Generation of Bis(ferrocenyl)silylenes from Siliranes †

Divalent silicon species, the so-called silylenes, represent attractive organosilicon building blocks. Isolable stable silylenes remain scarce, and in most hitherto reported examples, the silicon center is stabilized by electron-donating substituents (e.g., heteroatoms such as nitrogen), which results in electronic perturbation. In order to avoid such electronic perturbation, we have been interested in the chemistry of reactive silylenes with carbon-based substituents such as ferrocenyl groups. Due to the presence of a divalent silicon center and the redox-active transition metal iron, ferrocenylsilylenes can be expected to exhibit interesting redox behavior. Herein, we report the design and synthesis of a bis(ferrocenyl)silirane as a precursor for a bis(ferrocenyl)silylene, which could potentially be used as a building block for redox-active organosilicon compounds. It was found that the isolated bis(ferrocenyl)siliranes could be a bottleable precursor for the bis(ferrocenyl)silylene under mild conditions.

Molecules 2020, 25, x 2 of 12 steric protection, the products from the aforementioned reactions usually afford complex mixtures of oligosilanes. Conversely, most cases of the reduction of sterically hindered dihalosilanes or the photolysis of trisilanes that bear bulky substituents result in the facile formation of a disilene, i.e., the formal dimer of a silylene [33][34][35][36][37][38][39]. To prepare a stable silylene precursor, it should thus be required to introduce very bulky substituents on the silicon atom to avoid facile dimerization of the generated silylenes [33]. On the other hand, too bulky substituents can be expected to negatively affect the inherent reactivity of the resulting silylene. In order to find the right balance in this trade-off relationship, transient silylenes with a combination of very bulky and less hindered groups have been designed. Our group has already reported the design and synthetic utility of the sterically demanding 2,5-bis(3,5-di-t-butylphenyl)-1-ferrocenyl (Fc*) group [40][41][42], and this group should be an appropriate substituent for bis(ferrocenyl)silylene 1 (Chart 1), which bears Fc* and ferrocenyl (Fc) groups on the silicon center [43]. Silylene 1 can be expected to (i) be very reactive due to its low-lying LUMO level, and (ii) show intriguing redox activity on account of its two ferrocenyl moieties.
As the reduction of diorganodichlorosilanes represents an efficient method for the generation of silylenes [4,33], we speculated that the reduction of bis(ferrocenyl)dichlorosilane 4 might furnish silylene 1. However, the generation of 1 by the reduction of 4 was not observed when common metal reductants such as KC8, Na, or Li were employed, and only complex mixtures including Fc*H were As the reduction of diorganodichlorosilanes represents an efficient method for the generation of silylenes [4,33], we speculated that the reduction of bis(ferrocenyl)dichlorosilane 4 might furnish silylene 1. However, the generation of 1 by the reduction of 4 was not observed when common metal reductants such as KC 8 , Na, or Li were employed, and only complex mixtures including Fc*H were obtained [41]. Conversely, when 4 was treated with lithium naphthalenide in THF at −78 • C, 2-naphthylsilane 5 was isolated in 30% yield, and its structure was unambiguously determined by spectroscopic and XRD analyses [46]. The formation of 5 should most likely be interpreted in terms of a C-H insertion of the reactive intermediate of silylene 1 with naphthalene. It should be noted here that it has previously been reported that isolable or transient silylenes react with aromatic compounds such as naphthalene and benzene to give the corresponding silepine or silirane derivatives via C-C insertion and [1+2]cycloaddition reactions, respectively [33,[47][48][49][50]. Thus, the formation of 5 suggests the unique reaction manner of a silylene with an aromatic compound. Moreover, an isolable zwitterionic silylene has been reported to cleave a C-H bond of benzophenone via a 1,3-hydrogen shift [51,52].
