Ferrocenophanes with gallium and silicon as alternating bridges

Article abstract of DOI:10.1039/c2cc32967k

[1.1]Ferrocenophanes with gallium and silicon in bridging positions have been prepared in yields of 29 and 41%, respectively. From the same reactions, polymer-containing fractions were isolated (31% in each case) and shown to be comprised of linear and cy

Full text of DOI:10.1039/c2cc32967k

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2012, 48, 7823–7825
www.rsc.org/chemcomm
COMMUNICATION
Ferrocenophanes with gallium and silicon as alternating bridgesw
Bidraha Bagh,a Nora C. Breit,a Joe B. Gilroy,b Gabriele Schattec and Jens Mu
¨
ller*a
Received 25th April 2012, Accepted 7th June 2012
DOI: 10.1039/c2cc32967k
[1.1]Ferrocenophanes with gallium and silicon in bridging positions
have been prepared in yields of 29 and 41%, respectively. From the
same reactions, polymer-containing fractions were isolated (31%
in each case) and shown to be comprised of linear and cyclic
species with up to 16 ferrocene units (MALDI-TOF analysis).
During the last two decades, [1]ferrocenophanes ([1]FCPs;
Scheme 1) were developed as a class of precursors for the
preparation of well-defined metallopolymers.1 The most researched
strained sandwich compounds are silicon-bridged [1]FCPs, which
were also the first reported [1]FCPs.2 The related [1.1]FCPs
(Scheme 1) are an even older class of compounds, which were
described as early as 1956.3 During the 1980s, carbon-bridged
[1.1]FCPs were intensively investigated as it was found that the
syn conformer of the CH2-bridged species catalyzed the
formation of dihydrogen in aqueous acidic solutions.4 To
date, the family of [1.1]FCPs has a significant number of
members with heteroatom-bridged species known for B, Al,
Ga, In, Si, Sn, Pb, P, As, S, Zn, and Hg.5
Scheme 2 Linear and cyclic poly(ferrocenes) with alternating bridges.5
(3m) with up to 20 ferrocene units.5 Cyclic poly(ferrocenes)
of this size are very rare and only photocontrolled ring-
opening polymerization (ROP) of the Me2Si-bridged [1]FCP
(Scheme 1) has yielded larger macrocycles (with more than 40
repeating units) to date.6
Metal-containing polymers have shown utility as precursors
to functional ceramic materials. For example, cross-linked
poly(ferrocenylsilane)-based materials yielded shape-retaining
ceramics with tunable magnetic properties.7 Similarly, pyrolysis
of short cylindrical micelles with a cross-linked polyisoprene
shell and a poly(ferrocenylsilane) core resulted in nanoscale
magnetic ceramics.8 More recently, low-molecular-weight
ferrocene-platinum-containing polymers were employed to
pattern surfaces with FePt alloy nanoparticles for ultrahigh-
density magnetic data storage applications.9 In each case, the
predetermined composition of the polymers employed played
an important role in the properties of the resulting ceramics.
Although recent results in this area have been impressive, there
remains a need for synthetic methods that allow for variation
in the composition of pre-ceramic metal-containing polymers.
For example, pyrolysis of metal-containing polymers containing
gallium and iron may lead to magnetostrictive ceramic
materials that undergo changes in shape or dimensions during
magnetization.10
Recently, we developed a synthetic methodology that allowed
the preparation of the first [1.1]FCPs bridged by different
elements.5 The first silastanna[1.1]ferrocenophanes were syn-
thesized (Scheme 2), but could only be obtained in low yields
[21 (R = Et, R0 = Me): 3%; 21 (R = R0 = Me): 7%].
Furthermore, [1.1.1.1]FCPs (22) with alternating silicon and
tin in bridging positions were isolated and MALDI-TOF mass
analysis showed the presence of cyclic (2n) and linear polymers
Scheme 1 [1]Ferrocenophanes and [1.1]ferrocenophanes.
