Plasticity of the Nickel(II) coordination environment in complexes with hemilabile phosphino thioether ligands

Article abstract of DOI:10.1021/ja109624m

A series of homoligated Ni(II) complexes formed from two phosphino thioether (P,S) chelating ligands has been synthesized and characterized. Interestingly, this included octahedral Ni(II) complexes which, unlike previously characterized d⁸ Rh(I), Pt(II), and Pd(II) analogues, exhibit in situ exchange processes centered around chloride ligand dissociation. This was verified and studied through the controlled abstraction from and introduction of chloride ions to this system, which showed that these processes proceed through complexes with square pyramidal, tetrahedral, and square planar geometries. These complexes were studied with a variety of characterization methods, including single-crystal X-ray diffraction studies, solution ³¹P{¹H} NMR spectroscopy, UV-vis spectroscopy, and DFT calculations. A general set of synthetic procedures that involve the use of coordinating and noncoordinating counteranions, as well as different hemilabile ligands, to mediate geometry transformations are presented.

Full text of DOI:10.1021/ja109624m

ARTICLE
pubs.acs.org/JACS
Plasticity of the Nickel(II) Coordination Environment in Complexes
with Hemilabile Phosphino Thioether Ligands
Charles W. Machan, Alexander M. Spokoyny, Matthew R. Jones, Amy A. Sarjeant, Charlotte L. Stern, and
Chad A. Mirkin*
Department of Chemistry and the International Institute for Nanotechnology, Northwestern University, 2145 Sheridan Road,
Evanston, Illinois 60208, United States
S Supporting Information
b
ABSTRACT: A series of homoligated Ni(II) complexes formed
from two phosphino thioether (P,S) chelating ligands has been
synthesized and characterized. Interestingly, this included octahe-
dral Ni(II) complexes which, unlike previously characterized d8
Rh(I), Pt(II), and Pd(II) analogues, exhibit in situ exchange
processes centered around chloride ligand dissociation. This was
verified and studied through the controlled abstraction from and
introduction of chloride ions to this system, which showed that
these processes proceed through complexes with square pyrami-
dal, tetrahedral, and square planar geometries. These complexes
were studied with a variety of characterization methods, including
single-crystal X-ray diffraction studies, solution 31P{1H} NMR spectroscopy, UV-vis spectroscopy, and DFT calculations. A general set of
synthetic procedures that involve the use of coordinating and noncoordinating counteranions, as well as different hemilabile ligands, to
mediate geometry transformations are presented.
Scheme 1. Scheme for the Formation and Reversible
Allosteric Control of the Conformations of Supramolecular
Macrocycles Using the WLA
’ INTRODUCTION
Several coordination-chemistry-based methods for preparing
supramolecular macrocycles, tweezers, and cage structures have
been developed.1-11 One of these is the weak-link approach
(WLA),12 which can be used to synthesize inorganic complexes
of various and complex geometries from readily available hemi-
labile ligands 1 and metal precursors (Scheme 1).13-20 Impor-
tantly, the species formed from the WLA can be interconverted
between rigid and flexible structures (2 and 3, respectively)
through small-molecule reactions at the metal sites, thus provid-
ing a powerful platform for constructing abiotic allosteric enzyme
mimics.21-23 The approach is extremely general and has been
applied to many different metal centers,24-27 which has substantially
increased the scope of its utility in catalysis and biomimetic systems.
The ability to form closed complexes with a variety of metal centers
(mostly d8) and appropriate hemilabile ligands has created a library
of chemical reactions that involve the use of small molecules or
elemental anions to control the in situ closing and opening of such
complexes, often into catalytically active forms.28-30
Interestingly, Ni(II) is conspicuously absent from the list
of metal centers that has been explored with respect to the
WLA. Since Ni(II) is known to have a wide range of stable
geometries,31-39 we decided to investigate the potential of
using it to form homoligated structures from phosphinoethyl
thioether (P,S) hemilabile ligands.40-44 In this paper, we focus
on model ligands to map out the possible reactions that could
be extended to larger and more complex ligands capable of
forming tweezers and macrocycles, similar to the ones studied
in the context of other transition metal centers.13-18,20,24
Herein, we report the discovery of new P,S Ni(II) bis-chelate
complexes, which have coordination modes that can be reversibly
switched between octahedral, square pyramidal, tetrahedral, and
square planar geometries via the controlled introduction and
subsequent abstraction of equivalents of halide ion(s), respec-
tively (Scheme 2). Analogous reactions have been explored with
Received: October 26, 2010
Published: February 14, 2011
r
2011 American Chemical Society
3023
dx.doi.org/10.1021/ja109624m J. Am. Chem. Soc. 2011, 133, 3023–3033
|

Journal of the American Chemical Society
ARTICLE
Table 1. Selected Bond Length and Bond Angle Data for the
Crystallographically Derived Structures of 5, 9a, and 10a
bond lengths (Å)
5
9a
10a
P1-Ni
P2-Ni
S1-Ni
S2-Ni
Cl1-Ni
Cl2-Ni
2.4719(3)
2.4719(3)
2.4520(3)
2.4520(3)
2.3674(2)
2.3674(2)
2.1848(12)
2.1832(12)
2.2207(12)
2.2149(12)
n/a
2.2172(4)
2.2172(4)
2.1830(4)
2.1830(4)
n/a
Figure 1. Single-crystal X-ray crystallographically derived structure of 5.
Thermal ellipsoids are drawn at 50% probability. Hydrogen atoms have
been omitted for clarity; P = orange, S = yellow, Ni = red, C = gray, and
Cl = green.
n/a
n/a
bond angles (deg)
5
9a
10a
P1-Ni-P2
P1-Ni-S1
P1-Ni-S2
P2-Ni-S1
P2-Ni-S2
S1-Ni-S2
180.000(6)
84.643(6)
95.357(6)
95.357(6)
84.643(6)
180.000(7)
98.90(5)
86.72(4)
172.78(5)
174.39(5)
87.60(4)
86.80(4)
180.0
Scheme 2. Reaction Scheme for Formation of 5, 8a-d, 9a,b,
d, and 10a-ca
87.204(14)
92.795(14)
92.795(14)
87.205(14)
180.0
spectroscopic and elemental analysis data are consistent with the
proposed solution structure for 5. Single crystals of 5, suitable for
X-ray diffraction analysis, were grown by layering diethyl ether
(Et2O) on top of 5 dissolved in CH2Cl2 in an NMR tube
(Figure 1). The solid-state structure of 5 consists of a Ni center
chelated by 2 equiv of ligand 4. The phosphine and thioether
moieties are in a trans arrangement, with two chloride atoms
coordinated to the open axial positions of the metal, giving rise to
an octahedral geometry. Although there are no crystallographically
characterized octahedral Ni complexes with bidentate phosphinoalk-
yl thioether hemilabile ligands, the P-Ni and S-Ni bond
lengths are consistent with analogous bond lengths in previously
reported structures with monodentate phosphines and thioether
ligands.40,42,45-48 Whether 5 should be referred to as a 20-
electron complex is debatable,49 but regardless, the lability of both
the Ni-Cl and Ni-S interactions is unambiguous in solution,
which indicates that the solid-state structure is not representative of
the species in solution. The Ni-Cl bonds in 5 are relatively
unchanged from the reported Ni-Cl length of 2.38 Å for nickel(II)
chloride hexahydrate50 [2.3674(2) Å]. The P1-Ni-S1 and
P2-Ni-S1 angles in the equatorial plane around the Ni(II) center
deviate from 90° by approximately 5° (Table 1), likely a conse-
quence of the bite angle of chelating ligand 4.51
In order to determine the reason for the observed broadening
in the 31P{1H} NMR spectrum of 5, we conducted a variable
temperature (room temperature to 213 K) NMR study of this
complex in CD2Cl2. As the sample temperature is decreased, the
initially broad resonance at δ 32 resolves into two resonances at
253 K with a major peak at δ 57 and a minor one at δ 30. This
process is completely reversible (Figure 2). One possible ex-
planation for the appearance of two resonances at low tempera-
ture in the 31P{1H} NMR spectrum of 5 relates to the lability of
the Cl- ligands and their exchange from inner to outer sphere on
a time scale too fast to be observed by 31P{1H} NMR spectros-
copy at room temperature. This result is similar to that observed
for an analogous mononuclear Rh(I) tweezer complex.52 When
5 is dissolved in a more polar solvent (such as ethanol), the
31P{1H} NMR spectrum shows three broad resonances at δ 54,
32, and -14 (Figure SI-1, Supporting Information) instead of
the single broad resonance observed in CD2Cl2. The resonance
at δ -14 is downfield from the one observed for the free ligand in
ethanol (δ -16) and is thought to arise from a complex with
a (i) EtOH, 1 h; (ii) 8a, 2 equiv of LiClO4 in EtOH/CH2Cl2, 1 h; 8b, 2
equiv of NaBArF in CH2Cl2, 1 h; 8c, 2 equiv of TlPF6 in CH2Cl2,
1 h; 8d, 2 equiv of AgBF4 in CH2Cl2; (iii) 2 equiv of bis(triphenyl-
phosphine)iminium chloride (PPNþCl-) in CH2Cl2; (iv) single crystals
of 9a, 9d, 10b, 10c grown from CH2Cl2 solutions of 8a, 8b, 8c, and 8d
layered with Et2O; (v) 10a, excess LiClO4 in CH2Cl2, 3 days, singlecrystals
were grown by layering the reaction solution with Et2O.
