Key words
Wittig reaction - stereoselectivity - redox catalysis -
meso-desymmetrization - atropisomers
1
Introduction
Kalipada Jana,from West Bengal, India, completed his MSc at the Indian Institute of Technology Kanpur. He earned his PhD from the University of Münster (WWU) under the guidance of Prof. Armido Studer. He then joined the University of Basel in Switzerland as a postdoctoral researcher, where he worked with Prof. Christof Sparr on PIII/PV=O redox catalysis.
Christof Sparr received his PhD from the ETH Zurich working in the group of Prof. Ryan Gilmour. He subsequently joined Prof. Dieter Seebach (ETH Zurich) and Prof. Steven V. Ley (University of Cambridge) for postdoctoral research. He then moved to the University of Basel and started his independent research as habilitand mentored by Prof. Karl Gademann in 2013. In 2016 Christof became assistant professor and in 2021 associate professor of organic chemistry. He is recipient of the ETH medal, a SNSF Starting Grant, the 2017 Werner Prize of the Swiss Chemical Society, the 2018 Ruzicka Prize of the ETH Zurich, and an ERC Consolidator Grant.
Since its discovery in 1954, the Wittig reaction has become an essential method for the synthesis of alkenes by the reaction of phosphorus ylides with aldehydes or ketones.[1]
[2]
[3]
[4] Despite its versatility, the reaction inherently produces stoichiometric amounts of phosphine oxide as a byproduct, which limits its overall economy. In response, recent research focused on developing a catalytic Wittig process that operates through a PIII/PV=O redox cycle. Initial efforts by the O’Brien group demonstrated that the use of catalytic amounts of phosphine oxide combined with a silane-based reductant at elevated temperatures generates the active phosphine catalyst.[5] Building on this work, silanes in conjunction with acid or base additives have been successfully employed in PIII/PV=O redox catalysis, enabling the reaction to proceed at low temperature.[6]
[7]
[8]
[9]
[10]
[11] By this advancement, Wittig reactions became amenable to complex molecule synthesis with control over newly formed stereogenic units, such as stereocenters or stereogenic axes of atropisomers.
2
meso-Desymmetrizations
In 2014, the Werner group reported a significant discovery, the first enantioselective catalytic Wittig reaction, which enabled the intramolecular desymmetrization of meso-dicarbonyl compounds 1 (Scheme [1]). Their groundbreaking approach employed Me-DuPhos (3), a chiral phosphine as the catalyst, coupled with a silane reductant and Na2CO3 as a base, in a reaction carried out at 125 °C. Additionally, the Werner group identified alternative reaction conditions where high enantioselectivities could be achieved using butylene oxide as a capped base in combination with PhSiH3 as the reductant at further elevated temperatures of 150 °C.[12] In 2017, the Christmann group expanded upon Werner’s findings by applying this catalytic protocol to the total synthesis of propellane natural products. Their work highlighted the versatility and applicability of the enantioselective catalytic Wittig reaction in complex molecule synthesis, showcasing its potential for broader applications in natural product synthesis.[13]
[14] Although this enantioselective Wittig reaction was successfully developed and further applied in total synthesis, the conditions remained rather harsh.
Scheme 1 First enantioselective Wittig reactions
In 2019, the Kwon group demonstrated that a catalytic stereoselective Staudinger–aza-Wittig reaction could be achieved under mild conditions using a chiral l-hydroxyproline-derived strained HypPhos catalyst 8 (Scheme [2]).[15] To promote the PIII/PV=O catalytic cycle at room temperature with phenyl silane employed as a terminal reductant, nitrobenzoic acids were used serving as co-catalysts. This method produced a range of aza-heterocycles with chiral quaternary centers in high yield and enantioselectivity, showcasing the efficiency of stereoselective Staudinger–aza-Wittig reactions in the synthesis of chiral heterocycles. This pioneering research opened new avenues for catalytic enantioselective PIII/PV=O redox catalysis. Remarkably, stereoselective Staudinger–aza-Wittig reactions were recently achieved by the Tang group even when using a C
2-symmetric chiral bisphosphine (Duanphos) at slightly elevated temperature of 85 °C.[16]
Scheme 2 Enantioselective Staudinger–aza-Wittig reaction
Notably in 2022, the Kwon group further demonstrated the synthesis of oxindoles featuring a quaternary carbon center, starting from azidoaryl malonates via a catalytic stereoselective Staudinger–aza-Wittig reaction (Scheme [3]).[17] A bulkier HypPhos oxide catalyst, in conjunction with [Ir(cod)Cl]2 as an additive, was employed to achieve high enantioselectivities. The use of a sterically more demanding phosphine oxide was critical to preventing self-quenching between the phosphine and the iridium catalyst, ensuring efficient catalytic turnover. The iridium co-catalyst thereby played a dual role in the reaction mechanism. First, it facilitated the silane-mediated reduction of phosphine oxides, likely promoting the PIII/PV=O redox cycling necessary for the catalytic process. Second, it enhanced the electrophilicity of the malonate, thereby accelerating the aza-Wittig reaction and contributing to the overall enantioselectivity and efficiency of the transformation.
