57Fe-M?ssbauer spectroscopic study of monomeric and polymeric ferrocenyl- and octamethylferrocenyl-substituted ethynes

Article abstract of DOI:10.1016/S0022-328X(00)00280-1

A series of four monomeric and four polymeric ferrocenyl- and octamethylferrocenyl-substituted 1-alkynes, ethynylferrocene (1) [1271-47-2], 2-ethynylphenylferrocene (2) [1292-14-4], [2-[ethynylphen-2-yl]ethen-1-yl]-1′,2,2′,3,3′,4,4′,5- octamethylferrocene (3), novel [2-[ethynylphen-2-yl]ethen-1-yl]-1′,2,2′,3,3′,4,4′,5- octamethylferricinium hexafluorophosphate (4) and poly(ethynylferrocene) (5) [33410-56-9], poly(2-ethynylphenylferrocene) (6) [205865-61-8], poly([2-[ethynylphen-2-yl]ethen-1-yl]-1′,2,2′,3,3′,4,4′, 5-octamethylferrocene) (7), and novel poly([2-[ethynylphen-2-yl]ethen-1-yl]-1′,2,2′,3,3′,4,4′, 5-octamethylferricinium hexafluorophosphate) (8), have been studied by temperature-dependent ⁵⁷Fe-M?ssbauer spectroscopy. The relevant M?ssbauer parameters have been determined and are compared with the data obtained for the model compound octamethylferrocene (OMF). The data obtained by ⁵⁷Fe-M?ssbauer spectroscopy reflect a small decrease of s-electron density from the ferrocene moiety to the conjugated backbone, as is demonstrated by the smaller IS of the monomers 1-3 in comparison with the corresponding polymers 5-7. As may be deduced from the temperature dependence of the recoil-free fraction and the calculated lattice temperatures, the polymers are much 'softer' (within the definitions of M?ssbauer spectroscopy) than the corresponding monomers. Finally, an unusual phase transition has been observed in OMF some 80°C below the melting point, which appears to lead to a large, sudden, and unexpected increase in the mean-square-amplitude of vibration of the metal atom in methyl substituted ring systems. This anomalous T-dependence of the recoil-free fraction, which is associated with the presence of the ring methyl groups, is not observed when one (or more) of the ring protons is replaced by a sterically bulky side chain, as in 1-3.

Full text of DOI:10.1016/S0022-328X(00)00280-1

www.elsevier.nl/locate/jorganchem
Journal of Organometallic Chemistry 612 (2000) 1–8
57Fe-Mo¨ssbauer spectroscopic study of monomeric and polymeric
ferrocenyl- and octamethylferrocenyl-substituted ethynes
Herwig Schottenberger a, Michael R. Buchmeiser b, Rolfe H. Herber c,*
a Institute of General, Inorganic and Theoretical Chemistry, Uni6ersity of Innsbruck, Innrain 52 a, A-6020 Innsbruck, Austria
b Institute of Analytical Chemistry and Radiochemistry, Uni6ersity of Innsbruck, Innrain 52 a, A-6020 Innsbruck, Austria
c Racah Institute of Physics, Hebrew Uni6ersity, 91904 Jerusalem, Israel
Received 11 January 2000; accepted 19 April 2000
Abstract
A series of four monomeric and four polymeric ferrocenyl- and octamethylferrocenyl-substituted 1-alkynes, ethynylferrocene (1)
[1271-47-2], 2-ethynylphenylferrocene (2) [1292-14-4], [2-[ethynylphen-2-yl]ethen-1-yl]-1%,2,2%,3,3%,4,4%,5-octamethylferrocene (3),
novel [2-[ethynylphen-2-yl]ethen-1-yl]-1%,2,2%,3,3%,4,4%,5-octamethylferricinium hexafluorophosphate (4) and poly(ethynylferrocene)
(5) [33410-56-9], poly(2-ethynylphenylferrocene) (6) [205865-61-8], poly([2-[ethynylphen-2-yl]ethen-1-yl]-1%,2,2%,3,3%,4,4%,5-oc-
tamethylferrocene) (7), and novel poly([2-[ethynylphen-2-yl]ethen-1-yl]-1%,2,2%,3,3%,4,4%,5-octamethylferricinium hexafluorophos-
phate) (8), have been studied by temperature-dependent 57Fe-Mo¨ssbauer spectroscopy. The relevant Mo¨ssbauer parameters have
been determined and are compared with the data obtained for the model compound octamethylferrocene (OMF). The data
obtained by 57Fe-Mo¨ssbauer spectroscopy reflect a small decrease of s-electron density from the ferrocene moiety to the
conjugated backbone, as is demonstrated by the smaller IS of the monomers 1–3 in comparison with the corresponding polymers
5–7. As may be deduced from the temperature dependence of the recoil-free fraction and the calculated lattice temperatures, the
polymers are much ‘softer’ (within the definitions of Mo¨ssbauer spectroscopy) than the corresponding monomers. Finally, an
unusual phase transition has been observed in OMF some 80°C below the melting point, which appears to lead to a large, sudden,
and unexpected increase in the mean-square-amplitude of vibration of the metal atom in methyl substituted ring systems. This
anomalous T-dependence of the recoil-free fraction, which is associated with the presence of the ring methyl groups, is not
observed when one (or more) of the ring protons is replaced by a sterically bulky side chain, as in 1–3. © 2000 Elsevier Science
B.V. All rights reserved.
