Nickel 2-iminopyridine N-oxide (PymNox) complexes: Cationic counterparts of salicylaldiminate-based neutral ethylene polymerization catalysts

Article abstract of DOI:10.1021/om800548y

2-Iminopyridine N-oxides (PymNox) constitute a promising class of ligands that can be considered as the neutral counterparts of the well-known sahcylaldiminate system. A series of nickel PymNox complexes displaying different substitution patterns in the ligand have been synthesized, and their activity as ethylene polymerization catalysts has been studied. While an electron-donor group (OMe) at the remote position 4 of the pyridine ring causes a moderate decrease of the catalytic activity and increase of the polyethylene molecular weight, a strongly electron-withdrawing group (NO₂) in this position shifts the catalyst selectivity from ethylene polymerization to oligomerization. The introduction of a phenyl substituent next to the pyridine nitrogen (position 6) causes a significant increase of the catalytic activity and the polymer molecular weight. Although aldimino PymNox catalysts are inactive in ethylene-methyl acrylate copolymerization, we observed that acetaldimino (displaying the Me-C=NAr group) catalysts display a small but significant activity on this account, giving rise to copolymers incorporating ca. 1% methyl acrylate in their structure. The trends observed in the PymNox catalytic system are strongly reminiscent of those of nickel sahcylaldiminates (although the former are considerably more active), demonstrating for the first time that substitution of the widely used phenoxo anionic group by the neutral pyridine N-oxide fragment is a useful possibility in the design of late transition metal catalysts for olefin polymerization.

Full text of DOI:10.1021/om800548y

Organometallics 2008, 27, 4711–4723
4711
Nickel 2-Iminopyridine N-Oxide (PymNox) Complexes: Cationic
Counterparts of Salicylaldiminate-Based Neutral Ethylene
Polymerization Catalysts
´
M. Brasse, J. Ca´mpora,* P. Palma, and E. Alvarez
Instituto de InVestigaciones Qu´ımicas, Consejo Superior de InVestigaciones Cient´ıficas-UniVersidad de
SeVilla, Spain
V. Cruz and J. Ramos
GIDEM, Instituto de Estructura de la Materia, Consejo Superior de InVestigaciones Cient´ıficas,
Serrano 119, 28006 Madrid, Spain
M. L. Reyes
Centro de Tecnolog´ıa Repsol-YPF, Carretera de Extremadura NV, Km 18, 28930 Mostoles, Madrid, Spain
ReceiVed June 12, 2008
2-Iminopyridine N-oxides (PymNox) constitute a promising class of ligands that can be considered as the
neutral counterparts of the well-known salicylaldiminate system. A series of nickel PymNox complexes
displaying different substitution patterns in the ligand have been synthesized, and their activity as ethylene
polymerization catalysts has been studied. While an electron-donor group (OMe) at the remote position 4 of
the pyridine ring causes a moderate decrease of the catalytic activity and increase of the polyethylene molecular
weight, a strongly electron-withdrawing group (NO2) in this position shifts the catalyst selectivity from ethylene
polymerization to oligomerization. The introduction of a phenyl substituent next to the pyridine nitrogen (position
6) causes a significant increase of the catalytic activity and the polymer molecular weight. Although aldimino
PymNox catalysts are inactive in ethylene-methyl acrylate copolymerization, we observed that acetaldimino
(displaying the Me-CdNAr group) catalysts display a small but significant activity on this account, giving
rise to copolymers incorporating ca. 1% methyl acrylate in their structure. The trends observed in the PymNox
catalytic system are strongly reminiscent of those of nickel salicylaldiminates (although the former are
considerably more active), demonstrating for the first time that substitution of the widely used phenoxo anionic
group by the neutral pyridine N-oxide fragment is a useful possibility in the design of late transition metal
catalysts for olefin polymerization.
catalysts for the production of functionalized polyethylenes.
However, these attempts met serious difficulties, since even late
transition metal catalysts are often readily poisoned by such
co-monomers. Experimental3 and computational4 investigations
have uncovered a number of mechanisms for catalyst deactiva-
tion, e.g., blocking of the metal center by σ-coordination of the
polar co-monomer. These have shown that palladium complexes
are in general superior to those of nickel for polar co-monomer
incorporation, but replacing palladium with nickel might be
essential for economical reasons, if large-scale application
processes are to be achieved someday. This is by no means a
trivial task, which still will require extensive catalyst development.
Introduction
Much of the interest that has arisen in recent years by late
transition metal-based olefin polymerization catalysts stems from
their tolerance to polar molecules, which in some cases enables
controlled copolymerization of neutral and polar comonomers.1
The first example of copolymerization of ethylene with methyl
acrylate (a monomer of technical interest) by palladium R-di-
imine, reported by Brookhart in 1996,2 triggered widespread
attention to new group 10 complexes as potentially useful
* Corresponding author. E-mail: campora@iiq.csic.es.
(1) (a) Ittel, S. D.; Johnson, L. K.; Brookhart, M. Chem. ReV. 2000,
100, 1169. (b) Gibson, V. C.; Spitzmesser, S. K. Chem. ReV. 2003, 103.
(2) Johnson, L. K.; Mecking, S.; Brookhart, M. J. Am. Chem. Soc. 1996,
118, 267.
In the design of new olefin polymerization catalysts, electric
charge is probably the variable with a deeper impact on their
tolerance to polar substances. Mild electrophilicity provided by
electrically neutral complexes favors this kind of tolerance. The
long-known capability of the neutral SHOP catalysts to oligo-
merize and polymerize olefins was soon recognized as a
potentially useful feature in polar co-monomer incorporation.5
Neutral nickel salicylaldiminate catalysts, first reported by
Grubbs,6 laid a new milestone in the field. These display
relatively high catalytic activities in ethylene homopolymeri-
zation, are compatible with polar solvents (even aqueous
media7), and are capable of incorporation of certain types of
(3) (a) Mecking, S.; Johnson, L. K.; Wang, L.; Brookhart, M. J. Am
Chem. Soc. 1998, 120, 888. (b) Michalak, A.; Ziegler, T. Organometallics
2001, 20, 1521. (c) Kang, M.; Sen, A.; Zakharov, L.; Rheingold, A. L.
J. Am. Chem. Soc. 2002, 124, 12080. (d) Foley, S. R., Jr.; Shen, A.; Jordan,
R. F. J. Am. Chem. Soc. 2003, 125, 4350. (e) Waltman, A. W.; Younkin,
T. R.; Grubbs, R. H. Organometallics 2004, 23, 5121. (f) Li, W.; Zhang,
X.; Meetsma, A.; Hessen, B. J. Am. Chem. Soc. 2004, 126, 12246. (g)
Williams, B. S.; Leatherman, M. D.; White, P. S.; Brookhart, M. J. Am.
Chem. Soc. 2005, 127, 5132. (h) Groux, L. F.; Weiss, T.; Reddy, D. N.;
Chase, P. A.; Piers, W. E.; Ziegler, T.; Parvez, M.; Benet-Buchholz, J. J. Am.
Chem. Soc. 2005, 127, 1841. (i) Wu, F.; Foley, S. R.; Burns, C. T.; Jordan,
R. F. J. Am. Chem. Soc. 2005, 127, 1841. (j) Stojcevic, G.; Prokopchuk,
E. M.; Baird, M. C. J. Organomet. Chem. 2005, 690, 4349.
10.1021/om800548y CCC: $40.75
2008 American Chemical Society
Publication on Web 08/27/2008

4712 Organometallics, Vol. 27, No. 18, 2008
Brasse et al.
Chart 1. Representation of the Propagating Species in the Neutral Grubbs Salicylaldiminate-Based Catalyst System (A) and the
Neutral PymNox Catalysts (B)
Scheme 1
The presence of bulky substituents at position 6 of the central
aryl group is a crucial feature of salicylaldiminate nickel
polymerization catalysts, to achieve high catalytic activities.17
An extension of the method for compound 2 was used for the
synthesis of the quinoline derivative 3 and ligand 4, the latter
bearing a phenyl substituent at position 6 in the pyridine ring.
Pyridines displaying bulky substituents at the neighboring ring
positions are often difficult to oxidize,14,18 but somewhat
surprisingly, this step proceeds readily in both cases, especially
for 3.
functional monomers.8 More recently, other electrically neutral
group 10 catalysts have been successfully allowed co-polym-
erization of ethylene and acrylic monomers.9-12 However,
achieving high polymerization activities usually demands more
electrophilic, cationic metal centers. An obvious approach to
solve this dilemma is to adopt a compromise allowing a
fractional positive charge to sit at the metal center. We reasoned
that a modification of the original Grubbs’ salicylaldiminate
catalyst design (Chart 1, A) by replacing the anionic aryloxide
moiety by a neutral pyridine N-oxide fragment would lead to
highly active cationic polymerization catalysts, while it could
maintain to some degree their tolerance to polar substances. This
substitution leads to the isostructural 2-iminopyridine N-oxide
(PymNox) family of ligands (B). It may be anticipated that the
N-oxide fragment might contribute to efficiently delocalize part
of the positive charge on the ligand,13 leading to an intermediate
situation between typically cationic and neutral polymerization
catalysts. This concept is not new: Dyker14 has previously relied
on it for the design of a neutral analogue of Jacobsen’s Salen
ligand, and Erker has recently reported the synthesis of a number
of PymNox ligands and the corresponding complexes of Ni,
Pd, Co, and Cu, hinting at their potential in homogeneous
catalysis.15 However, this potential remains largely undeveloped
and in the case of olefin polymerization is still unexplored. In
this contribution, we begin our study of PymNox complexes
and their behavior as ethylene polymerization catalysts, provid-
ing some preliminary results on the capability of these com-
pounds to copolymerize ethylene with methyl acrylate.
Scheme 2 also shows the preparation of the starting material
required for 4, 2-phenyl-6-methylpyridine. The key step is a
Kumada coupling reaction using 2-bromo-6-methylpyridine,
(4) (a) Philipp, D. M.; Muller, R. P.; Goddard, W. A., III; Storer, J.;
McAdon, M.; Mullins, M. J. Am. Chem. Soc. 2002, 124, 10198. (b)
Michalak, A.; Ziegler, T. Organometallics 2003, 22, 2069. (c) Michalak,
A.; Ziegler, T. Organometallics 2003, 22, 2660. (d) Deubel, D. V.; Ziegler,
T. Organometallics 2003, 22, 2660. (e) Szabo, M. J.; Jordan, R. F.;
Michalak, A.; Piers, W. E.; Weiss, T.; Yang, S.-Y.; Ziegler, T. Organo-
metallics 2004, 23, 5565.
(5) (a) Klabunde, U.; Mulhaupt, R.; Herskovitz, T.; Janowicz, A. H.;
Calabrese, J.; Ittel, S. D. J. Polym. Sci., Part A: Polym. Chem. 1987, 25,
1989. (b) Klabunde, U.; Ittel, S. D. J. Mol. Catal. 1987, 41, 123. (c)
Klabunde, S. U.S. Patent 4716205, 1987.
(6) (a) Wang, C. M.; Friedrich, S.; Younkin, T. R.; Li, R. T.; Grubbs,
R. H.; Bansleben, D. A.; Day, M. W. Organometallics 1998, 17, 3149. (b)
Younkin, T. R.; Connor, E. F.; Henderson, J. I.; Friedrich, S. K.; Grubbs,
R. H.; Bansleben, D. A. Science 2000, 287, 460.
(7) (a) Bauers, F. M.; Mecking, S. Macromolecules 2001, 34, 1165. (b)
Bauers, F. M.; Mecking, S. Angew. Chem. Int. Ed. 2001, 40, 3020. (c)
Korthals, B.; Go¨ttker-Schnetmann, I.; Mecking, S. Organometallics 2007,
26, 1311. (d) Go¨ttker-Schnetmann, I.; Wehrmann, P.; Ro¨hr, C.; Mecking,
S. Organometallics 2007, 26, 2348. (e) Mecking, S.; Held, A.; Bauers, F. M.
Angew. Chem., Int. Ed. 2002, 41, 544.
(8) Connor, E. F.; Younkin, T. R.; Henderson, J. I.; Hwang, S.; Grubbs,
R. H.; Roberts, W. P.; Litzau, J. J. J. Polym. Sci. A 2002, 40, 2842.
(9) (a) Drent, E.; van Dijk, R.; van Ginkel, R.; van Oort, B.; Pugh, R. I.
Chem. Commun 2002, 744. (b) Weng, W.; Shen, Z.; Jordan, R. F. J. Am.
Chem. Soc. 2007, 129, 15450.
(10) (a) Kochi, T.; Yoshimura, K.; Nozaki, K. Dalton Trans. 2001, 25.
(b) Nakamura, A.; Munakata, K.; Kochi, T.; Nozaki, K. J. Am. Chem. Soc.
2008, 130, 8128.
(11) Li, X.-F.; Li, Y.-G.; Li, Y.-S.; Chen, Y.-X.; Hu, N.-H. Organo-
metallics 2005, 24, 2502.
(12) (a) Diamanti, S. J.; Khanna, V.; Hotta, A.; Coffin, R. C.; Yamakawa,
D.; Kramer, E. J.; Fredrickson, G. H.; Bazan, G. C. Macromolecules 2006,
39, 3270. (b) Coffin, R. C.; Diamanti, S. J.; Hotta, A.; Khanna, V.; Kramer,
E. J.; Fredrickson, G. H.; Bazan, G. C. Chem. Commun. 2007, 3350. (c)
Rojas, R. S.; Barrera Galland, G.; Wu, G.; Bazan, G. C. Organometallics
2007, 26, 5339.
(13) (a) Albini, A.; Pietra, S. Heterocyclic N-Oxides; CRC Press: Boca
Raton, FL, 1991. (b) Katritzky, A. R.; Lagowski, J. M. Chemistry of
Heterocyclic-N-oxides; Academic Press: London, 1971.
(14) Dyker, G.; Ho¨lzer, B.; Henkel, G. Tetrahedron 1999, 10, 3297.
(15) Nienkemper, K.; Kotov, V. V.; Kehr, G.; Erker, G.; Fro¨hlich, R.
Eur. J. Inorg. Chem. 2006, 366.
(16) Jerchel, D.; Heider, J.; Wagner, H. Justus Von Liebig Ann. Chem.
1958, 613, 153.
(17) Connor, E. F.; Younkin, T. R.; Henderson, J. I.; Waltman, A. W.;
Grubbs, R. H. Chem. Commun. 2003, 2272.
(18) Mongin, O.; Rocca, P.; Thomas-dit-Dumont, L.; Tre´court, F. M. A. G.
F.; Que´guiner, G. J. Chem. Soc., Perkin Trans. 1 1995, 19, 2503.
Results and Discussion
Synthesis of PymNox Ligands. N-Oxides of pyridine and
their derivatives are usually prepared by oxidation of the
corresponding heterocycles. Peroxocarboxylic acids, particularly
m-chloroperbenzoic acid (mCPBA), are the reagents of choice
for this kind of transformation. As previously shown by Erker,15
2-acetylpryridine is oxidized in this fashion to afford the
corresponding oxide (Scheme 1). The sensitive aldehyde group
of pyridine-2-carbaldehyde does not survive these oxidation
conditions, and a different approach must be used in order to
obtain the corresponding N-oxide. Selective oxidation of the
methyl group of 2-picoline N-oxide (readily obtained from
2-picoline) by selenium dioxide in pyridine under reflux provides
an adequate method to obtain the required compound (Scheme
2).16 2-Acetyl and 2-carbaledehyde pyridine N-oxide derivatives
were condensed with 2,6-diisopropylaniline in a Dean-Stark
apparatus to afford ligands 1 and 2.

