A novel access to phenylnickel-phosphinophenolate trimethylphosphine complexes as single component oligo- or polymerization catalysts

Article abstract of DOI:10.1002/zaac.200400435

A novel access to phenylnickel-2-phosphinophenolate trimethylphosphine complexes, which unlike the "P-ylide method" is not limited to P-arylphosphino derivatives, consists of the reaction of 2-phosphinophenols and diphenylnickel-bis(trimethylphosphine), easily prepared from phenyllithium and NiCl₂(PMe₃)₂. [PhNi(2-Ph₂PC ₆H₄O)(PMe₃)], [PhNi(2-cHex₂PC ₆H₄O)(PMe₃)] and [PhNi(2-Ph₂P-4- MeOC₆H₃O)(PMe₃)], the latter two with more basic dialkylphosphino or 4-methoxy substituents, were synthesized and tested in the oligomerization of ethylene. The molecular structure of [PhNi(2-cHex ₂PC₆H₄O)(PMe₃)] in the crystal is reported.

Full text of DOI:10.1002/zaac.200400435

A Novel Access to Phenylnickel-phosphinophenolate Trimethylphosphine
Complexes as Single Component Oligo- or Polymerization Catalysts
Joachim Heinicke*,a, Normen Peuleckea, Markus K. Kindermanna, and Peter G. Jonesb
a Greifswald, Institut für Chemie und Biochemie, Ernst-Moritz-Arndt-Universität
b Braunschweig, Institut für Anorganische und Analytische Chemie der Technischen Universität
Received October 11th, 2004.
Professor Manfred Meisel zum 65. Geburtstag gewidmet
Abstract. A novel access to phenylnickel-2-phosphinophenolate tri-
methylphosphine complexes, which unlike the “P-ylide method” is
not limited to P-arylphosphino derivatives, consists of the reaction
of 2-phosphinophenols and diphenylnickel-bis(trimethylphos-
phine), easily prepared from phenyllithium and NiCl2(PMe3)2.
[PhNi(2-Ph2PC6H4O)(PMe3)], [PhNi(2-cHex2PC6H4O)(PMe3)] and
[PhNi(2-Ph2P-4-MeOC6H3O)(PMe3)], the latter two with more
basic dialkylphosphino or 4-methoxy substituents, were synthe-
sized and tested in the oligomerization of ethylene. The molecular
structure of [PhNi(2-cHex2PC6H4O)(PMe3)] in the crystal is re-
ported.
Keywords: Nickel complex; O,P chelate; Crystal structure; Homo-
geneous catalysis; Oligomerization
Ein neuer Zugang zu Phenylnickel (2-phosphinophenolat)(trimethylphosphin)-
Komplexen als Einkomponentenkatalysatoren für Oligomerisierungs- oder
Polymerisierungs-Reaktionen
Inhaltsübersicht. Ein neuer Zugang zu Phenylnickel(2-phosphino-
phenolat)(trimethylphosphin)-Komplexen, im Gegensatz zur „P-
Ylid Methode” nicht auf P-Arylphosphinoderivative beschränkt,
besteht in der Reaktion von 2-Phosphinophenolen und Diphenyl-
nickel-bis(trimethylphosphin), das in einfacher Weise aus Phenyl-
lithium und NiCl2(PMe3)2 erhalten wird. Die Verbindungen
[PhNi(2-Ph2PC6H4O)(PMe3)]
sowie
[PhNi(2-cHex2PC6H4O)-
(PMe3)] und [PhNi(2-Ph2P-4-MeOC6H3O)(PMe3)] mit basischeren
Dialkylphosphino- bzw. 4-Methoxy-Substituenten wurden herge-
stellt und als (Pre)Katalysatoren der Oligomerisierung von Ethylen
getestet. Die Molekülstruktur von [PhNi(2-cHex2PC6H4O)(PMe3)]
im Kristall wird vorgestellt.
Introduction
been obtained from o-benzochinone-2-ylidene-triphenyl-
phosphorane, Ni(COD)2 and PR3 (Scheme 1) [12, 15]. The
ylide route was also applicable to the synthesis of phe-
nylnickel 2-alkylphenylphosphinoenolates from acylidene-
alkyldiphenylphosphoranes by migration of the phenyl
group [16], but 2-dialkylphosphino derivatives have not
been reported so far. Investigations on in situ generated or-
ganonickel 2-phosphinophenolate polymerization and olig-
omerization catalysts revealed that nickel 2-dialkylphosphi-
nophenolates give rise to products with higher molecular
weights than do related nickel diphenyphosphinophenolates
[10, 17, 18]; we have therefore studied an alternative access
to phenylnickel 2-phosphinophenolate complexes that does
not depend on the migration of a P-substituent. We report
here on a novel, convenient and generally applicable syn-
thesis of the single component catalysts, represented by
three derivatives with different basicity at phosphorus or
oxygen. Some screening tests in the oligo/polymerization of
ethylene allowing a comparison with the activity and selec-
tivity of related in situ generated catalysts and an estimation
of the influence of the basicity of the donor atoms are
also described.