Keywords: Ferrocenyl polymers; Octamethyl ferrocenyl polymers; T-dependent; Mo¨ssbauer spectroscopy; Ethynes
polymerizations using well-defined Schrock-type
metathesis catalysts [21–31]. In general, the actual
number of conjugated double bonds (usually expressed
as Neff) in these materials strongly depends on the
electronic, as well as steric, nature of the substituents in
the starting alkyne. Additionally, 1,3-interactions were
found to play an important role [30]. Conjugated
organometallic materials (such as polyenes), which con-
tain pendant metallocenes, are believed to have great
potential for high second- and third-order nonlinearities
for three main reasons: (i) metallocenes show M–L
(metal-to-ligand) and L–M (ligand-to-metal) charge
transfer absorptions in the UV–vis region, transitions
that are often associated with large second order non-
linearities; (ii) metallocenes often have high transition
dipole moments, which lead to high second-order non-
1. Introduction
Polymeric metallocenes, including their cationic
forms [1], have attracted significant interest during the
last few years. In general, fully conjugated polymers
based on poly-ynes [2], poly-enynes [3] or poly-enes [4]
represent interesting materials for many applications in
the fields of organic conductivity [5–9], optoelectronics
and photonics [10–16]. Besides the use of substituted
cycloactatetraenes [17–20], another attractive access to
conjugated polyenes lies in the use of terminal acetyle-
nes. These compounds are polymerized preferably by
* Corresponding author. Tel.: +972-2-6584244; fax: +972-2-
6586347.
E-mail address: herber@vms.huji.a.il (R.H. Herber).
0022-328X/00/$ - see front matter © 2000 Elsevier Science B.V. All rights reserved.
PII: S0022-328X(00)00280-1

2
H. Schottenberger et al. / Journal of Organometallic Chemistry 612 (2000) 1–8
linearities, and (iii) metallocenes are often stable in
oxidation states that differ by one electron, and there-
fore can give rise to charged polymers. Due to the fact
that ferrocene- and especially octa- and nonamethylfer-
rocene derivatives are known to possess a pronounced
Lewis base character, a significant charge transfer from
the metallocene moiety to the conjugated backbone
may be expected. In the case where such an electron
transfer takes place, and where the corresponding
polyene possesses sufficient coplanarity, bathochromic
shifts in absorption associated with large values for the
extinction coefficient (m) may be observed [29,30]. Nev-
ertheless, besides cyclic voltammetry (CV) there exist
few analytical tools which are capable of providing
quantitative information concerning the electron donat-
ing part of the metallocene or which at least allow an
estimate of the actual electronic influence of such a
pendant metallocene unit onto a conjugated polymer.
Recently, 57Fe-Mo¨ssbauer spectroscopy has been
used to elucidate the significance and extent of direct
interactions between the iron atom and the correspond-
ing bridge in ring-tilted [2]ferrocenophane systems [32]
as well as in the investigation of ferrocenyl-substituted
allenes and cummulenes [33,34]. In this contribution,
the effect of the iron atom on the transformation of the
monomeric compound into a polymer has been investi-
gated by means of 57Fe-Mo¨ssbauer spectroscopy, par-
ticularly the differences between monosubstituted and
nonasubstituted monomeric and polymeric ferrocenes,
as well as the lattice-dynamical consequences of the
polymerization. In addition, the Mo¨ssbauer data of
related polycationic species have been acquired, and will
be described in detail below.
2. Experimental
2.1. Monomers and polymers
The ethynyl group containing monomers discussed in
the present study are summarized graphically in Fig. 1,
and the polymers in Fig. 2, and their preparation is
discussed below. In addition, octamethyl ferrocene
(OMF) [59568-28-4] was repeatedly recrystallized from
hexane and showed a melting point of 431 K. Com-
pounds 1 [1271-47-2], 2 [1292-14-4] and 3 as well as
compounds 5 [33410-56-9], 6 [205865-61-8] and 7 were
prepared as described previously [29,30,35,36]. Poly-
mers 5–7 were prepared in benzene at room tempera-
ture (r.t.) from monomers 1–3 using Mo(N-2,6-
Me2C6H3)(CHCMe2Ph)(OCMe(CF3)2)2 as an initiator.
