Self-Assembled Organometallic Nickel Complexes as Catalysts for Selective Dimerization of Ethylene into 1-Butene

Article abstract of DOI:10.1021/acs.organomet.5b00055

Sulfonamido-phosphorus and aminophosphine ligands self-assemble to readily form active and stable nickel catalysts that are highly selective for the dimerization of ethylene to 1-butene. The self-assembled allyl-nickel complexes are zwitterionic and are stabilized by hydrogen bond interactions between the two ligands. These organometallic cis-diphosphine complexes rearrange under an ethylene atmosphere to give trans-diphosphine catalysts, with one monoanionic P,O METAMORPhos ligand and an aminophosphine. (Chemical Presented).

Full text of DOI:10.1021/acs.organomet.5b00055

Communication
pubs.acs.org/Organometallics
Self-Assembled Organometallic Nickel Complexes as Catalysts for
Selective Dimerization of Ethylene into 1‑Butene
Pierre Boulens,†,‡ Emmanuel Pellier,† Erwann Jeanneau,§ Joost N. H. Reek,*,‡
Helene Olivier-Bourbigou,*,† and Pierre-Alain R. Breuil*,†
́
̀
†IFP Energies nouvelles, Rond-point de l’ec
́
hangeur de Solaize, BP 3, 69360 Solaize, France
‡van ‘t Hoff Institute for Molecular Sciences, Faculty of Sciences, University of Amsterdam, Science Park 904, 1098 XH Amsterdam,
The Netherlands
§Centre de Diffractomet
́
̂ ̀ ́
rie Henri Longchambon, Site CLEA-Bat. ISA, 3eme etage, 5 rue de La Doua, 69100 Villeurbanne, France
S
* Supporting Information
ABSTRACT: Sulfonamido-phosphorus and aminophosphine ligands
self-assemble to readily form active and stable nickel catalysts that are
highly selective for the dimerization of ethylene to 1-butene. The self-
assembled allyl-nickel complexes are zwitterionic and are stabilized by
hydrogen bond interactions between the two ligands. These
organometallic cis-diphosphine complexes rearrange under an ethyl-
ene atmosphere to give trans-diphosphine catalysts, with one
monoanionic P,O METAMORPhos ligand and an aminophosphine.
ince the discovery by Keim and co-workers that nickel
Sulfonamido-phosphorus ligands (METAMORPhos) were
recently introduced as a family of highly versatile building
blocks for late-transition-metal complexes (Figure 1).6−11 They
display interesting adaptive coordination behavior, as they
coordinate in P and P,O chelating forms and in both neutral
and anionic states of the ligand. Tuning of the substituents
allows the optimization of specific catalytic properties: e.g., a
more acidic character of R1−SO2−NH−R2 is anticipated to
facilitate complex formation and to disfavor the reductive
elimination reaction, leading to the neutral ligand and catalyst
deactivation, resulting in an improved catalyst lifetime.
Moreover, these ligands proved to be particularly suited to
construct supramolecular bidentate or tridentate complexes
through hydrogen bonding. As it is known that the additional
PPh3 ligand coordinated to the SHOP catalyst A displayed in
Figure 1 has a great influence on catalyst stability and product
distribution,4a,b we anticipated that nickel complexes based on
METAMORPhos and aminophosphine ligands would form
supramolecular pincer ligands that due to the reversible nature
of the supramolecular bonds in the complexes would favor the
catalyst stability but at the same time retain the vacant site
required for adequate catalytic activity. While self-assembled
ligands by hydrogen bonding, metal−ligand, ionic, and stacking
interactions have successfully been developed for noble
transition metals such as rhodium, palladium, and platinum
and have been applied in various catalytic transformations,12
this approach has hardly been applied for first-row transition
metals. To the best of our knowledge, the in situ generated
Ni(0) complex that was used as a catalyst in the hydrocyanation
S
complexes can be highly active catalysts for the
oligomerization of ethylene (A in Figure 1),1,2 this reaction
Figure 1. Representative anionic P,O ligand that forms the active
nickel based SHOP catalyst (left) and a typical coordination mode of
METAMORPhos, an adaptive sulfonamido-phosphorus ligand (right).
has been one of the showcase examples of homogeneous
catalysis, leading to key industrial processes such as the Shell
Higher Olefin Process (SHOP).3 Due to the commercial
success of these processes, the nickel-catalyzed ethylene
oligomerization reaction was studied in detail at the
fundamental level.4 When traditional nickel catalysts giving a
broad Schulz−Flory product distribution are employed, 1-
butene is only a minor product and its production directly
depends on the market for the higher linear α-olefins.
