Silicon analogues of carboxylic acids: Synthesis of isolable silanoic acids by donor-acceptor stabilization

Article abstract of DOI:10.1002/anie.201002970

How to cope with instability? Storable complexes of silacarboxylic acids were prepared through addition of H2S and H2O B(C6F5)3, respectively, to the silanone precursor complex 1. Metalation of the OH group in the acid-base adduct 2 gave the novel silathi

Full text of DOI:10.1002/anie.201002970

Communications
DOI: 10.1002/anie.201002970
Silacarboxylate Ligands
Silicon Analogues of Carboxylic Acids: Synthesis of Isolable Silanoic
Acids by Donor–Acceptor Stabilization**
Yun Xiong, Shenglai Yao, and Matthias Driess*
In memory of Herbert Schumann and Pascal LeFloch
The carbonyl group in ketones, carboxylic acids, and related
compounds is one of the most widely occurring functional
groups in chemistry, biochemistry, and material science. In
=
contrast, its silicon analogues Si X (X = chalcogen) and their
=
related functional groups (e.g. Si( X)OH) are difficult to
realize or are even completely unknown in the condensed
ꢀ
phase owing to their intrinsically weaker pp pp bonding and
Scheme 1. N-donor-stabilized germanium and silicon analogues of
carboxylic acids A, B, and the silanoic amide C. X=O, S, Se;
R=2,6-iPr2C6H3.
=
higher polarities of the Si X subunit, which gives rise to facile
isomerization and self-polymerization reactions. However, in
the last decades, milestones have been achieved for the
synthesis and isolation of silanechalcogenones (heavier sila-
ketones) by taking advantage of thermodynamic and kinetic
stabilization. Among the early examples, isolable silaketone-
siloxy silylene precursors.[7d] Despite the presence of four-
coordinate silicon atoms, the latter esters possess significant
[7c]
=
=
=
like systems featuring donor-stabilized Si S and Si Se bonds
with four-coordinate silicon atoms have been reported by
Corriu and co-workers[1a] and more recently by the research
groups of Wagler,[1b] West,[1c] and Roesky.[1d] Moreover, even
silanechalcogenones with three-coordinate silicon atoms have
been synthesized and structurally characterized by the
research groups of Okazaki, Tokitoh, Iwamoto, Tꢀrring, and
Si X bond character.
Furthermore, we have synthesized peculiar types of
=
donor-supported Si X systems starting from the zwitterionic
=
N-heterocyclic silylene L’Si (L’ = CH[C(Me)(C CH2)(NR)2],
R = 2,6-iPr2C6H3).[7e] Thus, the latter species reacts directly
[8]
=
with H2S to give the isolable silathioformamide, LSi( S)H.
Moreover, starting from N-heterocyclic carbene (NHC)
adducts of L’Si, a series of NHC-stabilized heavier ketone
Kira.[2–5] However, stable compounds containing a Si O bond
=
=
and three-coordinate silicon atoms are still unknown. Like-
wise, experimental evidence about the existence of silacar-
analogues L’Si( X)(NHC) (X = O, S, Se, Te) have been
successfully synthesized, including a silanone adduct.[9a–b]
Accordingly, use of the readily accessible dmap complex of
L’Si (dmap = 4-dimethylaminopyridine) enabled us to syn-
thesize the stable silanone complex, L’Si(=O)(dmap) 1[9c]
(Scheme 2). Because the dmap ligand is only weakly coordi-
nated, compound 1 can serve as a convenient silanone
=
boxylic (silanoic) acids RSi( O)OH is scarce and limited to
low temperature studies under argon-matrix isolation con-
ditions.[6a–c] Use of the monoanionic b-diketiminate chelate
ligand L (L = CH[C(Me)NR]2, where R = 2,6-iPr2C6H3) led to
the first isolable donor-supported germanium analogues of
=
carboxylic acids, LGe( X)OH A (X = S, Se; Scheme 1),
which were prepared and structurally characterized by
Roesky and co-workers.[6d–e] To our knowledge, the corre-
=
sponding silicon analogues LSi( X)OH B (Scheme 1) are still
unknown. However, our research group has reported a
complete series of stable silacarboxylic silylesters
LSi(=X)OR’ (X = O, S, Se, Te; R’ = silyl),[7a–c] which are
accessible through direct chalcogenation of the corresponding
[*] Dr. Y. Xiong, Dr. S. Yao, Prof. Dr. M. Driess
Technische Universitꢀt Berlin, Institut fꢁr Chemie
Metallorganische Chemie und Anorganische Materialien
Strasse des 17. Juni 135, 10623 Berlin (Germany)
Fax: (+49)30-314-29732
E-mail: matthias.driess@tu-berlin.de
Homepage: http://www.driess.tu-berlin.de
[**] Financial support from the Deutsche Forschungsgemeinschaft and
the Fonds der Chemischen Industrie is gratefully acknowledged.