End-capping of the living polymers was achieved using
ferrocenecarbaldehyde. The general properties of these
polymers have previously been reported [35] and a
summary of the relevant polymer data is included in
Table 2. The purity of monomers 1, 2, and 3 was
checked by NMR and elemental analysis. Polymers 5–7
were fully characterized by GPC, NMR and FTIR.
2.2. [2-[Ethynylphen-2-yl]ethen-1-yl]-
1%,2,2%,3,3%,4,4%,5-octamethylferricinium
hexafluorophosphate (4)
Fig. 1. Monomers 1–4.
Compound 3 (100 mg, 0.236 mmol) was suspended in
acetonitrile (5 ml) and a solution of ferricinium hex-
afluorophosphate (120 mg, 0.230 mmol) in 2 ml of
acetonitrile was added. The reaction mixture was stirred
at r.t. for 1 h, diethyl ether (10 ml) was added and the
precipitate was collected by means of filtration. Exten-
sive washing with diethyl ether yielded 4 as a black
solid. IR (KBr): 2295 w, 1150 m, 960 w, 850 b, vs (PF6),
750 m. Anal. Calc. for C28H32F6FeP (Mw=569.150): C,
59.04; H, 5.67; N, 0.00. Found: C, 58.6; H, 5.4; N,
0.0%.
2.3. Poly([2-[ethynylphen-2-yl]ethen-1-yl]-
1%,2,2%,3,3%,4,4%,5-octamethylferri-cinium
hexafluorophosphate) (8)
This compound was prepared starting from 7 accord-
ing to the procedure described for the preparation of 4,
Fig. 2. Polymers 5–8.

H. Schottenberger et al. / Journal of Organometallic Chemistry 612 (2000) 1–8
3
Table 1
Mo¨ssbauer data for monomers 1–4
Compound
Ferrocene
Octamethyl-
ferrocene
Ethynyl-
ferrocene (1) ferrocene (2)
2-Ethynylphenyl- [2-[Ethynylphen-2-yl]- [2-[Ethynylphen-2-yl]-
ethen-1-yl]-octame-
ethen-1-yl]-octame-
thylferricinium PF6−
(4)
thylferrocene (3)
IS (90) (mm s−1
)
0.53190.001 0.52090.003 0.52490.003 0.51690.001
0.50790.003
2.39590.002
Non-linear
0.6290.02
QS (90) (mm s−1
)
0.249190.001 2.57790.00
3.29290.074 4.24190.080
2.34390.001 2.35290.002
3.8690.09
dIS/dT (mm s−1 K×104)
dQS/dT (mm s−1 K×104)
-d ln (A)/dT (K−1×103)
Meff (Da)
1.86190.276
8.5090.42
126
2.9390.13
5.0590.37
98.1
125
431
7.97+6 0.22
7.9390.52
108
102
325
105
99
344–349
[
M
(K)
85
M.p. (K)
Table 2
Mo¨ssbauer data for polymers 5–8 a
Compound
Poly(ethynyl-
ferrocene)70 (5)
Poly(2-ethynylphenyl- Poly([2-[ethynylphen-2-yl]ethen- Poly([2-[ethynylphen-2-yl]ethen-1-
ferrocene)55 (6)
1-yl]-octamethylferrocene)25 (7) yl]-octamethylferricinium PF-6)2+5 (8)
Polymer data
Mw=15 000 (D) Mw=19 200 (D)
Mw=9800 (D)
PDI=1.90
Mw=14 555 (Calc.) (D)
PDI=n.d.
PDI=1.50
umax=440
PDI=1.22
umax=511 b (nm)
umax=500 b (nm)
umax=330, 390 (shoulder) c (nm)
(shoulder) b (nm)
0.53590.002
2.39690.002
IS (90) (mm s−1
)
0.53190.001
2.40590.001
3.8290.07
1.5690.15
11.0190.21
109
0.51390.001
2.39690.004
3.3390.18
Non-linear
14.290.7
ꢀ137
0.57690.010
QS (90) (mm s−1
)
dIS/dT (mm s−1 K×104) 3.59790.133
dQS/dT (mm s−1 K×104) 1.93690.159
−d ln (A)/dT (K−1×103) 9.70990.287
Meff (Da)
(K)
116
[
83.2
80.4
ꢀ63
M
M.p. (K)
a Fc=ferrocene.
b In THF.
c In methylene chloride.
using a solvent mixture of methylene chloride and
acetonitrile (1:1) as a solvent. IR (KBr): 2295 w, 1152
m, 966 w, 845 b, vs (PF6), 754 m.
lite orientations with respect to the optical axis of the g
ray experiment, the microcrystalline powders were thor-
oughly mixed with roughly equal weight amounts of
BN prior to packing in sample holders.
