Lilu Zhang

Asymmetric photoredox transition-metal catalysis activated by visible light

Asymmetric catalysis is seen as one of the most economical strategies to satisfy the growing demand for enantiomerically pure small molecules in the fine chemical and pharmaceutical industries. And visible light has been recognized as an environmentally friendly and sustainable form of energy for triggering chemical transformations and catalytic chemical processes. For these reasons, visible-light-driven catalytic asymmetric chemistry is a subject of enormous current interest. Photoredox catalysis provides the opportunity to generate highly reactive radical ion intermediates with often unusual or unconventional reactivities under surprisingly mild reaction conditions. In such systems, photoactivated sensitizers initiate a single electron transfer from (or to) a closed-shell organic molecule to produce radical cations or radical anions whose reactivities are then exploited for interesting or unusual chemical transformations. However, the high reactivity of photoexcited substrates, intermediate radical ions or radicals, and the low activation barriers for follow-up reactions provide significant hurdles for the development of efficient catalytic photochemical processes that work under stereochemical control and provide chiral molecules in an asymmetric fashion. Here we report a highly efficient asymmetric catalyst that uses visible light for the necessary molecular activation, thereby combining asymmetric catalysis and photocatalysis. We show that a chiral iridium complex can serve as a sensitizer for photoredox catalysis and at the same time provide very effective asymmetric induction for the enantioselective alkylation of 2-acyl imidazoles. This new asymmetric photoredox catalyst, in which the metal centre simultaneously serves as the exclusive source of chirality, the catalytically active Lewis acid centre, and the photoredox centre, offers new opportunities for the ‘green’ synthesis of non-racemic chiral molecules.

Steering Asymmetric Lewis Acid Catalysis Exclusively with Octahedral Metal-Centered Chirality

Catalysts for asymmetric synthesis must be chiral. Metal-based asymmetric catalysts are typically constructed by assembling chiral ligands around a central metal. In this Account, a new class of effective chiral Lewis acid catalysts is introduced in which the octahedral metal center constitutes the exclusive source of chirality. Specifically, the here discussed class of catalysts are composed of configurationally stable, chiral-at-metal Λ-configured (left-handed propeller) or Δ-configured (right-handed propeller) iridium(III) or rhodium(III) complexes containing two bidentate cyclometalating 5-tert-butyl-2-phenylbenzoxazole (dubbed IrO and RhO) or 5-tert-butyl-2-phenylbenzothiazole (dubbed IrS and RhS) ligands in addition to two exchange-labile acetonitriles. They are synthetically accessible in an enantiomerically pure fashion through a convenient auxiliary-mediated synthesis. Such catalysts are of interest due to their intrinsic structural simplicity (only achiral ligands) and the prospect of an especially effective asymmetric induction due to the intimate contact between the chiral metal center and the metal-coordinated substrates or reagents. With respect to chiral Lewis acid catalysis, the bis-cyclometalated iridium and rhodium complexes provide excellent catalytic activities and asymmetric inductions for a variety of reactions including Michael additions, Friedel-Crafts reactions, cycloadditions, α-aminations, α-fluorinations, Mannich reactions, and a cross-dehydrogenative coupling. Mechanistically, substrates such as 2-acyl imidazoles are usually activated by two-point binding. Exceptions exist as for example for an efficient iridium-catalyzed enantioselective transfer hydrogenation of arylketones with ammonium formate, which putatively proceeds through an iridium-hydride intermediate. The bis-cyclometalated iridium complexes catalyze visible-light-induced asymmetric reactions by intertwining asymmetric catalysis and photoredox catalysis in a unique fashion. This has been applied to the visible-light-induced α-alkylation of 2-acyl imidazoles (and in some instances 2-acylpyridines) with acceptor-substituted benzyl, phenacyl, trifluoromethyl, perfluoroalkyl, and trichloromethyl groups, in addition to photoinduced oxidative α-aminoalkylations and a photoinduced stereocontrolled radical-radical coupling, each employing a single iridium complex. In all photoinduced reaction schemes, the iridium complex serves as a chiral Lewis acid catalyst and at the same time as precursor of in situ assembled photoactive species. The nature of these photoactive intermediates then determines their photochemical properties and thereby the course of the asymmetric photoredox reactions. The bis-cyclometalated rhodium complexes are also very useful for asymmetric photoredox catalysis. Less efficient photochemical properties are compensated with a more rapid ligand exchange kinetics, which permits higher turnover frequencies of the catalytic cycle. This has been applied to a visible-light-induced enantioselective radical α-amination of 2-acyl imidazoles. In this reaction, an intermediate rhodium enolate is supposed to function as a photoactivatable smart initiator to initiate and reinitiate an efficient radical chain process. If a more efficient photoactivation is required, a rhodium-based Lewis acid can be complemented with a photoredox cocatalyst, and this has been applied to efficient catalytic asymmetric alkyl radical additions to acceptor-substituted alkenes. We believe that this class of chiral-only-at-metal Lewis acid catalysts will be of significant value in the field of asymmetric synthesis, in particular in combination with visible-light-induced redox chemistry, which has already resulted in novel strategies for asymmetric synthesis of chiral molecules. Hopefully, this work will also pave the way for the development of other asymmetric catalysts featuring exclusively octahedral centrochirality.

