Synthesis and Cycloaddition Reactions of 1-Azido-1,1,2,2-tetrafluoroethane

A new fluorinated azidoethane—1-azido-1,1,2,2-tetrafluoroethane—was prepared in quantitative yield by the addition of an azide anion to tetrafluoroethylene in a protic medium. The title azide was shown to be thermally stable and insensitive to impact. Copper(I)-catalyzed [3 + 2] cycloaddition with alkynes afforded 4-substituted N-tetrafluoroethyl-1,2,3-triazoles which underwent rhodium(II)-catalyzed transannulation with nitriles to novel N-tetrafluoroethylimidazoles or the reaction with triflic acid to enamido triflates. [3 + 2] Cycloaddition of the title azide with primary amines afforded novel 5-difluoromethyl tetrazoles.


■ INTRODUCTION
−4 Introduction of fluorine atoms or fluorine-containing groups to a molecule of a drug candidate is one of the most promising strategies in the development of modern pharmaceuticals.In the last 8−10 years, 20−50% of approved small-molecule drugs 5−8 and 50−70% of agrochemicals 9,10 contained a fluorine atom or atoms in the molecule of the active ingredient.Thus, the development of new procedures to obtain fluorinecontaining small molecules is a continuing effort of high practical value.
One group of recently introduced fluorinated reagents and building blocks are α-fluorinated azidoalkanes. 11Known oneand two-carbon members of this family are shown in Figure 1.Their unusually high stability (except azidofluoromethane) and synthetic utility have been demonstrated on copper(I)catalyzed azide−alkyne cycloaddition (CuAAC), followed by transformations to novel fluorinated heterocycles, 12 enamides, 13 imidoyl halides, 14 and ketenimines. 15espite the importance of tetrafluoroethyl-and tetrafluoroethylene-containing compounds in synthesis and in various applications, 16 azidotetrafluoroethane ( 1) is yet unreported.It was briefly mentioned in literature without experimental details and characterization 17 and in situ generated by us also without characterization. 18Because of the established reactivity of polyfluorinated alkenes with nucleophiles, including the azide anion, 11,19 our synthetic approach to 1 is based on the reaction of tetrafluoroethylene with a nucleophilic azide source and quench of the resulting fluorinated carbanion with a proton source.

