Unsuccessful attempts to add alcohols to transient 2-amino-2-siloxy- silenes-leading to a new benign route for base-free alcohol protection

Article abstract of DOI:10.1039/c0dt00323a

Thermolytic formation of transient 1,1-bis(trimethylsilyl)-2-dimethylamino- 2-trimethylsiloxysilene (2) from N,N-dimethyl(tris(trimethylsilyl)silyl) methaneamide (1) in presence of a series of alcohols was investigated. The products are, however, not the expected alcohol-silene addition adducts but silylethers formed in nearly quantitative yields. Thermolysis of 1 in the presence of both alcohols (MeOH or iPrOH) and 1,3-dienes (1,3-butadiene or 2,3-dimethyl-1,3-butadiene) gives alkyl-tris(trimethylsilyl)silylethers and the [4+2] cycloadducts between the silene and diene, which confirms the presence of 2 and that it is unreactive towards alcohols. The observed silylethers are substitution adducts where the amide group of the silylamide is replaced by an alkoxy group, and the reaction time is reflected in the steric bulk of the alcohol. Indeed, the formation of silylethers from the reaction of alcohols with silylamide represents a new base-free method for protection of alcohols. The protection reactions using 1 progresses at elevated temperatures, or alternatively, under acid catalysis at ambient temperature, and similar protections can be carried out with N-cyclohexyl(triphenylsilyl)methaneamide and N,N-dimethyl(trimethylsilyl)methaneamide. The latter silylamide can be used under neutral conditions at room temperature. The only by-products are formamides (N,N-dimethylformamide (DMF) or N-cyclohexylformamide), and the reactions can be performed without solvent. In addition to alcohols we also examined the method for protection of diols, thiols and carboxylic acids, and also these reactions proceeded in high yields and with good selectivities. The Royal Society of Chemistry.

Full text of DOI:10.1039/c0dt00323a

View Article Online / Journal Homepage / Table of Contents for this issue
This article is published as part of the Dalton Transactions themed issue entitled:
New Horizon of Organosilicon Chemistry
Guest Editor: Professor Mitsuo Kira
Tohoku University, Japan
Published in issue 39, 2010 of Dalton Transactions
Image reproduced with the permission of Akira Sekiguchi
Articles in the issue include:
PERSPECTIVES:
Silylium ions in catalysis
Hendrik F. T. Klare and Martin Oestreich
Dalton Trans., 2010, DOI: 10.1039/C003097J
Kinetic studies of reactions of organosilylenes: what have they taught us?
Rosa Becerra and Robin Walsh
Dalton Trans., 2010, DOI: 10.1039/C0DT00198H
π-Conjugated disilenes stabilized by fused-ring bulky “Rind” groups
Tsukasa Matsuo, Megumi Kobayashi and Kohei Tamao
Dalton Trans., 2010, DOI: 10.1039/C0DT00287A
Organosilicon compounds meet subatomic physics: Muon spin resonance
Robert West and Paul W. Percival
Dalton Trans., 2010, DOI: 10.1039/C0DT00188K
HOT ARTICLE:
Novel neutral hexacoordinate silicon(IV) complexes with two bidentate monoanionic benzamidinato
ligands
Konstantin Junold, Christian Burschka, Rüdiger Bertermann and Reinhold Tacke
Dalton Trans., 2010, DOI: 10.1039/C0DT00391C
Visit the Dalton Transactions website for more cutting-edge inorganic and organometallic research
www.rsc.org/dalton

PAPER
www.rsc.org/dalton | Dalton Transactions
Unsuccessful attempts to add alcohols to transient 2-amino-2-siloxy-silenes -
leading to a new benign route for base-free alcohol protection†
Tamaz Guliashvili,a,b Julius Tibbelin,a Jiyeon Ryua and Henrik Ottosson*a
Received 17th April 2010, Accepted 6th July 2010
DOI: 10.1039/c0dt00323a
Thermolytic formation of transient 1,1-bis(trimethylsilyl)-2-dimethylamino-2-trimethylsiloxysilene (2)
from N,N-dimethyl(tris(trimethylsilyl)silyl)methaneamide (1) in presence of a series of alcohols was
investigated. The products are, however, not the expected alcohol-silene addition adducts but silylethers
formed in nearly quantitative yields. Thermolysis of 1 in the presence of both alcohols (MeOH or
iPrOH) and 1,3-dienes (1,3-butadiene or 2,3-dimethyl-1,3-butadiene) gives alkyl-tris(trimethylsilyl)-
silylethers and the [4+2] cycloadducts between the silene and diene, which confirms the presence of 2
and that it is unreactive towards alcohols. The observed silylethers are substitution adducts where the
amide group of the silylamide is replaced by an alkoxy group, and the reaction time is reflected in the
steric bulk of the alcohol. Indeed, the formation of silylethers from the reaction of alcohols with
silylamide represents a new base-free method for protection of alcohols. The protection reactions using
1 progresses at elevated temperatures, or alternatively, under acid catalysis at ambient temperature, and
similar protections can be carried out with N-cyclohexyl(triphenylsilyl)methaneamide and
N,N-dimethyl(trimethylsilyl)methaneamide. The latter silylamide can be used under neutral conditions
at room temperature. The only by-products are formamides (N,N-dimethylformamide (DMF) or
N-cyclohexylformamide), and the reactions can be performed without solvent. In addition to alcohols
we also examined the method for protection of diols, thiols and carboxylic acids, and also these
reactions proceeded in high yields and with good selectivities.
