4978

Organometallics 2001, 20, 4978-4992

Cover Essay

Organometallics 2001.20:4978-4992.
Downloaded from pubs.acs.org by 132.247.249.242 on 08/31/15. For personal use only.

Dimethyldichlorosilane and the Direct Synthesis of

Methylchlorosilanes. The Key to the Silicones Industry
When on May 10, 1940, in the Research Laboratory
of the General Electric Company in Schenectady, NY,
Eugene G. Rochow carried out an experiment in which
he passed gaseous methyl chloride through a crushed
50% Cu-Si mixture (previously activated with gaseous
HCl) in a tube furnace at 370 C and obtained as
products a mixture of methylchlorosilanes (among them
(CH3)2SiCl2, our cover molecule), he planted the proverbial acorn from which grew the proverbial mighty
oak, the modern silicones industry.
Before we consider this important reaction and its farreaching consequences in detail, it is useful to go back
in time, more than 175 years, to provide the necessary
introductionsthe work of earlier chemists that led,
finally, to this reaction which we can write in oversimplified form in eq 1.

2CH3Cl(g) + Si/Cu(s) f (CH3)2SiCl2(l)

(1)

The reaction requires elemental silicon, and this is

where our story begins. The most abundant elements
in the earths crust are oxygen (49.5%) and silicon
(25.7%). However, there is no free silicon in the earths
crust: it is all very firmly bound to oxygen in the form
of silica and metal silicates. Because of the high Si-O
bond strength of 108 kcal/mol, it is not easy to force
silicon out of its oxygen connection and for this reason,
despite its high abundance, elemental silicon is a relative newcomer to the chemical scene.1 It was not until
1824 that Jons Jakob Berzelius prepared silicon as an
amorphous brown solid.2 Friedrich Wohler was a student of Berzelius when this work was carried out (Nov
1823-Sept 1824) and in later years reported on that
time:3
Dedicated to Eugene G. Rochow on the occasion of his 92nd
birthday, October 4, 2001, with respect, admiration, and affection.
(1) A detailed, fully referenced account of the early attempts to
isolate elemental silicon and of the various successful preparations,
starting with that of Berzelius in 1824, can be found in: Gmelin
Handbook of Inorganic Chemistry, 8th ed.; Springer-Verlag: New York,
1984; Si, Silicon, Part A1, History, pp 7-50.
(2) Berzelius, J. J. Ann. Phys. Chem. [2] 1824, 1, 169. (Berzelius
(1779-1848), 50 years a professor at Stockholm. He originated the
present day symbols of the elements and their use in the formulas of
compounds. Pioneer in the determination of atomic weights. Proposer
and champion of a dualistic or electrochemical theory and of the
radical theory. Discovered ceria, selenium, silicon, and thorium. His
textbook and annual reports on the progress of chemistry (Jahresberichte) were very influential.)
(3) Wohler, F. Ann. 1856, 97, 266. (Wohler (1800-1882), student of
Gmelin and Berzelius, professor at Gottingen. Remembered today
mainly for his conversion of ammonium cyanate to urea in 1828, which
demolished the theory of a vital force responsible for the formation
of organic compounds and ended the strict separation between
inorganic and organic chemistry. Isolated aluminum, beryllium,
calcium carbide, siloxene from calcium silicide, trichlorosilane and
other inorganic silicon compounds, impure titanium, organic compounds such as hydroquinone. With Liebig, proposed the concept of
the benzoyl radical from which grew the short-lived radical theory.)

Silicon, without doubt, is one of the most

remarkable elements of our planet because it is
one of the main substances which has served in
its formation. It, therefore, is well worth the
effort of determining its properties as completely
as possible. As known, it first was prepared and
isolated by Berzelius in 1824 by the decomposition of gaseous fluorosilicic acid or of potassium
fluorosilicate by potassium. [The latter reaction
is: K2SiF6 + 4K f Si + 6KF.]
It was my good fortune to be his student at
this time when he was engaged in these instructive investigations and to help him by preparing
the required potassium...Berzelius examined and
described all of the characteristic properties of
this silicon with his usual keenness and precision. However, he obtained it only in amorphous
form, as a dull, brown powder. He commented
repeatedly how interesting it would be to become
acquainted with this material in a dense and
crystalline state.4
An 1846 chemistry text on my bookshelf5 describes
the Berzelius preparation of amorphous silicon:
To prepare silicon, (there) is selected...the
double fluoride of silicon and potassium (2SiF3
+ 2KF), which is a white powder...; a quantity
of this substance is to be mixed with nearly its
own weight of potassium, cut into little bits, and
placed in an iron cylinder, or in a tube of hard
glass, which may be held over the flame of a
spirit-lamp. As soon as the bottom of the tube
has been heated to redness, vivid ignition occurs
by the decomposition, which spreads with little
need of external heat, throughout the entire
mass; when cool, the residual brown matter is
to be washed carefully with water: fluoride of
potassium dissolves, and the silicon remains
behind. The silicon so obtained is a dull brown
powder, which, when heated in air or in oxygen,
takes fire and burns, forming silicic acid
Today this reaction would be called a self-propagating, high-temperature synthesis.
It was only in 1854 that crystalline silicon was first
reported by Henri-E
tienne Sainte-Claire Deville, who
(4) This amorphous silicon should not be confused with todays
amorphous hydrogenated silicon. Actually, the brown amorphous
silicon prepared by the procedure of Berzelius, which visually appears
amorphous, even in the microscope under 1000 magnification, is not
amorphous (or not completely so), since X-ray diagrams showed a
diamond lattice: Manchot, W. Z. Anorg. Allg. Chem. 1922, 124, 333;
1922, 120, 277.
(5) Kane, R. Elements of Chemistry; American edition by Draper,
J. W.; Harper and Brothers: New York, 1846; p 321.

10.1021/om0109051 CCC: $20.00 2001 American Chemical Society

Publication on Web 11/19/2001

Organometallics, Vol. 20, No. 24, 2001 4979

isolated it from aluminum melts in which it was present

as an impurity (sometimes up to 10%) by treating the
melt with hot hydrochloric acid.6 The first report of the
reduction of silica by carbon at high temperatures was
that of Henri Moissan,7 and it is this process (eq 2),

Organometallics 2001.20:4978-4992.
Downloaded from pubs.acs.org by 132.247.249.242 on 08/31/15. For personal use only.

SiO2 + 2C f Si + 2CO

(2)

carried out in an electric furnace using graphite electrodes at 3000 C, which has served, since the beginning
of the last century, for the large-scale production of
silicon.8 When pure, white quartzite rock and a pure
form of carbon are used, material of 98% purity is
obtained. Crystalline silicon is a covalent, nonmetallic
solid with a diamond lattice, density 2.33 g/cm3, and mp
1414 C. The commercial silicon has a shiny, blue-gray,
metallic appearance. As Wohler had noted,3 silicon
without doubt is one of the most remarkable elements.
In addition to its long-established applications as a
component in useful ferrous alloys and in aluminum and
magnesium alloys, there are the more modern ones,
based on ultrapure siliconsin the preparation of semiconductors (the chip revolution) and the fabrication
of solar cells. Now there are even are light-emitting
forms of silicon (porous silicon). However, we are
concerned here with elemental silicon as a reactant, and
there are some reactions of elemental silicon worthy of
note, as well as some chemistry of silicon compounds,
which require discussion before we return to eq 1.
The first chlorosilane, SiCl4, was obtained by Berzelius in 1824 when he found that his amorphous silicon
ignited when heated in a stream of chlorine and was
completely consumed. More relevant to the subject of
this essay is Buff and Wohlers reaction of crystalline
silicon (for whose preparation Wohler had developed an
improved procedure3,9) with anhydrous, gaseous hydrogen chloride.10 This reaction was carried out by passing
the HCl through a long glass tube in which the silicon
was spread out throughout its length and which was
surrounded by glowing coals (no tube furnaces with
temperature controllers in those days). A U-tube connected to the exit of the reaction tube, cooled with an
ice-salt mixture, served to condense the volatile products. Any volatiles not trapped in the U-tube were
passed into a large volume of water, in which a large
amount of white solid was formed during the reaction.
The condensed liquid usually was turbid and appeared
to be a mixture of several products. Temperature
control, difficult under these circumstances, was important. (At red heat, SiCl4 was by far the major product.)
Distillation of the contents of the U-trap gave as the
major product a colorless liquid boiling at 40-43 C
which had an irritating smell and fumed strongly in air.
Its vapors were found to ignite as readily as those of
(6) Sainte-Claire Deville, H. E. Compt. rend. acad. sci. 1854, 39, 321.
Curiously, however, it appears that Wohler was the first to prepare
silicon in observable crystalline form. In a letter written in 1843 to
his friend Justus Liebig, he reported that when a mixture of H2 and
SiCl4 vapors was brought to red heat, silicon was produced in the form
of black crystals. But, alas, Wohler apparently never published this
observation: Hofmann, A. W., Ed. Justus Liebig und Friedrich Wo hlers
Briefwechsel in den Jahren 1829/73; Braunschweig, Germany, 1888;
Vol. 1, p 230.
(7) Moissan, H. Bull. Soc. Chim. Fr. 1895, 13, 972.
(8) Dosaj, V. In Kirk-Othmer Encyclopedia of Chemical Technology,
4th ed.; Wiley: New York, 1997; pp 1104-1108.
(9) Wohler, F. Compt. rend. acad. sci. 1856, 42, 48.
(10) Buff, H.; Wohler, F. Ann. 1857, 104, 94.

