New tripodal iminophosphorane-based ethylene oligomerization catalysts. Part II. Catalytic behavior

Article abstract of DOI:10.1016/j.molcata.2007.12.012

Seven transition-metal complexes of general formula RC (C H₂ NP R^′₃)₃) M X₂ based on new tripodal iminophosphorane ligands were investigated as initiators for the oligomerization of ethylene in the presence of aluminum co-catalysts using high-throughput techniques. In all cases, ethylene consumption peaked at ca. 30 °C and was not drastically affected by varying the nature of the metal (M = Ni, Fe, Pd, Cu), the aluminum co-catalyst (MMAO, Et₂AlCl, or EtAlCl₂) or the substituents of the tris(iminophosphorane) ligand (R = Me, Ph; R′ = cyclopentyl, Ph). Structural modifications of the organometallic complexes, either at the metal center or within the tripodal ligand, however, had a significant impact on the oligomer distribution obtained. In particular, Pd-based catalyst (PhC(CH₂NPPh₃)₃)PdCl₂ displayed an excellent selectivity toward hexene formation.

Full text of DOI:10.1016/j.molcata.2007.12.012

Available online at www.sciencedirect.com
Journal of Molecular Catalysis A: Chemical 283 (2008) 77–82
New tripodal iminophosphorane-based
ethylene oligomerization catalysts
Part II. Catalytic behavior
Laurence Beauforta, Federica Benvenutib, Lionel Delaudea,∗, Alfred F. Noelsa
a Laboratory of Macromolecular Chemistry and Organic Catalysis (CERM), Institut de Chimie (B6a),
Universite´ de Lie`ge, Sart-Tilman par 4000 Lie`ge, Belgium
b Solvay S.A., rue de Ransbeek, 310, 1120 Bruxelles, Belgium
Received 24 October 2007; received in revised form 10 December 2007; accepted 10 December 2007
Available online 23 December 2007
Abstract
Seven transition-metal complexes of general formula RC(CH2NPR3ꢀ )3)MX2 based on new tripodal iminophosphorane ligands were investigated
as initiators for the oligomerization of ethylene in the presence of aluminum co-catalysts using high-throughput techniques. In all cases, ethylene
consumption peaked at ca. 30 ◦C and was not drastically affected by varying the nature of the metal (M = Ni, Fe, Pd, Cu), the aluminum co-catalyst
(MMAO, Et2AlCl, or EtAlCl2) or the substituents of the tris(iminophosphorane) ligand (R = Me, Ph; Rꢀ = cyclopentyl, Ph). Structural modifications
of the organometallic complexes, either at the metal center or within the tripodal ligand, however, had a significant impact on the oligomer
distribution obtained. In particular, Pd-based catalyst (PhC(CH2NPPh3)3)PdCl2 displayed an excellent selectivity toward hexene formation.
© 2007 Elsevier B.V. All rights reserved.
Keywords: Cu complexes; Fe complexes; Ni complexes; Pd complexes; Tripodal ligand
1. Introduction
In 1975, Appel and Voltz prepared and characterized the
first complexes based on 1,2-bis(triphenyl-phosphoranylidene-
amino)ethane-(Ph3P N (CH2)2 N PPh3) and metal halides
[6]. This report was later followed by a few publications on
the synthesis and catalytic applications of other type A metal
complexes for ethylene oligomerization [7,8], allylic alkylation
[9,10], or cyclopropanation reactions [11]. Recourse to this
kind of transition metal catalysts in polymer chemistry or for
organic synthesis remains, however, uncommon. In contrast,
ligands possessing two N P functionalities linked together
via the phosphorus atoms (structure B in Scheme 1) have
been thoroughly investigated [12–15] with an emphasis on
bis(iminophosphoranyl)methane derivatives, which proved
suitable ligands for ethylene polymerization [16,17].
Another class of well-studied bidentate ligands are
difunctional compounds that contain one iminophosphorane
functionality and a second donor site, which is in most cases
a phosphorus or arsenic atom (structures C and D in Scheme 1).
