Insight into the Thermal Ring-Opening Polymerization of Phospha[1]ferrocenophanes

Article abstract of DOI:10.1002/chem.201603000

A mixture of cis/trans isomers of phospha[1]ferrocenophanes equipped with one iPr group at the α position to the bridging PhP moiety was prepared. Both isomers (cis-4 and trans-4) were obtained as racemates and could be separated so that their thermal properties were investigated individually. The molecular structure of cis-4 was determined by single-crystal X-ray analysis showing a tilt angle α=26.35(8)°. Interconversion between both isomers occurred in the melt at elevated temperatures and revealed that the trans isomer is thermodynamically more stable. Structural and thermodynamic data was complemented by DFT calculations (B3PW91/6-311+G(d,p) and B3PW91-D3(BJ)/6-311+G(d,p)). Performance of thermal ring-opening polymerization (ROP) of trans-4 at 230 °C gave polymers and cyclic oligomers. Gel permeation chromatography (GPC) of the sulfurized polymer resulted in a molecular weight of 62.5 kDa (M_w) and a polydispersity index of 1.39 (PDI). Mass spectrometric analysis of the oligomers showed the presence of cyclic species from dimers to heptamers. After sulfurization, preparative thin layer chromatography led to the separation of three isomeric dimers. Structural characterization of these dimers by single-crystal X-ray analysis led to the conclusion that the Fe?Cp bond breaks during the thermal ROP process. A mechanism similar to the known mechanism of the photolytic ROP of ferrocenophanes is proposed.

Full text of DOI:10.1002/chem.201603000

DOI: 10.1002/chem.201603000
Full Paper
&
Sandwich Complexes |Hot Paper|
Insight into the Thermal Ring-Opening Polymerization of
Phospha[1]ferrocenophanes
Elaheh Khozeimeh Sarbisheh,[a] Jose Esteban Flores,[a] Jianfeng Zhu,[b] and Jens Mꢀller*[a]
Abstract: A mixture of cis/trans isomers of phospha[1]ferro-
cenophanes equipped with one iPr group at the a position
to the bridging PhP moiety was prepared. Both isomers (cis-
4 and trans-4) were obtained as racemates and could be
separated so that their thermal properties were investigated
individually. The molecular structure of cis-4 was determined
by single-crystal X-ray analysis showing a tilt angle a=
26.35(8)8. Interconversion between both isomers occurred in
the melt at elevated temperatures and revealed that the
trans isomer is thermodynamically more stable. Structural
and thermodynamic data was complemented by DFT calcu-
lations (B3PW91/6-311+G(d,p) and B3PW91-D3(BJ)/6-311+
G(d,p)). Performance of thermal ring-opening polymerization
(ROP) of trans-4 at 2308C gave polymers and cyclic oligo-
mers. Gel permeation chromatography (GPC) of the sulfur-
ized polymer resulted in a molecular weight of 62.5 kDa (Mw)
and a polydispersity index of 1.39 (PDI). Mass spectrometric
analysis of the oligomers showed the presence of cyclic spe-
cies from dimers to heptamers. After sulfurization, prepara-
tive thin layer chromatography led to the separation of
three isomeric dimers. Structural characterization of these
dimers by single-crystal X-ray analysis led to the conclusion
that the FeÀCp bond breaks during the thermal ROP pro-
cess. A mechanism similar to the known mechanism of the
photolytic ROP of ferrocenophanes is proposed.
Introduction
Shortly after the first [1]ferrocenophanes ([1]FCPs) were de-
scribed in 1975 (Figure 1;
A with ERx =SiPh2, Sifc; fc=
(H4C5)2Fe),[1] reports about the first phosphorus-bridged [1]FCPs
(A with ERx =PPh, PMe) appeared.[2] Early attempts to polymer-
ize the phosphorus-bridged species PhPfc (A with ERx =PPh)
using LiPh as an anionic initiator revealed that ring-opening
occurs to give oligomers, but failed to give polymers.[3] Surpris-
ingly, in a series of salt-metathesis reactions between PhPCl2
and dilithioferrocene–tmeda (tmeda=tetramethylethylenedia-
mine), in some cases, polymers were obtained with molecular
weights that were above expected values for a step-growth
polymerization.[3]
Figure 1. [1]Ferrocenophanes A, polyferrocenes B, planar-chiral [1]ferroceno-
phanes C, E, and distortion angles in [1]ferrocenophanes D.
Later it was speculated that in these experiments anionic
ring-opening polymerization (ROP) of in situ formed PhPfc had
occurred.[4] It took another decade before it was shown that
high-molecular-weight polyferrocenes (B) can indeed be pre-
pared from FCPs by ROP.[5] The first report of a successful ROP
of a phospha[1]ferrocenophane appeared in 1995, which was
an investigation of the thermal ROP of PhPfc.[4a] In subsequent
years, anionic and photolytic ROP, respectively, were developed
for phospha[1]ferrocenophanes. As living polymerizations,
these processes allow for the control of molecular weights
and molecular-weight distributions.[4b,6] To date, a significant
number of publications exist that deal with the synthesis and
characterization of phosphorus-bridged [1]FCPs, their reactions
with metal complexes, and their use as precursors for metallo-
polymers.[2–4,6–7] Moreover, the early discovery that the [1]FCP
PhPfc reacts with LiPh to form a lithiated diphenylphosphino-
ferrocene, (LiH4C5)Fe[C5H4(PPh2)],[2d] led to a series of publica-
tions with the majority being concerned with the preparation
and application of new ferrocene-based ligands.[8] Several as-
pects of phosphorus-bridged [1]FCPs are noteworthy. First, in
the large group of known [1]FCPs[9] they are among the high-
[a] E. Khozeimeh Sarbisheh, J. Esteban Flores, Prof. Dr. J. Mꢀller
Department of Chemistry, University of Saskatchewan
110 Science Place, Saskatoon, SK, S7N 5C9 (Canada)
E-mail: jens.mueller@usask.ca
[b] Dr. J. Zhu
Saskatchewan Structural Sciences Centre
University of Saskatchewan
110 Science Place, Saskatoon, SK, S7N 5C9 (Canada)
Supporting information for this article and ORCID for one of the authors
are available on the WWW under http://dx.doi.org/10.1002/
chem.201603000.
Chem. Eur. J. 2016, 22, 1 – 13
1
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
est strained species (A with ERx =PR) as indicated by tilt angles
a between 26 and 288 (D),[2b,d,4a,6b,e,7a,f,h,p,t] which for [1]FCPs
are only surpassed by sulfur[10] or boron[11] in bridging positions
(a>308). Second, the donor-ability of phosphorus in mono-
mers, oligomers, and polymers has the potential to be applica-
ble in catalysis.[2a,c,d,7a,d,h,q,s,12] In light of their potential applica-
tion as ligands, we recently started to prepare phosphorus-
bridged [2]FCPs[13] and [1]FCPs,[7t,14] with the latter species
being prepared as enantiomerically pure compounds of type C
(Figure 1). The C2-symmetric ferrocene moiety of these species
resulted from our efforts to tailor the bulkiness adjacent to the
bridging element, aiming to develop suitable aluminum- and
gallium-bridged ferrocenophanes for ROP.[9f,11c] The planar-
chiral ferrocene moiety in species C was introduced using Ugi
amine chemistry;[15] however, the first known planar-chiral
phospha[1]ferrocenophanes were already published in 1983
and synthesized as racemates (E: R’ =CHMe(NMe2), ERx =PPh,
PtBu; R’ =CHiPr(NMe2), ERx =PPh).[7a]
A 4:6 mixture of cis-4 to trans-4 was obtained as a dark-red
oil with a yield of 50% after chromatography on alumina,
which failed to separate the two isomers. After a flask-to-flask
condensation, crystallization from hexanes at À208C resulted
in a pink-red powder of trans-4 that contained only 6% of the
cis isomer. Further crystallization from the mother liquor at
À808C gave a solid, a 7:3 mixture of cis-4 to trans-4, from
which dark-red crystals of the cis isomer were picked for struc-
tural analysis (Figure 2). Both species can easily be differentiat-
Within this report, the synthesis and properties of new phos-
phorus-bridged [1]FCPs of type E are discussed. For the first
time, we show how the asymmetry of the ferrocene moiety
provides insight into the mechanism of the thermal ROP of
phospha[1]ferrocenophanes.
Figure 2. Molecular structure of cis-4 with thermal ellipsoids at the 50%
probability level. Hydrogen atoms are omitted for clarity. For bond lengths
and angles see Table S2 (Supporting Information).
Results and Discussion
Synthesis and characterization of monomers
ed by 1H NMR spectroscopy. The most distinguished signals
are those of the methine proton of the isopropyl groups at
d=3.45 ppm for trans-4 and at d=2.13 ppm for cis-4. The sig-
nificant difference in the chemical shift is likely caused by the
phenyl group, which shields the methine proton in the cis
isomer. On the other hand, the methine signal of the trans
isomer is a septet of doublets, whereas that of the cis species
is just a septet. The additional coupling in trans-4 is due to the
presence of phosphorus (J=2.5 Hz) and likely mediated by the
lone pair that is in proximity of the methine proton.
The racemic Ugi amine 1[15–16] was prepared based on pub-
lished procedures (see the Supporting Information) and trans-
formed into the dibromide 2 (Scheme 1). Following known
Species 4 is not the first reported isomeric mixture of phos-
pha[1]ferrocenophanes equipped with one alkyl group. In the
course of exploring the thermal ROP of PhPfc, the monomer
(nBuH3C5)PPh(H4C5)Fe, equipped with one nBu group, was pre-
pared and isolated in ca. 95% purity. As the precursor for the
needed dilithioferrocene derivative for the salt-metathesis ap-
proach was n-butylferrocene, this phospha[1]ferrocenophane
was obtained as a mixture of four racemates with nBu either in
a- or b position.[4a] The authors employed the alkyl group in
order to obtain polymers of increased solubility compared to
those obtained from the monomer of PhPfc. A similar ap-
proach toward improved solubilities was already taken in the
investigation of ROP of the sulfur-bridged FCP S3fc by Rauch-
fuss et al.[5a]
Scheme 1. Synthesis of the racemic [1]FCPs trans-4 and cis-4.
chemistry,[7t,11c,17] species 2 was converted in two steps to the
isopropyl-substituted ferrocene dibromide 3. Applying condi-
tions[18] that we used for similar dibromoferrocenes be-
fore,[7t,11c–e,19] the dilithio derivative of 3 was prepared by a lithi-
um–bromine exchange reaction and in situ reacted with
PhPCl2 to give the targeted [1]FCPs 4 as a mixture of cis and
trans isomers (Scheme 1). It should be noted that species 2 is
a 1:1 mixture of the (R,Sp) and (S,Rp) isomers and, for reasons of
simplification, only one enantiomer of this racemate is shown
in Scheme 1. Similarly, the racemate of the planar-chiral com-
pounds 3, cis-4, and trans-4 are each illustrated by one enan-
tiomer.