Scheme 2. Synthesis of 2 from 1 and its metalation to the silathiocar-
boxylate 3. dmap=4-dimethylaminopyridine, R=2,6-iPr2C6H3,
THF=tetrahydrofuran.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201002970.
6642
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 6642 –6645

Angewandte
Chemie
precursor. Strikingly, 1 is capable of reacting with ammonia
ꢀ
under replacement of the dmap ligand and subsequent N H
=
addition to the Si O bond to give the corresponding sila-
hemiaminal L’Si(OH)NH2 and its silacarboxylic amide tau-
tomer C LSi( O)NH2 (Scheme 1).[9c] The remarkable forma-
=
tion of C is due to the pronounced basicity of the exocyclic
methylene group of the C3N2 chelate ligand L’.
The latter results prompted us to probe whether 1 is
capable of activating H2S and H2O to form the corresponding
silacarboxylic acids B (X = S, O; Scheme 1). Herein, we
report the facile synthesis and structural characterization of
the first stable silathiocarboxylic acid–base adduct and its
metalation to form a silathiocarboxylate. In addition, we
describe for the first time the formation of a donor–acceptor
stabilized silacarboxylic acid.
Treatment of a pale yellow suspension of 1 in THF with an
excess of dry hydrogen sulfide gas at room temperature led to
a clear yellow solution, from which complex 2 could be
obtained as yellow crystals in 90% yield (Scheme 2).
The mechanism is unknown but we assume that 2 is
formed via the intermediate 2’, which results from the
Figure 1. Molecular structure of 2. Thermal ellipsoids are drawn at
50% probability level. Hydrogen atoms, except that at O1, are omitted
for clarity. The bridging hydrogen atom at O1 could be localized in the
difference Fourier maps. Selected bond lengths [ꢂ] and angles [8]: Si1–
S1 1.993(1), Si1–O1 1.620(2), Si1–N2 1.809(2), Si1–N1 1.827(2), C1–
C2 1.514(3), C2–C3 1.394(3), C3–C4 1.389(3), C4–C5 1.499(3), O1–N3
2.703; O1-Si1-N2 102.63(9), O1-Si1-N1 102.27(8), N2-Si1-N1 97.46(8),
O1-Si1-S1 120.49(7), N2-Si1-S1 114.90(7), N1-Si1-S1 115.76(7), O1-
H1-N3 172.0.
=
addition of H2S to the Si O bond in 1. Subsequent isomer-
ization of 2’ by a 1,5-proton shift from the thiol group to the
exocyclic methylene group affords 2 as the final product. The
driving force for the tautomerization process originates both
from the pronounced Brønsted acidity of the SH group and
the basicity of the methylene group of the C3N2 ligand in 2’.
bond lengths of 1.827(2) and 1.809(2) ꢁ, respectively, are
similar to the corresponding values observed in the silathio-
carboxylic silylester LSi( S)OR (1.819(3), 1.833(4) ꢁ).[7b] The
=
ꢀ
ꢀ
bond lengths of C1 C2 (1.514(3) ꢁ) and C4 C5 (1.499(3) ꢁ)
confirm the presence of two exocyclic methyl groups.
