Organocuprate Cross–Coupling Reactions with Alkyl Fluorides

Abstract

Cross-coupling of alkyl fluorides and organocuprates is accomplished via aluminum halide mediated C–F bond activation and subsequent C_sp2–C_sp3 and C_sp3–C_sp3 bond formation. Relatively mild conditions allow for smooth activation of notoriously challenging primary and secondary alkyl fluorides while competing alkyl chain rearrangement, HF elimination and homocoupling reactions are effectively controlled. The utility and functional group tolerance is demonstrated with 23 examples and a variety of coupling products obtained in up to 88% yield.

Graphical Abstract

The development of methods that achieve carbon-carbon bond formation with fluorinated organic compounds has remained one of the most challenging tasks in modern synthetic chemistry. This is particularly true for alkyl fluorides which are widely considered chemically inert and a synthetic roadblock. Although a broad selection of aliphatic organofluorines is nowadays commercially available or easily prepared, the synthetic utility is largely underdeveloped compared to the plethora of methods that exploit the more reactive alkyl chlorides, bromides and iodides as valuable starting materials in an impressive range of routinely performed transformations. The lack of synthetic applications of the C_sp3–F bond inarguably originates from the inherent chemical inertness and the high bond dissociation energy exceeding 100 kcal/mol unless it is activated, for example in allylic or benzylic positions. Several groups have explored C–F bond activation strategies that are compatible with conditions typically used in cross-coupling reactions. To this end, the pioneering efforts by Kambe and others have successfully demonstrated the scope and general utility of alkyl fluorides in Kumada-Corriu couplings with Grignard reagents.^1–14 Our laboratory has introduced chemodivergent C–F bond activation pathways that enable selective alkyl fluoride functionalization and carbon-carbon bond construction under mild conditions.^15–18 We recently reported that palladium and nickel catalyzed Suzuki cross-coupling with alkyl fluorides is possible when lithium iodide is used to activate the notoriously inert C_sp3–F bond.¹⁹ As the potential of carbon-carbon bond formation with alkyl fluorides is becoming increasingly apparent, the introduction of cross-coupling chemistry with other organometallic reagents seems most likely to draw attention among synthetic chemists to the emerging chemical space that becomes accessible with alkyl fluorides.

More than eight decades ago, the first report on conjugate additions of organocopper reagents to α,β-unsaturated ketones emerged when Grignard reagents were used in the presence of a Cu(I) salt.²⁰ Recognizing the synthetic value of cuprates, Gilman and co-workers developed the first R₂CuLi species in 1952.²¹ Since then the so-called Gilman reagents have become a popular choice,²² in particular for conjugate additions to enones²³ and substitution reactions with alkyl and aryl halides.²⁴ In 1981, higher-order organocuprates, R₂CuCNLi₂, that give superior results with secondary alkyl halides were introduced by Lipshutz’ group.^25,26 The substitution reactions are widely believed to involve a Cu(I)/Cu(III) redox pathway^27,28 and this reaction has been particularly successful with primary and secondary electrophiles bearing bromide and iodide leaving groups.²⁹ We now wish to introduce a method that allows C_sp2–C_sp3 and C_sp3–C_sp3 bond construction with alkyl fluorides and a variety of higher-order alkyl, vinyl and aryl organocuprates (Scheme 1). This is accomplished with activated and unactivated substrates under generally mild conditions that effectively minimize competing migration, elimination and homocoupling side reactions. We use aluminum triiodide for the initial C–F bond activation step which results in halide exchange while successfully outcompeting intrinsically fast rearrangement processes, at least in part through a concerted mechanism, and establishes a thermodynamic driving force for the alkyl fluoride cleavage by concomitant generation of AlF₃.

Scheme 1.

Scope and challenges of cross-coupling reactions with alkyl fluorides and organocuprates.

