Immobilization of bis(imino)pyridyliron(ii) complexes on silica

Article abstract of DOI:10.1021/om000939t

The well-known catalyst precursor 2,6-bis[1-((2,6-diisopropylphenyl)imino) ethyl]pyridine (1) is functionalized by various alkenyl groups to yield the corresponding alkenyl-functionalized 2,6-bis[1-((2,6-diisopropylphenyl)imino) ethyl]pyridine ligand. Rea

Full text of DOI:10.1021/om000939t

74
Organometallics 2002, 21, 74-82
Im m obiliza tion of Bis(im in o)p yr id ylir on (II) Com p lexes
on Silica
Franz A. R. Kaul, Gerd T. Puchta, Horst Schneider, Frank Bielert,
Dimitrios Mihalios, and Wolfgang A. Herrmann*
Anorganisch-chemisches Institut der Technischen Universita¨t Mu¨nchen, Lichtenbergstrasse 4,
D-85747 Garching, Germany
Received November 3, 2000
The well-known catalyst precursor 2,6-bis[1-((2,6-diisopropylphenyl)imino)ethyl]pyridine
(1) is functionalized by various alkenyl groups to yield the corresponding alkenyl-
functionalized 2,6-bis[1-((2,6-diisopropylphenyl)imino)ethyl]pyridine ligand. Reaction of these
novel functionalized bis(imino)pyridyl ligands with iron(II) chloride results in the formation
of precatalysts 5-7, which possess different spacer moieties. The possibility of self-
immobilization has been investigated with respect to the chain length of the alkenyl moiety.
Additionally, experiments to heterogenize these functionalized iron(II) complexes on modified
silica via hydrosilation are presented (8-10). The immobilization has been studied by IR,
and the resulting polymers have been characterized by GPC and static light scattering.
In tr od u ction
polymerization catalysts. A different heterogenization
strategy was applied by Alt et al., who provided met-
allocenes with alkene functionalities and used these
functionalized metallocenes as comonomers in the po-
lymerization process.7 To the best of our knowledge,
there have been no experiments to apply the latter
heterogenization techniques to late-transition-metal
catalysts.
Herein we report the versatile heterogenization tech-
nique to immobilize [2,6-bis[1-(2,6-(diisopropylphenyl)-
imino)ethyl]pyridyl]iron(II) chloride, which exhibits ex-
cellent activity in olefin polymerization after activation
with MAO or MMAO but severely suffers from the
insolubility of the produced polymers, resulting in a
dramatic reactor fouling during the polymerization
process.
Since the discovery of Brookhart1 and Gibson2 that
complexes of late transition metals such as iron and
cobalt are capable of polymerizing R-olefins, the interest
in polymerization catalysts other than metallocenes has
greatly increased. Because of their lower oxophilicity,3
these “post-metallocene catalysts” are well-suited for
application in the copolymerization of R-olefins with
polar monomers such as alkyl acrylates.3,4 Due to
problems (e.g. reactor fouling and an extremely exo-
thermic polymerization process) arising from the ho-
mogeneous nature of the polymerization catalysts, the
application of these polymerization catalysts in gas-
phase or slurry reactions still is comparatively prob-
lematic. Up to now, use of the [bis(imino)pyridyl]iron(II)
complexes in a continuous process has been difficult, as
the polyolefins produced will deposit at the reactor walls
and the stirring device, a problem which might be
overcome by tailoring the polymer morphology via
heterogenization of the homogeneous compounds.
Most often, inorganic support materials such as silica
or alumina,5 and also organic supports such as polysty-
rene and starch,6 have been used to heterogenize soluble
Exp er im en ta l Section
Gen er a l Con sid er a tion s. All experiments were carried out
under an atmosphere of dry, purified nitrogen or argon using
glovebox or standard Schlenk techniques. All solvents were
dried and deoxygenated as described in ref 8 and kept under
argon and molecular sieves (4 Å). NMR spectra were recorded
on J EOL J NM GX-400 and Bruker DPX-400 instruments. The
solvent signals were used for internal calibration. All spectra
were obtained at room temperature unless otherwise stated.
Polymer samples were dissolved in a mixture of 1,2,4-trichlo-
robenzene and C2D2Cl4 and analyzed via NMR spectroscopy
at a temperature of 120 °C. Elemental analyses were per-
formed in the microanalytical laboratory of our institute.
The obtained polymers were analyzed using a combination
of a Waters 150 CV high-temperature GPC at 135 °C with
1,2,4-trichlorobenzene as solvent and a modified Whyatt
Minidawn light-scattering instrument. The calibration was
performed with polystyrene standards and conversion into PE
and PP calibration curves using Mark-Houwink parameters.
(1) (a) Small, B. L.; Brookhart, M.; Bennett, A. M. A. J . Am. Chem.
Soc. 1998, 120, 4049-4050. (b) Small, B. L.; Brookhart, M. J . Am.
Chem. Soc. 1998, 120, 7143-7144.
(2) (a) Britovsek, G. J . P.; Gibson, V. C.; Kimberley, B. S.; Maddox,
P. J .; McTavish, S. J .; Solan, G. A.; White, A. J . P.; Williams, D. J .
Chem. Commun. 1998, 849-850. (b) Britovsek, G. J . P.; Gibson, V.
C.; Kimberley, B. S.; Maddox, P. J .; McTavish, S. J .; Solan, G. A.; White,
A. J . P.; Williams, D. J .; Bruce, M.; Mastroianni, S.; Redshaw, C.;
Stro¨mberg, S. J . Am. Chem. Soc. 1999, 121, 8728-8740. (c) Britovsek,
G. J . P.; Gibson, V. C.; Wass, D. F. Angew. Chem. 1999, 111, 448-
468; Angew. Chem., Int. Ed. 1999, 38, 428-447.
(3) J ohnson, L. K.; Mecking, S.; Brookhart, M. J . Am. Chem. Soc.
1996, 118, 267-268.
(4) Younkin, T. R.; Connor, E. F.; Henderson, J . I.; Friedrich, S. K.;
Grubbs, R. H.; Bansleben, D. A. Science 2000, 287, 460-462.
(5) (a) J ackson, R.; Ruddlesden, J .; Thompson, D. J .; Whelan, R. J .
Organomet. Chem. 1977, 125, 57-62. (b) Kaminsky, W. Macromol.
Chem. Phys. 1996, 197, 3907-3945 and references therein.
(6) Boussie, T. R.; Coutard, C.; Turner, H.; Murphy, V.; Powers, T.
S.; Angew. Chem. 1998, 110, 3472-3475.
(7) Alt, H. G. J . Chem. Soc., Dalton Trans. 1999, 1703-1709.
(8) Pangborn, A. B.; Giradello, M. A.; Grubbs, R. H.; Rosen, R. K.;
Timmers, F. J . Organometallics 1996, 15, 1518-1520.
10.1021/om000939t CCC: $22.00 © 2002 American Chemical Society
Publication on Web 12/05/2001

[Bis(imino)pyridyl]iron(II) Complexes
Organometallics, Vol. 21, No. 1, 2002 75
3
The infrared spectra of the homogeneous compounds were
recorded on a Perkin-Elmer 1600 FT-IR instrument as KBr
pellets. The supported precatalyst samples were prepared
under an inert atmosphere as self-supporting pellets and
analyzed using NaCl windows. All BET surface determinations
and BJ H pore-size distributions were carried out using a
Micromeritics ASAP 2000 nitrogen porosimeter.
