Efficient N -Trifluoromethylation of Amines with Carbon Disulfide and Silver Fluoride as Reagents

Open AccessCCS ChemistryRESEARCH ARTICLES6 Aug 2024Efficient N-Trifluoromethylation of Amines with Carbon Disulfide and Silver Fluoride as Reagents Haixia Song, Qin Wang, Xiaoying Wang, Yinbin Zhang, Xin-Hua Duan and Mingyou Hu Haixia Song School of Chemistry, Xi'an Key Laboratory of Sustainable Energy Material Chemistry, Engineering Research Center of Energy Storage Materials and Devices, Ministry of Education, Xi'an Jiaotong University, Xi'an 710049 , Qin Wang School of Chemistry, Xi'an Key Laboratory of Sustainable Energy Material Chemistry, Engineering Research Center of Energy Storage Materials and Devices, Ministry of Education, Xi'an Jiaotong University, Xi'an 710049 , Xiaoying Wang School of Chemistry, Xi'an Key Laboratory of Sustainable Energy Material Chemistry, Engineering Research Center of Energy Storage Materials and Devices, Ministry of Education, Xi'an Jiaotong University, Xi'an 710049 , Yinbin Zhang Department of Oncology, The Second Affiliated Hospital of Xi'an Jiaotong University, Xi'an 710004 , Xin-Hua Duan School of Chemistry, Xi'an Key Laboratory of Sustainable Energy Material Chemistry, Engineering Research Center of Energy Storage Materials and Devices, Ministry of Education, Xi'an Jiaotong University, Xi'an 710049 and Mingyou Hu *Corresponding author: E-mail Address: [email protected] School of Chemistry, Xi'an Key Laboratory of Sustainable Energy Material Chemistry, Engineering Research Center of Energy Storage Materials and Devices, Ministry of Education, Xi'an Jiaotong University, Xi'an 710049 Cite this: CCS Chemistry. 2024;0:1–11https://doi.org/10.31635/ccschem.024.202404432 SectionsSupplemental MaterialAboutAbstractPDF ToolsAdd to favoritesDownload CitationsTrack Citations ShareFacebookTwitterLinked InEmail The installation of a trifluoromethyl group onto a nitrogen atom can effectively modulate the basicity of amines owing to the strong electron-withdrawing effect of fluorine. Nevertheless, efficient and operationally simple methods for N-trifluoromethylation of amines are yet to be developed. This protocol involves the use of readily available secondary amines as starting materials, along with CS2 and AgF as the N-trifluoromethylation reagents, enabling the target molecules to be synthesized in a single step. The versatility of our method was demonstrated by successfully synthesizing N,N-dialkyl and N-(hetero)aromatic N-CF3-containing compounds with various substituents. Moreover, this methodology has been successfully applied to the late-stage modification of complex bioactive molecules, facilitating the synthesis of N-CF3 drug bioisosteres and N-CF3-tailored amino acids, which would broadly stimulate the drug discovery of N-CF3 containing molecules. Download figure Download PowerPoint Introduction Fluorinated compounds have gained significant attention in the fields of emerging pharmaceuticals and agrochemicals due to their enhanced lipophilicity, membrane permeability, and metabolic stability compared to their nonfluorine bioisosteres.1–5 Particularly, since the trifluoromethyl group is one of the most common motifs in numerous pharmaceuticals, its introduction has been extensively studied in synthetic organic chemistry.6–9 In the modern pharmaceutical domain, a significant proportion of biologically active molecules contain at least one nitrogen atom, with a specific focus on amines and nitrogen heterocycles. These nitrogen-containing compounds have the potential to form hydrogen bonds with specific binding sites, acting as either acceptors or donors.10–13 The incorporation of fluorine into amino compounds can modulate their basicity, lipophilicity, and conformational orientation, thereby enhancing or otherwise amending their bioavailability (Figure 1).3,14–16 Consequently, the incorporation of fluorinated moieties, such as the N-CF3 group, holds