Jia-Lan
Zhang
a,
Jin-Xiu
Xiong
a,
Lu-Qi
Fei
a,
Fei-Hu
Cui
a,
Shu-Hui
Li
*a,
Hai-Tao
Tang
a,
Yan-Zhao
Xie
*b and
Ying-Ming
Pan
*a
aState Key Laboratory for Chemistry and Molecular Engineering of Medicinal Resources, Key Laboratory for Chemistry and Molecular Engineering of Medicinal Resources (Ministry of Education of China), Collaborative Innovation Center for Guangxi Ethnic Medicine, School of Chemistry and Pharmaceutical Sciences, Guangxi Normal University, China. E-mail: panym@mailbox.gxnu.edu.cn; gxnulsh@gxnu.edu.cn
bFood Nutrition and Materials Research Center, Fujian Polytechnic Normal University, China. E-mail: xieyanzhaofz@126.com
First published on 2nd May 2024
Fluoroalkylated allenes are potential compounds for drug and materials development. Herein, we report electrochemically mediated trifluoromethylallene synthesis through the rearrangement of trifluoromethyl-containing 1,3-enynes under the action of silicon or boron radicals. A series of late functionalization reactions of natural product derivatives were explored to generate corresponding complex allylsilane compounds, and the transformation of homoallenylborates synthesized useful building blocks.
![]() | ||
| Scheme 1 Electrochemically mediated synthesis of trifluoromethyl allenes by rearrangement of trifluoromethyl-containing 1,3-enynes. | ||
The trifluoromethyl (CF3) group functions as an excellent bioisostere of the methyl group in medicinal chemistry and can improve the metabolic stability, bioavailability, and cellular membrane permeability of molecules.7 Thus, introducing trifluoromethyl groups into molecular skeletons is of great importance. Among many trifluoromethyl substituents, trifluoromethyl-containing allene skeletons are preferred because of their good plasticity. Current methods for synthesizing trifluoromethylated allenes are generally divided into two categories: introducing exogenous trifluoromethyl groups8 and rearranging alkynes or enynes containing trifluoromethyl groups.9
In general, the construction of trifluoromethyl-substituted allene skeletons through an exogenous introduction strategy requires not only transition metal catalysis but also a trifluoromethyl source, which is extremely expensive. A trifluoromethyl source can efficiently facilitate the construction of trifluoromethyl-substituted multifunctional allene skeletons through intramolecular rearrangement. Nevertheless, the rearrangement of enynes containing trifluoromethyl groups is limited to transition metal catalyzed reactions thus far (Scheme 1c).
Organic electrochemical synthesis is an environmentally friendly, mild, and sustainable method for efficient conversion and chemical synthesis.10 In recent years, Rueping's group has successfully synthesized allenes from 1,3-enynes through organic electrosynthesis involving free radicals.11 Thus, trifluoromethyl enynes may be rearranged into trifluoromethyl allenes through the radical addition reaction of silicon or boron radicals. This reaction has been widely reported in unsaturated systems under electrochemical conditions.12 Herein, a method for synthesizing trifluoromethylated allenes by intramolecular rearrangement through silicon or boron radicals attacking 1,3-enynes is successfully developed. The method does not require metal catalysts or exogenous oxidants and exhibits good chemical and regioselectivity. The introduction of boron promotes the derivatization of trifluoroallenes (Scheme 1d).