In this case, the weakest bond of the chelate may dissociate
reversibly without total detachment of the ligand, thereby open-
ing vacant coordination sites on the metal. Complexes based
on such ligands have been extensively studied by Cavell and co-
In recent years, bidentate donor species containing sp2-
hybridized nitrogen atoms have attracted much attention as
ligands for transition metal-catalyzed transformations [1]. An
important development in this field was reported by Brookhart
and co-workers who showed that ␣-diimines were efficient lig-
ands for the nickel- or palladium-catalyzed polymerization of
ethylene [2,3]. Iminophosphoranes are compounds of general
formula RN PRꢀ3. They possess a highly polarized N P dou-
ble bond and coordinate to transition metals via the lone pair of
their nitrogen atom to give stable complexes [4,5]. They offer
a steric environment relatively similar to that of diimines, but
their electronic characteristics (donor strength and ␲-acceptor
capability) are clearly different. So far, homo- and heterobiden-
tate derivatives of types A–D have been the most studied ligands
incorporating the iminophosphorane moiety (Scheme 1). Teth-
ering of this group may occur via the nitrogen (structures A and
C) or the phosphorus atom (structures B and D).
∗
Corresponding author. Tel.: +32 4 3663496; fax: +32 4 3663497.
E-mail address: L.Delaude@ulg.ac.be (L. Delaude).
1381-1169/$ – see front matter © 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.molcata.2007.12.012

78
L. Beaufort et al. / Journal of Molecular Catalysis A: Chemical 283 (2008) 77–82
tilled under argon prior to use. Triiminophosphorane ligands and
metal complexes derived thereof were prepared according to lit-
erature [27]. Complexes 5 and 7 were isolated in 81% and 96%
yield, respectively, both as ocher solids. Their spectroscopic
features were similar to those of (PhC(CH2NPPh3)3)CuBr
[27] and they afforded satisfactory elemental analyses. The
high-throughput screening experiments were carried out using
an Endeavor parallel pressure reactor system (Argonaut
Technologies, now part of Biotage AB), which had eight
mechanically stirred stainless steel reactors equipped with glass
liners.
Scheme 1. Homo- and heterobidentate chelates incorporating the iminophos-
phorane moiety.
2.2. Typical oligomerization procedure
2.2.1. Endeavor conditioning
Eight Reaction Vessels (RV) were weighted and inserted in
an 8-position stainless steel RV rack. The rack, 8 impellers and
2 needles were heated in an oven at 70 ◦C for at least 1 h (usually
overnight). Each hot RV was then inserted in the Endeavor fol-
lowing the procedure described by Argonaut. Air was removed
by applying a nitrogen flow through a Teflon tube down to the
bottom of the RV and the RV was covered with a nitrogen flow
coming from the Endeavor manifold by opening the appropri-
ate valve. All the impellers were placed on the stirrer assembly
and the Endeavor was closed after having stopped the nitrogen
flow to avoid a bad positioning of the O-rings. A seal check
was achieved by pressurizing all the reactors (21 bar), closing
each reactor valves, and venting the manifold. The pressure
drop should not exceed 0.14 bar in 15 min. After venting all
the reactors, the Endeavor conditioning procedure consisting in
10 nitrogen purges at 80 ◦C was launched for ca. 30 min.
Scheme 2. Transition-metal complexes bearing triiminophosphorane ligands
used in this work.
workers. Rhodium [18–23] and palladium [24,25] served mostly
as metal centers, but a range of other early or late transition
metals, including rhenium, molybdenum, platinum, and iridium,
were also investigated [19,22,26].
The small number of complexes based on type A ligands, in
which the tethering motive is formally derived from a diamine,
prompted us to undertake the study of this poorly explored
area, and more specifically, of the hitherto uncharted field of
metal complexes bearing tripodal iminophosphorane ligands
(Scheme2). Inapreviousarticle, wedisclosedthefirstpartofour
work in this direction, i.e., the multi-step synthesis of four new
triiminophosphorane ligands and their use for the preparation of
some representative transition-metal complexes [27].
In this paper, we report on the oligomerization of ethy-
lene promoted by selected metal complexes of Fe(II), Ni(II),
Pd(II), and Cu(II) based on the four tridentate iminophospho-
rane ligands depicted in Scheme 2. The ligands differ from each
other by the nature of their bridgehead R substituents (methyl
or phenyl groups) and their phosphoranylidene Rꢀ substituents
(cyclopentyl or phenyl rings). The reactions were carried out
using a high-throughput experimental device called “Endeavor”
supplied by Argonaut Technologies (now part of Biotage AB).
It is a parallel autoclave bench with eight independent reactors
that allows eight parallel experimentations under pressure with
different conditions for each experiment.