At the onset of this study we investigated the possibility of alkyl fluoride activation under conditions that are compatible with organocuprate cross-coupling. The formation and use of organocopper reagents is typically achieved in ether solvents which inhibit the LiI activation pathway previously incorporated by us in Suzuki cross-couplings with alkyl fluorides in toluene.¹⁹ Using the unactivated primary and secondary alkyl fluorides 1 and 2 as our model compounds, we began screening several main group and transition metal salts as well as Gilman and higher-order cuprates (Table 1 and SI). As expected, unsatisfactory conversion of 1 toward 3 was observed using one equivalent of either LiI, NaI or KI together with Ph₂Cu(CN)Li₂ at room temperature but these early results suggested the feasibility of alkyl fluoride organocuprate coupling (entries 1–3). We then found that aluminum bromide and iodide are much more efficient and the addition of 50 mol% of these activators enables almost quantitative conversion in a variety of ethereal solvents (entries 4–9). The same outcome was obtained with the Gilman agent Ph₂CuLi while the conversion dropped from 95% to only 40% when PhCu(CN)Li was used (entries 10 and 11). The analysis of the reaction with the secondary alkyl fluoride 2 showed that the combination of AlI₃ and the higher-order organocuprate Ph₂Cu(CN)Li₂ gives the best results (entries 12–15 and SI).

Table 1.

Optimization of the alkyl fluoride cross-coupling with organocuprates.

Entry	Substrate	Additive	Cuprate	Solvent	Conversion (%)a
1	1	LiI	Ph2Cu(CN)Li2	DBE	19b
2	1	NaI	Ph2Cu(CN)Li2	DBE	19b
3	1	KI	Ph2Cu(CN)Li2	DBE	20b
4	1	AlBr3	Ph2Cu(CN)Li2	Ether	60
5	1	AlI3	Ph2Cu(CN)Li2	Ether	46
6	1	AlI3	Ph2Cu(CN)Li2	THF	15
7	1	AlI3	Ph2Cu(CN)Li2	MTBE	95
8	1	AlI3	Ph2Cu(CN)Li2	Dioxane	95
9	1	AlI 3	Ph 2 Cu(CN)Li 2	DBE	95
10	1	AlI3	PhCu(CN)Li	DBE	40
11	1	AlI3	Ph2CuLi	DBE	95
12	2	AlI3	Ph2CuLi	DBE	60
13	2	AlBr3	Ph2CuLi	DBE	20
14	2	AlI3	PhCu(CN)Li	DBE	0
15	2	AlI 3	Ph 2 Cu(CN)Li 2	DBE	90

Reaction conditions: A mixture of the substrate and the additive (50 mol%) was stirred for two hours at 25 °C in the specified solvent. The cuprate THF solution was then added and stirring continued at room temperature for 16 hours.

^aConversions were determined by GC-MS.

^bStoichiometric amounts of the metal iodide were used. DBE = dibutyl ether, MTBE = methyl tert-butyl ether.

Since the optimization study revealed high conversion of both primary and secondary alkyl fluorides with higher-order organocuprates and AlI₃ as C–F bond activator in dibutyl ether at room temperature, we decided to use this protocol and investigate the general scope of this reaction (Scheme 2). We found that the phenylation of (4-fluorobutyl)benzene affords 3 in 78% yield and 4 was isolated in 71% yield. We obtained 5 in 62% yield which shows that this method can also be applied to α-branched primary alkyl fluorides. Other long-chain primary alkyl fluorides gave similar results. Compounds 6 and 7 were produced with no signs of rearrangement and with less than 5% elimination by-products in 68% and 75% yields, respectively. A variety of primary alkyl fluorides containing halogen, amine and ester functionalities undergo C–C bond formation toward 8-11 with yields ranging from 64–82%. It is noteworthy that the C_sp2–F bond proved inert under the reaction conditions. Benzylic fluorides can also be used as well as nitrogen and sulfur containing heterocycles. The corresponding products 12-15 were isolated in 59–72% yield. We were particularly pleased with the successful formation of 10 and 14 as the alkyl fluorides used in these reactions have a high propensity for HF elimination to conjugated styrenes which is a common problem in cross-coupling reactions with similar alkyl halide scaffolds and often outperforms the desired C–C bond formation pathway. However, the C_sp3–C_sp3 bond construction was achieved with 60–64% yield using our standard protocol. The reaction yields are somewhat compromised by the formation of elimination by-products but neither hydride nor aryl migration was observed. The use of secondary alkyl fluorides required ambient temperatures to allow for facile C–C coupling while limiting the amount of competing HF elimination and rearrangements, and we obtained 4 and 16 in 71% and 85% yield, respectively. The coupling of secondary alkyl fluorides with secondary alkyl cuprates was unsuccessful and outcompeted by elimination.

