Ethylene dimerization into 1-butene using 2-pyridylphosphole nickel catalysts

Article abstract of DOI:10.1016/j.jcat.2004.05.008

2-(2-Pyridyl) phospholes have been evaluated as ligands for the nickel-catalyzed oligomerization of ethylene. The ethylene pressure has a dramatic effect on the selectivity of the catalytic system. The total amount of butenes (C₄ fraction) did not vary with the ethylene pressure and was always > 90%. The catalytic system yielded almost only butene (97%) and the selectivity in 1-butene was remarkably high (80%). Nickel catalysts containing heteroditopic P,N ligands gave unusually high selectivity in 1-butene and constituted a demonstration of the potential of 2-pyridylphospholes as ligands in homogeneous catalysis.

Full text of DOI:10.1016/j.jcat.2004.05.008

Journal of Catalysis 226 (2004) 235–239
www.elsevier.com/locate/jcat
Research Note
Ethylene dimerization into 1-butene using 2-pyridylphosphole
nickel catalysts
Roberto F. de Souza a,∗, Katia Bernardo-Gusmão a, Guilherme A. Cunha a, Christophe Loup b,
François Leca b, Régis Réau b,∗
a
Instituto de Química, UFRGS, Av. Bento Gonçalves, 9500, P.O. Box 15003, CEP 91501-970, Porto Alegre, Brazil
Institute of Chemistry, UMR 6509, University of Rennes 1, Campus de Beaulieu, 35042 Rennes cedex, France
b
Received 2 March 2004; revised 30 April 2004; accepted 6 May 2004
Abstract
2-(2-Pyridyl)phospholes have been evaluated as ligands for the nickel-catalyzed oligomerization of ethylene. Under mild homogeneous
reaction conditions, high catalytic activities (> 15 cycles per second) were recorded. The selectivity depends on the ethylene pressure.
At 41 bar, high C fraction contents (until 97%) and high 1-butene selectivities (80% of the C fraction) were reached. This behavior is
4
4
interpreted as a consequence of the steric hindrance of the intermediate cationic 2-pyridylphosphole nickel hydrides.
2004 Elsevier Inc. All rights reserved.
Keywords: Nickel–phosphole; Ethylene; Oligomerization
1. Introduction
mogeneous [6] and heterogeneous phases [7]. Hence, the
development of new ligands in order to obtain selective Ni
catalysts is of great interest.
α-Olefins are key intermediates in the chemical indus-
try [1]. The industrial synthesis1 of 1-butene is nowadays
achieved using nickel or titanium catalysts in large indus-
trial processes like Alfabutol from IFP [3]. Alternatively
α-olefins are obtained in the SHOP process developed by
Shell [4]. The selectivity of the oligomerization process is
determined by the control of the carbon–carbon bond for-
mation and of the extension of the parallel double-bond
migration isomerization. The control of these reactions is
extremely attractive from both academic and technological
viewpoints. High-performance Ni catalysts that selectively
produce oligomers have been obtained via tailoring of the
surrounding ligands [5]. However, despite the intense re-
search efforts devoted to this area, there are still considerable
limitations. For example, very few catalytic systems are able
to produce selectively 1-butene from ethylene in both ho-
2-Pyridylphospholes 1 (Scheme 1) form tightly bonded
P,N chelates with transition-metal centers such as Pd(II) or
Ru(II) [8]. These donors are versatile ligands for performing
fundamental studies on the control of catalyst behavior since
their electronic and steric properties can be easily tuned by
varying the substitution patterns (Scheme 1). Furthermore,
the presence of two donors with different stereo- and elec-
tronic properties can induce selective processes in the metal
coordination sphere [8]. Therefore, we have systematically
investigated these P,N donors for the Ni-catalyzed oligomer-
ization of ethylene.
2. Experimental
All experiments have been performed under argon, previ-
ously dried on 3 Å molecular sieves and purified on BASF
R3-11 catalyst, using standard Schlenk tube techniques.