ether, burning with a weak, luminous, green flame and

releasing a smoke of silica and HCl. Pyrolysis of a
sample of the distillate in a glowing glass tube gave
brown amorphous silicon (a reaction used about one
hundred years later, with pure material and under more
controlled conditions, to produce highly pure, crystalline
silicon). Hydrolysis of the distillate proceeded vigorously
and exothermically, giving HCl and a solid silicon oxide
different in appearance from SiO2. Analysis of the
distillate proved to be difficult because of its volatility,
poor thermal stability, and great moisture sensitivity.
The product, of course, was trichlorosilane, HSiCl3, a
compound of technological importance today in the
silicones industry and in silicon-based material science.
What Wohler called fractional distillation was a simple
one-plate distillation, and it was not sufficient to effect
good separation from HSiCl3 of the SiCl4 also produced.
Also, the cold trap was not very efficient, as indicated
by the formation of a large amount of white solid when
the exit gas was passed into water. The elemental
analyses of the distillate consequently were high in Cl
and low in Si. Wohler wrote the formula [Si2Cl3 + 2 HCl]
for the HCl + Si reaction product (the atomic weight of
Si was 14, not 28, in those days). A few years later, for
reasons not germane to this discussion, Wohler doubled
the formula to Si6Cl10H4 but admitted that, because of
the problem of its purity, the constitution of the HCl +
Si product remained unsettled. Wohler was taken to
task by Friedel and Ladenburg 10 years later.11 How
could he, they asked, assign a formula of Si6Cl10H4 to a
compound of reported bp 42 C when SiCl4 boiled at 59
C? On the basis of the tetraatomicity of silicon, the
expected formula was SiCl3H, and Friedel and Ladenburg proceeded to prove this by very careful isolation
and characterization studies.
Relevant to the main subject of this essay is that later
workers found that some metal silicides (those of Mg,
Fe, V, and Cu) also reacted with gaseous HCl to produce
HSiCl3. The use of a commercial copper silicide (20 parts
of Cu to 100 of Si) by Combes is noteworthy.12 A reaction
temperature of 300 C gave HSiCl3 in high yield; higher
temperatures favored formation of SiCl4. Later work by
Ruff and Albert showed copper silicide to be the most
effective of the metal silicides examined.13
We shall return to reactions of elemental silicon, but
it is instructive to consider first the birth and early
development of organosilicon chemistry. The first known
organosilicon compound, tetraethylsilane, was prepared
by two chemists whose names one finds in all organic
textbooks for another important discovery, the aluminum chloride catalyzed acylation of aromatic hydrocarbons, the Friedel-Crafts reaction. Inspired by the
syntheses of organotin, -mercury, -arsenic, etc. compounds by Frankland and others in prior years,14
Charles Friedel and his American co-worker, James
Mason Crafts, began research whose goal was the
development of organosilicon chemistry along the lines
used by these earlier workers who had opened up
organometallic chemistry. A reaction of diethylzinc with
SiCl4 in a sealed tube, Friedel and Crafts found,
(11) Friedel, C.; Ladenburg, A. Ann. 1867, 143, 118.
(12) Combes, C. Compt. rend. acad. sci. 1896, 122, 531.
(13) Ruff, O.; Albert, K. Ber. Dtsch. Chem. Ges. 1905, 38, 2222.
(14) See our last cover molecule essay: Seyferth, D. Organometallics
2001, 20, 2940.

4980

Organometallics, Vol. 20, No. 24, 2001

commenced at around 140 C and was complete after 3

h at 160 C. The evolution of a significant amount of
gas when the tube was opened and the presence of
metallic zinc in the ZnCl2 formed in the reaction (eq 3)

Organometallics 2001.20:4978-4992.
Downloaded from pubs.acs.org by 132.247.249.242 on 08/31/15. For personal use only.

2(C2H5)2Zn + SiCl4 f (C2H5)4Si + 2ZnCl2

(3)

indicated that some diethylzinc had decomposed at the

high temperatures used. The liquid products included
a hydrocarbon (probably n-butane), unreacted SiCl4, and
a liquid with bp 152-154 C, which elemental analysis
and vapor density measurements showed to be the
desired tetraethylsilane.15 The preparation of tetramethylsilane was much more difficult. First of all,
preparation of (CH3)2Zn by Franklands high-temperature, no-solvent procedure14 gave poor yields. Then the
(CH3)2Zn/SiCl4 (no solvent) reaction required longer
heating of the sealed tube at 200 C. The product was a
volatile liquid, bp 30-31 C; its C, H analysis and vapor
density were in agreement with its identity as
(CH3)4Si.16,17 Also reported was a study that demonstrated the much greater kinetic stability of the Si-C
bond in tetraethylsilane compared to the Sn-C bond
in tetraethyltin with respect to cleavage by Cl2.18
Friedel carried out further research on inorganic and
organosilicon chemistry with Albert Ladenburg, while
Crafts focused on organic chemistry (the Friedel-Crafts
reaction was discovered in 1877). Of interest was the
synthesis of the first hexaorganodisilane, (C2H5)3SiSi(C2H5)3, by the reaction of diethylzinc with hexaiododisilane.19 Ladenburg continued research on organosilicon chemistry when he left Paris to take an academic
position in Germany. His studies of the ethylation of
tetraethoxysilane were most interesting and are relevant to the subject of this essay.20 The reaction of
diethylzinc with tetraethoxysilane was sluggish, even
when the reaction mixture was heated, and did not go
to completion. A rapid reaction was observed in the
presence of a stoichiometric quantity of sodium, no
doubt due to the formation of the much more reactive
NaZn(C2H5)3.21 Some heating still was required, and
Ladenburg makes the point that the round-bottomed
reaction flask should be warmed with a flame-heated
(15) Friedel, C.; Crafts, J. M. Ann. 1863, 127, 28. (Friedel (18321899), pupil of Wurtz, his successor at the Sorbonne in 1884. Research
on ketones, pinacone, lactic acid; synthesis of glycerol, Friedel-Crafts
reaction 1877, silicon alkoxides, established dimeric formulas of AlCl3,
GaCl3, and FeCl3; high temperature and pressure synthesis of minerals. J. M. Crafts (1839-1917) studied at Harvard; in Europe with
Plattner, Bunsen and Wurtz. Afterwards, academic positions at Cornell
and MIT. In Paris 1874-91, research on organic and organosilicon
chemistry with Friedel. Professor of Organic Chemistry at MIT 189297; President of MIT 1897-1900.)
(16) (a) Friedel, C.; Crafts, J. M. Ann. 1865, 136, 203. (b) Friedel,
C.; Crafts, J. M. Bull. Soc. Chim. Fr. 1865, 3, 356.
(17) References 15 and 16 are preliminary communications. A long,
detailed full paper that includes these preparations as well as an
account of the reactivity of these R4Si compounds, including the
preparation of the first siloxane, (C2H5)3SiOSi(C2H5)3, appeared a few
years later: Friedel, C.; Crafts, J. M. Ann. Chim. Phys. 1870, 19, 334.
(18) Friedel, C.; Crafts, J. M. Ann. 1866, 137, 19.
(19) (a) Friedel, C.; Ladenburg, A. Compt. rend. acad. sci. 1869, 68,
920; (b) Friedel, C.; Landenburg, A. Justus Liebigs Ann. Chem. 1880,
203, 241; Ann. Chim. Phys. 1880 [5], 19, 390.
(20) Ladenburg, A. Ann. 1872, 164, 300. (Ladenburg (1842-1911)
studied with Bunsen, Kirchhoff, Carius, and Kekule and, in Paris, with
Wurtz and Friedel. Privatdozent, then associate professor in Heidelberg, professor in Kiel (1872) and Breslau (1880-1909). Research on
organosilicon and -tin compounds; benzene and its derivatives. His
most important work was on alkaloids.)
(21) See the discussion of this compound in the previous essay.14

asbestos bath (forerunner of our electric heating

mantles), since direct heating of the flask, in which large
quantities of metallic zinc had precipitated, with an
open flame tended to cause cracking of the flask. Since
it contained sodium, diethylzinc, and NaZn(C2H5)3, the
result must have been spectacular. Fractional distillation of the product mixture was impossible: Si(OC2H5)4,
bp 166.5 C; C2H5Si(OC2H5)3, bp 159 C; (C2H5)2Si(OC2H5)2, bp 155.5 C; (C2H5)3SiOC2H5, bp 153 C.
However, since the ethylation proceeded in discrete
steps, it was possible to adjust reaction conditions so
that one of them was the major product. Thus, all could
be isolated, but this required large-scale reactions. That
the (C2H5)2Zn/Na/Si(OC2H5)4 reaction proceeded by way
of a substituent exchange process was shown by using
Si(OCH3)4 in place of tetraethoxysilane: the product
was C2H5Si(OCH3)3.22a,b Among the reactions Ladenburg tried with his products was that of (C2H5)2Si(OC2H5)2 with acetyl and with benzoyl chloride. The
organosilicon product was (C2H5)2SiCl2 (analytically
pure!). On treatment of the latter with water, HCl was
formed as well as a viscous, almost odorless, Cl-free
syrup. The same product was obtained by the action of
aqueous HI on (C2H5)2Si(OC2H5)2 (eq 4). Its analysis

n(C2H5)2Si(OC2H5)2 + 2nHI f
[(C2H5)2SiO]n + 2nC2H5I + nH2O (4)
agreed with the empirical formula (C2H5)2SiO, and
Ladenburg called it siliciumdiathyloxid. In fact, this
product had been obtained in 1866 by Friedel and Crafts
by the oxidation of tetraethylsilane.23 Therefore, not
only had Friedel and Crafts prepared the first organosilicon compound, but they had also prepared the first
polysiloxane! Ladenburg had provided a practical synthesis. Had Ladenburg carried out similar experiments
with dimethylzinc, (CH3)2SiCl2, our cover molecule,
would have been known in 1872. Ladenburg found
(C2H5)2SiO to be very thermally stable and very high
boiling (leaving an analytically pure residue at 330 C);
it did not solidify at -15 C. Ladenburg noted the formal
similarity to diethyl ketone but commented on the great
differences in properties. Attempts to hydrogenate
(C2H5)2SiO to obtaine a silicocarbinol were unsuccessful.
Ladenburg also prepared the first silicone resin,
[C2H5SiO1.5], which he called silicopropionic acid and
wrote as C2H5SiOOH, by hydrolysis of C2H5SiCl3 which
he had prepared in a similar manner from C2H5Si(OC2H5)3. The disiloxane was prepared by acid hydrolysis of (C2H5)3SiCl; treatment of the latter with aqueous
ammonia at room temperature gave the silanol, (C2H5)3SiOH.
Ladenburg did indeed carry out a reaction of dimethylzinc with Si(OC2H5)4.22a,c To do so, he developed a
much better procedure for the preparation of dimethylzinc: reaction of methyl iodide (120 parts) with zinc
filings (90 parts) in the presence of 1% sodium amalgam
(100 parts) and a few drops of ethyl acetate at atmospheric pressure at a temperature up to 90 C; the
(22) (a) Ladenburg, A. Justus Liebigs Ann. Chem. 1874, 173, 143.
(b) Ber. Dtsch. Chem. Ges. 1872, 5, 1081. (c) Ber. Dtsch. Chem. Ges.
1873, 6, 1029.
(23) Friedel, C.; Crafts, J. M. Ann. Chim. Phys. 1866 [4], 9, 5.