Allyl bromide, 4-bromo-1-butene, 5-bromo-1-pentene, and
lithium diisopropylamide (2 M in toluene) were purchased from
Aldrich Chemicals and were used without further purification.
The “Karstedt catalyst” (platinum-divinyltetramethyldisilox-
ane complex in vinyl-terminated poly(dimethylsiloxane)) was
obtained from ABCR and used as received.
As a support, silica (Aeroperl 300/40, Degussa-Hu¨ls AG)
with a BET surface area of 300 m2/g, a pore volume of 1.9 cm3/
g, and a main particle size of 40-120 µm was used. Co-
methylaluminoxane (Witco, 10% in heptane) was used as
cocatalyst. Ethene (AGA Gas GmbH, grade 3.5) and propene
(Linde AG, grade 2.8) were purified by passing them through
two purification columns containing activated BTS catalyst
and molecular sieves (4 Å) before feeding the reactor.
7.15-7.18 (m, 4H, Hm aryl), 7.89 (t, 1H, J HH ) 8.2, Hp py),
3
3
8.40 (d, 1H, J HH ) 8.2, Hm py), 8.43 (d, 1H, J HH ) 8.2, Hm
py). 13C{1H} NMR (CDCl3; δ): 17.130 (MeCdN), 17.236 (MeCd
N), 22.177 (CHMe2), 22.893 (CHMe2), 23.223 (CHMe2), 23.547
(CHMe2), 25.857 (CH2), 28.301 (CHMe2), 30.045 (CH2), 34.265
(CH2), 114.820 (CH2dCH), 122.080 (Cm py), 122.814 (Cm py),
123.006 (Cp aryl), 123.483 (Cm aryl), 123.628 (Cm aryl), 135.558
(Cquat), 135.818 (Cquat), 136.997 (Cp py), 137.944 (CHdCH2),
145.951 (Cquat), 146.367 (Cquat), 154.645 (quat), 155.090 (Cquat),
166.991 (CdN), 168.922 (CdN). IR (cm-1): ν˜ 3062.7 m, 2980.3
s, 2923.1 m, 2867.2 m, 1641.9 s, 1568.4 m, 1455.8 s, 1434.7 m,
1363.6 m, 1325.6 m, 1254.5 w, 1237.5 m, 1193.3 m, 11.09.8
m, 1077.3 w, 994.0 w, 910.6 w, 829.5 w, 768.4 s, 688.1 w. MS
(m/z:) 535.5 (M+), 481.5 (M+ - butenyl), 466.5 [M+ - (butenyl,
Me)].
Syn th esis of 2-[1-(2,6-(Diisopr opylph en yl)im in o)eth yl]-
6-[1-(2,6-(d iisop r op ylp h e n yl)im in o)-6-h e p t e n yl]p yr i-
d in e (4). In an analogous manner 1.65 g (72.4%) of 4 could be
obtained as a yellow powder from 2 g (4.15 mmol) of 2,6-bis-
[1-(2,6-(diisopropylphenyl)imino)ethyl]pyridine, 2.2 mL (4.4
mmol, 1.05 equiv) of lithium diisopropylamide (2 M in toluene),
and 1.0 mL (8.3 mmol, 2 equiv) of 5-bromo-1-pentene. Mp: 155
°C. Anal. Calcd for C38H51N3: C, 83.06; H, 9.29; N, 7.65.
Found: C, 82.63; H, 9.46; N, 7.42. 1H NMR (CDCl3; δ (J , Hz)):
1.12-1.21 (m, 24H, CHMe2), 1.29-1.38 (m, 2H, CH2), 1.50-
1.55 (m, 2H, CH2), 1.88-1.93 (m, 2H, CH2), 2.24 (s, 3H, MeCd
N), 2.66-2.86 (m, 2H, CH2), 2.72-2.80 (m, 4H, CHMe2), 4.80-
4.84 (m, 2H, CH2dCH), 5.58-5.68 (dq, 1H, CHdCH2), 7.06-
7.12 (m, 2H, Hp aryl), 7.14-7.18 (m, 4H, Hm aryl), 7.91 (t, 1H,
3J HH ) 7.7, Hp py), 8.41 (d, 1H, 3J HH ) 7.7, Hm py), 8.48 (d, 1H,
3J HH ) 7.7, Hm py). 13C{1H} NMR (CDCl3; δ): 17.232 (MeCd
N), 22.188 (CHMe2), 22.897 (CHMe2), 23.224 (CHMe2), 23.537
(CHMe2), 26.055 (CH2), 28.310 (CHMe2), 29.523 (CH2), 30.479
(CH2), 33.393 (CH2), 114.536 (CH2dCH), 122.042 (Cm py),
122.797 (Cm py), 123.008 (Cp aryl), 123.452 (Cm aryl), 123.616
(Cm aryl), 135.540 (Cquat), 135.806 (Cquat), 136.978 (Cp-Py),
138.351 (CHdCH2), 146.016 (Cquat), 146.408 (Cquat), 154.680
(Cquat), 155.062 (Cquat), 166.978 (CdN), 169.492 (CdN). IR
(cm-1): ν˜ 3062.8 m, 2960.7 s, 2927.1 s, 2866.7 s, 1635 s (ν CdN),
1588.6 w, 1568.6 w, 1456.9 s, 1435.2 s, 1382.0 w, 1362.8 s,
1325.4 m, 1254.4 m, 1236.6 m, 1192.5 m, 1109.3 s, 1043.1 w,
994.0 w, 911.2 w, 818.4 m, 765.0 s, 684.7 w, 529.8 w. MS (m/
z): 549.8 (M+), 534.7 (M+ - Me), 481.6 (M+ - pentenyl), 466.5
[M+ - (pentenyl, Me)].
Syn th esis of 2-[1-(2,6-(Diisopr opylph en yl)im in o)eth yl]-
6-[1-(2,6-(d iisop r op ylp h e n yl)im in o)-4-p e n t e n yl]p yr i-
d in e (2). To a vigorously stirred solution of 2 g (4.15 mmol) of
2,6-bis[1-(2,6-(diisopropylphenyl)imino)ethyl]pyridine (1) in 60
mL of absolute THF was added 2.2 mL (1.05 equiv) of lithium
diisopropylamide (2 M in toluene) via syringe. Immediately a
change in color from yellow to dark green-black was observed.
The reaction mixture was heated under reflux for 12 h, and 6
equiv of allyl bromide (24.9 mmol, 3.0 g, 2.15 mL) was added.
After 30 min a brightening of the solution was observed and
the reaction mixture was heated to 60 °C overnight. Removal
of the solvent and of the excess allyl bromide by evaporation
resulted in a pale yellow powder, which was extracted with
pentane (3 × 20 mL). The solution was concentrated and stored
at room temperature for 24 h and was afterward filtered again.
Compound 2 was isolated by removal of the solvent in vacuo
as 1.4 g of a yellow powder in 65% yield. Mp: 160 °C. Anal.