Nickel 2-Iminopyridine N-Oxide Complexes
Organometallics, Vol. 27, No. 18, 2008 4713
Scheme 2
Scheme 3
previously described by Braunstein.19 Even though 2-bromo-
6-methylpyridine was obtained in one step by selective monom-
ethylation of 2,6-dibromopyridine with methyllithium,20 we
found the method described by Schubert et al. more advanta-
geous. The latter relies on a diazotation of a commercially
available precursor, 2-amino-6-methylpyridine.21
Table 1. IR and Structural Data for Compounds 1-6
bond distances (Å)a
ν(N-O)
ring
ring
R/Z/R′b
cm-1 N1-O1 C1-C2 C1dN2 avc av SDd
5
3
4
1
2
6
H/NO2/H
benzo/H/H
Ph/H/H
1275
1245
1250
1255
1250
1250
1.290(2) 1.469(2) 1.265(2) 1.374
1.292(1) 1.466(1) 1.273(1) 1.394
1.296(2) 1.462(2) 1.255(2) 1.363
1.303(2) 1.493(2) 1.277(2) 1.379
1.306(2) 1.460(3) 1.265(3) 1.374
1.310(7) 1.444(9) 1.247(8) 1.365
0.01
0.03
0.006
0.01
0.01
0.02
In addition to the mentioned steric effects, electron-attracting
(NO2) and -releasing groups (OMe) at position 4 in the aromatic
ring of the salicylaldiminato ligands have a dramatic effect on
the activity of the corresponding nickel catalysts.7 In order to
investigate the occurrence of similar electronic effects in the
PymNox system, we were interested in preparing ligands
displaying the same substitution pattern in the pyridine ring.
Fortunately, pyridine N-oxides are prone to a wide variety of
both electrophilic and nucleophilic substitution reactions,13
which considerably simplified our task. 2-Picoline N-oxide is
readily and selectively nitrated at the 4 position using a
procedure adapted from the literature,22 as illustrated in Scheme
3. The same methods described above allowed the transforma-
tion of 4-nitro-2-picoline into 5. The methoxy derivative 6 was
directly accessible from 5, by nucleophilic displacement of the
nitro group with sodium methoxide.
H/H/Me
H/H/He
H/OMe/H
a See Figure1 numbering scheme. b See Chart2. c Average of the
pyridine C-C and C-N bonds. d Standard deviation of the ring bond
average. e Average distances for two crystallographically independent
molecules.
configuration, with the CdN bond pointing outward from the
N-O group and with the imino and the heterocyclic moieties
lying in an approximately coplanar position. The only significant
deviation from planarity is found for ligand 1, where the imino
group and the pyridine N-oxide fragment adopt a dihedral angle
of 40.8° in order to avoid steric repulsions. Additional metric
parameters are listed in Table 1, together with the corresponding
ν(N-O) frequencies.
The electronic structures of heteroaromatic N-oxides have
often been explained in terms of mesomeric structures; some
of them are represented in Chart 2. Electron-releasing substit-
uents in the 2, 4, or 6 position favor the localization of partial
negative charge on the oxygen atom of the oxide, decreasing
the N-O bond order (form A), while electron-withdrawing
groups produce the opposite effect (form C). Groups 2-formylimi-
The electronic structure of pyridine N-oxides, and therefore
their ability to coordinate and stabilize metal complexes, is
strongly influenced by the substituents attached to the hetero-
cyclic ring.13 It is expected that the influence of these groups
could correlate to some extent with certain spectroscopic
parameters, such as the IR ν(N-O) band frequency or the
intramolecular bond distances. Reliable identification of the
N-oxide absorption is difficult because it occurs at 1200-1300
cm-1, a spectral region populated with many other bands. In
spite of this, we have attempted an assignation of these bands
on the basis of their medium to strong intensity and comparisons
between the different spectra for consistency. In a further effort
to characterize the properties of ligands 1-6, their crystal
structures have been determined. An ORTEP view of compound
2 in shown in Figure 1, and full lists of bond lengths and angles
for all the remaining ligands are provided as Supporting
Information. The structure of ligand 1 was previously reported
by Erker.14 The six molecules were found in the same
(19) Speiser, F.; Braunstein, P. Organometallics 2004, 23, 3373.
(20) Basu, B.; Frejd, T. Acta Chim. Scand. 1996, 50, 316.
(21) Schubert, U. S.; Eschbaumer, C.; Hellert, M. Org. Lett. 2000, 2,
3373.
Figure 1. ORTEP view of the crystal structure of compound 2.

4714 Organometallics, Vol. 27, No. 18, 2008
Brasse et al.
substituted complex Ni5 (∆ν ) 20 cm-1) and largest for the
methoxy-bearing derivative Ni6 (∆ν ) 58 cm-1), confirming
that electron-donor groups favor a stronger interaction of the
ligand with the metal center.23 The composition of Ni1-Ni6
contrasts with those of the complexes reported by Erker, which
display an additional ligand, either a solvent molecule (THF)
or a second, monodentate ligand unit.14 The crystal structure
of complex Ni1, shown in Figure 2, shows that in this case
nickel achieves pentacoordination by formation of a dimeric
molecule bridged by bromide ligands. The metal center is found
in a distorted square-pyramidal environment, with the basal
Ni-Br2′ distance (2.5708(3) Å) being appreciably longer than
Ni-Br1 and Ni-Br2 (2.4356(3) and 2.4506(3) Å, respectively).
The Ni-O distance, 1.9774(12) Å, is slightly longer than in
nickel salicylaldiminate complexes (ca. 1.91 Å),24 but the
difference is small, considering the different coordination
number of nickel in the two complexes. As compared with the
free ligand, the effect of coordination is more evident in the
N1-O1 bond, which elongates by 0.023 Å, than in the imine
C1dN2 bond, which is increased by only 0.01 Å. In contrast
with the nickel salicyladiminate complexes, which invariably
display strictly planar structures, the PymNox chelate adopts a
folded configuration. Since the N2dC1-C2-C3 dihedral angle
is not very different in the complex and in the free ligand (ca.
27°), this effect can be in part attributed to the natural tendency
of the ligand to twist in order to avoid steric interactions between
the imine methyl and the pyridine ring. However, the even wider
dihedral angle Ni-O1-N1-C2 (52.3°) suggests that the chelate
ring system is inherently more flexible than that of the
salicyladiminate complexes.
Chart 2
no and acetaldimino are also electron-withdrawing, and their
effect can be described by form D. Increasing the negative
charge on the oxygen makes the pyridine N-oxide more akin to
an anionic phenoxide fragment, enhancing its donor properties.
Since the N-O bond order is related to the relative weights of
forms A and C, the frequencies of the IR ν(N-O) bands are
expected to decrease on going from the less to the more electron-
rich ligands, while the N-O bond lengths are expected to
display the opposite trend.
The expected trends are approximately matched by the
ligands. Thus, ligand 5, with a strongly electron-withdrawing
group (NO2), has the higher ν(N-O) frequency and shorter
N-O distance, while 6, with the electron-donor methoxy OMe
substituent, is at the opposite extreme of the series. However,
variation of these parameters is less pronounced in the remaining
compounds. Frequencies of the IR ν(N-O) bands were found
to be virtually identical for 4, 3, and 6, and the range of N-O
bond lengths is only 0.02 Å wide. The latter suggests that the
methoxy-substituted derivative 6 is actually similar to the
unsubstituted ligand 2, while compounds 3 and 4, bearing mild
electron-withdrawing aryl groups, are closer to 5. The shape of
the pyridine ring confirms that electronic effects have little
influence on the ligand structure. Mesomeric forms A and C
suggest that such effects might cause some degree of bond
alternation in the six C-C and C-N distances of the heteroaro-
matic ring; however these are quite similar in almost all cases.
As can be seen, the standard deviation of the average C-C/
C-N bond length remains small and almost the same for five
of the six compounds. The ring displays a somewhat less regular
shape in 3 due to the presence of the fused benzo ring of the
quinoline system. Bond lengths within the imino moiety are
also very similar from one ligand to the other, except for 1,
where the C1-C2 bond is significantly longer, probably due to
the lack of coplanarity of this group and the heteroaromatic
fragment causes a lesser degree of electronic conjugation.
In spite of their paramagnetic character, complexes Ni1, Ni3,
1
and Ni4 exhibit useful H NMR spectra at room temperature,
which have been assigned on the basis of the signal intensities
and width. All the signals could be located, except for that of
the aldimino H atom of Ni3 and Ni4. At room temperature, the
spectra of the remaining nickel compounds show very broad
signals that are difficult to assign, presumably due to the
occurrence of fluxional processes. Rapid exchange also occurs
in the case of Ni1, whose 1H spectrum is simpler than expected
on the basis of its solid-state structure, displaying a single set
of isopropyl signals. The isopropyl methyl groups are diaste-
reotopic, indicating that this apparent simplicity is not due to
the rapid rotation of the aryl groups, but probably involves either
a rapid exchange of the apical and bridging bromide ligands or
dimer dissociation. The situation is the same in Ni4, but
1
somewhat surprisingly, Ni3 shows a more complex H NMR
spectrum, with two independent sets of i-Pr signals, which
suggests that a dimeric structure is rigidly maintained in solution
in this case.
In contrast with the facile formation of the nickel derivatives,
the PymNox ligands fail in general to react with common
palladium precursors such as Pd(MeCN)2 or PdCl2(cod) at room
temperature. However, ligand 4 reacts directly with PdCl2 in
warm (40 °C) acetonitrile,14 affording Pd4 as a yellow
Syntheses of PymNox Complexes. Nickel halocomplexes
Ni1-Ni6 are readily prepared in good yields by treatment of
NiBr2(dme) (dme ) 1,2-dimethoxyethane) with a slight excess
of the corresponding ligand in dichloromethane, followed by
precipitation with hexane (Scheme 4). They were isolated as
red-brown (Ni1, Ni2, Ni5, Ni6), pink (Ni4), or green (Ni3)
solids. They are all high-spin, with µeff ) 2.9-3.5 µB at room
temperature. In general they are quite stable and can be handled
under air except for the nitro complex Ni5, which decays on
exposure to moist air. Their IR spectra display a medium to
strong absorption assigned to ν(N-O) in the proximity of 1200
cm-1. The ca. 40 cm-1 shift to lower frequency (as compared
to the free ligands) is consistent with a significant loss of the
N-O π-bonding. This shift is less pronounced for the nitro-
1
diamagnetic compound, whose H and 13C{1H} NMR spectra
(22) Tre´court, F.; Gervais, B.; Mongin, O.; Gal, C. L.; Que´guinier, G.
J. Org. Chem. 1998, 63, 2892.
(23) (a) Garvey, R. G.; Nelson, J. H.; Ragsdale, R. O. Coord. Chem.
ReV. 1968, 3, 375. (b) Orchin, M.; Schmidt, J. Coord. Chem. ReV. 1968, 3,
345. (c) Karayannis, N. M.; Oytlewski, L. L.; Mikulski, C. M. Coord. Chem.
ReV. 1973, 11, 932.
(24) (a) Zuideveld, M. A.; Wehrmann, P.; Ruˆhr, C.; Mecking, S. Angew.
Chem. Int. Ed. 2004, 43, 869. (b) Chen, Q.; Yu, J.; Huang, J. Organome-
tallics 2007, 26, 617. (c) Hu, T.; Li, Y.-G.; Liu, J.-Y.; Li, Y.-S.
Organometallics 2007, 26, 2609.

