The first publication using (E)-1,1,1,4,4,4-hexafluorobut-2-ene (1a) was published in the middle of the 20th century . Despite this, the real opportunity to study the properties of 1,1,1,4,4,4-hexafluorobut-2-ene appeared only after (Z)-1,1,1,4,4,4-hexafluorobut-2-ene (1b) began to be produced on an industrial scale . These hydrofluoroolefins belong to the newest 4th generation of fluorocarbon refrigerants and are promising compounds and starting materials. Due to this, interest in the use of (E)- and (Z)-butenes 1a,b as synthons in various organic transformations has recently grown significantly. One of the new and budding directions in recent years is the stereoselective olefin metathesis processes based on catalysis by complexes of molybdenum, tungsten and ruthenium . The first publications have recently appeared that molybdenum complexes can catalyze cross-metathesis of butene 1b. Wherein various alkyl and aryl olefins, including those that contain Lewis basic esters, carbamates and amines or α-branched moieties, may be used in efficient and exceptionally Z-selective cross-metathesis reactions . A few years ago, some publications devoted to the cleavage of the C–F bond in butenes 1a,b have been presented in the literature. First, Crimmin et al. investigated the reaction of an aluminum(I) complex with fluoroalkenes. Unlike all the presented fluoroolefins, the reaction of the Al(I) complex with (Z)-butene 1b did not allow isolating the intermediate organoaluminum compound, but led to the elimination of two fluorine atoms with the formation of 1,1,4,4-tetrafluorobuta-1,3-diene . It was also shown that the reactions of the boron reagent (CAACMe)BH2Li(thf)3 with hydrofluoroolefins, including (Z)-1,1,1,4,4,4-hexafluorobut-2-ene, results in defluoroborylation to form the corresponding =CF2 containing products . In addition to complexes of aluminum and boron, several magnesium and lithium silyl reagents were prepared and proved to be good nucleophiles in reactions with (Z)-1,1,1,4,4,4-hexafluorobut-2-ene, as a result of which the corresponding defluorosilylation product was obtained . In a related study of the hydrosilylation reaction of olefins 1a,b, it was shown that, depending on the catalyst used, platinum or rhodium compounds, along with the products of the addition of silane to the double bond, the elimination of the fluorine atom occurs with the formation of the corresponding olefin . Another area of application of olefins 1a,b is based on C=C double bond addition reactions. As early as 1968, Atherton and Fields showed that (Z)- and (E)-butenes 1a,b reacted with diazotrifluoroethane to give 3,4,5-tris(trifluoromethyl)pyrazoline . A few decades later, several new publications in this direction appeared in the literature. In a Chemours’ patent, it has been shown that the oxidation of (E)-butene 1a with sodium hypochlorite in the presence of tetrabutylphosphonium bromide leads to the formation of a bistrifluoromethyl containing oxirane . A related study has demonstrated that (E)-butene 1a reacts with potassium persulfate to form 4,5-bistrifluoromethyl)-1,3,2-dioxathiolane 2,2-dioxide . In 2021 Petrov published an article on the interaction of fluorinated olefins with fluorinated thioketones. In this publication it was demonstrated that 1,1,1,4,4,4-hexafluorobut-2-ene reacts with dithietane, sulfur and KF with the formation of the corresponding 1,3-dithiole . Also, a recent patent presents a method for the preparation of 5,6-bis(trifluoromethyl)-1,2,4-triazine-3-carboxylic acid ethyl ester starting from 1,1,1,4,4,4-hexafluorobut-2-ene and an oxalamide hydrazone .

In the present study, we investigated the reactions of commercially available butenes 1a,b with halogens, as well as subsequent transformations of the resulting compounds.

In 1952 Haszeldine found that the reaction of bromine with (E)-1,1,1,4,4,4-hexafluorobut-2-ene (1a) under UV irradiation leads to the formation of 2,3-dibromo-1,1,1,4,4,4-hexafluorobutane (2) . Subsequent dehydrobromination of compound 2 by treatment with alcoholic potassium hydroxide formed a mixture of isomers 2-bromo-1,1,1,4,4,4-hexafluorobut-2-ene (3a,b) with a yield in two stages of 48%. It should be noted, that only boiling points and elemental analysis data were given for the obtained substances. Ten years later Knunyants and co-workers also synthesized compound 2 and density, refractive index and mass spectra were added to the already available data . However, until now there was no information about the structure and spectral characteristics of the obtained compounds. We have now synthesized these compounds, fully characterized them, and studied some of their transformations.

Scheme 1: Synthesis of 2,3-dibromo-1,1,1,4,4,4-hexafluorobutane (2).

