O-Acylated 2-Phosphanylphenol Derivatives - Useful ligands in the Nickel- Catalyzed polymerization of ethylene

Article abstract of DOI:10.1002/ejic.200801121

The title ligands were prepared by O-acylation of 2-di- phenylphosphanyl-4-methylphenol (1) or directly by double lithiation of 2-bromo-4-methylphenol and stepwise coupling with ClPPh₂ and ClP(O)Ph₂ or RC(O))C1 (R = Me, tBu, Ph, 4-MeOC₆H ₄) to afford diphenylphosphinate 2 and carboxylic esters 3a-d. X-ray crystal structure analyses of 3b-d show conformations in which the P-phenyl substituents are rotated away from the ester group and the C(O)O π planes are nearly perpendicular to the phenol ring π plane. O-Acylated phosphanylphenols 2 and 3a-d form highly active catalysts with Ni(l,5-cod) ₂ (as does 1) for polymerization of ethylene, whereas phosphanylphenyl ethers do not give catalysts under the same conditions. The reason is the cleavage of the O-acyl bond upon heating with nickel(O) precursor compounds in the presence of ethylene. The precursors are P- coordinated Ni ⁰ complexes, which are formed at room temperature, such as 4d obtained from 3d and Ni(cod)₂ in a 2:1(molar ratio), and characterized by multinuclear NMR spectroscopy. Upon heating in the presence of ethylene, the pre-catalysts are activated. Catalysts2_Ni and 3a-d _Ni convert ethylene nearly quantitatively, 2_Ni slowly, and 3a-d_Ni rapidly, into linear polyethylene with vinyl and methyl end groups, and in the latter case, C(O)R end groups are also detectable. This proves insertion of Ni⁰into the O-C(O)R bond of 3a-d ligands for formation of the primary catalyst. Termination of the first chain growing cycle by β-hydride elimination changes the mechanism to the phosphanylphenolate- NiH initiated polymerization providing the main body of the polymer. A small retardation in the ethylene consumption rate with 3a-d_Ni, catalysts relative to that observed for l_Ni and stabilization of the catalyst, which gives rise to reproducibly high ethylene conversion, is observed.

Full text of DOI:10.1002/ejic.200801121

FULL PAPER
DOI: 10.1002/ejic.200801121
O-Acylated 2-Phosphanylphenol Derivatives – Useful Ligands in the Nickel-
Catalyzed Polymerization of Ethylene
Dmitry G. Yakhvarov,*[a,b] Kaleswara R. Basvani,[a] Markus K. Kindermann,[a]
Alexey B. Dobrynin,[b] Igor A. Litvinov,[b] Oleg G. Sinyashin,[b] Peter G. Jones,[c] and
Joachim Heinicke*[a]
Keywords: Phosphanes / Nickel / Polymerization / P ligands
The title ligands were prepared by O-acylation of 2-di-
phenylphosphanyl-4-methylphenol (1) or directly by double
lithiation of 2-bromo-4-methylphenol and stepwise coupling
with ClPPh2 and ClP(O)Ph2 or RC(O)Cl (R = Me, tBu, Ph, 4-
MeOC6H4) to afford diphenylphosphinate 2 and carboxylic
esters 3a–d. X-ray crystal structure analyses of 3b–d show
conformations in which the P-phenyl substituents are rotated
away from the ester group and the C(O)O π planes are nearly
perpendicular to the phenol ring π plane. O-Acylated phos-
phanylphenols 2 and 3a–d form highly active catalysts with
Ni(1,5-cod)2 (as does 1) for polymerization of ethylene,
whereas phosphanylphenyl ethers do not give catalysts un-
der the same conditions. The reason is the cleavage of the
O-acyl bond upon heating with nickel(0) precursor com-
pounds in the presence of ethylene. The precursors are P-
coordinated Ni0 complexes, which are formed at room tem-
perature, such as 4d obtained from 3d and Ni(cod)2 (in a 2:1
molar ratio), and characterized by multinuclear NMR spec-
troscopy. Upon heating in the presence of ethylene, the pre-
catalysts are activated. Catalysts 2Ni and 3a–dNi convert eth-
ylene nearly quantitatively, 2Ni slowly, and 3a–dNi rapidly,
into linear polyethylene with vinyl and methyl end groups,
and in the latter case, C(O)R end groups are also detectable.
This proves insertion of Ni0 into the O–C(O)R bond of 3a–d
ligands for formation of the primary catalyst. Termination of
the first chain growing cycle by β-hydride elimination
changes the mechanism to the phosphanylphenolate–NiH in-
itiated polymerization providing the main body of the poly-
mer. A small retardation in the ethylene consumption rate
with 3a–dNi catalysts relative to that observed for 1Ni and sta-
bilization of the catalyst, which gives rise to reproducibly
high ethylene conversion, is observed.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2009)
phosphane additives.[4b,6c] P-Basic dicyclohexylphosphanyl-
phenolate catalysts are somewhat less selective and allow
limited incorporation of α-olefins yielding copolymers with
various substituents at the main chain.[7] Cationic methal-
lylnickel 2-phosphanylphenol complexes are more active
but unselective and give rise to isomer mixtures, mainly of
butenes and hexenes.[8] Neutral tertiary and secondary 2-
phosphanylphenyl ethers were also screened in situ with
Ni(cod)2 (cod = 1,5-cyclooctadiene) but neither of these
compounds formed catalysts.[4a,6b] This suggests that the
selectivity mediated by neutral 2-phosphanylphenolate
nickel catalysts requires a leaving group at oxygen. For use
of phosphanylphenols a mechanism involving formation of
NiH catalysts was postulated and experimentally supported
Introduction
Organometallic nickel complexes bearing chelating phos-
phanyl–enolate ligands are of great importance in the large-
scale production of linear α-olefins in the Shell Higher Ole-
fin Process (SHOP),[1,2] but also allow the preparation of
polyethylene and ethylene–olefin copolymers.[2,3] Organo-
nickel 2-phosphanylphenolate catalysts[4,5] or related in situ
prepared catalyst systems[5,6] likewise accomplish selective
oligo- or polymerization of ethylene to linear α-olefin
chains and allow tuning of the molecular weights over a
wide range for polymers by variation of the substituents at
phosphorus[4a] and for low-molecular-weight oligomers by
[a] Institut für Biochemie, Ernst-Moritz-Arndt-Universität Greifs- by NMR spectroscopic data of precatalyst solutions gener-
wald,
ated at room temperature from 2-diphenylphosphanyl-
Felix-Hausdorff-Str. 4, 17487 Greifswald, Germany
phenol and Ni(cod)2.[4a] Signals similar to those of [Ni0(2-
Fax: +49-3834-864377
phosphanylphenol)(PMe3)3][9] and methallylnickel(II) 2-
E-mail: heinicke@uni-greifswald.de
[b] A.E. Arbuzov Institute of Organic and Physical Chemistry,
phosphanylphenolate complexes were detected.[4a] Ad-
Arbuzov str. 8, 420088 Kazan Russia
ditional support should be possible with 2-phosphanyl-
phenol derivatives, which are less inert than 2-phosphanyl-
phenyl ethers and, similarly to the O–H compounds, allow
insertion of nickel(0) into an O–E bond (E = electrophile).
Fax: +7-843-2732253
E-mail: yakhvar@iopc.knc.ru
[c] Institut für Anorganische und Analytische Chemie, Technische
Universität Braunschweig,
Postfach 3329, 38023 Braunschweig, Germany
1234
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2009, 1234–1242

Useful Ligands in the Nickel-Catalyzed Polymerization of Ethylene
In this case, the O-substituent should be detectable in the are almost coplanar and form a delocalized π system,
oligomer or polymer fraction. Furthermore, the nature of whereas the π planes of COO and the phenol ring are
the O-substituents might influence the catalyst and conse- strongly rotated to positions slightly more than perpendicu-
quently the product properties. We report here on our ende- lar with the double-bonded oxygen atom towards C6 for 3b
avors to address these aspects by synthesis of 2-phosphanyl- (ca. 100°) and even further for 3c and 3d (C2–C1–O–C ca.
phenyl esters, screening in the nickel-catalyzed ethylene po- 121 and 129°, respectively) to minimize electron–electron
lymerization and structural characterization of the ligands repulsion to the P lone electron pair.
and oligomers.