Calcd for C36H47N3: C, 82.92; H, 9.02; N, 8.06. Found: C, 82.96;
H, 8.55; N, 8.03. 1H NMR (CDCl3; δ (J , Hz)): 1.13-1.22 (m,
24H, CHMe2), 2.24 (s, 3H, MeCdN), 2.27-2.31 (m, 2H, CH2),
2.73-2.80 (m, 6H, CH2 + CHMe2), 4.83-4.89 (m, 2H, CH2d
C), 5.66-5.73 (dq, 1H, CH2dCH), 7.02-7.12 (m, 2H, Hp aryl),
7.15-7.18 (t, 4H, Hm aryl), 7.92 (t, 1H, 3J HH ) 7.7, Hp py), 8.40
3
3
Syn th esis of 2-[1-(2,6-(Diisopr opylph en yl)im in o)eth yl]-
6-[1-(2,6-(d iisop r op ylp h en yl)im in o)-4-p en ten yl]p yr id yl-
ir on (II) Ch lor id e (5). A 200 mg portion of ligand 2 (0.38
mmol) was dissolved in 20 mL of THF, and 0.95 equiv (72 mg)
of FeCl2‚4H2O was added with continuous, vigorous stirring.
The blue solution was stirred at room temperature for 20 h.
Afterward, compound 5 was precipitated with 30 mL of
pentane. The solution was filtered off and the remaining blue
solid washed with pentane and toluene and thoroughly dried
in vacuo. This synthesis afforded compound 5 as a blue powder
in 83% yield (194 mg). Mp: >250 °C dec. Anal. Calcd for
C36H47N3FeCl2: C, 66.69; H, 7.26; N, 6.48. Found: C, 66.21;
H, 7.26; N, 6.26. 1H NMR (CD2Cl2; δ): -39.5 (s, 3H, MeCd
N), -26.5 (br, 2H, CHMe2), -21.7 (br, 2H, CHMe2), -11.4 (s,
1H, Hp aryl), -10.5 (s, 1H, Hp aryl), -9.2 (br, 6H, CHMe2),
-6.8 (s, 6H, CHMe2), -6.6 (s, 6H, CHMe2), -5.1 (s, 6H,
CHMe2), -4.4 (br, 2H, CH2), 4.0 (s, 1H, CHdCH2), 4.8 (s, 1H,
CHdCH2), 6.1 (br, 1H, CHdCH2), 6.4 (br, 2H, CH2), 14.0 (s,
2H, Hm aryl), 15.4 (s, 2H, Hm aryl), 79.5 (s, 1H, HPy), 83.2 (s,
1H, HPy), 86.4 (s, 1H, HPy). IR (cm-1): ν˜ 3062.3 w, 2962.9 s,
2925.8 m, 2867.0 m, 1624.4 w, 1577.1 s, 1465.0 s, 1383.7 m,
1363.1 m, 1325.1 m, 1269.6 m, 1203.9 m, 1186.4 m, 1104.0 m,
1056.9 w, 936.7 m, 816.0 m, 794.5 m, 768.2. MS (m/z): 647.0
(M+), 612.1 (M+ - Cl), 577.0 (M+ - 2Cl), 522.0 (M+ - FeCl2).
Syn th esis of 2-[1-(2,6-(Diisopr opylph en yl(im in o)eth yl]-
6-[1-(2,6-(diisopr opylph en yl)im in o)-5-h exen yl]pyr idylir on -
(II) Ch lor id e (6). With ligand 3 (200 mg; 0.37 mmol) as the
(d, 1H, J HH ) 7.7, Hm py), 8.45 (d, 1H, J HH ) 7.7, Hm py).
13C{1H} NMR (CDCl3; δ): 17.238 (MeCdN), 22.220 (CHMe2),
22.904 (CHMe2), 23.217 (CHMe2), 23.515 (CHMe2), 28.316
(CHMe2), 29.945 (CH2), 30.854 (CH2), 114.699 (CH2dCH),
122.097 (Cm py), 122.853 (Cm py), 123.006 (Cp aryl), 123.544
(Cm aryl), 123.624 (Cm aryl), 135.487 (Cquat), 135.800 (Cquat),
137.044 (Cp py), 137.786 (CHdCH2), 145.515 (Cquat), 146.380
(Cquat), 154.911 (Cquat), 155.089 (Cquat), 166.968 (CdN), 168.812
(CdN). IR (cm-1): ν˜ 3062.3 m, 2960.9 s, 2926.2 s, 2867.1 m,
1633.9 s, 1568.7 m, 1454.8 s, 1435.7 s, 1382.0 m, 1363.7 s,
1325.7 m, 1254.1 m, 1238.0 m, 1191.4 m, 1109.8 s, 1076.8 w,
1058.1 w, 1044.3 w, 994.5 w, 935.5 w, 912.5 m, 820.3 m, 780.3
m, 758.3 s, 689.5 w, 531.5 w. MS (m/z): 521 (M+), 506.4 (M+
- Me), 478.5 (M+ - allyl).
Syn th esis of 2-[1-(2,6-(Diisopr opylph en yl)im in o)eth yl]-
6-[1-(2,6-(d iisop r op ylp h en yl)im in o)-5-h exen yl]p yr id in e
(3). By using the procedure described above, 1.75 g (79%) of 3
was obtained as a yellow powder with 2 g (4.15 mmol) of 2,6-
bis[1-(2,6-(diisopropylphenyl)imino)ethyl]pyridine (1), 2.2 mL
(4.4 mmol, 1.05 equiv) of lithium diisopropylamide (2 M in
toluene) and 1.0 mL (8.3 mmol, 2.0 equiv) of 4-bromo-1-butene.
Mp: 150 °C. Anal. Calcd for C37H49N3: C, 82.99; H, 9.16; N,
7.85. Found: C, 82.66; H, 9.27; N, 7.51. 1H NMR (CDCl3; δ (J ,
Hz)): 1.12-1.21 (m, 24H, CHMe2), 1.57-1.62 (m, 2H, CH2),
1.95-2.00 (m, 2H, CH2), 2.24 (s, 3H, MeCdN), 2.67-2.71 (m,
2H, CH2), 2.74-2.80 (m, 4H, CHMe2), 4.81-4.86 (m, 2H, CH2d
CH), 5.59-5.67 (dq, 1H, CHdCH2), 7.06-7.12 (m, 2H, Hp aryl),

76 Organometallics, Vol. 21, No. 1, 2002
Kaul et al.
starting material and using a procedure analogous to the one
described above, 6 could be obtained as a blue powder in 79%
yield (184 mg). Mp: >250 °C dec. Anal. Calcd for C37H49N3-
FeCl2: C, 67.10; H, 7.40; N, 6.35. Found: C, 66.85; H, 7.23;
1.98%. BET surface: 202 m2/g. Pore volume (BJ H, cumulative,
desorption, 17-1000 Å): 1.59 cm3/g.
Im m obiliza tion of Com p ou n d s 5-7 on Silica (8-10).
A 100 mg portion (0.15 mmol) of 5, 6, or 7 was combined with
500 mg of modified silica in a Schlenk tube. On addition of 6
mL of absolute THF and 0.1 mL of the Karstedt catalyst (3-
3.5 wt % Pt in xylene), the reaction mixture was stirred at 55
°C for 4 days. After that time, the resulting solid was separated
via filtration, thoroughly washed with toluene, and dried in
vacuo. The Fe content on the silica was determined to be 1.15
wt % Fe for precatalyst 8, 1.10 wt % Fe for precatalyst 9, and
0.99 wt % Fe for precatalyst 10.
