English

• The selective decoration of organic molecules with fluorine has emerged as an effective strategy for drug design and discovery owing to its profound impact on the activity, metabolic stability, bioavailability, and other physicochemical parameters [1-4]. According to the statistics in 2022, approximately 30% of the top-selling small-molecule drugs bear at least one fluorine atom [5]. Among the diverse fluorinated structures, 2-trifluoromethylquinolines are valuable and prominent frameworks that are widely encountered in numerous bioactive pharmaceuticals (Fig. 1) [6-11]. For example, mefloquine is clinically used as an antimalarial drug. SCH-351591 is a potent and highly selective PDE4 inhibitor. T-10106 is investigated as an inflammation drug. In addition, 2-trifluoromethylquinolines can also be applied as fluorescence probes [12,13] and ligands [14,15] for metal-catalyzed reactions. Because of their biological activity and broad applications, the synthesis of 2-trifluoromethylquinolines has attracted considerable attention. Traditional synthesis of these scaffolds mainly relied on the Skraup reaction [16,17] and Friedlander reaction [18]. Recently, the straightforward cross-coupling or C—H functionalization of the quinoline core with CF₃ reagents has shown promise in obtaining 2-trifluoromethylquinolines (Scheme 1a, path Ⅰ) [19-25]. In addition, intermolecular [5 + 1] [26,27], [4 + 2] [28,29], and [3 + 3] [30] annulations (Scheme 1a, paths Ⅱ-Ⅳ) have been described for the formation of 2-trifluoromethylquinolines. In spite of notable achievements, these transformations have been largely limited to construct 4-unsubstituted, 4-alkyl- and 4-aryl-substituted 2-trifluoromethylquinolines. To the best of our knowledge, catalytic or non-catalytic one-step methodology leading directly to 2-CF₃-quinoline-4-carboxamides have not been reported. Their preparation remains a formidable challenge owing to the requirement for multistep sequences to form the amide bond in the presence of condensing agents [31]. As such, developing a practical, green, transition-metal-free, and innovative approach to address this issue by using readily available starting materials is highly desirable due to the great influence of carboxamide substituents at the 4-position of quinoline ring on its biological activities [31].

  Figure 1

  

  Figure 1.  Selected drugs containing 2-trifluoromethylquinoline core.

  DownLoad: Full-Size Img PowerPoint

  Scheme 1

  

  Scheme 1.  Research background and our present reaction design.

  DownLoad: Full-Size Img PowerPoint

  Isatins are fascinating and versatile building blocks that are extensively utilized to synthesize various biologically active heterocyclic compounds and medicinal alkaloids [32-34]. In particular, isatins have been explored for the construction of quinoline-4-carboxylic acids by the condensation with carbonyl compounds. This reaction is generally known in organic chemistry as the Pfitzinger reaction [35-37]. However, its procedure generally requires a large excess of KOH or NaOH and acid in a two-step process, and is unable to form 2-CF₃-quinoline-4-carboxamides. Therefore, despite its synthetic utility and its discovery over 138 years ago (first recorded discovery by Pfitzinger in 1886 [35]), no truly catalytic approach has hitherto been reported for the Pfitzinger reaction. The key challenge arises from not only the high resonance stabilization of amide group [38-41] but also the poor leaving ability of -OH anion, eventually leading to inability to achieve a catalytic cycling process. In fact, a literature survey indicated that the breakage of the amide C—N bond in a catalytic manner was still rare. Current methods for the direct activation of the amide C—N bond employ transition metal complexes (via oxidative addition) [42-49], squaramides (via dual hydrogen-bonding interaction) [50,51] or triazolium N-heterocyclic carbenes (via nucleophilic substitution at the amide carbon) as catalysts (Scheme 1b) [52]. Inspired by these transformations, we hypothesized that a tertiary amine catalyst with high nucleophility and good leaving ability might be employed to address the aforementioned issue. As illustrated in Scheme 1d, tertiary amine might initiate the amide C—N bond cleavage through direct nucleophilic substitution at the C2 position of isatin, forming the zwitterionic intermediate Int-Ⅰ. Intramolecular cyclization of the 2-CF₃–1-azaallyl anion with the carbonyl group led to the formation of dihydroquinoline intermediate Int-Ⅱ, which was then converted into the key quinoline-based acyl-ammonium Int-Ⅲ by loss of a molecule of water. Intermediate Int-Ⅲ finally underwent nucleophilic acyl substitution with amines to produce the desired 2-CF₃-quinoline-4-carboxamides and regenerate the tertiary amine catalyst.

  Although the proposal hypothesized above is deceivingly simple, the following difficulties should not be ignored in developing such a catalytic Pfitzinger-type reaction. First, the amide C—N bond cleavage has a high activation energy [38-41] and its selective scission in the presence of amines is challenging because the reaction of isatins with amines conventionally leads to imines (Scheme 1c). Second, intermolecular reaction of zwitterionic intermediate Int-Ⅰ with isatins might strongly compete with the desired intramolecular cyclization (Scheme 1d). As a result, to date, such a methodology has not been realized although such research could not only benefit the understanding of catalytic mechanism but maximize the atom economy and application range of Pfitzinger reaction.