Results and Discussion
Synthesis and Structure of the Ligands
First hints of the applicability of O-substituted 2-phos-
phanylphenol derivatives in the nickel-catalyzed ethylene
polymerization were found in initial screening tests of 2-
(isopropylphenylphosphanyl)phenyl silyl ether, which gave
polymers similar to those afforded by the corresponding
phosphanylphenol. Steric hindrance, for example, in 2-di-
phenylphosphanyl-4,6-di-tert-butylphenyl
trimethylsilyl
ether, prevented activation. Therefore, in this study we in-
vestigated derivatives without substituents at the 6-position.
We focused on easily accessible 2-diphenylphosphanyl-4-
methylphenol derivatives, synthesized by bromination of p-
cresol, dilithiation, coupling with chlorodiphenylphos-
phane, and final O-substitution with chlorotrimethylsilane,
ClP(O)Ph2, or carboxylic acid chlorides RC(O)Cl (R = Me,
tBu, Ph, 4-MeOC6H4) (Scheme 1). The trimethylsilyl ether
was cleaved by methanol to provide 2-diphenylphosphanyl-
4-methylphenol (1) as the reference ligand. Diphenylphos-
phinate 2 and various carboxylic esters 3a–d were selected
to determine the influence of O-substituents on the forma-
tion of catalysts and their activity and to detect, by charac-
Figure 1. Molecular structure of 3b in the crystal (T = 293 K, ellip-
soids with 50% probability).
1
teristic 31P, H, or 13C NMR signals, the transfer of the O-
substituent to lower oligomer byproducts or to the polymer.
Figure 2. Molecular structure of 3c in the crystal (T = 100 K, ellip-
soids with 50% probability).
Scheme 1. Synthesis of 2-phosphanylphenol 1 and esters 2 and 3.
New esters 2 and 3a–d were structurally characterized by
1
31P, H, and 13C NMR spectroscopy and in some cases by
single-crystal X-ray diffraction (Figures 1, 2, and 3; Tables 1
and 3). The crystal structure analyses of 3b–d show that in
all three compounds the same conformation is preferred.
The P-phenyl groups are turned away from the ester group
to form a clockwise-turning propeller with one P–C axis
nearly in the phenol ring plane (torsion angles ca. –165 to
–178°) and the other nearly perpendicular (ca. 89–75°). The
Figure 3. Molecular structure of 3d in the crystal (T = 100 K, ellip-
aryl and COO π planes of the benzoyl group of 3c and 3d soids with 50% probability).
Eur. J. Inorg. Chem. 2009, 1234–1242
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1235

D. G. Yakhvarov, J. Heinicke et al.
FULL PAPER
Table 1. Selected bond lengths [Å], angles [°], and torsion angles [°] of 3b–d.
Assignment
3b
3c
3d
P–C2
P–CPh
C1–O
1.829(4)
1.833(4), 1.819(4)
1.412(4)
1.8406(12)
1.8303(12), 1.8329(13)
1.4065(15)
1.8412(11)
1.8311(11), 1.8316(11)
1.4014(13)
(Me)O–C
O–C
C=O
–
–
1.3654(14)
1.3668(14)
1.1979(15)
1.369(4)
1.190(5)
1.3641(16)
1.1953(17)
C–P–C
101.26(17)–103.92(17)
117.8(3)
125.5(3)
118.0(3)
119.5(3)
–88.4(3), 165.7(3)
–71.9(3), 119.5(3)
–25.9(4), 154.9(3)
–100.6(4)
82.5(4)
–4.0(5)
–
100.99(5)–103.44(5)
117.34(9)
125.44(9)
116.24(11)
121.06(11)
76.36(10), –177.18(9)
86.25(10), –98.06(11)
26.04(12), –151.66(10)
120.77(13)
–63.08(16)
5.1(2)
100.80(5)–103.84(5)
117.10(8)
125.47(8)
115.48(10)
122.06(10)
75.26(9), –177.91(8)
85.59(9), –99.71(9)
23.80(10), –152.07(8)
128.89(11)
–56.02(15)
5.14(19)
P–C2–C1
P–C2–C3
O–C1–C2
O–C1–C6
C1–C2–P–C
C2–P2–C12–C13/17
C2–P2–C18–C23/19
C2–C1–O–C
C6–C1–O–C
C1–O–C=O
O=C–C–C
13.0(2), –166.38(16)
4.2(2), –176.09(14)
Ethylene Polymerization
ing and heating with ethylene. Whereas the yellow color
Screening of 2-phosphanylphenyl esters 2 and 3a–d as turned rapidly to brown for the precatalyst solution of 1Ni,
ligands in the nickel-catalyzed polymerization of ethylene precatalyst solutions of 2Ni or 3a–dNi became only slightly
was performed as a batch procedure with an ethylene start- deeper yellow. This hints at differences in the formation of
ing pressure of 40–50 bar, for studying the fate of the acyl the precatalysts. For Ni(cod)2 and 2-diphenylphosphanyl-
groups at 20 bar, both at a bath temperature of 100 °C phenol, closely related to 1, at room temperature, the for-
(Table 2). These conditions were optimized for comparison mation of phosphanylphenol–nickel(0) and phosphanyl-
of variously substituted 2-phosphanylphenols[4a] and allow phenolate–nickel(II) complexes as major and minor precat-
an extension of the comparison to the new O-acyl deriva- alysts was indicated by 1H, 13C, and 31P NMR spectra
tives. As mentioned above, 1 served as a reference ligand. [∆δ(31P) = 44.8 and 57.7 ppm].[4a] Reaction of Ni(cod)2 with
The precatalysts were generated in situ by mixing toluene 3d, representing the behavior of ligands of type 3, leads at
solutions of Ni(1,5-cod)2 and 1, 2, or 3a–d at 0 °C (10 min) room temperature only to a nickel(0) complex (Scheme 2).
and stirring for 5 min at room temperature before pressuriz- When the components were mixed in a 1:2 molar ratio in
Table 2. Polymerization of ethylene with catalysts generated from 1–3 and Ni(cod)2.[a]
No.
Catalyst (µmol),[a]
Pstart (bar),
amount of C2H4 (g)
Conversion (%),
TON, TOFmax
PE (g),[c] m.p. (°C), d (gcm–3)
η (dLg–1),
Vin/olefin,
Me/C=C,
(gmol–1),
[b]
[d]
Mvis
,
[e]
MNMR (gmol–1)
Me/1000C[e]
1
2
3
4
5
6
7
8
1/Ni (100), 50, 14.6
1[f]/Ni (100), 50, 13.2
2/Ni (100), 50, 14.2
2/Ni (100), 50, 13.3
3a/Ni (100), 50, 13.0
3a/Ni (100), 50, 14.2
3b/Ni (120), 50, 12.3
3c/Ni (100), 40, 9.2
3c/Ni (100), 50, 14.6
3d/Ni (100), 40, 9.0
3d/Ni (100), 50, 12.9
3c/Ni (500), 20, 3.6
3d/Ni (500), 20, 3.5
2/Ni (200), 40, 10.2
2/Ni (500), 20, 3.5
74, 3875, n.d.
99, 4670, 7100
77, 3900, 250
99, 4700, n.d.
98, 4552, n.d.
98, 4955, 5600
98, 3600, 15500
97, 3180, n.d.
96, 4880, 30700
100, 3200, n.d.
100, 4600, 12500
97, 220, n.d.
100, 240, n.d.
30, 550, n.d.
0
10.9, 118–121, 0.958
13.0, 124.5–125.5, 0.950
11.1, 116–120, 0.955
0.15, 3600, –
–, –, 2450
0.92, 1.6, 7.6
0.82, 1.6, 9.0
0.95, 1.7, 9.3[g]
0.88, 1.5, 8.6[g]
0.83, 1.8, 10.5
–, –, 2620
13.1, 95–125 (127.5–129.5),[h] 0.950 0.16, 3940, 2440
12.8, 107–120 (127–129),[h] 0.960
13.9, 110–119, 0.958
12.0, 122.9–124.5, 0.946
8.2, 124.5–125.4, 0.950
13.7, 125–126, 0.955
8.6, 124.5–125.5, 0.945
12.6, 124.5–125.5, 0.965
3.1, 120.5–123.5, 0.956
3.3, 120.5–124.8, 0.950
2.8, 120–122, 0.949
–, –, 2400
0.18, 4620, 2200
–, –, 2100
–, –, 3400
–, –, 2150
–, –, 2150
–, –, 2650
–, –, 2050
–, –, 1530
–, –, 2050
–
[i]
0.91, 1.4, 9; 1/25[i]
0.84, 2.0, 13.5[j]
0.93, 2.8, 11.6
9
0.91, 1.9, 12.4[i]
0.87, 1.6, 10.5; 1/40[k]
0.84, 1.7, 9.0; 1/30[k]
0.75, 1.8, 11.8; 1/40[l]
0.88, 1.3, 11.9; 1/16[k]
0.96, 1.4, 9.3; 1/10[g]
–
10
11
12
13
14
15
–
[a] Equimolar amounts of 1, 2, or 3a–d and Ni(cod)2 were each dissolved in toluene (10 mL), united, and transferred to the autoclave.