Organometallics 2001.20:4978-4992.
Downloaded from pubs.acs.org by 132.247.249.242 on 08/31/15. For personal use only.

Organometallics, Vol. 20, No. 24, 2001 4981

Figure 1. Frederic Stanley Kipping (reproduced courtesy

of the Library and Information Centre, Royal Society of
Chemistry).

dimethylzinc yield was nearly quantitative. During six

(CH3)2Zn/Si(OC2H5)4 reactions, carried out in sealed
tubes at temperatures up to 300 C, the tubes exploded
in four of the experiments. From the contents of the two
remaining tubes, CH3Si(OC2H5)3 could be isolated.
Treatment of this product with aqueous HI gave
(CH3SiO1.5)n, which Ladenburg, believing it was CH3SiOOH, called methylsiliconsaure. The disubstituted
product, (CH3)2Si(OC2H5)2, was not obtained.
Ladenburg reported another first: the preparation of
the first trialkylsilane, (C2H5)3SiH, as a byproduct
(together with ethylene) of the ethylation of (C2H5)3SiOC2H5 with the (C2H5)2Zn/Na system.20 Its reactivity
was in line with its formulation as a silicon hydride: it
was attacked rapidly by bromine and concentrated
sulfuric acid and explosively by fuming nitric acid.
These then were the beginnings of organosilicon
chemistry. Toward the end of the 19th century and at
the beginning of the 20th, other researchers became
active in this area: there were papers by Pape, Polis,
Dilthey, and Schlenk, while Ladenburg continued research in this area. However, it was Frederic Stanley
Kipping (Figure 1) (1863-1949) who dominated organosilicon research, from his first communication in 1899
until he retired in 1936, publishing over 50 papers.
Kipping was born in England in 1863, the year that the
first organosilicon compound was reported. He joined
von Baeyers laboratory in Munich in 1886, where he
engaged in research on cyclic carbon compounds as the
first student of W. H. Perkin, Jr. (the son of the famous
English organic chemist), who was an Assistent. After

obtaining his Ph.D., Kipping returned to England,

obtaining a D.Sc. from London University in 1887. He
joined Perkin, now in Edinburgh, to begin his academic
career, working on projects in organic chemistry. In
1897 he was appointed to the chair in organic chemistry
in University College, Nottingham, where he stayed
until his retirement. It was at Nottingham that he
started organosilicon research, which became his main
area of interest. However, other projects in organic
chemistry were continued. One of Kippings primary
interests was the preparation of a silicon compound of
the type SiRRRR and its resolution into its optically
active d and l isomers. Also, he was interested in seeing
how similar the organic derivatives of silicon were to
the analogous compounds of carbon. (Of course, Friedel
and Ladenburg had already observed some differences.)
At the time Kipping began his research in organosilicon chemistry at Nottingham, the zinc alkyls of
Frankland14 still were the only practical sources of
nucleophilic alkyl groups. Soon thereafter things
changed: in 1900, Victor Grignard published his first
paper on the RMgX reagents, which came to bear his
name.24 The Grignard reagents found rapid acceptance
as useful aids in organic synthesis, and in 1904 two
papers were published on their application in the
synthesis of organosilicon compoundssby Kipping25 and
by Dilthey and Eduardoff26 in Zurich. This procedure
continues to find useful application in organosilicon
preparations today. Kipping carried out the reaction of
C2H5MgI with silicon tetrachloride and found it to be
quite unselective, all possible products being formed:
C2H5SiCl3, (C2H5)2SiCl2, (C2H5)3SiCl, and (C2H5)4Si.
However, a 1:1 reaction of C2H5MgBr with SiCl4 gave
almost exclusively C2H5SiCl3. As Kipping later found,27
silicon tetrachloride reacts with diethyl ether even at
room temperature (certainly at reflux), probably resulting in ether cleavage and formation of products such
as Cl3SiOC2H5. This, he said, accounts for previous
difficulties in preparing various organic derivatives of
silicon. Thus from the product of the interaction of
magnesium, ethyl bromide and the tetrachloride, it may
not be hard to obtain ethylsilicon trichloride, but the
isolation of the di- and tri-ethyl derivatives by fractional
distillation is a very troublesome task. Nevertheless,
Kipping and others continued to use the Grignard
reagents to good advantage, the difficulties apparently
being restricted to the silicon tetrahalides and avoided
by working at lower temperature.
Much of Kippings work dealt with arylsilicon compoundsstheir preparation and reactivity: the reactivity
of aryl and benzyl substituents on silicon toward nitration and sulfonation and the reactivity of Si-Cl bonds
in Ar2SiCl2 and Ar3SiCl compounds toward sodium,
which resulted in Si-Si bond formation. Ph2SiCl2 gave,
among other products, the cyclic tetramer, (Ph2Si)4. As
mentioned already, a primary interest was in the
possibility of preparing optically active organosilanes
and, by successive Grignard reactions, he and his
students prepared compounds of the type RRRRSi.28
(24) Grignard, V. Compt. rend. acad. sci. 1900, 128, 110.
(25) Kipping, F. S. Proc. Chem. Soc. 1904, 20, 15.
(26) Dilthey, W.; Eduardoff, F. Ber. Dtsch. Chem. Ges. 1904, 37,
1129.
(27) Kipping, F. S.; Murray, A. G. J. Chem. Soc. 1927, II, 1401.

Organometallics 2001.20:4978-4992.
Downloaded from pubs.acs.org by 132.247.249.242 on 08/31/15. For personal use only.

4982

Organometallics, Vol. 20, No. 24, 2001

Of interest to the subject of this essay is that Kipping

hydrolyzed (C2H5)2SiCl2, obtaining an oil with the
consistency of glycerol, with molecular weight 604.29 The
hydrolysis of many other chlorosilanes was studied.
Those of the type R3SiCl in many cases gave the silanol,
R3SiOH, which could be condensed to the disiloxane,
R3SiOSiR3. Hydrolysis of the diorganodichlorosilanes
proceeded via the silanediols, R2Si(OH)2, which in some
cases could be isolated and which underwent condensation to give oligosiloxanes, (R2SiO)n. The organotrichlorosilanes on hydrolysis gave products of the type
[RSiO1.5]n (which Kipping wrote as RSiOOH, thinking
them to be analogues of carboxylic acids). From Kippings point of view as an organic chemist, organosilicon
compounds were organic compounds that happened to
contain silicon, and so he used organic-type nomenclature: R4Si ) silicane; R3SiOH ) silicanol; R2Si(OH)2 )
silicanediol; R2SiO ) silicone; RSiOOH ) siliconic
acidsin analogy to alkane, alcohol, alkanediol, ketone,
and carboxylic acid. This point of view led him to expect
actual similarities. Kippings insistence on forcing organosilicon compoundsstheir properties and reactionss
into the framework of organic chemistry was a bias that
actually proved to be a handicap to his research in
organosilicon chemistry. Superficially, there are many
similarities (but also some differences) between compounds of type R4Si (already well-known in Kippings
time) and R4C compounds or alkanes in general. However, comparisons between what he called silicones (a
name that has stuck) and ketones showed them to be
very different. About the hydrolysis product of PhCH2(C2H5)SiCl2, Kipping said the following:31
...as benzylethylsilicon dichloride is decomposed by water, giving benzylethylsilicone, we
have studied the behavior of this silicone in order
to ascertain whether it shows any similarities to
the corresponding ketone. We may say at once
that it does not; benzyl ethyl ketone boils at 226
under atmospheric pressure; benzylethylsilicone
at 305-315 under a pressure of 22 mm. This
very high boiling point of the silicone doubtless
indicates molecular complexity, and the results
of ebullioscopic experiments bear out this indication, the values obtained in acetic acid and in
acetone pointing to the termolecular formula,
(BzEtSiO)3. ...dibenzylsilicone...is also represented by the molecular fomula (Bz2SiO)3; and
judging from its high boiling point (above 360),
diethylsilicone has an analogous molecular complexity. It would seem, therefore, that silicones,
as a class, differ from the ketones in readily
forming comparatively stable molecular aggregates, but whether the latter are to be regarded
as composed of loosely associated, or of chemically united, molecules, we have as yet no
satisfactory evidence before us.
This association, polymerization, or union of
simple silicone molecules is probably one of the
reasons, but not the only one, why in other
(28) A good summary of the individual Kipping papers can be found
in: Post, H. W. Silicones and Other Organosilicon Compounds;
Reinhold: New York, 1949; pp 13-31.
(29) Martin, G.; Kipping, F. S. J. Chem. Soc. 1909, 95, 302.
(30) Kipping, F. S. J. Chem. Soc. 1912, 101 II, 2106.
(31) Robison, R.; Kipping, F. S. J. Chem. Soc. 1908, 93, 439.