2.2.2. Catalyst sample preparation
A metal complex (ca. 5 mg) was weighted under air in a
1.5 mL vial stored in a dessicator. In a Schlenk flask conditioned
under nitrogen, the appropriate amount of toluene (ca. 23 mL,
the exact volume was calculated to reach 25 mL after addition
of the co-catalyst) was introduced. The vial containing the cat-
alyst was dropped in the toluene flask to dissolve its content.
The appropriate amount of an aluminum co-catalyst solution
was added and the resulting mixture was kept for ca. 10 min
(precontact time). During that period, the 2 needles were taken
out of the oven, connected to a 1 mL and a 5 mL glass syringes,
respectively, and flushed with nitrogen.
2.2.3. Oligomerization reaction
Each reaction vessel was loaded with 5 mL of n-heptane
containing 30 ppm of an aluminum scavenger (prepared by
mixing 100 mL of n-heptane with 0.2 mL of 10% triisobutylalu-
minum solution). Next, 0.8 mL of n-nonane (internal standard)
and 0.2 mL of catalyst solution were added using the same
syringe. The system was pressurized with 10 bar of ethylene.
The Endeavor took control of all the operations and monitoring
until the final venting of the reactor vessels. The reactions mix-
tures were quenched with methanol (0.5 mL) and analyzed by
gas chromatography.
2. Experimental
2.1. General information
All manipulations were carried out under an atmosphere of
dry nitrogen using standard Schlenk and cannula techniques.
Solvents were refluxed over appropriate drying agents and dis-

L. Beaufort et al. / Journal of Molecular Catalysis A: Chemical 283 (2008) 77–82
79
3. Results and discussion
3.1. Oligomerization of ethylene with Ni(II) complexes
Since nickel is the metal most frequently employed for the
oligomerization of ethylene at the industrial scale [28,29], we
first investigated the catalytic activity of complexes 1–4 that
had been prepared by chelating NiBr2(dme) with our four
new tripodal ligands (dme is 1,2-dimethoxyethane). Modified
methylaluminoxane (MMAO) was added as a co-catalyst with
an aluminum-to-nickel molar ratio of 800 and the reactions
were carried out in n-heptane under 10 bar of ethylene for 1 h.
Under these conditions, the oligomerization activities of the
four triiminophosphorane nickel complexes remained modest
and roughly unaffected by the exact nature of the R and Rꢀ
substituents. In all four cases, consumption of ethylene did not
exceed 150 g C2H4 mmol−1 h−1 at 30 ◦C (Fig. 1). Although this
figure is not exceptional, it is nevertheless superior to the one
recorded with the corresponding bis(iminophosphorane) nickel
complexes, which afforded – in the best cases – an oligomer-
ization activity of ca. 20 g C2H4 mmol−1 h−1 when activated
by MMAO at the same temperature (Beaufort et al., unpub-
lished results). Increasing the temperature up to 45 ◦C resulted
in a slight decrease of the oligomerization activity. A much
sharper drop of catalyst efficiency occurred between 45 and
50 ◦C. Activity fell then to around 30 g C2H4 mmol−1 h−1 and
remained at this level until 80 ◦C (Fig. 1). Hence, at 50–80 ◦C
the oligomerization activity of triiminophosphorane complexes
1–4 became similar to that of the analogous bidentate species
at 30 ◦C, suggesting a higher thermal stability of the tridentate
complexes.
Fig. 2. Influence of the bridgehead and phosphine substituents on the catalytic
activity of Ni(II) complexes 1–4 at various temperatures.
tricyclopentylphosphine-based derivatives, the reverse tendency
was observed since the complex with a methyl-capped ligand
(3) was found slightly more active than its phenyl-substituted
analogue (4), except at 45 ◦C.
We also put side-by-side species sharing the same bridge-
head on their iminophosphorane ligands. The activity of
catalyst 1 bearing a methyl bridgehead and a triphenylphos-
phine moiety was around 130 g C2H4 mmol−1 h−1, down
from 150 g C2H4 mmol−1 h−1 with complex 3 sporting a tri-
cyclopentylphosphine unit (Fig. 2). However, the opposite
trend was observed with a phenyl group at the bridgehead
position. In this case, the triphenylphosphine-based com-
plex (2) always afforded a higher oligomerization activity
than its tricylopentylphosphine counterpart (4) (∼140 vs.
120 g C2H4 mmol−1 h−1).