While Figure 2 shows the molecular structure of cis-4, crystal
and structural refinement data can be found in Table S1 in the
Supporting Information. The structure of trans-4 could not be
obtained experimentally as all our attempts to obtain single
crystals failed. Therefore, geometries of both isomers were op-
&
&
Chem. Eur. J. 2016, 22, 1 – 13
www.chemeurj.org
2
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
timized at the B3PW91/6-311+G(d,p) level (see the Experimen-
tal Section). The most interesting structural parameters of
[1]FCPs are the angles that are commonly used to describe dis-
tortions in these strained compounds (D; Figure 1); Table 1
compiles these angles for cis- and trans-4, and for the closely
related species PhPfc.[2b,c,7a]
Table 1. Distortion angles [8] in cis-4, trans-4, and PhPfc.
Distortion angles[a] cis-4
Exptl.[b]
cis-4
Calcd[c]
trans-4
Calcd[c]
PhPfc[d]
Exptl.[d]
a
b/b’[e]
26.35(8)
26.21
26.78
26.7
33.9(1)/33.5(1) 33.89/32.93 32.28/32.41 33/32
d
160.30(2)
91.27(5)
161.35
90.87
160.92
90.54
160
90.6(3)
q
[a] Definition of distortion angles (D; Figure 1): a=angle between the
least-squares planes of Cp rings; b=1808À(CpcentroidÀCipsoÀE); d=Cpcentroid
À
FeÀCpcentroid; q=CipsoÀEÀCipso. [b] See Figure 2. [c] B3PW91/6-311+G(d,p)
[d] fc=(h5-C5H4)2Fe; data taken from reference [2b]; the structure was
published again in 1983[7a] with the following values [8]: a=26.9; b=32.3
(mean value); d=159.8; q=90.7(2). [e] b is defined by the Cp centroid of
the iPr-substituted Cp ring.
Figure 3. Optimized geometries of cis- and trans-4 (B3PW91/6-311+G(d,p)).
The measured tilt angle a of 26.35(8)8 for cis-4 fits right in
the known range of 26 to 288 for RP-bridged
[1]FCPs.[2b,d,4a,6e,7a,f,h,p,t] Table 1 shows an excellent match be-
tween the measured and calculated distortion angles of cis-4,
revealing the suitability of the chosen B3PW91/6-311+G(d,p)
level of theory. This gives us confidence that the calculated
values of trans-4 would also match measured values, if they
were available. The molecular structure of the PhP-bridged
compound PhPfc, the first reported phospha[1]ferroceno-
phane,[2] has been published twice, and tilt angles of 26.7[2b]
and 26.98[7a] were reported (Table 1), which match the calculat-
ed value of 26.788 of the trans-4 better than that of the cis
isomer (a=26.35(8) or 26.218). Even though the difference be-
tween the a angles is small, from the close match between
the set of distortion angles of PhPfc and trans-4 one gets the
impression that the iPr group of trans-4 has less of a structural
effect compared to the iPr group of cis-4. This would make
sense as, in contrast to the trans isomer, the iPr and Ph groups
are in close proximity in the cis isomer and could sterically in-
teract. That this is indeed the case is illustrated in Figure 3, in
which the optimized geometries of both isomers are displayed.
From each set of bond angles around phosphorus one can see
that they are slightly widened for the cis isomer (Figure 3).
Also, the iPr and the Ph groups of cis-4 are turned away from
each other, obviously driven by minimizing steric repulsions.
This effect shows up in differences in the torsion angle C1-C2-
C11-C12, which is À79.908 for trans-4 and À82.258 for cis-4.
The difference for the conformation of the Ph group is more
pronounced and can be shown by the two torsion angles C1-
P1-C14-C15 and C6-P1-C14-C19, which are À47.39 and 46.158
for trans-4 and 62.47 and À30.178 for cis-4 (see the Supporting
Information for calculated geometries). Admittedly, these struc-
tural effects are small; however, they are significant and indi-
cate steric repulsion between Ph and iPr in the cis isomer that
is absent in the trans species. Therefore, it is expected that
trans-4 is thermodynamically more stable than cis-4, which was
confirmed both experimentally and theoretically. With inclusion
of Grimme’s dispersion in DFT calculations, the trans isomer is
4.8 kJmolÀ1 more stable than the cis isomer (DG8 at B3PW91-
D3(BJ)/6-311+G(d,p)); without dispersion the value increases
to 9.5 kJmolÀ1 (see the Experimental Section). If a solid sample
of the cis isomer was heated in a sealed NMR tube at 1378C
for 8 h, a red oil formed composed of a 9:1 mixture of trans-
to cis-4. If the experiment was started with a sample of the
trans isomer, a mixture of the same composition resulted (see
the Experimental Section). 1H and 31P NMR spectroscopy did
not reveal any species other than the two isomers. Under
these conditions, a thermal equilibrium forms, which is illus-
trated in Figure 4 in which the calculated transition state (TS4)
for this process is also displayed. The measured equilibrium
mixture of 9:1 refers to a difference in free energy of
Figure 4. Thermal equilibrium between trans-4 and cis-4.
Chem. Eur. J. 2016, 22, 1 – 13
www.chemeurj.org
3
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
5 kJmolÀ1, a value that fits very well to the calculated differ-
ence when dispersion was included.
As expected, TS4 exhibits a planar-coordinated phosphorus
atom, that is, the sum over the three C-P-C angles is 360.08
(Figure 4). With dispersion corrections the free activation
energy for this isomerization is 143.9 kJmolÀ1 (134.0 kJmolÀ1
without dispersion; see the Experimental Section). This value is
in the expected range of 122–149 kJmolÀ1,[20] which had been
experimentally determined for a series of acyclic phosphines
R1R2PMe.[21,22] The authors of the report about the first phos-
pha[1]ferrocenophane, PhPfc, already noted that inversion at
phosphorus was expected, but proton NMR spectroscopy
(C6D6; 100 MHz) did not reveal any evidence of such a process
at temperatures up to 708C.[2a]
Scheme 2. Thermal ROP of trans-4.
polymer 4n as a yellow powder (55% yield). In order to make
the polymer 4n insensitive to oxygen and suitable for analysis
by gel permeation chromatography (GPC), a common sulfuriza-
tion was performed (Scheme 2).[4a] GPC analysis of (4S)n with
a triple detection system gave an absolute molecular weight
of Mw =62.5 kDa (PDI=1.39; Figure S40 and Table S6 in the
Supporting Information). Species (4S)n showed the expected
Thermal ROP of 4
1
broad signals in the H NMR spectrum and a single broad peak
at d=37.8 ppm in the 31P NMR spectrum. The latter value is
very similar to the published chemical shift that was measured
for [PhP(=S)fc]n that lacks the iPr group (d=37.5 ppm, in
CD2Cl2).[4a]
Differential scanning calorimetry (DSC) was performed on each
isomer and resulted in values for DHROP of À89(Æ2) kJmolÀ1
for trans-4 and À88(Æ2) kJmolÀ1 for cis-4 (Figure 5; Figur-
Approximately half of the reaction crude mixture did not
precipitate and stayed in solution. After this soluble part was
also sulfurized, a 31P NMR spectrum exhibited a wealth of sig-
nals (Figure S25 in the Supporting Information). A mass spec-
trum showed peaks that referred to cyclic oligomers from
dimers up to heptamers,[25] with the signal for dimers dominat-
ing the mass diagram. Applying preparative thin layer chroma-
tography (PTLC), six fractions were isolated. According to 1H
and 31P NMR spectroscopy, three of the six fractions contained
a single species (fractions A, B, and F; Experimental Section),
one was fairly clean containing predominantly one species
(fraction E), while the other two fractions (C, D) were unre-
solved mixtures. MS data showed that the latter two fractions
contained dimers to heptamers, whereas the four cleaner frac-
tions (A, B, E, F) were all dimers with the same empirical formu-
la [{PhP(=S)C10H7(iPr)}Fe]2. The fact that the NMR spectra of
these four fractions were all different revealed that isomeric
dimers had been obtained. From a theoretical point of view,
many different isomers can be constructed from two ferrocene
moieties, two PhP=S bridging units, and two iPr groups. This
will be discussed with the help of Figure 6, in which a [1.1]FCP
is drawn in its anti and syn conformation. In each structure
two sets of a positions are indicated by R1 (exo) and R2 (endo).
For both isomers we display the same configuration at phos-
Figure 5. Differential scanning calorimetry (DSC) thermogram of trans-4
showing a melting endotherm (Tmax =118 8C) and a ROP exotherm
[Tmax =2558C; À89(Æ2) kJmolÀ1].
es S38–39 in the Supporting Information). Both measured en-
thalpies are identical within experimental error. These values
are expectedly larger than the known value of Me2Sifc (DHROP
of ca. À80 kJmolÀ1),[5b] as phosphorus in bridging positions
tilts Cp rings stronger than silicon (a=20.8(5)8 (Me2Sifc),[23]
26.35(8)8 (cis-4)). Moreover, the measured values for cis- and
trans-4 are lower than that determined for the highly strained
sulfur-bridged [1]FCP (DHROP =À130(Æ20) kJmolÀ1
;
a=
31.05(10)8 (Sfc)).[10a,24] The only DHROP values published for
phospha[1]ferrocenophanes are À68(Æ5) kJmolÀ1 for PhPfc,[7g]
À61(Æ5) kJmolÀ1 for (MePhP)fc+,[7g] and À60 kJmolÀ1 for
phosphine-borane adduct H3B-PhPfc.[7i] In light of our findings
and the a angles in these species, these measured values
seem to be too low.
