Catalytic properties of several supported Pd(II) complexes for Suzuki coupling reactions

Article abstract of DOI:10.1016/j.tetlet.2005.08.029

Pd(II) complexes with N-ligands were synthesized and tested for Suzuki coupling reaction. These complexes were also heterogenized on silica. The resulting site-isolated catalysts showed a high catalytic activity. The most stable complex was supported N-(propyl)ethylenediamine, which did not display any notable leaching. The effects of reaction conditions and the nature of the boronic acid derivative on the conversion of the starting compounds were studied.

Full text of DOI:10.1016/j.tetlet.2005.08.029

Tetrahedron
Letters
Tetrahedron Letters 46 (2005) 6865–6869
Catalytic properties of several supported Pd(II) complexes for
Suzuki coupling reactions
Oleksiy Vassylyev, Jengshiou Chen, Anthony P. Panarello and Johannes G. Khinast*
Department of Chemical and Biochemical Engineering, Rutgers, The State University of New Jersey, 98 Brett Road,
Piscataway, NJ 08854, USA
Received 30 June 2005; revised 28 July 2005; accepted 2 August 2005
Abstract—Pd(II) complexes with N-ligands were synthesized and tested for Suzuki coupling reaction. These complexes were also
heterogenized on silica. The resulting site-isolated catalysts showed a high catalytic activity. The most stable complex was supported
N-(propyl)ethylenediamine, which did not display any notable leaching. The effects of reaction conditions and the nature of the
boronic acid derivative on the conversion of the starting compounds were studied.
ꢀ 2005 Elsevier Ltd. All rights reserved.
NH
Palladium(II) complexes with nitrogen-containing
ligands are well-known catalysts for Suzuki coupling
1
CH3O
CH3O
NH2
[PdCl2(PhCN)2]
Si
Pd
Cl
reactions, which are of increasing importance for the
formation of C–C bonds.1 We recently used Suzuki cou-
plings for the synthesis of bis-indenyl ligands for a chiral
titanocene catalyst.2 Homogeneous Pd complexes pos-
sess high activity for this reaction; however, their disad-
vantage is that they are hard to remove and recycle,
leading to a contamination of products with the catalyst.
Thus, heterogenization of the catalysts is a logical step
to avoid these problems, as it allows for easy removal
from the reaction mixture. However, inorganic hetero-
geneous Pd catalysts show low stability due to leaching
of Pd from the surfaces.3 In fact, in detailed studies it
was shown that in most cases, the leached homogeneous
Pd species is responsible for the catalytic activity.4 A
promising strategy to obtain stable heterogeneous cata-
lysts is the synthesis of organic/inorganic hybrid materi-
als, containing Pd complexes with an organic ligand
anchored covalently on the support. This support may
be an organic polymer or an inorganic oxide (e.g.,
silica).5 When these catalysts are immobilized on a cata-
lytically active material (e.g., zeolites), bifunctional
catalysts may be obtained, where active sites of one kind
are located in the zeolite pores and the Pd moieties on
the external surface.6
OCH3
Cl
3
4
2
Cl
Cl
Pd
N
N
NH2
NH2
Si
Pd
Cl
Cl
Si
Si
Si
CH3O
CH3O
OCI
CH3O
CH3O
OCH3
OCH3
CH3O
OCH3
CH3O
OCH3
OCH3
6
5
Scheme 1. Synthesis of three different Pd(II)-complexes.
In this study, three ligands containing functional groups
for their immobilization were chosen, that is, N-[3-(tri-
methoxysilyl)propyl]-ethylenediamine (1), 2-(20-pyridyl)-
ethyltrimethoxysilane (2), and 3-aminopropyltrimethoxy-
silane (3). Their Pd(II) complexes were prepared by
the reaction of [PdCl2(PhCN)2] with the corresponding
ligand (Scheme 1) under reflux with i-PrOH.7 All three
ligand–Pd complexes were not reported before for use
as Suzuki coupling catalysts.
In order to obtain a heterogeneous catalyst, the ligands
were first immobilized on the silica surface in toluene
(Scheme 2) and then metalated with [PdCl2(PhCN)2].8
This strategy was chosen due to the very low solubility
of two of the resulting complexes (i.e., 4 and 6) in most
Keywords: Suzuki couplings; Catalysis; Pd leaching.