several P,S ligand systems, where the coordination geometry of
the complexes can be modulated by the deliberate and controlled
introduction of halide ion (vide infra). Given the generality of
these reactions with respect to Ni(II), this system provides a basis
for constructing new Ni(II)-based allosterically regulated cata-
lysts and abiotic enzyme mimics, with chemical reactivities that
complement the other d8 systems studied to date.1,20,21,25-28,30
’ RESULTS AND DISCUSSION
Synthesis of Complex 5. To evaluate the reactivity of Ni(II)
in the context of the WLA and known P,S ligands, we reacted
nickel(II) chloride hexahydrate (NiCl2 6H2O) in ethanol at
3
room temperature with 2 equiv of ligand 425 (Scheme 2). After
1 h, complex 5, trans-[κ2-(Ph)2PCH2CH2SCH3]2Ni(Cl)2, formed
in near quantitative yield as evidenced by 31P{1H} NMR spectros-
copy and a color change from pale green to deep red. The 31P{1H}
NMR spectrum of 5 in CD2Cl2 revealed a broad singlet at
δ 32, which is approximately 50 ppm downfield with respect
to free ligand 4, consistent with ligand coordination. All other
3024
dx.doi.org/10.1021/ja109624m |J. Am. Chem. Soc. 2011, 133, 3023–3033

Journal of the American Chemical Society
ARTICLE
Figure 2. 31P{1H} VT-NMR spectra of 5 in CD2Cl2 and HOMO level comparison of 5, 6, and 7a based on DFT calculations using ADF 2009.01
software. Kohn-Sham representations of frontier orbitals were generated by the ADF 2009.01 GUI.
a
Scheme 3. Reaction Scheme for Formation of 7a0
observed by 31P{1H} NMR spectroscopy (and the identity of
these intermediates), we have explored the electronic structures
of 5, 6, and 7a and some possible intermediates (7b-d, Figures 3
and 4) with DFT calculations (see Experimental Section).
Interestingly, the primary difference between the HOMO and
HOMO-1 levels of 5 is the location of the antibonding orbitals
between Ni and the coordinating S and Cl moieties (compare
Figure 3B with 3C). In the case of the HOMO level, the occupied
antibonding orbitals are located on the Ni and S atoms, whereas
in the HOMO-1 level, the occupied antibonding orbitals are
located on the Ni and Cl atoms. The LUMO level is composed of
orbitals involved in antibonding interactions between Ni, P, and
Cl. Since a five-coordinate complex is a likely intermediate in the
observed 31P{1H} NMR exchange processes, we calculated the
energies of several such complexes (6, 7a-d). The energy
calculated for the HOMO level of a five-coordinate complex
where a single Cl- has been removed (7a) is approximately 77
kcal/mol lower than the analogous level for a five-coordinate
complex where a Ni-S bond has been broken (6, Figure 2). Upon
the basis of pure thermodynamic arguments (inherent intermediate
complex stability), the DFT results suggest an exchange centered
around Cl- ligand dissociation, as opposed to the breaking of P-Ni
or S-Ni bonds.
a (i) 1 equiv of LiClO4 in EtOH/CH2Cl2, 1 h; (ii) 1 equiv of PPNþCl-
in CH2Cl2.
a monodentate version of ligand 4 formed via exchange with
ethanol at the metal-bound phosphine site based on literature
precedent for this type of reaction.53 The other two resonances
are assigned to 5 and 7a, respectively (Figure 2). The assignment
of these resonances will be corroborated further in the section
below describing DFT calculations.
When 5 is dissolved in a less polar solvent such as chlorocyclo-
hexane, a single well-resolved resonance is observed at δ 27. Since
less polar solvents stabilize charged intermediates less effectively than
polar solvents,52,54 rapid room-temperature exchange between 5, 6,
and 7a is prevented. In addition, we synthesized compound 7a0, an
analogue of 7a, by the abstraction of a Cl- ligand from 5 with 1 equiv
of lithium perchlorate (LiClO4, Scheme 3) (SAFETY NOTE: metal
perchlorate salts are potentially explosive and should only be
handled with care). Significantly, complex 7a0, when dissolved in
CD2Cl2, exhibits a low-temperature 31P{1H} NMR resonance
almost identical to the one assigned to 7a in CD2Cl2 (at 213 K δ
57 for 7a0, δ57 for 7a). At room temperature, compound 7a0 exhibits
a broad resonance at δ 52 in its 31P{1H} NMR spectrum, which is
not observed for 5. This observation may be due to an exchange
between five-coordinate species in the presence of only a single
chloride ligand in 7a0, but this has not been fully characterized. In 7a,
the extra Cl- ligand can participate in a series of exchange reactions
involving the displacement of either the phosphine, Cl-, or thioether
at the metal center, causing further signal broadening.53 Although we
were unable to grow single crystals of 7a, all other characterization
data are consistent with its proposed formulation. In addition, an
analogous complex 13, which will be discussed later, has been
synthesized and characterized in solution and the solid-state. Its
NMR data are almost identical to that for 7a0.
Taking the 31P{1H} VT-NMR spectroscopic data and DFT
calculations collectively into consideration allows us to propose a
set of reasonable exchange processes involving 5, 6, and 7a-d
(Figure 4). Upon cooling a solution of 5 in CD2Cl2, one begins to
observe different isomeric structures as a result of slower
exchange at lower temperatures. The upfield resonance at δ
30, which appears at 253 K (Figure 2), is assigned to a static
structure (5), based on the similarity to the sharp 31P{1H} NMR
resonance of 5 observed in chlorocyclohexane (δ 27) at room
temperature. If 6 exists under these conditions, it is in rapid
exchange with the other isomers, since one would expect multi-
ple 31P{1H} NMR resonances for the magnetically inequivalent
P atoms in this complex. At temperatures below 253 K, in
addition to the resonance at δ 30, a single downfield resonance
is observed at δ 57, which has been assigned to 7a upon the basis
of its similarity to 7a0 (δ 57). The DFT calculations suggest that
isomers 7b-d are also possible intermediates, but if these exist,
they are in rapid exchange with 7a at low temperature. However,
both of the resonances for 5 and 7a are clearly resolved at 213 K,
DFT Calculations on Complex 5. In order to more fully
understand 5 and its relation to the possible intermediates
3025
dx.doi.org/10.1021/ja109624m |J. Am. Chem. Soc. 2011, 133, 3023–3033

Journal of the American Chemical Society
ARTICLE
Figure 3. Kohn-Sham representations of the HOMO-LUMO orbitals of 5 were generated from DFT calculations made using the ADF 2009.01
software suite (See Computational Details, Supporting Information): (A) LUMO, -67.3 kcal/mol; (B) HOMO, -86.0 kcal/mol; (C) HOMO-1,
-94.9 kcal/mol.
Figure 4. Kohn-Sham representations of HOMO energy levels in the geometry-optimized structures of 5, 6, and 7a-d were obtained with DFT
calculations using the ADF 2009.01 software suite.
the lowest temperature probed. The broad resonance for 5 at
δ 30 is not observed until the sample is warmed to room tem-
perature. Note that the DFT calculations serve as a qualitative
way to explain these processes and do not attempt to quantitatively
assess the complexity of the exchange processes for 5, 6, and 7a-d.
Therefore, we hypothesize that this downfield resonance at δ 57 can
be assigned to 7a instead of 7b (the lowest energy species) since for
7b one would expect two unique resonances in the 31P{1H} NMR
spectra, due to its magnetically inequivalent phosphorus moieties.
We cannot dismiss the possibility that this resonance at δ 57
represents an averaged signal of two or more penta-coordinate
isomers derived from 5 (7a-d) exchanging rapidly even at 213 K.
Synthesis and Characterization of Complexes 8a, 9a, and
10a by the Abstraction of Cl- Ligands from 5. Since both
theoretical and experimental data suggest a potentially labile
Ni-Cl interaction, we decided to study the chemical abstraction
of Cl- from 5 to better understand the aforementioned exchange
processes and reactivity of 5. Two equivalents of lithium per-
chlorate (LiClO4), dissolved in ethanol, were added to a solution
of complex 5 dissolved in CH2Cl2 at room temperature
Figure 5. 31P{1H} NMR spectra of 8a (A), 9a (B), and 10a (C).
(see SAFETY NOTE). The reaction mixture was vigorously
stirred for 1 h prior to solvent removal under vacuum. The
resulting red-orange powder was dissolved in CH2Cl2 to form a
red mixture, which was filtered to remove LiCl and then taken to
dryness in vacuo, leaving a red-orange powder, 8a. The 31P{1H}
3026
dx.doi.org/10.1021/ja109624m |J. Am. Chem. Soc. 2011, 133, 3023–3033

Journal of the American Chemical Society
ARTICLE
Figure 6. Single-crystal X-ray crystallographically derived structures of 9a (A), 10a (B), and 10b (C) with thermal ellipsoids drawn at 50% probability.
Hydrogen atoms and some counteranions are omitted for clarity; P = orange, S = yellow, Ni = red, C = gray, Cl = green, O = purple, F = light green, and
B = light pink.
Scheme 4. Reaction Scheme for Formation of 8a, 9a, and 10a
Scheme 5. Reaction Scheme for the Formation of 9a and 9d,
with Noncoordinating Counteranion Nickel(II) Saltsa
- a
(A- = ClO4
)
a (i) EtOH, 1 h.
to 9a, but no significant conversion of 9a to 10a takes place under
the conditions studied.