At this point, we suspected that 5 was formed by the reaction of silylene 1 with naphthalene via the corresponding [1+2]cycloadduct, silirane 6, which would be a conceivable intermediate in the reaction of a silylene with naphthalene. In order to clarify the reaction mechanism, the potential energy surfaces (PESs) of the reactions of silylene 1 with naphthalene were calculated at the B3PW91/6-311G(2d,p) level of theory (E ZPE : relative energies including zero-point energy corrections) [53]. The [1+2]cycloaddition of 1 with naphthalene was found to be slightly exothermic (∆E ZPE = 6.12 kcal/mol) to give silirane 6 with a barrier of ∆E ZPE ‡ = 21.6 kcal/mol. Unexpectedly, silirane 6 was not connected to 5 on the PES in these calculations. Instead, the highly exothermic (∆E ZPE = 37.4 kcal/mol) direct C-H insertion pathway of the reaction of 1 with naphthalene to furnish 5 was found to have a higher barrier of ∆E ZPE ‡ = 29.5 kcal/mol. The PES also showed that silirane 6 could give 1 with concomitant elimination of naphthalene via the reverse reaction of the formation of 6 (∆E ZPE ‡ = 22.3 kcal/mol). Thus, it can be concluded 5 and 6 should be the thermodynamic and kinetic products, respectively, in the reaction of silylene 1 with naphthalene. At this stage, however, other plausible pathways for the formation of 5, e.g., by the reaction of dichlorosilane 4 with lithium naphthalenide, which would not include the generation of silylene 1, cannot be excluded from consideration (Scheme 2). However, based on the results of the theoretical calculations, silirane derivatives with a Si-C-C three-membered ring skeleton could potentially be considered appropriate precursors for silylenes such as 1.
Molecules 2020, 25, x 3 of 12 obtained [41]. Conversely, when 4 was treated with lithium naphthalenide in THF at −78 °C, 2naphthylsilane 5 was isolated in 30% yield, and its structure was unambiguously determined by spectroscopic and XRD analyses [46]. The formation of 5 should most likely be interpreted in terms of a C-H insertion of the reactive intermediate of silylene 1 with naphthalene. It should be noted here that it has previously been reported that isolable or transient silylenes react with aromatic compounds such as naphthalene and benzene to give the corresponding silepine or silirane derivatives via C-C insertion and [1+2]cycloaddition reactions, respectively [33,[47][48][49][50]. Thus, the formation of 5 suggests the unique reaction manner of a silylene with an aromatic compound. Moreover, an isolable zwitterionic silylene has been reported to cleave a C-H bond of benzophenone via a 1,3-hydrogen shift [51,52]. At this point, we suspected that 5 was formed by the reaction of silylene 1 with naphthalene via the corresponding [1+2]cycloadduct, silirane 6, which would be a conceivable intermediate in the reaction of a silylene with naphthalene. In order to clarify the reaction mechanism, the potential energy surfaces (PESs) of the reactions of silylene 1 with naphthalene were calculated at the B3PW91/6-311G(2d,p) level of theory (EZPE: relative energies including zero-point energy corrections) [53]. The [1+2]cycloaddition of 1 with naphthalene was found to be slightly exothermic (ΔEZPE = 6.12 kcal/mol) to give silirane 6 with a barrier of ΔEZPE ‡ = 21.6 kcal/mol. Unexpectedly, silirane 6 was not connected to 5 on the PES in these calculations. Instead, the highly exothermic (ΔEZPE = 37.4 kcal/mol) direct C-H insertion pathway of the reaction of 1 with naphthalene to furnish 5 was found to have a higher barrier of ΔEZPE ‡ = 29.5 kcal/mol. The PES also showed that silirane 6 could give 1 with concomitant elimination of naphthalene via the reverse reaction of the formation of 6 (ΔEZPE ‡ = 22.3 kcal/mol). Thus, it can be concluded 5 and 6 should be the thermodynamic and kinetic products, respectively, in the reaction of silylene 1 with naphthalene. At this stage, however, other plausible pathways for the formation of 5, e.g., by the reaction of dichlorosilane 4 with lithium naphthalenide, which would not include the generation of silylene 1, cannot be excluded from consideration (Scheme 2). However, based on the results of the theoretical calculations, silirane derivatives with a Si-C-C three-membered ring skeleton could potentially be considered appropriate precursors for silylenes such as 1. When a THF/Et2O solution of dichlorosilane 4 was reduced with sodium in the presence of an excess of cyclohexene and a small amount of naphthalene, a mixture of siliranes 9a and 9b was obtained (86% yield, 9a:9b = 66:34). Even though it was difficult to purify each product by common separation techniques such as gel permeation