respects also the silicones show no relationship

with the ketones. ...it is possible to account for
the results on the assumption that the group
Si:O of the simple silicone does not exist in the
associated molecule...
In his further research, mainly with diarylsilicon
systems, Kipping was able to isolate and identify pure,
covalently bonded cyclic tri- and tetrasiloxanes and 1,3disiloxanediols. Throughout his research he was plagued
by nonvolatile oils and gums and other materials of
higher molecular weight formed in the chlorosilane
hydrolysis reactions, which could not be crystallized and
which he considered a nuisance.
One reaction of his silicones, however, misled Kipping
into thinking that there was indeed an analogy between
his silicones and ketones. The reaction of a Grignard
reagent with a ketone to give, after hydrolytic workup,
a tertiary alcohol, i.e., R2CdO + RMgX f RR2COH,
was by 1911 a well-known reaction. As Kipping was able
to report:
As a matter of fact, the silicones were found
to react with the Grignard reagents in a normal
manner, and it is perhaps hardly too much to
say that this is the first instance in which the
silicones have been proved to show any analogy
to the ketones in chemical behavior.
A reaction reported was
H 2O

[(PhCH2)2SiO]3 + 3CH3MgI f 98
3(PhCH2)2CH3SiOH (5)
Kippings explanation, a reasonable one, considering
how he thought about his silicones, was
As these preparations gave ..good yields of the
desired products, it seemed to follow that the
termolecular silicones were resolved into the
unimolecular compounds [i.e., R2Si:O] by the
action of the Grignard reagents.
Of course, we know better now.
Kipping laid the groundwork for the explosive growth
of organosilicon chemistry that was to come, and he
lived to see its early, most important stages. The
obviously oligomeric nature and high thermal stability
of silicones such as [(C2H5)2SiO]n and [(PhCH2)(CH3)SiO]n, the adhesive properties of benzylsiliconic acid
(sticking to glass, paper, and porcelain), and the filmforming properties of diphenylsilicone that he had
observed did not move Kipping to think seriously about
possible applications. In his Bakerian Lecture,33 delivered on Dec 19, 1936, in which he reviewed his 37 years
of organosilicon research, he was not optimistic: ...the
few [organosilicon compounds] which are known are
very limited in their reactions, the prospect of any
immediate and important advances in this section of
organic chemistry does not seem to be very hopeful.
Kipping was a pure academic, a vanishing species in
todays academic chemistry research environment. However, these properties aroused the interest of others who
had specific applications in mind. For instance, in his
1935 book, The Chemistry of Synthetic Resins II, C. Ellis
(32) Kipping, F. S.; Hackford, J. E. J. Chem. Soc. 1911, 99, 138.
(33) Kipping, F. S. Proc. R. Soc. (London), A 1937, 159, 139.

Organometallics 2001.20:4978-4992.
Downloaded from pubs.acs.org by 132.247.249.242 on 08/31/15. For personal use only.

Organometallics, Vol. 20, No. 24, 2001 4983

Figure 2. Early form of Stocks high-vacuum system (from ref 36b).

wrote about Kippings work: The further development

of organosilicon chemistry could lead to the production
of cheap and easily preparable compounds which could
be used as resins or adhesives. In view of Kippings
important contributions to organosilicon chemistry, it
certainly is fitting that there is an American Chemical
Society award called the Frederic Stanley Kipping
Award in Organosilicon Chemistry.
Frankland had pointed out a formal relationship of
the hydrogen derivatives of the main-group elements
and their alkyl compounds, e.g., AsH3 and As(C2H5)3.
Thus, in that sense, H2SiCl2 can be viewed as the first
member of the (CnH2n+1)2SiCl2 series (n ) 0), in which
our cover molecule, (CH3)2SiCl2, is the first (alkyl)2SiCl2
(n ) 1) member. The chemistry of H2SiCl2 is indeed
similar in many respects to that of (CH3)2SiCl2 (except
for the special reactivity of its Si-H bonds). This
consideration brings us to the work of Alfred Stock.
Stock is noted especially for his discovery of the boron
hydrides and the development of their chemistry. Although Wohler and Buff were the first to prepare SiH4,34
it was Stock and his students who developed and
systematized the silicon hydrides and their chemistry.35
To develop the chemistry of the volatile (gaseous and
liquid), highly reactive, usually pyrophoric boron and
silicon hydrides, Stock had to invent a completely new
methodology for their handling. Stock, like Bunsen and
Frankland, was an inventive and expert experimentalist
and, what was important, an expert glassblower. To deal
with the hydrides of boron and silicon, he devised and
built an elaborate high-vacuum system with mercury
valves, in which all parts were fused together and in
which the chemicals with which one is dealing come in
contact only with glass and mercury (Figure 2). Performance of chemical reactions, distillation, sublimation,
analysis, determination of physical properties, and

storage, were all possible on the milligram scale in this

flexible system and its clever ancillary equipment.36
Stock and Somieski prepared H2SiCl2 by the AlCl3catalyzed gas-phase reaction of SiH4 with 2 molar equiv
of HCl.37 In one example,37b a reaction of 284 cm3 of SiH4
(all volumes are of the respective gaseous reactant or
product) with 594 cm3 of HCl in the presence of a small
amount of AlCl3 at 100 C for 10 days in a bulb on the
high-vacuum system gave a mixture of H2, 2 cm3 of
SiH4, 63 cm3 of HCl, 39 cm3 of H3SiCl, and 223 cm3 of
H2SiCl2. Isolation of pure H2SiCl2 from this mixture was
effected by fractional distillation under high vacuum, a
process Stock characterized as exceedingly laborious and
time-consuming.38 Of special interest is the gas-phase
hydrolysis of H2SiCl2,37b carried out, in one example,
with 49.6 cm3 of water vapor and 46.9 cm3 of gaseous
H2SiCl2 in the chromic acid-washed 8 L bulb shown in
Figure 3. The expected reaction was that shown in eq
6. The initial pressure increase suggested to Stock that

(34) Wohler, F.; Buff, H. Ann. 1857, 103, 218.

(35) Stock, A. Hydrides of Boron and Silicon; Cornell University
Press: Ithaca, NY, 1932; Chapter II. (Stock (1876-1946), chemistry
study in Berlin; a year with Moissan in Paris. Privatdozent in Berlin,
1900-1909; professor in University of Breslau, 1909-1916; Berlin
(Kaiser Wilhelm Institut), 1916-1926; University of Karlsruhe, 19261936; Berlin as emeritus professor to 1943. Best known for his
thorough, exacting studies of the boron hydrides.)

(36) (a) For a fascinating, detailed description of this ingenious

equipment and how it was used, see Chapter XXX in ref 35. (b) In
German: Stock, A. Ber. Dtsch. Chem. Ges. 1917, 50, 989; 1918, 51,
983; 1920, 53, 751.
(37) (a) Stock, A.; Somieski, C. Ber. Dtsch. Chem. Ges. 1919, 52, 695.
(b) 1919, 52, 1851.
(38) Now pure H2SiCl2 is available commercially in gas cylinders
containing up to 250 lb.

H2SiCl2(g) + H2O(g) f H2SiO(g) + 2HCl(g) (6)

monomeric H2SiO (which he called prosiloxane) had
been formed in the very dilute gas phase. (Had it
polymerized immediately, a large pressure decrease
would have been observed.) With time, the pressure in
the bulb decreased and [H2SiO]n polymer coated the
glass surface. This coating, as expected, evolved hydrogen on treatment with aqueous NaOH. Thus, the parent
silicone apparently had been prepared. The prosiloxane,
in other experiments, could be isolated (0.05 mL) as a
mobile, clear liquid that quickly became viscous and
finally gelled. Years later, our study at MIT showed that
the hydrolysis of H2SiCl2, either with a stoichiometric
amount of water in dichloromethane at -30 to -20 C
or by slow, controlled addition of a slight excess of water

4984

Organometallics, Vol. 20, No. 24, 2001

Organometallics 2001.20:4978-4992.
Downloaded from pubs.acs.org by 132.247.249.242 on 08/31/15. For personal use only.

Figure 3. Stocks apparatus used in the H2SiCl2 hydrolysis experiment. (from ref 37b).

at 0 C to a H2SiCl2 solution in dichloromethane, results

in formation of (H2SiO)n siloxanes.39 In the volatile
fraction, cyclic oligomers with n from 4 to 23 were
detected and those with n ) 4-6 were isolated by gas
chromatography and characterized by 1H and 29Si NMR
spectroscopy and infrared and mass spectroscopy: this
is much easier when you have all these modern tools
and can purchase a cylinder of H2SiCl2. Stock and his
students deserve our unstinted admiration! Stock and
Somieski also put Franklands dimethylzinc into play,
using it, again in a gas-phase reaction, to convert
H2SiCl2 to (CH3)2SiH2.37a The alkaline hydrolysis of
(CH3)2SiH2, followed by acidification, gave an oil and
some crystals. Evaporation of a benzene solution of
these products left a syrup. Stock suggested the sequence (CH3)2SiH2 f (CH3)2Si(ONa)2 f (CH3)2Si(OH)2
f [(CH3)2(HO)Si]2O f [(CH3)2SiO]n had taken place.
This was only a very small scale experiment (using 5.57
cm3 of gaseous (CH3)2SiH2), but poly(dimethylsiloxane)
had been prepared for the first time! Another compound
that we will meet again later, CH3SiHCl2, also was
prepared by Stock and Somieski during this study.
Up until this time, and for a few years more, all
organosilicon research was carried out in academic
laboratories, where nobody seemed at all interested in
possible applications of the chemistry that had been
uncovered.40 And at this point our introduction (a rather
long one) ends.
The scene now shifts to industry.41 Industrial interest
and activity arose from definite needs. The first industrial research in organosilicon chemistry was carried out
in the 1930s by J. Franklin Hyde (1903-1999) at the
Corning Glass Works. When Hyde was hired by Corning
in 1930, after Ph.D. studies at the University of Illinois
with Roger Adams and postdoctoral work at Harvard
with J. B. Conant, it was suggested to him that there
(39) Seyferth, D.; Prudhomme, C.; Wiseman, G. H. Inorg. Chem.
1983, 22, 2163.
(40) An exception was K. A. Andrianov (1904-1978) in the USSR,
who concluded that organosilicon compounds containing Si-O-Si
linkages should be useful materials of high thermal stability for
applications in electrical insulation. Beginning in 1937, he initiated
efforts to develop such materials, first investigating tetraalkoxysilanes
and then also Grignard synthesis derived organochlorosilanes. Andrianov was one of the leaders in the development of organosilicon
chemistry in the USSR after World War II. See: Zhdanov, A. A. Zh.
Obshch. Khim. 1979, 49, 462.