Introduction
Compounds with Si C double bonds, so-called silenes,1 are
highly air and moisture sensitive and this hampers the facile
application of reactions involving these compounds in target
directed synthesis. Often the regiochemistry of water and alcohol
additions is strongly in favor of the Si–O bonded adducts,2
and Veszpre´mi, Kira and co-workers verified through quantum
chemical calculations that water addition to the Si of the parent
silene is a strongly exothermic process that proceeds with a low
activation energy.3 The addition of the O atom to the C end, on
the other hand, was calculated to pass over a prohibitively high
barrier for the reaction to proceed at ambient temperature.
Whereas the parent silene readily adds water to the Si end it has
been concluded that reducing the Sid+ Cd- bond polarity is one
possible means to temper the reactivity of silenes.4,5 This reduction
of the Si C bond polarity can be effected through the reverse-
polarized resonance structures II and III of Scheme 1, and the
isolable silene with X = OSiMe3 and Y = 1-Ad formed by Brook
and co-workers in 1981 is a first example of such a silene.6 The
importance of reversed Si C bond polarization for the kinetic
stabilization of silenes was highlighted by Apeloig and co-workers
as they showed through computations that the activation energy
needed for water addition across the Si C bond, leading to the
Scheme 1
Si–O bonded adduct, increases gradually with the influence of
reversed polarization.7 Silenes influenced by reversed Sid- Cd+
polarization should therefore be less sensitive to moisture than
naturally Sid+ Cd- polarized silenes, possibly facilitating their
formation and handling by conventional synthetic techniques.
Indeed, Leigh and co-workers have verified in a series of studies
on transient silenes formed through laser flash photolysis that
their reactions with methanol markedly depends on the polarity
of the Si C bond.8 Whereas p-donor/s-acceptor substituents at
Si increase the electrophilicity of the Si C bond, p-acceptor/s-
donor substituents at Si and p-donor substituents at C have the
opposite effect.
Noteworthy, in all systems examined by Leigh and co-workers
the regiochemistry was in accord with the addition of the alkoxy
moiety to Si. Yet, it has been observed in two other cases that
water and alcohol additions to silenes can proceed with opposite
regiochemistry to the normal, yielding C–O instead of Si–O
bonded adducts.9,10 These findings were reported by Kira and co-
workers for the addition of methanol to a 4-silatriafulvene,9 and
by the group of Sekiguchi for water addition to a 1-silaallene.10
A common feature of the 4-silatriafulvenes and 1-silallenes is
that they are influenced by reversed Si C bond polarization,11,12
aDepartment of Biochemistry and Organic Chemistry, Uppsala University,
Box 576, 751 23, Uppsala, Sweden. E-mail: henrik.ottosson@kemi.uu.se;
Fax: +46 (0)18 471 3818; Tel: +46 (0)18 471 3809
bGE Power and Water, Water & Process Technologies, 4636 Somerton Road,
Trevose, PA, 19053, USA
† Electronic supplementary information (ESI) available: Additional exper-
imental details. See DOI: 10.1039/c0dt00323a
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 9379–9385 | 9379
©

but this influence is moderate as their Si atoms have planar
or nearly planar configurations and their Si C bond lengths,
to be carried out in solvent, and/or the reactions give by-
products that are not easily removed. An early example of a
more benign method is Kuwajima’s nearly solvent-free method
for alcohol and ketone protection using ethyl trimethylsilylacetate
(ETSA) with tetrabutylammonium-fluoride (TBAF) as catalyst.33
More recently, the base and catalyst free silylation using silyl
methallylsulfinates, described by Vogel,34 is promising, however,
gives sulfur dioxide as a by-product.
Herein, we report on studies of the fundamental reactivity
of a strongly reverse-polarized silene in comparison with the
reactivities of naturally polarized silenes. A spin-off finding from
our observations is that the unexpected reactions likely can be
exploited in synthesis in a wider sense.
˚
measured as 1.704–1.755 A, are closer to those of normal Si
C
˚
double bonds than normal Si–C single bonds (~1.70 and ~1.87 A,
respectively).13,14 This is in contrast to a silene which is strongly
influenced by the reverse polarized resonance structures II and III
because this species will be Si–C single rather than Si C double
bonded and its Si should be markedly pyramidal.5 Silenolates
are closely related to such strongly reverse polarized silenes as
their negative charge is predominantly localized to the Si atom,5,15
verified through our crystal structure of [18]crown-6 potassium
1,1-bis(trimethyl)silyl-2-tert-butylsilen-2-olate.16
Earlier, we formed the transient 1,1-bis(trimethylsilyl)-2-
dimethylamino-2-trimethylsiloxysilene (2, Scheme 2) through
thermolysis of tris(trimethylsilyl)silyl-N,N-dimethylamide (1).17
Quantum chemical B3LYP/6-31G(d) calculations indicated that
this silene is dominated by resonance structures II and III as it has
Experimental
The starting materials for the silylamides and the alcohols, thiols,
diols, and carboxylic acids used in the study were commercially
available, as were the solvents used. All chemicals were used as
purchased except THF and toluene which were dried over sodium
with sodium benzophenone ketyl as indicator, and CH2Cl2 which
was dried over calcium chloride. All reactions were performed
under inert atmosphere (argon). All purified compounds below
were purified on preparative TLC plates or by silica column
chromatography, depending on the amount of product (eluent:
pentane–EtOAc (98 : 2)), if nothing else noted.
