Neutral Ni(II) complexes based on keto-enamine salicylideneanilines active for selective dimerization of ethylene

Article abstract of DOI:10.1016/j.jorganchem.2009.01.017

Single component organonickel(II)phosphino neutral complexes from salicylideneaniline based ligands bearing N-O chelate sites, were synthesized, which were found efficient in dimerizing ethylene selectively to various butene products.

Full text of DOI:10.1016/j.jorganchem.2009.01.017

Journal of Organometallic Chemistry 694 (2009) 1254–1258
Contents lists available at ScienceDirect
Journal of Organometallic Chemistry
journal homepage: www.elsevier.com/locate/jorganchem
Communication
Neutral Ni(II) complexes based on keto-enamine salicylideneanilines active
for selective dimerization of ethylene
Deepak Chandran a, Cheolbeom Bae b, InYong Ahn a, Chang-Sik Ha a, Il Kim a,
*
a Division of Engineering, Pusan National University, Jangjeon-dong, Geumjeong-gu, 609-735 Busan, Republic of Korea
b R&BD Center, Kumho Petrochemical Co., Daejeon, Republic of Korea
a r t i c l e i n f o
a b s t r a c t
Article history:
Single component organonickel(II)phosphino neutral complexes from salicylideneaniline based ligands
bearing N–O chelate sites, were synthesized, which were found efficient in dimerizing ethylene selec-
tively to various butene products.
Received 3 November 2008
Received in revised form 12 January 2009
Accepted 13 January 2009
Available online 20 January 2009
Ó 2009 Elsevier B.V. All rights reserved.
Keywords:
Neutral Ni(II) complexes
Keto-enamine salicylideneaniline ligand
Ethylene oligomerization
Selective dimerization
1. Introduction
2. Results and discussion
Advances in synthesis of zero valent nickel complexes and the
successive development of synthetic protocol for single component
organonickel(II)phosphino neutral complexes from them have at-
tained great academic interests [1–10]. These complexes produced
dramatic changes in their catalytic properties when they are hav-
ing properly designed organic ligands. Also they produced an
opportunity for better understanding of the role of complexes in
catalysis mechanisms. Fabrication of ligands for these single com-
ponent organonickel(II)phosphino neutral complexes starts from
the report of Klein et al. on the first ever acylnickel compound con-
taining three monodentate neutral ligands [2]. It was followed by
methylnickel salicylaldiminato complexes with trimethylphos-
phine minor ligand [3]. Based on such complexes, Grubbs and
coworkers have developed neutral Ni(II) based complexes active
in ethylene polymerization reaction [4]. Modification of this ligand
environment was successfully carried out by Brookhart by intro-
ducing a keto-enamine scaffold [5]. Even though the keto-enamine
structure was found promising, the advances in developing such
new molecules as ligands are fewer in number [5–8]. In this regard
we have developed a new neutral nickel complex (Scheme 1) with
N–O chelates in keto-enamine form, capable to mediate an indus-
trially important selective dimerization reaction of ethylene [9,10].
2.1. Synthesis of the complexes and their characterizations
Ligands were synthesized from triformylphloroglucinol and 2,6-
dialkyl aniline (see Appendix A, Supplementary material, for de-
tails). The ligands showed multiple peaks between d 8.0–8.5 and
12.2–13.6 ppm in 1H NMR spectrum indicating the presence of
possible geometrical isomer a of C3h symmetry and b of Cs symme-
try (Fig. 1) for both 1 and 2. Coupling of these multiplets in 1H–1H
COSY spectra with multiplets at 8.0–8.5 ppm region clarifies struc-
ture of ligands as these couplings are the outcome of interaction
between enamine protons and NH of a keto-enamine scaffold
[11]. The two geometrical isomers of 2 were found in a ratio of
2.9:1 (C3h:Cs) from proton NMR spectrum in CDCl3. A BLYP/DNP
DFT calculation on different forms of compound 2 predicts that
keto-enamine tautomer is more favorable by 11.86 kcal/mol than
enolate-imine and among the keto-enamine tautomers the geo-
metrical isomer 2a is more stable than 2b by about 1.2 kcal/mol.