a Department of Chemistry, University of Saskatchewan,
110 Science Place, Saskatoon, Saskatchewan S7N 5C9, Canada.
E-mail: jens.mueller@usask.ca; Fax: (+)306 966 4730;
Tel: (+)306 966 2466
Within this report we present a novel synthetic method
towards unsymmetrically bridged [1.1]FCPs containing silicon
and gallium bridging moieties, which also yield linear and
cyclic polymers of similar composition. These results comple-
ment our efforts to develop gallium- and aluminum-containing
metallopolymers through ROP methodologies.11
b School of Chemistry, University of Bristol, Bristol BS8 1TS, UK
c Saskatchewan Structural Sciences Centre,
University of Saskatchewan, 110 Science Place, Saskatoon,
Saskatchewan S7N 5C9, Canada
w Electronic Supplementary Information (ESI) available: Details of
experimental procedures and characterization methods, NMR spectra,
and crystallographic details. CCDC 878280 (6a) and 878281 (6b). For
ESI and crystallographic data in CIF format see DOI: 10.1039/
c2cc32967k
As illustrated in Scheme 3, lithiation of the Me2Si-
bridged dibromide 1Me followed by the addition of gallium
c
This journal is The Royal Society of Chemistry 2012
Chem. Commun., 2012, 48, 7823–7825 7823

View Article Online
Scheme 4 anti-to-anti Isomerization of the known Me2Si-bridged
[1.1]FCP, resulting in swapping of positions of related H0 and H0 0 atoms.
Scheme 3 Synthesis of silagalla[1.1]ferrocenophanes 6a and 6b.
which have been coined ‘‘molecular acrobats’’.4,15 However,
the known Ar0E-bridged [1.1]FCPs (E = Al, Ga, In) do not
exhibit higher symmetries on time-average in proton NMR
spectra (500 MHz). For the indium-bridged species an anti-to-
anti isomerization still occurs at ambient temperature, which
was shown by exchange proton NMR spectroscopy (EXSY).16
For group-13-bridged [1.1]FCPs equipped with intra-
molecularly coordinating ligands, the anti-to-anti isomeriza-
tion must involve a breakage of the E–N donor bond, rotation
of the ligand about the E–Cipso bond, and reformation of the
E–N donor bond which is accompanied by an inversion at the
group 13 element. Such a dynamic process is faster for indium
compared with aluminum or gallium species, as indium forms
the weakest donor bond with nitrogen within the series.
Against this background, we measured 1H NMR spectra of
6a and 6b in [D8]toluene in the temperature range of r.t. to
80 1C (500 MHz), but could not find any indication of
fluxional behavior. Thus, we conclude that the gallium-
containing moiety does not allow for a fast isomerization,
even though the Me2Si linkage is very flexible.
Fig. 1 Molecular structure of 6b with thermal ellipsoids at the 50% prob-
ability level. Hydrogen atoms are omitted for clarity. Selected atom-atom
distances [A] for 6b (see ESI for 6a): Ga1-N1 2.1626(16); Ga1-C1 1.973(2);
Ga1-C20 1.958(2); Ga1-C35 1.9566(19); Fe1ꢀꢀꢀFe2 5.3147(4).
dichlorides (4 or 5), equipped with non-encumbered ligands,
12
gave the targeted silagalla[1.1]ferrocenephanes 6a and 6b.z
The isolated yields of 29 (6a) and 41% (6b) are significantly
higher than those of their silicon-tin counterparts 21 (Scheme 2;
3 and 7%).5 Single-crystal X-ray analysis revealed that com-
pounds 6a and 6b exist as anti isomers in the solid state (Fig. 1
and Scheme 1). This is expected as their symmetrically bridged
cousins, the disila[1.1]ferrocenophane (ERx = SiMe2)13 and the
digalla[1.1]ferrocenophane (ERx = GaAr0),14 exhibit the same
conformation. The bond lengths around the two bridging
elements are very similar to those of the symmetrically bridged
species.13,14 Species 6b is unsymmetrically bridged by a short Si
linkage (C25ꢀ ꢀ ꢀC30 3.120(3) A) and a slightly longer Ga linkage
(C20ꢀ ꢀ ꢀC35 3.347(3) A). The Cp rings of each ferrocenediyl unit
deviate from coplanarity: tilt angles between least square planes
of the C atoms of the Cp rings are 5.07(12) (Fe1) and 4.35(12)1
(Fe2). The molecular structure of 6a in the solid state is very
similar to that of 6b (see ESI).
Cyclic voltammetry of 6a and 6b at r.t. revealed the presence
of two well-resolved oxidation waves, showing an expected
sequential oxidation of the two iron centers (Fig. 2). The separa-
tion of these waves, DE10, provides a measure of the interaction
between the two iron-redox centers.17 Species 6a and 6b are
hybrids of the symmetrically bridged disila[1.1]ferrocenophane
(ERx = SiMe2) and digalla[1.1]ferrocenophane (ERx = GaAr0)
and it is not surprising that their DE10 of 0.27 V falls between that
of 0.25 V (ERx = SiMe2)13b and 0.30 V (ERx = GaAr0).14
Proton NMR spectra of reaction mixtures from which the
[1.1]FCPs 6a and 6b were extracted into hexane (29 and 41%;
Scheme 3) showed broad peaks indicating the formation of
oligomers or polymers. The hexane insoluble fractions were
purified by precipitation from toluene solutions into hexane,
resulting in isolated yields of 31% in both cases. GPC analysis of
these fractions, 6ax and 6bx, showed broad molecular weight
distributions and average molecular weights Mw of 3.08 kD (6ax)
and 7.05 kD (6bx). A clearer picture of the character of these
mixtures could be uncovered by MALDI-TOF mass analysis,
Species 6a and 6b both show similar pattern in their NMR
1
spectra. For example, the H NMR spectra show 8 signals of
equal intensity for all 16 Cp protons revealing the presence of a
two-fold symmetry element. This is consistent with the molecular
structures found in the solid state, if one considers that
inversion of the envelope conformation of the chelating five-
membered rings (Ga1-C1-C2-C7-N1) occurs fast in solution,
resulting in time-averaged Cs-symmetrical species. The known
dimethylsila[1.1]ferrocenophane13 is a fluxional molecule in
solution, exhibiting fast anti-to-anti isomerization (Scheme 4).
