Highly efficient nickel-catalyzed 2-methyl-3-butenenitrile isomerization: Applications and mechanistic studies employing the TTP ligand family

Article abstract of DOI:10.1021/om200164f

A series of sterically and electronically fine-tuned, chelating diphosphine ligands were synthesized. The ligands are analogues of Triptyphos (TTP, 1), all based upon a variably 9,10-two-carbon-bridged 9,10-dihydroanthracene scaffold. These new TTP-type l

Full text of DOI:10.1021/om200164f

ARTICLE
pubs.acs.org/Organometallics
Highly Efficient Nickel-Catalyzed 2-Methyl-3-butenenitrile
Isomerization: Applications and Mechanistic Studies Employing the
TTP Ligand Family
Michael E. Tauchert, Daniel C. M. Warth, Sebastian M. Braun, Irene Gruber, Alexandra Ziesak,
Frank Rominger, and Peter Hofmann*
Organisch-Chemisches Institut, Universit€at Heidelberg, Im Neuenheimer Feld 270, D-69120 Heidelberg, Germany
S Supporting Information
b
ABSTRACT: A series of sterically and electronically fine-tuned,
chelating diphosphine ligands were synthesized. The ligands
are analogues of Triptyphos (TTP, 1), all based upon a varia-
bly 9,10-two-carbon-bridged 9,10-dihydroanthracene scaffold.
These new TTP-type ligands were employed in the Ni(0)-
catalyzed isomerization of 2-methyl-3-butenenitrile (2M3BN),
one of the key steps of industrial adiponitrile production by the
DuPont process. The reaction showed a surprising preference
for ligands bearing electron-donating substituents, such as
methoxy or methyl groups, in the phenyl para position of the
Ni-ligating PPh2 units. Octyltriptyphos (3) afforded the highest 2M3BN-isomerization turnover rate yet reported. A series of
deuterium-labeling experiments was performed to investigate the possibility of an isomerization mechanism consisting of a cascade
of de- and rehydrocyanation steps, which could be excluded. Using the ethano-bridged ligand 4, complex 16a (4-κP:κP0)Ni(η3-
C4H7)CN (supposedly an intermediate of the 2M3BN-isomerization reaction) was isolated, and its solid-state structure was
determined by X-ray diffraction analysis. The complete catalytic cycle of 2M3BN isomerization with ligand 4, as suggested by the
available experimental evidence, was modeled using DFT methods.
’ INTRODUCTION
in situ from the catalyst, HCN, and the Lewis acid. Thanks to the
meticulous work of Tolman et al., many of the mechanistic
aspects of the nickel-catalyzed butadiene hydrocyanation are well
understood, at least for catalysts containing monodentate phos-
phite ligands (Scheme 2).5 According to this mechanism, the
catalyst precursor NiL4 (L = P(OEt)3) first loses one of its
ligands, forming NiL3 and thus allowing the oxidative addition of
HCN, which affords cyano hydride complex A. Further ligand
dissociation (complex B) promotes the insertion of butadiene
into the nickelꢀhydride bond, yielding the methallyl complex C.
This species is then transformed into the linear or branched
cyanoolefin complex D or F, respectively, by reductive elimina-
tion. Following an associative mechanism via complexes E or G
the catalyst can reenter the catalytic cycle with release of 2M3BN
and t3PN. According to Tolman’s work, all of these steps are
reversible.
The nickel-catalyzed isomerization of 2-methyl-3-buteneni-
trile is one of the key steps of the so-called DuPont process, one
of the large-scale applications of homogeneous transition metal
catalysis. During this reaction, butadiene is converted to adipo-
nitrile (ADN) by a sequence of hydrocyanation and isomeriza-
tion steps (Scheme 1).1 ADN is the precursor for the nylon-6.6
monomer hexamethylenediamine (1,6-diaminohexane). The
demand for ADN is therefore closely related to that for nylon-
6.6. In 2005 the global demand was estimated to be about 1.3
million tons, 67% of which was covered by the DuPont process.2
The process was established by Drinkard et al. (Du Pont)
using nickel phosphite complexes as catalysts.3 In a three-step
process, HCN is first added to butadiene, resulting in a mixture of
2-methyl-3-butenenitrile (2M3BN) and 3-pentenenitrile (3PN).
In a second step, the hydrocyanation catalyst is used in the
absence of HCN to remove the undesired branched 2M3BN by
isomerization to 3PN isomers. Usually trans-3-pentenenitrile
(t3PN) is favored over its isomer cis-3-pentenenitrile (c3PN).
During the third and last step 3PN is isomerized in situ to
4-pentenenitrile (4PN), which is then quickly hydrocyanated to
yield ADN. A Lewis acid is usually necessary for completion of
this last step.4 The isomerization between t3PN and 4PN (eq III)
formally does not involve HCN. Its mechanism is assumed,
however, to involve a cationic NiꢀH complex, which is formed
The 2M3BN isomerization appears to follow the same mechan-
istic pathways. Complexation of 2M3BN with NiL3 (complex E)
allows oxidative addition of the CꢀCN bond of the cyanoolefin,
yielding complex C. Reductive elimination eventually leads to the
thermodynamically favored 3PN. A deuterium-labeling experi-
ment by Druliner suggests the occurrence of β-hydride
Received: February 21, 2011
Published: April 21, 2011
r
2011 American Chemical Society
2790
dx.doi.org/10.1021/om200164f Organometallics 2011, 30, 2790–2809
|

Organometallics
ARTICLE
Scheme 1
Scheme 2
Scheme 4
Scheme 3
elimination (Scheme 3).6 In this experiment 2M3BN-d, which was
deuterium-enriched at its methyl group, was submitted to isomer-
ization conditions using Ni[P(O-p-tolyl)3]4 as a catalyst and
ZnCl2 as a cocatalyst at 110 °C. A statistical deuteration at the
C-2 and C-5 positions of the 3PN-d product was observed,
indicating a cascade of de- and rehydrocyanation as a mechanistic
pathway.
mechanistic insight into this reaction will also help in under-
standing the former one.
In addition to phosphites, in the past few years phosphonite7
as well as phosphine ligands have also been the object of intensive
research.8 Using dppb (1,4-bis(diphenylphosphino)butane) as a
ligand, Chaudret and Sabo-Etienne et al. were able to synthesize,
isolate, and structurally characterize the π-methallyl cyanonickel
complex (dppb)Ni(η3-C4H7)CN.9 The role of such methallylic
nickel complexes in the isomerization and the first hydrocyana-
tion step had been proposed almost 20 years before.5,6,10 On the
Due to the close relationship between the hydrocyanation of
butadiene and the isomerization of 2M3BN, the latter reaction
is a very interesting target for mechanistic investigations, since
2791
dx.doi.org/10.1021/om200164f |Organometallics 2011, 30, 2790–2809

Organometallics
ARTICLE
Figure 1. Representatives of the Triptyphos family.
Scheme 5
Scheme 6
basis of DFT calculations (B3LYP) using PH3 as a model ligand
the authors suggested a 1,3-allylic CN group migration via π-
allylic species as the reaction pathway (Scheme 4). A mechanism
involving intermediate HCN complexes was excluded as being
too high in energy. Detailed and definite experimental proof
which pathway is followed, however, is still lacking. In the
following years π-methallyl cyanonickel complexes have also
been described for dpe-phos (bis(2-diphenylphosphinophenyl)
ether),11 dppf (1,10-bis(diphenylphosphino)ferrocene),12 dcype
(1,2-bis(dicyclohexylphosphino)ethane),13 and tBuppf (1-diphe-
nylphosphino-10-(di-tert-butylphosphino)ferrocene).14 It is inter-
esting to note that for some of the phosphine ligands, in contrast to
the monodentate phosphite ligands, a decrease in activity has been
reported when adding Lewis acids such as ZnCl2.12,14 Also, in the
case of dippe (1,2-bis(diisopropylphosphino)ethane) it has been
shown that, depending on the choice of solvent, the 2M3BN
isomerization can be steered toward either 2-methyl-2-buteneni-
trile or linear 3-pentenenitrile.15 The mechanism of both isomer-
ization types has been recently investigated by DFT calculations
using dmpe (1,2-bis(dimethylphosphino)ethane) as a model
ligand.16
It is known for several examples that ligands originally devel-
oped for efficient rhodium-catalyzed hydroformylation also
function efficiently in nickel-catalyzed hydrocyanation reactions.
Phosphites derived from 2,20-dihydroxybiphenyls,17 arylpho-
sphonites with cycloalkane or triptycene spacers7a,b and Tripty-
phos 1 (TTP, Figure 1) from our own group and others,18,19 self-
associating phosphine ligands,20 and Xantphos21 and its deriva-
tives must be also mentioned in this context.
The impetus for the synthesis of Triptyphos 1 was based on
observations made for Xantphos-type ligands developed by van
Leeuwen et al. for rhodium-catalyzed hydroformylation.22 The
excellent selectivity of Xantphos and its derivatives in the hydro-
formylation of terminal alkenes is believed to originate from the
rigid xanthene backbone, from a wide bite angle of around 120°,
and from what was called advantageous “secondary interactions”
between the diphenylphosphino groups and the metal fragment.23
However, DFT calculations by Landis et al. and by Carbꢀo
et al. revealed that the flat but nonplanar roof-shaped xanthene
2792
dx.doi.org/10.1021/om200164f |Organometallics 2011, 30, 2790–2809

Organometallics
ARTICLE
backbone has a low inversion barrierand that thisfeature, as wellas
the practically free rotation of the phenyl rings at the phosphorus
atoms (not hampered by the xanthene backbone), can lead to
numerous ligandconformersandvariouscoordinationgeometries,
as shown for the trigonal-bipyramidal model complex 7
(Scheme 5).24,25 This conformational flexibility has also been
observed by van Leeuwen et al. when studying nickel-Thixantphos
complexes.21 Since this nonrigidity, which allows more facile
structural response of the ligand sphere toward different reaction
pathways than more rigid scaffolds, might lower catalyst selectivity
by failing to sufficiently differentiate between alternative geome-
tries of catalytic intermediates and/or transition states, we sought
to design a ligand system retaining the main features of Xantphos
and its congeners, while suppressing the backbone inversion and
free phenyl group rotation. Decreasing the number of possible
conformers of the catalyst was also expected to greatly simplify
mechanistic analysis.
Features along the lines of our working hypothesis were
achieved by introducing a two-carbon bridge across the 9 and
10 positions of a 9,10-dihydroanthracene (similar to xanthenes)
backbone, resulting in Triptyphos 1 and its derivatives 2ꢀ6
(Figure 1). These ligands indeed showed very high activity
combined with exceptional chemoselectivity and n-selectivity
(above 99% for 1) in the rhodium-catalyzed low-pressure
hydroformylation of 1-alkenes.18aꢀe Also, its phosphonite, phos-
phite, and dialkylphosphine analogues have been applied suc-
cessfully in catalysis.7a,26
Figure 2. Thermal ellipsoid plot of ligand 1 at the 50% probability level.
Hydrogen atoms are omitted for clarity. Selected bond lengths (Å) and
angles (deg): P1ꢀC1 = 1.845(2), P1ꢀC15 = 1.830(2), P1ꢀC21 =
1.836(3), C6ꢀC1ꢀC2 = 117.4(2), C15ꢀP1ꢀC21 = 101.88(11),
C15ꢀP1ꢀC1 = 103.31(11), C21ꢀP1ꢀC1 = 102.99(11).28.
Figure 3. Superposition of solid-state structures (front view) of Xant-
phos 10 (white) and Triptyphos 1 (black).
Table 1. 2M3BN to t3PN/c3PN Isomerization Reaction Conditions.a
entry
ligand
Ni [%]
T [°C]
t [h]
conv [%]b
2M3BN [%]b
t3PN [%]b
c3PN [%]b
1
3
0.20
120
100
80
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
3.0
20.0
1.5
3.0
20.0
1.5
1.5
1.5
1.5
1.5
1.5
96.1
96.7
79.2
16.7
96.0
96.0
95.6
46.5
82.3
96.1
47.6
80.0
95.6
72.1
71.2
55.6
68.0
71.7
17.4
3.9
3.3
85.6
93.5
77.6
14.9
89.1
92.6
92.0
45.5
80.7
87.2
46.3
78.2
87.3
69.0
69.1
52.0
65.5
68.9
15.6
8.2
1.5
0.9
0.0
5.3
2.2
2.2
0.4
1.0
8.0
1.3
1.8
8.3
0.9
0.4
1.3
0.9
0.6
0.0
2
3
0.20
3
3
0.20
20.8
83.4
4.0
4
3
0.20
60
5
3
0.10
120
120
120
120
120
120
120
120
120
100
100
100
100
100
100
6
3
0.05
4.0
7
3
0.026
0.011
0.011
0.011
0.010
0.010
0.010
0.20
4.4
8
3
53.5
17.7
3.9
9
3
10
11
12
13
14
15
16
17
18
19
3
2
52.4
20.0
4.4
2
2
5
28.0
28.8
44.4
32.0
28.3
82.6
4
0.20
6a
6b
6c
6d
0.20
0.20
0.20
0.20
a Reaction conditions: 10.0 μmol of Ni(cod)2, 10.5 μmol of P-ligand, 100 μL of DEG (diethyleneglycole diethyl ether), required amount of 2M3BN,
4 mL of toluene, 10 min preformation of catalyst. b Determined by calibrated GC analysis with DEG as an internal standard.
2793
dx.doi.org/10.1021/om200164f |Organometallics 2011, 30, 2790–2809

Organometallics
ARTICLE
Here we report our synthesis of the new representatives of the
Triptyphos ligand family 2ꢀ627 (Figure 1) and a study of their
performance in the technically relevant nickel-catalyzed isomer-
ization of 2M3BN, showing structural and electronic impacts of
the ligand. Furthermore, details of the mechanism determined
from deuterium-labeling experiments, X-ray structures of rele-
vant intermediates involved, and DFT calculations are discussed.
the phenyl rings bearing the PPh2 substituents (Xantphos: 151°,
1: 121°) and the shorter distance between the P-moieties in
Xantphos (ranging from 4.045(1) to 4.155(1) Å,29 depending on
solid state structure).
2. Catalysis of 2M3BN Isomerization. The combination of
Octyltriptyphos 3 and Ni(cod)2 (cod = 1,5-cyclooctadiene, 3/Ni
molar ratio = 1.05, 0.2 mol % Ni loading) in toluene at 100 °C
proved to be the most suitable set of reaction conditions for the
efficient isomerization of 2M3BN (Table 1, entry 2), affording
t3PN in 93.5% yield at 96.7% conversion. In all cases complete
conversion was achieved within 20 h. More detailed information
(e.g., conversion after 3 and 20 h) may be found in the
Supporting Information. Lowering the reaction temperature to
80 and 60 °C resulted in a significant drop in activity (Table 1,
entries 3 and 4). However, conversion greater than 96% was still
achieved after 3 and 20 h, respectively. Although the total
conversion was identical at 100 and 120 °C (Table 1, entries 2
and 1), the yield of the desired product t3PN was higher at
100 °C. This is probably due to a subsequent isomerization
reaction converting t3PN to cis-3-pentenenitrile (c3PN), an
isomerization step also described in several patents.30 The
isomerization to c3PN was also observed when pure t3PN was
treated under identical conditions and did not occur without
catalyst or with just the nickel precursor Ni(cod)2. Conversion
after 1.5 h at 100 °C was unchanged when altering the 3/Ni ratio
to 2, indicating that catalyst deactivation by bis-chelation of the
Ni(0) center is not an issue in this reaction.
At a reaction temperature of 120 °C, the amount of c3PN
(7%) present indicated that the 2M3BN isomerization was
complete well before 1.5 h. We therefore reduced the catalyst
loading stepwise down to 0.011 mol %, achieving a TOF of 2850
hꢀ1 after 46% conversion (Table 1, entries 8ꢀ10). At this
catalyst loading the reaction achieved 82% conversion after 3 h
and finally 96% conversion after 20 h, representing a TON of
9400, which indicates an extremely robust catalytic system.
Addition of 1000 equiv (relative to catalyst) of 2M3BN to a
reaction mixture that had already been run for 20 h at 120 °C (0.2
mol % catalyst) led to full conversion within 1.5 h. For an
investigation of structural influences of the ligand backbone, the
reaction temperature was fixed at 100 °C in order to suppress the
formation of the undesired secondary product c3PN while
simultaneously avoiding a loss of activity at lower temperatures.
The 9,10-bridge of the ligand backbone was varied, proceeding
from the presumably most rigid 3 (96.5% after 1.5 h) via the
more flexible systems 4 (71.2%) and 5 (72.1%) to the probably
least rigid 6a (55.6%) (Table 1, entries 2, 15, 14, 16). From these
data it was concluded that a drop in activity is caused by an
increase in flexibility of the backbone, the relatively rigid tripty-
cene backbone yielding the most active ligand system.
’ RESULTS AND DISCUSSION
1. Synthesis of Diphosphine Ligands. Diphosphine ligands
1ꢀ6 were synthesized by lithiation of the corresponding bridged
1,8-dichloro-9,10-dihydroanthracene 9, followed by quenching
with ClP(p-X-C6H4)2, affording overall yields of 62ꢀ84% for X =
H and 13ꢀ23% for X = Me, OMe, and CF3. Precursor 9 was
easily produced from commercially available 1,8-dichloro-
anthrachinone 8 using a procedure reported recently by our group
(Scheme 6).7a,18aꢀ18e Triptyphos 1 was obtained in 39% overall
yield from 8. Thus introduction of the phosphine moieties by
electrophilic phosphonylation using ClPPh2 or ClPR2 in general,
as disclosed earlier,18aꢀe offers a more convenient and superior
alternative to the recently reported synthetic route featuring
nucleophilic substitution of KPPh2 with 1,8-difluorotriptycene
(20% overall yield for 1 starting from 8).19
The solid-state structures of 1 (Figure 2), 4, 5, 6c, and 6d (see
the Supporting Information) were determined by X-ray diffraction.
The PꢀP distances in the free ligands are affected by the ligand
backbone (1: 4.339 Å, 4: 4.646 Å) as well as by the para substituents
on the phosphine phenyl groups (6c: 4.396 Å, 6d: 4.628 Å).
Figure 3 depicts superposition of the solid-state structures of
Xantphos29 (10) and 1, which emphasizes the unique structural
characteristics of these ligands. The most striking difference
observed is the smaller angle between the subtending planes of
Table 2. Solvent and Lewis Acid Effects on Catalytic Activitya
entry
ligand
additive
solvent
conversion [%]b
1
2
3
4
5
6
7
3
3
3
3
4
4
4
toluene
95.0
95.0
8.0
0.4% BPh3
0.4% ZnCl2
0.4% ZnCl2
toluene
toluene
1,4-dioxane
toluene
4.3
71.2
96.2
40.0
1,4-dioxane
1,4-dioxane
10% Z-2M2BN
a Reaction conditions: 10.0 μmol of Ni(cod)2, 10.5 μmol of P-ligand,
5.00 mmol of 2M3BN, 100 μL of DEG, 4 mL of solvent, T = 100 °C,
t = 1.5 h. b Determined by calibrated GC-analysis using DEG as an
internal standard.
Figure 4. Xantphos derivatives studied for 2M3BN isomerization.
2794
dx.doi.org/10.1021/om200164f |Organometallics 2011, 30, 2790–2809

