Zhijun
Wang
,
Lanfeng
Wei
,
Jichang
Liu
*,
Yu
Wei
* and
Liang
Xu
*
School of Chemistry and Chemical Engineering/State Key Laboratory Incubation Base for Green Processing of Chemical Engineering, Shihezi University, Shihezi, 832003, China. E-mail: xuliang4423@shzu.edu.cn
First published on 9th November 2022
A method for direct oxidation of boronic esters to the carbonyl groups has been established herein. Under the irradiation of visible light and O2 atmosphere, ketones and aldehydes can be obtained in moderate to excellent yields. Using tetrabutylammonium tribromide (TBATB) as the catalyst, this procedure avoids using of metal-based (photo)catalysts. α-Borylalkyl radicals have been proposed as the key intermediates of this oxidation process.
A close inspection of the literature revealed several existing strategies. In 2016, Floreancig, Deiters and their coworkers disclosed that oxidation of α-O-substituted alkylboron with H2O2 would release aldehydes/ketones through the collapse of hemiacetal intermediates (Scheme 1a).4 In 2019, Partridge group reported a copper catalyzed method for the selective oxidation of benzylic boronic esters to benzoyl moieties, using air as the terminal oxidant (Scheme 1b).5 Later, Penhoat group realized continuous photocatalyzed aerobic oxidation of benzylic trifluoroborates to benzaldehydes, using UV irradiation and Ir-based photocatalyst to generate benzylic radical intermediates (Scheme 1c).6 In spite of these achievements,7 generally, from alkylborons to the carbonyl moieties, sequential oxidation processes are still required via the intermediate of alcohols.8 This situation leaves ample space to develop more step-economical pathways that directly connect alkylborons and the carbonyl groups in a single vessel.
On the other hand, the generation and application of α-borylalkyl radicals have been receiving increasing attention from the organic community,9 along with the prosperity of organoboron chemistry and radical reactions10 in the last decade. As depicted in Scheme 1, α-boryl radicals are usually stabilized by delocalization onto the empty p-orbital of the sp2-boron atom through conjugation effect,11 which renders their generation relatively easier and thus enables diverse downstream transformations,12 including the construction of α-heteroatom-substituted alkylborons.13 For example, in 1970, Brown group realized α-bromination of organoboranes via α-boryl radicals that were generated from hydrogen atom transfer (HAT) to bromine radicals.13d
Grounded on this, it was questioned whether α-borylalkyl radicals could be utilized to forge the carbonyl groups. To the best of our knowledge, no such processes had been described in the literature. As part of our efforts in the development of photocatalytic oxidation reactions,14 herein we disclose a metal-free visible-light induced protocol to oxidize boronic esters to ketones/aldehydes directly, using tetrabutylammonium tribromide (n-Bu4N+Br3−, TBATB) as the catalyst (Scheme 1).
As indicated in Table 1, the optimal reaction conditions were established by systematically evaluating reaction parameters, including the dosages of TBATB, light sources, reaction time, and solvents. With tetrahydrofuran (THF) as the solvent, under an O2 atmosphere and irradiation of 10 W blue LEDs (455 nm), the HPLC yield of the target product 2a could reach 98% after 16 hours’ stirring, in the presence of 10 mol% of TBATB (Table 1, entry 1). The crucial roles of the above-mentioned parameters were then validated by control experiments. In the absence of TBATB, oxygen, or visible light, the signal of 2a could not be detected by HPLC analysis, indicating this transformation was a photoinduced TBATB-catalyzed aerobic process (entries 2–4). Furthermore, only trace 2a could be observed when the oxygen atmosphere was replaced by an air atmosphere (entry 6), implying a high concentration of oxygen was necessary to enable this transformation. Finally, the effect of different solvents was examined (entries 7–13). While alcoholic solvents, NMP and MeCN resulted in 67–93% yields of 2a, the application of DMSO and H2O was found detrimental, totally suppressing the desired transformation.
Entry | Variations from the ‘standard’ conditions | Yieldb (%) |
---|---|---|
a Standard reaction conditions: 1a (0.20 mmol), TBATB (0.02 mmol, 10 mol%), O2 atmosphere, solvent (1.0 mL), rt = room temperature, 10 W 455 nm blue LEDs, 16 hours. b HPLC yield. c N.D. = not detected. | ||
1 | None | 98 |
2 | Without TBATB | N.D.c |
3 | Without O2 (under N2) | N.D. |
4 | Without LEDs | N.D. |
5 | White LEDs instead of blue LEDs | 69 |
6 | Air instead of O2 | Trace |
7 | EtOH instead of THF | 87 |
8 | n-PrOH instead of THF | 93 |
9 | i-PrOH instead of THF | 82 |
10 | NMP instead of THF | 67 |
11 | MeCN instead of THF | 85 |
12 | DMSO instead of THF | Trace |
13 | H2O instead of THF | Trace |
Under the above-mentioned optimal conditions, we investigated the substrate scope of this photoinduced oxidation reaction (Scheme 2). As shown, acetophenone 2a could be obtained in 94% isolated yield.
