J. Blanchard, M. In, B. Schandel, C. Sanchez
FULL PAPER
in n-butanol. The ratio [Hϩ]/[Ti] was kept at 0.02, while the hy-
The 17O spectra of 17O-enriched titanium oxo-polymers have
drolysis ratio was varied from 1 to 10. The concentration of the been reported previously[28]. They are characterized by four broad
resulting sol was 0.18 mol·lϪ1
.
bands, which can be assigned, on the basis of the published litera-
ture[29][30], as follows:
The samples used for recording 17O-NMR spectra were prepared
as follows:
• δ ϭ 650Ϫ850: µ2O oxo-bridges
• δ ϭ 450Ϫ600: µ3O oxo-bridges
• δ ϭ 300Ϫ400: µ4O oxo-bridges
• δ ϭ 200Ϫ300: µ5O oxo-bridges
Sample ᕃ was prepared following procedure (i) with both the
[acac]/[Ti] and [H2O]/[Ti] ratios equal to 1. The water used for this
first hydrolysis was 5%-enriched with oxygen-17. After 1 h at room
temperature, the sample was further hydrolyzed with a solution of
5% enriched water in n-butanol. The total hydrolysis ratio of the
sample was thus equal to 2.
• the signal appearing at around δ ϭ 0 corresponds to the re-
sidual water.
Two additional peaks can be observed in the presence of acetyl-
acetone[28]
:
Sample ᕄ was prepared according to the same procedure as
sample ᕃ, but the first hydrolysis was performed with unenriched
water and the second hydrolysis with 10%-enriched water. Thus,
after the second hydrolysis, the two systems have the same chemical
composition and the same enrichment in oxygen-17.
• one broad peak at δ ϭ 300Ϫ400, attributable to acac bonded
to titanium
• one narrow peak at δ ϭ 273, attributable to the enol form of
free acetylacetone.
Zirconium Oxide Based sols and gels were prepared according to
a procedure analogous to (i) above: Zr(OnPr)4 was first diluted with
n-propanol, and then acetylacetone was added. The [acac]/[Zr] ratio
was either 0.9 or 1. Hydrolysis was performed by the dropwise ad-
dition of a solution of water in n-propanol. The [H2O]/[Zr] ratio
was varied from 1 to 10. The concentration of the resulting sols
These two peaks appear a few hours after hydrolysis, and are
only observed at complexation ratios higher than 1 and with a hy-
drolysis ratio higher than 2[16]. They result from the isotopic 17O-
exchange between enriched H2O and the free acetylacetone released
by the hydrolysis of the acac ligands[28]
.
was 0.44 mol·lϪ1
.
TiϪacac ϩ H2O Ǟ TiϪOH ϩ acacH
acacH ϩ H2O* Ǟ acac*H ϩ H2O
TiϪOH ϩ acac*H Ǟ TiϪacac* ϩ H2O
Silicon Oxide Based sols and gels were prepared following pro-
cedure (ii): Si(OEt)4 was first diluted with ethanol. A solution of
water acidified with HCl (pH 0 or pH 2) in ethanol was then added
dropwise. The hydrolysis ratio was varied from 0.5 to 10 and the
(the reaction leading to enrichment of the acetylacetone in oxy-
gen-17 is described in Scheme 1).
concentration of the resulting sol was 0.54 mol·lϪ1
.
The recording of oxygen-17 NMR spectra necessitates a pre-ac-
quisition delay in order to eliminate the trailing edge of the high-
power pulse. This delay gives rise to a rolling baseline of sin x/x
shape. A baseline correction is then performed, using a cubic
splines baseline function (WinNMR 1D). The total intensity of the
NMR signal has been measured for all the spectra reported herein,
in order to check that this area does not vary with the ageing and
the preparation of the sample solution. The areas of the spectra of
all the samples after the second hydrolysis are equal, within an
experimental error of ±10%. A calibration has also been performed
using a solution of 10%-enriched water in n-butanol, at the same
concentration as in the samples of oxo-polymers. The total intensit-
ies of the spectra of the oxo-polymer samples before the second
hydrolysis, after the second hydrolysis, and after 24 hours of ageing
are all consistent with the intensity of the water peak of the cali-
bration solution, within an experimental error of ±10%. This result
indicates that the formation of large polymers, which would result
in the partial loss of the signal of the oxo-polymers and, therefore,
in quantitativity problems, is negligible.
Characterizations
Water Titrations were performed by the Karl-Fischer method
using a 701 KF Titrino apparatus. The solvent used for the ti-
trations was Hydranal Solvent K, rather than methanol, in order
to prevent side reactions of ketones and metal hydroxyls[27]. The
¨
reagent was Hydranal-composite 5K (Riedel-de-Haen).
The 29Si-NMR spectra were recorded on a Bruker MSL 400
spectrometer operating at 79.48 MHz. The sample solution was
placed in an 8 mm tube, which was placed inside a 10 mm tube
containing C6D6 as a lock solvent. Tetramethylsilane was used as
reference (δ ϭ 0).
Silicon-29 is a spin 1/2 nucleus with a natural abundance of 4.7%.
A disadvantage of this nucleus is its high spin lattice relaxation
time T1, which results in a long recycle delay. Thus, the relaxation
agent chromium acetylacetonate Cr(acac)3 was added in order to
reduce this delay. Cr(acac)3 has previously been used in similar sys-
tems, where it was shown that it does not interfere with the poly-
merization process[7]
.
Calorimetric Study of the Complexation and Hydrolysis Reac-
tions: Four alkoxides were studied: Ti(OnBu)4, Ti(OEt)4,
Ti(OiPr)4, Zr(OnPr)4.
A second disadvantage of this nucleus is the presence of back-
ground Q4 silicon units present in the silica tube, which result in a
large peak at about δ ϭ 110. In order to suppress this peak, we
used a π.2ϪτϪπ echo sequence.The experimental NMR conditions
were as follows: first pulse 12.8 µs; τ ϭ 40 ms; second pulse 25.6
µs; recycle delay 30 s.
The 17O-NMR spectra were recorded on a Bruker MSL 400
spectrometer operating at 54.2 MHz. Again, the sample solutions
were placed in an 8 mm tube, which was immersed in C6D6, as a
lock solvent, in a 10 mm tube. H2O was used as reference (δ ϭ 0).
The experimental NMR conditions were as follows: pulse 19.6 µs;
pre-acquisition delay (time between the end of the RF pulse and
the beginning of the acquisition) 24 µs; recycle delay 0.2 s, number
of scans 1000.
First, m1 g of the alkoxide (or acac-modified alkoxide) was di-
luted with mЈ1 g of its parent alcohol. This solution was placed in
a Dewar vessel and its temperature T was measured. A solution of
m2 g of acetylacetone in mЈ2 g of the parent alcohol (or the hydroly-
sis solution of water in parent alcohol) at temperature T2 was then
added. The final temperature Tf was measured about 1 min. after
the addition of acetylacetone (or water). The [acac]/[M] ratio of the
resulting solution was varied from 0.5 to 2 [or to 4 for Zr(OnPr)4].
The hydrolysis ratio was varied from 0.5 to 4. The molar enthalpy
∆H of the reaction was calculated using the equation:
(m1∆H)/M ϩ Cp(Ma ϩ m1 ϩ mЈ1)(Tf Ϫ T1) ϩ Cp(m2
ϩ mЈ2)(Tf Ϫ T2) ϭ 0
1124
Eur. J. Inorg. Chem. 1998, 1115Ϫ1127