipso)) are observed for L = OTf, H, respectively the
weakest and strongest σ-donor ligands, but both of them have
little or no π-donor capacity. Good π-donors, such as F, NH2,
OH, and OMe are not grouped together but scattered along the
series, and the effect of Ph, which has some π-acceptor
character, is comparable to that of Me. Furthermore, Δδ(13C-
(ipso)) correlates with the electronegativity of atomic ligands,
i.e., halides and hydride (Figure 8B). This would not be
expected if the π-donor capacity of these ligands were playing a
substantial role, but it is reasonable if Δδ(13C(ipso)) is
controlled by the σ-donor capacity of the trans ligand. The
effect of anionic ligands on Δδ(13C(ipso)) can be related to the
familiar concept of trans influence, since this is also linked to σ-
donor capacity. The relationship between the electronegativity
of ligands and their trans influence has been noted before.25a
Apparently, the progressive deshielding of the 13C resonance
of the ipso carbon as the ligand L becomes a stronger σ-donor
contradicts basic ideas on the origin of the NMR chemical shift.
However, it must be recalled that, for nuclei heavier than H,
chemical shifts are determined not only by the diamagnetic
screening exerted by the surrounding electron density but also
by a paramagnetic contribution arising from the mixing of the
molecule ground and excited states containing unpaired
electrons. The intensity of this paramagnetic contribution
depends inversely on the energy required to promote the
electrons from the ground to the excited states. In part, the
typical deshielding of the signals originated by aromatic systems
(relative to saturated organic fragments) is due to the low
energy of the empty aromatic π* orbitals. When an electron is
promoted into them, orbital symmetry dictates that the effect of
the paramagnetic contribution will be opposite to that of the
diamagnetic screening, shifting the NMR signals toward higher
frequencies. The smaller the energy gap between occupied and
empty orbitals, the larger the chemical shifts of the aromatic
nuclei. This explains, for example, the very high frequency of
the ipso 13C carbon resonance in strongly polar aryllithium
derivatives (ca. 200 ppm).26 The polarized C−Li bond can be
assimilated to a high-energy electron pair sitting on a carbon
sp2 hybrid orbital; therefore, promotion of these electrons is
relatively easy. Pregosin used the same argument to explain the
unusually high frequency of the 13C resonances of aromatic
carbon atoms attached to Pd(II) centers.27 As shown
schematically in Figure 7B, when the ligand placed trans to
the carbon atom is a strong σ-donor (e. g., H or Me), it raises
the electron density at the σ Pd−C bond, shifting the 13C
toward higher frequency. In contrast, electronegative ligands
remove electron density from the trans position, stabilizing the
bonding pair. In the limit for very electronegative ligands, the
M−L bond is essentially ionic and Δδ(13C(ipso)) is not
influenced by the nature of L−. This is in good agreement with
the exponential form of the correlation between the Δδ(13C-
(ipso)) and ligand electronegativity (Figure 8B).
Since the behavior of the 13C resonance of the ipso carbon is
controlled by the σ-donor capacity of the anionic ligands, it
offers no indication of π effects such as pπ−dπ repulsions.
However, other spectroscopic parameters could provide some
indications regarding such an effect. Classic NMR studies on
arylpalladium and arylplatinum complexes have shown that
chemical shifts of the nuclei at the para positions of the aryl
group (either 13C for ring carbon atoms or 19F for attached
fluorine substituents) can be interpreted in terms of inductive
(σ) and resonance (π) effects caused by the metal fragment.13
Since chemical shifts at the meta position are determined
essentially by inductive effects, these are expected to cancel out
in the differences δpara − δmeta, which were then used to gauge
the π interaction of the metal fragment with the aromatic ring.
These differences show that Pt(II) or Pd(II) centers act as π
donors toward the aryl ligand, although this interaction is rather
weak compared to that of organic groups with strong π-
electron-donor properties such as NR2 and OR. The 13C
chemical shifts of the meta and para carbons in the
[(iPrPCP)M-L] system are not very different from those of
the free ligand; hence, the corresponding coordination shifts
Δδ(13C) are small. In spite of this, the differences Δδ(13C-
(para)) − Δδ(13C(meta)) show some interesting trends (Table
2). It is worth noting that the stronger π-donor ligands, F, NH2,
OH, and OMe, are characterized by negative and relatively
large values of this parameter, while the parameter is almost
zero for H or Me, suggesting that π effects are small for the
latter (note that for the free ligand Δδ(13C(para)) −
Δδ(13C(meta)) is zero by definition). Although these figures
have to be taken with due caution, the trend is not likely to be
entirely fortuitous and supports the ability of strongly π-
donating ligands such as fluoride to destabilize the filled metal d
orbitals increasing π back-donation to the aromatic ring, as
previously proposed by other authors. The NMR data suggest
that the π-donor strength of fluoride is comparable in
1431
dx.doi.org/10.1021/om2009793 | Organometallics 2012, 31, 1425−1438

Organometallics
Article
methoxides [(iPrPCP)M-OMe] as the only products in solution,
from which these products can be isolated in good yield. This
reaction can be useful for in situ generation of reactive alkoxides
that may be difficult to isolate otherwise.
Table 2. Difference between Coordination Shifts of the Meta
and Ipso Carbons (Δδ(13C(para)) − Δδ(13C(meta))) for
Different Complexes
ligand
Ni
Pd
As previously mentioned, methyl iodide reacts with
[(iPrPCP)M-L] complexes (M = Ni, Pd), cleanly affording
the corresponding iodide complexes. In order to facilitate the
identification of the organic products, we investigated the
reactions of the fluoride complexes with dodecyl iodide at 100
°C in C6D6. For comparison, the reactions of dodecyl iodide
with nickel bromide and palladium chloride complexes were
studied as well (Scheme 4). NMR, GC, and GC-MS analyses of
the reaction mixtures showed that the exchange reactions
proceed cleanly in all cases, affording equimolar amounts of the
corresponding iodide complexes and alkyl halides (fluoride,
chloride, and bromide). The course of the reactions was
monitored by 31P{1H} NMR. In the presence of a 10/1 excess
of dodecyl iodide, these exhibit pseudo-first-order kinetics. The
apparent rate constants, expressed as half-lives, are shown in
Scheme 4. Fluoride complexes are significantly more reactive
than the bromide or chloride derivatives. The difference is
particularly striking in the case of palladium: while 50%
conversion of the chloro derivative takes nearly 1 day, the
fluoride is fully transformed within 40 min. The nickel fluoride
complex is less reactive than its palladium analogue but reacts
significantly more quickly than the corresponding bromide. It is
worth noting that the reaction rate of the nickel fluoride with
dodecyl iodide increases significantly when carried out in THF
instead of C6D6.
From a mechanistic point of view, these halide exchange
reactions are probably more intricate than their apparent
simplicity may suggest. In the case of palladium complexes, the
mechanism could involve unstable Pd(IV) intermediates
formed by oxidative addition of the alkyl iodide, which would
undergo reductive elimination to afford the Pd(II) iodide
complex and alkyl fluoride. The recently observed decom-
position of palladium(IV) fluoride complexes by C−F reductive
coupling adds support to this proposal.4a,d Although such a
mechanism is less likely in the case of the nickel complexes, an
oxidative process involving paramagnetic Ni(III) intermediates
cannot be ruled out in principle.28,29 According to either of
these mechanisms, the reactivity of complexes [(iPrPCP)M-L]
would be higher when L is a strong σ donor, because this
facilitates the oxidation of the metal center. This conclusion is
not supported by the high reactivity of fluoride complexes,
since fluoride is a very weak σ-donor ligand. The experimental
trend of reactivity would be better understood if the mechanism
is assumed to be ligand- rather than metal-centered. In this
view, the highly polarized Mδ+−Fδ− retains part of the strong
nucleophilicity of the naked fluoride anion. However, this does
not explain why the palladium fluoride is much more reactive
than its nickel analogue, since the polarities of the M−F bonds
probably are not very different. The dπ−pπ repulsions theory
could be useful here, because these interactions are probably
more intense for Pd, due to the larger radial extension of the
metal d orbitals in second-row transition elements.30
OH
NH2
OMe
F
−1.7
−1.7
−1.4
−1.3
−0.9
−0.8
−0.7
−0.7
−0.4
−1
−1.9
−1.8
−1.8
−1.9
−1.6
−1.5
−1.5
−1.2
−1.4
−0.8
−0.6
−0.6
−0.2
Cl
Br
ONO2
I
OTf
Oac
Me
Ph
−0.5
−0.2
−0.1
H
magnitude with those of the OH and NH2 groups, and at the
same time it behaves as a weak σ donor. As previously
suggested in the literature,1b,7a the combination of these two
factors tends to destabilize the metal−ligand bond, rendering
the fluoride complexes of Ni and Pd especially reactive in
comparison with other halide complexes.
Reactivity of Ni and Pd Pincer Fluoride Complexes.
Due to the high reactivity associated with the M−F bond in
late-transition-metal complexes, nickel and palladium pincer
fluoro complexes can be considered as sources of either the
cationic [(iPrPCP)M]+ fragment or the anionic fluoride anion.
As such, they may find useful applications in synthesis, both of
new pincer organometallic complexes and of organic fluorides.
We selected two simple reactions of the [(iPrPCP)M-F]
complexes for illustrating both aspects of their reactivity:
displacement of fluoride by alkoxide at the metal center and the
reaction with an alkyl iodide leading to halide exchange at the
carbon center (eq 2 and Scheme 4).
