Ethylene oligomerization and propylene dimerization using cationic (α-diimine)nickel(II) catalysts

Article abstract of DOI:10.1021/om980736t

Nickel-catalyzed ethylene oligomerization and propylene dimerization reactions are described. A series of aryl-substituted α-diimine ligands 1 (ArN=C(R)C(R)=NAr, R,R ≡ 1,8-naphth-diyl, Ar = XC₆H₄, X = p-CF₃ (a), p-H (b), p-Me (c), p-OMe (d), o-Me (e)) and their corresponding Ni(II) dibromide complexes 2 were prepared. Treatment of the Ni(II) dibromide complexes 2 with aluminum alkyl activators such as MAO (methylalumoxane), modified MAO (MMAO), or Et₂AlCl in toluene generates active cationic catalysts in situ that oligomerize ethylene to a Schulz-Flory distribution of linear α-olefins. Reaction conditions can be adjusted that lead to selectivities as high as 96% for linear α-olefins. These catalysts are highly active, with ethylene turnover frequencies as high as 1.4 × 10⁵ mol of C₂H₄/((mol of Ni)h) observed. Schulz-Flory α values range from 0.59 to 0.81 and are dependent on the reaction temperature, ethylene pressure, and nature of the aluminum cocatalyst. The active catalysts dimerize propylene, generating product mixtures that have roughly equal compositions of n-hexenes and 2-methylpentenes. 2,3-Dimethylbutenes are minor products, usually comprising less than 10% of the total product distribution. Propylene dimers and their associated isomerization products are the only reaction products observed under the standard reaction conditions used. Propylene turnover frequencies as high as 2 × 10⁴ mol of C₃H₆/ ((mol of Ni)h) were observed. Control experiments indicate that terminal olefins are slowly isomerized to internal olefins under propylene dimerization reaction conditions but dimer production is fast relative to product isomerization. Higher olefins such as 1-butene and 4-methyl-1-pentene are dimerized very slowly by these catalysts.

Full text of DOI:10.1021/om980736t

Organometallics 1999, 18, 65-74
65
Eth ylen e Oligom er iza tion a n d P r op ylen e Dim er iza tion
Usin g Ca tion ic (r-Diim in e)n ick el(II) Ca ta lysts
Steven A. Svejda and Maurice Brookhart*
Department of Chemistry, University of North Carolina at Chapel Hill,
Chapel Hill, North Carolina 27599-3290
Received August 31, 1998
Nickel-catalyzed ethylene oligomerization and propylene dimerization reactions are
described. A series of aryl-substituted R-diimine ligands 1 (ArNdC(R)C(R)dNAr, R,R ≡ 1,8-
naphth-diyl, Ar ≡ XC6H4, X ) p-CF3 (a ), p-H (b), p-Me (c), p-OMe (d ), o-Me (e)) and their
corresponding Ni(II) dibromide complexes 2 were prepared. Treatment of the Ni(II) dibromide
complexes 2 with aluminum alkyl activators such as MAO (methylalumoxane), modified
MAO (MMAO), or Et2AlCl in toluene generates active cationic catalysts in situ that
oligomerize ethylene to a Schulz-Flory distribution of linear R-olefins. Reaction conditions
can be adjusted that lead to selectivities as high as 96% for linear R-olefins. These catalysts
are highly active, with ethylene turnover frequencies as high as 1.4 × 105 mol of C2H4/((mol
of Ni)h) observed. Schulz-Flory R values range from 0.59 to 0.81 and are dependent on the
reaction temperature, ethylene pressure, and nature of the aluminum cocatalyst. The active
catalysts dimerize propylene, generating product mixtures that have roughly equal composi-
tions of n-hexenes and 2-methylpentenes. 2,3-Dimethylbutenes are minor products, usually
comprising less than 10% of the total product distribution. Propylene dimers and their
associated isomerization products are the only reaction products observed under the standard
reaction conditions used. Propylene turnover frequencies as high as 2 × 104 mol of C3H6/
((mol of Ni)h) were observed. Control experiments indicate that terminal olefins are slowly
isomerized to internal olefins under propylene dimerization reaction conditions but dimer
production is fast relative to product isomerization. Higher olefins such as 1-butene and
4-methyl-1-pentene are dimerized very slowly by these catalysts.
In tr od u ction
Olefin dimerization9 and oligomerization10 reactions
are typically carried out using homogeneous late-transi-
tion-metal catalysts, particularly nickel-based catalysts.
A wide variety of ligands have been employed success-
fully with these systems,11-20 and careful tuning of
ligands can be used to vary product composition from
lower oligomers to polymer.21-24 Neutral nickel catalysts
based on phosphorus-oxygen chelates are particularly
Transition-metal-catalyzed oligomerization of ethyl-
ene and dimerization of R-olefins are important pro-
cesses that are used to convert these basic feedstocks
into useful value-added products. Ethylene oligomer-
ization by nickel catalysts is the first step of the Shell
Higher Olefin Process (SHOP).1-4 The SHOP process
is highly selective for producing linear R-olefins in the
C4-C20+ range, which are used as comonomers to
synthesize linear low-density polyethylene and in the
production of lubricants, plasticizers, surfactants, and
detergents.5 The IFP Dimersol process utilizes a cationic
nickel catalyst to nonselectively dimerize and codimerize
propylene and 1-butene; the dimer products are used
as gasoline additives.6-8
(9) Muthukumaru Pillai, S.; Ravindranathan, M.; Sivaram, S. Chem.
Rev. 1986, 86, 353-399.
(10) Skupinska, J . Chem. Rev. 1991, 91, 613-648.
(11) Keim, W. J . Mol. Catal. 1989, 52, 19-25.
(12) Keim, W. New J . Chem. 1987, 11, 531-534.
(13) Keim, W.; Behr, A.; Gruber, B.; Hoffmann, B.; Kowaldt, F. H.;
Ku¨rschner, U.; Limba¨cker, B.; Sistig, F. P. Organometallics 1986, 5,
2356-2359.
(14) Keim, W.; Behr, A.; Limba¨cker, B.; Kru¨ger, C. Angew. Chem.,
Int. Ed. Engl. 1983, 22, 503.
(1) Keim, W.; Kowaldt, F. H.; Goddard, R.; Kru¨ger, C. Angew. Chem.,
Int. Ed. Engl. 1978, 17, 466.
(15) Bonnet, M. C.; Dahan, F.; Ecke, A.; Keim, W.; Schulz, R. P.;
Tkatchenko, I. J . Chem. Soc., Chem. Commun. 1994, 615-616.
(16) Matt, D.; Huhn, M.; Fischer, J .; DeCian, A.; Kla¨ui, W.; Tkatch-
enko, I.; Bonnet, M. C. J . Chem. Soc., Dalton Trans. 1993, 1173-1178.
(17) Kissin, Y. V.; Beach, D. L. J . Polym. Sci., Part A: Polym. Chem.
1989, 27, 147-155.
(2) Keim, W.; Hoffmann, B.; Lodewick, R.; Peuckert, M.; Schmitt,
G.; Fleischhauer, J .; Meier, U. J . Mol. Catal. 1979, 6, 79-97.
(3) Peuckert, M.; Keim, W. Organometallics 1983, 2, 594-597.
(4) Keim, W. Angew. Chem., Int. Ed. Engl. 1990, 29, 235-244.
(5) Vogt, D. Oligomerization of Ethylene to Higher Linear R-Olefins.
In Applied Homogeneous Catalysis with Organometallic Compounds;
Cornils, B., Herrmann, W. A., Eds.; VCH: New York, 1996; Vol. 1, pp
245-258.