Altogether, changing the nature of the bridgehead and phos-
phine substituents had only a limited influence on the ethylene
consumption promoted by complexes 1–4. Yet, and rather
intriguingly, the two structural features seemed to be linked and
interdependent, since associating the next of kin, either aromatic
(phenyl/phenyl) or aliphatic (methyl/cyclopentyl), usually led to
the highest catalytic efficiencies.
In order to better apprehend the influence of the bridgehead
and phosphine substituents on the outcome of the reaction, the
catalytic efficiencies of complexes 1–4 at temperatures below
50 ◦C were examined in detail (Fig. 2). Comparison of the
triphenylphosphine-iminophosphorane Ni(II) complexes sub-
stituted either by a methyl (1) or a phenyl group (2) at the
bridgehead carbon atom revealed that the latter catalyst pre-
sented a slightly higher activity, except at 35 ◦C. For the two
GC analysis of the reaction mixtures indicated that the
oligomers obtained with complexes 1–4 were mainly butenes
and hexenes (Table 1). Whatever the reaction temperature and
catalyst used, 1-butene and isobutene formed the bulk of the C4
fraction, while 1-hexene was the most important component of
the C6 fraction. Contrary to the total consumption of ethylene,
the relative amounts of butenes and hexenes shifted markedly
with the nature of the iminophosphorane ligand employed.
Indeed, catalyst precursor (PhC(CH2NPCp3)3)NiBr2 (4) was
most selective toward hexene formation, whereas complex 2
favored dimerization into butenes and complex 3 led to a larger
proportion of higher oligomers. Increasing the temperature, on
the other hand, had a much more limited influence on the
oligomer distribution, except with (PhC(CH2NPPh3)3)NiBr2
(2).
Fig. 1. OligomerizationofethylenecatalyzedbyNi(II)complexes1–4atvarious
temperatures.

80
L. Beaufort et al. / Journal of Molecular Catalysis A: Chemical 283 (2008) 77–82
Table 1
Oligomer distributions obtained with Ni(II) complexes 1–4 at various temperatures
Complex
T (◦C)
Oligomer distribution (mol%)a
C4
C6
>C6
15
15
18
10
25
33
3
(MeC(CH2NPPh3)3)NiBr2 (1)
(MeC(CH2NPPh3)3)NiBr2 (1)
(PhC(CH2NPPh3)3)NiBr2 (2)
(PhC(CH2NPPh3)3)NiBr2 (2)
(MeC(CH2NPCp3)3)NiBr2 (3)
(MeC(CH2NPCp3)3)NiBr2 (3)
(PhC(CH2NPCp3)3)NiBr2 (4)
(PhC(CH2NPCp3)3)NiBr2 (4)
30
60
30
60
30
80
30
70
21
19
49
62
35
29
12
15
64
65
33
28
40
37
85
82
3
Reaction conditions: 0.5 mmol of nickel complex in the presence of MMAO (Al/Ni = 800), 10 bar of C2H4 (constant pressure), 5 mL of n-heptane for 1 h.
a
Determined by GC analysis.
3.2. Influence of the co-catalyst
3.3. Influence of the metal
Using nickel complexes 1 and 2 as catalysts precursors, we
have carried out the oligomerization of ethylene at various tem-
peratures in the presence of diethylaluminum chloride (DEAC,
Et2AlCl) instead of MMAO (Fig. 3). On the whole, this change
of alkylating agent did not result in any significant enhancement
of the catalyst efficiency. Only at 30 ◦C in the case of complex 1,
did DEAC somewhat outperform MMAO, as the catalytic activ-
ity peaked at ca. 160 g C2H4 mmol−1 h−1. This effect did not
persist at higher temperatures, however. With both co-catalysts,
the ethylene oligomerization activity was maximal at 30 ◦C and
dramatically decreased above 45 ◦C. Similar oligomer distribu-
tions were observed with each of the two catalyst precursors
(MeC(CH2NPPh3)3)NiBr2 (1) and (PhC(CH2NPPh3)3)NiBr2
(2) when tested with DEAC as activator. Both systems afforded
ca. 35–50% of butenes at 30 ◦C with no remarkable oligomer
selectivity.
Triiminophosphorane complexes of Fe(II), Pd(II), and
Cu(II) were prepared by complexing the tripodal ligand
PhC(CH2NPPh3)3 with FeCl2, PdCl2, and Cu(OTf)2, respec-
tively. These species were tested as catalyst precursors for
ethylene oligomerization in the presence of MMAO at vari-
ous temperatures (Fig. 4). The screening was carried out with
the Endeavor high-throughput system using the experimental
procedure set up for nickel complexes. Results showed that
the metal had only a limited influence on the outcome of the
reaction. This was a rather unexpected observation. For all cat-
alysts under scrutiny (2, 5–7), consumption of ethylene evolved
in a similar fashion within the 30–80 ◦C temperature range.