Nickel 2-Iminopyridine N-Oxide Complexes
Organometallics, Vol. 27, No. 18, 2008 4715
Scheme 4
are fully consistent with the proposed structure. The less ready
formation of the palladium complexes suggests that the PymNox
ligands bind less strongly the softer Pd(II) center than the Ni(II)
ion. This is apparently supported by the smaller coordination
frequency shift of the ν(N-O) band observed for Pd4 (1230
cm-1, ∆ν ) 25 cm-1) than for its Ni analogue (∆ν ) 46 cm-1).
Ethylene Polymerization with PymNox Complexes. On
treatment with MMAO or diethylaluminum chloride (DEAC),
all PymNox nickel complexes become active ethylene polym-
erization or oligomerization catalysts. In contrast, the palladium
complex Pd4 is found to be inactive under these conditions.
Table 2 collects activity data for these catalysts, and polymer
characterization data are displayed in Table 3. All experiments
were carried out in Fisher-Porter glass reactors with magnetic
stirring, immersed in a water thermostatic bath. The reactor,
charged with solvent and catalyst under N2 atmosphere, was
flushed three times with ethylene and allowed to stabilize at
the selected pressure and temperature before starting the
experiment by injection of the co-catalyst solution through a
septum-capped port. Isobaric ethylene consumption was con-
tinuously monitored along the experiments. In some of them,
the reactor was also equipped with a thermocouple probe in
order to record changes in the internal temperature. Figure 3
shows a typical activity-temperature profile, corresponding to
entry 9 of Table 2. As can be seen, the temperature curve
parallels that of the catalytic activity, the latter calculated from
ethylene consumption data. In general, catalytic activity reaches
a maximum about 3-10 min after the beginning of the
experiment, which is followed by a smooth decay. These
features can be analyzed to provide some parameters that can
be used to give a description of the experiments in a qualitative
context. While the time elapsed until the activity peak is attained
can be linked to the rate of catalyst activation, the height of the
maximum measures the optimum catalyst performance under
the given experimental conditions. On the other hand, the
declining part of the activity curve can be approximately fitted
by an exponential function, providing apparent half-life times
for the catalyst that can be considered an indication of catalyst
stability. The three parameters (maximum activity, time at which
it occurs, and catalyst half-life) are collected in Table 2. In
addition, the activity curves provide an accurate indication of
how long the catalyst remains active in each experiment, and
this time rather than the actual duration of the experiment,25
has been used to calculate the average activity of the cata-
lyst.
Catalysts N1 and Ni2 produce similar low molecular weight
polyethylenes (Mn ≈ 1000-2000) with small polydispersity
indexes of 1.2-1.7. GC analyses of the reaction mixtures reveal
the presence of oligomers with a flat Flory-Schulz molecular
weight distribution (R ≈ 0.85), which is consistent with the low
molecular weight fraction of the polymer rather than a dif-
ferentiated oligomeric product. The NMR spectra of the polymer
produced by Ni1 indicate that this is more branched than that
obtained with Ni2. This is also reflected in the Tm and
crystallinity indexes indicated in the DSC data that are lower
for the former catalyst. Higher productivities are achieved with
catalyst Ni2 than with Ni1, although the recorded catalytic
activities show a different dependency on the cocatalyst dosage.
The latter benefit of high amounts of cocatalyst, while the former
has a performance optimum for an Al/Ni ratio of 750. In both
cases, there is a direct relationship between the Al/Ni ratio and
the height of the catalytic activity maximum, suggesting that
the efficiency of the catalyst activation process might be a
limiting factor for the recorded activities. In the case of Ni2,
the productivity falls at the higher MMAO concentrations, which
appears to be due to shorter catalyst decay times (entry 6, Table
2). However, at difference with other catalysts in this study (see
below), Ni1 and Ni2 are fairly tolerant to high Al/Ni ratios.
Interestingly, attempts to activate Ni2 with DEAC caused
Figure 2. ORTEP view of compound Ni1. Selected bond distances
(Å) and angles (deg): N1-O1, 1.3258(17); C2-C1, 1.486(2);
C1-N2, 1.286(2); Ni-O1, 1.9774(12); Ni-N2, 2.0911(13); Ni-Br1,
2.4356(3); Ni-Br2, 2.4506(3); Ni-Br2′, 2.5708(3); Ni-O1-N1,
114.82(9); N2-C1-C2-N1, 29.06; Ni-O1-N1-C2, 52.34.
(25) The experiment duration was set to 20-25 min in most cases.
Shorter times were taken when the catalyst activity was observed to drop
to an inappreciable level (less than 90-95% of the initial activity).

4716 Organometallics, Vol. 27, No. 18, 2008
Brasse et al.
Table 2. Ethylene Polymerization Experiments
exptl conditionsa
results
calc
av activg activh activi
cat.
entry no.
[Ni]
time
press ext temp max. temp PE yield
max. time max.
k
R/Z/R′b
µMc [Ni]/[Al]d (min) (bar)
(°C)e
(°C)f
(grs)
(min)j
t1/2
23
13
15
15
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
Ni1 Me/H/H
Ni1 Me/H/H
Ni1 Me/H/H
Ni1 Me/H/H
Ni1 Me/H/H
Ni2 H/H/H
Ni2 H/H/H
Ni2 H/H/H
Ni2 H/H/H
Ni2 H/H/H
Ni2 H/H/H
Ni2 H/H/H
Ni5 H/NO2/H
Ni5 H/NO2/H
Ni5 H/NO2/H
Ni6 H/OMe/H
Ni6 H/OMe/H
Ni6 H/OMe/H
Ni3 Benz/H/H
Ni4 Ph/H/H
Ni4 Ph/H/H
Ni4 Ph/H/H
Ni4 Ph/H/H
Ni4 Ph/H/H
Ni4 Ph/H/H
Ni4 Ph/H/H
Ni4 Ph/H/H
80
80
80
80
80
100
80
100
100
100
80
2000
1500
1000
500
100
2000
1000
750
250
50
36
22
21
25
26
17
62
14
19
14
120
30
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
3
7
5
5
30
30
30
30
30
30
30
30
30
30
30
30
30
30
30
30
30
30
30
30
30
30
30
30
30
50
14
2.00
0.84
0.78
0.67
0.58
2.15
3.00
3.10
2.65
1.72
0.48
0.00
0.00
0.00
0.00
0.81
1.30
2.00
1.00
0.30
4.20
2.20
3.50
0.04
3.60
0.30
5.4
832
572
556
404
336
760
724
1328
836
736
60
800
524
652
456
336
984
460
1420
812
872
184
0
792
168
20
452
340
424
424
72
1220
944
1080
812
4
5
4
6
5
5
5
3
3
540
n.d.m
8
40
2196
1028
1980
1280
1092
616
0
876
244
132
792
976
1572
668
472
2472
1100
2028
60
5760
152
32
14
11
32
39
37
34
6
10
100l
10l
n.d.m
80
80
80
80
0
0
0
0
2000
1000
250
2000
1000
50
1000
75
50
50
50
50
50
32
32
31
33
35
33
9
4
2
3
3
4
3
1
5
6
4
9
2
4
11
n.d.m
25
23
19
21
22
29
109
27
14
20
26
30
7
n.d.m
13
100
100
100
80
100
100
100
100
100
100
100
100
232
356
476
136
68
1800
660
808
8
17
30
>50
31
40
36
40
31
53
51
36
2
4
7
10
1472
588
720
30
2412
28
n.d.m
1n
3084
76
2700
50
50
24
12
8
1624
3772
n.d.m
a Solvent, toluene. b Ligand substitution pattern, see Chart2. c Experiments with [Ni] ) 80 µM carried out in 50 mL (4 µmol) and [Ni] ) 100 µM, in
100 mL (10 µmol). d Co-catalyst ) MMAO unless otherwise stated. e Temperature of the external bath. f Maximum internal temperature recorded in the
experiment. g Average productivity from PE weight (kg PE/mol Ni · h). h Average activity from ethylene consumption (kg /mol Ni · h). i Maximum
instantaneous activity recorded in the experiment. j Time (min) at the maximum activity. k Catalyst half-life, estimated from the activity decay rate.
l DEAC co-catalyst. m No suitable exponential decay curve observed. n Approximate value.
Table 3. Polymer and Oligomer Characterization Data
exptl conditions
polymer data
Branches (13C NMR).
GC
GPC
DSC
cat.
[Ni] [Ni]/ press ext temp Flory- Schulz Mn Mw
entry no. R/Z/R′b µMc [Al]d (bar) (°C)
R
10-3 10-3 PDI Me/103C Et/103C Bu/103C Pen103C Hex+/103C MeT/103C Tm % cryst
1
2
3
4
8
9
Ni1 H/H/Me
Ni1 H/H/Me
Ni1 H/H/Me
Ni1 H/H/Me
Ni2 H/H/H
Ni2 H/H/H
8
8
8
8
8
8
8
8
2000
1500
1000
100
2000 11
750 11
250 11
50 11
5
5
5
5
30
30
30
30
30
30
30
30
30
30
30
30
30
30
30
30
30
50
14
53.9
57.2
59.8
58.9
7.6
5.6
6.1
6.6
1.5
1.6
1.7
1.9
3.0
3.5
4.4
4.5
15.0
15.2
14.9
16.4
80.9
83.1
87.0
88.0
53
55
3
6
1.85 3.05 1.65
2a
0.81
0.85
0.87
0.89
1.01 1.75 1.74
1.35 2.22 1.64
1.49 2.36 1.58
2.28 2.63 1.15
2.25 2.80 1.24
2.21 2.65 1.20
2.07 2.42 1.17
21.79 92.35 4.24b
27.10 77.00 2.84
34.06 90.91 2.67
6.52 68.61 10.5c
20.16 57.21 2.84
16.62 45.5 2.74
76.99 382.3 4.97d
35.9
31.1
29.3
6.2
5.4
4.6
3.8
8.2
5.0
5.5
7.2
1.9
3.1
2.4
2.5
2.9
2.4
1.2
1.8
1.2
1.1
2.4
1.7
1.6
2.1
0.7
0.6
0.8
0.4
0.6
0.8
0.7
6.2
5.5
4.9
5.5
6.8
6.2
7.1
0.9
0.7
0.9
0.7
0.6
0.7
0.7
16.2
12.4
11.8
13.7
13.5
13.5
14.9
2.5
2.4
2.2
3.7
3.2
64.5
55.0
51.0
36.0
70.2
68.0
76.4
27.9
28.3
29.3
28.5
30.2
33.2
17.5
57
57
58
31
38
46
10 Ni2 H/H/H
11 Ni2 H/H/H
17 Ni6 H/OMe/H 10 2000
18 Ni6 H/OMe/H 10 1000
19 Ni6 H/OMe/H 10
20 Ni6 H/OMe/H 10
21 Ni4 Ph/H/H
22 Ni4 Ph/H/H
23 Ni4 Ph/H/H
24 Ni4 Ph/H/H
25 Ni4 Ph/H/H
26 Ni4 Ph/H/H
27 Ni4 Ph/H/H
5
5
5
5
5
5
5
3
7
5
5
0.92
0.91
43.3
41.1
45.1
21.9
21.4
23.2
21.2
23.0
25.2
13.9
250
50
75
50
50
40
50
50
50
10
10
10
10
10
10
10
107
106
129
45
42
44
103
98
130
39
35
56
4.0
1.1
a Soluble fraction. b Ligand substitution pattern, see Chart2. c Experiments with [Ni] ) 8 µM carried out in 50 mL (4 µmol) and [Ni] ) 10 µM, in
100 mL (10 µmol). d From 13C NMR.
ethylene consumption, but no solid or liquid products could be
isolated on workup. Standard GC analysis showed the absence
of significant amounts of oligomers heavier than C6,; therefore
DEAC causes a shift in the activity from polymerization to
dimerization or trimerization, which suggests a profound
alteration of the catalyst structure.
Comparison of Ni2 with the pyridine-substituted catalysts Ni5
(Z ) NO2) and Ni6 (Z ) OMe) provides some information on
the influence of electronic effects on the catalytic performance
of the PymNox catalysts. The catalytic properties of compound
Ni6 are not too different from those of Ni1 and Ni2, but the
former gives rise to a somewhat less productive catalyst than
those originated from the latter two. The lower activity maxima
and smaller exotherms recorded for Ni6 confirm that the
introduction of the methoxy group gives rise to an intrinsically
less active catalyst. This system is rather sensitive to the MMAO
activator, but at low Al/Ni ratios (≈50), it is remarkably stable,
with an apparent half-life of 30 min. In spite of the low co-
catalyst load, the precatalyst is rapidly activated, which con-
tributes decisively to maintain a relatively good activity average.