(6) Chauvin, Y.; Olivier, H. Dimerization and Codimerization. In
Applied Homogeneous Catalysis with Organometallic Compounds;
Cornils, B., Herrmann, W. A., Eds.; VCH: New York, 1996; Vol. 1, pp
258-268.
(7) Andrews, J . W. Hydrocarbon Process. 1982, 61, 110.
(8) Chauvin, Y.; Gaillard, J . F.; Quang, D. V.; Andrews, J . W. Chem.
Ind. (London) 1974, 375-378.
(18) Stro¨mberg, S.; Oksman, M.; Zhang, L.; Zetterberg, K. Acta
Chem. Scand. 1995, 49, 689-695.
(19) Braunstein, P.; Pietsch, J .; Chauvin, Y.; Mercier, S.; Saussine,
L.; DeCian, A.; Fischer, J . J . Chem. Soc., Dalton Trans. 1996, 3571-
3574.
(20) Desjardins, S. Y.; Cavell, K. J .; J in, H.; Skelton, B. W.; White,
A. H. J . Organomet. Chem. 1996, 515, 233-243.
(21) Ostoja Starzewski, K. A.; Witte, J . Angew. Chem., Int. Ed. Engl.
1985, 24, 599-601.
(22) Ostoja Starzewski, K. A.; Witte, J . Angew. Chem., Int. Ed. Engl.
1987, 26, 63-64.
10.1021/om980736t CCC: $18.00 © 1999 American Chemical Society
Publication on Web 12/15/1998

66 Organometallics, Vol. 18, No. 1, 1999
Svejda and Brookhart
Sch em e 1. r-Diim in e Liga n d s 1a -e a n d Syn th esis of Cor r esp on d in g Nick el(II) Dibr om id e Com p lexes 2a -e
well-developed and are employed in industrial SHOP
reactors.3,4,25 A series of extremely active Ni(II) ethylene
oligomerization catalysts containing dithioacetylaceto-
nate chelating ligands have been reported,26-28 and the
mechanism by which they dimerize propylene has been
investigated.29 Recently, we reported highly active iron-
based catalysts for ethylene oligomerization.30
The strong propensity of late-transition-metal-based
catalysts to undergo â-hydrogen elimination reactions
has largely relegated this class of catalysts to applica-
tions in dimerization or oligomerization processes. We
recently reported the development of cationic Ni(II) and
Pd(II) R-diimine complexes that polymerize ethylene and
R-olefins (eq 1).31,32 The key to the ability of these late-
addition, the effects of the aluminum alkyl cocatalyst
on catalyst activity, catalyst lifetime, and product
composition have been examined. We have discovered
that these sterically unhindered Ni(II) catalysts dimer-
ize propylene and have investigated the effects of
R-diimine ligand electronics coupled with variations in
reaction conditions on the dimer product distribution.
Proposed mechanisms for both ethylene oligomerization
and propylene dimerization are discussed.
Resu lts a n d Discu ssion
A. Ca ta lyst P r ep a r a tion . Extensive studies on the
synthesis and reactivity of a wide variety of d8 R-diimine
complexes have been reported by Elsevier et al.34-38
Using similar synthetic methods, a series of p-aryl-
substituted R-diimine ligands and their corresponding
nickel(II) dibromide complexes were prepared as shown
in Scheme 1. Synthesis and characterization of ligands
1b-e and nickel(II) dibromide complexes 2b-e have
been previously reported.37,39,40 Acenaphthoquinone and
slightly more than 2 equiv of the appropriate aniline
were combined in toluene containing camphorsulfonic
transition-metal catalysts to form polymer lies in the
steric bulk of the R-diimine ligands. The aryl rings of
the R-diimine ligands are approximately orthogonal to
the plane formed by the metal and coordinated nitrogen
atoms. This ring orientation places the bulky substit-
uents in axial sites, which effectively retards the rate
of chain transfer relative to chain propagation. In a
recent communication we reported that when diimine
ligands lacking bulky aryl ortho substituents are used,
the resulting Ni(II)-based systems produce a Schulz-
Flory distribution of ethylene oligomers (eq 2).33 We
(23) Klabunde, U.; Mulhaupt, R.; Herskovitz, T.; J anowicz, A. H.;
Calabrese, J .; Ittel, S. D. J . Polym. Sci., Part A: Polym. Chem. 1987,
25, 1989-2003.
(24) Hirose, K.; Keim, W. J . Mol. Catal. 1992, 73, 271-276.
(25) Keim, W.; Schulz, R. P. J . Mol. Catal. 1994, 92, 21-33.
(26) Abeywickrema, R.; Bennett, M. A.; Cavell, K. J .; Kony, M.;
Masters, A. F.; Webb, A. G. J . Chem. Soc., Dalton Trans. 1993, 59-
68.
(27) Brown, S. J .; Masters, A. F. J . Organomet. Chem. 1989, 367,
371-374.
(28) Cavell, K. J .; Masters, A. F. Aust. J . Chem. 1986, 39, 1129-
1134.
(29) Brown, S. J .; Clutterbuck, L. M.; Masters, A. F.; Sachinidis, J .
I.; Tregloan, P. A. Appl. Catal. 1989, 48, 1-11.
(30) Small, B. L.; Brookhart, M. J . Am. Chem. Soc. 1998, 120, 7143-
7144.
(31) J ohnson, L. K.; Killian, C. M.; Brookhart, M. J . Am. Chem. Soc.
1995, 117, 6414-6415.
(32) Brookhart, M. S.; J ohnson, L. K.; Killian, C. M.; Arthur, S. D.;
Feldman, J .; McCord, E. F.; McLain, S. J .; Kreutzer, K. A.; Bennett,
A. M.; Coughlin, B. E.; Ittel, S. D.; Parthasarathy, A.; Tempel, D. J .
Pat. Appl. WO 96/23010, 1996.
(33) Killian, C. M.; J ohnson, L. K.; Brookhart, M. Organometallics
1997, 16, 2005-2007.
(34) van Asselt, R.; Rijnberg, E.; Elsevier, C. J . Organometallics
1994, 13, 706-720.
(35) van Asselt, R.; Gielens, E. E. C. G.; Ru¨lke, R. E.; Vrieze, K.;
Elsevier, C. J . J . Am. Chem. Soc. 1994, 116, 977-985.
(36) van Asselt, R.; Elsevier, C. J .; Smeets, W. J . J .; Spek, A. L.
Inorg. Chem. 1994, 33, 1521-1531.
(37) van Asselt, R.; Elsevier, C. J .; Smeets, W. J . J .; Spek, A. L.;
Benedix, R. Recl. Trav. Chim. Pays-Bas 1994, 113, 88-98.
(38) van Asselt, R. Ph.D. Thesis, Universiteit van Amsterdam,
Amsterdam, The Netherlands, 1993.
(2)
(39) tom Dieck, H.; Svoboda, M.; Grieser, T. Z. Naturforsch. 1981,
36b, 823-832.
(40) Svoboda, M.; tom Dieck, H. J . Organomet. Chem. 1980, 191,
321-328.
have subsequently undertaken a series of detailed
studies investigating the electronic effects of the R-di-
imine ligand on the oligomerization of ethylene. In

Cationic (R-Diimine)nickel(II) Catalysts
Organometallics, Vol. 18, No. 1, 1999 67
Ta ble 1. Oligom er iza tion of Eth ylen e in 200 m L of
Tolu en e w ith 2/Al Coca ta lysta
of an oligomer distribution with a high Schulz-Flory R
(e.g. R > 0.8) will consist of higher oligomers. The olefins
produced are primarily linear R-olefins, with the re-
mainder consisting of linear 2-alkenes. Branched olefins
are not observed when reaction times are 1 h or less.
Very long reactions (e.g. 3 h or more) with the most
active catalysts occasionally have very small amounts
of branched products present (<2% of total products),
which we believe are formed from reincorporation of
oligomerization products that build to high concentra-
tions after long reaction periods.