Organometallics
ARTICLE
To fine-tune the electronic properties of the P-donor atoms of
the Triptyphos-type ligands, the cyclopentano-bridged deriva-
tives 6aꢀd were investigated, since the lower activity of Ni
complexes derived from this backbone motif allowed a more
facile experimental observation of changes in activity. Substitu-
ents were introduced into the para positions of the PPh2 phenyl
groups in order to avoid steric influences. The methyl- and the
methoxy-substituted derivatives 6b (68.0%) and 6c (71.7%)
showed significantly higher conversion than unsubstituted 6a
(55.6%) (Table 1, entries 16ꢀ18). The poor results obtained for
the electron-withdrawing CF3-substituted derivative (17.4%)
apparently were caused by rapid decomposition of the preformed
(6d)Ni(cod) complex after the addition of 2M3BN.31 Similar
electronic effects on activity have been recently reported for the
related nickel-catalyzed hydrocyanation of styrene.20
Although ligand 2 combined the favorable steric and electronic
features of 3 and 6b, no significant enhancement in catalytic
activity was observed relative to 3 at 120 °C (Table 1, entries 11
and 8). Conversion at 100 °C was therefore monitored for both 2
and 3 in short intervals, and TOF values of 977 and 868 hꢀ1 were
derived for 3 and 2, respectively.
Industrial 2M3BN isomerization using phosphite ligands
usually requires a Lewis acid as a cocatalyst. However, for
example, in the case of a dppf/Ni catalyst, addition of ZnCl2 or
BEt3 decreases activity in 2M3BN-isomerization reactions
significantly.12 In the current study it was found that 3/Ni-
catalyzed 2M3BN conversion in toluene or dioxane remained
below 10% after 1.5 h in the presence of ZnCl2 (Table 2, entries 3
and 4). Conversely, activity was unaffected in the presence of
BPh3 (Table 2, entry 2). It is possible that BPh3 is sterically too
demanding or not Lewis acidic enough to interact sufficiently
with the active complexes involved in 2M3BN isomerization.
It is interesting to note that catalytic conversion for 4 increased
from 71.2% to 96.2% when 1,4-dioxane instead of toluene was
used as a solvent (Table 2, entries 5 and 6). A similar observation
in terms of activity was described for 1/Ni-catalyzed butadiene
hydrocyanation when changing from toluene to 1,4-dioxane.19
The appearance of the same solvent effect in 2M3BN isomeriza-
tion suggests that the acceleration might not be attributable to a
better solubility of butadiene in dioxane, as previously reported.19
Isomerization was also performed using 4 as a ligand in the presence
of KCN.32 The reaction rate remained unaffected, indicating an
intramolecular CNꢀ transfer in the isomerization mechanism.
In 2M3BN isomerization the formation of R,β-unsaturated
like 2-pentenenitrile (2PN) or E/Z-2-methyl-2-butenenitrile
(E/Z-2M2BN) as byproducts is a serious issue not only since
their formation affects yields but also because they act as catalyst
inhibitors. Absence of any such undesirable isomers under our
reaction conditions is especially noteworthy, since they have
been found as a major byproduct for several other chelating
phosphine ligands (Figure 4) such as DBP-Xantphos (59.6%
2PN), tBu-Xantphos (27.6% 2PN), POP-Xantphos (25.1%
2PN), and Xantphos (50.1% 2PN).7c For the alkylphosphine
ligand dtbpe (1,2-bis(di-tert-butylphosphino)ethane) exclusive
formation of 2-pentenenitriles has been reported.13 Thus, the
introduction of a bridge into the Xantphos-related 9,10-dihy-
droanthracene scaffold improves selectivity for 3PN dramatically
while at the same time significantly lowering the amount of
catalyst needed (Xantphos: 2.5% catalyst). For our ligand 4, the
addition of 10% Z-2M2BN decreased the TOF from 440 hꢀ1 to
200 hꢀ1 (determined by kinetic runs in both cases) and the
conversion from 96.2% to 40.0% (Table 2, entry 7). We interpret
this observation as indicative of Z-2M2BN acting as a reversible
inhibitor. More kinetic data or a direct 31P NMR spectroscopic
investigation of the binding of Z-2M2BN to the Ni fragment
(4)Ni(0) would be needed, however, to confirm this assumption.
3. Identity and Properties of Nickel Complexes. Treatment
of ligands 2ꢀ6 with 1 equiv of Ni(cod)2 in benzene-d6 in all
cases resulted in the formation of (PꢀP)Ni(cod) complexes
(Table 3). These one-to-one complexes were stable in solution
for several days at room temperature before forming two-to-one
bis-chelate (PꢀP)2Ni complexes. Addition of 2 equiv of 3 to
Ni(cod)2 resulted in the quantitative formation of (3)2Ni. This
Table 3. 31P{1H} NMR Chemical Shifts of (PꢀP)Ni(cod)
Complexes.a
ligand
δ [ppm]b
J [Hz]
3
21.6 (s)
4
19.6 (s)
5
16.6 (d), 17.9 (d)
18.4 (d), 19.5 (d)
16.7 (d), 17.7 (d)
15.4 (d), 16.5 (d)
18.9 (d), 20.0 (d)
44.0
44.8
43.9
45.8
31.5
6a
6b
6c
6d
a 31P{1H} NMR, 101.3 MHz, benzene-d6. b s = singlet, d = doublet.
Scheme 7
2795
dx.doi.org/10.1021/om200164f |Organometallics 2011, 30, 2790–2809

Organometallics
ARTICLE
bis-chelation mode was not reversed upon addition of another
To shed further light onto the origin of the enhanced activity
of the TTP system relative to that employing Xantphos, we
prepared the TTP analogue of (Thixantphos)NiCl2 (13), the
solid-state structure of which is available in the literature from
the work by van Leeuwen et al.21 Complex (4)NiCl2 (14)
was therefore prepared by mixing (dme)NiCl2 (dme = 1,2-
dimethoxyethane) and 4 in tetrahydrofuran. Isothermal evapora-
tion of the THF solvent led to precipitation of crystals suitable for
X-ray diffraction analysis (Figure 5). Comparing the structures of
13 and 14 showed little difference in geometry at the coordinat-
ing phosphine sites (Figure 6), suggesting that our original
assumption of enhanced backbone rigidity and blocked phenyl
group rotation in 14 could be the reason for its superior catalytic
performance. As in case of the free ligands, the most character-
istic attribute is the distinctively more bent (less flat) backbone
of 4.
equivalent of Ni(cod)2.33 By contrast, Thixantphos 11 readily
forms bis-chelate complexes showing fluxional behavior.21 1
H
NMR spectra of (4)Ni(cod) (12) revealed an η2-coordination
mode of the cod ligand (δ 2.08 (4H), 2.21 (4H), 4.30 (2H), and
5.58 (2H); see Scheme 7). DFT calculations (Turbomole,
B3LYP/TZVP//BP86/SV(P)) also supported the preference
of a trigonal-planar (4)Ni(η2-cod) over a tetrahedral (4)Ni(η4-
cod) complex.34
Figure 5. Thermal ellipsoid plot of complex 14 at the 50% probability
level. Hydrogen atoms are omitted for clarity. Selected bond lengths (Å)
and angles (deg): Ni1ꢀP1 = 2.3343(8), Ni1ꢀP2 = 2.3336(7), Ni1ꢀ
Cl1 = 2.2059(8), Ni1ꢀCl2 = 2.2070(8) ; Cl1ꢀNi1ꢀCl2 = 129.97(3)
Cl1ꢀNi1ꢀP2 = 107.68(3) Cl2ꢀNi1ꢀP2 = 99.41(3) Cl1ꢀNi1ꢀP1 =
106.81(3) Cl2ꢀNi1ꢀP1 = 100.50(3) P2ꢀNi1ꢀP1 = 112.00(3).35
Figure 6. Superposition of solid-state structures (front view) of com-
plexes 13 (white)21 and 14 (black).
Figure 7. Left: VT 31P{1H} NMR of 16a/b in THF-d8. Right: 31P{1H} NMR of 16a in CD2Cl2 at 298 K.
2796
dx.doi.org/10.1021/om200164f |Organometallics 2011, 30, 2790–2809

Organometallics
ARTICLE
Spectroscopic investigation using NMR and in situ IR spec-
troscopy allowed us to gather some insight into the nature of the
nickel complexes involved in the 2M3BN isomerization reaction
(Scheme 7). In situ IR measurements in THF at ꢀ20 °C
suggested that addition of 1 equiv of 2M3BN (ν~(CN) =
2241 cmꢀ1) to a solution of 12 very quickly resulted in formation
of 15 (ν~(CN) = 2236 cmꢀ1). Within one hour the orange
complex 15 was replaced by the red methallyl complex 16
(ν~(CN) = 2108 cmꢀ1). 31P NMR (THF-d8, ꢀ40 °C) revealed
two diastereotopic phosphorus sites (δ 21.3 and 22.1 ppm, J =
35.0 Hz) for 15, indicative of an η2-olefin complex. These
observations suggest that oxidative addition of 2M3BN to
complex 12 proceeds via the η2-olefin complex 15, rather than
through a side-on or end-on nitrile complex. Formation of
methallyl complex 16 was also observed upon the addition of
t3PN to 12, emphasizing the reversibility of the isomerization
reaction.
Since 16b was predominantly observed in the coordinating
solvent THF rather than in DCM, we wondered whether the
observed dynamic process might be caused by cyanide dissocia-
tion. The cationic complex (4)Ni(η3-C4H7)BF4 (19) was there-
fore prepared by cyanide abstraction of 16 with AgBF4. However,
the observed 31P NMR chemical shifts and the PꢀP coupling
constant of 19 (δ ꢀ0.40 and 4.17 ppm, J = 33.3 Hz, 298 K, THF-
d8) did not match those of 16b, and thus cyanide dissociation was
discarded as an explanation.
Dynamic behavior for L2Ni(η3-C4H7)CN complexes is not
unprecedented. Whereas (dppb)Ni(η3-C4H7)CN (17) under-
goes phosphine exchange with PPh3 in the presence of excess
monodentate ligand,9,41 variable-temperature 31P NMR spec-
troscopy of (dpe-phos)Ni(η3-C4H7)CN (20) also reveals the
presence of two isomers in solution.11 The authors suggest an
equilibrium between the syn and the anti isomers of the allyl
ligand, via πꢀσꢀπ rearrangement, as an explanation. However,
our own NOESY experiments (233 K, THF-d8) suggested a syn
configuration for the methallyl ligand in 16a and 16b. Moreover,
DFT calculations predicted that the syn isomer is favored over the
anti isomer by 12.1 kJ/mol (BP86/TZVP//BP86/SV(P), vide
infra).34
We therefore attribute the two isomers to rotamers with the
methyl group of the allylic ligand being oriented anti (16a) or syn
(16b) with respect to the backbone bridge (Figure 8). DFT
calculations (gas phase, vide infra) suggested a preference of the
anti conformer by 13.8 kJ/mol over the syn conformer. On the
basis of this analysis we assigned the conformer that is predomi-
nant in DCM as the anti conformer 16a. Reductive elimination of
these two rotamers would result in the linear cyano olefin or the
branched one, respectively, assuming that reductive elimination
occurs directly from the η3-methallyl complex.
31P NMR spectroscopy of 16 showed dynamic behavior in
THF-d8 (Figure 7). At room temperature a broad signal at δ 16.8
was observed, which at 253 K split into two pairs of doublets with
an intensity distribution of 1:1. These isomers will be referred to
as 16a (δ 13.9 and 18.8, J = 119.0 Hz) and 16b (δ 7.6 and 16.9,
J = 144.9 Hz).36 When a sample of isolated complex 16 was
dissolved in CD2Cl2 or toluene-d8, the isomer 16a was present
almost exclusively.37 Addition of THF to a solution of 16 in
CD2Cl2 led to the emergence of 16b again; however, a consider-
able excess of THF was needed.
1H NMR (213 K, THF-d8) chemical shifts for 16a/b differed
significantly, with 16b showing an unexpected hydrogen upfield
shift at C1 of the methallyl ligand (16b: δ ꢀ0.34 (C1), 0.00
(C1), 1.83 (C4), 4.02 (C3), 4.91 (C2); 16a: δ 1.35 (C1), 1.41
(C4) 2.25 (C1), 3.14 (C3), 4.29 (C2)).38,39 HMQC NMR(233 K,
THF-d8) showed chemical shifts very similar to complexes
(dppb)Ni(η3-C4H7)CN (17) and (PPh3)2Ni(η3-C4H7)CN
(18) (Table 4).9,40 The 13C NMR chemical shifts for 16, 17,
and 18 reflect a distorted coordination mode of the methallyl
ligand, induced by its methyl substituent, toward a significant
weight of the σ,π-resonance form (Scheme 7).
31P NMR spectra for (3)Ni(η3-C4H7)CN42 (21) revealed
slightly different dynamics than those displayed by 16. A solution
of 21 in toluene-d8 contained two isomers in a ratio of 3:7 (δ 10.3
and 17.7, J = 137.2 Hz vs δ 16.5 and 20.4, J = 131.5 Hz, all peaks
considerably broadened). In THF-d8 very broad signals corre-
sponding to the two diastereomers were already observed at
room temperature. Both experiments indicated a lowering of the
activation barrier for the allyl rotation upon replacing ligand 4
with the catalytically more active ligand 3.
To our delight, we were able to obtain crystals from a solution
of 16 suitable for X-ray diffraction analysis (Figure 9). Structural
analysis revealed only one rotamer, 16a, with the methallyl ligand
slightly disordered about its two enantiomeric positions (10%
and 90%).43
Table 4. 13C NMR Data of (ligand)Ni(η3-C4H7)CN
Complexes
position
16a/ba
22a
17b
18c
C1
C2
C3
C4
48.2/36.9
99.5
41.8
101.3
93.9
44.8
96.2
81.6
20.5
51.7
108.7
86.5
The molecular structure of 16a confirmed the distortion of the
C4H7 group toward a [η1:η2] coordination mode as derived from
the 13C NMR data (Table 5 and Scheme 7). The Ni1ꢀC63 bond
is elongated significantly compared to the Ni1ꢀC61 (2.17 vs
2.02 Å) bond, and the C61ꢀC62 bond (1.44 Å) is 7 pm longer
87.37/82.9
19.9
20.9
19.6
a HMQC, 125.7 MHz, THF-d8, 213 K. b 13C{1H} NMR, 100.6 MHz,
THF-d8, 263 K.9 c 13C{1H} NMR, 100.6 MHz, toluene-d8, 273 K.9
Figure 8. Rotamers of complex 16.
2797
dx.doi.org/10.1021/om200164f |Organometallics 2011, 30, 2790–2809