chromatography (GPC), column chromatography, or recrystallization, a few crystals of 9a and 9b were obtained by recrystallization from hexane and When a THF/Et 2 O solution of dichlorosilane 4 was reduced with sodium in the presence of an excess of cyclohexene and a small amount of naphthalene, a mixture of siliranes 9a and 9b was obtained (86% yield, 9a:9b = 66:34). Even though it was difficult to purify each product by common separation techniques such as gel permeation chromatography (GPC), column chromatography, or recrystallization, a few crystals of 9a and 9b were obtained by recrystallization from hexane and pentane, respectively. Thus, siliranes 9a and 9b were fully characterized by spectroscopic analyses, while 9a was structurally characterized by XRD analysis [46]. The similar up-field shifted 29 Si NMR chemical shifts of 9a (δ Si = −67.2) and 9b (δ Si = −69.8) also indicated a silirane structure for 9b, suggesting 9b could be a stereoisomer of 9a (Scheme 3). At this stage, 9a and 9b could not be separated from each other due the close similarity of their chemical properties. Heating a cyclohexene solution of a mixture of 9a and 9b (94:6 ratio) shifted the compositional ratio in favor of 9b (9a:9b = 45:55), suggesting that 9a and 9b represent the kinetic and thermodynamic products in the reaction of 1 with cyclohexene, respectively. Both 9a and 9b are stable under ambient conditions in the solid state and in C 6 D 6 solution. Interestingly, the conversion of 9a to 9b occurred in the solid state. After heating a mixture of solid 9a and 9b (ca. 74:26) at 120 • C for 30 min under reduced pressure, the 1 H NMR spectrum of the solid dissolved in C 6 D 6 showed a 9a:9b ratio of ca. 45:55, suggesting a transformation of 9a to 9b even in the solid state.
while 9a was structurally characterized by XRD analysis [46]. The similar up-field shifted 29 Si NMR chemical shifts of 9a (δSi = −67.2) and 9b (δSi = −69.8) also indicated a silirane structure for 9b, suggesting 9b could be a stereoisomer of 9a (Scheme 3). At this stage, 9a and 9b could not be separated from each other due the close similarity of their chemical properties. Heating a cyclohexene solution of a mixture of 9a and 9b (94:6 ratio) shifted the compositional ratio in favor of 9b (9a:9b = 45:55), suggesting that 9a and 9b represent the kinetic and thermodynamic products in the reaction of 1 with cyclohexene, respectively. Both 9a and 9b are stable under ambient conditions in the solid state and in C6D6 solution. Interestingly, the conversion of 9a to 9b occurred in the solid state. After heating a mixture of solid 9a and 9b (ca. 74:26) at 120 °C for 30 min under reduced pressure, the 1 H NMR spectrum of the solid dissolved in C6D6 showed a 9a:9b ratio of ca. 45:55, suggesting a transformation of 9a to 9b even in the solid state.
The XRD structure of 9a (Figure 1) shows that the cyclohexane moiety of 9a is oriented toward the crowded space close to the Fc* group; 9b could thus potentially exhibit a more stable geometry wherein the cyclohexyl group is oriented toward the less bulky ferrocenyl moiety. Theoretical calculations at B3PW91-D3(BJ)/6-311G(3d) level of theory suggest that 9b is by 0.78 kcal/mol more stable than 9a. Scheme 3. Synthesis of bis(ferrocenyl)siliranes 9a and 9b. Figure 1. Molecular structures of (a) 5, (b) 9a, and (c) 14 with thermal ellipsoids at 50% probability; hydrogen atoms other than that of the Si-H moiety are omitted for clarity.
As described above, heating of the cyclohexene solution of siliranes 9a and 9b at 75 °C resulted in the conversion of 9a to 9b with concomitant formation of minor amounts of Fc*FcSiH(OH) (10) [46], suggesting the generation of silylene 1 in the equilibrium state. The formation of small amounts of 10 should most likely be interpreted in terms of a hydrolysis of silylene 1 generated by thermolysis of 9a and/or 9b due to the inevitable contamination with a small amount of moisture. Dissolving a mixture of 9a and 9b (66:34) in an excess amount of methanol at r.t. afforded 11 (28%), 12 (10%), and 13 (15%) as shown in Scheme 4 [6,46]. The formation of 11 suggests the generation of silylene 1 at r.t. However, siliranes 9a and/or 9b would probably undergo alcoholysis with MeOH to give 12 and 13 The XRD structure of 9a ( Figure 1) shows that the cyclohexane moiety of 9a is oriented toward the crowded space close to the Fc* group; 9b could thus potentially exhibit a more stable geometry wherein the cyclohexyl group is oriented toward the less bulky ferrocenyl moiety. Theoretical calculations at B3PW91-D3(BJ)/6-311G(3d) level of theory suggest that 9b is by 0.78 kcal/mol more stable than 9a.