The oligomerization mechanism is proposed to be that
shown in Scheme 2 and is similar to that proposed for
polymerization using analogous Ni(II) complexes.31
Evidence suggests the intermediate cationic nickel alkyl
complexes exist as â-agostic species.46-48 Metal migra-
tion along the chain occurs as shown through olefin
hydride species such as III.49 When chain transfer
occurs from species II/III, only R-olefins can be pro-
duced. Internal olefins (2-alkenes) can be produced only
when chain transfer occurs from secondary alkyl agostic
species IV/IV′ (R-olefins could arise from these species
as well). We have suggested that chain transfer may
occur through associative displacement of olefin from
an olefin hydride species such as III. Theoretical
calculations suggest that chain transfer may occur via
a direct â-H migration from the alkyl group to bound
ethylene in a species such as I.50,51 No experimental
evidence is currently available which distinguishes
these possible chain transfer mechanisms.
Some general trends are observed in the data in Table
1, particularly the effects of ethylene pressure and
temperature, which can be rationalized with the mech-
anism shown in Scheme 2. Increasing the ethylene
pressure results in an increase in the selectivity for
R-olefins, as well as an increase in the Schulz-Flory R
constant (entries 1-4, 10-12, Table 1). According to
Scheme 2, an increase in ethylene concentration (pres-
sure) should favor increased selectivity for R-olefins over
internal olefins because the rate of chain transfer from
species II/III relative to the rate of chain isomerization
to species IV/IV′ increases with increasing ethylene
concentration. The fact that the Schulz-Flory R values
decrease with decreasing [C2H4] suggests that the ratio
of chain transfer to propagation (kct/kp) is much greater
for secondary alkyl complexes IV/IV′ than for primary
alkyl species II. This is supported by the observation
that no branched R-olefins are produced and all internal
olefins produced are 2-alkenes (cis- and trans-CH3CHd
CHR). These observations suggest that once metal
migration occurs to give IV/IV′ no ethylene insertion
occurs and no further metal migration occurs. Thus, the
secondary alkyl species IV/IV′ either undergo chain
entry
no.
press. temp time TONb
(atm) (°C) (min)
10-3
×
% R- Schulz-
olefinc Flory R
catalyst
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
2a /MMAO
2a /MMAO
2a /MMAO
2a /MMAO
2a /MMAO
2a /MMAO
2a /MMAO
2b/MMAO
2c/MMAO
2d /MMAO
2d /MMAO
2d /MMAO
2a /MMAO
2a /MAO
1
15
28
56
56
56
56
56
56
15
28
56
28
28
28
28
35
35
35
35
15
55
75
35
35
35
35
35
35
35
35
35
60
60
60
60
60
60
60
60
60
60
60
60
30
30
30
30
22
116
136
113
26
103
74
49
57
81
87
91
94
88
88
92
91
84
88
95
94
94
96
90
0.50
0.59
0.61
0.68
0.71
0.66
0.63
0.70
0.71
0.64
0.67
0.74
0.68
0.68
0.67
0.81
45
102
114
50
65
80
2a /MAO-IP
2a /Et2AlCl
23
51
a
Reaction conditions: entries 1-12, [Ni] ) 1.00 × 10-4 M, 240
equiv of Al/equiv of Ni; entries 13-16, [Ni] ) 2.50 × 10-5 M, 200
b
equiv of Al/equiv of Ni. Turnover number ) mol of ethylene
consumed/mol of Ni catalyst. c Remainder of product is 2-alkenes.
acid (ca. 2-3 mol %), and the mixture was heated to
reflux. Water evolution was observed by collecting the
toluene/water azeotrope using a Dean-Stark trap.
Electron-rich anilines react smoothly under these condi-
tions to form the corresponding diimines, but even at
these temperatures the reaction using electron-poor
4-(trifluoromethyl)aniline is sluggish. Efforts to prepare
the diimine derived from 4-nitroaniline and acenaph-
thoquinone using these reaction conditions proved
unsuccessful. Crude ligands were purified via precipita-
tion from chloroform with pentane, followed by several
pentane washes.
The nickel(II) dibromide complexes 2a -e were pre-
pared from the ligands 1a -e by reaction of a slight
excess of the free diimine ligand with (1,2-dimethoxy-
ethane)nickel(II) bromide in methylene chloride (Scheme
1). The nickel complexes precipitated from solution and
were purified by washing away the 1,2-dimethoxyethane
and excess ligand with diethyl ether. Complexes 2a -e
are air-stable solids and are insoluble in common
organic solvents, including CH2Cl2, CHCl3, Et2O, tolu-
ene, and hydrocarbon solvents. They are sparingly
soluble in 1,2-difluorobenzene.
B. Oligom er iza tion of Eth ylen e. Treatment of
catalyst precursors 2a -d with MAO, modified MAO
(MMAO), MAO-IP,41 or Et2AlCl in toluene generates
active ethylene oligomerization catalysts. The results
of a series of these oligomerization reactions are sum-
marized in Table 1. A Schulz-Flory distribution42-45 of
oligomers is produced, which is characterized by a
constant, R, where R represents the probability of chain
propagation (eq 3). An oligomer distribution with a
(46) Tempel, D. J .; Brookhart, M. Organometallics 1998, 17, 2290-
2296.
(47) McLain, S. J .; Feldman, J .; McCord, E. F.; Gardner, K. H.;
Teasley, M. F.; Coughlin, E. B.; Sweetman, K. J .; J ohnson, L. K.;
Brookhart, M. Polym. Mater. Sci. Eng. 1997, 76, 20.
(48) McLain, S. J .; Feldman, J .; McCord, E. F.; Gardner, K. H.;
Teasley, M. F.; Coughlin, E. B.; Sweetman, K. J .; J ohnson, L. K.;
Brookhart, M. Macromolecules 1998, 31, 6705-6707.
(49) Theoretical studies suggest that such species do not lie in
potential energy minima and are thus transition states for intercon-
version of agostic species.
(50) Musaev, D. G.; Froese, R. D. J .; Morokuma, K. Organometallics
1998, 17, 1850-1860.
(51) Deng, L.; Woo, T. K.; Cavallo, L.; Margl, P. M.; Ziegler, T. J .
Am. Chem. Soc. 1997, 119, 6177-6186.
(3)
small Schulz-Flory R (e.g. R < 0.5) will therefore consist
mainly of dimers and lower oligomers, while the bulk
(41) MAO-IP is a new proprietary product sold by Akzo Nobel.
(42) Flory, P. J . J . Am. Chem. Soc. 1940, 62, 1561-1565.
(43) Schulz, G. V. Z. Phys. Chem., Abt. B 1935, 30, 379-398.
(44) Schulz, G. V. Z. Phys. Chem., Abt. B 1939, 43, 25-46.
(45) Henrici-Olive´, G.; Olive´, S. Adv. Polym. Sci. 1974, 15, 1-30.

68 Organometallics, Vol. 18, No. 1, 1999
Sch em e 2. P r op osed Mech a n ism of Olefin Oligom er iza tion
Svejda and Brookhart
solvent polarity at higher ethylene concentrations52,53
or inhibition due to formation of an inactive (or less
active) five-coordinate intermediate.
Sch em e 3. P r op osed Mech a n ism of Ch a in
Isom er iza tion /Tr a n sfer for Secon d a r y Alk yl
Sp ecies
In addition to ethylene pressure effects, the oligomer-
ization reaction is strongly affected by the temperature.
