Cross-linked polystyrene resins containing triorganotin-4-vinylbenzoates: Assessment of their catalytic activity in transesterification reactions

Article abstract of DOI:10.1016/j.jorganchem.2006.03.016

Three new resins are synthesized by radical co-polymerization of triorganotin-4-vinylbenzoates (substituent at tin = Me, Bu or Ph) with styrene and 1,4-divinylbenzene. The products prepared have been characterized by FT-IR and NMR spectroscopy both in the solid state and as swollen samples, showing a predominantly tetracoordinated tin atom. The catalytic activity of the above resins in a transesterification reaction has been tested using ethyl acetate and primary, secondary or tertiary alcohol, showing good results with the former, but not with the latter ones. Focussing on the different triorganotin substituents, a comparison between the prepared resins and low molecular weight analogues evidences a lower activity of the resins, due to their inhomogeneous operating conditions. However, the triphenyltin functionalized resin shows a transesterification activity comparable to the corresponding model compound. The reaction mechanism and the effects of Lewis acidity of the different groups linked to tin have been investigated by ¹H and ¹¹⁹Sn hr-MAS NMR.

Full text of DOI:10.1016/j.jorganchem.2006.03.016

Journal of Organometallic Chemistry 691 (2006) 3043–3052
www.elsevier.com/locate/jorganchem
Cross-linked polystyrene resins containing
triorganotin-4-vinylbenzoates: Assessment of their catalytic activity
in transesterification reactions
a,
a
a
Luigi Angiolini *, Daniele Caretti , Laura Mazzocchetti ,
a
b
b
Elisabetta Salatelli , Rudolph Willem , Monique Biesemans
a
Dipartimento di Chimica Industriale e dei Materiali, University of Bologna, Viale Risorgimento 4, I-40136 Bologna, Italy
b
High Resolution NMR Centre (HNMR), Vrije Universiteit Brussel, Pleinlaan 2, B-1050 Brussel, Belgium
Received 23 February 2006; received in revised form 8 March 2006; accepted 8 March 2006
Available online 14 March 2006
Abstract
Three new resins are synthesized by radical co-polymerization of triorganotin-4-vinylbenzoates (substituent at tin = Me, Bu or Ph)
with styrene and 1,4-divinylbenzene. The products prepared have been characterized by FT-IR and NMR spectroscopy both in the solid
state and as swollen samples, showing a predominantly tetracoordinated tin atom. The catalytic activity of the above resins in a transe-
sterification reaction has been tested using ethyl acetate and primary, secondary or tertiary alcohol, showing good results with the for-
mer, but not with the latter ones. Focussing on the different triorganotin substituents, a comparison between the prepared resins and low
molecular weight analogues evidences a lower activity of the resins, due to their inhomogeneous operating conditions. However, the tri-
phenyltin functionalized resin shows a transesterification activity comparable to the corresponding model compound. The reaction mech-
1
anism and the effects of Lewis acidity of the different groups linked to tin have been investigated by H and 119Sn hr-MAS NMR.
ꢀ 2006 Elsevier B.V. All rights reserved.
Keywords: Catalyst; Sn NMR spectroscopy; Organotin; Resins; Transesterification
1. Introduction
has been dedicated to the separation of the catalyst from
the final products, the few reported processes being expen-
sive, due to the use of extraction solvents [11], or leading to
loss of catalytic activity because of the drastic conditions
needed for its recovery [12].
To address this issue, a useful approach consists of
grafting the tin derivative onto a polymeric cross-linked
support, since the catalyst obtained in this way at the
solid–liquid interface can be easily separated by filtration
of the support to which it is grafted, thus reducing, or even
completely preventing the presence of toxic organotin res-
idues in solution and giving rise to clean and potentially
recyclable organotin reagents.
Since their first biocidal application in the early 1920s [1],
organotin compounds are among the most widely used
organometallic chemicals [2]. Their applications cover a
wide range of different fields, being used as antifouling
paints [3–5], PVC stabilizers [6] and anti-tumour drugs [7]
as well as anion carriers in electrochemical membrane design
[8] and, last but not least, as homogeneous catalysts [9].
In spite of the well-known homogeneous catalytic activ-
ity of organotins in transesterifications, owed to the pres-
ence of the Lewis acid metal centre [10], their toxicity,
coupled to their difficult removal from the reaction mix-
ture, constitute a serious drawback. Only limited research
Several insoluble tri- and dialkyltin grafted polymers
have already been synthesized, starting from commercial
polystyrene resins cross-linked with 1,4-divinylbenzene
[13–18].
*
Corresponding author. Tel./fax: +39 051 2093687.
E-mail address: luigi.angiolini@unibo.it (L. Angiolini).
0022-328X/$ - see front matter ꢀ 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2006.03.016

3044
L. Angiolini et al. / Journal of Organometallic Chemistry 691 (2006) 3043–3052
All the synthetic procedures used in previous reports
The knowledge of the features of the related low molec-
ular weight model compounds, namely trimethyl-, tributyl-
and triphenyl-tin p-isopropylbenzoate (TMTIB, TBTIB
and TPTIB, respectively), described elsewhere [23], plays a
key role for the structural characterization of the cross-
linked polymeric derivatives, since their insolubility pre-
vents the application of common analytical techniques. In
this respect, the use of proton and tin high-resolution Magic
Angle Spinning (hr-MAS) NMR appears of determinant
relevance to the characterization of such cross-linked mate-
rials [15] (see Section 3). Moreover, the comparison with the
catalytic performances of low molecular weight models
allows a rough evaluation of the activity changes associated
with the use of an insoluble cross-linked macromolecular
structure, a feature which is believed to affect the accessibil-
ity to the catalytic sites.
[18] have the common disadvantage to start from an
insoluble reactant, which makes the reaction sluggish
and difficult to control. Even more hampering is that
the organotin functionalization degree cannot always be
selected a priori.
In the course of investigations on their potentiometric
anion response, tributyltin carboxylates were observed to
show interesting properties in transesterification catalysis
as well. Since this observation appears likewise related to
the Lewis acidity of the tin atom, the natural idea raises
that next to grafted dialkyltin dihalides [18], triorganotin
carboxylates grafted to the insoluble macromolecular sup-
port have the potency to offer an alternative, attractive
class of interfacial organotin catalysts.
