Silicon, belonging to Group Ⅳ elements, ranks as the second most abundant element in both the Earth's crust surface and soils [
1-
3]. Silicone–based organosilicon compounds possess distinct advantages such as controllable sol–gel condensation, lipophilicity, and low surface tension owing to their unique structure [
4-
6]. As a result, silicone compounds, serving as crucial raw materials, have assumed an indispensable role in materials science, polymer chemistry, and agrochemistry (
Fig. 1A) [
7-
12]. With the development of organic photo/electrocatalysis, chemoselective silylation
via silyl radicals holds great potential in the industrial realm (
Fig. 1B, left) [
13,
14]. Typically, silyl radicals are generated through the direct activation of Si–H bonds [
15-
23]. Silylboronic pinacol esters, containing Si–B bonds, possess advantages such as good stability, commercial availability, and ease of preparation [
24,
25], and they have the potential to generate silane radicals. Nevertheless, silylboronic pinacol esters have a high oxidation potential (approximately +2.0 V
vs. SCE). This high oxidation potential makes it more challenging to cleave the Si–B bond under photoredox conditions [
26]. The conventional method of activating Si–B bonds usually involve introducing nucleophilic reagents to coordinate with boron in order to lower the redox potential (
Fig. 1B, right) [
26]. Amine radicals, frequently used as radical transfer reagents in photocatalysis, have been demonstrated to drive C–B bond activation to generate carbon-radicals [
27-
32]. We believe that they also hold excellent synthetic potential for use in the activation of silylboronic pinacol esters.
Fluorine–containing compounds endow drug molecules with advantages such as enhanced lipophilicity, specific conformation, and improved bioavailability [
33-
40]. In recent years, with the continuous maturation of synthetic techniques, the number of fluorine–containing drugs and pesticides featuring complex structures has been on the rise [
40].
gem–Difluorovinyl compounds have attracted extensive interest from researchers due to their isosteric relationship with carbonyl groups [
41-
48].
Given the unique properties of silicon–containing organic compounds, we postulate that introducing both "Si" atoms and
gem–difluoroalkene fragments into the same molecule may offer medicinal chemists a novel perspective. Inspired by the research advancements, we devise a radical transfer strategy for generating silane radicals. Under visible–light irradiation, the catalyst transforms the amine into amine radicals, which in turn induce the cleavage of the Si–B bond to generate silyl radicals. This approach successfully enables the defluorination of trifluoroolefins for the synthesis of fluorine–containing organosilicon compounds under mild reaction conditions (
Fig. 1C). This study synthesized novel difluoromethylsilane compounds, showing potent fungicidal activity (inhibition rates > 60%) against major plant pathogens at 50 μg/mL. Their EC
50 values compared favorably with commercial fungicides. DFT and MEP analyses linked activity to molecular orbital distributions and electrostatic properties.
In our initial experiments, we selected 1-(benzyloxy)-4-(3,3,3-trifluoroprop-1-en-2-yl)benzene (
1a) and triethyl(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)silane (
2a) as model substrates (
Table 1). A solution of these substrates in dry acetonitrile ([
1a] = 0.1 mol/L) was prepared, containing 4CzIPN (2 mol%) as a photocatalyst, morpholine (1.5 equiv.) as an activation reagent, and K
3PO
4 (2 equiv.) as a base. This solution was irradiated with a 456 nm LED at room temperature (~25 ℃) under an argon atmosphere for 24 h. As a result, the desired addition product
3 was obtained in an 85% yield (entry 1).
We tested various other photocatalysts, such as Ir[dF(CF3)ppy]2(dtbbpy)PF6 (entry 2; for more detailed information, please refer to Supporting information for details). However, 4CzIPN was found to be the optimal choice. Different activation reagents were also evaluated. Piperidine and n–butylamine led to lower yields of product 3 (31% and trace amounts, respectively; entries 3 and 4). Solvent screening indicated that dichloromethane gave slightly lower yields of 3 compared to acetonitrile (compare entry 1 with entry 5), while toluene gave a relatively low yield, with only trace amounts detected. The other solvents tested yielded < 45% (see Supporting information for details). When we replaced K3PO4 with Na2CO3 or Cs2CO3 as the base, the yields of 3 decreased to 65% and 82%, respectively (entries 7 and 8). Control experiments confirmed that the activation reagent, the photocatalyst, light, and the exclusion of oxygen were all essential for this transformation (entries 9–11).