  To test the feasibility of our hypothesis depicted in Scheme 1d, N-[(α-trifluoromethyl)vinyl]isatin 1a and commercially available 4-chloroaniline 2a were selected as the model substrates to establish optimized reaction conditions (Table 1). The initial screening of tertiary amine catalysts in EtOAc at 60 ℃ revealed that triethylamine (Et₃ N), 1,2,2,6,6-pentamethylpiperidine (PMP), and 1,4-diaza bicyclo[2.2.2]octane (DABCO) were ineffective in this reaction because no desired 2-CF₃-quinoline-4-carboxamide 3 was detected after 24 h (Table 1, entries 1–3). However, when using superbases such as 1,8-diazabicyclo[5.4.0] undec–7-ene (DBU), 1,5,7-triazabicyclo[4.4.0]dec–5-ene (TBD), and 2–tert–butyl–1,1,3,3-tetramethylguanidine (BTMG), the reaction proceeded smoothly and delivered the desired product 3 in 40%, 35%, and 63% yields, respectively (Table 1, entries 4–6). These results demonstrated that the strength and nucleophilicity of tertiary amines were essential for the success of this reaction. Encouraged by this result, a series of solvents were examined by employing BTMG as the catalyst. Unfortunately, solvents such as CH₂Cl₂, ClCH₂CH₂Cl, THF, CH₃CN, or toluene gave reduced yields (entries 7–11). Subsequent investigation of the reaction conditions showed a lower yield (51%) when lowering the BTMG loading to 10 mol% (entry 12), but increasing the DABCO content (40 mol%) also led to a similar result (entry 13). Moreover, examination of the additive effect showed that 3 Å MS was optimal (entries 15–17). It should be noted that no reaction occurred without BTMG (entry 18). This result showed that the nucleophilic Lewis base catalyst played a crucial role in this reaction. Finally, the best outcome was obtained by running the reaction for 24 h under the catalysis of BTMG (30 mol%) at 60 ℃ in EtOAc with 3 Å MS as the additive, delivering product 3 in 71% yield (entry 15).

  Table 1

  

  Table 1.  Optimization of the reaction conditions.^a

  DownLoad: CSV

  Entry	Cat (X mol%)	Solvent	Additive	Yield (%)b
  1	Et3 N (30)	EtOAc	–	NR
  2	DABCO (30)	EtOAc	–	NR
  3	PMP (30)	EtOAc	–	NR
  4	DBU (30)	EtOAc	–	40
  5	TBD (30)	EtOAc	–	35
  6	BTMG (30)	EtOAc	–	63
  7	BTMG (30)	CH2Cl2	–	42
  8	BTMG (30)	ClCH2CH2Cl	–	48
  9	BTMG (30)	THF	–	30
  10	BTMG (30)	CH3CN	–	24
  11	BTMG (30)	Toluene	–	53
  12	BTMG (10)	EtOAc	–	51
  13	BTMG (20)	EtOAc	–	58
  14	BTMG (40)	EtOAc	–	64
  15	BTMG (30)	EtOAc	3 Å MS	71
  16	BTMG (30)	EtOAc	4 Å MS	68
  17	BTMG (30)	EtOAc	5 Å MS	67
  18	–	EtOAc	3 Å MS	NR
  a Standard reaction conditions: 1a (0.1 mmol), amine 2a (0.3 mmol), BTMG (10–40 mol%) in EtOAc, 60 ℃, 24 h. PMP: 1,2,2,6,6-pentamethylpiperidine; DABCO: 1,4-diaza bicyclo[2.2.2]octane; DBU: 1,8-diazabicyclo[5.4.0]undec–7-ene; TBD: 1,5,7-triazabicyclo[4.4.0]dec–5-ene; BTMG: 2–tert–butyl–1,1,3,3-tetramethyl-guanidine; NR = No Reaction. b Isolated yields.

  Having optimal reaction conditions in hand, we first examined the substrate scope and limitations of this reaction by utilizing various amines 2 and N-[(α-trifluoromethyl)vinyl]isatin 1a. As revealed in Scheme 2, aromatic amines reacted smoothly with isatin 1a to deliver a range of desired 2-CF₃-quinoline-4-carboxamides 3–18 in 56%−84% yields. Various substituents at different positions of the phenyl ring were tolerated in this transformation including chloro (3, 9, 13, 15), fluoro (4), bromo (5), iodo (10), nitro (6, 12), trifluoromethyl (7), and acetyl (8) groups, as well as electron-donating groups such as methyl (16) and methoxy (17 and 18) groups. As expected, a series of heteroaromatic amines could all be applied, affording the products 19–23 in 47%−84% yields. The structure of 23 was further unambiguously elucidated by X-ray crystallography (CCDC: 2278432).

  Scheme 2

  

  Scheme 2.  Substrate scope with respect to the synthesis of 2-CF₃, 2-CF₂H, and 2-CO₂Me-quinoline-4-carboxamides. Reaction conditions: 1 (0.10 mmol), 2 (0.30 mmol), 3 Å MS (50 mg), and BTMG (0.03 mmol) in EtOAc (0.8 mL) were stirred at 60 ℃ for 24–48 h. Isolated yields. ^a 50 mol% of BTMG was used.

  DownLoad: Full-Size Img PowerPoint

  Besides different aromatic and heteroaromatic amines, simple primary and secondary aliphatic amines were also suitable reaction partners under the current conditions, furnishing the corresponding products 24–30 in moderate to good yields. Notably, aqueous ammonia was also compatible with this process and gave the product 31 in 52% yield. Then, a series of isatin derivatives bearing substituents at different positions of the phenyl ring were further tested and these substrates all provided the products 32–35 with high yields. Furthermore, 2-CF₂H-quinoline-4-carboxamide 36 and 2-CO₂Me-quinoline-4-carboxamide 37, which are difficult to prepare using the available methods, could also be obtained in moderate yields. These results demonstrated the excellent functional group tolerance and the generality of this methodology.