ꢀ
The 1,5-proton shift is akin to that observed for the formation
Accordingly, the short Si S bond length of 1.993(1) ꢁ is
[8]
of C[9c] (Scheme 1) and the silathioformamide, LSi( S)H.
akin to those in the silathiocarboxylic silylester LSi( S)OR
=
=
[8]
(1.980(2) ꢁ)[7b] and LSi( S)H (1.985(1) ꢁ), and indicates a
=
Compound 2 is soluble in THF and dichloromethane,
marginally soluble in toluene but insoluble in n-hexane. The
constitution of 2 could be determined by multinuclear NMR
and IR spectroscopy, and its composition was proven by
electron-impact (EI) mass spectrometry and elemental anal-
ysis. The EI mass spectrum of 2 shows a base signal for the
=
significant Si S bond character. In contrast, the
ꢀ
Si1 O1 distance of 1.620(2) ꢁ in 2 is much larger than the
=
Si O length in the starting material 1 (1.545(2) ꢁ) and thus
ꢀ
represents a Si O single bond.
Notably, 2 does not undergo a 1,3-proton shift to give the
=
=
dmap-free species at m/z 495 [{LSi( S)OH} + H]. Mean-
possible LSi( O)SH···dmap isomer. It seems unlikely that the
while, the H and 13C NMR spectra of 2 (see the Supporting
Information) for the b-diketiminate ligand L exhibit similar
isomerization is prevented by the presence of a base because
1
base-free A remains stable in solution.[6d] In contrast,
=
chemical shifts to those reported for the germanium analogue
thiocarboxylic acids RC( S)OH are in equilibrium with the
=
A. The proton of the OH group in
2
resonates at
corresponding isomers RC( O)SH in solution through rapid
d = 6.35 ppm, that is, at a significantly lower field than that
proton migration between the oxygen and the sulfur atoms.[10]
The preference of LSi( S)OH 2 over LSi( O)SH is most
[6d]
in LGe( S)OH A (d = 2.0 ppm). The 29Si NMR spectrum
=
=
=
ꢀ
of 2 reveals a resonance signal at d = ꢀ30.0 ppm akin to the
likely a result of the weaker pp pp bond and the higher bond
[7c]
=
=
=
value observed for the silathiocarboxylic silylester LSi(
polarity of the Si O compared with the Si S subunit.
Compound 2 reacts easily with organometallic bases to
form metal silathiocarboxylates. Thus, treatment of 2 with one
molar equivalent of AlMe3 in toluene at ambient temperature
S)OR, which bears the same ß-diketiminate ligand L
(d=ꢀ41 ppm).[7b]
The molecular structure of 2 has been elucidated by
single-crystal X-ray diffraction analysis. The compound
crystallized in THF as a mono-THF solvate in the triclinic
=
furnished the silathiocarboxylate LSi( S){OAlMe2(dmap)} 3
with the dmap coordinated at the aluminum atom
(Scheme 2), as confirmed by 1H NMR spectroscopy. Reaction
of 2 with MeLi or LiN(SiMe3)2 resulted only in the formation
of inseparable mixtures. Complex 3 was isolated in high yield
(91%) in the form of colorless crystals and could be fully
characterized (1H, 13C, and 29Si NMR, IR, EI-MS, and
elemental analysis). Compound 3 is soluble in benzene,
toluene, and THF, but insoluble in n-hexane. In the 1H NMR
spectrum of 3, a typical high-field singlet can be observed for
the two methyl groups of the AlMe2 moiety at d = ꢀ0.57 ppm.
The 29Si NMR spectrum shows a slightly up-field-shifted
¯
space group P1. The compound bears one dmap ligand
ꢀ
connected to the silathiocarboxylic acid through an O H···N
hydrogen bond, and the O···N distance of 2.703 ꢁ falls into
ꢀ
the normal range of O H···N hydrogen bonds (Figure 1).