Because of our inability to characterize 8a crystallographically,
we decided to probe independent routes to these complexes
via alternative syntheses. The addition of 2 equiv of ligand 4
dissolved in ethanol to an ethanol solution of nickel(II)
a (i) 2 equiv of LiClO4 in EtOH/CH2Cl2, 1 h; (ii) 2 equiv of PPNþCl-
in CH2Cl2; (iii) solvent layering a CH2Cl2 solution of 8a with Et2O; (iv)
10a: excess of LiClO4 in CH2Cl2, 3 days; single crystals were grown by
layering the reaction solution with Et2O. The arrow lengths between 8a,
9a, and 10a represent the relative stability of each species with regard to
the others.
perchlorate hexahydrate (Ni[ClO4]2 6H2O) (see SAFETY
3
NOTE) resulted in the formation of a yellow powder that
precipitated out of solution over the course of 1 h while the
mixture was stirred vigorously (Scheme 5). After separation
from the supernatant by filtration and drying, this powder was
dissolved in CD2Cl2 and then characterized by 31P{1H} NMR
spectroscopy, which showed a single sharp resonance at δ 58
(9a, Figure 5B). Notably, this 31P{1H} NMR resonance at δ
58 is distinct from that of 8a isolated by chloride abstraction
(δ 53) and identical to that seen from dissolving crystals grown
from a solution of 8a (δ58). Single crystals grown by solvent layering
of a CH2Cl2 solution of 9a from this alternate synthesis (Scheme 5)
with Et2O were characterized by single-crystal X-ray diffraction
analysis to reveal again the structure of 9a (Figure 6A). Interestingly,
when crystals from the same batch were dissolved in CH2Cl2, the
observed 31P{1H} NMR resonance matched the resonance observed
before crystallization (9a, δ 58). These observations confirm that the
broad unresolved resonance seen at δ 53 (8a) when 5 is reacted with
2 equiv of LiClO4 arises from a chemically identical, yet structurally
unique species from 9a and 10a (Scheme 4).
NMR spectrum of 8a dissolved in CD2Cl2 exhibits a single broad
resonance at δ 53 (Figure 5A). Interestingly, when we attempted
to grow crystals of 8a by layering Et2O on top of a CD2Cl2
solution of 8a, we obtained crystals of a new pale yellow
compound, 9a (Scheme 4). A single-crystal X-ray diffraction
study of 9a shows that its Ni center is free of Cl- ligands and has a
nearly square planar geometry. To our surprise, ligands 4 in 9a
are in a cis configuration, unlike the trans configuration observed
in octahedral 5. As expected, when the yellow crystals of 9a were
dissolved in CD2Cl2, the color and the observed 31P{1H} NMR
spectrum did not match those for the CH2Cl2 solution of 8a.
The solution of 9a was light yellow in color, rather than red as
with 8a, and the observed 31P{1H} NMR spectrum revealed
a single, well-resolved resonance at δ 58 (9a, Figure 5B) com-
pared to δ 53 (8a). Interestingly, we noticed that by stirring 5
with an excess of LiClO4 for a prolonged period of time (3 days,
instead of 1 h), a new product formed, 10a. Complex 10a exhibits
a sharp resonance at δ 38 in its in situ 31P{1H} NMR spectrum
(Figure 5C). The solid-state identity of 10a was confirmed by an
X-ray crystallographic study of single crystals grown from the
reaction mixture by layering Et2O over a CH2Cl2 solution of this
material. In the structure of 10a, we observed a trans arrange-
ment of ligands 4 around the Ni metal center (Figure 6B).
Overall, these observations lead us to conclude that the forma-
tion of isomers 8a, 9a, and 10a from 5 is a result of their energetic
similarities. Compound 8a can be viewed as a metastable
intermediate between 9a and 10a. Complex 9a is thermodyna-
mically more stable than 10a; indeed, 10a will convert over time
Since the independent synthesis of square planar complex 9a gave
indirect proof of the existence of a different structural isomer with
identical chemical composition, 8a, we decided to characterize 5, 8a,
9a, and 10a using UV-vis spectroscopy. Each of these complexes
has a different color (vide supra) and a distinct UV-vis signature.
Crystallographically characterized square planar complexes 9a and
10a each exhibit a diagnostic d-d transition band at ∼420 nm, while
complex 8a exhibits a red-shifted band at 499 nm with a substan-
tially higher extinction coefficient (Figures SI-2, SI-3, and SI-4
and Table SI-1, Supporting Information), which can be an indicator
of tetrahedral Ni(II) complexes.55 Indeed, a higher extinction
3027
dx.doi.org/10.1021/ja109624m |J. Am. Chem. Soc. 2011, 133, 3023–3033

Journal of the American Chemical Society
ARTICLE
coefficient is often indicative of the higher prevalence of antibonding
and bonding orbital overlap between the ligands and d-orbitals of a
Ni(II) center (P-Ni and S-Ni interactions) in a tetrahedral
geometry.56 In addition, this transition in 8a (499 nm) is red-shifted
from the square planar 9a (424 nm) in solution, further evidence of
the proposed tetrahedral structure (considering that d-d transitions
become closer in energy in a tetrahedral species as compared to a
square planar species).55-60 Therefore, the proposed tetrahedral
species 8a can be viewed as an intermediate in the interconversion of
9a and 10a (Scheme 4). Note that we considered the potential for a
weak interaction between the “non-coordinating” counterions within
8a and the Ni center, but even at low temperature (203 K in CD2Cl2)
with BF4- and PF6- as the counterions, we saw no evidence of such
interaction by 11B, 19F, and 31P NMR spectroscopies. This does not,
however, completely discount the formation of five-coordinate
intermediates with the counteranion.
We also decided to conduct further UV-vis experimentation
on 7a0 in order to compare it to the other complexes (Scheme 3).
We found that the extinction coefficient for 7a0 was higher than
those of the octahedral complex 5 and square planar complexes
9a and 9d but lower than those of the proposed tetrahedral
complexes 8a and 8d (Table SI-1, Supporting Information). In
addition to this, the UV-vis spectrum shows more similarity to
that of the trans-10a, than the proposed tetrahedral 8a (Figure
SI-2, Supporting Information), which is consistent with the trans
arrangement of the 2 equiv of ligand 4 in the proposed structure
of 7a0. To further extend our understanding of this system, we
proceeded to study other counterions in order to identify what
conditions could control the formation of specific isomers during
chloride abstraction.
Figure 7. Single-crystal X-ray crystallographically derived structures of
9d (A) and 10c (B) with 50% thermal-ellipsoid probability. Hydrogen
atoms are omitted for clarity; P = orange, S = yellow, Ni = red, C = gray,
Cl = green, O = purple, F = light green, and B = light pink.
is exacerbated (Table 1) in the case of 10b, which is likely a
consequence of the larger BArF counteranion and corresponding
crystal packing forces. While the Ni-P bond distances in the trans
complexes 10a and 10b are slightly longer than the Ni-S bonds, the
opposite is true in the cis complex 9a, a consequence of the relative
trans-donor abilities of the phosphine and thioether,58 respectively.
When a CD2Cl2 solution of 5 was stirred with a slight excess of
NaBArF, two resonances were observed by 31P{1H} NMR
spectroscopy at δ 61 and 35 after only 1 h (Scheme 2). These
are attributed to 9b and 10b, which are the cis and trans isomers,
respectively. These assignments are consistent with our previous
ones for 9a and 10a. The distribution of these isomers at room
temperature is likely not representative of a thermodynamic
equilibrium, since upon sequential cooling of these mixtures to
213 K the ratio of the isomers is not altered. Heating, unfortu-
nately, results in decomposition.
The observation of the strong dependence of the geometry at
the Ni(II) center on the counteranion in both solution and the
solid-state prompted us to investigate this phenomenon further.
Complexes 8c and 10c were prepared under conditions identical
to those used to prepare 8a, 9a, and 10a, but 2 equiv of
thallium(I) hexafluorophosphate (TlPF6) (SAFETY NOTE:
Tl(I) salts are very toxic) were used as the abstracting agent in
place of LiClO4. In this case, a 31P{1H} NMR spectroscopic
study revealed only one broad resonance at δ 53 (8c). Interest-
ingly, the single crystals obtained using the same layering
technique as previously described for 9a and 10a-b provided
10c, which had a trans arrangement of the 2 equiv of ligand 4
around the metal center (Figure 7B). Dissolving these crystals of
10c in CD2Cl2 and observing the resulting red solution by
31P{1H} NMR spectroscopy revealed a single broad resonance
at δ 57, indicating that, in solution, the less stable trans product
converts into the cis isomer 9c, similar to the conversion of 10a
into 8a.
All solution characterization of 10a was limited to in situ
techniques, where 10a was generated from 5 by Cl- ligand
abstraction, since 10a is less stable than both 8a and 9a. Indeed, if
10a was prepared in CD2Cl2 and left in an NMR tube over a
period of several weeks, the sharp single resonance observed by
31P{1H} NMR spectroscopy at δ 38 began to broaden with a
gradual color change from yellow to red and growth of a new
broad resonance at δ 54, assigned to 8a (Scheme 2). This
indicates that the stability of trans-10a is less than that of either
proposed tetrahedral 8a or cis-square planar 9a.
