Hao-Ni
Qin
a,
Hao-Wen
Jiang
a,
Yi
Zhao
b,
Saira
Qurban
a,
Ke-Chun
Wang
a and
Peng-Fei
Xu
*a
aState Key Laboratory of Applied Organic Chemistry, College of Chemistry and Chemical Engineering, Lanzhou University, Lanzhou 730000, Gansu, China. E-mail: xupf@lzu.edu.cn
bTecon Pharmaceutical Co., Ltd, No. 109, Jintun Road, Urumqi City 830023, Xinjiang, China
First published on 13th January 2025
Substantial advancements have been achieved in the field of photocatalytic borylation utilizing 4c-7e Lewis base-boryl radicals. However, the utilization of 3c-5e neutral boryl radicals for C–B bond formation remains relatively underexplored due to their inherent instability. In this study, we successfully demonstrated the direct construction of C–B bonds using sodium tetraarylborate as a key reagent. This was accomplished by effectively stabilizing diaryl boryl radicals with nitrile compounds, thereby facilitating the synthesis of valuable boron-containing compounds. Overall, our research elucidates the significant role played by sodium tetraarylborate in enabling an efficient and versatile approach for synthesizing of 1,4,2-diazaborole analogs through a photocatalyzed [3 + 2]-annulation reaction. This mild and adaptable methodology expands synthetic strategies for obtaining diverse derivatives of 1,4,2-diazaboroles, with the RCN–BAr2 complex serving as an effective boron–nitrogen synthon that opens up pathways to multiple boron–nitrogen heterocycles. Furthermore, this breakthrough significantly enhances the applicability of sodium tetraarylborate in photoredox catalysis.
Ongoing research on the analogs of 1,4,2-diazaborole has been primarily focused on its exploration and synthesis optimization, with various boron reagents developed to date (Scheme 1B). As early as 1984, the Sadhu group devised a strategy utilizing α-iodoboronic ester in combination with amidine and n-BuLi to synthesize a single compound 1,4,2-diazaborole, achieving moderate yields.29 Subsequently, the Pawelke group synthesized the key boron reagents known as azoniaboratacyclopropanes which underwent ring-opening and expansion reactions to yield several derivatives of 1,4,2-diazaborole.30 Later on, the Erker group established a mild three-component sequential protocol using their described FmesBH2 along with isocyanides and nitriles. This showcased an efficient synthetic strategy.36 Recently, the Miyata group has developed a mild synthetic strategy using potassium α-aminoborate and azirine, thereby providing opportunities for further exploration.37 Despite recent advancements, there is still a strong demand for a more concise, efficient, and step-economical synthesis of 1,4,2-diazaborole analogs from readily available starting materials. We propose investigating the utilization of an economical and commercially available boron reagent for their direct synthesis.
In the field of photocatalytic borylation, significant advancements have been made in the realm of boryl radical chemistry in recent years.41–68 While Lewis base-boryl radicals, such as N-heterocyclic carbene (NHC) boryl radicals46–54 and amine boryl radicals,55–62 have been extensively studied, our group has also conducted relevant studies in this field.57,59 However, compared to the well-studied 4c-7e boryl radicals, the inherently unstable 3c-5e neutral boryl radicals have received considerably less attention. Amongst the precursors for neutral boryl radical, sodium tetraarylborate has garnered attention since the 1960s due to its affordability and stability.63,64 Nevertheless, its role in photochemistry has primarily focused on biaryl compound synthesis while leaving unexplored potential for generating neutral boryl radicals.65 In 2022, the Xia group achieved a significant breakthrough by employing neutral diaryl boryl radicals for the activation of the C–O bond in benzyl alcohol.66 The Wang group further employed diaryl boryl radicals as precursors for aryl radicals in reactions.67 Subsequently, the Sharma group demonstrated a core strategy by using diphenyl boryl radicals for C–Br bond activation.68 The aforementioned research studies successfully employed neutral boryl radicals derived from sodium tetraarylborate in photoredox catalysis to generate key carbon radicals; however, their application in C–B bond formation for boron-containing compounds remains unexplored (Scheme 1C).