Eth en e a n d P r op en e P olym er iza tion a n d An a lytica l
P r oced u r es. P olym er iza tion P r oced u r es for Hom oge-
n eou s a n d H et er ogen eou s [Bis(im in o)p yr id yl]ir on (II)
P r eca ta lyst. A 0.5 L Bu¨chi glass autoclave, equipped with a
pressflow gas controller, was filled with approximately 230 mL
of dry and oxygen-free toluene and the appropriate amount of
MMAO (10% in heptane; Al:Fe ) 1000). The autoclave was
pressurized with ethene or propene and thermostated to the
desired reaction temperature. The desired amount of precata-
lyst was dissolved or suspended in dry and oxygen-free toluene,
the precatalyst was transferred into the injection system via
a syringe or a cannula, the injection system was closed, the
solution/suspension was injected into the autoclave, and the
injection system was washed with an additional 5 mL of
toluene. The pressure was kept constant during the polymer-
ization, and the amount of ethene or propene consumed by
the catalyst during the reaction was measured. The polymer-
ization was quenched after 1 h by adding 10 mL of methanol,
and the resulting polymer suspension was poured into 250 mL
of methanol/HCl (3:1) to precipitate the polymer and stirred
for 16 h. After filtration the polymer was dried at 60 °C in
vacuo.
1
N, 5.98. H NMR (CD2Cl2; δ): -41.2 (s, 3H, MeCdN), -26.5
(br, 2H, CHMe2), -22.3 (br, 2H, CHMe2), -11.5 (s, 1H, Hp
aryl), -10.3 (s, 1H, Hp aryl), -9.1 (br, 6H, CHMe2), -6.7 (s,
12H, CHMe2), -5.3 (br, 2H, CH2), -5.0 (br, 6H, CHMe2), -1.5
(br, 2H, CH2), 3.0 (s, 2H, CH2), 5.5 (s, 2H, CH2dCH), 5.9 (s,
1H, CHdCH2), 14.0 (s, 2H, Hm aryl), 15.5 (s, 2H, Hm aryl),
79.3 (s, 1H, HPy), 83.6 (s, 1H, HPy), 85.7 (s, 1H, HPy). IR
(cm-1): ν˜ 3061.4 w, 2962.1 s, 2923.1 m, 2866.8 m, 1577.4 s,
1463.6 s, 1383.8 m, 1368.8 m, 1323.8 m, 1264.0 m, 1204.4 m,
1187.9 m, 1103.8 m, 1057.2 w, 989.7 w, 938.0 w, 916.8 m, 810.2
m, 778.7 m. MS (m/z): 660.8 (M+), 625.9 (M+ - Cl), 609.9 [M+
- (Cl, Me)], 590.1 (M+ - 2Cl), 536.0 (M+ - FeCl2).
Syn th esis of 2-[1-(2,6-(Diisopr opylph en yl)im in o)eth yl]-
6-[1-(2,6-(d iisop r op ylp h en yl)im in o)-6-h ep ten yl]p yr id yl-
ir on (II) Ch lor id e (7). Starting with ligand 4 (200 mg; 0,36
mmol) and using the procedure described above, compound 7
could be obtained as a blue powder in 75% yield (173 mg).
Mp: >250 °C dec. Anal. Calcd for C38H51N3FeCl2: C, 67.48;
1
H, 7.55; N, 6.22. Found: C, 67.02; H, 7.55; N, 5.96. H NMR
(CD2Cl2; δ): -41.1 (s, 3H, MeCdN), -26.1 (br, 2H, CHMe2),
-22.2 (br, 2H, CHMe2), -11.4 (s, 1H, Hp aryl), -10.3 (s, 1H,
Hp aryl), -9.1 (br, 6H, CHMe2), -6.6 (s, 12H, CHMe2), -5.3
(br, 2H, CH2), -4.9 (br, 6H, CHMe2), -1.7 (br, 2H, CH2), 1.8
(br, 2H, CH2), 2.3 (s, 2H, CH2), 5.5 (s, 2H, CH2dCH), 6.1 (s,
1H, CHdCH2), 14.0 (s, 2H, Hm aryl), 15.5 (s, 2H, Hm aryl),
79.2 (s, 1H, HPy), 83.4 (s, 1H, HPy), 85.1 (s, 1H, HPy). IR
(cm-1): ν˜ 3062.3 w, 2962.0 s, 2923.1 m, 2866.7 m, 1578.2 s,
1463.9 s, 1389.8 m, 1389.0 m, 1324.8 m, 1272.3 m, 1204.4 m,
1103.7 m, 1057.1 w, 914.4 m, 809.9 m, 777.4 m. MS (m/z):
675.7 (M+), 640.7 (M+ - Cl), 625.6 [M+ - (Cl, Me)], 605.6 (M+
- 2Cl), 550.7 (M+ - FeCl2).
Resu lts
Self-Im m obiliza tion of Com p ou n d s 5-7 on Silica (8-
10). A 15 mg portion (0.0225 mmol) of 5, 6, or 7 was dissolved
in 20 mL of absolute toluene and cooled to 0 °C. On addition
of 100 equiv of MMAO the color of the solution turned from
deep brown to green-brown. Dry and oxygen-free ethylene was
slowly bubbled through the solution, until the solution turned
opaque and small quantities of polymer began to precipitate.
The reaction vessel was transferred into a glovebox and the
polymer separated via centrifugation, thoroughly washed with
toluene and hexane, and dried in vacuo. The obtained self-
immobilized catalyst was used in olefin polymerization.
Syn t h eses of Alk en e-F u n ct ion a lized Liga n d s.
With the 2,6-bis[1-(2,6-(diisopropylphenyl)imino)ethyl]-
pyridine ligand (1) reported by Brookhart and Gibson1,2
as the starting material, the functionalization is achieved
by deprotonation of the bis(imino)pyridine ligand with
lithium diisopropylamide. Immediately the yellow solu-
tion turns dark green, almost black. After the mixture
is kept under reflux for 12 h, addition of an excess of
allyl bromide to the stirred solution leads to the forma-
tion of the allyl-modified ligand 2 within a reaction time
of 1 day (Scheme 1).
When the reaction is finished, the color of the solution
returns to yellow. In analogous manner the butenyl- (3)
and pentenyl-modified (4) ligands can be prepared in
good yields. The reaction time of the nucleophilic
substitution depends on the alkenyl bromide and in-
creases with increasing chain length, because of the
greater steric hindrance of the longer chains, but using
a greater excess of the bromoalkene accelerates the
reaction.
The modified ligands show an increased solubility
compared with 1, especially in pentane, which is used
to separate the product from the unreacted educt. The
pentenyl-substituted bis(imino)pyridyl ligand is the
most soluble one, followed by the butenyl- and the allyl-
modified bis(imino)pyridyl ligands.
After the preparation of the ligands, the iron(II)
complexes are synthesized according to published meth-
ods1,2 by addition of FeCl2‚4H2O to a stirred solution of
the ligands 2-4 in THF (Scheme 2). The resulting deep
blue iron complexes 5-7 can be precipitated with
pentane.
For NMR experiments the self-immobilized catalyst was
hydrolyzed with methanol and the resulting polymer thor-
oughly washed with methanol, THF, toluene and hexane and
dried in vacuo.