Using this DSC thermogram with an onset temperature of
2308C as a guide (Figure 5), neat trans-4 was heated for 2 h at
that temperature in a sealed NMR tube (Scheme 2). This pro-
cess resulted in a highly viscous melt, which solidified to
a glassy solid at ambient temperature. The crude was com-
pletely dissolved in thf and precipitation into hexanes gave the
Figure 6. Illustration of exo (R1) and endo (R2) a positions in diphospha[1.1]-
ferrocenophanes.
&
&
Chem. Eur. J. 2016, 22, 1 – 13
www.chemeurj.org
4
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
phorus with the larger phenyl group oriented away from the
ferrocene moieties. This is based on the structures of the
parent [1.1]FCPs [PhP(=S)fc]2, which had been isolated by
Miyoshi et al. in the course of exploring the photolytic ROP of
the [1]FCP PhPfc.[7q] For [PhP(=S)fc]2 an anti and a syn isomer
was crystallographically characterized (R1 =R2 =H).
The parent anti isomer (R1 =R2 =H) has C2h symmetry (group
order h=4). As all Cp protons are in general positions,[26] the
16 Cp protons fall into four sets of chemically equivalent pro-
tons, two sets in a positions (R1 and R2) and two sets in b posi-
tions. If one systematically places two iPr groups in a positions
the following picture evolves. Distributing both iPr groups
over exo positions (R1) must result in isomers with symmetries
that cover all possible subgroups of the group order h=2.
That means, three exo,exo isomers of C2, Cs, or Ci point-group
symmetry result. The same applies to distributing two iPr
groups over endo positions (R2): three endo,endo isomers of C2,
Cs, or Ci point-group symmetry result. If one iPr is in an exo
and one in an endo position, the group order of the parent
species is lowered from 4 to 1 because the substitution of only
one position in a set of four equivalent positions destroys the
symmetry. Therefore, one can expect four different exo,endo
isomers, all with C1 point-group symmetry. However, one of
the four isomers has both iPr groups on the same Cp ring and
can be excluded as the formation of this species would involve
the breakage of a CÀC bond during the ROP experiment of
species 4. In short, nine isomers are theoretically possible for
a disubstituted anti isomer: three times exo,exo (C2, Cs, or Ci),
three times endo,endo (C2, Cs, or Ci), and three times exo,endo
(3ꢂC1). Applying the same systematic approach to the syn
isomer (Figure 6) gives a similar picture because the group
order of the parent C2v-symmetric syn isomer is the same as for
the anti isomer (h=4). That means, nine syn isomers are theo-
retically possible: three times exo,exo (C2, 2ꢂCs), three times
endo,endo (C2, 2ꢂCs), and three times exo,endo (3ꢂC1). These
18 possible isomers all contain two planar-chiral elements, that
is, the iPr-substituted Cp rings. While the eight species of Cs or
Ci symmetry contain both planar-chiral configurations (Sp and
Rp) in the same molecule (meso compounds), the ten species
of C2 or C1 symmetry contain only one type of planar-chiral
configuration (Sp or Rp) in the same molecule (rac compounds).
We expect that isomers where possible steric interactions be-
tween iPr groups are minimized will thermodynamically be
more stable, and therefore, might also be dimers that prefera-
bly form. With this in mind, one can sort the 18 isomers into
three groups with an expected decreasing stability in the fol-
lowing order: six times exo,exo isomers > six times exo,endo
isomers > six times endo,endo isomers.
Figure 7. Molecular structure of syn-C2-(4S)2 with thermal ellipsoids at the
50% probability level. Hydrogen atoms are omitted for clarity. For bond
lengths and angles see Table S5 (Supporting Information).
Figure 8. Molecular structure of anti-Ci-(4S)2 with thermal ellipsoids at the
50% probability level. Hydrogen atoms are omitted for clarity. For bond
lengths and angles see Table S3 (Supporting Information).
From the three fractions (A, B, and F) that contained a single
species, crystals were obtained and molecular structures were
solved by single-crystal X-ray analysis (Figures 7–9; Tables S1
and S3–S5 in the Supporting Information). The structurally
characterized dimers are syn-C2-(4S)2 (Figure 7; fraction F), anti-
Ci-(4S)2 (Figure 8; fraction A), and anti-C2-(4S)2 (Figure 9; frac-
tion B) and are three of the possible six isomers with the two
iPr groups in exo positions. These types of compounds are not
new as the parent anti and syn isomers of [PhP(=S)fc]2 were
Figure 9. Molecular structure of anti-C2-(4S)2 with thermal ellipsoids at the
50% probability level. Hydrogen atoms are omitted for clarity. For bond
lengths and angles see Table S4 (Supporting Information).
characterized in 2005.[7q] As expected, the syn and anti isomers
characterized in our work are very similar to the known spe-
cies, which should just be illustrated by comparing the PÀP
and FeÀFe distances, respectively. The published values of the
syn isomer are 4.632 (PÀP) and 4.706 ꢃ (FeÀFe), which are
Chem. Eur. J. 2016, 22, 1 – 13
www.chemeurj.org
5
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
comparable to those of syn-C2-(4S)2 (Figure 7) with 4.593 (PÀP)
and 4.784 ꢃ (FeÀFe). While for the known anti isomer 4.847
(PÀP) and 5.014 ꢃ (FeÀFe) were measured, for anti-Ci-(4S)2,
4.871 (PÀP) and 5.010 ꢃ (FeÀFe) (Figure 8) and for anti-C2-(4S)2
4.892 (PÀP) and 5.071 ꢃ (FeÀFe) were found (Figure 9). We re-
frain from a more detailed structural discussion and like to
refer an interested reader to the crystallographic data in the
Supporting Information[27] and the detailed discussion by
Miyoshi et al.[7q]
Scheme 4. First, monomer 4 gets transformed to the intermedi-
ate L24 through an h5 to h1 haptotropic shift in which the loss
of four electrons at iron is compensated by addition of two
neutral two-electron donor molecules L. One can imagine that
monomer 4, which is a monodentate phosphine, takes up the
role of the needed donor L. The activated species L24 might
further react to the polar, ring-opened product L34. The latter
species ring-opens monomer 4 by a nucleophilic attack at iron
with its dangling Cp anion. Again, monomer 4 with two differ-
ent FeÀCp bonds can be ring-opened in two different ways;
only one is illustrated in Scheme 4. The resulting species L342
is a higher homologue of L34 and the first product of the
chain growth. Repeated addition of monomer 4 to the key in-
termediate L342 leads to polymers L34n or, through backbiting,
a [1.1]FCP is produced. For the case shown in Scheme 4, the
formation of the anti-C2-42 is illustrated; this compound was
isolated in the form of its sulfurized species anti-C2-(4S)2
(Figure 9). Of course, at various stages of the chain growth,
species L34n could produce cyclic species through backbiting,
which explains the formation of cyclic oligomers that were
identified by mass spectrometry ([1m]FCPs with m=2–7).[25]
As mentioned before, monomer 4 has two different FeÀCp
bonds and both will break with a certain ratio. Hence, different
variations of the specific mechanism shown in Scheme 4 could
be drawn resulting in species with different distributions of iPr
groups. Therefore, the formation of all possible [1.1]FCPs can
be explained, including those that led to the structurally char-
acterized species syn-C2-(4S)2 (Figure 7) and anti-Ci-(4S)2
(Figure 8). However, only the structure of anti-C2-(4S)2
(Figure 9) reveals that the FeÀCp cleavage has happened.
In the following are known results from the literature that
formed the basis of the proposed mechanism in Scheme 4.
During the investigation of the reactivity of PhPfc with metal
complexes, Miyoshi et al. discovered, for the first time, that
polymerization of phosphorus-bridged [1]FCPs occur under UV
radiation.[6c] Shortly after, the first evidence of possible inter-
mediates in this photolytic ROP was published.[6d] As PhP(=S)fc
in thf was completely converted into polymers under irradia-
tion, the same reaction in the presence of P(OMe)3 lead to the
isolation of compound F (Figure 10). Species F was structurally
characterized and is related to our proposed intermediate L24
(Scheme 4).[6d]
The most interesting structure is that of anti-C2-(4S)2
(Figure 9) as both iPr groups are bound to the same ferrocene
moiety. This proves that in the course of the formation of this
species FeÀCp bonds were broken. From a formal viewpoint,
anti-C2-42 is
a head-to-head dimer of the monomer 4
(Scheme 3).
Scheme 3. Formal head-to-head dimerization of 4 resulting in anti-C2-42.
The proposed reaction mechanism shown in Scheme 4 for
the thermal ROP of species 4 is based on the current knowl-
edge about the photolytic ROP.[4b,6c,d,7r,28] Monomer 4 has two
different FeÀCp bonds and one can assume that both will
break with a certain ratio. One possibility, the initial cleavage
of the FeÀCp bond of the nonalkylated ring, is illustrated in
Scheme 4. Proposed mechanism of the thermal ROP of 4 (see discussion for
details).
Figure 10. Characterized species in investigations of photolytic ROP.