Nickel 2-Iminopyridine N-Oxide Complexes
Organometallics, Vol. 27, No. 18, 2008 4717
of this catalyst is its low tolerance to the activator agent.
Negligible activities were recorded when Al/Ni ratios above 75
are used (entry 20, Table 2). Activity curves are characteristic,
featuring high maxima and strong exotherms, followed by rapid
decay, indicating that the catalyst is intrinsically very active,
but relatively short lived under the selected experimental
conditions. The need to initiate the reaction with very small
amounts of cocatalyst and the rather limited catalyst stability
lead to low experimental reproducibility (compare entries 21-23
in Table 2), but on average this complex gives rise to higher
productivities than any of the remaining catalysts.
In order to improve the behavior of the Ni4-based catalytst,
we studied the influence of temperature and pressure. Increasing
the initial reaction temperature to 50 °C nearly causes the
shutdown of the catalytic activity (entry 26, Table 2) and affords
lower molecular weight polymer (entry 26, Table 3). The same
temperature effects are often observed with late transition metal
catalysts and are usually attributed to the increase of the catalyst
decay rate.27,28 The activity record for this experiment displays
a typical profile with a low maximum (just 150 kg/mol Ni · h),
but catalyst decay does not appear to be substantially faster than
in the 30 °C experiment. In contrast, this catalyst attains its
maximum activity level when activation is started at 14 °C (entry
27, Table 2). In this experiment, catalytic activity is observed
to gradually increase over time, with the maximum level being
reached at the end of the experiment. This is consistent with
slow but efficient catalyst activation, together with inappreciable
decay. The parallel rise of the reactor temperature (from 14 to
32 °C) complicates the measure of the monomer consumption
due to the pressure compensation effect, and the weight of the
polymer is actually significantly larger than that calculated from
the ethylene consumption data. The inhomogeneous temperature
conditions during the experiment are also reflected by the broad
polymodal polymer molecular weight distribution that displays
two main peaks at 103.000 and 800.000 amu.
Entries 23, 24, and 25 (Table 2) illustrate the dramatic
influence of the ethylene pressure on the catalytic activity of
the Ni4 catalyst. On going from 3 to 7 bar, the average
productivity experiences a 100-fold increase. This is clearly a
nonlinear effect and cannot be uniquely related to the variation
of monomer concentration in the solution. In fact, rapid catalyst
deactivation was observed at both the highest and lowest
pressures, suggesting that the interplay of different factors, such
as the dependency of the catalyst stability on the monomer
concentration, the efficiency of the activation process, and the
reactor temperature, leads to a very complicated situation. At 7
bar, very high catalytic activities are rapidly achieved. The
causes of the rapid catalyst decay in this experiment are unclear.
Thermal deactivation (the reactor temperature rises above
53 °C) or encapsulation of the catalyst in the polymer and
precipitation are possible explanations.
Figure 3. Typical activity and temperature profile for an ethylene
polymerization experiment (entry 9, Table 2). Instantaneous activi-
ties have been calculated from ethylene uptake data.
It is possible that the interaction of the MeO group with the
Lewis acidic Al centers is responsible for the co-catalyst
sensitivity observed in this system.
Complex Ni6 affords significantly heavier (Mw ≈ 2700) and
slightly more branched polyethylenes than Ni2, also with a low
polydispersity (PDI) ratio (entries 17-20, Table 3). As com-
pared to the latter, the polymer obtained with Ni6 displays higher
molecular weight, in line with the even flatter Flory-Schulz
distribution (R ≈ 0.9) of the soluble oligomeric fraction. Similar
electronic effects have been observed in palladium R-diimine
system.26 Likewise, it would be expected that compound Ni5,
which exhibits a strongly electron-withdrawing NO2 group in
the pyridine ring, should produce lower molecular weight
products. Indeed, activation of Ni5 with MMAO exclusively
affords light, volatile oligomers (C4, C6), as observed in the
Ni2/DEAC system. As already mentioned, ligand 5 appears to
be the weakest in the series, and the dramatic selectivity shift
observed for the Ni5 system is likely to be due to some
fundamental alteration of its nature (as previously suggested
for the Ni2/DEAC system), rather than to selectivity modulation
by the electron-withdrawing group. Somewhat surprisingly, the
activity profiles recorded with both oligomerization catalysts
shows that these are long-lived systems.
The influence of substituent groups next to nitrogen in the
pyridine ring was explored with compounds Ni3 and Ni4. As
already mentioned, Grubbs’ salicylaldiminate complexes benefit
from this kind of substitution, which provides the most active
catalysts and higher molecular weight polymers. Compound Ni3
is disappointing on this regard, displaying a very low activity
level. Only by resorting to long polymerization times can a
significant amount of polymer be isolated. The activity profile
recorded in a 109 min run (entry 19, Table 2) shows little or no
catalyst decay, indicating that, in spite of its low catalytic
activity, this is a very stable catalyst. On the contrary, complex
Ni4 shows the expected positive influence of the 6-phenyl
substituent, achieving very high activity levels (up to 1800 kg
PE/mol Ni · h at 5 bar ethylene pressure) and producing medium
to high molecular weight polyethylene with a low level of
Although the polymer produced at the lower pressure has a
broad, polymodal distribution (possibly due to the irregular
catalyst performance), neither the molecular weight nor the
number and type of branches appear to be significantly altered
by pressure between 3 and 7 bar. This suggests that monomer
concentration has little effect on the chain propagation/walking
rate balance, in contrast with the observations in the R-diimine
branching and narrow molecular weight distributions (Mw
)
(27) Gates, D. P.; Svejda, S. A.; On˜ate, E.; Killian, C. M.; Johnson,
L. K.; White, P. S.; Brookhart, M. Macromolecules 2000, 33, 2320.
(28) Britovsek, G. J. P.; Bruce, M.; Gibson, V. C.; Kimberley, B. S.;
Maddox, P. J.; Mastroianni, S.; McTavish, S. J.; Redshaw, C.; Solan, G. A.;
Stro¨mberg, S.; White, A. J. P.; Williams, D. J. J. Am. Chem. Soc. 1999,
121, 8728.
(70-90) × 103, PDI 2.7-2.8). The DSC data for these products
(Tm and crystallinity degree) are typical for LLDPE. A drawback
(26) Popeney, C.; Guan, Z. Organometallics 2005, 24, 1145.

4718 Organometallics, Vol. 27, No. 18, 2008
Brasse et al.
nickel system.27 However our data are clearly insufficient on
this regard, and more experiments are necessary to clarify this
point.
catalyst activation process, methyl acrylate (1 mL) was intro-
duced ca. 1-3 min after triggering the polymerization with
MMAO. For most of the catalysts, this procedure results in
complete activity quench. The exception is catalyst Ni1, which
produces a small amount of polymer.
As a final consideration in this section, it is interesting to
compare the catalytic features of the cationic PymNox and
neutral salicylaldiminate system. The exact salicylaldiminate
counterparts of the PymNox ligands 2, 4, 5, and 6 have been
used by Grubbs in polymerization catalysis.7 It must be
considered that the different experimental conditions (particu-
larly the higher pressures employed in the studies on the neutral
system) make difficult or impossible an accurate comparison
between the two systems. In addition, since Grubbs’ catalyst
precursors contain a reactive Ni-C bond and a labile PPh3
ligand, they can behave as single-component catalysts. However,
activation of the salicylaldiminato complexes with suitable
phosphine scavengers (Ni(cod)2 or B(C6F)3) renders the neutral
catalysts more akin to those formed by the halide PymNox
complexes/alumoxane combination.7a Under these conditions
the neutral catalysts are approximately 1 order of magnitude
less active than the cationic ones, and the difference is even
more marked if the pressure effects are included in the activity
units. At the same time, polymers produced by the cationic
system have comparable molecular weights and branching
degree. For example, upon activation with Ni(cod)2, the neutral
salicylaldiminato analogue of Ni2 displays a modest activity
of just 40 kg/mol Ni · h and produces low molecular weight PE
(Mw ) 4000) at 7 bar (Ni2: 1300 kg/mol Ni · h and Mw ≈ 2000
at 5 bar). However, there are evident analogies in the influence
of catalyst structure. For example, the introduction of aryl
substituents at position 6 in the pyridine ring is a crucial factor
in both systems, improving both activity and product molecular
weight. In the neutral system, phenyl substitution at position 6
causes a 4-fold increase in activity with respect to the unsub-
stituted catalyst and produces PE with Mw ) 23 500 (at the
same pressure, 7 bar, Ni4 has an activity of ca. 3000 kg/mol
Ni · h, and the Mw is 57000). The presence of an electron-donor
methoxy substituent in the pyridine ring also has a similar impact
in the two catalytic systems, leading to a moderate increase of
the molecular weight and some decrease of activity. The effect
of the electron-withdrawing NO2 group marks one of the most
obvious differences between the salycylaldiminato and PymNox
systems: while it leads to high molecular weight polyethylene
in the neutral catalysts (with broad PDI), the cationic PymNox
counterpart (Ni5) affords only light oligomers. The striking
difference between these two catalysts supports our interpreta-
tion that the ethylene oligomerization activity by Ni5 is actually
due to some kind of catalyst decomposition or alteration.
Although significantly less active, the salicylaldiminato catalysts
appear to benefit from better thermal stability than PymNox
catalysts on a general basis. This is particularly true when the
former are allowed to act as single-component catalysts in the
absence of any activating agent,7b although the catalytic
activities are still lower under such conditions.
The ethylene-methyl acrylate copolymer is considerably
more soluble than polyethylene, hampering its quantitative
isolation of small amounts, and this made it difficult to estimate
the catalytic activity for short reaction times. However, the yield
of isolated product in the 3 h run, ca. 650 mg, indicates an
average activity of ca. 20 kg/mol · h. In order to maximize the
amount of polymer recovered, the product was simply washed
with aqueous HCl to remove any traces of aluminum. In
favorable cases, the homogeneity of the product was checked
by re-precipitation from a dichloromethane solution with
tetrahydrofuran (see Experimental Section). We have not carried
out full characterization of the products, but their spectroscopic
features and higher solubility strongly support that these are
ethylene-methyl acrylate copolymers. Thus, they display a
1
characteristic IR absorption band at 1716 cm-1 and their H
NMR spectra show a characteristic signal at δ 3.67 ppm, due
to the methoxycarbonyl group. A two-dimensional DOSY
(Diffusion Ordered SpectroscopY; see Supporting Information)
experiment showed that the methoxycarbonyl group displays
the same diffusion coefficient as the bulk polyethylene frag-
ments, and very likely they belong to a true copolymer and not
to a mixture of homopolymers. From the relative intensity of
the two groups of signals, a methyl acrylate incorporation of
ca. 0.7 mol % can be deduced. The overall intensity of the vinyl
groups (>80% internal) allows an estimation of Mn of about
1900, very similar to the Mn value (GPC) of the homopolymer
produced by the same catalyst. Direct observation of the 13C
resonances belonging to the inserted acrylate units other than
those of the OMe group (δ 51.9 ppm) and the carboxylate carbon
(δ 175.0 ppm) is difficult, due to the low acrylate content of
the polymer and the wealth of signals in the 10-40 ppm region,
corresponding to the branched polyethylene framework. How-
ever, the 2D long-range C-H heterocorrelation experiment
(HMBC) display several cross-peaks that related the 13C
carbonyl signal (δ 175 ppm) to the 1H resonances of the
neighboring groups. Besides an intense correlation with the
methoxycarbonyl group, another four cross-peaks can be located
at δ 2.3, 2.0, 1.7, and 1.6 ppm. These find correspondences
with 13C spectral zones in the one-bond HSQC heterocorrelation
experiment at δ 41.4, 39.7, 34.4, and 30.1 ppm, respectively.
The signal couples 2.3 (H)/41.4 (C) ppm and 1.6 (H)/30.1 (C)
are not very different from those observed by Drent for the
ethylene-methyl acrylate copolymer obtained with palladium
phosphinosulfonato catalysts9a (2.28 (H)/45.7 (C) (>CHCO2-
Me); 1.56 (H), 32.4 (...CH2CH(CO2Me)CH2...)) and attributed
to methyl acrylate units directly inserted into the polyethylene
chain. On the other hand, the pairs 2.0 (H)/39.7 (C) and 1.7
(H)/34.4 (C) compare well with the signals observed for the
>CH-CH2CO2Me unit in methyl 3-ethylhexanoate or methyl
3-ethyloctanoate (CH2: 2.15 (H)/38.5 (C); CH: 1.73 (H), 36.1
(C) ppm)30 and therefore suggest the presence of short
CH2CO2Me branches. However, longer branches ending in
carboxylate groups, similar to those obtained with the R-diimine
palladium catalyst system,3a are apparently absent in this
copolymer, since the characteristic cross-peaks for the terminal
CH2CO2Me methylene groups (2.2 (H), 34.2 (C) in methyl
undecanoate) could not be located.
Copolymerization of Ethylene and Methyl Acrylate. An
important objective of this work is to investigate the ability of
PymNox catalysts to copolymerize ethylene with methyl acry-
late. Systematic copolymerization tests have been performed
for each of the nickel catalysts under optimum conditions found
for ethylene homopolymerization. In these experiments, the
methyl acrylate concentration was set to 0.24 M, while the
ethylene concentration under these conditions was ca. 0.5 M.29
In order to avoid any interference of the co-monomer in the
(30) Lo´pez, F.; Harutyunyan, S. R.; Meetsma, A.; Minnaard, A. J.;
Feringa, B. L. Angew Chem., Int. Ed. 2005, 117, 2812.
(29) Lee, L.; Ou, H.; Hsu, H. Fluid Phase Equilibria 2005, 231, 221.