2.4. 57Fe Mo¨ssbauer spectroscopy
57Fe Mo¨ssbauer spectroscopy was carried out using a
source of 57Co in a rhodium matrix (Isoflex, Moscow),
as described previously [34]. All IS are reported with
respect to the centroid of a r.t. spectrum of an 18 mg
cm−2 metallic iron absorber, which was also used for
spectrometer calibration. For T in the range 90–200 K,
an excess of 106 counts per chanel (of 256) were
recorded in each spectrum. At higher temperatures,
where the f-factor becomes very small, up to 8×106
counts were recorded to reduce the statistical errors
inherent in the data. Experimental errors of the re-
ported parameters were calculated using standard prop-
agation of error techniques and reflect a91|
confidence level. In order to maximize random crystal-
3. Results and discussion
The 57Fe Mo¨ssbauer spectra of all of the compounds
involved in the present study — except for the
cationic complexes 4 and 8 — consist of well resolved
doublets at all temperatures over the range 90–300 K,
and a typical spectrum is shown in Fig. 3. The most
important parameters, which may be extracted from
57Fe Mo¨ssbauer spectroscopy are the isomeric shift (IS),
the quadrupole splitting (QS) and the corresponding
temperature-dependent values, dIS/dT, dQS/dT, as well
as −d ln A/dT. Moreover, using these parameters it is
possible to extract [37] an ‘effective vibrating mass’,
Meff, and a ‘Mo¨ssbauer lattice temperature’, [M, which

4
H. Schottenberger et al. / Journal of Organometallic Chemistry 612 (2000) 1–8
help to characterize the compound in question. The
isomer shift can be correlated with the s-electron den-
sity at the metal atom nucleus; an increase in the IS
implies a decrease in the electron density, and vice
versa. Consequently, any Lewis-base-related change in
the electron density from the ferrocenyl moiety to the
conjugated backbone should be reflected in the IS. The
detailed results of the Mo¨ssbauer measurements for
both the monomers and polymers will be briefly consid-
ered below, and the pertinent parameters are summa-
rized in Table 1. The Mo¨ssbauer data for ferrocene,
included in Table 1 for comparison, have been reported
in detail in an earlier publication [32].
3.1. Octamethyl ferrocene (OMF)
The 57Fe-Mo¨ssbauer parameters of this compound
have been investigated in considerable detail, since
OMF can serve as a reference (model) compound with
respect to the other monomers and polymers included
in this study, as well as the fact that OMF displays a
number of interesting properties in its own right.
A typical spectrum of 1 (at 90 K) is shown in Fig. 3,
and the 90 K hyperfine parameters are included in
Table 1. The temperature dependence of the IS is
shown in Fig. 4, in which a small amount of curvature
is evident. However, fitting these data to a linear regres-
sion (−dIS/dT=(4.24190.080)×10−4 mm s−1 K−1
with a correlation coefficient of 0.996 for 26 data
points) leads to a value of Meff=98 Da, in reasonable
agreement with data for the ring-substituted ferrocenes
reported earlier [37]. The temperature dependence of
the QS hyperfine interaction can be fitted by a reason-
Fig. 3. Mo¨ssbauer spectrum of octamethyl ferrocene (OMF) at 90 K.
The intensity asymmetry is temperature independent and arises from
preferential crystallite orientation with respect to the optical axis.
ably linear regression (−dQS/dT=(2.93+
6 0.13)×
10−4 mm s−1 −1, cc:0.98 for 25 data points) over the
K
entire temperature range. While these data are unexcep-
tional for ring-substituted ferrocenes, the temperature
dependence of the recoil-free fraction (as reflected in the
area under the resonance curve for a thin absorber) is
markedly anomalous, as indicated in Fig. 5, in which
the abscissa is expressed in terms of the reduced tem-
perature TR=T/TMP. From this figure it is seen that
while ln A(T) is reasonably linear in the range 905
T5347 K, the resonance effect abruptly disappears
Fig. 4. Temperature dependence of the isomer shift for OMF. The
error bars are those of the individual data points. The solid line
represents the second-order polynomial regression through the data
points.
above 350 K (TR 0.82), that is, some 80°C below Tmp
.