1H NMR (C2D2Cl4, 390 K; δ): 0.87 (s, polymer CH3), 1.1-
1.9 (s, polymer CH2), 1.1-1.9 (s, CHMe2), 1.1-1.9 (s, CH2),
1.1-1.9 (s, CH2), 1.1-1.9 (s, CH2), 2.22 (s, MeCdN), 2.76 (s,
CHMe2), 7.07-7.24 (m, H aryl), 7.91 (t, H py), 8.31-8.52 (m,
H py). 13C{1H} NMR (C2D2Cl4, 390 K; δ): 16.828 (MeCdN),
21.117-25.013 (polymer CH3), 21.117-25.013 (CHMe2), 21.117-
25.013 (CHMe2), 21.117-25.013 (CHMe2), 21.117-25.013
(CHMe2), 25.980 (CH2), 28.532 (CHMe2), 28.885 (CH2), 30.577
(CH2), 29.655-34.234 (polymer CH2), 35.091 (CH2), 121.042-
125.142 (C py, C aryl), 133.841-139.221 (C py, C aryl),
144.992-147.002 (C aryl), 153.553-155.996 (C aryl), 167.287
(CdN).
Mod ifica tion of th e Silica Su p p or t. To a suspension of
2 g of dehydroxylated silica (500 °C, 15 h, 2 × 10-4 mbar) in
20 mL of hexane was added 0.8 g (6 mmol) of tetramethyl-
disilazane dropwise at room temperature. The resulting
suspension was heated to 85 °C for 4 h. The solid was
separated via filtration and washed three times with 20 mL
of hexane. The support was dried in vacuo at room tempera-
ture for 30 h. The carbon content was determined as high as

[Bis(imino)pyridyl]iron(II) Complexes
Organometallics, Vol. 21, No. 1, 2002 77
Ta ble 1. Resu lts of Eth en e P olym er iza tion Ru n s
Sch em e 1. P r ep a r a tion of th e
Alk en yl-F u n ction a lized 2,6-Bis[1-(2,6-
(d iisop r op ylp h en yl)im in o)eth yl]p yr id yl Liga n d s
2-4a
w ith P r eca ta lysts 5-7a
activityb
precatalyst
T (°C) after 5 min
after 45 min
average
5
0
20
40
60
80
4.0 × 104
5.3 × 104
8.5 × 104
9.2 × 104
2.1 × 104
1.8 × 104
2.40 × 104
2.14 × 104
2.08 × 104
1.15 × 104
3.38 × 103
7.7 × 103
0
0
0
6
0
20
40
60
80
4.5 × 105
3.6 × 105
3.5 × 105
2.5 × 105
8.1 × 104
2.6 × 104
1.05 × 105
7.72 × 104
6.90 × 104
2.49 × 104
1.01 × 104
7.2 × 103
0
0
0
7
0
20
40
60
80
2.8 × 105
2.0 × 105
2.2 × 105
2.1 × 105
6.9 × 104
4.7 × 104
1.07 × 105
5.98 × 104
4.13 × 104
3.33 × 104
1.08 × 104
1.4 × 104
0
0
0
a
Reaction conditions: 2.0 bar of ethene pressure; toluene as
b
solvent; cocatalyst MMAO (Al:Fe ) 1000); reaction time 1 h. In
units of kg of PE/((mol of Fe) h bar).
Sch em e 3. Self-Im m obiliza tion of
Alk en yl-F u n ction a lized [Bis(im in o)p yr id yl]ir on (II)
Com p lexes
a
LDA ) lithium diisopropylamide.
Sch em e 2. P r ep a r a tion of th e
Alk en yl-F u n ction a lized [Bis(im in o)p yr id yl]ir on (II)
Com p lexes 5-7
polymerization, decreasing from 0 to 80 °C. Whereas 6
and 7 are strongly affected by increasing temperature
(decrease of activity from 105 000 to 10 100 kg of PE/
((mol of Fe) h bar) and from 107 000 to 10 800 kg of PE/
((mol of Fe) h bar), respectively), only precatalyst 5
shows more constant, but considerably lower, average
activities at temperatures of 0-60 °C.
The catalysts are rapidly deactivated during the
polymerization procedures at higher temperatures
(Tables 1 and 2).
At temperatures above 20 °C, a complete deactivation
of the catalysts can be observed within 40 min and even
faster times (within 20-30 min) are obtained at tem-
peratures above 60 °C. The rapid deactivation of the
catalysts may also originate from the self-immobiliza-
tion of the catalysts via copolymerization of the alkenyl
moiety of the [bis(imino)pyridyl]iron(II) catalysts with
the olefin during the polymerization experiment. Figure
1 demonstrates the dependence of the average activities
of precatalysts 5-7 on the polymerization temperature.
Due to faster deactivation of the catalyst at elevated
temperatures, the calculated average polymerization
activities of the catalysts 5-7 are significantly lower
than at low temperatures, although the activities at the
beginning of the polymerization are comparable at all
applied polymerization temperatures.
P olym er iza tion . The precatalysts were activated by
MMAO and used in the polymerization of ethene and
propene. As shown in Table 1 and Scheme 3, all
precatalysts exhibit an extremely high ethene polym-
erization activity of 24 000-107 000 kg of PE/((mol of
Fe) h bar) (at 0 °C). Precatalyst 5 shows the lowest
polymerization activity, which increases for 6 and 7. All
precatalysts show a strong temperature dependence of
their activity and of the average lifetime during ethene
The obtained polymers were analyzed by high-tem-
perature GPC and NMR.

78 Organometallics, Vol. 21, No. 1, 2002
Kaul et al.
Ta ble 2. Resu lts of Eth en e P olym er iza tion Ru n s
w ith P r eca ta lysts 5-7a
amt of catalyst
used (µmol)
amt of polymer
obtained (g)
precatalyst
T (°C)
5
0
20
40
60
80
0.617
0.617
0.617
0.922
1.543
26.73
24.56
22.12
21.89
10.01
6
0
20
40
60
80
0.227
0.604
0.302
0.227
0.756
4.65
81.83
37.11
9.45
13.11
7
0
20
40
60
80
0.355
0.296
0.296
0.296
0.740
0.54
28.59
22.05
17.13
13.99
F igu r e 2. Typical molecular weight distribution of the
polymers obtained using precatalyst 6 at different polym-
erization temperatures: (a) 0 °C; (b) 20 °C; (c) 40 °C; (d)
60 °C; (e) 80 °C. The molecular weight distributions of the
polymers obtained by the precatalysts 5 and 7 are analo-
gous.
a
Reaction conditions: 2.0 bar of ethene pressure; toluene as
solvent; cocatalyst MMAO (Al:Fe ) 1000); reaction time 1 h.
erization temperature is increased (approximately 40 000
at 0 °C, approximately 2000 at 80 °C). At intermediate
temperatures the molecular weight distribution broad-
ens due to the coincidence of the lower and higher
molecular weight fractions (Figure 2).
Although the produced polymers are very similar in
molecular weights, polydispersities, and molecular weight
distributions, it can be observed that the alkenyl-
functionalized catalyst with the shortest spacer moiety
produces the highest molecular weights of the polymers,
whereas an increase of spacer length decreases the
molecular weights of the produced polymers slightly.
As the NMR spectra clearly show, the catalysts
succeed in producing linear polyethylene at low and
intermediate temperatures up to 40 °C, as predomi-
nantly peaks originating from CH2 groups can be
F igu r e 1. Dependence of the polymerization activity of
precatalysts 5-7 on polymerization temperature. The
activity is given in units of kg of PE/((mol of Fe) h bar).