Then the autoclave was pressurized and heated. [b] Conversion of C2H4 to oligomers and polymers, TON in mol/mol, TOFmax in mol/
0.73 [17]
molh. [c] Isolated and purified polymer. [d] Mvis calculated from intrinsic viscosity (in decalin at 135 °C) [η] = 3.8ϫ10–4Mvis
.
[e] Determined by 1H NMR integration. [f] Trace impurity by P-oxide 6. [g] Additional trace 1H signals at δ = 2.45 (t), 3.8 (br) ppm;
relative intensity 0.1/C=C groups. [h] Polymer purified with aqueous methanolic hydrochloric acid; m.p. in parenthesis after additional
1
1
extraction with MeOH. [i] Additional trace H signals at δ = 2.03 (s, Me), 2.45 (t, CH2) ppm. [j] Additional trace H signal at δ = 2.53
(t, CH2) ppm. [k] Additional trace 1H signals at δ = 2.98 (t, CH2), 3.76 (OMe) ppm; relative intensity ≈2:3; δ = 6.97, 8.06 (d, 3J = 8.8 Hz,
1
1
m-, o-H) ppm; 1/x = ratio CH2C(O)R to olefin groups estimated by H integration. [l] Additional trace H signal at δ = 2.98 (t, CH2)
ppm.
1236
www.eurjic.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2009, 1234–1242

Useful Ligands in the Nickel-Catalyzed Polymerization of Ethylene
THF, yellow crystals could be obtained after several days and nearly quantitative conversion of ethylene (Table 2).
by overlaying the concentrated solution with n-hexane. Pro- The rate of the reaction is, however, different for 1Ni, 2Ni,
longed storage led to an additional colorless precipitate and and 3a–dNi, as visualized by pressure–time plots (Figures 4
a few orange-brownish crystals of a NiII P,O-chelate com- and 5). The polymerization with 1Ni gives higher ethylene
plex. The yellow crystals were not suitable for X-ray diffrac- conversion (Table 2, Entries 1 and 2) than with the 2-di-
tion, but allowed characterization by multinuclear NMR phenylphosphanylphenol/Ni(cod)2 catalysts studied ear-
spectroscopy. [D8]THF solution NMR spectroscopic data lier.[4a] As in the latter, the variation in the conversions is
of crystals, washed with n-hexane and dried in vacuo, are larger than that for catalysts 3a–dNi and depends sensitively
assignable to a Ni0(3d)2 complex (4d), accompanied by 3d, on deviations from a 1:1 molar ratio of ligand and Ni.[6c]
probably by ligand dissociation in solution. According to The polymerization with catalyst 2Ni (Table 2, Entries 3 and
1H NMR integration, the ratio of 4-Me and 4-MeO groups 4) differs from those with 1Ni and 3a–dNi (Table 2, En-
is each 2:1, corresponding to a 1:1 molar ratio of 4d and tries 5–13) by a very long induction period and slow conver-
3d. Cleavage of the O-acyl or O-C(phenyl) bond by sion. The formation of the active catalyst from 2 is the criti-
nickel(0) can be excluded at this stage, as most of the signals cal step and also led in some experiments with 2Ni to low
of 3d were found to be only slightly shifted in 4d as well. conversion or inactivity, in particular for lower ethylene/cat-
Exceptions are nuclei involved in the coordination to nickel alyst ratios (Table 2, Entries 14 and 15), intended to trace
and adjacent carbon nuclei. Phosphorus displays a coordi- the fate of the Ph2P(O) group. The selectivity for linear vi-
nation chemical shift of ∆δ(31P) = 45.1 ppm, which is close nyl- and methyl-terminated polyethylene and for α-olefins
to ∆δ
=
42.8 ppm for [Ni0(2-Ph2P-4,6-tBu2C6H4OH- in the small amount of volatile oligomers, separated along
η1P)(PMe3)3],[9] whereas the ∆δ(31P) values for 2-phosphan- with 1,5-cod and solvent from the polymer by flash distil-
ylphenol or 2-phosphanylphenolate nickel(II) P,O-chelate lation, is however the same as with catalysts 1Ni. This hints
complexes are slightly (∆δ = 48.2 ppm)[10] or significantly at a closely related mechanism apart from the initial step.
larger (∆δ = 58–70 ppm).[4,9] Phosphanylphenolate coordi- The phosphorus signals of unconsumed 2 in the methanol
nation is additionally excluded for 4d by downfield shifts of extract of polyethylene obtained with 2Ni (Table 2, Entry 3)
the 13C-2 and 13C-i(Ph) nuclei compared to 3d (∆δ = 7.4, suggest slow cleavage of 2, and two small, equally intensive
3.1 ppm) and relatively small one-bond P–C coupling con- phosphorus singlets at δ = 27.7 ppm and δ = 32.2 ppm in
1
stants (1JP,C-2 = 11.9 Hz, JP,C-i = 31.9 Hz). For nickel(II) the same region as 31P resonances of Ph2P(O)OR com-
phosphanylphenolates, 13C-2 and 13C-i(Ph) signals are up-
field shifted, and the one-bond P–C couplings are consider-
ably larger (1JP,C = 45–51 Hz).[4,9] The coordination of a
second ligand at nickel is indicated by a doublet of doublet
for 13C-i(Ph). Additional interactions of the carbonyl oxy-
gen atom to Ni0 are suggested by strong downfield shift of
the ester 13C(O) resonance (δ = 199.7 ppm, ∆δ = 36.1 ppm).
Coordinated carbon monoxide is improbable, as no other
degradation products of 3d were observed, likewise a ben-
zoyl group at nickel, which in addition should display a
much stronger downfield shift.[11] A closer investigation of
compounds of type 4 will be subject of a separate study. It
is assumed that they are formed stepwise and that during
Figure 4. Pressure–time plots for batch polymerization of ethylene
with catalysts prepared in situ from Ni(cod)2 and 1 (solid), 2
(dashed), or 3a (dotted) in toluene; bath temperature 100 °C.
the short time of catalyst formation only one cod ligand is
replaced.
Scheme 2. Reaction of Ni(cod)2 with 3d at room temperature.
Figure 5. Pressure–time plots for batch polymerization of ethylene
with catalysts prepared in situ from Ni(cod)2 and 3a (solid), 3b
(dashed), 3c (dotted), or 3d (dashed/dotted) in toluene; bath tem-
Apart from the different nature of the precatalyst solu-
tions of 1Ni and 2Ni or 3a–dNi, heating with ethylene under
pressure led in almost all experiments to active catalysts perature 100 °C.
Eur. J. Inorg. Chem. 2009, 1234–1242
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1237