Organometallics
ARTICLE
than C62ꢀC63 (1.37 Å). Furthermore, the structure features a
significant PꢀNi bond length difference of 17 pm, which is in
excellent agreement with the observed distinction of the 31P
NMR resonances of 16.
signals at 8.8 and 17.1 ppm (J = 126.7 Hz) and the other at 11.0
and 21.1 ppm (J = 94.1 Hz), in a ratio of 1:9 at room temperature.
IR spectroscopy showed a higher value for the CN vibration
(ν~(CN) = 2132 cmꢀ1) compared to 16, which is in good
agreement with a Lewis acid coordinating to the cyanide ligand.
The identity of this complex was also confirmed by X-ray
diffraction analysis,45 revealing a μ2-chloro-bridged dimer
Similar bonding properties (NiꢀP, NiꢀC) are also known
for ((dpe-phos)Ni(η3-C4H7)CN ZnCl2 EtOH (22)11 and
3
3
(dppf)Ni(η3-C4H7)CN BEt3 (23);12 however, in the current
3
[(4)Ni(η3-C4H7)CN ZnCl2]2, 24 (Figure 10). The methallyl
system these features are far more pronounced, possibly as a
result of the larger PꢀP distance (16a: 3.84 Å, 22: 3.60 Å, 23:
3.61 Å), which enforces, due to the rigid ligand backbone, a rather
obtuse PꢀNiꢀP angle (16a: 116.2°, 22 and 23: 106.1°).
To investigate the question of why isomerization is retarded in
the presence of ZnCl2, a solution of complex 16 in THF-d8 was
stirred with 1 equiv of ZnCl2 for 1 h. Two complexes were
subsequently observed by 31P NMR spectroscopy, one giving
3
ligand is disordered in a ratio of 61:39.
The core geometries of 16a and 24 were found to be nearly
identical (Figure 11). However, close analysis of the bonding
properties around the Ni center of 24 revealed a smaller
difference in the NiꢀP bond lengths of only 10 pm, as well as
shortened C61ꢀC62 and Ni1ꢀC63 bonds and elongated
C62ꢀC63 and Ni1ꢀC61 bonds, indicating a more clear-cut
η3-coordination mode than in complex 16a.
The chloro nickel analogue of 16 was prepared by adding rac-
3-chloro-1-butene to a solution of complex 12 in benzene,
yielding 25 (Scheme 8). The 31P NMR spectrum of 25 indicated
a dynamic behavior similar to its cyanide analogue 16 by a broad
pair of doublets in THF-d8 (δ 2.19 and 13.23, J = 78.0 Hz) and
toluene-d8 (δ 1.65 and 13.50, J = 99.1 Hz) at room temperature.
The observed broad resonances split at 213 K (THF-d8) into two
pairs of doublets (δ ꢀ4.5 and 21.2, J = 98.0 Hz/δ ꢀ1.8 and 12.5,
J = 89.3 Hz), possibly caused by rotation of the allylic ligand
featuring rotamers 25a/b, as observed in 16a/b. 13C NMR
resonances in toluene-d8 for the allylic ligand were very similar
to those observed for 16a/b (δ 37.0 (C1), 102.8 (C2), 88.5
(C3), 19.1 (C4)). The molecular structure (Figure 12)47 of
isolated complex 25a is very similar to that of 16a, but the
methallyl ligand displays an even slightly more pronounced η3-
coordination mode.
4. Deuterium-Labeling Experiments. In Druliner’s mecha-
nistic experiment (Scheme 3, vide supra), ZnCl2 acted as a cocatalyst
for the Ni[P(O-p-tolyl)3]4-catalyzed isomerization of 2M3BN.
However, as mentioned above, addition of ZnCl2 retards the
isomerization catalyzed by [(PꢀP)Ni(cod)] complexes derived
from bidentate phosphine ligands such as 3, 4, or dppf12,14
(Table 2). This observed inhibition is in contrast to expectations
one would associate with a mechanism involving a cascade of de-
and rehydrocyanation as proposed for Ni[P(O-p-tolyl)3]4. To
shed light on this issue, we isomerized the deuterated system49
2M3BN-d at 80 °C in benzene with catalytic amounts of 3/
Ni(cod)2 (0.2 mol % Ni(cod)2, 3/Ni = 1.05). The reaction was
Figure 9. Thermal ellipsoid plot of complex 16a at the 50% probability
level. Hydrogen atoms are omitted for clarity. Selected bond lengths (Å)
and angles (deg): Ni1ꢀP1 = 2.3756(15), Ni1ꢀP2 = 2.2073(15),
Ni1ꢀC61 = 2.018(13), Ni1ꢀC62 = 2.020(6), Ni1ꢀC63 = 2.171(6),
Ni1ꢀC65 = 1.900(6), C61ꢀC62 = 1.436(13), C62ꢀC63 = 1.367(9),
C63ꢀC64 = 1.535(12), C65ꢀN65 = 1.135(7); C65ꢀNi1ꢀC63 =
92.1(2), C65ꢀNi1ꢀP2 = 93.41(17), C65ꢀNi1ꢀP1 = 96.61(18),
C63ꢀNi1ꢀP2 = 142.7(2), C63ꢀNi1ꢀP1 = 99.7(2), P2ꢀNi1ꢀP1 =
116.21(6).44
Table 5. Selected Calculated (BP86/SV(P)) Atomic Distances (in Å) of Complexes Involved in the Catalytic Cycle of 2M3BN
Isomerization
parameter
A
TS1
B
TS2
C
D
TS3
E
TS4
F
NiꢀC1
NiꢀC2
NiꢀC3
NiꢀC4
NiꢀC5
NiꢀH(C4)
C84ꢀH
C1ꢀC2
C2ꢀC3
C3ꢀC4
C5ꢀC1
1.9798
1.9982
3.0227
4.3687
3.1154
2.1601
2.0078
2.2248
3.5449
1.8894
2.0641
2.0288
2.1962
3.3013
1.8636
2.0672
2.0445
2.2295
3.4250
1.9597
2.0857
2.0235
2.1684
3.3337
1.8644
3.7648
1.1111
1.4161
1.4228
1.5102
2.6265
4.0579
3.0059
2.0280
2.1642
1.8366
1.6715
1.2195
1.3556
1.4637
1.4828
3.6915
4.1034
3.0759
2.1527
2.1185
1.8683
1.5004
1.7231
1.3546
1.4613
1.4187
3.7403
4.1675
3.1616
2.2726
2.1701
1.8810
1.4841
2.2193
1.3545
1.4601
1.3940
2.1518
2.0274
2.2840
3.4486
1.8538
3.7316
1.1122
1.4791
1.3980
1.5074
1.8036
3.0087
1.9993
1.9985
3.0447
3.1636
1.4258
1.5359
1.5504
2.9952
1.4057
1.4766
1.5258
2.6233
1.4321
1.4101
1.5042
1.4206
1.4166
1.5035
1.5328
1.4265
1.5180
1.4677
2798
dx.doi.org/10.1021/om200164f |Organometallics 2011, 30, 2790–2809

Organometallics
ARTICLE
monitored by 2H NMR spectroscopy after 0.5, 1.0, 1.5, 3, and 20 h.
Full conversion to 3PN-d was achieved within 3 h. After 20 h at
80 °C no resonances at δ 3.06, indicative for D-scrambling, were
observed (Figure 13).
However, a detailed analysis of the mechanistic steps most
likely involved revealed that a de- and rehydrocyanation mechan-
ism does not necessarily produce D-scrambling (Scheme 9). In
addition to β-hydride elimination to produce 27 from 26,
isomerization to 28—either by butadiene dissociation and
recoordination or by an intramolecular process (e.g., a hapto-
tropic shift)—is essential to observe D-scrambling. Without a 27
to 28 isomerization, reinsertion of butadiene into the NiꢀD
bond will eventually produce only 31.
In order to detect intermediate 27 indirectly, the isomerization
reaction was performed in the presence of 1 equiv of 1,1,4,4-
tetradeuteriobuta-1,3-diene (BD-d) in relation to the amount of
2M3BN substrate (Scheme 10, catalyst: 0.2 mol % 3/Ni(cod)2,
80 °C, toluene). Within 3 h full conversion was achieved. GC/
MS analysis of the reaction mixture showed no increased
incorporation of deuterium into the 3-pentenenitrile product,
and H NMR spectroscopy, using benzene (C6H6) for calibra-
tion, also revealed no significant deuterium enrichment. On the
basis of these two D-labeling experiments we feel confident that
2M3BN isomerization does not proceed via a cascade of de- and
rehydrocyanation steps.
2
To search for intermediate Ni(η1-butenyl) complexes, we
chose 25 as a model for 16. The averaged 1H NMR resonances
(toluene-d8, room temperature) for the allylic ligand’s C1 posi-
tion in rotamers 25a/b were well separated (δ 1.16 and 1.92,
toluene-d8). The observed dynamic behavior (Scheme 8) could
be caused either by rotation of the allylic ligand or by πꢀσꢀπ
interconversion. We figured that use of rac-(Z)-1-deuterio-3-
chloro-1-butene instead of rac-3-chloro-1-butene to produce 25
from 12 could provide a convenient way to detect intermediate
[(4)Ni(η1-butenyl)CN] complexes involved in πꢀσꢀπ inter-
conversion indirectly. If the interconversion proceeded by
rotation only, addition of rac-(Z)-1-deuterio-3-chloro-1-butene
to 12 should produce 32 and 34 exclusively. Involvement
of η1-complex 33, however, is expected to also yield 35
(Scheme 11). For the actual experiment we used rac-(Z)-1-
deuterio-3-chloro-1-butene, which was 40% overdeuterated at
the C1 position (Scheme 11). One equivalent was added to
complex 12 in benzene or toluene at room temperature, which
led to a mixture of 32 and 35, indicated in the 2H NMR by two
very broad overlapping resonances at δ 1.28 and 1.92. Integra-
tion suggested a 1:1 ratio, indicative for the involvement of 33 as
part of an πꢀσꢀπ interconversion mechanism between 32 and
34, rather than a 2.5:1 ratio (due to overdeuterated starting
material), which would indicate the absence of 33. However, due
Figure 10. Solid-state structure of dimer complex 24. Hydrogen atoms
are omitted for clarity. Selected bond lengths (Å) and angles (deg):
Ni1ꢀP1 = 2.3528(9), Ni1ꢀP2 = 2.2533(9), Ni1ꢀC61 = 2.051(19),
Ni1ꢀC62 = 2.008(6), Ni1ꢀC63 = 2.152(6), Ni1ꢀC65 = 1.878(4),
C61ꢀC62 = 1.412(11), C62ꢀC63 = 1.373(9), C63ꢀC64 = 1.552(14),
C65ꢀN65 = 1.152(4), N65ꢀZn1 = 1.960(3), Zn1ꢀCl1 = 2.3604(10),
Zn1ꢀCl2 = 2.1863(10); C65ꢀNi1ꢀP2 = 91.54(10), C65ꢀNi1ꢀP1 =
97.95, Ni1ꢀP1 = 112.79(4).46
Figure 11. Superposition of monomer complex 16a (black) and the
dimer complex 24 (white). Only one unit of 24 is shown for clarity.
Scheme 8
2799
dx.doi.org/10.1021/om200164f |Organometallics 2011, 30, 2790–2809

Organometallics
ARTICLE
to the of use of 40% overdeuterated starting material, the
possibility that these broad and overlapping signals hide an
unknown impurity that might falsify the integrals cannot be
excluded.
5. DFT Calculations. On the basis of the experimental results,
which indicated an intramolecular mechanistic pathway with the
involvement of complexes 15 and 16, we used QM calculations
(DFT) in order to model the catalytic cycle of the 2M3BN
isomerization reaction, with the ethano-bridged Rucaphos (4) as
a ligand.34
Figure 14 depicts the free energy profile found for the reaction.
Relevant bonding parameters for all minima and transition states
involved are given in Table 5, and the atomic coordinates for all
structures may be found in the Supporting Information. The
Gibbs free energy, ΔG, of the overall 2M3BN to t3PN isomer-
ization was calculated as ꢀ16.6 kJ/mol. Starting from the
trigonal-planar 2M3BN olefin complex A (15), oxidative addi-
tion was predicted to proceed through a late transition state TS1
(ΔG‡ = þ51.1 kJ/mol) to the methallyl complex B (16a). The
calculated structural parameters of complex B were found to be in
good agreement with the experimentally determined structure of
16a; however, the distortion toward a σ,π-coordination mode of
the organic moiety (Scheme 7) was less pronounced than in the
crystal structure.
For the transition from B to C (= 16b), two alternative
mechanisms were reasonable: a πꢀσꢀπ rearrangement, as
it was found to be possible for chloro-complex 25 (Scheme 11),
or simple rotation of the methallyl ligand. Extensive screening
for (4)Ni(η1-nBu)CN species revealed exclusively complexes
stabilized by agostic interactions such as in D, which were found
unsuitable for the B to C transformation (ΔG‡ values varied
from þ75.0 to þ118.5 kJ/mol). By contrast, simple rotation of
the methallyl ligand (TS2) featured a low activation barrier
ΔG‡ value of þ51.1 kJ/mol. The fact that a suitable πꢀσꢀπ
rearrangement transition state could not be located does not
exclude its existence; on the other hand, the calculated ΔG values
for all (4)Ni(η1-C4H7)CN minima found were well above the
Figure 12. Thermal ellipsoid plot of complex 25a at the 50% probability
level. Most hydrogen atoms are omitted for clarity. Selected bond
lengths (Å) and angles (deg): Ni1ꢀP1 = 2.396(2), Ni1ꢀP2 =
2.249(2), Ni1ꢀC61 = 2.033(8), Ni1ꢀC62 = 1.998(8), Ni1ꢀC63 =
2.198(9), Ni1ꢀCl1 = 2.273(2), C61ꢀC62 = 1.369(11), C62ꢀC63 =
1.387(11), C63ꢀC64 = 1.551(10); C61ꢀNi1ꢀC63 = 67.8(3),
C62ꢀNi1ꢀP2 = 107.2(2), C62ꢀNi1ꢀCl1 = 112.7(3), C63ꢀNi1ꢀ
P2 = 144.3(2), Cl1ꢀNi1ꢀP1 = 93.82(7), P2ꢀNi1ꢀP1 = 113.83(8).48
Figure 13. 2H{1H} NMR (46.1 MHz, C6H6, room temperature); 2M3BN-d isomerization after 20 and 0.5 h (inset).
2800
dx.doi.org/10.1021/om200164f |Organometallics 2011, 30, 2790–2809

Organometallics
ARTICLE
Scheme 9
Scheme 10
Scheme 11
energy of TS2, and therefore in our view the rotation mechanism
appears to be the most likely pathway, at least for an intramolecular
process.
From complex C (16b) onward, the reaction pathway can
proceed via either dehydrocyanation (D, TS3, E) or reductive
elimination (TS4). Dehydrocyanation is initiated by the
2801
dx.doi.org/10.1021/om200164f |Organometallics 2011, 30, 2790–2809

