V. Nissinen et al. / Journal of Molecular Catalysis A: Chemical 413 (2016) 94–99
95
and (110)) and thus the structure of MgCl2 [15]. Both internal
and external electron donors also have a great influence on the
properties of Ziegler–Natta catalysts and polymers produced. The
productivity and hydrogen response of the catalyst as well as the
stereoregularity, molar mass and molar mass distribution of the
polymers formed are affected by the electron donors [2,15]. Elec-
tron donors regulate the content and distribution of active species
by competing from the same surface sites and also affect the stere-
ospesificity of active sites due to steric effects [13,16,17]. Typical
electron donors in Ziegler–Natta catalysts are e.g. aromatic esters,
phthalates, diethers and succinates [18]. Diethers are perhaps the
most interesting group of electron donors due to their ability to pro-
duce highly isotactic polypropylene with a very high productivity
without the presence of an external electron donor [19].
products formed were isolated and washed in the glove box and
then dried as mentioned before.
2.3. Characterization of the solid products
Magnesium content of solid products was determined by
an EDTA (ethylenediaminetetraacetic acid) titration. Amounts of
organic molecules bound to the solid products were determined
1
by H NMR spectroscopy (Bruker AMX-400 spectrometer). For this
analysis, solids were dissolved in a 10% (V/V) D2SO4/D2O solu-
tion and sodium acetate was used as an internal standard. The
chlorine content of the products was obtained on the basis of the
determined mole fractions of magnesium, organic moieties and
unreacted donor compounds.
Solid state 13C NMR spectra of the products were recorded. Mea-
surements were conducted with a Bruker AMX-400 spectrometer
using a cross polarization and a magic angle spinning with the fol-
lowing parameters: a spin rate 4500 Hz, a relaxation delay 4 s, a
contact time 3.0 ms and a number of scans 10000. Glycine was
used as an external standard. IR spectra of solid products were
measured using a Nicolet Impact 400D spectrometer. Spectra were
recorded using a DRIFT (diffuse reflectance infrared Fourier trans-
formation) unit mounted inside a glove box. Number of scans was
In our studies, MgCl2 supports have been synthesized with
Grignard–Wurtz reaction in the presence of different ethers as elec-
tron donors [20–22]. In the case of diethers, we have observed a
cleavage of the C O bond during the Grignard–Wurtz synthesis.
The cleavage reaction is undesired if the aim is the synthesis of
MgCl supports containing diethers as electron donors. The cleav-
2
age reaction can also be adverse if it takes place during the other
steps of catalyst preparation (e.g. addition of TiCl ) or even during
4
polymerization. The aim of this study is to gain insight into the
cleavage reaction of diethers, especially 1,3-dimethoxypropane,
which is the simplest of 1,3-diethers. The cleavage reactions of
simple ethers have been studied [23–25] in the past decades, but
according to our knowledge no reports on the cleavage of diethers
in the Grignard–Wurtz synthesis have been published.
−
1
32 and resolution 2 cm
.
2.4. Characterization of the liquid and gas phases
Also liquid and gas phases of some reaction mixtures were ana-
lyzed. Liquid phases were analyzed with 1H NMR spectroscopy.
Samples were taken directly from the liquid phases and measured
2
. Experimental
in CDCl . TMS was used as an internal standard. Gas phases of
3
the reaction mixtures were analyzed using IR spectroscopy and
gas chromatography. An IR spectrum was recorded with a Bruker
Vertex 70 spectrometer. Number of scans was 16 and resolution
2
.1. Materials
Octane (reagent grade, 98%), 1-chlorobutane (ReagentPlus, 99%),
-methoxy-1-propene (97%) and butylmagnesium chloride (2.0 M
−
1
2
cm . A gas cuvette was approximately 16 cm long and equipped
3
with ZnSe windows. A gas chromatogram was recorded using a
Hewlett-Packard 6890 Series GC system which was equipped with
a Chromosorb column and a hot wire detector. Helium was used
as a carrier gas. For analysis of the gas phase of the reaction mix-
ture, the autoclave was thoroughly cooled with an ice bath after
the heating. Samples were taken from the autoclave through a long
copper tube which was also cooled with an ice bath.
solution in diethyl ether) were purchased from Sigma–Aldrich
and 1,3-dimethoxypropane (98%) from CHEMOS GmbH. Ethers, 1-
chlorobutane and octane were dried and stored over activated 3 Å
molecular sieves. Magnesium turnings (99.9+%) were purchased
◦
from Acros Organics and dried in an oven at 110 C for two days
prior to the synthesis. All reactions and handling of products were
strictly done under an argon or nitrogen atmosphere using a glove
box and Schlenk techniques.
3. Results and discussion
2.2. Grignard–Wurtz synthesis
3.1. Grignard–Wurtz synthesis with 1,3-dimethoxypropane as an
electron donor
Syntheses were mostly carried out in an autoclave. Follow-
ing amounts of reagents were used: 0.5 g (0.021 mol) magnesium;
some small crystals of iodine; 6.45 ml (0.062 mol) 1-chlorobutane;
Grignard–Wurtz synthesis was performed in the presence of
1,3-dimethoxypropane as an electron donor in an autoclave using
magnesium and 1-chlorobutane as starting materials and octane
as a solvent. The molar ratio of 1,3-dimethoxypropane to mag-
nesium was 1:2. The powder X-ray diffraction analysis indicated
that the solid product formed was highly disordered. The diffrac-
togram of the product differed notably from that of ı-MgCl2 (the
diffractograms are presented in Fig. SA1 in Supplementary informa-
tion). According to solid state CP/MAS 13C NMR spectrum (Fig. 1),
the product contained 1,3-dimethoxypropane and another com-
pound, which was preliminarily identified as methanol or methoxy
groups. The relative content of 1,3-dimethoxypropane was signif-
icantly lower than that of methanol or methoxy groups. In the IR
spectrum of the solid product (Fig. SA2), the typical O H stretch-
1
.27 ml (0.010 mol) 1,3-dimethoxypropane and 20 ml octane.
Reagents were packed into the autoclave in the glove box. After
◦
packing, the autoclave was heated at 130 C for 2 h with a constant
mixing. The solid product formed was separated by a filtration in
the glove box where the product was also washed with octane (3
times 10 ml). The product was then dried in vacuum at room tem-
perature at least for two days. After washing and drying the product
was fine white powder.
Some syntheses were also carried out in glassware at an atmo-
spheric pressure in order to take samples from the reaction mixture
during the reaction. The reagents were packed in the glove box
into a two-necked flask equipped with a reflux condenser. The
◦
−1
flask was then heated at 130 C for 2 h with an oil bath and mixed
ing vibration at 3350 cm for a possible methanol had a very low
−
1
with a magnetic stirrer. A constant argon flow through the system
ensured an oxygen and moisture free atmosphere. Samples were
taken from the reaction mixture using two-headed needles. Solid
intensity compared to the C O stretching vibration at 1000 cm
.
Thus, it was highly probable that the solid product formed con-
tained methoxy groups instead of methanol. 1H NMR spectrum