  The facile construction of 2-CF₃, 2-CF₂H, and 2-CO₂Me-quinoline-4-carboxamides using the current strategy prompted us to further investigate the substrate scope. To our delight, as shown in Scheme 3, different substituted primary, secondary, propargylic, and allylic alcohols were found to be adaptable nucleophilic partners for this process, and all combinations of various alcohols and N-[(α-trifluoromethyl)vinyl]isatin 1a worked well to give the desired 2-CF₃-quinoline-4-carboxylic esters 39–48 in moderate to good yields except for t-BuOH, which afforded the target product 42 with a lower yield (13%) possibly due to the large steric hindrance of t-Bu group. Complex alcohols derived from natural products or chiral catalysts could also effectively participate in the reactions, furnishing the products 49–51 in 45%−64% yields. The synthetic potential of this approach was further highlighted by the direct synthesis of deuterated product 52 by using CD₃OD as the nucleophile. Further evaluation of amino alcohols bearing two nucleophilic sites (NH₂ and OH) indicated that the reaction exclusively occurred on the NH₂ site, probably because the nucleophilicity of NH₂ is greater than OH. In addition, N-[(α-difluoromethyl)vinyl]isatin and N-[(α-carbomethoxy)vinyl]isatin also proved to be compatible with this reaction, converting to the target products 54–55 with moderate yields. However, attempts to extend the substrate scope to phenols led to the decomposition of N-[(α-trifluoromethyl)vinyl]isatin. In addition, other nucleophiles, such as carboxylic acids and thiols are also tested. But unfortunately, the reactions are complex under standard conditions and no desired products are detected by GM-MS analysis of the reaction mixture, which constitutes the major limitation of this methodology. This is likely because carboxylic acids can react with BTMG and thus deactivate the catalyst. In the case of thiols, the acyl-ammonium intermediate (Int-Ⅱ, Scheme 1) is considered to be a hard Lewis acid, which prefers to be attacked by the harder nucleophiles such as alcohols and amines rather than thiols according to hard-soft acid-base theory.

  Scheme 3

  

  Scheme 3.  Substrate scope with respect to the synthesis of 2-CF₃, 2-CF₂H, and 2-CO₂Me-quinoline-4-carboxylic esters. Reaction conditions: 1 (0.10 mmol), 2 (0.30 mmol) and BTMG (0.03 mmol) in EtOAc (0.8 mL) were stirred at 60 ℃ for 24–48 h under air atmosphere. Isolated yields.

  DownLoad: Full-Size Img PowerPoint

  Having developed an efficient and general approach, we subsequently turned our attention to demonstrating its synthetic utility. As illustrated in Scheme 4, treating compound 3 with BH₃·THF and Lawesson’s reagent led to the formation of 2-trifluoromethylquinolines 57 and 58 in 70% and 84% yields, respectively. In addition, some selective reduction reactions could be conducted with compound 39. For example, the CO₂Me group of compound 39 could be easily reduced using DIBAL-H in the presence of THF at 0 ℃ to form compound 59, which can further react with ketenimine 60 to afford product 61 in 61% overall yield. Whereas both the CO₂Me group and the pyridine ring of quinoline 39 were reduced by simply replacing the reducing agent DIBAL-H with LiAlH₄ to provide the tetrahydroquinoline 62 as a single diastereomer in 51% yield. Encouraged by these results, the late-stage modification of biologically-active compounds was performed. Delightedly, the amine drugs such as norquetiapine, fluoxetine, and mexiletine all reacted successfully with N-[(α-trifluoromethyl)vinyl]isatin 1a to deliver the corresponding products 63–66 in 37%−61% yields. Likewise, perphenazine bearing a terminal alcohol (-OH) group could be efficiently converted to the desired 2-trifluoromethyl quinoline 66. Furthermore, 3,5-di(9H-carbazol-9-yl)aniline 67 and 3,5-bis(9,9-dimethylacridin-10(9H)-yl)aniline 70, which were widely used as the donor units in the field of optoelectronic material, were found to be accommodated to provide products 68 and 69 in 51% and 46% yield, respectively, showcasing the utility and practicality of this strategy.

  Scheme 4

  

  Scheme 4.  Synthetic applications of this approach.

  DownLoad: Full-Size Img PowerPoint

  In addition, to discover the possible bioactivities of the products, we investigated the inhibitory activity against PTP1B of selected 2-CF₃-quinoline-4-carboxamides and 4-carboxylic esters by testing their biochemical function on DiFUMP assay. As summarized in Fig. 2, several products display strong inhibitory activity against PTP1B, with IC₅₀ values ranging of 2.17–27.41 µmol/L. It was discovered product 6 displays the strongest inhibition against PTP1B, with IC₅₀ value of 2.17 µmol/L. PTP1B inhibitors are known to have potential in gaining anti-cancer immunity. As a result, biological evaluation demonstrates that these 2-CF₃-quinoline-4-carboxamides and 2-CF₃-quinoline-4-carboxylic esters should have potential applications in medicinal chemistry.

  Figure 2

  

  Figure 2.  Some selected 2-CF₃-quinoline-4-carboxamides and carboxylic esters with promising inhibitory activity against PTP1B.