Organonickel PʝOϪ monochelate complexes are highly ac-
tive oligo- or polymerization catalysts and have found large
scale industrial application in the Shell Higher Olefin Pro-
cess [1]. They have also been the subject of considerable
academic interest [2, 3], which was recently revitalized by
the discovery of novel nickel chelate polymerization cata-
lysts [4Ϫ7] and their tolerance of water and functional
groups [8Ϫ10]. The standard synthesis of single component
phenylnickel 2-diphenylphosphino-enolate phosphine cata-
lysts is the reaction of acylidenetriphenylphosphoranes with
one equivalent of nickelbis(1,5-cyclooctadiene) (Ni(COD)2)
and tertiary phosphines [11Ϫ15]. Phenylnickel 2-diphenyl-
phosphinophenolate phosphine complexes have similarly
* Prof. Dr. J. Heinicke
Inst. f. Chemie u. Biochemie der Universität
Soldmannstr. 16
D-17487 Greifswald
Tel.: ϩ49 (0)3834/86-4318
fax: ϩ49 (0)3834/86-4319
E-mail address: heinicke@uni-greifswald.de
Z. Anorg. Allg. Chem. 2005, 631, 67Ϫ73
DOI: 10.1002/zaac.200400435
© 2005 WILEY-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim
67

J. Heinicke, N. Peulecke, M. K. Kindermann, P. G. Jones
46-47 Hz for 2a,b, 38.7 Hz for 2c) and C1 (2JPC ϭ 26-28 Hz
for 2a,b, 22.2 Hz for 2c) and moderate for C6 (3JPC ϭ 8.0-
10.2 Hz) and C4 (3JPC ϭ 5.1-6.6 Hz), indicates the PʝOϪ-
2
3
chelate coordination mode of nickel(II). JPC-1 and JPC-6
are markedly lower for η1-P-coordination of phosphino-
phenols [20]. The phenyl ligand in 2a-c is indicated by its
downfield shifted C-i and upfield shifted C-p signals, the
trimethylphosphine ligand by its characteristic doublet with
1JPC ϭ 24.5-25.8 Hz. The trans-position of the two phos-
phorus atoms follows from the large P-P coupling con-
stants, at Ϫ60 °C 303 and 304.6 Hz for 2a and 2c and some-
what smaller (293.3 Hz) for 2b with the more basic PʝOϪ-
ligand. 2a and 2c display two broad singlets at 25 °C and
need low-temperature measurements to produce sharp 31P
doublets. Similarly the 13C doublets of C-1 of the PʝOϪ-
ligand and C-i of NiPh couple only at low temperature with
the phosphorus nucleus of PMe3. In contrast, 2b shows
sharp 31P doublets and coupling of carbon nuclei C-2, C-i
and C-a even at 25 °C. This indicates increased fluxionality
of the trimethylphosphine ligand in the diphenylphosphino-
as compared to dicyclohexylphosphino-phenolate phe-
nylnickel complexes and is surprising, as the opposite be-
havior was observed by us in related methylnickel trimeth-
ylphosphine complexes. Methylnickel trimethylphosphine
diisopropylphosphinophenolates (sec-alkyl groups at phos-
phorus as in cHex2PC6H4OϪ) are more fluxional than
methylnickel trimethylphosphine diphenylphosphinopheno-
lates and need lower temperature (Ϫ60 versus Ϫ30 °C) to
afford sharp doublets in the 31P NMR spectra (each in tolu-
ene-d8) [18]. The reason for this unusual behavior is not yet
clear; it may be associated with the different hybridization
of the carbon atom bonded to nickel and with cis-effects.
No signals from free PMe3 were observed. Signal averaging
is assumed, e.g. from 2, free PMe3 and a pentacoordinate
complex formed by trapping of PMe3 by another molecule
of 2. Thus, the pentacoordinate methylnickel-phosphino-
phenolate bis(trimethylphosphine) complex 3, characterized
by crystal structure analysis, was accessible from the re-
spective square-planar mono-PMe3 complex and PMe3 [18].
3 is highly fluxional and displays the 31PMe3 signal upfield
from the corresponding square-planar PMe3 complex (∆δ
ca. 20). The phosphinophenolate nickel chelate unit in 2
(and 3) is quite rigid and not fluxional. The hemilabile
character observed in complexes with phosphinopheno-
lether ligands [21] or in cationic phosphinophenol com-
plexes [22] is strongly reduced by the negative charge of the
ligand and is limited to slow processes such as the E/Z-
isomerization of moderately bulky P-substituted nickel
bis(2-phosphinophenolates) [20].
Scheme 1
Scheme 2
Results and Discussion
Synthesis
For the conversion of 2-phosphinophenols 1 to phe-
nylnickel 2-phosphinophenolate trimethylphosphine com-
plexes 2, the direct replacement of a phenyl group of di-
phenylnickel complexes by the acidic hydroxyl group of
phosphinophenols was found to be a superior method. Be-
cause of the general stabilization of organometallic com-
pounds of transition metals by trimethylphosphine [19],
Ph2Ni(PMe3)2 was chosen as reagent. This compound, to
the best of our knowledge not yet reported in the literature,
is available easily and in good yield from two equivalents of
phenyllithium and NiCl2(PMe3)2. Heating with equimolar
amounts of 2-phosphinophenols 1a-c in toluene / THF pro-
vides the complexes 2a-c in high yields and without by-
products (Scheme 2), showing the applicability of the
method for diarylphosphino- and dialkylphosphinophenols
as well as methoxy-substituted derivatives. Small amounts
of unconverted starting compounds were removed by
recrystallization or by extraction with hexane.
Structure elucidation
Structural evidence of the phenylnickel 2-phosphinophenol-
ate trimethylphosphine complexes 2a-c was provided by
multinuclear NMR studies, in particular by the character-
istic 13C and 31P NMR data. The 13C chemical shifts are in
accordance with the substitution pattern. The strong down-
field shifts of the 13C-1 nuclei as compared to those of the
corresponding phosphinophenols 1a-c (∆δ ϭ 17.2-19.9) are
attributable to deprotonation and coordination, and con-
firm the formation of chelates. Small but indicative down-
field coordination shifts (deshielding) are observed for the
C-6 nuclei (∆δ ϭ 2.9-4.3) while C-2 and C-4 are shifted
upfield (C-2: 2a,b ∆δ ഠ Ϫ2.7, 2c ∆δ ϭ Ϫ11.8; C-4: 2a,b
∆δ ഠ Ϫ6.6, 2c ∆δ ϭ Ϫ3.9). The stronger shielding of C-2
at the expense of C-4 in 2c demonstrates the ϩM-effect of
the 4-methoxy group and the increased basicity of the phe-
nolate oxygen atom in this complex. The magnitude of the
coupling constants with the 31P nuclei, large for C2 (1JPC ϭ
Closer structural information on complexes of type 2 is
provided by the crystal structure analysis of 2b (Figure 1,
Tables 1 and 2). To the best of our knowledge it is the first
structure report of a four-coordinate σ-organonickel phos-
phinophenolate complex. Selected bond lengths and angles
are compiled in Table 1, for comparison together with the
corresponding data of 3 and the vinylogous phenylnickel
phosphinoenolate phosphine complexes 4 [23] and 5 [11].