Molecules 2020, 25, x 4 of 12 pentane, respectively. Thus, siliranes 9a and 9b were fully characterized by spectroscopic analyses, while 9a was structurally characterized by XRD analysis [46]. The similar up-field shifted 29 Si NMR chemical shifts of 9a (δSi = −67.2) and 9b (δSi = −69.8) also indicated a silirane structure for 9b, suggesting 9b could be a stereoisomer of 9a (Scheme 3). At this stage, 9a and 9b could not be separated from each other due the close similarity of their chemical properties. Heating a cyclohexene solution of a mixture of 9a and 9b (94:6 ratio) shifted the compositional ratio in favor of 9b (9a:9b = 45:55), suggesting that 9a and 9b represent the kinetic and thermodynamic products in the reaction of 1 with cyclohexene, respectively. Both 9a and 9b are stable under ambient conditions in the solid state and in C6D6 solution. Interestingly, the conversion of 9a to 9b occurred in the solid state. After heating a mixture of solid 9a and 9b (ca. 74:26) at 120 °C for 30 min under reduced pressure, the 1 H NMR spectrum of the solid dissolved in C6D6 showed a 9a:9b ratio of ca. 45:55, suggesting a transformation of 9a to 9b even in the solid state. The XRD structure of 9a (Figure 1) shows that the cyclohexane moiety of 9a is oriented toward the crowded space close to the Fc* group; 9b could thus potentially exhibit a more stable geometry wherein the cyclohexyl group is oriented toward the less bulky ferrocenyl moiety. Theoretical calculations at B3PW91-D3(BJ)/6-311G(3d) level of theory suggest that 9b is by 0.78 kcal/mol more stable than 9a.  As described above, heating of the cyclohexene solution of siliranes 9a and 9b at 75 °C resulted in the conversion of 9a to 9b with concomitant formation of minor amounts of Fc*FcSiH(OH) (10) [46], suggesting the generation of silylene 1 in the equilibrium state. The formation of small amounts of 10 should most likely be interpreted in terms of a hydrolysis of silylene 1 generated by thermolysis of 9a and/or 9b due to the inevitable contamination with a small amount of moisture. Dissolving a mixture of 9a and 9b (66: 34) in an excess amount of methanol at r.t. afforded 11 (28%), 12 (10%), and 13 (15%) as shown in Scheme 4 [6,46]. The formation of 11 suggests the generation of silylene 1 at r.t. However, siliranes 9a and/or 9b would probably undergo alcoholysis with MeOH to give 12 and 13 hydrogen atoms other than that of the Si-H moiety are omitted for clarity.
As described above, heating of the cyclohexene solution of siliranes 9a and 9b at 75 • C resulted in the conversion of 9a to 9b with concomitant formation of minor amounts of Fc*FcSiH(OH) (10) [46], suggesting the generation of silylene 1 in the equilibrium state. The formation of small amounts of 10 should most likely be interpreted in terms of a hydrolysis of silylene 1 generated by thermolysis of 9a and/or 9b due to the inevitable contamination with a small amount of moisture. Dissolving a mixture of 9a and 9b (66: 34) in an excess amount of methanol at r.t. afforded 11 (28%), 12 (10%), and 13 (15%) as shown in Scheme 4 [6,46]. The formation of 11 suggests the generation of silylene 1 at r.t. However, siliranes 9a and/or 9b would probably undergo alcoholysis with MeOH to give 12 and 13 under these conditions. Thus, siliranes 9a and 9b could be appropriate thermal precursors for silylene 1 on heating, although they are sensitive toward protic solvents. Conversely, heating of a mixture of 9a and 9b (25:86) at 60 • C for 41 h in the presence of an excess of 2,3-dimethyl-1,3-butadiene afforded silolene 14 as the corresponding [1+4]cycloadduct of silylene 1 with 2,3-dimethyl-1,3-butadiene in 44% isolated yield, suggesting that the thermolysis of both 9a and 9b affords silylene 1 at this temperature. Thus, it can be concluded that siliranes 9a and 9b could work as synthetic precursors for bis(ferrocenyl)silylene 1 in such pericyclic reactions.