Stannylating carboxylic acids in cross-linked resins
appeared previously to be not achievable quantitatively
[19]. Co-polymerization to cross-linked resins involving
triorganotin carboxylate monomers is believed to address
this issue and is presented in the present paper as an alter-
native strategy. More particularly, we have addressed the
synthesis of cross-linked resins, having a structure closely
related to that of Amberlite IRC-86 [19], starting, however,
from triorganotin p-vinylbenzoates rather than acrylates as
suitable monomers, since cross-linked polystyrenes proved
previously to be convenient insoluble supports giving rise
to satisfactory catalytic activities in other organotin sys-
tems [13–18], unlike Amberlite IRC-86, scoring less satis-
factorily. In fact, although it is known [20] that
organometallic monomers can inhibit the polymerization
process, this synthetic route is nevertheless explored since
it enables one to skip further functionalization steps,
potentially limited by unfavourable diffusion due to the
cross-linking, as observed with Amberlite IRC-86 deriva-
tives [19]. In addition, in order to minimise the effect of
tin on the chain growth, we designed a monomer having
the polymerizable vinyl group and the triorganotin func-
tionality conveniently separated. It is established that inhi-
bition by the organotin moiety strongly depends on the
relative position of the polymerizable C@C bond with
respect to the organotin moiety in the molecule, the inter-
calation of polar or polarizable groups such as phenyl or
carboxyl ones, leading to a considerably weaker inhibiting
action [21,22].
2. Results and discussion
2.1. Synthesis and characterization of monomers
As reported in Scheme 1, the first step of the synthesis of
p-vinylbenzoic acid (pVBA) implies the formation of a
Grignard derivative of p-chlorostyrene which is quenched
with CO2 in order to achieve the carboxylic moiety, after
treatment with aqueous H2SO4.
pVBA was characterized by FT-IR and 1H and 13C
NMR spectroscopy. The conversion of p-vinylbenzoic acid
into the desired triorganotin p-vinylbenzoate monomers
was then achieved by direct triorganostannylation with
the suitable triorganotin derivative (Scheme 2) in the pres-
ence of a small amount of hydroquinone as a polymeriza-
tion inhibitor.
Direct triorganostannylation of a carboxylic acid with a
bis(triorganotin) oxide is one of the most common methods
[24] to synthesize triorganotin carboxylates. As an alterna-
tive, the reaction of the carboxylic acid with a suitable
triorganotin hydroxide can likewise be used [25]. The
water, produced as co-product in both cases, can be conve-
niently removed from the reaction mixture by azeotropic
distillation in toluene, in order to conduct the reaction to
completion.
As represented in Scheme 2, we applied this synthetic
pathway to the production of both the tributyltin (TBTVB)
and triphenyltin (TPTVB) p-vinylbenzoates, starting from
pVBA and bis(tributyltin)oxide or triphenyltin hydroxide,
respectively.
Thus, we have submitted triorganotin derivatives of p-
vinylbenzoic acid (pVBA) to co-polymerization with sty-
rene and 1,4-divinylbenzene, the carboxylic function of
pVBA being functionalized with different R3Sn organotin
residues (R = Me, Bu, Ph). Since the catalytic activity is
likely to be affected by the Lewis acidity of the tin atom,
the organic substituents were selected in order to investi-
gate their influence on the catalytic performances.
Mg
CO2
H+
THF
The different tin containing monomeric compounds
have been respectively co-polymerized with 1,4-divinylben-
zene, as cross-linking bifunctional monomer, and styrene,
the latter aiming at limiting the content of stannylated
co-units in the final resin.
MgCl
Cl
COOH
pVBA
Scheme 1.

L. Angiolini et al. / Journal of Organometallic Chemistry 691 (2006) 3043–3052
3045
is well established for low molecular weight compounds
bearing the triphenyltin carboxylate moiety, and also the
tributyltin analogue, although two examples of pentacoor-
dinated tributyltin carboxylates were recently ascertained
by the X-ray crystal structure of tributyltin p-ethyl and p-
isopropyl phenylacetate [27].
The tin pentacoordination in TMTVB is evidenced by
its FT-IR spectrum, displaying asymmetric and symmetric
stretching bands at 1587 and 1378 cmꢀ1, respectively, in
agreement with a trigonal bipyramidal structure with a
diaxial configuration of the carboxylic oxygen of one
ligand and the carbonyl oxygen of a neighbouring triorg-
anotin carboxylate molecule [28].
The FT-IR absorptions of the monomers (Table 1) are
in good agreement with those of the related p-isopropyl
tin benzoate models [23], reflecting the same behaviour as
regards the coordination at tin. The presence of the absorp-
tion at 1629 cmꢀ1 related to the carbonyl stretching of
TPTIB (Fig. 1), allows also to assign the band at
1620 cmꢀ1 of TPTVB to the same vibration, as the exten-
sion of conjugation of a carbonyl moiety with the aromatic
system lowers the frequency of the stretching absorption
[29]. As a consequence, the absorption at 1635 cmꢀ1 in
TPTVB is assigned to the vinyl stretching vibration. It
can also be noticed that the absorption frequency of the
Me3SnOH
Acetone
TMTVB
TBTVB
TPTVB
COOSnMe3
COOSnBu3
COOSnPh3
Bu3SnOSnBu3
Toluene
COOH
pVBA
Ph3SnOH
Toluene
Scheme 2.
The synthesis of the trimethyltin p-vinylbenzoate
(TMTVB) was also achieved by direct trimethylstannyla-
tion of pVBA with the corresponding trimethyltin hydrox-
ide, using in this case acetone as solvent. Since trimethyltin
hydroxide sublimates at 80 ꢁC, the solvent was chosen so as
to enable a sufficiently low refluxing temperature in order
to prevent sublimation of the reactant. The water obtained
Table 1
Carbonyl stretching frequencies of triorganotin p-vinyl and p-isopropyl
benzoates
˚
as a by-product was removed by using 5 A molecular sieves
Tin substituent
m(C@O) (cmꢀ1
)
added to the reaction mixture. In this case the purification
from the organometallic reagent in excess was simply
achieved by gentle heating under reduced pressure in order
to sublimate any unreacted trimethyltin hydroxide.
p-Vinyl benzoate
p-Isopropyl benzoatea
Me
Bu
Ph
1587
1630
1620
1595
1636
1629
Whereas the final purification of TMTVB and TPTVB
was easily achieved by crystallization of the product from
ethanol/water, this was not possible with the tributyltin
derivative, the final crude product appearing as a viscous
oil. Purification of TBTVB from the corresponding unre-
acted tin oxide was thus achieved by treatment of the reac-
tion mixture with a weak ionic exchange resin (Fluka
Amberlite IRC-86) consisting of poly(acrylic acid) cross-
linked with 1,4-divinylbenzene, which is able to react with
the organometallic reagent in excess, as previously reported
[25], thus leading to pure TBTVB after filtration of the
resin from the reaction mixture and solvent evaporation.