Scheme 4
Independently of what might be the chemical behavior of the
[(iPrPCP)M-F] complexes, their high reactivity could be
exploited in catalytic reactions. Thus, we decided to investigate
if these compounds could catalyze fluoride exchange between
alkyl halides and AgF, as shown in Scheme 5. Although AgF
does react directly with alkyl iodides, the synthetic interest of
this reaction is limited, because it is not selective and alkyl
The reaction of [(iPrPCP)M-F] (M = Ni, Pd) with lithium
methoxide in THF (eq 2) is straightforward. The 31P{1H}
spectra of the reaction mixtures indicate that these reactions are
clean and quantitative. The resulting light precipitate of LiF is
readily removed by centrifugation, leaving the corresponding
1432
dx.doi.org/10.1021/om2009793 | Organometallics 2012, 31, 1425−1438

Organometallics
Article
latter. This difference is not unexpected, since [(iPrPCP)Pd-F]
is much more reactive toward dodecyl iodide than its nickel
analogue. When the palladium fluoride complex is used as a
catalyst, formation of dodecanol and other minor byproducts is
almost suppressed and the only significant byproduct is 1-
dodecene. The best results are obtained with [(iPrPCP)Pd-I]:
dodecyl iodide is fully converted within 2.3 h, producing
dodecyl fluoride and a small amount of 1-dodecene (10%) as
the only detected byproduct. Although there is no obvious
reason for the better performance of the iodide compared to
the fluoride, this could be due to the fact that [(iPrPCP)Pd-F]
easily captures traces of water or other proton-donor impurities
that may have some effect in the reaction.
GC monitoring of the catalyzed reactions confirmed that
nickel complexes slightly accelerate the rate of formation of
dodecyl fluoride and inhibit the formation of n-dodecanol. In
contrast, palladium complexes not only improve the selectivity
of the reaction but also catalyze actively the formation of
dodecyl fluoride. This suggests that the effect of the pincer
complexes on the course of the halide exchange reaction is 2-
fold. First, they inhibit the formation of products arising by a
carbocationic pathway, probably because they are much better
halide donors toward Ag+ than dodecyl iodide. Second, they
catalyze C−F bond formation via reaction of the fluoride
complex with the alkyl halide. The fact that the rate of dodecyl
fluoride formation increases only slightly in the presence of Ni
complexes suggests that in this case most of this product is still
formed by direct reaction with AgF. This is in line with the slow
reaction rate of the reaction of [(iPrPCP)Ni-F] with dodecyl
iodide, which requires 6.5 h to produce 0.5 equiv of the alkyl
fluoride. The metal-catalyzed fluoride transfer reaction is
significantly accelerated only in the case of the palladium
complexes, due to the higher reactivity of [(iPrPCP)Pd-F].
The 31P{1H} spectra of the reaction mixtures of catalysis
experiments with Ni or Pd complexes are usually simple,
displaying one or two signals in the 54−60 ppm region. These
signals show evident line broadening and are reminiscent of
those of impure samples of the fluoride complexes. They
probably correspond to rapidly exchanging mixtures of fluoride-
containing species, such as bifluorides. This observation
indicates that the fluoride framework [(iPrPCP)M] survives
the catalytic process, which may allow catalyst recycling or
achieving higher turnover numbers if the experimental
conditions are optimized.
Scheme 5
fluorides are formed together with a number of byproducts.31,32
Notwithstanding, the reaction has found some application in
the synthesis of alkyl fluorides.33 Very recently, Doyle et al.
reported a Pd-catalyzed method for the stereoselective
fluorination of allyl chlorides, involving an outer-sphere fluoride
from AgF to a (π-allyl)palladium intermediate.34
The most important byproducts of the reactions of AgF with
alkyl iodides are alcohols, but significant amounts of olefins and
isomerization products are formed as well. Hence, a mechanism
involving a carbocationic intermediate has been suggested for
these reactions.32 In agreement with this, the reaction of AgF
with dodecyl iodide in toluene at 100 °C yields a complex
mixture containing roughly similar amounts of 1-dodecanol, 1-
dodecene, and 1-fluorododecane and minor amounts of
isomers of these substances, which were identified by GC-
MS. The conversion reaches 75% in 3.3 h. Figure 9 compares
Figure 9. Conversion and selectivity of the reaction of dodecyl iodide
with AgF in the presence of 10 mol % of [(iPrPCP)M-X] (X = F, Br, I).
Experimental conditions: toluene, 100 °C, 200 min, except for Pd-F
(135 min).
the effect of catalytic amounts of nickel and palladium halo
complexes [(iPrPCP)M-L] (L = F, Br, I) on the latter reaction.
The catalyst dose was adjusted to 10 mol % (with regard to R-
I) to match the conditions used in stoichiometric halogen
exchange experiments. Since the nickel and palladium fluoride,
bromide, and chloride complexes are converted by dodecyl
iodide into the corresponding iodides, the choice of the catalyst
precursor has only a minor influence on the catalytic process.
The halide complexes [(iPrPCP)M-X] do influence the rate
of the reaction of AgF with dodecyl iodide, but while this is
appreciably accelerated by the palladium complexes, the overall
rate of consumption of dodecyl iodide decreases in the presence
of the nickel derivatives. The selectivity of the reaction
improves significantly in either case: formation of byproducts
is inhibited, and the yield of dodecyl fluoride increases.
Actually, the net rate of production of alkyl fluoride is higher
in the presence of both nickel and palladium complexes,
although the acceleration is much more pronounced with the
CONCLUSIONS
■
A series of nickel and palladium complexes containing the
iPrPCP pincer ligand and different anions, including triflate,
nitrate, alkyl, phenyl, hydride, and halide, were prepared and
characterized. Together with the previously reported hydroxo,
methoxo, and amido derivatives, the new complexes constitute
one of the widest known series of isostructural complexes with
simple monoanionic anions. This facilitated the comparison of
the electronic properties of fluoride with those of many other
anionic ligands. Significant trends were observed in the 13C
NMR data corresponding to the aromatic ring of the pincer
ligand. The position of the metal-bound carbon signal
(C(ipso)) is very sensitive to the nature of the anionic ligand
in the complex, shifting over a spectral interval of ca. 35 ppm.
The magnitudes of the coordination shift are nearly identical
for the nickel and palladium complexes and are related to the
qualitative σ-donor capacity of the anionic ligands. This is
supported by a good correlation between the coordination
1433
dx.doi.org/10.1021/om2009793 | Organometallics 2012, 31, 1425−1438

Organometallics
Article
Solvent 1H and 13C{1H} resonances were used as the internal
standard, but the chemical shifts are reported with respect to TMS,
and 31P spectra are referenced to external PPh3 in C6D6 (δ −6.0 ppm).
Abbreviations for multiplicities are as follows: br, broad; s, singlet, d,
doublet; t, triplet; m, mutiplet; v, virtual; dvt, doublet of virtual triplets.
IR spectra were recorded in Nujol mulls on Bruker Vector 22 and
Tensor 27 spectrophotometers. GC and GC-MS analyses were carried
out in Agilent 6890 and Thermoquest Trace GC chromatographs
equipped with an FID detector and an AutomassMulti mass
spectrometer, respectively, using capillary polymethylsiloxane columns.
Microanalyses were performed by the Microanalytical Service of the
shifts of C(ipso) and the electronegativity of the monatomic
ligands (hydride and halides). This relationship can be used to
gauge the σ-donor capacity of the anionic ligands toward the Ni
and Pd centers, which increases in the order OTf < NO3 < F <
OAc < OMe < Cl < OH ≈ Br < I < NH2 < Ph < Me < H.
Analysis of the chemical shifts of the para and meta resonances
revealed a completely different pattern that could be related to
the π-donor capacity of the ligands. The differences between
the chemical shifts of the para and meta carbons are more
negative in complexes containing strong π-donor ligands (OH,
F, NH2, and OMe) than in those without this capacity (H, Me).
This is consistent with an enhancement of π back-donation
from the metal to the aryl ring caused by nonbonding pπ−dπ
interactions. These results are largely in agreement with
longstanding ideas on the electronic properties of simple
anionic ligands (e.g., fluoride is a weak σ-donor ligand but a
stronger π donor than the rest of the halides), but the new data
offer the opportunity to directly compare very different ligands
that are seldom found within the same system.
Two methods have been applied for the synthesis of the
fluoride derivatives [(iPrPCP)M-F]: halide exchange with AgF
and protonation of the methyl derivatives [(iPrPCP)M-Me]
with TREAT-HF. Fluoride complexes [(iPrPCP)M-F] exhibit
an interesting reactivity pattern. The fluoride ligand is readily
displaced from nickel or palladium by lithium alkoxides, cleanly
leading to the corresponding pincer alkoxo derivatives. The
fluoride complexes also react with dodecyl iodide, affording
iodide complexes and alkyl fluoride. Although analogous halide
exchange reactions also take place between dodecyl iodide and
chloride or bromide pincer complexes, these react much more
slowly than the fluorides, indicating that the latter are more
nucleophilic. Furthermore, the reaction of the fluorides with
dodecyl iodide is at least 20 times faster for the palladium than
for the nickel derivative. It is difficult to justify this difference
on the basis of the degree of polarization of the M−F bonds,
since this is probably similar in the Ni and Pd complexes.
However, the higher nucleophilicity of the palladium fluoride
can be qualitatively explained by the greater degree of
destabilization of the fluoride lone pairs by the filled metal d
orbitals, which are more spatially extended in second-row than
in first-row transition elements.
́
Instituto de Investigaciones Quımicas.
Synthesis of Complexes. [(iPrPCP)Ni-Br]. A solution of iPrPCHP
in THF (0.5 M, 10 mL, 5 mmol) was added to a suspension of
NiBr2(dme) (1.54 g, 5 mmol; dme = dimethoxyethane) in THF (60
mL) at −78 °C; the resulting mixture was stirred and allowed to reach
room temperature. Et3N (5.5 mL, 50 mmol) was added and the
reaction mixture heated to 125 °C for 3 h. The solvent was evaporated
under reduced pressure and the residue extracted with Et2O (3 × 40
mL). The combined extracts were concentrated and cooled to −20 °C,
1
affording [(iPrPCP)Ni-Br] as a yellow solid in 90% yield. H NMR
(CD2Cl2, 25 °C, 300 MHz): δ 1.17 (dvt, 12H, 3JHH ≈ J*HP ≈ 6.9 Hz,
CH3), 1.44 (dvt, 12H, 3JHH ≈ J*HP ≈ 7.7 Hz, CH3), 2.36 (m, 4H, CH),
3.09 (vt, 4H, J*HP = 4.0 Hz, CH2), 6.90 (m, 3H, 2 × CarHm, CarHp).