The complexes 3 and 4 were synthesized from the mono-so-
dium salts of 1 and 2 by treating them with one equivalent of
trans-[NiBr(o-tolyl)(PPh3)2] in benzene (Scheme 1). According to
the NMR spectra of the complexes, the diamagnetic Ni(II) com-
plexes adopt a square planar geometry as in the case of salicylaldi-
mine ligand based Ni complexes [4]. The repeated presence of
multiple peaks between d 8.0–8.5 and 12.2–13.6 ppm in 1H NMR
spectra and the coupling of doublets in these regions in 1H–1H
COSY spectra (Fig. 2) of these complexes reveals that the keto-en-
amine core remains intact even after metallation. In addition the
* Corresponding author. Tel.: +82 51510 2399; fax: +82 51512 8563.
E-mail address: ilkim@pusan.ac.kr (I. Kim).
0022-328X/$ - see front matter Ó 2009 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2009.01.017

D. Chandran et al. / Journal of Organometallic Chemistry 694 (2009) 1254–1258
1255
Scheme 1. Synthesis of complexes 3 and 4.
Single crystal XRD analysis of complex 4a (Fig. 3) provided sub-
stantial evidences to prove that the ligand sustains keto-enamine
structure in the complex. The bond length and bond angle values
obtained from the DFT calculations for keto-enamine and eno-
late-imine forms were compared with that obtained for complex
4a from its crystal structure. The values obtained for bond lengths
O1–C2, C2–C3 and C3–C4 (1.281(5), 1.457(6) and 1.458(7) Å,
respectively) from the crystal data are more close to the values ob-
tained for NH form (1.271, 1.472 and 1.463 Å, respectively) than
that obtained for OH form (1.343, 1.425 and 1.428 Å, respectively)
from DFT calculations. The average C–C bond length 1.45 Å of cen-
tral cyclohexanetrione core is considerably longer than the average
bond length of 1.38 Å found in benzene. Conversion of a keto-en-
amine structure, originated from an imine reaction on trifo-
rmylphloroglucinol, to an enolate-imine and its subsequent
trapping with BF3 ꢀ Et2O had been reported along with its crystal
structure for a tris(N-salicylideneaminoferrocene) type complex
and this gives an excellent reference molecule for comparison
[12]. For this enolate structure, bond length values reported for
C2–C3 and C3–C4 in the core aromatic ring and O1–C2 of the eno-
late are 1.412(3), 1.408(3) and 1.318(3) Å, respectively from the
single crystal XRD data. A comparatively shorter C–O and longer
Fig. 1. The ligands 1a and 2a with the two possible geometric isomers 1b and 2b.
DFT calculations support keto-enamine structure of complex 4 by
an energy factor of 7.45 kcal/mol. As the keto-enamine core re-
mains unchanged, the two geometrical isomers a and b of 4 were
found in a ratio of 2.4:1 from proton NMR spectrum in CDCl3.
Fig. 2. 1H–1H COSY NMR spectrum (CDCl3) of complex 4 showing the coupling between vinylic and NH protons.

1256
D. Chandran et al. / Journal of Organometallic Chemistry 694 (2009) 1254–1258
Fig. 3. ORTEP diagram representing the molecular structure of complex 4. (A) Top view. (B) Side view detailing the PPh3 on Ni. (C) Side view detailing the o-tolyl group on Ni.
The hydrogen atoms and solvent entities are omitted for clarity.
C–C bond lengths of the central C6O3 core for complex 4a justify
the keto-enamine structure of ligands in these complexes.
oxygen of salicylaldiminato Ni(II) complexes. The Ni–P bond is
considerably longer than that of neutral Ni(II) pyridine carboxylate
complexes [13], but slightly longer than neutral salicylaldiminato
Ni(II) complexes [4], suggesting a steric free atmosphere of the tri-
phenylphosphine minor ligand. The central cyclohexanetrione ring
consisting of six carbon atoms can be observed as flat within the
error. It can be the fractional crystallization which produced the
crystals only of 4a exclusively.