might be some field worth looking at between glass and

organic polymers. As it happened, Hyde did not get to
organosilicon chemistry right away, but he did read
Kippings papers and soon he prepared some phenylchlorosilanes by the Grignard procedure. In further
investigations, he found that their resinous, low-melting
hydrolysis products were very thermally stable. In the
mid-30s, Corning was trying to apply its glass fiber
technology to the fabrication of thermally stable insulating tape for electrical applications. A thermally stable
polymer was needed to coat the fibers and to provide
the matrix for the glass fibers in fabricating the tape.
Kippings polysiloxanes seemed to be good candidates,
and so Hyde began preparing the required starting
materials, various organochlorosilanes, by the Grignard
procedure and studying their hydrolysis. This now is
industrial research, and if journal publications come at
all, they come at a later date, after patent applications
have been filed or, more often, only after the patents
have been issued. Hydes synthetic work was the subject
of a 1941 publication.42 The compounds prepared were
Ph(C2H5)SiCl2, (C2H5)2SiCl2, Ph(CH3)SiCl2, Ph2SiCl2,
and (CH3)2SiCl2. All were hydrolyzed under mild conditions to give liquid products of relatively low molecular
weight, except for Ph2SiCl2, which gave the solid
(Ph2SiO)3. Such cyclic trisiloxanes appeared to be
present also in the hydrolysis products of the other
dichlorosilanes. The oligosiloxanes were converted to
resinous polymers either by heating with aqueous HCl
in the case of the phenyl-containing siloxanes (which
resulted in Ph-Si cleavage) or, in the case of the
phenylalkyldichlorosilanes and dialkyldichlorosilanes,
by air oxidation at high temperature, which replaced
alkyl groups by Si-O-Si links. The properties of the
hydrolysis/condensation/oxidation products of these
R2SiCl2 compounds must have looked promising, because these studies were continued with applications
in mind. With the beginning of World War II, the US
Navy, which required thermally stable insulation for the
electric motors used in submarines, took an active
interest in this research. This ultimately led to the
formation of the Dow Corning Corporation in 1942, a
joint venture of the Dow Chemical Co., which could
provide the magnesium required for the preparation of
the Grignard reagents to be used in the organochlorosilane synthesis as well as plant-scale chemical engineering expertise, and the Corning Glass Works, which had
developed the chemistry. Dow Cornings research, development, and manufacturing operations were located
in Midland, MI, the location of Dow Chemical Co. Hyde
and colleagues from Corning and the Corning Mellon
Institute Fellowship moved to Midland. Hyde had a
distinguished research and development career at Dow
Corning, remaining active in research on silicones well
past his formal retirement in 1969.
(41) Accounts of the beginnings of industrial organosilicon chemistry
are given in several books. (a) Liebhafsky, H. A. Silicones under the
Monogram; Wiley-Interscience: New York, 1978. (b) Rochow, E. G.
Silicon and Silicones; Springer: Berlin, Heidelberg, 1987. (c) McGregor,
R. R. Silicones and Their Uses; McGraw-Hill: New York, 1954. (d)
Warrick, E. Forty Years of Firsts; McGraw-Hill: New York, 1990. (e)
See also the article: Hyde, J. F. Organic Chemist in a Glass Factory.
Chem. Heritage 1992, 9(2), 12; 1992-3, 10(1), 13. References 41a,b
describe developments at the General Electric Co. and 41c- e at the
Corning Glass Works, at the Mellon Institute, where Corning supported a research project, and at Dow Corning Corp.
(42) Hyde, J. F.; DeLong, R. C. J. Am. Chem. Soc. 1941, 63, 1194.

Organometallics 2001.20:4978-4992.
Downloaded from pubs.acs.org by 132.247.249.242 on 08/31/15. For personal use only.

Organometallics, Vol. 20, No. 24, 2001 4985

Before this happened, however, there were other

important developments that are described in more
detail in ref 41a,b. After a visit in January 1938 to the
Corning laboratories, during which they heard about
Hydes work, chemists from the Schenectady Research
Laboratory of the General Electric Co. became interested in silicones for their own applications as insulation
of high thermal stability in electric motors. In fact, W.
I. Patnode, a GE chemist, had been trying to apply
condensation products based on tetraethoxysilane to
this purpose. A decision to investigate polyorganosiloxanes at GE followed this visit. A young chemist who had
joined the GE Research Laboratory after earning his
Ph.D. at Cornell, Eugene G. Rochow, took on this project
on his own. After initial work with the thermally cured
hydrolysis/condensation product of diphenyldichlorosilane, which did not seem very promising, Rochow
changed direction. He reasoned that polydialkylsiloxanes might be worth looking at but that those which
contained C-C bonds should be avoided, since they
should be less stable thermally. Also, their thermal
decomposition would leave conducting carbonaceous
residues in the electric motor. That left poly(dimethylsiloxane) as the only candidate, and in August 1938,
Rochow began research on methylpolysiloxanes. The
Grignard synthesis that Kipping had used was applied
to the reaction of CH3MgBr with SiCl4 in diethyl ether.
The problem with this synthesis is that it gives a
mixture of products: CH3SiCl3, (CH3)2SiCl2, (CH3)3SiCl,
and even some (CH3)4Si. In general, the susceptibility
of the chlorosilanes toward nucleophilic attack decreases
in the order SiCl4 RSiCl3 > R2SiCl2 . R3SiCl; the
process is not selective. The proportion of any one
product can be optimized by adjusting the molar ratio
of the reactants and the reaction conditions, but mixtures will result in any case. Added to this problem was
the possibility of some Si-Br for Si-Cl exchange when
CH3MgBr was used (it being much easier to prepare and
more ether-soluble than CH3MgCl) and the fact that the
components present in the final product mixture were
extremely difficult to separate by fractional distillation
(boiling points: CH3SiCl3, 65.7 C; (CH3)2SiCl2, 70.0 C;
(CH3)3SiCl, 57.3 C; SiCl4, 57.6 C; (CH3)4Si was not a
problem with bp 26.5 C). In view of these difficulties,
it was the mixed methylchlorosilane product which was
hydrolyzed and processed to give methylsilicone resins
with CH3/Si ratios of 1.3 to 1.5.43 These showed remarkable thermal and oxidative stability: stable when
heated at 550 C under vacuum for 16 h and only slow
oxidation in air at 300 C. These properties were very
encouraging, and research on polymethylsiloxanes continued, with other chemists at the GE laboratories
becoming involved. Attempts to prepare pure (CH3)2SiCl2
and CH3SiCl3 by the reaction of CH3MgCl with SiCl4
in di-n-butyl ether followed by careful fractional distillation (120 plate column under extremely stable conditions) gave products of empirical composition (CH2.98)2.03SiCl1.84 and (CH2.89)1.13SiCl2.98 (by elemental analysis).44
This now allowed a study of each chlorosilane as a pure
compound.
The major stumbling block in the commercialization
of the polymethylsiloxanes by GE was the fact that there
(43) Rochow, E. G.; Gilliam, W. F. J. Am. Chem. Soc. 1941, 63, 798.
(44) Gilliam, W. F.; Liebhafsky, H. A.; Winslow, A. F. J. Am. Chem.
Soc. 1941, 63, 801.

was no practical, commercially reasonable procedure for

the synthesis of the required methylchlorosilanes as
individually pure compounds. The only available synthetic route at the time was Kippings organomagnesium procedure, and some of the problems associated
with its use already have been noted above. Carrying
the Grignard synthesis out on a plant scale would not
be impossible, but minor problems associated with the
laboratory-scale reaction would be less minor on the
plant scale (e.g., handling large volumes of flammable
solvent, disposing or recycling of magnesium salts, and
purifying the individual methylchlorosilanes by fractional distillation). An added difficulty was that the Dow
Chemical Co. controlled magnesium production in the
USA, thus greatly favoring its joint venture with the
Corning Glass Works. GE either had to give up the idea
of a polymethylsiloxane business or find a new route to
the methylchlorosilanes. The GE methylsilicone project
was put on hold. However, Rochow was permitted to
work part time on alternate routes to the methylchlorosilanes.
Among the experiments Rochow carried out were
repetitions of the reactions of gaseous HCl at high
temperature with elemental silicon and its alloys, with
the idea of somehow converting the HSiCl3 product to
CH3SiCl3. All attempts to do this were unsuccessful.
However, there came the following idea: if HCl reacts
with Si/Cu, why not try CH3Cl? In his first experiment,
passing a mixture of HCl and CH3Cl through a heated
tube furnace containing ferrosilicon, only HSiCl3 and
SiCl4 appeared to have been formed. However, after
carrying out their hydrolysis in diethyl ether solution,
Rochow noted that the flask that had contained the
ether solution had a slippery feelsas though a film of
methyl silicone had formed there.41b
Rochows research notebook, as quoted in ref 41a,b,
records what he did next:
May 9, 1940, contd
Copper-silicon
I crushed some of the Niagara Falls Smelting Co.
50% Cu-Si in the jaw-crusher, and packed a
Nonex tube with the material (size is about 1/4
down to fine powder). I arranged the tube in the
furnace and arranged to admit both CH3Cl &
HCl. Single CO2 condensing tube on the outlet
end (see Figure 5.1). [Figure 4 in this essay.]
May 10, 1940
I heated the tube in the furnace to 370 C and
kept it there. I passed through some HCl at first,
to attack the alloy superficially, then passed in
CH3Cl slowly. Let it run all day.
At 4:40 P.M. I stopped the stream of CH3Cl.
About 5 cm3 of liquid had collected in the
condenser, plus some in the cold end of the
furnace tube. I put it all in ice water having a
layer of ether on top, and stirred. The material
hydrolyzed with some cloudiness, but not large
volumes of silica; there seemed to be little CH3Cl
either.
I decanted some of the ether solution into a
Petri dish and evaporated the ether. A clear
thick glycerol-like substance resulted. This liquid
is sticky to the touch, acts very much like the
methyl silicone.