Synthesis and Characterization of Cl- Abstracted Complexes
With Other “Noncoordinating” Counteranions: 8b-d, 9b,d,
10b,c. Proposed tetrahedral complex 8b was generated by reacting
octahedral complex 5 in CH2Cl2 with 2 equiv of sodium tetrakis-
[3,5-bis(trifluoromethyl)phenyl]borate (NaBArF) followed by vig-
orous stirring for 1 h at room temperature (Scheme 2). The 31P{1H}
NMR spectrum of this orange-red solution exhibits a single broad
resonance at δ 60, consistent with quantitative formation of 8b.
Filtering this solution of 8b to remove precipitated sodium chloride
(NaCl), followed by solvent evaporation in vacuo, produced 10b as a
red-orange solid. This powder was redissolved in CH2Cl2 and
layered with Et2O in order to grow single crystals suitable for
X-ray diffraction studies. The solid-state structure of 10b like 10a
shows a trans arrangement of 2 equiv of ligand 4 around the Ni(II)
center (Figure 6C). By comparing the structures of 10a and 10b, one
can observe significant differences in bond angles and lengths around
the metal center, which are likely the result of the difference in the
steric bulk of the counterions and crystal packing (Table 1).61-64
The Ni(II) centers in 10a and 10b adopt square planar geometries
and share certain attributes. In particular, like 5, the P1-Ni-S1
angle is smaller than the P1-Ni-S2 angle, due to the influence of
the bite angle of ligand 4.51 This difference in observed angles
As a final comparison in probing the effect of noncoordinating
counteranion contributions on the geometry of this Ni(II) system,
we investigated the syntheses of the square planar and proposed
tetrahedral complexes using tetrafluoroborate (BF4-) salts
(Scheme 2). Tetrahedral complex 8d was prepared by adding 2
equiv of silver(I) tetrafluoroborate (AgBF4) to a solution of 5 in
CH2Cl2 and stirring vigorously for 1 h (Scheme 4). Removal of
solvent in vacuo revealed a red powder, which exhibits a broad
singlet at δ 55 in CD2Cl2 in its 31P{1H} NMR spectrum. This
resonance is consistent with the formation of the tetrahedral
complex (vide supra). Much like the structural change seen for
8a converting to 9a under solvent layering crystallization condi-
tions, when a CH2Cl2 solution of 8d was layered with Et2O on top
3028
dx.doi.org/10.1021/ja109624m |J. Am. Chem. Soc. 2011, 133, 3023–3033

Journal of the American Chemical Society
ARTICLE
Figure 8. Single-crystal X-ray crystallographically derived structures of 12 (A), 13 (B), and 14 (C) with 50% thermal-ellipsoid probability. Hydrogen
atoms are omitted for clarity; P = orange, S = yellow, Ni = red, C = gray, and Cl = green. Blue corresponds to a refined C disorder in the naphthalene
group.
Scheme 6. Reaction Scheme for the Formation of 12-14,
and SI-4 and Table SI-1, Supporting Information). These data are
analogous to what was observed for 8a and 9a and consistent with
literature model complexes and spectroscopic assignments.55
The complexes formed with the various counterions exhibit
qualitatively similar chemistry (but with different transformation
rates between the three isomers 8, 9, and 10; Scheme 2), in
addition to some deviations in their 31P{1H} NMR spectra.
Although we were unable to determine a specific reason for these
observations, these counteranions have inherent differences in
size and degree of electron delocalization. Others have previously
observed how these properties affect reactivity in related co-
ordination complexes.61-64
Where R = (2-Naphthalene)ethyla
One important feature of all the Cl--abstracted Ni(II) com-
plexes in this study (7a0, 8a-d, 9a-d, 10a-c) is that they can be
quantitatively converted back to the starting material 5 by the
addition of bis(triphenylphosphine)iminium chloride (PPNþCl-)
(Scheme 2). In addition to providing further evidence for the pro-
posed structures, these reactions show how elemental anions such as
Cl- can be used to control the orientation of pendant groups in
tweezer complexes. Such capabilities are important in the design of
more sophisticated allosteric structures, and the cis to trans iso-
merization reactions common to this family of Ni complexes may
become a powerful new pathway for controlling architecture that is
not characteristic of the previously studied isolectronic WLA Rh(I),
Pt(II), and Pd(II) systems.1,20,21,25-28,30
Synthesis and Characterization of Complexes 12, 13, and
14. In the interest of using these previous findings to probe the
generality of the aforementioned reactions with respect to
functionalized phosphinoalkyl thioether ligands, we synthesized
a series of model complexes 12-14 (Scheme 6). Similar to the
synthesis of 5, by reacting 2 equiv of a (2-(2-naphthalene)-
thioethyl)ethyldiphenylphosphine ligand, 11 (P,S-NAP; see Ex-
a (i) EtOH, 1 h; (ii) 2 equiv of Tl(CF3SO3) (14) or 1 equiv of TlPF6
(13) in CH2Cl2; (iii) 1 or 2 equiv of PPNþCl- in CH2Cl2.
in an NMR tube, pale yellow crystals formed. An X-ray diffraction
study conducted on these crystals revealed the structure of 9d,
which has a cis-square planar arrangement of 2 equiv of ligand 4
(Figure 7A). Similar to the crystal structure of the cis-square planar
complex 9a, 9d exhibits shorter Ni-P bonds than Ni-S bonds,
again in contrast to the observed trend for the trans complexes
10a-c. The trend of the bite angle contributing to a smaller
P1-Ni-S1 bond angle than the P1-Ni-P2 bond angle also
holds. Finally, the bite angle in 9d is nearly identical to the one in 9a.
Complex 9d also can be synthesized directly by reacting 2
equiv of ligand 4 with an ethanol solution of Ni[BF4]2 6H2O, a
3
reaction analogous to the one used to prepare ClO4--containing
complex 9a from Ni[ClO4]2 6H2O. After the mixture contain-
3
ing 2 equiv of ligand 4 and 1 equiv of Ni[BF4]2 6H2O was
3
vigorously stirred for 1 h at room temperature (Scheme 5), a
microcrystalline yellow powder precipitated and was recovered
by filtration. 31P{1H} NMR spectroscopic analysis of this powder
dissolved in CD2Cl2 showed a single resonance at δ 58,
consistent with the formation of 9d. An X-ray diffraction study
of single crystals grown by layering Et2O over a CH2Cl2 solution
of this material confirmed the identity of this complex as 9d.
When these single crystals are dissolved in CD2Cl2 and char-
acterized by 31P{1H} NMR spectroscopy, they show a singlet at
δ 58, which is diagnostic of 9d. Consistent with the results for
single crystals of 9a grown from a solution of 8a, dissolving
crystals of 9d grown from a CH2Cl2 solution of 8d and
subsequent characterization by 31P{1H} NMR spectroscopy
shows a single resonance at δ 58. A comparison of the UV-vis
spectra for 8d and 9d respectively shows that proposed tetra-
hedral complex 8d exhibits a band at 506 nm that has a
significantly higher extinction coefficient than the analogous
band at 420 nm for the square planar complex 9d (Figures SI-3
perimental Section), with NiCl2 6H2O, we prepared and were
3
able subsequently to isolate a new complex, 12 (Scheme 6). The
solid-state structure of 12 was determined by a single-crystal X-ray
diffraction study carried out with crystals grown by layering Et2O on
top of a CH2Cl2 solution of this material. In the solid state, complex
12 has an octahedral geometry similar to that observed for 5, and its
fluxional behavior as observed by 31P{1H} NMR spectroscopy shows
remarkable similarities as well (Figures 8A and SI-5, Supporting
Information). All other characterization data are consistent with the
proposed structural formulation of 12.
Interestingly, the mono-Cl- adduct complex 13 can be syn-
thesized from 12 by the addition of 1 equiv of TlPF6 (see
SAFETY NOTE) inCH2Cl2. This red compound, when dissolved in
CD2Cl2, showed a single resonance in its 31P{1H} NMR spec-
trum at δ 51. Single crystals of 13 suitable for an X-ray diffraction
study were grown by layering Et2OontopofaCH2Cl2 solution of 13.
3029
dx.doi.org/10.1021/ja109624m |J. Am. Chem. Soc. 2011, 133, 3023–3033

Journal of the American Chemical Society
Table 2. Selected Bond Length and Bond Angle Data for the Crystallographically-Derived Structures of 10b,c, 9d, and 12-14
ARTICLE
bond lengths (Å)
10b
10c
9d
12
13
14
P1-Ni
P2-Ni
S1-Ni
S2-Ni
Cl1-Ni
Cl2-Ni
2.2124(4)
2.2124(4)
2.1934(4)
2.1934(4)
n/a
2.2320(11)
2.2320(11)
2.1940(14)
2.1940(14)
n/a
2.1779(16)
2.1710(16)
2.2133(17)
2.2140(16)
n/a
2.509 (2)
2.509 (2)
2.454 (2
2.203 (3)
2.197 (3)
2.211 (3)
2.234 (3)
2.456 (3)
n/a
2.1740(6)
2.1777(7)
2.2155(7)
2.2138(6)
n/a
2.454 (2)
2.3635(17)
2.3636(17)
n/a
n/a
n/a
n/a
bond angles (deg)
10b
10c
9d
12
13
14
P1-Ni-P2
P1-Ni-S1
P1-Ni-S2
P2-Ni-S1
P2-Ni-S2
S1-Ni-S2
180.0
180.0
98.24(6)
86.77(6)
171.88(7)
174.04(6)
87.94(6)
87.39(6)
180.0
166.85(13)
87.07(12)
90.81(13)
90.44(12)
87.70(12)
162.51(12)
99.43(2)
86.31(2)
170.15(3)
170.69(3)
86.97(2)
88.35(3)
86.779(15)
93.221(15)
93.221(15)
86.779(15)
180.0
87.22(5)
92.78(5)
92.78(5)
87.22(5)
180.0
84.38(8)
95.62(8)
95.62(8)
84.38(8)
180.0
A single-crystal X-ray diffraction study shows that 13 (Figure 8B) is a
square-pyramidal complex with two fully chelated equivalents of
phosphinoalkyl thioether ligand 11 around the Ni center in a trans
arrangement with a single apical Cl- ligand. Since 5 and 12 are
isoelectronic and coordinatively isostructural, structure 13 provides
further evidence of 7a0 (compare Schemes 3 and 6). The X-ray crystal
structure of 13 exhibits a Ni-Cl bond longer [2.456(3) Å, Table 2]
than the Ni-Cl bonds in 12 [2.3635(17) Å, Table 2]. This Ni-Cl
bond length is the longest observed for all crystallographically
characterized Ni(II) complexes discussed in this paper. Interest-
ingly, in 13 the two Ni-S bond lengths are almost identical to the
two Ni-P bond lengths.