13C{1H} NMR (CD2Cl2, 25 °C, 300 MHz): δ 18.3 (s, CH3), 19.0 (s,
CH3), 24.2 (vt, J*CP = 11.2 Hz, CH), 33.0 (vt, J*CP = 12.9 Hz, CH2),
122.4 (vt, J*CP = 8.9 Hz, CarHm), 125.5 (s, CarHp), 152.8 (vt, J*CP
=
13.2 Hz, Car‑o), 159.5 (t, 2JCP = 15.9 Hz, Car‑i). 31P{1H} NMR (CD2Cl2,
25 °C, 300 MHz): δ 60.5. Anal. Calcd for C22H35BrNiP2: C, 50.46; H,
7.41. Found: C, 50.35; H, 7.32.
[(iPrPCP)Pd-Br]. A 12 mL portion of a 0.085 M solution of NaCp in
THF (1.02 mmol) was added to a solution of the palladium chloro
allyl dimer [Pd(η3-C3H5)(μ-Cl)]2 (0.183 g, 0.5 mmol) in 40 mL of
THF stirred at −80 °C. The resulting red mixture was stirred for 1 h at
room temperature and then cooled again to −80 °C. A solution of
iPrPCBrP (0.5 M in THF, 2 mL, 1 mmol) was added. The mixture was
warmed to room temperature and stirred for 12 h. The solvent was
evaporated under vacuum, and the resulting solid was extracted with
diethyl ether (3 × 20 mL). The solution was taken to dryness again
and the residue purified by chromatography on a silica gel column,
using diethyl ether/hexane 5/1 as eluent. The product was isolated as
a pale yellow solid in 40% yield. 1H NMR (CD2Cl2, 25 °C, 300 MHz):
δ 1.14 (dvt, 12H, 3JHH ≈ J*HP ≈ 7.0 Hz, CH3), 1.37 (dvt, 12H, 3JHH
≈
J*HP ≈ 7.0 Hz, CH3), 2.34 (m, 4H, CH), 3.19 (vt, 4H, J*HP = 4.0 Hz,
CH2), 6.91 (t, 1H, 3JHH = 7.4 Hz, CarHp), 7.01 (d, 2H, 3JHH = 7.6 Hz, 2
× CarHm). 13C{1H} NMR (CD2Cl2, 25 °C, 300 MHz): δ 18.1 (s,
The nucleophilic reactivity of the fluoride pincer complexes
was exploited in a catalytic halide exchange reaction between
AgF and RI to afford alkyl fluorides. Both nickel and palladium
catalysts improve the selectivity of the reaction of AgF with
dodecyl iodide, probably by avoiding the formation of reactive
carbocationic intermediates. However, only the palladium
complexes lead to a significant acceleration of the reaction,
presumably due to the much greater nucleophilic reactivity of
the fluoride complex.
CH3), 18.9 (s, CH3), 24.4 (vt, J*CP = 11.2 Hz, CH), 33.4 (vt, J*CP
=
11.5 Hz, CH2), 122.9 (vt, J*CP = 10.8 Hz, CarHm), 125.2 (s, CarHp),
151.0 (vt, J*CP = 10.7 Hz, Car‑o), 159.3 (s, Car‑i). 31P{1H} NMR
(CD2Cl2, 25 °C, 300 MHz): δ 61.4. Anal. Calcd for C22H35BrNiP2: C,
50.12; H, 7.36. Found: C, 50.42; H, 7.43.
[(iPrPCP)Ni-I]. To a solution of (iPrPCP)Ni-Br (0.095 g, 0.2 mmol) in
THF (5 mL) was added MeI (0.152 mL, 1 mmol), and the resulting
mixture was heated to 60 °C for 72 h. The solvent was removed under
reduced pressure and the residue extracted with hexane. The product
was obtained as a yellow solid after evaporating the solvent under
reduced pressure. Yield: 70%. 1H NMR (CD2Cl2, 25 °C, 300 MHz): δ
1.15 (dvt, 3JHH ≈ J*HP ≈ 6.9 Hz, 12H, CH3), 1.43 (dvt, 3JHH ≈ J*HP
≈
EXPERIMENTAL SECTION
■
7.7 Hz, 12H, CH3), 2.48 (m, 4H, CH), 3.14 (vt, J*HP = 4.1 Hz, 4H,
CH2), 6.95 (m, 3H, 2 × CarHm, CarHp). 13C{1H} NMR (CD2Cl2, 25
°C, 300 MHz): δ 18.4 (s, CH3), 19.3 (s, CH3), 25.2 (vt, J*CP = 11.8
Hz, CH), 34.1 (vt, J*CP = 13.0 Hz, CH2), 122.4 (vt, J*CP = 8.9 Hz,
CarHm), 125.6 (s, CarHp), 152.5 (vt, J*CP = 12.9 Hz, Car‑o), 163.8 (t,
2JCP = 15.9 Hz, Car‑i). 31P{1H} NMR (CD2Cl2, 25 °C, 300 MHz): δ
63.1. Anal. Calcd for C20H35INiP2: C, 45.93; H, 6.74. Found: C, 45.59;
H, 5.81.
All procedures and chemical manipulations were carried out under Ar
or N2 using Schlenk or glovebox techniques. Solvents were rigorously
dried and degassed before use. Commercially available reagents were
used as received. The pincer ligand precursor 2,6-bis-
(diisopropylphosphino)benzene35
(
iPrPCHP) and the complex
[(iPrPCP)Pd-Cl]15b were prepared according to literature methods.
The brominated precursor 1-bromo-2,6-bis(diisopropylphosphino)-
benzene36 (iPrPCBrP) was prepared analogously to iPrPCHP from 1-
bromo-α,α′-dibromoxylene. NMR spectra were recorded on Bruker
DRX 300, 400, and 500 MHz spectrometers. Assignment of signals
was assisted by combined one-dimensional (gated 13C) and two-
dimensional techniques (HSQC). Chemical shifts (δ) are in ppm.
[(iPrPCP)Pd-I]. MeI (0.152 mL, 1 mmol) was added to a solution of
[(iPrPCP)Pd-Cl] (0.095 g, 0.2 mmol) in THF (5 mL), and the
resulting mixture was heated to 60 °C for 24 h. The solvent was
removed under reduced pressure, the residue was extracted with
1434
dx.doi.org/10.1021/om2009793 | Organometallics 2012, 31, 1425−1438

Organometallics
Article
3
hexane, and the solution was taken to dryness, affording [(iPrPCP)Pd-
J*HP ≈ 6.6 Hz, 12H, CH3), 1.25 (dvt, JHH ≈ J*HP ≈ 7.5 Hz, 12H,
CH3), 2.24 (m, 4H, CH), 3.20 (vt, J*HP = 4.0 Hz, 4H, CH2), 6.82 (t,
1
I] as a colorless solid. Yield: 80%. H NMR (CD2Cl2, 25 °C, 300
MHz): δ 1.10 (dvt, 3JHH ≈ J*HP ≈ 7.2 Hz, 12H, CH3), 1.37 (dvt, 3JHH
≈ J*HP ≈ 8.0 Hz, 12H, CH3), 2.44 (m, 4H, CH), 3.26 (vt, J*HP = 4.3
Hz, 4H, CH2), 6.97 (t, 3JHH = 7.4 Hz, 1H, CarHp), 7.07 (d, 3JHH = 7.6
Hz, 2H, CarHm). 13C{1H} NMR (CD2Cl2, 25 °C, 75 MHz): δ 18.1 (s,
3
3JHH = 7.3 Hz, 1H, CarHp), 6.95 (d, JHH = 7.5 Hz, 2H, CarHm).
13C{1H} NMR (CD2Cl2, 25 °C, 75 MHz): δ −17.8 (t, 2JCP = 21.7 Hz,
NiCH3), 18.4 (s, CH3), 19.0 (vt, J*CP = 2.3 Hz, CH3), 23.8 (vt, J*CP
=
10.3 Hz, CH), 37.0 (vt, J*CP = 13.1 Hz, CH2), 121.0 (vt, J*CP = 8.5 Hz,
CarHm), 124.4 (s, CarHp), 151.2 (vt, J*CP = 13.8 Hz, Car‑o) 176.6 (t, 2JCP
= 13.5 Hz, Car‑i). 31P{1H} NMR (CD2Cl2, 25 °C, 121 MHz): δ 62.4.
Anal. Calcd for C21H38P2Ni: C, 61.34; H, 9.32. Found: C, 31.34; H,
9.32. The complex [(iPrPCP)Pd-Me] was obtained as a white solid
CH3), 19.3 (s, CH3), 25.3 (vt, J*CP = 11.9 Hz, CH), 34.9 (vt, J*CP
=
11.6 Hz, CH2), 122.7 (vt, J*CP = 10.7 Hz, CarHm), 125.4 (s, CarHp),
150.8 (vt, J*CP = 10.6 Hz, Car‑o), 165.2 (s, Car‑i). 31P{1H} NMR
(CD2Cl2, 25 °C, 121 MHz): δ 62.4. Anal. Calcd for C20H35IP2Pd: C,
42.09; H, 6.18. Found: C, 42.27; H, 5.81.
