Monometallic Ni0 and Heterobimetallic Ni0/AuI Complexes of Tripodal Phosphine Ligands: Characterization in Solution and in the Solid State and Catalysis

Article abstract of DOI:10.1002/chem.201500187

The tridentate chelate nickel complexes [(CO)Ni{(PPh₂CH₂)₃CMe}] (2), [(CO)Ni{(PPh₂CH₂CH₂)₃SiMe}] (6), and [Ph₃PNi{(PPh₂CH₂CH₂)₃SiMe}] (7), as well as the bidentate complex [(CO)₂Ni{(PPh₂CH₂)₂CMeCH₂PPh₂}] (3) and the heterobimetallic complex [(CO)₂Ni{(PPh₂CH₂)₂CMeCH₂Ph₂PAuCl}] (4), have been synthesized and fully characterized in solution. All ¹H and ¹³C NMR signal assignments are based on 2D-NMR methods. Single crystal X-ray structures have been obtained for all complexes. Their ³¹P CP/MAS (cross polarization with magic angle spinning) NMR spectra have been recorded and the isotropic lines identified. The signals were assigned with the help of their chemical shift anisotropy (CSA) data. All complexes have been tested regarding their catalytic activity for the cyclotrimerization of phenylacetylene. Whereas complexes 2-4 display low catalytic activity, complex 7 leads to quantitative conversion of the substrate within four hours and is highly selective throughout the catalytic reaction.

Full text of DOI:10.1002/chem.201500187

DOI: 10.1002/chem.201500187
Full Paper
&
Nickel Complexes
Monometallic Ni0 and Heterobimetallic Ni0/AuI Complexes of
Tripodal Phosphine Ligands: Characterization in Solution and in
the Solid State and Catalysis
Kyle J. Cluff, Nattamai Bhuvanesh, and Janet Blümel*[a]
Abstract: The tridentate chelate nickel complexes [(CO)Ni-
{(PPh2CH2)3CMe}] (2), [(CO)Ni{(PPh2CH2CH2)3SiMe}] (6), and
[Ph3PNi{(PPh2CH2CH2)3SiMe}] (7), as well as the bidentate
complex [(CO)2Ni{(PPh2CH2)2CMeCH2PPh2}] (3) and the het-
erobimetallic complex [(CO)2Ni{(PPh2CH2)2CMeCH2Ph2PAuCl}]
(4), have been synthesized and fully characterized in solu-
tion. All 1H and 13C NMR signal assignments are based on
2D-NMR methods. Single crystal X-ray structures have been
obtained for all complexes. Their 31P CP/MAS (cross polariza-
tion with magic angle spinning) NMR spectra have been re-
corded and the isotropic lines identified. The signals were
assigned with the help of their chemical shift anisotropy
(CSA) data. All complexes have been tested regarding their
catalytic activity for the cyclotrimerization of phenylacety-
lene. Whereas complexes 2–4 display low catalytic activity,
complex 7 leads to quantitative conversion of the substrate
within four hours and is highly selective throughout the cat-
alytic reaction.
Introduction
catalyst are needed; c) no gases are involved, as in hydrogena-
tion reactions,[3] and therefore no gas diffusion effects will blur
the experimental results; d) there are two isomers as reaction
products, which allow for the simultaneous determination of
the activity and selectivity; e) the cyclotrimerization reaction
proceeds under ambient pressure and at moderate tempera-
tures, and there is already a solid basis of data for compari-
son.[5]
The general research interest of our group is to immobilize ho-
mogeneous catalysts on insoluble oxide supports, so that the
catalyst can easily be separated from the reaction mixtures
and recycled.[1,2] The primary remaining problem with both ho-
mogeneous and immobilized catalysts is ligand detachment,
which leads to leaching and nanoparticle formation. Over the
years, realizing that the linkers are crucial for the success of im-
mobilized catalysts, many systems with refined linkers have
been obtained, which have shown record activities and life-
times. These species include rhodium hydrogenation cata-
lysts,[3] Pd0/CuI Sonogashira systems for CÀC cross-coupling re-
actions,[4] and nickel catalysts for acetylene cyclotrimerization[5]
and hydrophosphonation.[6]
We have previously demonstrated that one general feature
of immobilized catalysts is the potential formation of nanopar-
ticles, which can also lead to an active catalyst, as is the case
with rhodium catalysts for hydrogenation.[3a,c] However, in the
case of nickel catalysts, nanoparticle formation may lead to
less active catalysts, and a nickel mirror at the wall of a Schlenk
flask lacks catalytic activity.[5d] Therefore, it is important to use
a favorable ligand that shields the metal center from other
complexes, thus impeding nanoparticle formation, but still al-
lowing substrate access. With respect to later immobilization it
will also become important to shield the metal center from
a reactive oxide support surface, which can be accomplished
with the right ligands.
Nickel catalysts in general merit exploration because they
are less expensive than rhodium or palladium catalysts,
a factor that is important for large-scale experiments and for
industrial applications.[6] In order to blaze a trail for the next
generation of immobilized nickel complexes with a new type
of linker, molecular analogues are studied in a first step.
The nickel complexes described herein are tested for their
catalytic activity with respect to the cyclotrimerization of phe-
nylacetylene. This reaction is important,[7] for example, in natu-
ral product syntheses,[7d] and it offers many advantages, includ-
ing the following: a) The reaction does not require a change of
the metal oxidation state, in contrast, for example, to the So-
nogashira reaction;[4] b) only one substrate and no base or co-
Immobilized catalysts that are tethered to oxide supports via
bidentate[4a,5a] and tripodal chelating phosphine ligands[3a,c,8]
have shown superior recycling characteristics as compared to
analogous catalysts bound by monodentate ligands. The reten-
tion of the metal center is improved by the increased coordi-
nating strength, as we recently demonstrated by using in situ
high-resolution magic angle spinning (HRMAS) NMR spectros-
copy.[4a,c] Furthermore, the formation of double layers of link-
ers,[4a,5c] which enhances metal leaching, cannot occur with tri-
podal phosphine ligands with only one ethoxysilane group.
Although immobilizable versions of tripodal ligands incorpo-
rating ethoxysilane groups[3b,8] or phosphonium salts[3a] can
[a] K. J. Cluff, Dr. N. Bhuvanesh, Prof. J. Blümel
Department of Chemistry, Texas A&M University
College Station, TX, 77842-3012 (USA)
E-mail: bluemel@tamu.edu
Chem. Eur. J. 2015, 21, 10138 – 10148
10138
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
easily be synthesized, for the sake of solubility and ease of
handling without having to prevent cross-linking, the original
tripodal phosphine ligands 1 and 5 (Scheme 1 and Scheme 2)
with methyl groups and carbon or silicon as center atoms
have been used here. Ligand 5 was chosen because it is closer
electronically to the ethoxysilane versions to be considered for
later immobilization.[3b,8]
their characteristic CSA values.[22] Therefore, one of our goals in
this contribution is to enlarge the CSA database of nickel phos-
phine complexes to provide a foundation for further investiga-
tion of these catalysts.
Herein we seek to identify a correlation between the X-ray
structural, the solid-state NMR, and catalysis results. This com-
bined approach should provide a better insight into the factors
that govern acetylene cyclotrimerization catalysis with nickel
complexes.
In addition to their potential application as linkers for immo-
bilized catalysts, tripodal phosphine ligands are of fundamental
interest in coordination chemistry, due to their potential for dy-
namic and fluxional ligand scenarios.[9] Furthermore, they
might allow the selective synthesis of heterobimetallic com-
plexes[10] that are not only of phenomenological but also prac-
tical importance because they might display synergistic effects
in catalysis. Heterobimetallic complexes also have the potential
to provide metal atom assemblies as preconditioned systems
for the formation of nanoparticles with a well-defined stoichi-
ometry of metal atoms.
Results and Discussion
The presented results will be outlined according to the follow-
ing. The starting point is the synthesis of the compounds, fol-
lowed by the characterization in solution. Then the discussion
of the X-ray structures in combination with the solid-state
NMR results follows. Finally the complexes will be investigated
regarding their catalytic activity.