It was also possible to isolate an analogue of 9a and 9d
for complex 12 under complete chloride-abstracting conditions.
The addition of 2 equiv of thallium(I) triflate [Tl(CF3SO3)] (see
SAFETY NOTE) to a solution of 12 in CH2Cl2 yielded a deep
red solution, which when filtered and reduced to dryness under
vacuum gave a red powder, 14. When this powder was dissolved
in CD2Cl2 and characterized by 31P{1H} NMR spectroscopy, it
showed a single resonance at δ 58. Layering Et2O on top of a
solution of 14 in CH2Cl2 yielded single crystals suitable for an
X-ray diffraction study. This structural study showed a square-
planar complex with a cis arrangement of 2 equiv of 11 around
the Ni center, similar to that observed in the complexes with
ligand 4 (9a, 9d). The crystal structure of 14 exhibited Ni-P and
Ni-S bond lengths, which were very similar to the isoelectronic
analogues 9a and 9d. As mentioned above, the conformational
flexibility of the ethyl spacer and asymmetric attachment of the
naphthalene to the S moiety on ligand 11 allows for multiple
orientations, which creates disorder in the solid-state. Indeed,
two dominant conformations were resolved in the refined crystal
structure of 14 (the secondary positions for the naphthalene
groups are shown in blue in Figure 8C).
assigned to the Cl-- dissociated adduct 7a. Similarly, the
resonances at δ 55 and 31 observed when cooling 12 to 245 K
are assigned to the analogous octahedral and square-pyramidal
complexes 12 and 13, respectively. When the system is cooled to
223 K, a third resonance appears and resolves at δ 52 and remains
significantly broader than the other resonances at δ 55 and
31 down to 203 K. The broad resonance at δ 52 is likely
due to slowly exchanging structural isomers (with the square
pyramidal complex 13, there at least four likely conformational
arrangements).
Compared to 5, the solid-state structure of complex 12 shows
slightly longer Ni-P bonds [2.509(2) Å, Table 2] and slightly
shorter Ni-Cl bonds [2.3635(17) Å, Table 2]. All of the bond
angles in 12 follow the same trends observed in 5, with the angles
between the coordinating moieties of ligand 11 (P1-Ni-S1,
P2-Ni-S2) being slightly smaller than the angles between the
coordinating moieties of the separate ligands (P1-Ni-P2,
S1-Ni-S2).
’ CONCLUSIONS
We have synthesized and studied a new system consisting of
Ni(II) complexes formed from P,S-alkyl hemilabile ligands and
Ni(II) salts wherein transformations of complex geometry can be
induced by the introduction or abstraction of a coordinating
halide. These processes originate from an octahedral Ni complex
(5) in which the Cl- ligands and thioether moieties possess a high
degree of lability. Consequently, the solid-state structure of 5
represents an average of the many subspecies that exist in solution,
which arise from several exchange pathways. Some of these
subspecies can be obtained directly by the controlled abstraction
of Cl- ligands, which provides a basis for regulated geometric
switching around the Ni(II) center. It is also clear that the
counterions in each of these complexes can have a significant
effect on the isomeric distribution and exchange pathways based
on the differences observed between complex behavior in the
solution and solid states. These transformations are fundamentally
different from the series of reactions previously studied involving
halide-induced switchingin analogous systems usingRh(I), Pt(II),
and Pd(II). It is evident from these experiments that the Ni(II)
chemistry in this context is much richer and more complex than
the previously studied WLA systems, with a variety of new
geometries and interconversions available, where many can be
Sequential cooling of 12 in CD2Cl2 (room temperature to
203 K) and concomitant monitoring by 31P{1H} NMR spec-
troscopy showed evidence for the same behavior observed for
5 under identical conditions (Figure SI-5, Supporting In-
formation). The initially broad singlet observed at room tem-
perature (δ 31) for 12 begins to sharpen with decreasing
temperature until 245 K, where a second resonance appears at
δ 55, similar to the resonances observed for 5 at δ 57 and 30 at
these temperatures. Recall that the latter resonance is assigned to
octahedral complex 5 and the former resonance at δ 57 was
3030
dx.doi.org/10.1021/ja109624m |J. Am. Chem. Soc. 2011, 133, 3023–3033

Journal of the American Chemical Society
ARTICLE
obtained in a controlled fashion. On the basis of our studies, a
mechanism that explains these geometrical transformations is
proposed, where it is likely that a four-coordinate P,S-alkyl Ni(II)
complex undergoes an isomerization via a transient tetrahedral
intermediate (with or without a weak interaction from the
noncoordinating ion) between a cis- and trans-square planar
arrangement. The presence of Ni complexes in distorted tetra-
hedral geometries in addition to ones with square planar geome-
tries is not without precedence; indeed, the work of Miedaner et
al.51 showed that Ni(II) and Ni(I) complexes with phosphine
chelates (nonhemilabile) exhibit distinct tetrahedral and square
planar geometries. They argue that since the potential energy
surface of the Ni(II) system is fairly soft, a balance between these
steric and electronic forces can create a “compromise” and allow
for the formation of multiple geometries, as observed here.
In many ways this complicated puzzle highlights the plasticity of
the Ni(II) environment and the fact that, while the chemistry is rich,
characterization and application of these reactions can be a daunting
task. This study provides a glimpse at the key compounds and
intermediates that should prove useful in understanding the neces-
sary reaction pathways for application in functionalized WLA
complexes. For these reasons, Ni(II) chemistry provides a signifi-
cant challenge and also an opportunity for use in a nontraditional
WLA chemistry. The set of reactions discovered may allow one to
use halide ions (and possibly other coordinating molecules) to
allosterically trigger the interaction between two functional groups
appended to the chalcogen moieties via a cis-trans isomerization.
Ultimately, future success in these endeavors will depend on an in-
depth understanding of how to control these switching processes by
reaction conditions and judicious ligand design.
corrections used ZORA (zero-order relativistic approximation), and
core electrons were unrestricted.
Single crystals suitable for X-ray diffraction studies were mounted
using oil (Infineum V8512) on a glass fiber. All measurements were
made on a Bruker APEX-II CCD with graphite-monochromated Mo KR
radiation (5, 9d, 10a-c, 14) or with graphite-monochromated Cu KR
radiation (9a, 12, 13). The data were collected at a temperature of
100(2) K using an APEX2 V2.1-4 (Bruker, 2007) detector and
processed using SAINTPLUS from Bruker. The data were corrected
using their respective linear absorption coefficients, μ, before being
corrected for Lorentz and polarization effects. The structures were
solved by direct methods and expanded using Fourier techniques.
Neutral atom scattering factors were taken from Cromer and Waber.68
Anomalous dispersion effects were included in Fcalc;
69 the values for Df 0
and Df 00 were those of Creagh and McAuley.70 The values for the mass
attenuation coefficients are those of Creagh and Hubbell.71 All calcula-
tions related to these corrections were performed using the Bruker
SHELXTL3 crystallographic software package.
trans-[K2-(Ph)2PCH2CH2SCH3]2Ni(Cl)2 (5). A flask equipped
with a magnetic Teflon-coated stir bar was charged with 0.059 g of
NiCl2 6H2O (0.25 mmol) in 5 mL of absolute EtOH. To this green
3
solution was added 0.130 g of 4 (0.5 mmol). The mixture turned deep
red in color almost immediately and was stirred for 1 h before all volatiles
were reduced in vacuo, leaving a crystalline green powder in 99% yield.
1H NMR (CD2Cl2, 400 MHz): δ 12.12, 7.73-7.52 (br m, 20H, ArH), δ
5.10 (br s, 8H, -CH2-), δ 2.47 (br s, 6H, -CH3). 13C{1H} NMR
(CD2Cl2, 100 MHz): δ 159.1 (br), δ 140.3 (br), δ 132.7 (sh), δ 132.5
(sh), δ 132.2 (sh), δ 129.8 (sh), δ 35.2 (br), δ 21.0 (br), δ 9.9 (br).
31P{1H} NMR (CD2Cl2, 161 MHz): δ 31.8 (br s). ESIMS (m/z) for
[M - Cl]þ: calcd 613.0619, found 613.0265. Anal. Calcd for C30H34-
Cl2NiP2S2: C, 55.41; H, 5.27; Cl, 10.90. Found: C, 53.42; H, 5.16; Cl,
10.62.
