Figure 1 A brief concept of reaction design
Pyridine is the No. 1 aromatic heterocycle found in pharmaceutical entities. In particular, the introduction of pyridine into the chiral center is able to bring 3D features to the pharmacore. There have been some successful works to build chiral stereogenic centers bearing pyridine using stoichiometric chiral reagents or asymmetric catalysis. On the other hand, asymmetric allylic pyridinylation has not been achieved to date. The absence of the corresponding transformation is partially due to the lack of nucleophilic pyridine reagents. We reported the radical pyridinylation of ketones under thermal and electrochemical conditions. This chemistry, along with other studies using 4-CN-pyridine, produced fruitful structures bearing chiral centers bearing pyridinyl groups, however, as racemates. We presumed that the combination of asymmetric allylic functionalization and 4-CN-pyridine could provide an approach to chiral allyl pyridine compounds. It was envisioned that under electrochemical conditions, the Pd complex could act not only as a transition metal catalyst but also as an electron transfer catalyst. With this design, Pd could shuttle electrons from the cathode to 4-CN-pyridin. The radical 4-CN-pyrdine forms a radical with known chemistry. At this stage, Pd could obtain the second electron from the cathode and undergo a rebound with the radical of 4-CN-pyrdine. Reductive elimination could give rise to the product. If a suitable chiral ligand is present, the enantioselectivity can be regulated.
Figure 2 Proposed reaction pathway of Pd-catalyzed asymmetric allylic pyridinylation reaction under electrochemical conditions.
The Pd-catalyzed asymmetric allylic pyrinylation reaction was explored under electrochemical conditions. Ph.D Student Mr. Weijie Ding started to test the feasibility of the reaction design. After a series of experiments, Ding found that the C2-symmetric diphosphine chiral ligand could fulfil the task and provided branched allyl pyridine in 60% ee. Ding further optimized the parameters of the reaction, including solvent, temperature, supporting electrolyte, electrode and chiral ligands. Finally, 94% ee was achieved by using 10 mol% PdCl2/12 mol% DTBM-SEGPHOS in MeOH/MeCN with Zn(+)/graphite felt (-) electrodes at 35 °C. Meanwhile, the branched product was confirmed as the major outcome.
Next, Ding extended the study with more substrates with various structural features at both aryl and alkyl units. A number of products bearing substituents on aromatic rings are compatible with these enantioselective conditions. Furthermore, it was found that the product with internal alkenes could be achieved as well, and the enantioselectivity was at the same level. In addition, Ding conducted the reaction at the gram scale and obtained the desired product in comparable yield, branch/linear selectivity, and enantioselectivity. The application of this protocol was further demonstrated with two syntheses of pharmaceutical compounds as their chiral versions.
In this study, understanding the role of the Pd complex, electrode, and chiral ligand is important to disclose the introduction of 4-CN-pyrdine onto the allyl group. First, Ding independently prepared a chiral R-SEGPHOS-Pd-allyl complex and confirmed it as an effective catalyst. In addition, the high-resolution mass analysis of the reaction mixture at the initial stage revealed that the diphosphine-Pd-allyl complex formed predominantly and quickly. Thus, Ding conducted the electrochemical analysis of this complex in the presence of 4-CN-pyridine. A clear catalytic current was observed, which was confirmed by a worked reaction conducted at a controlled cathodic potential lower than that required to reduce 4-CN-pyridine. These results suggested the multiple roles of the Pd complex. Therefore, a DFT computation study was carried out to find the following pathway where the chiral ligand regulates the selectivity. It was found that in the transition state to the desired branched R-product, aromatic stacking is present, which benefits the corresponding transition state in comparison to TSs to the linear product and the opposite configuration of the branched product.
Ding then performed NMR titration of 4-CN-pyridine with ZnCl2 and found that the Zn cation had a strong coordination with the nitrogen atom in the pyridine ring but a negligible interaction with nitrogen in the CN group. A controlled reaction showed that the anode with other reductive metals, such as Al, Mg, and Fe, did not lead to detectable products. If 20 mol% ZnCl2 was added, the desired conversion took place to give the product in appreciable yield. These results showed that Zn not only act as a sacrificial anode, but also releases Zn2+ anions to activate 4-CN-pyridine.
Inspiration during the review process.
Three Reviewers gave us a series of important suggestions regarding the substrate scope, reaction conditions, and mechanism. For example, the elucidation of the charge distribution of the Pd complex prompted us to carry out a DFT computation study that gave fruitful information not only on the [L-Pd]+ complex but also on the regioselectivity and enantioselectivity. Another comment suggested using a charge-tagged chiral ligand to investigate the reaction. In this case, a new BINAP ligand was tagged with imidazolium salt. The high-resolution mass analysis of the crude mixture reaction using this ligand showed that the CN anion could form a stable complex with Pd during the conversion, which could be a reason for the diminished enantioselectivity observed. This study also demonstrated the importance of Zn cations that could quench the CN anion. The comment from another reviewer prompted us to study why the alkene in the product is detected as the E configuration only. By using substrates bearing E and Z alkene separately, we obtained the same outcome. In turn, the 31P NMR study revealed that the fast equilibrium of the allyl complex eliminated the difference in the substrate. DFT studies showed that the syn-allyl-Pd complex corresponding to the E product is more favorable. The third reviewer gave an important comment on the Zn anode. This comment leads to the conclusion that Zn not only acts as the sacrificial anode to protect the phosphine ligand but also acts as the donor for the Zn cation to activate 4-CN-pyridine and quench the CN anion.
The link to the publication:
Journal: Nature Communications.
DOI : 10.1038/s41467-022-28099-w
Title : Palladium-catalyzed asymmetric allylic 4-pyridinylation via electroreductive substitution reaction