1
after evaporating the solvent under reduced pressure. Yield: 62%. H
[(iPrPCP)Ni-ONO2]. A solution of [(iPrPCP)Ni-Br] (2.05 g, 4.3
mmol) in THF (50 mL) was added to a flask charged with AgNO3
(0.750 g, 4.4 mmol). The reaction mixture was stirred at room
temperature for 36 h. The solvent was evaporated under vacuum and
the residue extracted with Et2O (3 × 50 mL). The complex
[(iPrPCP)Ni-ONO2] was obtained as yellow crystals after evaporating
the solvent under reduced pressure and crystallizing from hexane at
−30 °C. 1H NMR (CD2Cl2, 25 °C, 300 MHz): δ 0.91 (dvt, 12H, 3JHH
≈ J*HP ≈ 7.0 Hz, CH3), 1.25 (dvt, 12H, 3JHH ≈ J*HP ≈ 7.7 Hz, CH3),
1.79 (m, 4H, CH), 2.58 (vt, 4H, J*HP = 4.1 Hz, CH2), 6.73 (d, 2H,
NMR (CD2Cl2, 25 °C, 300 MHz): δ −0.34 (t, 3JHP = 5.2 Hz, PdCH3),
1.09 (dvt, 3JHH ≈ J*HP ≈ 6.9 Hz, 12H, CH3), 1.23 (dvt, 3JHH ≈ J*HP
≈
7.8 Hz, 12H, CH3), 2.25 (m, 4H, CH), 3.35 (vt, J*HP = 4.1 Hz, 4H,
CH2), 6.86 (t, 3JHH = 7.4 Hz, 1H, CarHp), 7.05 (d, 3JHH = 7.4 Hz, 2H,
CarHm). 13C{1H} NMR (CD2Cl2, 25 °C, 75 MHz): δ −18.6 (t, 2JCP
=
10.1 Hz, PdCH3), 18.2 (s, CH3), 19.1 (vt, J*CP = 2.8 Hz, CH3), 24.5
(vt, J*CP = 10.8 Hz, CH), 38.7 (vt, J*CP = 11.9 Hz, CH2), 121.1 (vt,
J*CP = 9.6 Hz, CarHm), 124.1 (s, CarHp), 149.8 (vt, J*CP = 10.8 Hz,
Car‑o), 178.2 (t, JCP = 4.1 Hz, Car‑i). 31P{1H} NMR (CD2Cl2, 25 °C,
2
121 MHz): δ 59.2. Anal. Calcd for C21H38P2Pd: C, 54.96; H, 8.35.
Found: C, 54.90; H, 8.30.
3
3JHH =7.4 Hz, CarHm), 6.95 (t, 1H, JHH = 7.4 Hz, CarHp). 13C{1H}
[(iPrPCP)Ni-Ph and [(iPrPCP)Pd-Ph]. A solution of PhLi in hexane
(2.0 M, 0.15 mL, 0.3 mmol) was added to a solution of [(iPrPCP)Pd-
Cl] (0.149 g, 0.31 mmol) or [(iPrPCP)Ni-Br] (0.148 g, 0.31 mmol) in
Et2O (25 mL) at −80 °C. The cooling bath was removed, and the
mixture was stirred until it reached room temperature. The solvent was
evaporated under reduced pressure, and the remaining residue was
extracted with Et2O (3 × 10 mL). The extract was evaporated under
vacuum, and the residue was taken up in hexane (20 mL). The
products crystallized when the hexane solutions were concentrated to
NMR (CD2Cl2, 25 °C, 300 MHz): δ 17.6 (s, CH3), 18.2 (s, CH3), 23.5
(vt, J*CP = 10.1 Hz, CH), 30.4 (vt, J*CP = 13.6 Hz, CH2), 122.4 (vt,
J*CP = 8.6 Hz, CarHm), 125.6 (s, CarHp), 149.3 (t, 2JCP = 16.6 Hz, Car‑i),
152.6 (vt, J*CP = 12.5 Hz, Car‑o) . 31P{1H} NMR (CD2Cl2, 25 °C, 300
MHz): δ 58.6 ppm. Anal. Calcd for C20H35NO3P2Ni: C, 52.43; H,
7.70. N, 3.06. Found: C, 52.51; H, 7.89. N, 2.97. IR (Nujol mull):
ν(NO2) 1456, ν(NO2) 1282, ν(NO) 1026 cm−1.
[(iPrPCP)Ni-OAc] and [(iPrPCP)Pd-OAc]. A solution of [(iPrPCP)Ni-
Br] (0.238 g, 0.5 mmol) or [(iPrPCP)Pd-Cl] (0.240 g, 0.5 mmol) in
THF (25 mL) was transferred to a flask charged with AgOAc (0.083 g,
0.5 mmol), and the resulting mixture was stirred at room temperature
for 12 h. The solvent was evaporated under reduced pressure and the
residue extracted with Et2O (3 × 25 mL). The complex [(iPrPCP)Ni-
OAc] was obtained as a yellow crystalline solid after addition of some
hexane and cooling to −30 °C. Yield: 75%. 1H NMR (CD2Cl2, 25 °C,
1
1 or 2 mL and cooled to −30 °C. [(iPrPCP)Ni-Ph]: yellow solid. H
NMR (CD2Cl2, 25 °C, 300 MHz): δ 0.86 (dvt, 3JHH ≈ J*HP ≈ 7.4 Hz,
12H, CH3), 1.09 (dvt, 3JHH ≈ J*HP ≈ 6.9 Hz, 12H, CH3), 2.04 (m, 4H,
3
CH), 3.18 (vt, J*HP = 4.0 Hz, 4H, CH2), 6.61 (t, JHH = 7.2 Hz, 1H,
CPhHp), 6.80 (m, 3H, CarHp, and 2 × CPhHm), 6.91 (d, 3JHH = 7.4 Hz,
3
1
2H, CarHm), 7.42 (dd, 2H, JHH = 7.8 Hz, JHH = 1.3 Hz, CPhHo).
13C{1H} NMR (CD2Cl2, 25 °C, 75 MHz): δ 17.7 (s, CH3), 18.5 (s,
CH3), 23.6 (vt, J*CP = 11.0 Hz, CH), 36.9 (vt, J*CP = 13.0 Hz, CH2),
120.5 (s, CPhHp), 121.1 (vt, J*CP = 8.7 Hz, CarHm), 124.8 (s, CarHp),
3
300 MHz): δ 1.24 (dvt, 12H, JHH ≈ J*HP ≈ 6.9 Hz, CH3), 1.36 (dvt,
3
12H, JHH ≈ J*HP ≈ 8.6 Hz, CH3), 1.73 (s, 3H, OCOCH3), 2.18 (m,
4H, CH), 3.02 (vt, 4H, J*HP = 4.1 Hz, CH2), 6.79 (m, 3H, 2 × CarHm,
CarHp). 13C{1H} NMR (CD2Cl2, 25 °C, 75 MHz): δ 18.1 (s, CH3),
18.7 (s, CH3), 24.1 (vt, J*CP = 10.0 Hz, CH), 24.3 (s, COOCH3), 31.4
(vt, J*CP = 13.4 Hz, CH2), 122.2 (vt, J*CP = 8.6 Hz, CarHm), 125.1 (s,
CarHp), 153.1 (vt, J*CP = 13.1 Hz, Car‑o), 153.6 (t, 2JCP = 17.3 Hz, Car‑i),
175.9 (s, OCOCH3). 31P{1H} NMR (CD2Cl2, 25 °C, 121 MHz): δ
56.6 ppm. Anal. Calcd for C22H38NiO2P2: C, 58.05; H, 8.41. Found: C,
58.09; H, 8.37. The complex [(iPrPCP)Pd-OAc] was obtained as a
colorless microcrystalline solid after evaporating the solvent under
reduced pressure. Yield: 75%. 1H NMR (CD2Cl2, 25 °C, 300 MHz): δ
1.19 (dvt, 3JHH ≈ J*HP ≈ 7.3 Hz, 12H, CH3), 1.28 (dvt, 3JHH ≈ 7.9 Hz,
J*HP ≈ 7.9 Hz, 12H, CH3), 1.85 (s, 3H, OCOCH3) 2.29 (m, 4H, CH),
3.15 (vt, J*HP = 4.3 Hz, 4H, CH2), 6.88 (t, 3JHH = 8.6 Hz, 1H, CarHp),
125.5 (s, CPhHm), 140.1 (vt, J*CP = 2.5 Hz, CPhHo), 152.1 (vt, J*CP
=
10.8 Hz, Car‑o), 164.2 (t, 2JCP = 26.1 Hz, Car‑i), 174.3 (t, 2JCP = 13.2 Hz,
Car‑i). 31P{1H} NMR (CD2Cl2, 25 °C, 121 MHz): δ 56.3. Anal. Calcd
for C26H40P2Ni: C, 65.99; H, 8.52. Found: C, 65.95; H, 8.63.
1
[(iPrPCP)Pd-Ph]: white solid, 62% yield. H NMR (CD2Cl2, 25 °C,
3
300 MHz): δ 1.02 (dvt, JHH ≈ J*HP ≈ 7.7 Hz, 12H, CH3), 1.13 (dvt,
3JHH ≈ J*HP ≈ 7.2 Hz, 12H, CH3), 2.13 (m, 4H, CH), 3.41 (vt, J*HP
=
3
4.2 Hz, 4H, CH2), 6.77 (t, JHH = 7.3 Hz, 1H, CPhHp), 6.94 (m, 1H,
3
3
CarHp), 6.93 (t, JHH = 7.4 Hz, 2H, CPhHm), 7.08 (dd, 2H, JHH = 7.4
3
1
Hz, CarHm), 7.51 (dd, 2H, JHH = 7.6 Hz, JHH = 1.3 Hz, CPhHo).