Organometallics
ARTICLE
Figure 14. Calculated (BP86/TZVP//BP86/SV(P)) free energy (ΔG) profile for the catalytic cycle of 2M3BN isomerization using Rucaphos (4) as a
ligand.
formation of an η1-iBu complex D, which is stabilized by an
agostic interaction. Facilitated by this agostic interaction, β-
hydride elimination can proceed via TS3, resulting in butadiene
complex E. It is interesting to note that complex E, according to
our DFT results may lose 1,3-butadiene in an exergonic reaction
with a very low reaction barrier of ca. 9 kJ/mol, supporting the
results of our exchange experiment with deuterated butadiene
(Scheme 10). Such a butadiene loss has been reported for
Triphos (bis(diphenylphosphinoethyl)phenylphosphine) and is
often invoked as a catalyst decomposition pathway in 2M3BN
isomerization.50 Rehydrocyanation of E will again lead to C
via D. A relatively facile reductive elimination finally yields the
t3PN complex F.
’ EXPERIMENTAL SECTION
All reactions were carried out under an atmosphere of dry argon with
standard Schlenk-tube and syringe-cannula techniques. Toluene, THF,
and n-pentane were dried by distillation from sodium/benzophenone,
dichloromethane by distillation from CaH2, and acetone by distillation
from K2CO3. Prior to use solvents (including H2O) were saturated with
argon. Ni(cod)2,51 ClP(4-Me-C6H4)2,52 ClP(4-OMe-C6H4)2,52 ClP(4-
CF3ꢀC6H4)2,52 1,8-dichloroanthracene,7a 1,8-dichlorotriptycene,7a
1,8-dichloro-10-n-octyltriptycene,7a 1,8-dichloro-9,10-(ethano)-9,10-
dihydroanthracene,7a 1,8-dichloro-9,10-(cis-1,2-cyclopentano)-9,10-
dihydroanthracene,7a and 1,1,4,4-tetradeuteriobuta-1,3-diene53 were
prepared according to published procedures. Lithium powder (sodium
content 2.4%, dispersion in hexanes, Chemetall GmbH),54 2M3BN, and
authentic samples of the other nitriles were donations of BASF SE.
2M3BN and diethyleneglycol diethyl ether (DEG) were distilled under
an argon atmosphere, degassed trice applying the pumpꢀfreeze tech-
nique, and dried over molecular sieves (4 Å). High-pressure
DielsꢀAlder reactions were performed in a steel autoclave with Teflon
insert (100 mL, Berghoff). NMR spectra were recorded using a Bruker
DRX 250, DRX 300, or DRX 500 spectrometer with both the 31P and
13C spectra being measured in the 1H decoupled mode. Chemical shifts
are given in ppm referenced to solvent (1H, 13C) or an 85% solution of
H3PO4 (31P). Abbreviations used are s = singlet, d = doublet, t = triplet,
dt = doublet of triplets, m = multiplet, bs = broad signal, p = pseudo.
’ CONCLUSION
In a straightforward way a series of ligands of the Triptyphos
family have been synthesized and applied to the industrially
relevant 2M3BN isomerization reaction. With a turnover fre-
quency of up to 2850 hꢀ1 under optimized conditions, ligand 3 is,
to the best of our knowledge, the most active catalyst for this
reaction reported to date. The reaction proved to be highly
sensitive to the steric and (partially) to the electronic properties
of the ligand.
Relevant intermediates of the reaction have been isolated and
characterized structurally and spectroscopically. Deuterium-la-
beling experiments clearly show that the widespread general
assumption of cascade de/rehydrocyanation steps proposed to
operate for Ni[P(O-p-tolyl)3]4/ZnCl2 as a catalyst system is not
the mechanistic pathway for our types of ligands. Furthermore, a
Gibbs free energy profile of the reaction was predicted by DFT
for a Rucaphos-containing catalyst, which is in good agreement
with our experimental observations.
1
Assignments of resonances in the H NMR are orientated to IUPAC
nomenclature. However, in some cases assignments depart from IUPAC
for the sake of readability. Relevant 1H NMR spectra with illustrations
clarifying the assignments used may be found in the Supporting
Information. Gas chromatography was performed on a Varian auto-
sampler chromatograph equipped with a CP-Wax 58 CB column (30 m,
carrier gas 50 kPa of N2, FID detector). 2M3BN and t3PN, were
calibrated against DEG as an internal standard. FT-IR spectra were
recorded on a Bruker Equinox 55. Mass spectroscopy was performed on
a Jeol JMS-700 (CI, FAB (matrix: nitrobenzyl alcohol), LIFDI55).
Melting points were measured using a melting point apparatus by
Dr. Tottoli and are uncorrected.
2802
dx.doi.org/10.1021/om200164f |Organometallics 2011, 30, 2790–2809

Organometallics
ARTICLE
1,8-Bis(diphenylphosphino)triptycene (1). A suspension of
Li powder (380 mg, 54.8 mmol, 12 equiv, containing 2.4% of sodium) in
THF (20 mL) was cooled to ꢀ78 °C, and a cooled solution of 1,8-
dichlorotriptycene (1.50 g, 4.60 mmol, 1 equiv) in THF (40 mL) was
added. The resulting slurry was stirred for 6 h at ꢀ78 °C. Then excess Li
powder was removed by filtration followed by the addition of chloro-
diphenylphosphine (2.04 g, 9.25 mmol, 2 equiv). The solution was
allowed to warm to room temperature overnight. After evaporation of
the solvent the resulting solid was suspended in degassed H2O (35 mL)
and filtrated. The solid was dissolved in dichloromethane (50 mL) and
dried over MgSO4. After removal of MgSO4 and evaporation of the
solvent, the solid was washed with acetone (25 mL) and dried in vacuo.
Yield: 2.30 g (3.66 mmol, 80%) of a white solid. Mp: 275 °C. 1H NMR
(500.1 MHz, CD2Cl2): δ 5.43 (s, 1H, H10), 5.81 (d, 1H, J = 7.4 Hz, H9),
6.50 (m, 2H), 6.56 (t, 1H, J = 3.7 Hz), 6.60 (dt, 1H, J = 7.7 Hz, J = 1.0 Hz,
H13), 6.89 (m, 3H), 7.23 (m, 12H), 7.23 (m, 11H). 13C{1H} NMR
1,8-dichloro-10-n-octyltriptycene (3.0 g, 6.9 mmol, 1 equiv) in THF
(40 mL) was added. The resulting slurry was stirred for 7 h at ꢀ78 °C.
Then excess Li powder was removed by filtration followed by the addition
of chlorodiphenylphosphine (3.04 g, 13.8 mmol, 2 equiv). The solution
was allowed to warm to room temperature overnight. After evaporation of
the solvent the resulting solid was suspended in H2O (50 mL) and
filtrated, and thesoliddissolved indichloromethane (50 mL). The organic
was layer was extracted with H2O (3 ꢁ 50 mL) and dried over MgSO4.
After the separation of MgSO4 and evaporation of the solvent the solid
was washed with acetone (10 mL) and n-pentane (15 mL). Yield: 3.13 g
(4.26 mmol, 62%) of a white solid. Mp: 176ꢀ177 °C. 1H NMR (500.1
MHz, CD2Cl2): δ 0.90 (t, 3H, J = 7.0 Hz, H24), 1.34 (m, 4H, H23, H22),
1.41 (m, 2H, H21), 1.51 (m, 2H, H20), 1.76 (dt, 2H, J =7.7 Hz, J = 7.4 Hz,
H19), 2.09 (m, 2H, H18), 2.88 (dd, 2H, J = 7.4 Hz, J = 8.0 Hz, H17), 5.77
(d, 1H, J = 7.0 Hz, H9), 6.48 (d, 2H, J = 7.7 Hz, H4, H5), 6.56 (t, 1H, J =
7.4 Hz), 6.63 (dd, 1H, J = 4.4 Hz, J = 4.0 Hz), 6.89 (m, 4H), 7.19ꢀ7.37
(m, 19H), 7.35 (d, 2H, J = 7.4 Hz, H2, H7). 13C{1H} NMR (125.7 MHz,
CD2Cl2): δ 14.2 (C24), 23.0 (C23), 25.6 (C18), 28.7 (C17), 29.7 (20),
30.0 (C21), 32.0 (C22), 32.3 (C19), 49.81 (t, C9, J = 21.7 Hz), 54.27
(C10), 122.0, 123.2, 124.5 (d, J = 9.4 Hz), 124.6, 125.0, 128.8 (pt, J = 3.8
Hz), 128.9 (pt, J = 3.8 Hz), 129.0, 129.1, 129.2, 132.8 (dt, Cquart, J = 6.6
Hz, J = 7.5 Hz), 134.5 (pt, J = 10.4 Hz), 134.6 (pt, J = 10.4 Hz), 137.0 (pt,
Cquart, J = 65.7 Hz), 137.5 (pt, Cquart, J = 65.7 Hz), 145.7 (Cquart), 147.2
(Cquart), 151.2 (Cquart). 31P{1H} NMR (202.5 MHz, CD2Cl2): δ ꢀ15.5
(s). MS (FABþ): m/z 734.3 (Mþ). IR (KBr): ~ν/cmꢀ1 3053, 2952, 2923,
2852, 1585, 1475, 1433, 734. Anal. Calcd for C52H48P2: C, 84.99; H, 6.58;
P, 8.43. Found: C, 84.81; H, 6.60; P, 8.16.
1,8-Bis(diphenylphosphino)-9,10-ethano-9,10-dihydro-
anthracene (4). At ꢀ78 °C a solution of 1,8-dichloro-9,10-(ethano)-
9,10-dihydroanthracene (3.0 g, 11 mmol, 1 equiv) in THF (60 mL)
was added to a suspension of Li powder (1.0 g, 0.14 mmol, 13 equiv,
sodium content 2.4%) in THF (40 mL) and stirred for 6 h at this
temperature. After removal of the Li powder by filtration at ꢀ78 °C,
chlorodiphenylphosphine (4.77 g, 21.8 mmol, 2 equiv) was added by
syringe. The solution was then allowed to warm to room temperature
overnight. After removal of all volatile compounds the resulting white
solid was elutriated in degassed H2O (50 mL) and filtrated, and the
remaining solid dissolved in CH2Cl2 (100 mL) and dried over
MgSO4. The product was obtained after crystallization from acetone
and final washing with pentane as a white solid (4.3 g, 7.48 mmol,
69%). Mp: 230 °C. 1H NMR (500.1 MHz, CD2Cl2): δ 1.09 (m, 2H,
H11), 1.56 (m, 2H, H12), 4.32 (s, 1H, H10), 5.73 (m, 1H, H9), 6.63
(m, 2H, H2, H7), 6.97 (m, 2H, H3, H6), 7.15 (m, 4H), 7.22 (d, 2H,
H4, H5, J = 7.7 Hz), 7.25ꢀ7.32 (m, 16 H). 13C{1H} NMR (75.5
MHz, CDCl3): δ 24.2 (C11), 26.2 (C12), 39.2 (C9), 44.6 (C10),
123.8, 125.5, 128.3, 129.9, 129.2, 132.1 (C17, C23, C29, C33), 133.9,
137.2 (C3, C5), 137.7 (C2, C7), 144.0 (C1, C8), 147.6 (C4a, C10a),
147.9 (C8a, C9a). 31P{1H} NMR (202.5 MHz, CD2Cl2): δ ꢀ16.88.
MS (EIþ): m/z 574.2 (Mþ), 246.0 (Mþ ꢀ C2H4). IR (KBr): ν~/cmꢀ1
3053 (m), 2950 (m), 2584 (m), 1477 (m), 1433 (s), 1416 (m), 745 (s).
Anal. Calcd for C40H32P2: C, 83.61; H, 5.61; P, 10.78. Found: C, 83.48;
H, 5.68; P, 10.80.
3
(125.8 MHz, CD2Cl2): δ 49.8 (t, C9, JP,C = 19.8 Hz), 54.6 (C10),
124.2, 124.6, 124.8, 125.0, 128.7 (pt, J = 3.8 Hz), 128.8 (pt, J = 2.9 Hz),
128.9, 129.0, 129.2, 133.2 (dd, Cquart, J = 8.5 Hz, J = 7.4 Hz), 134.5 (td,
J = 10.4 Hz, J = 4.7 Hz), 136.7 (pt, Cquart, J = 5.7 Hz), 137.2 (pt, Cquart, J =
5.7 Hz), 143.7 (Cquart), 145.7 (Cquart), 150.0 (pt, Cquart, J = 2.8 Hz),
149.3 (pt, Cquart, J = 14.5 Hz). 31P{1H} NMR (202.5 MHz, CD2Cl2):
δ ꢀ14.7 (s). MS(CIþ): m/z 622.3 (Mþ), 439.3(Mþ ꢀ PPh2). IR(KBr):
ν~/cmꢀ1 3052, 1948, 1583, 1433. Anal. Calcd for C44H32P2: C, 84.87; H,
5.18; P, 9.95. Found: C, 84.70; H, 5.21; P, 9.89.
1,8-Bis[di(4-methylphenyl)phosphino]-10-n-octyltripty-
cene (2). A suspension of Li powder (689 mg, 99.3 mmol, 8 equiv,
containing 2.4% of sodium) in THF (40 mL) was cooled to ꢀ78 °C, and
a cooled solution of 1,8-dichloro-10-n-octyltriptycene (5.38 g, 12.4
mmol, l equiv) in THF (80 mL) was added. The resulting slurry was
stirred for 6 h at ꢀ78 °C. Then excess Li powder was removed by
filtration followed by the addition of chlorodi(4-methylphenyl)phos-
phine (6.15 g, 24.7 mmol, 2 equiv) in THF (10 mL). The solution was
allowed to warm to room temperature overnight. After evaporation of
the solvent the resulting solid was suspended in CH2Cl2 (60 mL),
washed with degassed H2O (3 ꢁ 50 mL), and dried over MgSO4. After
removal of MgSO4 and evaporation of the solvent the solid was
crystallized from acetone and n-pentane. The product was obtained as
a white solid (2.12 g, 2.68 mmol, 22%). Mp: 166 °C. 1H NMR (300.1
3
MHz, CD2Cl2): δ 0.95 (t, J = 7.0 Hz, 3H, H24), 1.40 (m, 4H, H21,
H22), 1.55 (m, 4H, H21, H20), 1.80 (m, 3J = 7.5 Hz, 2H, H19), 2.13 (m,
2H, H18), 2.36 (pd, J = 4.1 Hz, 12H, Ar-Me), 2.91 (t, 3J = 7.6 Hz, 2H,
H17), 5.84 (d, 3J = 7.2 Hz, 1H, H12 or H15), 6.52 (d, 3J = 7.0 Hz, 2H,
H2/7 or H4/5), 6.61 (t, J = 7.6 Hz, 1H, H9), 6.62 (m, 1H, H13 or H14),
6.90 (t, 3J = 7.5 Hz, 1H, H14 or H13), 6.91 (t, 3J = 7.6 Hz, 2H, H3, H6),
7.03ꢀ7.20 (m, 16H, H25, H26), 7.32 (d, 3J = 7.5 Hz, 1H, H15 or H12),
7.37 (d, 3J = 7.3 Hz, 2H, H2/7 or H4/5). 13C{1H} NMR (75.5 MHz,
CD2Cl2): δ 14.27 (C24), 21.33 (Ar-Me), 21.33 (Ar-Me), 23.07 (C23),
25.63 (C18), 28.80 (C17), 29.71 (C21), 30.07 (C20), 32.06 (C19),
3
32.30 (C22), 49.71 (t, JPC = 20.6 Hz, C9), 54.26 (Cquart10), 122.03
(C12 or C15), 122.91 (C2/7 or C4/5), 124.30, 124.58 (C15 or C12),
124.62, 124.89, 128.98, 129.55 (dd, J = 3.7 Hz, J = 3.1 Hz, C25/26),
134.12 (pt, J = 4.8 Hz, Cquart), 134.49 (dd, J = 10.5 Hz, J = 9.9 Hz, C25/
26), 133.45 (Cquart), 138.97 (d, J = 7.4 Hz, Cquart), 145.83 (Cquart),
147.18 (Cquart), 147.65 (Cquart), 150.99 (pt, J = 14.8 Hz, Cquart).
1,8-Dichloro-9,10-(exo-cis-1,2-norbornano)-9,10-dihydro-
anthracene. A 100 mL steel autoclave was charged with a solution of
1,8-dichloroanthracene (10.0 g, 48.6 mmol, 1 equiv) and 2-norbornene
(15.0 g, 138 mmol, 3 equiv) in toluene (35 mL), pressurized with 40 bar
of argon, and heated to 135 °C for 5 days. After the removal of all volatile
compounds the crude product was purified by flash chromatography
(petroleum ether, ambient conditions), giving the product as a white
31P{1H} NMR (121.5 MHz, CD2Cl2): δ ꢀ17.33 (s). MS (LIFDI):
þ
m/z 790.3 (Mþ 3 ), 806.13 (M þ O), 822.25 (Mþ þ2O). IR (KBr):
ν~/cmꢀ1 808 (s), 1399 (m), 1454 (m), 1496 (m), 1596 (w), 1908 (w),
2360 (w), 2850 (m), 2918 (s), 2962 (m), 3011 (w), 3067 (w). Anal.
Calcd for C56H56P2: C, 85.03; H, 7.14; P, 7.83. Found: C, 84.87; H, 7.16;
P, 7.71.
3
3
1
solid (14.1 g, 41.3 mmol, 85%). Mp: 173 °C. H NMR (300.1 MHz,
CDCl3): δ ꢀ0.34 (d, 1H, H17, 2J = 10.6 Hz), ꢀ0.43 (d, 1H, H17, 2J =
10.7 Hz), 1.05 (m, 2H, H14), 1.38 (m, 2H, H15), 1.88 (dt, 1H, H13, 3J =
8.4 Hz, 3J = 2.3 Hz), 1.90 (dt, 1H, H16, 3J = 8.8 Hz, 3J = 2.6 Hz), 2.01 (s,
1,8-Bis(diphenylphosphino)-10-n-octyltriptycene (3). A sus-
pension of Li powder (479 mg, 69.0 mmol, 10 equiv, containing 2.4% of
sodium) in THF (40 mL) was cooled to ꢀ78 °C, and a cooled solution of
3
1H, H12), 2.14 (s, 1H, H11), 4.28 (d, 1H, H10, J = 2.6 Hz), 5.38
2803
dx.doi.org/10.1021/om200164f |Organometallics 2011, 30, 2790–2809