After identifying the optimized conditions, we explored the substrate scope for the synthesis of fluorinated organosilicon compounds through the defluorination functional-group reaction between
α-trifluoromethyl arylalkene and silylboronic pinacol esters derivatives (
Fig. 2). These conditions were widely applicable to
1 and demonstrated excellent selectivity and functional-group compatibility. Electron–withdrawing groups attached to the phenyl ring were well–tolerated, resulting in good yields (45%‒51%) of the corresponding
gem–difluoroalkenes
4–
5. We observed that when the para-position of the phenyl ring in the arylalkene was substituted with electron–donating groups such as methyl,
n-propyl, or methoxy, the corresponding products
6–
14 were produced with yields ranging from moderate to excellent (64%–81%).
However, when the
meta–position was substituted with an aryl ring or a methoxy group, the yields decreased slightly (
15–
16, 61%–62%). By placing bisubstituted groups at different positions on the benzene ring, the desired products
17–
21 were obtained with moderate yields (52%–68%). Alkenes with trifluoromethyl groups located on the benzene ring were poorly tolerated in this reaction system, yielding the product in only 41% (
22). When
α-trifluoromethyl alkenes containing heterocycles such as dibenzo[
b, d]thiophene-4-yl and naphthalene-2-yl moieties were used, the corresponding products
23–
24 were obtained in moderate yields (49%−55%). Next, we explored the reaction scope of silylboronic pinacol esters using compound
1a as the free–radical acceptor (
Fig. 2). As shown, various silylboronic pinacol esters bearing different sterically diverse substituents, including triisopropyl, dimethylethyl, and dimethyl–
tert–butyl groups, as well as some primary silylboronic pinacol esters, successfully reacted with
α-trifluoromethyl arylalkene
1a. This reaction afforded the desired products
25–
32 in yields ranging from 51% to 83%. Subsequently, for further evaluating the applicability of this reaction, various complex
α-trifluoromethyl arylalkene-derived from natural products
33–
37 were subjected to the methodology.
To further demonstrate the synthetic potential of our method, we carried out a gram-scale continuous-flow reaction to synthesize the difluoroketone compound (
Fig. 3). With a flow rate of 1.5 mL/min and a residence time of 24 h, we were able to scale up the reaction to 4 mmol.
Plant-pathogenic fungi pose a serious threat to global food security and ecosystems. Chemical control is an effective means of preventing and managing fungal diseases [
49,
50]. However, the repeated long-term use of fungicides with a single mode of action can lead to resistance issues. An effective approach to developing new fungicides is the construction of new fungicide skeletons through organic synthesis methodology [
51-
54]. Compounds containing silicone that are environmentally compatible and exhibit synergistic effects have attracted considerable attention in the field of drug discovery and development [
1-
12]. Using this method, we successfully synthesized a series of novel
gem–difluoroallylsilane compounds and investigated their fungicidal properties.
The mycelial growth rate method was used to test the fungicidal activity of the
gem–difluoroallylsilane skeleton compounds against 10 pathogenic fungi at a concentration of 50 μg/mL. Commercial fungicides pyraclostrobin and boscalid were selected as positive controls (Table S6 in Supporting information). Some of the target compounds exhibited potent fungicidal activity against
Botrytis cinerea, Setosphaeria turcica and
Rhizoctonia solani. Most of these compounds exhibited inhibitory activity against
B. cinerea of greater than 60%, with compounds
13–15 inhibiting it by > 75%. Compounds
3, 8, 11, 18 and
19 exhibited an inhibition rate of over 70% against
S. turcica, while compounds
13, 14, 15, 23, 26 and
33 demonstrated an inhibition rate of over 60% against
R. solani. Compound
13 exhibited relatively broad-spectrum antifungal activity, with inhibition rates exceeding 60% against four phytopathogenic fungi. These results suggest that the
gem–difluoroallylsilane skeleton has fungicidal properties. Compounds with an inhibition rate greater than 75% were selected for EC
50 activity determination (
Table 2). For
B. cinerea, the EC
50 values of compounds
13–15 ranged from 2.39 μg/mL to 3.09 μg/mL, which are lower than those of the commercial fungicides pyraclostrobin (1.30 μg/mL) and pydiflumetofen (0.62 μg/mL). Compound
8 (2.58 μg/mL) exhibited comparable activity to boscalid (2.97 μg/mL) against
S. turcica, but was less effective than pyraclostrobin (1.77 μg/mL). Compounds
8 and
13 showed good antifungal activity and could serve as lead compounds for further structural optimization and mode-of-action studies. For
R. solani, the EC
50 values of compound
14 and
33 were determined to be 14.64 and 8.31 μg/mL, respectively. Although both compound
14 and
33 demonstrated inhibitory effects against
R. solani, their potency was lower than that of the commercial fungicides boscalid (1.63 μg/mL) and pyraclostrobin (0.03 μg/mL). This class of compounds can serve as lead fungicides for structural optimization and mechanism-of-action studies.
