Iridium-Catalyzed Synthesis of Primary Allylic Amines from Allylic Alcohols: Sulfamic Acid as Ammonia Equivalent

Link: http://www3.interscience.wiley.com/cgi-bin/abstract/114179538/ABSTRACT

From Prof. Erick M. Carreira's group at ETH Zürich, Zürich, Switzerland

A report in ACIEE EarlyVIew presents a new and intriguing method of converting allylic alcohol directly to allylic amine. The features of this method includes a direct conversion of the allylic alcohol to the allylic amine without prior manipulation of the alcohol, the use of sulfamic acid as the NH2 group equivalent leading to the product as primary ammonium salt and the use of iridium phosphoramidite catalyst.

Ligands L1 to L4 were screened with substrates 1-3 as test models. Low conversions were observed with substrates 1 and 2. However, it was discovered by accident that substrate 3 as the free alcohol could be converted to the corresponding allylic amine.

The iridium catalyst used in the process was [{Ir(cod)Cl}2] (1.5 mole%). When ligand L3 was used (3 mol%) with the Ir-catalyst in DMF as the solvent of choice, substrate 3 could be converted to the corresponding allylic amine in the presence of sulfamic acid in >99% conversion after 24 h at 23 C. The double bond in ligand L3 was crucial as the comparison studie using ligand L4 (missing double bond) only led to 20% conversion. Ligand L3 could be accessed easily from reaction of 2,2'-BINOL, PCl3 and 5H-dibenzo-[b,f]azepine in one step. This was an early example of sulfamic acid being utilized as amino-group equivalent in organic synthesis.

The scope of the reaction was studied and the results are summarized below.

From the table, besides obtaining the products as the free amine salts, it could also be obtained as N-Bz, N-Boc or N-amide protected amino group (entries 2-4).

In proton NMR experiments using DMF-d7 as the solvent, the allylic alcohol 3 was quickly converted to the corresponding allylic amine 4 very quickly (after two hours). However, when the excess sulfamic acid was used and when the reaction was allowed to react longer, the expected product 5 was formed.

Based on the spectroscopic observations, the following conclusions were made:

- When an excess of sulfamic acid was employed in the Ir-catalyzed process, the amine 4 initially produced underwent partial sulfamation with the second equivalent of sulfamic acid to form 6. This sulfamation only began after 3 had been entirely transformed into 4
- Throughout the course of the Ir-catalyzed process, no signals corresponding to the sulfate ester 5 were observed. Given the long reaction time required to perform the sulfation with sulfamic acid (eight hours), it is unlikely that the sulfate ester 5 serves as activated intermediate in the Ir-catalyzed reaction
- The absence of signals corresponding to 6 leads the authors to suspect that sulfamic acid does not act as a nucleophile in the catalytic process

The mechanism of the process was proposed as followed. It is believed that the Vilsmeier-type intermediate 7 was formed from DMF and sulfamic acid. This intermediate then reacted with the allylic alcohol to give the intermediate 8, which underwent oxidative addition with Ir-catalyst. This pi-allyl iridium complex 9 then reacted with ammonia in the reaction to form the desired product.

Besides the achiral version of the reaction, the process was also shown to undergo asymmetric catalysis process to produce chiral allylic amine. That is when ligand L5 was used with the iridium catalyst system, the cyclohexyl allylic alcohol was converted to the corresponding allylic ammonium chloride salt in 70% yield and 70% ee, demonstrating the utility of this method in asymmetric catalysis. This was also the first asymmetric example of the process.