  DownLoad: Full-Size Img PowerPoint

  With facile access to highly substituted and extended π-conjugated 2-trifluoromethyl quinolines, which might be of great interest for their physical and material properties, we therefore further performed an investigation of the detailed photophysical and device characterizations of compounds 68 and 69 bearing an unsymmetric donor-acceptor (D-A) structure. To begin with, the frontier molecular orbitals (FMOs) of these two compounds were theoretically evaluated by density functional theory (DFT) calculations, which are depicted in Fig. 3a for the highest occupied molecular orbital (HOMO) and the lowest unoccupied molecular orbital (LUMO) distributions. It is noticed that HOMO and LUMO electrons of 68 and 69 are mainly positioned on donor and accepter units of them, while just a little extent of wavefunction overlap occurs. Accordingly, it is anticipated that the emission of them is dominated by charge transfer (CT) excited state in nature [53]. According to the calculated HOMO and LUMO levels, the corresponding bandgaps E_g(cal.) are determined to be 2.98 eV for 68 and 2.79 eV for 69, respectively. Therefore, 68 displays the slightly wider bandgap as compared to that of 69. Such estimation is basically consistent with the results of their photophysical experiments. As depicted in Fig. 3b, the corresponding absorption bandgap Eg(exp.) of 68 and 69 (derived from absorption onset) is determined to be 3.51 eV and 3.34 eV, respectively. For the photoluminescent (PL) emission features of 68 and 69, the PL emission peaks (λ_PL) are located at 526 and 596 nm, along with a distinct wide full width at half maximum (FWHM) values of 132 nm and 155 nm, respectively. Significantly, both of them show a broad and featureless PL spectrum. It is sharply different from common fluorophores, such as perylene or rhodamine, which shows a PL spectrum characteristic of much narrower FWHM and a significant refined structure due to locally excited (LE) state in nature [54]. In this regard, PL emission from 68 to 69 should be CT state in practice, which is well correlated with the abovementioned DFT results. The PL transient experiments of 68 and 69 were further performed (Fig. S1 and Table S1 in Supporting information), which leads to an average exciton lifetime (τ_ave) of 18.5 and 16.9 ns, respectively. These measured τ_ave values are higher than that of common fluorophores as disclosed [54]. Compared the ground-state transition from S₀ → S₁, i.e., the first absorption peak, the first excite-state transition from S₁ → S₀, i.e., PL peak, displays an obvious red-shift for 68 and 69, corresponding to a stokes shift of 0.58 cm⁻¹ and 0.35 cm⁻¹, respectively. Such phenomena indicates that a distinct polarization of PL is involved owing to a large transition moment of these CT state under excitation. The corresponding PL quantum efficiencies (φ_PL) of 68 and 69 were measured to be 5.36% and 5.96% in the lightly 3 wt% doped film samples in 1,3-di(9H-carbazol-9-yl)benzene (mCP) matrix.

  Figure 3

  

  Figure 3.  HOMO/LUMO distributions (a), absorption and PL spectra (b) of 68 and 69, respectively.

  DownLoad: Full-Size Img PowerPoint

  Subsequently, monochromatic solid-state organic light-emitting diodes (OLEDs) [55] were fabricated by using those two compounds as the fluorophores (see Supporting information for the more details about device fabrications), in which the device structures were indium tin oxide (ITO)/poly(ethylenedioxythiophene): poly(styrene sulfonate) (PEDOT: PSS) (ca. 50 nm)/mCP: 68 or 69 (x = 3, 5, 7, 10, 15, 20 wt%) (ca. 30 nm)/bis[2-(diphenylphosphino)phenyl]ether oxide (DPEPO) (10 nm)/1,3,5-tri(m-pyrid-3-yl-phenyl)-benzene (TmPyPB) (50 nm)/LiF (1 nm)/Al (100 nm), in which ITO, PEDOT: PSS, DPEPO, TmPyPB, LiF and Al layers were used as transparent anode, hole injection layer(HIL), exciton blocking layer(EBL), electron transport layer(ETL), electron injection layer(EIL) and reflective cathode, respectively.

  The emissive layer (EML) was composed of a wide bandgap mCP as the host matrix and 68 or 69 as the light-emissive fluorophores. The corresponding device structures and related EL characteristics were shown in Fig. 4. To optimize EL performance of 68 and 69, the doping concentrations of EMLs were tailored from 3 wt% to 20 wt%, which indicates that the optimized EL performance was achieved at the lowest doping level of 3 wt%, corresponding to a maximum external quantum efficiency (EQE_max.) of 1.02% and 1.93% for 68 and 69, respectively, along with a Commission Internationale de l’Eclairage (CIE) chromaticity coordinate of (0.22, 0.29) and (0.35, 0.45) and λEL of 473 nm and 534 nm. Either of these OLEDs shows stable and bright EL, in which EL emissions were purely from 68 or 69 themselves. Noticeably, the FWHM of 69 device is as wide as 167 nm, which stems from its intense CT excited state as discussed aforementioned. Such wide EL spectrum from solely one compound is rarely reported. As increasing the doping concentrations, energy transfer process from mCP to 68 and 69 is gradually enhanced (Fig. S2 in Supporting information). However, EQE performance was progressively lowered from 1.02% to 0.31% for 68 and 1.93% to 0.27% for 69, along with gradual red-shifted EL spectra (Figs. S3-S6, Table S2 and S3 in Supporting information). Such decline in EQE performance should be ascribed to aggregation caused quenching (ACQ) issue of these fluorophores.

  Figure 4

  

  Figure 4.  (a) device structures of single-color of OLEDs using 68 or 69, and the corresponding (b) current density (J) - luminance (L) -voltage (V) curves, (c) external quantum efficiency (EQE)-J curves, (d) EL spectra of devices at a voltage of 7 V.

  DownLoad: Full-Size Img PowerPoint

  Considering the wide FWHM of EL spectra of 69, it is highly expected that white-color OLEDs can be prepared by adding extra blue emitters into the EML of 69-based OLEDs [56]. To confirm such speculation, the corresponding attempts were accomplished by constructing EML composed of mCP: iridium(Ⅲ) [bis(4,6-difuorophenyl)-pyridinato-N, C2]-picolinate (FIrpic): 69, in which FIrpic is a well-known blue phosphor emitter (Fig. 5a) [57]. With the purpose to tailor EL spectra of those white-color OLEDs, the different doping concentrations were used, in which the concentration of FIrpic was fixed at 20 wt% and the concentration of 69 was monotonically changed from 20 wt% to 50 wt%. The corresponding EL characteristics were depicted in Figs. 5b-f, Fig. S7 and Table S4 (Supporting information). It is clearly shown that complementary two-color-type white OLEDs can be realized, in which blue and orange color come from FIrpic and 69, respectively. As increasing the doping concentration of 69, the relative contribution from orange component is gradually enhanced. Such results are reasonable since in this way more exciton can be formed on 69 through the direct exciton generation pathway. Very meaningfully, those white-color OLEDs show satisfactory white-color stability. As shown in Fig. 5f, the CIE coordinates are just slightly changed from (0.41, 0.46) at 5 V to (0.42, 0.46) at 10 V, which well meets the requirements of color-stability levels for general lighting applications [58-60]. The corresponding lighting image is shown as an inset in Fig. 5f, which demonstrates a typical white illumination characteristic with a high color rendering index (CRI) of 76. Accordingly, these results are combined to illustrate that 68 and 69 are qualified fluorochromes, which is capable of being used in solid-state OLED displays or lighting fields. After further optimization of their chemical structures towards the higher φ_PL, much higher performance OLEDs could be anticipated.