All complexes in this work have been characterized by solid-
state NMR spectroscopy.[11,12] This method grows in importance
for the characterization not only of materials like metal–organ-
ic frameworks (MOFs), porous organic frameworks (POFs),[13]
and polymers,[14] but it is increasingly applied to insoluble mo-
lecular metal complexes.[15–17] In this field, solid-state NMR is
particularly important, because complexes and ligands might
not crystallize readily, and only in some cases are the polycrys-
talline materials of high enough quality for powder diffrac-
tion.[18] Naturally, solid-state NMR spectroscopy is indispensable
for investigating immobilized catalysts on amorphous supports
such as oxides, which do not allow powder diffraction.[3c] Fur-
thermore, dynamic effects that otherwise complicate the analy-
sis of coordination compounds can be conveniently studied
when “frozen” in the solid state.[17]
Syntheses of the nickel complexes
Nickel complex 2 was obtained as a yellow powder in pure
form and in high yield from the reaction of the tripodal phos-
phine ligand 1 with [Ni(CO)2(PPh3)2] in toluene at elevated tem-
peratures (Scheme 1). Alternatively, heating complex 3 to 908C
for a few hours leads to the quantitative conversion of 3 to 2
(Scheme 1).
An additional option for generating 2 in a straightforward
and quantitative manner is to melt complex 3 and thus pro-
voke the loss of one CO group. Complex 3, in turn, can be ob-
tained selectively and in high yield from the reaction of ligand
1 with [Ni(CO)2(PPh3)2] at ambient temperatures under exclu-
sion of light (Scheme 1). Besides the NMR analysis (see below),
the presence of one uncoordinated phosphine group in 3 can
be proven by binding it to a gold center (Scheme 1). The color-
less heterobimetallic Ni0/AuI complex 4 is obtained in a good
overall yield of 58%. It has a shelf life of several months at am-
The spectra presented herein were obtained by using mag-
netization transfer from protons to 31P nuclei (cross polariza-
tion, CP)[11,12] to boost the signal intensities. Magic angle spin-
ning (MAS) was applied to reduce anisotropic line-broadening
effects in the solid state.[11]
Besides providing chemical shift information, in favorable
cases of clean and polycrystalline materials, solid-state NMR
spectra result in a wealth of chemical shift anisotropy (CSA)[12]
data. CSA values provide valuable insight about the symmetry
of the electronic surroundings of a nucleus, which is of general
interest.[16] Most importantly, dynamic processes in molecules[17]
and mobilities of adsorbed species on surfaces can be identi-
fied and quantified by a reduced or fully eliminated CSA. We
could, for example, distinguish polycrystalline from silica-ad-
sorbed metallocenes[19] by their vastly different CSA values,
and the interactions of phosphine oxides[20] with silica surfaces
and their dynamic behavior could be studied through their
CSA data. Finally, from a practical perspective, the CSA is char-
acteristic for each species and thus allows the identification of
functional groups in a material or compound. For example, our
group has used CSA values to distinguish surface-bound phos-
phine oxides from phosphonium salts.[21] More recently H2O2
adducts of phosphine oxides could be described and distin-
guished from the corresponding pure phosphine oxides by
Scheme 1. Syntheses of the monometallic Ni0 complexes 2 and 3 and of the
heterobimetallic Ni0/AuI complex 4. tht=tetrahydrothiophene.
Chem. Eur. J. 2015, 21, 10138 – 10148
www.chemeurj.org
10139
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
bient temperatures under an inert atmosphere, which is sur-
prising, taking into account that the less noble Ni in an oxida-
tion state of 0 is neighboring an intramolecular AuI center. De-
composition only occurs upon heating 4 up to its melting
point. This heterobimetallic Ni0/AuI combination within one
molecule is, to our knowledge, unprecedented, although
a wealth of NiII/AuI species have been described.
and 3.[2,25] These are expected, as we have found and de-
scribed virtual couplings in tripodal complexes earlier.[2,8]
The assignments of the 31P NMR signals in the spectra of 3,
4, and 7 in solution are straightforward due to the signal inten-
1
sities and J(31P,31P) couplings. All H and 13C NMR signal assign-
ments were supported by 2-dimensional NMR methods. For
1
example, the H–1H NOESY[26] spectrum of 3 in C6D6 proves the
The tripodal phosphine ligand 5 was synthesized by hydro-
phosphination using azobisisobutyronitrile (AIBN), as described
previously for analogous compounds incorporating ethoxysi-
lane groups.[3b] The nickel complex 6 (Scheme 2) cannot be ob-
signal assignments for the syn/anti protons in position 3
(Figure 1, solid circle). Only the 3anti proton, which points in the
same direction as the methyl protons in position 1, is close
enough spatially to give a NOESY cross peak. The 3syn proton
trans to the methyl group is too far away to give a cross peak
(Figure 1, broken line circle).
Scheme 2. Structures of the ligand 5 and the Ni0 complexes 6 and 7.
tained from the nickel dicarbonyl precursor in an analogous
manner to the synthesis of 2 because a multitude of insepara-
ble oligomeric byproducts result from this approach. However,
it could be obtained through ligand exchange by the reaction
of
5 (Scheme 2) with the monocarbonyl complex [Ni-
(CO)(PPh3)3] cleanly and in a yield of 25%. The synthesis of [Ni-
(CO)(PPh3)3] according to literature procedures proved to be
nontrivial.[23] Therefore, we developed a more straightforward
synthesis for this complex by making use of a molten mixture
of the components [Ni(CO)2(PPh3)2] and PPh3 in the absence of
any solvent. This procedure allowed us to obtain [Ni-
(CO)(PPh3)3] quickly and with a good yield of 62%.
Figure 1. 1H–1H NOESY NMR spectrum of 3 in C6D6.
Complex 7 (Scheme 2) could be synthesized by reaction of
the NiII pyridine complex NiCl2py4 with ligand 5 while offering
the corresponding amount of PPh3 under reducing conditions
achieved by adding Zn powder. Hereby, the yield of orange-
red 7 was excellent. Additionally, compound 7 is perfectly shelf
stable at ambient temperatures over months if kept under an
inert atmosphere.
Characterization of the nickel complexes in the solid state
As expected, the IR spectra of the monocarbonyl nickel com-
plexes 2 and 6 show one CO stretching band each in the
wavenumber region corresponding to terminal carbonyl
groups. The dicarbonyl complexes 3 and 4 display two CO
bands each within this region.
Characterization of the nickel complexes in solution
Due to the symmetries of compounds 1–7, the numbers of
NMR signals are small and their signal assignments are
straightforward, using signal intensities and coupling patterns.
The chemical shifts and coupling constants are in the anticipat-
ed ranges for nickel complexes with bidentate[5,15,24] and tri-
podal phosphine ligands.[3a,8] As in the earlier cases, there is no
indication of dynamic processes involving dissociation and re-
coordination of phosphine groups. Special attention has been
given to complex 3 in this respect, but it was found to be
static and stable also in solution. Regarding the characteriza-
tion of the dissolved complexes, the most interesting feature is
the occurrence of virtual couplings in the 13C NMR spectra of 2
Next, the single crystal X-ray analyses of all nickel complexes
were evaluated with respect to their corresponding 31P solid-
state NMR spectra. The single crystal X-ray structure and unit
cell of 2[27] are shown in Figure 2 and Figure 3, respectively.
The 31P CP/MAS spectra of 2 are displayed in Figure 4. The
structure shows that the nickel center is tetrahedrally coordi-
nated by the three phosphines of the tripodal ligand and one
CO group. The NiÀP bond lengths are 2.182, 2.181, and
2.186 Š, and seem to be mainly dominated by the steric de-
mands of the ligand. The same accounts for the P-Ni-P angles
which are smaller than the expected tetrahedral angle and
amount to 92.45, 93.91, and 95.958.[27]
Chem. Eur. J. 2015, 21, 10138 – 10148
www.chemeurj.org
10140
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
three 31P NMR signals in the solid-state NMR spectrum. Two
phosphine groups have their phenyl substituents pointing
downward, whereas one phosphine has them point upward.