*
Corresponding author. Tel.: +1 7324452970; fax: +1 7324452581;
e-mail: khinast@rutgers.edu
0040-4039/$ - see front matter ꢀ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2005.08.029

6866
O. Vassylyev et al. / Tetrahedron Letters 46 (2005) 6865–6869
0.00
L
Si
OH
O
O
OH
1 - 3
O
OH
0.02
0.04
0.06
0.08
0.10
Silica
7 - 9
C
7
8
Si
O
O
[PdCl2(PhCN)2]
O
10 - 12
10
11
= (CH2)3NH(CH2)2NH2 (7), (CH2)2Py (8), (CH2)3NH2 (9)
L
C
= (CH2)3NH(CH2)2NH2-PdCl2 (10), (CH2)2Py-PdCl2 (11),
(CH2)3NH2-PdCl2 (12)
vPy
vCH2
vPy
Scheme 2. Preparation of supported Pd(II)-complexes.
vPy
Table 1. Composition of immobilized catalysts10
1500
1000
500
Wavenumber, cm-1
Ligand
Loading (mmol/g)
Surface density
(molecules/nm2)
Figure 1. IR spectra of immobilized ligands 7–8 and Pd-complexes 10–
11.
Ligand
Pd(II)
Ligand
Pd(II)
1
2
3
1.40
1.20
1.19
1.38
0.99
0.93
2.52
2.16
2.14
2.48
1.78
1.68
1457 cmꢀ1). In contrast to 4, the corresponding doublet
in the spectra of 5 shifted to 1448–1483 cmꢀ1. Band of
the Py stretching vibrations shifted to 1568–1604 cmꢀ1
In addition, a new band at 770 cmꢀ1 appeared.
.
solvents, which made it very difficult to immobilize
them. As can be seen from Table 1, the loading of the
ligands in all cases was similar (somewhat larger in the
case of 4). Assuming the surface area of silica to be
330 m2/g and the estimated number of silanol groups
on the surface to be 4.9 0.8 molecules/nm29 we con-
clude that most of the silanol groups were replaced by
the ligand molecules.
After immobilization of 1 and 3 on the surface, the band
771–807 disappeared due to replacement of the Si–O–C
bonds onto Si–O–Si with a band at 797 cmꢀ1 (Fig. 1).
Only a weak shoulder can be seen that corresponds to
a small number of remaining methoxy groups. For
immobilized 2, the band of the pyridine ring vibrations
was observed, but shifted to 743 cmꢀ1. The bands at
1434–1473 cmꢀ1 shifted insignificantly (1434–1477
cmꢀ1 in the immobilized ligand 8 and 1437–1478 cmꢀ1
in the immobilized complex 11), but the bands at
1568–1591 cmꢀ1 changed their position to 1576–1598
cmꢀ1 (8) and 1569–1608 cmꢀ1 (11), which is an evidence
of complexation with Pd. For 9 and 12, these bands
appeared too weak to be detected.
The yields of the surface complexes of Pd(II) were
between 78% and 98% (the highest loading of Pd was
found for complex 10). Considering the density of the
ligand coverage (an average distance between their mole-
cules is 0.8–0.9 nm), it is evident that the Pd atoms in the
surface complexes 11–12 are bound to only one ligand
molecule. This distance excludes influence of neighbor-
ing catalyst molecules on the active site.
R
Br
Immobilization of Pd(II) was confirmed by IR studies of
the samples. Spectra of the parent ligands contain strong
bands related to the stretching vibrations of Si–O–C
bonds at 1189 (asymmetric) and 771–807 (symmetric,
doublets) cmꢀ1. In addition, the spectrum of ligand 2
contains a strong band at 789 cmꢀ1 related to the vibra-
tions of the pyridine ring, and medium doublets at 1434–
1473 (overlapping of CH2 bending and scissoring, and
Py stretching vibrations) and 1568–1591 (Py stretching)
R
B(OH)2
+
COOH
COOH
Tests of catalysts 4–6 for the Suzuki coupling reaction in
an isopropanol/water solution showed that the com-
plexes with monodentate ligands 5–6 have a higher
activity, while the complex with the bidentate ligand 4
appeared less active (Table 2). The activity of the immo-
bilized version (catalyst 10) was found significantly high-
er, which can be attributed to the effect of Ôsite isolationÕ,
that is, the absence of interactions between catalyst mole-
cules,12 which is followed by a decomposition of the
cmꢀ1
.