The rate of oligomerization was highest toward the lowest end
(ca. 140 g C2H4 mmol−1 h−1 at 30 ◦C) and slowly decreased
until 50 ◦C. Above that threshold, it sharply plunged to about
20 g C2H4 mmol−1 h−1. Thus, the temperature–activity profiles
recorded by varying the metal center closely paralleled those
obtained when the tripodal ligand was changed (cf. Fig. 1).
Noteworthyly, the Pd-based complex 6 was slightly more active
than its counterparts with a maximum consumption of ca.
180 g C2H4 mmol−1 h−1 at 35 ◦C.
Catalyst activation by ethylaluminum dichloride (EADC,
EtAlCl2), which is more acidic than DEAC, was also investi-
gated in the case of complex 3. Recourse to this alkylating agent
yieldedanoligomerizationactivitysimilartothoseobservedpre-
viously with MMAO or DEAC, although it decreased somewhat
more slowly as temperature increased (Table 2).
Fig. 3. Influence of the co-catalyst on the oligomerization of ethylene catalyzed
by (MeC(CH2NPPh3)3)NiBr2 (1) or (PhC(CH2NPPh3)3)NiBr2 (2) at various
temperatures.
Fig. 4. Oligomerization of ethylene catalyzed by triiminophosphorane com-
plexes of Ni(II) (2), Fe(II) (5), Pd(II) (6), and Cu(II) (7) at various temperatures.

L. Beaufort et al. / Journal of Molecular Catalysis A: Chemical 283 (2008) 77–82
81
Table 2
Influence of the co-catalyst on the oligomerization activity of complex 3 at various temperatures
Co-catalyst
Oligomerization activity (g C2H4 mmol−1 h−1
)
30 ◦C
35 ◦C
40 ◦C
45 ◦C
MMAO
DEAC
EADC
150
123
153
121
110
116
78
79
103
97
110
113
Reaction conditions: 0.5 mmol of (MeC(CH2NPCp3)3)NiBr2 (3) in the presence of an aluminum co-catalyst (Al/Ni = 800), 10 bar of C2H4 (constant pressure), 5 mL
of n-heptane for 1 h.
Table 3
Oligomer distributions obtained with various transition-metal complexes at 30 ◦C
Complex
Oligomer distribution (mol%)a
C4
C6
>C6
(PhC(CH2NPPh3)3)NiBr2 (2)
(PhC(CH2NPPh3)3)FeCl2 (5)
(PhC(CH2NPPh3)3)PdCl2 (6)
(PhC(CH2NPPh3)3)Cu(OTf)2 (7)
49
50
4
33
26
93
50
18
23
3
30
20
Reaction conditions: 0.5 mmol of catalyst in the presence of MMAO (Al/M = 800), 10 bar of C2H4 (constant pressure), 5 mL of n-heptane for 1 h.
a
Determined by GC analysis.
Although the four metal complexes examined in this study
reacted with ethylene at about the same rate, they afforded
mixtures of oligomers (mainly butenes and hexenes) with signif-
icantly different selectivities (Table 3). Nickel and iron catalytic
species (2 and 5, respectively) favored butene formation (up
to 50 mol% of the product distribution) together with a lesser
amount of higher oligomers (mostly hexenes), whereas palla-
dium and copper complexes 6 and 7 produced more hexenes
than butenes. Palladium complex 6, in particular, displayed
an impressive selectivity, affording almost exclusively hexenes
from ethylene.
References
[1] A. Togni, L.M. Venanzi, Angew. Chem. Int. Ed. Engl. 33 (1994)
497–526.
[2] S.D. Ittel, L.K. Johnson, M. Brookhart, Chem. Rev. 100 (2000)
1169–1203.
[3] G.J. Domski, J.M. Rose, G.W. Coates, A.D. Bolig, M. Brookhart, Progr.
Polym. Sci. 32 (2007) 30–92.
[4] K. Dehnicke, M. Krieger, W. Massa, Coord. Chem. Rev. 182 (1999)
19–65.
[5] K. Dehnicke, J. Stra¨hle, Polyhedron 8 (1989) 707–726.
[6] R. Appel, P. Volz, Z. Anorg. Allg. Chem. 413 (1975) 45–50.
