The Photochemical irradiation of R3SiH in the presence of [(γ5-C5H5)Fe(CO)2CH 3] in DMF leads to disiloxanes not disilanes

Full text of DOI:10.1002/anie.200903021

Communications
DOI: 10.1002/anie.200903021
Disiloxanes
The Photochemical Irradiation of R SiH in the Presence of
3
5
[
(h -C H )Fe(CO) CH ] in DMF Leads to Disiloxanes not Disilanes**
5
5
2
3
Hemant K. Sharma and Keith H. Pannell*
[
3]
We recently reported that the photochemical irradiation of
ꢁ21.5 ppm. Figure 1 illustrates this clean process for silane
5
tBu SnH2 in the presence of a catalytic amount of [(h -
1.
2
C H )Fe(CO) CH ] ([FpMe]) led to the efficient synthesis of
The GC–MS analysis of the product of this reaction also
confirmed the formation of the disiloxane and the absence of
5
5
2
3
the distannane HtBu SnSntBu H in hydrocarbon solutions
2
2
[
1]
[
Eq. (1)].
hn
!
2
tBu
2
SnH
2
HtBu
2
SnꢁSntBu
2
H þ H
2
ð1Þ
½
FpMeꢀ;C6 D6
It was also recently noted that a similar irradiation of the
tertiary silanes R SiH with the same catalyst [FpMe] led to
3
the high-yield formation of the corresponding disilane
R SiSiR , but only in the presence of N,N-dimethylformamide
3
3
[
2]
(
DMF) as solvent [Eq. (2)].
hn
2
R
3
SiH
!
R
3
SiꢁSiR
3
þ H
2
ð2Þ
2
9
½
FpMeꢀ;DMF
Figure 1. Si NMR spectroscopic monitoring showing the disappear-
ance of silane 1 in the presence a 5 mol% [FpMe] catalyst in DMF.
The ability of DMF to react with the hydrogen formed in
this dehydrocoupling process to produce Me N was used to
rationalize such interesting chemistry.
any disilanes. We illustrate the gas chromatograph of
PhMe SiOSiMe Ph along with that of a commercial product
3
2
2
In our hands, the irradiation of the tertiary silanes R SiH
(Gelest) in Figure 2. The equivalence is clear, with retention
3
reported in reference [2] (e.g. Ph MeSiH (1), PhMe SiH (2),
2
2
Et SiH (3)) in the presence of the catalyst [FpMe] in DMF
3
solution results in the high-yield formation of the correspond-
ing disiloxanes R Si-O-SiR and no apparent formation of the
3
3
disilane.
In our experiments, 130 mg samples of the tertiary silanes
Gelest) were irradiated in dry DMF (Aldrich) with 5 mol%
(
of the catalyst [FpMe] in sealed NMR tubes at a distance of
cm from a 400 W low-pressure mercury lamp. The reactions
3
2
9
were monitored by Si NMR spectroscopy. The reactions
with 1, 2, and 3 behaved similarly, with the disappearance of
the resonance of the starting silane (d = ꢁ17.5 ppm (1) in
4
8 h; d = ꢁ17.2 ppm (2) in 1.5 h; d = 0.5 ppm (3) in 10 h)
along with the appearance of a single resonance associated
with the formation of the corresponding disiloxane
R SiOSiR (d = ꢁ9.1, 0.6, and 8.4 ppm for the reactions of 1,
3
3
Figure 2. GC retention time data for PhMe SiOSiMe Ph, synthesized
2
2
2, and 3, respectively). No resonances associated with the
herein (top) and an authentic sample (Gelest; bottom).
corresponding disilanes were observed, for example
[
3]
(
Ph MeSi)2 at d = ꢁ22.6 ppm
or (PhMe Si)2 at d =
2
2
times of 10.14 and 10.16 min, respectively. The small amount
of material at a retention time of 11.28 min is the trisiloxane
PhMe SiOSiMe OSiMe Ph, which curiously is present both in
[*] Dr. H. K. Sharma, Prof. K. H. Pannell
2
2
2
Department of Chemistry, University of Texas at El Paso
El Paso, TX 79968-0513 (USA)
E-mail: kpannell@utep.edu
the commercial material and the material synthesized in our
experiments.
