Organofunctional alkoxysilanes in dilute aqueous solution: New accounts on the dynamic structural mutability

Article abstract of DOI:10.1016/S0022-328X(01)00650-7

The condensation of alkyltrialkoxysilanes was studied using time-dependent ¹H- and ²⁹Si-NMR spectroscopy. The main goal was to correlate the results of these investigations with standard conditions in technical applications of the most commonly used commercial silane adhesion promoters. Thus, experiments with 3-aminopropyltriethoxysilane (DYNASYLAN AMEO), 3-glycidyloxypropyltrimethoxysilane (DYNASYLAN GLYMO) and 3-methacryloxypropyltrimethoxysilane (DYNASYLAN MEMO) were carried out in aqueous solutions at a selected range of concentrations. This investigation reflects a convenient way to determine the total concentration of reactive silanol groups during the course of the reaction. Moreover, the activity of the solution in adhesion promoting processes can be monitored and measured.

Full text of DOI:10.1016/S0022-328X(01)00650-7

Journal of Organometallic Chemistry 625 (2001) 208–216
www.elsevier.nl/locate/jorganchem
Organofunctional alkoxysilanes in dilute aqueous solution: new
accounts on the dynamic structural mutability
Fatmir Beari b, Marcus Brand d, Peter Jenkner a, Robert Lehnert d,
Hans Ju¨rgen Metternich c,*, Jaroslaw Monkiewicz a, H.W. Siesler b
a Degussa-Hu¨ls AG, Werk Rheinfelden, Untere Kanalstraße 3, D-79603 Rheinfelden, Germany
b Uni6ersita¨t-GH Essen, Institut fu¨r Physikalische Chemie, Schu¨tzenbahn 70, D-45117 Essen, Germany
c Infracor GmbH, Analytisch Technische Ser6ices, Paul Baumann Straße 1, D-45764 Marl, Germany
d Hu¨ls StructurA GmbH, Friedrich 6on Heyden Platz 1, D-01610 Nu¨nchritz, Germany
Received 20 September 2000; received in revised form 4 December 2000; accepted 12 December 2000
Abstract
1
The condensation of alkyltrialkoxysilanes was studied using time-dependent H- and 29Si-NMR spectroscopy. The main goal
was to correlate the results of these investigations with standard conditions in technical applications of the most commonly used
commercial silane adhesion promoters. Thus, experiments with 3-aminopropyltriethoxysilane (DYNASYLAN® AMEO), 3-gly-
cidyloxypropyltrimethoxysilane (DYNASYLAN® GLYMO) and 3-methacryloxypropyltrimethoxysilane (DYNASYLAN®
MEMO) were carried out in aqueous solutions at a selected range of concentrations. This investigation reflects a convenient way
to determine the total concentration of reactive silanol groups during the course of the reaction. Moreover, the activity of the
solution in adhesion promoting processes can be monitored and measured. © 2001 Elsevier Science B.V. All rights reserved.
Keywords: Alkoxysilanes; Hydrolysis; Reaction rates; Condensation; Siloxanes
The wide range of applications is based on the bi-
functionality of organofunctional silanes. These contain
a structurally inherent silicon-alkoxy function and a
functional organic substituent bonded via a C3-alkyl
group. In practice, on the one hand, chemical binding
to a generally inorganic material can be established by
exchange of the alkoxy groups during use. On the other
hand, the organic functionality permits an interaction
with organic polymers. In this simple model, a perma-
nent bond between inorganic substrate and organic
matrix is produced by means of stable chemical bonds
(Fig. 2).
1. Introduction
Organofunctional silanes, such as 3-glycidyloxy-
propyltrimethoxysilane (DYNASYLAN® GLYMO; 1),
3-methacryloxypropyltrimethoxysilane
(DYNASY-
LAN® MEMO; 2) or 3-aminopropyltriethoxysilane
(DYNASYLAN® AMEO; 3) [1–4], are used on an
industrial scale as adhesion promoters, for surface
modification, as crosslinking agents and for mechanical
reinforcement of ceramic surfaces (Table 1; Fig. 1).