significant importance in the fields of pharmaceuticals and agrochemicals.17–20 In this regard, medicinal chemists have formulated a range of N-CF3-containing drug bioisosteres as candidates for kinase inhibitors and antitumor agents (Figure 1).21–25 Figure 1 | Drug candidates containing the N-CF3 motif and β-fluoroalkylamine unit. Download figure Download PowerPoint With regard to the synthesis of N-CF3 compounds, conventional methods typically entail intricate multistep procedures.18–20 These methods often involve deoxygenative/desulfurizative-fluorination and fluorine/halogen exchange of amine derivatives.26–44 Additionally, isothiocyanates have been effectively transformed into N-CF3 entities through desulfurizative-fluorination with AgF (Scheme 1a).45–55 In contrast, direct nitrogen trifluoromethylation has been pursued using electrophilic trifluoromethylation agents like Umemoto reagent56 and Togni reagent57–59 reagents, or via metal-mediated/catalyzed trifluoromethylation employing the Ruppert-Prakash reagent (TMSCF3)60–73 (Scheme 1b). Nonetheless, these methods have encountered constraints such as the necessity for prefunctionalized amines, harsh reaction conditions, intricate multistep synthesis, and challenges in product isolation. Scheme 1 | (a–e) Synthesis approaches for N-CF3 compounds. Download figure Download PowerPoint The reaction between AgF and carbon disulfide (CS2) is known to yield AgSCF3 (Scheme 1c).74,75 Moreover, the "−SCF3" anion has a tendency to generate electrophilic thiocarbonyl fluoride (S=CF2).26 On the other hand, when an amine substrate is combined with a trifluoromethylthiolate, it produces thiocarbamoyl fluoride, which readily undergoes desulfurizative-fluorination upon treatment with AgF, resulting in the formation of the N-CF3 product (Scheme 1d).26–31 Nevertheless, the direct N-trifluoromethylation of amines using AgF and an in situ "SCF3" source has not been reported previously. Building upon previous significant discoveries, we hypothesized that employing CS2 and AgF could provide a straightforward N-trifluoromethylation strategy. This concept was rooted in the following considerations: (1) amines readily react with CS2 to form carbamodithioic acid adducts; (2) the high affinity between silver ion and sulfide (Ksp of Ag2S = 6.3 × 10−50 at room temperature), AgF exhibits robust desulfurization and fluorination capabilities76 (Scheme 1e). Considering the comparably lower cost of AgF to alternative fluorination/fluoroalkylation reagents, along with its relatively low toxicity, we anticipate that the incorporation of CS2 and AgF in straightforward processes for constructing N-CF3 functionalities would effectively overcome the constraints associated with existing methods. This advancement holds the potential to promote the broader application of N-CF3 compounds across diverse preparative realms. Experimental Methods General procedure for the syntheses of N-CF3 dialkyl amines A sealed tube was charged with dialkyl amines (0.20 mmol), AgF (0.90 mmol, 114.2 mg), 1,4-diazabicyclo[2.2.2]octane (DABCO; 0.10 mmol, 11.2 mg), ethyl acetate (EA, 1.5 mL), and a solution of CS2 (0.20 mmol, 15.2 mg) in EA (0.5 mL) under N2 atmosphere. The mixture was stirred at 80 °C for 6 h. When the reaction was completed, the mixture was filtered, and then the filtrate was concentrated in vacuo. The crude product was purified by column chromatography on silica gel (8–10 cm length) using a mixture of hexanes and EA as eluent to give the desired N-CF3 dialkyl amines. More experimental details and characterization are available in the Supporting Information. General procedure for the syntheses of N-CF3 alkylaryl amines A sealed tube was charged with alkylaryl amine (0.2 mmol), AgF (1.2 mmol, 148.6 mg), 4-pyrrolidinopyridine ( B1, 0.2 mmol, 29.6 mg), EA (1.5 mL), and a solution of CS2 (0.24 mmol, 18.3 mg) in EA (0.5 mL) under N2 atmosphere. The mixture was stirred at 40 °C for 20 