4986

Organometallics, Vol. 20, No. 24, 2001

CH3SiCl3 + 3H2O f CH3Si(OH)3 + 3HCl

(CH3)2SiCl2 + 2H2O f (CH3)2Si(OH)2 + 2HCl
and a small amount of

SiHCl3 + 3H2O f HSi(OH)3 + 3HCl

Organometallics 2001.20:4978-4992.
Downloaded from pubs.acs.org by 132.247.249.242 on 08/31/15. For personal use only.

The methyl silicols undergo partial condensation

immediately to form the viscous intermediate
products:

Figure 4. Apparatus used in Rochows first CH3Cl + Si/

Cu direct synthesis experiment: (A) HCl generator; (B)
bubble counter; (M) CH3Cl input; (F) tube furnace surrounding reaction tube; (C) condenser; (R1) ) receiver at 0
C; (R2) receiver at -80 C (from ref 41b, by permission of
Springer-Verlag).

Some of the thick liquid product from the

evaporation of the ether solution was warmed by
the rays of the projection lamp for 1 h. At the
end of this time, it was a colorless, sticky, almost
solid resin. The behavior suggests condensation
of the methyl silicols which I believe are produced during the hydrolysis.
The hydrolysis product of the materials resulting from the reaction of CH3Cl and coppersilicon, then, resembles methyl silicone produced
by another method, and I believe it to be methyl
silicone.
The reactions which I believe occur are as
follows. First HCl is passed through the tube:

Si + 3HCl f SiHCl3 + H2
Only a small amount of HCl is passed through,
and this is done principally to etch the surface
of the alloy. Small amounts are later mixed with
the CH3Cl, in the ratio of perhaps 1 part to 50
parts of CH3Cl. The CH3Cl reacts in this way:

3CH3Cl + Si f CH3SiCl3 + C2H6

2CH3Cl + Si f (CH3)2SiCl2
and, to a much smaller extent, this might occur:

CH3Cl + 2HCl + Si f CH3SiCl3 + H2

The liquid products, which I believe are methyl
silicon chlorides, condense in the cooler portions
of the tube containing the alloy and are also
distilled out into the condensing tube kept at -80
C. The colorless liquid so collected (in the
condensing tube) does not bubble much when
warmed to room temperature, hence does not
contain much CH3Cl.
Upon hydrolysis of the combined liquid products,

This goes on until sticky liquid products result.

On warming, condensation proceeds further,
splitting off more water (which evaporates in
part or stays behind in globules). The end result
is a clear resinous body which I believe to be
methyl silicone.
/S/ E.G. Rochow
May 10, 1940
As Herman Liebhafsky said41a about this experiment: The most important single experiment and the
best single days work in the history of the silicone
industry. (Figure 5) This was the breakthrough for
which GE had been hoping: a proprietary methylchlorosilane synthesis that, one might hope, even at this
early stage, could be scaled up and commercialized, a
reaction that does not require a preformed organometallic reagent and a flammable solvent and does not
generate large quantities of a magnesium halide.
Further work showed that the use of the silicon/
copper alloy (vs pure silicon) definitely was advantageous: it accelerated the rate of the reaction so that
lower reaction temperatures could be used. A fair
selectivity favoring (CH3)2SiCl2 was observed, but substantial amounts of CH3SiCl3 also were formed during
the latter stages of the reaction. Of course, a patent
application was filed that covered this new chemistry;
the patent (U.S. Patent 2,380,995) was issued in Sept
1945 (application on Sept 26, 1941, but issued a few
years late since it had been under wartime secrecy
restriction), and in June 1945 Rochow published his
results45 (Figure 6). As noted in this paper, the direct
RX + Si/Cu synthesis could be extended to RX ) CH3Br,
C2H5Cl, and C2H5Br but gave only poor results with
chlorobenzene.45 However, good results (satisfactory
yields and fair selectivity to Ph2SiCl2 vs PhSiCl3) were
obtained when pellets containing 90% silicon and 10%
silver which had been sintered in hydrogen were used.46
A short account concerning E. G. Rochow (Figure 7)
should be of interest at this point. He was born on Oct
4, 1909, in Newark, NJ. He obtained his undergraduate
degree at Cornell University, where he was associated
with Professor L. M. Dennis, whose research focused
on fluorine and its compounds, and group 13 and
germanium organometallic chemistry. Because of the
severe depression in the USA at that time, Rochow
(45) Rochow, E. G. J. Am. Chem. Soc. 1945, 67, 963.
(46) Rochow, E. G.; Gilliam, W. F. J. Am. Chem. Soc. 1945, 67, 1772.

Organometallics 2001.20:4978-4992.
Downloaded from pubs.acs.org by 132.247.249.242 on 08/31/15. For personal use only.

Organometallics, Vol. 20, No. 24, 2001 4987

Figure 5. Rochow with his more elaborate direct synthesis apparatus in the GE Research Laboratory (photo by J. P.
McNally Photography; American Chemical Society photo archives).

Figure 6. Rochows 1945 J. Am. Chem. Soc. paper.45 It

was concise, only three pages in length, but it was one of
the most important papers in the history of organosilicon
chemistry.

stayed at Cornell for his graduate studies, continuing

research with Dennis. He worked on the electrochemical
preparation of fluorine and studied the oxyacids of
fluorine but also, in a separate project, worked on the
synthesis and properties of trimethylindium and triethylthallium. He became acquainted with silicon chemistry when he was a special assistant of Alfred Stock
during the year that Stock was the George Fisher Baker
Lecturer at Cornell. (In his book,35 Stock thanks Rochow
for preparing most of the illustrative drawings in the
book.) In 1935, Rochow began his career at GE, first
spending several years in the area of ceramics. Then
came his extensive research in organosilicon chemistry.
During these years he continued work on the direct

Figure 7. Eugene G. Rochow (courtesy of Prof. E. G.

Rochow).

synthesis of methyl- and ethylchlorosilanes and extended it to the synthesis of methyl- and ethylchlorogermanes by reaction of CH3Cl and C2H5Cl, respectively,
with a Ge/Cu reaction mass at 320-350 C.47 He also
found that methanol reacted with a powdered 90% Si/
10% Cu mixture at 250 C to give (CH3O)4Si as the
(47) (a) Rochow, E. G. J. Am. Chem. Soc. 1947, 69, 1729. (b) 1950,
72, 198.

Organometallics 2001.20:4978-4992.
Downloaded from pubs.acs.org by 132.247.249.242 on 08/31/15. For personal use only.

4988

Organometallics, Vol. 20, No. 24, 2001

major product.48 The reaction, however, gave also minor

products containing Si-H and Si-CH3 bonds.
Rochow left GE in 1948 to join the faculty of the
Harvard University Chemistry Department. He taught
courses in general and inorganic chemistry. His freshman general chemistry lectures were entertaining and
instructive and were very popular with the students.
His research activities were quite varied. They included
topics such as the use of ClF3 as a fluorinating agent,
electrode potentials in silicate melts, and extensions of
the direct synthesis to the reactions of CH3Cl with
molten Sn/Cu at 300-350 C to produce methyltin
chlorides (with very high selectivity to (CH3)2SnCl2)49
and with arsenic- and antimony/copper mixtures at
350-375 C to give the monomethyl dichlorides and
dimethyl chlorides of these elements.50 Further studies
of the ROH + Si/Cu reaction also were carried out, and
the reaction of Si/Cu with ethers at high temperature
was investigated. Organosilicon polymers continued to
be of interest. Of note were the first low-resolution
(broadline) NMR spectroscopic studies of molecular
motion in organosilicon polymers and the first syntheses
of polysilazanes. Application of the then new highresolution NMR spectroscopy to the determination of
the relative electronegativities of the group 14 elements
led to further work, whose culmination was the AllredRochow electronegativity scale, which can be found in
every inorganic chemistry textbook. This distinguished
research career was recognized with many awards,
including the 1965 Frederic Stanley Kipping Award in
Organosilicon Chemistry and the Alfred Stock Medal
of the German Chemical Society. A Chemical Reviews
article in 1947 on The Present State of Organosilicon
Chemistry51 and especially his book, An Introduction
to the Chemistry of the Silicones,52 were very influential
worldwide by pointing out the known and potential
applications of organosilicon chemistry; they attracted
many chemists into this field. Rochow retired from his
position on the Harvard faculty in 1970, but his interest
in organosilicon chemistry and in chemistry in general
has not waned. He continued his writing (four books
since 1970) and he has attended and spoken at many
of the national and international conferences devoted
to organosilicon chemistry since 1970. (For a biographical sketch of E. G. Rochow, published on the occasion
of his 70th birthday, see Seyferth, D. J. Organomet.
Chem. 1979, 178, ix-xii.)
Unknown to the chemists at GE and Corning, the
direct synthesis of methylchlorosilanes was discovered
independently, but after Rochows discovery, by Richard
Muller in Germany. Muller had been trying to improve
the synthesis of HSiCl3 in the early 1930s for possible
use as a military smoke agent. He found that the
reaction of gaseous HCl with a high-silicon (>80%)
ferrosilicon admixed with a copper compound (CuO,
CuCl, CuCl2, CuCO3) proceeded rapidly below 300 C
to give HSiCl3 in good yield. Even better results were
(48) Rochow, E. G. J. Am. Chem. Soc. 1948, 70, 2170.
(49) Smith, A. C., Jr.; Rochow, E. G. J. Am. Chem. Soc. 1953, 75,
4103.
(50) Maier, L.; Rochow, E. G.; Fernelius, W. C. J. Inorg. Nucl. Chem.
1961, 16, 213.
(51) Burkhard, C. A.; Rochow, E. G.; Booth, H. S.; Hartt, J. Chem.
Rev. 1947, 41, 97.
(52) Rochow, E. G. An Introduction to the Chemistry of the Silicones;
Wiley: New York, 1946 (1st ed.); 1951 (2nd ed.).