  Figure 5

  

  Figure 5.  Device structure (a) and EL characteristics of white-color OLEDs by using the EMLs of mCP: FIrpic (20 wt%): 69 (x wt%) (b-d), the corresponding white EL spectra at 7 V (e) and one respective bias-dependent EL spectra of device using 69 at 30 wt% (f).

  DownLoad: Full-Size Img PowerPoint

  To shed light on the mechanistic details, a set of exploratory experiments between N-[(α-trifluoromethyl)vinyl]isatins 1 and several superbases were conducted. We initially investigated the reaction of N-[(α-trifluoromethyl)vinyl]isatin 1a and BTMG in the presence of commercially available EtOAc (containing 5% H₂O), which unexpectedly afforded 2-CF₃-quinoline-4-carboxylic acid 71 in 51% yield (Scheme 5a). Interestingly, when substrates 1a and 1b were treated with DBU (1.0 equiv.) under the same reaction conditions, 2-CF₃-quinoline-4-carboxamides 72 and 73 bearing an unusual 7-membered lactam unit were obtained in 72% and 67% yield, respectively (Scheme 5b). Similarly, with TBD as the reaction partner, substrates 1a and 1b could also be converted into tetrahydro-2(1H)-pyrimidinone-tethered 2-CF₃-quinoline-4-carboxamides 74 and 75 in 68% and 62% yield, respectively (Scheme 5c). These observations indicated that the quinoline-based acyl-ammonium Int-Ⅲ shown in Scheme 2d should be the key intermediate involved in this transformation. In the case of BTMG, quinoline-based acyl-ammonium species underwent nucleophilic acyl substitution with H₂O to form 71. Whereas in the cases of DBU and TBD, acyl-ammonium species underwent ring-opening reaction to form 72–75. To our surprise, the treatment of substrate 1a with 1,3-diarylguanidines 76a and 76b in EtOAc at 50 ℃ resulted in the formation of the acylguanidine-tethered 2-trifluoromethyl quinolines 77 and 78. This result provided further support for our hypothesis that the transient acyl-ammonium species was the key intermediate. The molecular structures of 72 and 78 were confirmed by NMR spectroscopy and X-ray analysis (72: CCDC 2321247; 78: CCDC 2321250). Moreover, it is worth noting that DBU and TBD are traditionally known as highly stable organocatalysts in a range of applications due to their ability to delocalize charge over the nitrogen atoms. As a result, ring openings of DBU and TBD have been scarcely reported in the literature [61-64]. In particular, their participation in Pfitzinger reaction as nucleophiles rather than catalysts toward quinoline synthesis accompanied by the ring-opening process is rather challenging and has never been realized. This finding thus represents an important breakthrough in Pfitzinger reaction and superbase chemistry.

  Scheme 5

  

  Scheme 5.  Reactions of N-[(α-trifluoromethyl)vinyl]isatins with organic superbases.

  DownLoad: Full-Size Img PowerPoint

  Based on the above experimental results and the literature pecedents [37,52], we proposed possible reaction pathways for the formation of products 3 and 72. As depicted in Scheme 6a, BTMG could act as a strong nucleophilic Lewis base catalyst to attack the C2 position of isatin 1a to generate the corresponding zwitterionic intermediate Int-Ⅰ and then undergo the intramolecular cyclization and dehydration process to give the key quinoline-based acyl-ammonium species Int-Ⅴ. Finally, the nucleophilic acyl substitution between Int-Ⅴ and 4-chloroaniline 2a would occur to deliver the desired 2-CF₃-quinoline-4-carboxamide 3 after eliminating the catalyst BTMG. However, in the absence of amine 2a, the reaction of isatin 1a and DBU in a 1:1 molar ratio would initially produce zwitterionic intermediate Int-Ⅰ’, which could be converted to acyl-ammonium intermediate Int-Ⅴ’ through a similar process shown in Scheme 6a. The electrophilic nature of the resulting iminium ion carbon in Int-Ⅴ’ was then trapped by a molecule of H₂O present in the reaction mixture to form the tertiary amino alcohol Int-Ⅵ’. Finally, a base-triggered ring-opening process of the cyclic amidine ring motif generated the functionalized quinoline 72.

  Scheme 6

  

  Scheme 6.  Proposed reaction mechanism for the formation of products 3 and 72.