D. G. Yakhvarov, J. Heinicke et al.
FULL PAPER
pounds (δ = 25–29 ppm)[12] and phosphanylphenolate ene/catalyst ratio in the batch procedure. The products are
nickel complexes indicate cleavage of the O–P bond and mainly low molecular weight (MNMR 2100–4000) linear
formation of a phosphanylphenolate catalyst. In the poly- polyethylenes with CH3 and vinyl end groups, but a small
mers, 31P signals could not be detected. Nickel complexes amount of the polymers possess acyl end groups, as men-
with Ph2P(CH2)nP(O)Ph2 or Ph2PC6H4P(O)Ph2 ligands, tioned above, derived from ligand 3 employed in the cata-
which like 2 possess a phosphanyl and (P=)O donor group lyst. The ethylene consumption rate is not greatly different
but are unable to degrade to P–C=C–O– moieties, form cat- for catalysts 3a–dNi, but a clear dependence on the size of
ionic nickel catalysts that behave differently towards ethyl- the substituent R can be seen in the pressure–time plots
ene and provide isomeric mixtures of lower oligomers (Figure 5). The system with the bulky pivaloyl group re-
(mainly C4–C8).[13]
acted fastest, followed by the benzoyl- and 4-methoxyben-
The short induction period for catalysts 3a–dNi is attrib- zoyl systems, whereas the small acetyl group led to the slow-
uted to rapid formation and low thermal stability of com- est reaction. It is assumed that these findings are due to
plexes of type 4. It is known that Ni(cod)2 reacts with ben- intermolecular interactions of polymer R–C(O) end groups
zoic acid esters in the presence of triphenylphosphane un- with catalytically active nickel centers. They are facilitated
der mild heating (55–70 °C) mainly with scission of the for small acyl groups and cause slight retardation of the
acyl–OR bond, although for some R groups, acylO–R ethylene consumption, but stabilize the catalyst against de-
cleavage is also observed.[14] Complexes of type 4 prefer the activation routes and thus give higher conversions than
first route, acyl–OR cleavage (Scheme 3). Evidence is given most 2-phosphanylphenol-derived catalysts. In this light,
by detection of acyl end groups in the ethylene polymers the slow reaction with 2Ni may be attributable not only to
and the presence of 5 and 6 in the methanol extracts of the slow catalyst formation, as supported by detection of un-
crude polyethylenes. Compounds 5 and 6 were found also consumed 2, but also to stronger interactions of the
in the extracts of polyethylene obtained with 1Ni and char- Ph2P(O)OR species with the catalyst center.
acterize the ligand of the active catalyst by trapping in a
The acyl end groups in the polymers obtained with cata-
stable complex or by air oxidation. The NMR spectroscopic lysts 3aNi–dNi are detected in the 1H NMR spectra by char-
data of 5 and 6 are identical to those of independently syn- acteristic trace signals in addition to the usual signals for
thesized samples.
the linear vinyl-terminated polyethylene. The signals for
Me3CC(O) end groups are superimposed by strong polymer
methyl signals, but a tiny triplet for α-CH2 attached to the
terminal acyl group was detected at δ = 2.45 (at acetyl) and
2.53 (at pivaloyl) ppm, respectively (Table 2). For linear
alkyl aryl ketones the α-CH2 group is more strongly deshi-
elded and appears at δ = 2.95 ppm.[15] The polymers ob-
tained with 3cNi and 3dNi display this group clearly, in par-
ticular for a low ethylene/catalyst ratio, and allow a rough
estimation of the ratio of primarily formed polymer chains
with terminal acyl groups to Ni–H initiated chains with a
methyl end group formed in later polymerization cycles
(estimations see Table 2). In the polymers obtained with
3dNi, small signals for the 4-OMe group (δ = 3.76 ppm) and
the doublets for o- and m-aryl protons were also detected
3
(δ = 6.68, 8.06 ppm, d, J = 8.8 Hz, m-, o-H) (Figure 6).
Mechanistic Aspects
Polyethylene obtained with catalyst 2Ni is formed more
slowly but displays the same microstructure, linear chains
with methyl and vinyl end groups, as in polyethylene pre-
pared by means of catalyst 1Ni. Apart from the above-men-
tioned different initiation step, the mechanisms will thus be
the same. For catalysis with 3a–dNi, the occurrence of two
types of linear polyethylenes hints at two different chain-
growing cycles. The formation of small amounts of poly-
mers with R–C(O) end groups suggests that, on heating,
Scheme 3. Proposed mechanism of ethylene polymerization by 3a–
dNi
.
The polymerization of ethylene with catalyst systems 3a– Ni0 of the primary complexes of type 4 inserts into the O-
dNi generally proceeds rapidly, almost quantitatively within acyl bond and generates 2-phosphanylphenolate Ni–C(O)R
1–1.5 h (including initiation period) and with good repro- species that allow addition and insertion of ethylene. The
ducibility. The turnover numbers were limited by the ethyl- growth of this chain is terminated by β-hydride elimination,
1238
www.eurjic.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2009, 1234–1242

Useful Ligands in the Nickel-Catalyzed Polymerization of Ethylene
1
ylsilane for H and 13C and H3PO4 (85%) for 31P. Coupling con-
stants refer to JH,H in 1H and JP,C in 13C NMR spectroscopic data
unless indicated otherwise. Assignment numbers of the phenol ring
follow the nomenclature, P-phenyl nuclei are indicated by o, m, p,
i, benzoic acid nuclei by oЈ, mЈ, pЈ, iЈ. IR spectra were recorded
with an FTIR spectrometer System 2000 (Perkin–Elmer), and mass
spectra on a single-focusing mass spectrometer AMD40 (Intectra).
Melting points were determined in a capillary and are uncorrected.
Elemental analyses were carried out with a CHNS-932 analyzer
from LECO by using standard conditions. 1H NMR measurements
of polyethylenes were carried out as described in ref.[4a] [η] was
determined in tetralin at 120 °C (Dr. D. Lilge, BASELL).[17]
2-Diphenylphosphanyl-4-methylphenyl Diphenylphosphinate (2). A
solution of BuLi (1.6  in hexane, 62.5 mL, 0.1 mol) was added
dropwise to 2-bromo-4-methylphenol (6.0 mL, 49.7 mmol) dis-
solved in diethyl ether (100 mL), followed by dropwise addition of
a solution of Ph2PCl (9.0 mL, 50.1 mmol) in diethyl ether (10 mL).
After stirring overnight, Ph2P(O)Cl (9.7 mL, 50.9 mmol) in diethyl
ether (10 mL) was added to the suspension without cooling. A large
amount of insoluble material precipitated. Stirring was continued
for 4 h, the solvent was evaporated in vacuo, and the residue was
extracted with toluene (150 mL). After evaporation of toluene in
vacuo, the residue was crystallized from ethanol to give 9.6 g (39%)
of colorless 2. M.p. 148–150 °C. C31H26O2P2 (492.49): calcd. C
Figure 6. Characteristic 1H NMR signals (in C6D5Br, 104 °C) of 4-
MeOC6H4COCH2 in addition to signals from vinyl end groups in
polyethylene obtained with 3dNi
.
which generates vinyl end groups and NiH starting catalysts
for the subsequent chain growing cycles yielding CH3 and
vinyl-terminated linear polyethylenes as the main body of
the polymer product.
1
75.60, H 5.32; found C 75.46, H 5.57. H NMR: δ = 2.07 (s, 3 H,
3
4
CH3), 6.46 (m, 1 H, 3-H), 6.97 (dd, J = 8.3 Hz, J = 2.2 Hz, 1 H,
5-H), 7.27–7.35 (m, 14 H), 7.42 (“tq”, 3J = 7.5, 7.3 Hz, J = 2.8,
1.4 Hz, 2 H, pЈ-H), 7.49 (ddd, 3J = 8.2 Hz, JPH = 4.6 Hz, J =
4
Conclusions
0.9 Hz, 1 H, 6-H), 7.76 (br. dd, JP,H = 12.6 Hz, 3J = 7 Hz, 4J ≈
3
1.4 Hz, 4 H, oЈ-H) ppm. 13C{1H, CH COSY} NMR: δ = 20.7 (s,
2-Phosphanylphenyl carboxylic acid esters (3) and
Ni(1,5-cod)2 form highly active catalysts for the polymeriza-
3
1
3
CH3), 119.0 (d, J = 3.0 Hz, CH-6), 127.8 (dd, J = 14.1 Hz, J =
tion of ethylene to linear polymers with vinyl and methyl 7.1 Hz, Cq-2), 128.2 (d, 3J = 13.5 Hz, 4 CH-mЈ), 128.5 (d, 3J =
7.1 Hz, 4 CH-m), 128.8 (s, 2 CH-p), 129.9 (s, Cq-4), 130.8 (s, CH-
5), 131.8 (dd, J = 10.6 Hz, J = 2.1 Hz, 4 CH-oЈ), 132.1 (d, J =
end groups. Detection of a small amount of the acyl end
group instead of the methyl end group shows that the cata-
lyst for the first chain growth cycle is formed by insertion
of nickel(0) into the O-acyl bond of the ligands. Termina-
tion by β-hydride elimination then starts NiH-initiated po-
lymerization cycles yielding methyl- and vinyl-terminated
chains. The acyl groups in the polymer interact weakly with
the catalyst metal, which leads to a small retardation of the
ethylene consumption rate tBu Ͻ Ph, 4-MeOC6H4 Ͻ Me,
but also to stabilization of the catalyst and very high con-
version of ethylene. 2-Phosphanylphenyl phosphinate 2 also
forms ethylene polymerization catalysts with Ni(1,5-cod)2,
but the reaction rate is low. Slow O–P(O)Ph2 cleavage to
2
6
4
1
2
2.9 Hz, 2 CH-pЈ), 132.7 (d, J = 144.5 Hz, 2 Cq-iЈ), 134.0 (d, J =
1
21.1 Hz, 4 CH-o), 134.1 (s, CH-3), 135.7 (d, J = 9.7 Hz, 2 Cq-i),
151.5 (dd, 2J = 17, 8 Hz, Cq-1) ppm. 31P{1H} NMR: δ = –15.7,
30.4 (2 d, JP,P = 4.2 Hz, PIII, PV) ppm. IR (KBr): ν
region =
4
˜
P=O
1232 (vs), 1204 (vs) cm–1. MS (EI, 70 eV): m/z (%) = 492 (100)
[M+], 415 (64) [M+ – Ph], 308 (36) [M+ – PPh2], 274 (46) [M+
–
OP(O)Ph2], 214 (18), 202 (47) [P(O)Ph2+], 186 (13) [PPh2+], 78 (28)
[Ph+].