13C{1H} NMR (CD2Cl2, 25 °C, 75 MHz): δ 17.9 (s, CH3), 18.6 (s,
CH3), 24.4 (vt, J*CP = 11.3 Hz, CH), 38.2 (vt, J*CP = 12.1 Hz, CH2),
121.0 (s, CPhHp), 121.3 (vt, J*CP = 9.3 Hz, CarHm), 124.6 (s, CarHp),
3
6.96 (d, JHH = 7.5 Hz, 2H, CarHm). 13C{1H} NMR (CD2Cl2, 25 °C,
75 MHz): δ 18.0 (s, CH3), 18.8 (s, CH3), 24.2 (br s, OCOCH3), 24.7
(vt, J*CP = 10.9 Hz, CH), 32.4 (vt, J*CP = 11.7 Hz, CH2), 122.7 (vt,
J*CP = 10.6 Hz, CarHm), 125.8 (s, CarHp), 151.4 (vt, J*CP = 10.6 Hz,
Car‑o), 155.4 (s, Car‑i), 175.7 (s, OCOCH3). 31P{1H} NMR (CD2Cl2,
121 MHz): δ 59.5. IR (Nujol mull): ν(CO) 1580 cm−1. Anal. Calcd
for C22H38O2P2Pd: C, 52.54; H, 7.62. Found: C, 52.50; H, 7.48.
[(iPrPCP)Ni-Me] and [(iPrPCP)Pd-Me]. A solution of MeLi in Et2O
(1.6 M, 0.29 mL, 0.47 mmol) was added to a solution of [(iPrPCP)Ni-
Br] (0.148 g, 0.31 mmol) in Et2O (25 mL) or [(iPrPCP)Pd-Cl] (0.149
g, 0.31 mmol) at −50 °C. The resulting mixture was stirred for 2 h at
room temperature. Methanol (0.29 mL) was added and the reaction
mixture stirred for an additional 15 min. The mixture was evaporated
under reduced pressure and the residue extracted with Et2O (3 × 15
mL). The complex [(iPrPCP)Ni-Me] was obtained as an orange solid
after evaporating the solvent under reduced pressure and was
126.1 (s, CPhHm), 140.6 (vt, J*CP = 2.5 Hz, CPhHo), 150.8 (vt, J*CP
=
10.8 Hz, Car‑o), 164.4 (t, 2JCP = 12.1 Hz, CPh‑i), 175.1 (t, 2JCP = 4.1 Hz,
Car‑i). 31P{1H} NMR (CD2Cl2, 25 °C, 121 MHz): δ 55.1. Anal. Calcd
for C26H40P2Pd: C, 59.94; H, 7.74. Found: C, 59.99; H, 7.77.
[(iPrPCP)Ni-OTf]. A solution of trifluoromethanesulfonic (triflic) acid
(0.7 M, 0.3 mL, 0.21 mmol) was added to a solution of [(iPrPCP)Ni-
Me] (0.082 g, 0.2 mmol) in Et2O (15 mL) at −78 °C. The product
precipitated as a red solid when the mixture was warmed to room
temperature. The solution was filtered off, and the solid was washed
with Et2O (10 mL), dried under vacuum, and recrystallized from
1
dichloromethane/toluene. Yield: 50%. H NMR (C6D6, 25 °C, 300
MHz): δ 0.81 (dvt, 12H, 3JHH ≈ J*HP ≈ 6.9 Hz, CH3), 1.45 (dvt, 12H,
3JHH ≈ J*HP ≈ 7.9 Hz, CH3), 2.16 (m, 4H, CH), 2.52 (vt, 4H, J*HP
=
=
3
3
3.7 Hz, CH2), 6.66 (d, 2H, JHH =7.4 Hz, CarHm), 6.91 (t, 1H, JHH
1
recrystallized from hexane. Yield: 73%. H NMR (CD2Cl2, 25 °C,
7.4 Hz, CarHp). 13C{1H} NMR (C6D6, 25 °C, 75 MHz): δ 18.2 (s,
3
3
300 MHz): δ −0.77 (t, JHP = 8.0 Hz, 3H, NiCH3), 1.11 (dvt, JHH
≈
CH3), 19.2 (s, CH3), 23.8 (vt, J*CP = 10.4 Hz, CH), 29.5 (vt, J*CP =
1435
dx.doi.org/10.1021/om2009793 | Organometallics 2012, 31, 1425−1438

Organometallics
Article
13.9 Hz, CH2), 120.2 (q, J*CF = 318.4 Hz, CF3), 122.9 (vt, J*CP = 8.5
Hz, CarHm), 126.3 (s, CarHp), 144.7 (t, 2JCP = 16.2 Hz, Car‑i), 152.8 (vt,
J*CP = 12.5 Hz, Car‑o). 31P{1H} NMR (C6D6, 25 °C, 121 MHz): δ 60.1
ppm. IR (Nujol mull): ν(SO) 1320, 1203, 1167 cm−1. Anal. Calcd for
C21H35F3NiOP2S: C, 46.21; H, 6.45; S, 5.88. Found: C, 46.21; H, 6.47;
S, 5.98.
IR (C6D6 solution): νmax 1735 cm−1 (ν(Ni−H)). Anal. Calcd for
C20H36P2Ni: C, 60.49; H, 9.14. Found: C, 60.21; H, 9.01.
[(iPrPCP)Pd−H]: colorless solid, 90% yield. 1H NMR (C6D6, 300
MHz, 25 °C): δ −3.77 (t, 1H, 2JHP = 17.0 Hz, Pd-H), 0.91 (dvt, 3JHH
=
7.2 Hz, J*HP = 7.2 Hz, 12H, CH3), 1.17 (dvt, 3JHH = 7.8 Hz, J*HP = 7.8
Hz, 12H, CH3), 1.88 (m, 4H, CH), 3.24 (vt, J*HP = 4.1 Hz, 4H, CH2),
3
3
[(iPrPCP)Pd-OTf]. A solution of trifluoromethanesulfonic (triflic)
acid (0.7 M, 0.3 mL, 0.21 mmol) was added to a solution of
[(iPrPCP)Pd-Me] (0.092 g, 0.2 mmol) in Et2O (15 mL) at −78 °C,
and the reaction mixture was warmed to room temperature. The
solvent was evaporated under reduced pressure, the residue was
extracted with CH2Cl2, and the solution was filtered and taken to
dryness. The white residue was recrystallized from Et2O at −30 °C to
7.23 (t, JHH = 7.4 Hz, 1H, CarH), 7.28 (d, JHH = 7.4 Hz, 2H, CarH).
13C{1H} NMR (C6D6, 75 MHz, 25 °C): δ 18.7 (CH3), 20.1 (vt, J*CP
=
3.4 Hz, CH3), 25.3 (vt, J*CP = 11.6 Hz, CH), 39.8 (tv, J*CP = 11.7 Hz,
CH2), 121.4 (tv, J*CP = 9.6 Hz, CarHm), 125.0 (CarHp), 150.3 (vt, J*CP
= 10.7 Hz, Car‑o), 179.7 (t, 2JCP = 3.8 Hz, Car‑i). 31P{1H} NMR (C6D6,
121 MHz): δ 71.7. IR (Nujol mull): 1787 and 1763 cm−1 (ν(Pd−H).
IR (C6D6 solution): νmax 1721 cm−1 (ν(Pd−H)). Anal. Calcd for
C20H36P2Pd: C, 54.00; H, 8.16. Found: C, 53.49; H, 8.24.
1
afford the product as colorless crystals. Yield: 40%. H NMR (C6D6,
25 °C, 400 MHz): δ 0.76 (dvt, 12H, 3JHH ≈ J*HP ≈ 7.2 Hz, CH3), 1.33
[(iPrPCP)Ni-F]. (a). From [(iPrPCP)Ni-Br]. A solution of the nickel
bromide complex (0.381 g, 0.8 mmol) in THF (20 mL) was added to
a suspension of AgF (0.203 g, 1.6 mmol) in toluene (10 mL), and the
resulting suspension was sonicated for 12 h. The solvent was removed
under reduced pressure and the residue extracted with hexane (3 × 25
mL). The combined extracts were centrifuged, concentrated, and
cooled to −30 °C, affording the product as yellow crystals in 80%
yield.
3
(dvt, 12H, JHH ≈ J*HP ≈ 7.7 Hz, CH3), 2.15 (m, 4H, CH), 2.58 (vt,
4H, J*HP = 4.7 Hz, CH2), 6.79 (d, 2H, 3JHH = 7.5 Hz, CarHm), 6.94 (t,
1H, 3JHH = 7.5 Hz, CarHp). 13C{1H} NMR (C6D6, 25 °C, 100 MHz): δ
18.0 (s, CH3), 19.1 (s, CH3), 24.8 (vt, J*CP = 11.0 Hz, CH), 30.8 (vt,
J*CP = 11.9 Hz, CH2), 121.2 (c, J*CF = 318.7 Hz, CF3), 123.5 (vt, J*CP
2
= 10.5 Hz, CarHm), 125.9 (s, CarHp), 149.5 (t, JCP = 11.2 Hz, Car‑i),
151.1 (vt, J*CP = 10.3 Hz, Car‑o). 31P{1H} NMR (C6D6, 25 °C, 161
MHz): δ 63.2 ppm. Anal. Calcd for C21H35F3OP2PdS: C, 42.54; H,
5.95; S, 5.41. Found: C, 42.50; H, 5.78; S, 5.14. IR (Nujol mull):
ν(SO3) 1314, 1203, 1161 cm−1.
(b). From [(iPrPCP)Ni-Me]. A solution of TREAT-HF in THF (0.25
M, 6.0 mL, 1.5 mmol) was added to a solution of [(iPrPCP)Ni-Me]
(1.23 g, 3 mmol) in THF (80 mL) at −78 °C. The reaction mixture
was warmed to room temperature. The Et3N·3HF excess was
eliminated transferring the solution to a flask charged with CaH2
(2.0 g), and the resulting suspension was stirred for 24 h. The reaction
mixture was centrifuged and filtered to afford a yellow solution. The
solvent was evaporated under reduced pressure, the residue extracted
with hexane (3 × 50 mL), and the solution taken to dryness. Yield:
Synthesis of [(iPrPCP)Ni-H] and [(iPrPCP)Pd-H]. The synthesis of
these two compounds involves a protocol for the consecutive
generation of the amides [iPr(PCP)M-NH2]5b,c and methoxides
[iPr(PCP)M-OMe],5b,c which are then reacted with trimethoxysilane.