:
1), Et4NBF4 was used as the electrolyte (0.05 M), B2oct2 was used as the boron reagent, and stainless-steel electrodes had a constant current of 10 mA and charge of 2 F mol−1. We screened a variety of solvents (Table 1, entries 2–5). The optimum reaction result occurred when methanol and hexane were combined. The yield was affected by the methanol content because methanol concentration directly influenced the methoxide anion concentration and intermediate oxidation. No significant change in reaction yield was observed after the amount of 2a was increased (Table S1,† entries 14 and 15). Different electrolytes considerably influenced the reaction. When LiClO4 was used as the electrolyte, the target product was not obtained. When tetrabutylammonium bromide was used as the electrolyte, the yield was low (see the details in the ESI†). The use of other electrodes at the cathode and anode was detrimental to the reaction (Table 1, entries 6 and 7). For the electrode materials, the oxidation of the substrate had a high degree of specificity. There have been previous reports of pinacol borane oxidizing on stainless-steel electrodes. When it was added to the reaction, sodium methoxide resulted in a slightly reduced yield (Table S1,† entries 16–20). When the amount of electricity was constant, either a decrease or an increase in current slightly reduced the yields (Table 1, entries 11 and 12). When the reaction was carried out without current, no product was obtained (Table 1, entry 15).
| Entry | Variation from standard conditions | Yieldb/% |
|---|---|---|
a Standard conditions: 1 (0.2 mmol), 2a (0.4 mmol), Et4NBF4 [0.05 M] in hexane/CH3OH (1 : 2, 6 mL), rt, stainless-steel electrodes (cathode and anode), 10 mA, 2 F mol−1, 5 mA cm−2, undivided cell, under Ar.
b Yield of isolated product.
|
||
| 1 | None | 81 |
| 2 | CH3CN instead of hexane : MeOH (1 : 2) |
0 |
| 3 | CH3OH instead of hexane : MeOH (1 : 2) |
66 |
| 4 | 1,4-Dioxane/CH3OH(1: 2) instead of hexane : MeOH(1 : 2) |
50 |
| 5 | CH3CN/CH3OH (9 : 1) instead of hexane : MeOH (1 : 2) |
30 |
| 6 | SS (+)|Pt (–) instead of SS (+)|SS (–) | 40 |
| 7 | SS (+)|C (–) instead of SS (+)|SS (–) | 0 |
| 8 | 1 eq. CH3ONa as base | 20 |
| 9 | n Bu4NBr instead of Et4NBF4 | 20 |
| 10 | n Bu4NPF6 instead of Et4NBF4 | 54 |
| 11 | 5 mA instead of 10 mA | 50 |
| 12 | 15 mA instead of 10 mA | 56 |
| 13 | 6 h instead of 4 h | 60 |
| 14 | 8 h instead of 4 h | 50 |
| 15 | No current | 0 |
Under the optimized conditions, we examined the scope and the limitations of the electrochemical 1,4-protoborylation of CF3-substituted conjugated enynes reaction (Scheme 2). The substrate range of electron-donor substituents was expanded. All the 2-trifluoromethyl-1,3-enynes generated the desired products in good to excellent yields (Scheme 2, 3a–3l). In addition to simple 2-trifluoromethyl-1,3-enyne, enynes with various substitutions, such as methyl, methoxyl, and tertbutyl, at the para site formed the corresponding allenes (3a–3d) in high yields (81%–69%). Halogen-containing enynes (3e–3h) were obtained in good yields (78%–69%). When a chlorine substituent was in the para and meta positions, the target products (3f and 3g) were obtained in good yields. Notably, when the chlorine atom was in the ortho position, we did not obtain satisfactory results. Therefore, the steric hindrance influenced our reaction. Subsequently, different boranes were tested. B2hex2 and B2pin2 were converted into the target products (3i and 3j) in moderate yields (57%–50%) under the optimized conditions. Finally, we explored different reaction sites. When the substrate was 1k, two molecules were generated by the boron-addition rearrangement of the allene product (3k). When the reaction substrate was 1l, 3l was generated by the addition rearrangement of the alkyl fragment and boron. This result indicated that boron radicals are reactive for alkyl fragments. To further explore the substrates of alkynes containing alkyl substituents, 1m and 1n have been selected as the substrates. The experimental results indicate that when 1m was used, the selectivity to the products was 3m
:
4m = 37
:
63. When the substrate was 1n, a complex mixture was obtained. Furthermore, the target product cannot be obtained when the substrate contains heterocyclic and electron-withdrawing substituents (1o and 1p). Therefore, the experimental results indicate that boron radicals preferentially react with higher electron density alkynes, which may be related to the reaction properties of boron radicals themselves.