This is in marked contrast to the related octamethyl
ferrocenyl complexes as will be discussed. The onset of
the reduction of the resonance effect is unusually sharp,
and appears to involve a hysteresis, since it is necessary
to cool the sample (from slightly above 350 K to about
330 K) before the resonance effect can again be ob-
served. These observations suggest the presence of a
first-order transition, but it should be noted that this is
probably not a simple crystallographic phase transition,
since there is no shattering of the crystallites when these
are warmed through the 350 K point. The drastic
decrease in the resonance effect on warming must be
associated with the onset of a motion (or motions) of
the Cp-rings, which permits a large increase in the
mean-square-amplitude of vibration of the metal atom
above 350 K. At this point it might be speculated that
this Cp-ring motion is associated with the presence of
two rings, neither of which has a ring substituent larger
Fig. 5. The temperature dependence of the recoil-free fraction for
OMF and 1.The abscissa is the reduced temperature, TR=T/TMP
,
where the melting points are those of OMF and 1, respectively. The
large error bars of the highest temperature measurements reflect the
very small values of the resonance effect magnitude, although in
excess of 8×106 counts per channel were acquired in the individual
experiments.

H. Schottenberger et al. / Journal of Organometallic Chemistry 612 (2000) 1–8
5
resonance curve — is summarized graphically in Fig.
5 (in which the abscissa is the reduced temperature) and
provides a comparison with the corresponding data for
OMF. For 1, the ln A(T) data are nearly linear up to
ꢀ315 K (TRꢀ0.97), that is, to within less than 10 K of
the melting point (324.5 K), reflecting the fact that no
low frequency modes (libration or rotation) arising
from ring motions play a dominant role in the metal
atom motion, which is governed primarily by the
FeꢀCp bonding interaction.
3.3. Compound 2
The Mo¨ssbauer data for 2-ethynylphenylferrocene,
for which a detailed crystal structure has been pub-
lished [29], are very similar to those summarized for 1,
above. Although as usual there is some curvature in the
IS(T) data, a linear regression (−dIS/dT=(3.969
0.14)×10-4, with a correlation coefficient of 0.985 for
28 data points) leads to a value of Meff=105 Da, and
thus the addition of the phenyl group ‘spacer’ plays no
significant role in the bonding of the metal atom to the
Cp ligands. The temperature dependence of the recoil-
free fraction is nearly linear (905T5340 K) up to a
reduced temperature of 0.99 (Tmp=344–346 K), and
shows no anomalous behavior over this interval. Ap-
plying a linear regression to these data leads to a value
of [M=99 K.
In the asymmetric unit of 2 there are two indepen-
dent molecules, which differ greatly in their conforma-
tion — the two dihedral angles are 26.6(5) and
30.2(4)°. In the former conformation the triple bond
points in the direction of the iron atom, in the latter,
the molecule is twisted in the opposite direction. The
distance between the acetylinic C atoms (C(17) and
C(18) [29]) and the metal atom are 3.813(8) and
Scheme 1. Preparation of polymers 5–7.
than a methyl group, but appears not to be observed in
the absence of these ring substituent methyl groups.
This point will be more fully addressed below.
The structures of monomers 1–4 as shown in Fig. 1
were chosen because they represent two sets of com-
pounds. The first set consists of the simplest metallo-
cenylalkyne, 1, and a spacered homologue, 2. The
second represents the octamethyl-substituted analogues,
3 and 4. In terms of polymer structures it is worth
noting that poly-1 (5) stems from an a insertion of 1
into the initiator, while poly-2 and poly-3 result from a
b insertion [27,29,31,38]. A description of the polymer-
ization and the resulting polymer structures is given in
Scheme 1.
3.2. Compound 1
,
4.229(9) A in one conformer, and 4.658(1) and 5.379(1)
in the other. These distances are too long to permit a
meaningful bonding interaction between the metal
atom and the olefinic moiety and, not unexpectedly, the
Mo¨ssbauer parameters are insensitive to the difference
between the two conformers, and the data can be
treated in terms of a single type of metal atom
environment.
The Mo¨ssbauer spectrum at 90 K of ethynyl fer-
rocene consists of a normal well-resolved doublet at all
temperatures considered. The temperature dependence
of the IS is similar to that shown in Fig. 4, and the
slight curvature noted in these data suggest the partici-
pation of several vibrational modes in the motion of the
central metal atom. Using a linear approximation to
the data (−dIS/dT=(3.8690.09)×10−4 mm s−1
K−1 with a correlation coefficient of 0.996 for 15 data
points) leads to a value of Meff=108 Da. The differ-
ence between this value and the ‘bare’ iron atom mass
of 57 Da reflects the covalent bonding of the metal
atom to the Cp framework. The QS parameter for 1 is
essentially temperature independent over the range
905T5273 K and has a mean value of 2.34390.001
3.4. Compound 3
The Mo¨ssbauer parameters for 1-octamethyl ferro-
cenyl 2 (2-ethynylphenyl)ethylene are summarized in
Table 1. Again, although there is some curvature in the
IS(T) data, a linear regression analysis (−dIS/dT=
(3.3090.13)×10−4 mm s−1
K
−1, cc:0.990 for 15 data
points) leads to an Meff=126 Da. The QS hyperfine
interaction is essentially T-independent, and has a
mean value of 2.39590.003 mm s−1 over the range
90BTB328 K. The temperature dependence of the
mm s−1
.