As Table 3 clearly shows, there is a strong dependence
of the molecular weight of the produced polymers on the
polymerization temperatures. At low polymerization
temperatures the molecular weight distribution exhibits
a significant bimodality, which decreases toward higher
polymerization temperatures. The high-molecular-
weight fraction of the polymer, which seems to be
predominant for the polymer produced at 0 °C, disap-
pears at temperatures above 40 °C (at 20 °C only traces
of the high-molecular-weight fraction can be detected).
In addition, the low-molecular-weight fraction shifts
toward even lower molecular weights when the polym-
1
detected at 30.1 ppm in 13C NMR and 1.30 ppm in H
NMR, respectively. Peaks originating from terminal
methyl groups or unsaturated terminal groups are
extremely small and confirm the high selectivity of the
[bis(imino)pyridyl]iron(II) catalysts for linear polymers
and the high molecular weights of the produced poly-
mers.
For higher polymerization temperatures of 60 °C and
above additional peaks for terminal methyl groups and
Ta ble 3. Da ta for th e P olym er s P r od u ced by P r eca ta lysts 5-7
T (°C) Mw Mn
mol wt distribna
B (40%/60%)
precatalyst
D
5
0
20
40
60
80
7.90 × 105/7.82 × 104
5.54 × 105 / 3.70 × 104
3.41 × 105/2.65 × 104
2.19 × 104/2.78 × 103
6.05‚103
1.42/2.11
1.84/2.09
3.53/1.39
5.84
B (15%/85%)
B (5%/95%)
M
6.28 × 105/5.54 × 104
7.73 × 104/3.88 × 104
3.53 × 104
3.34 × 103
1.64 × 103
2.03
6
7
0
20
40
60
80
B (50%/50%)
B (12%/88%)
B (2%/97%)
M
7.39 × 105/5.87 × 104
7.57 × 105/5.66 × 104
7.61 × 105/4.00 × 104
2.57 × 104
4.72 × 105/2.91 × 104
5.96 × 105/1.70 × 104
6.68 × 105/9.46 × 103
3.71 × 103
1.56/2.02
1.27/3.33
1.14/4.23
6.92
M
2.29 × 103
1.41 × 103
1.63
0
20
40
60
80
B (53%/47%)
B (21%/79%)
B (4%/96%)
M
6.14 × 105/4.01 × 104
7.16 × 105/4.55 × 104
6.78 × 105/3.25 × 104
1.15 × 104
3.64 × 105/1.94 × 104
5.04 × 105/1.50 × 104
5.32 × 104/8.83 × 103
2.77 × 103
1.69/2.07
1.42/3.04
1.27/7.68
4.17
M
2.96 × 103
1.66 × 103
1.78
a
Legend: B, bimodal; M, monomodal. Results from GPC at 135 °C.

[Bis(imino)pyridyl]iron(II) Complexes
Organometallics, Vol. 21, No. 1, 2002 79
1
terminal alkenyl groups can be observed in 13C and H
NMR. The signal for the terminal methyl groups can
be found at 22.85 ppm in 13C NMR and at 0.87 ppm in
1H NMR. Signals for the unsaturated groups can be
detected at 114.38 and 137.61 ppm in 13C NMR and at
Sch em e 4. Im m obiliza tion of th e
Alk en yl-F u n ction a lized [Bis(im in o)p yr id yl]ir on (II)
Com p lexes on a Mod ified Silica Su r fa ce
1
4.97 and 5.77 ppm in H NMR.
These observations give rise to the assumption that
the chain termination at higher temperatures predomi-
nantly occurs through â-H transfer, resulting in alkenyl-
terminated polymer chains, whereas at low polymeri-
zation temperatures no indication for alkenyl-terminated
polymer chains could be found.
Self-Im m obiliza tion . For self-immobilization the
catalysts 5-7 are activated with MMAO and ethylene
is bubbled through the solution for a short period of
time: e.g., 5 min. The polymeric catalysts can be
isolated via filtration and, after extensive workup to
separate the homogeneous catalyst, used in ethylene
polymerization, giving considerably lower activity than
the monomeric species.
The incorporation of the catalyst into the polymer
chain can be recognized by the green brownish color of
the polymer, which spontaneously disappears on expo-
sure to air and moisture.
Due to the air and temperature sensitivity of ac-
tivtated [bis(imino)pyridyl]iron(II) compounds, an ex-
amination of the incorporation of the activated alkenyl-
functionalized [bis(imino)pyridyl]iron complexes into the
polymer chain can best be achieved by an hydrolysis of
the self-immobilized activated catalyst and an NMR
examination of the obtained polymer, which contains
the self-immobilized catalyst ligand framework.
As the NMR data clearly show, nearly identical
signals are obtained for the homogeneous bis(imino)-
pyridine ligand and the polymer-incorporated ligand
framework after hydrolysis of the active catalyst. Al-
though the signals of the bis(imino)pyridine ligand are
small compared to the signals obtained by the CH2 and
CH3 groups of the polymer chain and are partially
overlaid by the polymer signals, the incorporation of the
ligand moiety can be demonstrated by 1H and 13C NMR.
As experiments showed, the activated [bis(imino)-
pyridyl]iron(II) complexes are not stable in the absence
of olefins and decompose rather quickly at room tem-
perature. Therefore, the polymerization activity of the
polymeric species, obtained by the self-immobilization
of the monomeric alkenyl-functionalized [bis(imino)-
pyridyl]iron(II) compounds, decreases rapidly when the
polymeric catalysts are isolated and especially when
they are stored over a longer period of time.
alkene-functionalized [bis(imino)pyridyl]iron(II) com-
plexes with an SiR2H-modified surface via hydrosilation
of the alkenyl moiety of the homogeneous precatalyst
with Si-H groups of the modified silica surface. There-
fore, in a first step, the Si-OH groups of the dehydroxy-
lated support have to be functionalized with Si-H
surface groups. This transformation can be performed
by treating the support with tetramethyldisilazane
(TMDS), which nearly selectively converts free (and to
a minor extend bonded) surface OH groups into OSiMe2H
groups.
In agreement with data found for the monomeric
species 5-7, the self-immobilized catalysts exhibit
higher polymerization activities, longer catalyst life-
times, and higher overall polymer molecular weights at
low temperatures than at higher (>40 °C) temperatures.
To rule out the influence of the Karstedt catalyst,
control experiments have been started in order to find
polymerization activity by adding to a solution of
Karstedt catalyst various amounts of MMAO. Neither
this reaction mixture nor addition of iron(II) chloride
or iron(III) chloride showed any ethene polymerization
activity.
In a second step, the alkenyl-functionalized [bis-
(imino)pyridyl]iron(II) complexes can be reacted with
the surface Si-H groups with the use of Karstedt
catalyst as a coupling reagent (Scheme 4) to yield the
covalently tethered [bis(imino)pyridyl]iron(II) com-
plexes. To ensure the intended formation of a covalent
bond between the alkenyl-functionalized homogeneous
[bis(imino)pyridyl]iron(II) complexes and the SiR2H-
modified silica surface via hydrosilation, infrared spec-
troscopy has been applied (Figure 3).
To monitor the covalent immobilization of the cata-
lysts, IR spectra of the SiMe2H-modified support (Figure
3, spectrum A) and the catalysts (Figure 3, spectra B-D)
have been recorded. The decrease of the intensity of the
Im m obiliza tion on Silica . An easy access to sup-
ported polymerization catalysts is the coupling of the

80 Organometallics, Vol. 21, No. 1, 2002
Kaul et al.