Molecules 2020, 25, x 5 of 12 under these conditions. Thus, siliranes 9a and 9b could be appropriate thermal precursors for silylene 1 on heating, although they are sensitive toward protic solvents. Conversely, heating of a mixture of 9a and 9b (25:86) at 60 °C for 41 h in the presence of an excess of 2,3-dimethyl-1,3-butadiene afforded silolene 14 as the corresponding [1+4]cycloadduct of silylene 1 with 2,3-dimethyl-1,3-butadiene in 44% isolated yield, suggesting that the thermolysis of both 9a and 9b affords silylene 1 at this temperature. Thus, it can be concluded that siliranes 9a and 9b could work as synthetic precursors for bis(ferrocenyl)silylene 1 in such pericyclic reactions. Subsequently, we performed theoretical calculations on the dissociation energies of siliranes bearing several organic substituents at B3PW91-D3(BJ)/6-311G(3d) level of theory (Scheme 5). The dissociation energies of 9a and 9b are ca. 17 kcal/mol, which are smaller than those of phenyl-and methyl-substituted siliranes 17 (26.0 kcal/mol) and 19 (27.6 kcal/mol). Considering the small dissociation energy of less bulky bis(ferrocenyl)silirane 15 (14.0 kcal/mol), the facile generation of silylene 1 from 9a and 9b under mild conditions could be explained, not by the steric hindrance due to the bulky Fc* group, but the electronic effect of the ferrocenyl groups. Even though the electronic perturbation from the ferrocenyl groups toward the silirane moiety of 9a and 9b remains unclear at present, the electron-donating properties of the ferrocenyl groups can be expected to lower the dissociation energy of the silirane skeleton. Subsequently, we performed theoretical calculations on the dissociation energies of siliranes bearing several organic substituents at B3PW91-D3(BJ)/6-311G(3d) level of theory (Scheme 5). The dissociation energies of 9a and 9b are ca. 17 kcal/mol, which are smaller than those of phenyl-and methyl-substituted siliranes 17 (26.0 kcal/mol) and 19 (27.6 kcal/mol). Considering the small dissociation energy of less bulky bis(ferrocenyl)silirane 15 (14.0 kcal/mol), the facile generation of silylene 1 from 9a and 9b under mild conditions could be explained, not by the steric hindrance due to the bulky Fc* group, but the electronic effect of the ferrocenyl groups. Even though the electronic perturbation from the ferrocenyl groups toward the silirane moiety of 9a and 9b remains unclear at present, the electron-donating properties of the ferrocenyl groups can be expected to lower the dissociation energy of the silirane skeleton.
Molecules 2020, 25, x 5 of 12 under these conditions. Thus, siliranes 9a and 9b could be appropriate thermal precursors for silylene 1 on heating, although they are sensitive toward protic solvents. Conversely, heating of a mixture of 9a and 9b (25:86) at 60 °C for 41 h in the presence of an excess of 2,3-dimethyl-1,3-butadiene afforded silolene 14 as the corresponding [1+4]cycloadduct of silylene 1 with 2,3-dimethyl-1,3-butadiene in 44% isolated yield, suggesting that the thermolysis of both 9a and 9b affords silylene 1 at this temperature. Thus, it can be concluded that siliranes 9a and 9b could work as synthetic precursors for bis(ferrocenyl)silylene 1 in such pericyclic reactions. Subsequently, we performed theoretical calculations on the dissociation energies of siliranes bearing several organic substituents at B3PW91-D3(BJ)/6-311G(3d) level of theory (Scheme 5). The dissociation energies of 9a and 9b are ca. 17 kcal/mol, which are smaller than those of phenyl-and methyl-substituted siliranes 17 (26.0 kcal/mol) and 19 (27.6 kcal/mol). Considering the small dissociation energy of less bulky bis(ferrocenyl)silirane 15 (14.0 kcal/mol), the facile generation of silylene 1 from 9a and 9b under mild conditions could be explained, not by the steric hindrance due to the bulky Fc* group, but the electronic effect of the ferrocenyl groups. Even though the electronic perturbation from the ferrocenyl groups toward the silirane moiety of 9a and 9b remains unclear at present, the electron-donating properties of the ferrocenyl groups can be expected to lower the dissociation energy of the silirane skeleton.

