Cycloaddition reactions of aldehydes to tetramesityldisilene and tetramesitylgermasilene: Evidence for a biradical intermediate

Article abstract of DOI:10.1021/ja9812002

The mechanism of the addition of aldehydes to disilenes find germasilenes was investigated by examining the reaction between trans-(2- phenylcyclopropyl)carboxaldehyde and tetramesityldisilene or tetramesitylgermasilene. [Dimesityl(2-cis-4-trans-1-oxa-5-

Full text of DOI:10.1021/ja9812002

J. Am. Chem. Soc. 1998, 120, 11049-11053
11049
Cycloaddition Reactions of Aldehydes to Tetramesityldisilene and
Tetramesitylgermasilene: Evidence for a Biradical Intermediate
Craig E. Dixon,† Donald W. Hughes,‡ and Kim M. Baines*,†
Contribution from the Department of Chemistry, UniVersity of Western Ontario, London, Ontario, Canada
N6A 5B7, and Department of Chemistry, McMaster UniVersity, Hamilton, Ontario, Canada L8S 4M1
ReceiVed April 9, 1998
Abstract: The mechanism of the addition of aldehydes to disilenes and germasilenes was investigated by
examining the reaction between trans-(2-phenylcyclopropyl)carboxaldehyde and tetramesityldisilene or
tetramesitylgermasilene. [Dimesityl(2-cis-4-trans-1-oxa-5-phenylpentadienyl)silyl]dimesitylsilane (1) and
2,2,3,3-tetramesityl-4-phenyloxa-2,3-disilacyclohept-6-ene (2) were formed in the reaction between the aldehyde
and the disilene, and the analogous [dimesityl(2-cis-4-trans-1-oxa-5-phenylpentadienyl)silyl]dimesitylgermane
(4) and 2,2,3,3-tetramesityl-4-phenyloxa-2,3-silagermacyclohept-6-ene (5) were formed in reaction with the
germasilene. It is proposed that compounds 1, 2, 4, and 5 are the result of either disproportionation or ring
closure of an intermediate biradical derived from initial bond formation between the oxygen atom of the aldehyde
and the doubly bonded silicon atom of the dimetallene.
Introduction
ring is often inferred from the isolation of products derived from
the opening of the cyclopropyl ring. We realized that the nature
of the reactive intermediate involved (if any) in the addition of
carbonyl compounds to dimetallenes could be probed by
examination of the products from the reaction between trans-
(2-phenylcyclopropyl)carboxaldehyde (a variation on the probes
developed by Newcomb) and a dimetallene, specifically, tet-
ramesityldisilene or -germasilene. The results of our investiga-
tions are presented here.
Since their discovery over 15 years ago, much has been, and
continues to be, learned about the reactivity of group 14
dimetallenes.1 However, despite the obvious synthetic potential
of these compounds, relatively few mechanistic investigations
of the reactions of dimetallenes have been carried out. There
have been a few reports concerning the mechanism of the
addition of alcohols,2-6 and alkynes2,7 to disilenes, and recently,
we reported our preliminary results on the mechanism of alkene
additions to tetramesitylgermasilene and tetramesityldisilene.8
In contrast, the mechanism for the addition of carbonyl
compounds such as aldehydes and ketones to dimetallenes has
not been examined, despite being one of the most extensively
investigated synthetic reactions of dimetallenes.9
Results
When trans-(2-phenylcyclopropyl)carboxaldehyde was al-
lowed to react with tetramesityldisilene at -60 °C in hexanes,
two adducts, 1 and 2, were produced in approximately a 1:1
There are numerous mechanistic probes available for explor-
ing the possibility of the intermediacy of radicals in chemical
reactions.10 One class of probes, developed by Newcomb and
co-workers, takes advantage of the extremely fast ring opening
reaction of phenyl-substituted cyclopropylcarbinyl radicals.11
The intermediacy of a radical center adjacent to the cyclopropyl
* To whom correspondence should be addressed. E-mail: kbaines2@
julian.uwo.ca.
† University of Western Ontario.
mixture as determined by 1H NMR spectroscopy. The 1H NMR
‡ McMaster University.
(1) Reviews: (a) Okazaki, R.; West, R. AdV. Organomet. Chem. 1996,
39, 231-273. (b) Raabe, G.; Michl, J. In The Chemistry of Organic Silicon
Compounds; Patai, S., Rappoport, Z., Eds.; Wiley; New York, 1989; pp
1015-1142. (c) West, R. Angew. Chem., Int. Ed. Engl. 1987, 26, 1201-
1211. (d) Baines, K. M.; Stibbs, W. G. AdV. Organomet. Chem. 1996, 39,
275-324. (e) Escudie´, J.; Couret, C.; Ranaivonjatovo, H.; Satge´, J. Coord.