The solvents were distillated under argon on sodium/benzo-
phenone (cyclohexane, ethylic ether, tetrahydrofurane), on
P2O5 (dichloromethane, pentane, acetonitrile) or on 3 Å
molecular sieves (chlorobenzene), immediately before use.
Solids were dried under reduced pressure. Phospholes 1a–bꢀ
*
Corresponding authors.
E-mail addresses: rfds@iq.ufrgs.br (R.F. de Souza),
regis.reau@univ-rennes1.fr (R. Réau).
1
1-Butene can be obtained from refinery operations (C fraction distil-
4
lation, steam cracking or dehydrogenation of butane) or by synthesis using
oligomerization processes. For details, see [2].
0021-9517/$ – see front matter
doi:10.1016/j.jcat.2004.05.008
2004 Elsevier Inc. All rights reserved.

236
R.F. de Souza et al. / Journal of Catalysis 226 (2004) 235–239
Scheme 1. Synthesis of Ni(II)-complexes bearing 2-pyridylphosphole ligands.
[8d,9] and NiCl2(DME) (DME = 1,2-dimetoxyethane) [10]
were prepared as described in the literature. Diethylalu-
minum chloride (AlEt2Cl) was purchased from Aldrich and
used as received. High-resolution mass spectra were ob-
tained on a Varian MAT 311 or ZabSpec TOF Micromass
at CRMPO, University of Rennes, France. Elemental analy-
ses were performed by the CRMPO, University of Rennes,
France. The reaction products were quantitatively deter-
mined by gas chromatography using a 100-m-long capillary
Petrocol (polimethylsilicon) column, working between 36
and 250 ◦C (5 ◦C/min) in a Varian 3400CX equipment.
3. Results and discussion
Complexes 2a–bꢀ were readily obtained by reacting 2(2-
pyridyl)phospholes 1a–bꢀ with one equivalent of (DME)
NiCl2 in dichloromethane at room temperature (Scheme 1).
The paramagnetic complexes 2a–bꢀ were characterized by
high-resolution mass spectrometry and give satisfactory ele-
mental analyses.
Complexes 2a–bꢀ were evaluated as catalyst precursors
for ethylene oligomerization under homogeneous phase con-
ditions in the presence of an alkylaluminum cocatalysts. Un-
der mild reaction conditions (1.1 bar, 0 ◦C), all complexes
exhibit good to high catalytic activities, with turnover fre-
quencies (TOF) ranging from 1.9 to 3.4 s−1 (Table 1, entries
1–4). Surprisingly, the catalytic activity is not influenced
by the nature of the P substituent (R2) (entries 1 vs 3 and
2 vs 4) but varies with the nature of the R1 substituents (en-
tries 1 vs 2 and 3 vs 4). It is well known that thienyl and
phenyl groups possess very different electron-donatingprop-
erties [11], and we have established that the dienic moiety of
the phosphole ring is highly polarizable [12]. Thus, it seems
very likely that the substituent at the C5 position of the phos-
phole ring influences the donor ability of the pyridyl donor,
and thus the behavior of the catalyst. The highest catalytic
activities are observed when R1 is a thienyl group (entries
2 vs 1, 4 vs 3). In all cases, the major products are butenes
(71–89%); however, under such reaction conditions, the se-
lectivity in 1-butene is extremely low (1–2%) (Table 1).
The influence of the ethylene pressure was studied with
complex 2b, a precursor leading to one of the most active
catalyst. Increasing the pressure from 1.1 to 21.0 bar results
in a continuous enhancement of the catalytic activity (Ta-
ble 1, entries 2, 5, and 6) until 21 bar. Further augmentation
of the pressure has a negative effect on the TOF (Table 1,
entries 7–9). Such effects of the ethylene pressure on the
activity have already been observed with other Ni-based cat-
alytic systems [13].