2-Diphenylphosphanyl-4-methylphenyl Acetate (3a): A solution of
BuLi (1.6  in hexane, 125 mL, 0.2 mol) was added at –50 °C to 2-
bromo-4-methylphenol (12.0 mL, 99.3 mmol) dissolved in diethyl
ether (200 mL), the first half dropwise, the second part rapidly
form the active phosphanylphenolate catalyst is assumed to without cooling. After stirring for 4 h at 20 °C, the reagent was
cooled to –50 °C and a solution of Ph2PCl (17.9 mL, 99.7 mmol) in
diethyl ether (20 mL) was added dropwise. Stirring was continued
overnight, followed by dropwise addition of a solution of acetyl
chloride (8.6 mL, 120.9 mmol.) in diethyl ether (10 mL) at room
temperature. The precipitate was filtered off and washed with di-
ethyl ether. The solvent was evaporated from the filtrate. The resid-
ual oil was distilled under high vacuum (b.p. 155–160 °C/
3.8ϫ10–4 Torr) to give 18.4 g (55%) of 3a. The compound solidi-
fied slowly at 20 °C, faster at 40–50 °C. M.p. 86–90 °C. C21H19O2P
be the main reason, as the microstructure of the polyethyl-
ene is the same as with 1Ni catalysts.
Experimental Section
General Methods: All manipulations and reactions were carried out
under an atmosphere of dry argon by using Schlenk techniques and
freshly distilled dry solvents. The acyl chlorides and KOH-dried
triethylamine were recondensed or distilled before use. 2-Diphenyl-
phosphanyl-4-methylphenol (1)[16] was prepared as reported earlier.
Other chemicals were purchased. Ethylene (99.5%, Air Liquide)
was used without further treatment. NMR spectra were recorded
at 25 °C with a multinuclear FT-NMR spectrometer ARX300
(Bruker) at 300.1 (1H), 75.5 (13C), and 121.5 (31P) MHz in CDCl3
unless indicated otherwise. Chemical shift references are tetrameth-
1
(334.35): calcd. C 75.44, H 5.73; found C 75.96, H 5.46. H NMR:
3
δ = 1.95 (s, 3 H, CH3), 2.20 (s, 3 H, 4-CH3), 6.64 (dd, JP,H
=
=
4
4.7 Hz, 4J = 2.1 Hz, 1 H, 3-H), 7.01 (dd, 3J = 8.2 Hz, JP,H
3
4
4.3 Hz, 1 H, 6-H), 7.17 (dd, J = 8.2 Hz, J = 2.1 Hz, 1 H, 5-H),
7.28–7.36 (m, 10 H, phenyl) ppm. 13C{1H} NMR: δ = 20.4, 20.90
3
(2 s, CH3), 122.2 (s, C-6), 128.45 (d, J = 7.5 Hz, 4 C-m), 128.9 (s,
2 C-p), 129.7 (d, J = 15.1 Hz, C-2), 130.54 (s, C-5), 130.7 (d, J =
1
3
Eur. J. Inorg. Chem. 2009, 1234–1242
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1239

D. G. Yakhvarov, J. Heinicke et al.
FULL PAPER
3.8 Hz, C-4), 133.8 (d, 2J = 20.4 Hz, 4 C-o), 135.57 (d, 1J = 7.5 Hz,
2 C-i), 135.63 (d, J = 2.3 Hz, C-3), 150.5 (d, J = 17.4 Hz, C-1),
(2.5  in hexane, 21.6 mL, 54.0 mmol) at –70 and 20 °C, stirred for
4 h at 20 °C, followed by dropwise addition of Ph2PCl (4.8 mL,
26.8 mmol) dissolved in diethyl ether (10 mL) at –70 °C. After stir-
2
2
169.0 (CO) ppm. 31P{1H} NMR: δ = –15.3 ppm. IR (KBr): ν
˜
C=O
= 1765 (s) cm–1. MS (EI, 70 eV, 90 °C): m/z (%) = 335 (16), 334 ring overnight at room temperature, a solution of 4-methoxyben-
(65) [M+], 320 (21), 319 (100) [M+ – CH3], 292 (61), 291 (96) [M+
–
zoyl chloride (4.33 mL, 32.0 mmol) in diethyl ether (10 mL) was
added dropwise at 20 °C. The mixture was stirred at this tempera-
ture for 3 d. Then the precipitate was filtered off and washed with
diethyl ether (3ϫ30 mL) and dried in vacuo. The filtrate contained
mainly impurities, only a small portion of product. Pure product
(7.38 g, 65%) was extracted from the precipitate with dichloro-
methane (50 mL). M.p. 133.8–134.7 °C. C27H23O3P (426.44): calcd.
COCH3], 214 (76) [M+ – COCH3 – Ph], 202 (59), 184 (50) [PPh2+],
108 (37) [M+ – COCH3 – PPh2], 44 (79).
2-Diphenylphosphanyl-4-methylphenyl 2,2-Dimethylpropanoate (3b):
A solution of pivaloyl chloride (0.46 mL, 3.8 mmol) was added
dropwise at 0 °C to a solution of 1 (0.73 g, 2.5 mmol) and Et3N
(0.52 mL, 3.8 mmol) in diethyl ether (20 mL). Stirring was contin-
ued overnight, and the precipitate was filtered off and washed with
diethyl ether (3ϫ). Ether was removed in vacuo leaving 0.87 g
(92%) of crude NMR spectroscopic pure 3b, which was crystallized
from hexane. M.p. 109–110 °C. C24H25O2P (376.44): calcd. C 76.58,
H 6.69; found C 76.37, H 6.90. 1H NMR: δ = 1.12 (s, 9 H, CMe3),
1
C 76.04, H 5.44; found C 75.68, H 5.35. H NMR: δ = 2.23 (s, 3
3
4
H, 4-CH3), 3.85 (s, 3 H, 4-OCH3), 6.63 (dd, JP,H = 4.6 Hz, J =
2.0 Hz, 1 H, 3-H), 6.81 (mAAЈBBЈ
,
3J = 8.9 Hz, 2 H, mЈ-H), 7.16
3
4
3
(dd, J = 8.1 Hz, JP,H = 3.8 Hz, 1 H, 6-H), 7.21 (dd, J = 8.1 Hz,
4J = 1.8 Hz, 1 H, 5-H), 7.29–7.36 (m, 10 H, phenyl), 7.76 (mAAЈBBЈ
,
3J = 8.9 Hz, 2 H, oЈ-H) ppm. 13C{1H} NMR: δ = 20.98 (4-CH3),
55.34 (p-OMe), 113.40 (2 CH-mЈ), 121.56 (Cq-iЈ), 122.29 (CH-6),
3
2.17 (s, 3 H, 4-CH3), 6.51 (ddd, JP,H = 4.2 Hz, J = 2.2, 0.5 Hz, 1
4
H, 3-H), 6.99 (dd, 3J = 8.2 Hz, JP,H = 4.4 Hz, 1 H, 6-H), 7.16
3
1
128.41 (d, J = 6.8 Hz, 4 CH-m), 128.84 (s, 2 CH-p), 130.0 (d, J
= 13.0 Hz, Cq-2), 130.49 (s, CH-5), 132.16 (2 CH-oЈ), 133.69 (2 Cq-
i), 134.00 (d, 2J = 21.0 Hz, 4 C-o), 135.48 (CH-3), 135.64 (Cq-4),
150.65 (d, 2J = 16.1 Hz, Cq-1), 163.56 (CO), 164.05 (Cq-pЈ) ppm.