Therefore, two of the three signals should have similar CSA,
whereas that of the third should differ.
Indeed, the solid-state NMR spectra of polycrystalline 2
(Figure 4) can be interpreted in a straightforward manner. Re-
Figure 2. Single crystal X-ray structure of 2.[27] Thermal ellipsoids are shown
at the 50% probability level. Hydrogen atoms are omitted for clarity.
As described above, in solution complex 2 displays only one
31P NMR resonance at 11.54 ppm, due to the symmetry of the
molecule. However, in the solid state even this rather spherical-
ly shaped complex cannot tumble, but has to assume a fixed
position within the lattice. Therefore, the number of 31P solid-
state NMR signals is determined by the orientation of the mol-
ecules within the unit cell.[11] The maximal number of signals
equals the number of magnetically inequivalent 31P nuclei in
the unit cell. In the case of similar orientations of functional
groups with respect to the external magnetic field, the corre-
sponding lines might overlap, thus reducing the overall
number of signals. The CSA parameters[11,12] should reflect the
orientation of the functional groups in the unit cell with re-
spect to the external magnetic field. The CSA values are more
reliable criteria for making signal assignments in solid-state
NMR spectra, because the isotropic chemical shifts can differ
substantially from the solution chemical shift values and vary
widely, as demonstrated earlier for metal complexes with che-
late ligands.[8]
Figure 4. 31P CP/MAS NMR spectra of polycrystalline 2. Vertical arrows indi-
cate the isotropic lines.
cording the spectra at different rotational speeds of 4 and
7 kHz under optimized Hartmann–Hahn conditions[15] reveals
the isotropic lines that do not migrate but retain the same
chemical shifts. Three lines can be distinguished and, after de-
convolution of the two overlapping isotropic lines and simula-
tion,[28] the CSA data for 2 are obtained.
The CSA data of the polycrystalline complexes 2–4 and 7 are
listed in Table 1. We suggest that the isotropic line at 8.1 ppm
corresponds to the phosphine with both its phenyl groups
pointing upward in all molecules of the unit cell (Figure 3), be-
cause its CSA of 179.5 ppm deviates by more than 20 ppm
from the values of the resonance at 13.5 ppm. The difference
of the CSA values between the signals at 13.5 and 7.1 ppm is
The unit cell of 2 contains four molecules (Figure 3).[27]
Hereby, assemblies of two molecules each point in the same
direction. Looking along the vertical axis of the unit cell as dis-
played in Figure 2, one can determine that there should be
Table 1. 31P CP/MAS chemical shift anisotropy NMR data [ppm] for the
polycrystalline complexes 2–4 and 7.[28]
Complex[a]
diso
13.5
8.1
7.1
d11
d22
25.6
16.0
19.3
d33
CSA[b]
2
107.4
93.9
96.9
À92.3
À85.6
À94.8
199.7
179.5
191.7
3
4
7
16.0
9.6
À31.9
106.4
86.5
17.8
0.9
17.2
À47.3
À59.4
À74.8
À66.1
165.8
161.4
83.9
19.4
14.1
8.6
93.2
71.3
74.2
8.0
29.8
20.6
À42.9
À58.7
À69.0
143.2
130.0
143.2
23.6
10.3
73.8
79.2
68.7
34.6
À71.5
À83.0
145.3
162.3
[a] For structures of complexes, see Schemes 1 and 2; [b] the span of the
wideline signals (d11–d33) is listed according to the convention described
by Duncan.[12]
Figure 3. Unit cell of 2.[27] Thermal ellipsoids are shown at the 50% probabili-
ty level. Hydrogen atoms and phenyl rings, except for the ipso carbons, are
omitted for clarity.
Chem. Eur. J. 2015, 21, 10138 – 10148
www.chemeurj.org
10141
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 5. Single crystal X-ray structure of 3.[29] Thermal ellipsoids are shown
at the 50% probability level. Hydrogen atoms are omitted for clarity.
Figure 7. 31P CP/MAS NMR spectra of polycrystalline 3. Vertical arrows indi-
cate the isotropic lines.
only 8 ppm. Therefore, we assume that these two signals
belong to the phosphine groups of 2, which each have one
phenyl group pointing up and one down in the unit cell dis-
played in Figure 3.
chemical shifts of 9.6 and 16.0 ppm and the CSA values of
165.8 and 161.4 ppm for the other two phosphine resonances
are very similar, and therefore, the resonances of the coordi-
nated phosphines cannot be assigned. This is not surprising
given the similar orientation of the coordinated phosphine
groups as seen in the unit cell of 3, for example, when looking
along the vertical axis in the display (Figure 6).
The X-ray structure of 3 shows that only two phosphine
groups of the tripodal ligand are coordinated to the nickel
center. The third phosphine group remains uncoordinated
(Figure 5), although the orientation of its lone pair seems to
predestine it to replace one of the two CO groups. The relative
stability of 3 might be due to steric constraints of the tripodal
ligand and thus hindrance of the attack of the phosphine at
the metal center by the syn-positioned CO group. Furthermore,
the chair conformation of the six-membered ring, which is
best seen in the unit cell display (Figure 6), incorporating the
nickel atom has to change into a less favorable boat conforma-
tion to coordinate the third phosphine group.
Interestingly, the X-ray structure of the heterobimetallic
complex 4[30] differs little from that of 3, except for the addi-
tional presence of the AuCl moiety (Figure 8). The unit cell[30]
(Figure 9) shows a similar chair conformation of the six-mem-
bered ring incorporating the nickel atom. The AuCl moiety is
pointed towards the nickel center, with a Ni···Au distance of
7.085 Š. This heterobimetallic complex is rather stable. The
The interpretation of the 31P CP/MAS spectra of 3 (Figure 7)
is rather straightforward. The resonance at À31.9 ppm corre-
sponds to the uncoordinated phosphine group, because its
chemical shift is closest to that found for 3 in solution
(À27.79 ppm). Even more indicative, however, is its small CSA
value of 83.9 ppm (Table 1). This is in accordance with the fact
that the CSA values of phosphines are much smaller than
those of their metal-coordinated analogs.[2,4b,12,15] Both the
Figure 8. Single crystal X-ray structure of 4.[30] Thermal ellipsoids are shown
at the 50% probability level. Hydrogen atoms are omitted for clarity.
Figure 6. Unit cell of 3.[29] Thermal ellipsoids are shown at the 50% probabili-
ty level. Hydrogen atoms and phenyl rings, except for the ipso carbons, are
omitted for clarity.
Figure 9. Portion of the unit cell of 4.[30] Thermal ellipsoids are shown at the
50% probability level. Hydrogen atoms, solvent molecules, and phenyl rings,
except for the ipso carbons, are omitted for clarity.
Chem. Eur. J. 2015, 21, 10138 – 10148
www.chemeurj.org
10142
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
steric constraints discussed above for 3 probably also play
a major role in 4.
The packing of complex 4 in the unit cell (Figure 9) again re-
sembles that of 3 (Figure 6). Comparing the 31P CP/MAS spec-
tra of polycrystalline 3 and 4, the expectation is that the reso-
nance for the uncoordinated phosphine group vanishes and
a new signal appears. The signals of the phosphines coordinat-
ed to the nickel center should display similar chemical shifts
and CSA values.
At first sight, any signal assignment in the 31P CP/MAS spec-
tra of polycrystalline 4 (Figure 10) seems impossible. However,
Figure 11. Single crystal X-ray structure of 7.[32] Thermal ellipsoids are shown
at the 50% probability level. Hydrogen atoms are omitted for clarity.
the coordinated tripodal phosphine ligand. This steric factor
might indicate a facilitated release of the PPh3 ligand and thus
explain the high catalytic activity of 7, as discussed below.
Regarding the 31P CP/MAS spectra of polycrystalline
7
(Figure 12), the signal assignment is again straightforward. The
Figure 10. 31P CP/MAS NMR spectra of polycrystalline 4. Vertical arrows indi-
cate the isotropic lines.
two of the resonances have very similar chemical shifts, 19.4
and 8.6 ppm (Table 1), to those for the nickel-coordinated
phosphine groups in 3 (16.0, 9.6 ppm). Furthermore, the
CSA values are the same (143.2 ppm) for both resonances.