All the reactions were monitored by FT-IR spectroscopy
and kept under reflux until no more absorption bands
related to the carboxylic acid carbonyl stretching at
1678 cmꢀ1 could be detected.
a
From Ref. [23].
The asymmetric stretching vibration of the triorganotin
carboxylate displays an absorption band in the range
1620–1635 cmꢀ1 for both TBTVB and TPTVB, while the
symmetric one lies around 1330 cmꢀ1. As previously stated,
these bands refer to a carbonyl moiety free from any inter-
action with the metal atom [26], indicating the presence of
tetracoordinated tin atom in the solid state. This behaviour
Fig. 1. FT-IR spectra of model compound TPTIB (a) and of triphenyltin-
p-vinyl benzoate monomer TPTVB (b).

3046
L. Angiolini et al. / Journal of Organometallic Chemistry 691 (2006) 3043–3052
carbonyl moiety in the triphenyltin derivatives is lower
than in the analogous non-coordinated tributyltin ones
both in the monomers and the models, as reported in Table
1.
tin and high oxygen contents (deviations from expected
up to 50%) were found. Consequently, this synthetic proce-
dure was abandoned.
An alternative route to cross-linked derivatives was then
applied by performing the radical polymerization of the
monomers in dry THF, using AIBN as a thermal radical
initiator, and keeping the reaction mixture sealed under
vacuum for 5 days at 60 ꢁC (Scheme 3).
The monomeric triorganotin p-vinylbenzoates were also
characterized by 1H and 13C NMR spectroscopy, using
two-dimensional COSY and HSQC experiments where
necessary, in particular for the proton resonances of
TPTVB, due to the presence of two overlapping aromatic
spin systems. All the investigated triorganotin carboxylate
monomers clearly display the expected multiplet pattern
of vinyl protons analogously to the pVBA precursor, con-
firming that the vinyl group remained unaltered after the
triorganostannylation of the starting carboxylic acid. The
119Sn NMR spectra of TMTVB, TBTVB and TPTVB in
non-coordinating CDCl3 display a single resonance with
chemical shift values typical for tetracoordinated tin [30]
(Table 2). This allows to conclude that the monomeric
triorganotin p-vinylbenzoates are exempt from any unre-
acted organotin reagent after the triorganostannylation
reaction.
The obtained cross-linked triorganotin ter-polymers,
labelled as S-ter-X (Table 3), where S refers to solution
and X to the substituent at tin, were recovered by simple
TMTVB
R
S-ter-Me
COOSnMe3
COOSnMe3
TBTVB
2.2. Synthesis and characterization of cross-linked polymers
R
COOSnBu3
S-ter-Bu
S-ter-Ph
AIBN
THF
A first unsuccessful attempt to polymerize the synthe-
sized monomers with 1,4-divinylbenzene and styrene was
performed by radical polymerization in emulsion. This pro-
cedure allows to operate at short reaction times, in addi-
tion, at least in principle, to a better control over the
final resin particle size, therefore leading to a product suf-
ficiently easy to handle. The polymerization was carried
out in aqueous medium, in the presence of dodecylsulphate
sodium salt as surfactant, and sodium persulphate as the
radical initiator. The reaction mixture was allowed to react
for 24 h at 80 ꢁC under vigorous stirring and the obtained
polymers were recovered from the emulsion by filtration,
after coagulation with a concentrated aqueous Al2(SO4)3
solution.
+
COOSnBu3
TPTVB
R
COOSnPh3
COOSnPh3
a
b
R
c
=
COOSnR3
The occurrence of polymerization involving the vinyl
functionality was proved by infrared spectra, showing the
disappearance of the typical CH2 bending band of the vinyl
group around 1400 cmꢀ1. However, the presence of an
absorption band around 1690 cmꢀ1, a value typical for
the carbonyl vibration of carboxylic acids for the tributyl
and trimethyl derivatives, suggested the occurrence of a
partial hydrolysis of triorganotin carboxylate during the
polymerization. The thermogravimetric analysis, coupled
to elemental analysis, confirmed indeed the occurrence of
hydrolysis for all the ter-polymers, since abnormally low
a
COOSnR3
Scheme 3.
Table 3
Polymerization data for cross-linked terpolymers S-ter-X
Polymer Styrene
Divinylbenzene Organotin Yielda Functiona-
(% mole) (% mole)
monomer (%)
(% mole)
lization
degreeb
(% mole)
S-ter-Me 85
S-ter-Bu 85
S-ter-Ph 85
a
10
10
10
5
5
5
52
62
54
5.7
7.1
7.7
Table 2
119Sn NMR chemical shifts of triorganotin monomers in CDCl3
Compound
d
119Sn (ppm)
Determined as (g of polymer/g of feeding monomers) · 100.
b
Determined by a statistical processing method of elemental data
TMTVB
TBTVB
TPTVB
137
112
ꢀ112
analysis of C, H, Sn, and O, as described elsewhere [18]. The functional-
ization degree is defined as the percent molar content of triorganotin
carboxylate monomeric units present in the final polymeric products.

L. Angiolini et al. / Journal of Organometallic Chemistry 691 (2006) 3043–3052
3047
filtration and purified from unreacted monomers and solu-
ble oligomers by swelling the mixture in different solvents
followed by filtration. As reported in Table 3, the content
of triorganotin carboxylate units in the products results
higher than expected on the basis of the corresponding
monomer feeds. This would indicate a significantly higher
reactivity of the tin functionalized monomer with respect
to the other co-monomers, at least for the tributyl- and
the triphenyltin derivatives, in contrast with results
obtained for similar derivatives where the organotin
functionality was connected to the styrenic residue through
a dimethylene spacer [31]. The rather low polymerization
yields obtained, however, suggest that an abundant frac-
tion of soluble product, probably constituted by styrene
moieties, was lost in the manipulation of the crude mate-
rial, thus increasing the relative content of organotin co-
units. Indeed, DSC investigations on the resins obtained
allowed to exclude the presence of polystyrene homopoly-
mer in the purified material, no glass transition being
observed in the thermograms.
line with the FT-IR data. By contrast, S-ter-Me displays
only a major 117Sn resonance at 121 ppm, typical for tetra-
coordinated tin, and a second minor resonance at
ꢀ57 ppm. The latter is notably different from that observed
for the corresponding trimethyltin p-isopropylbenzoate
model in the solid state, showing a chemical shift of
+26 ppm [23], attributed to pentacoordinated tin origi-
nated by intermolecular interaction with the neighbouring
carbonyl moieties [30]. The higher level of shielding indi-
cated by the value of ꢀ57 ppm for the minor tin species
suggests an even stronger pentacoordination than in the
model, the exact structural nature of which remains
unclear. Integration of these signals supplies a rough eval-
uation of the relative amounts of the two differently coor-
dinated tin species, the pentacoordinated form being
estimated as 8% of the total, a content which reasonably
explains why the related carbonyl stretching could not be
detected by FT-IR, and which is too low to enable further
investigations regarding the nature of this type of pentaco-
ordination at tin.