Figure 8. Richard Muller (courtesy of Professor H.

Schmidbaur).

obtained when the reaction was carried out under

pressure.53 Further experiments on similar reactions of
Si/Cu with chloroform gave, as Muller said,54 no
organochlorosilanes in the ordinary sense, and, therefore, we abandoned that method. A reaction of methyl
chloride with Si/Cu was tried only some years later. It
is not exactly clear just when this independent discovery
of the CH3Cl + Si/Cu-based methylchlorosilane process
was made, since details are lacking. One article says
that it was in 1941/1942.55 Muller, in his short history
of organosilicon chemistry,54 said I sometimes wonder
why the idea took me such a long time so that E. G.
Rochow could precede me by nine months. But earlier
in this paper, he puts Rochows discovery as occurring
in 1941, not in 1940..... Mullers discovery was disclosed
in a German patent application (DRP Anm. C57 411,
with a secrecy restriction) in June 1942. After that, the
project was discontinued (the German bureaucrats
involved did not recognize the importance of this work)
and was revived again only after the end of the war, in
the early days of the German Democratic Republic
(DDR).
A few words about Richard Muller (1903-1999)
(Figure 8) are warranted.56 He was born in Saxony and
studied at the University of Leipzig. He obtained his
Ph.D. in 1931, having carried out his dissertation
research on the system nickel oxide/oxygen/hydrogen.
Employment in industry (Chemische Fabrik von Heyden in Radebeul, near Dresden) followed. He worked on
various projects; research in silicon chemistry began in
(53) Muller, R. Chem. Tech. 1950, 2, 7, 41. German patents were
applied for on this work in 1934 and 1938 and granted in 1936 and
1939, respectively, but were put under a secrecy restriction by the
German government and not made public.
(54) Muller, R. J. Chem. Educ. 1965, 42, 41 (entitled One Hundred
Years of Organosilicon Chemistrys an eight-page historical review
with pictures of all the silicon chemists involved: Friedel, Crafts,
Ladenburg, Kipping, Dilthey, Stock, R. Schwarz, Rochow, Hyde,
Patnode, Andrianov, Dolgov). German version: Wiss. Z. Tech. Univ.
Dresden 1963, 12, 1633.
(55) Reuther, H. Chem. Techn. 1953, 5, 297.
(56) For brief biographical sketches, see ref 55 and an obituary:
Schmidbaur, H. Nachr. Chem. Techn. Lab. 1999, 47, 1261.

Organometallics, Vol. 20, No. 24, 2001 4989

Organometallics 2001.20:4978-4992.
Downloaded from pubs.acs.org by 132.247.249.242 on 08/31/15. For personal use only.

Figure 9. Eugene G. Rochow and Richard Muller at the

Munich Silicon Days Conference, August 1992 (by permission, Doris Wacker, Wacker-Chemie GmbH).

1934 and continued until 1942. After the end of World

War II, under the DDR, Muller was instrumental in
getting silicone production in Nunchritz, near Radebeul,
started. He resumed research and development activity
on organosilicon chemistry. By 1955, the direct synthesis of methylchlorosilanes was operative in the plant in
Nunchritz and Muller became director of the new
Institut fur Silikon- und Fluorkarbon-Chemie in Radebeul. In 1954, he was appointed Professor in the
Technical University of Dresden, where he lectured on
silicone, fluorocarbon, and polymer chemistry. He was
relieved of all these positions in 1968, at age 65she had
not been a friend of the DDR regime. His research
activity in the postwar years dealt primarily with the
synthesis and diverse applications of the silicones. He
also carried out research in other areas of organosilicon
chemistry, returning to the reaction of chloroform with
Si/Cu at 300 C in 1958. Interesting trichlorosilyl
derivatives were found: (Cl3Si)2CH2, Cl3SiCH2SiHCl2,
Cl3SiCH(SiHCl2)2, (Cl3Si)2CHSiHCl2, and (Cl3Si)3CH.57
A similar reaction of CCl4 gave C(SiCl3)4, Cl3SiCtCSiCl3,
Cl3SiCCldCClSiMe3, and smaller amounts of Cl3SiCCldCCl2 and (Cl3Si)2CdCdC(SiCl3)2.58 Also of interest
were studies of the clathrates of linear polysilanes,
n-SinH2n+2,59a and of linear mono- and dialkylsilanes
with urea and thiourea,59b but his most extensive
research activities outside of silicone chemistry were
devoted to the preparation and chemistry of organofluorosilanes. Of special interest was his pioneering work
on the preparation and applications in synthesis of
hypercoordinate organofluorosilicates that contained
anions of the types [RSiF4]- and [RSiF5]2-, which were
found to be effective sources of nucleophilic alkyl and
aryl groups in aqueous solution (green chemistry
before its time).60 In 1992, during a meeting in Munich
whose purpose was to celebrate the 50th anniversary
of the direct synthesis of methylchlorosilanes, Eugene
Rochow and Richard Muller were each honored with a
Wacker-Silikon-Preis (Figure 9).
We return now to the GE Research Laboratory in
order to follow our cover molecule from the laboratory
(57) Muller, R.; Seitz, G. Chem. Ber. 1958, 91, 22.
(58) Muller, R.; Beyer, H. Chem. Ber. 1959, 92, 1018.
(59) (a) Muller, R.; Meier, G. Z. Anorg. Allg. Chem. 1965, 337, 268.
(b) 1964, 332, 81.
(60) (a) Muller, R. Z. Chem. 1965, 5, 220. (b) Organomet. Chem. Rev.
1966, 1, 359.

Figure 10. Stirred bed reactor used initially in the CH3Cl + Si/Cu direct synthesis (from ref 41a, by permission of
John Wiley & Sons, Inc.).

to the plant.41a,b It is a long way from the first successful

experiment to the plantsfrom Rochows initial 5 cm3 of
product to commercial quantities of methylchlorosilanes.
More chemists were assigned to the silicone project.
There was much to do in research, scale-up in the pilot
plant, and process development. Very quickly, a smallscale pilot operation was started in the Research
Laboratory to investigate the problems of larger scale
synthesis and to produce the methylchlorosilanes in
larger quantities for study of their chemistry. A better
Si/Cu reaction mass was developed; but many other
metals besides copper were tested to make sure that
nothing would be missed. As in the case of the HCl +
Si reaction, copper turned out to be the best one for the
methyl chloride reaction. Addition of other gases was
studied. Dilution of methyl chloride with nitrogen
proved to be beneficial, and addition of hydrogen
increased the yield of byproduct CH3SiHCl2, which later
was found to have useful applications. In these early
studies, zinc was discovered to be a useful additive to
the Si/Cu reaction mass; it was a promotor, and it made
the catalyst more effective. The first studies concerning
the mechanism of the direct reaction were carried out.
(Such studies are still being pursued now, 60 years
later.) There was some urgency to all these efforts
because there were important real and potential applications of some silicones in the war effort. Silicone
fluids and silicone rubber had been discovered. The
methylchlorosilane hydrolysis products had been found
to be good water repellents and, as such, had found an
application in the waterproofing of insulators in radios
in military aircraft.
In addition to the methylchlorosilanes, (CH3)2SiCl2
(the most desirable product), CH3SiCl3, and (CH3)3SiCl,
many other compounds were found to be present in the

Organometallics 2001.20:4978-4992.
Downloaded from pubs.acs.org by 132.247.249.242 on 08/31/15. For personal use only.

4990

Organometallics, Vol. 20, No. 24, 2001

Figure 11. Fluid-bed reactor with associated fluid-energy mill used in the CH3Cl + Si/Cu direct synthesis: Powdered
Si/C feed at A; CH3Cl feed at B (from ref 41b, by permission of Springer-Verlag).

Figure 12. Distillation system used to obtain the pure methylchlorosilanes (from ref 41b, by permission of SpringerVerlag).

reaction product mixture, most of them minor compared

to the three above, and these had to be separated and
identified. As a result of various studies worldwide, the
following compounds were found: CH3SiHCl2, SiCl4,
(CH3)4Si, (CH3)2SiHCl, HSiCl3, H2SiCl2, RnSiCl4-n
(R > CH3); disilanes ((CH3)3SiSi(CH3)3, Cl3SiSiCH3Cl2,
(CH3)3SiSi(CH3)2Cl, (CH3)2ClSiSiCl(CH3)2, Cl3SiSiCl3,
Cl2(CH3)SiSi(CH3)2Cl, Cl2(CH3)SiSi(CH3)Cl2); silaalkanes (Cl3SiCH2SiCl3, [(CH3)nSiCl3-n]2CH2, (CH3)n(SiCH2Si)Cl6-n, (CH3)n(SiCH2CH2Si)Cl6-n, (CH3)n(SiCH2SiCH2Si)Cl8-n); siloxanes ((CH3)2HSiOSiH(CH3)2, (CH3)3SiOSi(CH3)3, [(CH3)nSiCl3-n]2O, (CH3)nCl6-nSi2O); some
hydrocarbons (CH4, C2H6, C2H4); H2.61 The compounds
containing two or more silicon atoms constitute the

high boiling residue, the disilane fraction of which can

be converted to useful methylchlorosilanes by several
different methods. The direct reaction of methyl chloride
with Si/Cu obviously is a process of some complexity.
Over the years, at GE, Dow Corning, Union Carbide,
and the other methylchlorosilane producers, many studies devoted to obtaining an understanding of the mechanism of the CH3Cl + Si/Cu reaction were carried out.
Additives to the Si/Cu contact mass were investigated:
elemental metals and metal compounds. Tin in combination with zinc was found to be a good promoter; lead
(61) Voorhoeve, R. J. H. Organohalosilanes. Precursors to Silicones;
Elsevier: Amsterdam, 1967.