As the temperature is increased, both selectivity for
R-olefins and the Schulz-Flory constant R decrease
(entries 4-7, Table 1). These results indicate that the
rate of chain transfer relative to the rate of chain
propagation increases with temperature. Since ethylene
solubility in toluene decreases with increasing temper-
ature,54 the loss of selectivity for R-olefins and concomi-
tant decrease in the Schulz-Flory R constant as the
temperature increases could be due in part to lower
concentrations of ethylene in solution. Thus, the same
arguments given for ethylene pressure effects on these
parameters apply to the observed temperature effects.
Furthermore, the first-order chain isomerization reac-
tion of II should show a stronger temperature depen-
dence than the second-order reaction with ethylene
required to effect chain transfer. This same trend is seen
with increased branching of polyethylene as the tem-
perature is increased at constant pressure.31 Turnover
frequencies also increase with temperature (compare
entries 4 and 5), but significant catalyst decomposition
occurs over 60 min when the reaction temperature is
raised above 55 °C (compare entries 6 and 7 with entry
4).
transfer or isomerize back to II, as shown in Scheme 3.
Schulz-Flory R values should track the percentage of
R-olefin produced. Inspection of Table 1 shows that this
trend is followed, with the exception of the oligomer-
ization carried out with Et2AlCl as activator (run 16;
this activator behaves differently from MAO activators,
see below).
Turnover frequencies are also ethylene-dependent,
reaching a maximum value when the ethylene pressure
is near 30 atm. Further increases in ethylene pressure
result in a decrease in the apparent turnover frequency.
The reason for this decrease in turnover frequency is
not understood. Possibilities include a decrease in the
No substantial ligand electronic effects on selectivity
for R-olefins or the Schulz-Flory R constant are ob-
served. There does appear to be a trend of increasing
turnover frequencies as the metal center becomes more
electrophilic. For example, as the para substituents on
(53) Further evidence of turnover frequency dependence on solvent
polarity is that turnover frequencies decrease as the solvent is changed
from chlorobenzene to toluene to a 1/1 mixture of toluene and pentane.
(54) Fogg, P. G. T.; Gerrard, W. Solubility of Gases in Liquids;
Wiley: West Sussex, England, 1991.
(52) Longo, P.; Oliva, L.; Grassi, A.; Pellecchia, C. Makromol. Chem.
1989, 190, 2357-2361.

Cationic (R-Diimine)nickel(II) Catalysts
Organometallics, Vol. 18, No. 1, 1999 69
F igu r e 2. Ethylene flow profiles in oligomerization reac-
tions with various ratios of 2a to Et2AlCl (15 atm of
ethylene, 35 °C).
F igu r e 1. Ethylene flow profiles in oligomerization reac-
tions with 2a and various cocatalysts (400 equiv of Al/equiv
of Ni, 15 atm of ethylene, 35 °C).
runs with 2a /MAO-IP show that the catalyst is still
active after 12 h, although the catalyst activity at this
point is less than 10% of the activity observed 30 min
into the reaction.
the ligand aryl groups are varied from methyl to
trifluoromethyl, turnover frequencies double (entries 4,
8, 9, Table 1). The electron-poor trifluoromethyl sub-
stituent is expected to lead to a more electrophilic metal
center relative to that when more electron-rich substit-
uents are present, which leads to an increased rate of
ethylene insertion, the turnover-limiting step. However,
the methoxy substituent leads to turnover frequencies
that are higher than expected (entries 10-12, Table 1).
The reason for this behavior is not clear, but formation
of Lewis acid-base complexes of the methoxy group
with excess aluminum cocatalyst would be expected to
reduce electron density on the ligand and therefore lead
to a more electrophilic metal center than would nor-
mally be expected.
The choice of aluminum cocatalyst has a significant
effect on the ethylene oligomerization reactions. MAO,
MMAO, MAO-IP, and Et2AlCl are all effective cocata-
lysts. Reactions activated by MAO, MMAO, and MAO-
IP all show similar Schulz-Flory values, but those
activated by Et2AlCl produce oligomer mixtures with a
higher Schulz-Flory R constant (entries 13-16, Table
1).
Catalyst lifetimes were investigated by measuring the
ethylene flow into the reactor with a mass flowmeter.
The lifetimes of catalysts generated by activation of 2a
with various cocatalysts in ethylene oligomerization
reactions are shown in Figure 1. The high ethylene flow
rates indicated during the initial 5-10 min of each
reaction reflect the process of ethylene saturation of the
toluene solvent. Overall, Et2AlCl generates the most
active catalyst over 3 h reaction periods. However,
comparison of entries 13-16 (Table 1) shows that, over
the first 30 min, the MAO-activated catalyst, for ex-
ample, produces more turnovers than the Et2AlCl-
activated catalyst. J udging from the flow profiles (Fig-
ure 1), this is likely due to a more rapid catalyst
activation by MAO and thus more turnovers at early
times. MAO and MMAO cocatalysts produce catalysts
that are most active during the first 20 min of the
reaction, after which time catalyst activity steadily
declines. After 3 h, the catalyst produced from MMAO
is almost completely inactive. MAO-IP generates the
least active catalyst during early reaction times, but this
catalyst shows more constant activity throughout 3 h
reactions than reactions using MAO or MMAO. Longer
The effects of various Al:Ni ratios in ethylene oligo-
merizations were investigated using 2a and Et2AlCl,
and the results are shown in Figure 2. Generally, a
threshold amount of cocatalyst is needed to effectively
activate the precatalyst, presumably to scavenge impu-
rities that may poison the active catalyst. Figure 2
suggests that a period of 15-25 min is required for full
activation of the precatalyst using Et2AlCl, depending
on the Al:Ni ratio used. No similar induction period is
seen when using MAO-type activators.55 The activation
periods when using Et2AlCl as the cocatalyst were found
to be variable and not well-defined. Since turnover
numbers shown in Table 1 were determined by measur-
ing the amount of reaction products formed over set
reaction periods, the combination of variable activation
periods and catalyst deactivation makes turnover num-
ber comparisons between various cocatalysts difficult
and very qualitative at best.
Higher amounts of cocatalyst (e.g. >1000 equiv of Al/1
equiv of Ni) in the ethylene oligomerization reactions
lead to faster degradation of the active catalyst. The
reason for this trend is not understood, but it was
observed with all cocatalysts tested. Cavell and Masters
noted that contamination of Et2AlCl with Et3Al reduced
the activity of cationic Ni catalysts in olefin oligomer-
izations.26 Similarly, Kissin and Beach have reported
that sulfonated nickel ylide ethylene oligomerization
catalysts are reduced to inactive Ni(0) species by Et3-
Al.17 Attempts to use Me3Al to activate dibromide
precatalyst 2d in the presence of ethylene failed to form
an active catalyst. These observations may suggest that
trialkylaluminum impurities in the cocatalyst could in
part be responsible for catalyst degradation.
Adding certain ortho substituents to the aryl rings of
the R-diimine ligands leads to much longer-lived cata-
lysts.56 In addition, the ethylene flow experiments
described above show that lower Al:Ni ratios result in
(55) In addition to the mass flow data, reactions initiated with MAO-
type cocatalysts are exothermic at the start of the reaction, while
reactions initiated with Et2AlCl exhibit maximum exotherms 10-20
min after the reaction start.
(56) Svejda, S. A.; Onate, E.; Gates, D. P.; Brookhart, M. Manuscript
in preparation.