2.1. Synthesis of the nickel complexes 2a–bꢀ
General procedure: A solution of 1-phenyl-2-(2-pyridyl)-
5-(phenyl)phosphole 1a (0.33 g, 0.90 mmol) in CH2Cl2
(5 mL) was added, at room temperature, to a solution of
(DME)NiCl2 (0.19 g, 0.90 mmol) in CH2Cl2 (5 mL). The
solution was stirred 3 h at room temperature, and the volatile
materials were removed under vacuum. The residue was
washed with diethyl ether (3 × 10 mL) and dried under vac-
uum. Complex 2a was obtained as yellow solid (0.39 g,
89%); HR-MS (FAB-mNBA): m/z: 460.0532 [M–Cl]+;
calcd for C25H22NPNiCl2 (497.025): C 60.41, H 4.46,
N 2.82; found: C 60.33, H 4.37, N 2.89.
All complexes have been synthesized using the same pro-
cedure and have been characterized by high-resolution mass
spectrometry and elemental analyses.
2.2. Oligomerization reactions
Ethylene oligomerization reactions were performed us-
ing a 120-mL glass reactor (ambient pressure reactions) or
a 250-mL Pressure autoclave, double walled and with mag-
netically driven mechanical stirring (for reaction performed
until 41.0 bar). The reaction temperature was controlled by
an external circulation bath set at the reaction run temper-
ature. In a typical run, 33 µmol of the nickel complex and
50 mL of chlorobenzene were feed, the system was saturated
with ethylene, and then 1.2 mL of a 1.8 mol/L solution of
the alkylaluminum cocatalyst was added (Al/Ni molar ra-
tio of 70). The ethylene pressure was then enhanced to the
run value and kept constant during the reaction. After the de-
sired reaction time the reactor was cooled to −10 ◦C, and the
products were withdrawn and analyzed by gas chromatogra-
phy.
Of particular interest, the ethylene pressure has a dra-
matic effect on the selectivity of the catalytic system. The
total amount of butenes (C4 fraction, column 8) does not
vary with the ethylene pressure and is always higher than
90%. However, the selectivity in 1-butene increases from 1%
at 1.1 bar (entry 2) to 73% at 31 bar (entry 8). This trend
is also observed with precursor 2a as illustrated in Fig. 1.
Increasing the pressure from 1.1 to 41 bar at 10 ◦C with
2a results in an increase of the 1-butene selectivity from 1
to 80% (Table 1, entries 11–14). The last result is partic-
ularly noteworthy since the catalytic system yields almost

R.F. de Souza et al. / Journal of Catalysis 226 (2004) 235–239
237
Table 1
ꢀ
Catalytic performance of complexes 2a–b in ethylene oligomerization: Effect of the reaction conditions and of the nature of the 2(2-pyridyl)phosphole ligand
a
Entry
Cat
Pressure
(bar)
Temperature
Time
(h)
Weight
(g)
TOF
Selectivity (%)
b
◦
−1
c
d
e
( C)
(s
)
C
4
1-Butene
C
6
Linear
C
8
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
2a
2b
2a
1.1
1.1
1.1
1.1
6.0
21.0
25.0
31.0
41.0
21.0
1.1
21.0
31.0
41.0
21.0
0
0
0
0
0
0
0
0
0
1
1
1
1
1
1
1
1
1
1
2
2
2
2
1
7.6
10.6
6.2
11.4
24.7
52.0
50.6
26.0
5.5
12.8
14.4
81.7
17.7
9.7
2.3
3.2
1.9
3.4
7.4
15.6
15.2
7.8
1.6
3.8
2.2
12.3
2.7
1.5
6.0
79
89
71
82
94
90
92
97
91
95
90
89
94
97
96
2
1
1
11
11
24
17
6
8
8
3
8
5
9
10
6
3
39
40
40
32
58
75
60
70
75
71
43
69
87
69
63
10
0
5
1
0
2
0
0
1
0
1
1
0
0
0
ꢀ
ꢀ
2b
2
2b
2b
2b
2b
2b
2a
2a
2a
2a
2a
31
55
66
73
73
66
1
36
76
80
65
0
10
10
10
10
0
ꢀ
2b
20.0
4
a
Catalytic precursor as defined in Scheme 1.
b
c
d
e
C
C
fraction, defined as the amount of C olefins (1-butene + 2-butene) divided by the total amount of olefins ×100.