Scheme 2 Substrate scope for photoinduced oxidation of boronic esters to ketones/aldehydes. aNMP as solvent, TBATB (20 mol%). bNMP as solvent, TBATB (20 mol%), 30 hours. |
Substrates with diverse substituents were well compatible with the conditions. Generally, the reactivity of benzylic boronic esters with electron-donating groups on the phenyl skeletons are higher than those with electron-withdrawing groups, no matter for the synthesis of ketones or aldehydes. For example, para-OMe-acetophenone 2k and para-OMe-benzaldehyde 2t were obtained in 88% and 92 yields, respectively, while their para-Cl counterparts were isolated in only 55% and 68% yields under the same conditions. When para-CF3-acetophenone was applied, only trace oxidized product was observed, further confirming the negative effect of electron-withdrawing groups. On the macro level, it was well-known electron-rich substrates were more susceptible to be oxidized. On the molecular level, this difference might be rationalized by the captodative or push–pull stabilization effect on the involved benzylic radical intermediates, which were substituted with electron-accepting functionalities like α-boryl. In addition, among ortho-, meta-, and para-methyl-substituted boronic esters, the ortho-substituted case resulted in the lowest yield (2d, 51% vs.2b, 74% vs.2c, 73%), indicating the negative influence of steric hindrance.
For the substrates with halogen atoms as substituents, the corresponding products 2e, 2f, 2g, 2h were obtained in moderate yields, leaving opportunities for further decoration of these products. Furthermore, naphthyl starting materials with more extended conjugated systems afforded 2m and 2n in 66% and 75% yields, respectively. When R2 was benzyl, 2p was obtained in 64% yield. Cyclic 1-indanone 2q could also be accessed via this protocol in 80% yield. When R2 was hydrogen atoms, the corresponding aromatic aldehydes would be exclusively obtained in moderate to excellent yields, without the formation of the over-oxidized carboxylic acids.
The preliminary attempts to oxidize inactivated aliphatic boryl species under these conditions reached no success. Also, trifluoroborates were found to be not suitable for this reaction system, probably due to the absence of vacant p-orbitals.
Subsequently, a series of control experiments were carried out to shed light on the reaction mechanism (Scheme 3). In the beginning, ethylbenzene 3 without substituents at the benzylic position was treated under the standard reaction conditions. Acetophenone 2a could not be detected by the HPLC analysis of the reaction mixture, thus ruling out the possibility of an initial deborylative protonation and subsequent benzylic oxidation pathway (Scheme 3a). Then, in the reaction was added water-18O (5.0 equiv.), which would not affect the overall yield of the oxidized product significantly (87%). However, mass spectrum analysis of the obtained product revealed the absence of 18O-labeled 2a. Taking the necessity of O2 for the conversion (Table 1, entries 3 and 6) into consideration, it was proposed the oxygen of 2a came from O2. Furthermore, when a typical radical scavenger, 2,2,6,6-tetramethylpiperidinooxy (TEMPO, 2.0 equiv.), was added, the formation of 2a was completely suppressed, indicating the reaction might go through a free radical process. Finally, after adding t-butyl alcohol (TBA, 2.0 equiv.), a hydroxyl radical scavenger, 2a was still generated in 58% yield, implying the hydroxyl radical might not be involved in the reaction mechanism.
Furthermore, (1-bromoethyl)benzene 5 and para-tert-butylbenzyl bromide 6 were treated under the optimized conditions (Scheme 4). The corresponding carbonyl product 2a and 7 could be isolated in moderate yields, indicating the reactions using benzyl bromides as the starting materials might share similar intermediates with the reactions using benzyl boronic esters.
Accordingly, the possible reaction pathways have been depicted in Scheme 5. The reaction began with taking away the benzylic hydrogen atom from 2a by the bromine radical (Br˙), which originated from the homolytic cleavage of Br–Br bonds in TBATB under illumination.16 Under the oxygen atmosphere, the α-borylbenzyl radical IM1 would trap one O2 to generate peroxyl radical IM2. Then, IM2 and hydrogen bromide (HBr) underwent a hydrogen atom transfer (HAT) process to afford IM3 and regenerate the bromine radical (Br˙), thus completing the catalytic cycle. Finally, intramolecular elimination of HOBpin from IM3 occurred, leading to the product 2a. The above-mentioned process should also be applicable for the conversion of benzylic bromides, via the intermediates IMA and IMB.
Footnote |
† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d2qo01457b |
This journal is © the Partner Organisations 2023 |