Organometallics, Vol. 20, No. 24, 2001 4991

Table 1. Properties of the Polysiloxanes

Organometallics 2001.20:4978-4992.
Downloaded from pubs.acs.org by 132.247.249.242 on 08/31/15. For personal use only.

(1) high thermal stability

(2) unusually weak intermolecular forces
lower than expected boiling points
remarkably low freezing and pour points
relatively low surface tension (spreading ability)
small variation of physical properties with temperature
(viscosity: very low Eact for viscous flow)
high compressibility
low glass transition temperatures: (Me2SiO)n -120 C
vs polypropylene -35 C, poly(methyl methacrylate)
+105 C
(3) higher permeability (gases)
(4) water repellency
(5) nontoxic

was a poison. Aluminum, antimony, arsenic, bismuth,

and phosphorus were found to have beneficial effects.
The study of the mechanism of the reaction of a gas with
a solid consisting of two different elements and involving
several phases is fraught with difficulties. Obviously,
the initial reaction must occur at the surface of the solid
phase; therefore, a good understanding of the nature of
the surface is required. Where does the initial attack
occursat the copper, at silicon, or at the silicon-copper
boundary? What happens then? What surface intermediates are involved? How are the products formed? What
determines the selectivity? There are many questions,
all difficult to answer. Collaborative surface science
analytical studies by Lewis and co-workers at the Union
Carbide Silicones Division Research Laboratory and
Falconer and co-workers at the University of Colorado
in Boulder62 provided strong evidence that the CH3Cl
+ Si/Cu direct synthesis proceeds by way of surfaceconfined silylenes: CH3SiCl as the (CH3)2SiCl2 precursor and SiCl2 as the precursor for CH3SiCl3, with all
other organosilicon products except (CH3)4Si originating
from silylene intermediates. Key experimental support
was provided by the observation, by mass spectrometry,
of a desorbed species of mass 78 (CH3SiCl) following
CH3Cl chemisorption on silicon-enriched surfaces of a
Si/Cu catalyst.
Further scale-up at GE was a chemical engineering
project, and to accomplish this task, Charles A. Reed,
an assistant professor of chemical engineering at MIT,
was hired in 1942. A larger pilot plant was built in the
Schenectady laboratory which contained 12 vertical
reactors, each of which produced 5.5-6.5 lb/h of operation and allowed a thorough process study to be carried
out. The separation of the pure individual methylchlorosilanes was an important and very difficult problem
because of their close boiling points, as noted above. In
his book,41a Liebhafsky describes the development of the
distillation process as unexpectedly painful, frustrating
and costly. However, in the end, the individual methylchlorosilanes were obtained in the required purity.
In December 1943, GE made the decision to go into
the silicones business: a silicone production and pilot
plant facility was to be built. The plant went on stream
in Waterford, NY, in 1947, seven years after Rochows
first experiment, with Reed as first General Manager
of the GE Silicone Products Department. At the begin(62) (a) Lewis, K. M.; McLeod, D.; Kanner, B.; Falconer, J. L.; Frank,
T. C. In Catalyzed Direct Reactions of Silicon; Lewis, K. M., Rethwisch,
D. G., Eds.; Elsevier: Amsterdam, 1993; Chapter 16. (b) Lewis, K. M.
Abstracts, 34th Organosilicon Symposium, White Plains, NY, May 3-5,
2001; pp HR-2 to HR-6.

Table 2. Some Applications of Silicones (from Ref

41a with Permission of John Wiley & Sons, Inc.)
Automotive
special lubricants
hydraulic bumpers
truck hose

wire insulation
transmission seals
spark-plug boots

Electrical/Electronic
motor and transformer insulation transistor encapsulants
wire and cable insulation
circuit encapsulants
circuit board laminates
television insulation
telephone wire connectors
rubber tapes (adhesive)
aircraft seals
firewall insulation

Military/Aerospace
special lubricants
heat shields
Paper

antistick surfaces

process defoamers
Textiles
dyeing-process defoamers

water repellents
fabric softeners

Rubber
tire release coatings
Food
coffee defoamers
bread pan coatings

milk-carton release coatings

cooking-process defoamers

Construction
window and building sealants
weather-durable paints
roof coatings
heat-resistant paints
masonry water repellents
furniture molding
vinyl shoe molding
RTV sealants
tile grout
shoe water repellents

Plastic Tooling
jewelry molding
Consumer Products
eye-glass tissues
lubricant sprays

Chemical Specialties and Cosmetics

auto and furniture polish
hand creams
antiperspirants
bath oils
hair sprays
foaming agents
prostheses
artificial organs and skin
facial reconstruction

Medical
contact lenses
catheters
drug delivery systems

ning, the direct reaction of CH3Cl with Si/Cu was carried

out in a stirred-bed reactor (Figure 10) and later using
a fluidized bed reactor (Figure 11). The separation of
the low-boiling products of the direct synthesis could
be accomplished satisfactorily in the plant, despite the
close boiling points of the components (a schematic
diagram of the distillation unit used is shown in Figure
12). Dimethyldichlorosilane could be obtained in 99.9%
purity. This is important for its further utilization in
the production of linear silicone fluids and silicone
rubber gum, where the presence of the trifunctional
CH3SiCl3 or the tetrafunctional SiCl4 would result in
unwanted cross-linking. The other methylchlorosilanes
have found various uses, but it is (CH3)2SiCl2, our cover
molecule, which is the most important: it is the molecule
which made the silicone industry a reality.
GE was not alone in the silicones business. Dow
Corning, because of the initial military applications,
with decisive support from the US Navy, had the plant
and the opportunity to start producing and selling
silicones before GE began to do so. Initially, their
methylchlorosilanes were prepared on a plant scale by
the Grignard route, but later they licensed GEs direct

4992

Organometallics, Vol. 20, No. 24, 2001

process technology. Over the years, their chemists have

made outstanding and important contributions to the
progress of organosilicon chemistrysin research and in
the development of many new and useful silicone
products. They became the largest silicone producer in
the world.

Organometallics 2001.20:4978-4992.
Downloaded from pubs.acs.org by 132.247.249.242 on 08/31/15. For personal use only.

Union Carbide Corp. also entered the silicones business, as did companies in other countries (among them
Wacker-Chemie, Bayer, Shin-Etsu, Toshiba Silicones,
and Rhone-Poulenc). The direct synthesis today is being
practiced all over the world: in the USA, Japan,
Germany, France, Russia, the UK, India, Australia, and
China. Over the years, the silicones, as a result of their
useful and, in many cases, unique physical and chemical
properties (Table 1), have found many important ap(63) Yajima, S. Am. Ceram. Soc. Bull. 1983, 62, 893.
(64) Burkhard, C. A. J. Am. Chem. Soc. 1949, 71, 963.
(65) (a) Kruger, C. R.; Rochow, E. G. J. Polym. Sci. A 1964, 2, 3179.
(b) Kruger, C. R.; Rochow, E. G. Angew. Chem., Int. Ed. Engl. 1962, 1,
458. (c) Rochow, E. G. Monatsh. Chem. 1964, 95, 750.
(66) Readers who wish to read more about the silicones and their
applications and about organosilicon chemistry in general are referred
to the following books: (a) Reference 52. (b) Eaborn, C. Organosilicon
Compounds; Butterworth: London, 1960. (c) Petrov, A. D.; Mironov,
V. F.; Ponomarenko, V. A.; Chernyshev, E. A. Synthesis of Organosilicon Monomers; Consultants Bureau: New York, 1964. (d) Noll, W.
Chemistry and Technology of the Silicones; Academic Press: New York,
1968. (e) Brook, M. A. Silicon in Organic, Organometallic and Polymer
Chemistry; Wiley: New York, 2000. (f) The Chemistry of Organic
Silicon Compounds; Patai, S., Rappaport, Z., Eds.; Wiley: Chichester,
U.K., 1989; Vol. 1. The Chemistry of Organic Silicon Compounds;
Rappaport, Z., Apeloig, Y., Eds.; Wiley: Chichester, U.K., 1998; Vol.
2. For a more complete listing, see ref 66e, pp 5-9. Reference 66d gives
a good account of the utilization of (CH3)2SiCl2 in the preparation of
silicones: its hydrolysis and the further processing of the resulting
HO(SiMe2O)nH linear polymers and (Me2SiO)n cyclic oligomers to give,
ultimately, the many diverse useful silicone and silicone-derived
products.

plications in industry and commerce and also in our

everyday lives (Table 2).
Today, 61 years later, Rochows little acorn has grown
into a mighty oak indeed. The annual production of
(CH3)2SiCl2 is around 1.4 million metric tons. The annual
production of the derived [(CH3)2SiO]n hydrolysate is
around 800 000 metric tons; this has a value of 3-5
billion dollars. However, (CH3)2SiCl2 has found many
uses outside of the silicone area: in the preparation of
silicon carbide fibers by a complex process,63 the first
step of which is its conversion to polydimethylsilylene,
[(CH3)2Si]n, by sodium condensation;64 its use in the
synthesis of polysilazanes (which was pioneered by
Rochow at Harvard),65 and its use as a starting material
in the synthesis of many thousands of organosilicon
compounds over the years throughout the world.66
Acknowledgment. My thanks go to Dr. Bela Prokai
and Dr. Kenrick M. Lewis of OSi Specialties for information on the silicones business, to Professor Hubert
Schmidbaur, Ms. Joyce W. Berger (ACS Library and
Information Center) and Ms. N. Best (RSC Library and
Information Centre), for photographs, to Professor Arnold L. Rheingold for the cover molecule picture, and
to Ms. Rhonda Saunders for the cover design.
Dietmar Seyferth
Department of Chemistry
Massachusetts Institute of Technology
Cambridge, Massachusetts 02139
Received October 17, 2001
OM0109051