4
4
fraction, defined as the amount of C olefins (hexenes + methylpentenes) divided by the total amount of olefins ×100.
6
6
Linearity of the C fraction, defined as the amount of hexenes divided by the total amount of C olefins (hexenes + methylpentenes) ×100.
6
6
C
8
and plus fraction, defined as the amount of C plus C , C and C olefins divided by the total amount of olefins ×100.
8
10 12
14
2b gives low amounts of C6 (3 to 11%) but complexes 2aꢀ
and 2bꢀ, at low ethylene pressure, give 24 and 17%, respec-
tively. This results suggests that the higher steric hindrance
of the complex containing the cyclohexyl fragment com-
pared with the phenyl one, combined with the high steric
hindrance of the pyridyl ligand, is responsible for a decrease
in the amount of β-elimination process. This explanation is
nowadays well accepted as the reason why sterically hin-
dered nickel–diimine complexes are able to polymerize eth-
ylene [14].
The comparison of the catalytic behavior of the nickel–
2-pyridylphosphole catalysts herein described with other
nickel catalysts is somehow difficult due to the large amount
of systems that have been described in the open and patent
literature [15–17]. Despite this difficulty, some general fea-
tures can be drawn:
Fig. 1. Dependence of the selectivity in 1-butene with the ethylene pressure
for 2a (!) and 2b (ꢀ).
only butenes (97%) and the selectivity in 1-butene is re-
markably high (80%). The positive influence of the ethylene
pressure on the 1-butene selectivity is also recorded with 2bꢀ
(see entries 4 and 15, Table 1), indicating that this behavior
is characteristic of these novel nickel–pyridylphosphole cat-
alysts. These results shows that nickel catalysts containing
heteroditopic P,N ligands give unusually high selectivity in
1-butene and constitute a demonstration of the potential of
2-pyridylphospholes as ligands in homogeneous catalysis.
A second important point to note is that the C6 fraction is
mainly constituted by linear hexenes at high pressure (col-
umn 11, Table 1). It can be seen that at 1.1 bar the C6
fraction contains 32 to 40% (entries 1–4, Table 1) of lin-
ear hexenes and that this amount increases with the ethylene
pressure up to 87% (entry 14, Table 1). The total amount of
the C6 fraction has very low variations. Complexes 2a and
(1) A large group of catalytic systems including Ni(acac)2/
AlEt2Cl [18], Ni(π-allyl)2 or Ni(π-allyl)Cl/AlEt2Cl
[19], Ni(acac)2/EtAlCl2 [20,21], or PhNi(PPh3)2Br/
AgClO4 [22] give a majority of dimmers (90–98%)
which are mainly internal olefins (1-butene < 5%).
These examples were the first reports of nickel-catalyzed
oligomerization of ethylene and have been shown to in-
volve nickel hydrides as active species.
(2) A second group comprises nickel complexes bearing
chelating P,O ligands like Ni(Ph)PPh3(Ph2PC6H4COO)
[23–25], which conduct to terminal (α) linear olefins
with a large molecular weight distribution. It seems that
the chelate ligand minimizes the double-bond migration
isomerization, but with the corresponding enlargement
of the product distribution.

238
R.F. de Souza et al. / Journal of Catalysis 226 (2004) 235–239
Acknowledgments
Financial support from CNPq and Fapergs (Brazil),
CNRS (France), and Capes-Cofecub (international exchange
program) is gratefully acknowledged.