Ta ble 5. Resu lts of Eth en e P olym er iza tion Ru n s
w ith Im m obilized Ca ta lysts 8-10a
activityb
precatalyst t (°C) after 5 min after 45 min
average
8
0
20
40
60
2.8 × 104
2.4 × 104
3.0 × 104
2.0 × 104
9.0 × 103
2.7 × 103
1.7 × 103
1.4 × 103
6.61 × 103
4.53 × 103
4.33 × 103
2.17 × 103
400
500
9
0
20
40
60
80
3.7 × 104
2.9 × 104
2.8 × 104
3.8 × 104
8.2 × 103
5.1 × 103
2.6 × 103
1.5 × 103
9.31 × 103
7.52 × 103
4.53 × 103
3.77 × 103
700
40
500
F igu r e 3. Infrared spectra of the hetergeneous precata-
lysts: (A) Si(Me)2H-modified support; (B) anchored catalyst
8; (C) anchored catalyst 9; (D) anchored catalyst 10.
10
0
20
40
60
80
3.3 × 104
2.4 × 104
2.6 × 104
2.4 × 104
9.5 × 103
4.6 × 103
9.71 × 103
7.00 × 103
4.05 × 103
3.17 × 103
2.8 × 103
1.2 × 103
Ta ble 4. BET/BJ H Da ta a n d F e Con ten t of
500
200
Ca ta lysts 8-10
900
a
heterogeneous
precatalyst
Reaction conditions: 2.0 bar of ethene pressure; toluene as
b
solvent; cocatalyst MMAO (Al:Fe ) 1000); reaction time 1 h. In
units of kg of PE/((mol of Fe) h bar).
support +
Karstedt catalyst
8
9
10
surface BET (m2 g-1
)
178
1.1
1.15
189
1.3
1.10
178
1.1
0.99
198
1.5
0
Ta ble 6. Resu lts of Eth en e P olym er iza tion Ru n s
a
pore vol (cm3 g-1
Fe (wt %)
)
w ith Im m obilized Ca ta lysts 8-10a
amt of catalyst
used (µmol of Fe)
amt of polymer
obtained (g)
a
BJ H, desorption, cummulative, 17-1000 Å.
precatalyst
T (°C)
Si-H band (2152 cm-1) after the immobilization clearly
shows the formation of a covalent bond between the
alkenyl-functionalized [bis(imino)pyridyl]iron(II) com-
plex and the SiMe2H-modified surface. The appearance
of aromatic bands around 3068 cm-1 and new aliphatic
bands around 2870 cm-1 in the catalyst spectra (Figure
3, spectra B-D) confirms the presence of aromatic and
aliphatic moieties of the supported catalyst. To rule out
any influence of the Karstedt catalyst on polymerization
results, control experiments under identical hydrosila-
tion conditions using the modified support and various
amounts of hydrosilation catalyst have shown nearly no
difference concerning the Si-H band at 2152 cm-1
compared with spectrum A. This result leads to the
assumption that practically no covalent bonding be-
tween the Karstedt catalyst and the modified support
occurs.
As presented in Table 4, the BET surfaces of the
heterogenized precatalysts 8-10 decrease by approxi-
mately 10% during the immobilization of the [bis(imino)-
pyridyl]iron(II) complex and a decrease of the pore
volume from 1.6 cm3 g-1 to around 1.2 cm3 g-1 can be
observed. Both the IR data and the BET data give rise
to the assumption that the covalent anchoring of the
alkenyl functionalized iron(II) complexes 5-7 was suc-
cessful and that the catalysts are anchored not only on
the outer surface but also in the pores of the support.
In contrast, the BET and BJ H numbers of the control
experiments (support + Karstedt catalyst) are close to
the results obtained from modified support (Table 4).
Therefore, we suggest no or just a very small amount
of anchored Karstedt catalyst.
5
0
20
40
60
80
4.180
4.283
4.777
4.345
4.819
51.23
33.87
38.19
16.14
3.01
6
0
20
40
60
80
3.920
4.373
4.235
4.136
8.529
59.98
61.23
35.51
28.67
6.20
7
0
20
40
60
80
3.394
4.325
3.989
4.308
4.148
59.17
56.15
27.36
23.44
6.31
a
Reaction conditions: 2.0 bar of ethene pressure; toluene as
solvent; cocatalyst MMAO (Al:Fe ) 1000); reaction time 1 h.
loading of iron. Elemental analysis gave no evidence of
self-immobilization of the Pt hydosilation catalyst on the
support. Also, control experiments have been started
wherein various amounts of the Karstedt catalyst and
the modified silica were mixed together under identical
hydrosilation conditions. Neither a suspension of this
mixture nor addition of MMAO, iron(II) chloride, or iron-
(III) chloride showed any catalytic activity toward
polymerization of ethene.
P olym er iza tion Resu lts. As shown in Tables 5 and
6 and Figure 4, all heterogenized precatalysts show a
good polymerization activity in the range of 103-104 kg
of PE/((mol of Fe) h bar). Analogous to the homogeneous
precatalysts, the compound with the shortest alkenyl
moiety, the allyl-functionalized compound 8 exhibits the
lowest activity, whereas the activity increases with
increasing length of the alkenyl chain (compounds 9 and
10 exhibit a higher activity). The activities decrease at
higher temperatures for all heterogenized precatalysts,
reaching a minimum activity of only 500 kg of PE/((mol
of Fe) h bar) at high temperatures of 80 °C. However,
in comparison to the homogeneous precatalysts 5-7,
which provide no activity after 45 min at temperatures
Elemental analysis shows that the percent load of iron
on the silica (Table 4) varies between 0.99 wt % for
catalyst 10 and 1.15 wt % for the allyl-functionalized
catalyst 8. This roughly corresponds to the size of the
Si-H band in the IR spectra, as the size of the signal
of the Si-H moiety decreases most in the case of the
heterogeneous precatalyst 8, which exhibits the highest

[Bis(imino)pyridyl]iron(II) Complexes
Organometallics, Vol. 21, No. 1, 2002 81
Ta ble 7. Da ta for th e P olym er s P r od u ced by 8-10
precatalyst
T (°C)
mol wt distribn
Mw
Mn
D
8
0
20
40
60
80
B (74%/26%)
B (75%/15%)
B (79%/21%)
B (84%/16%)
B (9%/91%)
8.97 × 105/4.45 × 104
7.34 × 105/1.27 × 104
6.95 × 105/4.75 × 103
5.10 × 105/1.72 × 103
4.17 × 104/690
3.96 × 105/ 2.63 × 104
2.31 × 105/7.65 × 103
1.13 × 105/2.51 × 103
5.99 × 104/1.04 × 103
2.76 × 104/570
2.27/1.69
3.18/1.66
6.13/1.89
8.52/1.65
1.51/1.21
9
0
20
40
60
80
B (72%/28%)
B (78%/22%)
B (82%/18%)
B (84%/16%)
B (8%/92%)
8.11 × 105/1.81 × 104
7.36 × 105/8.42 × 103
6.79 × 105/2.87 × 103
4.49 × 105/1.61 × 103
3.41 × 104/600
3.01 × 105/8.37 × 103
2.17 × 105/4.66 × 103
1.05 × 105/1.43 × 103
4.64 × 105 / 1.01 × 103
2.35 × 104/520
2.69/2.16
3.40/1.81
6.44/2.01
9.59/1.59
1.45/1.15
10
0
20
40
60
80
B (70%/30%)
B (78%/22%)
B (82%/18%)
B (84%/16%)
B (8%/92%)
8.77 × 105/1.78 × 104
6.75 × 105/8.88 × 103
4.44 × 105/2.83‚103
4.21 × 105/1.59 × 103
2.35 × 104/600
3.10 × 105/9.19 × 103
2.16 × 105/4.76 × 103
7.75 × 104/1.46 × 103
4.51 × 104/930
2.76/1.94
3.13/1.87
5.73/1.94
9.33/1.71
1.47/1.20
1.60 × 104/500
higher than 20 °C, the heterogeneous precatalysts
exhibit higher activities at elevated temperatures. Pre-
catalyst 8 even shows good activity at 60 °C and remains
active after 45 min; precatalysts 9 and 10 exhibit
activity after 45 min at even higher temperatures of
80 °C.
