The Enantioselective α-Trifluoromethylation of Aldehydes via Photoredox Organocatalysis

Organofluorine compounds possess unique physical properties that are exploited in a wide range of applications such as dyes, polymers, agrochemicals, and pharmaceuticals.¹ In medicinal chemistry, for example, valuable physiological properties are often conferred on “drug-like” molecules via the incorporation of CF₃ groups that enhance binding selectivity, elevate lipophilicity and/or improve metabolic stability.² It is not surprising therefore, that broad research efforts have been focused on the enantioselective construction of trifluoromethyl-containing stereogenicity.^3,4 While significant progress has been made in the arena of nucleophilic 1,2-trifluoromethylation of ketones,⁵ to date, the enantioselective α-alkylfluorination of carbonyl derived enolates (or enolate equivalents) remains an elusive goal.^6,7 Herein we describe a conceptually new approach to the asymmetric α-trifluoromethylation of aldehydes via the successful merger of enamine⁸ and organometallic photoredox^9,10 catalysis.

Design Plan

Recently, our laboratory introduced a new mode of organocatalytic activation, termed photoredox organocatalysis,⁹ whose mechanistic foundation relies on the propensity of electrophilic radicals (derived from the reduction of an alkyl halide by a photoredox catalyst (e.g. 1)) to combine with facially biased enamine intermediates (derived from aldehydes and chiral amine catalyst 2). Inspired by this strategy, we hypothesized that the enantioselective α-trifluoromethylation of aldehydes should also be possible by the marriage of two similar activation pathways. As detailed in Scheme 1, we anticipated that Ir(ppy)₂(dtb-bpy)⁺ 1,^11,12 previously employed as a photosynthesis mimic, should readily accept a photon from a variety of light sources within the visible spectrum (such as a household fluorescent bulb) to populate the *Ir(ppy)₂(dtb-bpy)⁺ 7 excited state. Given its known tendency towards reductive quenching, we presumed that *Ir(ppy)₂(dtb-bpy)⁺ 7 would readily accept a single electron from a sacrificial quantity of enamine (0.5 mol%) to form a strong reductant Ir(ppy)₂(dtb-bpy) 8 (−1.51 V vs. SCE in CH₃CN).¹³ At this stage we anticipated that this electron-rich iridium system 8 would participate in single electron transfer (SET) with trifluoromethyl iodide (−1.22 V vs. SCE in DMF)¹⁴ to render the electrophilic radical 3, while regenerating the photoredox catalyst 1. In concert with this radical formation pathway, we expected that an organocatalytic cycle would initiate by condensation of the imidazolidinone catalyst 2 with an aldehyde substrate to form the enamine 4. The merger of the two activation pathways would then occur in the key alkylation step via rapid addition of the trifluoromethyl radical 3 to the π -rich olefin 4 to form the α-amino radical 5. Given that radical 5 should have a low barrier to oxidation,¹⁵ a second electron transfer event with the *Ir(ppy)₂(dtb-bpy)⁺ 7 excited state would close the photoredox cycle and deliver the iminium ion 6. Rapid hydrolysis of iminium 6 would then reconstitute the organocatalyst 2 while rendering the optically enriched α-CF₃ aldehyde.

Scheme 1.

Proposed Merger of Catalytic Cycles for CF₃-alkylation.

As a critical design element, we anticipated high levels of enantiocontrol resulting from positioning of the 4π-electron system of activated enamine DFT-4 away from the bulky tert-butyl group, while also adopting an (E)-configuration to minimize non-bonding interactions.¹⁶ In this arrangement, the methyl group on the imidazolidinone scaffold effectively shields the Re face, leaving the Si face exposed towards electrophilic radical addition.

Our photoredox fluoroalkylation was first evaluated using octanal and trifluoromethyl iodide along with imidazolidinone catalysts 2 or 10, photoredox catalysts 1 or 9, and a 26 W fluorescent household lamp (Table 1). Initial experiments revealed that the proposed alkylation reaction was indeed possible using a combination of Ru(bpy)₃²⁺ (9) and the amine catalyst 10 (entry 1, 51% yield), albeit to render a racemic product. Importantly, exclusion of light from this protocol resulted in almost complete loss of catalyst efficiency (<5% yield), in accord with the photoredox mechanism outlined in Scheme 1. While the use of Ir(ppy)₂(dtb-bpy)⁺ 1 allowed a significant increase in reaction yield (entry 3, 85%), it was not until sub–ambient temperatures were employed that enantioinduction was observed (entry 4, −20 °C, 52% ee). Moreover, implementation of the trans-tert-butyl-methyl imidazolidinone catalyst 2 (along with photocatalyst 1) provided almost perfect enantiocontrol in the trifluoromethylation step without detectable post-reaction racemization (entry 6, 79% yield, 99% ee).¹⁷ The superior levels of induction and efficiency exhibited by the combination of organocatalyst 2 with photoredox catalyst 1 in DMF at −20 °C prompted us to select these conditions for further exploration.^{18, 19}

Table 1. Enantioselective α-Trifluoromethylation: Initial Studies.

entry	organocat.	photocat.	light	temp (°C)	% yield	% eea
1	10	9	yes	23	51	0
2	10	9	no	23	<5	0
3	10	1	yes	23	85	0
4	10	1	yes	−20	92	52
5	2	9	yes	−20	67	87
6	2	1	yes	−20	79	99

^aEnantiomeric excess by chiral GC analysis of the corresponding alcohol.