&
&
Chem. Eur. J. 2016, 22, 1 – 13
www.chemeurj.org
6
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
It was speculated that P(OMe)3 replaces the weaker coordi-
nating thf, which is the iron-stabilizing ligand in the polymeri-
zation. However, if P(OMe)3 was replaced by the stronger coor-
dinating ligand PMe3, the ring-open product G was identified
by NMR spectroscopy.[6d] On the basis of these discoveries,
Manners et al. have shown that the Me2Si-bridged [1]FCP
Me2Sifc (Figure 1; A with ERx =SiMe2) polymerizes in a living
process in thf under irradiation with NaCp as an anionic initia-
tor.[28a] Exchange of the anionic initiator by a tridentate terpyri-
dine gave polymers and the dimer (Me2Sifc)2.[28b] The proposed
mechanism involved the terpyridyl stabilized intermediate H
(Figure 10), an analogue of compound L342 proposed in
Scheme 4. Further evidence for the initial h5 to h1 haptotropic
shift was found for certain [1]- and [2]FCPs, as well as for
carbon-bridged [2]ruthenocenophanes.[28c,d] In these studies,
PMe3 and Me2PCH2CH2PMe2 had been applied as ligands. It
was discovered that less strained species (tilt angle a~19–228)
require photoactivation for the h5 to h1 haptotropic shift,
which is reversible, whereas for the highly strained [2]rutheno-
cenophanes (tilt angle a~29–318) the h5 to h1 haptotropic
shift occurs thermally and is irreversible. Several species of
type F, which are related to L24, were isolated and structurally
characterized. Please note that for the proposed intermediate
L24 only one of many feasible isomers is shown in Scheme 4;
these isomers differ with respect to the position of the allylic
proton on the h1 bound Cp ring. Moreover, structural evidence
for the species of type G, and hence for L34, was obtained by
the structural characterization of I, which was obtained from ir-
radiating a solution of PMe3 and dicarba[2]ferrocenophane in
EtOH, followed by crystallization as a tetraphenylborate salt.[28d]
Importantly, the three phosphine ligands in I were completely
substituted by CpÀ when a solution of I was irradiated in the
presence of NaCp. This result provided evidence for the back-
biting process that leads to cyclic compounds, for example,
L342 to anti-C2-42 (Scheme 4).
very well with that from B3PW91-D3(BJ)/6-311+G(d,p) calcula-
tions. The thermal ROP of trans-4 at 2308C was investigated re-
vealing that approximately half of the monomer transformed
into polymers while the other half gave cyclic oligomers. The
most interesting result of our study came about when, after
sulfurization, separation of the cyclic oligomers was attempted
with PTLC. From the obtained six fractions, three contained
a single species and their structures were solved by single-crys-
tal X-ray analysis revealing the presence of isomeric dimers.
The formation of one of these dimers, namely anti-C2-(4S)2
(Figure 9), can only be rationalized if one assumes that FeÀCp
bonds in the starting monomer 4 broke. Based on this insight,
a polymerization mechanism is proposed (Scheme 4), which is
similar to that of the known photolytic ROP of [n]FCPs. Where-
as in photolytic ROP absorbed light weakens the FeÀCp bond
by excitation of an electron from the HOMO to an antibonding
LUMO to allow bond breakage at low temperature (e.g.
58C),[6e,30] thermal ROP of 4 requires higher temperatures (e.g.
2308C) to allow breakage of unweakened FeÀCp bonds. In
either case, donor ligands are needed as a substitute for the
shifting Cp moiety. For photolytic ROP, solvent molecules or
added initiators are the known donor ligands. In the case of
the thermal ROP of 4, the identity of the donor ligand L is un-
known (Scheme 4). However, the melt of the ROP process con-
tains a wealth of different phosphines from the monodentate
monomer 4 to polydentate cyclic oligomers or polymers and
we speculate that they take up the role of the needed donor
ligands. This mechanism, which involves the FeÀCp bond
cleavage, contrasts that of the Me2Si-bridged [1]FCP, which
occurs through cleavage of SiÀCipso bonds during the thermal
ROP process.[31]
Experimental Section
General methods
Manners et al. investigated the thermal ROP of dicarba[2]fer-
rocenophanes and discovered that, depending on the substitu-
ents on the bridging carbon atoms, either CÀC or FeÀCp bond
cleavage occurs.[29] To the best of our knowledge, this is the
only account of a thermal ROP in which the occurrence of FeÀ
Cp bond cleavage was revealed.[29b]
If not mentioned otherwise, all syntheses were carried out using
standard Schlenk and glovebox techniques. Solvents were dried
using an MBraun Solvent Purification System and stored under ni-
trogen over 3 ꢃ molecular sieves. All solvents for NMR spectrosco-
py were prepared by pump-freeze-thaw cycles and stored under
nitrogen over 3 ꢃ molecular sieves. Unless otherwise noted, tem-
peratures refer to that of the bath (e.g., dry ice/acetone bath for
À788C). Flash chromatography was performed with neutral alumi-
num oxide and silica gel 60; mixed solvent eluents are reported as
vol/vol solutions. For controlled addition of a solution of PhPCl2
a syringe pump was used (Sage Instruments, model 355).
Conclusion
The two racemic, phosphorus-bridged [1]FCPs cis- and trans-4
were prepared using the common salt-metathesis approach
(Scheme 1). Making use of differences in solubility and physical
appearance of both solids, cis-4 and trans-4 could be separat-
ed from each other. This allowed the investigation of their
thermal properties individually and revealed that trans-4 is
thermodynamically more stable than cis-4. At elevated temper-
ature thermal equilibrium between both isomers exists that ex-
pectedly occurs through inversion at the bridging phosphorus
atom. DFT calculations revealed a convincing agreement be-
tween the experimental and calculated structure of cis-4. Fur-
thermore, the experimentally determined difference of
5 kJmolÀ1 between the free energies of both isomers agreed
Characterization methods
1H, 13C, and 31P NMR spectra were recorded on a 500 MHz Bruker
Avance, a 500 MHz Bruker Avance III HD, and a 600 MHz Bruker
Avance III HD NMR spectrometer at 258C in C6D6 and CDCl3. 1H
chemical shifts were referenced to the residual protons of the deu-
terated solvents (d=7.15 ppm for C6D6; d=7.26 ppm for CDCl3);
13C chemical shifts were referenced to the C6D6 signal at d=
128.00 ppm and the CDCl3 signal at d=77.00 ppm. Coupling con-
stants are reported to the nearest 0.5 Hz (1H NMR spectroscopy) or
rounded to integer values in Hz (13C and 31P NMR spectroscopy).
31P NMR chemical shifts were reported relative to 85% H3PO4 in
Chem. Eur. J. 2016, 22, 1 – 13
www.chemeurj.org
7
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
D2O used as an external reference. Assignments were supported
by additional NMR experiments (1, 3, cis-4, trans-4: DEPT, HMQC,
COSY; (4S)n: HMQC, COSY). As signals of Cp protons show a fine
structure, all signals were called multiplets. HRMS were measured
on a JEOL AccuTOF GCv 4G using field desorption ionization (FDI).
For the isotopic pattern, only the mass peak of the isotopoloque
or isotope with the highest natural abundance is given. DSC analy-
ses were performed at a heating rate of 108C minÀ1 (Q20 from TA
Instruments). The samples, sealed in hermetic aluminum pans,
were tared using a balance with a repeatability of 0.01 mg (AB204-
S Mettle Toledo); around 3 mg of sample were used per measure-
ment. DSC data were analyzed with the TA Instruments Universal
Analysis 2000 software. Absolute molecular weights were obtained
by triple detection gel permeation chromatography (GPC) employ-
ing a Viscotek 350 HT-GPC system (Malvern) that was used at
a column temperature of 458C (thf; flow rate=1.0 mLminÀ1). The
instrument was equipped with the following Viscotek components:
autosampler (Model 430 Vortex), degasser (model 7510), two
pumps (model 1122), 7 and 908 light-scattering detectors, refrac-
tometer, and viscometer. GPC columns cover the range of Mw of
500 to 10,000,000 gmolÀ1 (three main columns: Plgel 10 mm
MIXED-B LS 300ꢂ7.5 mm; one guard column: 10 mm GUARD 50ꢂ
7.5 mm; Agilent Technologies). Samples were dissolved in thf,
stirred overnight, and filtered through 0.45 mm PTFE syringe filters
before GPC analysis. Elemental analyses were performed on
a Perkin–Elmer 2400 CHN Elemental Analyzer.
ed, the aqueous phase was extracted with Et2O (3ꢂ120 mL), and
the organic layers were washed with brine. The organic layers
were combined, dried over anhydrous Na2SO4, filtered, and concen-
trated using a rotatory evaporator. The crude material was purified
by flash column chromatography (hexanes/EtOAc; 2:1; plus 5 vol%
NEt3) and afforded the product as an orange oil (1.78 g, 72%).
1H NMR (CDCl3, 500 MHz): d=1.52 [d, 3H, J=7.0 Hz, CH(NMe2)CH3],
2.14 [s, 6H, CH(NMe2)CH3], 3.77 [q, 1H, J=7.0 Hz, CH(NMe2)CH3],
4.06 (m, 1H, Cp), 4.09 (m, 1H, Cp), 4.14 (m, 1H, Cp), 4.20 (pst, 1H,
J=2.5 Hz, Cp), 4.37 (m, 2H, Cp), 4.47 ppm (m, 1H, Cp); 13C{1H} NMR
(CDCl3, 126 MHz): d=16.6 [CH(NMe2)CH3], 40.9 [CH(NMe2)CH3], 55.9
[CH(NMe2)CH3], 67.8 (CH of Cp), 69.1 (CH of Cp), 70.90 (CH of Cp),
70.94 (CH of Cp), 72.5 (CH of Cp), 73.0 (CH of Cp), 73.4 (CH of Cp),
78.5 (Cipso of Cp), 80.2 (Cipso of Cp), 87.8 ppm, (Cipso of Cp); HRMS
(FDI; m/z): [M+] calcd for C14H17Br2FeN: 412.9077; found: 412.9075.
Synthesis of rac-2-[1-(acetoxy)ethyl]-1,1’-dibromoferrocene
A mixture of 2 (1.78 g, 4.30 mmol) and acetic anhydride (6.6 mL,
70 mmol) was thoroughly degassed and stirred for 10 h at 708C.
All volatiles were removed at 508C under high vacuum to give the
product as a brown oil (1.77 g, 96%).
Note: Due to the sensitivity of this species to eliminate AcOH fur-
ther characterization or purification was not performed. 1H NMR
(CDCl3, 500 MHz): d=1.67 [d, 3H, J=6.5 Hz, CH(OCOCH3)CH3], 2.02
[s, 3H, CH(OCOCH3)CH3], 4.09 (m, 1H, Cp), 4.14 (m, 1H, Cp), 4.26
(m, 1H, Cp), 4.28 (m, 1H, Cp), 4.40 (m, 2H, Cp), 4.51 (m, 1H, Cp),
6.01 ppm [q, 1H, J=6.5 Hz, CH(OCOCH3)CH3].