Therefore, proposing on-purpose ethylene dimerization pro-
cesses is of prime importance to serve the growing demand for
1-butene (global demand in 2011, 1.6 million metric tons;
estimated demand in 2025, 3.5 million metric tons).5 Several
nickel complexes have been reported to produce 1-butene, but
these catalysts either also produce 2-butene (via isomerization)
or only produce small amounts of 1-butene because of activity
and/or lifetime issues. As such, re-exploring nickel-based
catalysts with a different approach is scientifically and
industrially challenging.
Received: January 20, 2015
Published: March 26, 2015
© 2015 American Chemical Society
1139
DOI: 10.1021/acs.organomet.5b00055
Organometallics 2015, 34, 1139−1142

Organometallics
Communication
reaction reported by Breit et al. is the only example.13
Considering the frequent use of nickel for industrial catalytic
transformations, there is, however, still a great deal of potential
for the use of self-assembled ligands. We report here such a
supramolecular ligand approach for the formation of stable
nickel complexes based on hydrogen bonds. In the presence of
a nickel precursor as a template, METAMORPhos and
aminophosphine ligands form a complex in which the ligands
are organized via a hydrogen bond between the two ligands.
The organometallic nickel complexes are remarkably stable and
very active and can be tuned to favor selective ethylene
dimerization to 1-butene.
Mixing equimolar amounts of nickel(0) bis(1,5-cyclo-
octadiene) (Ni(COD)2), Ph2P-NH-iPr, and 1·NEt3 (or 2) in
chlorobenzene solution led to the selective formation of
nickel(II) complex 3 (or 4), in which METAMORPhos
coordinates as an anionic ligand. During the formation of the
complex, the COD ligand was converted to the π-allyl species
(Figure 2). Such complexes can be formed after oxidative
Figure 3. ORTEP plot (50% probability displacement ellipsoids) of
complex 3. Hydrogen atoms have been omitted for clarity (except for
the NH moiety). Selected bond lengths (Å) and angles (deg): Ni1−
P1, 2.222(2); Ni1−P2, 2.206(2); N2−H2, 0.859; N1- - -H2, 2.994;
O1- - -H2, 2.844; P2−Ni1−P1, 104.55(9).
Figure 2. Synthesis of supramolecular nickel complexes 3 and 4.
addition of the acidic sulfonamide ligand and subsequent
insertion of the hydride in the double bond of the COD
fragment. The supramolecular nickel complexes were isolated
as yellow powders (yields up to 57%) and characterized by 1H,
13C, and 31P NMR and the molecular structures were confirmed
by X-ray analyses (see the Supporting Information).
Figure 4. ORTEP plot (50% probability displacement ellipsoids) of
complex 4. Hydrogen atoms have been omitted for clarity (except for
the NH moiety). Selected bond lengths (Å) and angles (deg): Ni1−
P1, 2.2165(17); Ni1−P2, 2.2007(13); N2−H2, 0.852; N1- - -H2,
2.190; O1- - -H2, 2.588; P2−Ni1−P1, 103.96(6).
Crystals of complexes 3 and 4 suitable for X-ray analysis were
obtained by slow diffusion of pentane in a toluene solution of
the complex. The complexes 3 and 4, displayed in Figures 3
and 4, adopt a square-planar coordination geometry, with the
phosphorus ligands in cis positions with respect to one another.