Fig. 1. Structure of the silanes.
* Corresponding author. Tel.: +49-2365-492401; fax: +49-2365-
497350.
E-mail address: hans-juergen.metternich@infracor.de (H.J. Metter-
nich).
Fig. 2. Principle of adhesion promotion.
0022-328X/01/$ - see front matter © 2001 Elsevier Science B.V. All rights reserved.
PII: S0022-328X(01)00650-7

F. Beari et al. / Journal of Organometallic Chemistry 625 (2001) 208–216
209
Table 1
Structure, name and designation of the alkyltrialkoxysilane investigated
No.
Name
X
Ref.
1
2
3
4
3-Glycidyloxypropyltrimethoxysilane
3-Methacryloxypropyltrimethoxysilane
3-Aminopropyltriethoxysilane
–(CH2)3–O–CH2–CH(O)CH2
–(CH2)3–O–C(O)–C(CH3)ꢀCH2
–(CH2)3–NH2
[4]
[1]
[2,3]
3-(Triethoxysilylpropyl)ammonium acetate
–(CH2)3–NH+3
1
3. H- and 29Si-NMR spectroscopy of the hydrolysates
Owing to their efficiency as adhesion promoters,
functional organosilanes are now an integral part of
high-performance composites and ‘intelligent’ coating
systems.
products
The hydrolysis of organofunctional alkyltrialkoxysi-
lane was studied by means of H-NMR spectroscopy
1
[4,8,9]. Monitoring of the alkoxy groups bonded to
the central silicon atom permits an unambiguous dif-
ferentiation of the different hydrolysis intermediates
-monosilanol and silanediol- and, owing to the fast
measuring procedures, observation of even rapid hy-
drolysis reactions. However, nothing can be said
about the completely hydrolysed silanetriols and pos-
sible secondary oligomerization products since the de-
cisive information is not present in the 1H-NMR
spectra.
2. Reactivity of organofunctional silanes
In practice, organofunctional silanes are usually ap-
plied in a concentration of 0.2–2% by weight as addi-
tives as such or from aqueous or water–alcohol
solutions as a primer. As a result of the reaction with
water or moisture, the methoxy groups of 1 and 2 are
eliminated as methanol or the ethoxy groups of 3 as
ethanol and are replaced by hydroxyl units (silanols)
[4–7]. Below, the term ‘silanols’ is used as a sim-
plified description of the monomeric compounds bear-
ing at least one Si–OH function, i.e. monosilanol,
silanediol and silanetriol in the case of trialkoxysi-
lanes.
In the 29Si-NMR spectra, all monomeric units are
detectable in the initial stage of a hydrolysis reaction
(see Fig. 3).
However, silanols are metastable units which un-
dergo condensation to thermodynamically more stable
siloxane units by elimination of water.
These Si–OH functions prove to be very reactive in
establishing the bond between material and
organofunctional silanes. In addition to a fast hydrol-
ysis of the ethoxy- or methoxy-silicon bond, the
availability of the reactive hydroxyl units on the cen-
tral silicon atom significantly influences the reactivity
of the organofunctional silane. In this context, it has
long been postulated that exclusively silanetriols, i.e.
the completely hydrolysed, monomeric silanes, are re-
sponsible for the activity of the adhesion promoter
and these monomeric units are stable for a few hours
to a few days in aqueous solutions [5–7]. Subse-
quently, the activity of the aqueous mixtures of silane
hydrolysis products decreases as a result of crosslink-
ing to give insoluble, polymeric siloxanes (gel struc-
tures), which is evident from a substantial decrease in
the adhesion-promoting effect in the composite.
Knowledge of the silanol concentration and of the
degree of oligomerization is therefore of decisive im-
portance for the practical use of the silane formula-
tion. Detailed investigations into the type and
concentration of the structures present in the mixture
and their changes as a function of time are not avail-
able to date.