68
© 2005 WILEY-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim
zaac.wiley-vch.de
Z. Anorg. Allg. Chem. 2005, 631, 67Ϫ73

A Novel Access to Phenylnickel-phosphinophenolate Trimethylphosphine Complexes
˚
Table 1 Selected bond lengths/A and angles/° of 2b and comparison
with the corresponding data of 3-5.
2b
3 [18]
4 [23]
5 [11]
Ni-C(1)
Ni-P(2)
Ni-P(1)
Ni-O
P(1)-C(11)
P(1)-C(21)
P(1)-C(31)
C(11)-C(12)
O-C(12)
1.8929(4)
2.1917(4)
2.1775(4)
1.9108(10)
1.8003(14)
1.8393(14)
1.8495(14)
1.4182(18)
1.3241(17)
1.981(5)
2.224(2), 2.245(2)
2.191(2)
1.984(3)
1.797(5)
1.818(5)
1.831(5)
1.411(6)
1.332(5)
1.867(6)
2.184(2)
2.189(2)
1.914(4)
1.797(6)
1.822(6)
1.826(6)
1.392(7)
1.308(7)
1.893
2.230
2.168
1.914
1.769
1.813
1.829
1.385
1.313
C(1)-Ni-P(1)
C(1)-Ni-P(2)
O-Ni-P(2)
92.77(4)
88.56(4)
92.37(3)
87.12(3)
92.6(2)
93.0(2)
89.6(2)
93.6(1)
85.2(1)
101.5(2)
108.7(5)
119.9(4)
173.1(2)
167.1(1)
90.6
91.7
91.5
86.5
99.3
113.8
118.9
91.7(2), 88.9(2)
93.43(11), 87.24(11)
86.35(10)
99.2(2)
114.3(4)
118.9(3)
177.7(2)
114.02(6), 120.71(7)
O-Ni-P(1)
Ni-P(1)-C(11)
P-C(11)-C(12)
Ni-O-C(12)
C(1)-Ni-O
99.49(5)
111.58(10)
119.75(8)
176.32(5)
166.990(16)
P(1)-Ni-P(2)
products depend on the substituents of the catalysts and
increase in the order 2c < 2a < 2b (Table 3). Catalysis with
2a leads to roughly equal amounts of liquid and waxy olig-
omers along with a smaller fraction of volatiles (mainly 1-
butene, small amount of hexene). 2c, which differs from 2a
in the 4-methoxy group (ϩM-effect) and a more basic phe-
nolate oxygen (cf. ∆δC-1), gives rise to formation of signifi-
cantly lower oligomers, mainly butenes and lower liquid oli-
gomers. This is attributed to the increased basicity of the
chelate-oxygen atom. Increased basicity at phosphorus, in
2b accomplished by cyclohexyl instead of phenyl groups,
causes the opposite effect, markedly longer chains and al-
most exclusive formation of low molecular weight poly-
ethylene. As compared to the polyethylenes obtained with
related methallylnickel phosphinophenolates or allyl-type
catalysts formed in situ from 1a-c and Ni(COD)2 [10], how-
ever, the molecular weights of the products from exp. 1-3
are very low, because of the effect of trimethylphosphine.
Conversions of ethylene with catalysts of 2a or 2b and equi-
molar amounts of Ni(COD)2, which trap a major part of
the trimethylphosphine (see below), lead to significantly in-
creased molecular weights but the values of polyethylenes
formed with 1/Ni(COD)2 are not even approached. Only
the Mw value of the small high molecular weight fraction
of the bimodal polyethylene obtained in exp. 5 resembles
that of polyethylene formed in presence of 1/Ni(COD)2.
Possibly the transfer of PMe3 to the catalyst site is less easy
than that of ethylene after a critical chain length is reached.
Altogether, the oligomerization of ethylene by 2a-c can be
explained by a competition of ethylene and PMe3 for coor-
dination at nickel and insertion of coordinated ethylene into
the Ni-C bond (chain propagation) (Scheme 3).
Figure 1 Molecular structure of 2b (ellipsoids with 50 % probabi-
lity).
The Ni-phenyl ring is arranged perpendicular (interplanar
angle 85.2°) to the chelate ring. The Ni-C σ-bond in 2b is
significantly shorter than in 3 but nearly identically with
that in 4 and 5 and is typical for Ni-C bonds with an sp2-
hybridized carbon atom, whereas Ni-C(sp3) bonds are usu-
˚
ally found in the range 1.92-1.97 A [24]. The Ni-P(2) bond
is similar to that in 4 but shorter than in the triphenylphos-
phine complex 5. The bonds C(11)-C(12) are longer in 2b
and 3 as compared to 4 and 5 but the other bond lengths
and angles of the rigid planar five-membered chelate rings
are similar and give evidence for close structural relations
of phosphinophenolate and phosphinoenolate complexes.
Oligomerization of ethylene
The phenylnickel phosphinophenolate trimethylphosphine
complexes 2a-c, dissolved in toluene and heated with ethyl-
ene (110 °C) under pressure (batch procedure, pstart ca.