The nickel atom is formally cationic, whereas the negative
charge is delocalized on the NSO fragment of the
METAMORPhos ligand, as is also clear from the P−N and
N−S bond lengths (intermediate between single and double
bonds) and the S−N−P angles (typically between sp2- and sp1-
hybridized nitrogen: 131.0° for 3 and 134.17° for 4). The
anionic NSO site forms a good hydrogen bond acceptor, and
indeed there is a hydrogen bond formed with the NHP of the
adjacent ligand. Interestingly, for complex 3 the nitrogen−
hydrogen bond distance is significantly longer (2.994 Å) in
comparison to that found in complex 4 (N2−H- - -N1 bond of
2.190 Å, consistent with the literature),9 suggesting that there is
a weaker interaction between the ligands in complex 3. The
difference in steric bulk between the two METAMORPhos
ligands likely accounts for this. Inspection of the structures
reveals that in complex 4 the two isopropyl substituents are
transversal to the coordination plane, whereas in complex 3 the
o-tolyl substituents are opposite to the allyl moiety. The latter
geometry results in a rotation of the sulfonamido fragment,
increasing the distance to the hydrogen bond donor. This
geometry difference in the two complexes is further supported
by the switch of the dihedral angle P−Ni−P−NMETMAMORPhos
from 28.5° for 4 to 79.3° for 3.
These diamagnetic complexes also form in solution, as is
1
clear from the 31P NMR and H NMR spectra. The two
doublets observed in 31P NMR are shifted downfield with
respect to the corresponding ligands, and the small coupling
constant is in agreement with a cis geometry (JPP = 30 Hz). In
1
the H NMR spectra the signals for the π-allyl fragment are
clearly observed at δ (ppm, C6D6) 3.71 (2H) and 5.09 (1H) for
complex 3 and 3.36 (1H), 4.19 (1H), and 4.48 (1H) for
complex 4. Moreover, the NH proton of the coligand Ph2P−
NH−iPr, initially appearing at 1.56 ppm (C6D6), was shifted to
2.85 ppm for complex 3 and to 5.85 ppm for complex 4, in
agreement with a H-bonding interaction between the two
ligands.
To clarify the role of hydrogen bonding in the complex, we
performed control experiments in which we replaced Ph2P−
NH-iPr by similar ligands that do not have a H-donor group.
When the ligand Ph2P−N(iPr)2 or PPh3 was stirred in a
1140
DOI: 10.1021/acs.organomet.5b00055
Organometallics 2015, 34, 1139−1142

Organometallics
Communication
solution with Ni(COD)2 and METAMORPhos 2, the reaction
mixture instantly turned black, indicating complex decom-
position. These control experiments support the importance of
hydrogen bonding between the two ligands, as it improves
complex stability, allowing isolation.
These complexes were evaluated as catalysts in the ethylene
oligomerization reaction at 40 °C under an ethylene pressure of
30 bar, in the absence of any additional activator. The reactions
ran for 90 min, which allowed the accumulation of significant
amounts of oligomers (>10 g), to confirm the productivity per
gram of nickel and stability of the catalyst. In addition, at large
production a proper mass balance (>90%) can be reached,
leading to very reproducible catalytic results. High selectivity
for the formation of 1-butene (up to 84 wt %/all products) and
good productivity (24 kgoligo/(gNi·h)) were obtained with
steady ethylene uptake over a period of 90 min for complex 3
(Table 1; see also the Supporting Information). On the basis of
of 1-butene (from 99.0% at 40 °C to 97.6% at 80 °C) are
favored.
More information on the active species was obtained from in
situ NMR, revealing the rearrangement of both complexes
under ethylene pressure. Indeed, at room temperature and
under 5 bar of ethylene, the original complex 3 solution turned
from orange to green and a new complex formed with two
phosphines in trans positions, as evidenced by the large
coupling 31P NMR (55.5 ppm (d, J = 271 Hz); 69.2 ppm (d, J
1
= 271 Hz)). In the H NMR no hydride was observed. Similar
reactivity was observed with nickel complex 4, leading to a new
species (53.5 ppm (d, J = 275 Hz); 89.8 ppm (d, J = 275 Hz)).