The intermediate complexes have been reported before,5 except for
the palladium amide derivative. Due to its thermal instability, this
compound was characterized by NMR only, and data are reported
below.
1
3
62%. H NMR (CD2Cl2, 25 °C, 300 MHz): δ 1.28 (dvt, 12H, JHH
≈
3
J*HP ≈ 7.0 Hz, CH3), 1.43 (dvt, 12H, JHH ≈ J*HP ≈ 7.7 Hz, CH3),
2.19 (m, 4H, CH), 2.99 (vt, 4H, J*HP = 5.1 Hz, CH2), 6.73 (d, 2H,
(a). Generation of Amide Complexes [iPr(PCP)M-NH2]. A solution
of [(iPrPCP)Pd-Cl] (1.59 g, 3.3 mmol) or [(iPrPCP)Ni-Br] (1.57 g, 3.3
mmol) in THF (60 mL) was transferred to a flask charged with
NaNH2 (1.29 g, 33.2 mmol), and the resulting suspension was
sonicated for 2 h. The initially colorless solution turned red. NaNH2 in
excess was eliminated by centrifugation. For NMR characterization,
the solution containing [(iPrPCP)Pd-NH2] was taken to dryness and
dissolved in C6D6. 1H NMR (C6D6, 25 °C, 400 MHz): δ −0.42 (s, 2H,
3JHH = 6.9 Hz, CarHm), 6.78 (t, 1H, JHH = 6.6 Hz, CarHp). 13C{1H}
3
NMR (CD2Cl2, 25 °C, 75 MHz): δ 18.2 (s, CH3), 18.9 (s, CH3), 23.3
(vt, J*CP = 9.7 Hz, CH), 30.7 (vt, J*CP = 12.9 Hz, CH2), 122.2 (vt, J*CP
2
= 8.7 Hz, CarHm), 124.7 (s, CarHp), 152.1 (bs, Car‑i), 153.3 (t, JCP
=
10.9 Hz, Car‑o). 31P{1H} NMR (CD2Cl2, 25 °C, 121 MHz): δ 54.3
ppm (d, 2JFP = 36.7 Hz). 19F{1H} NMR (CD2Cl2, 25 °C, 376 MHz): δ
−287.0 ppm. Anal. Calcd for C20H35FNiP2: C, 57.86; H, 8.50. Found:
C, 57.99; H, 8.48.
3
NH2), 0.93 (dvt, 12H, JHH ≈ J*HP ≈ 6.9 Hz, CH3), 1.27 (dvt, 12H,
3JHH ≈ J*HP ≈ 7.6 Hz, CH3), 1,89 (m, 4H, CH), 2.95 (vt, 4H, J*HP
=
[(iPrPCP)Pd-F]. (a). From [(iPrPCP)Pd-I] and AgF. A solution of the
palladium iodide complex (0.457 g, 0.8 mmol) in THF (20 mL) was
added to a flask charged with AgF (0.203 g, 1.6 mmol) and the
resulting suspension was sonicated for 12 h. The solvent was removed
under reduced pressure, and the residue was extracted with hexane (3
× 25 mL). The complex [(iPrPCP)Pd-F] was obtained as a colorless
solid after cooling to −30 °C.
3.8 Hz, CH2), 7.09 (m, 3H, 2 × CarHm, CarHp). 13C{1H} NMR (C6D6,
25 °C, 100 MHz): δ 18.1 (s, CH3), 19.1 (s, CH3), 24.1 (vt, J*CP = 10.6
Hz, CH), 35.9 (vt, J*CP = 12.0 Hz, CH2), 122.0 (vt, J*CP = 10.1 Hz,
CarHm), 124.0 (s, CarHp), 149.7 (vt, J*CP = 10.9 Hz, Car‑o), 164.4 (s,
Car‑i). 31P{1H} NMR (C6D6, 25 °C, 161 MHz): δ 58.8 ppm.
(b). Generation of Methoxide Complexes [iPr(PCP)M-OMe]. After
removal of NaH2, THF solutions of the amido complexes were treated
with 0.20 mL of methanol (5 mmol) at room temperaure. The
resulting solutions were taken to dryness and the residue was taken up
in 20 mL of THF or hexane and used for the synthesis of the hydrides
as follows.
(b). From [(iPrPCP)Pd-Me and TREAT-HF. For the success of this
method, the correct adjustment of the reagent ratio is essential. Since
the weighing of TREAT-HF may involve significant error, we used a
standard solution of this reagent in THF. The TREAT-HF content of
this solution was estimated by reacting a sample of this solution with
half the required amount of [(iPrPCP)Pd-Me] at 50 °C for 2 h and
then analyzing the ratio of product to starting material by 31P{1H}
NMR. Once the concentration of the TREAT-HF is accurately known,
the synthesis can be carried out as follows: a solution of TREAT-HF in
THF (0.1 M, 3.4 mL, 0.34 mmol) was added to a solution of
[(iPrPCP)Pd-CH3] (0,467 g, 1.02 mmol) in THF (10 mL) at room
temperature. The resulting mixture was heated to 50 °C for 2 h. The
solvent was evaporated under reduced pressure and the residue
extracted with hexane (3 × 25 mL). The complex [(iPrPCP)Pd-F] was
(c). Reaction of Methoxide Complexes with Triethoxysilane. A
solution of the appropriate methoxide complex in THF (Ni) or hexane
(Pd), prepared as indicated above, was cooled to 0 °C and treated with
an excess of triethoxysilane (0.9 mL, 5 mmol). The mixture was
allowed to warm at room temperature and the volatiles were removed
under vacuum. The nickel hydride [(iPrPCP)Ni-H] was obtained as
yellow needles after recrystallization from pentane at −30 °C. Yield,
2
60%. 1H NMR (C6D6, 25 °C, 400 MHz): δ −9.92 (t, JHP = 55.4 Hz,
3
1H, Ni-H), 0.96 (dvt, 12H, JHH ≈ J*HP ≈ 6.9 Hz, CH3), 1.20 (dvt,
3
1
12H, JHH ≈ J*HP ≈ 7.6 Hz, CH3), 1,92 (m, 4H, CH), 3.15 (vt, 4H,
obtained as a colorless solid after cooling to −30 °C. Yield: 42%. H
J*HP = 3.9 Hz, CH2), 7.24 (m, 3H, 2 × CarHm, CarHp). 13C{1H} NMR
(C6D6, 25 °C, 100 MHz): δ 18.4 (s, CH3), 19.5 (s, CH3), 24.6 (vt,
NMR (C6D6, 25 °C, 300 MHz): δ 1.00 (dvt, JHH ≈ J*HP ≈ 7.2 Hz,
3
12H, CH3), 1.34 (dvt, 3JHH ≈ J*HP ≈ 7.9 Hz, 12H, CH3), 1.96 (m, 4H,
3
J*CP = 11.9 Hz, CH), 37.5 (vt, J*CP = 12.3 Hz, CH2), 121.0 (vt, J*CP
=
CH), 2.75 (vt, J*HP = 4.0 Hz, 4H, CH2), 6.91 (d, JHH = 7.4 Hz, 2H,
10.1 Hz, CarHm), 124.7 (s, CarHp), 151.3 (vt, J*CP = 13.7 Hz, Car‑o),
CarHm), 7.00 (t, 3JHH = 7.4 Hz, 1H, CarHp). 13C{1H} NMR (C6D6, 25
°C, 100 MHz): δ 18.2 (s, CH3), 19.0 (s, CH3), 24.1 (vt, J*CP = 10.4
Hz, CH), 32.4 (vt, J*CP = 10.3 Hz, CH2), 122.8 (vt, J*CP = 10.3 Hz,
2
178.1 (t, JCP = 11.1 Hz, Car‑i). 31P{1H} NMR (C6D6, 25 °C, 161
MHz): δ 76.9 ppm. IR (Nujol mull) 1761 and 1743 cm−1 (ν(Ni−H)).