˚
an Si–C bond length of 1.870 A and a pyramidal Si with Ra(Si) of
336.6◦. Hence, 2 is a suitable species for studying the reactivity of
a silene strongly influenced by reversed SiC bond polarity toward
alcohols. At this point it should be noted that strongly reverse-
polarized silenes formed through [1,3]-silyl shifts from suitable
acylsilane precursors will likely only be transient species as the
barrier for the back-transfer of the TMS group will be low as a
result of the strong single bond character of the formal Si
double bond.18
C
The (thermolytic) reactions of silylamides with alcohols, thiols,
diols and carboxylic acids (substrates) were performed according
to either of nine different reaction protocols listed below. Further
details of the protocols are given in the supporting information. In
all protection reactions similar amounts of formamides and pro-
tected substrates were formed according to 1H NMR spectroscopy.
Protocol 1: One equivalent of N,N-dimethyl(tris(trimethyl-
silyl)silyl)methaneamide (1) and four equivalents of alcohol were
placed in a dry NMR tube and dissolved in toluene-d8. The NMR
tube was sealed and placed in an oil bath and heated to 120
◦C. The complete consumption of 1 was monitored by 1H NMR
spectroscopy.
Protocol 2: The same reaction procedure as in protocol 1 was
also performed at 180 ◦C.
Protocol 3: The same reaction and amounts as in protocol 1, with
one drop of triflic acid added. The reaction started immediately at
ambient temperature.
Protocol 4: The same reaction as protocol 3, with the difference
that one drop Hu¨nig’s base (iPr2EtN) was used instead of acid.
Higher temperature (120 ◦C) was needed to initiate the reaction.
Protocol 5: One equivalent of alcohol and 1.3 equivalents of
silylamide 1 were placed in sealed tubes containing toluene, and
the reactions were run at 180 ◦C. The fastest protections (< 7.5 h)
were also performed in refluxing toluene for one night, with yields
similar as when run in sealed tubes.
Scheme 2
Whereas applications of unsaturated Si compounds in target-
directed synthesis is an emerging research area of current
organosilicon chemistry,19 the applications of various silyl reagents
for protection of alcohols, thiols, and carboxylic acids is presum-
ably the most well-explored and mature topic within the field.20,21
As will be shown, the results given herein can, however, also
expand this last field by providing a benign and base-free alcohol
protection protocol.
The most common protection methods use trialkylsilyl halides
or triflates,19–22 and reactions with these reagents are generally
base-promoted or sometimes catalysed by, e.g., magnesium or
lithium sulfide. A range of other silyl reagents exist,23–32 but
nearly all of these reagents require a catalyst, or they need
Protocol 6: Same conditions as in protocol 5, except that a
temperature of 240 ◦C was used instead of 180 ◦C.
Protocol 7: Monofunctionalized substrates were protected
at ambient temperature with 1.3 equivalent of N,N-dimethyl-
(trimethylsilyl)methaneamide (10), whereas difunctionalized sub-
trates were protected with 1.1 equivalents of 10. The reactions
were performed in sealed NMR-tubes at ambient temperature in
CH2Cl2 and with a few drops of CDCl3 added. The protected
9380 | Dalton Trans., 2010, 39, 9379–9385
This journal is
The Royal Society of Chemistry 2010
©

Table 1 Reactions of alcohols with N,N-dimethyl(tris(trimethylsilyl)-
silyl)methaneamide (1)a
substrates were not purified due to the poor stability towards
aqueous and silica conditions of a few of them. A few also
possess low boiling points. Moreover, a limited selection of the
reactions were performed neat, and in other solvents such as THF,
pentane, toluene and diethyl ether with similar reaction times and
conversions as reported for the reactions in CH2Cl2.
Protocol 8: The same reaction and reaction conditions as in
protocol 7, with DMF instead of CH2Cl2 as solvent.
Protocol 9: Protection with the triphenylsilyl group. N-
cyclohexyl(triphenylsilyl)methaneamide (11) was applied in
1.3 times excess over the substrates, and the reactions were run
in toluene in 1.5 ml sealed tubes at 180 ◦C.
ROH
T/◦C
Time/h
Yieldb (%) NMR/isolated
MeOH (a)
MeOH (a)
EtOH (b)
EtOH (b)
iPrOH (c)
iPrOH (c)
tBuOH (d)
PhOH (e)
PhOH (e)
BnOH (f)
BnOH (f)
AllylOH (g)
AllylOH(g)
120
180
120
180
120
180
180
120
180
120
180
120
180
4
0.5
8
0.8
11
1.1
352
1.1
0.5
1
96/92
95/92
94/90
93/89
93/87
92/85
—/18
—/40
—/60
76/68
89/84
70/63
85/78
0.25
1
0.25
Results and discussion
a Conditions: Sealed NMR tubes, toluene-d8, [silylamide]:[ROH] = 1 : 4
(protocol 1 (120 ◦C) and protocol 2 (180 ◦C)). b Conversions determined
by 1H NMR spectroscopic measurements on sealed NMR tubes, and yield
after purification by preparative TLC.
The
methylsiloxysilene (2, Scheme 2) was formed thermally from
N,N-dimethyl-tris(trimethylsilyl)silylmethaneamide (1) by
transient
1,1-bis(trimethylsilyl)-2-dimethylamino-2-tri-
heating at either 120 or 180 ◦C. Formation of the silene at the
higher temperature was previously verified through trapping
with 2,3-dimethyl-1,3-butadiene giving the silacyclohexene 4
through a [4+2] cycloaddition reaction between the silene and
1,3-diene followed by a permutation of the substituents of the
initial [4+2] cycloadduct 3 (Scheme 2).17,35 We could now prove
formation of 2 also at the lower temperature as the silene was
trapped with 1,3-butadiene, a diene which is more reactive in
the [4+2] cycloadditions with 2-amino-2-siloxy-silenes than 2,3-
dimethyl-1,3-butadiene, thus indicating that the reactions follow
the reactivity pattern of inverse-electron-demand Diels–Alder
reactions.38
Since silene 2 is formed at both temperatures we first carried
out the attempted alcohol-silene reactions at 120 ◦C as this should
lead to less by-products. The reactions were carried out in toluene
solutions in sealed tubes containing 1 and either methanol (a),
ethanol (b), iso-propyl alcohol (c), tert-butyl alcohol (d), phenol
(e), allyl (f) or benzyl alcohol (g). The ratio of the silylamide
to alcohol was set at 1 : 4, and the progress of the reactions
1
was monitored by H and 13C NMR spectroscopy. As seen in
Table 1, the reactions with all alcohols except tBuOH and phenol
were clean yielding only single products in yields that are nearly
quantitative. However, the reaction times vary significantly and
correlate with the steric bulk of the alcohol.