3
4
(dd“t”, J = 8.2 Hz, J = 2.2 Hz, J ≈ 0.5 Hz, 1 H, 5-H), 7.25–7.35
(m, 10 H, phenyl) ppm. 13C{1H} NMR: δ = 21.0 (s, 4-Me), 26.9 (s,
CMe3), 39.1 (s, CMe3), 121.9 (s, C-6), 128.5 (d, 3J = 7.1 Hz, 4C-
m), 128.8 (s, 2 C-p), 129.5 (d, 1J = 14.5 Hz, C-2), 130.5 (s, C-5),
31P{1H} NMR: δ = –15.2 ppm. IR (KBr): ν
= 1732 (s) cm–1.
˜
C=O
3
2
131.8 (d, J = 3.6 Hz, C-4), 133.8 (d, J = 21.2 Hz, 4 C-o), 134.0
MS (EI, 70 eV, 120 °C): m/z (%) = 428 (7), 427 (23) [M+ + H], 351
(13), 350 (45), 334 (17), 213 (6), 183 (4), 136 (10), 135 (100), 197
(12), 77 (16).
1
2
(d, J = 7.5 Hz, 2 C-i), 135.4 (s, C-3), 151.0 (d, J = 17.5 Hz, C-1),
176.4 (CO) ppm. 31P{1H} NMR: δ = –15.7 ppm. IR (KBr): ν
˜
C=O
= 1747 (s) cm–1. MS (EI, 70 eV): m/z (%) = 376 (24) [M+], 375 (78),
361 (18), 322 (64), 321 (99), 295 (73), 294 (100), 273 (48), 213 (78),
57 (81).
Reaction of 3d with Ni(cod)2 to form 4d: Solid 3d (0.45 g,
1.06 mmol) was added to a solution of Ni(cod)2 (0.14 g, 0.51 mmol)
in THF (20 mL). After stirring for 12–15 h the solution was con-
centrated in vacuo and overlaid with n-hexane. After several days
small orange-yellow column-shaped crystals and small white crys-
tals deposited. The orange-yellow crystals displayed very weak X-
ray diffraction and multiply split reflections that, despite repeated
attempts, did not allow a crystal structure analysis. NMR spectra
of the orange-yellow crystals separated from the mixture, washed
with n-hexane and dried in vacuo, hint at a Ni0(3d)3 complex,
which in solution dissociates into Ni0(3d)2 (4d) and 3d (ratio of 4-
2-Diphenylphosphanyl-4-methylphenyl Benzoate (3c): As described
for 3a,
a
solution of 2-bromo-4-methylphenol (3.23 mL,
26.7 mmol) in diethyl ether (50 mL) was dilithiated with BuLi
(2.5  in hexane, 21.6 mL, 54.0 mmol) at –50 and 20 °C, stirred for
4 h at 20 °C, followed by dropwise addition of Ph2PCl (4.8 mL,
26.8 mmol) dissolved in diethyl ether (5 mL) at –50 °C. After stir-
ring overnight at room temperature, a solution of benzoyl chloride
(3.70 mL, 32.0 mmol) in diethyl ether (5 mL) was added dropwise
at 20 °C. The mixture was stirred at this temperature for 1 d. Then
the precipitate was filtered off and washed with diethyl ether
(3ϫ30 mL). The filtrate was washed with air-free distilled water
(3ϫ) and then dried with Na2SO4. After evaporation of 60% of the
solvent and cooling to +4 °C, 3.0 g of 3c crystallized. M.p. 125.1–
125.6 °C. A further 3.8 g portion of product was isolated by extrac-
tion of the precipitate with dichloromethane (50 mL). Total yield:
6.8 g (64%). C26H21O2P (396.42): calcd. C 78.78, H 5.34; found C
1
Me and 4-MeO signals each 2:1 by H NMR integration). A trace
amount of noncoordinated 1,5-cod was detected as an impurity.
Data for the yellow crystals: C72H75NiO6P3 (1187.98): calcd. C
72.79, H 6.36; found C 73.04, H 6.77. Complex 4d: 1H NMR ([D8]-
3
THF): δ = 2.07 (s, 4-CH3), 3.77 (s, 4-OCH3), 6.69 (mAAЈBBЈ, J =
8.9 Hz, mЈ-H), 7.0–7.2 (m, 5 H, aryl), 7.27–7.31 (m, 2 H, aryl),
7.35–7.41 (m, 3 H, aryl), 7.34 (mAAЈBBЈ,
3J = 8.8 Hz, oЈ-H) ppm.
13C{1H} NMR ([D8]THF): δ = 20.81 (4-Me), 55.50 (p-OMe),
113.67 (2 CH-mЈ), 122.53 (Cq-iЈ), 124.42 (CH-6), 128.55 (d, 3J =
11.1 Hz, 4 CH-m), 129.22 (s, 2 CH-p), 131.53 (CH-5), 132.73 (2
1
3
78.53, H 5.26. H NMR: δ = 2.23 (s, 3 H, 4-CH3), 6.65 (dd, JP,H
= 4.8 Hz, 4J = 1.8 Hz, 1 H, 3-H), 7.17 (dd, 3J = 8.4 Hz, JP,H
=
4
3
4
2
3.8 Hz, 1 H, 6-H), 7.22 (dd, J = 8.4 Hz, J = 1.8 Hz, 1 H, 5-H),
CH-oЈ), 133.57 (d, J = 15.2 Hz, 4 CH-o), 135.85 (s, Cq-4), 136.37
7.28–7.37 (m, 12 H, phenyl), 7.52 (tt, J ≈ 7.4 Hz, 4J = 1.3 Hz, 1
3
(d, 2J = 17.7 Hz, CH-3), 137.39 (d, 1J = 11.9 Hz, Cq-2), 137.65 (dd,
1J = 31.9 Hz, 3J = 4.0 Hz, 2 Cq-i), 150.79 (s, Cq-1), 163.90 (Cq-
pЈ), 199.70 (br. t-shape, CO) ppm. 31P{1H} NMR ([D8]THF): δ =
29.88 ppm.
H, pЈ-H), 7.52 (dm, 3J = 8.3 Hz, 2 H, oЈ-H) ppm. 13C{1H} (DEPT)
NMR: δ = 21.0 (4-CH3), 122.2 (d, 4J = 1.8 Hz, CH-6), 128.2 (2
CH-mЈ), 128.5 (d, 3J = 7.2 Hz, 4 C-m), 128.9 (s, 2 CH-p), 129.2
(Cq-4), 130.1 (d, 1J = 14.8 Hz, Cq-2), 130.1 (s, 2 CH-oЈ), 130.5 (CH-
5), 133.2 (CH-pЈ), 133.8 (d, 2J = 2.1 Hz, C-3), 134.0 (d, 2J =
20.6 Hz, 4 C-o), 135.6 (d, 1J = 14.4 Hz, 2 C-i), 135.6 (Cq, C-iЈ),
Complex 5: A solution of 1 (1.6 g, 5.48 mmol) in methanol (5 mL)
was added to a solution of NiCl2(H2O)6 (650 mg, 2.74 mmol). An
excess amount of triethylamine (1.1 g) was added until the green
color disappeared and all material was orange. Volatiles were re-
moved in vacuo, triethylamine hydrochloride was extracted with
water, and the residue crystallized from dichloromethane/methanol
(1:1) to give 1.1 g (80%) of 4. C38H32NiO2P2 (641.30): calcd. C
150.6 (d, J = 16.7 Hz, C-1), 164.4 (CO) ppm. 31P{1H} NMR: δ =
2
–15.2 ppm. IR (KBr): ν
= 1739 (s) cm–1. MS (EI, 70 eV,
˜
C=O
160 °C): m/z (%) = 398 (6), 397 (23), 396 (12) [M+], 368 (9), 320
(50) [M+ + 1 – Ph], 292 (20) [M+ + 1 – COPh], 213 (33), 183 (16)
[PPh2+], 105 (100).