70 Organometallics, Vol. 18, No. 1, 1999
Svejda and Brookhart
Ta ble 2. Oligom er iza tion of Eth ylen e w ith
2a /MMAOa
Turnover frequencies at 2 atm of propylene pressure
increase as the temperature is raised from -15 to 0 °C
but decline as the temperature is raised further (entries
2, 6, and 7, Table 3). We believe this observed decrease
in turnover frequency is due to a lower concentration
of propylene in solution at 28 °C than at 0 °C, since the
solubility of propylene in toluene decreases dramatically
with increasing temperature.54 The observation that
maximum turnover frequency at 2 atm occurs near 0
°C therefore indicates the point of balance between
monomer concentration in solution and the expected
increasing rate of insertion with increasing tempera-
ture. The product distribution is only slightly affected
by temperature changes, but the degree of isomerization
of the reaction products was nearly double for reactions
run at 28 °C as opposed to those run at -15 °C (compare
entries 6 and 7).
entry
no.
time
(min)
TON
%
Schulz-
Flory Rb
× 10-3
R-olefin
1
2
3
4
15
30
60
41
75
113
185
94
93
91
90
0.67
0.67
0.68
0.68
120
Reaction conditions: [Ni] ) 1.00 × 10-4 M, 240 equiv of Al/
a
equiv of Ni, 200 mL of toluene, 56 atm of ethylene, 35 °C.
b
Determined from ratios of 1-tetradecene/1-dodecene GLC areas.
more active and longer-lived catalyst systems. These
results seem to indicate that catalytic intermediates
formed from precursors containing diimine ligands
lacking steric bulk near the metal center could possibly
deactivate through reaction with excess cocatalyst.
The oligomer composition and distribution at various
points during the reaction was investigated by running
a series of ethylene oligomerization reactions catalyzed
by 2a /MMAO using identical reaction conditions but
different reaction times. The results are summarized
in Table 2. Total turnover numbers are not proportional
to reaction time; i.e., turnover frequencies appear to
drop as the reaction progresses, which supports the data
obtained in the monomer flow measurement experi-
ments, indicating significant catalyst decay occurs in
1/2-2 h. Although catalyst performance degrades with
time, selectivity for R-olefins and the Schulz-Flory R
value remain essentially unchanged throughout the
duration of catalysis. Relatively constant R-olefin to
internal olefin ratios indicate no significant isomeriza-
tion of the reaction products is taking place at longer
reaction times, and these data support a previous
control experiment where no observable isomerization
of added 1-nonene was seen.33
C. Dim er iza tion of P r op ylen e. Treatment of pre-
catalysts 2a -e with MMAO or Et2AlCl in toluene
generates active catalysts which dimerize propylene.
The results of a series of propylene dimerization reac-
tions with these catalysts are summarized in Table 3.
Under the reaction conditions described in Table 3,
these catalysts are greater than 99% selective for
production of propylene dimers as determined by GLC.
No trimers or higher oligomers are observed in 30 min
runs. The products referred to in Table 3 are shown in
Scheme 5.
At low propylene concentrations, the turnover fre-
quencies are dependent on the propylene concentration.
As the propylene concentration is increased, the turn-
over frequency becomes independent of propylene con-
centration (entries 1-5, Table 3). This behavior may be
due to an equilibrium between an alkyl agostic resting
state species and an alkyl olefin resting state species
at lower propylene concentrations. This equilibrium
would shift in favor of the alkyl olefin species at higher
propylene concentrations. The turnover-limiting step is
believed to be migratory insertion of the alkyl olefin
complex; therefore, the observed rate of dimerization
should be independent of propylene concentration when
the sole resting state species is the alkyl olefin species.
The dimer product distribution is essentially unaffected
by changes in the propylene concentration, although the
degree of isomerization of the reaction products de-
creases with increasing propylene concentration.
Turnover frequencies increase as the electrophilic
nature of the catalyst increases (entries 1, 8-10, Table
3). Dimerizations catalyzed by 2a /MMAO proceed at
more than twice the rate as those catalyzed by 2d /
MMAO over the first 30 min following catalyst activa-
tion. The dimer product distribution is relatively inde-
pendent of electronic effects, but the degree of product
isomerization increases as the catalyst becomes more
electrophilic.
The nature and quantity of aluminum cocatalyst
affects the outcome of the propylene dimerization reac-
tions. For example, use of Et2AlCl as the cocatalyst
results in turnover frequencies that are 1 order of
magnitude lower than when MMAO is used (compare
entries 2 and 11, Table 3). Since long induction periods
were observed when Et2AlCl was used as a cocatalyst
in ethylene oligomerizations at 35 °C, it is likely that
the apparent low activity of 2a /Et2AlCl in dimerizing
propylene is due to an even longer induction period,
since this reaction was run at 0 °C. While the overall
product distribution is essentially unchanged, the de-
gree of isomerization observed when Et2AlCl is used is
substantially lower. Varying the amount of MMAO used
led to different turnover frequencies; 1300 equiv of Al/
equiv of Ni was found to be the optimal ratio of
cocatalyst to precatalyst for maximizing turnover fre-
quency (entries 1, 12-13, Table 3). In contrast, lower
cocatalyst:precatalyst ratios led to optimal turnover
frequencies in the ethylene oligomerization reactions
(see above).
The isomerization of dimer products was probed in a
control experiment in which 1-pentene was added to a
propylene dimerization reaction catalyzed by 2a /MMAO
(Table 4). Over the course of 2 h, 1-pentene was slowly
isomerized to the internal isomers, which suggests that
isomerization of the reaction products takes place to
some extent after chain transfer. Although isomeriza-
tion is taking place, it is slow as compared to the rate
of dimer production.
Even though some isomerization of the dimerization
products takes place during the reaction, some conclu-
sions about the regiochemistry of propylene insertion
can be drawn on the basis of product distributions. A
commonly advanced mechanism for nickel-catalyzed
dimerization assumes a propylene-free nickel hydride

Cationic (R-Diimine)nickel(II) Catalysts
Ta ble 3. Dim er iza tion of P r op ylen e in 50 m L of Tolu en e w ith 2/Al Coca ta lysta
product distribn, mol %b
Organometallics, Vol. 18, No. 1, 1999 71
entry no.
catalyst
press. (atm)
temp (°C)
TOFc
path A
path B
path C
path D
isomerizn productsd
1
2
3
4
5
6
7
8
9
10
11
12
13
14
2a /MMAO
2a /MMAO
2a /MMAO
2a /MMAO
2a /MMAO
2a /MMAO
2a /MMAO
2b/MMAO
2c/MMAO
2d /MMAO
2a /Et2AlCl
2a /MMAOf
2a /MMAOh
2e/MMAO
1.1
2
3
4
5
0
0
0
0
0
4410
14600
19600
19500
19400
2760
19
19
19
18
19
18
21
20
24
20
19
17
14
13
38
41
42
40
43
37
38
39
41
38
40
34
28
13
8
9
8
8
8
8
8
9
9
10
8
8
8
6
21
20
20
23
21
29
18
20
16
22
25
25
29
43
14
11
11
11
9
2
2
-15
28
0
8
10000
3730
15
12
10
10
8
16
21
25
1.1
1.1
1.1
2
1.1
1.1
1.1
0
0
0
0
0
0
3310e
1800
1520
3140g
2640g
230
a
b
Reaction conditions: 1300 equiv of Al/equiv of Ni; 30 min reaction times. Product distribution according to pathways shown in
Scheme 5. c TOF ) turnover frequency (mol of C3H6/((mol of Ni)h)). Percentage of product mixture containing double-bond shift isomers
d
of the dimerization products. e 35 min reaction time. f 870 equiv of Al/equiv of Ni. 60 min reaction time. 1700 equiv Al/equiv of Ni.
g
h
Ta ble 4. Isom er iza tion of 1-P en ten e by 2a /MMAO
u n d er P r op ylen e Dim er iza tion Con d ition sa
Sch em e 5. P r op ylen e Dim er iza tion P a th w a ys
rel mol %b
isomer
5 min
1 h
2 h
1-pentene
trans-2-pentene
cis-2-pentene
100
0
0
79
12
9
71
16
13
a
Reaction conditions: [Ni] ) 4.00 × 10-5 M, 1300 equiv of Al/
b
equiv of Ni, 50.0 mL of toluene, 0 °C, 1 atm of propylene. Areas
relative to an octane internal standard as determined by GLC.