Chem. ReV. 1994, 130, 427-480.
spectrum of 1 is completely consistent with the assigned
(9) (a) Boudjouk, P.; Han, B.-H.; Anderson, K. R. J. Am. Chem. Soc.
1982, 104, 4992-4993. (b) Weidenbruch, M.; Scha¨fer, A.; Thom, K. L. Z.
Naturforsch. B. 1983, 38, 1695-1696. (c) Fink, M. J.; DeYoung, D. J.;
West, R.; Michl, J. J. Am. Chem. Soc. 1983, 105, 1070-1071. (d) Scha¨fer,
A.; Weidenbruch, M.; Pohl, S. J. Organomet. Chem. 1985, 282, 305-313.
(e) Batcheller, S. A.; Masamune, S. Tetrahedron Lett. 1988, 29, 3383-
3384. (f) Ando, W.; Tsumuraya, T. J. Chem. Soc., Chem. Commun. 1989,
770-772. (g) Fanta, A. D.; DeYoung, D. J.; Belzner, J.; West, R.
Organometallics 1991, 10, 3466-3470. (h) Fanta, A. D.; Belzner, J.; Powell,
D. R.; West, R. Organometallics 1993, 12, 2177-2181.
(2) DeYoung, D. J.; Fink, M. J.; West, R.; Michl, J. Main Group Met.
Chem. 1987, 10, 19-43.
(3) Sekiguchi, A.; Maruki, I.; Sakurai, H. J. Am. Chem. Soc. 1993, 115,
11460-11466.
(4) Budaraju, J.; Powell, D. R.; West, R. Main Group Met. Chem. 1996,
19, 531-537.
(10) Newcomb, M. Tetrahedron 1993, 49, 1151-1176.
(5) Apeloig, Y.; Nakash, M. J. Am. Chem. Soc. 1996, 118, 9798-9799.
(6) Apeloig, Y.; Nakash, M. Organometallics 1998, 17, 1260-1265.
(7) Nakadaira, Y.; Sato, R.; Sakurai, H. Chem. Lett. 1985, 643-646.
(8) Dixon, C. E.; Baines, K. M. Phosphorus, Sulfur Silicon Relat. Elem.
1997, 124/125, 123-132.
(11) (a) Newcomb, M.; Manek, M. B. J. Am. Chem. Soc. 1990, 112,
9662-9663. (b) Newcomb, M.; Johnson, C. C.; Manek, M. B.; Varick, T.
R. J. Am. Chem. Soc. 1992, 114, 10915-10921. (c) Martin-Esker, A. A.;
Johnson, C. C.; Horner, J. H.; Newcomb, M. J. Am. Chem. Soc. 1994, 116,
9174-9181.
10.1021/ja9812002 CCC: $15.00 © 1998 American Chemical Society
Published on Web 10/20/1998

11050 J. Am. Chem. Soc., Vol. 120, No. 43, 1998
Dixon et al.
Scheme 1
hexamesitylsiladigermirane at -70 °C in the presence of Et3-
SiH, yielded three products, (Et3Si)GeMes2H (3),12 4, and 5, in
1
structure. Perhaps the most striking feature in the H NMR
spectrum is the presence of resonances entirely consistent with
the oxydienic moiety (see the Experimental Section). The
chemical shifts of these signals, along with their respective
coupling constants, suggest the cis-trans orientation for the
diene substructure in compound 1. The 13C, 1H COSY, 1H/29-
a ratio of 2:1:1, respectively, as determined by 1H NMR
spectroscopy. The features of the 1H NMR spectrum of
compound 4 are completely compatible with the assigned
structure. The 1H COSY NMR spectrum confirmed the
expected couplings for the diene moiety. The relative chemical
shift of the signals, along with their respective coupling
constants, suggest the cis-trans orientation for the diene
substructure in compound 4. The 13C and 1H/29Si HMBC NMR
spectra all support the assigned structure for compound 4. For
example, the 1H/29Si HMBC spectrum shows a two-bond
correlation between the signal at 3.75 ppm in the 29Si dimension
and the signal assigned as the Ge-H resonance at 5.65 ppm in
1
Si HSQC, and H/29Si HMBC NMR spectra all support the
1
assigned structure for compound 1. For example, the H/29Si
1
HMBC and H/29Si HSQC NMR spectra indicated two-bond
(2JSi,H ) 12.1 Hz) and one-bond (1JSi,H ) 180.9 Hz) Si-H
couplings, respectively, which is in agreement with the structure
assigned.
Compound 2 was also identified by NMR spectroscopy.