In contrast with the derivatives prepared by emulsion
polymerization, the FT-IR spectra of S-ter-X did not exhi-
bit any absorption at 1690 cmꢀ1 related to free carboxylic
acid, indicating the absence of triorganotin carboxylate
hydrolysis when the procedure in organic solution is
adopted.
The carbonyl absorption frequency of S-ter-Me moved
with respect to monomer TMTVB from 1587 cmꢀ1 (Table
1) to 1639 cmꢀ1, a value typical for non-coordinated triorg-
anotin carboxylate [27]. It appears therefore that the pres-
ence of a stiff cross-linked polymeric structure prevents the
interactions between organotin carboxylate groups, which
are anyhow highly unlikely due to the low content of tin
containing units in the material.
To confirm the result of IR spectra, tin NMR experi-
ments were also performed on the cross-linked derivatives,
in the solid state, by CP-MAS (Cross Polarization-Magic
Angle Spinning), as well as in heterogeneous medium at
the solid–liquid interface, with the derivatives swollen in
CDCl3, using tin hr-MAS NMR spectroscopy (Table 4)
[15]. The 119Sn hr-MAS spectra show, in agreement with
the behaviour of the model compounds in solution, only
resonances typical for tetracoordinated tin [30] (Table 4).
Similarly, for both the tributyl and the triphenyltin deriva-
tives, the 117Sn CP-MAS NMR spectra in the solid state
reveal the presence of fully tetracoordinated tin atom, in
In order to assess the thermal stability of these materials
in view of possible applications, the thermogravimetric
analysis (TGA) was performed both on the polymeric
and model compounds, with the aim of evaluating the
influence of the macromolecular cross-linked chains on
the thermal behaviour. As reported in Fig. 2 and Table 5,
all the cross-linked materials display a notably higher
initial degradation temperature with respect to the mono-
meric models, a contribution to the increase of Td values
deriving also from the low concentration of tin containing
units in the resins when compared to the triorganotin
p-isopropylbenzoate model compounds. As previously
established [32], the thermal stability of polystyrenic
macromolecules increases upon lowering the triorganotin
co-unit content, as the triorganotin carboxylate group sub-
stantially favours the decomposition process.
2.3. Assessment of catalytic activity in transesterification
reactions
In order to investigate the catalytic performance, the
activity of S-ter-X resins has been assessed in a transesteri-
fication reaction of ethyl acetate with differently hindered
alcohols, similarly to the previously described Amberlite
derivatives [19]. A comparison with the catalytic perfor-
mances of the related low molecular weight models [23]
under the same conditions has also been carried out. The
amount of catalyst in each experimental run was calculated
so as to have a 1% mole ratio of tin containing units with
respect to the selected alcohol, 1-octanol, cyclohexanol or
3-ethyl-3-pentanol. All the reactions, when carried out in
the absence of catalyst gave, as expected, no conversion
at all of ethyl acetate.
Table 4
119Sn NMR chemical shifts (ppm) of S-ter-X cross-linked polymers
swollen in CDCl3 and in the solid state
Polymer
d (Sn) (ppm)
119Sn hr-MAS
117Sn solid state
121(92); ꢀ57(8)a
The alcohol conversion in each catalytic run was
assessed by gas chromatography (Table 6), determining
the relative amounts of transesterified product and unre-
acted starting alcohol after 24 and 48 h.
S-ter-Me
S-ter-Bu
S-ter-Ph
a
132
101
ꢀ121
97
ꢀ114
Values in brackets refer to the percentage of each integrated resonance.

3048
L. Angiolini et al. / Journal of Organometallic Chemistry 691 (2006) 3043–3052
Fig. 2. TGA curves of S-ter-Me (a), S-ter-Bu (b), S-ter-Ph (c) and the related models TMTIB (d), TBTIB (e) and TPTIB (f).
Table 5
extent the accessibility to the tin atom, actually does not
Initial decomposition temperature of the S-ter-X resins and the related low
affect the transesterification rate. By contrast, the trialkyl-
tin substituted resins S-ter-Me and S-ter-Bu display a
remarkable decrease of the transesterification conversion
with respect to corresponding model compounds, the
obtained values being in good agreement with those
obtained with the Amberlite IRC86 organotin derivatives
[19]. It is not yet clear why this effect does not have relevant
consequences for the phenyl derivative, S-ter-Ph having
quite hindering sustituents at tin. Presumably, this indi-
cates that a balance between Lewis acidity, as outlined pre-
viously for the model compounds [23], and steric hindrance
of the cross-linked backbone is operative.
In order to obtain a better insight in factors governing
the catalysis, 119Sn hr-MAS spectra of the catalysts were
recorded in the actual reaction mixture. The 119Sn chem-
ical shifts of the cross-linked polymers in the presence of
transesterification reagents, either ethyl acetate or a mix-
ture of ethyl acetate and n-octanol, summarized in Table 7,
reveal a low frequency shift with respect to the chemical
molecular weight compounds
a
Polymer
Td (ꢁC)
Model
Td (ꢁC)
S-ter-Me
S-ter-Bu
S-ter-Ph
a
361
350
370
TMTIB
TBTIB
TPTIB
150
169
183
Temperature of the first weight loss.
The catalytic performances offered by the cross-linked
derivatives are significantly more promising than the previ-
ously described organotin modified Amberlite IRC-86 [19].