The middle signal at 14.1 ppm has a distinct CSA of only
130.0 ppm. Therefore, we attribute this signal to the phosphine
group coordinated to the gold atom in 4.
Figure 12. 31P CP/MAS NMR spectra of polycrystalline 7. Arrows indicate the
isotropic lines.
Complex 6 has been crystallized from different solvents. The
best single crystal X-ray structure of 6[31] could be solved and it
proves the structure in Scheme 2. However, it shows twin crys-
tals and lower quality than the other structures shown herein.
Consequently, polycrystalline 6 was not a favorable candidate
for solid-state NMR analysis.
resonances at 23.6 and 10.3 ppm correspond to the phosphine
groups of coordinated PPh3 and tripodal ligand, respectively.
This follows from the comparison with the chemical shifts in
solution (22.50 and 12.10 ppm) and the signal intensities of
about 1:3.5, including the signal intensities of the rotational
sidebands. Although, in principle, solid-state NMR signals
cannot be integrated when CP is used,[11,15] in the case of 7 it
is obvious that there are more alkyl protons close to the phos-
phine nuclei in the tripodal ligand than in PPh3. Since alkyl pro-
tons facilitate the transfer of magnetization to the phosphorus
nuclei, the found ratio of the signal intensities is not exactly
1:3.
Fortunately, complex 7 crystallizes readily and the single
crystal analysis resulted in a structure of high quality[32]
(Figure 11). It shows that the nickel center is coordinated tetra-
hedrally by the phosphine atoms, with the P-Ni-P angles of the
tripodal ligand assuming the values 107.13, 112.93, and
109.358. The Ph3P-Ni-P angles amount to 111.37, 107.42, and
108.468. Hereby, the phenyl groups of the tripodal ligand
shield the metal center from interacting with other nickel com-
plexes, which might inhibit the formation of nickel nanoparti-
cles and premature deactivation of the catalyst, as discussed
below. The steric demand of the phenyl groups of the tripodal
ligand also interferes with the space occupied by the PPh3
ligand, which might be the reason for the slightly elongated
NiÀPPh3 bond (2.234 Š) in the molecule, as compared to 2.212,
2.207, and 2.217 Š for the other three NiÀP bond lengths of
As in the cases discussed above, the CSA can be used to ad-
ditionally corroborate the signal assignment. The electronic
surroundings of the 31P nucleus in the PPh3 ligand is expected
to be more symmetric than in the coordinated alkyldiphenyl-
phosphine groups of the tripodal ligand.[21b] Therefore, the CSA
value of the PPh3 signal, 145.3 ppm, is much smaller than that
of the tripodal phosphine signal (162.3 ppm, Table 1).
Chem. Eur. J. 2015, 21, 10138 – 10148
www.chemeurj.org
10143
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Catalysis
site needed for the cyclotrimerization to proceed at the metal
center (see below), thus enhancing the catalytic activity.
To test the catalytic activity of the nickel complexes shown in
Schemes 1 and 2, the cyclotrimerization of phenylacetylene
was employed (Scheme 3 and Experimental Section). As de-
scribed above, the results can readily be compared with
a large database of earlier research.[5]
Indeed, nickel complex 7 leads to nearly quantitative conver-
sion of phenylacetylene into the cyclotrimers within only
about four hours. This activity compares very favorably to ear-
lier results, as the highest catalytic activity of a molecular
nickel complex reported to date corresponds to a quantitative
conversion of the substrate at 908C within 4.5 h.[5a] We assume
that this high activity, which does not coincide with the forma-
tion of a typical nickel mirror at the glass wall of the Schlenk
tube, is based on the precarious position of the PPh3 ligand in
7. As the X-ray structure[32] (Figure 11) shows, the phenyl
groups of the tripodal phosphine ligand are oriented towards
the PPh3 ligand, thus presumably facilitating its departure from
the metal center and in this way allowing the substrate to co-
ordinate. To test this hypothesis, the nickel complex
[{CH3C(CH2PPh2)3}NiPPh3] (8)[33] was applied to the catalytic re-
action under identical conditions (Figure 13). As anticipated, 8
is more active than the carbonyl nickel complexes 2–4. Howev-
er, it does not quite reach the catalytic activity of 7. In addition
to the electronic effects mentioned above, the slightly in-
creased lengths of the alkyl chains of the tripodal ligand in 7
to (CH2)2 might facilitate further detachment of one or more
phosphine groups from the metal center. In order to proceed
with the cyclotrimerization reaction, at least two coordination
sites have to be liberated at the metal center. However, the
ligand detachment seems to be reversible enough to prevent
the formation of nickel nanoparticles or a nickel mirror.
Scheme 3. Nickel-catalyzed cyclotrimerization of phenylacetylene to give the
isomers 1,2,4-triphenylbenzene and 1,3,5-triphenylbenzene.
Interestingly, all tripodal phosphine ligands with short alkyl
chain lengths [(CH2)1], namely 2, 3, and 4, displayed only mini-
mal catalytic activities, even at the elevated temperature of
658C (Figure 13). One should also note that the presence of
AuI in 4 alongside the Ni0 center did not lead to a synergistic
effect, neither enhancing, nor decreasing the catalytic activity
of this heterobimetallic complex.
The high selectivity of 7 regarding the two possible prod-
ucts (Scheme 3) is unprecedented. Nickel catalysts with mono-
and bidentate phosphine ligands studied earlier became unse-
lective during the catalytic run,[5a] producing equal amounts of
the two isomers. Catalyst 7, however, reaches a constant ratio
of 1,2,4- to 1,3,5-triphenylbenzene of 16:1 after about 15 min,
with an initial product ratio of already 11:1. In comparison, cat-
alyst 8 requires about 30 min to reach a constant ratio of 15:1
with an initial product ratio of 9:1. This scenario shows that,
for 7, the active species is formed most quickly, for example,
by releasing the PPh3 ligand, and that the selectivity is then
pushed further towards the product that requires a less steri-
cally demanding transition state.
Figure 13. Conversion of phenylacetylene into 1,2,4-triphenylbenzene and
1,3,5-triphenylbenzene at 658C in cyclooctane catalyzed by the selected
nickel complexes 2–4, 7, and 8.
Putting these results into perspective and comparing the ad-
vantages of 7 with those of previously reported catalyst sys-
tems is not quite straightforward because each catalyst is usu-
ally optimized towards a specific goal. Regarding the selectivity
of the unsymmetric versus the symmetric cyclotrimerization
products, a [CoBr2(diimine)] complex in combination with Zn
and ZnI2 gave the highest ratio of 97:3.[35] A ruthenium com-
plex was also found to cyclotrimerize phenylacetylene, al-
though 4 mol% of the catalyst had to be applied and the reac-
tion took five hours for completion.[36] A very versatile rhodium
complex that is efficient for a variety of terminal and internal
acetylenes[37] needs, however, 40 h to produce the cyclotrimers
of phenylacetylene with a yield of only 68% in a 1:3 ratio. The
most selective catalyst systems for producing the symmetric
versus the unsymmetric cyclotrimerization product are based
on niobium and tantalum.[38] Isomer ratios of >99:1 have been
The lower catalytic activities of metal complexes coordinated
by ligands with short alkyl chains follows a trend observed ear-
lier for immobilized rhodium hydrogenation catalysts of the
Wilkinson type.[3a] At that time the low catalytic activity was at-
tributed to increased decomposition of the catalysts on the re-
active silica surface, assuming that the short linker chains do
not provide sufficient distance to the support surface. The new
results suggest, however, that the coordination of the tripodal
ligands might be too strong to allow access of the substrate to
the metal center. This general scenario might also explain the
low catalytic activity of 2–4. Comparing the IR band for n(CO)
of 2 (1886.5 cmÀ1) with that of 6 (1911.5 cmÀ1) corroborates
the assumption that the longer alkyl chains of the tripodal
ligand 5 render it a poorer electron donor than 1. Therefore,
ligand 5 can more easily liberate the additional coordination
Chem. Eur. J. 2015, 21, 10138 – 10148
www.chemeurj.org
10144
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
a 4 mm multinuclear MAS probehead and ZrO2 rotors. For the
obtained, but the overall yield was in general lower than 20%.