Sch em e 4. Gen er a l Ca ta lytic Cycle for P r op ylen e
Dim er iza tion
intermediate,57,58 which in this case would be (N∧N)-
Ni-H+. As discussed above, it is unlikely such a discrete
hydride intermediate is generated; the proposed general
catalytic dimerization cycle is shown in Scheme 4.
Scheme 5 outlines the specific steps which lead to
formation of various hexene isomers. As shown in
Scheme 5, the isomers formed depend on (1) whether
n-propyl or isopropyl group migration occurs, (2) whether
migration occurs in a 1,2- or 2,1-fashion, and (3) what
the direction of â-H elimination is when two possibilities
exist (routes B and D). On the basis of the analysis of
ethylene oligomerization above and the observation of
branching in ethylene polymerization by similar Ni
catalysts as well as recent observations concerning
cationic palladium alkyl complexes,46 it is likely that
the cationic nickel propyl complexes V and VI as well
as the nickel propyl propylene complexes VII and VIII
partially or fully equilibrate prior to migratory insertion.
Some isomerization of primary dimerization products
takes place (see Table 3). Since 2-methyl-2-pentene is
the major isomerization product and it can arise from
either pathway A or D, the fraction of reaction through
one or both of these pathways may be underestimated
using ratios of primary products. However, since isomer-
ization products account for only 10% of total product,
qualitative conclusions can be drawn from primary
product ratios. Clearly pathway B, involving 2,1-inser-
tions of the n-propyl complex, is the major pathway for
dimerization, accounting for 40% of primary products
observed. Products from migration of an n-propyl group
(both 2,1 and 1,2) are favored over products from
migration of an isopropyl group by ca. 2:1 (paths A +
(57) Mu¨ller, U.; Keim, W.; Kru¨ger, C.; Betz, P. Angew. Chem., Int.
Ed. Engl. 1989, 28, 1011-1013.
(58) Collman, J . P.; Hegedus, L. S.; Norton, J . R.; Finke, R. G.
Principles and Applications of Organotransition Metal Complexes;
University Science Books: Mill Valley, CA, 1987; pp 578-596.

72 Organometallics, Vol. 18, No. 1, 1999
Svejda and Brookhart
Ta ble 6. Dim er iza tion of 4-Meth yl-1-p en ten e by
2a /MMAO in Tolu en e
product distribn, mol %
species
1 h aliquot
2 h aliquot
4-methyl-1-pentene
cis-4-methyl-2-pentene
trans-4-methyl-2-pentene
2-methyl-1-pentene
2-methyl-2-pentene
dimersa
89.6
1.3
1.4
1.4
1.5
4.8
205
86.9
1.6
1.8
1.7
1.9
6.1
270
dimer TON (mol of dimers/
mol of Ni)
a
Includes all C12 alkenes.
but even after 3 h they comprise less than 2% of the
total product mixture. No tetramers or higher oligomers
were detected by gas chromatography. Similar results
were obtained from the dimerization reaction catalyzed
by 2c/MMAO.
D. Dim er iza tion of High er Olefin s. Dimerization
reactions of 1-butene and 4-methyl-1-pentene run under
conditions similar to the propylene dimerization reac-
tions were analyzed by GLC and found to proceed only
very slowly as compared to propylene dimerizations.
1-Butene is dimerized by 2a /MMAO at a rate less than
10% of that observed for propylene dimerizations run
under similar reaction conditions.
The results of dimerization of 4-methyl-1-pentene by
2a /MMAO are listed in Table 6. While small amounts
of dimers are produced, this catalyst system actually
isomerizes 4-methyl-1-pentene at a faster rate than it
dimerizes it. Turnover frequencies are less than 5% of
those measured for propylene dimerizations carried out
under similar conditions. Overall, these results suggest
that uptake of olefin products in ethylene oligomeriza-
tions should occur much more slowly than ethylene, and
this is experimentally borne out by both the observed
Schulz-Flory oligomer distributions and the lack of
branched olefin products.
F igu r e 3. Propylene turnover number (TON) vs time
using 2a /MMAO and 2c/MMAO in toluene at 0 °C.
Ta ble 5. Dim er iza tion of P r op ylen e w ith
2a /MMAOa
dimer product distribn, mol %
reaction
2-methyl- 2,3-dimethyl- isomerizn
time (min) n-hexenes pentenes
butenes
products
20
40
60
90
120
180
44
45
45
46
45
45
47
46
46
45
46
46
9
9
9
9
9
9
12
12
12
12
12
13
a
Reaction conditions: 2.0 × 10-6 mol of 2a , 1300 equiv of Al/
equiv of Ni, 50 mL of toluene, 1.1 atm of propylene, 0 °C.
B: paths C + D). However, since the ratios of propyl to
propylene complexes (VII:VIII) undergoing insertion is
not known (and they may be in equilibrium), it is not
valid to conclude that the barrier to migration of the
n-propyl group is less than that of the isopropyl group.
Overall, 2,1-insertion is favored over 1,2-insertion by ca.
2:1 (paths B + D: paths A + C).
It is also instructive to examine the 2,1:1,2 insertion
ratio as a function of which group, n-propyl or isopropyl,
is migrating. Again, on the basis of ratios of primary
products, the n-propyl group exhibits a ca. 2:1 ratio of
2,1:1,2 insertion (path B:A) while the corresponding
isopropyl group ratio (path D:C) is ca. 2.5:1. Thus, 2,1-
insertion is somewhat preferred for the secondary alkyl
group. This preference may be somewhat greater when
considering that the 2-methyl-2-pentene isomerization
product, no matter which pathway it forms from, either
decreases the B:A ratio or increases the D:C ratio.
The catalyst lifetimes under typical propylene dimer-
ization conditions used to construct Table 3 were
examined by analyzing aliquots of dimerization reac-
tions catalyzed by 2a /MMAO and 2c/MMAO and plot-
ting the total propylene turnover number vs time
(Figure 3). Both catalysts demonstrate relatively steady
activity for the first 40 min following activation, but
subsequently their performance degrades. Following 2
h of reaction time, the catalysts are still functioning,
but slowly. The dimer product distribution remains
constant throughout the duration of the experiment, as
well as the extent to which these products are isomer-
ized (Table 5). During the early periods of the reaction
the sole reaction products are dimers. At longer reaction
times (e.g. after 2 h), small amounts of trimers appear,
Con clu sion s
The findings that sterically unhindered (R-diimine)-
nickel(II) catalysts actively oligomerize ethylene and
dimerize propylene further demonstrate the versatility
of this class of catalysts and illustrate their sensitivity
to steric tuning. The temperature and pressure sensitiv-
ity of these catalysts in ethylene oligomerization reac-
tions allows control of the reaction products through
simple variation of these reaction conditions. Ethylene
oligomer mixtures with longer average chain lengths
can be produced by increasing the ethylene pressure,
lowering the reaction temperature, and using Et2AlCl
as the cocatalyst, while oligomer mixtures with shorter
average chain lengths can be produced by lowering the
ethylene pressure, raising the reaction temperature, and
using MAO or MMAO as the cocatalyst. Lowering the
electron density on the metal center through addition
of electron-withdrawing groups on the ligand produces
more active catalysts but leads to slightly lower selec-
tivity for R-olefins. Catalyst longevity in ethylene oli-
gomerization reactions is dependent upon the nature
and concentration of aluminum cocatalyst, and MAO-
IP was found to be the best cocatalyst for generating
long-lived catalysts.