The identities of the reaction products were next to be deter-
mined, and surprisingly these do not correspond to the expected
adducts 6 and/or 7 resulting from trapping of the silene by the
alcohols. Instead, the major products are in all cases, silylethers 5
resulting from replacement of the amide group in 1 by the alkoxy
group of the alcohol (Scheme 3). The yields of the silylethers
reported in Table 1 were determined both from 1H NMR spectra of
the sealed NMR tubes as well as on the isolated compounds. Apart
from silylethers, formation of N,N-dimethylformamide (DMF)
in very similar yields to the silylethers was also observed. With
regard to tBuOH, this alcohol reacted considerably slower than
the others. In contrast, phenol is the alcohol which is most rapidly
consumed, but its yield for the silylether formation is lower than
for the aliphatic alcohols. The side-products could not be fully
characterized, but could result from trapping of silene 2 by phenol
yielding 6 and/or 7. Here, it should be noted that phenol which
Scheme 3
reacts the fastest also is more acidic than the aliphatic alcohols, a
factor that influences the reaction rate (vide infra).
In order to probe if silene 2 is formed at all during the thermol-
ysis of 1 when alcohols are present we carried out thermolysis run
in precence of both iPrOH and 2,3-dimethyl-1,3-butadiene in a
1.1 : 10 ratio. Indeed, both iso-propyl-tris(trimethylsilyl)silylether
and the [4+2] cycloadduct between silene and diene could be
observed in yields of 66 and 27%, respectively. No other side-
products were observed, and it can thus be concluded that the
silene is present but that the activation energies for the two possible
iPrOH additions to the silene, yielding 6 and 7, are so high that
these reactions do not occur. Instead, the silene reacts with the
diene and the silylamide reacts with iPrOH (giving 5). A similar
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 9379–9385 | 9381
©

experiment in presence of both methanol and 1,3-butadiene was
also performed, giving both silylether and [4+2] cycloadduct.
To our knowledge the formation of silylethers 5 instead of the
silene trapping adducts 6 and 7 represents the first example where a
silene is formed which does not react with an alcohol. The reduced
reactivity of silene 2 towards alcohols is in line with previous
experimental observations of Leigh and co-workers that the rate
of methanol addition to silenes decreases gradually with increased
p-donor ability of the substituents at the C terminus. The finding
also agrees with the conclusions of Apeloig and co-workers that the
activationenergyfor water additiontosilenes is stronglyinfluenced
by the degree of reversed Si C bond polarity since it becomes
successively higher when resonance structures II and III increase in
importance.7 Clearly, in the transient silene 2 the extent of reversed
Si C bond polarization has reached a point where the rate for
alcohol addition is so slow that it cannot compete with the reaction
of the silylamide in which the C( O)NMe2 moiety of 1 is replaced
by a methoxy group.
Realizing the poor leaving group ability of the amide group
the substitution reaction is most likely not a regular nucleophilic
substitution reaction but a process that proceeds by a more
complex mechanism, presumably coupled with initial protonation
of 1. Indeed, it is likely that protonation of 1 is the rate limiting
step because the reaction rate is increased very significantly when
triflic acid is added to a toluene solution of 1 and MeOH in
1 : 4 : 0.001 ratio of 1, MeOH and acid. Interestingly, this reaction
ran at room temperature and was completed within 30 min to
give the silylether in equally high yield as when run without acid
at high temperature. However, the reaction without acid needs
4h at 120 ◦C for completion. As can be expected, addition of
base (iPr2EtN) to the reaction of 1 with MeOH lead to a reduced
reaction rate, supporting that protonation of 1 is an important
step in the mechanism.
An in-depth analysis of the mechanism is outside the scope of
the present report, and results from computational studies will
be given in due course. However, based on the earlier finding of
Apeloig and co-workers in a gradual increase in the activation
energies for ROH addition as the silenes become gradually more
reverse polarized,7 it can easily be realized that the barrier for
ROH addition to the 2-amino-2-siloxysilenes will be very high, so
high that other reactions occur instead. At this point, it should
be remarked that Leigh et al. recently showed that the addition of
MeOH to three different 1,1-disilylsilenes proceed via methanol
dimers rather than via monomers.36 Still, strongly reverse polarized
silenes such as 2-amino-2-siloxysilenes do not exhibit the marked
electrophilic character of naturally polarized silenes. This implies
a low moisture sensitivity in line with our earlier observation
that the transient 1,1-bis(trimethyl-silyl)-2-N,N-diphenylamino-
2-tri-methylsiloxysilene (9) can be formed and trapped with 2,3-
dimethyl-1,3-butadiene through reflux in non-distilled (wet) THF
giving the corresponding silacyclo-hexene (cf . 4, Scheme 2) in the
same high yield as when run in dry toluene in a sealed tube with
inert atmosphere.17
silyl)silyl)methaneamide (8) in presence of alcohols can yield also
the silene-alcohol adducts in addition to the silylethers. Indeed,
when carrying out the thermolysis of 8 at 120 ◦C together
with methanol at a silylamide to alcohol ratio of 1 : 4, the
formation of the silylether was not observed at all. However, the
reaction mixture was too complex to separate, possibly a result of
degradation of silene-alcohol adducts, compounds that are heavily
substituted and that therefore could be prone to rearrange and/or
decompose. Interestingly, when 2,3-dimethyl-1,3-butadiene is also
included so that the ratio between the silylamide, methanol, and
diene is 1 : 2 : 2, then the [4+2] cycloadduct is exclusively formed.