1
71.17, H 5.03; found C 70.45 (incomplete combustion), H 5.15. H
2-Diphenylphosphanyl-4-methylphenyl 4-Methoxybenzoate (3d): As
described for 3a, a solution of 2-bromo-4-methylphenol (3.23 mL,
26.7 mmol) in diethyl ether (50 mL) was dilithiated with BuLi
NMR: δ = 2.05 (s, 3 H, 4-Me), 6.47 [br. t (A of AMXXЈ, X = XЈ
3
4
= P), |J+JЈ| = 10–11 Hz, 2 H, 6-H], 6.89 (dt, J = 8.5 Hz, J +5JP,H
3
= 5.6 Hz, 2 H, 5-H), 6.97 (br. d, JP,H = 8.7 Hz, 2 H, 3-H), 7.10
1240
www.eurjic.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2009, 1234–1242

Useful Ligands in the Nickel-Catalyzed Polymerization of Ethylene
3
3
4
4
(br. t, J = 7.4–7.6 Hz, 4 H, m-H), 7.32 (br. t, J = 7.3–7.5 Hz, 2 10.5 Hz, 4 C-o), 132.5 (d, J = 2.6 Hz, C-p), 135.3 (d, J = 2.4 Hz,
H, p-H), 7.42 (td, JPH ≈ 5.3 Hz, 4 H, o-H) ppm. 31P{1H} NMR: C-4), 161.81 (d, J = 2.7 Hz, Cq-2) ppm. 31P NMR: δ = 39.7 ppm.
δ = 34.3 ppm.
3
2
MS (EI, 70 eV, 180 °C): m/z (%) = 309 (35), 308 (85) [M+], 307
(100), 229 (20), 213 (16), 201 (14), 199 (16), 183 (14), 152 (11), 77
(21).
2-(Diphenylphosphoryl)-4-methylphenol (6): As described for 3a, a
solution of 2-bromo-4-methylphenol (6.0 mL, 49.7 mmol) in di-
ethyl ether (100 mL) was dilithiated with BuLi (1.6  in hexane,
62.5 mL, 0.1 mol) at –70 and 20 °C, stirred for 4 h at 20 °C. Then
a solution of Ph2P(O)Cl (9.0 mL, 47.2 mmol) in diethyl ether
(10 mL) was added dropwise at –50 °C. The suspension was stirred
overnight at 20 °C and then acidified with a solution of glacial
acetic acid (3.3 g, 0.6 mol) in diethyl ether (10 mL). The white pre-
cipitate was filtered off and washed with diethyl ether, and the resi-
due was extracted with warm absolute ethanol. The crystals formed
overnight were separated, washed with cold EtOH, and dried in
vacuo to give 5.1 g (35%) of 5. M.p. 188 °C. The yield is not opti-
mized. The filtrate contains more product along with another phos-
phorus compound with δ(31P) = 32.0 ppm. C19H17O2P (308.32):
Ethylene Polymerization: The corresponding phosphanylphenol de-
rivative (Table 2) and Ni(cod)2 were dissolved in toluene (each in
10 mL, unless indicated otherwise), cooled to 0 °C (10 min), and
mixed. The resulting slightly deeper yellow (with 2 and 3) or brown
(with 1) solution was stirred at room temperature for 5 min. If indi-
cated in Table 2 an additive was added. The mixture was transfer-
red to an argon-filled stainless steel autoclave (75 mL) (details see
in ref.[4a]). The autoclave was pressurized with ethylene (see Table 2)
and placed into a heating bath at 100 °C. After heating for ca.
15 h, cooling to room temperature and weight control, unconverted
ethylene was allowed to escape through a cooling trap (–78 °C).
The polymer with solvent and oligomers was transferred to a flask,
and all volatiles were flash-distilled at 3–4 mbar/70–80 °C to a cool-
ing trap (–196 °C) for GC analysis. The polymer residue was ex-
tracted in most samples (except Entries 4–6) first with methanol
1
calcd. C 74.02, H 5.56; found C 74.15, H 5.99. H NMR: δ = 2.18
(s, 3 H, 5-Me), 6.73 (ddd, JP,H = 13.4 Hz, J = 1.6, 0.5 Hz, 1 H, 6-
H), 6.90 (dd, J = 8.5 Hz, JPH = 4.9 Hz, 1 H, 3-H), 7.21 (dm, J
3
3
4
3
= 8.6, 1.6 Hz, 1 H, 4-H), 7.44–7.53 (m, 4 H, m-H), 7.55–7.62 (m, (50 mL) and then methanol (36.5 mL)/concentrated aqueous HCl
3
4
2 H, p-H); 7.69 (m, JP,H = 12.3 Hz, J = 1.5 Hz, 2 H, o-H), 10.9 (13.5 mL) by stirring overnight at room temperature. The polymer
(v br, 1 H, OH) ppm. 13C{1H} NMR: δ = 20.5 (4-Me), 110.5 (d,
1J = 103.6 Hz, C-i), 118.5 (d, 3J = 8.5 Hz, C-3), 128.1 (d, 3J =
12.4 Hz, Cq-5), 128.7 (d, 3J = 12.8 Hz, 4 C-m), 131.5 (d, 2J =
10.0 Hz, C-6), 131.9 (d, 1J = 104.8 Hz, 2 C-i), 132.0 (d, 3J =
was thoroughly washed with methanol and dried in vacuo. Conver-
sion and characteristic polymer data are given in Table 2. GC
analysis of the flash distillate apart from toluene indicated small
amounts of 1,5-cod and linear α-olefins C4–C12 (mostly C6 Յ C8
Table 3. Crystallographic data of compounds 3b–d.
Compound
3b
3c
3d
Empirical formula
Formula weight
Temperature [K]
Wavelength [Å]
Crystal system
Space group
a [Å]
b [Å]
c [Å]
α [°]
β [°]
C24H25O2P
376.41
293
C26H21O2P
396.40
100
0.71073
Monoclinic
P21/n
9.1570(6)
24.1370(16)
10.0608(6)
90
110.021(3)
90
2089.3(2)
4
1.260
C27H23O3P
426.42
100
0.71073
Triclinic
0.71073
Triclinic
¯
¯
P1
P1
7.792(2)
11.195(3)
13.483(4)
70.74(3)
81.82(2)
73.80(2)
1064.6(5)
2
9.1385(7)
9.8116(7)
13.2947(9)
87.572(4)
81.590(4)
69.623(4)
1105.39(14)
2
γ [°]
Volume [Å3]
Z
Dcalcd. [Mgm–3]
Absorption coefficient [mm–1]
F(000)
1.174
0.144
400
1.281
0.151
448
0.151
832
Crystal size [mm3]
θ range for data collection
Index ranges
0.60ϫ0.40ϫ0.35
2.7 to 26.3
–9ՅhՅ0
–13ՅkՅ13
–16ՅlՅ16
4576
4256 [R(int) = 0.049]
98.6 to θ =
26.29
0.20ϫ0.20ϫ0.10
2.51 to 30.51
–13ՅhՅ13
–34ՅkՅ34
–14ՅlՅ14
53765
6379 [R(int) = 0.0482]
100 to θ =
30.50
0.2ϫ0.1ϫ0.1
2.64 to 30.51°
–13ՅhՅ13
–14ՅkՅ14
–18ՅlՅ18
27382
6725 [R(int) = 0.0310]
99.8 to θ =
30.50
Reflections collected
Independent reflections
Completeness [%] to
θ = 30.00°
Absorption correction
Refinement method
Data/restraints/parameters
Goodness-of-fit on F2
Final R indices
None
None
None
Full-matrix least-squares on F2 Full-matrix least-squares on F2 Full-matrix least-squares on F2
4256/0/248
0.991
R1 = 0.0583
wR2 = 0.1445
for I Ͼ 2σ(I)
R1 = 0.2020
wR2 = 0.1871
6379/0/263
1.044
6725/0/282
1.019
R1 = 0.0432
wR2 = 0.1047
for I Ͼ 2σ(I)
R1 = 0.0560
wR2 = 0.1126
0.494 and –0.234
R1 = 0.0377
wR2 = 0.0973
for I Ͼ 2σ(I)
R1 = 0.0489
wR2 = 0.1049
0.411 and –0.239
R indices (all data)
Largest diff. peak and hole [eÅ–3] 0.633 and –0.211
Eur. J. Inorg. Chem. 2009, 1234–1242
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1241

D. G. Yakhvarov, J. Heinicke et al.
FULL PAPER
J. Heinicke, M. He, A. Dal, H.-F. Klein, O. Hetche, W. Keim,
U. Flörke, H.-J. Haupt, Eur. J. Inorg. Chem. 2000, 431–440; c)
G. A. Nesterov, V. A. Zakharov, G. Fink, W. Fenzl, J. Mol.
Catal. 1991, 69, 129–136; d) J. Heinicke, N. Peulecke, M. K.