h. When the reaction was completed, the mixture was filtered, and then the filtrate was concentrated in vacuo. The crude product was purified by column chromatography on silica gel (8–10 cm length) using a mixture of hexanes and EA as eluent to give desired N-CF3 alkylaryl amines. More experimental details and characterization are available in the Supporting Information. Results and Discussion Reaction design and development Initially, we commenced with our investigation by reacting AgSCF3 with dibenzylamine 1a, with or without 4-dimethylaminopyridine (DMAP) as a base. In both scenarios, we did not detect the product N,N-dibenzyl-1,1,1-trifluoromethanamine 2a; however, we observed a minor amount of dibenzylcarbamothioic fluoride ( 2a-1) (Scheme 2a). Subsequently, treating 1a and CS2 adduct dibenzylcarbamodithioic acid ( 1a-1) with AgF led to a 18% yield of 2a in the presence of DMAP and a 9% yield without DMAP (Scheme 2b). These results suggested that dibenzylcarbamodithioic acid could be an intermediate, albeit with low efficiency. These initial findings inspired us to proceed with one-step reactions involving 1a, CS2, and AgF; however, we did not observe the formation of 2a without a base. In contrast, the presence of DMAP as a base facilitated a 72% yield of 2a (Scheme 2c). Scheme 2 | (a–c) Experimental attempts for direct N-trifluoromethylation. Download figure Download PowerPoint Investigation of reaction conditions Building upon the earlier experimental results, we proceeded to investigate the reaction parameters by employing dibenzylamine ( 1a) as a representative substrate and using CS2 and AgF as reagents (Table 1). The reaction was carried out in MeCN under a nitrogen atmosphere at 80 °C for 8 h. During this process, we examined a range of bases, such as N,N-diisopropylethylamine (DIPEA), triethylamine, DABCO, and 1,8-diazabicyclo[5,4,0]undec-7-ene (DBU), which effectively promoted the reaction, affording 2a in the range of 45%–74% yields (Table 1, entries 1–4). However, when NaOH was used as a promoter or no base was employed, the desired product was not observed (Table 1, entries 5–6). Next, using DABCO as the preferred base, we investigated the impact of base loading but, unfortunately, altering the amount did not improve the yield (Table 1, entries 7–8). Encouragingly, by slightly increasing the amount of AgF from 4.0 to 4.5 and 5.0 equiv, the yields of 2a increased to 81% and 85%, respectively (Table 1, entries 10–11). Notably, by utilizing EA as the solvent and 4.5 equiv of AgF, the yield was enhanced to 94% (Table 1, entry 12). Further experimentation revealed that the reaction could be completed within 6 h (Table 1, entry 13). Table 1 | Optimization of Reaction Conditionsa Entry Base (equiv) Solvent AgF (equiv) Yield ( 2a, %) 1 DIPEA (0.5) MeCN 4.0 66 2 Et3N (0.5) MeCN 4.0 72 3 DABCO (0.5) MeCN 4.0 74 4 DBU (0.5) MeCN 4.0 45 5 NaOH (0.5) MeCN 4.0 0 6 None MeCN 4.0 0 7 DABCO (0.2) MeCN 4.0 56 8 DABCO (1.0) MeCN 4.0 47 9 DABCO (0.5) MeCN 4.5 81 10 DABCO (0.5) MeCN 5.0 85 11 DABCO (0.5) EA 4.5 94 12 DABCO (0.5) EA 4.5 95 (90)b aReaction conditions: The reaction was performed in a sealed tube, 1a (0.2 mmol), CS2 (0.2 mmol), AgF, base, solvent (2 mL), N2, 8 h. The yield was determined by 19F NMR with benzotrifluoride as an internal standard. bThe reaction time was 6 h, datum in the parentheses refers to isolated yield. Substrate scope With the optimal reaction conditions established (Table 1, entry 12), a variety of secondary alkyl amines were investigated. Encouragingly, a wide range of secondary alkyl amines underwent efficient transformation into N-CF3 products, providing the corresponding N-trifluoromethylation products in moderate to excellent yields (Scheme 3). When tetrahydroisoquinolines were used as substrates, the presence of strong electron-donating methoxy group substitution on the aromatic ring greatly favored the reaction, resulting in a 96% yield of the desired