Obviously, the hydrogen bridge stabilizes the system and
prevents the dimerization process, which is usually observed
for carboxylic acids and has also been observed for the “base-
free” germanium analogues A.[6d] The six-membered SiN2C3-
ring in 2 is strongly puckered and the silicon atom possesses a
ꢀ
distorted tetrahedral coordination environment. The Si N
Angew. Chem. Int. Ed. 2010, 49, 6642 –6645
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 6643

Communications
[9c]
=
resonance at d = ꢀ40.0 ppm for 3 compared to that of 2 (d =
even close to the Si O distance in 1 (1.545(2) ꢁ), and this
similarity presumably results from the strong electron dona-
tion of the dmap ligand to the aluminum center which
weakens the Al–O interaction and hence significantly induces
a stronger Si–O electrostatic interaction. A similar effect
ꢀ30.0 ppm). While the molecular-ion signal of 3 and (3-dmap)
did not appear in the EI mass spectrum, however, the
+
=
fragment ion [LSi( S)(OAlMe2)] , which indicated loss of a
methyl group, could be detected as the most abundant species
at m/z 535.
Compound 3 crystallized at 48C from THF as yellow
needles in the monoclinic space group P21/c. The aluminum
atom in 3 is tetrahedrally coordinated to the N3 atom of the
dmap ligand, the O1 atom, and the two carbon atoms of the
=
induced by dmap has also been observed for related Si O!
Al complexes, which resulted from the addition reactions of
AlMe3 to 1.[14]
Attempts to synthesize the analogous silacarboxylic acid–
base complex 4 by exposing a pale yellow solution of 1 in THF
to H2O vapor at room temperature were unsuccessful
(Scheme 3). Instead, the parent b-diketiminate ligand LH
and “free” dmap resulted, along with the quantitative
precipitation of SiO2. At low temperature (< 0 8C), no
significant reaction progress of 1 with H2O could be observed.
By increasing the acidity of the water molecule, a much faster
addition reaction occurs even at low temperature; as shown
ꢀ
methyl residues (Figure 2). The Al1 S1 distance of 3.967 ꢁ
ꢀ
by using the water–borane adduct H2O B(C6F5)3. Thus,
ꢀ
conversion of 1 with H2O B(C6F5)3 furnished the first
=
isolable LSi( O)OH complex in the form of
[LSi(OH···dmap)(=O!B(C6F5)3)] 5 in almost quantitative
yield (Scheme 3). It is very likely that formation of 5 occurs
via intermediate 5’, which immediately undergoes
a
=
1,5-proton shift to form a Lewis acid supported Si O bond.
The latter process is akin to the formation of the borane-
[7d]
=
stabilized silaformamide complex LSi( O!B(C6F5)3)H.
The composition and constitution of 5 have been unambig-
uously proven by multinuclear NMR spectroscopy (1H, 13C,
19F, and 29Si), IR spectroscopy, and elemental analysis (see the
Figure 2. Molecular structure of 3. Thermal ellipsoids are drawn at
50% probability level. Hydrogen atoms are omitted for clarity. Selected
bond lengths [ꢂ] and angles [8]: Si1–S1 1.991(1), Si1–O1 1.589(1), Si1–
N1 1.810(2), Si1–N2 1.819(2), Al1–O1 1.742(1), Al1–N3 1.961(2), Al1–
C31 1.967(2), Al1–C30 1.967(2), C5–C4 1.501(3), C4–C3 1.388(3), C3–
C2 1.3973, C2–C1 1.488(3); O1-Si1-N1 106.2(1), O1-Si1-N2 104.9(1),
N2-Si1-N1 98.0(1), O1-Si1-S1 120.9(1), N1-Si1-S1 111.6(1), N2-Si1-S1
112.7(1), O1-Al1-N3 104.2(1), O1-Al1-C31 111.2(1), N3-Al1-C31
106.9(1), O1-Al1-C30 113.4(1), N3-Al1-C30 103.8(1), C31-Al1-C30
116.2(1), Si1-O1-Al1 139.0(1).
Supporting Information). The H, 13C, and 19F NMR spectra
1
of 5 observed for the chelate ligand L and the borane subunit
=
are similar to the respective values reported for LSi( O!
B(C6F5)3)H.[7d] The 1H NMR spectrum further proves that the
ꢀ
dmap ligand remains coordinated by an O H···N hydrogen
bond and the chemical shift for the OH proton appears at d =
3.02 ppm. The 29Si NMR spectrum of 5 shows a resonance
signal at
d = ꢀ73.2 ppm, which is close to the values observed for the
respective silylesters (d = ꢀ85.1, ꢀ85.8 ppm).[7b] Compound 5
is fragile in aprotic organic solutions and decomposes slowly
clearly shows that there is no S!Al coordination. The bond
ꢀ
ꢀ
length of 1.742(1) ꢁ for Al1 O1 falls in the normal range for
to give LH, SiO2, and dmap B(C6F5)3.