Nickel 2-Iminopyridine N-Oxide Complexes
Organometallics, Vol. 27, No. 18, 2008 4719
(TLC (SiO2, CH2Cl2/acetone, 1:1) Rf ) 1 organic residues, 0.8
mCPBA, 0.2 product) using first CH2Cl2 as eluent to remove the
excess mCPBA as the first fraction and then acetone to extract the
Conclusion
2-Iminopyridine N-oxides (PymNox) are promising ligands
for applications in catalysis. One of the advantages of these
ligands is that their synthesis is considerably facilitated by the
rich chemistry of heteroaromatic N-oxides. They form stable
complexes with nickel and palladium, the former becoming
active catalysts for the ethylene polymerization or oligomer-
ization on treatment with alumoxanes or DEAC. This demon-
strates that the electrically neutral N-oxide functionality suc-
cessfully substitutes anionic oxygen donor groups widespread
in olefin polymerization or oligomerization catalysts, e.g., the
SHOP-type or salicylaldiminate catalysts, giving rise to func-
tional cationic counterparts. PymNox nickel catalysts are more
active than their neutral salicylaldiminato-based analogues, but
otherwise the two systems display many analogies: they give
rise to similar polymers in terms of both molecular weights and
branching degree, and display similar trends in the influence of
ligand structure on catalyst performance. Thus, electronic effects
modulate catalytic activity, with electron-donor groups leading
to somewhat decreased activities. More importantly, substitution
at position 6 in the aromatic ring with aromatic groups leads to
dramatic increases in both activity and polymer molecular
weight. Similarly to salicylaldiminate catalysts, 2-aldiminopy-
rdine N-oxide catalysts are not compatible with methyl
acrylate.3d However, the ketimino derivative Ni1 copolymerizes
this comonomer with ethylene. NMR analysis suggests that the
resulting ethylene-methyl acrylate copolymer displays both
directly inserted acrylate units and short CH2CO2Me branches.
Further research directed to the optimization of the PymNox
ligands and catalysts for olefin polymerization and copolym-
erization reactions are currently being carried out in our
laboratories.
1
product in the second fraction. H NMR (400 MHz, CDCl3, 298
K): δ 8.17 (d, 3JHH ) 6.8Hz, 1H, Ar meta), 7.67 (d, 3JHH ) 10Hz,
1H, Ar ortho 3), 7.34 (t, 1H, Ar para), 7.30 (t, 1H, Ar ortho 5),
2.77 (s, 3H, Me). Anal. Calcd (%) for C7H7NO2 (137.14 g/mol): C
61.31; H 5.14; N 10.21. Found: C 60.98; H 5.15; N 10.54.
2-[N′-(2,6-Diisopropylphenyl)acetaldimino]pyridine N-Oxide
(1). A mixture of 2-acetylpyridine N-oxide (1.37g, 10 mmol) and
2,6-diisopropylaniline (2.83 mL, 15 mmol) was dissolved in dry
toluene (20 mL) and refluxed for 2 h in a Dean-Stark apparatus.
The solvent was removed under vacuum, and the resulting oil was
purified by flash chromatography with CHCl3 to extract the excess
of aniline and with acetone to extract the product. After removing
the solvent under vacuum 2.5 g (89%) of the product was afforded
as a brown oil. Mp ) 77 °C. 1H NMR (500 MHz, CDCl3, 298 K):
3
3
δ 8.22 (d, JHH ) 7.5 Hz, 1H,CH Py 6); 7.62 (d, JHH ) 9.6 Hz,
1H, CH Py 3); 7.34 (t, 1H, CH Py 4); 7.32 (t, 1H, CH Py 5); 7.15
3
3
(d, JHH ) 7,5 Hz, 2H, m-CH(Ar)); 7.10 (t, JHH ) 7.5 Hz, 1H,
p-CH(Ar)); 2.87 (m, 2H, CH iPr); 2.18 (s, 3H, CH3); 1.21 (d, 3JHH
3
) 7.15 Hz, 6H, CH3 (iPr)); 1.15 (d, JHH ) 7.15 Hz, 6H, CH3
(iPr)). 13C{1H} NMR (75 MHz, CDCl3, 298 K): δ 165.3 (1C,
CdN); 149.9 (1C, Cq Py 2); 144.9 (1C, Cipso Ar); 140.5 (1C, CH
Py 6); 136.4 (2C, Cq-iPr); 126.5 (1C, CH Py 5); 126.5 (1C, CH
Py 4); 126.3 (1C, CH Py 3); 124.6 (1C, p-CH Ar); 123.5 (2C, o-CH
Ar); 28.4 (2C, CH (iPr)); 23.8 (2C, CH3 (iPr)); 23.2 (2C, CH3 (iPr));
20.2 (1C, CH3 (Py)). IR (Nujol, cm-1): 1635.7 ν(CdN); 1254.5
ν(N-O). UV/vis (CH2Cl2, c ) 10-5 M (ꢀ, 105 mol-1 · L · cm-1):
363 (0.337); 290 (1.175); 248 (2.93). MS (ESI, MeOH) m/z: 296.17
[M+]. Anal. Calcd (%) for C19H24N2O (296.41 g/mol): C 76.99; H
8.16; N 9.45. Found: C 76.29; H 8.10; N 9.24.
General Procedure B: 2-Pyridinecarbaldehyde N-Oxide.
Adapted from the literature.19,21 A mixture of picoline N-oxide (5.45
g, 50 mmol) and selenium dioxide (5.54 g, 50 mmol) in pyridine
(25 ml) was heated under reflux for 5 h. The solid material was
filtered off, the filtrate was concentrated in vacuo, and the residual
oil was extracted four times with hot toluene (50 ml). TLC (silica,
acetone/ether, 2:1): Rf ) 0.4 product; 0.08 picoline N-oxide. Flash
chromatography with acetone/ether (2:1) afforded the product as
the first fraction. The solvent was removed in vacuo, and the product
was obtained as a yellow solid (5.16 g, 83%). 1H NMR (500 MHz,
CDCl3, 298 K): δ 10.61 (s, 1H, CHO); 8.19 (d, 3JHH ) 6.5Hz, 1H,
CH Py 2); 7.80 (d, 3JHH ) 7Hz, 1H, CH Py 5); 7.43 (t, 3JHH ) 6.5
Hz, 1H, CH Py 3); 7.30 (t, 3JHH ) 7Hz, 1H, CH Py 4). Anal. Calcd
(%) for C6H5NO2 (123.10 g/mol): C 58.54; H 4.09; N 11.38. Found:
C 58.06; H 4.11; N 11.25.
Experimental Section
All the complexes’ synthesis and catalyst preparations were
carried out under a dry nitrogen atmosphere by conventional
Schlenk techniques. Solvents were rigorously dried by refluxing
over sodium benzophenone (toluene, hexane, diethyl ether, tet-
rahydrofuran) or calcium hydride (CH2Cl2) and freshly distilled prior
to use. Deuterated solvents (Aldrich) were dried over CaH2 (CD2Cl2,
CDCl3) or sodium benzophenone (C6D6) and then distilled. Pyridine
derivatives and other organic reagents were purchased from Aldrich
and used without further purification. NiCl2(dme) was synthesized
as reported in the literature.31 MMAO (1.8 M in heptane) was
purchased from Akzo-Nobel and used without further purification.
IR spectra were recorded on a Bruker Vector 22, and UV-vis on
a Perkin-Elmer model Lambda-12. NMR spectra were obtained on
Bruker Avance DRX 500 (500 MHz), Bruker Avance DRX 400
(400 MHz), and Bruker Avance 300 (300 MHz) spectrometers. The
1H and 13C{1H} resonances of the solvent were used as internal
standard, but the chemical shifts (ppm) are reported with respect
to TMS. Elemental analyses were performed by the Micro-analytical
Service of the Instituto de Investigaciones Qu´ımicas. Molecular
weights of the polymers were determined by gel permeation
chromatography employing universal calibration in a Waters 150c
instrument with differential refractive index (DRI) detector and a
Viscotek 150R DV detector.
General Procedure C: Aniline Condensation. 2[-(N′-2,6-
Diisopropylphenyl)carbaldimino]pyridine N-Oxide (2). A mix-
ture of 2-carboxyaldehyde pyridine N-oxide (5.16 g, 42 mmol),
2,6-diisopropylaniline (8 mL, 42 mmol), and a catalytic amount of
acetic acid in methanol (200 mL) was refluxed for 4 h. The mixture
was concentrated under vaccuum and allowed to crystallize upon
cooling in hexane. The product was filtered off and recovered as a
yellow crystalline solid (9.43 g, 79% at this step, 66% for the whole
1
synthesis). Mp ) 130.4 °C. H NMR (400 MHz; CDCl3; 298 K):
3
δ 8.89 (s, 1H, CHdN); 8.23 (d, JHH ) 4Hz, 1H, CH Py 2); 8.18
(d, 3JHH ) 7Hz, 1H, CH Py 5); 7.35 (m, 2H, CH Py 3 and 4); 7.13
(m, 3H, CH Ar); 2.91 (m, 2H, CH iPr); 1.16 (d, 3JHH ) 7Hz, 12H,
CH3 iPr). 13C{1H} NMR (75 MHz, CDCl3, 298 K): δ 155.4 (1C,
CHdN); 148.6 (1C, Cq-NO); 145.8 (Cq-NdC); 140.3 (1C, CH 6
py); 137.5 (2C, Cq-iPr); 127.6 (1C, CH 5 Py); 125.8 (1C, CH p-Ar);
125.2 (1C, CH 4 Py); 124.8 (1C, CH 3 Py); 123.4 (2C, CH o-Ar);
28.2 (2C, CH(CH3)2); 23.7 (4C, CH(CH3)2). IR (Nujol): ν 1628
cm-1 (CdN); 1603 (CdN); 1251 (N-O). UV/vis (CH2Cl2, 10-4
M, λ ) nm, ꢀ ) 10-4 mol-1 · L · cm-1): 363 (0.337); 290 (1.175);
248 (2.93). MS (ESI; MeOH) m/z: 321 [M + K]+; 305 [M + Na]+;
Ligand Synthesis. General Procedure A: N-Oxidation of
2-Acetylpyridine. 2-Acetylpyridine N-Oxide. A mixture of
2-acetylpyridine (20 mmol) and mCPBA (70 mmol) in CHCl3 (25
mL) was refluxed for 4 h. The solvent was removed under vacuum,
and the resulting product was purified by flash chromatography
(31) (a) Cotton, F. A. Inorg. Synth. 1971, 13, 162. (b) Cotton, F. A.
Inorg. Synth. 1971, 13, 47.