[K2-(Ph)2PCH2CH2SCH3]2NiCl[ClO4] (7a0). A flask equipped
with a magnetic Teflon-coated stir bar was charged with 0.100 g of 5
(0.15 mmol) in 5 mL of CH2Cl2. To the formed light red solution was
added slowly 0.033 g (0.30 mmol) of LiClO4 dissolved in a minimal
amount of absolute EtOH. After 1 h of vigorous stirring, the mixture
turned deep red in color and was filtered to remove LiCl before the
solvent was removed in vacuo to reveal a crystalline red powder (0.107 g,
98%). 1H NMR (CD2Cl2, 400 MHz): δ 7.23-7.51 (br s, 20H, ArH), δ
3.06 (br s, 8H, -CH2-), δ 2.77 (br s, 6H, -CH2-), δ 2.02 (br s, 3H,
-CH3). 13C{1H} NMR (CD2Cl2, 100 MHz): δ 133.9 (m), δ 132.6
(sh), δ 129.4 (sh), δ 127.4 (br), δ 34.0 (sh), δ 31.2 (sh), δ 21.6 (sh).
31P{1H} NMR (CD2Cl2, 161 MHz): δ 52 (br s). ESIMS (m/z) for
[M - Cl]þ: calcd 613.0619, found 613.0613. Anal. Calcd for C30H34Cl2O4-
NiP2S2: C, 50.45; H, 4.80. Found: C, 50.34; H, 4.87.
[K2-(Ph)2PCH2CH2SCH3]2Ni[ClO4]2 (8a, 9a, 10a). 8a: A flask
equipped with a magnetic Teflon-coated stir bar was charged with 0.144
g of 5 (0.25 mmol) in CH2Cl2. To this was added 0.053 g of LiClO4 (0.5
mmol) dissolved in absolute EtOH and the reaction was stirred for 1 h
before all volatiles were removed in vacuo to yield a red solid in 98%
yield. 1H NMR (CD2Cl2, 400 MHz): δ 7.96-7.10 (br m, 20H, ArH), δ
5.50 (br s, 8H, -CH2-), δ 3.24-1.21 (br s, 6H, -CH3). 13C{1H}
NMR (CD2Cl2, 100 MHz): δ 140.5 (br), δ 134.0.8 (m), δ 131.8 (sh), δ
130.8 (sh), δ 129.3 (sh), δ 128.7 (sh), δ 30.1 (br), δ 25.5 (br), δ 16.2
(br). 31P{1H} NMR (CD2Cl2, 161 MHz): δ 53 (br s). Anal. Calcd for
8a, C30H34Cl2NiO8P2S2: C, 46.30; H, 4.40. Found: C, 45.75; H, 4.82.
9a: A flask equipped with a magnetic Teflon-coated stir bar was
charged with 0.130 g of 4 (0.5 mmol) in CH2Cl2. To this was added
’ EXPERIMENTAL SECTION
Materials and Methods. All chemicals were purchased from
Sigma-Aldrich and Strem Chemicals and used without further purifica-
tion, unless otherwise specifically mentioned. All solvents (CH2Cl2,
CH3CH2OCH2CH3, CH3CH2OH) were purchased from Sigma-
Aldrich as anhydrous grade and used as received. The Ni salts containing
a noncoordinating anion (Ni[BF4]2 6H2O and Ni[ClO4]2 6H2O) and
3
3
respective thallium(I) salts [TlPF6 and Tl(CF3SO3)] were purchased from
Strem Chemicals and used without further purification. The P,S-Me ligand
was prepared according to previously published methods.25 2-(2-Naphtha-
lene)ethyl bromide was prepared from (2-(2-naphthalene)ethanol by
bromination under nitrogen with excess PBr3 in anhydrous CH2Cl2 over-
night in 67% yield.65 All NMR spectra were recorded on a Bruker Avance
400 MHz equipped with a broad-band probe and TopSpin software.
UV-vis spectra were obtained on a Varian Cary 5000 UV-vis-NIR
spectrophotometer. Flash chromatography was preformed using a Biotage
HPFC SP4 Flash Purification System on silica. All manipulations with
Ni(II) complexes were done at ambient conditions, and phosphine ligands
were synthesized and stored under inert atmosphere of N2 using Schlenk
techniques or a glovebox (Vacuum Atmospheres, Nexus).
Bond energies and orbital structures for 5, 6, and 7a were obtained
from DFT calculations based on the geometry-optimized version of the
crystallographically determined structure of 5, which was edited for 6
and 7a with the ADF 2009.01 GUI. Complexes 7b-d were constructed
in the ADF 2009.01 GUI before geometry optimization. The optimiza-
tion used basis sets containing triple-ζ functions with one polarization
function (TZP), the local density approximation of Becke66
and Perdew67 with the associated default exchange and correlation
corrections, and the density functional methods available in the
Amsterdam Density Functional (ADF2009.01) suite on a 16-core
Parallel Quantum Solutions (PQS) computational cluster. Relativistic
0.091 g of Ni[ClO4]2 6H2O (0.25 mmol) and the reaction was stirred
3
for 1 h during which time a yellow microcrystalline solid precipitated
from solution. This solid was removed from the supernatant by filtration
and dried on a glass frit. 1H NMR (CD2Cl2, 400 MHz): δ 7.55-7.43
(br m, 20H, ArH), δ 3.25-2.68 (br m, 16H, -CH2-, -CH3). 13C{1H}
3031
dx.doi.org/10.1021/ja109624m |J. Am. Chem. Soc. 2011, 133, 3023–3033

Journal of the American Chemical Society
ARTICLE
NMR (CD2Cl2, 100 MHz): δ 133.9 (sh), δ 129.8 (sh), δ 123.5 (br), δ
35.8 (br), δ 33.0 (br), δ 24.2 (br). 31P{1H} NMR (CD2Cl2, 161 MHz):
δ 57 (br s). Anal. Calcd for 9a, C30H34Cl2NiO8P2S2: C, 46.30; H, 4.40.
Found: C, 45.91; H, 4.13.
CD2Cl2, 22 °C): δ 7.78-7.31 (br, m, 15H, ArH), δ 2.99-2.98 (br, m,
2H, -CH2-), δ 2.89-2.87 (br, m, 2H, -CH2-), δ 2.65-2.61 (br, m,
2H, -CH2-), δ 2.38-2.35 (br, m, 2H, -CH2-). 31P{1H} NMR (161
MHz, CD2Cl2, 22 °C): δ -17.0 (s).
trans-[K2-(Ph)2PCH2CH2SCH2CH2-(2-naphthalene)]2Ni(Cl)2
(12). A flask equipped with a magnetic Teflon-coated stir bar was charged
10a: A flask equipped with a magnetic Teflon-coated stir bar was
charged with 0.050 g of 5a (0.076 mmol) in CH2Cl2. To this was added
0.050 g of LiClO4 (0.46 mmol) and the reaction was stirred for 3 days
until yellow powder precipitated from solution. The supernatant solu-
tion was separated and characterized by NMR, and single crystals
suitable for X-ray diffraction were grown from this solution enabling
the confirmation of the structure of 10a. 1H NMR (CD2Cl2, 400 MHz):
δ 7.73-7.51 (br m, 20 H, ArH), δ 2.66 (br m, 8H, -CH2-), δ 2.12 (br
m, 8H, -CH2-), δ 1.27 (br s, 6H, -CH3). 13C{1H} NMR (CD2Cl2,
100 MHz): δ 206.5 (sh), δ 133.6 (sh), δ 132.8 (m), δ 131.0 (br), δ
129.5 (sh), δ 129.0 (sh), δ 30.6 (sh), δ 29.7 (sh). 31P{1H} NMR
(CD2Cl2, 161 MHz): δ 39 (br s).
K2-(Ph)2PCH2CH2SCH3]2Ni[BArF]2 (8b). A flask equipped with
a magnetic Teflon-coated stir bar was charged with 0.144 g of 5a (0.25
mmol) in CH2Cl2. To this was added 0.428 g of NaBArF (0.5 mmol) and
the reaction was stirred for 1 h before being reduced under vacuum to reveal a
yellow solid. 1HNMR(CD2Cl2, 400MHz):δ7.73-7.22 (br m, 44H, ArH),
δ 2.98-2.66 (br m, 8H, -CH2-, -CH3) δ 1.98 (br s, 6H, -CH3).
31P{1H} NMR (CD2Cl2, 161 MHz): δ 60. Anal. Calcd for C94H58B2F48-
with 0.500 g (2.1 mmol) of nickel(II) chloride hexahydrate (NiCl2 6H2O)
3
in ethanol at room temperature. To this mixture was added 2 equiv of ligand
11 (P,S-NAP) (1.686 g, 4.2 mmol) dissolved in a minimum amount of
CH2Cl2. The resulting mixture was stirred for 1 h, causing the color of the
solution to change from pale green to deep red. The solvent was removed in
vacuo to yield a fine crystalline green powder (1.9 g, 97%). 1H NMR (400
MHz, CD2Cl2, 22 °C): δ 7.78-7.73, δ 7.44 (br, s, ArH, 34 H); δ 3.7-1.25
(br, s, -CH2-, 16 H). 13C{1H} NMR (CD2Cl2, 100 MHz): δ 168.5 (br),
156.5 (br), 147.6 (br), 138.9 (br), 134.7 (sh), 134.2 (sh), 133.0 (sh), 132.8
(sh), 132.5 (sh), 131.6 (m), 130.5 (m), 130.0 (m), 129.6 (sh), δ 128.6
(sh), δ 128.3 (sh), δ 128.0 (sh), δ 127.8 (sh), δ 127.5 (sh), δ 126.8 (sh), δ
126.7 (sh), δ 126.2 (sh), δ 126.0 (sh), δ 41.3 (br), δ 39.1 (br), δ 36.7 (br),
δ 32.9 (br), δ 23.2 (br). 31P{1H} NMR (161 MHz, CD2Cl2, 22 °C):
δ 30.9 (br, s). ESIMS (m/z) [M - Cl]þ: calcd 893.1871, found 893.1895.