The temperature dependence of the recoil-free frac-
tion for 1 — as reflected in the area (A) under the

6
H. Schottenberger et al. / Journal of Organometallic Chemistry 612 (2000) 1–8
relaxation program used earlier [37] (ISꢀ0.503(5) mm
s−1 and QSꢀ −0.19(1) mm s−1 at 90 K). The near
collapse of the quadrupole interaction is due to the
removal of one d-electron from the e2g orbitals, which,
in the case of the ferrocene–ferricinium absorbers, ef-
fectively removes 0.8064 of a [3d0\] electron, follow-
ing the molecular orbital scheme of Dahl and
Ballhausen [39,40]. However, this d-electron removal is
expected to decrease the IS of the oxidized species
relative to the neutral complex, whereas in the present
case, the isomer shift increases by ꢀ0.11 mm s−1
going from 3 to 4 both at 90 K. This increase in IS is
in good agreement with the data for the ring-substi-
tuted ferrocene systems reported earlier [37].
Fig. 6. The temperature dependence of the recoil-free fraction for
compounds 3 and 7 showing the ‘softer’ lattice in the latter. The data
have been normalized to the respective 90 K data points. The error
bars reflect the errors in the individual data points.
The relaxation noted above is due to the interaction
of the dilute unpaired electron spins with the lattice,
similar to those which have been observed [37] for
related ferricinium cationic absorbers and is presumed
due to a Raman process (high T limit). Assuming a
magnetic hyperfine field of 520 kOe (typical of Fe+3),
the calculated relaxation time at 90 K is 1.0×10−9 s.
Finally, it is worth noting that the resonant effect in 4
is readily observable at temperatures well above 360 K,
in contrast to the behavior of OMF detailed above,
despite the presence of the eight methyl groups on the
two Cp rings in both monomers. Moreover, the T-de-
pendence of the recoil-free fraction for 3 and 4 are
essentially identical; the one-electron oxidation, as well
as the presence of the PF−6 groups apparently has little
effect on the metal atom dynamics in these two com-
pounds. It is worth noting in this context that the
corresponding data for OMF+PF6− similarly closely
mimic those for neutral OMF. The coincidence of these
two sets of data have a profound implication for the
mechanism of the anomalous recoil-free fraction behav-
ior, noted above for OMF, since it is clear that this
phenomenon must be associated with an intra- rather
recoil-free fraction for 3 is included graphically in Fig.
6
(vide infra, a comparison with the respective
monomer/polymer data) from which it is noted that —
although there is some ‘softening’ of the lattice starting
at about 25°C below Tmp (330 K) — the resonance
effect is readily observed to within 2°C of Tmp. This, of
course, is in marked contrast to the data cited for
OMF, above, in which the same two tetramethyl substi-
tuted ring moieties are present. In 3, as is obvious, the
presence of the large 2-ethynylphenylacetylene side
chain apparently inhibits the onset of the ring motion,
which has been identified with the large increase in the
mean-square-amplitude of the metal atom motion
above 350 K in OMF.
3.5. Compound 4
The 90 K Mo¨ssbauer spectrum of 4 is shown in Fig.
7. The spectral line shape is one which is characteristic
of a relaxation spectrum with its broad non-Lorentzian
appearance, and the data have been fitted with the
Fig. 7. Mo¨ssbauer Spectrum of 4 at 90 K typical of spin-lattice relaxation and a non-zero quadrupole hyperfine interaction in the solid. The
magnetic hyperfine field assumed in the fitting procedure was 396 kOe.

H. Schottenberger et al. / Journal of Organometallic Chemistry 612 (2000) 1–8
7
than an intermolecular process, such as ring librational
4. Summary and conclusions
motion. The detailed structural implications of these
observations are currently being studied in these
laboratories.