[11]
ꢀ
ꢀ
an Al O single bond, while the bond lengths for Al1 C30
In summary, the formation and structural characterization
ꢀ
and Al1 C31 (1.967(2) ꢁ) are comparable to those found in
of the first silathiocarboxylic acid–base adduct 2 has been
Al2Me6 (1.956(2), 1.949(2) ꢁ)[12] and in the related
MeC(=O)(OAlMe3) anion (2.01(1) to 2.05(1) ꢁ).[11b]
ꢀ
As expected, the bond length of 1.961(2) ꢁ for Al1
ꢀ
N3 is longer than an Al N bond in aluminum amides
and thus indicative of a dative N!Al interaction.[13]
Similar to the situation in precursor 2, the tetrahedrally
coordinated silicon atom in 3 is bonded to two nitrogen
atoms from the chelate ligand, the sulfur atom, and an
ꢀ
oxygen atom. The Si N distances (1.810(2),
1.819(2) ꢁ) do not vary significantly from those in 2
=
(1.809(2), 1.827(2) ꢁ). The same is true for the Si S
bond length (1.991(1) ꢁ vs. 1.993(1) ꢁ in 2), which
indicates no electronic change in that subunit. In
ꢀ
contrast, a slightly shorter Si O distance (1.589(1) ꢁ)
can be observed in 3 compared with that in 2
ꢀ
ꢀ
(1.620(2) ꢁ). This value for the Si O bond length is Scheme 3. Reaction of 1 with H2O and H2O B(C6F5)3. R=2,6-iPr2C6H3.
6644 www.angewandte.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 6642 –6645

Angewandte
Chemie
Matsuhashi, R. Okazaki, M. Goto, J. Am. Chem. Soc. 1991, 113,
7047; c) H. Suzuki, N. Tokitoh, R. Okazaki, S. Nagase, M. Goto,
J. Am. Chem. Soc. 1998, 120, 11096.
described featuring the unique Si(=S)OH subunit. Although
ꢀ
the OH group in 2 is O H···N coordinated, the moiety can be
metalated using AlMe3 to give, under concomitant deproto-
nation and migration of the dmap ligand to the Al center, the
unprecedented silathiocarboxylate 3. While similar attempts
to form the corresponding silacarboxylic acid by addition of
[4] J. Hoeft, F. J. Lovas, E. Tiemann, Z. Tꢀrring, Z. Naturforsch.
1969, 24A, 1422.
[5] T. Iwamoto, K. Sato, S. Ishida, C. Kabuto, M. Kira, J. Am. Chem.
Soc. 2006, 128, 16914.
ꢀ
H2O to 1 failed, conversion of 1 with the H2O B(C6F5)3
[6] a) R. Whitnall, L. Andrews, J. Am. Chem. Soc. 1985, 107, 2567;
b) A. Patyk, W. Sander, J. Gauss, D. Cremer, Angew. Chem.
1989, 101, 920; Angew. Chem. Int. Ed. Engl. 1989, 28, 898; c) M.
Remko, M. Smiesko, P. T. van Duijnen, Mol. Phys. 2000, 98, 709;
d) L. W. Pineda, V. Jancik, H. W. Roesky, R. Herbst-Irmer,
Angew. Chem. 2004, 116, 5650; Angew. Chem. Int. Ed. 2004, 43,
5534; e) L. W. Pineda, V. Jancik, R. B. Oswald, H. W. Roesky,
Organometallics 2006, 25, 2384.
[7] a) S. Yao, Y. Xiong, M. Brym, M. Driess, J. Am. Chem. Soc. 2007,
129, 7628; b) S. Yao, Y. Xiong, M. Brym, M. Driess, Chem. Asian
J. 2008, 3, 113; c) J. D. Epping, S. Yao, Y. Apeloig, M. Karni, M.
Driess, J. Am. Chem. Soc. 2010, 132, 5443; d) S. Yao, M. Brym, C.
van Wꢃllen, M. Driess, Angew. Chem. 2007, 119, 4237; Angew.