Many of the signals attributed to the mesityl substituents and
the hydrogen atoms of the oxadisilacycloheptene ring were
broadened in the NMR spectra of this compound, and thus, some
spectra were recorded at higher temperatures to sharpen the
signals. Central to assigning the structure for compound 2 was
confirming the hydrogen and carbon atom coupling interactions
that would be expected for the oxadisilacycloheptene ring. The
1H/13C HSQC NMR spectrum indicated that the 6.33 and 4.76
ppm signals assigned to the enolic hydrogen atoms in the1H
dimension correlated with signals found at 140.6 and 111.8 ppm
in the 13C dimension, respectively. These 13C NMR resonances
are in the expected chemical shift range for oxygen-substituted
vinylic carbon atoms. The COSY NMR spectrum showed the
expected coupling interactions for these hydrogen atoms and
demonstrated that the signal at 4.76 ppm (dCHsCHH) was
coupled to two broad signals at 3.48 ppm (dCHsCHH) and
2.44 ppm (dCHsCHH, overlapped by Mes CH3 signals). The
1H/13C HSQC NMR spectrum indicated a correlation between
1
the H dimension. Compound 5 was also identified by NMR
spectroscopy. The spectral features of this compound were very
similar to those previously discussed for compound 2. The 1H/
29Si HMBC spectrum shows a correlation between the signal
at 2.43 ppm in the 29Si dimension and the signal assigned to
the OCHdCH resonance (6.35 ppm) in the 1H dimension,
supporting the assigned regiochemistry of 5.
Discussion
The structures of compounds 1, 2, 4, and 5 indicate that the
cyclopropyl ring found in the aldehyde has opened during the
course of the reaction. We believe the ring opening is the result
of the rearrangement of an intermediate biradical (Scheme 1).
Initial addition of the carbonyl double bond to the Si-M double
bond of germasilene or the disilene would lead to biradical 6.
We expect the cyclopropylcarbinyl radical moiety in 6 to
undergo rapid rearrangement to the secondary benzyl/homoallyl
radical, 7. Following rearrangement, 7 can disproportionate to
give 1 or 4 or can undergo ring closure to yield compound 2 or
5.
Biradical 6 contains an oxy-substituted cyclopropylcarbinyl
radical. Although the rate constant for the ring opening of oxy-
substituted cyclopropylcarbinyl radicals is unknown, Newcomb
and co-workers have shown that the cyclization rate constant
of the 1-methoxy-6,6-diphenyl-5-hexenyl radical is nearly equal
to the related radical lacking the methoxy group.13 Thus, it is
not unreasonable to assume that the oxy-substituted cyclopro-
pylcarbinyl radical will open at a rate similar to that of the
unsubstituted analogue. The following two examples give
precedents for the rapid ring opening of an oxycyclopropyl-
carbinyl radical. Thermolysis of trans-(2-phenylcyclopropyl)-
carboxaldehyde and tributyltin hydride in the presence of AIBN
gave a 2.5:1 mixture of 4-phenylbutanal and (2-phenylcyclo-
propyl)carbinol (Scheme 2).14 The tributyltin radical is believed
to add to the carbonyl oxygen of the aldehyde to give the
1
the 3.48 and 2.44 ppm signals in the H dimension with a 13C
signal found at 33.0 ppm. Taken together, these coupling
interactions support the assignment of the 3.48 and 2.44 ppm
signals to the methylene moiety for compound 2. Finally, the
COSY NMR spectrum indicated a correlation between the
methylene hydrogen atoms and a signal at 4.02 ppm (-CHPh).
The 4.02 ppm signal was also found to correlate with a
resonance at 43.5 ppm in the 13C dimension of the 1H/13C HSQC
NMR spectrum. The NMR spectral data reported here indicate
the appropriate coupling interactions for the hydrogen and
carbon atoms of the proposed seven-membered ring structure
1
of compound 2. The H/29Si HMBC NMR spectrum at 67 °C
indicated a correlation between the signal at -0.5 ppm in the
1
29Si dimension and the H signal at 6.33 ppm (OsCHdCH),
and thus, this signal is assigned to the silicon of the silyl enol
ether moiety. The chemical shift of this signal agrees well with
that observed in the analogous compound derived from the
germasilene (2.4 ppm; see below) and the silyl enol ether 1
(1.2 ppm). The signal at -8.5 ppm (shifted from -11.9 ppm
at 30 °C) correlated to the methine hydrogen atom at 4.03 ppm,
and thus was assigned to the silicon atom attached to the carbon
atom bearing the phenyl substituent.
(12) Baines, K. M.; Cooke, J. A.; Vittal, J. J. J. Chem. Soc., Chem.