product 2b, while the electron-withdrawing bromo group delivered a 82% yield of compound 2c. Piperazine, a frequently encountered structural fragment in complex drug molecules, exhibited good compatibility under optimal conditions, delivering N-trifluoromethylated products in yields ranging from 77% to 98% for piperazines with diverse substituents ( 2d- 2l). These results demonstrated that functionalities such as alkoxy, halogen, ketonic carbonyl, cyano, nitro, and trifluoromethyl groups are well-tolerated, and the electronic characteristics of substituents on the aryl rings had minimal influence on the reaction. Moreover, when aromatic heterocycles such as pyridyl, furyl, thienyl, indolyl, isothiazyl, and isoxazoly were included as substrates, they underwent successful transformation under the reaction conditions ( 2j– 2k, 2m– 2q). Notably, amide and carbamate functionalities were well-tolerated in the reaction ( 2k– 2l). In the case of substrates bearing a hydroxyl group, the N-trifluoromethylated product 2r was obtained in 56% yield, and, importantly, no O-trifluoromethylated byproduct was detected in this reaction. Unlike dialkyl secondary amines, primary amine ( 1s) and amide ( 1t) are ineffective substrates. These advantages provide a reliable platform for the synthesis of N-CF3-containing complex bioactive molecules. The successful and easily scalable synthesis of compounds 2f and 2l (as exemplified) highlights the significant synthetic potential of our method for industrial production. Scheme 3 | Scope of dialkyl amines. Reaction conditions: 1 (0.2 mmol), CS2 (0.2 mmol), AgF (0.9 mmol), DABCO (0.1 mmol), EA (2 mL), N2, 80 °C, 6 h. Isolated yields. Download figure Download PowerPoint In contrast to the methodology employed by Tlili, which involved the use of diethylaminosulfur trifluoride (DAST) as the fluorination reagent (acidic conditions),31 our current approach (neutral conditions) simplifies the isolation of the N-CF3 products. Interestingly, we observed that electron-rich compounds 2b, 2m, and 2o underwent slow decomposition into carbamoyl fluoride under acidic conditions. To further investigate the acid-base stability, we performed experiments with compound 2f. The results revealed that 2f was prone to decomposition under medium to strong acidic conditions, while remaining stable under weak acidic, neutral, and basic conditions ( Supporting Information Table S12). These findings provide valuable insights for the future development of bioactive molecules containing N-CF3 groups. Subsequently, we proceeded to investigate the synthesis of N-CF3 compounds using secondary aromatic amines. However, we observed a remarkable decrease in reaction efficiency under the same conditions. This result could be attributed to the reduced nucleophilicity of aromatic amines compared to aliphatic amines. To address this issue, we turned to seek a stronger base and gratifyingly discovered that 4-(pyrrolidin-1-yl)pyridine ( B1) provided the most satisfactory results (for details, see Supporting Information Table S7). We then scrutinized the reaction with aromatic amines and obtained moderate to excellent yields of target products (Scheme 4). Switching from N-benzyl aniline to N-methyl-4-methylaniline, the yield increased from 78% to 92%, likely due to reduced steric hindrance ( 4a- 4b). When the 4-methylaniline motif was replaced with the more electron-rich 4-methoxyaniline, the yield dropped to 70% ( 4c). Comparing the results with 4-methoxy, 3-methoxy, and 2-methoxyaniline, the decreased yield of 2-methoxyaniline further confirmed the sensitivity of the reaction to steric hindrance ( 4c- 4e). Using N-butylaniline, the desired product 4f was obtained in 82% yield, while N-isopropylaniline afforded product 4g in only 21% yield. Further increasing the steric hindrance with substrate 3h almost completely shut down the reaction, however, by removing the 2-methoxy group from the phenyl ring to