Although the average lifetimes of the heterogeneous
precatalysts are higher than those of the homogeneous
precatalysts, a rise of polymerization temperature
strongly reduces the overall lifetimes of all heteroge-
neous precatalysts and therefore the average polymer-
ization activity of the catalysts. A possible reason for
the slower deactivation of the heterogeneous catalyst
might be the impossibility of a deactivation of the
alkenyl-substituted homogeneous catalyst via self-im-
mobilization, as the alkenyl moiety is covalently bonded
to the surface and therefore no longer available for a
copolymerization with olefins, which, as experiments
showed,9 partially reduces polymerization activity dur-
ing the polymerization.
(the polymers produced by the homogeneous precata-
lysts were dominated by low molecular weights, and the
high-molecular-weight fraction completely disappeared
at polymerization temperatures >20 °C). In accordance
with the polymers produced by 5-7, the molecular
weights slightly shift toward lower values, when the
polymerization temperature is increased from 0 to 60
°C, and shift very drastically (to a molecular weight of
102-103) when the temperature is further increased to
80 °C. In addition, the molecular weight distribution
becomes nearly monomodal at higher temperatures, but
the transition from mono- to bimodal distribution only
occurs at very high polymerization temperatures (mono-
modal molecular weight distributions are obtained at
temperatures >40 °C for the homogeneous catalysts and
>80 °C for the heterogeneous [bis(imino)pyridyl]iron-
(II) catalysts).
The highest molecular weights are obtained with
precatalyst 8, which has the shortest alkenyl moiety.
An increase of chain length of the alkenyl moiety
decreases the molecular weight of the produced poly-
mers slightly, so that the lowest molecular weights are
obtained by using the heterogeneous precatalyst 10.
This tendency becomes more obvious when the polym-
erization is conducted at higher temperatures, whereas
the polymers produced by the heterogeneous 8-10 at
low temperatures are nearly identical.
The obtained polymers were analyzed by high-tem-
perature GPC (Table 7, Figure 5) and NMR.
The polymers produced by the heterogeneous pre-
catalysts 8-10 are substantially different from those
produced by the homogeneous precatalysts 5-7. As can
be seen in Table 7 and Figure 5, the polymers obtained
with the precatalysts 8-10 are of bimodal molecular
weight distribution, but unlike the polymers produced
by 5-7, the total molecular weight is dominated by the
high molecular fraction of the polymer with a molecular
mass of approximately 105-106 g/mol (at temperatures
of 0-40 °C). Only a small quantity of the polymer shows
a lower molecular mass of approximately 103-104 g/mol
F igu r e 5. Dependence of the molecular weight of polymers
obtained at different temperatures: (a) 0 °C; (b) 20 °C; (c)
40 °C; (d) 60 °C; (e) 80 °C. The molecular weight distribu-
tions of the polymers obtained by precatalysts 8-10 are
analogous.
F igu r e 4. Dependence of the polymerization activity of
precatalysts 8-10 on polymerization temperature. The
activity is given in units of kg of PE/((mol of Fe) h bar).

82 Organometallics, Vol. 21, No. 1, 2002
Kaul et al.
NMR experiments showed that, in contrast to the
homogeneous catalysts, no signals for unsaturated
terminal groups could be found in 13C and 1H NMR.
Even at higher temperatures no signals for alkenyl
when the polymerization is conducted at very low
temperatures, whereas no activity can be detected at
polymerization temperatures above 20 °C.
The heterogeneous catalysts failed to show any pro-
pene polymerization activity, so that no activity could
be detected for any polymerization conditions.
1
groups were detected in 13C and H NMR.
Analogous to the case for the homogeneous catalysts,
larger signals for terminal methyl groups can be de-
tected at higher temperatures, a fact which correlates
well with the lower molecular weights of the polymers
produced.
Con clu sion s
The presented modification of ligand 1 is a straight-
forward approach to alkenyl-functionalized [bis(imino)-
pyridyl]iron(II) complexes. The peripheral alkenyl moi-
ety allows a variety of useful heterogenization reactions.
They include the self-immobilization of the alkenyl-
functionalized [bis(imino)pyridyl]iron(II) catalysts and
the immobilization on a SiMe2H-modified silica surface
via hydrosilation. The covalent bonding to the SiMe2H-
modified silica support can be monitored by IR spec-
troscopy.
All three catalysts show good activities at high
temperatures, even in contrast to the unsupported
systems 5-7, which provide no activity after 45 min at
temperatures higher than 20 °C. Catalyst 8 shows good
activity in the beginning at 60 °C and remains active
after 45 min. At 80 °C catalysts 9 and 10 are still active
after 45 min. In general, the activities of the supported
catalysts are approximately one order of magnitude
lower than those of their homogeneous counterparts. In
case of the homogeneous precatalysts the variation of
the spacer length results in an increase of activity from
the allyl-modified catalyst (5) to the butenyl- (6) or
pentenyl-modified catalyst (7). The same trend is ob-
served for the supported catalysts 8-10.
The immobilization of the modified [bis(imino)pyridyl]-
iron(II) complexes 5-7 via hydrosilation results in
highly active and temperature-stable catalyst precursors
and can be activated with MAO and MMAO. As an
important advantage they lack any reactor fouling. As
further examination showed, the modification with
butenyl or pentenyl groups is more favorable than allyl
spacers, at least with regard to catalyst activity.
For all three supported catalysts no reactor fouling
was observed during the whole polymerization process
at all polymerization conditions, in contrast to the case
for the homogeneous precatalysts 5-7.
Propene polymerization experiments with the homo-
geneous alkenyl-functionalized [bis(imino)pyridyl]iron-
(II) catalysts only gave very low starting activities and
a complete inactivity of the catalysts after a few minutes
of polymerization. Analogous to the ethene polymeri-
zation, a higher polymerization activity can be obtained
Ack n ow led gm en t. We are grateful to our co-worker
A. Osterauer for experimental support and Mr. M. Barth
for analytical support. Dr. H. W. Go¨rlitzer is thanked
for helpful discussions. This work was generously sup-
ported by the Deutsche Forschungsgemeinschaft and
the Bayerische Forschungsverbund Katalyse FORKAT.
(9) Wieczorek, K.; Puchta, G. T.; Herrmann, W. A. Unpublished
results.
OM000939T