Final Report Summary - NEWSILICON (Low-valent silicon complexes:Transition metal-like catalysts)
Silicon (25.8 %) is second most abundant element in the earth’s crust. This quantity of silicon is indeed enormous compared with those of transition metals [Pt (3x10-7 %), Au (1.1x10-7 %), Ru (1x10-7 %), Pd (6x10-8 %), Rh (2x10-8 %)]. The project concerns the investigation of silicon complexes with transition metal like behaviors. Ultimate objective of this project is to find performant silicon complexes with an appropriate ligand system which can replace precious transition metals in catalysis.
Recently we have discovered the easy synthesis of various stable divalent silicon [Si(II) or silylene] complexes with a phosphine ligand. In those complexes, silicon and phosphine fragments are connected by a bridging backbone with a moderate flexibility, which is very important to control their stability and reactivity. Indeed, the obtained Si(II) complexes are perfectly stable but, due to the weak Si-P interaction, they keep a high reactivity similar to that of transient silylenes. This is a very important discovery and thanks to this weak interaction, those silicon complexes can activate various small molecules even those thought to be impossible to be activated without transition metals such as non-polarized olefins or silanes. More importantly, the ethylene binding to the silylene center was found to be labile and reversible at room temperature. This is indeed an ordinary phenomenon in transition metal chemistry but it is unusual in main group element chemistry. Moreover, in the case of complexes featuring a labile substituent such a hydride (-H) or stannyl group (-SnMe3), the olefin inserts into the Si(II)-H or -Sn bond to regenerate a new Si(II)-complex. Of particular importance, in some cases, the reaction is even reversible at room temperature, which can be regarded as an insertion/beta-elimination process (Angew. Chem. 2013, 52, 8437). This process is a key step in several catalytic reactions such as Heck reaction. Furthermore, they also reversibly react with silanes (R3Si-H) at RT via oxidative addition / reductive elimination, which are two of the most fundamental and important processes for transition metal complexes (Angew. Chem. 2016, 55, 14355).
Another important and fruitful discovery was the synthesis of the first base-stabilized sila-cyclopropylidene a three-membered cyclic silylene (Angew. Chem. 2012, 52, 605). This molecule with a highly strained small cyclic structure shows a versatile reactivity and unique properties. Particularly, it behaves as a strongly nucleophilic ligand towards transition metals. Of special interest, the corresponding Pt(0)-complex exhibits a high stability and a high level of selectivity and catalytic activity in hydrosilylation reactions (Patent 2013, WO2015004397, Inorg. Chem. 2016, 55, 8234). In addition, under thermal or photolytic conditions this sila-cyclopropylidene can be isomerized to give an original silene (silicon analogue of olefins) possessing unique properties (Angew 2013, 52, 10840) and being an excellent ligand for transition metals (the corresponding Pt(0)-complex shows an exceptional catalytic activity for the hydrosilylation of olefins - patent 2013, WO2015004396).
Furthermore, the oxidation of the sila-cyclopropylidene affords rare base-stabilized silicon oxides. It should be noted that silicon oxides are extremely reactive species and stable silicon analogues of ketones are still unknown. We have demonstrated that the oxidation of the cyclic silylene complex with N2O leads the first stable sila-cyclopropanone complex (Angew. Chem. 2013, 52, 4426). In contrast, the use of dioxygen (O2) as reagent allowed the synthesis of the first silicon containing four-membered cyclic ester (sila-beta-lactone) (Angew. Chem. 2013, 52, 8980). Interestingly, silanone functions make significantly fragile small rings. Indeed, the four-membered cyclic silanone decomposes even at room temperature via a [2+2]-retro-cycloaddition, just like the metalacylic intermediate of the olefin metathesis reaction! (JACS, 2016, 138, 2965). In fact, we have demonstrated that silaketene derivatives can be involved in an olefin metathesis type reaction, via the transient formation of a four membered ring fragelized by a silanone fragment (Chem. Eur. J. 2016, 22, 10247). The development of appropriate silaketene complexes should allow to realize the metal-free catalytic olefin metathesis reactions. A similar [2+2]-retro-cycloaddition reaction of the related sila-beta-lactone allowed the isolation of the first stable SiO2-complex (a monomer of sand stabilized by a donor/acceptor ligand system) (Angew. Chem. 2017, 56, 3935).
