Highly diastereoselective reduction of ferrocene bis-imines with methyllithium and the formation of C2-symmetric Zr complexes

Article abstract of DOI:10.1039/b308360h

Alkylation of ferrocene bis-imines Fc[N=C(H)Ar]₂ (Ar = Ph, p-Tol) with MeLi was found to proceed in a highly diastereoselective fashion producing the C₂-symmetric ferrocene diamines Fc[NC(H)(Me)Ar] ₂-H₂ with 90% diastereomeric excess. This process allowed for the synthesis of C₂-symmetric zirconium complexes Fc[NC(Me)Ar]ZrBn₂.

Full text of DOI:10.1039/b308360h

Highly diastereoselective reduction of ferrocene bis-imines with
methyllithium and the formation of C2-symmetric Zr complexes†
Alexandr Shafir, Dorothea Fiedler and John Arnold*
Department of Chemistry, University of California, Berkeley, and the Chemical Sciences Division, Lawrence
Berkeley National Laboratory, Berkeley, CA, 94720-1460
Received (in West Lafayette, IN, USA) 19th July 2003, Accepted 28th August 2003
First published as an Advance Article on the web 19th September 2003
Alkylation of ferrocene bis-imines Fc[NNC(H)Ar]2 (Ar =
Ph, p-Tol) with MeLi was found to proceed in a highly
diastereoselective fashion producing the C2-symmetric fer-
rocene diamines Fc[NC(H)(Me)Ar]2–H2 with 90% diaster-
eomeric excess. This process allowed for the synthesis of C2-
symmetric zirconium complexes Fc[NC(Me)Ar]ZrBn2.
The most widely investigated aspects of olefin polymerization
chemistry relate to factors controlling stereochemistry of the
resulting polymer. In a site-control mechanism of Ziegler–Natta
olefin polymerization, the initiator geometry and symmetry
dictate the stereochemical outcome of each successive olefin
insertion.1 Based on the studies of metallocene complexes of
Group 4 metals, relationships have been established between
the symmetry of the initiator and the tacticity of the resulting
polymer. In particular, production of an isotactic polymer can
often be rationalized on the basis of the C2-symmetric structure
of the initiator.2 A large number of C2-symmetric metallocene
initiators have been reported. Their synthesis is often compli-
cated by the formation of a Cs-symmetric meso byproduct.
Separation of the isomers can be tedious and it is desirable to
generate the C2-symmetric rac isomer selectively. Since the
complex need not be enantiopure, the term “racemo-selective”
has been used to describe the diastereoselectivity of this
method.3 Although several highly diastereoselective methods of
producing C2-symmetric Group 4 metallocenes have been
developed, the process is not readily generalized.
Scheme 1
with p-MeC6H4CHO produced the corresponding bis-imine 2 in
95% yield.
Addition of MeLi to a THF suspension of 1 at 260 °C led to
the formation of a dark-blue mixture, which, upon warming to
above approx. 28 °C, afforded a red solution. Hydrolysis of the
mixture led to a yellow solution from which diamine 3 was
obtained as an orange air-sensitive solid in 92% yield (Scheme
2). The reduction of the bis-imine 2 proceeded in a similar
fashion and afforded, upon hydrolysis, the corresponding
diamine 4 in 78% yield.
Transition to non-metallocene initiators creates the possibil-
ity of using inherently chiral racemic C2-symmetric ligands for
the synthesis of C2-symmetric Group 4 metal complexes. A
number of non-metallocene systems of this type have already
been investigated as potential olefin polymerization catalysts.4
In particular, several C2-symmetric diamide Ti and Zr com-
Scheme 2
plexes incorporating a chiral backbone were reported.5–8
A
Although in principle the diamine 3 is expected to form as a
mixture of the trans and the cis isomers (Fig. 1), only one major
product was observed by 1H NMR spectroscopy.
recent report by Long and coworkers concerning the reduction
of a ferrocene bis-imine Fc[NNC(H)Ph] to the corresponding
diamine Fc(NCH2Ph)2–H2 using LiAlH49 raised the possibility
of using the related reaction with MeLi to produce a C2-
symmetric diamine. This report describes the highly diaster-
eoselective formation of racemic ferrocene diamines and the
subsequent conversion of these diamines to C2-symmetric Zr
complexes.
Several Schiff-base complexes of 1,1A-diaminoferrocene
were reported recently,9,10 including the preparation of a simple
bis-benzaldimine Fc[NNC(H)Ph]2 (1) through Schiff base
condensation of 1,1A-diaminoferrocene with benzaldehyde in
toluene.9 We found that conducting the condensation reaction in
EtOH resulted in the isolation of the analytically pure
burgundy-colored bis-imine 1‡ in 93% yield (Scheme 1). The
1H NMR spectrum of the product contained a set of two
ferrocene resonances (virtual triplets) at 4.59 and 4.10 ppm, a
pattern characteristic of C2v-symmetric ferrocene derivatives.
In an analogous fashion, the reaction of 1,1A-diaminoferrocene
The product displayed four ferrocene ABCD multiplets,
consistent with the formation of a ferrocene derivative with two
chiral substituents in the 1,1A-positions. The resonances of the
newly incorporated methyl group show coupling to the methine
hydrogen and appear as doublets. A closer analysis of the
spectrum revealed the presence of the minor isomer. The two
diastereomers are found in a 90 : 10 ratio, which corresponds to
a diastereomeric excess (de) of 80%. The 1H NMR spectrum of
diamine 4 showed the same ratio of two diastereomers.
† Electronic supplementary information (ESI) available: experimental
procedures and characterization data for all new compounds. See http://
www.rsc.org/suppdata/cc/b3/b308360h/
Fig. 1
2598
CHEM. COMMUN., 2003, 2598–2599
This journal is © The Royal Society of Chemistry 2003