4720 Organometallics, Vol. 27, No. 18, 2008
Brasse et al.
283 [M + H]+. Anal. Calcd (%) for C18H22N2O (282.38 g/mol): C
76.56; H 7.85; N 9.92. Found: C 75.94; H 7.95; N 9.81.
bromide (7.8 mL, 1 M in diethyl ether) over a period of 10 min.
After the solution was heated overnight in a closed ampule under
argon, it was hydrolyzed by pouring it into 20 mL of diluted
hydrochloric acid. The solution was stirred for 15 min before the
organic phase was separated. The aqueous phase was extracted twice
(15 mL) with diethyl ether, and the ether was again discarded. The
aqueous phase was neutralized by slow addition of Na2CO3 until
no further CO2 evolution was observed and a pale yellow precipitate
was formed. The basic solution was extracted three times with
CH2Cl2 (20 mL), and the organic phase was dried over MgSO4.
Then the CH2Cl2 fraction was evaporated under reduced pressure,
yielding a NMR-pure brown oil, which was used directly (1.63 g,
94%). 1H NMR (400 MHz; CDCl3; 298 K): δ 2.65 (s, 3H, py-
Quinaldine N-Oxide. The general procedure A using quinaldine
(1.35 mL; 10 mmol). TLC (SiO2, ether/petroleum ether, 1:1). The
product crystallizes in ether upon cooling, affording a pale brown
solid (1.4 g, 90%). 1H NMR (400 MHz; CDCl3; 298 K): δ 8.74 (d,
1H, CH 3); 7.79 (d, 1H, CH 6); 7.70 (t, 1H, CH 8); 7.64 (d, 1H,
CH 9); 7.55 (t, 1H, CH 7); 7.29 (d, 1H, CH 4); 2.70 (s, 3H, CH3).
Anal. Calcd (%) for C10H9NO (159.18 g/mol): C 75.45; H 5.70; N
8.80. Found: C 74.87; H 5.64; N 8.71.
Quinoleine-2-carbaldehyde N-Oxide. The general procedure
B using quinaldine N-oxide (1.4 g, 9 mmol). TLC (silica, ether):
Rf ) 0.6, product; 0, quinaldine N-oxide. Flash chromatography
with ether afforded the product as a yellow solid (1.16 g, 74% at
3
3
CH3); 7.09 (d, JHH ) 7.6 Hz, 1H, py H5); 7.41 (dd, JHH ) 8.6
1
4
3
this step, 63.5% starting from quinaldine). H NMR (300 MHz;
Hz, JHH ) 1.7 Hz, 1H, Ph Hp); 7.46 (d, JHH ) 7.8 Hz, 1H, py
CDCl3; 298 K): δ 10.83 (s, 1H, CHO); 8.76 (d, 1H, CH 3); 7.88
(d, 1H, CH), 7.77 (m, 4H, CH). Anal. Calcd (%) for C10H7NO2
(173.16 g/mol): C 69.36; H 4.07; N 8.09. Found: C 70.10; H 4.08;
N 8.15.
H3); 7.50 (d, 3JHH ) 8.6 Hz 2H, Ph Hm); 7.62 (d, 3JHH ) 7.8 Hz,
1H, py H4); 7.7 (dd, JHH ) 8.7 Hz, JHH ) 1.7 Hz, 2H, Ph Ho).
Anal. Calcd (%) for C12H11N (169.22 g/mol): C 85.17; H 6.55; N
8.28. Found: C 84.95; H 6.68; N 8.02.
3
4
2-Methyl-6-phenylpyridine N-Oxide. The general procedure A
using 2-methyl-6-phenylpyridine (1.63g; 9.6 mmol). TLC (SiO2,
ether/petroleum ether, 1:1). The product crystallizes in ether upon
2-(N′-2,6-Diisopropylphenylimino)methyl]carbaldiminoqui-
noleine N-Oxide (5). The general procedure C using 2-carboxy-
aldehyde quinoline N-oxide (1.16 g, 6.35 mmol). The product was
recovered as a yellow crystalline solid after filtration (1.9 g, 90%
in this step, 57% for the whole synthesis). Mp ) 85.8 °C. 1H NMR
1
cooling, affording a pale yellow solid (1.16 g, 65.5%). H NMR
(400 MHz; CDCl3; 298 K): δ 2.56 (s, 3H, Me); 7.2 (d, 1H, H3
Py); 7.23 (t, 1H, H4 Py); 7.29 (dd, 1H, H6 Py); 7.24 (t, 1H, Hp
Ph); 7.43 (dd, 2H, Hm Ph); 7.76 (dd, 2H, Ho Ph). Anal. Calcd (%)
for C12H11NO (185.22 g/mol): C 77.81; H 5.99; N 7.56. Found: C
77.50; H 5.89; N 7.36.
2-Carbaldehyde-6-phenylpyridine N-Oxide. The general pro-
cedure B using 2-methyl-6-phenylpyridine N-oxide (1.16 g, 6.27
mmol). TLC (silica, ether): Rf ) 0.7 (2-carbaldehyde-6-phenylpy-
ridine N-oxide), 0.16 (2-methyl-6-phenylpiridine N-oxide). Flash
chromatography with ether afforded the product as the first fraction
(440 mg, 35%), and with ethyl acetate the starting product was
extracted in a second fraction (550 mg). 1H NMR (400 MHz;
CDCl3; 298 K): δ 10.67 (s, 1H, HCdN); 7.78 (d, 1H, Py 3); 7.75
(d, 2H, Ph o); 7.59 (d, 1H, Py 5); 7.53 (t, 1H,Py 4); 7.5 (d, 2H, Ph
m); 7.35 (t, 1H, Ph p). Anal. Calcd (%) for C12H9NO2 (199.20
g/mol): C 72.35; H 4.55; N 7.03. Found: C 71.96; H 4.40; N 6.95.
3
(400 MHz; CDCl3; 298 K): δ 9.13 (s, 1H, CHdN); 8.77 (d, JHH
3
) 8.8Hz, 1H, CH 3); 8.22 (d, JHH ) 8.8Hz, 1H, CH 4); 7.89 (d,
3
3JHH ) 8Hz, 1H, CH 6); 7.78 (m, 2H, CH 8 and 9); 7.70 (t, JHH
) 8Hz, 1H, CH 7,); 7.16 (m, 3H, CH Ar); 2.96 (m, 2H, CH(CH3)2);
1.18 (d, 3JHH ) 6.8Hz, 12H, CH(CH3)2). 13C{1H} NMR (75 MHz;
CDCl3; 298 K): δ 156.1 (1C, CHdN); 148.9 (1C, Cq Py 2); 142.4
(1C, Cipso Ar); 142.2 (1C, Cq 10 quin.); 137.6 (2C, Cq-iPr); 131.3
(1C, Cq 5 quin.); 130.9 (1C, CH 8 quin.); 130.0 (1C, CH 4 quin.);
128.6 (1C, CH 3 quin.); 125.7 (1C, CH 7 quin.); 125.3 (1C, CH
p-Ar); 123.4 (2C, CH o-Ar); 120.2 (1C, CH 9 quin.); 119.7 (1C,
CH 6 quin.); 28.3 (2C, CH(CH3)2); 23.8 (4C, CH(CH3)2). IR (Nujol,
cm-1): 1726; ν(CdN) 1618,8; ν(N-O) 1230. UV/vis (CH2Cl2,
10-4 M, λ ) nm, ꢀ ) 10-4 mol-1 · L · cm-1): 381 (0.423); 325
(0.587); 282 (2.59); 247 (2.33). MS (ESI; MeOH) m/z: 355.1 [M
+ Na]+. Anal. Calcd (%) for C22H24N2O (332.44 g/mol): C 79.48;
H 7.27; N 8.42. Found: C 79.53; H 7.39.
(2-[1-(2,6-Diisopropylphenylimino)methyl]-6-phenyl)pyri-
dine N-Oxide (4). The general procedure C using 2-carboxyalde-
hyde-6-phenylpyridine N-oxide (440 mg, 2.2 mmol). The product
crystallized upon cooling in hexane and was recovered after
filtration as a yellow crystalline solid (709 mg, 90%). Mp ) 132.5
°C. 1H NMR (500 MHz; CDCl3; 298° K): δ 1.18 (s, 12H,
2-Methyl-6-bromopyridine. Adapted from literature.21 Pow-
dered 2-aminopicoline (5.4 g, 50 mmol) was added in portions to
48% hydrobromic acid (25 mL) at 20 to 30 °C with vigorous
stirring. After the entire compound was dissolved, the mixture was
cooled at -20°C. To this suspension was added cooled bromine
(7.5 mL, 150 mmol) dropwise over 30 min, maintaining the
temperature at -20 °C. The resulting paste was stirred for 90 min
at this temperature. Then sodium nitrite (10g, 150 mmol) in water
(100 mL) was added dropwise. After that the reaction mixture was
allowed to warm to 15 °C over 1 h and was stirred for an additional
45 min. The mixture was cooled to -20 °C and treated with cooled
aqueous NaOH (33 g, 50 mL of H2O). During the addition the
temperature was kept at -10 °C maximum. The mixture was
allowed to warm to room temperature and stirred for 1 h. The
mixture was extracted with ethyl acetate, the organic phase was
dried with MgSO4, and the solvent was removed in vacuo. The
residue was purified by flash chromatography (TLC: SiO2 ether/
hexane, 1:8; Rf ) 0.58 impurities; 0.29 product) to yield the desired
2-bromopicoline as a brown oil (3.53g, 40%). 1H NMR (300 MHz;
3
CH(CH3)2); 2.96 (m, 2H, CH(CH3)2); 7.12 (t, JHH ) 7.5Hz, 1H,
Ar-p); 7.16 (d, 3JHH ) 7.5Hz, 2H, Ar-m); 7.36 (t, 3JHH ) 8Hz, 1H,
3
Py 4); 7.46 (t, 1H, Ph-p); 7.47 (m, 2H, Ph-m); 7.51 (dd, JHH
)
8Hz, 4JHH ) 2Hz, 1H, Py 5); 7.78 (d, 3JHH ) 8Hz, 2H, Ph-o); 8.16
3
4
(dd, JHH ) 8Hz, JHH ) 2Hz, 1H, Py 3); 8.94 (s, 1H, HCdN).
13C{1H} NMR (125 MHz; CDCl3; 298 K): δ 23.5 (4C, CH(CH3)2);
28.0 (2C, CH(CH3)2); 123.1 (2C, Ar-m); 123.5 (1C, CH Py 3);
124.8 (1C, CH Ar-p); 124.8 (1C, CH Py 4); 128.4 (2C, CH Ph-o);
128.7 (1C, CH Py 5); 129.4 (2C, CH Ph-m); 129.7 (1C, CH Ph-p);
132.4 (1C, Cipso Ph); 137.3 (2C, Ar Cq-(CH(CH3)2)); 146.2 (1C,
Ar Cq-N); 148.4 (1C, Py Cq-CHN); 150.0 (1C, Py Cq-C6H5); 156.0
(1C, CHdN). IR (Nujol): ν 1623, 1586, 1321, 1292, 1250, 1234,
1190, 1160. UV/vis (CH2Cl2, 10-5 M, λ ) nm, ꢀ ) 10-5
mol-1 · L · cm-1): 369 (0.107); 278 (0.781); 227 (3.17). MS (ESI;
MeOH) m/z: 359.2 [M + H]+; 381.2 [M + Na]+; 397.2 [M +
K]+. Anal. Calcd (%) for C24H26N2O (358.47 g/mol): C 80.41; H
7.31; N 7.81. Found: C 80.44; H 7.35; N 7.83.
3
CDCl3; 298 K): δ 2.51 (s, 3H, Me); 7.08 (d, JHH ) 7.5 Hz, 1H,
H5); 7.26 (d, 3JHH ) 7.5 Hz, 1H, H3); 7.38 (t, 3JHH ) 7.5 Hz, 1H,
H4). Anal. Calcd (%) for C6H6BrN (172.02 g/mol): C 41.89; H
3.52; N 8.14. Found: C 42.01; H 3.65; N 8.15.
4-Nitro-2-picoline N-Oxide. Adaptation from literature.22
A
2-Methyl-6-phenylpyridine. Adapted from literature.19 To a
mixture of 2-bromo-6-methylpyridine (1.76 g, 10.2 mmol), [Ni(a-
cac)2] (32 mg, 0.13 mmol), and P(tBu)3 (0.13 mL, 0.13 mmol) in
20 mL of diethyl ether in an ice bath was added phenylmagnesium
solution of 2-picoline N-oxide (5.45 g, 50 mmol) and KNO3
(15 g) in concentrated H2SO4 (50 mL) was heated for 20 h at 80
°C. The mixture was neutralized by addition of Na2CO3 until the
compound precipitated. Filtration and recrystallization from CH2Cl2