Commun. 1992, 1484-1485.
(13) Johnson, C. C.; Horner, J. H.; Tronche, C.; Newcomb, M. J. Am.
Chem. Soc. 1995, 117, 1684-1687.
Addition of trans-(2-phenylcyclopropyl)carboxaldehyde to a
solution of tetramesitylgermasilene, produced by photolysis of
(14) Zelechonok, Y.; Silverman, R. B. J. Org. Chem. 1992, 57, 5785-
5787.

Cycloaddition Reactions of Aldehydes
J. Am. Chem. Soc., Vol. 120, No. 43, 1998 11051
Scheme 2
Scheme 3
cyclopropylmethyloxonium 12 to the secondary benzyl car-
bocation 13 is required (Scheme 3). However, a comparison
of the relative stabilizing abilities of the groups substituted on
the carbocationic center (oxy > phenyl ≈ cyclopropyl) suggests
that the cyclopropylmethyloxonium intermediate would be the
lower energy species relative to the secondary benzyl carboca-
tion. Rearrangement to a higher energy species seems unrea-
sonable. The notion that the three-membered ring in a
cyclopropylmethyloxonium ion does not readily open under
ionic conditions is also supported by reports that the carbonyl
group of trans-(2-phenylcyclopropyl)carboxaldehyde can be
successfully deprotected under acidic conditions without any
evidence for opening of the cyclopropyl ring.18,19 Despite the
many precedents in support of a biradical (rather than zwitte-
rionic) intermediate, we continue to design and execute experi-
ments which will allow us to distinguish between these two
possibilities.
oxycyclopropylcarbinyl radical which either abstracts a hydrogen
atom from the tin hydride or undergoes ring opening before
hydrogen atom abstraction. The observed products are obtained
after solvolysis (ethanol) of the stannyl ethers.
Furthermore, Castellino and Bruice reported that the dicy-
clopropyl ketone 8 gave, among other products, compound 9
Summary
In summary, the addition of trans-(2-phenylcyclopropyl)-
carboxaldehyde to tetramesitylgermasilene or-disilene yielded
products which are consistent with the formation of cyclopro-
pylcarbinyl radical intermediates. The formation of these
products suggests that the addition of aldehydes to tetramesi-
tylgermasilene and -disilene proceeds via a biradical intermedi-
ate. The results detailed here are interesting in light of a recent
report by Bradaric and Leigh on the mechanism of the ene
addition of acetone to several 1,1-diphenylsilene derivatives.20
The results of their study are consistent with a mechanism
involving initial, reversible formation of a zwitterionic silene-
ketone complex which collapses to product by rate-controlling
proton transfer. Furthermore, it is suggested that the formal [2
+ 2] addition of nonenolizable carbonyl compounds to reactive
silenes might also proceed through the intermediacy of silene-
carbonyl complexes.20,21 Our results, however, indicate that it
is unlikely there is an initial dimetallene-carbonyl zwitterionic
complex formed in the addition of aldehydes to dimetallenes,
but rather, the cycloaddition most likely proceeds through a
stepwise, biradical pathway without the involvement of the
oxygen lone pair. Once again, it appears that the doubly bonded
silicon atom, at least in disilenes, is not as Lewis acidic22 as in
silenes23 and silanimines.24
when irradiated in the presence of Bu3SnH, followed by addition
of HBr to the reaction mixture.15 Compound 9 is the result of
the ring opening rearrangement of the phenyl-substituted
cyclopropyl moieties. The rearrangement involves the genera-
tion of an oxycyclopropylcarbinyl radical, 10, which rearranges
to the secondary benzyl/homoallyl radical, 11. The rate of this
rearrangement was estimated to be at least 2 × 1010 s-1. Given
these precedents, as well as the investigations by Newcomb on
the rapid ring opening reactions of phenyl-substituted cyclo-
propylcarbinyl radicals,11 it is reasonable to propose a similar
pathway for the addition of trans-(2-phenylcyclopropyl)carbox-
aldehyde to tetramesitylgermasilene and -disilene.
Many rearrangements that may occur from a biradical
intermediate may also arise from a zwitterionic intermediate,
so despite the precedents that favor a radical mechanism, the
ionic pathway was also considered.16 For an ionic mechanism,
the first intermediate formed would most reasonably be 12.17
To arrive at compound 1, 2, 4, or 5, rearrangement from the
(17) Nucleophilic addition of the dimetallene to the aldehyde to give a
zwitterion with a positive charge at Si (or Ge) and an oxy anion would
most probably rapidly close to the dimetallaoxetane and not lead to
cyclopropyl ring opening.