reduce steric hindrance, the reaction proceeded to give the target product 4i in 40% yield. Substrates bearing fluoro, chloro, bromo, and iodo substituents on the phenyl ring were all compatible and provided good yields of the products ( 4j- 4m). Other functionalities such as nitro, alkenyl, cyano, quinolone and cyclopropyl were well-tolerated, indicating that a radical process was less likely involved ( 4o- 4r). Cyclic aromatic amines were also applied smoothly to deliver the desired products in moderate to good yields ( 4s- 4x). From the results of 4n, 4o, and 4w, it can be observed that electron-poor anilines afforded inferior results compared to electron-neutral and electron-rich substrates. Regrettably, the desired product 4y was not attained when the amine molecule featured a phenolic hydroxyl group. Scheme 4 | Scope of alkylaryl amines. Reaction conditions: 3 (0.2 mmol), CS2 (0.24 mmol), AgF (1.2 mmol), B1 (0.2 mmol), EA (2 mL), N2, 40 °C, 20 h. Isolated yields. Download figure Download PowerPoint To showcase the practical utility of our method, a variety of complex drug molecules, drug precursors, and amino acid derivatives were subjected to the optimal reaction conditions for N-trifluoromethylation, and generally good to excellent yields were obtained (Scheme 5). Firstly, we performed N-trifluoromethylation on drug molecules, including fluoxetine ( 5a), sertraline ( 5b), troxipide ( 5c), desloratadine ( 5d), amoxapine ( 5e), ciprofloxacin methyl ester ( 5f), and fasudil ( 5g), affording the corresponding products in 72%–93% yields. These results demonstrated the feasibility of late-stage N-trifluoromethylation modification of complex molecules to enhance their lipophilicity, metabolic stability, and bioavailability. Next, we explored the N-CH3 modification of drug molecules by an N-CF3 group, that is, replacing the N-CH3 moiety with N-CF3. N-CF3 drug bioisosteres of commercially available drugs such as nefopam ( 5h), mirtazapine ( 5i), clozapine ( 5j), sildenafil ( 5k), terbinafine ( 5l), and naftifine ( 5m) were successfully synthesized with yields ranging from 76% to 92%. However, the N-CF3 analogue of imatinib ( 6n) was obtained in only 15% yield, likely due to the chelating effect of nitrogen atoms to Ag+, which impeded the N-trifluoromethylation reaction. Furthermore, we investigated the N-trifluoromethylation of amino acid derivatives, resulting in the corresponding N-CF3 analogues 6o- 6t in 51–92% yields. These compounds ( 6a- 6t) hold potential as drug candidates with possibly improved pharmacokinetic and pharmacodynamic profiles. Scheme 5 | N-Trifluoromethylation of drug molecules and bioisosteres. For 6a–6s, conditions A: 5 (0.2 mmol), CS2 (0.2 mmol), AgF (0.9 mmol), DABCO (0.1 mmol), EA (2 mL), N2, 80 °C, 6 h. For 6t, conditions B: 5t (0.2 mmol), CS2 (0.24 mmol), AgF (1.2 mmol), B1 (0.2 mmol), EA (2 mL), N2, 40 °C, 20 h. Isolated yields. Download figure Download PowerPoint Proposed mechanism Based on these findings, we propose a plausible reaction mechanism (Scheme 6). The reaction likely starts with the nucleophilic addition of an amine to CS2 at the carbon center, followed by deprotonation with a base to form thioamide species I. The high affinity between Ag+ and sulfide facilitates the desulfurization of I, leading to the formation of carbamoyl fluoride II. Further desulfurization and fluorination by AgF result in the formation of the N-CF3 product. Scheme 6 | Proposed mechanism. Download figure Download PowerPoint Conclusion In summary, we have developed a versatile method for the direct synthesis of N-CF3 compounds by using secondary amines, CS2, and AgF. This approach offers operational simplicity and utilizes readily available starting materials, overcoming the limitations associated with previous methods that relied on amine derivatives and complex reagents. The robustness of our method was demonstrated through the synthesis of N,N-dialkyl and N-(hetero)aromatic