  DownLoad: Full-Size Img PowerPoint

  In conclusion, we have firstly developed a BTMG-catalyzed system which is capable of activating amide C—N bonds of N-[(α-trifluoromethyl)vinyl]isatins for Pfitzinger-type reactions with amines and/or alcohols to produce a wide range of valuable 2-CF₃-quinoline-4-carboxamides and/or 2-CF₃-quinoline-4-carboxylic esters in a single step. In addition, this methodology enables the facile incorporation of other functional groups, i.e., CF₂H and CO₂Me, into the C2 position of quinoline skeleton. More importantly, reactions of N-[(α-trifluoromethyl)vinyl]isatins with superbases in a 1:1 molar ratio to form 2-trifluoromethyl quinolines containing lactam, tetrahydro-2(1H)-pyrimidinone, or guanidine moiety in moderate to good yields were also achieved, which represents a rare example of ring-opening reactions involving DBU and TBD. The synthetic utility of this protocol was demonstrated by the green and environmentally benign conditions, broad substrate scope, the late-stage modification of commercial drugs, and the diverse derivatization of quinoline framework. Moreover, preliminary studies on biological activities and photophysical properties of the obtained products revealed that several compounds display strong inhibitory activity against PTP1B, while some extended π-conjugated compounds could serve as the promising optoelectronic materials towards solid-state lighting or display devices. Therefore, this green and streamlined synthetic methodology might find broad applications in the fields of organic synthesis, medicinal chemistry, and materials science. Efforts to develop new reactions of N-[(α-trifluoromethyl)vinyl]isatins are underway in our laboratory.

  Declaration of competing interest

  The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

  CRediT authorship contribution statement

  Cong-Bin Ji: Writing – original draft, Investigation. Ding-Xiong Xie: Investigation. Mei Chen: Investigation. Ye-Ying Lan: Investigation. Bao-Hua Zhang: Project administration, Formal analysis. Ji-Ying Yang: Investigation. Zheng-Hui Kang: Writing – review & editing, Funding acquisition. Shu-Jie Chen: Investigation. Yu-Wei Zhang: Project administration, Formal analysis. Yun-Lin Liu: Writing – review & editing, Writing – original draft, Investigation, Formal analysis, Data curation, Conceptualization.

  Acknowledgments

  The National Natural Science Foundation of China (Nos. 22171056, 22122402, 21801050), Outstanding Youth Project of Guangdong Natural Science Foundation (Nos. 2024B1515020036, 2021B1515020048), Guangdong Natural Science Foundation (Nos. 2023A1515011313, 2021A1515010510, 2024A1515030037), Tertiary Education Scientific Research Project of Guangzhou Municipal Education Bureau (No. 202235305), the Open Fund from Key Laboratory of Organofluorine Chemistry, and Shanghai Engineering Research Center of Molecular Therapeutics and New Drug Development are gratefully acknowledged for financial support.

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2024.110598.

  
  
• 1. [1]

     S. Purser, P.R. Moore, S. Swallow, V. Gouverneur, et al., Chem. Soc. Rev. 37 (2008) 320–330.

  2. [2]

     N.A. Meanwell, J. Med. Chem. 61 (2018) 5822–5880. doi: 10.1021/acs.jmedchem.7b01788

  3. [3]

     J. Han, L. Kiss, H. Mei, et al., Chem. Rev. 121 (2021) 4678–4742. doi: 10.1021/acs.chemrev.0c01263

  4. [4]

     D.S. Deng, S.Q. Tang, Y.T. Yuan, et al., Chin. Chem. Lett. 35 (2024) 109417.

  5. [5]

     N.A. McGrath, M. Brichacek, J.T. Njardarson, J. Chem. Educ. 87 (2010) 1348–1349. doi: 10.1021/ed1003806

  6. [6]

     J. Dade, O. Provot, H. Moskowitz, J. Mayrargue, E. Prina, Chem. Pharm. Bull. 49 (2001) 480–483.

  7. [7]

     Y. Zhou, J. Wang, Z. Gu, et al., Chem. Rev. 116 (2016) 422–518. doi: 10.1021/acs.chemrev.5b00392

  8. [8]

     V. Gouverneur, K. Seppelt, Chem. Rev. 115 (2015) 563–565. doi: 10.1021/cr500686k

  9. [9]

     J. Mao, H. Yuan, Y. Wang, et al., J. Med. Chem. 52 (2009) 6966–6978. doi: 10.1021/jm900340a

  10. [10]

      M. Billah, N. Cooper, F. Cuss, et al., Bioorg. Med. Chem. Lett. 12 (2002) 1621–1623.

  11. [11]

      R.E. Lutz, C.J. Ohnmacht, A.R. Patel, J. Med. Chem. 14 (1971) 926–928. doi: 10.1021/jm00292a008

  12. [12]

      P. Jia, D. Liu, Z. Zhuang, et al., Sens. Actuators B: Chem. 320 (2020) 128583.

  13. [13]

      H. Zhu, C. Liang, X. Cai, et al., Anal. Chem. 92 (2020) 1883–1889. doi: 10.1021/acs.analchem.9b04009

  14. [14]

      P. Evans, P. Hogg, R. Grigg, et al., Tetrahedron 61 (2005) 9696–9704.

  15. [15]

      D.M. Pearson, N.R. Conley, R.M. Waymouth, Organometallics 30 (2011) 1445–1453. doi: 10.1021/om101037k

  16. [16]

      A. Weyesa, E. Mulugeta, RSC Adv. 10 (2020) 20784–20793. doi: 10.1039/d0ra03763j

  17. [17]

      G.A. Ramann, B.J. Cowen, Molecules 21 (2016) 986. doi: 10.3390/molecules21080986

  18. [18]

      M. Fallah-Mehrjardi, Mini-Rev. Org. Chem. 14 (2017) 187–196.

  19. [19]

      M. Oishi, H. Kondo, H. Amii, Chem. Commun. 45 (2009) 1909–1911. doi: 10.1039/b823249k

  20. [20]

      A. Lishchynskyi, M.A. Novikov, E. Martin, et al., J. Org. Chem. 78 (2013) 11126–11146. doi: 10.1021/jo401423h

  21. [21]

      A. Lishchynskyi, V.V. Grushin, J. Am. Chem. Soc. 135 (2013) 12584–12587. doi: 10.1021/ja407017j

  22. [22]