(18) (a) Mori, A.; Arai, I.; Yamamoto, H. Tetrahedron 1986, 42, 6447-
6458. (b) Barrett, A. G. M.; Doubleday: W. W.; Tustin, G. J.; White, A.
J. P.; Williams, D. J. J. Chem. Soc., Chem. Commun. 1994, 1783-1784.
(19) Abdallah, H.; Gre´e, R.; Carrie´, R. Bull. Soc. Chim. Fr. 1985, 794-
802.
(20) Bradaric, C. J.; Leigh, W. J. Organometallics 1998, 17, 645-651.
(21) (a) Wiberg, N.; Link, M. Chem. Ber. 1995, 128, 1231-1240. (b)
Wiberg, N.; Hwang-Park, H.-S.; Lerner, H.-W.; Dick, S. Chem. Ber. 1996,
129, 471-478.
(15) Castellino, A. J.; Bruice, T. C. J. Am. Chem. Soc. 1988, 110, 7512-
7519.
(16) On the basis of the oxidation potential of Mes2SidSiMes2 (Ep,ox
+0.38 V; Shepherd, B. D.; West, R. Chem. Lett. 1988, 183-186) and the
reduction potential of trans-(2-phenylcyclopropyl)carboxaldehyde (Ep,red
-2.27 V at 200 mV/s), SET between these two species is highly
endothermic.
)
)

11052 J. Am. Chem. Soc., Vol. 120, No. 43, 1998
Dixon et al.
[Dimesityl(2-cis-4-trans-1-oxa-5-phenylpentadienyl)silyl]dimesi-
tylsilane (1): IR (thin film, cm-1) 3020 (m), 2958 (m), 2926 (s), 2859
(m), 2134 (Si-H, m), 1641 (m), 1607 (s), 1497 (m), 1453 (s), 1409
(m), 1279 (m), 1238 (s), 1151 (s), 1063 (s); 1H NMR (ppm) 7.64 (dd,
Experimental Section
All experiments were carried out in flame-dried glassware under an
atmosphere of argon. Toluene, benzene, THF, and Et2O were freshly
distilled from sodium benzophenone ketyl prior to use, whereas pentane,
hexanes, and Et3SiH were distilled from LiAlH4 prior to use. trans-
(2-Phenylcyclopropyl)carboxylic acid was obtained from the Aldrich
Chemical Co. and used without additional purification. Chromatog-
raphy was carried out on silica gel plates using a Chromatotron
(Harrison Research) or on conventional silica gel preparative plates.
Photolyses were carried out using a Rayonet photochemical reactor.
Low-temperature photolyses were carried out by cooling the sample
using an Endocal model ULT-70 low-temperature external bath
circulator to force cold (-60 or -70 °C) methanol through a vacuum
jacketed quartz (254 nm) or Pyrex (350 nm) immersion well. NMR
spectra were recorded on a Varian Gemini 200, a Varian XL or Gemini
300, or a Bruker Avance DRX-500 using benzene-d6 as a solvent. The
3
3
1 H, OsCHdCH-CHdCHsPh, J ) 15.7 Hz, J ) 11.2 Hz), 7.48
(d, 2 H, Ph, 3J ) 7.4 Hz), 7.26 (t, 2 H, Ph, 3J ) 7.7 Hz), 7.07 (t, 1 H,
Ph, 3J ) 7.4 Hz), 6.75 (s, 4 H, Mes CH), 6.66 (s, 4 H, Mes CH), 6.39
(d, 1 H, OsCHdCHsCHdCHsPh, 3J ) 16.0 Hz), 6.23 (d, 1 H, Os
CHdCHsCHdCHsPh, 3J ) 5.7 Hz), 5.66 (s, 1 H, SisH), 5.38 (dd,
3
3
1 H, OsCHdCHsCHdCHsPh, J ) 5.8 Hz, J ) 10.9 Hz), 2.33
(br s, 24 H, Mes o-CH3), 2.14 (s, 6 H, Mes p-CH3), 2.07 (s, 6 H, Mes
p-CH3); 13C NMR (ppm) 145.51, 144.96 (Mes C and Ph C), 141.62
(OsCHdCHsCHdCHsPh), 139.75, 138.99, 131.52, 130.48 (Mes
C and Ph C), 129.81, 129.11 (Mes CH), 128.0 (OsCHdCHsCHd
CHsPh), 128.92, 127.02, 126.47 (Ph CH), 124.20 (OsCHdCHsCHd
CHsPh), 110.82 (OsCHdCHsCHdCHsPh), 24.46 (br s), 23.94 (br
1
s), 21.11, 20.99 (Mes CH3); 29Si NMR (ppm) -56.5 (Si-H, JSi-H
)
1
standards were as follows: residual C6D5H 7.15 ppm for H spectra;
180.9 Hz, JSi-H ) 12.1 Hz), 1.2 (Si-O); MS (m/z) 678 (M+, 0.3),
559 (M+ - Mes, 0.5), 533 (Mes2Si - SiHMes2, 1.4), 430 (Mes2SiO -
SiHMes, 29), 411 (Mes2SiOC4H4Ph, 100), 291 (MesSiOC4H3Ph, 19);
high-resolution MS calcd for C46H54Si2O 678.3713, found 678.3686.