Nickel 2-Iminopyridine N-Oxide Complexes
Organometallics, Vol. 27, No. 18, 2008 4721
1
gave 4 g (50%) of the pure product. H NMR (300 MHz; CDCl3;
(Me,H)NNO-NiBr2, Ni1. The reaction of 1 (326 mg, 1.1 mmol)
with NiBr2(dme) afforded the product Ni1 as a red-brown powder
(386 mg; 75%). 1H NMR (300 MHz; CD2Cl2; 298 K): δ 27.7 (1H,
∆ν1/2 ) 15 Hz; CH Py); 21.8 (2H, ∆ν1/2 ) 27 Hz, CH Ar); 21.1
(1H, ∆ν1/2 ) 28 Hz, CH Ar); 11.9 (2H, ∆ν1/2 ) 440 Hz, CHMe2);
4.2 (6H, ∆ν1/2 ) 22 Hz, CHMeMe); 1.4 (6H, ∆ν1/2 ) 59 Hz,
CHMeMe); -5.9 (1H, ∆ν1/2 ) 26 Hz, CH Py); -16.7 (1H, ∆ν1/2
) 26 Hz, CH Py); -17.4 (1H, ∆ν1/2 ) 25 Hz, CH Py); -26.7
(3H, ∆ν1/2 ) 33 Hz, CH3). IR (Nujol, ν cm-1): 1617.6 (CdN);
1213.3 (N-O). UV/vis (CH2Cl2, 10-4 M, λ ) nm, ꢀ ) 10-4
mol-1 · L · cm-1): 234.21 (2.00). Magnetic moment µc ) 3.48 µB.
Anal. Calcd (%) for C19H24Br2N2NiO (514.91 g/mol): C 44.32; H
4.70; N 5.44. Found: C 45.01; H 5.02; N 5.29.
(H,H)NNO-NiBr2, Ni2. The reaction of 2 (310 mg, 1.1 mmol)
with NiBr2(dme) afforded the product Ni2 as a red powder (400
mg; 80%). IR (Nujol, ν cm-1): 1632.5 (CdN); 1209 (N-O). UV/
vis (CH2Cl2, 10-4 M, λ ) nm, ꢀ ) 10-4 mol-1 · L · cm-1): 324
(0.40); 283 (0.784); 241 (1.90). MS (ESI, CH2Cl2/MeOH) m/z:
421.1 [NiBr(1)]+. Magnetic moment µc ) 3.28 µB. Anal. Calcd
(%) for C18H22Br2N2NiO (500.88 g/mol): C 43.16; H 4.43; N 5.59.
Found: C 43.29; H 4.82; N 5.71.
(H,Qui)NNO-NiBr2, Ni3. The reaction of 3 (332 mg, 1.1 mmol)
with NiBr2(dme) afforded the product Ni3 as a green powder (418
mg; 76%). 1H NMR (300 MHz; CD2Cl2; 298 K): δ 23.1 (1H, ∆ν1/2
) 76 Hz, CH Py); 14.2 (1H, ∆ν1/2 ) 61 Hz, CHMeMe); 11.9 (1H,
∆ν1/2 ) 30 Hz, CH Ar); 11.0 (1H, ∆ν1/2 ) 64 Hz, CHMeMe); 8.9
(3H, ∆ν1/2 ) 98 Hz, CHMeMe); 8.5 (1H, ∆ν1/2 ) 20 Hz, CH Qui);
8.2 (1H, ∆ν1/2 ) 20 Hz, CH Qui); 7.7 (2H, ∆ν1/2 ) 74 Hz, CH
Ar); 5.0 (3H, ∆ν1/2 ) 73 Hz, CHMeMe); 4.0 (3H, ∆ν1/2 ) 85 Hz,
CHMeMe); 3.4 (3H, ∆ν1/2 ) 98 Hz, CH Qui); 2.7 (1H, ∆ν1/2 ) 24
Hz, CH Qui); 2.0 (3H, ∆ν1/2 ) 49 Hz, CHMeMe); -15.9 (1H,
∆ν1/2 ) 56 Hz, CH Py). IR (Nujol, ν cm-1): 1703, 1629, 1613,
1307, 1254 (N-O), 1196, 1140, 1102, 887, 748, 722. UV/vis
(CH2Cl2, 10-4 M, λ ) nm, ꢀ ) 10-4 mol-1 · L · cm-1): 325 (0.57);
277 (1.55); 258 (1.77); 223 (3.18). Magnetic moment µc ) 3.44
µB. MS (ESI, CH2Cl2 + MeOH) m/z: 801.2 [NiBrL2]+; 469
[NiBrL]+; 333.2 [L + H]+. Anal. Calcd (%) for C22H24Br2N2NiO
(550.94 g/mol): C 47.96; H 4.39; N 5.08. Found: C 47.64; H 4.94;
N 4.75.
(H,Ph)NNO-NiBr2, Ni4. The reaction of 4 (358 mg, 1.1 mmol)
with NiBr2(dme) afforded the product Ni4 as a pink powder (502
mg; 90%). 1H NMR (300 MHz; CD2Cl2; 298 K): δ 31.2 (1H, ∆ν1/2
) 30 Hz, CH Py); 21.0 (2H, ∆ν1/2 ) 45 Hz, CH Ar); 17.7 (1H,
∆ν1/2 ) 71 Hz, CH Ar); 11.9 (2H, ∆ν1/2 ) 213 Hz, CHMeMe);
10.6 (2H, ∆ν1/2 ) 30 Hz, CH Ph); 8.7 (1H, ∆ν1/2 ) 24 Hz, CH
Py); 4.5 (6H, ∆ν1/2 ) 35 Hz, CHMeMe); 2.37 (6H, ∆ν1/2 ) 61
Hz, CHMeMe); 0.5 (2H, ∆ν1/2 ) 30 Hz, CH Ph); -11.4 (1H, ∆ν1/2
) 44 Hz, CH Py); -14.4 (1H, ∆ν1/2 ) 33 Hz, CH Py). IR (Nujol,
ν cm-1): 1630; 1599; 1408; 1315; 1278; 1234; 1205 (N-O); 1166;
1101; 763. UV/vis (CH2Cl2, 10-4 M, l ) nm, ꢀ ) 10-4
mol-1 · L · cm-1): 336 (0.87); 289 (1.73); 250 (2.53). Magnetic
moment µc ) 3.52 µB. MS (ESI, CH2Cl2 + MeOH) m/z: 495
[LMBr]+. Anal. Calcd (%) for C24H26Br2N2NiO (576.97 g/mol):
C 49.96; H 4.54; N 4.86. Found: C 49.53; H 4.69; N 4.51.
(H,H,NO2)NNO-NiBr2, Ni5. The reaction of 5 (360 mg, 1.1
mmol) with NiBr2(dme) afforded the product Ni5 as a brown
powder (408 mg; 75%). IR (Nujol, ν cm-1): 1628, 1606, 1583,
1345, 1277, 1255 (N-O), 1198, 1167, 1074, 951, 932, 811, 758.
UV/vis (CH2Cl2, 10-4 M, λ ) nm, ꢀ ) 10-4 mol-1 · L · cm-1):
338 (0.40); 275 (0.48); 223 (3.2). Magnetic moment µc ) 2.93 µB.
Anal. Calcd (%) for C18H21Br2N3NiO3 (545.87 g/mol): C 39.60; H
3.88; N 7.70. Found: C 40.08; H 4.66; N 7.39.
298 K): δ 8.28 (d, 3JHH ) 7Hz, 1H, H2); 8.1 (d, 4JHH ) 3Hz, 1H,
4
3
H5); 7.96 (dd, JHH ) 3Hz; JHH ) 7Hz, 1H, H3); 2.53 (s, 3H,
Me). IR (Nujol): ν 1681 cm-1 (CdN); 1610 (CdN); 1286 (N-O).
Anal. Calcd (%) for C6H6N2O3 (154.12 g/mol): C 46.76; H 3.92;
N 18.18. Found: C 46.78; H 3.83; N 18.58.
4-Nitro-2-carbaldehydepyridine N-Oxide. The general proce-
dure B, using 4-nitropicoline N-oxide (4 g, 26 mmol). TLC (silica,
acetone/ether, 1:3): Rf ) 0.7 product; 0.07 picoline N-oxide. Flash
chromatography with acetone afforded the product as a yellow solid
1
(1.74 g, 40%). H NMR (300 MHz; CDCl3; 298 K): δ 10.46 (s,
1H, CHO); 8.58 (s, 1H, H5); 8.28 (d, 1H, H2); 8.11 (d, 1H, H3).
Anal. Calcd (%) for C6H4N2O4 (168.10 g/mol): C 42.87; H 2.40;
N 16.66. Found: C 42.65; H 2.53; N 16.25.
4-Nitro-2-[N′-(2,6-diisopropylphenylimino)carbaldimino]pyridine
N-Oxide (5). The general procedure C using 4-nitro-2-carboxyal-
dehyde pyridine N-oxide (1.74 g, 10.35 mmol). The product was
purified by flash chromatography with CH2Cl2/ether/hexane, 1:1:
2. TLC: silica (CH2Cl2/ether/hexane, 1:1:2) (Rf ) 0.89 aniline; 0.46
product; 0.1 picoline N-oxide). The product was recovered as a
yellow crystalline solid (1.5 g, 45% for the whole synthesis). Mp
) 123.1 °C. 1H NMR (300 MHz; CDCl3; 298 K): δ 8.91 (d, 4JHH
3
) 7Hz 1H, H3); 8.77 (s, 1H, CHN); 8.28 (d, JHH ) 7.2Hz, 1H,
3
4
H6); 8.16 (dd, JHH ) 7.2Hz, JHH ) 3.4Hz, 1H, H5); 2.86 (m,
2H, CH(CH3)2); 1.17 (s, 6H, CH(CH3)2); 1.16 (s, 6H, CH(CH3)2).
13C{1H} NMR (75 MHz; CDCl3; 298 K): δ 153.6 (1C, CHdN);
148.0 (1Cq, C2-CHN); 146.8 (1Cq, Cipso Ar); 142.5 (1Cq, C4-NO2);
141.5 (1C. CH Py 5); 137.5 (2Cq, C-CH(CH3)2); 125.8 (1C, CH
Ar p); 123.5 (2C, CH Ar m); 121.3 (1C, CH Py 3); 119.4 (1C, CH
Py 6); 28.3 (2C, CH(CH3)2); 23.8 (4C, CH(CH3)2). IR (Nujol): ν
) cm-1 1630 (CdN); 1608 (CdN); 1343 (NO2); 1274 (N-O).
UV/vis (CH2Cl2, 10-4 M, λ ) nm; ꢀ ) 10-4 mol-1 · L · cm-1):
342 (0.863); 278 (1.558); 221 (3.199). MS (ESI; MeOH) m/z: 328.2
[M + H]+; 350.2 [M + Na]+; 366.2 [M + K]+. Anal. Calcd (%)
for C18H21N3O3 (327.37 g/mol): C 66.04; H 6.47; N 12.84. Found:
C 65.84; H 6.51; N 12.70.
4-Methoxy-2-[N′-(2,6-diisopropylphenylimino)car-
baldimine]pyridine N-Oxide (6). Adapted from the literature.26
Under an inert atmosphere, to a solution of 4-nitro-2-iminometh-
ylpyridine N-oxide (5, 500 mg, 1.5 mmol) in dry methanol (5 mL)
was added dropwise a solution of MeONa (100 mg, 1.5 mmol) in
dry methanol. The mixture was heated for 1 h at 40 °C, cooled,
and hydrolyzed with water (20 mL). Extraction with CH2Cl2
afforded the product as a yellow solid (338 mg, 72%). Mp ) 108.8
1
°C. H NMR (300 MHz; CDCl3; 298 K): δ 8.91 (s, 1H, CHdN);
8.13 (d, 3JHH ) 7.2Hz, 1H, CH Py 2); 7.68 (d, 4JHH ) 3.3 Hz, 1H,
3
CH Py 3); 7.14 (m, 3H, Ar-p and -o); 6.91 (dd, JHH ) 7.2Hz;
4JHH ) 3.3 Hz, 1H, CH Py 5); 3.94 (s, 3H, O-CH3); 2.88 (m, 2H,
CH(CH3)2); 1,16 (d, 12H, CH(CH3)2). 13C{1H} NMR (75 MHz;
CDCl3; 298 K): δ 157.51 (1C, Cq-OMe Py 4); 155.63 (1C, CHdN);
148.43 (1C, Cq-CHN Py 2); 145.94 (1C, Cipso Ar); 141.12 (1C,
CH Py 6); 137.47 (2C, Cq-CH(CH3)2); 125.27 (1C, CH Ar-p);
123.39 (2C, CH Ar-o); 115.18 (1C, CH Py 5); 108.02 (1C, CH Py
3); 56.57 (1C, O-CH3); 28.26 (2C, CH(CH3)2); 23.74 (2C,
CH(CH3)2). IR (Nujol): ν 1619 (CdN); 1591 (CdN); 1298 (OMe);
1193 (N-O). UV/vis (CH2Cl2, 10-4 M, λ ) nm, ꢀ ) 10-4
mol-1 · L · cm-1): 388 (0.372); 293 (1.422); 243 (2.697); 226
(3.267). MS (ESI; MeOH) m/z: 313.2 [M + H]+; 335.2 [M + Na]+;
351.2 [M + K]+. Anal. Calcd (%) for C19H24N2O2 (312.40 g/mol):
C 73.05; H 7.74; N 8.97. Found: C 73.19; H 8.10; N 8.39.
General Procedure for NiBr2(PymNox) Complexes. To a
solution of (dme)NiBr2 (308 mg, 1 mmol) in CH2Cl2, quickly
filtered to a cooled Schlenk (-20 °C), was added dropwise the
corresponding ligand (1.1 mmol). After stirring at room temperature
for 30 min the mixture was concentrated and the product was
precipitated by addition of hexane. Filtration and washing with
hexane (3 × 15 mL) afforded the product as a powder.
(H,H,OMe)NNO-NiBr2, Ni6. The reaction of 6 (312 mg, 1.1
mmol) with NiBr2(dme) afforded the product Ni6 as a brown
powder (451 mg; 85%). IR (Nujol, ν cm-1): 3318, 1637, 1613,
1489, 1310, 1191, 1165, 1107, 1022, 944, 796. UV/vis (CH2Cl2,
10-4 M, λ ) nm, ꢀ ) 10-4 mol-1 · L · cm-1): 331 (0.69); 283