50 bar) catalyze the oligomerization of ethylene with high
selectivity for linear α-olefins (Vin/Σ(CϭC) > 90-98 %). The
content of branched products, based on the Me/olefin ratio,
is low (Table 3). The oligomerization of ethylene with 2a,
however under constant ethylene pressure, has already been
reported [15]. With respect to selectivity and molecular
weights the results of both methods, batch and flow pro-
cedure, are similar and show that the results can be com-
pared in this respect. The average molecular weights of the
For the chain termination, β-hydride eliminations are as-
sumed, which yield α-olefins and Ni-H species, starting new
chain growth cycles. This is suggested by the detection of
trace amounts of styrene in the above ethylene oligomeriz-
ation experiments and with [PhNi(Ph2PC6H4O)(PPhMe2)]
Z. Anorg. Allg. Chem. 2005, 631, 67Ϫ73
zaac.wiley-vch.de
© 2005 WILEY-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim
69

J. Heinicke, N. Peulecke, M. K. Kindermann, P. G. Jones
Figure 2 Pressure time plots for batch polymerization of ethylene
with a) 2a (129 µmol, exp.1; solid), b) 2b (90 µmol, exp.2; dash) or
c) 2c (100 µmol, exp.3; dot) in toluene (20 mL), d) with 1a/Ni-
(COD)2/PMe3 (each 100 µmol [17]; dash-dot) in toluene (20 mL)
at 100 °C.
Figure 3 Pressure time plots for batch polymerization of ethylene
with a) 2a and Ni(COD)2 (exp. 4; solid), b) 2b and Ni(COD)2 (exp.
5; dash), c) 1a/Ni(COD)2 (dot [10]) each in toluene (20 mL).
purchased. Ethylene (99.5 %, Air Liquide) was used without
further treatment. NMR spectra were recorded on a multinuclear
FT-NMR spectrometer ARX300 (Bruker) at 299.6/300.1 (1H),
75.3/75.5 (13C), and 121.3/121.5 (31P) MHz. Shift references are
tetramethylsilane for 1H and 13C and H3PO4 (85 %) for 31P. Assign-
ment numbers are given partly in Scheme 2; to denote proton or
carbon nuclei of P-phenyl or P-cyclohexyl groups i, o, m, p or α,
β, γ, δ, for Ni-Ph iЈ, oЈ, mЈ, pЈ are used, respectively. Pseudotriplets
(X of AAЈX; A,AЈ ϭ P) of 1H or 13C nuclei formed by coupling
with two 31P nuclei are referred to as τ. Coupling constants refer
to JHH in 1H and JPC in 13C NMR data unless stated otherwise.
Assignments are supported by DEPT and in part by CH-COSY
experiments. Melting points were determined with a Sanyo Gallen-
kamp melting point apparatus, elemental analysis with a CHNS-
932 analyzer from LECO using standard conditions. 1H NMR
spectra of polymers were measured on a multinuclear FT-NMR
spectrometer ARX300 (Bruker) (300.1 MHz) at 100 °C, acquisition
time 4.9-5.4 s, delay 1.0 s, using concentrated solutions of the poly-
ethylene samples prepared by swelling for 1 d at 120 °C under ar-
gon in C6D5Br (p-CH, δ ϭ 7.23 as reference). Molecular weight
determination by GPC was performed as reported in ref. [10]. GC
analyses were carried out using a gas chromatograph Hewlett Pack-
ard 5890, column HP-5(30m) (crosslinked 5 %PhMe silicone), 40-
150 °C, 10 min isotherm, 4 °C/min.
[15] and the generation of ethylene polymerization catalysts
from 1a, nickel bromide and sodium hydride [10], respec-
tively, but other possibilities should not be excluded.
The coordination of PMe3 in competition with ethylene
has a strong influence on the reaction rates, as illustrated
by the pressure-time plots (Figures 2 and 3). The conversion
of ethylene by 2a is slow, by 2b and 2c with more basic
phosphorus or oxygen even slower (Figure 2a-c, Table 3),
but the oligomerization by the catalyst formed in situ from
1a / Ni(COD)2 and PMe3 [17] is much slower (Figure 2d).
The latter might be caused by incomplete reaction of 1a
and Ni(COD)2 and thus excess PMe3. The replacement of
PMe3 by ethylene occurs only to a small extent or is slow
despite its flexibility, shown in the NMR-spectra of 2a and
2c at normal pressure. Thus, bubbling ethylene (10 min)
through a solution of 2b in D8-toluene at 60 °C, 20 °C
above the boiling point of PMe3 (38-40 °C), did not lead to
a detectable replacement of PMe3. Only a small amount of
uncoordinated ethylene (δ13C ϭ 123.1) was observed in the
subsequent NMR-measurement. After addition of
Ni(COD)2 (25 µmol) to the solution of 2b (66 µmol) and
subsequent reaction with ethylene in D8-toluene under simi-
lar conditions, both coordinated (δ13C ϭ 49.7 (br)) and un-
coordinated ethylene were observed. The trapping of the
major part of PMe3 by Ni(COD)2 leads not only to the
above-mentioned increase of the molecular weights but also
to a strong acceleration of the ethylene consumption rate
((Figure 2).
Nickel complexes
Diphenylnickel-bis(trimethylphosphine).
Phenyllithium (0.87 mL 2M in diethylether, 1.74 mmol) was added
at Ϫ60 °C to a stirred solution of [NiCl2(PMe3)2] (0.245 g,
0.87 mmol) in Et2O/THF (15 mL). The mixture was allowed to
warm to room temperature (color change to pale yellow), and the
solvent was removed. Repeated extraction by warm hexane af-
forded 191 mg (61 %) of spectroscopically pure pale yellow crystals,
mp. 138-139 °C. (The compound was used immediately after syn-
thesis; it decomposes slowly at room temperature with color change
to black.)