Clearly, the reaction between the silene and the diene is more rapid
than the reaction between the silene and methanol. This finding is
in line with our earlier observation that silene 9 could be formed
and trapped with 2,3-dimethylbutadiene to the cycloadduct under
moist conditions.17
The silylethers 5 formed through reaction between 1 and
various alcohols are also interesting in their own respect as the
tris(trimethylsilyl)silyl group, the sisyl group, has been shown by
Brook and co-workers to be a protecting group of primary and
secondary alcohols which is resistant towards fluoride deprotect-
ing reagents such as KF in 18-crown-6 and CsF.38,39 However,
the TMS3Si group is easily removed photolytically (l = 254
nm), giving it an orthogonal reactivity to other silyl groups.
Thus, our unsuccessful attempts to add ROH to the transient
and strongly reverse polarized 2-amino-2-siloxysilene 2, formed
thermally from silylamides, instead represent a new protocol for
protecting alcohols with a synthetically interesting silyl group.
However, when we examined the reactivity of the transient silene
toward ROH we carried out these reactions in fourfold excess of
alcohol when compared to silylamide, whereas for the subsequent
alcohol protections we used the silylamide in slight excess. For
the monofunctional substrates (alcohols, thiols, and carboxylic
acid) we used a 1.3 : 1 ratio of silylamide vs. substrate, whereas for
the difunctional substrates we used a ratio of 1.1 : 1. In addition
to the alcohols of Table 1 we also included octanol (h) and 1-
methylcyclohexanol (i), as well as isovaleric acid as a carboxylic
acid (j) (Table 2).
As shown in Table 2, the reactions all proceed in reason-
able times, and in all cases the conversions were relatively
consistent on scales ranging from 20 to 500 mg. A potential
Table 2 Protection of various alcohols and isovaleric acid using N,N-
dimethyl(tris(trimethylsilyl)silyl)methaneamide (1)a
Substrate
Time/h
Yield (%)
MeOH (a)
2
6
93
EtOH (b)
96
iPrOH (c)
27
89c
tBuOH (d)
0.17b
96b,c
73, 94b
93c
PhOH (e)
3, 0.17b
AllylOH (f)
BnOH (g)
4
4
7
91c
Octanol (h)
1-Methylcyclohexanol (i)
Isovaleric acid (j)
91c
0.17b
2
92b
95c
However, with regard to silene 9 we previously found its
NPh2 group to be rotated partially out of conjugation with
the Si C double bond so that this silene is less influenced by
reversed Si C bond polarity than 2.37 It should therefore have
a more electrophilic Si. Of this reason, one could expect that
thermolytic formation of 9 from N,N-diphenyl(tris(trimethyl-
a Conditions: Sealed NMR tubes, toluene-d8, [silylamide]:[ROH] = 1.3 : 1.0
run at 180 ◦C (protocol 5). b Microwave irradiated to 240 ◦C (protocol 6).
1
c Conversion determined through H NMR spectroscopic measurements
on sealed NMR tubes.
9382 | Dalton Trans., 2010, 39, 9379–9385
This journal is
The Royal Society of Chemistry 2010
©

Table 3 Conversions and reaction times for protections of different substrates using 10 in CH2Cl2 or DMF at ambient temperature
Substrate
CH2Cl2 Con./Sel. (%)b
Time/h
DMFa Con./Sel. (%)b
Time/h
iPrOH (c)
94/—
97/—
99/—
90/—
97/—
87/—
96/—
87/—
84/—
96/85
95/82
86/83
93/83
25
48
30
15
13
18
53
6
40
27
15
28
20
96/—
92/—
81/—
86/—
91/—
88/—
86/—
91/—
82/—
91/82
94/82
83/78
87/76
28
49
33
19
16
15
56
9
42
17
16
28
19
tBuOH (d)
PhOH (e)
AllylOH (f)
BnOH (g)
Octanol (h)
1-Methylcyclohexanol (i)
Isovaleric acid (j)
nBuSH (k)
PhCHOHCH2OHc (l)
4-HOC6H4CH2OHc (m)
HOCH2CH2OHc (n)
HSCH2CH2OHc (o)
a A small amount of CDCl3 was added to each sample, to allow the reaction to be followed by NMR. b Monofunctional substrate run at [substrate]:[10] =
1 : 1.3, and difunctional substrates run at [substrate]:[10] = 1 : 1.1 (protocols 7 and 8). c In all cases main selectivity for primary alcohol.
Table 4 Protection of alcohols using N-cyclohexyl(triphenylsilyl)-
methaneamide (11) at 180 ◦Ca
ROH
Time/h
Yield (%)
MeOH (a)
EtOH (b)
iPrOH (c)
PhOH (e)
AllylOH (f)
BnOH (g)
Octanol (h)
1
94
93
88
85
90
87
93
1.5
2.5
2.5
1.75
1.75
2
a Conditions: Sealed NMR tubes, toluene-d8, [silylamide]:[ROH] = 1.3 : 1.0
Scheme 4 Protection of alcohols, thiols, and carboxylic acids using three
different silylamides.
run at 180 ◦C (protocol 9).