In the case of S-ter-Ph in particular, the conversion of pri-
mary alcohol compares remarkably well with the corre-
sponding model compound TPTIB [23], the grafted
catalyst at the liquid–solid interface displaying a catalytic
activity very close to that of the model counterpart in
homogeneous solution. The presence of the macromolecu-
lar cross-linked backbone, expected to hamper to some
Table 6
Conversions obtained in catalysed transesterification reaction of ethyl acetate with differently hindered alcohols, as determined by gas chromatography
Catalyst
Alcohol conversion (%)
1-Octanol
24 h
Cyclohexanol
24 h
3-Ethyl-3-pentanol
24 h
48 h
48 h
48 h
TMTIB
TBTIB
TPTIB
S-ter-Me
S-ter-Bu
S-ter-Ph
19
76
95
10
8
43
95
100
18
28
96
5
7
10
3
2.5
3
6
9
17
3
7
4.5
–
–
–
–
–
–
Trace
Trace
Trace
Trace
Trace
Trace
90

L. Angiolini et al. / Journal of Organometallic Chemistry 691 (2006) 3043–3052
3049
Table 7
hr-MAS 119Sn chemical shifts (ppm) of the model compounds and the corresponding cross-linked polymers in various solventsa
Compound
CDCl3
EtOAc
Dd (EtOAc vs. CDCl3)
EtOAc/octanolb
Dd (EtOAc/oct. vs. EtOAc)
TMTIB
TBTIB
TPTIB
S-ter-Me
S-ter-Bu
S-ter-Ph
a
133c
109c
–115c
132
101
–121
95
88
–127
90
84
–128
–38
–21
–12
–42
–17
–7
78
82
–130
78
81
–131
–17
–6
–3
–12
–3
–3
Pure octanol does not swell the polymers and consequently no 119Sn spectrum is obtained in this solvent.
EtOAc/octanol is 4/1 for the model compounds and 7/1 for the polymers.
From Ref. [23].
b
c
shifts in CDCl3, similar to the effects observed for the
fore confirming the data obtained both with the low molec-
ular weight model compounds and the Amberlite IRC86
derivatives [19].
model compounds. Although no spectra could be
obtained in pure n-octanol as a result of insufficient swell-
ing, the more pronounced low frequency shift observed
with a mixture octanol/ethyl acetate than with the ester
alone, implies that both reagents coordinate the tin atom.
As for the model compounds, the explanation for the
transesterification proceeding better with the triphenyltin
polymer, where coordination of both the ester and alcohol
is obviously weaker, is that although complex formation
is necessary, the complex has to be sufficiently labile in
order to achieve fast exchange between incoming reac-
tants and outgoing products. Compared to the model
compounds, a difference in behaviour is noticed for the
tributyltin compound. Whereas the model compound
TBTIB scores rather well, only little conversion is
observed for the cross-linked polymer S-ter-Bu (Table 6).
Careful inspection of the 119Sn hr-MAS chemical shift
data reveals that the relative increase in 119Sn chemical
shift difference (Dd) in chloroform and ethyl acetate
between S-ter-Bu and S-ter-Ph (ꢀ17 compared to
ꢀ7 ppm, i.e., 24 times) is somewhat stronger than between
the corresponding model compounds (ꢀ21 compared to
ꢀ12 ppm, i.e., 1.8 times), suggesting that ethyl acetate
coordinates somewhat stronger the tin atom in the poly-
mer than in the corresponding model compound. The
addition of n-octanol results in the same relative increase
(ꢀ3 ppm) for both polymers, while a significantly more
pronounced effect (2 times) is noticed for the related
model compound TBTIB, suggesting n-octanol coordi-
nates somewhat less the tin atom in S-ter-Bu than in
the model compound (Table 7). These subtle changes in
coordination, may lay at the basis of the change in catal-
ysis kinetics, but no further evidence can be provided to
this.
3. Experimental
3.1. Materials
Chemicals were supplied by Sigma–Aldrich and gener-
ally used as received. Solvents were purified using standard
techniques [33] and stored under nitrogen. Commercial sty-
rene was treated with 5% aq. NaOH solution, washed with
water, dried for 24 h on anhydrous MgSO4 and distilled at
16 mmHg (b.p. = 35 ꢁC). The obtained monomer was
stored under nitrogen at 0 ꢁC. Commercial tetrahydrofuran
(THF) was kept in contact with KOH for 8 h, refluxed with
metallic K and Na in the presence of benzophenone and
then distilled (b.p. = 65 ꢁC) under nitrogen atmosphere;
the dried solvent was freshly distilled every time, immedi-
ately prior to synthetic use. a,a0-azobisisobutyronitrile
(AIBN) was purified by crystallization from absolute etha-
nol before use. Commercial 1,4-divinylbenzene was treated
with 5% aq. NaOH solution, washed with water, dried for
24 h on anhydrous MgSO4 and distilled at 4 mmHg
(b.p. = 40 ꢁC). The monomer obtained was stored under
nitrogen at 0 ꢁC.
3.2. Measurements
Infrared spectra in KBr pellets or in chloroform solu-
tion were recorded on a Perkin–Elmer 1750 FT-IR spec-
1
trometer. H and 13C NMR spectra were recorded both
on a Varian Gemini 300 NMR and Varian Mercury 400
NMR using CDCl3 solutions and tetramethylsilane as
internal standard. Notations in NMR spectra: Ci = ipso
carbon of the functionalized phenyl group, bound to the
variable triorganotin carboxylate; Co = ortho; Cm = meta;
Cp = para. Solid state 117Sn NMR spectra were recorded
at 89.15 MHz using a Bruker Avance 250 NMR spec-
trometer, equipped with a 7 or 4 mm MAS broad-band
probe using tetracyclohexyltin (ꢀ97.4 ppm relative to
(CH3)4Sn), as an external standard. The 1H and 119Sn
hr-MAS NMR spectra of heterogeneous samples, consist-
ing of cross-linked resins to which organotin compounds
When the secondary alcohol cyclohexanol is used in the
transesterification reaction, a low final conversion is found
even with S-ter-Ph, which now scores similarly to the trial-
kyltin substituted resins. This suggests that the steric hin-
drance of incoming alcohol becomes the determinant
factor, since also in homogeneous solution the conversions
are very low, even after 48 h, being at most 17% with
TPTIB. The transesterification of ethyl acetate with 3-
ethyl-3-pentanol practically does not proceed at all, there-

3050
L. Angiolini et al. / Journal of Organometallic Chemistry 691 (2006) 3043–3052
are covalently grafted and swollen up with an appropriate
solvent [15], were recorded on a Bruker AMX500 NMR
spectrometer equipped with three channels and gradient
solid, further purified by crystallization in ethanol/water
mixture. Yield: 0.39 g (46%).
Elemental analysis calcd (%) for C12H16O2Sn (310.95): C
46.35, H 5.19, O 10.29; found: C 46.21, H 5.17, O 10.32.