The only difference was that under UV irradiation, the reaction proceeded faster. In both cases, product 2 represented a mixture of stereoisomers in 2:1 ratio. After isolation by distillation, 2,3-dibromo-1,1,1,4,4,4-hexafluorobutane (2) was characterized by 1H, 19F, 13C NMR and mass spectra.

Table 1: The reaction of 2,3-dibromo-1,1,1,4,4,4-hexafluorobutane (2) with bases.

Scheme 2: Synthesis of (E)-butene 3a.

Scheme 3: Isomerization reaction of (E)-butene 3a to (Z)-butene 3b.

Scheme 4: Synthesis of 2-chloro-3-iodo-1,1,1,4,4,4-hexafluorobutane (5).

Pure final product 5 was isolated by distillation at 112 °C in 85% yield. NMR analysis as in the case of bromo derivative 2 showed a mixture of stereoisomers with a 2:1 ratio.

Scheme 5: Dehydrohalogenation reaction of 2-chloro-3-iodo-1,1,1,4,4,4-hexafluorobutane (5).

The chloro- and iodobutenes were separated by distillation in 52% yield for 6a,b and 34% yield for 7a,b. NMR analysis showed the mixtures of (E)- and (Z)-isomers in both cases with the ratio E/Z = ≈4:5 for 6a,b and E/Z = ≈1:2 for 7a,b. The configuration of the isomers was determined by the 5JFFcis coupling constant in the 19F NMR spectrum (ca. 0 Hz for (Z)-isomer and ≈11 Hz for (E)-isomer). The 2-chloro-1,1,1,4,4,4-hexafluorobut-2-enes were first described by Haszeldine in 1952 and only then the (Z)-isomer. We report here the isolation and complete characterization of both (E)- and (Z)-isomers. The spectral characteristics of product 6b obtained by us fully correspond to the literature data . The 1H and 19F NMR spectra of compounds 7a,b also corresponded to the data given in the literature . We present here the spectral data for isomer 6a, as well as the missing data of 13C NMR spectra for iodoolefins 7a,b.

Scheme 6: The reaction of silane 8 with I2/KF.

The mixture was stirred at 30 °C for several days until none of the starting 8 was detected in 19F NMR spectra. The desired iodoolefin 7a together with byproduct triethylfluorosilane (Et3SiF) were removed from sulfolane under vacuum and after double distillation with column 7a was isolated in 67% yield. The 19F NMR spectra of 7a, showed the coupling constant of CF3 and CF3 to be 11.3 Hz, suggesting strongly that product 7a like the original silane 8 has (E)-configuration.

Although some olefins presented above were obtained several decades ago, there was almost no information about their reactivity in the literature. Therefore, we became interested in exploring the synthetic potential of halo-bistrifluoromethyl-containing olefins 3, 6 and 7, which are readily available and can be synthesized in the laboratory in appreciable amounts.

Fluorinated organic compounds are important synthons, which are widely used in agrochemicals, pharmaceuticals and other fields . Fluoroorganic lithium and Grignard reagents have been obtained by the metalation reactions of organofluorine compounds containing bromine and iodine atoms with alkyllithium and Grignard reagents.

Although olefin 3 has been available for many decades, only one paper describes its lithiation with methyllithium and the subsequent reaction of the lithium compound with trifluoroacetophenone . We began our research on the reactivity of the bromobutenes 3a,b with isopropylmagnesium chloride (iPrMgCl) and butyllithium (BuLi), as well as the reactions of the resulting organometallic compounds.

Scheme 7: The reaction of 3a with iPrMgCl and 4-fluorobenzaldehyde (9).

Pure final alcohol 10 was isolated by column chromatography on SiO2 in 46% yield and 1H, 19F and 13C NMR spectra were in full agreement with the published data .

The most interesting outcome from our point of view was the formation of 1,1,4,4,4-pentafluorobuta-1,2-diene (11). 1,1-Difluoroallenes are building blocks for a great number of valuable transformations . Therefore, the synthesis of new fluorinated allenes continues to be relevant. One of the methods for the synthesis of allenes was based on the interaction of bromoolefins with organolithium compounds, followed by the elimination of lithium fluoride . It was logical to assume that in our case a similar reaction of the Grignard reagent 12 with aldehyde 9, elimination of MgBrF results in the formation of allene 11. To confirm our hypothesis, we studied the reaction of haloolefins 3 and 7 with iPrMgCl and BuLi.

Scheme 8: The reaction of olefin 3a with iPrMgCl.

Unfortunately, due to difficulties in separating from olefin 1b and diethyl ether, allene 11 was not isolated in a pure state. Therefore, we turned to study the reaction of adduct 3a with butyllithium.