The formation of the silanols and of the oligomeric
structures formed by condensation is influenced both
by the functional group in the organic substituent
and by the general conditions of the hydrolysis reac-
tion (pH, temperature, concentration or catalyst) [4].
The present studies show that the partially hydrolysed
silanols undergo condensation and hence oligomeriza-
tion products are formed as early as in the hydrolysis
step, as shown schematically in Fig. 4. Individual
oligomer structures can be identified without prob-
lems in dilute aqueous solutions resembling those in
practice, by means of 29Si-NMR spectroscopy (Fig.
5). In the experiments, the addition of chromium(III)
acetylacetonate proved useful, accelerating the relax-
ation of the excited silicon nuclei. Consequently, it
Fig. 3. 29Si-NMR spectrum at beginning of hydrolysis of 1.

210
F. Beari et al. / Journal of Organometallic Chemistry 625 (2001) 208–216
four resonance signals occur, these signals correspond-
ing to the unhydrolysed silane and to the monosilanol,
silanediol and silanetriol fraction (Fig. 3). Depending
on the degree of hydrolysis at the individual silicon
atoms, a large number of different environments, which
are detectable as broad signal peaks in the 29Si-NMR
spectrum, result in the region of the M-, D- and in
particular T-structures.
4. Results and discussion
The literature [15,17] reports on reactions of
organofunctional alkyltrialkoxysilanes with equimolar
amounts of water. On the basis of the work on the
hydrolysis of organofunctional trialkoxysilanes [4], the
present work focuses on a very close relationship with
industrial use in which the organofunctional tri-
alkoxysilanes are generally reacted at low concentra-
tions (0.2–2%) with water. The inorganic surface to be
modified is then brought into contact with the dilute
silane solution, and a chemical bond can rapidly form
between silane and workpiece. In general, the solvent
water and the alcohol fractions liberated during the
hydrolysis reaction are then removed. The result is an
increase in the concentration of the dissolved silicon
compound and the condensation of present silanol
groups. In this way, a covalently bonded, crosslinked
polysiloxane layer forms on the material surface. A
decisive factor in practice is that this process takes
place as completely as possible, since only very exten-
sive coverage of the material surface guarantees opti-
mum properties (adhesion promotion, surface
modification). For this reason, fairly dilute aqueous
systems were used in the present work (silane concen-
trations 4–10%), in order to be able to investigate the
oligomerization during the hydrolysis reaction in a
manner closely resembling that in practice. The experi-
ments at higher concentrations (50% of silane) serve as
Fig. 4. Course of hydrolysis and condensation.
Fig. 5. Oligomers of 3 (analytically monitored in the 29Si-NMR
spectrum).
was possible on the one hand to obtain spectra with
optimum sensitivity for reliable, quantitative 29Si-NMR
evaluation and, on the other hand, to achieve the short
measuring times required for detecting the reaction
kinetics. Due to the considerably longer measuring
1
times compared with H-NMR experiments, a determi-
nation of the initial intermediate hydrolysis steps is
possible only to a limited extent. The hydrolysis prod-
ucts were monitored spectroscopically over a period of
up to 3 months.
With the aid of 29Si-NMR spectroscopy, it is possible
to obtain information about the structure and amount
of oligomers found [10–15]. For the characterization of
the oligomer structures, the term M-structures is used
for monocrosslinked units, D-structures for di-
crosslinked units and T-structures for tricrosslinked
units in general (see Figs. 4 and 5). For the case of a
tetraalkoxysilane (e.g. tetramethoxysilane), additional
signals for the siloxane units in which all four silanol
units are oligomerized additionally occur (Q-structure)
[11,16].
The chemical shift which initially characterize the
fundamental environment of the silicon atom, i.e. the
amount of siloxane bridges, is determined by a large
number of different resonance lines. These signal lines
indicate structural units of different environments. In
the region of the uncrosslinked silane, a maximum of
Fig. 6. Course of hydrolysis/oligomerization of 1 in a 10% aqueous
solution (pH 6).