The mass spectral portion of the GC–MS analysis of the
three siloxane materials formed in our studies are illustrated
in Figure 3. The inverted mass spectrum on the bottom of
each part of Figure 3 is that derived from the standard library
of materials and demonstrates absolute equivalence of the
[
**] We thank Dr. Shizue Mito for useful discussions and Dr. Ernesto
Nakayasu and the U. T. El Paso Border Biomedical Research Center
facility for access to, and training on, the GC–MS instrumentation.
We also thank the Welch Foundation, Houston, TX for support of
this research.
7052
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 7052 –7054

Angewandte
Chemie
exposed the appropriate disilane to the same photochemical
conditions, in the same solvent, either dry or used “as
received”, and detected no disiloxane production [Eq. (3)].
hn
PhMe SiꢁSiMe Ph
!
no reaction
ð3Þ
2
2
½
FpMeꢀ;DMF
This result rules out initial formation of the disilanes and
subsequent oxidation. At present we are unable to identify
the specific features of our experimental setup that could
result in the production of the siloxanes as opposed to
disilanes.
The ability of DMF to remove the hydrogen produced to
form Me N was suggested as a crucial feature for the solvent-
3
specific process; however, this provokes the question as to the
fate of the oxygen atom. In 1985 Voronkov and co-workers
reported that treatment of various silanes R R’SiH (R R’ =
2
2
Cl Me, Cl Et, Et Me, and Et ) with DMF in the presence of
2
2
2
3
metal species (NO) PtCl or [Me NH ][Rh(CO) Cl ] led to
2
6
2
2
2
2
[
7]
the formation of the corresponding disiloxanes. In that
study, and from the product yield/time relationship of the
recent “disilane” synthesis, it was concluded that electron-
withdrawing groups slowed down the process. We can confirm
the same relationship using 1, 2, and Ph SiH. Indeed, in our
3
hands the photolysis of Ph SiH with [FpMe] failed to yield a
3
significant amount of Ph SiOSiPh , contrary to the reportedly
3
3
[2]
slow reaction but high recovered yield of Ph SiSiPh .
3
3
Figure 3. Mass spectra of a) Ph MeSiOSiMePh , b) PhMe SiOSiMe Ph,
2
2
2
2
As to a mechanism for this interesting new chemistry,
there are several possibilities. One certainly involves the
formation of a bis(silyl) iron complex originally proposed for
and c) Et SiOSiEt formed in the reactions presented (top) and the
3
3
literature spectra (bottom).
[2]
the formation of the disilanes. However, the fact that
disiloxane formation was noted by Voronkov using metal
complexes of Pt and Rh, for which bis(silyl) complexes are
less obviously accessible leads us to suggest the mechanism
outlined in Scheme 1.
[
4]
product to the reported standards. No remaining R SiH, nor
3
any disilane R SiSiR , was detected. In the report on the
3
3
[
2]
formation of disilanes, neither structural nor spectroscopic
evidence of the disilanes was reported, only GC analyses. In
our hands, the use of GC to distinguish the disilane from the
disiloxane product resulted in retention times of 10.14 min for
PhMe SiOSiMe Ph and 10.35 min for PhMe SiSiMe Ph
The key feature of this mechanism involves formation of
the DMF metal complex and a subsequent hydrosilylation–
reductive elimination process. There is significant precedent
for the photochemical formation of DMF metal carbonyl
[
8]
2
2
2
2
complexes, in which the electron-deficient ketone carbon
[
9]
(
Gelest) under our instrumental conditions. We note that in
atom is activated towards hydrosilylation chemistry.
the supporting information associated with reference [2], the
reported elemental analyses correspond more closely to
disiloxane products than disilanes. Furthermore, the chemical
As noted in Scheme 1, it is possible that the hydrosilylated
product R SiOCH NMe could be eliminated and itself react
3
2
2
with R SiH to form the disiloxane and Me N. This suggestion
3
3
2
9
shift in the Si NMR spectrum presented for poly[(tetrame-
comes from a related precedent from Mironov and co-
workers involving disiloxane elimination reactions of this
5
5
thyldisilanylene)ferrocenylene]
[{(h -C H )Fe(h -C H -
5 4 5 4
SiMe SiMe )} ] was d = 0.78 ppm, close to that of the model
product with chlorosilanes and related species [Eq. (4), X =
2
2
n
5
5
[10]
disiloxane complex [{(h -C H )Fe(h -C H )} SiMe OSiMe ],
Cl, NR etc.].