Bis(ferrocenyl)dichlorosilane 4
Ferrocene (1.00 g, 5.38 mmol) was dissolved in THF (4.0 mL) and cooled to 0 • C. During 30 min, a pentane solution of t-BuLi (3.6 mL, 1.6 M in pentane, 5.76 mmol) was added dropwise. Then, hexane (10 mL) was added to the reaction mixture, and the solution was kept at −78 • C for 15 min. The resulting orange precipitate including ferrocenyl lithium was filtered off, and then washed with small portions of hexane. The orange precipitate was then dissolved in toluene (4 mL), and the solution was added to SiCl 4 (884 mg, 5.27 mmol). After stirring the mixture for 3 h at room temperature, the resulting inorganic salts were removed and the solvent of the filtrate was evaporated under reduced pressure to give ferrocenyltrichlorosilane 2 as an orange solid that was used for the subsequent reactions without further purification. An ether solution (2 mL) of Fc*Br (635 mg, 994 µmol) was treated with n-BuLi (0.5 mL, 2.60 M in hexane, 1.3 mmol) at 0 • C. After stirring for 6 h at room temperature, the solution of the resulting Fc*-Li (3) was added to 2 at 0 • C. After stirring for 12 h at room temperature, the solution was filtered and the filtrate was evaporated under reduced pressure. The obtained orange solid was subjected to GPC (toluene) to give bis(ferrocenyl)dichlorosilane 4 (366 mg, 434 µmol, 44%).

Siliranes 9a and 9b
A mixture of sodium (dispersion in mineral oil, 7.2 mg, 313 µmol), naphthalene (25.5 mg, 199 µmol), and cyclohexene (0.2 mL, 1.85 mmol) was added to a THF/Et 2 O (1:1, 0.4 mL) solution of 4 (92.3 mg, 109 µmol). After stirring for 12 h at room temperature, the solvent and residual cyclohexene were removed under reduced pressure, before the residue was dissolved in cyclohexane. After filtration, the solvent of the filtrate was evaporated under reduced pressure. The obtained orange solid was subjected to GPC (toluene) to give a mixture of 9a and 9b (66:34 ratio as judged by the 1 H-NMR spectrum, 80.5 mg, 94.0 µmol, total yield: 86%). Continuous recrystallization of the mixture from hexane afforded a few crystals of 9a, which could be isolated by the filtration. After evaporating the solvent of the filtrate, the residue was heated at 120 • C for a few hours under the reduced pressure. Spectroscopic and XRD analyses of the single crystals thus obtained enabled us to identify 9a. Silirane 9b was identified based on the spectral data including 29 Si NMR data, which was similar to those of 9a.

Thermolysis of Siliranes 9a and 9b in Cyclohexene
Cyclohexene (0.7 mL) was added to a mixture of 9a and 9b (94:6 ratio as judged by the 1 H NMR spectrum). After heating at 75 • C for 30 min, the residual cyclohexene was removed under reduced pressure. The 1 H NMR spectrum of the residue in C 6 D 6 showed the signals for 9a and 9b (45:55 ratio).

Thermolysis of Siliranes 9a and 9b in the Solid State
A mixture of the orange solids of 9a and 9b (74:26 ratio as judged by the 1 H-NMR spectrum) was heated at 120 • C for 30 min under evacuation in an NMR tube (5 mm diameter) equipped with a J-Young© tap. The 1 H-NMR spectrum of the residue in C 6 D 6 showed the signals for 9a and 9b (45:55 ratio).

Computational Methods
The level of theory and the basis sets used for the structural optimization are given in the main text. Frequency calculations confirmed minimum energies for all optimized structures. All calculations were carried out on the Gaussian 16 (Revision C.01) program package [53]. Computational time was generously provided by the Supercomputer Laboratory in the Institute for Chemical Research of Kyoto University.

Conclusions
Bis(ferrocenyl)siliranes 9a and 9b were prepared by the reduction of the corresponding dichlorosilane with lithium naphthalenide in the presence of an excess of cyclohexene. Siliranes 9a and 9b are appropriate precursors for bis(ferrocenyl)silylenes upon heating under mild conditions, i.e., they can be considered as bottleable synthetic precursors for silylenes. Further investigations into the creation of redox-active organosilicon compounds that bear ferrocenyl moieties by using siliranes 9a and b as silylene precursors are currently in progress in our laboratory.