[7] M. Sauthier, F. Leca, R.F. de Souza, K. Bernardo-Gusma˜o,
L.F.T. Queiros, L. Toupet, R. Re´au, New J. Chem. 26 (2002)
630–635.
[8] K. Bernardo-Gusma˜o, L.F.T. Queiros, R.F. de Souza, F. Leca, C. Loup, R.
Re´au, J. Catal. 219 (2003) 59–62.
[9] O. Tardif, B. Donnadieu, R. Re´au, C. R. Acad. Sci., Ser. IIc: Chim. (1998)
661–666.
4. Conclusion
The catalytic activity of seven transition-metal complexes
bearing new tripodal iminophosphorane ligands was investi-
gated in the oligomerization of ethylene using high-throughput
techniques. In all cases, ethylene consumption peaked at
ca. 30 ◦C and was not drastically affected by varying the
nature of the metal (Ni, Fe, Pd, Cu), the aluminum co-
catalyst (MMAO, DEAC, or EADC) or the substituents of
the tris(iminophosphorane) ligand. Structural modifications of
the organometallic complexes, either at the metal center or
within the tripodal ligand, however, had a significant impact on
the oligomer distribution obtained. In particular, the Pd-based
catalyst (PhC(CH2NPPh3)3)PdCl2 (6) displayed an excellent
selectivity toward hexene formation.
[10] M. Sauthier, J. Fornie´s-Camer, L. Toupet, R. Re´au, Organometallics 19
(2000) 553–562.
[11] M.T. Reetz, E. Bohres, R. Goddard, J. Chem. Soc., Chem. Commun. (1998)
935–936.
[12] M.W. Avis, C.J. Elsevier, N. Veldman, H. Kooijman, A.L. Spek, Inorg.
Chem. 35 (1996) 1518–1528.
[13] C. Ong, D.W. Stephan, J. Am. Chem. Soc. 121 (1999)
2939–2940.
[14] R.G. Cavell, R.P.K. Babu, A. Kasani, R. McDonald, J. Am. Chem. Soc.
121 (1999) 5805–5806.
[15] M.S. Hill, P.B. Hitchcock, Dalton Trans. (2003) 4570–4571.
[16] S. Al-Benna, M.J. Sarsfield, M. Thornton-Pett, D.L. Ormsby, P.J. Mad-
dox, P. Bre`s, M. Bochmann, J. Chem. Soc., Dalton Trans. (2000)
4247–4257.
[17] K. Kreischer, J. Kipke, M. Bauerfeind, J. Sundermeyer, Z. Anorg. Allg.
Chem. 627 (2001) 1023–1028.
[18] K.V. Katti, R.G. Cavell, Organometallics 7 (1988) 2236–2238.
[19] K.V. Katti, R.G. Cavell, Organometallics 8 (1989) 2147–2153.
[20] K.V. Katti, B.D. Santarsiero, A.A. Pinkerton, R.G. Cavell, Inorg. Chem.
32 (1993) 5919–5925.
Acknowledgments
[21] R.W. Reed, B.D. Santarsiero, R.G. Cavell, Inorg. Chem. 35 (1996)
4292–4300.
This work was generously supported by a BP-Solvay grant
to L.B.

82
L. Beaufort et al. / Journal of Molecular Catalysis A: Chemical 283 (2008) 77–82
[22] M.S. Balakrishna, B.D. Santarsiero, R.G. Cavell, Inorg. Chem. 33 (1994)
3079–3084.
[23] J. Li, R. McDonald, R.G. Cavell, Organometallics 15 (1996) 1033–1041.
[24] K.V. Katti, R.J. Batchelor, F.W.B. Einstein, R.G. Cavell, Inorg. Chem. 29
(1990) 808–814.
[26] K.V. Katti, R.G. Cavell, Inorg. Chem. 28 (1989) 3033–3036.
[27] L. Beaufort, L. Delaude, A.F. Noels, Tetrahedron 63 (2007) 7003–7008.
[28] J. Skupinska, Chem. Rev. 91 (1991) 613–648.
[29] D. Vogt, in: B. Cornils, W.A. Herrmann (Eds.), Applied Homogeneous
Catalysis with Organometallic Compounds, vol. 1, 2nd ed., Wiley-VCH,
Weinheim, 2002, pp. 240–253.
[25] K.V. Katti, R.G. Cavell, Organometallics 10 (1991) 539–541.