Experimental Section
General
All reactions were carried out under carefully dried, oxygen-free
argon, using Schlenk techniques and freshly distilled ketyl-dried
solvents. The phosphinophenols 1a [25], 1b and 1c [10] and
Ni(COD)2 [26] were synthesized as reported, other chemicals were
1H NMR (C6D6, 25 °C): δ ϭ 0.62 (s br, 18 H, PMe3), 7.02 (m, 2 H, H-p),
7.19 (m, 4 H, H-m), 7.60 (d, 3J ϭ 6.8 Hz, 4 H, H-o). 13C NMR (D8-THF,
70
© 2005 WILEY-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim
zaac.wiley-vch.de
Z. Anorg. Allg. Chem. 2005, 631, 67Ϫ73

A Novel Access to Phenylnickel-phosphinophenolate Trimethylphosphine Complexes
Scheme 3
Table 2 Crystal data and structure refinement of 2b.
C27H28OP2Ni (489.16); found C 66.18 (calc. 66.30); H, 6.01
(5.77) %.
Empirical formula
Formula weight
Temperature
Wavelength
Crystal system
Space group
C27H40NiOP2
501.24
2
1H NMR (D8-toluene, Ϫ60 °C): δ ϭ 0.74 (d, JPH ϭ 9.3 Hz, 9 H, PMe3),
6.53 (dd, 3J ϭ 7.6, JPH ϭ 3.9 Hz, 1 H, 3-H or 6-H), 6.79-7.47 (ArH). 1H
NMR (D8-THF, 25 °C): δ ϭ 1.10 (d, JPH ϭ 9.3 Hz, 9 H, PMe3), 6.28 („t“,
133(2) K
0.71073 A
2
˚
3J ϭ 7.2-7.3 Hz, 1 H, ArH), 6.50-6.64 (m, 4 H, ArH), 6.93-7.02 (m, 4 H,
ArH), 7.23-7.37 (m, 10 H, ArH). 13C NMR (D8-THF, 25 °C): δ ϭ 11.5 (d,
1J ϭ 25.7 Hz; PMe3), 114.3 (d, 3J ϭ 5.3 Hz; C-4), 118.3 (d, 1J ϭ 46.8 Hz;
C-2), 119.8 (d, 3J ϭ 8.3 Hz; C-6), 121.6 (C-pЈ), 126.2 (2 C-mЈ), 128.8 (d, 3J ϭ
9.8 Hz; 4 C-m), 130.4 (2 C-p), 132.2 (d, 1J ϭ 45.3 Hz; 2 C-i), 133.1, 133.2
(C-3, C-5), 133.8 (d, 2J ϭ 10.6 Hz; 4 C-o), 138.0 (d, 3J ϭ 3.0 Hz; 2 C-oЈ),
150.3 (d, 2J ϭ 29.4 Hz; C-iЈ), 179.0 (d, 2J ϭ 26.4 Hz; C-1). 31P{1H} NMR
(toluene-d8, Ϫ60 °C): δ ϭ 23.6 (d), Ϫ13.4 (d), JPP ϭ 303.0 Hz; at 25 °C:
δ ϭ 23.1 (br), Ϫ14.2 (br). 31P{1H} NMR (THF-d8, 25 °C): δ ϭ 22.7 (br),
Ϫ12.9 (br).
Monoclinic
P21/c
a ϭ 9.3475(6) A
˚
˚
˚
Unit cell dimensions
α ϭ 90°
β ϭ 97.065(4)°
γ ϭ 90°
b ϭ 9.8384(6) A
c ϭ 28.593(2) A
3
˚
Volume
Z
Density (calculated)
Absorption coefficient
F(000)
Crystal size
Theta range for data collection
Index ranges
2609.6(3) A
4
1.276 Mg/m3
0.883 mmϪ1
1072
0.4 x 0.3 x 0.1 mm3
1.44 to 30.51°
Ϫ13<ϭh<ϭ13,
Ϫ14<ϭk<ϭ14,
Ϫ40<ϭl<ϭ40
50734
Phenylnickel-2-(dicyclohexylphosphino)phenolate(O,P)
(trimethylphosphine) (2b).
Reflections collected
A solution of 1b (200 mg, 0.685 mmol) and Ph2Ni(PMe3)2 (250mg,
0.685 mmol) in THF/toluene (20mL), prepared as described for 2a,
was heated for 3 d to 60 °C (slight color change to yellow). The
solvent was removed in vacuum, and the residue was washed with
hexane affording 320 mg (92 %) of 2b as pale yellow powder, mp.
189-191 °C, contaminated by a trace of Ph2Ni(PMe3)2. Pure single
crystals were grown from n-hexane.
Independent reflections
Completeness to theta ϭ 30.00°
Absorption correction
Max. and min. transmission
Refinement method
7964 [R(int) ϭ 0.0391]
99.9 %
Semi-empirical from equivalents
0.9169 and 0.7500
Full-matrix least-squares on F2
7964 / 0 / 283
Data / restraints / parameters
Goodness-of-fit on F2
1.067
Final R indices [I>2sigma(I)]
R indices (all data)
Largest diff. peak and hole
R1 ϭ 0.0314, wR2 ϭ 0.0713
R1 ϭ 0.0463, wR2 ϭ 0.0778
0.486 and Ϫ0.242 e·A
C27H40OP2Ni (501.25), found: C 64.25 (calc. 64.70); H 7.90
(8.04) %.