It can be seen from Fig. 3 and Table 1 that the minor ligand,
PPh3, occupies the position trans to the bulky 2,6-diisopropyl-ben-
zamino moiety. N(1)–Ni(1)–P(1) bond angle is 170.27(13)° and
hence can be considered as positioned nearly linear. The C(64)–
Ni(1)–O(1), that is the angle between o-tolyl group and O, through
Ni, is more open and hence the bonds making this angle are ar-
ranged more close to be linear (173.67(18)°). The linear orientation
of these angles, which are separated from each other at their inter-
cept by an average angle value of 90.41°, makes the metal center
more planar. The Ni(1)–N(1) and Ni(1)–C(64) bond distances are
similar to those in known nickel complexes [4]. The Ni(1)–O(1)
bond is 0.027 Å longer than that in neutral salicylaldiminato Ni(II)
complexes [4], indicating the sp2 hybridization of this oxygen and
hence a reduced availability of electrons than a sp3 hybridized
The use of multinucleating ligands as a means to impose close
spacial confinement on multimetal centers has been widely docu-
mented for a variety of naturally occurring metalloenzymes [14]
and for catalysts with a range of metal-mediated processes [15–
17]. For polymerization applications, good catalytic performances
have been observed and cooperative effects by the neighboring ac-
tive metal centers have been suggested [16,17]. In this sense, at-
tempts to prepare bi and tri-sodium salts of ligands 1a and 2a
were successful in high yields but the metallations using two and
three equivalent of trans-[NiBr(o-tolyl)(PPh3)2] were found ineffec-
tive based on their FAB-mass spectra. A computational investiga-
tion showed that when a metal moiety is attached at only one
N–O chelate the larger space filled by this metal moiety pushes
the two nearby Ar(iPr)2 away. As a result, it makes the dihedral an-
gle between the central C6O3(CHNH)3 and Ar(iPr)2, i.e., the angle
between the central plane and the edge plane, expanded at one
center while contracting at the other. Accordingly, it reduces avail-
able volume for the two other centers, making any possible new in-
take of metal moiety into the complex difficult. The calculations on
reaction energetics for the formation of mono-, bi-, and trinuclear
complexes, with ꢁ263.8, ꢁ170.3 and ꢁ110.3 kcal/mol, respec-
tively, revealed that the formation of bi- and trinuclear complexes
are energetically more demanding than the mononuclear one. De-
tails of the spatial confinements and related energy calculations
are described in Appendix A, Supplementary material.
Table 1
Selected bond distances and angles from crystallographic data.
Bond lengths (Å)
Bond angles (°)
C1–C2
C2–C3
C3–C4
C4–C5
C5–C6
C6–C1
C2–O1
C4–O2
C6–O3
C7–N1
O1–Ni1
N1–Ni1
C64–Ni1
P1–Ni1
1.420(6)
1.457(6)
1.458(7)
1.436(6)
1.443(7)
1.469(7)
1.281(5)
1.264(6)
1.267(5)
1.311(6)
1.943(3)
1.932(4)
1.899(5)
2.1807(15)
O1–Ni1–C64
O1–Ni1–P1
O1–Ni1–N1
N1–Ni1–P1
N1–Ni1–C64
P1–Ni1–C64
Ni1–O1–C2
Ni1–N1–C7
C1–C2–C3
C2–C3–C4
C3–C4–C5
C4–C5–C6
C5–C6–C1
C6–C1–C2
173.67(18)
89.07(10)
91.73(15)
170.27(13)
93.6(19)
86.25(15)
129.5(3)
123.2(3)
119.6(4)
121.3(4)
117.8(5)
121.9(5)
119.0(4)
120.3(4)

D. Chandran et al. / Journal of Organometallic Chemistry 694 (2009) 1254–1258
1257
2.2. Ethylene oligomerizations
Reaction towards ethylene were carried out with complexes 3
and 4 in combination with methylalumoxane (MAO). Both com-
plexes selectively gave dimerization of ethylene along with a little
amount of trimerization product and trace amount of polymer (Ta-
ble 2). At 30 °C both complexes 3 and 4 gave high butene content
which is above 90% of the total oligomer content. A 96% of total bu-
tene content (Run No. 4 of Table 2) given by complex 4 at 30 °C re-
cords the highest. When temperature of oligomerization was
increased the selectivity towards dimerization was found dimin-
ished. An 83% of total butene content by complex 3 (Run No. 3 of
Table 2), at 70 °C is the lowest recorded value of butene content
of this group. Fig. 4 shows the rate (Ro) versus time plots carried
out at various temperatures. The profiles at 30 °C are stable with
slight decay for both 3 and 4 (curves A and D), suggesting that
the active species are stable under oligomerization conditions at
30 °C. Being less crowded, complex 3 shows higher activity
(5.10 ꢂ 105 g-oligomer/mol-Ni h bar) than that of 4 (4.33 ꢂ 105 g-
oligomer/mol-Ni h bar). Complex 3 and 4 show very high initial
activities at higher temperatures (50 and 70 °C) and then the rate
decreased monotonously, giving a reduced overall activities of
1.22 ꢂ 105 and 1.42 ꢂ 105 g-oligomer/mol-Ni h bar, respectively
at 70 °C. These results indicate the active species become thermally
unstable at high temperature. The low solubility of ethylene mono-
mer and the accumulation of vapor pressure of the oligomer prod-
ucts are another factors to decrease the activity at high
temperatures.