Cationic (R-Diimine)nickel(II) Catalysts
Organometallics, Vol. 18, No. 1, 1999 73
1
product was isolated as a yellow solid (1.06 g, 80% yield). H
These catalysts are very active propylene dimerization
catalysts and demonstrate some selectivity for linear
products. Maximum turnover frequencies are realized
using electrophilic precatalysts activated with MMAO
at temperatures near 0 °C and propylene pressures
above 2 atm. Slow dimer product isomerization takes
place under typical reaction conditions and can be
minimized by lower temperatures and higher propylene
pressures. The dimer product distribution indicates a
2:1 preference for an initial 1,2-propylene insertion
(versus a 2,1-insertion). Furthermore, a 2:1 preference
for the second propylene inserting in a 2,1-fashion is
observed. Higher olefins such as 1-butene and 4-methyl-
1-pentene are dimerized very slowly relative to propy-
lene, which is further evidence that reincorporation of
higher olefin reaction products in both ethylene oligo-
merization and propylene dimerization reactions is not
favorable.
NMR (CDCl3, 400 MHz, 30 °C): δ 7.94 (d, J ) 8.1 Hz, 2 H),
7.73 (d, J ) 8.1 Hz, 4 H), 7.42 (pseudo t, 2 H), 7.21 (d, J ) 8.1
Hz, 4 H), 6.84 (d, J ) 7.2 Hz, 2 H); 13C NMR (CDCl3, 100 MHz,
30 °C): δ 161.3, 154.5, 142.0, 131.4, 129.6, 128.0, 127.9, 126.8
(q, J C-F ) 3.3 Hz), 126.7 (q, J C-F ) 33.4 Hz), 124.4 (q, J C-F
) 271 Hz, CF3), 124.1, 118.3; 19F NMR (CDCl3, 282 MHz, 20
°C) δ -62.3 (s, CF3). Anal. Calcd for C26H14F6N2: C, 66.67; H,
3.01; N, 5.98. Found: C, 66.69; H, 3.08; N, 5.90.
3
2
1
Syn t h esis of (Ar NdC(An )C(An )dNAr )NiBr 2 (Ar
)
p-CF 3C6H4; 2a ). (DME)NiBr2 (77 mg, 0.25 mmol) and 1a (150
mg, 0.32 mmol) were combined in a Schlenk flask under an
argon atmosphere. CH2Cl2 (20 mL) was added, and the
reaction mixture was stirred at room temperature for 20 h.
The supernatant liquid was removed, and the product washed
with 3 × 10 mL of Et2O and dried in vacuo. The product was
isolated as an olive green powder (141 mg, 82% yield). MS
(FAB): m/z [%] 610, 609, 608, 607, 606, 605 [12, 35, 32, 100,
23, 72, M+ - Br], 529, 528, 527, 526 [6, 15, 13, 35, M+ - Br2],
470, 469, 468 [10, 26, 8, M+ - NiBr2], 450, 449 [17, 52, M+
-
NiBr2F]. Anal. Calcd for C26H14Br2F6N2Ni: C, 45.46; H, 2.05;
N, 4.08. Found: C, 45.31; H, 2.70; N, 3.75.
Exp er im en ta l Section
Gen er a l P r oced u r e for Eth ylen e Oligom er iza tion . A
mechanically stirred 1000 mL Parr autoclave was heated
overnight to 100 °C under vacuum and then cooled to 30 °C
under an ethylene atmosphere. The autoclave was charged
with 198 mL of toluene and MMAO. The autoclave was sealed,
and ethylene was added (100 psig). The solution was stirred
for 10 min, during which time the desired reaction temperature
was established. The ethylene pressure was then released, and
a suspension of the appropriate catalyst precursor (2a -d , 2.0
× 10-5 mol in 2 mL of toluene) was added to the reaction
mixture via cannula. The reactor was then sealed and pres-
surized with ethylene to the desired reaction pressure. The
reaction mixture was stirred under constant ethylene pressure
for 1 h, after which time the pressure was released and the
catalyst quenched with acetone and water. An aliquot of the
reaction mixture was analyzed by GLC to determine the
Schulz-Flory R constant. The integrated areas of the C12 and
C14 oligomers were used to calculate the Schulz-Flory R
constants. The GLC conditions used are as follows: injector
and detector temperatures, 250 °C; oven temperature program,
100 °C/4 min, 8 °C/min ramp, 250 °C/70 min. Oligomer peaks
in the C4-C26 range were resolved under these analytical
conditions. The solvents were removed on a rotary evaporator,
and the residual toluene was removed in vacuo overnight. The
mass of the remaining olefin products was obtained, and lower
olefins lost during workup were calculated using the Schulz-
Flory R constant and a gas chromatograph taken of the olefin
products following workup. R-Olefin selectivity was determined
by comparing the ratios of terminal olefin proton areas to
internal olefin proton areas as measured by 1H NMR spec-
troscopy.
Eth ylen e F low Ra te Mea su r em en ts. A mechanically
stirred 1000 mL Parr autoclave under an ethylene atmosphere
was charged with 198 mL of toluene and the appropriate
aluminum cocatalyst. The reactor was sealed and pressurized
with ethylene to 200 psig, and the temperature was allowed
to equilibrate at 35 °C. Once the system reached equilibrium,
the ethylene flow baseline was established with a mass
flowmeter positioned between the ethylene supply cylinder and
the autoclave. A small flame-dried Schlenk tube under an
argon atmosphere was charged with 2a (1.0 × 10-5 mol, 6.9
mg) and 2 mL of toluene to form a precatalyst slurry. The
reactor was vented, and the precatalyst slurry was transferred
into the reactor via cannula. The reactor was repressurized
to 200 psig with ethylene, and the temperature was main-
tained at 35 °C. Ethylene flow rates were recorded at 5-15 s
intervals throughout the duration of the reaction.
Gen er a l Meth od s. All manipulations of air- and/or water-
sensitive compounds were performed using standard high-
vacuum or Schlenk techniques. Argon was purified by passage
through columns of BASF R3-11 catalyst (Chemalog) and 4 Å
molecular sieves. Solid organometallic compounds were trans-
ferred in an argon-filled Vacuum Atmospheres drybox. 1H, 13C,
and 19F NMR spectra were recorded on Varian XL-400 MHz
or Gemini 300 MHz spectrometers. Chemical shifts are
reported relative to residual CHCl3 (δ 7.24 for 1H), CDCl3 (δ
77.00 for 13C), and CCl3F (δ 0.00 for 19F). The dimer and
oligomer products were analyzed on a Hewlett-Packard 5890A
gas chromatograph using a Supelco Petrocol DH 50.2 capillary
column (50 m, 0.20 mm i.d., 0.50 µm film thickness) and a
flame ionization detector. Ethylene flow measurements were
made with an Omega FMA-869 mass flowmeter and recorded
with an Omega RD-853-T paperless recorder. Elemental
analyses were performed by Atlantic Microlab of Norcross, GA.
Ma ter ia ls. Toluene, diethyl ether, pentane, and methylene
chloride were purified using procedures recently reported by
Pangborn et al.59 Polymer-grade ethylene and propylene were
purchased from National Specialty Gases. Propylene was
passed through a Drierite gas drying jar (6.7 cm × 29 cm
column) before use, while ethylene was used without further
purification. We purchased 7% Al (wt %) and 6.4% Al solutions
of modified methylalumoxane (MMAO) in heptane containing
23-27% isobutyl groups from Akzo Nobel. MAO-IP (13.5 wt
% Al in toluene) was purchased from Akzo Nobel. Diimine
ligands 1b-e and corresponding nickel(II) bromide complexes
2b-e were prepared according to modified literature proce-
dures.31,37,39,40 Acenaphthoquinone, 4-(trifluoromethyl)aniline,
(DME)NiBr2, MAO, Et2AlCl, and 1-pentene were purchased
from Aldrich Chemical Co. and used as received. The hexene
isomer standards were purchased from Aldrich, Lancaster, and
Fluka.