Ϫ3
˚
2
1H NMR (C6D6, 25 °C): δ ϭ 0.83 (d, JPH ϭ 9.0 Hz, 9 H, PMe3), 0.91-2.29
(multiplets, 22 H, Cy), 6.59 („t“, 3J ϭ 7.0, 7.3 Hz, 1 H, ArH), 6.87 („t“, J ഠ
37.0 Hz, 1 H, ArH), 7.05 („t“, 3J ϭ 7.4-7.5 Hz, 2 H, ArH), 7.08-7.22 (m,
2 H, ArH), 7.26 („t“, 3J ϭ 7-8 Hz, 1 H, ArH), 7.56 (d, 3J ϭ 7.4 Hz, 2 H,
ArH). 13C NMR (C6D6, 25 °C): δ ϭ 11.9 (d, 1J ϭ 24.5 Hz; PMe3), 27.0 (C-
δ), 27.8 (d, 3J ϭ 10.4 Hz; C-β or γ), 28.14 (d, 3J ϭ 13.3 Hz; C-β or γ), 28.06
(C-γ or β), 28.5 (d, 2J ϭ 3.3 Hz; C-γ or β), 32.7 (d, 1J ϭ 24.2 Hz; C-α),
114.3 (d, 3J ϭ 5.1 Hz; C-4), 116.2 (d, 1J ϭ 38.7 Hz; C-2), 120.0 (d, 3J ϭ
8.0 Hz; C-6), 122.3 (C-pЈ), 127.1 (2 C-mЈ), 132.0, 133.6 (C-3, C-5), 138.6 (br,
2 C-oЈ), 151.7 („t“, 2J ഠ 32, 33.5 Hz; C-iЈ), 180.2 (dd, 2J ϭ 22.2, 3J ϭ 9.5 Hz;
C-1). 31P{1H} NMR (C6D6): 25 °C δ ϭ 29.8 (d), Ϫ15.6 (d), 2JPP ϭ 295.7 Hz;
Ϫ60 °C: δ ϭ 13.2 (τ, Η1Jϩ3JΗ ϭ 28.5 Hz, PMe3), 121.1 (C-p), 126.3 (C-m),
137.8 (C-o), 151.4 (τ, Η2JEϩ2JЈZΗ ϭ 63.4 Hz, C-i). 31P{1H} NMR (C6D6):
δ ϭ Ϫ10.5.
Phenylnickel-2-(diphenylphosphino)phenolate(O,P)
(trimethylphosphine) (2a).
2
Ϫ60 °C δ ϭ 29.9 (d), Ϫ14.9 (d), JPP ϭ 293.3 Hz.
A solution of 1a (450 mg, 1.62 mmol) in THF (5 mL) was added
to a stirred solution of Ph2Ni(PMe3)2 (800 mg, 2.19 mmol) in
15 mL of a THF/toluene mixture. The orange solution was heated
to 50 °C and stirred for 4h. After cooling to 20 °C, the solvent was
removed in vacuum, and the residue was washed with n-hexane.
The solution of the residue in a small amount of toluene was over-
layered with n-hexane to give yellow-ochre crystals of 2a (635 mg,
81 %), mp. 189-192 °C.
Phenylnickel-2-diphenylphosphino-4-
methoxyphenolate(O,P) (trimethylphosphine) (2c).
A solution of 1c (244 mg, 0.79 mmol) and Ph2Ni(PMe3)2 (288 mg,
0.789 mmol) in THF / toluene (20 mL), prepared as described for
2a, was heated for 4 h at 50 °C. Then the solvent was removed in
Z. Anorg. Allg. Chem. 2005, 631, 67Ϫ73
zaac.wiley-vch.de
© 2005 WILEY-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim
71

J. Heinicke, N. Peulecke, M. K. Kindermann, P. G. Jones
Table 3 Oligomerization of ethylene with catalysts formed from 2a-c.a)
b)
Exp.
Catalyst
(µmol)
Conversion
of C2H4
[g] (%)
TON
Products
C6, C8, C10, C12
Solid polymer data:
[mol·molϪ1
TOFmax
]
(% of liquid oligomers) b)
mp. [°C], D [g cmϪ3
]
c)
mol. weights [g·molϪ1
]
[mol·molϪ1·hϪ1
]
α-olefins (%), Me/Vin
1
2
3
4
5
2a (129)
2b (90)
9.2(72)
9.7 (78)
8.6 (66)
7.8(59)
10.3 (83)
2540
1000
1.6 g volatiles
4.0 g oligomers
3.6 g wax
42, 38, 16, 2
28, 32, 21, 4
69, 23, 3, <1
64, 24, 9, 2
0
83-85
MNMR 420
>98, 1.6
3840
800
0.7 g oligomers
9.0 g polyethylene
113-115, 0.934
MNMR 800
>98, 1.5
2c (100)
3070
600
3.1 g volatiles (C4)
5.1 g oligomers
0.4 g soft wax
Ϫ
2a (120), Ni(COD)2
(120)
2320
14000
0.5 g volatiles (C4)
1.3 g oligomers
6.0 g polyethylene
120-122, 0.943
MNMR 1650
94, 1.3
2b (110), Ni(COD)2
(100)
3340
13000
< 0.1 g oligomers
10.3 g polyethylene
128-130, 0.952
Mw 8.900, Mn 4.600 d)
MwЈ 90.000, MnЈ 82.000
>98, 1.3
a) Catalysis in toluene (20 mL), p(C2H4)start ca. 50 bar, bath temperature 100 °C. b) Dissolved butenes in the flash distillate (>90 % 1-butene) added to volatiles
(mainly C4); percentages refer to 1-olefines; deviations from the Schultz-Flory distribution may be due to loss of hexene (to volatile fraction) and the solubility
c)
of C10 and C12 in the polymer.
band separation.
M
based on 1H NMR integration. d) Bimodal molecular weight distribution determined by GPC, Mw and Mn after
NMR
vacuum, and the yellow-orange residue was washed with n-hexane
to yield 0.36 g (88 %) of 2c, mp. 157-160 °C, slightly contaminated
with Ph2Ni(PMe3)2.
C28H30O2P2Ni (519.18); found: C 64.40 (calc. 64.78); H 5.75
(calc. 5.82) %.