GC and GC/MS analyses of the NMR solution revealed the
presence of short-chain olefins (butenes and hexenes) and
vinylcyclooctene. According to these experiments, we propose
as a mechanism for catalyst rearrangement the ethylene
insertion in catalyst precursor 3 (and 4) with subsequent β-H
elimination or β-H transfer with ethylene, leading to vinyl-
cyclooctene and the nickel−ethyl complex as the resting state
(see Figure 5). Concomitantly, the rearrangement of the
METAMORPhos ligand under the monoanionic P,O chelating
ligand is proposed.
a
Table 1. Catalytic Evaluation of Complexes 3, 4, and A
c
product distribution
b
d
+
complex nNi (μmol) T (°C) prod
C4
C6
C8
1-C4
3
4
A
10
10
50
10
40
40
50
80
24
12
5
85
35
1
13
28
2
2
99.0
99.7
76.7
97.6
37
97
6
e
3
63
73
21
a
Test conditions unless specified otherwise: 30 bar of C2H4, solvent
b
c
toluene (55 mL), 90 min. Productivity in kgoligo/(gNi h). In wt %,
d
determined by GC. 1-C4 wt % in C4 fraction, determined by GC.
e
Solvent toluene (100 mL).
these results, we formed a new class of nickel complexes which
are, to our knowledge, the most robust and efficient
organometallic nickel catalysts for 1-butene formation.4,14−21
The high selectivity for short terminal olefins (1-C4 > 99.0%),
i.e. little isomerization, was also observed when complex 4 was
applied as the catalyst; however, a lower productivity and a
clear shift in selectivity to a larger linear α-olefin distribution
(Schulz−Flory with KSF = 0.45) was, however, observed. The
difference in selectivity observed between catalysts 3 and 4 is
likely caused by the difference in electron density at the P atom
of the METAMORPhos. In comparison, the representative
benchmark complex (A; Figure 1) required higher concen-
tration to produce a significant amount of oligomers and
slightly higher temperature for activation (50 °C). Under these
reaction conditions, this complex led to a very large Schulz−
Flory distribution with a low productivity (KSF > 0.90).
Moreover, a break in the ethylene consumption curve for the
benchmark complex after 40 min, representative of catalyst
deactivation, reinforces the importance of hydrogen bonding to
the catalyst lifetime in complexes 3 and 4, for which stable
activities were noticed over 90 min. At 80 °C, an important
exotherm was observed with catalyst 3 at the beginning of the
reaction (up to 123 °C; see the Supporting Information)
despite an increased volume of toluene for a better heat
dispersion. The temperature is rapidly stabilized at 80 °C, and
then a remarkable steady ethylene consumption was observed
Figure 5. Suggested mechanism for catalyst rearrangement.
In conclusion, we have reported the synthesis and detailed
characterization of stable nickel complexes supported by
supramolecular bidentate ligands based on sulfonamido-
phosphorus and aminophosphine ligands. The hydrogen
bond between the ligands in the zwitterionic nickel complexes
was unambiguously proven in two X-ray structures. It was
established that the hydrogen bond is essential for the stability
of the complex during ethylene dimerization reactions.
Importantly, this novel class of complexes provides highly
active catalysts that display unprecedented selectivity (1-C4
>99.0% in the C4 fraction). The lifetime of these catalysts is
excellent, even at high temperature, and these species form
during ethylene oligomerization mainly 1-butene (up to 84 wt
%) along with small amounts of hexenes and octenes. This high
selectivity for short linear α-olefins is interesting, considering
the market demand for such products, and as such these results
may renew interest in the development of a new generations of
nickel catalysts. In situ NMR experiments under ethylene
pressure suggest the rearrangement of these structures to the
proposed nickel complex as the resting state chelated by a
monoanionic P,O ligand, which may explain the specific
over more than 1 h. The global productivity is high (63 kgoligo
/
(gNi·h)), as is the activity after temperature control (35 kgC H
/
2
4
(gNi·h); see the Supporting Information). Not surprisingly, at
higher temperature, as the ethylene concentration decreases,22
side reactions such as codimerization and slight isomerization
1141
DOI: 10.1021/acs.organomet.5b00055
Organometallics 2015, 34, 1139−1142

Organometallics
Communication
(12) For selected reviews, see: (a) Wilkinson, M. J.; van Leeuwen, P.