5
5
5
4
2
2
2
2
[
5]
which we reported at d = 0.48 ppm. In contrast, we also
reported a model disilane complex of this type, that is, [{(h -
C H )Fe(h -C H )} SiMe SiMe ], and noted that its Si NMR
5
Me SiOCH NMe þ Me SiX ! Me SiOSiMe þ XCH NMe
ð4Þ
5
29
3
2
2
3
3
3
2
2
5
5
5
4
2
2
2
[
6]
spectrum contained a single resonance at d = ꢁ22.4 ppm.
Thus we have clearly demonstrated that only disiloxane
formation takes place under the photochemical conditions
We are continuing a detailed mechanistic study of this
system, including studies of R SiOCH NMe compounds.
3
2
2
[2]
reported above.
Regardless of the exact process, this catalytic process may be
5
We performed a series of experiments to determine
whether any variation of the reaction conditions could
account for our observations. For example, reactions with
either dry or wet DMF yielded the same results. We also
very general, as exchanging [FpMe] for [(h -
C H )Mo(CO) CH ] and other related complexes accom-
5
5
3
3
[11]
plishes the same transformation of silane to disiloxane.
A
recent discussion on the role of DMF in activating the
Angew. Chem. Int. Ed. 2009, 48, 7052 –7054
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
7053

Communications
Scheme 1. Proposed catalytic mechanism for transformation of R SiH to R SiOSiR .
3
3
3
2
9
observed at d = ꢁ21.8 ppm and in C D at d = ꢁ21.48 ppm. Si
dehydrocoupling of silanes with hydroxylic reagents illus-
6
6
[
12]
chemical shift of the disilane Ph
2
MeSiꢁSiMePh
2
synthesized
trates the complexity of the DMF–silane interactions.
from equimolar amounts of Ph MeSiLi and Ph MeSiCl in THF is
2
2
observed at d = ꢁ22.56 ppm in C D .
6
6
Received: June 4, 2009
Published online: August 22, 2009
[
[
4] GC–MS analysis was performed using a trace GC connected to a
Polaris Q mass spectrometer (Thermo Fisher Scientific). The
library search algorithm used was the NIST MS Search 2.0,
which comes in Xcalibur 1.4 SR1 (Thermo Fisher Scientific)
package.
Keywords: homogeneous catalysis · iron · oxygenation · silanes
.
5] K. H. Pannell, H. K. Sharma, Organometallics 1997, 16, 3077.
[
1] a) K. H. Pannell, H. K. Sharma, R. Arias, Abstracts of Papers,
35th ACS National Meeting, New Orleans, LA, United States,
[6] K. H. Pannell, H. Sharma, Organometallics 1991, 10, 954.
[7] L. I. Kopylova, N. D. Ivanova, M. G. Voronkov, Zhur. Obshch.
Khim. 1985, 55, 1649.
2
April 6 – 10, 2008, INOR-015; b) H. K. Sharma, R. Arias-Ugarte,
A. J. Metta-Magana, K. H. Pannell, Angew. Chem. 2009, 121,
[8] I. W. Stolz, G. R. Dobson, R. K. Sheline, Inorg. Chem. 1963, 2,
323.
[9] a) S. Diez-Gonzꢀlez, S. P. Nolan, Organic Preparations and
Procedures International 2007, 39, 523; b) R. H. Morris, Chem.
Soc. Rev. 2009, 38, 2282.
6
427; Angew. Chem. Int. Ed. 2009, 48, 6309.
2] a) M. Itazaki, K. Ueda, H. Nakazawa, Angew. Chem. 2009, 121,
363; Angew. Chem. Int. Ed. 2009, 48, 3313; b) “Disilane bond-
[
3
containing polymers and preparation thereof”: H. Nakazawa, M.
Itazaki, Jpn. Kokai Tokkyo Koho, 2009; c) “Preparation of SiꢁSi
[10] V. P. Kozyukov, V. P. Kozyukov, V. F. Mironov, Zh. Obshch.
Khim. 1983, 53, 119.
bond-bearing compounds”: H. Nakazawa, M. Itazaki, U.S. Pat.
Appl. Publ., 2009.
[11] R. Arias, unpublished results.
[12] J. J. Chru s´ ciel, Can. J. Chem. 2005, 83, 508.
2
9
[
3] Si chemical shift in the NMR spectra of a commercial sample
Gelest) of PhMe SiꢁSiMe Ph in DMF (without lock) is
(
2
2
7054
www.angewandte.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 7052 –7054