Ethylene oligomerization / polymerization
2a, 2b or 2c (quantities see Table 3) were dissolved in toluene. In
experiments 4 and 5, additionally a solution of Ni(COD)2 (quantit-
ies see Table 3) in toluene was added at 0 °C. In each experiment
the total volume of the solvent(s) was 20 mL. The precatalyst solu-
tion was transferred to an argon-filled stainless steel autoclave
(75 mL), equipped with Teflon-coated magnetic stirrer, gas and
sample inlet valves, a safety diaphragm and an HEJU pressure
sensor (1-100 bar, Juchheim) for automatic pressure registration.
The autoclave was pressurized (ethylene ca. 50 bar), placed into a
preheated bath (100 °C) and heated overnight (ca. 15 h). After
cooling to 0-5 °C, unconverted ethylene was allowed to escape and
butenes were condensed in a cooling trap (Ϫ78 °C). The remaining
product was transferred to a flask, and volatiles were flash distilled
(80-100 °C / 1 Torr) into a cooling trap (Ϫ196 °C). Residual waxes
or polymers were stirred for 1 d with methanol / hydrochloric acid
(1:1), washed with methanol and dried. Conversion, oligomer com-
position and characteristic data of polymers are given in Table 3.
The maximum turnover frequency (TOFmax) was determined from
the slope of the range with the steepest roughly linear pressure de-
crease.
2
1H NMR (D8-THF, Ϫ60 °C): δ ϭ 1.05 (d, JPH ϭ 9.6 Hz, 9 H, PMe3), 3.43
3
(s, 3 H, OMe), 6.40 (d, 3J ϭ 9.0 Hz, 1 H, H-6), 6.42 (dd, JPH ϭ 8.8, 4J ഠ
3 Hz, 1 H, H-3), 6.48-6.60 (m, 4 H, ArH), 6.64 (dd, 3J ϭ 9.0, 4J ഠ 3 Hz,
1 H, H-5), 6.83 (t, 3J ϭ 7.3 Hz, 1 H, PhH), 6.96 (d, 3J ϭ 7.1 Hz, 2 H, PhH),
7.25-7.40 (m, 8 H, PhH). 13C NMR (D8-THF, Ϫ60 °C): δ ϭ 11.1 (d, 1J ϭ
25.8 Hz; PMe3), 55.1 (OMe), 113.8 (C-5 or C-3), 116.5 (d, 1J ϭ 45.7 Hz; C-
2), 119.8 (d, 3J ϭ 10.2 Hz; C-6), 121.35 (C-3 or C-5), 121.44 (C-pЈ), 126.1 (2
C-mЈ), 129.0 (d, 3J ϭ 9.8 Hz; 4 C-m), 130.7 (2 C-p), 131.9 (d, 1J ϭ 45.9 Hz;
2 C-i), 133.9 (d, 2J ϭ 10.9 Hz; 4 C-o), 138.1 (2 C-oЈ), 149.6 (d, 2J ϭ 6.6 Hz;
C-4), 151.4 (d, 2J ϭ 34.4 Hz; C-iЈ), 172.7 (d, 2J ϭ 28.1 Hz; C-1). 31P{1H}
NMR (D8-THF): 25 °C: δ ϭ 23.5 (br), Ϫ12.9 (br); Ϫ60 °C δ ϭ 24.2 (d),
Ϫ12.0 (d), JPP ϭ 304.6 Hz.
Crystal Structure Analysis.
Data collection: A crystal of 2b was mounted on a glass fiber in
inert oil and transferred to the cold gas stream of the diffractometer
(Bruker SMART 1000 CCD). Data were collected with monochro-
mated Mo-Ka radiation and corrected for absorption (program SA-
DABS).
Acknowledgments. We thank Prof. Dr. Dr. h.c. W. Keim (RWTH
Aachen) for the encouragement of the catalytic studies, Dr. Peter
Montag (Polymer Standard Service, Mainz) for the determination
of the molar mass distribution and B. Witt and S. Siegert for nu-
merous NMR measurements. The Fonds der Chemischen Industrie
is acknowledged for gift of chemicals.
Structure refinement: The structure was refined anisotropically on
F2 using the program SHELXL-97 [27]. Hydrogen atoms were re-
fined using a riding model or rigid methyl groups.
For selected bond lengths and angles of 2b see Table 1. The crystal-
lographic data are listed in Table 2. Complete crystallographic data
for 2b have been deposited with the Cambridge Crystallographic
Data Centre as supplementary publication no. CCDC-252210.
Copies of the data can be obtained free of charge on application
to CCDC, 12 Union Road, Cambridge CB2 1EZ, Uk [Fax: (int.
code) ϩ44-(1223)/336-033, e-mail: deposit@ccdc.cam.ac.uk].
References
[1] W. Keim, in Industrial Applications of Homogenous Catalysis
(Eds.: A. Mortreux, F. Petit), Dordrecht, 1988, p. 335Ϫ347.
[2] W. Keim, Angew. Chem. 1990, 102, 251Ϫ260; Angew. Chem.
Int. Ed. Engl. 1990, 29, 235Ϫ244; W. Keim, New J. Chem.
1994, 18, 93Ϫ96.
72
© 2005 WILEY-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim
zaac.wiley-vch.de
Z. Anorg. Allg. Chem. 2005, 631, 67Ϫ73

A Novel Access to Phenylnickel-phosphinophenolate Trimethylphosphine Complexes
[3] K. A. Ostoja-Starzewski, J. Witte, K. H. Reichert, G. Vasiliou,
in Transition Metals and Organometallics as Catalysts for Ole-
fin Polymerization [Proceedings of the International Sym-
posium], (Eds.; W. Kaminsky, H. Sinn), Springer, Berlin, 1988,
349Ϫ360; K. A. Ostoja-Starzewski, J. Witte, in Transition Me-
tal Catalyzed Polymerizations [Proc. Int. Symp.], 2nd ed., (Ed.:
R. P. Quirk), Cambridge University Press, Cambridge (UK),
1988, 472Ϫ496.