Scheme 2. Reaction Scope of Organocuprate Cross-Coupling with Primary and Secondary Fluorides.^a.

^aR-F (1 equiv.) and AlI₃ (0.5 equiv.) stirred in 600 μL DBE at room temperature. Ph₂CuCNLi₂ (2 equiv.) added and stirred for 18 hours at room temperature. ^b0.75 equiv. AlI₃ used. ^cAlI₃ and R-F heated at 50 °C for two hours. ^dCuprate addition at 35 °C, 5 days. All reported yields are for isolated products. See SI for details.

To further extend the scope of our alkyl fluoride cross-coupling method, we continued with varying the organocuprate reagent. We were able to prepare organolithium precursors through deprotonation or halogen-lithium exchange, producing a wide range of organocuprates that proved viable for C–C coupling with our protocol. Aryl organocuprates containing heteroatoms or a naphthyl ring showed good conversion with no evidence of competing rearrangement and small amounts of HF elimination products. The corresponding C_sp2–C_sp3 cross-coupling products 17-19 we obtained in 65–70% yield (Scheme 3). Moreover, the formation of 20 demonstrates that the addition of a vinyl unit is possible. The utilization of n-butyllithium and sec-butyllithium as cuprate precursors enabled us to introduce primary and secondary alkyl fluorides to C_sp3–C_sp3 coupling which gave 21-23 in good yields. Interestingly, deprotonation of acetonitrile and subsequent organocuprate formation allows cross-coupling of primary and secondary alkyl fluorides with an acetonitrile moiety toward 24 and 25.

Scheme 3. Diversification of Organocuprates in the Cross-Coupling with Alkyl Fluorides.^a.

^aR-F (1 equiv.) and AlI₃ (0.5 equiv.) stirred in 600 μL DBE at 25 °C. Ph₂CuCNLi₂ (2 equiv.) added and stirred for 18 hours at room temperature. ^bCuprate addition at 50 °C. ^c0.75 equiv. of AlI₃. ^dCuprate addition performed at 35 °C, 5 days. All reported yields are for isolated products. See SI for details.

We finally decided to investigate the aliphatic fluoride activation with aluminum triiodide. The high efficiency of this step with substoichiometric amounts of aluminum triiodide showing little to no indication of rearrangement or elimination products suggests the possibility of a concerted mechanism bypassing carbocation formation or a stepwise pathway with a short-lived ion pair that can successfully outcompete side reactions. To verify this, aluminum triiodide was added to a solution of enantioenriched (S)-(3-fluorobutyl)benzene 26 (er = 24:1) synthesized by following a literature procedure.³⁰ This reaction gave (R)-(3-iodobutyl)benzene 27 (er = 3:1) in 90% yield (Scheme 4), demonstrating inversion of configuration and the likelihood of a concerted S_N2-type mechanism. As the enantiomeric ratio significantly decreased, a stepwise mechanism via a carbocation intermediate cannot be ruled out and it is likely that both pathways occur concurrently. When enantioenriched 26 was subjected to our general C–C coupling method almost racemic 1,3-diphenylbutane 4 (er = 1.2:1) was produced. This outcome is in agreement with previous reports. In fact, the conversion of (R)-28 to 29 is known to coincide with complete racemization.²⁶

Scheme 4. Organocuprate C–C Coupling with Enantioenriched Fluoride 26.

See SI for details.

In conclusion, we have developed a C-F bond functionalization method with unactivated primary and secondary aliphatic fluorides using commercially available aluminum triiodide to enable C-C coupling with organocuprate reagents. A likely combination of stepwise and concerted activation paths is believed to set the stage for the reaction with organocuprates while rearrangement and elimination by-product formation can be successfully suppressed. This one-pot protocol tolerates several functional groups and achieves C_sp2–C_sp3 and C_sp3–C_sp3 cross-coupling with alkyl, vinyl and aryl organocuprates in 52–88% yield. The application scope of C–C bond formation with cuprates, previously demonstrated with more reactive alkyl halides, can now be extended to C–F bond functionalization with generally considered chemically inert alkyl fluorides.

Data Availability Statement.

The data underlying this study are available in the published article and its online supplementary material.