F. Beari et al. / Journal of Organometallic Chemistry 625 (2001) 208–216
211
At the same time, the opening of the terminal oxirane
to give the diol (Fig. 8) can be observed, although much
more slowly [18,19].
Surprisingly, a so far unknown reaction product of
diol and 3-glycidyloxypropylsilanols or OH-functional
oligomers is detected after only 1 h (‘X component’),
the concentration of which increases to \25% during
the period of observation (1600 h) [20,21] (Fig. 9).
The proportion of free, monomeric silanetriol de-
creases in the course of the hydrolysis from initially
about 75% (15 h) to B10% (1600 h).
From the available data the absolute concentration
of remaining reactive Si–OH functions can be deter-
mined as a function of the reaction time. This concen-
tration being decisive with regard to the reactivity with
a substrate, in particular in the case of pretreatment
with dilute aqueous silane solutions (primer applica-
tions). It is to be expected that even proportionately
oligomerized trialkoxysilanes having free Si–OH func-
tions are capable of binding to inorganic surfaces. In
principle, a more uniform coating can be expected when
a material surface is coated with, for example, short-
chain, doubly bridged oligomers (D-structures). A ho-
mogeneous coverage of the material surfaces has many
advantages even for industrial manufacture, since
shorter coating times can be achieved.
Fig. 7. Course of hydrolysis/oligomerization of 1 in a 50% aqueous
solution (pH 6).
The following equation is used for determining the
proportion of acti6e Si–OH functions:
Active SiꢁOH functions (mol%)
3 · mol%Silanetriol+2 · mol%M+mol%D
=
3
Fig. 8. Ring opening of 1 in a 10% aqueous solution (pH 6).
In a 10% solution, starting from 85% (15 h), it is still
\70% after a standing time of 160 h (corresponding to
about 2.1 hydroxyl units per silicon atom). In a 50%
solution, a maximum hydroxyl group concentration is
reached after only 8.5 h (\93%) and decreases to
B15% (corresponding to about 0.4 hydroxyl unit per
silicon atom) after a standing time of 550 h (Figs. 6–8).
a model for describing reactions in the concentration
phase during drying on the material surface.
4.1. 3-Glycidyloxypropyltrimethoxysilane (1)
In demineralized water (pH 6), hydrolysis to the
corresponding silanols and consecutive oligomerization
take place, as expected, relatively slowly, since the rate
of the reaction is decisively influenced by the H+
concentration [4].
4.2. 3-Methacryloxypropyltrimethoxysilane (2)
The time-dependent concentration curve of the re-
sulting structures of monomers and oligomers (M-/D-/
Fig. 9. Possible structural elements of the ‘X component’.

212
F. Beari et al. / Journal of Organometallic Chemistry 625 (2001) 208–216
T-structures; Fig. 5) during hydrolysis and condensa-
tion of 2 is essentially comparable with 1 at pH 6. In an
earlier paper [4], it was possible to show that the rate of
the reaction (kinetics: pseudo-first order) is influenced
foremostly by the pH. By way of example, the course of
the hydrolysis to the monomeric silanols is shown in
which is complete in a few minutes, is comparable with
that in a strongly acidic medium. Another special fea-
ture is that the amino function causes both the unhy-
drolysed silane and its hydrolysis and condensation
products to have excellent water solubility. The
aqueous solutions of 3 are therefore stable for months.
The results of the 29Si-NMR investigations on the
aqueous solutions of 3 with different initial concentra-
tions are illustrated in Figs. 14–17.
In all experiments, a rapid build-up of oligomers (D-
and T-structures) is detectable. The percentage of the
doubly and triply bridged units, and hence the rate of
the condensation, increases with increasing initial con-
centration as expected. Overall, depending on the initial
concentration of 3, a virtually steady state is reached
after a reaction time of about 15 h (50%), about 50 h
(10%) and about 100 h (4%). In the further course of
the reaction, no significant changes are observed in the
reaction solutions. The mixtures then remain stable
1
Fig. 10, as the result of the H-NMR analysis.