[4] G. J. P. Britovsek, V. C. Gibson, D. F. Wass, Angew. Chem.
1999, 111, 448Ϫ468; Angew. Chem. Int. Ed. 1999, 38,
428Ϫ447.
[5] S. D. Ittel, L. K. Johnson, M. Brookhardt, Chem. Rev. 2000,
100, 1169Ϫ1204.
[6] T. R. Younkin, E. F. Connor, J. I. Henderson, S. K. Friedrich,
R. H. Grubbs, D. A. Bansleben, Science 2000, 287, 460Ϫ462.
[7] Late Transition Metal Polymerization Catalysis, (Eds.: B. Rie-
ger, L. Saunders, S. Kacker, S. Striegler), Wiley-VCH,
Weinheim, 2003.
[8] S. Mecking, Angew. Chem. 2001, 113, 550Ϫ557; Angew. Chem.
Int. Ed. 2001, 40, 534Ϫ540; A. Held, F. M. Bauers, S. Meck-
ing, Chem. Commun. 2000, 301Ϫ302; M. O. Kristen, L. Man-
ders, S. Mecking, F. M. Bauers, (BASF A.G.), DE 199 61 340
(17. 12. 1999).
[9] R. Soula, C. Novat, A. Tomov, R. Spitz, J. Claverie, X. Dru-
jon, J. Malinge, T. Saudemont, Macromolecules 2001, 34,
2022Ϫ2026; A. Tomov, R. Spitz, T. Saudemont, X. Drujon,
(Elf Atochem S.A.) FR 98 12 476 (06. 10. 1998).
[10] J. Heinicke, M. Köhler, N. Peulecke, M. He, M. K. Kinder-
mann, W. Keim, G. Fink, Chem. Eur. J. 2003, 9, 6093Ϫ6107;
M. O. Kristen, J. Heinicke, W. Keim, M. Köhler, M. He,
(BASF-A.G.), DE 199 47519 (01. 10. 1999).
[13] U. Klabunde, R. Mühlhaupt, T. Herskovitz, A. H. Janowicz,
J. Calabrese, S. D. Ittel, J. Polym. Sci. Part A: Polym. Chem.
1987, 25, 1989Ϫ2003.
[14] P. Braunstein, Y. Chauvin, S. Mercier, L. Saussine, A. De
Cian, J. Fischer, Chem. Commun. 1994, 2203Ϫ2204.
[15] J. Pietsch, P. Braunstein, Y. Chauvin, New J. Chem. 1998,
467Ϫ472.
[16] K. Hirose, W. Keim, J. Mol. Catal. 1992, 73, 271Ϫ276.
[17] J. Heinicke, M. Köhler, N. Peulecke, W. Keim, J. Catal. 2004,
225, 16Ϫ23.
[18] J. Heinicke, M. He, A. Dal, H.-F. Klein, O. Hetche, W. Keim,
U. Flörke, H.-J. Haupt, Eur. J. Inorg. Chem. 2000, 431Ϫ440.
[19] H.-F. Klein, Angew. Chem.. 1980, 92, 362Ϫ375; Angew. Chem.
Int. Ed. Engl. 1980, 19, 362Ϫ375; H.-F. Klein, H. H. Karsch,
Chem. Ber. 1972, 105, 2628Ϫ2636.
[20] J. Heinicke, A. Dal, H.-F. Klein, O. Hetche, U. Flörke, H.-J.
Haupt, Z. Naturforsch. 1999, 54b, 1235Ϫ1243; J. Heinicke, M.
Köhler, N. Peulecke, W. Keim, P. G. Jones, Z. Anorg. Allg.
Chem. 2004, 630, 1181Ϫ1190.
[21] A. Bader, E. Lindner, Coord. Chem. Rev. 1991, 108, 27Ϫ110.
[22] J. Heinicke, M. Köhler, N. Peulecke, M. K. Kindermann, W.
Keim, M. Köckerling, Organometallics, submitted.
[23] U. Klabunde, T. H. Tulip, D. C. Roe, S. D. Ittel, J. Organomet.
Chem. 1987, 334, 141Ϫ156.
[24] Nickel Hydride, Alkyl and Aryl Complexes, P. W. Jolly, in Com-
prehensive Organometallic Chemistry, Vol. 6 (Eds.: G. Wilkin-
son, F. G. A. Stone, E. W. Abel), Pergamon Press, New York,
1982, pp. 37Ϫ100.
[25] T. B. Rauchfuss, Inorg. Chem. 1977, 16, 2966Ϫ2967.
[26] B. Bogdanovic, M. Kröger, G. Wilke, Liebigs Ann. Chem.
1966, 699, 1Ϫ23; W. A. Herrmann, C. Zybill, in Synthetic
Methods of Organometallic and Inorganic Chemistry, Vol. 1
(Eds.: W. A. Herrmann, A. Salzer), Thieme, Stuttgart, 1996,
157.
[11] W. Keim, F. H. Kowaldt, R. Goddard, C. Krüger, Angew.
Chem. 1978, 90, 493; Angew. Chem. Int. Ed. Engl. 1978, 17,
466Ϫ467.
[12] W. Keim, A. Behr, B. Gruber, B. Hoffmann, F. H. Kowaldt,
U. Kürschner, B. Limbäcker, F. P. Sistig, Organometallics 1986,
5, 2356Ϫ2359.
[27] SHELXL-97, a program for refining crystal structures. G. M.
Sheldrick, University of Göttingen, 1997.
Z. Anorg. Allg. Chem. 2005, 631, 67Ϫ73
zaac.wiley-vch.de
© 2005 WILEY-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim
73