Organometallics
ARTICLE
3
(d, 1H, H9, J = 2.6 Hz), 6.98ꢀ7.02 (m, 6H). 13C{1H} NMR (75.4
(d, 1H, H10, 3J = 2.8 Hz), 5.99 (pq, 1H, H9, 3J = 3.4 Hz, 4JPH = 5.1 Hz),
6.63 (ddd, 1H, H6, 3J = 7.9 Hz, 3J = 7.8 Hz, 4J = 1.0 Hz), 6.83 (ddd, 1H,
H3, 3J = 7.9 Hz, 3J = 7.8 Hz, 4J = 1.0 Hz), 6.94 (pt, 1H, H7, 3J = 7.7 Hz),
7.03 (m, 3H), 7.11 (m, 2H), 7.20 (m, 8H), 7.30 (m, 9H). 13C{1H} NMR
(125.7 MHz, CD2Cl2): δ 27.5 (C14), 30.34 (C13), 30.6 (C15), 44.4 (t,
C9, 3JPC = 21.7 Hz), 44.9 (C12), 45.3 (C11), 50.2 (C10), 124.4 (C5),
125.8 (C6), 125.9 (C3), 126.7 (C4), 128.4, 128.5 (m), 130.5 (C7),
131.6 (C2). 123.3 (C10a), 132.4 (C4a), 133.7, 134, 134.2, 134.3, 13.6
(d, Cquart, 1JPC = 12.3 Hz), 138.1 (d, Cquart, 1JPC = 12.3 Hz), 138.4 (d,
1JPC = 12.2 Hz), 142.7 (d, C8a, 2JPC = 6.6 Hz), 145.5 (d, C9a, 2JPC = 5.6
MHz, CDCl3): δ 30.7 (C14), 33.8 (C15), 33.2 (C17), 39.2 (C12), 39.6
(C11), 41.1 (C9), 47.8 (C10), 48.6 (C13), 49.1 (C16), 121.6 (C7),
122.8 (C2), 126.1 (C6), 126.5 (C5), 126.7 (C3), 127.0 (C4), 129.4
(C8), 130.7 (C1), 138.7(C8a), 140.8 (C9), 141.2 (C10), 146.5 (C4a).
MS (FABþ): m/z 341.2 (Mþ), 329.2 (Mþ ꢀ CH2), 307.3 (Mþ ꢀ Cl).
IR (KBr): ν~/cmꢀ1 3060 (w), 3025 (w), 2973 (s), 2956 (s), 2945 (s),
2898 (s), 2871 (s), 1598 (m), 1451 (s), 1434 (m), 1168 (m), 773 (s).
Anal. Calcd for C21H18Cl2: C, 73.91; H, 5.32; Cl, 20.78. Found: C, 73.81,
H, 5.33, Cl, 20.94.
1
1
Hz), 148.3 (d, C8, JPC = 32.9 Hz), 149.7 (d, C1, JPC = 30.2 Hz).
31P{1H} NMR (202.5 MHz, CD2Cl2): δ ꢀ18.74 (d, 1P, J = 13.6 Hz),
ꢀ20.90 (d, 1P, J = 13.6 Hz). MS (FABþ): m/z 615.3 (Mþ). IR (KBr):
ν~/cmꢀ1 3048 (m), 2949 (s), 2930 (s), 2902 (s), 2860 (m), 1584 (m),
1477 (m), 1432 (s), 1416 (s), 742 (s), 659 (s). Anal. Calcd for
C43H36P2: C, 84.02; H, 5.90; P, 10.08. Found: C, 83.81; H, 6.10;
P, 10.03.
1,8-Bis(diphenylphosphino)-9,10-(exo-cis-1,2-norbornano)-
9,10-dihydroanthracene (5). At ꢀ78 °C a solution of 1,8-dichloro-
9,10-(norbornano)-9,10-dihydroanthracene (1.5 g, 4.4 mmol, 1 equiv)
in THF (25 mL) was added to a suspension of Li powder (280 mg,
40.3 mmol, 9 equiv, sodium content 2.4%) in THF (25 mL) and stirred
for 6 h at this temperature. After removal of the Li powder by filtration at
ꢀ78 °C chlorodiphenylphosphine (1.93 g, 8.75 mmol, 2 equiv) was
added by syringe. The solution was then allowed to warm to room
temperature overnight. After removal of all volatile compounds the
resulting white solid was elutriated in degassed H2O (25 mL) and
filtrated, the remaining solid was dissolved in CH2Cl2 and dried over
MgSO4, and the solvent was removed in vacuo. The product was
obtained after crystallization from acetone and final washing with
pentane as a white solid (2.38 g, 3.7 mmol, 84%). Mp: 231ꢀ234 °C.
1H NMR (500.1 MHz, CD2Cl2): δ ꢀ0.43 (d, 1H, H17a, 2J = 10.4 Hz),
0.17 (d, 1H, H17b, 2J = 10.4 Hz), 0.66 (m, 1H, H14), 0.89 (m, 1H, H15),
1.04 (m, 1H, H14), 1.19 (d, 1H, H13, 3J = 3.7 Hz), 1.22 (m, 1H, H15),
1.47 (dd, 1H, H12, 3J = 8.8 Hz, 3J = 2.6 Hz), 1.74 (dd, 1H, H11, 3J = 8.8
Hz, 3J = 2.9 Hz), 1.90 (d, 1H, H16, 3J = 3.7 Hz), 4.22 (d, 1H, H10, J = 2.9
Hz), 6.02 (m, 1H, H9), 6.66 (ddd, 1H, H7, 3J = 7.6 Hz, 4J = 3.8 Hz, J =
1.3 Hz), 6.76 (ddd, 1H, H2, 3J = 7.6 Hz, 4J = 3.8 Hz, J = 1.3 Hz), 6.93 (pt,
1H, H6, 3J = 7.4 Hz, 3J = 7.7 Hz), 6.98 (pt, 1H, H3, 3J = 7.7 Hz, 3J = 7.4
Hz), 7.03 (m, 2H, H4, H5), 7.12ꢀ7.25 (m, 10H), 7.28ꢀ7.36 (m, 10H).
13C{1H} NMR (125.7 MHz, CD2Cl2): δ 30.9 (C14), 31.1 (C15), 33.6
(C17), 39.4 (C13), 40.1 (C16), 43.8 (t, 1C, C9, 3JPC = 22.6 Hz), 48.4
(C12), 49.0 (C11), 49.5 (C10), 124.1(C5), 125.3 (C6), 125.8 (C4),
126.1 (C3), 128.3, 128.4, 128.5, 128.5, 128.6, 128.6 (2C), 128.7, 128.8,
128.9, 130.4 (C7), 131.7 (C2), 132.1 (d, Cquart, JPC = 13.2 Hz), 133.1 (d.
1C, JPC = 13.2 Hz), 136.7, 136.8, 134.0, 134.2, 134.2, 134.4, 134.6, 134.7,
136.6 (pt, Cquart, JPC = 11.3 Hz), 138.1 (d, Cquart, JPC = 11.3 Hz), 138.4
(d, Cquart, JPC = 11.3 Hz), 142.9 (d, C8a, 2JPC = 5.6 Hz), 145.2 (d, C9a,
1,8-Bis[di(4-methylphenyl)phosphino]-9,10-(cis-1,2-cy-
clopentano)-9,10-dihydroanthracene (6b). Having been dried
over CaCl2 in the desiccator for several days and subsequently under
high vacuum for 12 h, 1,8-dichloro-9,10-(cis-1,2-cyclopentano)-9,10-
dihydroanthracene (2.00 g, 6.34 mmol, 1 equiv) was dissolved in THF
(25 mL). The solution was slowly added to a suspension of lithium
powder (440 mg, 63.0 mmol, 10 equiv, sodium content 2.4%) in THF
(20 mL) at ꢀ78 °C. The formerly colorless solution turned deep brown
within 30 min and was subsequently stirred for a further 6.5 h at ꢀ78 °C.
The reaction mixture was then filtered through a frit to remove the
excess of lithium, and at ꢀ78 °C chlorodi(4-methylphenyl)phosphine
(3.23 g, 13.0 mmol, 2 equiv) was added dropwise from a syringe over 20
min. The solution was warmed to room temperature overnight under
continuous stirring. The solvent was removed from the pale yellow
solution in vacuo, and the residue was dissolved in dichloromethane
(100 mL). The solution was washed with H2O (3 ꢁ 30 mL), dried over
MgSO4 for one hour, and filtered off through a frit. Subsequently, the
solvent was removed in vacuo, and the remaining yellow solid was
crystallized from acetone and repeatedly washed with pentane. The
product was obtained as a white powder, still containing 0.2 equiv of
acetone, which could not be removed even after several days under high
vacuum. Yield: 1.006 g (1.500 mmol, 23%). Mp: 258ꢀ260 °C. 1H NMR
(500.1 MHz, CD2Cl2): δ 0.66ꢀ0.79 (m, 2H, H13/H14), 0.83ꢀ0.97
(m, 1H, H15), 1.05ꢀ1.16 (m, 1H, H14), 1.24ꢀ1.36 (m, 1H, H13),
1.55ꢀ1.68 (m, 1H, H15), 2.04ꢀ2.10 (m, 1H, H12), 2.31 (s, 3H, Ar-Me),
2.32 (s, 3H, Ar-Me), 2.35 (s, 6H, Ar-Me), 2.39ꢀ2.41 (m, 1H, H11), 4.13
(m, 1H, H10), 5.96 (dd, 1H, H9, J = 7.5 Hz, J = 5.0 Hz), 6.63 (m, 1H),
6.82 (m, 1H), 6.89ꢀ6.95 (m, 3H), 6.98ꢀ7.05 (m, 7H), 7.13ꢀ7.24 (m,
10H). 13C{1H} NMR (125.8 MHz, CD2Cl2): δ 21.4 (Ar-Me), 27.7
(C14), 30.5 (C13), 30.8 (C15), 44.4 (t, C9, 3JP,C = 45 Hz), 45.0 (C12),
45.5 (C11), 50.3 (C10), 124.2 (C5), 125.7 (C6), 125.8 (C3), 126.7
(C4), 129.4, 129.4, 129.5, 129.5, 130.5 (C7), 131.6 (C2), 132.6 (d,
C10a, 3JPC = 10 Hz) 133.1 (d, C4a, 3JP,C = 13 Hz), 134.0 (d, JP,C = 34
Hz), 134.1 (d, JP,C = 33 Hz), 134.3 (d, JP,C = 37 Hz), 134.4ꢀ135.13 (m,
2JPC = 5.7 Hz), 147.9 (d, C10a, 3JPC = 31.1 Hz), 149.7 (d, C4a, 3JPC
=
30.1 Hz). 31P{1H} NMR (202.4 MHz, CD2Cl2): δ ꢀ20.85 (d, 1P, J =
17.4 Hz), ꢀ18.20 (d, 1P, J = 17.4 Hz). MS (FABþ): m/z 640.3 (Mþ). IR
(KBr): ν~/cmꢀ1 3053 (m), 2948 (m), 2953 (m), 2867 (m), 1585 (s),
1478 (m), 1434 (s), 740 (s). Anal. Calcd for C45H38P2: C, 84.90; H,
5.58; P, 9.52. Found: C, 84.64; H, 6.03; P, 9.60.
1,8-Bis(diphenylphosphino)-9,10-(cis-1,2-cyclopentano)-
9,10-dihydroanthracene (6a). At ꢀ78 °C a solution of 1,8-
dichloro-9,10-(cis-1,2-cyclopentano)-9,10-dihydroanthracene (1.5 g,
4.8 mmol, 1 equiv) in THF (70 mL) was added to a suspension of Li
powder (350 mg, 50.4 mmol, 10 equiv, sodium content 2.4%) in THF
(30 mL) and stirred for 6 h at this temperature. After removal of the Li
powder by filtration at ꢀ78 °C chlorodiphenylphosphine (2.1 g, 0.10
mol, 2 equiv) was added by a syringe. The solution was then allowed to
warm to room temperature overnight. After removal of all volatile
compounds the resulting white solid is elutriated in degassed H2O
(20 mL) and filtrated. The remaining solid was dissolved in dichlor-
omethane and dried over MgSO4, and the solvent removed in vacuo. The
product was obtained after crystallization from acetone and final washing
with pentane as a white solid (2.05 g, 3.34 mmol, 70%). Mp: 234 °C. 1H
NMR (500.1 MHz, CD2Cl2): δ 0.85 (m, 1H, H13a), 0.89 (m, 1H,
H15a), 0.93 (m, 1H, H14a), 1.27 (m, 1H, H15b), 1.67 (m, 1H, H13b),
1.73 (m, 1H, H14b), 2.40 (m, 1H, H11), 2.42 (m, 1H, H12), 4.16
C
quart), 138.4 (Cquart), 138.6 (Cquart), 138.9 (Cquart), 142.74 (d, 2JP,C
=
6.0 Hz, C8a), 145.5 (d, 2JC,P = 5.0 Hz, C9a), 148.2 (d, C8, 1JC,P = 31.0
Hz), 149.7 (d, C1, JC,P = 31.0 Hz). 31P{1H} NMR (202.5 MHz,
1
CD2Cl2): δ ꢀ20.7 (d, 1P, J = 13.0 Hz), ꢀ20.6 (d, 1P, J = 13.0 Hz). IR
(KBr): ν~/cmꢀ1 3064 (w), 2935 (s), 1496 (s), 1415 (m), 1185 (m), 1090
(m), 1019 (m), 804 (s). MS (FABþ): m/z 670.3 (Mþ). Anal. Calcd for
C27H44P2 0.2 C3H6O: C, 83.78; H, 6.68; P, 9.08. Found: C, 83.45; H,
3
6.65; P, 8.74.
1,8-Bis[di(4-methoxyphenyl)phosphino]-9,10-(cis-1,2-cy-
clopentano)-9,10-dihydroanthracene (6c). 1,8-Dichloro-9,10-(cis-
1,2-cyclopentano)-9,10-dihydroanthracene (2.40 g, 7.61 mmol, 1 equiv)
was dried, first for three days in a desiccator over CaCl2 and subsequently
overnight under high vacuum. Next, the substance was dissolved in THF
2804
dx.doi.org/10.1021/om200164f |Organometallics 2011, 30, 2790–2809