1H NMR (400 MHz, CDCl3, 20 ꢁC, TMS): d = 0.7 (s,
1
pulse hardware, using a Bruker triple H–13C–119Sn TXI
probe especially dedicated for NMR measurements at
the liquid–solid interface of organotin moieties grafted
to a macromolecular support. Thermogravimetric analy-
ses were carried out on a TGA-7 Perkin–Elmer thermo-
balance in air. Oxygen elemental analyses were
performed by REDOX s.n.c. (Milano). Carbon, Hydro-
gen and Tin elemental analyses were performed by Centre
National de la Recherche Scientifique (CNRS), Service
Central d’Analyse, Vernaison (France). Conversion
assessments in catalytic tests were made by gas chroma-
tography using an Agilent 6890 instrument equipped with
a capillary methylsilicone column and an Agilent 5973 M
mass detector.
3
9H, CH3), [2J(H–119/117Sn) = 58/56 Hz]; 5.35 (dd, J(Hcis,
H) = 11 Hz, 3J(Hcis,Htrans) = 0.8 Hz, 1H; @CHtransHcis),
5.85 (dd, 3J(Htrans,H) = 18 Hz, 3J(Htrans,Hcis) = 0.8 Hz,
1H; @CHtransHcis), 6.75 (dd, 3J(H,Hcis) = 11 Hz,
3
3J(H,Htrans) = 18 Hz 1H; @CH), 7.45 (d, J(H,H) = 8 Hz,
3
2H; CHm), 8.0 (d, J(H,H) = 8 Hz, 2H; CHo) ppm.
13CNMR (400 MHz, CDCl3, 20 ꢁC, TMS): d = ꢀ2.1
(CH3, 1J(13C–119/117Sn) = 397/380 Hz), 116.3 (@CH2),
126.2 (Cm), 126.4 (Ci), 130.7 (Co), 136.4 (@CH), 141.6
(Cp), 171.9 (COO) ppm.
IR (KBr): 3077–3003 (mCH arom.), 2915 (mCH aliph.),
1587 (mas COOSn), 1606 (mC–C ring), 1399 (d@CH2), 991, 916
(dCH@CH2), 863 (cCH1,4-subst. ring) cmꢀ1
.
3.3. Synthesis of monomers
3.3.3. Tributyltin p-vinylbenzoate (TBTVB)
3.3.1. p-Vinylbenzoic acid (pVBA)
In a 100 mL flask equipped with a Dean-Stark appara-
tus pVBA (0.40 g, 2.70 mmol) and hydroquinone (0.12 g)
are dissolved in toluene (20 mL) and bis(tributyltin) oxide
(BTBO) (0.7 mL, 1.35 mmol) is added. The mixture is
heated until no more water evolution is observed, the reac-
tion progress being monitored likewise by FT-IR spectros-
copy. The crude product appearing as a viscous liquid, its
purification from the unreacted BTBTO was achieved
through reaction in heterogeneous phase with dried
Amberlite IRC-86 for 24 h. The resin was filtered off and
washed with chloroform (3 · 20 mL), the combined solu-
tions were evaporated under reduced pressure to give a yel-
lowish oil which could not be crystallized. Yield: 0.63 g
(53%).
In a nitrogen flushed 250 mL three-necked, round-bot-
tomed flask, equipped with a magnetic stirrer and a drop-
ping funnel, pure magnesium (1.78 g, 71.3 mmol) is added
to freshly distilled THF (20 mL); subsequently, a solution
of p-chlorostyrene (5.0 mL, 41.7 mmol) in THF (10 mL)
is slowly dropped in. The mixture is heated at 35 ꢁC and
allowed to react for 2 h. Carbon dioxide is then bubbled
in the solution for 3 h, aqueous H2SO4 2 M (20 mL) is
added, and the solution extracted with diethyl ether
(2 · 20 mL). The organic phase is dried with Na2SO4.
The organic solvent is distilled off and the product is puri-
fied by crystallization with petroleum ether. Yield: 3.81 g
(62%).
1H NMR (300 MHz, CDCl3, 20 ꢁC, TMS): d = 5.4 (d,
3J(Hcis,H)= 11 Hz, 1H; @CHtransHcis), 5.9 (d, 3J(Htrans,H)=
18 Hz, 1H; @CHtransHcis), 6.8 (dd, 3J(H,Hcis) = 11 Hz,
Elemental analysis calcd (%) for C21H34O2Sn (437.19):
C 57.69, H 7.84, O 7.32; found: C 57.81, H 7.81, O 7.34.
1H NMR (400 MHz, CDCl3, 20 ꢁC, TMS): d = 0.9 (t,
3J(H,H) = 8 Hz, 9H; CH3Bu), 1.4–1.2 (m, 12H, CH2a-
Bu + CH2cBu), 1.6 (m, 3J(1H–119/117Sn) = 65 Hz, 6H;
3J(H,Htrans) = 18 Hz, 1H; @CH), 7.5 (d, J(H,H) = 8 Hz,
3
3
2H; CHm), 8.1 (d, J(H,H) = 8 Hz, 2H; CHo) ppm.
13C NMR (300 MHz, CDCl3, 20 ꢁC, TMS): d = 117.7
(@CH2), 126.7 (Cm), 129.0 (Ci), 131.3 (Co), 136.4 (@CH),
143.5 (Cp), 172.4 (COOH) ppm.
CH2bBu,), 5.35 (d, J(Hcis,H) = 11 Hz, 1H; @CHtransHcis),
3
3
5.85 (d, J(Htrans,H) = 17 Hz, 1H; @CHtransHcis), 6.75 (dd,
3
3J(H,Hcis) = 11 Hz, J(H,Htrans) = 17 Hz, 1H; @CH), 7.45
IR (KBr): 3006–2960 (mCH arom.), 1682 (mCO acid),
1627 (mC@C vinyl), 1408 (d@CH2), 990, 904 (dCH@CH2), 827
(d, 3J(H,H) = 8 Hz, 2H; CHm), 8.0 (d, 3J(H,H) = 8 Hz,
2H; CHo) ppm.
(c1,4-subst. ring arom.) cmꢀ1
.