Anal. Calcd for C52H50Cl2NiP2S2: C, 67.11; H, 5.42. Found: C, 65.28;
H, 5.22.
trans-[K2-(Ph)2PCH2CH2SCH2CH2-(2-naphthalene)]2NiCl[PF6]
(13). One equivalent of thallium(I) hexafluorophosphate (TlPF6) (0.038 g,
0.107 mmol) was added to a solution of 12 (0.100 g, 0.107 mmol) in CH2Cl2
and stirred for 1 h. The solution was filtered and solvent was removed in
vacuo, yielding a crystalline red powder (0.103 g, 93%). 1HNMR(400MHz,
CD2Cl2, 22 °C): δ 7.77, δ 7.75-7.08 (br, m, ArH, 34 H); δ 2.95, δ 2.80, δ
2.63 (br, m, -CH2-, 16 H). 31P{1H} NMR (161 MHz, CD2Cl2, 22 °C):
δ51.4 (br, singlet). ESIMS (m/z) [M - PF6]þ: calcd 893.1871, found
893.1859. Anal. Calcd for C52H50ClF6NiP3S2: C, 60.05; H, 4.85. Found: C,
62.75; H, 4.93.
cis-[K2-(Ph)2PCH2CH2SCH2CH2-(2-naphthalene)]2Ni[CF3SO3]2
(14). A flask equipped with a magnetic Teflon-coated stir bar was
charged with 2 equiv of thallium(I) triflate [Tl(CF3SO3)] (0.76 g, 0.21
mmol), in a solution of 12 (0.100 g, 0.107 mmol) in CH2Cl2,andwasallowed
to stir for 1 h. The resulting mixture was filtered and the solvent removed in
vacuo, leaving a red powder (0.214 g, 88%). 1H NMR (400 MHz, CD2Cl2,
22 °C): δ 7.48-7.13 (br, m, ArH, 34 H); δ 3.07-1.59 (br, m, -CH2-,
16 H). 31P{1H} NMR (161 MHz, CD2Cl2, 22 °C): δ 57.5 (br, s).
Anal. Calcd for C54H50F6NiO6P2S4: C, 56.02; H, 4.35. Found: C, 56.73;
H, 4.27.
NiP2S2 CH2Cl2: C, 47.73; H, 2.53. Found: C, 47.58; H, 2.32.
3
[K2-(Ph)2PCH2CH2SCH3]2Ni[PF6]2 (8c). A flask equipped with a
magnetic Teflon-coated stir bar was charged with 0.144 g of 5a (0.25
mmol) in CH2Cl2. To this was added 0.177 g of Tl(PF6) (0.5 mmol) and
the reaction was stirred for 1 h before being reduced under vacuum to
1
reveal a red solid. H NMR (CD2Cl2, 400 MHz): δ 8.66, 7.71-7.51,
δ 7.06 (br s, 20H, ArH), δ 6.29 (br s, 8H, -CH2), δ 2.74-2.49 (br m,
6H, -CH3). 31P{1H} NMR (CD2Cl2, 161 MHz): δ 51 (br, s, cis), δ
37.7 (br, s, trans). Anal. Calcd for C30H34F12NiP4S2: C, 41.45; H, 3.94.
Found: C, 41.26; H, 3.65.
[K2-(Ph)2PCH2CH2SCH3]2Ni[BF4]2 (8d, 9d). 8d: A flask equip-
ped with a magnetic Teflon-coated stir bar was charged with 0.144 g of 5
(0.25 mmol) in CH2Cl2. To this was added 0.097 g of AgBF4 (0.5 mmol)
and the reaction was stirred for 1 h before being reduced in vacuo,
leaving a red solid. 1H NMR (CD2Cl2, 400 MHz): δ 7.69-7.40 (br m,
20H, ArH), δ 3.70-1.30 (br m, 14H, -CH2-, -CH3). 13C{1H} NMR
(CD2Cl2, 100 MHz): δ 144.6 (br), δ 135.1 (m), δ 131.9 (sh), δ 130.8
(br), δ 129.3 (m), δ 130.8 (sh), δ 128.8 (sh), δ 20.2 (br), δ 13.57 (br).
31P{1H} NMR (CD2Cl2, 161 MHz): δ 54 (br s). Anal. Calcd for 8d
C30H34B2F8NiP2S2: C, 47.85; H, 4.55. Found: C, 48.03; H, 4.58.
9d: A flask equipped with a magnetic Teflon-coated stir bar was
charged with 0.130 g of 4 (0.5 mmol) in CH2Cl2. To this was added
’ ASSOCIATED CONTENT
0.091 g of Ni[ClO4]2 6H2O (0.25 mmol) and the reaction was stirred
3
S
Supporting Information. Additional NMR data, UV-vis
for 1 h during which time a yellow microcrystalline solid precipitated
from solution. This solid was removed from the supernatant by filtration
and dried on a glass frit. 1H NMR (CD2Cl2, 400 MHz): δ 7.56-7.44
(br m, 20H, ArH), δ 3.20 (br s, 4H, -CH2-), δ 2.70-2.59 (br m,
10H, -CH2-, -CH3). 13C{1H} NMR (CD2Cl2, 100 MHz): δ 133.8
(sh), δ 133.3 (sh), δ 129.8 (sh), δ 123.4 (m), δ 35.3 (m), δ 32.8 (m), δ
23.9 (m). 31P{1H} NMR (CD2Cl2, 161 MHz): δ 58 (br, s). Anal. Calcd
for 9d, C30H34B2F8NiP2S2: C, 47.85; H, 4.55. Found: C, 47.60; H, 4.12.
(2-(2-Naphthalene)thioethyl)ethyldiphenylphosphine (11).
2-(2-Naphthalene)ethyl bromide was dissolved in acetonitrile (20 mL)
and added to a solution 2-(diphenylphosphine)ethanethiol72 (1 equiv)
premixed with potassium tert-butoxide (1 equiv) in acetonitrile in a
glovebox for 1 h at room temperature. This suspension was stirred
overnight and then passed through a layer of Celite, and all volatile
residues were removed in vacuo. The resulting oil was subjected to a
chromatographic purification (Flash 40þM column, n-hexanes/dichlor-
omethane 8% f 66%, 1320 mL, flow rate 40 mL/min), which afforded 11
as a clear, analytically pure oil (3.4 g, 43%). 1H NMR (400 MHz,
b
spectra, a table of extinction coefficients, computational data, details
on X-ray experimental procedures, and CIF files. This material is
available free of charge via the Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION
Corresponding Author
chadnano@northwestern.edu
’ ACKNOWLEDGMENT
We thank Northwestern IMSERC for analytical instrumenta-
tion. M.R.J. acknowledges Northwestern University for a Ryan
Fellowship and the NSF for a Graduate Research Fellowship.
C.A.M. is grateful to NSF, ARO, and AFOSR-MURI for gener-
ous funding. The authors are grateful to Dr. Robert Kennedy for
useful discussions regarding the manuscript.
3032
dx.doi.org/10.1021/ja109624m |J. Am. Chem. Soc. 2011, 133, 3023–3033

Journal of the American Chemical Society
ARTICLE
’ REFERENCES
(1) Holliday, B. J.; Mirkin, C. A. Angew. Chem., Int. Ed. 2001, 40, 2022.
(2) Dietrich, B.; Guilhem, J.; Lehn, J.-M.; Pascard, C.; Sonveaux, E.
Helv. Chim. Acta 1984, 67, 91.
(3) Dietrich-Buchecker, C. O.; Sauvage, J. P.; Kern, J. M. J. Am.
Chem. Soc. 1984, 106, 3043.
(4) Jude, H.; Sinclair, D. J.; Das, N.; Sherburn, M. S.; Stang, P. J.
J. Org. Chem. 2006, 71, 4155.
(5) Johnson, D. W.; Raymond, K. N. Inorg. Chem. 2001, 40, 5157.
(6) Knight, L. K.; Freixa, Z.; van Leeuwen, P. W. N. M.; Reek, J. N. H.
Organometallics 2006, 25, 954.
(7) Liu, Y.; Flood, A. H.; Stoddart, J. F. J. Am. Chem. Soc. 2004, 126,
9150.
(8) McClure, B. A.; Rack, J. J. Angew. Chem., Int. Ed. 2009, 48, 8556.
(9) Swiegers, G. F.; Malefetse, T. J. Chem. Rev. 2000, 100, 3483.
(10) Carlucci, L.; Ciani, G.; Proserpio, D. M. Coord. Chem. Rev.
2003, 246, 247.
(11) Kay, E. R.; Leigh, D. A.; Zerbetto, F. Angew. Chem., Int. Ed.