Inspired by groundbreaking previous research, our focus is on the development of a catalytic strategy for synthesizing 1,4,2-diazaborole analogs, a distinct class of five-membered boron–nitrogen heterocycles. The [3 + 2]-annulation reaction serves as an efficient method for rapidly constructing five-membered carbon (hetero) cycles. Therefore, we propose to investigate whether neutral diphenyl boryl radicals can facilitate this reaction with suitable substrates, thereby enabling the rapid synthesis of 1,4,2-diazaborole analogs.
Building on the aforementioned literature, we propose that nitrile compounds, possessing a lone pair of electrons on the nitrogen atom, can function as Lewis bases by coordinating with electron-deficient diphenyl boryl radicals to yield intermediate II, specifically the MeCN–BPh2 complex.34,69 Intermediate II features two reactive sites: an electrophilic cyanide carbon and a nucleophilic boryl radical. This imparts significant potential for it to serve as a [3 atoms]-synthon in [3 + 2]-annulation reactions. Furthermore, we hypothesize that this synthon could interact with imines to construct the C–B bond and facilitate the synthesis of 1,4,2-diazaborole analogs through a [3 + 2]-annulation pathway.
Entry | PC |
1a![]() ![]() |
Solvent | Additive | Yielda,b |
---|---|---|---|---|---|
a Isolated yield. b Unless otherwise noted, reaction conditions are as follows: imine (0.05 mmol), NaBPh4 (x mmol), PC (3 mol%), MeCN (0.5 mL), 30 W blue LEDs, 0 °C, 12 h, under an argon atmosphere. c [Ir(dF(CF3)ppy)2dtbbpy]PF6 (4 mol%). d Reaction time extended to 24 h. e Reaction time extended to 30 h. f Reaction time extended to 27 h. | |||||
1 | PC1 | 1![]() ![]() |
MeCN | — | 47 |
2 | PC2 | 1![]() ![]() |
MeCN | — | 33 |
3 | PC3 | 1![]() ![]() |
MeCN | — | 22 |
4 | PC4 | 1![]() ![]() |
MeCN | — | 0 |
5 | PC5 | 1![]() ![]() |
MeCN | — | 0 |
6 | PC1 | 1![]() ![]() |
MeCN | — | 34 |
7 | PC1 | 1![]() ![]() |
MeCN | — | 49 |
8 | PC1 | 1![]() ![]() |
MeCN | — | 44 |
9 | PC1 | 1![]() ![]() |
MeCN | — | 51c |
10 | PC1 | 1![]() ![]() |
MeCN | 4 Å MS (20 mg) | 55c |
11 | PC1 | 1![]() ![]() |
MeCN + DCM | 4 Å MS (20 mg) | 60c |
12 | PC1 | 1![]() ![]() |
MeCN + DCM | 4 Å MS (20 mg) | 73c,d |
13 | PC1 | 1![]() ![]() |
MeCN + DCM | 4 Å MS (20 mg) | 58c,e |
14 | PC1 | 1![]() ![]() |
MeCN + DCM | 4 Å MS (20 mg) | 78c,f |
15 | PC1 |
3![]() ![]() ![]() |
MeCN + DCM | 4 Å MS (20 mg) | 91c , |
16 | Without PC | Trace | |||
17 | Without light | N.R. |
After establishing the optimal reaction conditions, we proceeded to expand our investigation in order to assess the substrate scope and generality of the reaction. Initially, a variety of imines derived from aromatic amines were subjected to testing. As indicated in Table 2, the incorporation of electron-donating groups such as methoxy (4aa), tert-butyl (4ab) and phenyl (4ac) onto the aniline aromatic ring consistently resulted in favorable results, thereby demonstrating compatibility within the reaction system. However, it should be noted that introducing an amido group (4ad) led to a modest reduction in yield. Furthermore, both unsubstituted aniline-derived imine (4ae) and 2-naphthylamine-derived imine (4af) exhibited commendable yields. In contrast, halogen-substituted anilines (4ag–4ai) showed a slight decrease in yields. Notably, electron-withdrawing groups such as acetyl (4aj), cyano (4ak), and ester groups (4al) were well tolerated, resulting in a slight reduction in yields. Next, different boron sources were investigated. Substituting the phenyl group in sodium tetraphenylborate with 4-methylphenyl (4ba) and 2-naphthyl (4bb) still afforded the target product, albeit with reduced yield. Subsequently, nitriles were tested. In addition to acetonitrile, cyclopropyl nitrile (4ca), isobutyronitrile (4cb) and 4-methoxybenzenecarbonitrile (4cc) also underwent successful transformation, demonstrating the method's broad applicability. The configuration of compound 4aa was determined by X-ray crystal structure analysis (CCDC 2381718).