2,2,3,3-Tetramesityl-4-phenyloxa-2,3-disilacyclohept-6-ene (2): IR
(thin film, cm-1) 3024 (m), 2927 (s), 2860 (m), 1750 (w), 1656 (m),
1611 (s), 1559 (w), 1454 (s), 1413 (m), 1387 (w), 1297 (m), 1136 (m),
1036 (m), 856 (s); 1H NMR (ppm) 6.89 (s, 5 H, Ph CH), 6.85 (br s, 2
H, Mes CH), 6.70 (s, 2 H, Mes CH), 6.59 (br s, 2 H, Mes CH), 6.39
2
C6D6 central transition for13C NMR spectra; and Me4Si as an external
standard, 0 ppm for 29Si. IR spectra were recorded (cm-1) as thin films
on a Perkin-Elmer System 2000 FT IR spectrometer. A Finnegan MAT
model 8230 instrument, with an ionizing voltage of 70 eV, was used
to obtain electron impact mass spectra (reported in mass-to-charge units,
m/z, with ion identity and peak intensities relative to the base peak in
parentheses).
Cyclic voltammetry was performed using a PAR 263A potentiostat
interfaced to a personal computer using PAR 270 electrochemistry
software. The electrochemical cell was maintained at 25 °C and
contained 0.1 mol/L tetraethylammonium perchlorate in N,N-dimeth-
ylformamide purged by argon. Cell iR compensation was adjusted to
at least 95% of the oscillation value. At the beginning of the
experiment, the 3 mm glassy carbon working electrode was freshly
polished with 1 µm diamond paste and ultrasonically cleaned in ethanol
for 15 min. The counter electrode was a platinum plate and the
reference a silver wire immersed in a glass tube containing the solvent
and 0.1 mol/L electrolyte with a fine sintered bottom. Ferrocene was
used as an internal redox reference, and the reported potential was
calibrated against the saturated calomel electrode (SCE).
3
(br s, 2 H, Mes CH), 6.33 (d, J ) 6.2 Hz, 1 H, OsCHdCH), 4.76
(ddd, 3J ) 6.2 Hz, 3J ) 8.1 Hz, 3J ) 6.2 Hz, 1 H, OsCHdCH), 4.02
(br m, 1 H, CHPh), 3.48 (br m, 1 H, dCsCHH), 2.90 (br s, 6 H, Mes
CH3), 2.60 (br s, 6 H, Mes CH3), ∼2.4 (vbr s, approximately 4 H total,
dCsCHH and Mes CH3), 2.30 (br s, 3 H, Mes CH3), 2.11 (s), 2.085
(s), 2.079 (s), 2.05 (s, 12 H total, Mes CH3), 1.76 (vbr s, 3 H, Mes
1
CH3), 1.59 (br s, 3H, Mes CH3); H NMR (ppm, 67.5 °C) 6.83-6.93
(m, Ph CH), 6.70 (s, Mes CH), 6.59 (br s, Mes CH), 6.32 (d, 3J ) 6.3
3
3
3
Hz, OsCH)CH), 4.74 (ddd, J ) 6.3 Hz, J ) 8.1 Hz, J ) 6.3 Hz
OsCHdCH), 4.03 (dd, 3J ) 7.3 Hz, 3J ) 2.6 Hz, CHPh), 3.36 (ddd,
3J ) 7.3 Hz, J ) 8.1 Hz, J ) -14.1 Hz, dCsCHH), 2.48 (ddd, J
3
2
3
3
2
) 2.6 Hz, J ) 6.3 Hz, J ) -14.1 Hz, dCsCHH), 2.56 (s), 2.10-
Dimesitylbis(trimethylsilyl)silane25 and hexamesitylsiladigermirane26
were prepared following the reported procedures. trans-(2-Phenylcy-
clopropyl)carboxylic acid was reduced with lithium aluminum hydride
in diethyl ether to give the corresponding alcohol27 which was
subsequently oxidized using Swern conditions to trans-(2-phenylcy-
clopropyl)carboxaldehyde.19
Addition of trans-(2-Phenylcyclopropyl)carboxaldehyde to Tet-
ramesityldisilene. Mes2Si(SiMe3)2 (100 mg, 0.24 mmol) was dissolved
in hexanes (8 mL) and photolyzed (254 nm) at -60 °C for 12 h. trans-
(2-Phenylcyclopropyl)carboxaldehyde was added dropwise to this
solution until the yellow color of the disilene had faded (approximately
5 drops). Following removal of the solvent, the reaction mixture was
purified by preparative thin-layer chromatography (30:70 CH2Cl2/
hexanes) to yield approximately a 1:1 ratio of compounds 1 and 2.