The corresponding free silanetriol cannot be detected
1
but is clearly quantifiable by combination of H-NMR
and 29Si-NMR data. The results of measurements on
10% solutions at pH 6 and pH 4 are compared in Figs.
11 and 12.
The experiments had to be terminated after 74 h (pH
4) and 188 h (pH 6) since the aqueous solution became
cloudy and therefore complete detection of all sec-
ondary hydrolysis and condensation products could no
longer be guaranteed (Figs. 11 and 12).
Nevertheless, the experimental data for 2 show a
picture analogous to 1, with high initial concentrations
of silanetriol (after 1.5 h: 88% at pH 4 and after 2.5 h:
95% at pH 6), which decreased to below 15% (pH 4)
and below 35% (pH 6). In contrast, an Si–OH concen-
tration of more than 60% is maintained in both cases
(i.e. on average about two hydroxyl functions per sili-
con atom), finally ensuring the reactivity of the formu-
lations. In comparison, the hydrolysis of 2 takes place
much more rapidly in acidic solution but the yield of
active Si–OH functions is higher at neutral pH and the
tendency to oligomerization is suppressed. This means
that acid catalyses both the hydrolysis and the conden-
sation of the silanols formed.
The supplementary hydrolysis experiments on 50%
aqueous solutions (pH 6) of 2 underline the relation-
ships presented (Fig. 13).
The loss of silanetriol and the higher proportion of
T-structures impressively illustrate the crosslinking ef-
fects which occur on the material surface, for example
in the drying step after application of the aqueous
primer solution, owing to concentration of the
organosilicon compounds present. The remaining Si–
OH concentration of about 45% after the solution has
been left to stand for 223 h (corresponding on average
to 1.35 hydroxyl radicals per silicon atom) guarantees
in part the activity of the primer formulation.
Fig. 10. Course of hydrolysis of 2 in a 2% aqueous solution (1H-
NMR; pH 6).
4.3. 3-Aminopropyltriethoxysilane (3) and
3-(triethoxysilylpropyl)ammonium acetate (4)
In contrast to most other organofunctional silanes,
the hydrolysis and condensation of silanes having pri-
mary amino functions in aqueous solutions are auto-
catalysed. This is evident from the pronounced
polarization of the molecule (electrostatic interaction of
NH2lSiOR) and from the strong interaction of the
NH2 group with the polar solvent itself. An aqueous
solution of 3 exhibits a pH of 11 so that the hydrolysis,
Fig. 11. Course of hydrolysis/oligomerization of 2 in a 10% aqueous
solution (pH 4).

F. Beari et al. / Journal of Organometallic Chemistry 625 (2001) 208–216
213
Fig. 15. Course of hydrolysis/oligomerization of 3 in a 10% aqueous
solution (pH 11).
Fig. 12. Course of hydrolysis/oligomerization of 2 in a 10% aqueous
solution (pH 6).
Fig. 13. Course of hydrolysis/oligomerization of 2 in a 50% aqueous
solution (pH 6).
Fig. 16. Course of hydrolysis/oligomerization of 3 in a 50% aqueous
solution (pH 11).
The absolute concentration of remaining reactive Si–
OH functions is 48% in the case of a 4% solution after
complete hydrolysis, increases to a maximum value of
\55% after 7 h and is still \47% (about 1.4 hydroxyl
units per silicon atom) after a standing time of 172
days. In contrast, following a brief rise to about 23%,
the percentage of silanetriol in the 4% solution of 3
decreases to 19% after 3 days and only 12% after 172
days (Fig. 18).
In 10% and 50% solution of 3 this threshold cannot
be reached even in the initial part. The maximum
hydroxyl group concentration is about 37% (corre-
sponding to about 1.1 hydroxyl units per silicon atom)
for the 10% solution and B26%, i.e. 0.8 hydroxyl unit
per silicon atom, in the case of the 50% solution.