Chem. Int. Ed. 2007, 46, 4159; e) M. Driess, S. Yao, M. Brym, C.
van Wꢃllen, D. Lentz, J. Am. Chem. Soc. 2006, 128, 9628.
[8] A. Meltzer, S. Inoue, C. Prꢄsang, M. Driess, J. Am. Chem. Soc.
2010, 132, 3038.
=
adduct furnished the first isolable LSi( O)OH complex 5.
The easily accessible silathiocarboxylic acid adduct 2, con-
=
=
taining the Si( S)(OH) subunit, and the Si( O)OH moiety in
adduct 5 represent versatile new functional groups in silicon
chemistry. With these compounds in hand, fundamental
aspects regarding the chemistry of silacarboxylic and thio-
silanoic acids (e.g. in metal coordination) as well as their
suitability for the synthesis of functional materials (e.g. metal-
organic-inorganic hybrid frameworks) could now be
explored. These studies are currently in progress.
Received: May 17, 2010
Published online: August 2, 2010
Keywords: carboxylate ligands · silanones · silicon ·
thiocarboxylic acids · trimethyl aluminum
[9] a) Y. Xiong, S. Yao, M. Driess, J. Am. Chem. Soc. 2009, 131,
7562; b) S. Yao, Y. Xiong, M. Driess, Chem. Eur. J. 2010, 16,
1281; c) Y. Xiong, S. Yao, R. Mꢃller, M. Kaupp, M. Driess, J. Am.
Chem. Soc. 2010, 132, 6912.
[10] a) F. Duus in Comprehensive Organic Chemistry, Vol. 3 (Eds.: D.
Barton, W. D. Ollis), Pergamon, Oxford, 1979, pp. 420 – 421; b) J.
Hine, Physical Organic Chemistry, McGraw-Hill, New York,
1962, p. 238.
[11] a) R. Mulhaupt, J. Calabrese, S. D. Ittel, Organometallics 1991,
10, 3403; b) J. L. Atwood, W. E. Hunter, K. D. Crissinger, J.
Organomet. Chem. 1977, 127, 403.
[12] J. C. Huffman, W. E. Streib, J. Chem. Soc. Chem. Commun. 1971,
911.
.
[1] a) P. Arya, J. Boyer, F. Carrꢂ, R. Corriu, G. Lanneau, J. Lapasset,
M. Perrot, C. Priou, Angew. Chem. 1989, 101, 1069; Angew.
Chem. Int. Ed. Engl. 1989, 28, 1016; b) E. Brendler, A. F. Hill, J.
Wagler, Chem. Eur. J. 2008, 14, 11300; c) A. Mitra, J. P. Wojcik,
D. Lecoanet, T. Mꢃller, R. West, Angew. Chem. 2009, 121, 4130;
Angew. Chem. Int. Ed. 2009, 48, 4069; d) C. W. So, H. W. Roesky,
R. B. Oswald, A. Pal, P. C. Jones, Dalton Trans. 2007, 5241.
[2] a) N. Tokitoh, R. Okazaki in The Chemistry of Organosilicon
Compounds, Vol. 2 (Eds.: Z. Rappoport, Y. Apeloig), Wiley,
New York, 1998, Chapter 17, pp. 1063 – 1103; b) N. Tokitoh, R.
Okazaki, Adv. Organomet. Chem. 2001, 47, 121.
[13] F. Thomas, S. Schulz, H. Mansikkamꢄki, M. Nieger, Organo-
metallics 2003, 22, 3471.
[14] Y. Xiong, S. Yao, M. Driess, Dalton Trans. 2010, DOI: 10.1039/
C0DT00148A.
[3] a) H. Suzuki, N. Tokitoh, S. Nagase, R. Okazaki, J. Am. Chem.
Soc. 1994, 116, 11578; b) N. Tokitoh, H. Suzuki, T. Matsumoto, Y.
Angew. Chem. Int. Ed. 2010, 49, 6642 –6645
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 6645