13C NMR (400 MHz, CDCl3, 20 ꢁC, TMS): d = 14.0
(CH3Bu), 16.9 (CH2aBu
,
1J(13C–119/117Sn) = 359/342 Hz),
3.3.2. Trimethyltin p-vinylbenzoate (TMTVB)
27.4 (CH2cBu, ,
3J(13C–119/117Sn) = 64 Hz), 28.2 (CH2bBu
In a 100 mL flask pVBA (0.40 g, 2.70 mmol), trimethyl-
tin hydroxide (0.50 g, 2.76 mmol) and hydroquinone
(0.12 g) are dissolved in dry acetone (70 mL). Molecular
sieves 5A are added to remove the water formed as by-
product. The mixture is heated to reflux, the reaction pro-
gress being monitored likewise by FT-IR spectroscopy. The
cloudy reaction mixture was finally filtrated over a Celite
path to remove traces of molecular sieves, then the solvent
was evaporated under reduced pressure to leave an oily res-
idue. This was pumped at high vacuum, leading to a white
2J(13C–119/117Sn) = 20 Hz), 116.0 (@CH2), 126.2 (Cm),
130.8 (Co), 131.7 (Ci), 136.7 (@CH), 141.4 (Cp), 171.6
(COO) ppm.
IR (neat): 3085–3037 (mCH arom.), 2956–2854 (mCH
aliph.), 1630 (mas COOSn), 1606 (mC–C ring), 1401 (d@CH2),
988, 912 (dCH@CH2), 863 (cCH1,4-subst. ring) cmꢀ1
.
3.3.4. Triphenyltin p-vinylbenzoate (TPTVB)
In a 250 mL flask equipped with a Dean-Stark appara-
tus pVBA (1.80 g, 11.2 mmol), hydroquinone (0.12 g) and

L. Angiolini et al. / Journal of Organometallic Chemistry 691 (2006) 3043–3052
3051
triphenyltin hydroxide (4.5 g, 12.2 mmol) are dissolved in
toluene (150 mL). The mixture is heated until no more
water evolution is observed, the reaction progress being
monitored likewise by FT-IR spectroscopy. The solvent
was evaporated under reduced pressure and the solid resi-
due purified by crystallization from ethanol/water. Yield:
3.64 g (64%).
cross-linked polymeric benzoates and allows a better con-
trol of the final composition in triorganotin carboxylate
moieties. Another advantage is that the synthesized resins
do not require any further functionalization, thus avoiding
the potentially unfavourable need of heterogeneous reac-
tion on cross-linked substrates.
The assessment of their catalytic properties in the reac-
tion of ethyl acetate with differently hindered alcohols indi-
cates an influence of the Lewis acidity of the metal,
confirming the trend outlined previously for the model
compounds [23]. The triphenyltin derivatives having lower
Lewis acidity result the most active, close to the related sol-
uble model compound, the methyltin catalyst with the
highest Lewis acidity [23], displaying a lower activity. This
confirms that although the Lewis acidity of the metal centre
is a prerequisite for catalytic activity, the interactions
between the tin atom and the incoming transesterification
reagents must not be too strong, because this would disfa-
vour a fast addition-release mechanism of the reagents at
the tin atom. The weak activity of the trimethyltin deriva-
tives can be related to higher affinity of the tin atom for the
carbonyl oxygen, which leads to a more stable tin complex,
thus slowing down the exchange kinetic associated with the
catalytic process.
Elemental analysis calcd (%) for C27H22O2Sn (497.16):
C 65.23, H 4.46, O 6.44; found: C 65.11, H 4.47, O 6.46.
1H NMR (400 MHz, CDCl3, 20 ꢁC, TMS): d = 5.3 (d,
3
3J(Hcis,H) = 11 Hz, 1H; @CHtransHcis), 5.8 (d, J(Htrans,H)
3
= 17 Hz, 1H; @CHtransHcis), 6.7 (dd, J(H,Hcis) = 11 Hz,
3J(H,Htrans) = 17 Hz, 1H; @CH), 7.4 (m, 11H; CHm +
CHm-SnPh + CHp-SnPh), 7.9 (bs, 8H, CHo + CHo-SnPh) ppm.
13C NMR (400 MHz, CDCl3, 20 ꢁC, TMS): d = 115.0
(@CH2), 125.2 (Cm), 127.8 (Cm-SnPh
68 Hz), 128.6 (Cp-SnPh), 129.6 (Co), 132.9 (Ci), 135.7
(@CH), 136.1 (Co-SnPh
2J(13C–119/117Sn) = 46 Hz), 139.7
,
3J(13C–119/117Sn) =
,
(Cp), 132.1 (Ci-SnPh), 169.6 (COO) ppm.
IR (KBr): 3065–2997 (mCH arom), 1635 (mC@C vinyl),
1620 (mas COOSn), 1606 (mC–C ring), 1401 (d@CH2), 1331 (ma
COOSn), 861 (cCH1,4-subst. ring) cmꢀ1
.
3.4. Polymeric compounds
All polymeric derivatives listed in Table 3 were prepared
following the same procedure reported here for S-ter-Me.
To TMTVB (0.32 g, 1.03 mmol), 1,4-divinylbenzene
(0.3 mL, 2.09 mmol) and styrene (2.0 mL, 17.4 mmol) dis-
solved in anhydrous THF (20 mL), AIBN (33.8 mg, 1%
w/w) is added in a vial kept under nitrogen atmosphere
and submitted to several freeze – thaw cycles in order to
remove any oxygen trace. The vials were left at 60 ꢁC for
5 days and the polymeric products purified by stirring in
diethyl ether and subsequently in pentane, followed by dry-
ing under vacuum till constant weight. Data related to
polymers synthesis are summarized in Table 3.
Acknowledgements
This work was financially supported by the University
of Bologna (Progetti Pluriennali di Ricerca and Ricerca
Finalizzata Orientata) and the Polo Scientifico – Didattico
di Rimini (Catalizzatori polimerici a basso impatto ambi-
entale per reazioni di transesterificazione). M.B. and
R.W. are indebted to the Fund for Scientific Research –
Flanders (Belgium) (FWO) (Grant G.0016.02) and to the
Research Council of the VUB (Grants GOA31, OZR362,
OZR875) for financial support.
References
3.5. Elemental analysis
[1] J.A. Thompson, M.G. Sheffer, R.C. Pierce, Y.K. Chau, J.J. Cooney,
W.P. Cullen, R.J. Maguire, Organotin Compounds in the Aquatic
Environment: Scientific Criteria for Assessing their Effects on
Environmental Quality, National Research Council Canada, Ottawa,
1985, p. 284.
[2] A.G. Davies, Organotin Chemistry, second ed., VCH, Weinheim,
2004.
[3] D. Seyferth, T.C. Masterman, Appl. Organomet. Chem. 8 (1994)
335–349, and references therein.
[4] A.A. Mahmoud, A.F. Shaaban, M.M. Azab, N.N. Messia, Eur.
Polym. J. 28 (1992) 555–559.