For the experiments with 4, in comparison with 3, an
approximately 25% decrease in the crosslinked struc-
tures can be detected at both concentrations (10% and
50%) for the hydrolysis at neutral pH in the first 3–5
days. In the case of hydrolysis in water containing
Fig. 14. Course of hydrolysis/oligomerization of 3 in a 4% aqueous
solution (pH 11).
over several weeks to months, as it was possible to
show by long-term experiments.

214
F. Beari et al. / Journal of Organometallic Chemistry 625 (2001) 208–216
Fig. 17. Course of hydrolysis/oligomerization of 3 in a 50% aqueous
Fig. 18. Comparison of the active Si–OH functions of 3 and 4.
solution (pH 11) in initial part.
acetic acid, the reduction is even \30%. Associated
with this is an increase in active Si–OH functions.
However, this increase corresponds only to a fraction
of the percentage to be observed in the crosslinked
structures, since the crosslinking starts somewhat re-
tarded. Accordingly, up to about 60 h in neutral and
acidic media, the content of M- and D-structures is
higher and the content of active Si–OH is only about
5% (pH 6) or 10% (pH 4) higher than in the investiga-
tions at pH 11.
The NMR results are backed up by LC-MS studies
on aqueous solutions of 3. The aim of the comparative
measurements was to detect the low molecular weight
silanols in the initial stage of the hydrolysis reaction
and of the molecular weight build-up as a consequence
of condensation reaction taking place in the solution
(Figs. 19 and 20).
Fig. 19. Time-dependent mass spectra of an 1% aqueous solution of
3 (reaction time 1 and 15 min).
A modified LC-MS technique with direct injection
was used in order on the one hand to rule out reaction
of the reactive Si–OH functions with the column mate-
rial of the LC separation and, on the other hand, to
suppress fragmentation in the mass spectrometer by the
use of a mild ionization procedure (APCI). The mass
spectra can therefore be used only for the qualitative
detection of silicon species and do not reflect the per-
centage distribution of the individual components.
While the molecular peaks of silane (M+H: m/e 222),
monosilanol (M+H: m/e 194), silanediol (M+H: m/e
166) and silanetriol (M+H: m/e 138) can be detected
in the spectra of 1% solutions of 3 after a hydrolysis
period of about 1 min, a substantial build-up of
oligomers is observable even after only 15 min. The
mass spectrum of a 10% solution of 3 recorded after
hydrolysis for 21 min shows structural units which
prominently indicate siloxanes with two to nine SiO
units.
Fig. 20. Time-dependent mass spectra of an 10% aqueous solution of
3 (reaction time 21 min).
reactivity of the aqueous silane solutions is associated
with the oligomer build-up in aqueous solutions of
organofunctional silanes. In contrast, through the pro-
portionate crosslinking, material surfaces to be
modified are offered with relatively large agglomerates
which permit homogeneous, complete coverage.
In comparison with the hydrolysis of 3, the two
non-amine trialkoxysilanes 1 and 2 have a significantly
lower tendency to crosslinking. Particularly in the first
1–3 days, substantially higher contents of silanetriol
and active Si–OH functions are associated with the
On the basis of the present investigations, and con-
trary to all postulates [5–7], initially no reduction in the

F. Beari et al. / Journal of Organometallic Chemistry 625 (2001) 208–216
215
a negative sign means a shift to high field relative to the
standard. To ensure an acceptable relaxation time, a
relaxation accelerator in an organic solvent had to be
added to the reaction mixture before the measurement.
5.1. Starting compounds
The organofunctional trialkoxysilanes (DYNASY-
LAN® GLYMO, DYNASYLAN® MEMO, DYNA-
SYLAN® MTES, DYNASYLAN® AMEO) used in the
hydrolysis experiments originate from the production of
Degussa-Hu¨ls AG, Rheinfelden Works. The purity was
\98% in each case.