Formation of the complexes resulted in some changes in
the IR spectra. In 4, the band of CH2 bending and scis-
soring vibrations shifted insignificantly (from 1459 to

O. Vassylyev et al. / Tetrahedron Letters 46 (2005) 6865–6869
Table 2. Activity of Pd catalysts in Suzuki coupling11
6867
Catalyst
R
Solvent
Base
T (ꢁC)
Conversion, % (time, h)
4
Phenyl
Phenyl
Phenyl
Phenyl
Phenyl
Phenyl
Phenyl
Phenyl
Phenyl
Phenyl
Phenyl
Phenyl
Phenyl
Phenyl
Phenyl
Phenyl
IPA/H2O (1:9)
DMA/H2O (1:9)
IPA/H2O (1:9)
IPA/H2O (1:9)
IPA/H2O (1:9)
IPA/H2O (1:9)
IPA/H2O (1:9)
IPA/H2O (1:9)
Toluene
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
Et3N
80
90
80
80
80
80
80
80
90
90
90
80
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
1.9 (2), 16.6 (19)
1.5 (2), 18.6 (19)
100 (2)
90.4 (2)
80.1 (2), 99.9 (16)
100 (2)
4
5
6
10
11
12
13
13
13
13
10
10
10
10
10
10
10a
10b
10
10
10
10
10
10
10
10
93.6 (2)
20.7 (2), 99.8 (16)
4.3 (2), 8.9 (22)
23.1 (2)
55.5 (2), 99.7 (22)
2.0 (22)
DMA/H2O (1:9)
DMA/H2O (1:9)
Dioxane
Cs2CO3
Na2CO3
Et3N
Toluene
Et3N
18.2(2)
DMA/H2O (1:9)
DMA/H2O (1:9)
DMA/H2O (1:9)
DMA/H2O (1:9)
DMA/H2O (1:9)
DMA/H2O (5:6)
DMA/H2O (1:9)
DMA/H2O (1:9)
DMA/H2O (1:9)
DMA/H2O (1:9)
DMA/H2O (1:9)
DMA/H2O (1:9)
DMA/H2O (1:9)
DMA/H2O (2:9)
Cs2CO3
Na2CO3
K2CO3
Et3N
89.6 (2), 97.7 (22)
98.3 (2), 100 (22)
85.5 (2), 99.8 (22)
28.3 (2), 94.6 (22)
100 (1)
81.8 (2), 100 (22)
89.7 (2), 99.8 (22)
97.6 (2), 100 (22)
97.3 (2)
89.1 (2)
100 (2)
46.1 (2), 57.1 (22)
2.0 (2)
22.3 (2), 49.8 (22)
Phenyl
Phenyl
Phenyl
Na2CO3
Na2CO3
Na2CO3
Na2CO3
Na2CO3
Na2CO3
Na2CO3
Na2CO3
Na2CO3
Na2CO3
3-Hydroxyphenyl
3-(Hydroxymethyl)phenyl
4-(Hydroxymethyl)phenyl
3-Carboxyphenyl
4-Carboxyphenyl
4-Formylphenyl
3-Thienyl
Vinylc
a With 4-iodobenzoic acid.
b With 4-iodoanisole.
c Dibutyl ester.
complex and formation of less active Pd nanoclusters.13
The significantly higher activity of the immobilized cat-
alyst proved that heterogeneous and not homogeneous
Pd contributed to its activity. The catalysts 11–12 dis-
played a higher activity as well, however a filtration
test14 indicated some leaching for them to occur (in con-
trast to catalyst 10, where the filtration test indicated no
detectable leaching). The concentration of homogeneous
Pd after reaction with catalyst 10 as determined by IPC
was found to be below reliable detection limits, while for
catalyst 11 it reached 3 ppm.
Comparison of different bases indicated that Na2CO3 is
the most favorable one. It seems that the nature of the
optimal base is unique for each catalyst and that no gen-
eral dependence exists (Artok and Bulut17 and Wesk-
amp et al.18 reported better performance of Cs2CO3 in
comparison to Na2CO3, while Phan et al.5a found a
reverse dependence. However, they found a much higher
activity in the presence of TEA). This difference in the
activating effect of the base is caused by the formation
of a transition complex with Pd, which depends on its
coordination sphere.