■ RESULTS AND DISCUSSION
Tetrafluoroethylene is a multiton chemical used industrially mainly for the manufacture of poly(tetrafluoroethylene) (PTFE) and copolymers with other alkenes. 20Tetrafluoroethylene is an ideal two-carbon building block for incorporating fluorinated moieties such as tetrafluoroethyl, tetrafluoroethylene, and trifluorovinyl.However, it is a suspected carcinogen, unstable when exposed to radicals, and prone to exothermic polymerization.These properties require caution when handling tetrafluoroethylene.
On an industrial scale, tetrafluoroethylene is formed via the dimerization of difluorocarbene formed from HCF 2 Cl, but on a laboratory scale, the preferred methods are the reduction of 1,2-dihalotetrafluoroethane with zinc, 21 vacuum pyrolysis of PTFE, 22 decarboxylation of sodium pentafluoropropionate 20 or from the Ruppert−Prakash reagent (TMSCF 3 ) and sodium iodide. 23,24We used the first two methods to access tetrafluoroethylene and, subsequently, form the target novel azide (Scheme 1).The results of optimization employing various azide and proton sources under different conditions are summarized in Table 1.
Performing the reaction in an autoclave with tetrafluoroethylene (ca. 10 bar) obtained by depolymerization of PTFE and the use of hydrazoic acid 25 did not lead to any product (entry 1).The same result was observed with sodium azide and tetrabutylammonium hydrogensulfate in aqueous ethanol under elevated temperature (entry 2).However, the use of aqueous THF and extremely long reaction time (14 days) afforded a low NMR yield of 1 when only 1 bar of tetrafluoroethylene formed by the reduction of BrCF 2 CF 2 Br was used (entry 3).Returning to the use of pressurized tetrafluoroethylene in an autoclave and the use of triethylammonium azide or tetrabutylammonium azide afforded good product yields using a suitable proton source (dihydrogen phosphate or ammonium sulfate as pH buffers, entries 5 and 6).Finally, the highest yield of 1 was obtained by using tetrabutylammonium azide and sodium dihydrogen phosphate in wet THF (entry 7).The product 1 was isolated in quantitative yield by codistillation with THF (Scheme 2).A pure sample of 1 (bp 30−32 °C) was obtained by distillation from the high boiling solvent NMP (experiment Table 1, entry 8).
Although other heavily fluorinated small organic azides were proven to be unusually stable and are even commercially available, it was necessary to establish the stability limits of 1 for safe use in synthesis.An indicative thermal stability test was performed by sealing a THF + CDCl 3 solution of 1 in a highpressure NMR tube.After heating the tube to 150 °C for 8 h, no decomposition was observed by 19 F NMR (see the Supporting Information for details).The Koenen test (sensitivity to heat) and fall-hammer test (sensitivity to impact) of a solution of 1 in THF (0.5 M) were both negative (see Supporting Information for details).We therefore conclude that azide 1 is safe to use on a laboratory scale in solution under ambient or moderately harsh conditions.
After developing an efficient and scalable synthesis of azidotetrafluoroethane 1, we evaluated its reactivity with a terminal alkyne using a CuAAC reaction.The application of conditions previously used in triazole formation from other azido(per)fluoroalkanes developed by us, 26 namely, THFsoluble catalyst copper(I) 3-methylsalicylate (CuMeSal), afforded exclusively 1,4-disubstituted-1,2,3-triazoles in good to high yields.The reaction was not limited to aryl (electronrich, neutral, or electron-poor) acetylenes; other competent substrates were alkyl and cycloalkyl acetylenes, containing various functional groups (ester, hydroxyl, protected amine), and even a functionalized steroid derivative (Scheme 3).The obtained triazoles are stable solids and were easily purified by column chromatography on silica gel or by crystallization.The triazole core and N-tetrafluoroethyl substitution of compound 2i survived acidic deprotection, affording triazole 2j with an amino function as a potentially useful building block.
Triazoles 2 were employed in a rhodium(II)-catalyzed transannulation reaction 12 with nitriles to afford N-tetrafluoroethyl-substituted imidazoles.Among nitrogen heterocycles used in medicinal chemistry, imidazoles are privileged scaffolds 27,28 and N-CF 2 CF 2 H-substituted imidazoles are rare. 29,30The method outlined above was applied to newly synthesized triazoles 2, and the corresponding imidazoles 3 were obtained (Scheme 4).Under microwave heating, the transannulation reaction proceeded well with triazoles bearing aryl groups in position 4 (not alkyl groups), including a neutral phenyl group and moderately electron-poor and electron-rich aryl groups.The triazole with strongly electron-acceptor aryl group in position 4 was unreactive (3c).The reaction proceeded well with benzonitrile and its derivatives; however, with acetonitrile, the reaction was less efficient.Triazole 2l having a steroidal structure was found to be unreactive.
To further demonstrate the synthetic potential of the synthesized triazole products, we investigated another denitrogenative transformation, this time mediated by a strong Brønsted acid.Previously, we have shown that N-fluoroalkylated 1,2,3-triazoles in the presence of triflic or fluorosulfonic acids afford β-enamido triflates or fluorosulfonates, respectively, which are stereoselectively functionalized N-alkenyl compounds useful in enamide synthesis. 13Indeed, the reaction of triazole 2a with an equimolar amount of triflic acid provided unstable enamine C via diazonium salt A and vinyl cation B.