Merger of Visible Light Induced Oxidation and Enantioselective Alkylation with a Chiral Iridium Catalyst

A single chiral octahedral iridium(III) complex is used for visible light activated asymmetric photoredox catalysis. In the presence of a conventional household lamp and under an atmosphere of air, the oxidative coupling of 2-acyl-1-phenylimidazoles with N,N-diaryl-N-(trimethylsilyl)methylamines provides aminoalkylated products in 61-93 % yields with high enantiomeric excess (90-98 % ee). Notably, the iridium center simultaneously serves three distinct functions: as the exclusive source of chirality, as the catalytically active Lewis acid, and as a central part of the photoredox sensitizer. This conceptionally simple reaction Scheme may provide new avenues for the green synthesis of non-racemic chiral molecules.

Duplex formation of the simplified nucleic acid GNA.

Glycol nucleic acid (GNA) has an acyclic backbone of propylene glycol nucleosides that are connected by phosphodiester bonds. This paper characterizes the duplex‐formation properties of this simplified nucleic acid. Although single and multiple GNA nucleotides are highly destabilizing if incorporated into DNA duplexes, the two enantiomeric oligomers (S)‐GNA and (R)‐GNA form antiparallel homoduplexes that are thermally and thermodynamically significantly more stable than analogous duplexes of DNA and RNA. The salt‐dependence and Watson–Crick‐pairing fidelity of GNA duplexes are similar to those of DNA duplexes, but, apparently, the 2′‐deoxyribonucleotide and the propylene glycol backbones are not compatible with each other. This conclusion is further supported by cross‐pairing experiments. Accordingly, both (S)‐ and (R)‐GNA strands do not generally pair with DNA. However, (S)‐GNA, but not (R)‐GNA, forms stable heteroduplexes with RNA in sequences that are low in G:C content. Altogether, the high stability and fidelity of GNA duplex formation in combination with the economical accessibility of propylene glycol building blocks for oligonucleotide synthesis render GNA an attractive candidate for the design of self‐assembling materials. They further suggest that GNA could be considered as a potential candidate for a predecessor of RNA during the evolution of life on Earth.

Insight into the High Duplex Stability of the Simplified Nucleic Acid GNA

More than 50 years after Watson and Crick unraveled the secondary structure of the canonical B-form DNA duplex, the function of the pyrimidine–purine nucleobase pairing and the importance of the charged phosphodiester backbone are well established. However, the chemical and evolutionary necessity for the rather complicated deoxyribose and ribose moiety in the backbone of DNA and RNA, respectively, is still uncertain. To address this question, researchers have investigated nucleic acids with alternative sugar residues. In the course of searching for structurally simplified nucleic acid backbones, we discovered recently that a glycol nucleic acid (GNA) with an acyclic propylene glycol phosphodiester backbone can form stable antiparallel duplexes in a Watson–Crick fashion. The constitution of the GNA backbone as well as a newly published GNA-duplex structure are shown in Figure 1. The glycol nucleotide building blocks contain just three carbon atoms and one stereocenter, and are connected by phosphodiester bonds. GNA combines structural simplicity and atom economy with a high duplex stability that significantly exceeds the stabilities of analogous duplexes of DNA and RNA. These features not only make GNA a possible genetic molecule for initial life on Earth but also an interesting scaffold for nucleic acid derived nanotechnology. For almost the last two decades, it had been widely assumed that nucleic acid analogues containing a phosphodiester backbone need to be cyclic to produce the required conformational preorganization of the individual strands for the formation of a stable duplex. 11] The high duplex stability of GNA therefore appears very surprising. In fact, GNA is to date the only known phosphodiester-based nucleic acid with an acyclic backbone that is capable of forming stable duplexes. Herein we present data that resolve this apparent discrepancy between the acyclic nature of the GNA backbone and the high stability of GNA duplexes. To gain insight into the reason for the high duplex stability of GNA, we began by determining the thermodynamic parameters of duplex formation. Accordingly, the values of DG (Gibbs free energy), DH (change in enthalpy), and DS (change in entropy) were obtained from van t Hoff plots by charting the reciprocal of the melting temperature, Tm, against the natural logarithm of varying duplex concentrations for three GNA duplexes and comparison with DNA duplexes with the same sequences (Table 1). As expected, the higher thermal stabilities of GNA duplexes correlate with higher thermodynamic stabilities