      T. Nishida, H. Ida, Y. Kuninobu, M. Kanai, Nat. Commun. 5 (2014) 3387.

  23. [23]

      T.T. Tung, S.B. Christensen, J. Nielsen, Chem. Eur. J. 23 (2017) 18125–18128. doi: 10.1002/chem.201704261

  24. [24]

      C. Lu, Y. Gu, J. Wu, Y. Gu, Q. Shen, Chem. Sci. 8 (2017) 4848–4852.

  25. [25]

      T. Shirai, M. Kanai, Y. Kuninobu, Org. Lett. 20 (2018) 1593–1596. doi: 10.1021/acs.orglett.8b00339

  26. [26]

      H. Xu, F. Yu, R. Huang, et al., Org. Chem. Front. 7 (2020) 3368–3373. doi: 10.1039/d0qo00709a

  27. [27]

      J. Nan, Y. Hu, P. Chen, Y. Ma, Org. Lett. 21 (2019) 1984–1988. doi: 10.1021/acs.orglett.9b00039

  28. [28]

      X. Dong, Y. Xu, J. Liu, et al., Chem. Eur. J. 19 (2013) 16928–16933. doi: 10.1002/chem.201303149

  29. [29]

      H. Amii, Y. Kishikawa, K. Uneyama, Org. Lett. 3 (2001) 1109–1112.

  30. [30]

      H. Huang, H. Wang, C. Gong, et al., Org. Chem. Front. 9 (2022) 413–419. doi: 10.1039/d1qo01478a

  31. [31]

      H. Siebeneicher, A. Cleve, H. Rehwinkel, et al., ChemMedChem 11 (2016) 2261–2271. doi: 10.1002/cmdc.201600276

  32. [32]

      F. Zhou, Y.L. Liu, J. Zhou, Adv. Syn. Catal. 352 (2010) 1381–1407. doi: 10.1002/adsc.201000161

  33. [33]

      G.S. Singh, Z.Y. Desta, Chem. Rev. 112 (2012) 6104–6155. doi: 10.1021/cr300135y

  34. [34]

      M.V. Mijangos, Y.A. Amador-Sánchez, L.D. Miranda, Eur. J. Org. Chem. 2021 (2021) 637–647. doi: 10.1002/ejoc.202001455

  35. [35]

      W. Pfitzinger, J. Prakt. Chem. 33 (1886) 100. doi: 10.1002/prac.18850330110

  36. [36]

      P.D. Popp, Adv. Heterocycl. Chem. 18 (1975) 1–58.

  37. [37]

      M.G.A. Shvekhgeimer, Chem. Heterocycl. Compd. 40 (2004) 257–294.

  38. [38]

      J.I. Mujika, J.M. Matxain, L.A. Eriksson, X. Lopez, Chem. Eur. J. 12 (2006) 7215–7224. doi: 10.1002/chem.200600052

  39. [39]

      G.C. Li, S.Y. Ma, M. Szostak, Trends Chem. 2 (2020) 914–928.

  40. [40]

      C.R. Kemnitz, M.J. Loewen, J. Am. Chem. Soc. 129 (2007) 2521–2528. doi: 10.1021/ja0663024

  41. [41]

      E. Kovács, B. Rózsa, A. Csomos, I.G. Csizmadia, Z. Mucsi, Molecules 23 (2018) 2859. doi: 10.3390/molecules23112859

  42. [42]

      J.E. Dander, N.K. Garg, ACS Catal. 7 (2017) 1413–1423. doi: 10.1021/acscatal.6b03277

  43. [43]

      T.B. Boit, A.S. Bulger, J.E. Dander, N.K. Garg, ACS Catal. 10 (2020) 12109–12126. doi: 10.1021/acscatal.0c03334

  44. [44]

      L. Hie, N.F.F. Nathel, T.K. Shah, et al., Nature 524 (2015) 79–83. doi: 10.1038/nature14615

  45. [45]

      G. Li, M. Szostak, Chem. Rec. 20 (2020) 649–659. doi: 10.1002/tcr.201900072

  46. [46]

      J. Hu, Y. Zhao, J. Liu, Y. Zhang, Z. Shi, Angew. Chem. Int. Ed. 55 (2016) 8718–8722. doi: 10.1002/anie.201603068

  47. [47]

      J. Hu, M. Wang, X. Pu, Z. Shi, Nat. Commun. 8 (2017) 14993.

  48. [48]

      H. Yue, L. Guo, S.C. Lee, X. Liu, M. Rueping, Angew. Chem. Int. Ed. 56 (2017) 3972–3976. doi: 10.1002/anie.201612624

  49. [49]

      A. Dey, S. Sasmal, K. Seth, G.K. Lahiri, D. Maiti, ACS Catal. 7 (2017) 433–437. doi: 10.1021/acscatal.6b03040

  50. [50]

      G. Wang, Q. Shi, W. Hu, et al., Nat. Commun. 11 (2020) 946. doi: 10.21037/atm-20-5438

  51. [51]

      W.T. Wang, S. Zhang, L.F. Tao, et al., Chem. Commun. 58 (2022) 6292–6295. doi: 10.1039/d2cc01625g

  52. [52]

      Y. Cai, Y. Zhao, K. Tang, et al., Nat. Commun. 15 (2024) 496.

  53. [53]

      S. Wang, H. Zhang, B. Zhang, Z. Xie, W.Y. Wong, Mater. Sci. Eng. R: Rep. 140 (2020) 100547.

  54. [54]

      J.R. Lakowice, Principles of Fluorescent Spectroscopy (third edition), Springer, 2006.

  55. [55]

      T.H. Han, M.R. Choi, C.W. Jeon, et al., Sci. Adv. 2 (2016) e1601428.