Fig. 4. Ethylene oligomerization profiles obtained by 3 at 30 (a), 50 (b) and 70 °C
(c), and by 4 at 30 (d), 50 (e) and 70 °C (f). Conditions: catalyst = 2.5
[Ni] = 200, PC
¼ 5:5 bar, and toluene = 40 mL.
lmol, [MAO]/
2H4
3.1. General method for synthesis of tris(N-salicylideneaniline)
substituted Ni complex (3 and 4)
Sodium salt of compound 1 (0.15 g, 0.27 mmol), and trans-[Ni-
Br(o-tolyl)(PPh3)2] (0.20 g, 0.27 mmol) in a Schlenk flask were dis-
solved in dry benzene (20 mL) and stirred well at rt for 6 h. The
reaction mixture was filtered using a Schlenk frit. The filtrate
was concentrated to ꢃ3 mL, and added dry pentane (30 mL). An or-
ange-brown solid precipitated from solution and it was isolated by
filtration using a Schlenk frit to yield complex 3 (0.19 g, 73%). 1H
NMR (500 MHz, CDCl3): d 13.18–12.30 (m, 2H, NH), 8.38–8.22
(m, 3H, HC-N), 7.73–7.14 (m, 28H, Ar), 2.31 (s, 18H, CH3), 2.14 (s,
3H, o-CH3). 13C NMR (125 MHz, CDCl3): d 186.1, 185.0, 184.0,
173.7, 168.0, 160.4, 157.8, 157.0, 156.8, 156.4, 149.4, 138.0,
137.4, 137.0, 135.0, 134.1, 133.9, 132.6, 132.3, 132.4, 131.1,
129.7, 129.2, 129.0, 128.4, 127.7, 127.2, 124.9, 124.2, 121.8, 29.0,
2.3. Conclusions
In summary, keto-enamine ligands with varying bulkiness di-
rected towards the chelation sites have synthesized and the oxy-
gen and nitrogen atoms of this molecule were exploited as
chelations for a nickel complex with triphenyl phosphine and o-to-
lyl group as minor ligands. The ligands were proved to retain the
keto-enamine structure even in the complex state. The metal cen-
ter found to have a square planar structure from the single crystal
XRD measurements. When activated with MAO, these complexes
showed high activity towards ethylene giving selective dimeriza-
tion with a minor amount of trimerization products.
18.9. 31P NMR (162 MHz, CDCl3):
d 29.5. Anal. Calc. for
C58H54N3NiO3P: C, 74.85; H, 5.85; N, 4.51. Found: C, 74.70; H,
5.64; N, 4.26%. MS(FAB+): m/z = 840 (M-tolyl).
3. Experimental
Sodium salt of compound 2 (0.19 g, 0.27 mmol), and trans-[Ni-
Br(o-tolyl)(PPh3)2] (0.20 g, 0.27 mmol) in a Schlenk flask were dis-
solved in dry benzene (20 mL) and stirred well at rt for 6 h. The
reaction mixture was filtered using a Schlenk frit. The filtrate
Synthesis of pre-ligands, ligands and their sodium salts are gi-
ven in Appendix A, Supplementary material, with details of their
characterizations.
Table 2
Summary of ethylene oligomerization results.a
Selectivity (%)b
Yield (g)
Ro,avg ꢂ 10ꢁ5
e
Run No.
Complex
Temperature (°C)
¼
¼
C4
C6
P
P
¼
¼
d
d
a=c
C4
a=c
C4
1
2
3
4
5
6
3
3
3
4
4
4
30
50
70
30
50
70
9
83
80
63
87
84
69
1
2
3
1
2
4
7
7.02
6.26
2.12
5.99
6.25
2.13
5.10
4.55
1.22
4.33
4.54
1.42
11
20
9
11
17
7
14
3
3
10
a
Oligomerization conditions: catalyst = 2.5
Determined by GC.
l
mol, [MAO]/[Ni] = 200, PC
¼ 5:5 bar, toluene = 40 mL, and time = 30 min.
2H4
b
c
a
-Olefin.
Sum of internal olefins other than
Average rate of oligomerization in g-oligomer/mol-Ni h bar.
d
e
a
-olefin.