After reduction of Pd(II) in the catalyst 10 to Pd(0),15
the activity of 13 was lower, and leaching of Pd was
detected. Thus, the stability of the Pd(II) complex
appeared higher than that of the Pd(0) complex. For
this reason, we continued to work with catalyst 10
only.
The activity of 10 was also studied for the conversion of
different substituted boronic acids. In order to elucidate
the effect of the substituent in the aromatic ring, we
tested the catalyst for the coupling of phenylboronic
acids with hydroxy, hydroxymethyl, and carboxy groups
in 3- and 4-positions. The position of a CH2OH group in
the ring did not affect the conversion. For an electron-
withdrawing COOH substituent, the reactivity of (4-
carboxyphenyl)boronic acid was higher (Table 2). In
the case of an electron-releasing OH group, a high con-
version was observed for the 3-substituted ring, while no
cross-coupling product was detected for (4-aminophen-
yl)boronic acid pinacoline ester and (4-hydroxyphen-
yl)boronic acid (not shown in Table 2). Non-identified
peaks in the HPLC chromatograms of the reaction mix-
tures for these cases probably belong to products of
homocoupling of boronic acid and its deboronation,
while (4-bromophenyl)acetic acid remained uncon-
sumed. Earlier, Lei and Zhang reported much higher
Effect of the solvent on the activity of 10 was studied
with different solvent systems. The highest conversion
of phenylboronic acid was found using a DMA/water
mixture. In contrast to the data reported by LeBlond
et al., high water content in this mixture did not lead
to the homocoupling of phenylboronic acid.16 The con-
version in i-PrOH/water mixture was somewhat lower,
but higher than in toluene. Surprisingly, a very low cat-
alytic activity was seen in dioxane, while no reaction
occurred in DMA and THF. In contrast to a mixture
of i-PrOH/water, some insignificant leaching of Pd was
detected in DMA/water.

6868
O. Vassylyev et al. / Tetrahedron Letters 46 (2005) 6865–6869
J. M.; Weck, M.; Jones, C. W. Adv. Synth. Catal. 2005,
347, 161–171.
ability of (4-methoxyphenyl)boronic acid to homocou-
pling in comparison with non-substituted substrate even
in non-aqueous solution.19 The possibility of deborona-
tion of arylboronic acids with an electron-donor group
in the benzene ring under mild conditions was men-
tioned by Lapert20 and Lange,21 and the promoting
effect of silica for this pathway was reported by Schilling
and Kaufmann.22 We assume that the OH group in the
4-position favored these side-reactions.
5. (a) Phan, N. T. S.; Brown, D. H.; Styring, P. Tetrahedron
Lett. 2004, 45, 7915–7919; (b) Wang, Y.; Sauer, D. R. Org.
Lett. 2004, 6, 2793–2796; (c) Buchmeiser, M. R.; Schare-
ina, T.; Kempe, R.; Wurst, K. J. Organometal. Chem.
2001, 634, 39–46; (d) Wang, B.; Gu, Y.-L.; Yang, L.-M.;
Suo, J.-S. Fenzi Cuihua 2003, 17, 468–480; (e) Kaneda, K.
Yuki Gosei Kagaku Kyokaishi 2003, 61, 436–444; (f)
Blanco, B.; Mehdi, A.; Moreno-Manas, M.; Pleixats, R.;
Reye, C. Tetrahedron Lett. 2004, 45, 8789–8791; (g)
Baleizao, C.; Corma, A.; Garcia, H.; Leyva, A. J. Org.
Chem. 2004, 69, 439–446; (h) Gurbuz, N.; Ozdemir, I.;
Cetinkaya, B.; Seckin, T. Appl. Organometal. Chem. 2003,
17, 776–780; (i) Corma, A.; Das, D.; Garcia, H.; Leyva, A.
J. Catal. 2005, 229, 322–331.
The conversion of other boronic acids was low as well.