Reagents
nBuLi (2.5m in hexanes), dichlorophenylphosphine (97%), ferro-
cene (98%), AlMe3 (2.0m in hexanes), and 1,2-dibromotetrachloro-
ethane (Acros, 97%) were purchased from Sigma–Aldrich. Triethyla-
mine (reagent grade, 99%) was purchased from Fisher Chemicals.
N,N,N’,N’-tetramethylethylenediamine (Alfa Aesar, 99%) and acetic
anhydride (EMD, ACS grade, 99%) were purchased from VWR. Ele-
mental sulfur (99%) and PTLC plates (glass plates (20ꢂ20 cm) pre-
coated (0.25 mm) with silica gel 60 F254) were purchased from
EMD. Silica gel 60 (EMD, Geduran, particle size 0.040–0.063 mm)
and aluminum oxide (Sigma–Aldrich, activated, neutral, Brockmann
I, 58 ꢃ pore size) were used for flash column chromatography. 1-
Acetylferrocene was prepared as described.[16,32] Species rac-[1-(hy-
droxy)ethyl]ferrocene, rac-[1-(acetoxy)ethyl]ferrocene, and rac-[1-
(dimethylamino)ethyl]ferrocene (1) were synthesized according to
the reported procedure with some alterations (see the Supporting
Information).[16]
Synthesis of rac-1,1’-dibromo-2-isopropylferrocene (3)
A solution of AlMe3 (8.3 mL, 2.0m in hexanes, 17 mmol) was added
dropwise to a solution of rac-2-[1-(acetoxy)ethyl]-1,1’-dibromoferro-
cene (1.77 g, 4.13 mmol) in CH2Cl2 (49 mL) at À788C. The reaction
mixture was stirred for 30 min at À788C, warmed up to rt, and
then stirred for an additional 30 min. The mixture was transferred
by a cannula into a saturated aqueous NaHCO3 solution (25 mL) at
08C, followed by addition of a saturated aqueous sodium potassi-
um tartrate solution (25 mL). From this point on manipulations
were done without inert-gas protection. All volatiles were removed
under high vacuum and the residue was dissolved in Et2O (25 mL).
After the resulting solution was stirred vigorously for 15 min, it was
acidified with 1m HCl (aq). The phases were separated and the
aqueous phase was extracted with Et2O (2ꢂ50 mL). The organic
layers were washed with saturated aqueous NaHCO3 solution,
water, and brine. The combined organic layers were dried over an-
hydrous Na2SO4, filtered, and concentrated using a rotatory evapo-
rator. The resulting red oil (1.53 g, 96%) was purified by flash
column chromatography (hexanes/Et2O; 20:1), yielding the product
as an orange oil. Further purification was done by three flask-to-
flask condensations (408C; p ~10À2 mbar) and gave 0.90 g of prod-
uct 3 (56%).
Synthesis of rac-2-[1-(dimethylamino)ethyl]-1,1’-dibromofer-
rocene (2)
The procedure is based on the published procedure of the enan-
tiomerically pure product.[33] To a stirred solution of 1 (1.55 g,
6.03 mmol) in Et2O (12 mL), nBuLi (2.5m in hexanes, 12 mL,
30 mmol) was added dropwise over 10 min at rt. Subsequently,
tmeda (2.3 mL, 15 mmol) was added dropwise over 5 min at rt.
After several minutes, the color of the reaction mixture changed
from orange to red and the mixture was stirred overnight. After
the addition of Et2O (60 mL) and cooling to À788C, a solution of
1,2-dibromotetrachloroethane (11.0 g, 33.8 mmol) in thf (100 mL)
was added dropwise to the reaction mixture by a cannula over
20 min at À788C. The resulting yellow suspension was stirred an-
other 30 min at À788C, followed by stirring for 2.5 h at rt, which
resulted in a change of color from yellow to brown. After the addi-
tion of H2O (100 mL) to the suspension, all following manipulations
were done without inert-gas protection. The phases were separat-
Note: Compound 3 contained ca. 1% of rac-1-bromo-2-isopropyl-
1
ferrocene as an impurity. H NMR (CDCl3, 500 MHz): d=1.09 [d, 3H,
J=7.0 Hz, CH(CH3)2], 1.38 [d, 3H, J=7.0 Hz, CH(CH3)2], 2.85 [sept,
1H, J=7.0 Hz, CH(CH3)2], 4.07 (m, 2H, Cp), 4.10 (m, 2H, Cp), 4.36
(m, 2H, Cp), 4.41 ppm (m, 1H, Cp); 13C{1H} NMR (CDCl3, 126 MHz):
d=21.5 [CH(CH3)2], 24.2 [CH(CH3)2], 26.3 [CH(CH3)2], 66.0 (CH of
Cp), 68.1 (CH of Cp), 70.3 (CH of Cp), 70.6 (CH of Cp), 72.32 (CH of
Cp), 72.33 (CH of Cp), 72.8 (CH of Cp), 78.5 (Cipso of Cp), 79.4 (Cipso
of Cp), 95.9 ppm (Cipso of Cp); HRMS (FDI) m/z: calcd for
C13H14Br2Fe: 383.8812 [M+]; found: 383.8823.
&
&
Chem. Eur. J. 2016, 22, 1 – 13
www.chemeurj.org
8
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
J(13C/31P)=23 Hz, iPr-Cipso of Cp], 127.6 (s, p-Ph), 128.8 [d, J(13C/
31P)=4 Hz, m-Ph], 130.8 [d, J(13C/31P)=14 Hz, o-Ph], 138.0 ppm [d,
J(13C/31P)=11 Hz, Cipso of Ph]; 31P{1H} NMR (C6D6, 202.5 MHz): d=
9.6 ppm.
Synthesis of phospha[1]ferrocenophane 4
Species 3 (0.470 g, 1.22 mmol) was dissolved in a solvent mixture
(12 mL; hexanes/thf; 9:1) and cooled to 08C. After a solution of
nBuLi (2.5m in hexanes, 1.0 mL, 2.5 mmol) was added dropwise,
the reaction mixture was stirred at this temperature for 30 min.
The orange reaction mixture was warmed up to 228C (water bath)
and a solution of PhPCl2 (0.55m in hexanes (0.506 g from 97%
PhPCl2 in 5.0 mL of hexanes), 2.4 mL, 1.3 mmol) was added drop-
wise within 10 min with a syringe pump at 228C. With every drop
of addition a color change from orange to dark-red and the forma-
tion of a white precipitate was observed. After the reaction mixture
was stirred for additional 25 min at 228C, it was transferred to
a column packed with alumina in the solvent mixture of hexanes/
EtOAc/Et3N of 10:1:1 under N2 pressure. Chromatography gave
a dark-red fraction, which yielded a dark-red oil (0.205 g, 50%)
after all volatiles were removed. 1H NMR spectroscopy revealed
that this mixture was composed of cis-4 (44%) and trans-4 (56%).
Analytically pure product was obtained through a flask-to-flask
condensation (768C; p ~10À2 mbar), which resulted in a mixture of
a crystalline and an oily substance (0.169 g, 42%; cis-4=36%;
trans-4=64%). This mixture was dissolved in hexanes (3.0 mL) and
left at À208C for 16 h, resulting in trans-4 as a pink-red powder
along with 6% of cis-4 (0.089 g, 22%). The dark-red mother liquor
solution was left at À808C for 16 h, resulting in a solid composed
of a 6.6:3.4 mixture of cis-4/trans-4 (0.068 g, 17%). From this mix-
ture, dark-red crystals were picked and single-crystal X-ray analysis
revealed their identity as cis-4. Further mechanical separation of
the dark-red crystals (cis-4) from this mixture led to fairly clean
samples of each isomer (see below). HRMS (FDI) m/z: calcd for
C19H19FeP: 334.0574 [M+]; found: 334.0573; elemental analysis
calcd (%) for C19H19FeP (334.06): C 68.29, H 5.73; found: C 68.38, H
5.57.
Isomerization of cis-4 to trans-4
Species cis-4 (3.30 mg, contaminated with 1.3% of trans-4) was
heated to 1378C for 22 h in a flame-sealed Pyrex NMR tube. The
dark-red crystalline powder turned into a red oil, which was dis-
solved in C6D6. The 1H NMR spectrum of this red oil revealed
a trans-4/cis-4 ratio of 9:1. Heating of cis-4 (3.90 mg, contaminated
with 1.3% of trans-4) at 1378C for 8 h gave the same ratio. Heating
of trans-4 (4.00 mg, contaminated with 1.8% of cis-4) to 1378C for
8 h gave the same ratio.
Thermal ROP of 4
Monomer 4 (75.4 mg; trans-4/cis-4 of 20:1) was heated to 2308C
for 2 h in a flame-sealed Pyrex NMR tube. After letting the sample
cool down to rt, a dark-red glassy solid was obtained that dis-
solved in thf (0.6 mL). Precipitation into hexanes (4 mL) in a Schlenk
flask afforded a yellow solid and an orange-colored supernatant.
Removal of all the volatiles from the solution resulted in an orange
powder (26.2 mg, 35%), which was used to separate cyclic species
(see below). The yellow solid was washed with hexanes (3ꢂ2 mL),
dried under high vacuum to give polymer 4n as a light-yellow
powder (41.7 mg, 55%). 1H NMR (C6D6, 600 MHz): d=0.45–1.74
[6H, CH(CH3)2], 2.77–3.32 [1H, CH(CH3)2], 3.38–4.79 (7H, Cp), 6.93–
7.42 (3H, Ph), 7.51–7.94 ppm (2H, Ph); 13C{1H} NMR (C6D6,
151 MHz): d=22.5–24.3, 25.1–26.2, 26.2–27.5 [CH(CH3)2 and
CH(CH3)2], 67.9–71.4, 71.4–73.8, 75.3–79.5 (CH of Cp and P-Cipso of
Cp], 101.9 (iPr-Cipso of Cp), 129.0, 134.7, 135.7, 140.0 ppm (CH of Ph
and Cipso of Ph); 31P{1H} NMR (C6D6, 243 MHz): d=À39.0 to
À30.6 ppm (see Figure S20 in the Supporting Information).