The protections of the alcohols, phenols, carboxylic acids, and
thiols with 10 were all performed at room temperature giving good
to excellent conversions (Table 3). Isovaleric acid reacted fastest
of the thirteen substrates (6–9 h), in line with the observation
presented above that acid catalyzes the reaction. While the tertiary
alcohols reacted slowest (48–56 h), their conversions remained
very high.
drawback is the elevated temperatures that are needed, but the
procedure avoids base which is normally required for alcohol
protection using silylchlorides or silyltriflates.40 Alternatively, and
as shown above for the protection of MeOH, the reactions can
be run at room temperature using acid catalysis. Indeed, the
protection of tBuOH with 1 can be facilitated by acid as the
reaction between 1, tBuOH, and triflic acid in a ratio of 1 : 4 : 0.001
ran to completion in 33 h at room temperature giving a 56%
yield.
With regard to the protection of the two diols, 1-phenyl-
ethane-1,2-diol and 4-(hydroxymethyl)phenol, high conversions
and good selectivities were achieved with preferred protection
of the primary alcohol. Protection of 1-phenyl-ethane-1,2-diol
in CH2Cl2 gave good conversion and selectivity for the primary
alcohol (96 and 85%, respectively). The result was similar for the
trimethyl-silylation of 4-(hydroxymethyl)phenol. The protection
of the two difunctionalised compounds, hydroxy acetic acid and
2-mercaptoethanol gave mainly protection of the alcohol moiety
in high yields. The protection of hydroxy acetic acid proceeds via
the silylester (kinetic product), which rearrange to the protected
alcohol (thermodynamic product), according to 1H NMR.
The reactions listed in Table 3 were performed in dry CH2Cl2
or DMF. A few of the substrates were also protected in other dry
solvents such as THF, toluene, acetonitrile and chloroform, as
well as neat, with similar conversions. As seen in Table 3 a slight
decrease in yield is found when changing solvent from CH2Cl2
to DMF. The only by-product is DMF, i.e., the solvent when the
reaction is performed in this medium.
Noteworthy, when run at neutral conditions at 180 ◦C the
yields for silylether formation are slightly higher than observed
previously (73–97%),38 although the earlier studies were carried
out at room temperature. With regard to the phenol and tertiary
alcohol protections by 1 under neutral conditions, excellent yields
are obtained after 10 min of microwave heating at 240 ◦C in
toluene. Finally, yields similar to those above are also obtained
for the protection of primary alcohols, phenols and carboxylic
acids by 1 when refluxed in toluene for one day.
Although the tris(trimethylsilyl)silyl group has interesting fea-
tures, the regular protecting groups are the trialkylsilyl groups.
To explore if our findings of a base-free protocol for protection
of alcohols can be expanded also to these groups we tested the
use of trimethylsilyl-N,N-dimethylamide 10 and triphenylsilyl-
N-cyclohexylamide 11 (Scheme 4).41,42 For 10 we expanded the
substrates examined so that we tested protection of alcohols, diols,
thiols and carboxylic acids (Table 3), whereas for 11 we used a more
limited selection (Table 4).
Reaction with silylamide 11, having a bulky Ph3Si group,
was performed in sealed NMR or microwave tubes at elevated
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 9379–9385 | 9383
©

temperatures above 100 ◦C (Table 4).† Higher temperatures are
required in order to achieve good conversions within reasonable
time as the t1/2 of 11 (~24 h) sets a limit for the reaction time. We
carried out the reactions with 11 at 180 ◦C to achieve reasonable
timings for most of the alcohols in the study. Reactions with 11
give N-cyclohexylformamide which is also easy to remove by water
extraction, preparative TLC, or silica gel chromatography.
There are some general trends for the three silylamides used here.
First, the yields of the protected alcohols at the given temperatures
are similar for the different silylamides independent of substrate.
Secondly, the trends in the relative reaction rates of the protection
reactions are similar for the three silylamides at given conditions
(room temperature for 10 and 180 ◦C for 11 and 1) and generally
follow the order: carboxylic acid > BnOH ~ allylOH > primary
alcohol > secondary alcohol > PhOH > thiol > tertiary alcohol.
However, for 1 the rate for protection of the secondary alcohol is
substantially slower than that of phenol.
alcohol protection chemistry and found to give positive results. In
conclusion, silylamides provide advantages compared to existing
alcohol protecting reagents since they allow for base and catalyst
free conditions, and their reactions give either solvent (DMF)
or other easily removable and benign by-products. The conver-
sions/yields are high to excellent, and the smallest silylamide (10)
used shows high selectivity for primary alcohols.
Acknowledgements
We first acknowledge Dr. Patrick Steel for many stimulating dis-
cussions on the topics of silenes and alcohol protections. Financial
support from the Swedish Research Council (Vetenskapsra˚det) is
gratefully appreciated.
Notes and references
As noted above, we do not yet have a full view of the
mechanism of these reactions, although protonation is clearly
a key step in the process. At a first glance, there are sim-
ilarities to the Si–(C O) bond cleavage of N-methyl-N-2,6-
dimethylphenyltriethylsilylamide reported by Murai et al., how-
ever, this process is base promoted.43 The same applies to
the finding of Brook and Gilman from 1955 on the reaction
of triphenylsilanecarboxylate with catalytic amounts of sodium
methoxide or ethoxide in the respective alcohol which gives the
corresponding alkoxytriphenylsilane in good yields besides carbon
monoxide and sodium methoxide (Scheme 5).44 However, both
these procedures require the use of a base, whereas our process is
catalyzed by acid, indicating distinctly different mechanisms.