As reported earlier, the low conversion of (4-formylphen-
yl)boronic acid did not depend on the experimental
conditions.23 Only an insignificant conversion of 3-thi-
enylboronic acid was detected. Surprisingly, no reaction
occurred with (4-vinylphenyl)boronic acid. Recently, we
found that for most catalysts (except Pd(PPh3)4), (4-
vinylphenyl)boronic acid does not react with arylbro-
mides in the presence of water, while its conversion in
non-aqueous solvent systems remains low.24 As to the
non-aromatic boronic acids, vinylboronic acid dibutyl
ester yielded moderate conversions, while cis-1-propen-
ylboronic acid did not react. (Reactions with 0% yield
are not shown in Table 2.)
6. (a) Corma, A.; Garcia, H.; Leyva, A.; Primo, A. Appl.
Catal. A: General 2004, 257, 77–83; (b) Corma, A.; Garcia,
H.; Leyva, A. Appl. Catal. A: General 2002, 236, 179–185.
7. Synthetic procedure for complexes 4–6. 0.011 mol of
ligands 1–3 was dissolved in 10 mL of CH2Cl2, and a
solution of 0.01 mol of [PdCl2(C6H5CN)2] in 10 mL of
CH2Cl2 was added at room temperature. The solvent was
removed in vacuum, the residue was washed by pentane
and diethyl ether. Dichloro(N-(30-trimethoxysilyl)propyl-
1,2-ethanediamine-N,N0)-palladium (4): white crystals
(dioxane/ethanol, decomp. >180 ꢁC); IR (cmꢀ1): 774 (Si–
O–C), 909, 1074, 1189 (Si–O–C), 1456 (CH2), 2838, 2941
(CH2). Dichloro-bis[2-(20-trimethoxysilyl)ethylpyridine]-
palladium (5): yellow crystals (dioxane, decomp.
The conversion of iodo-substituted aromatic com-
pounds was much higher than the bromo-substituted
ones. In the case of 4-iodobenzoic acid, the reaction
was completed within 1 h. For the reaction with 4-iodo-
anisole, we used a solvent system with lower water con-
tent in order to provide a sufficient solubility of all
reactants.
1
>200 ꢁC); H NMR (d): 1.61 (t, 2H, SiCH2), 3.73 (s, 9H,
OCH3), 4.03 (t, 2H, Py-CH2), 7.20–7.34 (m, 2H, Py-CH),
7.69 (t, 1H, Py-CH), 9.01 (d, 1H, Py-CH); IR (cmꢀ1): 747
(Si–O–C), 770, 791 (Py), 912, 1073, 1190 (Si–O–C), 1448,
1483 (CH2), 1568, 1604 (Py), 2839, 2940 (CH2). Dichloro-
bis[1-(30-trimethylsilyl)propaneamine]-palladium (6): white
crystals (dioxane/ethanol, decomp. >210 ꢁC); IR (cmꢀ1):
774 (Si–O–C), 911, 1024, 1189 (Si–O–C), 1487 (CH2), 2935
(CH2).
In summary, immobilized dichloro(N-(30-trimethoxysil-
yl)propyl-1,2-ethanediamine-N,N0)-Pd showed a high
activity for Suzuki coupling reactions, which makes it
a useful system for biaryl compound synthesis. Most
importantly, leaching appeared negligible. Future work
will be devoted to the understanding of leaching in
different solvent systems and to the use for other
reactions.
8. Preparation of immobilized complexes 10–12. 0.01 mol of
ligands 1–3 was dissolved in 25 mL of toluene. Two grams
of previously dehydrated silica was suspended, and the
reaction mixture was kept at 100 ꢁC for 18 h. The silica
was filtered, washed with hot toluene and dried in a
vacuum. Obtained hybrid materials 7–9 were suspended in
a solution of 0.01 mol of [PdCl2(C6H5CN)2] in 2 5 mL of
i-PrOH and refluxed for 4 h. The brown product was
filtered, washed with 2-C3H7OH, CH2Cl2, and dried in a
vacuum.
References and notes
9. Zhuravlev, L. T. Langmuir 1987, 3, 316–318.
10. Loadings of the ligands and Pd were calculated on the
basis of elemental analysis.
11. For all tests, a halide aryl compound (1 mmol), a boronic
acid (1 mmol) and a base (3 mmol) were dissolved in
10 mL of a solvent. A catalyst (amount containing
0.01 mmol of Pd) was then added to the reaction mixture.