Organometallics
ARTICLE
(60 mL) and the colorless solution was added dropwise to a stirred
suspension of lithium powder (0.56 g, 80 mmol, 10 equiv, sodium content
2.4%) in THF (30 mL) at ꢀ78 °C. The suspension was then stirred
at ꢀ78 °C for another six hours, during which the solution quickly turned
first yellow, then orange, and finally dark brown. The excess of lithium was
removed by filtration through a frit. At ꢀ78 °C a solution of chlorobis(4-
methoxyphenyl)phosphine (4.79 g, 17.1 mmol, 2.3 equiv) in THF
(20 mL) was added dropwise, and the then pale red solution was warmed
to room temperature overnight. Subsequently, the solvent was removed in
vacuo, the residue was diluted in dichloromethane (50 mL), and the
suspension was washed with degassed H2O (3 ꢁ 30 mL). The pale red
solution was dried over MgSO4, the solvent was removed in vacuo, and the
resulting solid was crystallized from acetone. The product was obtained as a
white powder. Yield: 0.73 g (0.99 mmol, 13%). Single crystals for X-ray
analysis were obtainedbycovering a saturatedsolution in dichloromethane
witha layerof pentane. Mp: 263ꢀ265 °C. 1H NMR(300.1 MHz, CDCl3):
δ 0.67 (m, 2H, H13, H14), 0.86 (m, 1H, H15), 1.05 (m, 1H, H14), 1.26
(m, 1H, H13), 1.56 (m, 1H, H15), 2.04 (m, 1H, H12), 2.32 (m, 1H, H11),
3.73 (s, 3H, Ar-OMe), 3.74 (s, 3H, Ar-OMe), 3.79 (s, 6H, Ar-OMe), 4.08
(m, 1H, H10), 5.92 (dd, 1H, H9, J = 7.8 Hz, 3J = 4.8 Hz), 6.61 (m, 1H),
6.72 (m, 5H), 6.85 (m, 4H), 6.95 (m, 6H), 6.21 (m, 6H). 13C{1H} NMR
(75.5 MHz, CDCl3): δ 27.6 (C14), 30.3 (C13), 30.6 (C15), 44.1 (t, C9,
3JC,P = 21.0 Hz), 44.8 (C12), 45.4 (C11), 50.2 (C10), 55.3 (Ar-OMe), 55.4
(Ar-OMe), 55.5 (Ar-OMe), 114.2 (d, 3JC,P = 3.0 Hz), 114.2 (d, 3JC,P = 2.0
Hz), 114.3 (d, 3JC,P = 4.0 Hz) 114.4 (d, 3JC,P = 4.0 Hz), 124.0 (C5), 125.6
(C6), 125.7 (C3), 126.4 (C4), 128.5, 128.7, 129.0, 129.1, 129.1, 129.9,
(m, C, CF3), 148.6 (d, C1/C8, 1JC,P = 33.0 Hz), 150.1 (d, C8/C1, 1JC,P
=
30.0 Hz). 31P{1H} NMR (202.5 MHz, CD2Cl2): δ ꢀ20.1 (d, 1P, J =
13.0 Hz), ꢀ18.5 (d, 1P, J = 13.0 Hz). 19F{1H} NMR (282.4 MHz,
CD2Cl2): δ ꢀ63.9 (s, 3F, Ar-CF3), ꢀ63.8 (s, 6F, Ar-CF3), ꢀ63.7 (s, 3F,
Ar-CF3). IR (KBr): ν~/cmꢀ1 3057 (w), 2955 (m), 1606 (m), 1418 (w),
1396 (m), 1325 (s), 1165 (s), 1122 (s), 1062 (m), 1017 (m). MS
(FABþ): m/z 886.2 (Mþ). Anal. Calcd for C47H32P2F12: C, 63.66; H,
3.64; P, 6.99. Found: C, 63.76; H, 3.70; P, 6.85.
(3)Ni(η2-cod). In a glovebox (argon) Ni(cod)2 (37.9 mg, 138 μmol,
1 equiv) was added as a solid to a solution of 3 (98.8 mg, 135 μmol, 1
equiv) in toluene (2.5 mL) in a Schlenk tube. The solution was stirred
for 5 min at room temperature and then set aside overnight, during
which an orange precipitate formed. The solid was filtered, washed with
n-pentane, and dried in vacuo. Yield: 67.7 mg (75.1 μmol, 56%) of an
orange solid contaminated with traces of [(3)2Ni]. Mp: 150 °C (dec).
1H NMR (500.1 MHz, C6D6): δ 0.95 (t, 3H, J = 6.9 Hz), 1.32 (bs, 8H),
1.58 (bs, 2H), 2.08 (s, 8H, CH2-CHdCH-CH2), 2.30 (bs, 2H), 2.80 (t,
2H, J = 7.4 Hz), 4.30 (s, 2H, CH2-CHdCH-CH2), 5.56 (s, 2H, CH2-
CHdCH-CH2), 6.62ꢀ6.73 (bs, 5H), 6.89 (m, 1H), 6.95 ꢀ 7.12 (bs,
14H), 7.26ꢀ7.37 (m, 8H), 7.78 (bs, 4H). 13C{1H} NMR (125.8 MHz,
C6D6): δ 14.8, 23.5, 26.1, 28.7 (CH2-CHdCH-CH2), 29.1, 30.1, 30.4,
31.2, 32.4, 32.7, 48.1 (t, J = 18.7 Hz), 55.1, 90.0 (CH2-CHdCH-CH2),
122.7, 123.4, 124.8, 125.3, 125.8, 128.9, 129.0, 129.2, 129.6, 129.8, 129.9,
131.13, 131.2, 131.4, 133.5, 133.5, 133.6, 136.0, 136.1, 136.1, 136.3,
136.5, 136.7. 31P{1H} NMR (202.5 MHz, C6D6): δ 21.14 (s). MS
(FABþ): m/z 824.4 [M ꢀ cod þ Cl]þ (40), 789.4 [M ꢀ cod]þ (100),
751.5 [3 þ O]þ (100), 734.5 [3]þ (25). IR (KBr): ν~/cmꢀ1 3053 (w),
3002 (m), 2955 (s), 2851 (m), 1637 (m), 1479 (m), 1457 (s), 1402 (m),
1225 (m), 1120 (m), 896 (w), 790 (m), 765 (s), 695 (vs), 545 (m), 479
(s).
129.4, 129.5, 130.0, 131.1, 131.1, 133.6, 133.8, 134.9, 135.2, 135.2 (d, 1JC,P
15.0 Hz), 135.5 (d, 1JC,P = 16 Hz), 135.6 (d, 1JC,P = 12 Hz), 135.9 (d, 1JC,P
=
=
12.0 Hz), 142.7 (d, C8a, 2JC,P = 7.0 Hz), 145.5 (d, C9a, 2JC,P = 5.0 Hz),
147.7 (d, C8, 1JC,P = 32.0 Hz), 149.2 (d, 1JC,P = 29.0 Hz), 160.3, 160.3,
160.5, 160.5. 31P{1H} NMR (121.5 MHz, CDCl3): δ ꢀ23.6 (d, J = 12.0
Hz), ꢀ22.1 (d, J = 12.0 Hz). IR (KBr): ~ν/cmꢀ1 3017 (w), 2947 (s), 1595
(m), 1567 (w), 1497 (s), 1290 (m), 1248 (s), 1175 (m), 1033 (w). MS
(FABþ): m/z 735.3 ([M þ H]þ). Anal. Calcd for C47H44P2O4: C, 76.81;
H, 6.05; P, 8.43. Found: C, 76.50; H, 5.92; P, 8.08.
General Procedure for [(ligand)Ni(cod)] Complex Pre-
paration. In a glovebox (argon) Ni(cod)2 and P-ligand were combined
in a 5 mL glass vial equipped with a magnetic stir bar. C6D6 or THF-d8
was added, and the solution was stirred for 5 min at room temperature.
The now yellow solution was transferred to an NMR tube, and spectra
were recorded within 20 min.
1,8-Bis[di(4-trifluoromethylphenyl)phosphino]-9,10-(cis-
1,2-cyclopentano)-9,10-dihydroanthracene (6d). 1,8-Di-
chloro-9,10-(cis-1,2-cyclopentano)-9,10-dihydroanthracene (2.42 g,
7.69 mmol, 1 equiv) was dried overnight under high vacuum and then
solved in THF (30 mL). At ꢀ78 °C the solution was slowly added to a
stirred suspension of lithium powder (350 mg, 50.0 mmol, 6.5 equiv,
sodium content 2.4%) in THF (20 mL). Subsequently, the suspension
was stirred for another six hours at ꢀ78 °C; during this time the formerly
colorless solution turned dark brown. The excess of lithium was
removed by filtration through a frit. Next, chlorobis(4-trifluoromethyl-
phenyl)phosphine (3.42 g, 9.59 mmol, 1.2 equiv) was added dropwise to
the stirred solution, and the reaction mixture was warmed to room
temperature overnight under continuous stirring. Subsequently the
solvent was removed in vacuo, the residue was diluted in dichloro-
methane (30 mL), and the suspension was then washed with degassed
H2O (2 ꢁ 30 mL). The now pale red solution was dried over MgSO4,
and the solvent was removed in vacuo. Finally the resulting solid was
crystallized from acetone, repeatedly washed in pentane, and dried under
high vacuum for several days. The product was obtained as a white
powder. Yield: 1.40 g (1.58 mmol, 21%). Single crystals for X-ray
analysis were obtained by slowly removing solvent from a saturated
solution of the product in dichloromethane at room temperature. Mp:
(4)Ni(η2-cod) (12). 4 (8.4 mg, 15 μmol, 1 equiv), Ni(cod)2 (4.1 mg,
15 μmol, 1 equiv), and C6D6 (0.5 mL) were used. 1H NMR (250.1 MHz,
C6D6): δ 1.61 (bs, 2H, H11 or H12), 1.80 (bs, 2H, H12 or H11), 2.08 (s,
4H, CH2-CHdCH-CH2), 2.21 (s, 12H, CH2-CHdCH-CH2), 3.99 (s,
1H, H10), 4.30 (bs, 2H, CH2-CHdCH-CH2), 5.58 (bs, 6H, CH2-
CHdCH-CH2), 6.53 (bs, 1H), 6.74ꢀ6.87 (bs, 4H), 7.05 (bs, 14H), 7.30
(bs, 3H), 7.48 (bs, 1H), 7.92 (bs, 4H). 31P{1H} NMR (101.3 MHz,
C6D6): δ 18.6 (s). MS(LIFDI): m/z 667.2 [4 þ Ni þ Cl]þ (100), 574.3
[4]þ (100).
(2)Ni(η2-cod). 2 (15.0 mg, 19.0 μmol, 1 equiv), Ni(cod)2 (5.2 mg,
19 μmol, 1 equiv), and C6D6 (0.5 mL) were used. 1H NMR (300.1 MHz,
C6D6): δ 0.94 (t, 3H, J = 6.8 Hz), 1.32 (bs, 8H), 1.58 (m, 2H), 1.95 ꢀ
2.09 (m, 8H, CH2-CHdCH-CH2), 2.02 (bs, 3H, Ar-Me), 2.04 (s, 3H,
Ar-Me), 2.07 (bs, 6H, Ar-Me), 2.21 (s, 8H, free cod), 2.38 (bs, 2H), 2.80
(t, 2H, J = 7.2 Hz, H17), 2.66ꢀ3.01 (bs, 1H, CH2-CHdCH-CH2),
3.32ꢀ3.69 (bs, 1H, CH2-CHdCH-CH2), 4.30 (s, 1H), 5.53ꢀ5.65 (m,
1H, CH2-CHdCH-CH2), 5.58 (s, 4H, free cod), 5.77 (d, 1H, J = 7.2 Hz,
CH2-CHdCH-CH2), 6.64 ꢀ 6.76 (bs, 5H), 6.81 ꢀ 7.01 (bs, 9H),
7.22ꢀ7.41 (bs, 7H), 7.61ꢀ7.71 (bs, 1H), 7.76 (bs, 3H), 8.02 (s, 1H).
13C{1H} NMR (75.5 MHz, C6D6): δ 14.8, 21.5, 21.6, 26.1, 28.7 (free
cod), 29.1, 30.1, 30.4, 31.2, 32.4 32.7, 70.7 (CH2-CHdCH-CH2), 121.7,
122.3, 123.8, 124.1, 124.2 (CH2-CHdCH-CH2), 124.3 (CH2-CHdCH-
CH2), 124.6, 127.7, 128.1, 128.4 (free cod), 128.8, 128.9, 135.5, 132.5,
135.2, 135.4. 31P{1H} NMR (101.3 MHz, C6D6): δ 22.1 (s).
(5)Ni(η2-cod). 5 (20.0 mg, 31.2 μmol, 1 equiv), Ni(cod)2 (8.6 mg,
31 μmol, 1 equiv), and C6D6 (0.5 mL) were used. 1H NMR (500.1 MHz,
C6D6): δ 0.18 (d, 1H, J = 11.3 Hz), 0.24 (d, 1H, J = 11.3 Hz), 0.30
(s, 2H), 0.56 (bs, 1H), 0.86 (bs, 1H), 1.02 (bs, 1H), 1.12 (bs, 3H), 1.80
1
265ꢀ266 °C. H NMR (300.1 MHz, CD2Cl2): δ 0.76 (m, 2H, H13,
H14), 0.93 (m, 1H, 15), 1.18 (m, 1H, H14), 1.43 (m, 1H, H13), 1.69 (m,
1H, H15), 2.20 (m, 1H, H12), 2.45 (m, 1H, H11), 4.24 (d, 1H, H10, 3J =
1.6 Hz), 5.98 (dd, 1H, H9, J = 7.8 Hz, 3J = 4.1 Hz), 6.63 (m, 1H), 6.84
(m, 1H), 7.02ꢀ7.50 (m, 16 H), 7.63 (m, 4H). 13C{1H} NMR (125.8
MHz, CD2Cl2): δ 27.6 (C14), 30.6 (C13), 30.7 (C15), 44.2 (t, C9,
3JC,P = 23.0 Hz), 44.6 (C12), 44.8 (C11), 50.2 (C10), 121.0ꢀ136.0
2805
dx.doi.org/10.1021/om200164f |Organometallics 2011, 30, 2790–2809