Scheme 4
Scheme 3
selectivity was achieved in toluene. Reaction of 1 with
methyllithium in toluene at 260 °C produced a 95 : 5 mixture
of the trans and cis isomers (90% de). This result allowed for a
highly selective synthesis of C2-symmetric Group 4 metal
complexes. The sequence is summarized in Scheme 4 showing
the formation of a chiral racemic zirconium complex 5 with a
90% diastereomeric excess.
In summary, a method for the highly diastereoselective
formation of C2-symmetric racemic diaminoferrocenes is
described. The compounds are obtained through a reaction of
methyllithium with ferrocene bis-imines. Under the optimized
reaction conditions the formation of the trans-diastereomer
proceeded with a 90% diastereomeric excess. The diamino-
ferrocenes thus obtained were subsequently used to generate
C2-symmetric zirconium diamide complexes, whose chemistry
is presently under investigation. Additional work is in progress
to establish the origin of the selectivity of the reduction of bis-
imines with MeLi.
It is important to note that in the case of diamines 3 and 4 the
assignment of diastereoisomers as trans or cis is not straightfor-
5
ward. In each isomer the two (h -C5H4)NCH(Me)(Ar) groups
are related by a symmetry element: a mirror plane in the cis
isomer and a two-fold rotation axis in the trans isomer. The 1H
NMR patterns exhibited by the two diastereomers are, therefore,
indistinguishable.11 The question of the assignment of the major
diastereomer was addressed by the synthesis of zirconium
dibenzyl complexes (Scheme 3). Reaction of diamine 3 with
ZrBn4 afforded the desired zirconium dibenzyl complex
{Fc[NCH(Me)(Ph)]2}ZrBn2 (5) (Scheme 3). As expected, the
1H NMR analysis of the product shows the presence of two
diastereomers in a 90 : 10 ratio. This ratio mirrors the isomer
distribution in diamine 3.
The Cs and the C2 symmetrical zirconium dibenzyl com-
plexes can be easily distinguished by the 1H NMR pattern of the
benzylic methylene protons. In a C2-symmetric compound, the
two benzyl groups are equivalent, but the protons on the same
benzyl group are diastereotopic and appear as two doublets.
Conversely, the two benzyl groups in a Cs-symmetric complex
are distinct, but each group contains a pair of equivalent
benzylic protons (Fig. 2). In this case, the spectrum would
consist of two singlets. The major diastereomer of 5 contains
two doublets at 2.69 and 2.65 ppm (2J = 10.4 Hz) correspond-
ing to the benzylic hydrogen atoms. This pattern identifies this
isomer as the C2-symmetric trans diastereomer. Therefore, the
remaining 10% of the product is the Cs-symmetric cis
We thank the U.S. Department of Energy for funding.
Notes and references
‡ Characterization details for 1, 3, and 5. Compound 1: Mp. 126–128 °C. 1H
NMR (C6D6), d 8.38 (s, 2H, NNCH), 7.68 (m, 4H, o-Ph), 7.10–7.02 (m, 6H,
m-Ph and p-Ph), 4.59 (virt t, 4H, Fc), 4.10 (virt t, 4H, Fc). 13C NMR (C6D6),
d 157.9, 137.4, 130.2, 128.7, 128.3, 106.1 (Fcipso), 68.8 (Fc), 64.8 (Fc).
Anal. Calcd for C24H20FeN2: C, 73.48; H, 5.14; N, 7.14. Found: C, 73.39;
H, 5.16; N, 7.15%. 3: Major product (3-trans): 1H NMR (C6D6), d 7.34 (m,
4H, o-Ph), 7.20 (m, 4H, m-Ph), 7.10 (m, 2, p-Ph), 4.09 (qd, JH–H(Me) = 6.7