This degenerate isomerization of two C2h-symmetrical
species creates a pseudo-mirror plane (sh) resulting in D2h
symmetrical molecules on time-average and 2 instead of the
expected 4 signals for all Cp protons are observed. Such a
fluxional behavior is characteristic of [1.1]FCPs and usually
observed for syn- as well as for anti isomers. This dynamic
behavior was first described for the carbon-bridged species,
Fig. 2 Cyclic voltammogram of 6a (1 mM) in CH2Cl2 (0.1 M
NBu4PF6; scan rate = 50 mV sꢁ1; E10 = 0.121 and 0.149 V).
c
7824 Chem. Commun., 2012, 48, 7823–7825
This journal is The Royal Society of Chemistry 2012

View Article Online
of 1Me so that chain growth termination can be suppressed,
as well as the evaluation of the polymers described as pre-
cursors to magnetostrictive ceramic materials.
We acknowledge the NSERC of Canada for support
(DG for J.M.; PDF for J.B.G.), the EU for a Marie Curie
PDF (J.B.G.), Dr C. L. Lund (LANXESS, London, ON) and
S. Dey (U of Saskatchewan) for contributions, and Prof.
I. Burgess (U of Saskatchewan) and Prof. I. Manners (U of
Bristol) for making instruments available for our studies.
Notes and references
z To improve solubilities of [1.1]FCPs, the Ar0 ligand (2-Me2NCH2C6H4)
was equipped with a p-Me3Si group. This tactic had been applied
successfully to [1.1]metallarenophanes by introducing a p-tBu
group (see ref. 12). However, the solubilities of 6a and 6b are very
similar.
1 (a) D. A. Foucher, B.-Z. Tang and I. Manners, J. Am. Chem. Soc.,
1992, 114, 6246–6248; (b) D. E. Herbert, U. F. J. Mayer and
I. Manners, Angew. Chem., Int. Ed., 2007, 46, 5060–5081;
(c) V. Bellas and M. Rehahn, Angew. Chem., Int. Ed., 2007, 46,
5082–5104.
2 A. G. Osborne and R. H. Whiteley, J. Organomet. Chem., 1975,
101, C27–C28.
3 A. N. Nesmeyanov and I. I. Kritskaya, Bull. Acad. Sci. USSR, Div.
Chem. Sci. (Engl. Transl.), 1956, 5, 243–244.
Fig. 3 MALDI-TOF mass spectrum of 6ax (* indicates unassigned
peaks).
which showed cyclic and linear polymers. Fig. 3 depicts the
mass spectrum for mixture 6ax with four different series
of species. Cyclic ferrocenophanes (6an) with up to 12 ferro-
cenediyl moieties (n = 6) were detected. Furthermore, three
series of linear species were found: one series with only Cp end
groups (7am; m = 1–8), one with one GaAr0Cl end group
(8am; m = 1–7), and one with one GaAr0Br end group (9am;
m = 1–7). Similar series of species were detected for the
mixture 6bx, showing compounds with up to 14 ferrocenediyl
moieties; however, only the series of cyclic species was less
pronounced than for 6ax (see ESI).
4 U. T. Mueller-Westerhoff, Angew. Chem., Int. Ed. Engl., 1986, 25,
702–717.
5 B. Bagh, N. C. Breit, S. Dey, J. B. Gilroy, G. Schatte, K. Harms
and J. Muller, Chem.–Eur. J., 2012, DOI: 10.1002/chem.
201200953.
¨
6 D. E. Herbert, J. B. Gilroy, W. Y. Chan, L. Chabanne, A. Staubitz,
A. J. Lough and I. Manners, J. Am. Chem. Soc., 2009, 131,
14958–14968.
7 M. J. MacLachlan, M. Ginzburg, N. Coombs, T. W. Coyle,
N. P. Raju, J. E. Greedan, G. A. Ozin and I. Manners, Science,
2000, 287, 1460–1463.