To further investigate the structural properties underlying the antifungal activity of target compounds, density functional theory (DFT) calculations were performed on the high-activity compounds
8 and
13, as well as the low-activity compound
30. The electron distributions of the HOMO and LUMO for the highly active compounds
8 and
13 were similar. In contrast, the LUMO of the low-activity compound
30 was mainly distributed over the benzene ring and the
gem–difluoroalkene moiety. This suggested that variations in substituents on the benzene ring significantly impact biological activity (
Fig. 4A).
Fig. 4B shows that, for compounds
8 and
13, the MEP maps indicate that the
gem–difluoroalkenyl group is located in the negative electrostatic potential region (blue), suggesting nucleophilic characteristics. Meanwhile, the silyl group and aromatic ring are situated in the positive potential region (red), indicating electrophilic properties. Optimizing the electrostatic potential around the active core region (
e.g., the aromatic ring) facilitates effective interaction with fungal receptors or enzyme active sites. These results imply that variations in factors such as energy levels, molecular orbital distribution and MEP characteristics may account for differences in antifungal activity by affecting receptor interactions.
Subsequently, the effect of active compound
13 on the mycelial morphology of
B. cinerea was observed using scanning electron microscopy. The results were shown in
Fig. 4C. After treatment with compound
13, the mycelia exhibited obvious collapse and shrinkage, while the mycelial structure in blank control maintained smooth and regular. This result indicated that compound
13 could disrupt the mycelial morphology, thereby interfering with the normal growth of
B. cinerea.
To gain insight into the reaction pathway, several control experiments were performed. We conducted a light/dark experiment (see Supporting information for details); the results revealed that in the absence of light, the reaction halted. This suggests that continuous visible-light irradiation is essential for this amination reaction to proceed. A series of cyclic voltammetry measurements were carried out with Ag/AgCl as a reference electrode in CH3CN using tetrabutylammonium perchlorate as a supporting electrolyte. The purpose was to determine the possible changes in oxidation-reduction potential after adding morpholine. The cyclic voltammetry results showed that the oxidation peak potential of morpholine (E = +0.95 V vs. SCE) was lower than that of the silylboronic pinacol esters (E = +1.94 V vs. SCE), indicating that morpholine would be preferentially oxidized (see Supporting information for details). Furthermore, adding morpholine to the silylboronic pinacol esters did not result in a shift in the NMR signal of [¹¹B], suggesting that no direct complex was formed between morpholine and the silylboronic pinacol esters (see Supporting information for details).
On the basis of the above-mentioned experiments and literature reports, we propose the reaction mechanism (
Fig. 5): Initially, photocatalyst is excited by light irradiation giving excited state of photocatalyst
35. morpholine quenches the excited-state photocatalyst to generate an amine radical
36 via the SET process and lead to the transformation from
35 to
37. Then amine radical
36 interacts with the boron species to liberate the corresponding silyl radical. Subsequently, the silyl radical undergoes a Michael addition reaction with the
α-trifluoromethyl alkenes to generate the CF
3-styrene carbon radical
38, Single-electron reduction of the CF
3-styrene carbon radical by
37 generates the CF
3-styrene carbanion intermediate, which undergoes a
β-fluoride elimination reaction to afford the target product and complete the photoredox cycle.
In conclusion, we have developed a photochemical method to generate silyl radicals from readily available silylboronic pinacol esters through homolytic substitution reactions of the boron group with amine radicals. Under mild conditions, this method excels in forming silicon-containing difluoroalkene compounds and can be scaled up to the gram level. Antifungal activity showed that at 50 μg/mL, some compounds showed > 60% inhibition against B. cinerea, S. turcica and R. solani, with EC50 values comparable to or lower than commercial fungicides boscalid and pyraclostrobin. DFT and MEP analyses revealed that high-activity compounds like 8 and 13 had distinct molecular orbital distributions and electrostatic properties compared to low-activity ones. Scanning electron microscopy indicated that compound 13 effectively inhibited the mycelial growth of B. cinerea. These compounds serve as promising leads for developing new fungicides with optimized structures and mechanisms.
PDF
Full
47
27
关闭全屏
No. 86 Xueyuan South Road, Haidian District, Beijing
010-62662699(Operation and Technical Support)
qkjq@cast.org.cn