The reaction's scope and versatility were evaluated by testing various imines derived from α-ketoesters. As indicated in Table 3, the reaction provided moderate to good yields for imines derived from α-ketoesters, with increased steric hindrance resulting in lower yields (4am–4ar). Subsequently, the influence of electron-donating substituents on the aromatic ring of α-ketoesters, such as methyl (4as), ethyl (4at), tert-butyl (4au), phenyl (4av), and methoxy groups (4aw), was investigated, yielding slightly lower results compared to the model substrate.
Remarkably, when an iodine substituent was introduced at the para position (4ax), a yield of 89% was achieved, demonstrating the robustness of the reaction pathway against this substitution.
To further broaden the applicability of our methods, sodium tetraphenylborate can also undergo photocatalytic [3 + 2]-annulation reactions with alkenes, leading to the generation of additional classes of boron–nitrogen heterocycles. This shows that introducing both electron-withdrawing group-substituted styrenes (6a–6c) and electron-rich 1,1-diphenylethylene (6d) into the template reaction yields the desired azaborole in moderate to good efficiency (Table 4). To demonstrate the practicality of our method, we conducted a scale-up experiment, successfully synthesizing 1.1 grams of the target product with a yield of 77% (Table 5a). These results highlight the advantages of our strategy for synthesizing diverse boron–nitrogen heterocycles, including mild reaction conditions and broad functional group compatibility.
We then conducted a series of experiments to elucidate the reaction mechanism. Under standard conditions, the reaction was inhibited by radical scavengers, and the corresponding product formed through radical trapping was identified using high-resolution mass spectrometry (HR-MS) (Table 5b). Control experiments demonstrated that both light irradiation and the presence of a photocatalyst are indispensable for the reaction to occur (Table 1: entries 16–17). A radical clock experiment employing 1-phenylvinylcyclopropane and sodium tetraphenylborate yielded the formation of a cyclopropane ring-opened product, which was identified by HR-MS (Table 5c). These experimental findings provide support for a photoredox radical process. Fluorescence quenching experiments revealed that both imine and sodium tetraphenylborate can quench the excited state of the photocatalyst; however, imine 1a exhibited a more pronounced effect (Table 5d, left). Further light on/off experiments reveal that any chain propagation process was transient and that light played a crucial role in the formation of the product (Table 5d, right). The existence of the radical intermediate II is indirectly confirmed by the hydrogen abstraction reaction (ESI Fig. S3†). Additionally, the dimeric product of the reduced imine, confirmed by HR-MS, provides evidence for the radical coupling pathway leading to the target product (Scheme 2). However, it cannot be completely ruled out that direct attack by the boryl radical on the imine via a single-electron transfer mechanism may occur (ESI Fig. S8†).
Based on the mechanistic studies, a plausible reaction mechanism is proposed in Scheme 2. Upon irradiation, an excited IrIII undergoes single-electron transfer (SET) on the imine to deliver the α-amino radical anion I and a resulting IrIV. Subsequently, this IrIV species is reduced back to its ground state by NaBPh4, while simultaneously oxidizing NaBPh4 to produce a diphenyl boryl radical. In the presence of acetonitrile, it is hypothesized that nitrogen initially coordinates with the diphenyl boryl radical, forming intermediate II—indirectly confirmed by HR-MS. Intermediate I and II then engage in a radical–radical cross-coupling reaction to form III. The nitrogen anion in III subsequently undergoes nucleophilic addition with the cyano group, ultimately resulting in the target 1,4,2-diazaborole product.
Footnote |
† Electronic supplementary information (ESI) available. CCDC 2381718. For ESI and crystallographic data in CIF or other electronic format see DOI: https://doi.org/10.1039/d4sc08085h |
This journal is © The Royal Society of Chemistry 2025 |