Scheme 2. Optimized Preparative Synthesis of Azidotetrafluoroethane 1
The Journal of Organic Chemistry Intermediate C hydrolyzed to the corresponding β-enamido triflate 4 in a good yield (Scheme 5).
Primary amines react with α,α-difluorinated azido alkanes to afford tetrazoles, 31 important polyazaheterocycles 32 displaying various bioactive properties. 33Because tetrazoles bearing the difluoromethyl moiety are unknown, we investigated the reaction of azide 1 with primary amines.Optimization of the reaction conditions revealed that n-butylamine reacted with 1 under mild conditions and full conversion of 5a was reached in 12 h at ambient temperature or in 2 h at 40 °C (Table 2).Amide side product 6a comes from the substitution of the azido group with amine and hydrolysis.Two equivalents of the triethylamine base are necessary for the neutralization of the 2 equiv of HF formed, and the reaction is water-tolerant.A small scope study revealed that alkyl-, cycloalkyl-, and benzyl-type primary amines were competent substrates and the corresponding tetrazoles 5 formed in good yields (Scheme 6).The structure of compound 5b was confirmed by X-ray analysis.Aniline, on the other hand, was an ineffective amine in the preparation of the tetrazoles, most likely owing to its low nucleophilicity.Small amounts of side products 6 were separated from 5 by column chromatography or by basic hydrolysis of 6. Tetrazoles 5 are highly resistant to basic hydrolysis and prolonged heating of the reaction mixture with NaOH (1 M) caused the hydrolysis of amide 6 but left 5 unchanged.
A single report describing the formation of 5-fluoroalkylsubstituted tetrazoles suggested the mechanism proceeding by the substitution of the α-fluorine atom of the azide with nitrogen nucleophiles (Scheme 7, route 1). 31However, this process is highly unlikely because halogen substitution on quaternary carbon atoms is difficult.We suggest that nucleophilic nitrogen of the primary amine attacks the terminal nitrogen of the azido moiety 34 to form intermediate D, which eliminates HF to produce intermediate E, whose cyclization leads to tetrazole 5 (Scheme 7, route 2).Both HF eliminations are facilitated by the presence of a base.Additionally, a control experiment of utilizing the highly nucleophilic azide anion NaN 3 or (Bu 4 N)N 3 revealed no reactivity with 1 in a substitution fashion even under prolonged heating to 80 °C (Scheme 7, route 3) which makes route 1 unlikely.
In conclusion, a method for the preparation of the new 1azido-1,1,2,2-tetrafluoroethane (1) based on the addition of the azide anion to tetrafluoroethylene in a protic environment is reported.The title azide is prepared on a multigram scale in quantitative yield and demonstrated a good thermal stability and a nonexplosive character.Azide 1 undergoes [3 + 2] cycloaddition with terminal alkynes catalyzed by Cu(I) salts to 4-substituted N-tetrafluoroethyl-1,2,3-triazoles.Subsequent rhodium(II)-catalyzed transannulation with nitriles provides the corresponding N-tetrafluoroethyl-containing imidazoles.Acid-mediated denitrogenation of triazole 2a affords βenamido triflate 4. 5-Substituted N-tetrafluoroethyltetrazoles 5 are prepared by the reaction of azide 1 with primary amines.The reaction proceeds by the attack of the nucleophilic nitrogen of primary amines on the terminal nitrogen of the azide moiety, followed by HF elimination and, finally, cyclization.
■ EXPERIMENTAL SECTION General Information.All reactions were carried out in ovendried vessels under a dry N 2 atmosphere.All chemicals were obtained from commercial sources and used as received.(Bu 4 N)N 3 was prepared using a published procedure. 35THF was freshly distilled Scheme 3. Synthesis of 4-Substituted N-Tetrafluoroethyl Triazoles a Prepared from 2i using HCl (6 equiv), Et 2 O, 0 °C to rt, 18 h.

The Journal of Organic Chemistry
over Na/benzophenone prior to use.CDCl 3 and DMF were dried using molecular sieves (3 and 4 Å, respectively). 1 H, 13 C, and 19 F NMR spectra were measured at ambient temperature using 5 mm diameter NMR tubes.The chemical shift values (δ) are reported in parts per million relative to internal Me 4 Si (0 ppm for 1 H and 13 C NMR) or residual solvents and internal CFCl 3 (0 ppm for 19 F NMR).

Scheme 5. Synthesis of β-Enamido Triflate 4
The Journal of Organic Chemistry was allowed to warm to room temperature and stirred overnight (10 bar).Then, the autoclave was cooled with an ice bath, the gaseous products were vented off, and citric acid (5.76 g, 40 mmol, 1 equiv) and Na 2 SO 4 (10 g, 70 mmol) were added.The reaction mixture was filtered and distilled at ambient pressure to afford 1 (oil bath temperature up to 90 °C, heating mantle) together with THF to a cooled (−78 °C) receiving flask containing PhCF 3 as an internal standard.The product was obtained as a THF solution.Caution!Excessive heating of the solution of 1 or neat 1 can cause explosion.Yield: 99% (57 mL of 0.7 M THF solution, 39.86 mmol).An analytically pure sample was obtained by repeating the procedure using NMP instead of THF and redistillation under ambient pressure giving a colorless liquid, bp 30−32 °C, IR (CDCl 3 film): 2991, 2163, 1292, 1273, 1236, 1130, 1065 cm −1 ; 1 H NMR (400 MHz, CDCl 3 ): δ 5.79 (tt, J = 53.0Hz, 2.3 Hz); 13  General Procedure for the Synthesis of Triazoles 2. Terminal alkyne (1 mmol) was added to a THF solution of azide 1 (1.2 mmol, 3 mL) in a screw-cap tube.CuMeSal (11 mg, 0.05 mmol, 5 mol %) was added, and the tube was closed.The reaction mixture was stirred overnight at 40 °C (aluminum heating block and heating mantle).The solvent was evaporated under reduced pressure, and the crude product was purified by column chromatography on silica gel.

Scheme 1 .
Scheme 1. Synthetic Approaches to the Synthesis of 1 via Tetrafluoroethylene

Table 1 .
Optimization of the Synthesis of 1 a e With added water (4 equiv).