2007, 46, 72.
(12) Farrell, J. R.; Mirkin, C. A.; Guzei, I. A.; Liable-Sands, L. M.;
Rheingold, A. L. Angew. Chem., Int. Ed. 1998, 37, 465.
(13) Dixon, F. M.; Eisenberg, A. H.; Farrell, J. R.; Mirkin, C. A.;
Liable-Sands, L. M.; Rheingold, A. L. Inorg. Chem. 2000, 39, 3432.
(14) Gianneschi, N. C.; Cho, S.-H.; Nguyen, S. T.; Mirkin, C. A.
Angew. Chem., Int. Ed. 2004, 43, 5503.
(15) Liu, X.; Stern, C. L.; Mirkin, C. A. Organometallics 2002, 21,
1017.
(16) Gianneschi, N. C.; Mirkin, C. A.; Zakharov, L. N.; Rheingold,
A. L. Inorg. Chem. 2002, 41, 5326.
(17) Eisenberg, A. H.; Ovchinnikov, M. V.; Mirkin, C. A. J. Am.
Chem. Soc. 2003, 125, 2836.
(18) Jeon, Y.-M.; Heo, J.; Brown, A. M.; Mirkin, C. A. Organome-
tallics 2006, 25, 2729.
(19) Slone, C. S.; Mirkin, C. A.; Yap, G. P. A.; Guzei, I. A.; Rheingold,
A. L. J. Am. Chem. Soc. 1997, 119, 10743.
(20) Khoshbin, M. S.; Ovchinnikov, M. V.; Mirkin, C. A.; Zakharov,
L. N.; Rheingold, A. L. Inorg. Chem. 2005, 44, 496.
(21) Wiester, M. J.; Ulmann, P. A.; Mirkin, C. A. Angew. Chem., Int.
Ed. 2010, 50 (1), 114–137.
(22) Kovbasyuk, L.; Kramer, R. Chem. Rev. 2004, 104, 3161.
(23) Takeuchi, M.; Ikeda, M.; Sugasaki, A.; Shinkai, S. Acc. Chem.
Res. 2001, 34, 865.
(24) Eisenberg, A. H.; Dixon, F. M.; Mirkin, C. A.; Stern, C. L.;
Incarvito, C. D.; Rheingold, A. L. Organometallics 2001, 20, 2052.
(25) Ulmann, P. A.; Brown, A. M.; Ovchinnikov, M. V.; Mirkin,
C. A.; DiPasquale, A. G.; Rheingold, A. L. Chem.-Eur. J. 2007, 13, 4529.
(26) Masar, M. S.; Mirkin, C. A.; Stern, C. L.; Zakharov, L. N.;
Rheingold, A. L. Inorg. Chem. 2004, 43, 4693.
(27) Brown, A. M.; Ovchinnikov, M. V.; Mirkin, C. A. Angew. Chem.
2005, 117, 4279.
(28) Yoon, H. J.; Mirkin, C. A. J. Am. Chem. Soc. 2008, 130, 11590.
(29) Yoon, H. J.; Kuwabara, J.; Kim, J.-H.; Mirkin, C. A. Science 2010,
330, 66.
(37) Holm, R. H.; Chakravorty, A.; Dudek, G. O. J. Am. Chem. Soc.
1963, 85, 821.
(38) Schumann, M.; Elias, H. Inorg. Chem. 1985, 24, 3187.
(39) Chakravorty, A.; Fennessey, J. P.; Holm, R. H. Inorg. Chem.
1965, 4, 26.
(40) James, T. L.; Smith, D. M.; Holm, R. H. Inorg. Chem. 1994, 33,
4869.
(41) Panda, R.; Zhang, Y.; McLauchlan, C. C.; Venkateswara Rao, P.;
Tiago de Oliveira, F. A.; M€unck, E.; Holm, R. H. J. Am. Chem. Soc. 2004,
126, 6448.
(42) James, T. L.; Cai, L.; Muetterties, M. C.; Holm, R. H. Inorg.
Chem. 1996, 35, 4148.
(43) Rigo, P.; Bressan, M.; Basato, M. Inorg. Chem. 1979, 18, 860.
(44) Dilworth, J. R.; Wheatley, N. Coord. Chem. Rev. 2000, 199, 89.
(45) Kim, J. S.; Reibenspies, J. H.; Darensbourg, M. Y. Inorg. Chim.
Acta 1996, 250, 283.
(46) Kim, J. S.; Reibenspies, J. H.; Darensbourg, M. Y. J. Am. Chem.
Soc. 1996, 118, 4115.
(47) Ge, P.; Riordan, C. G.; Yap, G. P. A.; Rheingold, A. L. Inorg.
Chem. 1996, 35, 5408.
(48) Jalil, M. A.; Fujinami, S.; Senda, H.; Nishikawa, H. J. Chem. Soc.,
Dalton Trans. 1999, 1655.
(49) Green, M. L. H. J. Organomet. Chem. 1995, 500, 127.
(50) Mizuno, J. J. Phys. Soc. Jpn. 1961, 16, 1574.
(51) Miedaner, A.; Haltiwanger, R. C.; DuBois, D. L. Inorg. Chem.
1991, 30, 417.
(52) Wiester, M. J.; Braunschweig, A. B.; Mirkin, C. A. Inorg. Chem.
2010, 49, 7188.
(53) Meier, P. F.; Merbach, A. E.; Dartiguenave, M.; Dartiguenave,
Y. Inorg. Chem. 1979, 18, 610.
(54) Basolo, F.; Johnson, R. C. Coordination Chemistry: The Chem-
istry of Metal Complexes; W. A. Benjamin, Inc.: New York, 1964.
(55) Nelson, S. M.; Shepherd, T. M. J. Chem. Soc. 1965, 3276.
(56) Ebsworth, E. A. V.; Rankin, D. W. H.; Cradock, S. Structural
Methods in Inorganic Chemistry; CRC Press: Boston, 1991.
(57) Hollas, J. M. Modern Spectroscopy; John Wiley & Sons, Ltd:
Hoboken, NJ, 2004.
(58) Huheey, J. E.; Keiter, E. A.; Keiter, R. L. Inorganic Chemistry:
Principles of Structure and Reactivity; Harper Collins College Publishers:
New York, 1993.
(59) Collman, J. P., Hegedus, L. S., Norton, J. T., Finke, R. G.
Principles and Applications of Organotranstion Metal Chemistry; Univer-
sity Science Books: Sausalito, CA, 1987.
(60) Miessler, G. L.; Tarr, D. A. Inorg. Chem.; 3rd ed.; Prentice Hall:
Upper Saddle River, NJ, 2004.
(61) Strauss, S. H. Chem. Rev. 1993, 93, 927.
(62) Reed, C. A. Acc. Chem. Res. 1998, 31, 133.
(63) Krossing, I.; Raabe, I. Angew. Chem., Int. Ed. 2004, 43, 2066.
(64) Macchioni, A. Chem. Rev. 2005, 105, 2039.
(65) Kim, D. W.; Hong, D. J.; Seo, J. W.; Kim, H. S.; Kim, H. K.;
Song, C. E.; Chi, D. Y. J. Org. Chem. 2004, 69, 3186.
(66) Becke, A. D. Phys. Rev. A 1988, 38, 3098.
(67) Perdew, J. P.; Yue, W. Phys. Rev. B 1986, 33, 8800.
(68) Cromer, D. T.; Waber, J. T. International Tables for X-ray
Crystallography; The Kynoch Press: Birmingham, England, 1974;
Vol. IV.
(69) Ibers, J. A.; Hamilton, W. C. Acta Crystallogr. 1964, 17, 781.
(70) Creagh, D. C.; McAuley, W. J. International Tables for Crystal-
lography; Kluwer Academic Publishers: Boston, 1992; Vol. C.
(71) Creagh, D. C.; Hubbell, J. H. International Tables for Crystal-
lography; Kluwer Academic Publishers: Boston, 1992; Vol. C.
(72) Chat, J.; Dilworth, J. R.; Schmultz, J. A.; Zubieta, J. A. J. Chem.
Soc., Dalton Trans. 1979, 1595.
(30) Gianneschi, N. C.; Nguyen, S. T.; Mirkin, C. A. J. Am. Chem.
Soc. 2005, 127, 1644.
(31) Koga, T.; Furutachi, H.; Nakamura, T.; Fukita, N.; Ohba, M.;
_
Takahashi, K.; O kawa, H. Inorg. Chem. 1998, 37, 989.
(32) Colpas, G. J.; Maroney, M. J.; Bagyinka, C.; Kumar, M.; Willis,
W. S.; Suib, S. L.; Baidya, N.; Mascharak, P. K. Inorg. Chem. 1991, 30,
920.
(33) Li, Y.; Zamble, D. B. Chem. Rev. 2009, 109, 4617.
(34) Barondeau, D. P.; Kassmann, C. J.; Bruns, C. K.; Tainer, J. A.;
Getzoff, E. D. Biochemistry 2004, 43, 8038.
(35) Trofimenko, S.; Calabrese, J. C.; Kochi, J. K.; Wolowiec, S.;
Hulsbergen, F. B.; Reedijk, J. Inorg. Chem. 1992, 31, 3943.
(36) Baidya, N.; Olmstead, M.; Mascharak, P. K. Inorg. Chem. 1991,
30, 929.
3033
dx.doi.org/10.1021/ja109624m |J. Am. Chem. Soc. 2011, 133, 3023–3033