Syn th esis of Ar NdC(An )C(An )dNAr (Ar ) p-CF 3C6H4;
1a ). Acenaphthoquinone (0.514 g, 2.82 mmol) and 4-(trifluo-
romethyl)aniline (1.00 g, 6.20 mmol) were added to 40 mL of
dry toluene in a round-bottom flask equipped with a magnetic
stir bar. Camphorsulfonic acid catalyst (15 mg) was added to
the reaction mixture, and a Dean-Stark trap was attached
to the flask. The solution was refluxed for 5 days. Toluene was
then removed on a vacuum line, leaving an orange solid. The
solid was dissolved in 5 mL of CHCl3, and 40 mL of pentane
was then added to precipitate the diimine. The precipitate was
washed with 2 × 20 mL of pentane and dried in vacuo. The
Gen er a l P r oced u r e for P r op ylen e Dim er iza tion . A dry
Fisher-Porter bottle was charged with a magnetic stir bar,
(59) Pangborn, A. B.; Giardello, M. A.; Grubbs, R. H.; Rosen, R. K.;
Timmers, F. J . Organometallics 1996, 15, 1518-1520.

74 Organometallics, Vol. 18, No. 1, 1999
Svejda and Brookhart
evacuated, and then filled with propylene. Toluene (47.0 mL)
and MMAO (1.4 mL of 7% heptane solution) were added to
the bottle via syringe. The reaction mixture was then cooled
to the reaction temperature while being stirred under constant
propylene pressure. A suspension of the appropriate catalyst
precursor (2a -d , 2.00 × 10-6 mol in 2.0 mL of toluene) and
50 µL of octane (internal standard) were transferred via
cannula to the reaction vessel. The bottle was then sealed and
pressurized with propylene. After a 30 min reaction period,
the reactor was depressurized, and an aliquot of the reaction
mixture was removed for GLC analysis. The GLC conditions
used: injector temperature. 220 °C; detector temperature, 250
°C; oven temperature program, 30 °C/18 min, 20 °C/min ramp,
150 °C/25 min.
P r op ylen e Dim er iza tion Lon gevity Exp er im en ts. The
general procedure described for propylene dimerization was
followed (vide supra). A slurry of precatalyst 2a (1.37 mg, 2.00
× 10-6 mol) or 2c (1.16 mg, 2.00 × 10-6 mol) in 2.0 mL of
toluene was added to MMAO (1.4 mL of 7% solution, 1300
equiv of Al/equiv of Ni) in 46.6 mL of toluene containing 50
µL of octane at 0 °C. The propylene pressure was maintained
at 1.1 atm throughout the duration of the experiment. Aliquots
(ca. 4 mL) of the reaction mixture were removed via cannula
and quantified by GLC (vide supra). The amounts of catalyst,
solvent, and dimer products removed with each aliquot were
determined using the octane internal standard and utilized
in subsequent turnover number calculations.
Dim er iza tion of 1-Bu ten e. Precatalyst 2a (5.0 × 10-6 mol,
3.4 mg) and 50 mL of toluene were combined in a 100 mL
Schlenk flask equipped with a magnetic stir bar. 1-Butene was
purged through the solution for 10 min, and then 1.2 mL of
MMAO (7% in heptane) was added via syringe. The reaction
was allowed to proceed at 25 °C for 75 min under 1 atm of
1-butene pressure. Aliquots were removed at 10 and 75 min,
quenched with acetone, and injected into the GLC (60 °C
isothermal). A mixture of butene dimers was observed. Com-
parison of total dimer GLC areas with propylene dimer
standards shows the turnover frequencies for 1-butene dimer-
ization are less than 10% of those observed for analogous
propylene dimerization reactions.
Dim er iza tion of 4-Meth yl-1-P en ten e. Dry, degassed
toluene (48 mL) was added to a dry, Ar-filled 100 mL Schlenk
flask equipped with a magnetic stir bar. 4-Methyl-1-pentene
(ca. 1 mL) was added to the reaction mixture via syringe,
followed by MMAO (1.5 mL of a 6.42% solution). Precatalyst
2a (2.00 × 10-6 mol, 1.37 mg), octane (52.5 µL, GLC internal
standard) and 2 mL of toluene were added to a flame-dried
Schlenk tube. The precatalyst slurry was transferred into the
reaction flask via cannula. The reaction was allowed to proceed
at 25 °C under a static Ar atmosphere for 2 h. Aliquots of the
reaction mixture were removed after 1 and 2 h, quenched in
acetone, and analyzed by GLC using the procedure described
above for quantitative analysis of propylene dimers. The
product distributions are listed in Table 6. Turnover numbers
based on dimer masses calculated using the 4-methyl-1-
pentene GLC calibration curve are 205 mol of dimers/mol of
Ni (1 h aliquot) and 270 mol of dimers/mol of Ni (2 h aliquot).
An a lysis of P r op ylen e Dim er s. Standard solutions of the
propylene dimers and octane were prepared by adding 2.5, 5.0,
10.0, and 15.0 µL of authentic samples of each compound to
toluene in 10.0 mL volumetric flasks. The solutions were
diluted to 10.0 mL with toluene, and GLC calibration curves
were obtained. Aliquots of the dimerization reaction mixtures
(ca. 5 mL) were removed via cannula and added to 10.0 mL
volumetric flasks containing about 4 mL of acetone. The
aliquots were warmed to 20 °C, diluted to volume with acetone,
and injected into the gas chromatograph. Each dimer was
qualitatively identified by retention time. Retention times for
the dimers under the analytical conditions used are as fol-
lows: 4-methyl-1-pentene, 10.92 min; 2,3-dimethyl-1-butene,
11.55 min; cis-4-methyl-2-pentene, 11.65 min; trans-4-methyl-
2-pentene, 11.85 min; 2-methyl-1-pentene, 13.19 min; 1-hex-
ene, 13.28 min; trans-3-hexene, 14.48 min; cis-3-hexene, 14.57
min; trans-2-hexene, 14.73 min; 2-methyl-2-pentene, 14.96
min; cis-2-pentene, 15.66 min; 2,3-dimethyl-2-butene, 17.59
min. The dimers were quantified by first determining the
volume of the aliquot withdrawn via comparison of the octane
internal standard area with an octane calibration curve and
then using the GLC areas of each propylene dimer versus their
respective standard calibration curves to determine the mass
of each dimer present in the reaction mixture.
1-P en ten e Isom er iza tion Con tr ol Exp er im en t. The
general procedure described for propylene dimerization was
followed (vide supra). A slurry of precatalyst 2a (1.37 mg, 2.00
× 10-6 mol) in 2.0 mL of toluene was added to MMAO (1.5
mL of 6.4% heptane solution, 1300 equiv of Al/equiv of Ni) in
46.5 mL of toluene containing 50 µL of 1-pentene and 50 µL
of octane at 0 °C. The propylene pressure was maintained at
1 atm throughout the duration of the experiment. Aliquots (ca.
5 mL) of the reaction mixture were removed at 5 min, 1 h,
and 2 h after catalyst activation and analyzed by GLC. The
areas of 1-pentene, cis-2-pentene, and trans-2-pentene relative
to an octane internal standard were determined for each
aliquot.
Ack n ow led gm en t. This work was supported by the
National Science Foundation and DuPont. S.A.S. thanks
the United States Air Force Institute of Technology
Civilian Institutions program (AFIT/CI) for sponsorship.
We thank Dr. David S. Frohnapfel for assistance with
400 MHz NMR spectra.
OM980736T