1436
dx.doi.org/10.1021/om2009793 | Organometallics 2012, 31, 1425−1438

Organometallics
Article
Table 3. Summary of Crystallographic Data and Structure Refinement Results for [(iPrPCP)Ni-F], [(iPrPCP)Ni-H],
[(iPrPCP)Ni-Me], [(iPrPCP)Ni-Ph], and [iPr(P(O)C(Me)P(O)]·H2O
[(iPrPCP)Ni-F]
[(iPrPCP)Ni-H]
[(iPrPCP)Ni-Me]
[(iPrPCP)Ni-Ph]
[
iPr(P(O)C(Me)P(O)]·H2O
chem formula
C20H35FNiP2
415.13
C20H36NiP2
397.14
C21H38NiP2
411.16
monoclinic
10.998(9)
13.794(9)
28.90(3)
90.00
C26H40NiP2
473.23
C21H38O2P2•H2O
402.47
formula mass
cryst syst
monoclinic
18.316(1)
19.001(2)
19.963(2)
90.00
monoclinic
18.038(4)
18.938(5)
20.036(5)
90.00
triclinic
monoclinic
13.5036(11)
12.3747(11)
14.2276(12)
90.00
a/Å
10.4777(10)
13.9169(14)
18.1873(18)
81.963(2)
76.658(1)
77.812(2)
2510.7(4)
100(2)
b/Å
c/Å
α/deg
β/deg
113.933(2)
90.00
113.393(4)
90.00
96.038(14)
90.00
107.045(2)
90.00
γ/deg
V/Å3
6350.2(10)
100(2)
P21/n
6282(3)
100(2)
P21/n
4360(6)
120(2)
P21/n
2273.0(3)
100(2)
temp/K
space group
P1
P21/c
̅
no. of formula units per unit cell, Z
radiation type
12
12
8
4
4
Mo Kα
1.076
Mo Kα
1.078
Cu Kα
2.648
Mo Kα
0.910
Mo Kα
0.208
−1
abs coeff, μ/mm
no. of rflns measd
no. of indep rflns
Rint
41 108
39 578
36 997
6775
30 622
12 620
0.0379
0.0383
0.0708
0.0560
0.0745
0.835
14 665
15 961
15 516
5757
0.0863
0.0937
0.0788
0.0476
0.1159
0.0586
0.1224
0.0546
a
2
final R1 values (F2 > 2σ(F ))
0.0679
0.0728
0.0470
b
final wR2(F2) values (F2 > 2σ(F2))
0.1428
0.1637
0.0687
a
final R1 values (all data)
0.1168
0.1359
0.0787
b
final wR2(F2) values (all data)
0.1647
0.1939
0.0743
c
goodness of fit on F2, S
1.003
0.993
0.997
0.980
a
b
c
2
2
2
R1(F) = ∑|Fo| − |Fc|)/∑|Fo|. wR2(F2) = {∑[w(Fo − Fc2)2]/∑(Fo )2}1/2. S = {∑[w(Fo − Fc2)2]/(n − p)}1/2 (n = number of reflections, p =
number of parameters).
CarHm), 124.6 (s, CarHp), 151.2 (vt, J*CP = 10.9 Hz, Car‑o), 155.2 (d,
2JCF = 48.2 Hz, Car‑i). 31P{1H} NMR (C6D6, 121 MHz): δ 57.5 (d, 2JFP
= 5.9 Hz). 19F{1H} NMR (C6D6, 25 °C, 376 MHz): δ −245.3 ppm.
Anal. Calcd for C20H35FP2Pd: C, 51.90; H, 7.62. Found: C, 51.52; H,
7.72.
Reactions of Fluoride Complexes with LiOMe. To a solution of
[(iPrPCP)Ni-F] (0.66 g, 1.6 mmol, see below) or [(iPrPCP)Pd-F]
(0.74, 1.6 mmol, see below) in THF (25 mL) was added a solution of
MeOLi (2.4 mmol) in THF (20 mL) at −78 °C. The resulting mixture
was warmed to room temperature and stirred for 20 min. According to
the 31P{1H} spectra of the reaction mixtures, the fluoride complexes
were cleanly and quantitatively converted in the corresponding
methoxides. Lithium fluoride was removed by centrifugation, the
solvent was evaporated under reduced pressure, and the residue was
extracted with hexane. The complexes [(iPrPCP)Pd-OMe]5c and
[(iPrPCP)Ni-OMe]5b were obtained after concentration and cooling to
−30 °C.
these new crystalline compounds is given in Table 3. Crystals coated
with dry perfluoro polyether were mounted on glass fibers and fixed in
a cold nitrogen stream (T = 100, 120 K). Intensity data were collected
on a Bruker SMART CCD diffractometer operating with graphite-
monochromated Mo Kα radiation (λ = 0.710 73 Å), except for
[(iPrPCP)Ni-Me], whose intensity data were collected on an Enraf-
Nonius FR590-Kappa CCD 2000 diffractometer equipped with a
graded multilayer mirror providing monochromated Cu Kα radiation
(λ = 1.541 78 Å). For the Bruker diffractometer the data were reduced
by SAINT,37 while for the Enraf-Nonius diffractometer the data were
reduced by HKL-Denzo and Scalepak.38 The data were in all cases
corrected for absorption effects by the multiscan method (SA-
DABS).36 The structures were solved by direct methods (SIR-2002,
SHELXS)39,40 and refined against all F2 data by full-matrix least-
squares techniques (SHELXTL-6.12)37 minimizing w[Fo − Fc2]2. All
2
non-hydrogen atoms were refined with anisotropic thermal parame-
ters. Hydrogen atoms were included in calculated positions and
allowed to ride on the attached carbon atoms with the isotropic
temperature factors (Uiso values) fixed at 1.2 times (1.5 times for the
methyl groups) those Ueq values of the corresponding carbon atoms.
Reaction of Halide Complexes with Dodecyl Iodide. Solutions
of the halide complexes [(iPrPCP)Ni-Br], [(iPrPCP)Ni-I], [(iPrPCP)Ni-
F], [(iPrPCP)Pd-I], and [(iPrPCP)Pd-Cl] (0.02 mmol) and n-
iododecane (50 μL, 0.2 mmol) in C6D6 (0.6 mL) containing a small
amount of p-cymene as internal standard for 1H integration were
transferred to NMR tubes capped with J. Young valves and placed in a
preheated oil bath at 100 °C. The course of the reactions was
ASSOCIATED CONTENT
■
1
monitored by H and 31P{1H} NMR. The reaction of [(iPrPCP)Pd−-
S
* Supporting Information
F] with dodecyl iodide was studied directly in the NMR probe at the
same temperature.
CIF files giving crystallographic data for all of the structure
studies in this paper. This material is available free of charge via
the Internet at http://pubs.acs.org.
Catalytic Fluorination of Dodecyl Iodide. To a suspension of
AgF (0.065 g, 0.52 mmol) in toluene (1 mL) was added a solution of
[(iPrPCP)Ni-Br], [(iPrPCP)Ni-I], [(iPrPCP)Ni-F], [(iPrPCP)Pd-I], or
[(iPrPCP)Pd-F] (0.035 mmol) in toluene (2 mL). The resulting
suspension was stirred for 3 h at 100 °C. Aliquots were taken out for
GC analysis. Yields and conversions were calculated by GC using p-
cymene as internal standard.
ACKNOWLEDGMENTS
■
Financial support from the DGI (Project CTQ2009-11721),
́
Junta de Andalucıa (Project FQM6276), and the European
X-ray Structure Analysis for [(iPrPCP)Ni-F], [(iPrPCP)Ni-H],
[(iPrPCP)Ni-Me], [(iPrPCP)Ni-Ph], and [iPr(P(O)C(Me)P(O)]·H2O. A
summary of crystallographic data and structure refinement results for
Union (FEDER funds) is gratefully acknowledged. L.M.M.-P. is
́
grateful for a predoctoral grant from the Junta de Andalucıa.
1437
dx.doi.org/10.1021/om2009793 | Organometallics 2012, 31, 1425−1438

Organometallics
Article
Acta 1995, 231, 195−200. (d) van der Boom, M. E.; Liou, S.-Y.;
Shimon, L. J. W.; Ben-David, Y.; Milstein, D. Inorg. Chim. Acta 2004,
357, 4015−4023.
REFERENCES
■
(1) (a) Vigalok, A. Organometallics 2011, 30, 4802−4810.
(b) Grushin, V. V. Acc. Chem. Res. 2010, 43, 160−171. (c) van der
Vulgt, J. I. Chem. Soc. Rev. 2010, 39, 2302−2322. (d) Gunnoe, B. T.
Eur. J. Inorg. Chem. 2007, 1185−1203. (e) Roesky, H. W.; Singh, S.;
Yusuff, K. K. M.; Maguire, J. A.; Hosmane., N. S. Chem. Rev. 2006, 106,
3813−3843. (f) Fulton, J. R.; Holland, A. W.; Fox, D. J.; Bergman, R.
G. Acc. Chem. Res. 2002, 35, 44−56. (g) Fryzuk, M. D.; Montgomery,
C. D. Coord. Chem. Rev. 1989, 95, 1−40. (h) Bryndza, H. E.; Tam, W.
Chem. Rev. 1988, 88, 1163−1188.
(2) (a) Lappert, M. F.; Power, P. P.; Protchenko, A. V.; Seeber, A. L.
Metal Amide Chemistry; Wiley: West Sussex, U.K., 2009. (b) Bradley,
D. C.; Mehrotra, R. C.; Rothwell, I. P.; Singh, A. Alkoxo and Aryloxo
Derivatives of Metals; Academic Press: Cornwall, U.K., 2001.
(3) (a) Baratta, W.; Ballico, M.; Del Zotto, A.; Herdtweck, E.;
Magnolia, S.; Peloso, R.; Siega, K.; Tonutti, M.; Zangrando, E.; Rigo,
P. Organometallics 2009, 28, 4421−4430. (b) Samec, J. S. M.; Backvall,
J. E.; Andersson, P. G.; Brandt, P. Chem. Soc. Rev. 2006, 35, 237−248.
(c) Sigman, M. S.; Jensen, D. R. Acc. Chem. Res. 2006, 39, 221.
Hartwig, J. F. Nature 2008, 455, 314−322. (d) Shannon, S. S. Angew.
Chem., Int. Ed. 2004, 43, 3400.
(17) Boro, B. J.; Dueslert, E. N.; Goldberg, E. I.; Kemp, R. A. Inorg.
Chem. 2009, 48, 5081−5087.
́
(18) del Rıo, D. Ph.D. Thesis, Universidad de Sevilla, 2003.
(19) Terheijden, J.; van Koten, G.; Vinke, I. C.; Spek., A. L.
Organometallics 1982, 1, 1366−1370.
(20) Pilon, M. C.; Grushin, V. V. Organometallics 1998, 17, 1774−
1781.
(21) Roe, D. C.; Marshall, W. J.; Davidson, F.; Soper, P. D.; Grusin,
V. V. Organometallics 2000, 18, 4575−4582.
(22) Hartwig, J. F. Organotransition Metal Chemistry. From Bonding to
Catalysis; University Science Books: Sausalito, CA, 2010; Chapter 4.