Hz, JH–H(NH) = 2.5 Hz, 2H, MeCH), 3.80 (m, 2H, Fc), 3.78 (m, 2H, Fc),
3.66 (m, 2H, Fc), 3.54 (m, 2H, Fc), 2.35 (d, J = 2.5 Hz, 2H, NH), 1.19 (d,
J = 6.7 Hz) ppm. 13C{1H} NMR (C6D6), d 146.5, 128.4, 128.2, 126.1,
109.4 (Fcispo), 63.4 (Fc), 63.3 (Fc), 57.6 (Fc), 57.4 (MeCH), 57.10 (Fc), 25.2
(MeCH) ppm. Anal. Calcd for C26H28FeN2: C, 73.59; H, 6.65; N, 6.60.
Found: C, 73.32; H, 6.45; N, 6.78. 5: 1H NMR (only the major trans
diastereomer is described) (C6D6), d 7.25–6.90 (m, 20H, Ar), 4.89 (q, 3J =
6.7 Hz, 2H, NCH), 3.77 (m, 2H, Fc), 3.75 (m, 2H, Fc), 2.90 (m, 2H, Fc),
2.81 (m, 2H, Fc), 2.66 (d, 2J = 10.4 Hz, 2H, ZrCH2), 2.62 (d, 2J = 10.4 Hz,
diastereomer.
The
zirconium
dibenzyl
complex
Fc[NCH(Me)(p-MeC6H4)]2ZrBn2 (6) formed from ZrBn4 and 4
again was shown by 1H NMR to contain mainly the trans
diastereomer.
The results obtained from these experiments indicate that the
reduction of ferrocene bis-imines 1 and 2 with methyllithium is
highly racemoselective. It is well established that high diaster-
eoselectivities are often achieved in the formation of vicinal
diamines through alkylation of glyoxal-derived bis-imines.11,12
With other bis-imines, the outcome of the alkylation varies from
the selective formation of the meso diamines13 to the selective
generation of the racemic diamines.14 However, the favorable
sense of the selectivity (rac) coupled with the proven ability of
diaminoferrocene-based complexes to polymerize ethylene
render the alkylation of ferrocene an attractive route to C2-
symmetric olefin polymerization catalysts.
2H, ZrCH2), 1.67 (d, 3J
40H40FeN2Zr: C, 69.04; H, 5.79; N, 4.03. Found: C, 68.70; H, 5.60; N,
3.79%.
= 6.7 Hz, 6H, Me) ppm. Anal. Calcd for
C
1 L. Resconi, L. Cavallo, A. Fait and F. Piemontesi, Chem. Rev., 2000,
100, 1253–1345.
2 G. W. Coates, Chem. Rev., 2000, 100, 1223–1252.
3 H. R. H. Damrau, E. Royo, S. Obert, F. Schaper, A. Weeber and H. H.
Brintzinger, Organometallics, 2001, 20, 5258–5265.
4 V. C. Gibson and S. K. Spitzmesser, Chem. Rev., 2003, 103,
283–315.
In order to further optimize the selectivity of the alkylation
step, solvent and temperature effects on the product distribution
were investigated. Conducting the reaction at room temperature
reduced the diastereomeric excess from 80% to 72%. The best
5 F. G. N. Cloke, T. J. Geldbach, P. B. Hitchcock and J. B. Love, J
Organomet. Chem., 1996, 506, 343–345.
6 P. N. O’Shaughnessy, K. M. Gillespie, C. Morton, I. Westmoreland and
P. Scott, Organometallics, 2002, 21, 4496–4504.
7 L. Ackermann and R. G. Bergman, Org. Lett., 2002, 4, 1475–1478.
8 T. I. Gountchev and T. D. Tilley, Inorg. Chim. Acta, 2003, 345,
81–88.
9 V. C. Gibson, N. J. Long, E. L. Marshall, P. J. Oxford, A. J. P. White and
D. J. Williams, Dalton Trans., 2001, 1162–1164.
10 A. Shafir, D. Fiedler and J. Arnold, Dalton Trans., 2002, 555–560.
11 W. L. Neumann, M. M. Rogic and T. J. Dunn, Tetrahedron Lett., 1991,
32, 5865–5868.
12 R. Bloch, Chem. Rev., 1998, 98, 1407–1438.
13 K. Henrick, P. M. Judd, P. G. Owston, R. Peters, P. A. Tasker and R. W.
Turner, Chem. Commun., 1983, 1253.
14 M. G. Gardiner and C. L. Raston, Inorg. Chem., 1995, 34,
4206–4212.
Fig. 2
CHEM. COMMUN., 2003, 2598–2599
2599