4722 Organometallics, Vol. 27, No. 18, 2008
Brasse et al.
Table 4. Polymer Yields and Activities for Ethylene-Methylacrylate
radiation λ(Mo KR1) ) 0.71073 Å, by means of ω and ꢀ scans
with a width of 0.30 and an exposure times of 10 to 30 s per frame,
in the range 3.58° < 2θ < 61.20°, with a detector distance of 37.5
mm. The data were reduced (SAINT) and corrected for Lorentz--
polarization effects and absorption by multiscan method applied
by SADABS.32,33 The structure was solved by direct methods (SIR-
2002)34 and refined against all F2 data by full-matrix least-squares
techniques (SHELXTL).35 All the non-hydrogen atoms were refined
with anisotropic displacement parameters. The hydrogen atoms were
included from calculated positions and refined riding on their
respective carbon atoms with isotropic displacement parameters.
Crystal data for 1: C19H24N2O, Mr ) 296.40, colorless block
crystal (0.28 × 0.15 × 0.15 mm3) from chloroform; triclinic, space
Copolymerization
Ni,
comon. addn. reaction
polym
entry µmol Al/Ni
time, mina
time, h yield, mg activ.b
1
2
3
4
4
4
10
10
1000
1000
800
3
3
1
1
2
2
3
30
30
650
800
4
4
22
3
800
24
a Delay for comonomer introduction after MMAO addition. b kg/mol
Ni · h.
(1.06); 240 (2.74). Magnetic moment µc ) 3.22 µB. MS (ESI,
CH2Cl2 + MeOH) m/z: 763.1 [NiBrL2]+; 451 [NiBrL]+; 313.2 [L
+ H]+. Anal. Calcd (%) for C19H24Br2N2NiO2 (530.90 g/mol): C
42.98; H 4.56; N 5.28. Found: C 43.13; H 5.08; N 5.44.
j
group P1 (no. 2), a ) 8.5986(6) Å, b ) 8.6704(6) Å, c )
11.4818(7) Å, R ) 97.006(2), ) 101.110(2)°, γ ) 97.648(2)°, V
) 822.85(10) Å3, Z ) 2, Fcalcd ) 1.196 g cm-3, F(000) ) 320, µ
) 0.074 mm-1; 6663 measured reflections, of which 3957 were
unique (Rint ) 0.0257); 204 refined parameters, final R1 ) 0.0473,
for reflections with I > 2σ(I), wR2 ) 0.1626 (all data), GOF )
1.039.
(H,Ph)NNO-PdCl2, Pd4. A mixture of ligand 4 (0.5mmol; 179.2
mg) and PdCl2 (0.5 mmol; 88 mg) in CH3CN was stirred at 40 °C
for 4 h. After removing the solvent under vacuum, the solid was
washed with hexane (2 × 15 mL) and then dried under reduced
pressure. Pd4 was obtained as a yellow powder (240 mg, 90%).
1H NMR (300 MHz; CD2Cl2; 298 K): δ 8.06 (dd, 3JHH ) 8.4 Hz,
Crystal data for 2: C18H22N2O, Mr ) 282.38, yellow plate
crystal (0.50 × 0.22 × 0.20 mm3) from hexane; monoclinic, space
group P21/n (no. 14), a ) 10.853(3) Å, b ) 22.624(5) Å, c )
3
4JHH ) 4 Hz, 2H, H o-Ph); 7.98 (t, JHH ) 7.6 Hz, 1H, H Py 4);
3
4
7.93 (dd, JHH ) 7.6 Hz, JHH ) 2 Hz, 1H, H Py 5); 7.89 (s, 1H,
CHN); 7.74 (dd, 3JHH ) 7.6 Hz, 4JHH ) 2 Hz, 1H, Py 3); 7.59 (m,
3H, H m and p Ph); 7.34 (t, 1H, H p Ar); 7.22 (d, 2H, H m Ar);
3.53 (m, 2H, CH iPr); 1.50 (d, 3JHH ) 6.4 Hz, 6H, Me iPr a); 1.19
13.360(3) Å, ) 101.549(6)°, V ) 3214.1(12) Å3, Z ) 8, Fcalcd
)
1.167 g cm-3, F(000) ) 1216, µ ) 0.073 mm-1; 25 438 measured
reflections, of which 5955 were unique (Rint ) 0.0853); 388 refined
parameters, final R1 ) 0.0535 for reflections with I > 2σ(I), wR2
) 0.1494 (all data), GOF ) 0.992.
3
(d, 6H, Me iPr b, JHH ) 6.4 Hz). 13C{1H} NMR (100 MHz;
CD2Cl2; 298 K): δ 156.84 (1C, CHN), 153.40 (1C, Cipso Ph);
145.37 (1C, Cq Py2); 142.06 (1C, Cq Py6); 140.42 (2C, Cq-iPr);
133.71 (1C, CH p Ph); 132.47 (1C, CH Py4); 132.13 (1C, CH Py3);
131.76 (1C, CH Py5); 130.62 (2C, CH o Ph); 129.10 (1C, Cipso
Ar), 128.59 (2C, CH m Ph), 128.58 (1C, CH p Ar), 123.63 (2C,
CH m Ar); 28.96 (2C, CH iPr); 24.31 (2C, Me iPr); 23.09 (2C,
Me iPr). IR (Nujol, ν cm-1): 1624; 1597; 1309; 1232; 1186; 971;
831; 808; 768; 696; 666. Anal. Calcd (%) for C24H26Cl2N2OPd
(535.80 g/mol): C 53.80; H 4.89; N 5.23. Found: C 54.01; H 4.95;
N 5.45.
Crystal data for 3: C22H24N2O, Mr ) 332.43, yellow prism
crystal (0.28 × 0.26 × 0.13 mm3) from diethyl ether/hexane;
triclinic, space group P (no. 2), a ) 7.8275(3) Å, b ) 10.8435(3)
Å, c ) 11.4514(4) Å, R ) 104.8940(10)°, ) 101.5370(10)°, γ
) 100.6490(10)°, V ) 891.24(5) Å3, Z ) 2, Fcalcd ) 1.239 g cm-3
,
F(000) ) 356, µ ) 0.076 mm-1; 22 670 measured reflections, of
which 5385 were unique (Rint ) 0.0223); 230 refined parameters,
final R1 ) 0.0429 for reflections with I > 2σ(I), wR2 ) 0.1219 (all
data), GOF ) 1.049.
Polymerization Procedure. Typical homopolymerization reac-
tion conditions: A Fischer Porter reactor was flushed with N2, and
the catalyst (4 or 10 mmol) was transferred in toluene (50 or 100
mL, respectively) and then charged with ethylene at the working
pressure and temperature. When the pressure and temperature were
stabilized, the reaction was started by addition of MMAO with a
syringe. The temperature was maintained by a water bath and
monitored during the experiment as well as the ethylene consumed.
The experiment time was typically about 20 min. After that time
the reaction was quenched by addition of 1 mL of a HCl/MeOH
solution, and then the polymers were precipitated in a methanol
solution.
Copolymerization Procedure. The reaction was set following
the same protocol employed in homopolymerization, with 5 bar of
ethylene pressure. Methyl acrylate (1 mL, 12 mmol) was injected
with a syringe 1 or 3 min after MMAO. After the specified time
the solvent was removed under vacuum, and the polymers were
washed with aqueous HCl and extracted with CH2Cl2. Solvents were
removed from this polymer solution in the rotary evaporator, and
then the solid polymers were washed with methanol, filtrated, and
finally dried under vacuum. Yields and catalytic activities are shown
in Table 4. The NMR and IR spectra of the polymers are similar
in all cases. One of the polymers (entry 4) was dissolved in CH2Cl2
and reprecipitated with THF in order to check the homogeinity of
the product. The IR and NMR spectra of the sample were not altered
by this procedure.
Crystal data for 4: C24H26N2O, Mr ) 358.47, yellow prism
crystal (0.50 × 0.44 × 0.42 mm3) from hexane; orthorhombic, space
group P212121 (no. 19), a ) 11.3512(5) Å, b ) 13.2373(6) Å, c )
14.0721(6) Å, V ) 2114.46(16) Å3, Z ) 4, Fcalcd ) 1.126 g cm-3
,
F(000) ) 768, µ ) 0.069 mm-1; 24 863 measured reflections, of
which 3175 were unique (Rint ) 0.0329); 248 refined parameters,
final R1 ) 0.0434 for reflections with I > 2σ(I), wR2 ) 0.1140 (all
data), GOF ) 1.031.
Crystal data for 5: C18H21N3O3, Mr ) 327.38, yellow prism
crystal (0.50 × 0.08 × 0.07 mm3) from diethyl ether/hexane;
monoclinic, space group P21 (no. 4), a ) 9.0486(7) Å, b )
6.0887(5) Å, c ) 15.6974(11) Å, ) 99.066(3)°, V ) 854.03(11)
Å3, Z ) 2, Fcalcd ) 1.273 gcm-3, F(000) ) 348, µ ) 0.088 mm-1
;
)
13 186 measured reflections, of which 1918 were unique (Rint
0.0309); 221 refined parameters, final R1 ) 0.0286 for reflections
with I > 2σ(I), wR2 ) 0.0697 (all data), GOF ) 1.037.
Crystal data for 6: C19H24N2O2, Mr ) 312.40, yellow needle
crystal (0.49 × 0.10 × 0.08 mm3) from hexane; monoclinic, space
group P21/n (no. 14), a ) 7.7362(18) Å, b ) 6.0752(17) Å, c )
35.758(9) Å, ) 90.740(14)°, V ) 1680.4(7) Å3, Z ) 4, Fcalcd
)
1.235 g cm-3, F(000) ) 672, µ ) 0.080 mm-1; 20 686 measured
reflections, of which 3807 were unique (Rint ) 0.0861); 214 refined
parameters, final R1 ) 0.0773 for reflections with I > 2σ(I), wR2
) 0.2093 (all data), GOF ) 1.054.
(32) Bruker Apex 2, version 2.1; Bruker AXS Inc.: Madison, WI, 2004.
(33) Bruker. SAINT and SADABS; Bruker AXS Inc.: Madison, WI,
2001.
(34) SIR2002: Burla, M. C.; Camalli, M.; Carrozzini, B.; Cascarano,
G. L.; Giacovazzo, C.; Polidori, G.; Spagna, R. J. Appl. Crystallogr. 2003,
36, 1103.
Crystal Structure Determination. A single crystal of suitable
size, coated with dry perfluoropolyether was mounted on a glass
fiber and fixed in a cold nitrogen stream, 100(2) K, to the
goniometer head. Data collection was performed on a Bruker-
Nonius X8Apex-II CCD diffractometer, using monochromatic
(35) SHELXTL 6.14; Bruker AXS, Inc.: Madison, WI, 2000-2003.

Nickel 2-Iminopyridine N-Oxide Complexes
Organometallics, Vol. 27, No. 18, 2008 4723
Crystal data for Ni1: C40H52Br4Cl4N4Ni2O2, [C38H48Br4N4Ni2O2,
2(CH2Cl2)], Mr ) 1199.72, orange prism crystal (0.31 × 0.30 ×
0.21 mm3) from dichloromethane; monoclinic, space group C2/c
(no. 15), a ) 20.3564(5) Å, b ) 10.6284(3) Å, c ) 21.7557(6) Å,
Acknowledgment. Financial support from the DGI (Project
CTQ2006-05527/BQU), Junta de Andaluc´ıa, and Repsol-
YPF is gratefully acknowledged. We thank Dr. J. Angulo
(IIQ) for recording two-dimensional DOSY spectra of
copolymers.
) 94.3540(10)°, V ) 4693.4(2) Å3, Z ) 4, Fcalcd ) 1.698 g cm-3
,
F(000) ) 2400, µ ) 4.473 mm-1; 34 984 measured reflections, of
which 7153 were unique (Rint ) 0.0256); 253 refined parameters,
final R1 ) 0.0246 for reflections with I > 2σ(I), wR2 ) 0.0636 (all
data), GOF ) 1.024.
CCDC 683748-683754 contain the supplementary crystal-
lographic data for this paper. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
Supporting Information Available: Full tables of bond and
distance angles for the PymNox ligands, along with spectroscopic
data for copolymers. This material is available free of charge via
the Internet at http://pubs.acs.org.
OM800548Y