Characterization of cis-4: Note that this sample contains 1.3% of
trans-4. 1H NMR (C6D6, 500 MHz): d=0.87 [d, 3H, J=6.5 Hz,
CH(CH3)2], 1.00 [d, 3H, J=7.0 Hz, CH(CH3)2], 2.13 [sept, 1H, J=
7.0 Hz, CH(CH3)2], 4.15 (m, 1H, Cp), 4.17 (m, 1H, Cp), 4.21 (m, 2H,
Cp), 4.38 (m, 1H, Cp), 4.53 (m, 2H, Cp), 7.00 (t, 1H, J=7.5 Hz, p-Ph),
7.10 (t, 2H, J=7.5 Hz, m-Ph), 7.71 ppm (t, 2H, J=7.0 Hz, o-Ph);
13C{1H} NMR (C6D6, 126 MHz): d=17.7 [d, J(13C/31P)=61 Hz, P-Cipso of
Cp], 18.7 [d, J(13C/31P)=49 Hz, P-Cipso of Cp], 21.2 [s, CH(CH3)2],
27.08 [d, J(13C/31P)=15 Hz, CH(CH3)2], 27.14 [s, CH(CH3)2], 75.1 [d,
J(13C/31P)=12 Hz, CH of Cp], 75.3 (s, CH of Cp), 76.2 [d, J(13C/31P)=
9 Hz, CH of Cp], 77.8 [d, J(13C/31P)=38 Hz, CH of Cp], 78.5 (s, CH of
Cp), 78.8 [d, J(13C/31P)=41 Hz, CH of Cp], 83.2 [d, J(13C/31P)=8 Hz,
CH of Cp], 107.8 [d, J(13C/31P)=9 Hz, iPr-Cipso of Cp], 127.7 (s, p-Ph),
128.6 [d, J(13C/31P)=4 Hz, m-Ph], 130.2 [d, J(13C/31P)=15 Hz, o-Ph],
138.2 ppm [d, J(13C/31P)=13 Hz, Cipso of Ph]; 31P{1H} NMR (C6D6,
202.5 MHz): d=11.3 ppm.
Sulfurization of 4n to (4S)n
Polymer 4n (41.7 mg) was dissolved in CH2Cl2 (3 mL) and allowed
to react with an excess amount of sulfur (36 mg) over 16 h. From
this point on the manipulation was done under air. After the reac-
tion mixture was filtered through a 0.2 mm syringe PTFE filter, the
solvent was removed under high vacuum. The resultant product
was dissolved in a minimum amount of CH2Cl2 and filtered again
through a 0.2 mm syringe PTFE filter. The CH2Cl2 solution was then
precipitated into hexanes (3 mL) to afford a yellow solid and a col-
orless supernatant. The yellow solid was washed with hexanes (3ꢂ
3 mL) and dried under high vacuum for 16 h, which resulted in
(4S)n as a yellow powder (42.7 mg, 93% for the sulfurization step;
1
52% relative to monomer 4). H NMR (CDCl3, 600 MHz): d=À0.14–
1.42 [6H, CH(CH3)2], 2.71–3.57 [1H, CH(CH3)2], 3.61–5.32 (7H, Cp),
7.32–8.30 ppm (5H, Ph); 13C{1H} NMR (CDCl3, 126 MHz): d=22.5
[CH(CH3)2], 24.1–25.6 [CH(CH3)2 and CH(CH3)2], 70.0–76.4 (CH of Cp
and P-Cipso of Cp), 102.0 (iPr-Cipso of Cp), 127.9 (CH of Ph), 130.4–
132.9 (CH of Ph), 132.9–135.1 ppm (Cipso of Ph); 31P{1H} NMR (CDCl3,
243 MHz): d=37.8 ppm; GPC: Mw =62.5 kDa, Mn =45.0 kDa, PDI=
1.39; elemental analysis calcd (%) for C19H19FePS (366.03): C 62.31,
H 5.23; found: C 61.81, H 4.78.
Characterization of trans-4: Note that this sample contains 1.8%
of cis-4. 1H NMR (C6D6, 500 MHz): d=1.21 [d, 3H, J=7.0 Hz,
CH(CH3)2], 1.29 [d, 3H, J=7.0 Hz, CH(CH3)2], 3.45 [sept of d, 1H,
J(1H/1H)=6.5 Hz, J(1H/31P)=2.5 Hz, CH(CH3)2], 4.13 (m, 1H, Cp), 4.17
(m, 1H, Cp), 4.19 (m, 2H, Cp), 4.21 (m, 1H, Cp), 4.26 (m, 1H, Cp),
4.29 (m, 1H, Cp), 7.04 (t, 1H, J=7.5 Hz, p-Ph), 7.14 (t, 2H, J=
7.5 Hz, m-Ph), 7.62 ppm (t, 2H, J=7.0 Hz, o-Ph); 13C{1H} NMR (C6D6,
126 MHz): d=15.5 [d, J(13C/31P)=55 Hz, P-Cipso of Cp], 19.1 [d, J(13C/
31P)=55 Hz, P-Cipso of Cp], 21.2 [s, CH(CH3)2], 27.3 [d, J(13C/31P)=
10 Hz, CH(CH3)2], 27.4 [s, CH(CH3)2], 75.1 [d, J(13C/31P)=3 Hz, CH of
Cp], 76.1 (s, CH of Cp), 76.9 [d, J(13C/31P)=8 Hz, CH of Cp], 77.3 (s,
CH of Cp), 77.4 [d, J(13C/31P)=8 Hz, CH of Cp], 77.9 [d, J(13C/31P)=
7 Hz, CH of Cp], 82.2 [d, J(13C/31P)=33 Hz, CH of Cp], 105.5 [d,
Isolation of cyclic phosphine sulfides
The orange powder (26.2 mg, 35%) obtained from the supernatant
of the thermal ROP of 4 was dissolved in CH2Cl2 (2 mL) and al-
lowed to react with excess amount of sulfur (36 mg) during 16 h.
Chem. Eur. J. 2016, 22, 1 – 13
www.chemeurj.org
9
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
From this point on the manipulation was done under air. After the
reaction mixture was filtered through a 0.2 mm syringe PTFE filter,
the solvent was removed under high vacuum. After the product
mixture was dissolved in a minimum amount of CH2Cl2 and filtered
again through a 0.2 mm syringe PTFE filter, all volatiles were re-
moved and the resultant orange solid was dried under high
vacuum (49.7 mg). The obtained solid was divided into two por-
tions and PTLC was performed using two glass plates (20ꢂ20 cm;
pre-coated (0.25 mm) with silica gel 60 F254). A solvent mixture of
hexanes/EtOAc (3:1) was applied as an eluent, followed by a sol-
vent mixture of hexanes/acetone (6:1). Materials were detected by
visualization under an ultraviolet lamp (l=254 nm). The corre-
sponding strips were scraped off and washed with a mixture of
10 vol% MeOH in CH2Cl2 (3ꢂ15 mL) and filtered through a fine frit-
ted funnel. Overall 6 fractions (strips) were obtained and named
from top to bottom of the glass plates as fractions A, B, C, D, E,
and F. From each fraction approximately 2 mg were obtained. Suit-
able crystals of A, B, and F for X-ray structural analysis were ob-
tained from CH2Cl2 solutions by slow evaporation at rt.
atoms. Crystallographic data are summarized in Table S1, Support-
ing Information, while bond lengths and bond angles are shown in
Tables S2–S5 (Supporting Information). Ellipsoid plots were pre-
pared using ORTEP-3 for Windows.[37] The common set of distor-
tion angles in cis-4 was calculated using the programs PLATON.[38]
The esds of all distortion angles that involve centroids of Cp rings
(b, d, and t) might be somewhat smaller than they should be, as
esds on the positions of centroids were not included in the calcula-
tion.
DFT calculations
All calculations were done employing the software package GAUS-
SIAN 09.[39] Geometries were optimized at the B3PW91/6–311+
G(d,p) level.[40] with and without inclusion of the D3(BJ)[41] disper-
sion corrections. The B3PW91 functional had been chosen based
on the benchmark investigation of Grimme et al.[42] and our recent
application to boron-bridged [1]FCPs.[9f] Frequency calculations
were used to confirm minima and saddle points and provided ther-
modynamic information. An ultrafine grid (int=ultrafine) and tight
requirements for geometry optimizations (opt=tight) were used
for all calculations. The notation used for free energies, DG8, indi-
cate standard conditions (p=1 atm.; T=298.15 K). Graphical illus-
trations of calculated results were done with the help of ORTEP-3
for Windows (version 2.02)[37] and CYLview (version 1b).[43] Extrac-
tion of structural parameters from the calculated coordinates were
done with the help of Mercury (version 3.7)[44] and CYLview (ver-
sion 1b).[43]
NMR data for the clean fractions are given below (see also Figur-
es S26–37 in the Supporting Information).
Fraction A: 1H NMR (CDCl3, 500 MHz):d=0.34 [d, 6H, J=6.5 Hz,
CH(CH3)2], 1.31 [d, 6H, J=7.0 Hz, CH(CH3)2], 3.77 [sept, 2H, J=
7.0 Hz, CH(CH3)2], 4.42 (m, 2H, Cp), 4.44 (m, 2H, Cp), 4.66 (m, 2H,
Cp), 4.77 (m, 2H, Cp), 4.93 (m, 2H, Cp), 5.63 (m, 2H, Cp), 5.91 (m,
2H; Cp), 7.17–7.24 (m, 6H, Ph), 7.62 ppm (m, 4H, Ph); 31P{1H} NMR
(CDCl3, 202.5 MHz): d=39.2 ppm.
Fraction B: 1H NMR (CDCl3, 500 MHz): d=0.21 [d, 6H, J=7.0 Hz,
CH(CH3)2], 1.25 [d, 6H, J=7.0 Hz, CH(CH3)2], 3.57 [sept, 2H, J=
7.0 Hz, CH(CH3)2], 4.45 (m, 2H, Cp), 4.47 (m, 2H, Cp), 4.64 (m, 2H,
Cp), 4.79 (m, 2H, Cp), 4.88 (m, 2H, Cp), 5.24 (m, 2H, Cp), 6.27 (m,
2H, Cp), 7.17–7.25 (m, 6H, Ph), 7.65 ppm (m, 4H, Ph); 31P{1H} NMR
(CDCl3, 202.5 MHz): d=38.3 ppm.