Organometallics
ARTICLE
(bs, 4H, CH2-CHdCH-CH2), 1.87 (s, 2H, CH2-CHdCH-CH2), 2.08
(s, 10H, CH2-CHdCH-CH2), 3.94 (bs, 1H), 4.30 (bs, 2H, CH2-
CHdCH-CH2), 5.58, (bs, 4H, CH2-CHdCH-CH2), 6.64 (bs, 1H),
6.71ꢀ6.79 (m, 4H), 6.87 (1H, s), 6.91 (d, 2H, J = 6.9 Hz), 6.98ꢀ7.13
(bs, 14 H), 7.30 (t, 2H, J = 7.8 Hz), 7.42 (t, 2H, J = 8.0 Hz), 8.02 (bs,
4H). 13C{1H} NMR (125.8 MHz, C6D6): δ 25.3, 28.7 (free cod), 30.1,
31.6, 31.5, 31.7, 34.7, 39.3, 41.2, 41.5 (t, J = 17.0 Hz), 46.4, 48.6, 49.8,
124.7, 125.1, 125.2, 125.4, 128.8, 128.9, 129.2 (free cod), 130.0, 130.1,
130.2, 131.0, 131.1, 132.2, 132.3, 133.1, 133.2, 133.3, 133.4, 136.9, 136.5,
136.8, 137.0, 137.3, 137.4, 137.5, 144.2, 146.5, 149.4, 149.5, 151.4, 151.5.
31P{1H} NMR (202.5 MHz, C6D6): δ 16.7 (d, J = 46.2 Hz), 18.0 (d, J =
46.2 Hz).
THF-d8, 213 K): δ 19.87 (C-D*, C-D), 21.89 (C11*, C11, C12), 25.94
(C12*), 40.48 (C9*), 43.72 (C10*), 48.17 (C-A), 36.85 (C-A*), 82.93
(C-C), 87.37 (C-C*), 99.50 (C-B*), 127.74, 126.08ꢀ132.67,
133.70ꢀ137.27. 31P{1H} NMR (202.46 MHz, THF-d8, 213 K): δ
7.64 (d, 1P, J = 144.9 Hz), 13.91 (d, 1P, J = 119.0 Hz), 16.87 (d, 1P, J =
144.9 Hz), 18.75 (d, 1P, J = 119.0 Hz). IR (KBr): ν~/cmꢀ1 3054, 2954,
2869, 2102 (CN), 1479, 1434, 1092, 1067, 746, 398, 518. MS (LIFDI):
m/z 667.1 (4NiClþ), 574.2 (4). Anal. Calcd for C45H39NNiP2 1.5
3
THF: C, 74.46; H, 6.25; P, 7.53. Found: C, 74.56; H, 6.45; P, 7.46.
(4)Ni(η3-C4H7)BF4 (19). A Schlenk tube was charged with a
solution of complex 16a (24 mg, 35 μmol) in THF (1 mL). A solution
of AgBF4 (ca. 6 mg, 31 μmol) in THF was added. Instantly a black
precipitate was formed, which was filtrated via canula. From the orange
solution all volatiles were removed in vacuo, and the resulting solid was
dissolved in THF-d8 (0.6 mL). 31P{1H} NMR (101.3 MHz, THF-d8):
δ ꢀ0.4 (d, J = 33.1 Hz), 4.2 (d, J = 33.1 Hz).
(6a)Ni(η2-cod). 6a (9.0 mg, 15 μmol, 1 equiv), Ni(cod)2 (4.8 mg,
17 μmol, 1 equiv), and C6D6 (0.5 mL) were used. 1H NMR (250.1 MHz,
C6D6): δ ꢀ0.11 (bs, 1H), 0.52 (bs, 1H), 0.93 (bs, 3H), 1.55 (bs, 1H),
1.82 (bs, 3H), 2.08 (s, 4H, CH2-CHdCH-CH2), 2.21 (bs, 3H, CH2-
CHdCH-CH2), 2.21 (s, 8H, free cod), 2.39 (bs, 1H, CH2-CHdCH-
CH2), 3.86 (s, 1H), 4.30 (bs, 2H, CH2-CHdCH-CH2), 5.58 (bs, 3H,
CH2-CHdCH-CH2), 5.58 (s, 4H, free cod), 6.70ꢀ7.07 (m, 20H), 7.40
(bs, 2H), 8.01 (bs, 4H). 31P{1H} NMR (101.3 MHz, C6D6): δ 18.5 (d,
J = 44.8 Hz), 19.5 (d, J = 44.8 Hz).
(3)Ni(η3-C4H7)CN (21). In a glovebox (argon) a flame-dried
Schlenk tube was charged with 3 (268.1 mg, 365.0 μmol, 1 equiv) and
Ni(cod)2 (100.4 mg, 365.0 μmol, 1 equiv). The solids were dissolved in
toluene (7 mL), and the solution was then overlaid with n-pentane
(2 mL). After 7 d at room tempertaure a red solid precipitated, which
was isolated by filtration, washed with n-pentane, and dried in vacuo (187
mg, 214 μmol, 59%). Mp: 133 °C (dec). 31P{1H} NMR (202.5 MHz,
toluene-d8): δ 10.3 (d, 1P, J = 137.2 Hz), 16.5 (d, 2P, J = 131.5 Hz), 17.7
(d, 1P, J = 137.2 Hz), 20.4 (d, 2P, J = 131.5 Hz). IR (KBr): ν~/cmꢀ1 3053
(m), 2962 (s), 2922 (s), 2851 (m), 2108 (w), 1631 (vs), 1601 (vs), 1479
(s), 1434 (vs), 1351 (s), 1261 (m), 1117 (vs), 1024 (vs), 875 (m),
801(s), 744 (m), 696 (m).
(6b)Ni(η2-cod). 6b (7.9 mg, 12 μmol, 1 equiv), Ni(cod)2 (3.2 mg,
12 μmol, 1 equiv), and C6D6 (0.5 mL) were used. 1H NMR (250.1 MHz,
C6D6): very broad signals. 31P{1H} NMR (101.3 MHz, C6D6): δ 16.7
(d, J = 43.9 Hz), 17.7 (d, J = 43.9 Hz).
(6c)Ni(η2-cod). 6c (9.8 mg, 14 μmol, 1 equiv), Ni(cod)2 (3.7 mg,
14 μmol, 1 equiv), and C6D6 (0.5 mL) were used. 1H NMR (250.1 MHz,
C6D6): very broad signals. 31P{1H} NMR (101.3 MHz, C6D6): δ 15.4
(d, J = 45.8 Hz), 16.4 (d, J = 45.8 Hz).
[(4)Ni(η3-C4H7)CN ZnCl2]2 (24). A Schlenk tube was charged
3
(6d)Ni(η2-cod). 6d (13.4 mg, 15 μmol, 1 equiv), Ni(cod)2 (4.2 mg,
15 μmol, 1 equiv), and C6D6 (0.5 mL) were used. 1H NMR (250.1 MHz,
C6D6): very broad signals. 31P{1H} NMR (101.3 MHz), C6D6): δ
15.5ꢀ18.3 (broad multiplet), 18.9 (d, J = 31.4 Hz), 20.1 (d, J = 31.4 Hz).
(4)NiCl2 (14). A Schlenk tube was charged with 4 (20 mg, 34 μmol)
and (dme)NiCl2 (7.6 mg, 34 μmol) as solids. These were dissolved in as
little THF as possible. The violet solution formed was covered with a
layer of n-pentane. From this solution violet crystals precipitated, which
were suitable for X-ray analysis.
with Ni(cod)2 (96.3, 0.350 mmol, 1 equiv) and 4 (201.3 mg, 0.350
mmol, 1 equiv) as solids, which were dissolved in dry THF (30 mL). An
orange-red solution was instantly formed, which was stirred for 5 min
before 2M3BN (72 μL, 0.35 mmol, 1 equiv) was added by means of a
Hamilton syringe. The solution turned instantly dark red and was stirred
for another 5 min before a solution of ZnCl2 (41.4 mg, 0.301 mmol, 0.85
equiv) in THF (3 mL) was added. The resulting solution was stored
overnight at ꢀ25 °C. About 25 mL of the volatile compounds was then
removed in vacuo and the solution again stored at ꢀ25 °C. A red solid
precipitated, which was isolated by filtration and washed with n-pentane.
1H NMR (500.1 MHz, THF-d8, 233 K): δ 0.49 (s, 1H, H11a), 0.76 (s,
1H, H11b), 1.40 (s, 3H, H-D), 1.63 (m, 2H, H12), 1.96 (bs, 1H, H-A),
2.59 (s, 1H, H-A), 3.43 (s, 1H, H-C), 4.31 (s, 1H, H-B), 5.59 (s, 1H,
H9), 6.89 ꢀ 8.20 (m, 26H). 13C{1H} NMR (HMQC, 125.7 MHz,
THF-d8, 233 K): δ 20.85 (C-D), 22.46 (C11), 26.64 (C12), 40.89 (t, J =
16.3 Hz, C9), 41.84 (C-A), 44.46 (C10), 93.86 (C-C), 101.27 (C-B),
125.90, 127.75, 128.97, 129.26, 130.25, 131.13, 131.27, 133.15, 133.70,
133.89, 135.02ꢀ136.90 (bs), 145.34, 145.58, 147.78, 148.21, 148.35,
155.89. 31P{1H} NMR (202.5 MHz, THF-d8, 233 K): δ 8.84 (d, 12P, J =
126.7 Hz), 11.02 (d, 100P, J = 94.1 Hz), 17.08 (d, 12P, J = 126.7 Hz),
21.12 (d, 100P, J = 94.05 Hz). (KBr): ν~/cmꢀ1 3055 (m), 2948 (m),
2132 (s), 4780 (m), 1435 (s), 1138 (w), 1093 (m), 1026 (w), 773 (s),
698 (s). Anal. Calcd for C45H39NNiP2ZnCl2: C, 63.53; H, 4.62. Found:
C, 64.77; H, 4.94.
(4)Ni(η2-2M3BN) (15). In a glovebox (argon) 4 (11.6 mg, 20.2
μmol, 1 equiv) and Ni(cod)2 (5.1 mg, 19 μmol, 1 equiv) were combined
in a 5 mL glass vial and dissolved in THF-d8 (0.5 mL). The solution was
stirred at room temperature for 5 min and then transferred to a Young
NMR tube. Outside the glovebox the NMR tube was cooled to ꢀ78 °C,
and in a stream of argon 2M3BN (10 μmol, excess) was added by syringe
to the cooled solution. 31P{1H} NMR (202.5 MHz, 231 K, THF-d8): δ
21.3 (d, J = 35.0 Hz), 22.1 (d, J = 35.0 Hz).
(4)Ni(η3-C4H7)CN (16). A flame-dried Schlenk tube was charged
with Ni(cod)2 (48 mg, 0.17 mmol, 1 equiv) and 4 (100 mg, 0.174 mmol,
1 equiv) as solids, which were dissolved in dry THF (15 mL). An orange-
red solution was instantly formed, which was stirred for 5 min before
2M3BN (35 μL, 0.35 mmol, 2 equiv) was added by Hamilton syringe.
After 3 h at room temperature the red solution was concentrated in vacuo
to 8 mL, and n-pentane (5 mL) was added. The solution was stored
at ꢀ40 °C, precipitating dark red crystals suitable for X-ray analysis. Mp:
129 °C (dec). 1H NMR (500.1 MHz, THF-d8, 213 K): δ ꢀ0.34 (m, 1H,
H-A), 0.00 (dd, 1H, J = 22.9 Hz, J = 10.9 Hz, H-A), 0.61 (bs, 2H, H11*),
1.11ꢀ1.31 (m, 4H, H11, H12), 1.35 (m, 1H, H-A*), 1.41 (bs, 3H,
H-D*), 1.58 (bs, 2H, H12*), 1.83b (m, 3H, H-D), 2.25 (pt, 1H, J = 7 Hz,
H-A*), 3.14 (bs, 1H, HꢀC*), 4.02 (bs, 1H, HꢀC), 4.29 (bs, 1H), 4.35
(s, 1H, H10*), 4.45 (s, 1H, H10), 4.74 (s, 1H, H9), 4.91 (bs, 1H), 5.49
(s, 1H, H9*), 6.56 (t, 1H, J = 7.6 Hz), 6.72 (t, 1H, J = 7.2 Hz), 6.83 (m,
3H), 7.12 (bs, 6H), 7.18ꢀ7.51 (m, 27H), 7.57 (bs, 2H), 7.67 (m, 2H),
7.74 (t, 2H, J = 7.7 Hz), 7.92 (bs, 2H), 8.09 (bs, 1H), 8.33 (pt, 1H, J = 8.6
Hz), 8.63 (bs, 1H), 8.79 (bs, 1H). 13C{1H} NMR (HMQC, 125.7 MHz,
(4)Ni(η3-C4H7)Cl (25). A flame-dried Schlenk was charged with 4
(200 mg, 348 μmol, 1 equiv) and Ni(cod)2 (96 mg, 0.35 mmol, 1 equiv).
The solids were dissolved in toluene (25 mL). Instantly an orange solution
was formed, which was stirred for 5 min at room temperature. Then
3-chloro-1-butene was added with a Hamilton syringe (100 μL). The
solution instantly turned dark violet and was stirred overnight at room
temperature. A dark red precipitate was formed, which was isolated
by filtration and washed with n-pentane (160 mg, 220 μmol, 63%).
1
Mp: 149 °C (dec). H NMR (600.1 MHz, toluene-d8): δ 0.96 (bs,
1H, H11 or H12), 1.16 (m, 1H, H-A), 1.18 (m, 1H, H11 or H12),
1.21 (bs, 3H, H-D), 1.92 (bs, 1H, H-A), 3.78 (bs, 1H, H-C), 3.98
2806
dx.doi.org/10.1021/om200164f |Organometallics 2011, 30, 2790–2809

Organometallics
ARTICLE
(s, 1H, H10), 5.07 (bs, 1H, H-B), 5.86 (s, 1H, H9), 6.81ꢀ7.14 (bs, 12H),
7.67 (bs, 4H), 7.79 (bs, 2H), 7.92 (bs, 2H). 13C{1H} NMR (HMQC, 150.9
MHz, toluene-d8): δ 19.1 (C6-D, 22.7 (C11 or C12), 26.6 (C12 or C11),
37.0 (m, C-A), 40.78 (t, J = 16.0 Hz, C9) 88.5 (t, J = 12.4 Hz, C-B), 102.8
(m, C-C), 127.6, 129.4, 130.4, 134.3, 134.9, 135.6, 136.3, 137.2, 144.7.
31P{1H} NMR (121.5 MHz, toluene-d8): δ 1.65 (d, J = 99.1 Hz), 13.50 (d,
J = 99.1 Hz). 31P{1H} NMR (101.3 MHz, THF-d8): δ 2.19 (bd, J = 78.0
Hz), 13.23 (bd, J = 78.0 Hz). MS (FABþ, NPOE): m/z 722.2 [M]þ, 687.2
[M ꢀ Cl]þ, 632.1 [M ꢀ Cl ꢀ C4H7]þ, 574.2 [4]þ. IR (KBr): ~ν/cmꢀ1
3053 (m), 2956 (m), 2876 (w), 1585 (w), 1493 (m), 1479 (s), 1434 (vs),
1422 (vs), 1304 (w), 1136 (m), 1092 (s), 784 (vs), 704 (vs). Anal. Calcd
was stirred for 10 min at ambient temperature. To the catalyst solution
was added 100 μL of diethyleneglycol diethyl ether via a Hamilton
syringe as a standard for GC analysis. From a separate vessel the required
amount of 2M3BN was transferred to the catalyst solution by means of a
Pasteur pipet using 3 ꢁ 1 mL of toluene to secure the quantitative
transfer of the substrate. The reaction mixture was then heated for 20 h,
and samples were taken after 1.5, 3.0, and 20 h for GC analysis. Every
reaction was performed twice in order to prove the reproducibility of the
data obtained. After 20 h of heating the reaction mixture was clear and
free from any precipitates in every case. In the case of isomerization
performed using very low catalyst concentrations the catalyst was added
from a prepared stock solution to ensure accurancy.
for C44H40ClNiP2 toluene: C, 75.07; H, 5.81; P, 7.59. Found: C, 74.96; H,
3
5.80; P, 7.52.
2-Trideuteromethyl-3-butenenitrile (2M3BN-d). To a
cooled solution of LDA (2.00 g, 18.7 mmol, 1 equiv) in THF
(38 mL) was added 3-butenenitrile (1.58 mL, 19.6 mmol, 1.05 equiv)
in a manner where the temperature of the solutions stayed lower than
ꢀ70 °C at all times. After 30 min at ꢀ80 °C CD3I (1.40 mL, 22.5 mmol,
1.2 equiv) was added by syringe, again without letting the temperature of
the solution rise above ꢀ70 °C. After 30 min at ꢀ80 °C the reaction was
quenched by the addition of NH4Cl (0.5 mL), and the solvent was
removed in vacuo at 0 °C. The residue was dissolved in CH2Cl2 (20 mL),
washed with H2O, and dried over MgSO4. Distillation at 114 °C yielded
260 mg (3.09 mmol, 17%) of 2M3BN-d along with 5% of Z-2M2BN-d.
Bp: 114 °C. 1H NMR (300.1 MHz, toluene-d8): δ 2.57 (m, 1H, H2),
’ ASSOCIATED CONTENT
S
Supporting Information. PDF file giving further detailed
b
information about the catalysis experiments performed, reproduc-
tions of relevant NMR spectra, as well as details for the DFT
calculations. CIF files giving X-ray crystallographic data for 1, 4, 5,
6c, 6d, 14, 16a, 24, and 25. This material is available free of charge
via the Internet at http://pubs.acs.org. CCDC 813674 (1), 813675
(4), 813676 (5), 813677 (6c), 813678 (6d), 813679 (14), 813680
(16a), 813681 (24), and 813682 (25) contain the supplementary
crystallographic data for this paper. These data can be obtained free
of charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
4.85 (ddd, 1H, J4,3 = 9.5 Hz, J4,2 = 2J = 1.2 Hz, H4), 5.12 (ddd, 1H, J4,3
=
17.0 Hz, J4,2 = 2J = 1.2 Hz, H4), 5.25 (ddd, 1H, J3,4 =17.1 Hz, J3,4 = 9.7
Hz, J3,2 = 4.5 Hz, H3). 2H{1H} NMR (46 MHz, toluene-d8): δ 0.56(s).
13C{1H} NMR (75.5 MHz, toluene-d8): δ 17.61 (m, 2-CD3), 28.67
(C2), 116.59 (C4), 120.21 (C1), 133.94 (C3). IR (liquid): ν~/cmꢀ1
3092 (w), 2989 (w), 2240 (s), 2216 (m), 2127 (w), 2084 (w), 1643(m),
1418 (m), 1055 (m), 989 (s), 931 (s). MS (HR-EIþ): m/z 84.0770
(calcd for C5H4D3N: 84.0764, err = þ3.6 ppm).
’ AUTHOR INFORMATION
Corresponding Author
*E-mail: ph@oci.uni-heidelberg.de.
rac-(Z)-1-Buten-3-ol-1D1.4. LiAlH4 (9.8 g, 0.26 mol, 1 equiv) was
suspended within a flame-dried three-necked flask (500 mL) equipped
with a magnetic stirrer and a reflux condenser in dry THF (250 mL).
But-3-yn-2-ol (15.8 mL, 0.200 mol, 1 equiv) was then added by syringe.
An exothermic reaction commenced, and the reaction mixture was
cooled with an ice bath. After the exothermic reaction stopped, the ice
bath was removed and the reaction mixture was stirred for another 18 h
at room temperature. The mixture was then cooled again with an ice bath
and quenched with D2O (25 mL). The slurry was filtered, the separated
solid was extracted with THF (3 ꢁ 50 mL), and the resulting solution
was dried over MgSO4. The title compound was isolated as a colorless
liquid from this solution by Vigreux colum distillation. Yield: 1.904 g,
95% purity, 40% overdeuteration for the terminal E-hydrogen
(H0.6D1.4CdCHꢀCH(OD)CH3). Bp: 98 °C. 1H NMR (250.1 MHz,
CDCl3): δ 1.30 (3H), 5.08 (0.6H, H4), 5.33 (1H, H3), 5.93 (1H, H2).
’ ACKNOWLEDGMENT
We thank the Deutsche Forschungsgemeinschaft (SFB 623)
and the Studienstiftung des Deutschen Volkes (fellowship to
S.M.B.) for financial support of this work. We thank BASF SE
for financial support of this work, for samples of pentenenitriles,
and for assistance in performing high-pressure DielsꢀAlder
reactions.
’ REFERENCES
(1) (a) Collman, J. P.; Hegedus, L. S.; Finke, R. G. Principles and
Applications of Organotransition Metal Chemistry; University Science
Books: Mill Valley, CA, 1987. (b) James, B. R. Comprehensive Organo-
metallic Chemistry; Wilkinson, G.; Stone, F. G. A.; Abel, E. W., Eds.;
Pergamon: Oxford, 1982. (c) Parshall, G. W. Homogeneous Catalysis;
Wiley: New York, 1980. (d) Parshall, G. W.; Ittel, S. D. Homogeneous
Catalysis; Wiley: New York, 1992. (e) Hartwig, J. F. Organotransition
Metal ChemistryꢀFrom Bonding to Catalysis; University Science Books:
Sausalito, CA, 2010. (f) Weissermel, K.; Arpe, H.-J. Industrial Organic
Chemistry; Wiley-VCH: Weinheim, 1997. (g) Elschenbroich, C. Orga-
nometallics; Wiley-VCH: Weinheim, 2006. (h) Behr, A. Angewandte
Homogene Katalyse; Wiley-VCH: Weinheim, 2008.
(2) Burridge, E. Eur. Chem. News 2005, 15 (April), 4–10.
(3) Drinkard, W. C.; Lindsey, R. V.(DuPont) US Pat., 2,496,217,
1970.
(4) Tolman, C. A.; Seidel, W. C.; Druliner, J. D.; Domaille, P. J.
Organometallics 1984, 3, 33.
rac-(Z)-3-Chloro-1-butene-1D1.4
.
rac-(Z)-1-Buten-3-ol-1D1.4
(1.8 g, 24 mmol, 1 equiv) was added to hexachloroacetone (8.8 mL,
58 mmol, 2.4 equiv) in a round flask equipped with a magnetic stir bar.
The mixture was cooled to 0 °C, and PPh3 (6.8 g, 26 mmol, 1 equiv) was
added slowly, with the mixture never exceeding 10 °C (20 min), while
the color changed to a very dark violet. The mixture was then warmed to
room temperature and stirred for another 20 min. rac-(Z)-3-Chloro-1-
butene-1D1.4 was isolated as a colorless liquid by isothermal distillation
(cooling trap, dry ice) at 100 mbar (1.13 g, 12.3 mmol, 51%; 40%
overdeuteration at C1). 1H NMR (250.1 MHz, CDCl3): δ 1.61 (d, 3H,
J = 6.6 Hz, H4), 4.54 (q, 1H, J = 6.8 Hz, H3), 5.09 (d, 0.6H, J = 10.7 Hz,
H1), 5.96 (m, 1H, H2). 2H{1H} NMR (92.2 MHz, toluene-d8): δ 4.90
(bs, 0.4D, D1), 5.05 (d, J = 1.4 Hz, 1D, D1).
(5) Tolman, C. A.; McKinney, R. J.; Seidel, W. C.; Druliner, J. D.;
Stevens, W. R. Adv. Catal. 1985, 33, 1.
Catalysis. Under an atmosphere of argon toluene (1 mL) was added
into a 20 mL flame-dried Schlenk tube containing Ni(cod)2 (2.8 mg)
and 1.05 equiv of the ligand as solids. A red solution formed instantly and
(6) Druliner, J. D. Organometallics 1984, 3, 205.
(7) (a) Tauchert, M. E.; Kaiser, T. R.; G€othlich, A. P. V.;
Rominger, F.; Warth, D. C. M.; Hofmann, P. ChemCatChem 2010, 2, 674.
2807
dx.doi.org/10.1021/om200164f |Organometallics 2011, 30, 2790–2809