W. N. M.; Reek, J. N. H. Org. Biomol. Chem. 2005, 3, 2371−2383.
(b) Meeuwissen, J.; Reek, J. N. H. Nat. Chem. 2010, 2, 615−621.
(c) Raynal, M.; Ballester, P.; Vidal-Ferran, A.; van Leeuwen, P. W. N.
M. Chem. Soc. Rev. 2014, 43, 1660−1733. (d) Raynal, M.; Ballester, P.;
Vidal-Ferran, A.; van Leeuwen, P. W. N. M. Chem. Soc. Rev. 2014, 43,
1734−1787. For selected publications, see: (e) Dubrovina, N. V.;
properties displayed by the catalyst. Complementary studies are
underway to understand more precisely what governs the
oligomerization selectivity. This provides a good starting point
for further development and a detailed understanding of this
new class of nickel catalysts.
ASSOCIATED CONTENT
Borner, A. Angew. Chem., Int. Ed. 2004, 43, 5883−5886. (f) Breit, B.
̈
■
Angew. Chem., Int. Ed. 2005, 44, 6816−6825. (g) Machut, C.;
Patrigeon, J.; Tilloy, S.; Bricout, H.; Hapiot, F.; Monflier, E. Angew.
Chem., Int. Ed. 2007, 46, 3040−3042. (h) Owens, S. B.; Gray, G. M.
Organometallics 2008, 27, 4282−4287. (i) Moteki, S. A.; Takacs, J. M.
Angew. Chem., Int. Ed. 2008, 47, 894−897. (j) Breuil, P.-A. R.;
Patureau, F. W.; Reek, J. N. H. Angew. Chem., Int. Ed. 2009, 48, 2162−
2165. (k) Hasegawa, Y.; Gridnev, I. D.; Ikariya, T. Angew. Chem., Int.
Ed. 2010, 49, 8157−8160. (l) Christiansen, A.; Selent, D.;
S
* Supporting Information
Text, figures, tables, and CIF files giving details of ligand and
complex syntheses, crystal structure determinations for
complexes 3 and 4 and ligand 1, NMR spectra, and catalytic
studies. This material is available free of charge via the Internet
at http://pubs.acs.org. CIF files have also been deposited with
the CCDC and can be obtained on request free of charge, by
quoting the publication citation and deposition numbers
1029754−1029756.
Spannenberg, A.; Kockerling, M.; Reinke, H.; Baumann, W.; Jiao,
̈
H.; Franke, R.; Borner, A. Chem. - Eur. J. 2011, 17, 2120−2129.
̈
(m) Pignataro, L.; Boghi, M.; Civera, M.; Carboni, S.; Piarulli, U.;
Gennari, C. Chem. - Eur. J. 2012, 18, 1383−1400. (n) Moteki, S. A.;
Toyama, K.; Liu, Z.; Ma, J.; Holmes, A. E.; Takacs, J. M. Chem.
Commun. 2012, 48, 263−265. (o) Gellrich, U.; Seiche, W.; Keller, M.;
Breit, B. Angew. Chem., Int. Ed. 2012, 51, 11033−11038.
(13) De Greef, M.; Breit, B. Angew. Chem., Int. Ed. 2009, 48, 551−
554.
AUTHOR INFORMATION
■
Corresponding Authors
*E-mai for J.N.H.R.: j.n.h.reek@uva.nl.
*E-mail for H.O.-B.: helene.olivier-bourbigou@ifpen.fr.
*E-mail for P.-A.R.B.: pierre-alain.breuil@ifpen.fr.
(14) Malinoski, J.; Brookhart, M. Organometallics 2003, 22, 5324−
5335.
(15) Ceder, R.; Muller, G.; Sales, J.; Vidal, J.; Neibecker, D.;
Tkatchenko, I. J. Mol. Catal. 1991, 68, 23−31.
(16) Boardman, B.; Bazan, G. C. Acc. Chem. Res. 2009, 42, 1597−
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
1606.
We thank IFP Energies nouvelles for financial support, David
́
(17) Albers, I.; Alvarez, E.; Cam
́
pora, J.; Maya, C. M.; Palma, P.;
chez, L. J.; Passaglia, E. J. Organomet. Chem. 2004, 689, 833−839.