  56. [56]

      L. Chen, Y.F. Chang, S. Shi, S. Wang, L. Wang, Mater. Horizons. 9 (2022) 1299–1308. doi: 10.1039/d1mh02060a

  57. [57]

      H. Wu, G. Zhou, J. Zou, et al., Adv. Mater. 21 (2009) 4181–4184. doi: 10.1002/adma.200900638

  58. [58]

      S. Wang, B. Zhang, X. Wang, et al., Adv. Opt. Mater. 3 (2015) 1349–1354. doi: 10.1002/adom.201500175

  59. [59]

      M.C. Gather, A. Köhnen, K. Meerholz, Adv. Mater. 23 (2011) 233–248. doi: 10.1002/adma.201002636

  60. [60]

      M.C. Gather, R. Alle, H. Becker, K. Meerholz, Adv. Mater. 19 (2007) 4460–4465. doi: 10.1002/adma.200701673

  61. [61]

      Z. Luo, J. Chen, M. Li, G.C. Tsui, Org. Lett. 25 (2023) 5656–5660. doi: 10.1021/acs.orglett.3c02050

  62. [62]

      D. Peixoto, G. Malta, H. Cruz, et al., J. Org. Chem. 84 (2019) 3793–3800. doi: 10.1021/acs.joc.8b02823

  63. [63]

      Z. Xiong, C.M. Hu, W.J. Zhu, et al., J. Org. Chem. 88 (2023) 8379–8386. doi: 10.1021/acs.joc.3c00405

  64. [64]

      H. Lammers, P. Cohen-Femandes, C.L. Habraken, Tetrahedron 50 (1994) 865–870.

• 

  Figure 1  Selected drugs containing 2-trifluoromethylquinoline core.

  下载: 全尺寸图片 幻灯片
  

  Scheme 1  Research background and our present reaction design.

  下载: 全尺寸图片 幻灯片
  

  Scheme 2  Substrate scope with respect to the synthesis of 2-CF₃, 2-CF₂H, and 2-CO₂Me-quinoline-4-carboxamides. Reaction conditions: 1 (0.10 mmol), 2 (0.30 mmol), 3 Å MS (50 mg), and BTMG (0.03 mmol) in EtOAc (0.8 mL) were stirred at 60 ℃ for 24–48 h. Isolated yields. ^a 50 mol% of BTMG was used.

  下载: 全尺寸图片 幻灯片
  

  Scheme 3  Substrate scope with respect to the synthesis of 2-CF₃, 2-CF₂H, and 2-CO₂Me-quinoline-4-carboxylic esters. Reaction conditions: 1 (0.10 mmol), 2 (0.30 mmol) and BTMG (0.03 mmol) in EtOAc (0.8 mL) were stirred at 60 ℃ for 24–48 h under air atmosphere. Isolated yields.

  下载: 全尺寸图片 幻灯片
  

  Scheme 4  Synthetic applications of this approach.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  Some selected 2-CF₃-quinoline-4-carboxamides and carboxylic esters with promising inhibitory activity against PTP1B.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  HOMO/LUMO distributions (a), absorption and PL spectra (b) of 68 and 69, respectively.

  下载: 全尺寸图片 幻灯片
  

  Figure 4  (a) device structures of single-color of OLEDs using 68 or 69, and the corresponding (b) current density (J) - luminance (L) -voltage (V) curves, (c) external quantum efficiency (EQE)-J curves, (d) EL spectra of devices at a voltage of 7 V.

  下载: 全尺寸图片 幻灯片
  

  Figure 5  Device structure (a) and EL characteristics of white-color OLEDs by using the EMLs of mCP: FIrpic (20 wt%): 69 (x wt%) (b-d), the corresponding white EL spectra at 7 V (e) and one respective bias-dependent EL spectra of device using 69 at 30 wt% (f).

  下载: 全尺寸图片 幻灯片
  

  Scheme 5  Reactions of N-[(α-trifluoromethyl)vinyl]isatins with organic superbases.

  下载: 全尺寸图片 幻灯片
  

  Scheme 6  Proposed reaction mechanism for the formation of products 3 and 72.

  下载: 全尺寸图片 幻灯片

  Table 1.  Optimization of the reaction conditions.^a

  Entry	Cat (X mol%)	Solvent	Additive	Yield (%)b
  1	Et3 N (30)	EtOAc	–	NR
  2	DABCO (30)	EtOAc	–	NR
  3	PMP (30)	EtOAc	–	NR
  4	DBU (30)	EtOAc	–	40
  5	TBD (30)	EtOAc	–	35
  6	BTMG (30)	EtOAc	–	63
  7	BTMG (30)	CH2Cl2	–	42
  8	BTMG (30)	ClCH2CH2Cl	–	48
  9	BTMG (30)	THF	–	30
  10	BTMG (30)	CH3CN	–	24
  11	BTMG (30)	Toluene	–	53
  12	BTMG (10)	EtOAc	–	51
  13	BTMG (20)	EtOAc	–	58
  14	BTMG (40)	EtOAc	–	64
  15	BTMG (30)	EtOAc	3 Å MS	71
  16	BTMG (30)	EtOAc	4 Å MS	68
  17	BTMG (30)	EtOAc	5 Å MS	67
  18	–	EtOAc	3 Å MS	NR
  a Standard reaction conditions: 1a (0.1 mmol), amine 2a (0.3 mmol), BTMG (10–40 mol%) in EtOAc, 60 ℃, 24 h. PMP: 1,2,2,6,6-pentamethylpiperidine; DABCO: 1,4-diaza bicyclo[2.2.2]octane; DBU: 1,8-diazabicyclo[5.4.0]undec–7-ene; TBD: 1,5,7-triazabicyclo[4.4.0]dec–5-ene; BTMG: 2–tert–butyl–1,1,3,3-tetramethyl-guanidine; NR = No Reaction. b Isolated yields.

  下载: 导出CSV
•