5.2. Characterization of the secondary hydrolysis
Fig. 21. Active Si–OH functions of 1–4 in comparison (10% aqueous
solutions).
products
Two, 10 or 50% of the trialkoxysilane to be investi-
gated were stirred with demineralized water in a glass
beaker at 500 rev min−1. At fixed time intervals, the
stirrer was switched off and a 2 ml sample was taken
from the reaction mixture. The 29Si-NMR spectroscopic
analysis was carried out after addition of 1 ml of
acetone-d6 with 3% of chromium(III) acetylacetonate as
an internal lock substance and relaxation accelerator.
At the same time, the homogeneity of the solution was
checked by means of a turbidity measurement.
reduced crosslinking of the hydrolysis products of 1
and 2 (Fig. 21).
The non-amine aqueous solutions of hydrolysis prod-
ucts of 1 and 2 have limited usability in terms of
adhesion promotion not because of insufficient Si–OH
concentration but as the result of the precipitation of
the active ingredient from the aqueous solution after
standing for a few days. The cause of the poor solubil-
ity of the oligomers of 1 and 2 is the loss of hydrophilic
functions which occurs with increasing condensation. A
minimum concentration of 50% of active Si–OH func-
tions, i.e. on average 1.5 hydroxyl groups per silicon
atom, proves to be a limit which can be used as a
neutral evaluation criterion for the activity of a silane
solution. Consequently, a period of 5–40 h can be
recommended for optimum processing of 1 and 2 at pH
3–4 and of 3 at pH 4–11. The present investigations
were carried out on trialkoxysilanes with very different
functional groups and are therefore sufficient to reflect
a broad cross section of possible functionalities. Fur-
ther work on the evaluation of the reaction behavior of
functional alkoxysilanes will be presented shortly.
5.3. Synthesis of 3-(triethoxysilylpropyl)ammonium
acetate
A total of 1000 g (4.52 mol) of 3-aminopropyltri-
ethoxysilane was placed in a 2 l three-necked stirrer
apparatus with a dropping funnel, KPG stirrer, reflux
condenser and thermometer, ice water and nitrogen
blanketing. The reactor was cooled in an ice-water
bath. A total of 271.7 g (4.52 mol) of glacial acetic acid
was added dropwise while stirring, at a rate such that a
bottom temperature of about 50°C was not exceeded.
After the end of the addition, stirring was continued for
about a further 30 min at room temperature.
5. Experimental
5.4. Spectroscopic data of
3-(triethoxysilylpropyl)-ammoniumacetate (4)
All experimental work was carried out in glass ap-
paratuses and NMR tubes pretreated with hexamethyl-
disiloxane, in order to suppress a reaction with the glass
pH 7.2; 29Si-NMR (C2D2Cl4): l= −46.8; 13C-NMR
(C2D2Cl4): l=178.7 (1C, CH3COO−), 59.0 (3C,
Si(OCH2CH3)3), 42.5 (1C, +NH3CH2CH2CH2Si), 25.1
(1C, CH3COO−), 22.3 (1C, +NH3CH2CH2CH2Si), 19.1
(3C, Si(OCH2CH3)3), 8.6 (1C, +NH3CH2CH2CH2Si);
1
surface. pH: Metrohm E 597/1; NMR: H: Bruker AC
250 (250 MHz) and DPX 360 (360 MHz), internal
standard TMS or TMS salt (hydrolysis experiments);
29Si: Bruker AC 250 (49 MHz) and DPX 360 (71 MHz),
relaxation accelerator chromium(III) acetylacetonate,
internal standard TMS. 29Si-NMR spectra were 1H-
broadband-decoupled; 1H-NMR spectra of alkyltri-
ethoxysilanes were recorded in selectively decoupled
form [22]. A positive sign means a shift to low field and
1H-NMR
(C2D2Cl4):
l=6.34
(m,
3H,
+NH3CH2CH2CH2Si), 3.74 (q, 6H, J=7.04 Hz,
Si(OCH2CH3)3), 2.75 (tq, 2H, J=8.3 and 7.3 Hz,
+NH3CH2CH2CH2Si), 1.84 (s, 3H, CH3COO−), 1.67
(quin, 2H, J=8.3 Hz, +NH3CH2CH2CH2Si), 1,15 (t,
9H, J=7.04 Hz, Si(OCH2CH3)3), 0.57 (t, 2H, J=8.3

216
F. Beari et al. / Journal of Organometallic Chemistry 625 (2001) 208–216
Hz, +NH3CH2CH2CH2Si); C11H27O5NSi (239.33):
Ber.: C, 46.93; H, 9.69; N, 4.98; Gef.: C, 45.89; H, 9.55;
N, 5.05%
[9] L. Delatire, F. Babonneau, Mat. Res. Soc. Symp. Proc. 346
(1994) 365.