Organometallics
ARTICLE
(b) G€othlich, A. P. V.; Tensfeldt, M.; Rothfuss, H.; Tauchert, M. E.; Haap,
D.; Rominger, F.; Hofmann, P. Organometallics 2008, 27, 2189. (c) van der
Vlugt, J. I.; Hewat, A. C.; Neto, S.; Sablong, R.; Mills, A. M.; Lutz, M.; Spek,
A. L.; Vogt, D. Adv. Synth. Catal. 2004, 346, 993.
(30) (a) Bartsch, M.; Baumann, R.; Kunsmann-Keitel, D. P.; Haderlein,
G.; Jungkamp, T.; Altmayer, M.; Siegel W. (BASF) DE 10136488 A1,
2003. (b) Bartsch, M.; Kunsmann-Keitel, D. P.; Baumann, R.; Haderlein,
G.; Siegel, W. (BASF) WO 0213964 A2 2002.
(8) (a) Bini, L.; M€uller, C.; Vogt, D. ChemCatChem 2010, 2, 590. (b)
Bini, L.; M€uller, C.; Vogt, D. Chem. Commun. 2010, 46, 8325.
(9) Chaumonnot, A.; Lamy, F.; Sabo-Etienne, S.; Donnadieu, B.;
Chaudret, B.; Barthelat, J.-C.; Galland, J.-C. Organometallics 2004,
23, 3363.
(31) The cod complex itself is stable at room temperature in solution
for several days in the absence of the substrate.
(32) Reaction performed in dioxane, 100 °C applying 0.2, 0.5, and
0.9 equiv of KCN with respect to the catalyst. Due to the poor solubility
of KCN in dioxane, 18-crown-6 was also present.
(10) (a) B€ackvall, J. E.; Andel, O. S. Organometallics 1986, 5, 2350.
(b) Tolman, C. A. J. Chem. Educ. 1986, 63, 199.
(33) This is in contrast to a previous report (ref 19) where bis-
(chelate) formation with ligand 1 was not observed.
(11) Wilting, J.; M€uller, C.; Hewat, A. D.; Tooke, D. L.; Vogt, D.
Organometallics 2005, 24, 13.
(12) Acosta-Ramírez, A.; Mu~noz-Hernꢀandez, M.; Jones, W. D.;
García, J. J. J. Organomet. Chem. 2006, 691, 3895.
(13) Acosta-Ramírez, A.; Flores-Gaspar, A.; Mu~noz-Hernꢀandez, M.;
Arꢀevalo, A.; Jones, W. D.; García, J. J. Organometallics 2007, 26, 1712.
(14) Acosta-Ramírez, A.; Mu~noz-Hernꢀandez, M.; Jones, W. D.;
García, J. J. Organometallics 2007, 26, 5766.
(15) Swartz, B. D.; Reinartz, N. M.; Brennessel, W. W.; García, J. J.;
Jones, W. D. J. Am. Chem. Soc. 2008, 130, 8548.
(16) Jones, W. D.; Li, T. Organometallics 2011, 30, 547.
(17) Baker, M. J.; Harrison, K. N.; Orpen, A. G.; Pringle, P. G.; Shaw,
G. J. Chem. Soc., Chem. Commun. 1991, 803.
(18) (a) Ahlers, W.; R€oper, M.; Hofmann, P.; Warth, D. C. M.;
Paciello, R. (BASF) WO 0158589, 2001. (b) Ahlers, W.; Paciello, R.;
Vogt, D.; Hofmann, P. (BASF) WO 02083695 A1, 2002. (c) Warth,
D. C. M. Diploma Thesis, University of Heidelberg, 1999. (d) Warth,
D. C. M. Ph.D. Dissertation, University of Heidelberg, 2004. (e)
Hofmann, P. Invited lecture, 14th International Symposium on Homo-
geneous Catalysis, Munich, 2004, abstract IL7. (f) Gelman and co-
workers, beginning in 2006, published a series of papers describing
interesting aspects of the transition metal chemistry and catalytic
applications of analogues of our Triptyphos ligand family; see for
example: Kaganovsky, L.; Rueck-Braun, K.; Gelman, D. J. Organomet.
Chem. 2010, 695, 260. (g) Grossmann, O.; Rueck-Braun, K.; Gelman, D.
Synthesis 2008, 4, 537, and references therein.
(19) Bini, L.; M€uller, C.; Wilting, J.; von Chrzanowski, L.; Spek,
A. L.; Vogt, D. J. Am. Chem. Soc. 2007, 129, 12622.
(20) de Greef, M.; Breit, B. Angew. Chem. 2009, 121, 559; Angew.
Chem., Int. Ed. 2009, 48, 551.
(21) Goertz, W.; Keim, W.; Vogt, D.; Englert, U.; Boele, M. D. K.;
van der Veen, L. A.; Kamer, P. C. J.; van Leeuwen, P. W. N. M. J. Chem.
Soc., Dalton Trans. 1998, 2981.
(22) Kranenburg, M.; van der Burght, Y. E. M.; Kamer, P. C. J.; van
Leeuwen, P. W. N. M. Organometallics 1995, 14, 3081.
(34) (a) All calculations were performed with the Turbomole
program package.34b Because of its robustness in different chemical
bonding situations, the DFT Becke-Perdew86 (BP86) level of theory34cꢀe
within the efficient RI-J approximation for the Coulomb two-electron
terms was used.34f,g For structure optimizations SV(P) basis sets were
employed.34h Single-point energies were calculated at the BP86 level of
theory with larger triple-zeta-valence plus polarization basis sets (BP86/
TZVP//BP86/SV(P)).34iꢀk Stationary points on the potential energy
surface were characterized as either minima or transition states by the
presence of (respectively) zero or one significantly imaginary frequency
in the BP86/SV(P) vibrational spectrum, obtained by second analytic
derivative calculations.34l,m All G values refer to 298.15 K and 0.1013 MPa
pressure. Also, the vibrational partition function contribution to G was
omitted, as the calculated vibrational spectra of the complexes were
found to exhibit extremely small restricted rotational frequencies. Effects
of anharmonicity and numerical accuracy, therefore, would lead to
artifacts in statistical thermodynamics.34n (b) Ahlrichs, R.; B€ar, M.;
H€aser, M.; K€olmel, C. Chem. Phys. Lett. 1989, 162, 165. (c) Vosko, S. H.;
Wilk, L.; Nusair, M. Can. J. Phys. 1980, 58, 1200. (d) Perdew, J. P. Phys.
Rev. B 1986, 33, 8822. (e) Becke, A. D. Phys. Rev. A 1988, 38, 3098. (f)
€
Eichkorn, K.; Treutler, O.; Ohm, H.; H€aser, M.; Ahlrichs, R. Chem. Phys.
Lett. 1995, 240, 283. (g) Eichkorn, K.; Weigend, F.; Treutler, O.;
Ahlrichs, R. Theor. Chim. Acta 1997, 97, 119. (h) Sch€afer, A.; Horn,
H.; Ahlrichs, R. J. Chem. Phys. 1992, 97, 2571. (i) Becke, A. D. J. Chem.
Phys. 1993, 98, 5648. (j) Lee, C.; Young, W.; Parr, R. G. Phys. Rev. B
1998, 73, 785. (k) Sch€afer, A.; Huber, C.; Ahlrichs, R. J. Chem. Phys.
1994, 100, 5829. (l) Deglmann, P.; Furche, F.; Ahlrichs, R. Chem. Phys.
Lett. 2002, 362, 511. (m) Deglmann, P.; May, K.; Furche, F.; Ahlrichs, R.
Chem. Phys. Lett. 2004, 384, 103. (n) Deglmann, P.; Ember, E.;
Hofmann, P.; Pitter, S.; Walter, O. Chem.—Eur. J. 2007, 13, 2864.
(35) Crystal data for complex 14: C40H32Cl2NiP2, Mr = 704.21,
monoclinic space group P21/c, T = 200(2) K, a = 16.2756(3) Å,
b = 10.8845(2) Å, c = 20.0927(4) Å, R = 90°, β = 102.290(1)°, γ =
90°, V = 3477.88(11) Å3, Z = 4, μ = 0.831 mmꢀ1, 34 007/7924 reflec-
tions collected/unique, R1 = 0.047, wR2 = 0.095, GOF = 1.05.
(36) This same distribution was monitored when the complex was
prepared from t3PN instead of 2M3BN.
(23) van Leeuwen, P. W. N. M.; Kamer, P. C. J.; Reek, J. N. H.;
Dierkes, P. Chem. Rev. 2000, 100, 2741.
(24) Landis, C. R.; Uddin, J. J. Chem. Soc., Dalton Trans. 2002, 729.
(25) Carbꢀo, J. J.; Maseras, F.; Bo, C.; van Leeuwen, P. W. N. M. J. Am.
Chem. Soc. 2001, 123, 7630.
(26) (a) Grossman, O.; Azerraf, C.; Gelman, D. Organometallics
2006, 25, 375. (b) Schnetz, T.; R€oder, M.; Rominger, F.; Hofmann, P.
Dalton Trans. 2008, 2238. (c) Schnetz, T.; Rominger, F.; Hofmann, P.
Acta Crystallogr. Sect. E 2010, E66, m453.(d) Rosendahl, T. Ph.D.
Dissertation, University of Heidelberg, 2007.
(27) For simplicity, in our lab these ligands are jargonized as follows:
Triptyphos (1), Octyltriptyphos (3), Rucaphos (4, from Ruperta Carola
= Ruprecht-Karls-University Heidelberg), Maophos (5, from Martin
August Otfried Volland, a former member of our group), and C5-phos
(6a).
(37) Isolated complex 16 contained 1.5 equiv of crystal THF.
(38) For atom labeling (C1, C2, C3, C4) of the methallyl ligand see
Scheme 7.
(39) 13C{1H} NMR of 2M3BN in CDCl3: δ 133.4 (C3), 120.6
(CN), 117.4 (C4), 28.9 (C2), 18.5 (2-Me). 13C{1H} NMR of t3PN in
CDCl3: δ 130.9 (C3), 118.6 (C4), 118.0 (CN), 20.4 (C2), 17.6 (C5)
(atom labeling according to IUPAC nomenclature).
(40) 18 reveals an extremely broad 31P NMR signal down to 183 K
and only one isomer reported at 263 K; 18 shows exchange with free
PPh3 with a decoalescence at 183 K (ref 9).
(41) 18 also displays very broad 31P NMR signals down to 183 K
without decoalesence (ref 9).
(42) Prepared in the same fashion as 16.
(28) Crystal data for ligand 1: C44H32P2, Mr = 622.64, orthorhombic
space group Pnma, T = 200(2) K, a = 14.3118(5) Å, b = 19.9092(7) Å, c
= 11.4573(4) Å, R = 90°, β = 90°, γ = 90°, V = 3264.6(2) Å3, Z = 4, μ =
0.165 mmꢀ1, 23 367/3044 reflections collected/unique, R1 = 0.045,
wR2 = 0.098, GOF = 1.01.
(43) Samples of these crystals were dissolved in THF and CH2Cl2,
revealing the same ratios of 16a/b as described above (31P NMR).
(44) Crystal data for complex 16a: C53H55NNiO2P2, Mr = 858.63,
monoclinic, space group C2/c, T = 200(2) K, a = 33.9156(5) Å,
b = 13.0276(2) Å, c = 25.7532(1) Å, R = 90°, β = 129.098(1)°, γ = 90°,
V = 8830.69(19) Å3, Z = 8, μ = 0.554 mmꢀ1, 33 402/7009 reflections
collected/unique, R1 = 0.067, wR2 = 0.176, GOF = 1.11.
(29) Structural data for Xantphos: Hillebrand, S.; Bruckmann, J.;
Kruger, C.; Haenel, M. W. Tetrahedron Lett. 1995, 36, 75.
2808
dx.doi.org/10.1021/om200164f |Organometallics 2011, 30, 2790–2809

Organometallics
ARTICLE
(45) Crystals suitable for X-ray analysis were obtained by slow
diffusion of a solution of ZnCl2 in Et2O into a saturated solution of
16 in toluene.
(46) Crystal data for complex 24: C52H47Cl2NNiP2Zn, Mr = 942.83,
monoclinic, space group C2/c, T = 200(2) K, a = 35.9521(1) Å, b =
13.3196(2) Å, c = 21.2920(1) Å, R = 90°, β = 116.503(1)°, γ = 90°, V =
9124.55(15) Å3, Z = 8, μ = 1.162 mmꢀ1, 45 408/10 450 reflections
collected/unique, R1 = 0.052, wR2 = 0.097, GOF = 1.04.
(47) Crystals suitable for X-ray analysis precipitated at ꢀ35 °C from
a saturated THF solution.
(48) Crystal data for complex 25: C51H47ClNiP2, Mr = 815.99,
triclinic, space group P1, T = 200(2) K, a = 10.0032(1) Å, b =
11.0104(4) Å, c = 19.3264(6) Å, R = 76.469(2)°, β = 87.728(2)°, γ =
89.092(2)°, V = 2067.84(10) Å3, Z = 2, μ = 0.646 mmꢀ1, 10 881/4754
reflections collected/unique, R1 = 0.060, wR2 = 0.097, GOF = 0.97.
(49) Prepared by deprotonation of 3-butenenitrile with LDA in
THF followed by addition of CD3I. 2M3BN readily isomerizes toward
the more stable Z-2M2BN in the presence of bases. A 5 mol % amount of
(Z)-2M2BN-d could not be avoided. The amount of (Z)-2M2BN did
not change during isomerization.
ꢀ
(50) Acosta-Ramírez, A.; Flores-Alamo, M.; Jones, W. D.; García,
J. J. Organometallics 2008, 27, 1834.
(51) Kindler, K.; Oelschl€ager, H.; Henrich, P. Chem. Ber. 1953,
86, 167.
(52) Casalnuovo, A. L.; RajanBabu, T. V.; Ayers, T. A.; Warren, T. H.
J. Am. Chem. Soc. 1996, 116, 9869.
(53) Stadler, R.; Stibal, E.; Gronski, W. Makromol. Chem., Rapid
Commun. 1986, 7, 443.
(54) Lithium powder was filtrated under argon, dried in vacuo, and
stored in a glovebox (argon).
(55) Gross, J. H.; Nieth, N.; Linden, H. B.; Blumbach, U.; Richter,
F. J.; Tauchert, M. E.; Tompers, R.; Hofmann, P. Anal. Bioanal. Chem
2006, 386, 52–58.
2809
dx.doi.org/10.1021/om200164f |Organometallics 2011, 30, 2790–2809