One recent report on the catalytic cyclotrimerization of phenyl-
acetylene describes the activity of a NiCl2/Zn system in the
presence of modified pyridines[39] using 20 mol% of catalyst to
obtain a yield of 94%. In comparison, only 0.5 mol% of 7 is
used and quantitative substrate conversion is achieved within
four hours.
1
31P CP/MAS measurements H high-power decoupling was applied
with a contact pulse of 5 ms. NH4H2PO4 was used to establish the
Hartmann–Hahn matching condition and as an external chemical
shift standard [d(31P)= +0.81 ppm]. The recycle delay was 3 s for
all CP/MAS spectra. The IR spectra of the neat powders were re-
corded on a Shimadzu IRAffinity-1 FTIR instrument using a Pike
Technologies MIRacle ATR plate. All reactions were carried out
using standard Schlenk techniques and a purified N2 atmosphere,
if not stated otherwise. Reagents purchased from Sigma Aldrich or
VWR were used without further purification. Tripodal ligand 1 was
obtained from Strem Chemicals and used as received. Solvents
were dried by boiling them over sodium, then they were distilled
and stored under purified nitrogen. Ligand 5 was synthesized ac-
cording to the procedure described previously.[3b] The 31P and
13C NMR data match those in the literature.[34]
In summary, regarding the complexes presented herein, the
most promising for future studies on homogeneous and im-
mobilized nickel catalyst systems are complexes with tripodal
phosphine ligands of medium length, as realized in 7, which
additionally incorporate a ligand that can be easily detached
from the metal center.
Nickel complex 8 has been obtained by combining [Ni(cod)2] with
the tripodal ligand 1 and PPh3 at ambient temperature in toluene.
The NMR data of 8 are in accordance with those given in the litera-
ture.[33]
Conclusion
In this work, new monometallic Ni0 complexes and one hetero-
bimetallic Ni0/AuI complex with tripodal phosphine ligands
have been described. They have been characterized thorough-
ly in solution and in the solid state. The single crystal X-ray
structures of all complexes have been obtained and discussed
in connection with their 31P CP/MAS spectra. The CSA values of
the complexes have been analyzed and all solid-state NMR sig-
nals could be assigned. One nickel complex displayed unprece-
dented activity and selectivity with respect to the cyclotrimeri-
zation of phenylacetylene, which could be interpreted based
on the structural features of the complex. The characteristics
of this nickel catalyst compare favorably with those of other
catalyst systems described in the literature.
Representative standardized catalytic procedure using complex 7:
Naphthalene (102.5 mg, 0.7997 mmol) and catalyst 7 (14.0 mg,
0.0140 mmol, 0.5 mol%) were stirred in cyclooctane (5 mL). The
mixture was heated to 658C and phenylacetylene (0.3 mL,
2.732 mmol) was added via syringe. Single drop aliquots were
taken periodically and filtered through a one inch thick layer of
silica in a Pasteur pipette and washed with toluene into a GC vial.
GC analyses of the catalysis products were obtained with a Shimad-
zu GC 2010 gas chromatograph equipped with a SHRXI-5 MS
column (15 m”0.25 mm”0.25 mm) and a flame ionization detector
(GC:FID). Naphthalene served as the intensity standard for the GC
signals.
Complex 2: [Ni(CO)2(PPh3)2] (213.8 mg, 0.3344 mmol) and Triphos
(208.0 mg, 0.3330 mmol) were dissolved in toluene (50 mL) and
stirred at 908C for 3.5 h. The solvent was removed from the yellow
solution under reduced pressure until solid began to precipitate.
Dry pentane (20 mL) was added. The resulting yellow precipitate
was filtered, washed with pentane (4”20 mL) and dried under
vacuum at 708C to afford 2 (175.5 mg, 0.2467 mmol, 74% yield
with respect to [Ni(CO)2(PPh3)2]). Single crystals suitable for X-ray
diffraction were obtained from a saturated toluene solution of 2 in
the course of a few days. The compound was synthesized previ-
ously starting from [Ni(CO)4],[40] but not characterized with more
recent analytical techniques. The complete data are given below.
Experiments to test the nickel complexes with tripodal phos-
phine ligands with respect to their catalytic activity in other re-
actions, for example, as inexpensive substitutes for the Pd
component of the Sonogashira reaction, are underway. Fur-
thermore, tripodal phosphine ligands incorporating the eth-
oxysilane group for immobilized versions of the nickel com-
plexes on oxide supports[3b,8] will be used to transfer the en-
couraging results on the homogeneous catalysts presented in
this paper to the world of catalysts immobilized on oxide sup-
ports.
Experimental Section
General remarks
The 1H, 13C, and 31P NMR spectra of liquids were recorded at
499.70, 125.66, and 202.28 MHz, respectively, on a 500 MHz Varian
1
spectrometer. The 13C and 31P NMR spectra were recorded with H
decoupling if not stated otherwise. Neat Ph2PCl [d(31P)=
+81.92 ppm] in a capillary centered in the 5 mm NMR tubes was
used for referencing the 31P chemical shifts of dissolved com-
1
M.p. (decomp.) 1588C; H NMR (499.69 MHz, C6D6): d=7.50 (t, 12H,
3J(1H,1H)=8.3 Hz, H6), 6.82–6.73 (m, 18H, H5, H7, overlapping), 2.13
(d, 6H, 2J(31P,1 H)=7.2 Hz, H3), 1.22 ppm (q, 3H, 4J(31P,1 H)=2.7 Hz,
H1); 13C NMR (125.66 MHz, C6D6): d=207.89 (q, 2J(31P,13C)=4.0 Hz,
C8), 140.62 (virt. m, 1J(31P,13C)=42.9 Hz, C4), 131.93 (d, 2J(31P,13C)=
1
pounds. For referencing the H and 13C chemical shifts, if not men-
tioned otherwise, the residual proton signals of the solvent C6D6
and the carbon signals were used [d(1H)=7.16 ppm, d(13C)=
128.00 ppm]. All signal assignments were based on comparisons
with analogous metal complexes with tripodal phosphine li-
gands[3a,b,8] and two-dimensional 1H–1H and 1H–13C correlation
spectroscopy. The 31P solid-state NMR spectra were measured with
a Bruker Avance 400 widebore NMR spectrometer equipped with
16.1 Hz, C5), 128.29 (s, C7), 127.91 (d, J(31P,13C)=9.8 Hz, C6), 38.17
3
(q, J(31P,13C)=10.9 Hz, C2), 36.73 (q, J(31P,13C)=6.5 Hz, C1), 36.74–
2
3
36.53 ppm (virt m, C3); 31P{1H} NMR (202.28 MHz, C6D6): d=
À1
~
11.54 ppm (s); IR: (CO) n=1886.5 cm
.
Chem. Eur. J. 2015, 21, 10138 – 10148
www.chemeurj.org
10145
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Complex 3: Triphos (198.1 mg, 0.3171 mmol) was dissolved in tolu-
ene (20 mL) and added to a solution of [Ni(CO)2(PPh3)2] (197.4 mg,
0.3088 mmol) in toluene (22 mL). The mixture was stirred at RT in
the dark for 15 h. The solution was concentrated under reduced
pressure, resulting in a slightly yellow precipitate, which was fil-
tered off and washed with hexanes (3”10 mL) to give 3 as a white
powder (156.3 mg, 0.2114 mmol, 69% yield with respect to [Ni-
(CO)2(PPh3)2]). Single crystals suitable for X-ray diffraction were
grown by layering a toluene solution of 3 with pentane.
powder when pentane (40 mL) was added. The solid was washed
with pentane (1”10 mL) to yield pure 4 as a white powder
(33.1 mg, 58% yield with respect to [AuCl(tht)]). Single crystals suit-
able for X-ray diffraction analysis were grown by slow diffusion of
liquid pentane into a THF solution of 4.