We next performed a series of experiments to determine the scope of the aldehydic component in this asymmetric trifluoromethylation protocol. As revealed in Table 2, these mild redox conditions are compatible with a wide range of functional groups including ethers, esters, amines, carbamates, and aromatic rings (entries 2–4, 6, 8, 9, 61–86% yield, 93–98% ee). Moreover, significant variation in the steric demand of the aldehyde substituent can be accommodated without loss in enantiocontrol (entries 5–7, 10, and 11, ≥96% ee). Notably, this protocol enables the formation of a benzylic-CF₃ α-formyl stereocenter without significant erosion in enantiopurity (entry 8, 61% yield, 93% ee).²¹ As highlighted in entries 10 and 11, exposure of enantiopure (R)-3-phenyl-butyraldehyde to catalyst (2S, 5R)-2 results in the diastereoselective production of the anti α,β-disubstituted isomer, while the use of (S)-3-phenyl-butyraldehyde with the same amine catalyst affords the corresponding syn adduct with high fidelity. These transformations clearly demonstrate the synthetic advantages of catalyst-enforced induction versus substrate-directed stereocontrol.

Table 2. Enantioselective α-Trifluoromethylation: Aldehyde Scope.

entry	producta	yield,b eec	entry	producta	yield,b eec
1		79% yield 99% ee	7		73% yieldd 90% ee
2		72% yield 95% ee	8		61% yield 93% ee
3		86% yield 97% ee	9		75% yield 97% ee
4		78% yield 98% ee	10		68% yield >20:1 dr 99% ee
5		X = CH2 70% yield 99% ee	11		62% yield >20:1 dr 99% ee
6		X = NBoc 70% yield 98% ee

^aStereochemistry assigned by chemical correlation or by analogy.

^bIsolated yields of the corresponding alcohol.

^cEnantiomeric excess determined by chiral SFC or HPLC analysis.

^dUsing catalyst 11; ref 20.

We have found that a broad range of perfluoroalkyl iodides and bromides also participate in this enantioselective alkylation reaction (Table 3). For example, n-perfluoroalkyl substrates of varying chain length undergo reductive radical formation and enamine addition without loss in enantiocontrol or reaction efficiency (entries 1–3 and 5–8, 66–88% yield, 96–99% ee). We have also found that aldehyde α-functionalization step can be performed with sterically demanding coupling partners such as perfluoro isopropyl iodides (entry 4, 72% yield, 98% ee). Moreover, benzylic, -ester, and α-ether difluoromethylene carbons are readily incorporated as part of this new enantioselective catalytic α-carbonyl alkylation.

Table 3. Enantioselective α-Perfluoroalkylation: Alkyl Iodide Scope.

entry	producta	yield,b eec	entry	producta	yield,b eec
1		73% yield 96% ee	5		85% yield 98% ee
2		69% yield 99% ee	6		71% yield 99% ee
3		67% yield 96% ee	7		68% yieldd 99% ee
4		72% yield 98% ee	8		89% yieldd 99% ee

^aStereochemistry assigned by chemical correlation or by analogy.

^bIsolated yields of the corresponding alcohol.

^cEnantiomeric excess determined by chiral HPLC analysis of corresponding 2-naphthoyl ester.

^dThe perfluoroalkyl bromide was employed as starting material.

We fully expect that the α-trifluoromethyl aldehyde products generated in this study will be of value for the production of a variety of organofluorine synthons. As shown in Scheme 2, reduction or oxidation of the formyl group is possible to generate β-hydroxy and as α-trifluoromethyl acids (the latter we expect will be a key building block for the formation of heterocycles that incorporate CF₃ at the benzylic position). Moreover, these aldehyde products can be employed in a reductive amination sequence without significant loss in enantioselectivity to produce β-trifluoromethyl amines. Last, and perhaps most important, aldehyde oxidation followed by a Curtius rearrangement allows enantioselective formation of α-trifluoromethyl amine containing stereocenters, a commonly employed amide isostere in medicinal chemistry.² In this case, careful selection of base and reaction temperature is essential to maintain the enantiopurity obtained in the initial alkylation step.

Scheme 2.

Access to Enantioenriched Organofluorine Synthons.

In summary, the first enantioselective, organocatalytic α-trifluoromethylation and α-perfluoroalkylation of aldehydes have been accomplished using a readily available iridium photocatalyst and a commercial imidazolidinone catalyst. Full details of this survey will be forthcoming.