Proriol and Seb
pressure experiments, and finally Fred
́ ́
Gloaguen, and Rosalba Bellini for helpful discussions.
́
astien Drochon for their help with NMR and
eric Terrade, Yann
San
́
(18) Younkin, T. R. Science 2000, 287, 460−462.
(19) Gutsulyak, D. V.; Gott, A. L.; Piers, W. E.; Parvez, M.
Organometallics 2013, 32, 3363−3370.
(20) Klein, H.-F.; He, M.; Hetche, O.; Rau, A.; Walther, D.;
Wieczorek, T.; Luft, G. Inorg. Chim. Acta 2005, 358, 4394−4402.
(21) Chandran, D.; Bae, C.; Ahn, I.; Ha, C.-S.; Kim, I. J. Organomet.
Chem. 2009, 694, 1254−1258.
(22) Fogg, P. G. T.; Gerrard, W. Solubility of Gases in Liquids; Wiley:
West Sussex, England, 1991.
REFERENCES
■
(1) Keim, W.; Kowaldt, F. H.; Goddard, R.; Kruger, C. Angew. Chem.,
Int. Ed. 1978, 17, 466−467.
(2) Keim, W. Angew. Chem., Int. Ed. 2013, 52, 12492−12496.
(3) (a) Breuil, P.-A. R.; Magna, L.; Olivier-Bourbigou, H. Catal. Lett.
2015, 145, 173−192. (b) Cornils, B.; Herrmann, W. A. In Applied
Homogeneous Catalysis with Organometallic Compounds; Wiley-VCH:
̈
Weinheim, Germany, 2008; pp 213−385. (c) Forestier
Bourbigou, H.; Saussine, L. Oil Gas Sci. Technol. 2009, 64, 649−667.
(4) (a) Kuhn, P.; Semeril, D.; Jeunesse, C.; Matt, D.; Neuburger, M.;
Mota, A. Chem. Eur. J. 2006, 12, 5210−5219. (b) Kuhn, P.; Semeril,
̀
e, A.; Olivier-
́
́
D.; Matt, D.; Chetcuti, M. J.; Lutz, P. Dalton Trans. 2007, 515−528.
(c) Kermagoret, A.; Braunstein, P. Dalton Trans. 2008, 33, 822−831.
(d) Mecking, S. Coord. Chem. Rev. 2000, 203, 325−351. (e) Speiser,
F.; Braunstein, P.; Saussine, L. Acc. Chem. Res. 2005, 38, 784−793.
(f) Wang, S.; Sun, W.-H.; Redshaw, C. J. Organomet. Chem. 2014, 751,
717−741.
(5) Nexant PERP 2011S11. Developments in LAO Comonomer
Technologies for Polyethylene; May 2012.
(6) Patureau, F. W.; Kuil, M.; Sandee, A. J.; Reek, J. N. H. Angew.
Chem., Int. Ed. 2008, 47, 3180−3183.
(7) Patureau, F. W.; de Boer, S.; Kuil, M.; Meeuwissen, J.; Breuil, P.-
A. R.; Siegler, M. A.; Spek, A. L.; Sandee, A. J.; de Bruin, B.; Reek, J. N.
H. J. Am. Chem. Soc. 2009, 131, 6683−6685.
(8) Leon, T.; Parera, M.; Roglans, A.; Riera, A.; Verdaguer, X. Angew.
Chem., Int. Ed. 2012, 51, 6951−6955.
́
(9) Patureau, F. W.; Siegler, M. A.; Spek, A. L.; Sandee, A. J.; Juge, S.;
Aziz, S.; Berkessel, A.; Reek, J. N. H. Eur. J. Inorg. Chem. 2012, 2012,
496−503.
(10) Terrade, F. G.; Lutz, M.; Reek, J. N. H. Chem. - Eur. J. 2013, 19,
10458−10462.
(11) Terrade, F. G.; Lutz, M.; van der Vlugt, J. I.; Reek, J. N. H. Eur.
J. Inorg. Chem. 2014, 2014, 1826−1835.
1142
DOI: 10.1021/acs.organomet.5b00055
Organometallics 2015, 34, 1139−1142