Kindermann, P. G. Jones, Z. Anorg. Allg. Chem. 2005, 631, 67–
73.
Ն 1.5-cod Ͼ C10) and trace amounts of isomers (Ͻ5%). In the
methanol extract small amounts of 1,5-cod and linear α-olefins
from C10 to C22 (C14 Ͼ C16 Ͼ C12 Ͼ others) were detected; the
isoolefin content was very low.
Crystal Structure Analysis of 3b–d: A crystal of 3b was measured
with an Enraf Nonius CAD-4 diffractometer at 20 °C (Mo-Kα radi-
ation, ω/2θ scan technique, 2θ Ͻ 52.6°). The structure was solved
by direct methods with the use of SIR[18] and refined by full-matrix
least-squares by using SHELXL97.[20] All non-hydrogen atoms
were refined anisotropically. The hydrogen atoms were refined by
using a riding model. All calculations were performed with the use
of WinGX.[21] Cell parameters, data collection, and data reduction
were performed by using MolEN.[19]
[5] Personal account: J. Heinicke, N. Peulecke, M. Köhler, M. He,
W. Keim, J. Organomet. Chem. 2005, 690, 2449–2457.
[6] a) J. Heinicke, E. Musina, N. Peulecke, A. A. Karasik, M. K.
Kindermann, A. Dobrynin, I. Litvinov, Z. Anorg. Allg. Chem.
2007, 633, 1995–2003; b) J. Heinicke, M. He, A. A. Karasik,
I. O. Georgiev, O. G. Sinyashin, P. G. Jones, Heteroat. Chem.
2005, 16, 379–390; c) J. Heinicke, M. Köhler, N. Peulecke, W.
Keim, J. Catal. 2004, 225, 16–23; d) C. Müller, L. J. Ackerman,
J. N. H. Reek, P. C. J. Kramer, P. W. N. M. Van Leeuwen, J.
Am. Chem. Soc. 2004, 126, 14960–14963; e) J. Heinicke, M.
Koesling, R. Brüll, W. Keim, H. Pritzkow, Eur. J. Inorg. Chem.
2000, 299–305; f) K. A. Ostoja-Starzewski, J. Witte, Angew.
Chem. 1987, 99, 76–77; Angew. Chem. Int. Ed. Engl. 1987, 26,
63–64; g) D. M. Singleton (Shell Oil Company), US 4 472 522,
1984; D. M. Singleton (Shell Oil Company), US 4 472 525,
1984.
Crystals of 3c and of 3d were measured with a Bruker APEX-2
diffractometer at low temperature (Mo-Kα radiation, ω-scans, 2θ
Ͻ 61°). The structures were solved by direct methods and refined
as above. H atom positions were refined by using a riding model
or rigid methyl groups.[20]
Selected bond lengths, angles, and torsion angles of 3b–d are pre-
sented in Table 1, and the crystal data is summarized in Table 3.
CCDC-709792 (for 3b), -709793 (for 3c), and -709794 (for 3d) con-
tain the supplementary crystallographic data for this paper. These
data can be obtained free of charge from The Cambridge Crystallo-
graphic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
[7] C.-Y. Guo, N. Peulecke, M. K. Kindermann, J. Heinicke, J. Po-
lym. Sci., Part A: Polym. Chem. 2009, 47, 258–266.
[8] J. Heinicke, M. Köhler, N. Peulecke, M. K. Kindermann, W.
Keim, M. Köckerling, Organometallics 2005, 24, 344–352.
[9] J. Heinicke, A. Dal, H.-F. Klein, O. Hetche, U. Flörke, H.-J.
Haupt, Z. Naturforsch., Teil B 1999, 54, 1235–1243.
[10] J. Heinicke, M. Köhler, N. Peulecke, M. K. Kindermann, W.
Keim, M. Köckerling, Organometallics 2005, 24, 344–352.
[11] a) H.-F. Klein, A. Bickelhaupt, T. Jung, G. Cordier, Organome-
tallics 1994, 13, 2557–2559; b) T. Yamamoto, T. Kohara, A.
Yamamoto, Bull. Chem. Soc. Jpn. 1981, 54, 2161–2168; c) K.
Maruyama, T. Ito, A. Yamamoto, J. Organomet. Chem. 1978,
157, 463–474.
[12] R. S. Edmundson in CRC Handbook of Phosphorus-31 Nuclear
Magnetic Resonance Data (Ed.: J. C. Tebby), Boca Raton,
1991, p. 360.
[13] a) I. Brassat, W. Keim, S. Killat, M. Mothrath, P. Mastrorilli,
C. F. Nobile, G. P. Suranna, J. Mol. Catal. A 2000, 157, 41–58;
b) I. Brassat, U. Englert, W. Keim, D. P. Keitel, S. Killat, G.-
P. Suranna, R. Wang, Inorg. Chim. Acta 1998, 280, 150–162.
Acknowledgments
We acknowledge financial support of this study by the Deutsche
Forschungsgemeinschaft (HE1997/8-2) and the Russian Founda-
tion for Basic Research (RFBR 06-03-32247-a) and gifts of chemi-
cals by the Fonds der Chemischen Industrie. We thank Dr. N. Peu-
lecke for initial catalytic tests, B. Witt (Greifswald) for gathering the
NMR spectroscopic data, and Dr. D. Lilge (BASELL, Ludwigs-
hafen) for viscosity measurements.
[1] a) W. Keim in Industrial Applications of Homogenous Catalysis [14] T. Yamamoto, J. Ishizu, T. Kohara, S. Komiya, A. Yamamoto,
(Eds.: A. Mortreux, F. Petit), Dordrecht, 1988, pp. 335–347; b)
W. Keim, Angew. Chem. 1990, 102, 251–260; W. Keim, Angew.
Chem. Int. Ed. Engl. 1990, 29, 235–244; c) W. Keim, Vysoko-
mol. Soedin., Ser. A 1994, 36, 1644–1652.
J. Am. Chem. Soc. 1980, 102, 3758–3764.
1
[15] J. Pouchert, J. Behnke, The Aldrich Library of 13C and H FT
NMR Spectra, Aldrich Chem. Comp. 1993, vol. 1, p. 642C, vol.
2, pp. 804A-C, 805A-C.
[2] a) S. D. Ittel, L. K. Johnson, M. Brookhart, Chem. Rev. 2000,
100, 1169–1204; b) V. C. Gibson, S. K. Spitzmesser, Chem. Rev.
2003, 103, 283–316; c) B. Rieger, L. Saunders, S. Kacker, S.
Striegler (Eds.), Late Transition Metal Polymerisation Catalysis,
Wiley-VCH, Weinheim, 2003; d) P. Braunstein, Chem. Rev.
2006, 106, 134–159.
[16] J. Heinicke, R. Kadyrov, M. K. Kindermann, M. Koesling,
P. G. Jones, Chem. Ber. 1996, 129, 1547–1560.
[17] R. Kuhn, H. Krömer, G. Roßmanith, Angew. Makromol.
Chem. 1974, 40/41, 361–389.
[18] A. Altomare, G. Cascarano, C. Giacovazzo, D. Viterbo, Acta
Crystallogr., Sect. A 1991, 47, 744–748.
[3] a) K. A. Ostoja-Starzewski, J. Witte, Angew. Chem. 1985, 97,
610–612; Angew. Chem. Int. Ed. Engl. 1985, 24, 599–601; b)
U. Klabunde, R. Mülhaupt, T. Herskovitz, A. H. Janowicz, J.
Calabrese, S. D. Ittel, J. Polym. Sci., Part A: Polym. Chem.
1987, 25, 1989–2003; c) U. Klabunde, S. D. Ittel, J. Mol. Catal.
1987, 41, 123–134.
[19] L. Straver, A. J. Schierbeek, MolEN, Structure Determination
System, Nonius B. V. 1994, vols. 1 and 2.
[20] G. M. Sheldrick, SHELXL-97: A Program for Refining Crystal
Structures, University of Göttingen, 1997.
[21] L. J. Ferrugia, WinGX Program System, University of Glas-
gow, 2001.
[4] a) J. Heinicke, M. Köhler, N. Peulecke, M. He, M. K. Kinder-
mann, W. Keim, G. Fink, Chem. Eur. J. 2003, 9, 6093–6107; b)
Received: November 18, 2008
Published Online: February 5, 2009
1242
www.eurjic.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2009, 1234–1242