M.p. (decomp.) 1338C; 1H NMR (499.69 MHz, C6D6): d=7.85 (dd,
8H, 3J(1H,1H)=7.6 Hz, 3J(31P,1 H)=9.2 Hz, H6, H10), 7.14 (tm, 4H,
3J(1H,1H)=7.8 Hz, H7), 7.10 (tm, 4H, 3J(1H,1H)=7.8 Hz, H11), 7.06
(ddt, 4H, 3J(31P,1 H)=12.8 Hz, 3J(1H,1H)=7.1 Hz, 4J(1H,1H)=1.5 Hz,
H16), 7.00 (dtt, 2H, 5J(31P,1 H)=1.2 Hz, 3J(1H,1H)=7.4 Hz, 4J(1H,1H)=
1.2 Hz, H8), 6.92 (dtt, 2H, 5J(31P,1 H)=1.2 Hz, 3J(1H,1H)=7.4 Hz,
4J(1H,1H)=1.2 Hz, H12), 6.89 (dtt, 2H, 5J(31P,1 H)=2.1 Hz, 3J(1H,1H)=
7.4 Hz, 4J(1H,1H)=1.6 Hz, H18), 6.82 (dtt, 4H, 4J(31P,1 H)=2.4 Hz,
3J(1H,1H)=7.5 Hz, 4J(1H,1H)=1.5 Hz, H17), 2.68 (dd, 2H, 2J(31P,1 H)=
8.2 Hz, 2J(1H,1H)=14.2 Hz, H3syn), 2.27 (dd, 2H, 2J(31P,1 H)=6.7 Hz,
2J(1H,1H)=14.2 Hz, H3anti), 2.01 (d, 2H, 2J(31P,1 H)=11.5 Hz, H13),
0.74 ppm (s, 3H, H1); 13C{1H,31P} NMR (125.66 MHz, C6D6): d=
201.25 (s, C19), 201.03 (s, C20), 139.09 (s, C5, C9), 134.16 (s, C15),
133.35 (s, C16), 132.92 (s, C10), 132.35 (s, C6), 131.40 (s, C18),
129.56 (s, C8), 129.43 (s, C12), 129.12 (s, C17), 128.88 (s, C11),
128.72 (s, C7), 43.06 (s, C13), 41.06 (s, C3), 38.67 (s, C2), 31.76 ppm
M.p. (quantitative transformation into 2) 1368C; 1H NMR
(499.69 MHz, C6D6): d=7.89 (t, 4H, 3J(31P,1 H)=8.8 Hz, 3J(1H,1H)=
7.5 Hz, H10), 7.59 (t, 4H, 3J(31P,1 H)=8.3 Hz, 3J(1H,1H)=7.5 Hz, H6),
7.35 (t, 4H, 3J(31P,1 H)=7.3 Hz, 3J(1H,1H)=7.6 Hz, 4J(1H,1H)=1.7 Hz,
H16), 7.10–6.92 (m, 18H, H7, H8, H11, H12, H17, H18), 2.56 (dd, 2H,
2J(1H,1H)=14.1 Hz, 2J(31P,1 H)=11.7 Hz, H3syn), 2.23 (dd, 2H,
4
(s, C1); 31P{1H} NMR (202.28 MHz, C6D6): d=14.04 (d, 2P, J(31P,31P)=
~
1.3 Hz, P4), 16.20 ppm (m, 1P, P14); IR: (CO) n=1999.3,
1933.7 cmÀ1
.
[Ni(CO)(PPh3)3]: [Ni(CO)(PPh3)3] was prepared by stirring a blend of
molten [Ni(CO)2(PPh3)2] (131.7 mg, 0.2060 mmol) and PPh3
(0.9976 g, 3.8034 mmol) at 1608C for 40 min while a slow stream
of N2 was passed over the mixture. The resulting bright yellow
liquid was allowed to cool, but not enough to solidify (ca. 1008C),
and dry pentane (10 mL) was added and the mixture shaken caus-
ing a precipitate to form. The latter was washed with pentane (4”
10 mL) and then dried under vacuum to give the product as
a bright yellow powder (111.7 mg, 0.1279 mmol, 62% yield). The
31P NMR chemical shift matches the literature values.[23c,d] 1H NMR
(499.69 MHz, C6D6): d=7.36 (bdd, 2H, 3J(1H,1H)=7.36 Hz,
2
4
2J(1H,1H)=14.2 Hz, J(31P,1 H)=3.2 Hz, H3anti), 2.04 (dd, 2H, J(31P,1 H)=
1.7 Hz, 2J(31P,1 H)=2.7 Hz, H13), 0.69 ppm (s, 3H, H1); 13C NMR
(125.66 MHz, C6D6): d=202.87 (s, C19), 200.03 (t, J(31P,13C)=3.6 Hz,
2
C20), 140.84 (virt. m, J(31P,13C)=36.3 Hz, C5*), 140.01 (d, J(31P,13C)=
1
12.7 Hz, C15), 138.93 (virt. m, J(31P,13C)=30.9 Hz, C9*), 133.40 (d,
2J(31P,13C)=20.2 Hz, C16), 132.90 (virt. m, 2/4J(31P,13C)=29.1 Hz, C10),
131.78 (virt. m, 2/4J(31P,13C)=28.2 Hz, C6), 129.50 (s, C8°), 128.86 (s,
C12°), 128.81–128.70 (m, C17°, C18†), 128.51 (virt. m, J(31P,13C)=
4.5 Hz, C7†, C11†), 49.11 (dt, 3J(31P,13C)=8.5 Hz, 1J(31P,13C)=17.5 Hz,
C13), 40.71 (ddd, 3J(31P,13C)=8.6 Hz, 3J(31P,13C)=8.0 Hz, 1J(31P,13C)=
11.3 Hz, C3), 39.09 (dt, 2J(31P,13C)=13.6 Hz, 2J(31P,13C)=4.7 Hz, C2),
29.29 ppm (dt, 3J(31P,13C)=10.0 Hz, 3J(31P,13C)=6.4 Hz, C1); *,°,† as-
signments interchangeable; 31P{1H} NMR (202.28 MHz, C6D6): d=
3
3J(31P,1 H)=7.36 Hz, Ho), 6.95 (t, 1H, J(1H,1H)=7.38 Hz, Hp), 6.84 ppm
(t, 2H, 3J(1H,1H)=7.49 Hz, Hm); 13C NMR (125.66 MHz, C6D6): d=
201.25 (q, 2J(31P,13C)=6.61 Hz, CO), 139.15 (virt. m, 1J(31P,13C)=
34.6 Hz, Ci), 134.09 (dd, 4J(31P,13C)=4.9 Hz, 2J(31P,13C)=9.08 Hz, Co),
128.35 (s, Cp), 127.87 ppm (dd, 5J(31P,13C)=2.9 Hz, 3J(31P,13C)=
5.58 Hz, Cm); 31P{1H} NMR (202.28 MHz, C6D6): d=31.90 ppm.
4
13.93 (d, 2P, 4J(31P,31P)=2.2 Hz, P4), À27.79 ppm (t, 1P, J(31P,31P)=
À1
~
2.2 Hz, P14); IR: (CO)=n=1998.3, 1931.8 cm
.
Complex 4: [AuCl(tht)] (19.0 mg, 0.0593 mmol; tht=tetrahydro-
thiophene) was added to a solution of 3 (45.9 mg, 0.0621 mmol) in
toluene (10 mL). The solution immediately assumed a dark purple
color. After stirring the reaction mixture for 20 h, the solvent was
removed under reduced pressure, yielding a purple solid. The solid
was triturated in ether (50 mL), stirred briefly, and allowed to settle
overnight. The clear supernatant solution was then collected and
concentrated under reduced pressure. The residue was then re-dis-
solved in THF (2 mL) and the product precipitated as a white
Complex 6: [Ni(CO)(PPh3)3] (0.1117 g, 0.1279 mmol) was dissolved
in toluene (10 mL) and combined with a solution of 5 (87.2 mg,
0.1277 mmol) in toluene (10 mL). This mixture was stirred for
75 min, after which the solvent was removed under reduced pres-
sure. The resulting solid was washed with pentane (3”10 mL), dis-
solved in toluene (3 mL), and precipitated by adding pentane (2”
10 mL). The resulting yellow powder was shown by 31P NMR spec-
troscopy to contain 6 (34.5 mg, 0.0448 mmol, 25% yield), alongside
a small amount of PPh3, and presumably oligomeric side products.