[5] J.R. Dharia, C.P. Pathak, G.N. Babu, S.K. Gupta, J. Polym. Sci. 26
(1988) 595–599.
S-ter-Me: found: C 85.04, H 7.34, Sn 5.01, O 1.96; calcu-
lated according to the functionalization degree (Table 3) as
described in Ref. [18]: C 85.20, H 7.39, Sn 5.87, O 1.58.
S-ter-Bu: found: C 83.69; H 7.80; Sn 5.82; O 1.65; calcu-
lated according to the functionalization degree (Table 3) as
described in Ref. [18]: C 83.85; H 7.77; Sn 6.60; O 1.78.
S-ter-Ph: found: C 84.46; H 6.91; Sn 6.19; O 2.18; calcu-
lated according to the functionalization degree (Table 3) as
described in Ref. [18]: C 84.60; H 6.81; Sn 6.77; O 1.82.
4. Conclusions
[6] R. Ga¨chter, H. Muller, in: R. Ga¨chter, H. Muller (Eds.), Plastics
¨
¨
Additives Handbook, Hanser Publisher, Munich, 1990, pp. 281–287.
[7] M. Gielen, Chim. Nouv. 76 (2001) 3333–3346.
[8] J.K. Tsagatakis, N.A. Chaniotakis, K. Jurkschat, S. Damoun, P.
Geerlings, A. Bouhdid, M. Gielen, I. Verbruggen, M. Biesemans,
J.C. Martins, R. Willem, Helv. Chim. Acta 82 (1999) 531–542.
[9] J. Otera, Chem. Rev. 93 (1991) 1449–1470.
New catalytic systems derived from convenient p-vinyl
triorganotin benzoate monomers, cross-linked by co-poly-
merization with 1,4-divinylbenzene and styrene, were
prepared. This synthetic pathway enables one to obtain
non-coordinated trimethyltin, tributyltin and triphenyltin
[10] O.A. Mascaretti, R.L.E. Furlan, Aldrichim. Acta 30 (1997) 55–68.

3052
L. Angiolini et al. / Journal of Organometallic Chemistry 691 (2006) 3043–3052
[11] T. Yokoyama, Y. Kawaragi, K. Murai, Mitsubishi Chemical Co., EP
0 881 209 A1, May 26, 1998.
[21] N.E. Ikladious, N.N. Messiha, A.F. Shaaban, Eur. Polym. J. 20
(1984) 625–628.
[12] L.E. Trapasso, P.L. Meisel, L.B. Meisel, B. Lee, W.K. Chwang, CPS
Chemical Company Inc., U.S. Patent 5,606,103, February 25, 1997.
[13] B. Delmond, G. Dumartin, in: M. Gielen, R. Willem, B. Wrackmeyer
(Eds.), Physical Organometallic Chemistry. Solid State Organome-
tallic Chemistry. Methods and Applications, Wiley, Chichester, 1999,
p. 445.
[22] S.S.S. Al-Diab, H.-K. Suh, J.E. Mark, H. Zimmer, J. Polym. Sci.
Part A: Polym. Chem. 28 (1990) 299–314.
[23] L. Angiolini, D. Caretti, L. Mazzocchetti, E. Salatelli, M. Biesemans,
R. Willem, J. Organomet. Chem. 691 (2006) 1965–1972.
[24] R.K. Ingham, S.D. Rosenberg, H. Gilman, Chem. Rev. 60 (1960)
459–539.
[14] M. Biesemans, F.A.G. Mercier, M. Van Poeck, J.C. Martins, G.
Dumartin, R. Willem, Eur. J. Inorg. Chem. 14 (2004) 2908–2913.
[15] J.C. Martins, F.A.G. Mercier, A. Vandervelden, M. Biesemans, J.-
M. Wieruszeski, E. Humpfer, R. Willem, G. Lippens, Chem. Eur. J.
8 (2002) 3431–3441.
[25] K.C. Molloy, K. Quill, S.J. Blunden, R. Hill, Polyhedron 5 (1986)
959–965.
[26] L. Angiolini, D. Caretti, L. Mazzocchetti, E. Salatelli, M. Biesemans,
R. Willem, J. Polym. Sci. Part A: Polym. Chem. 43 (2005) 3091–3104.
[27] L. Angiolini, D. Caretti, L. Mazzocchetti, E. Salatelli, C. Femoni, J.
Organomet. Chem. 689 (2004) 3301–3307.
[16] G. Deshayes, K. Poelmans, I. Verbruggen, C. Camacho-Camacho,
´
ˇ
´ˇ
ˇ
P. Degee, V. Pinoie, J.C. Martins, M. Piotto, M. Biesemans, R.
[28] J. Holecek, K. Handlır, A. Lycka, T.K. Chattopadhyay, B.
Majee, A.K. Kumar, Collect. Czech. Chem. Commun. 51 (1986)
1100–1111.
Willem, P. Dubois, Chem. Eur. J. 11 (2005) 4552–4561.
[17] C. Camacho-Camacho, M. Biesemans, M. Van Poeck, F.A.G.
Mercier, R. Willem, K. Darriet-Jambert, B. Jousseaume, T. Tou-
pance, U. Schneider, U. Gerigk, Chem. Eur. J. 11 (2005) 2455–2461.
[18] F.A.G. Mercier, M. Biesemans, R. Altmann, R. Willem, R. Pintelon,
R. Schoukens, B. Delmond, G. Dumartin, Organometallics 20 (2001)
958–962.
[19] L. Angiolini, D. Caretti, L. Mazzocchetti, E. Salatelli, M. Biesemans,
R. Willem, Appl. Organomet. Chem. 19 (2005) 841–847.
[20] V.V. Korshak, A.M. Polyakova, M.D. Suchkova, Vysokomol.
Soedin. 2 (1960) 13–19.
[29] R.M. Silverstein, G.C. Bassler, T.C. Morrill, Spectrometric Identifi-
cation of Organic Compounds, fourth ed., Wiley, New York, 1981, p.
120.
[30] B. Wrackmeyer, Ann. Rep. NMR Spectrosc. 16 (1985) 73–186.
[31] D.H. Hunter, C. McRoberts, Organometallics 18 (1999) 5577–5583.
[32] L. Angiolini, M. Biesemans, D. Caretti, E. Salatelli, R. Willem,
Polymer 41 (2000) 3913–3924.
[33] D.D. Perrin, W.L.F. Amarego, D.R. Perrin, Purification of Labora-
tory Chemicals, Pergamon Press Ltd., Oxford, 1966.