N-CF3 compounds with various substituents. Importantly, this methodology has been successfully applied to the late-stage modification of complex bioactive molecules, enabling the synthesis of N-CF3 drug bioisosteres and N-CF3-tailored amino acids. We anticipate that this work will find broad applications in medicinal chemistry and materials science, providing a valuable tool for the preparation of N-CF3 molecules with diverse applications. Supporting Information Supporting Information is available and includes details on experimental procedures, reaction optimizations, stability test, and compound characterizationdata. Conflict of Interest There is no conflict of interest to report. Funding Information This work was supported by National Natural Science Foundation of China (grant no. 21901196), the Natural Science Basic Research Plan in Shaanxi Province of China (grant no. 2020JQ-016), and Xi'an Jiaotong University (grant no. 71211920000001). Acknowledgments We thank Prof. Jinbo Hu (Shanghai Institute of Organic Chemistry) and Anis Tlili (Université de Lyon) for kind discussion. We thank Lu Bai and Chao Feng at the Instrument Analysis Center of Xi'an Jiaotong University for high-resolution mass spectrometry and nuclear magnetic resonance (NMR) analysis. We thank Prof. Aqun Zheng and Junjie Zhang at Xi'an Jiaotong University for NMR analysis. References 1. Reichenbächer K.; Süss H. I.; Hulliger J.Fluorine in Crystal Engineering-"The Little Atom that Could".Chem. Soc. Rev.2005, 34, 22–30. Google Scholar 2. Hagmann W. K.The Many Roles for Fluorine in Medicinal Chemistry.J. Med. Chem.2008, 51, 4359–4369. Google Scholar 3. Purser S.; Moore P. R.; Swallow S.; Gouverneur V.Fluorine in Medicinal Chemistry.Chem. Soc. Rev.2008, 37, 320–330. Google Scholar 4. Müller K.; Faeh C.; Diederich F.Fluorine in Pharmaceuticals: Looking Beyond Intuition.Science2007, 317, 1881–1886. Google Scholar 5. Ni C.; Hu J.The Unique Fluorine Effects in Organic Reactions: Recent Facts and Insights into Fluoroalkylations.Chem. Soc. Rev.2016, 45, 5441–5454. Google Scholar 6. Liu X.; Xu C.; Wang M.; Liu Q.Trifluoromethyltrimethylsilane: Nucleophilic Trifluoromethylation and Beyond.Chem. Rev.2015, 115, 683–730. Google Scholar 7. Ni C.; Hu M.; Hu J.Good Partnership Between Sulfur and Fluorine: Sulfur-Based Fluorination and Fluoroalkylation Reagents for Organic Synthesis.Chem. Rev.2015, 115, 765–825. Google Scholar 8. Xu X.-H.; Matsuzaki K.; Shibata N.Synthetic Methods for Compounds Having CF3-S Units on Carbon by Trifluoromethylation, Trifluoromethylthiolation, Triflylation, and Related Reactions.Chem. Rev.2015, 115, 731–764. Google Scholar 9. Liang of Fluorine and Google Scholar M.; S.; Diederich M.; Müller and in Google Scholar for and Drug Google Scholar of the and of Med. Google Scholar S.; is a Key of Med. Google Scholar of Fluorine in Medicinal Chemistry.J. Med. Google Scholar C.; S.; I.; S.; Fluorinated from to Google Scholar of on the of Google Scholar and Rev.2005, Google Scholar and Recent of the N-CF3 Google Scholar in the of N-CF3 to Google Scholar Feng Zhang on the Synthesis of the N-CF3 in Google Scholar K.; M.; K.; S.; K.; S.; of a of and of Med. Google Scholar I.; K.; M.; and of the Having a as a Med. Google Scholar S.; Fluoride Synthesis by of Med. Google Scholar Xu Wang Hu X.; Zhang R.; K.; Wang C.; Liu and of a and Med. Google Scholar M.; M.; by of Google Scholar K.; Synthesis of Amines through a from the Google Scholar of Fluoride from with Google Scholar Liang S.; Liu Synthesis of Amines and Amines with and Google Scholar K.; Tlili Synthesis of and Google Scholar Xu Liu M.; Fluoride Synthesis by of Google Scholar M.; Tlili A and Google Scholar of with in the of of Google Scholar and of with or Google Scholar M.; of and and with and and a Google Scholar M.; I.; I.; I.; of and Google Scholar M.; I.; 45, Google Scholar M.; Synthesis of by of Google Scholar M.; K.; K.; of and by Google Scholar K.;