1 For reviews on silenes see e.g. A. G. Brook and M. Brook, Adv.
Organomet. Chem., 1996, 39, 71; T. Mu¨ller, W. Ziche and N. Auner, in
The Chemistry of Organic Silicon Compounds, Vol. 2 (ed.: Z. Rappoport,
Y. Apeloig), Wiley Interscience, New York, 1998, pp. 857; T. L. Morkin
and W. J. Leigh, Acc. Chem. Res., 2001, 34, 129; L. E. Gusel’nikov,
Coord. Chem. Rev., 2003, 244, 149; H. Ottosson and P. G. Steel, Chem.–
Eur. J., 2006, 12, 1576; H. Ottosson and A. M. Eklo¨f, Coord. Chem.
Rev., 2008, 252, 1287.
2 H. Sakurai, in in The Chemistry of Organic Silicon Compounds, Vol.
2 (ed.: Z. Rappoport, Z., Y. Apeloig), Wiley Interscience, New York,
1998, pp. 827.
3 T. Veszpre´mi, M. Takahashi, B. Hajgato and M. Kira, J. Am. Chem.
Soc., 2001, 123, 6629.
4 Y. Apeloig and M. Karni, J. Am. Chem. Soc., 1984, 106, 6676.
5 H. Ottosson, Chem.–Eur. J., 2003, 9, 4144.
6 A. Brook, G. Abdesaken, B. Gutekust and R. K. Kallury, J. Chem. Soc.,
Chem. Commun., 1981, 191; A. G. Brook, S. C. Nyburg, F. Abdesaken,
B. Gutekunst, G. Gutekunst, R. Krishna, M. R. Kallury, Y. C. Poon,
Y. M. Chang and W. N. Winnie, J. Am. Chem. Soc., 1982, 104, 5667.
7 M. Bendikov, S. R. Quadt, O. Rabin and Y. Apeloig, Organometallics,
2002, 21, 3930.
8 W. J. Leigh, R. Boukherroub and C. Kerst, J. Am. Chem. Soc., 1998,
120, 9504; W. J. Leigh, C. Kerst, R. Boukherroub, T. L. Morkin, S. I.
Jenkins, K. Sung and T. T. Tidwell, J. Am. Chem. Soc., 1999, 121, 4744.
9 K. Sakamoto, J. Ogasawara, Y. Kon, T. Sunagawa, C. Kabuto and M.
Kira, Angew. Chem., Int. Ed., 2002, 41, 1402.
10 M. Ichinohe, T. Tanaka and A. Sekiguchi, Chem. Lett., 2001, 104, 4329.
11 K. Sakamoto, J. Ogasawara, H. Sakurai and M. Kira, J. Am. Chem.
Soc., 1997, 119, 3405; T. Veszpre´mi, M. Takahashi, J. Ogasawara,
K. Sakamoto and M. Kira, J. Am. Chem. Soc., 1998, 120, 2408; T.
Veszpre´mi, M. Takahashi, B. Hajgato´, J. Ogasawara, K. Sakamoto
and M. Kira, J. Phys. Chem. A, 1998, 102, 10530; M. Takahashi, K.
Sakamoto and M. Kira, Int. J. Quantum Chem., 2001, 84, 198; K.
Sakamoto, J. Ogasawara, Y. Kon, T. Sunagawa, C. Kabuto and M.
Kira, Angew. Chem., 2002, 114, 1460.
Scheme 5 The study by Brook and Gilman of reactions of triphenyl-si-
lanecarboxylate with sodium alkoxides.44 Yields in parentheses.
Conclusions
The
transient
1,1-bis(trimethylsilyl)-2-dimethylamino-2-tri-
methylsiloxysilene (2) with a reversed Si C bond polarity was
formed through thermolysis of N,N-dimethyl(tris(trimethyl-
silyl)silyl)methaneamide (1) in presence of alcohols. Surprisingly,
no products resulting from reaction of the silene with the alcohol
were observed, and this should represent the first example
of a silene that is persistent to alcohols. The amide group of
the silylamide is instead replaced by the alkoxy group. When
thermolyses are performed in the presence of both an alcohol and
a 1,3-diene, both the silylether and the [4+2] cycloadduct of the
reaction between silene and diene are observed.
Indeed, the reaction of alcohols with N,N-dimethyl-
(tris(trimethyl-silyl)silyl)methaneamide represent a new base-free
pathway for protection of alcohols with the tris(trimethyl-silyl)silyl
group, a group that has shown to be a fluoride resistant but
photochemically removable protecting group of alcohols.38 Two
other silylamides were also tested in the reaction of base-free
12 G. E. Miracle, J. L. Ball, D. R. Powell and R. West, J. Am. Chem. Soc.,
1993, 115, 11598; M. Trommer, G. E. Miracle, B. E. Eichler, D. R.
Powell and R. West, Organometallics, 1997, 16, 5737.
13 The Si C double bond length in the parent silene H2Si CH2 is 1.7039
˚
A as measured by millimetre wave spectroscopy S. Bailleux, M. Bogey,
J. Breidung, H. Bu¨rger, R. Fajgar, Y. Liu, J. Pola, M. Senzlober and
W. Thiel, Angew. Chem., Int. Ed. Engl., 1996, 35, 2513; S. Bailleux, M.
Bogey, J. Demaison, H. Bu¨rger, M. Senzlober, J. Breidung, W. Thiel, R.
Fajgar and J. Pola, J. Chem. Phys., 1997, 106, 10016.
14 M. Kaftory, M. Kapon and M. Botoshansky, in The Chemistry of
Organic Silicon Compounds, Vol. 2 (ed.: Z. RappoportY. Apeloig),
Wiley, Chichester, UK, 1998, 857.