Scheme 9: The reaction of (E)-butene 3a with BuLi.

The volatile product 11 was removed from the reaction mixture at 20 °C for 2 hours with a slow flow of argon through the system. Unfortunately, we failed to completely separate it from the hexane and allene 11 was produced with 80–85% purity. We managed to partially solve this problem by replacing hexane with the higher-boiling heptane. In this case, the desired product was isolated by double distillation at 7 °C in 50% yield and >90% purity and was fully characterized by 1H, 19F, 13C NMR and IR spectra. The presence of a characteristic band at 2038 cm−1 in the IR spectrum confirmed the allene structure of product 11. As expected, allene 11 is extremely reactive and therefore it decomposes rather quickly during storage both in solvent and in the individual state.

Since we were able to obtain both isomers 3a and 3b in the individual state, we investigated the reaction of (Z)-isomer 3b with BuLi. Thus, 3b also reacted with butyllithium to form allene 11. However, when studying the reaction of each of the isomers under the same conditions, it turned out that the (Z)-isomer reacts faster than the (E)-isomer. Therefore, in the case of the (E)-isomer, to achieve its complete conversion, it was necessary to increase the lithiation reaction time. Based on the results obtained, it was logical to assume that the reaction can be carried out with a mixture of 3a,b. We found that the reaction of 3a,b with 1.2 equivalents of BuLi in heptane at −80 °C for 1 hour, followed by warming to room temperature, led to the complete conversion of the original olefins and the formation of allene 11.

Thus, we have developed a method for the synthesis of the previously unknown allene 11. In addition, the possibility of using a mixture of olefins 3a,b made the allene 11 a more accessible synthon for studying its further transformations.

Scheme 10: The reaction of allene 11 with bromine.

The 19F NMR spectrum showed two signals with the integrated intensity ratio 3:2, a doublet for a CF3 group at δ −57.7 ppm with the coupling constant 3JFH = 6.8 Hz and the characteristic signal of a CF2Br group as a singlet at δ −50.1 ppm. In the 1H NMR spectrum, there was a signal attributed to the CH proton at 6.9 ppm as a quartet with the coupling constant of 3JHF = 6.8 Hz. The (Z)-configuration of product 13 was determined by the absence of the F–F constant in the 19F NMR spectra. It should be noted that the bromination reaction could also be carried out in diethyl ether. In this case, the bromo derivative 13 could be isolated in pure form only after several distillations, which significantly reduces its yield.

Scheme 11: The reaction of allene 11 with ICl.

It should be noted that we failed to isolate the pure product 15 from reaction mixtures because of the small difference in boiling points and volatiles between isomers 14a,b and product 15. Nevertheless, it was tentatively identified in the mixture, based on 1H, 19F and 13C NMR data. The 19F NMR spectrum showed three signals with the integrated intensity ratio 3:1:1, a doublet for a CF3 group at δ −71.5 ppm with a coupling constant of 3JFH = 6 Hz and two highly characteristic signals of geminal fluorine atoms of the CF2 group as doublets of multiplets at δ −66.7 and −70.3 ppm with a coupling constant of 2JFF = 15 Hz. In the 1H NMR spectrum, there was a signal attributed to CH proton at 4.9 ppm as a quartet of multiplets with a coupling constant of 3JHF = 6 Hz.

Thus, during the study of reactions of allene 11 with Br2 and ICl, we obtained new synthons, which due to the presence of several reaction centers, could be of particular interest in various kinds of transformations. Therefore, we decided to synthesize a bistrifluoromethyl containing olefin with bromine and chlorine atoms and explore the possibility of using it to obtain another allene.

Scheme 12: Synthesis of 2,3-dibromo-2-chloro-1,1,1,4,4,4-hexafluorobutane (16).

As in the case of alkanes 2 and 5, product 16 represented a mixture of stereoisomers in 2:1 ratio. After isolation by distillation butane 16 was characterized by 1H, 19F, 13C NMR and mass spectra.

Scheme 13: Synthesis of (Z, E)-2-bromo-3-chloro-1,1,1,4,4,4-hexafluorobut-2-enes (17a,b).

After completion of the reaction the mixture of isomers 17a,b was isolated by distillation at 79 °C and fully characterized. The configuration of the isomers was determined by the 5JFFcis coupling constant in the 19F NMR spectrum (ca. 0 Hz for (Z)-isomer and ca. 13 Hz for (E)-isomer).

Scheme 14: The reaction of olefins 17a,b with BuLi.

Although attempts to obtain allene 18 were unsuccessful, olefins 17a,b may be of great interest as synthons due to the presence of several reaction centers in the molecule.