The presence of bromide-containing end groups (Fig. 3;
series 9am) was unexpected. These species must have formed
from the respective chlorides (series 8am) through a Cl/Br
exchange reaction. Recently, we discovered that such a Cl/Br
exchange happened during the course of the synthesis of the
intramolecularly stabilized gallium compound MxGaCl2.18
A similar exchange was reported in the literature for the
Mes*GaCl2, where the authors speculated that unreacted starting
compound Mes*Br was the direct source of bromide.19 In the
course of the synthesis of MxGaCl2, we found some indication
that LiBr reacted with MxGaCl2 to give MxGaClBr and
MxGaBr2. Therefore, we speculate that in the reaction
mixtures (Scheme 3) LiBr formed and acted as the reagent
for the Cl/Br exchange. In analogy to the well-known reaction
of tBuLi and tBuBr, some fraction of species with lithiated
Cp groups might have reacted with the formed nBuBr to give
LiBr, butene and protonated Cp end groups. Alternatively,
substitution could occur to give LiBr and butylated Cp;
however, butyl-containing compounds were not detected.
The new methodology described here has allowed the
synthesis of the first examples of poly(ferrocene)s with alter-
nating silicon and gallium in bridging positions. Such species
would be very difficult to obtain through ROP of respective
sila- and galla[1]ferrocenophanes: the required gallium species
are unknown and, in addition, a perfect control over the
copolymerization would be needed. The discovered Cl/Br
exchange shows that unwanted side reactions occurred, which
probably lead to chain growth termination. Future activities
will be concentrated on further optimizations of the metallation
8 (a) X. S. Wang, A. Arsenault, G. A. Ozin, M. A. Winnik and
I. Manners, J. Am. Chem. Soc., 2003, 125, 12686–12687;
(b) X. Wang, K. Liu, A. C. Arsenault, D. A. Rider, G. A. Ozin,
M. A. Winnik and I. Manners, J. Am. Chem. Soc., 2007, 129,
5630–5639.
9 Q. Dong, G. J. Li, C. L. Ho, M. Faisal, C. W. Leung, P. W. T.
Pong, K. Liu, B. Z. Tang, I. Manners and W. Y. Wong, Adv.
Mater., 2012, 24, 1034–1040.
10 J. Atulasimha and A. B. Flatau, Smart Mater. Struct., 2011,
20, 043001.
11 (a) B. Bagh, G. Schatte, J. C. Green and J. Muller, J. Am. Chem.
¨
Soc., 2012, 134, 7924–7936; (b) B. Bagh, J. B. Gilroy, A. Staubitz
and J. Muller, J. Am. Chem. Soc., 2010, 132, 1794–1795.
¨
12 C. L. Lund, J. A. Schachner, I. J. Burgess, J. W. Quail, G. Schatte
and J. Muller, Inorg. Chem., 2008, 47, 5992–6000.
¨
13 (a) J. W. Park, Y. S. Seo, S. S. Cho, D. M. Whang, K. M. Kim and
T. Y. Chang, J. Organomet. Chem., 1995, 489, 23–25;
(b) D. L. Zechel, D. A. Foucher, J. K. Pudelski, G. P. A. Yap,
A. L. Rheingold and I. Manners, J. Chem. Soc., Dalton Trans.,
1995, 1893–1899.
14 J. A. Schachner, G. A. Orlowski, J. W. Quail, H.-B. Kraatz and
J. Muller, Inorg. Chem., 2006, 45, 454–459.
¨
15 (a) M. Lowendahl, O. Davidsson and P. Ahlberg, J. Chem. Res.,
¨
¨
Synop., 1993, 40–41; (b) J.-M. Lowendahl and M. Hakansson,
¨
Organometallics, 1995, 14, 4736–4741.
16 J. A. Schachner, C. L. Lund, I. J. Burgess, J. W. Quail, G. Schatte
and J. Muller, Organometallics, 2008, 27, 4703–4710.
¨
17 E10
=
12(Epa + Epc); see A. J. Bard and L. R. Faulkner,
Electrochemical Methods, John Wiley & Sons, Inc., New York,
2nd edn, 2001.
18 Mx: 2,4-di-tert-butyl-6-(dimethylamino)phenyl; see M. Yoshifuji,
M. Hirano and K. Toyota, Tetrahedron Lett., 1993, 34,
1043–1046.
19 S. Schulz, S. Pusch, E. Pohl, S. Dielkus, R. Herbst-Irmer, A. Meller
and H. W. Roesky, Inorg. Chem., 1993, 32, 3343–3346.
c
This journal is The Royal Society of Chemistry 2012
Chem. Commun., 2012, 48, 7823–7825 7825