The mixture was separated by additional preparative thin-layer chro-
matography (5:95 Et2O/hexanes) to give compound 1 (23 mg, 28%)
and compound 2 (15 mg, 18%) (combined yield 46%).28
2.45 (vbr s), 2.10 (s), 2.08 (s), 2.05 (s, dCsCHH and Mes CH3); 13
C
NMR (ppm) 146.04, 145.37 (Mes C and Ph C), 140.63 (OsCHdCH),
138.75, 138.64, 138.31, 137.76 (Mes C and Ph C), 131.06 (Ph CH),
130.28, 129.56, 128.83 (Mes CH), 127.01, 125.07 (Ph CH), 111.76
(OsCHdCH), 43.51 (CHPh), 33.03 (br s, CHH), 28.03 (br s), 26.88
(br s), 26.20 (br s), 25.85 (br s), 23.52 (br s), 20.88, 20.69 (Mes CH3);
29Si NMR (ppm) -0.53 (SisCHR), -11.83 (SisO); MS (m/z) 678
(M+, 14), 549 (Mes4Si2O, 14), 439 (Mes2SiOC4H4Ph, 28), 267 (Mes2-
Si + H, 100), 147 (MesSi, 100); high-resolution MS calcd for C46H54-
Si2O 678.3713, found 678.3602.
Addition of trans-(2-Phenylcyclopropyl)carboxaldehyde to Tet-
ramesitylgermasilene. SiGe2Mes6 (50 mg, 0.058 mmol) and Et3SiH
(0.5 mL, excess) were dissolved in toluene (4 mL) and photolyzed (350
nm) at -70 °C for 12 h. To this solution was added trans-(2-
phenylcyclopropyl)carboxaldehyde dropwise until the yellow color of
the germasilene had faded (approximately 5 drops). The yellow color
did not entirely fade upon addition of the aldehyde, so the reaction
mixture was allowed to warm to 22 °C and stand for an additional 2 h.
After this time, there was no further change in color. Following
removal of the solvent, the reaction mixture was separated by
preparative thin-layer chromatography (30:70 CH2Cl2/hexanes) to give
compounds 4 and 5 as a 1:1 mixture. This 1:1 mixture was separated
by additional preparative thin-layer chromatography (5:95 Et2O/
hexanes) to give compound 4 (13 mg, 31%) and compound 5 (8 mg,
19%) (combined yield 50%).28
(22) Wind, M.; Powell, D. R.; West, R. Organometallics 1996, 15, 5772-
5773.
(23) (a) Wiberg, N.; Wagner, G.; Mu¨ller, G.; Riede, J. J. Organomet.
Chem. 1984, 271, 381-391. (b) Wiberg, N.; Wagner, G.; Reber, G.; Riede,
J.; Mu¨ller, G. Organometallics 1987, 6, 35-41.
(24) (a) Wiberg, N.; Schurz, K.; Mu¨ller, G.; Riede, J. Angew. Chem.,
Int. Ed. Engl. 1988, 27, 935-936. (b) Wiberg, N.; Schurz, K. J. Organomet.
Chem. 1988, 341, 145-163. (c) Walter, S.; Klingebiel, U.; Schmidt-Ba¨se,
D. J. Organomet. Chem. 1991, 412, 319-326. (d) Denk, M.; Hayashi, R.
K.; West, R. J. Am. Chem. Soc. 1994, 116, 10813-10814.
(25) Fink, M. J.; Michalczyk, M. J.; Haller, K. J.; West, R.; Michl, J.
Organometallics 1984, 3, 793-800.
(26) Baines, K. M.; Cooke, J. A. Organometallics 1991, 10, 3419-3421.
(27) Smart, R. P.; Peelen, T. J.; Blankespoor, R. L.; Ward, D. L. J. Am.
Chem. Soc. 1997, 119, 461-465.