Acknowledgements
We thank the Natural Sciences and Engineering Research
Council of Canada (NSERC Discovery Grant, JM) for support.
We thank the Canada Foundation for Innovation (CFI) and the
government of Saskatchewan for funding of the X-ray and
NMR facilities in the Saskatchewan Structural Sciences Centre
(SSSC). For DFT calculations, we are grateful to WestGrid
(www.westgrid.ca) and Compute Canada (www.computecana-
da.ca) for support. We thank K. Thoms (University of Saskatche-
wan; CHN analysis and MS) for support and measurements. We
thank Prof. S. R. Foley (University of Saskatchewan) for making
the GPC instrument available for our studies.
Fraction F: 1H NMR (CDCl3, 500 MHz): d=0.44 [d, 6H, J=7.0 Hz,
CH(CH3)2], 1.33 [d, 6H, J=7.0 Hz, CH(CH3)2], 3.75 [sept, 2H, J=
7.0 Hz, CH(CH3)2], 4.01 (m, 2H, Cp), 4.37 (m, 2H, Cp), 4.40 (m, 2H,
Cp), 4.65 (m, 2H, Cp), 4.69 (m, 2H, Cp), 5.62 (m, 2H, Cp), 5.91 (m,
2H, Cp), 7.17–7.26 ppm (m, 6H, Ph), 7.50 ppm (m, 4H, Ph);
31P{1H} NMR (CDCl3, 202.5 MHz): d=36.1 ppm.
Crystal structure determinations of cis-4, syn-C2-(4S)2, anti-
[27]
C2-(4S)2, and anti-Ci-(4S)2
Keywords: ferrocenophanes · metallopolymers · phosphanes ·
phosphorus · strained sandwich complexes
Single crystals were coated with Paratone-N oil, mounted using
a micromount (MiTeGen-Microtechnologies for Structural Genom-
ics), and frozen in the cold stream of an Oxford Cryojet attached to
the diffractometer. Crystal data were collected on a Bruker APEX II
diffractometer at À1008C using monochromated MoKa radiation
(l=0.71073 ꢃ). An initial orientation matrix and cell was deter-
mined by w scans, and the X-ray data were measured using f and
w scans.[34] The frames were integrated with the Bruker SAINT soft-
ware package[35] and data reduction was performed with the
APEX2 software package.[34] A multiscan absorption correction
(SADABS) was applied.[35] The structures were solved by the Intrin-
sic Phasing method implemented with SHELXT and refined using
the Bruker SHELXTL software package.[36] Nonhydrogen atoms
were refined with independent anisotropic displacement parame-
ters. Hydrogen atoms were placed at geometrically idealized posi-
tions (riding model) and their displacement parameters were fixed
to be 20 or 50% larger than those of the attached nonhydrogen
[1] A. G. Osborne, R. H. Whiteley, J. Organomet. Chem. 1975, 101, C27–C28.
[2] a) A. G. Osborne, R. H. Whiteley, R. E. Meads, J. Organomet. Chem. 1980,
193, 345–357; b) H. Stoeckli-Evans, A. G. Osborne, R. H. Whiteley, J. Or-
ganomet. Chem. 1980, 194, 91–101; c) D. Seyferth, H. P. Withers, J. Orga-
nomet. Chem. 1980, 185, C1–C5; d) D. Seyferth, H. P. Withers, Organo-
metallics 1982, 1, 1275–1282.
[3] H. P. Withers, D. Seyferth, J. D. Fellmann, P. E. Garrou, S. Martin, Organo-
metallics 1982, 1, 1283–1288.
[4] a) C. H. Honeyman, D. A. Foucher, F. Y. Dahmen, R. Rulkens, A. J. Lough,
I. Manners, Organometallics 1995, 14, 5503–5512; b) V. Bellas, M.
Rehahn, Angew. Chem. Int. Ed. 2007, 46, 5082–5104; Angew. Chem.
2007, 119, 5174–5197.
[5] a) P. F. Brandt, T. B. Rauchfuss, J. Am. Chem. Soc. 1992, 114, 1926–1927;
b) D. A. Foucher, B.-Z. Tang, I. Manners, J. Am. Chem. Soc. 1992, 114,
6246–6248.
&
&
Chem. Eur. J. 2016, 22, 1 – 13
www.chemeurj.org
10
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
[6] a) C. H. Honeyman, T. J. Peckham, J. A. Massey, I. Manners, Chem.
Commun. 1996, 2589–2590; b) T. J. Peckham, J. A. Massey, C. H. Honey-
man, I. Manners, Macromolecules 1999, 32, 2830–2837; c) T. Mizuta, M.
Onishi, K. Miyoshi, Organometallics 2000, 19, 5005–5009; d) T. Mizuta, Y.
Imamura, K. Miyoshi, J. Am. Chem. Soc. 2003, 125, 2068–2069; e) S. K.
Patra, G. R. Whittell, S. Nagiah, C.-L. Ho, W.-Y. Wong, I. Manners, Chem.
Eur. J. 2010, 16, 3240–3250.
[7] a) I. R. Butler, W. R. Cullen, F. W. B. Einstein, S. J. Rettig, A. J. Willis, Orga-
nometallics 1983, 2, 128–135; b) W. R. Cullen, S. J. Rettig, T. C. Zheng, J.
Organomet. Chem. 1993, 452, 97–103; c) M. Viotte, B. Gautheron, M. M.
Kubicki, Y. Mugnier, R. V. Parish, Inorg. Chem. 1995, 34, 3465–3473; d) T.
Mizuta, T. Yamasaki, H. Nakazawa, K. Miyoshi, Organometallics 1996, 15,
1093–1100; e) L. Cao, M. A. Winnik, I. Manners, J. Inorg. Organomet.
Polymers 1998, 8, 215–224; f) M. Herberhold, F. Hertel, W. Milius, B.
Wrackmeyer, J. Organomet. Chem. 1999, 582, 352–357; g) T. J. Peckham,
A. J. Lough, I. Manners, Organometallics 1999, 18, 1030–1040; h) H.
Brunner, J. Klankermayer, M. Zabel, J. Organomet. Chem. 2000, 601,
211–219; i) C. E. B. Evans, A. J. Lough, H. Grondey, I. Manners, New J.
Chem. 2000, 24, 447–453; j) L. Cao, I. Manners, M. A. Winnik, Macromo-
lecules 2001, 34, 3353–3360; k) L. Cao, J. A. Massey, T. J. Peckham, M. A.
Winnik, I. A. Manners, Macromol. Chem. Phys. 2001, 202, 2947–2953;
l) T. Mizuta, M. Onishi, T. Nakazono, H. Nakazawa, K. Miyoshi, Organome-
tallics 2002, 21, 717–726; m) B. Wrackmeyer, O. L. Tok, A. Ayazi, F.
Miyoshi, J. Am. Chem. Soc. 2006, 128, 8068–8077; c) Y. Imamura, T.
Mizuta, K. Miyoshi, Organometallics 2006, 25, 882–886.
[13] S. Dey, W. Sun, J. Mꢀller, Inorg. Chem. 2016, 55, 3630–3639.
[14] S. Sadeh, Ph. D. thesis, University of Saskatchewan 2014.
[15] D. Marquarding, H. Klusacek, G. Gokel, P. Hoffmann, I. Ugi, J. Am. Chem.
Soc. 1970, 92, 5389–5393.
[16] G. W. Gokel, I. K. Ugi, J. Chem. Educ. 1972, 49, 294–296.
[17] J. Kang, J. H. Lee, K. S. Im, J. Mol. Catal. A 2003, 196, 55–63.
[18] W. F. Bailey, M. R. Luderer, K. P. Jordan, J. Org. Chem. 2006, 71, 2825–
2828.
[19] B. Bagh, S. Sadeh, J. C. Green, J. Mꢀller, Chem. Eur. J. 2014, 20, 2318–
2327.
[20] R. D. Baechler, K. Mislow, J. Am. Chem. Soc. 1970, 92, 3090–3093.
[21] A. Rauk, L. C. Allen, K. Mislow, Angew. Chem. Int. Ed. Engl. 1970, 9, 400–
414; Angew. Chem. 1970, 82, 453–468.
[22] As a recent example for an inversion at phosphorus of a cyclic phos-
phine see: N. Humbert, E. Larionov, L. Mantilli, P. Nareddy, C. Besnard, L.
Guenee, C. Mazet, Chem. Eur. J. 2014, 20, 745–751.
[23] W. Finckh, B.-Z. Tang, D. A. Foucher, D. B. Zamble, R. Ziembinski, A.
Lough, I. Manners, Organometallics 1993, 12, 823–829.
[24] J. K. Pudelski, D. P. Gates, R. Rulkens, A. J. Lough, I. Manners, Angew.
Chem. Int. Ed. Engl. 1995, 34, 1506–1508; Angew. Chem. 1995, 107,
1633–1635.
Hertel, M. Herberhold, Z. Naturforsch.
B
2002, 57, 305–308; n) C.
[25] Matching HRMS data was obtained for dimers to hexamers; however,
the peak of the heptamer was too weak preventing the collection of
HRMS data. Mizuta et al. obtained MS data of an oligomeric mixture of
dimers to hexamers from a photolytic ring-opening reaction of PhPfc
(see ref. [7s]).
Paquet, P. W. Cyr, E. Kumacheva, I. Manners, Chem. Mater. 2004, 16,
5205–5211; o) C. Paquet, P. W. Cyr, E. Kumacheva, I. Manners, Chem.
Commun. 2004, 234–235; p) P. W. Cyr, A. J. Lough, I. Manners, Acta Crys-
tallogr. Sect. A 2005, 61, M457–M459; q) T. Mizuta, Y. Imamura, K.
Miyoshi, Organometallics 2005, 24, 990–996; r) Y. Imamura, K. Kubo, T.
Mizuta, K. Miyoshi, Organometallics 2006, 25, 2301–2307; s) T. Mizuta, T.