These impurities were removed by repeated precipitation from
a toluene solution by pentane. Crystals suitable for X-ray diffraction
were obtained by slow diffusion of liquid pentane into a toluene
solution of 6.
M.p. (decomp.) 113 8C; 1H{31P} NMR (499.69 MHz, C6D6): d=À0.06
(s, 3H, H1), 0.74 (m, 6H, H2), 2.15 (m, 6H, H3), 6.95–6.99 (m, 18H,
H6, H8), 7.47–7.50 ppm (m, 12H, H7); 13C{31P,1 H} NMR (125.66 MHz,
[D8]THF, d(13C)=25.37): d=142.41 (C5), 132.35 (C6), 128.55 (C7),
128.31 (C8), 23.73 (C3), 8.40 (C2), 5.79 ppm (C1); 31P{1H} NMR
Chem. Eur. J. 2015, 21, 10138 – 10148
www.chemeurj.org
10146
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Keywords: homogeneous catalysis
·
nickel
·
NMR
spectroscopy · phosphane ligands · solid-state structures
[1] a) Heterogenized Homogeneous Catalysts for Fine Chemicals Production
(Eds.: P. Barbaro, F. Liguori), Springer, Heidelberg, 2010; b) Chiral Cata-
lyst Immobilization and Recycling (Eds.: D. E. De Vos, I. F. J. Vankelecom,
P. A. Jacobs), Wiley-VCH, Weinheim, 2000; c) G. Rothenberg, Catalysis:
Concepts and Green Applications, Wiley-VCH, Weinheim, 2008.
[2] J. Blümel, Coord. Chem. Rev. 2008, 252, 2410–2423.
[3] a) J. Guenther, J. Reibenspies, J. Blümel, Adv. Synth. Catal. 2011, 353,
443–460; b) J. Guenther, N. Bhuvanesh, J. Blümel, unpublished results;
c) R. Silbernagel, A. Diaz, E. Steffensmeier, A. Clearfield, J. Blümel, J. Mol.
Catal. A 2014, 394, 217–223; d) C. Merckle, J. Blümel, Adv. Synth. Catal.
2003, 345, 584–588; e) C. Merckle, J. Blümel, Top. Catal. 2005, 34, 5–
15; f) Y. Yang, B. Beele, J. Blümel, J. Am. Chem. Soc. 2008, 130, 3771–
3773; g) B. Beele, J. Guenther, M. Perera, M. Stach, T. Oeser, J. Blümel,
New J. Chem. 2010, 34, 2729–2731.
[4] a) J. C. Pope, T. Posset, N. Bhuvanesh, J. Blümel, Organometallics 2014,
33, 6750–6753; b) T. Posset, J. Blümel, J. Am. Chem. Soc. 2006, 128,
8394–8395; c) T. Posset, J. Guenther, J. Pope, T. Oeser, J. Blümel, Chem.
Commun. 2011, 47, 2059–2061.
[5] a) S. Reinhard, P. Soba, F. Rominger, J. Blümel, Adv. Synth. Catal. 2003,
345, 589–602; b) S. Reinhard, K. D. Behringer, J. Blümel, New J. Chem.
2003, 27, 776–778; c) K. D. Behringer, J. Blümel, Chem. Commun. 1996,
653–654; d) T. Posset, Dissertation Thesis, University of Heidelberg,
2006.
[6] a) P. Kumar, J. Louie in Transition-Metal-Mediated Aromatic Ring Construc-
tion (Ed.: K. Tanaka), Wiley, Hoboken, 2013, pp. 37–70; b) M. C. Biel, R.
Kessinger, J. Sommer, J. Blümel, U.S. Patent, 8318968 B2, 2012; [Chem.
Abstr. 2012, 2008, 1157541]; c) M. C. Biel, R. Kessinger, J. Sommer, J.
Blümel, U.S. Patent, WO 2008113777, 2008; [Chem. Abstr. 2008, 149,
402503].
~
(202.28 MHz, toluene): d=19.81 ppm (s, P4); IR: (CO) n=
1911.5 cmÀ1
.
Complex 7: [NiCl2py4] (103.8 mg, 0.2327 mmol) and ligand
5
(157.7 mg, 0.2309 mmol) were stirred in THF (40 mL). The opaque
reaction mixture turned red immediately and then the color faded
to white within 1 min. Subsequently, PPh3 (62.6 mg, 0.239 mmol)
was added, followed by Zn powder (2.000 g, 30.58 mmol). The re-
action mixture was degassed under vacuum for 2 min and then
stirred for 16 h. During this time a clear, orange-red solution
formed alongside solid matter. After settling, the solution was sep-
arated from the precipitate via syringe and the solvent was re-
moved under reduced pressure. The resulting residue was extract-
ed by stirring with toluene. The solids were allowed to settle and
the supernatant was collected, stripped of the solvent, and the res-
idue was washed with pentane (10 mL) and dried under vacuum,
giving the product as an orange-red powder (203.4 mg,
0.2026 mmol, 87% yield).
[7] a) D. L. J. Broere, E. Ruijter, Synthesis 2012, 2639–2672; b) R. Hua,
M. V. A. Abrenica, P. Wang, Curr. Org. Chem. 2011, 15, 712–729; c) Y. Shi-
bata, K. Tanaka, Synthesis 2012, 323–350; d) G. Domínguez, J. Perez-
Castells, Chem. Soc. Rev. 2011, 40, 3430–3444.
[8] M. Bogza, T. Oeser, J. Blümel, J. Organomet. Chem. 2005, 690, 3383–
3389.
[9] Selected references: a) R. Gilbert-Wilson, L. D. Field, M. M. Bhadbhade,
Inorg. Chem. 2012, 51, 3239–3246; b) P. J. Desrochers, D. S. Duong, A. S.
Marshall, S. A. Lelievre, B. Hong, J. R. Brown, R. M. Tarkka, J. M. Manion,
G. Holman, J. W. Merkert, D. A. Vicic, Inorg. Chem. 2007, 46, 9221–9233;
c) R. B. Bedford, M. Betham, C. P. Butts, S. J. Coles, M. Cutajar, T. Gelbrich,
M. B. Hursthouse, P. N. Scully, S. Wimperis, Dalton Trans. 2007, 459–466;
d) C. E. MacBeth, J. C. Thomas, T. A. Betley, J. C. Peters, Inorg. Chem.
2004, 43, 4645–4662.
[10] Selected recent references: a) D. Fernµndez-Anca, M. I. Garcia-Seijo,
M. E. Garcia-Fernandez, Dalton Trans. 2010, 39, 2327–2336; b) D.
Fernµndez-Anca, M. I. Garcia-Seijo, A. Castineiras, M. E. Garcia-Fernan-
dez, Inorg. Chem. 2008, 47, 5685–5695; c) A. B. Chaplin, Z. Beni, C. G.
Hartinger, H. H. Ben, A. D. Phillips, R. Scopelliti, P. J. Dyson, J. Cluster Sci.
2008, 19, 295–309; d) P. W. Miller, A. J. P. White, J. Organomet. Chem.
2010, 695, 1138–1145.
[11] C. A. Fyfe, Solid-State NMR for Chemists, CFC Press, Guelph, ON, 1983.
[12] T. M. Duncan, A Compilation of Chemical Shift Anisotropies, Farragut
Press, Chicago, IL, 1990.