Considering the variability and wide practicality of silicon reagents, we tested the silicon boron reagent 2b instead of 2a. Next, we studied the 1,4-protosilylation of trifluoromethyl substituted enynes. The scope of CF3-substituted conjugated enynes was examined (Scheme 3). Both electron-withdrawing and electron-donating enynes were tolerated under the standard reaction conditions. All the 2-trifluoromethyl-1,3-enynes afforded the desired products in moderate to excellent yields (Scheme 3, 5a–5o). In addition to simple 2-trifluoromethyl-1,3-enyne, enynes with various substitutions such as ester and nitro at the para- position produced allenes (5b and 5c) in good yields (68–64%). When halogen containing enynes were used (5d–5h), good yields (96–56%) were obtained. When a chlorine substituent was in the para-, meta- and ortho- positions (5e–5g), the target products were obtained in good yields (96–56%). When a chlorine substituent was in the ortho position, a single product (5g) was obtained, although the yield was not extremely high. This result showed that steric hindrance also had an important influence on the selectivity of the reaction. Then, for enynes containing thiophene or biphenyl, the target products (5i and 5j) could also be obtained in good yield (77–66%). Subsequently, we used a different silicon substrate and obtained a moderate yield (5k) with Ph2MeSi-Bpin. Finally, we used different natural product derivatives, such as menthol, borneol, and adamantyl alcohol derivatives, which produced the corresponding products (5m–5o) in good yields (61%–55%).
We conducted a series of control experiments to study the mechanism (Scheme 4). When 3.0 equivalents of (2,2,6,6-tetramethylpiperidin-1-yl)oxyl were employed in the reaction, the process was completely suppressed, and 3a was not detected. Adding 3.0 equivalents BHT into the reaction resulted in the production of trace amounts of 3a (Scheme 4-1). The HRMS analysis showed that there may be free radical trapping species 7a and 7b generated. These results indicated that the reaction may involve a radical pathway. To understand the proton source of hydrogenation in the reaction, we conducted labeling experiments on methanol. The results showed that the proton source of the reaction was methanol (Scheme 4-2). All the results suggested that the boryl radical intermediate was involved in the transformation. We performed cyclic voltammetry (CV) experiments to further understand the details of the oxidation process (Scheme 4-3). First, we tested B2oct2 in different solvents. The results showed that methanol was necessary. Subsequently, the solvent under the standard conditions was tested, indicating that hexane promoted the reaction. Second, we tested the CV of 2-trifluoromethyl-1,3-enyne under the standard conditions and without B2oct2 in the reaction system. According to the CV experiments, under electrochemical conditions, the results indicate that electrochemical mediated single electron transfer oxidation of silicon boron or boron reagents can easily occur in the form of borate through solvent activation.12h Finally, the reaction of homoallenylboronates with benzaldehyde produced dienyl alcohol (9a). The oxidative derivatization of 3a afforded the corresponding homoallenol (8a) in excellent yield (Scheme 4).
Based on mechanism studies and evidences13 regarding the electrochemical borylation reactions, we suggested the following plausible mechanism (Scheme 5). First, the methoxide anion generated by the reduction of methanol at the cathode reacted with bis(hexyleneglycolato)diboron and formed the borate species A. The subsequent anodic oxidation of borate A generated radical species B, which decomposed into octB–OMe and borate radical C. Then, the radical rapidly added to the double bond of the 1,3-enyne to produce a highly reactive propargyl radical D, which tautomerized with the allenyl radical intermediate D1,1a,9a and then the radicals were reduced to negative ions E and E1 at the cathode, which were protonated by the solvent (MeOH), generating the final products 4a and 3a. Finally, the main product allene derivatives and alkyne derivatives were obtained by protonation.
Footnote |
| † Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d4qo00332b |
| This journal is © the Partner Organisations 2024 |