References
[1] A.M. Aljarallah, J.A. Anabtawi, M.A.B. Siddiqui, A.M. Aitani, A.W.
Alsadoun, Catal. Today 14 (1992) 1.
[2] G. Lappin, J. Sauer (Eds.), Alpha-Olefins Applications Handbook,
Dekker, New York, 1989.
[3] Y. Chauvin, J.F. Gaillard, D.V. Quang, J.W. Andrews, Chem. Ind.
(1974) 375.
[4] E.R. Freitas, C.R. Gum, Chem. Eng. Prog. 75 (1979) 73;
W. Keim, A. Behr, B. Limbacker, C. Krüger, Angew. Chem. Int. Ed. 22
(1983) 503.
[5] J. Skupinska, Chem. Rev. 91 (1991) 613;
W. Keim, Angew. Chem. Int. Ed. 29 (1990) 235;
Y. Chauvin, H. Olivier, in: B. Cornils, W.A. Herrmann (Eds.), Applied
Homogeneous Catalysis with Organometallic Compounds, vol. 1,
Wiley–VCH, New York, 1996, p. 245;
Scheme 2. Mechanism of ethylene oligomerization catalyzed by Nickel hy-
dride species.
The catalytic behavior of the nickel–2-pyridylphosphole
systems is intermediate between these groups and can be ra-
tionalized as a classical mechanism involving a hydrido or
an alkyl-nickel species 3 and 4, respectively, formed from
the precatalysts 2a–bꢀ and the alkylaluminum cocatalyst un-
der ethylene pressure (Scheme 2).
The key intermediate 4 is formed by insertion of ethyl-
ene on the nickel–carbon bond at 3. This intermediate can
give 5 by β-elimination or 9 by an insertion reaction fol-
lowed by coordination of ethylene. The intermediate 6 can
produce 1-butene and complex 3 or reinsert the olefin giv-
ing the secondary alkyl-nickel species 7. Compound 7 can
product 1-butene or 2-butene through β-elimination, or can
coordinate another ethylene molecule giving 8, which will
afford methylpentenes. By the same reaction path, 9 can give
10 and 11, which led to linear hexenes. The key point for the
production of 1-butene is the selective formation of 6 over
7 from intermediate 5 (Scheme 2), a process clearly favored
by increasing ethylene pressure (Scheme 2).
Scheme 2 shows that the selectivity toward 1-butene is
determined by the steric hindrance of the cationic nickel–
phosphole intermediate which disfavors species containing
more hindered hydrocarbon fragments bonded to the nickel
center, as the case of intermediate 7 that would conduct to
the formation of 2-butenes.
In conclusion, we have described new nickel catalytic
systems using 2-pyridylphosphole ligands for the dimeriza-
tion of ethylene. These catalytic systems are highly effi-
cient and associate high catalytic activities under mild reac-
tion conditions with unusually high selectivities in 1-butene.
Ni–(2-pyridyl)phospholecomplexes are among the very rare
catalytic systems affording efficiency for 1-butene from eth-
ylene. These results hold very promising future prospects
since further variations of the ligand structures are possible.
The specific role played by the heteroditopic P,N ligand and
the use of these versatile catalytic systems in biphasic con-
ditions are under active investigation.
B. Bogdanovic, B. Spliethoff, G. Wilke, Angew. Chem. Int. Ed.
Engl. 19 (1980) 622;
S.M. Pillai, M. Ravindranathan, S. Sivaram, Chem. Rev. 86 (1986)
353.
[6] (a) H. Hu, B. Wu, H. Yang, N. Zhu, X. Leng, H. Wang, New J.
Chem. 26 (2002) 1474;
(b) J.M. Malinoski, M. Brookhart, Organometallics 22 (2003) 5324;
(c) Z.J.A. Komon, X. Bu, G.C. Bazan, J. Am. Chem. Soc. 122 (2000)
12379.