1
M.p. (decomp.) 388C; H NMR (499.69 MHz, C6D6): d=À0.25 (s, 3H,
H1), 0.70 (bs, 6H, H2), 2.43 (vbs, 6H, H3), 6.58 (dd, 6H, J(31P,1 H)=
3
7.6 Hz, 3J(1H,1H)=7.0 Hz, H11), 6.81 (t, 6H, 3J(1H,1H)=7.2 Hz, H12),
6.89 (t, 3H, 3J(1H,1H)=7.1 Hz, H13), 7.11 (bs, 24H, H6, H7), 7.30 ppm
(bs, 6H, H8); 13C{31P,1 H} NMR (125.66 MHz, [D8]THF, d(13C)=25.37):
d=5.18 (s, C1), 9.02 (s, C2), 27.82 (s, C3), 141.04 (s, C5), 149.86 (s,
C10), 134.82 (s, C11), 134.17 (bs, C6), 129.10 (s, C8), 128.21 (s, C13),
128.03 (s, C7), 127.81 ppm (s, C12); 31P{1H} NMR (202.28 MHz, C6D6):
d=12.01 (d, 3P, 2J(31P,31P)=18.1 Hz, P4), 22.50 ppm (q, 1P,
2J(31P,31P)=18.2 Hz, P9).
[13] W. Lu, D. Yuan, D. Zhao, C. I. Schilling, O. Plietzsch, T. Muller, S. Bräse, J.
Guenther, J. Blümel, R. Krishna, Z. Li, H.-C. Zhou, Chem. Mater. 2010, 22,
5964–5972.
[14] a) J. C. Pope, H.-J. Sue, T. Bremner, J. Blümel, Polymer 2014, 55, 4577–
4585; b) J. C. Pope, H.-J. Sue, T. Bremner, J. Blümel, J. Appl. Polym. Sci.
2015, 132, 1804–1816; c) J. Guenther, M. Wong, H.-J. Sue, T. Bremner, J.
Blümel, J. Appl. Polym. Sci. 2013, 128, 4395–4404.
Acknowledgements
[15] S. Reinhard, J. Blümel, Magn. Reson. Chem. 2003, 41, 406–416.
[16] a) J. W. Diesveld, E. M. Menger, H. T. Edzes, W. S. Veeman, J. Am. Chem.
Soc. 1980, 102, 7935–7936; b) K. Eichele, R. E. Wasylishen, J. F. Corrigan,
N. J. Taylor, A. J. Carty, K. W. Feindel, G. M. Bernard, J. Am. Chem. Soc.
2002, 124, 1541–1552.
This work was supported by the Robert A. Welch Foundation
(A-1706), the National Science Foundation (CHE-0911207, CHE-
1300208, CHE-0840464), and the APPEAL Consortium at Texas
A&M University.
[17] M. R. Chierotti, R. Gobetto, Eur. J. Inorg. Chem. 2009, 2581–2597.
Chem. Eur. J. 2015, 21, 10138 – 10148
www.chemeurj.org
10147
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[18] N. Masciocchi, S. Galli, M. Bogza, J. Blümel, Powder Diffr. 2007, 22, 55–
58.
quest/cif. C43H39NiO2P3; unit cell parameters: a 8.5795(12), b 24.844(3),
c 17.483(3); P21/n.
[19] a) K. J. Cluff, N. Bhuvanesh, J. Blümel, Organometallics 2014, 33, 2671–
2680; b) K. J. Cluff, M. Schnellbach, C. R. Hilliard, J. Blümel, J. Organomet.
Chem. 2013, 744, 119–124.
[20] C. R. Hilliard, S. Kharel, K. J. Cluff, J. A. Gladysz, J. Blümel, Chem. Eur. J.
2014, 20, 17292–17295.
[30] CCDC 1037017 (4) contains the supplementary crystallographic data for
this paper. These data can be obtained free of charge from The Cam-
bridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_re-
quest/cif. C43H39AuClNiO2P3, 0.9(C4H8O), 0.1(H2O), unit cell parameters:
a 38.030(3), b 11.0474(8), c 25.887(2); C2/c.
[21] a) J. Blümel, Inorg. Chem. 1994, 33, 5050–5056; b) J. Sommer, Y. Yang,
D. Rambow, J. Blümel, Inorg. Chem. 2004, 43, 7561–7563.
[22] C. R. Hilliard, N. Bhuvanesh, J. A. Gladysz, J. Blümel, Dalton Trans. 2012,
41, 1742–1754.
[23] Selected more recent references: a) T. Yamamoto, J. Ishizu, T. Kohara, S.
Komiya, A. Yamamoto, J. Am. Chem. Soc. 1980, 102, 3758–3764; b) S.
Gambarotta, C. Floriani, A. Chiesi-Villa, C. Guastini, Organometallics
1986, 5, 2425–2433; c) T. R. Gaffney, J. A. Ibers, Inorg. Chem. 1982, 21,
2860–2864; d) L. S. Isaeva, L. N. Morozova, D. N. Kravtsov, Inorg. Chim.
Acta 1990, 171, 29–33.
[31] CCDC 1039467 (6) contains the supplementary crystallographic data for
this paper. These data can be obtained free of charge from The Cam-
bridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_re-
quest/cif. C44H45NiP3Si; unit cell parameters: a 10.465(2), b 10.485(2),
c 18.592(4); P21/n.
[32] CCDC 1037018 (7) contains the supplementary crystallographic data for
this paper. These data can be obtained free of charge from The Cam-
bridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_re-
quest/cif. C61H60NiP4Si; 0.72(C5H12), 0.28(C3H7), 0.28(C2H5), unit cell pa-
rameters: a 13.2287(11), b 16.4601(14), c 13.7514(12); P21.
[33] J. Mautz, K. Heinze, H. Wadepohl, G. Huttner, Eur. J. Inorg. Chem. 2008,
1413–1422.
[24] F. Piestert, R. Fetouaki, M. Bogza, T. Oeser, J. Blümel, Chem. Commun.
2005, 1481–1483.
[25] a) W. H. Hersh, P. Xu, B. Wag, J. W. Yom, C. K. Simpson, Inorg. Chem.
1996, 35, 5453–5459; b) W. H. Hersh, J. Chem. Educ. 1997, 74, 1485–
1488.
[26] S. Brenna, T. Posset, J. Furrer, J. Blümel, Chem. Eur. J. 2006, 12, 2880–
2888.
[34] D. E. Hendriksen, A. A. Oswald, G. B. Ansell, S. Leta, R. V. Kastrup, Orga-
nometallics 1989, 8, 1153–1173.
[35] G. Hilt, T. Vogler, W. Hess, F. Galbiati, Chem. Commun. 2005, 1474–1475.
[36] B. O. Öztürk, S. Karabulut, Y. Imamoglu, Appl. Catal. A Gen. 2012, 433–
434, 214–222.
[27] CCDC 1037015 (2) contains the supplementary crystallographic data for
this paper. These data can be obtained free of charge from The Cam-
bridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_re-
[37] V. P. Anju, D. K. Roy, R. S. Anju, S. Ghosh, J. Organomet. Chem. 2013, 733,
79–81.
[38] M. Matsuura, T. Fujihara, M. Kakeya, T. Sugaya, A. Nagasawa, J. Organo-
met. Chem. 2013, 745–746, 288–298.
quest/cif. C42H39NiOP3; unit cell parameters:
a 17.94920.564(5),
b 10.117(3), c 16.976(5); Pna21.
[39] C. Xi, Z. Sun, Y. Liu, Dalton Trans. 2013, 42, 13327–13330.
[40] J. Chatt, F. A. Hart, J. Chem. Soc. 1965, 812–813.
[28] Program “dmfit”: D. Massiot, F. Fayon, M. Capron, I. King, C. S. Le, B.
Alonso, J.-O. Durand, B. Bujoli, Z. Gan, G. Hoatson, Magn. Reson. Chem.
2002, 40, 70–76.
Received: January 15, 2015
Revised: April 23, 2015
Published online on June 8, 2015
[29] CCDC 1037016 (3) contains the supplementary crystallographic data for
this paper. These data can be obtained free of charge from The Cam-
bridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_re-
Chem. Eur. J. 2015, 21, 10138 – 10148
www.chemeurj.org
10148
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim