Mechanochemical method of producing triethoxysilane

Article abstract of DOI:10.1007/s11172-019-2382-x

A mechanochemical method for synthesis of triethoxysilane from silicon-copper contact mass and ethyl alcohol in the developed vibration reactor is presented. It is shown that the process of a direct alkoxysilane synthesis in the vibro-boiling layer is affected by a series of control parameters such as the ratio between the contact mass and the mass of grinding bodies, the grinding body sizes and their ratios in a polydisperse mixture, power density. Optimization of these parameters allowed us to obtain HSi(OEt)₃ with a selectivity of 50% at a silicon conversion of 90% without the use of promoters.

Full text of DOI:10.1007/s11172-019-2382-x

Russian Chemical Bulletin, International Edition, Vol. 68, No. 2, pp. 270—274, February, 2019
270
Mechanochemical method of producing triethoxysilane*
M. N. Temnikov,a A. A. Anisimov,a S. M. Chistovalov,a P. V. Zhemchugov,a
D. N. Kholodkov,a S. N. Zimovets,a,b Yu. S. Vysochinskaya,a and A. M. Muzafarova,b
aA. N. Nesmeyanov Institute of Organoelement Compounds, Russian Academy of Sciences,
28 ul. Vavilova, 119991 Moscow, Russian Federation.
Fax: +7 (499) 135 5085. E-mail: temnikov88@gmail.com, aziz@ispm.ru
bN. S. Enikolopov Institute of Synthetic Polymeric Materials, Russian Academy of Sciences,
70 ul. Profsoyuznaya, 117393 Moscow, Russian Federation.
Fax: +7 (495) 335 9000
A mechanochemical method for synthesis of triethoxysilane from silicon-copper contact
mass and ethyl alcohol in the developed vibration reactor is presented. It is shown that the
process of a direct alkoxysilane synthesis in the vibro-boiling layer is affected by a series of
control parameters such as the ratio between the contact mass and the mass of grinding bodies,
the grinding body sizes and their ratios in a polydisperse mixture, power density. Optimization
of these parameters allowed us to obtain HSi(OEt)3 with a selectivity of 50% at a silicon con-
version of 90% without the use of promoters.
Key words: triethoxysilane, direct synthesis, alkoxysilanes, mechanochemistry, mechano-
chemical activation, vibro-boiling layer.
Polyorganosiloxanes (POSs) represent one of the most
chlorosilanes, it was not aimed at preparation of the most
valuable derivatives. Therefore, chlorosilanes remain the
most widely used monomers in the modern industry.13—15
Of the available direct synthesis procedures, the most
promising is the passing of alcohol vapor through a layer
of silicon powder with a catalyst. It results in the formation
of HSi(OAlk)3 and Si(OAlk)4 as products. The reactor for
implementing this method is called a fixed-bed reac-
tor.16—18 This synthetic approach is characterized by high
efficiency and simplicity of instrumentation. High con-
version and selectivity in respect to HSi(OMe)3 can
be achieved when using silicon powder with grains of
<60 m19 or some CuCl precursors.18,20 For instance,
triethoxysilane was obtained recently21 with a high selec-
tivity (97%) at a silicon conversion of 64%, however, the
process duration was about 30 h for a 5 g silicon loading.
In our previous study16 with the use of a fixed-bed
reactor, the silicon conversion did not exceed 40%, and
the reaction proceeded only with MeOH as an alcohol and
was displaced towards formation of Si(OMe)4. The process
selectivity could be controlled by addition of MeCl. It was
also shown22 that double promoting using the HF/EtCl
pair can be applied to increase conversion and selectivity
with respect to hydrosilane. However, the use of MeCl,
EtCl and HF is unwanted since scaling of the process on
going from laboratory to industrial level requires the use
of large amounts of these toxic compounds.
demanded classes of organometallic polymers. Because of
their thermal stabilities, a wide range of operating tem-
peratures, hydrophobicity, permeability to gases, bioinert-
ness, and high dielectric characteristics, they have found
wide application both in everyday life and in high-tech
developments.
The demand for products of POSs is increasing every
year. However, their manufacturing is limited by a number
of economic and environmental factors. It is necessary to
note that the major part of organosilicon polymers is
produced by hydrolysis of organochlorosilanes. In addition
to the negative impact on the environment, the further
disadvantage of this approach consists in the need to carry
out this process in the same enterprise where the mono-
mers are produced. Otherwise the starting chlorosilane
should be transported, and this imposes certain risks be-
cause of instability of these compounds in the air.
The search for alternative raw materials for the synthesis
of POSs has been conducted for a long time. The use of
alkoxysilanes proved to be the most successful choice.1—11
In this regard, the development of methods for the syn-
thesis of alkoxysilanes is of great importance.
One of the most promising methods for production of
alkoxysilanes is the direct synthesis. This heterogeneous
process was discovered already in the middle of the last
century.12 However in contrast to the direct synthesis of
As a rule, the principal parameters for controlling the
process in a fixed-bed reactor are the temperature, con-
centration and type of a catalyst, the rate of introduc-
* Dedicated to Academician of the Russian Academy of Sciences
A. I. Konovalov on the occasion of his 85th birthday.
Published in Russian in Izvestiya Akademii Nauk. Seriya Khimicheskaya, No. 2, pp. 0270—0274, February, 2019.
1066-5285/19/6802-0270 © 2019 Springer Science+Business Media, Inc.

Method of producing triethoxysilane
Russ. Chem. Bull., Int. Ed., Vol. 68, No. 2, February, 2019
271
tion of an alcohol and the addition of a halogen-contain-
ing promoters.
halide (RCl). Pyrolysis of the latter results in an elimina-
tion of Cl and its interaction with Cu. Thus, Cu3Si is re-
generated, and the active reaction centers are formed.
A fundamentally new method for synthesis of alkoxysi-
lanes has been developed in the Institute of Organoelement
Compounds of Russian Academy of Sciences. It consists
in the interaction of metallic silicon with aliphatic alcohol
in the presence of a catalyst at 200—300 С in a vibro-
boiling layer of grinding bodies at a vibrational acceleration
Scheme 2
above 26.6 m s–2 23
.
To implement the proposed method, a reactor for the
production of alkoxysilanes from silicon and alcohols by
the direct synthesis was created. The reactor consists of
a working chamber, which is equipped with grinding bodies,
an electric heater, and process tubing and installed on
a vibrodrive. Both the working chamber and grinding bod-
ies are made of brass, a copper alloy.24 The efficiency of
this method was demonstrated for the synthesis of tetra-
ethoxy- and tetramethoxysilanes,25 which were obtained
in high yields without the use of high-temperature heat
carriers, as well as without a number of preliminary stages
of preparation. However, a selective preparation of trialk-
oxysilanes using this method was not reported so far. At
the same time, triethoxysilane is a practically important
product, which has found its use in the production of
semiconductor silicon and a series of organosilicon mono-
mers such as APTES, triethoxyvinylsilane, 3-(triethoxy-
silyl)propyl methacrylate, etc.
In the case of the direct synthesis of alkoxysilanes, the
copper regeneration does not take place. This determines
the formation of Cu0 clusters in the course of the process.
Copper in this form catalyzes the reaction (see Scheme 1,
equation 2).17 As can be seen from Scheme 1, produced
triethoxysilane interacts with ethanol. In addition, a grad-
ual decrease in the active copper content results in the
deactivation of the contact mass.22 It explains the fact that
introduction of little amounts of allyl chloride or chlorine-
containing metal salts increases the yield of HSi(OR)3.
These compounds make up the lack of Cl, which is required
for formation of CuCl in the system (see Scheme 2).
Carrying out the direct synthesis in a vibro-boiling
layer makes it possible to avoid the use of compounds of
this type. The dependences of the concentrations of reac-
tion products and ethanol on the reaction time for
a typical experiment in a vibration reactor are depicted in
Fig. 1.
In this regard, the purpose of this work is to study and
optimize the factors affecting the yield of thriethoxysilane
in the reaction of silicon-copper contact mass with ethanol
in a vibro-boiling layer.
As follows from the data presented, the first stage of
the reaction is characterized by a low ethanol conversion
and a low concentration of triethoxysilane, i.e., there is
an induction period (see Fig. 1, period I). Apparently, it
is associated with the formation of active centers and the
increase in the specific surface of the contact mass in the
grinding process. During the induction period, a little
Results and Discussion
The reactions shown in Scheme 1 occur during the
interaction of ethanol with silicon in the presence of
copper.
C (%)
100
Scheme 1
I
II
III
IV
80
60
40
20
0
(1)
1
2
(2)
3
4
R = Me, Et
In this case, copper is in an active state, most likely, in
the form of the intermetallic phase Cu3Si. It is assumed
that this inermetallic compound participates in the Cl atom
transfer from copper to silicon (Scheme 2), producing
active centers —(SiCln)—.26,27 The main chlorine source
in the direct synthesis of organochlorosilanes is an organic
50 100 150 200 250 300 350 400 t/min
Fig. 1. Dependences of concentrations of starting ethanol (1),
HSi(OEt)3 (2), Si(OEt)4 (3) and condensation products (4) on
the reaction time (see Table 1, run 2).

272
Russ. Chem. Bull., Int. Ed., Vol. 68, No. 2, February, 2019
Temnikov et al.
amount of Si(OEt)4 is formed. Evidently, it is due to the
presence of Cu0 which catalyzes conversion of HSi(OEt)3
into Si(OEt)4 (see Scheme 1, equation 2), in the contact
mass at the beginning of the synthesis, The peak of the
reaction falls on period II (see Fig. 1), for which the
maximum concentration of HSi(OR)3 is observed, indicat-
ing the minimum content of inactive copper during this
period. At stages II and III, the ethanol content in the
distilled mixture approaches 0%. However, tetraethoxysi-
lane becomes a predominant product over time. It is be-
cause of accumulation of deactivated Cu0. In period IV,
the reaction stops as it follows from the increase in the
ethanol concentration to 100%. The concentration of
condensation products remains invariable during the
whole process since their formation, obviously, does not
depend on the contact mass activity in different periods
of the reaction.
Thus, the developed method for preparation of alk-
oxysilanes ensures high efficiency of the direct synthesis
with ethanol, as evidenced by practically complete conver-
sion of the alcohol at stages II and III of the process. It
should be emphasized that under the used conditions,
HSi(OEt)3 is formed without addition of alkyl chlorides
or HF during the process.
Performed trial evaluative experiments showed that the
process of the mechanochemical synthesis of alkoxysilane
can be controlled through a series of parameters, which
determine rate, efficiency and completeness of transforma-
tions in the course of mechanoactivation. The following
parameters can be attributed to the control parameters:
the volume ratio of silicon and grinding bodies, total mass
of the grinding bodies, dispersion of the grinding bodies,
the ratios between the grinding bodies of different sizes in
their polydisperse mixture, and the supplied energy, which
is associated with the acceleration created at different
frequencies and amplitudes of vibrations.
Experiments showed (Table 1, runs 1—4) that an in-
crease in contact mass loading (i.e., the change in the
ratio between volumes of the silicon under the treatment
and grinding bodies) from 12 to 60 g did not resulted in
a significant reduction in the silicon conversion and se-
lectivity with respect to triethoxysilane. The experiment
duration increased in proportion to the load and amounted
to 12 h. A further increase in the mass of the initial com-
ponents to 120 g led to the reduction in the silicon conver-
sion to 40% and selectivity to 17%.
In the course of further optimization carried out by
varying the control parameters listed above, it was found
that the selectivity of the process and the silicon conversion
are affected by the dispersion of grinding bodies (see Table 1,
runs 1 and 5). As seen from the data of Table 1, the selec-
tivity with respect to HSi(OEt)3 increased from 36 to 54%
when 1000 balls of 5 mm in diameter (see Table 1, run 5)
were used instead of two larger grinding balls of 40 mm in
diameter (run 1). Simultaneously, silicon conversion also
increased. Perhaps, this effect is related to an increase in
the number of points of contact (the number of contacts)
between grinding bodies. Since mechanoactivation takes
place at these points, its efficiency rises with the increase
in their number.
The best result was obtained by the increase in the
vibration acceleration (see Table 1, run 6). In this case,
almost complete conversion of silicon was achieved while
maintaining the same high selectivity with respect to
triethoxysilane. The reaction time was reduced to 4 h. The
possible reason of this effect is that the increase in ac-
celeration determines the larger number of collisions be-
tween grinding bodies per unit time.
The proposed by us mechanochemical method for the
direct synthesis of alkoxysilanes enlarges the number of
control parameters affecting the process rate, silicon con-
version, and selectivity with respect to HSi(OEt)3. For
a full assessment of the influence of the control parameters on
the nature and completeness of the process, it is necessary to
know what are the characteristics, including dispersion, of
silicon grains, with which alcohol preferably reacts at any
time during the process. In future, we plan to study kinetics
of dry (without an addition of alcohol) grinding, i.e.,
mechanoactivation, to obtain granulometric composition of
mechanically activated silicon with an interval of 5—10 min-
utes in a wide amplitude-frequency range at different ratios
of masses of silicon and grinding bodies with the use of
mono- and polydisperse grinding bodies at different de-
grees of filling the working chamber. This will establish
Table 1. Effects of control parameters on the production of triethoxysilane at Т = 250 С
Run
Vibrational
acceleration,
g
Loading
Si/CuCl
/g
Diameters and amounts of
grinding bodies*,
mm (pcs.)
Conversion
of Si (%)
Selectivity
to HSi(OEt)3
/Si(OEt)4 (%)
Time of
reaction
/h
1
2
3
4
5
6
7.5
7.5
7.5
7.5
7.5
19
10/2
20/4
10(500)+40(2)
10(500)+40(2)
10(500)+40(2)
10(500)+40(2)
5(1000)+10(500)
5(1000)+10(500)
65
63
50
40
77
90
36/64
38/62
41/59
17/83
54/46
50/50
15
16
12
17
15
14
50/10
100/20
10/2
10/2

Method of producing triethoxysilane
Russ. Chem. Bull., Int. Ed., Vol. 68, No. 2, February, 2019
273
where n is the amount of the reaction products (mol), nSi is the
amount of used silicon (mol), nSi-prod is the amount of the silicon-
containing product (mol).
a causal relationship between the control parameters and
the final result of the mechanochemical synthesis.
Thus, we proposed a mechanochemical method for
realization of the direct synthesis of alkoxysilanes in
a vibro-boiling layer. It has been shown that the process
of the direct synthesis of alkoxysilanes is affected by
a number of control parameters such as the ratio between
the contact mass and the mass of grinding bodies, the sizes
of grinding bodies and their ratios in their polydisperse
mixtures, and energy supply. The performed optimization
of these parameters made it possible to obtain HSi(OEt)3
with the 50% selectivity at the 90% silicon conversion
without the use of promoters.
This work was financially supported by the Russian
Science Foundation (Project No. 18-73-10153). The design
and assembly of the reactor with the vibro-boiling layer
were carried out under the financial support of the Russian
Science Foundation (Project No. 14-23-00231).
References
1. R. Wakabayashi, K. Kuroda, ChemPlusChem, 2013, 78, 764.
2. M. A. Brook, Chem. — A Eur. J., 2018, 8458.
3. M. A. Brook, J. B. Grande, F. Ganachaud, Adv. Polymer Sci.,
2010, 235, 161—183.
Experimental
4. A. A. Kalinina, D. N. Kholodkov, I. V. Meshkov, M. A.
Pigaleva, I. V. Elmanovich, Yu. A. Molodtsova, M. O. Gallya-
mov, A. M. Muzafarov, Russ. Chem. Bull., 2016, 65, 1104.
5. A. A. Kalinina, E. V. Talalaeva, A. I. Demchenko, N. G.
Vasilenko, Yu. A. Molodtsova, N. V. Demchenko, A. M.
Muzafarov, Russ. Chem. Bull., 2016, 65, 1013.
6. S. A. Milenin, A. A. Kalinina, V. V. Gorodov, N. G. Vasi-
lenko, M. I. Buzin, A. M. Muzafarov, Russ. Chem. Bull., 2015,
64, 2498.
7. A. Kalinina, N. Strizhiver, N. Vasilenko, N. Perov, N. De-
mchenko, A. Muzafarov, Silicon, 2014, 7, 95.
8. M. A. Soldatov, N. A. Sheremeteva, A. A. Kalinina, N. V.
Demchenko, O. A. Serenko, A. M. Muzafarov, Russ. Chem.
Bull., 2014, 63, 267.
9. M. A. Obrezkova, A. A. Kalinina, I. V. Pavlichenko, N. G.
Vasilenko, M. V. Mironova, A. V. Semakov, V. G. Kulichikhin,
M. I. Buzin, A. M. Muzafarov, Silicon, 2014, 7, 177.
10. S. A. Milenin, A. A. Kalinina, N. V. Demchenko, N. G.
Vasilenko, A. M. Muzafarov, Russ. Chem. Bull., 2013, 62, 705.
11. E. V. Egorova, N. G. Vasilenko, N. V. Demchenko, E. A.
Tatarinova, A. M. Muzafarov, Dokl. Chem., 2009, 424, 15.
12. E. G. Rochow, W. E. Newton, Inorg. Chem., 1970, 9, 1071.
13. W. Noll, Chemistry and Technology of Silicones, Academic
Press, New York, 1968.
14. K. A. Andrianov, L. M. Khananashvili, Technologiya elemento-
organocheskih monomerov i polymerov [Technology of Organo-
metallic Monomers and Polymers], Khimiya, Moscow, 1973,
400 pp. (in Russian).
15. E. G. Rochow, Mir kremniya [Silicon World], Khimiya,
Moscow, 1990, 147 (in Russian).
16. M. N. Temnikov, A. S. Zhiltsov, V. M. Kotov, I. V. Krylova,
M. P. Egorov, A. M. Muzafarov, Silicon, 2015, 7, 69.
17. M. Okamoto, H. Abe, Y. Kusama, E. Suzuki, Y. Ono,
J. Organomet. Chem., 2000, 616, 74.
Triethoxysilane (97%) and tetraethoxysilane (>97%) were
employed as standards for GLC. These compounds and СuCl
(«Acros», >97%) were used without additional purification.
Silicon KR-1 (98%; impurities: 0.7% Fe, 0.7% Al, 0.6% Ca;
characteristic size is 140 m) was subjected to a preliminary
treatment with an aqueous solution of HF to remove the oxide
layer. Ethanol was dried before the use over CaH2 and kept
over 3 Å molecular sieves; the final water content was less than
200 ppm.
The reactor for the synthesis of alkoxysilanes24,28 consisted
of a cylindrical working chamber installed on a vibrodrive.
Hermetic working chamber of 1l in volume was made of pipe
brass L59. It was filled by 70% with spherical grinding brass bod-
ies of 5, 10 and 40 mm in diameter taken in different ratios with
total mass of 2850 g. The working chamber was equipped with
a heater, thermocouple, inlet and outlet pipes.
Product analysis. Samples were analyzed on a Crystal Lux-
4000M chromatograph equipped with a catarometer at 50—280 С,
20 degree min–1, column (2 mm2 m) with 5% SE-30 on
Chromaton-N-AW-HMDS as a stationary phase. Helium
(30 ml min–1) was used as a carrier gas. The data were processed
using the NetChrom 2.0 program.
NMR spectra were recorded using BrukerAvance III HD 400
NMR spectrometer (Germany), operating frequencies are
400 and 79 MHz for 1H and 29Si nuclei, respectively. Tetra-
methylsilane was used as an internal standard, CDCl3 was used
as a solvent. Spectra were processed with the use of the Mestre-
nova program.
Experiment conducting. A specified amount of the contact
mass prepared by the reaction between Si and CuCl at 300 С
was loaded in the working chamber. The reactor was heated to
250 С. Then vibrodrive was turned on, and ethanol was supplied
into the chamber by a high pressure pump at a rate within
0.4—0.7 ml min–1 at a given vibrational acceleration in the range
from 7.5 to 19g. The formed products of the reaction between
silicon and ethanol were passed through a condenser to a re-
ceiver. Sampling and analysis of the samples of volatile products
were carried out every 30 min. Conversion (K) and selectivity (S)
were calculated by the following formulas:
18. L. Zhang, J. Li, K. Yang, C. Hu, S. Ge, C. Yang, Adv. Mater.
Res., 2011, 233—235, 1534.
19. E. Suzuki, Y. Ono, Chem. Lett., 1990, 19, 47.
20. G. J. Wang, F. X. Zhang, G. Y. Liu, X. N. Liu, Adv. Mater.
Res., 2012, 455—456, 80.
21. F. Chigondo, B. Zeelie, P. Watts, ACS Sustain. Chem. Eng.,
2016, 4, 6237.
22. Z. Lei, H. Sue, Y. Chunhui, L. Ji, Y. Kai, H. Chenfa,
G. Shibin, Appl. Organomet. Chem., 2011, 25, 508.
23. RF Pat. 2628299C1, 2017.
KSi (%) = (n/nSi)•100,
S (%) = (nSi-prod/nSi)•100,

274
Russ. Chem. Bull., Int. Ed., Vol. 68, No. 2, February, 2019
Temnikov et al.
24. RF Pat. 2671732C1, 2018.
28. S. M. Tchistovalov, V. M. Kotov, A. A. Anisimov, M. N.
Temnikov, P. V. Zhemchugov, A. M. Muzafarov, Khimiches-
koe i neftegasovoe mashinostroenie [Chemical and Oil and Gas
eEgineering], 2018, 10, 3 (in Russian).
25. M. Temnikov, A. Anisimov, P. Zhemchugov, D. Kholodkov,
A. S. Goloveshkin, A. Naumkin, S. Chistovalov, D. E.
Katsoulis, A. Muzafarov, Green Chem., 2018, 20, 1962.
26. A. I. Gorbunov, A. P. Belyi, G. G. Filippov, Chem. Rev.,
1974, 43, 676.
Received December 7, 2018;
in revised form December 24, 2018;
accepted December 26, 2018
27. R. A. Turetskaya, K. A. Andrianov, I. V. Trofimova, E. A.
Chernyshev, Chem. Rev., 1975, 44, 444.