Direct Visible-Light-Excited Asymmetric Lewis Acid Catalysis of Intermolecular [2+2] Photocycloadditions

A reaction design is reported in which a substrate-bound chiral Lewis acid complex absorbs visible light and generates an excited state that directly reacts with a cosubstrate in a highly stereocontrolled fashion. Specifically, a chiral rhodium complex catalyzes visible-light-activated intermolecular [2+2] cycloadditions, providing a wide range of cyclobutanes with up to >99% ee and up to >20:1 d.r. Noteworthy is the ability to create vicinal all-carbon-quaternary stereocenters including spiro centers in an intermolecular fashion.

Octahedral Ruthenium Complex with Exclusive Metal-Centered Chirality for Highly Effective Asymmetric Catalysis

A novel ruthenium catalyst is introduced which contains solely achiral ligands and acquires its chirality entirely from octahedral centrochirality. The configurationally stable catalyst is demonstrated to catalyze the alkynylation of trifluoromethyl ketones with very high enantioselectivity (up to >99% ee) at low catalyst loadings (down to 0.2 mol%).

Enantioselective Alkynylation of 2-Trifluoroacetyl Imidazoles Catalyzed by Bis-Cyclometalated Rhodium(III) Complexes Containing Pinene-Derived Ligands

Chiral rhodium(III) complexes containing two cyclometalating 2-phenyl-5,6-(S,S)-pinenopyridine ligands and two additional acetonitriles are introduced as excellent catalysts for the highly enantioselective alkynylation of 2-trifluoroacetyl imidazoles. Whereas the ligand-based chirality permits the straightforward synthesis of the complexes in a diastereomerically and enantiomerically pure fashion, the metal-centered chirality is responsible for the asymmetric induction over the course of the catalysis. For comparison, the analogous iridium congeners provide only low enantioselectivity, and previously reported benzoxazole- and benzothiazole-based catalysts do not show any catalytic activity for this reaction under standard reaction conditions.

Stereogenic‐Only‐at‐Metal Asymmetric Catalysts

Chirality is an essential feature of asymmetric catalysts. This review summarizes asymmetric catalysts that derive their chirality exclusively from stereogenic metal centers. Reported chiral-at-metal catalysts can be divided into two classes, namely, inert metal complexes, in which the metal fulfills a purely structural role, so catalysis is mediated entirely through the ligand sphere, and reactive metal complexes. The latter are particularly appealing because structural simplicity (only achiral ligands) is combined with the prospect of particularly effective asymmetric induction (direct contact of the substrate with the chiral metal center). Challenges and solutions for the design of such reactive stereogenic-only-at-metal asymmetric catalysts are discussed.

Catalytic Asymmetric Dearomatization by Visible‐Light‐Activated [2+2] Photocycloaddition

A novel method for the catalytic asymmetric dearomatization by visible-light-activated [2+2] photocycloaddition with benzofurans and one example of a benzothiophene is reported, thereby providing chiral tricyclic structures with up to four stereocenters including quaternary stereocenters. The benzofurans and the benzothiophene are functionalized at the 2-position with a chelating N-acylpyrazole moiety which permits the coordination of a visible-light-activatable chiral-at-rhodium Lewis acid catalyst. Computational molecular modeling revealed the origin of the unusual regioselectivity and identified the heteroatom in the heterocycle to be key for the regiocontrol.