1258
D. Chandran et al. / Journal of Organometallic Chemistry 694 (2009) 1254–1258
was concentrated to ꢃ3 mL, and added dry pentane (30 mL). A yel-
low solid precipitated from solution and it was isolated by filtra-
tion using a Schlenk frit to yield complex 4 (0.21 g, 72%). 1H NMR
(500 MHz, CDCl3): d 12.87–12.56 (m, 2H, NH), 8.4–8.0 (m, 3H,
HC-N), 7.68–7.19 (m, 28H, Ar), 3.20 (sept, 6H, CH), 2.10 (s, 3H, o-
CH3), 1.24 (d, 36H, CH3). 13C NMR (125 MHz, CDCl3): d 186.1,
184.7, 183.1, 172.4, 167.6, 158.2, 157.1, 156.9, 156.2, 144.4,
137.5, 136.4, 135.9, 134.9, 134.0, 133.9, 132.3, 132.2,131.4, 130.1,
128.9, 128.7, 128.4, 127.6, 126.4, 124.1, 123.8, 123.2, 106.8,
105.9, 102.9, 29.8, 28.7, 24.0. 31P NMR (162 MHz, CDCl3): d 29.4.
Anal. Calc. for C70H78N3NiO3P: C, 76.50; H, 7.15; N, 3.82. Found:
C, 76.90; H, 7.24; N, 3.54%. MS(FAB+): m/z = 1099 (M), 1007
(M-tolyl).
Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/
data_request/cif. Supplementary data associated with this article
can be found, in the online version, at doi:10.1016/
j.jorganchem.2009.01.017.
References
[1] Z. Zhang, Synlett 5 (2005) 877.
[2] H.-F. Klein, A. Bickelhaupt, B. Hammerschmitt, Organometallics 13 (1994)
2944.
[3] H-F. Klein, A. Bickelhaupt, Inorg. Chim. Acta 248 (1996) 111.
[4] C. Wang, S. Friedrich, T.R. Younkin, R.T. Li, R.H. Grubbs, D.A. Bansleben, M.W.
Day, Organometallics 17 (1998) 3149.
[5] F.A. Hicks, M. Brookhart, Organometallics 20 (2001) 3217.
[6] J.C. Jenkins, M. Brookhart, Organometallics 22 (2003) 250.
[7] L. Zhang, M. Brookhart, P.S. White, Organometallics 25 (2006) 1868.
[8] M. Okada, Y. Nakayama, T. Shiono, J. Organomet. Chem. 692 (2007) 5183.
[9] R.D. Broene, M. Brookhart, W.M. Lamanna, A.F. Volpe, J. Am. Chem. Soc. 127
(2005) 17194.
Acknowledgements
[10] M. Shiotsuki, P.S. White, M. Brookhart, J.L. Templeton, J. Am. Chem. Soc. 129
(2007) 4058.
[11] J.H. Chong, M. Sauer, B.O. Patrick, M.J. MacLachlan, Org. Lett. 5 (2003) 3823.
[12] Y.-K. Lim, S. Wallace, J.C. Bollinger, X. Chen, D. Lee, Inorg. Chem. 46 (2007)
1694.
[13] S.Y. Desjardins, K.J. Cavell, J.L. Hoare, B.W. Skelton, A.N. Sobolev, A.W. White,
W. Keim, J. Organomet. Chem. 544 (1997) 163.
[14] A.L. Gavrilova, B. Bosnich, Chem. Rev. 104 (2004) 349.
[15] L.E. Breyfigle, C.K. Williams, V.G. Young Jr., M.A. Hillmyer, W. Tolman, Dalton
Trans. (2006) 928.
The authors thank H.J. Lee (KBSI, Jeonju) and R. Thomas
(JNCASR, India) for assistance with X-ray crystallography and dis-
cussion. This work was supported by grants-in-aid for the World
Class University Program (No. R32-2008-000-10174-0) and the Na-
tional Core Research Center Program from MEST (No. R15-2006-022-
01001-0), and the Brain Korea 21 program (BK-21).
Appendix A. Supplementary material
[16] H. Li, L. Li, D.J. Schwartz, M.V. Metz, T.J. Marks, L. Liable-Sands, A.L. Rheingold, J.
Am. Chem. Soc. 127 (2005) 14756.
[17] M.L.H. Green, N.H. Popham, J. Chem. Soc., Dalton Trans. (1999) 1049.
CCDC 690677 contains the supplementary crystallographic data
for 4a. These data can be obtained free of charge from The