The hyperfine interactions and lattice dynamics of a
number of ferrocenyl and octamethyl ferrocenyl substi-
tuted ethynes has been examined of the range 90BTB
360 K, and an unusual motional anomaly has been
observed for OMF. The sudden increase in the mean-
square amplitude of vibration of the metal atom can be
associated with the presence of eight methyl sub-
stituents on the two Cp rings, but appears to be absent
when the ninth ring position is occupied by a bulky
organic group. Oxidation of the metal atom in
monomer 3 leads to spin-lattice interactions in both the
corresponding monomer cation complex, 4, as well as
in the corresponding polymer cation complex, 8. The
Mo¨ssbauer lattice temperature, [M, as probed by the
metal atom in these solids, decreases by 20–40% when
the monomer is polymerized into an extended chain
structure, presumably due to the influence of additional
low-frequency vibrational modes on the metal atom
motion. Polymerization of 1–3 to 5–7, respectively, is
reflected in a small decrease in the s-electron density at
the metal atom nucleus, while the oxidation of 3 to 4
leads to a more significant decrease of this density, in
agreement with earlier results on ring-substituted fer-
rocenes [37].
3.6. Compound 5
The Mo¨ssbauer parameters of 5 are included in Table
2 and are not otherwise exceptional. The IS(T), QS(T)
and ln A(T) data are well fitted by linear regressions,
and the Meff and [M data extracted from these depen-
dencies are included in Table 2. It is interesting to note
that these parameters are all very similar to those
observed for the monomer 1 with the exception of the
Mo¨ssbauer lattice temperature, which is some 20%
lower in the polymer, indicating a ‘softening’ of the
polymer lattice with respect to the monomer, pre-
sumably due to a larger density of states of low fre-
quency vibrational motions in the latter with respect to
the former. The small increase in the IS at 90 K in
going from 1 to 5, which is also noted for the other
monomer–polymer pairs, suggests a small decrease in
the s-electron density (at the metal atom site) in the
polymer with respect to the monomer. The reasons for
this change are not yet fully elucidated in the present
instance.
3.7. Compound 6
Acknowledgements
Comments similar to those articulated above for 5
apply to 6, including the apparent ‘softening’ of the
polymer lattice with respect to the monomer, and the
appropriate parameters are included in Table 2.
3.8. Compound 7
Financial support was provided by the Austrian Na-
tional Science Foundation (FWF Vienna, project num-
ber P-12963-GEN) and the ‘Jubila¨umsfond der
8
Osterreichischen Nationalbank’, Vienna, Austria, (pro-
ject number 7489). This research has also been sup-
ported by a grant from the Israel Academy of Sciences,
and is herewith gratefully acknowledged. The generous
assistance of Professor I. Nowik in connection with
evaluation of the relaxation spectra is also thankfully
acknowledged.
Again, the relevant hyperfine parameters for this
polymer are included in Table 2. However, because of
the significant curvature observed in both the IS(T) and
ln a(T) data for this material, it is not possible to
extract meaningful [M and Meff values in this case.
However, a direct comparison of the recoil-free fraction
data with those of the corresponding monomer 3 is
shown in Fig. 6, from which the ‘softening’ of the
polymer lattice relative to the monomer is clearly estab-
lished by the larger slope in the former.
References
[1] H. Schottenberger, M.R. Buchmeiser, Recent Res. Dev. Macro-
mol. Res. 3 (1999) 535.
[2] T. Ogawa, Prog. Polym. Sci. 20 (1995) 943.
3.9. Compound 8
[3] S. Patai, Z. Rappoport, in: S. Patai (Ed.), The Chemistry of
Triple-Bonded Functional Groups, Vol. Suppl. C2, Wiley,
Chichester, UK, 1983, pp. 917–980.
[4] R. Lytel, G.F. Lipscomb, J. Thackara, J. Altman, P. Elizondo,
M. Stiller, B. Sullivan, in: P.N. Prasad, D.R. Ulrich (Eds.),
Nonlinear and Electro-Optic Organic Devices, Plenum, New
York, 1987, p. 415.
[5] J. Zyss, D.S. Chemla, in: D.S. Chemla, J. Zyss (Eds.), Nonlinear
Optical Properties of Organic Molecules and Crystals, vol. 1,
Academic Press, Orlando, FL, 1987, p. 23.
The Mo¨ssbauer spectrum for this polymer is very
similar to that of the corresponding monomer 4, and
consists of a single broadened line at an isomer shift of
0.576 at 90 K. Each of the iron atoms in the monomer
has been oxidized to Fe+3, and the quadrupole splitting
has effectively collapsed, as noted above. The calcu-
lated relaxation rate at 90 K is 1.24×10−10 s.

8
H. Schottenberger et al. / Journal of Organometallic Chemistry 612 (2000) 1–8
[6] G. Yu, A. Heeger, The Physics of Semiconductors, vol. 1, World
[23] K.C. Wallace, A.H. Liu, W.M. Davis, R.R. Schrock,
Organometallics 8 (1989) 644.
Scientific, Singapore, 1996.
[7] J. Vohlidal, J. Sedla´cek, M. Pacovska´, O. Lavastre, P.H.
Dixneuf, H. Balcar, J. Pfleger, Polymer 38 (1997) 3359.