We have also found that phosphine-stabilized silylenes are extremely useful precursors for the synthesis of various unique molecules and new functionalities such as the first boron-based phosphonium ylide (bora-ylide) (Angew. Chem. 2017, 56, 4814). This boron-based chemical function presents an exceptionally strong electron donating character, which allows to stabilize various silicon-based highly reactive molecules (silylenes and silylium cations), which are generally impossible to be stabilized by other ways (Angew. Chem. 2017, 56, 10549). Using the combination of two different donating substituents (amino and phosphonium-ylide) we have isolated a new heterocyclic silylene which presents a high thermal stability and an exceptional strong nucleophilic character (Angew. Chem. 2016, 56, 16141). Moreover, the oxidation of this heterocyclic silylene led to the isolation of the first fairly stable crystalline silanone, derivatives which were known as Kipping’s dream (Angew. Chem. 2017, 56, 10481).
Although we have not yet realized any catalytic reactions using silicon based catalysts, we have successfully demonstrated that the related phosphine-stabilized germanium analogues (germanium is located just below silicon in the periodic table, and therefore it presents similar properties) can be efficient catalysts for the hydrosilylation of ketones (Patent 2014, WO2016075414 and 2016, FR 1600757). We have also synthesized a new stable heterocyclic amino(P-ylide)germylene showing an enhanced reactivity compared to classical N-heterocyclic germylenes (Angew. Chem. 2016, 55, 4753). Of particular interest, the corresponding borane-complex, presenting multi-reactive sites, is able to activate simultaneously two small molecules like transition metals (extremely rare for non-metallic species)! Taking advantage of this unique phenomenon, we have successfully realized the selective reduction of CO2 to diacetal derivative, which is difficult to be achieved even using transition metal complexes (Angew. Chem. 2017, 56, 1365). We actually try to extend it to silicon analogues.
Recently we have discovered the easy synthesis of various stable divalent silicon [Si(II) or silylene] complexes with a phosphine ligand. In those complexes, silicon and phosphine fragments are connected by a bridging backbone with a moderate flexibility, which is very important to control their stability and reactivity. Indeed, the obtained Si(II) complexes are perfectly stable but, due to the weak Si-P interaction, they keep a high reactivity similar to that of transient silylenes. This is a very important discovery and thanks to this weak interaction, those silicon complexes can activate various small molecules even those thought to be impossible to be activated without transition metals such as non-polarized olefins or silanes. More importantly, the ethylene binding to the silylene center was found to be labile and reversible at room temperature. This is indeed an ordinary phenomenon in transition metal chemistry but it is unusual in main group element chemistry. Moreover, in the case of complexes featuring a labile substituent such a hydride (-H) or stannyl group (-SnMe3), the olefin inserts into the Si(II)-H or -Sn bond to regenerate a new Si(II)-complex. Of particular importance, in some cases, the reaction is even reversible at room temperature, which can be regarded as an insertion/beta-elimination process (Angew. Chem. 2013, 52, 8437). This process is a key step in several catalytic reactions such as Heck reaction. Furthermore, they also reversibly react with silanes (R3Si-H) at RT via oxidative addition / reductive elimination, which are two of the most fundamental and important processes for transition metal complexes (Angew. Chem. 2016, 55, 14355).