Aotani, Y. Imamura, K. Kubo, K. Miyoshi, Organometallics 2008, 27,
2457–2463; t) E. Khozeimeh Sarbisheh, J. C. Green, J. Mꢀller, Organome-
tallics 2014, 33, 3508–3513, 6708.
[26] F. A. Cotton, Chemical Application of Group Theory, 3rd ed., Wiley-Inter-
science, 1990.
[27] CCDC 1486781 (cis-4), 1486784 (anti-Ci-(4S)2), 1486783 (anti-C2-(4S)2),
and 1486782 (syn-C2-(4S)2) contain the supplementary crystallographic
data for this paper. These data are provided free of charge by The Cam-
bridge Crystallographic Data Centre.
[8] a) I. R. Butler, W. R. Cullen, Can. J. Chem. 1983, 61, 147–153; b) I. R.
Butler, W. R. Cullen, Organometallics 1984, 3, 1846–1851; c) I. R. Butler,
W. R. Cullen, F. W. B. Einstein, A. C. Willis, Organometallics 1985, 4, 603–
604; d) I. R. Butler, W. R. Cullen, Organometallics 1986, 5, 2537–2542;
e) I. R. Butler, W. R. Cullen, S. J. Rettig, Organometallics 1987, 6, 872–
880; f) I. R. Butler, Organometallics 1992, 11, 74–83; g) I. R. Butler, Poly-
hedron 1992, 11, 3117–3121; h) T. J. Kim, Y. H. Kim, H. S. Kim, S. C. Shim,
Y. W. Kwak, J. S. Cha, H. S. Lee, J. K. Uhm, S. I. Byun, Bull. Korean Chem.
Soc. 1992, 13, 588–592; i) I. R. Butler, W. R. Cullen, S. J. Rettig, A. S. C.
White, J. Organomet. Chem. 1995, 492, 157–164; j) H. Brunner, M.
Janura, Synthesis 1998, 45–55; k) M. Laly, R. Broussier, B. Gautheron, Tet-
rahedron Lett. 2000, 41, 1183–1185; l) J. C. Hierso, F. Lacassin, R. Broussi-
er, R. Amardeil, P. Meunier, J. Organomet. Chem. 2004, 689, 766–769.
[9] a) H. Braunschweig, F. M. Breitling, E. Gullo, M. Kraft, J. Organomet.
Chem. 2003, 680, 31–42; b) D. E. Herbert, U. F. J. Mayer, I. Manners,
Angew. Chem. Int. Ed. 2007, 46, 5060–5081; Angew. Chem. 2007, 119,
5152–5173; c) H. Braunschweig, T. Kupfer, Eur. J. Inorg. Chem. 2012,
1319–1332; d) R. A. Musgrave, A. D. Russell, I. Manners, Organometallics
2013, 32, 5654–5667; e) A. D. Russell, R. A. Musgrave, L. K. Stoll, P. Choi,
H. Qiu, I. Manners, J. Organomet. Chem. 2015, 784, 24–30; f) H. Bhatta-
charjee, J. Mꢀller, Coord. Chem. Rev. 2016, 314, 114–133.
[28] a) M. Tanabe, I. Manners, J. Am. Chem. Soc. 2004, 126, 11434–11435;
b) W. Y. Chan, A. J. Lough, I. Manners, Chem. Eur. J. 2007, 13, 8867–
8876; c) M. Tanabe, S. C. Bourke, D. E. Herbert, A. L. Lough, I. Manners,
Angew. Chem. Int. Ed. 2005, 44, 5886–5890; Angew. Chem. 2005, 117,
6036–6040; d) D. E. Herbert, M. Tanabe, S. C. Bourke, A. J. Lough, I.
Manners, J. Am. Chem. Soc. 2008, 130, 4166–4176.
[29] a) D. E. Herbert, J. B. Gilroy, A. Staubitz, M. F. Haddow, J. N. Harvey, I.
Manners, J. Am. Chem. Soc. 2010, 132, 1988–1998; b) J. B. Gilroy, A. D.
Russell, A. J. Stonor, L. Chabanne, S. Baljak, M. F. Haddow, I. Manners,
Chem. Sci. 2012, 3, 830–841.
[30] M. Tanabe, G. W. M. Vandermeulen, W. Y. Chan, P. W. Cyr, L. Vanderark,
D. A. Rider, I. Manners, Nat. Mater. 2006, 5, 467–470.
[31] a) J. K. Pudelski, I. Manners, J. Am. Chem. Soc. 1995, 117, 7265–7266;
b) J. K. Pudelski, D. A. Foucher, C. H. Honeyman, P. M. Macdonald, I. Man-
ners, S. Barlow, D. Ohare, Macromolecules 1996, 29, 1894–1903.
[32] K. Tappe, P. Knochel, Tetrahedron: Asymmetry 2004, 15, 91–102.
[33] X. Feng, B. Pugin, E. Kꢀsters, G. Sedelmeier, H.-U. Blaser, Adv. Synth.
Catal. 2007, 349, 1803–1807.
[34] Bruker, 2014.3–0 ed., Bruker AXS Inc., Madison, Wisconsin, 2014.
[35] Bruker, v8.34a ed., Bruker AXS Inc., Madison, WI, USA, 2013.
[36] a) G. M. Sheldrick, University of Gçttingen, Germany; b) G. M. Sheldrick,
Acta Crystallogr. Sect. A 2008, 64, 112–122.
[10] a) R. Rulkens, D. P. Gates, D. Balaishis, J. K. Pudelski, D. F. McIntosh, A. J.
Lough, I. Manners, J. Am. Chem. Soc. 1997, 119, 10976–10986; b) N. S.
Ieong, W. Y. Chan, A. J. Lough, M. R. Haddow, I. Manners, Chem. Eur. J.
2008, 14, 1253–1263.
[37] L. J. Farrugia, J. Appl. Crystallogr. 1997, 30, 565.
[38] A. L. Spek, University of Utrecht, The Netherlands, 2011.
[39] M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R.
Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson, H. Na-
katsuji, M. Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G.
Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Ha-
segawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven,
J. J. A. Montgomery, J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E.
Brothers, K. N. Kudin, V. N. Staroverov, T. Keith, R. Kobayashi, J. Nor-
mand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi,
M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox, J. B. Cross, V.
Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev,
A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin, K. Moro-
kuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dap-
[11] a) A. Berenbaum, H. Braunschweig, R. Dirk, U. Englert, J. C. Green, F.
Jꢄkle, A. J. Lough, I. Manners, J. Am. Chem. Soc. 2000, 122, 5765–5774;
b) H. Braunschweig, R. Dirk, M. Mꢀller, P. Nguyen, R. Resendes, D. P.
Gates, I. Manners, Angew. Chem. Int. Ed. Engl. 1997, 36, 2338–2340;
Angew. Chem. 1997, 109, 2433–2435; c) S. Sadeh, G. Schatte, J. Mꢀller,
Chem. Eur. J. 2013, 19, 13408–13417; d) S. Sadeh, H. Bhattacharjee, E.
Khozeimeh Sarbisheh, J. W. Quail, J. Mꢀller, Chem. Eur. J. 2014, 20,
16320–16330; e) H. Bhattacharjee, J. D. Martell, E. Khozeimeh Sarbisheh,
S. Sadeh, J. W. Quail, J. Mꢀller, Organometallics 2016, 35, 2156–2164.
[12] a) J. D. Fellmann, P. E. Garrou, H. P. Withers, D. Seyferth, D. D. Traficante,
Organometallics 1983, 2, 818–825; b) W. Affo, H. Ohmiya, T. Fujioka, Y.
Ikeda, T. Nakamura, H. Yorimitsu, K. Oshima, Y. Imamura, T. Mizuta, K.
Chem. Eur. J. 2016, 22, 1 – 13
www.chemeurj.org
11
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
prich, A. D. Daniels, O. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski,
D. J. Fox, Revision E.01 ed., Gaussian, Inc., Wallingford CT, 2013.
Functional Theory: Recent Progress and New Directions (Eds.: J. F.
Dobson, G. Vignale, M. P. Das), Plenum, 1998, pp. 81–111.
[41] S. Grimme, S. Ehrlich, L. Goerigk, J. Comput. Chem. 2011, 32, 1456–
1465.
[42] L. Goerigk, S. Grimme, Phys. Chem. Chem. Phys. 2011, 13, 6670–6688.
[43] C. Y. Legault, Vol. 1.0b, Universitꢅ de Sherbrooke, 2009.
[44] Mercury, http://www.ccdc.cam.ac.uk/mercury.
[40] a) J. P. Perdew, in Electronic Structure of Solids ‘91 (Eds.: P. Ziesche, H.
Eschrig), Akademie Verlag, Berlin, 1991, pp. 11–20; b) J. P. Perdew, J. A.
Chevary, S. H. Vosko, K. A. Jackson, M. R. Pederson, D. J. Singh, C. Fiol-
hais, Phys. Rev. B 1992, 46, 6671–6687; c) J. P. Perdew, J. A. Chevary,
S. H. Vosko, K. A. Jackson, M. R. Pederson, D. J. Singh, C. Fiolhais, Phys.
Rev. B 1993, 48, 4978–4978; d) A. D. Becke, J. Chem. Phys. 1993, 98,
5648–5652; e) J. P. Perdew, K. Burke, Y. Wang, Phys. Rev. B 1996, 54,
16533–16539; f) K. Burke, J. P. Perdew, Y. Wang, in Electronic Density
Received: June 23, 2016
Published online on && &&, 0000
&
&
Chem. Eur. J. 2016, 22, 1 – 13
www.chemeurj.org
12
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
FULL PAPER
&
Sandwich Complexes
E. Khozeimeh Sarbisheh, J. Esteban Flores,
J. Zhu, J. Mꢀller*
Metallopolymers: A phospha[1]ferroce-
nophane was prepared as a mixture of
cis/trans isomers. These diastereomers
were separated from each other and
their thermal interconversion was inves-
tigated experimentally and theoretically.
The thermal ring-opening polymeri-
zation of the trans species resulted in
polymers and cyclic oligomers (see
scheme).
&& – &&
Insight into the Thermal Ring-Opening
Polymerization of
Phospha[1]ferrocenophanes
Chem. Eur. J. 2016, 22, 1 – 13
www.chemeurj.org
13
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