[7] (a) J.R. Sohn, W.C. Park, Appl. Catal. A 239 (2003) 269;
(b) V.I. Smetanyuk, A.V. Ivanyuk, M.A. Martynova, O.P. Komarova,
Petrol. Chem. 43 (2003) 28;
(c) F.X. Cai, C. Lepetit, M. Kemarec, D. Olivier, J. Mol. Catal. 43
(1987) 93.
[8] (a) C. Fave, M. Hissler, K. Sénéchal, I. Ledoux, J. Zyss, R. Réau,
Chem. Commun. (2002) 1674;
(b) F. Leca, M. Sauthier, V. Deborde, L. Toupet, R. Réau, Chem. Eur.
J. 9 (2003) 3785;
(c) A. Caballero, F. Jalon, B. Manzano, M. Sauthier, L. Toupet, R.
Réau, J. Organomet. Chem. 663 (2002) 118;
(d) M. Sauthier, F. Leca, L. Toupet, R. Réau, Organometallics 21
(2002) 1591;
(e) F. Leca, M. Sauthier, B. le Guennic, C. Lescop, L. Toupet, J.-F.
Halet, R. Réau, Chem. Commun. (2003) 1774.
[9] C. Hay, M. Hissler, C. Fischmeister, J. Rault-Berthelot, L. Toupet,
L. Nyulaszi, R. Réau, Chem. Eur. J. 7 (2001) 4222.
[10] F.A. Cotton, Inorg. Syntheses 13 (1972) 157.
[11] I.D.L. Albert, T.J. Marks, M.A. Ratner, J. Am. Chem. Soc. 119 (1997)
6575.
[12] (a) C. Hay, C. Fave, M. Hissler, J. Rault-Berthelot, R. Réau, Org.
Lett. 19 (2003) 3467;
(b) C. Fave, T.-Y. Cho, M. Hissler, C.-W. Chen, T.-Y. Luh, C.-C. Wu,
R. Réau, J. Am. Chem. Soc. 125 (2003) 9254.
[13] M. Sauthier, F. Leca, R.F. de Souza, K. Bernardo-Gusmão, L.F.T.
Queiroz, L. Toupet, R. Réau, New J. Chem. 26 (2002) 630.
[14] L.K. Johnson, C.M. Killian, M. Brookhart, J. Am. Chem. Soc. 117
(1986) 6414;
L.K. Johnson, S. Mecking, M. Brookhart, J. Am. Chem. Soc. 118
(1986) 267.
[15] R.F. de Souza, A.L. Monteiro, M. Seferin, L. Almeida, New J.
Chem. 17 (1993) 437.
[16] S.M. Pillai, M. Ravindranathan, S. Sivaram, Chem. Rev. 86 (1986)
353.
[17] J. Skupinska, Chem. Rev. 91 (1991) 613.

R.F. de Souza et al. / Journal of Catalysis 226 (2004) 235–239
239
[18] J. Ewers, Angew. Chem. 78 (1966) 593.
[22] M. Tsutsui, T. Koyano, J. Polym. Sci., Part A 1 (1967) 681.
[23] W. Keim, in: G. Wilkinson, F.G.A. Stone, E.W. Abel (Eds.), in: Com-
prehensive Organometallic Chemistry, vol. 8, Pergamon, Oxford,
1982, p. 647.
[19] G. Wilke, B. Bogdanovic, Angew. Chem. 78 (1966) 157.
[20] Y. Chauvin, N.H. Phung, N. Geuchard, G. Lefebvre, Bull. Soc. Chim.
Fr. (1966) 3223.
[24] E.R. Freitas, C.R. Gum, Chem. Eng. Prog. 75 (1979) 73.
[25] W. Keim, New J. Chem. 11 (1987) 531.
[21] Y.Sh. Feldblyum, N.V. Obeshchalova, A.I. Lescheva, Dokl. Akad.
Nauk SSSR 172 (1967) 111.