Another important and fruitful discovery was the synthesis of the first base-stabilized sila-cyclopropylidene a three-membered cyclic silylene (Angew. Chem. 2012, 52, 605). This molecule with a highly strained small cyclic structure shows a versatile reactivity and unique properties. Particularly, it behaves as a strongly nucleophilic ligand towards transition metals. Of special interest, the corresponding Pt(0)-complex exhibits a high stability and a high level of selectivity and catalytic activity in hydrosilylation reactions (Patent 2013, WO2015004397, Inorg. Chem. 2016, 55, 8234). In addition, under thermal or photolytic conditions this sila-cyclopropylidene can be isomerized to give an original silene (silicon analogue of olefins) possessing unique properties (Angew 2013, 52, 10840) and being an excellent ligand for transition metals (the corresponding Pt(0)-complex shows an exceptional catalytic activity for the hydrosilylation of olefins - patent 2013, WO2015004396).
Furthermore, the oxidation of the sila-cyclopropylidene affords rare base-stabilized silicon oxides. It should be noted that silicon oxides are extremely reactive species and stable silicon analogues of ketones are still unknown. We have demonstrated that the oxidation of the cyclic silylene complex with N2O leads the first stable sila-cyclopropanone complex (Angew. Chem. 2013, 52, 4426). In contrast, the use of dioxygen (O2) as reagent allowed the synthesis of the first silicon containing four-membered cyclic ester (sila-beta-lactone) (Angew. Chem. 2013, 52, 8980). Interestingly, silanone functions make significantly fragile small rings. Indeed, the four-membered cyclic silanone decomposes even at room temperature via a [2+2]-retro-cycloaddition, just like the metalacylic intermediate of the olefin metathesis reaction! (JACS, 2016, 138, 2965). In fact, we have demonstrated that silaketene derivatives can be involved in an olefin metathesis type reaction, via the transient formation of a four membered ring fragelized by a silanone fragment (Chem. Eur. J. 2016, 22, 10247). The development of appropriate silaketene complexes should allow to realize the metal-free catalytic olefin metathesis reactions. A similar [2+2]-retro-cycloaddition reaction of the related sila-beta-lactone allowed the isolation of the first stable SiO2-complex (a monomer of sand stabilized by a donor/acceptor ligand system) (Angew. Chem. 2017, 56, 3935).
We have also found that phosphine-stabilized silylenes are extremely useful precursors for the synthesis of various unique molecules and new functionalities such as the first boron-based phosphonium ylide (bora-ylide) (Angew. Chem. 2017, 56, 4814). This boron-based chemical function presents an exceptionally strong electron donating character, which allows to stabilize various silicon-based highly reactive molecules (silylenes and silylium cations), which are generally impossible to be stabilized by other ways (Angew. Chem. 2017, 56, 10549). Using the combination of two different donating substituents (amino and phosphonium-ylide) we have isolated a new heterocyclic silylene which presents a high thermal stability and an exceptional strong nucleophilic character (Angew. Chem. 2016, 56, 16141). Moreover, the oxidation of this heterocyclic silylene led to the isolation of the first fairly stable crystalline silanone, derivatives which were known as Kipping’s dream (Angew. Chem. 2017, 56, 10481).
Although we have not yet realized any catalytic reactions using silicon based catalysts, we have successfully demonstrated that the related phosphine-stabilized germanium analogues (germanium is located just below silicon in the periodic table, and therefore it presents similar properties) can be efficient catalysts for the hydrosilylation of ketones (Patent 2014, WO2016075414 and 2016, FR 1600757). We have also synthesized a new stable heterocyclic amino(P-ylide)germylene showing an enhanced reactivity compared to classical N-heterocyclic germylenes (Angew. Chem. 2016, 55, 4753). Of particular interest, the corresponding borane-complex, presenting multi-reactive sites, is able to activate simultaneously two small molecules like transition metals (extremely rare for non-metallic species)! Taking advantage of this unique phenomenon, we have successfully realized the selective reduction of CO2 to diacetal derivative, which is difficult to be achieved even using transition metal complexes (Angew. Chem. 2017, 56, 1365). We actually try to extend it to silicon analogues.