[Dimesityl(2-cis-4-trans-1-oxa-5-phenylpentadienyl)silyl]dimesi-
tylgermane (4): IR (thin film, cm-1) 3022 (m), 2967 (s), 2912 (s),
2857 (m), 2028 (Ge-H, w), 1732 (w), 1691 (w), 1609 (s), 1443 (s),
1061 (m), 1028 (m), 978 (m), 853 (m); 1H NMR (ppm) 7.64 (dd, 1 H,
3
3
OsCHdCHsCHdCHsPh, J ) 15.9 Hz, J ) 11.1 Hz), 7.48 (d, 2
(28) While the isolated yields of the adducts are low, they appear as the
major components in the crude reaction mixture, as determined by 1H NMR
spectroscopy. Loss of material appears to occur during the separation and
purification of the compounds by chromatography.
3
3
H, Ph, J ) 7.4 Hz), 7.26 (t, 2 H, Ph, J ) 7.7 Hz), 7.07 (t, 1 H, Ph,
3J ) 7.4 Hz), 6.77 (s, 4 H, Mes CH), 6.65 (s, 4 H, Mes CH), 6.39 (d,
3
1 H, OsCHdCHsCHdCHsPh, J ) 16.1 Hz), 6.22 (d, 1 H, Os

Cycloaddition Reactions of Aldehydes
J. Am. Chem. Soc., Vol. 120, No. 43, 1998 11053
CHdCHsCHdCHsPh, 3J ) 5.6 Hz), 5.65 (s, 1 H, Ge-H), 5.38 (dd,
4.13-4.22 (br m, CHPh), 3.44-3.60 (vbr m, dCsCHH), 2.86 (br s,
Mes CH3), 2.58 (br s, Mes CH3), 2.29 (vbr s, Mes CH3), 2.102 (s),
2.095 (s), 2.07 (s), (Mes CH3), 1.54 (br s, Mes CH3); 13C NMR (ppm)
145.60 (br s), 145.52, 144.12, 143.71 (br s), 143.20 (br s), 142.85 (br
s), 140.55, 139.02, 138.94, 137.52, 137.05, 135.25 (br s), 130.60,
129.75, 129.31, 128.93, 128.58, 127.06, 125.29, 112.28, 48.66, 33.01,
28.06 (br s), 26.32 (br s), 25.54 (br s), 20.89, 20.74; 29Si NMR (ppm)
2.43; MS (m/z) 724 (M+, 2), 594 (Mes2SiOGeMes2,19), 474 (MesSi-
OGeMes2 - H, 23), 411 (Mes2SiOC4H4Ph, 43), 385 (Mes3Si, 55), 312
(Mes2Ge, 90), 267 (Mes2SiH, 100), 192 (MesGe - H, 94); high-
resolution MS calcd for C46H54SiGeO 724.3156, found 724.3556.
3
3
1 H, OsCHdCHsCHdCHsPh, J ) 5.7 Hz, J ) 10.9 Hz), 2.36
(br s, Mes o-CH3), 2.34 (br s, Mes o-CH3) (24 H total), 2.15 (s, 6 H,
Mes p-CH3), 2.06 (s, 6 H, Mes p-CH3); 13C NMR (ppm) 144.75, 144.19
(Mes C and Ph C), 141.26 (OsCHdCHsCHdCHsPh), 139.98,
138.94, 138.05, 135.24, 131.08 (Mes C and Ph C), 129.93 (Mes CH),
128.96 (Mes CH and Ph CH, coincident peaks), 128.55 (OsCHd
CHsCHdCHsPh), 127.78, 126.50 (Ph CH), 123.99 (OsCHdCHs
CHdCHsPh), 111.03 (OsCHdCHsCHdCHsPh), 24.79, 23.88,
21.06, 20.99 (Mes CH3); 29Si NMR (ppm) 3.75 (Si-O); MS (m/z) 594
(Mes2SiO s GeMes2, 12), 474 (MesSiO s GeMes2 s H, 11), 411
(Mes2SiOCHCHCHCHPh, 100), 283 (Mes2SiO + H, 62).
2,2,3,3-Tetramesityl-4-phenyloxa-2,3-silagermacyclohept-6-ene
(5): IR (thin film, cm-1) 3024 (m), 2920 (s), 2867 (m), 1652 (m),
1611 (s), 1559 (w), 1454 (s), 1409 (w), 1383 (w), 1290 (m), 1133 (m),
Acknowledgment. We thank Natural Sciences and Engi-
neering Research of Canada for financial support. We also thank
Professors Martin Newcomb and Mark S. Workentin for useful
discussions and Robert L. Donkers for technical assistance.
1
1036 (m), 856 (m); H NMR (ppm) 6.84-6.97 (m, Ph CH and Mes
CH), 6.69 (br s, Mes CH), 6.61 (br s, Mes CH), 6.40 (br s), 6.35 (d,3J
) 6 Hz, Mes CH and OsCHdCH), 4.79-4.86 (m, OsCHdCH),
JA9812002