The reaction was monitored by HPLC (Shimadzu SCL-
10A with Discovery C18 column, 15 cm · 4.6 mm, 5 lm;
eluent: methanol/water, 60:40).
1. (a) Suzuki, A. Modern Arene Chemistry. In The Suzuki
Reaction with Arylboron Compounds in Arene Chemistry;
Astruc, D., Ed.; Wiley: Weinheim, Germany, 2002; pp 53–
106; (b) Bellina, F.; Carpita, A.; Rossi, R. Synthesis 2004,
2419–2440; (c) Blaser, H.-U.; Indolese, A.; Schnyder, A.;
Steiner, H.; Studer, M. J. Mol. Catal. A: Chem. 2001, 173,
3–18.
2. Panarello, A. P.; Vassylyev, O.; Khinast, J. G. Tetrahedron
Lett. 2005, 46, 1353–1356.
3. (a) Conlon, D. A.; Pipik, B.; Ferdinand, S.; LeBlond, C.
R., Jr.; Sowa, J. R., Jr.; Izzo, B.; Collins, P.; Ho, G.-J.;
Williams, J. M.; Shi, Y.-J.; Sun, Y. Adv. Synth. Catal.
2003, 345, 931–935; (b) Biffis, A.; Zecca, M.; Basato, M.
J. Mol. Catal. A: Chem. 2001, 173, 249–274; (c) Horniak-
ova, J.; Raja, T.; Kubota, Y.; Sugi, Y. J. Mol. Catal. A:
Chem. 2004, 217, 73–80; (d) Heidenreich, R. G.; Kohler,
K.; Krauter, J. G. E.; Pietsch, J. Synlett 2002, 1118–1122.
4. (a) Yu, K.; Sommer, W.; Weck, M.; Jones, C. W. J. Catal.
2004, 226, 101–110; (b) Yu, K.; Sommer, W.; Richardson,
12. The effect of site isolation was reported in (a) Grubbs, R.;
Lau, C. P.; Cukier, R.; Brubaker, C. J. Am. Chem. Soc.
1977, 99, 4517–4518; (b) Pugin, B. J. Mol. Catal. A: Chem.
1996, 107, 273–279.
13. (a) Shimizu, K.; Koizumi, S.; Hatamachi, T.; Yoshida, H.;
Komai, S.; Kodama, T.; Kitayama, Y. J. Catal. 2004, 228,
141–151; (b) Thathagar, M. B.; Beckers, J.; Rothenberg,
G. Adv. Synth. Catal. 2003, 345, 979–985.
14. The reaction was carried out until a conversion of 30–50%
was reached. Then, the catalyst was filtered and the

O. Vassylyev et al. / Tetrahedron Letters 46 (2005) 6865–6869
6869
reaction mixture was stirred at the same temperature for
22 h. The absence of further conversion proves the absence
of homogeneous Pd in the solution.
19. Lei, A.; Zhang, X. Tetrahedron Lett. 2002, 43, 2525–2528.
20. Lappert, M. F. Chem. Rev. 1956, 56, 959–1064.
21. Lange, S. The effect of the alkoxy chain position in PTPO6
isomers on the CPI induction phenomenon (final report),
University of Leeds, 2001.
22. Schilling, B.; Kaufmann, D. E. Eur. J. Org. Chem. 1998,
701–709.
23. Wendeborn, S.; Berteina, S.; Brill, W. K.-D.; De Mes-
maeker, A. Synlett 1998, 671–675.
15. Reduction was carried out by dropping a methanol
suspension of NaBH4 to the catalyst in CH3OH at 40 ꢁC
for 1 h. The black product was then filtered, washed with
CH3OH and CH2Cl2, and dried in a vacuum.
16. LeBlond, C. R.; Andrews, A. T.; Sun, Y.; Sowa, J. R. Org.
Lett. 2001, 3, 1555–1557.
17. Artok, L.; Bulut, H. Tetrahedron Lett. 2004, 45, 3881–
3884.
18. Weskamp, T.; Bohn, V. P. W.; Herrmann, W. A. J.
Organometal. Chem. 1999, 585, 348–352.
24. Panarello, A. Ethylene-bis(indenyl) catalytic complexes:
efficient synthesis, introduction of the active metal species
and immobilization. Ph.D. Thesis, Rutgers University,
Piscataway, NJ, 2005.