[10] S. Macomber, NMR Spectroscopy: Basic Principles and Appli-
cation, Harcourt Brace Javonowich, San Diego, 1988.
[11] E.A. Williams, NMR spectroscopy of organosilicon compounds,
in: The Chemistry of Organic Silicon Compounds, Wiley, Lon-
don, 1989.
[12] E.A. Williams, Recent advances in silicon-29 NMR spec-
troscopy, Annual Reports on NMR Spectroscopy 15, London,
1983.
References
[1] J. Monkiewicz, A. Frings, M. Horn, P. Jenkner, H.-J. Koetzsch,
F. Kopfgans, C.-D. Seiler, H.-G. Srebny, B. Standke, Process for
the Preparation of 3-Acryloxypropylalkoxysilanes, DE 4434200
A1 960328, 1996.
[2] C.-D. Seiler, H. Rauleder, H.-J. Koetzsch, H.-G. Srebny, Process
for the preparation of aminofunctional organosilanes with low
chlorine contamination, EP 702017 A1 960320, 1996.
[3] T. Kudota, M. Endo, T. Hirahara, Preparation of aminopropyl-
alkoxysilane derivates; JP 10017578 A2 980120 Heisei, 1998.
[4] M. Brand, A. Frings, P. Jenkner, R. Lehnert, H.J. Metternich, J.
Monkiewicz, J. Schram, Z. Naturforsch. 54b (1999) 155.
[5] E.P. Plueddemann, Silane Coupling Agents, Plenum Press, New
York, 1991.
[13] G. Engelhardt, H. Jancke, Polym. Bull. 5 (1981) 577.
[14] G.S. Caravajal, D.E. Leydon, G.R. Quinting, G.E. Maciel, Anal.
Chem. 60 (1988) 1776.
[15] E.A. Williams, J.D. Cargioli, Annual Reports on NMR Spec-
troscopy 9, London, 1979.
[16] S.E. Rankin, J. Sefcik, A.V. McCormieck, Ind. Eng. Chem. Res.
38 (1999) 191.
[17] J.K. Crandall, C. Morel-Fourrier, J. Organomet. Chem. 489
(1995) 5.
[18] G. Xue, J.L. Koenig, H. Ishida, D.D. Wheeler, Rubber Chem.
Technol. 64 (1991) 2.
[19] R. Nass, E. Arpac, W. Glaubitt, H. Schmidt, J. Non-Crystal.
Solids 121 (1990) 370.
[6] H. Ishida, J.L. Koenig, Appl. Spectrosc. 32 (1978) 469.
[7] E.P. Plueddemann, SPI 24th Ann. Tech. Conf. Reinf. Plast.
19-A, 1969.
[8] S. Savard, L.P. Blanchard, R. Leonard, R.E. Prudhomme,
Polym. Composites 5(4) (1984) 242.
[20] L. Lan, A. Montenero, G. Gnappi, E. Dradi, Gazzeta Chim.
Ital. 127 (1997) 505.
[21] G.R. Bogat, D.E. Leyden, J. Chromatogr. 483 (1989) 209.
[22] S. Braun, H.-O. Kalinowski, S. Berger, 150 and More Basis
NMR Experiments, VCH, Weinheim, 1998.
.