Scheme 3 Enantioselective synthesis of oxindoles
3
Atroposelective Catalysis
Based on these advances, our group recently reported the enantioselective synthesis of isoquinoline atropisomers via arene-forming Staudinger–aza-Wittig reactions under mild conditions (Scheme [4]). The catalytic transformation was achieved using Kwon’s commercially available HypPhos catalyst 14, with phenylsilane as the terminal reductant. The co-catalyst 2,4-dinitrobenzoic acid was essential for regenerating the phosphine from its oxide at lower temperatures. Notably, we observed that both E- and Z-azido cinnamates 12 could be selectively transformed into isoquinoline atropisomers 13 in high yields and enantioselectivities. Additionally, we demonstrated the synthesis of ligands such as QUINOL, QUINAP, and QUINOX from the resulting isoquinolines.[18]
Scheme 4 Atroposelective isoquinoline forming Staudinger–aza-Wittig reaction
In these reactions, the Staudinger–aza-Wittig reaction was achieved under mild conditions through PIII/PV=O redox catalysis (Schemes 2–4). This approach is facilitated by the direct formation of iminophosphoranes with nitrogen gas release, enabling compatibility with acid co-catalysis. Conversely, the classic Wittig reaction relies on a base to deprotonate phosphonium intermediates, forming phosphorus ylides as active species. However, the acidic environment required for efficient redox cycling is incompatible with these basic conditions, making it challenging to perform catalytic enantioselective Wittig reactions under mild conditions.
In 2019, the Voituriez group reported a catalytic stereoselective tandem Michael addition/intramolecular Wittig reaction for the synthesis of (trifluoromethyl)cyclobutenes via PIII/PV=O redox catalysis.[19] Kwon’s strained phosphine oxide catalyst (HypPhos oxide 18), in combination with phosphoric acid and phenyl silane, was employed to achieve high enantioselectivities under mild conditions (Scheme [5]). Mechanistically, the addition of the phosphine to the alkyne generates an intermediate, which deprotonates the diketone to form a nucleophilic species. Subsequently, a stereoselective Michael addition generates the phosphonium ylides under base-free conditions. Finally, an intramolecular Wittig olefination yields the cyclobutenes along with the phosphine oxide catalyst.[19]
Scheme 5 Enantioselective synthesis of cyclobutenes via a Michael addition–Wittig olefination reaction
Stereoselective arene formation offers a versatile means to construct structurally precise molecular frameworks. Building on our previous work on stereoselective arene formation by aldol reactions and alkene metathesis to achieve configurationally defined atropisomers,[20]
[21]
[22]
[23] we envisioned an atroposelective arene-forming Wittig reaction as an adaptable and complementary method for forming aromatics with catalyst stereocontrol.
Previous studies have shown that tert-butoxide can be efficiently released from tert-butyloxycarbonyl (Boc) MBH adducts in the presence of phosphines.[24]
[25]
[26]
[27]
[28]
[29] These findings encouraged us to explore Boc-MBH substrates for the situ base generation, compatible alongside the low-temperature PIII/PV=O redox cycling that permits mild stereoselective alkene bond formation in catalytic Wittig reactions.[30]
We hypothesized that a Boc-MBH substrate could undergo allylic substitution with a phosphine catalyst, producing a phosphonium salt. This salt would then release tert-butoxide through decarboxylation, creating an endogenous base to initiate deprotonation and form a phosphorus ylide, which would drive an intramolecular Wittig reaction under mild redox cycling conditions with acid co-catalysis. We further anticipated that the energy from aromatization in this process would provide an additional driving force, enabling an efficient stereoselective arene formation to access a diversity of pertinent atropisomers.
Using Boc-MBH substrate 22, we evaluated several commercially available Kwon’s chiral HypPhos phosphines (Scheme [6]). Gratifyingly, the electron-rich exo-methoxyphenyl phosphine catalyst 24 produced the product 23 with an 88% yield and an enantiomeric ratio (e.r.) of 76:24 in toluene at 50 °C after 24 h. Turning from toluene to acetonitrile resulted in an excellent yield and improved selectivity (88% yield, 90:10 e.r.). In developing the catalytic version of the reaction, we determined that using the Lewis acid [Ir(cod)Cl]₂ with phenyl silane provided optimal reaction conditions. In the catalytic reaction, a competing hydride addition pathway (side product 25) was observed alongside the desired allylic substitution in the Wittig reaction, where tert-butoxide is released as an endogenous base upon the addition of phosphine. The co-catalyst system [Ir(cod)Cl]₂/n-Bu₄NBF₄ successfully suppressed these competing reduction processes, enhancing substrate activation in coordination with the phosphine catalyst. Testing a range of substrates under these optimized conditions yielded satisfactory results with yields of up to 85% and selectivity ratios as high as 94:6.
Scheme 6 Atroposelective arene forming Wittig reaction
4
Conclusion
In conclusion, the advancements in PIII/PV=O redox catalysis have enabled the efficient and enantioselective synthesis of alkenes, aza-heterocycles, and atropisomers under mild conditions. These catalytic reactions are facilitated by in situ generation of iminophosphorane through N2 release or ylide formation via endogenous base release. Employing chiral HypPhos phosphine catalysts, silane reductants, and co-catalysts like [Ir(cod)Cl]₂ or carboxylic acids, these methods enable the selective and more sustainable synthesis of complex molecules. Based on PIII/PV=O redox catalysis, the atroposelective Wittig reaction was developed by using Boc-MBH adducts. The release of an endogenous base renders ylide formation compatible with acid co-catalysis to promote PIII/PV=O redox cycling of chiral phosphine catalysts under mild conditions. Such advancements in catalytic methodology are expected to significantly broaden the scope of complex molecule synthesis, providing new opportunities in fields of stereoselective catalysis and synthesis.