[8] P.N. Prasad, Polymer 32 (1991) 1746.
[9] T. Kobayashi, T. Hattori, A. Terasaki, K. Kurokawa, in: P.N.
Prasad, D.R. Ulrich (Eds.), Femtosecond Studies of Dephasing
and Phase Conjugation with Incoherent Light, Plenum, New
York, 1987, p. 137.
[10] F. Meyers, S.R. Marder, B.M. Pierce, J.L. Bre´das, J. Am. Chem.
Soc. 116 (1994) 10703.
[11] J.L. Bre´das, R. Silby, D.S. Boudreaux, R.R. Chance, J. Am.
Chem. Soc. 105 (1983) 6555.
[12] T.J. Marks, M.A. Ratner, Angew. Chem. Int. Ed. Engl. 107
(1995) 167.
[13] C.B. Gormann, S.R. Marder, Chem. Mater. 7 (1995) 215.
[14] D.R. Kanis, M.A. Ratner, T.J. Marks, Chem. Rev. 94 (1994)
195.
[15] J.L. Bre´das, Adv. Mater. 7 (1995) 262.
[16] F. Hide, M.A. D´ıaz-Garc´ıa, B.J. Schwartz, A. Heeger, J. Acc.
Chem. Res. 30 (1997) 430.
[17] C.B. Gorman, E.J. Ginsburg, R.H. Grubbs, J. Am. Chem. Soc.
115 (1993) 1397.
[24] R. Schlund, R.R. Schrock, W.E. Crowe, J. Am. Chem. Soc. 111
(1989) 8004.
[25] F.J. Schattenmann, R.R. Schrock, W.M. Davis, J. Am. Chem.
Soc. 118 (1996) 3295.
[26] H.H. Fox, R.S. Schrock, Organometallics 11 (1992) 2763.
[27] R.R. Schrock, S. Luo, N. Zanetti, H.H. Fox, Organometallics 13
(1994) 3396.
[28] M. Buchmeiser, Macromolecules 30 (1997) 2274.
[29] M.R. Buchmeiser, N. Schuler, N. Kaltenhauser, K.-H. Ongania,
I. Lagoja, K. Wurst, H. Schottenberger, Macromolecules 31
(1998) 3175.
[30] M. Buchmeiser, R.R. Schrock, Macromolecules 28 (1995) 6642.
[31] R.R. Schrock, S. Luo, J.C. Lee, Jr., N.C. Zanetti, W.M. Davis,
J. Am. Chem. Soc. 118 (1996) 3883.
[32] R.H. Herber, K. Temple, I. Manners, M. Buretea, T.D. Tilley,
Inorg. Chim. Acta 287 (1999) 152.
[33] V. Guillaume, P. Thominot, F. Coat, A. Mari, C. Lapinte, J.
Organomet. Chem. 565 (1998) 75.
[34] R.H. Herber, B. Bildstein, P. Denifl, H. Schottenberger, Inorg.
Chem. 36 (1997) 3586.
[35] M.R. Buchmeiser, N. Schuler, H. Schottenberger, I. Kohl, A.
Hallbrucker, K. Wurst, Macromolecules (1999), in press.
[36] K. Wurst, O. Elsner, H. Schottenberger, Synlett (1995) 833.
[37] R.H. Herber, T.P. Hanusa, Hyperfine Interact. 108 (1997) 563.
[38] H.H. Fox, M.O. Wolf, R. O’Dell, B.L. Lin, R.R. Schrock, M.S.
Wrighton, J. Am. Chem. Soc. 116 (1994) 2827.
[39] J.P. Dahl, C.F. Ballhausen, Mat. Fys. Medd. Dan. Vid. Selsk. 55
(1961) 5.
[18] C.B. Gorman, E.J. Ginsburg, M.J. Sailor, J.S. Moore, T.H.
Jozefiak, N.S. Lewis, R.H. Grubbs, S.R. Marder, J.W. Perry,
Synth. Metals 41 (1991) 1033.
[19] M.J. Sailor, E.J Ginsburg, C.B. Gorman, A. Kumar, R.H.
Grubbs, N.S. Lewis, Science 249 (1990) 1146.
[20] E.J. Ginsburg, C.B. Gorman, S.R. Marder, R.H. Grubbs, J. Am.
Chem. Soc. 111 (1989) 7621.
[21] K. Knoll, S.A. Krouse, R.R. Schrock, J. Am. Chem. Soc. 110
(1988) 4424.
[22] S.A. Krouse, R.R. Schrock, Macromolecules 21 (1988) 1885.
[40] N.N. Greenwood, T.C. Gibb, Mo¨ssbauer Spectroscopy, Chap-
man and Hall, London, 1971.
.