15 A. M. Eklo¨f and H. Ottosson, Tetrahedron, 2009, 65, 5521.
16 T. Guliashvili, I. El-Sayed, A. Fischer and H. Ottosson, Angew. Chem.,
Int. Ed., 2003, 42, 1640.
17 I. El-Sayed, T. Guliashvili, R. Hazell, A. Gogoll and H. Ottosson, Org.
Lett., 2002, 4, 1915.
18 A. M. Eklo¨f and H. Ottosson, Organometallics, 2008, 27, 5203.
9384 | Dalton Trans., 2010, 39, 9379–9385
This journal is
The Royal Society of Chemistry 2010
©

19 J. D. Sellars and P. G. Steel, Tetrahedron, 2009, 65, 5588; J. Bower, M. R.
Box, M. Czyzewski, A. E. Goeta and P. G. Steel, Org. Lett., 2009, 11,
2744; R. D. C. Pullin, J. D. Sellars and P. G. Steel, Org. Biomol. Chem.,
2007, 5, 3201; J. D. Sellars and P. G. Steel, Eur. J. Org. Chem., 2007, 3815;
N. J. Hughes, R. D. C. Pullin, M. J. Sanganee, J. D. Sellars and P. G.
Steel, Org. Biomol. Chem., 2007, 5, 2841; J. D. Sellars and P. G. Steel,
Org. Biomol. Chem., 2006, 4, 3223; J. D. Sellars, P. G. Steel and M. J.
Turner, Chem. Commun., 2006, 2385; M. B. Berry, R. J. Griffiths, M. J.
Sanganee, P. G. Steel and D. K. Whelligan, Org. Biomol. Chem., 2004,
2, 2381; M. J. Sanganee, P. G. Steel and D. K. Whelligan, Org. Biomol.
Chem., 2004, 2, 2393; M. B. Berry, R. J. Griffiths, M. J. Sanganee, P. G.
Steel and D. K. Whelligan, Tetrahedron Lett., 2003, 44, 9135; A. S.
Batsanov, I. M. Clarkson, J. A. K. Howard and P. G. Steel, Tetrahedron
Lett., 1996, 37, 2491.
28 T. Morita, Y. Okamoto and H. Sakurai, Tetrahedron Lett., 1980, 21,
835.
29 T. Suzuki, T. Watahiki and T. Oriyama, Tetrahedron Lett., 2000, 41,
8903.
30 X. Huang, C. Craita, L. Awad and P. Vogel, Chem. Commun., 2005,
1297.
31 L. Birkofer and P. Sommer, J. Organomet. Chem., 1975, 99, C1.
32 A. Iida, A. Horii, T. Misaki and Y. Tanabe, Synthesis, 2005, 2677 and
references there in.
33 E. Nakamura, T. Murofushi, M. Shimizu and I. Kuwajima, J. Am.
Chem. Soc., 1976, 98, 2346.
34 X. Huang, C. Craita, L. Awad and P. Vogel, Chem. Commun., 2005,
1297.
35 T. Guliashvili, Ph.D Thesis, Uppsala University, 2004.
36 W. J. Leigh, T. R. Owens, M. Bendikov, S. S. Zade and Y. Apeloig,
J. Am. Chem. Soc., 2006, 128, 10772.
37 H. Ottosson, T. Guliashvili and I. El-Sayed, in Organosilicon Chemistry
V: From Molecules to Materials, ed.: N. Auner and J. Weis, Wiley –VCH,
Weinheim, 2003, pp. 78.
20 P. J. Kocien´ski, Protecting Groups, Geroge Thime Verlag, Stuttgart,
New York, 3rd. edn 2005.
21 T. W. Greene and P. G. M. Wuts, Protective Groups in Organic Synthesis,
John Wiley & Sons, New York, 3rd. edn 1999.
22 G. A. Olah, A. Husain, B. G. B. Gupta, G. F. Salem and S. C. Narang,
J. Org. Chem., 1981, 46, 5212.
23 D. Amantini, F. Fringuelli, F. Pizzo and L. Vaccaro, J. Org. Chem.,
2001, 66, 6734.
24 F. Le Bideau, T. Coradin, J. He´nique and E. Samuel, Chem. Commun.,
2001, 1408.
38 M. A. Brook, C. Gottardo, S. Balduzzi and M. Mohamed, Tetrahedron
Lett., 1997, 38, 6997.
39 M. A. Brook, S. Balduzzi, M. Mohamed and C. Gottardo, Tetrahedron,
1999, 55, 10027.
40 M. A. Brook, Silicon in Organic, Organometallic, and Polymer Chem-
istry, Wiley Interscience, New York, 2000, pp. 198.
41 R. F. Cunico and J. Chen, Synth. Commun., 2003, 33, 1963.
42 J. E. Baldwin, A. D. Derome and P. D. Riordan, Tetrahedron, 1983, 39,
2989.
43 A. Orita, K. Ohe and S. Murai, Organometallics, 1994, 13, 1533.
44 A. G. Brook and H. Gilman, J. Am. Chem. Soc., 1955, 77, 2322.
25 M. Hayashi, Y. Matsuura and Y. Watanabe, Y., Tetrahedron Lett., 2004,
45, 1409.
26 Y. Tanabe, H. Okumura, A. Maeda and M. Murakami, Tetrahedron
Lett., 1994, 35, 8413.
27 Y. Tanabe, M. Murakami, K. Kitaichi and Y. Yoishida, Tetrahedron
Lett., 1994, 35, 8409.
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 9379–9385 | 9385
©