(23) (a) Holland, P. L.; Andersen, R. A.; Bergman, R. G.; Huang, J.;
Nolan, S. P. J. Am. Chem. Soc. 1997, 119, 12800−12814. (b) Holland,
P. L.; Andersen, R. A.; Bergman, R. G. Comments Inorg. Chem 1999,
21, 115−119.
(24) (a) Flemming, J. P.; Pilon, M. C.; Borbulevitch, O. Ya.; Antipin,
M. Yu.; Grushin, V. V. Inorg. Chim. Acta 1998, 280, 87−98.
(b) Marshall, W. J.; Thorn, D. L.; Grusin, V. V. Organometallics
1998, 17, 5427−5430.
(4) For leading references, see for example: (a) Furuya, T.; Benitez,
D.; Tkatchouk, E.; Strom, A. E.; Tang, P.; Goddard, W. A.; Ritter, T. J.
Am. Chem. Soc. 2010, 132, 3793−3807. (b) Gnanaprakasam, B.;
Zhang, J.; Milstein, D. Angew. Chem., Int. Ed. 2010, 49, 1468−1471.
(c) Hanley, P. S.; Markovi, D.; Hartwig, J. F. J. Am. Chem. Soc. 2010,
138, 6302−6303. (d) Ball, N. D.; Sanford, M. S. J. Am. Chem. Soc.
2009, 131, 3796−3797. (e) Neukom, J. D.; Perch, N. S.; Wolfe, J. P. J.
(25) (a) Hartley, B. R. Chem, Soc. Rev. 1973, 2, 145−161.
(b) Appleton, T. G.; Clark, H. C.; Manzer, L. E Coord. Chem. Rev.
1973, 10, 335−422.
(26) (a) Berger, S.; Fleischer, U.; Geletneky, C.; Lohrenz, J. C. W.
Chem. Ber 1995, 128, 1183−1186. (b) Reich, M. J.; Green, D. P.;
̈
Medina, M. A.; Goldenberg, W. S.; Gudmunsson, B. O.; Dykstra, R.
R.; Phillips, N. H.. J. Am. Chem. Soc. 1998, 120, 7201−7219.
(27) (a) Tschoerner, M.; Pregosin, P. S. Inorg. Chim. Acta 1999, 290,
95−99. (b) Tschoerner, M.; Kunz, P. W.; Pregosin, P. S. Magn. Reson.
́
Am. Chem. Soc. 2010, 132, 6276−6277. (f) Matas, I.; Campora, J.;
́
Palma, P.; Alvarez, E. Organometallics 2009, 28, 6515−6523.
(g) Smythe, N. A.; Grice, K. A.; Williams, B. S.; Goldberg, K. I.
Organometallics 2009, 28, 277−278. (h) Trend, R. M.; Stoltz., B. M. J.
Am. Chem. Soc. 2008, 130, 15957−15966. (i) Stambuli, J. P.; Weng, Z.;
Incarvito, C. D.; Hartwig, J. F. Angew. Chem., Int. Ed. 2007, 44, 7674−
7677. (j) Zhao, J.; Hesslink, H.; Hartwig, J. F. Organometallics 2001,
́
Chem. 1999, 37, 91−97. (c) Martınez-Viviente, E.; Pregosin, P. S.;
Tschoerner, M. Magn. Reson. Chem. 2000, 38, 23−28.
(28) Grove, D. M.; van Koten, G.; Mul, P.; Zoet, R.; Van der Linden,
J. G. M.; Legters, J.; Schmitz, J. E. J.; Murrall, N. W.; Welch, A. J. Inorg.
Chem. 1988, 27, 2466.
123, 7220−7227. (k) Hartwig, J. F.; Richards, S.; Baranano, D.; Paul,
̃
(29) Higgs, A. T.; Zinn, P. J.; Simmons, S. J.; Sanford, S. S.
Organometallics 2009, 28, 6142.
F. J. Am. Chem. Soc. 1996, 118, 3626−3633.
́
́
́
(5) (a) Campora, J.; Palma, P.; del Rıo, D.; Alvarez, E.
(30) Frenking, G.; Frohlich, N. Chem. Rev. 2000, 100, 717.
̈
Organometallics 2004, 23, 1652−1655. (b) Cam
́
pora, J.; Palma, P.;
(31) Manam, R. R.; McArthur, K. A.; Chao, T. H.; Weiss, J.; Lloyd,
G. K.; Palladino, M. A.; Neuteboom, S. T. C.; Macheria, V. R.; Potts,
B. C. M.; Ali, J. A.; Palombella, V. J.; Groll, M. J. Med. Chem. 2008, 18,
4227−4730.
́
́
del Rıo, D.; Conejo, M. M.; Alvarez, E. Organometallics 2004, 23,
́
́
5653−5655. (c) Melero, C.; Martınez-Prieto, L. M.; Palma, P.; del Rıo,
́
́
D.; Alvarez, E.; Campora, J. Chem. Commun. 2010, 46, 8851−8853.
(6) Clot, E.; Eisenstein, O.; Jasim, N.; Macgregor, S. A.; McGrady, J.
E.; Perutz., R. N. Acc. Chem. Res. 2011, 44, 333−348.
(32) San Filipo, J.; Romano, L. J. J. Org. Chem. 1974, 40, 782−787.
(33) See, for example: (a) Pretze, M.; Mamat, C.; Wuest, F.; Peppel,
T.; Koeckerling, M. Tetrahedron Lett. 2010, 51, 6410−6414. (b) Shilar,
P.; Radek, P.; Votruva, I.; Hocek, M. Org. Biomol. Chem. 2005, 3,
3001−3007. (c) Wei, Z.-L.; Xiao, Y.; Yuan, H.; Badyuk, M.; Petukhov,
P. A.; Musachio, J. L.; Kellar, K. J.; Kozikowski, A. P. J. Med. Chem.
2005, 48, 1721−1724.
(7) (a) Grushin, V. V. Chem. Eur. J. 2002, 8, 1006. (b) Watson, D. A.;
́
Su, M.; Teverovskiy, G.; Zhang, Y.; Garcıa-Fortanet, J.; Kinzel, T.;
Buchwald, S. L.. Science 2009, 325, 1661−1664.
(8) Grushin, V. V. Angew. Chem., Int. Ed. 1998, 37, 994−996.
́
(9) (a) Campora, J.; Matas, I.; Palma, P.; Graiff, C.; Tiripicchio, A.
́
Organometallics 2005, 24, 28217. (b) Campora, J.; Matas, I.; Palma, P.;
(34) (a) Katcher, M. H.; Doyle, A. G. J. Am. Chem. Soc. 2010, 132,
17402−17404. (b) Katcher, M.; Sha, A.; Doyle, A. G. J. Am. Chem. Soc.
2011, 133, 15902−15905.
́
Alvarez, E.; Graiff, C.; Tiripicchio, A. Organometallics 2007, 26, 5712.
(10) (a) Ball, N. D.; Kampf, J. W.; Sanford, M. S. Dalton Trans. 2009,
39, 632−640. (b) Huacuja, R.; Herbert, D. E.; Fafard, C. M.; Ozerov,
O. V. J. Fluorine Chem. 2010, 131, 1257.
(35) Montag, M.; Schwartsburd., L.; Cohen, R.; Leiturs, G.; Ben-
David, Y.; Martin, J. M. L.; Milstein, D. Angew. Chem., Int. Ed. 2007,
46, 1901−1904.
́
(11) Slander, N.; Szabo, K. Chem. Rev. 2011, 111, 2048−2076.
(12) (a) Caulton, K. G. New J. Chem. 1994, 18, 25−41. (b) Mayer, J.
M. Comments Inorg. Chem 1988, 8, 125−135.
(13) (a) Parshall, G. W. J. Am. Chem. Soc. 1966, 20, 704−708.
(b) Clark, H. C.; Ward, J. E. H. J. Am. Chem. Soc. 1974, 96, 1741−
1747. (c) Manna, J.; Kuehl, C. J.; Whiteford, J. A.; Stang, P. J.
Organometallics 1997, 16, 1897−1905.
(36) Rybtchinski, B.; Ben-David, Y.; Milstein, D. Organometallics
1997, 16, 3786−3793.
(37) Bruker programs: SMART, version 5.629; SAINT+, version 6.45;
SADABS, version 2.10; SHELXTL, version 6.12 Bruker AXS Inc.,
Madison, WI, 2003.
(38) HKL-Denzo and Scalepack: Otwinowski, Z.; Minor, W.
(14) Poulton, J. T.; Sigalas, M. P.; Folting, K.; Streib, W. E.;
Eisenstein, O.; Caulton, K.; G.. Inorg. Chem. 1994, 33, 1476−1485.
(15) (a) Ohff, M.; Ohff, A.; van der Boom, M. E.; Milstein, D. J. Am.
Chem. Soc. 1997, 119, 11687−11688. (b) Frech, C. M.; Shimon, L. J.
W.; Milstein, D. Angew. Chem., Int. Ed. 2005, 44, 1709−1711.
(16) (a) Moulton, C. J.; Shaw, B. L. J. Chem. Soc., Dalton Trans. 1976,
1020−1024. (b) Rimml, H.; Venanzi. L. M. J. Organomet. Chem. 1983,
259, C6-C7. (c) Kennedy, A. R.; Cross, R. J.; Muir., K. W. Inorg. Chim.
Methods Enzymol. 1997, 276, 307−326.
(39) Burla, M. C.; Camalli, M.; Carrozzini, B.; Cascarano, G. L.;
Giacovazzo, C.; Polidori, C.; Spagna, G. R. SIR2002: the program. J.
Appl. Crystallogr. 2003, 36, 1103.
(40) (a) Sheldrick, G. M. SHELX97: Program System for Crystal
Structure Determination; University of Gottingen, Gottingen, Germany,
1997.
̈
̈
1438
dx.doi.org/10.1021/om2009793 | Organometallics 2012, 31, 1425−1438