2-(Fluoromethyl)-4,7-dimethoxy-1-methyl-1H-benzimidazole

Selectfluor (1-chloromethyl-4-fluoro-1,4-diazoniabicyclo[2.2.2]octane bis(tetrafluoroborate)) substitutes the TEMPO free radical with fluorine on 4,7-dimethoxy-1-methyl-2-{[(2,2,6,6-tetramethylpiperidin-1-yl)oxy]methyl}-1H-benzimidazole to give the title compound in a 77% yield. A mechanism is proposed for the formation of this novel methylene fluoride.


Results and Discussion
Treatment of alkoxyamine 1 with Selectfluor (1.2 equiv) at 0 °C led to the rapid liberation of the TEMPO free radical, as indicated by GC-MS (Figure 2), and benzylic fluorination, to give 2-(fluoromethyl)-4,7-dimethoxy-1-methyl-1H-benzimidazole 3. The novel methylene fluoride 3 was isolated in a 77% yield. 2-Fluoromethylbenzimidazole (without the dimethoxy groups) was previously prepared by condensation of 1,2-phenylenediamine with fluoroacetic acid [8]. The expected electrophilic aromatic fluorination at the electron-rich p-dimethoxybenzene part of 1 to give 2 was not observed (Scheme 1). The displacement of TEMPO is apparent when comparing the 1 H NMR spectra ( Figure 3). There are no TEMPO-based peaks in the spectrum of isolated 3, and the methylene signal shifted downfield to 5.61 ppm with splitting into a doublet ( 2 JH-F = 48.2 Hz) due to 1 H-19 F coupling. The location and multiplicity of the methylene signal is in good agreement with signals reported for 2-(fluoromethyl)-1H-benzimidazole (5.64 ppm, d, 2 JH-F = 47.5 Hz) [8].
Selectfluor Scheme 1. Unexpected formation of the title compound 3.

Results and Discussion
Treatment of alkoxyamine 1 with Selectfluor (1.2 equiv) at 0 °C led to the rapid liberation of the TEMPO free radical, as indicated by GC-MS (Figure 2), and benzylic fluorination, to give 2-(fluoromethyl)-4,7-dimethoxy-1-methyl-1H-benzimidazole 3. The novel methylene fluoride 3 was isolated in a 77% yield. 2-Fluoromethylbenzimidazole (without the dimethoxy groups) was previously prepared by condensation of 1,2-phenylenediamine with fluoroacetic acid [8]. The expected electrophilic aromatic fluorination at the electron-rich p-dimethoxybenzene part of 1 to give 2 was not observed (Scheme 1). The displacement of TEMPO is apparent when comparing the 1 H NMR spectra (Figure 3). There are no TEMPO-based peaks in the spectrum of isolated 3, and the methylene signal shifted downfield to 5.61 ppm with splitting into a doublet ( 2 JH-F = 48.2 Hz) due to 1 H-19 F coupling. The location and multiplicity of the methylene signal is in good agreement with signals reported for 2-(fluoromethyl)-1H-benzimidazole (5.64 ppm, d, 2 JH-F = 47.5 Hz) [8]. The displacement of TEMPO is apparent when comparing the 1 H NMR spectra (Figure 3). There are no TEMPO-based peaks in the spectrum of isolated 3, and the methylene signal shifted downfield to 5.61 ppm with splitting into a doublet ( 2 J H-F = 48.2 Hz) due to 1 H-19 F coupling. The location and multiplicity of the methylene signal is in good agreement with signals reported for 2-(fluoromethyl)-1H-benzimidazole (5.64 ppm, d, 2 J H-F = 47.5 Hz) [8]. The methylene signal split into a doublet ( 1 JC-F = 165.5 Hz) in the 13 C NMR spectrum of 3 at 76.8 ppm (see Supplementary Materials for NMR spectra). 13 C-19 F NMR coupling also gave doublets for benzimidazole-C-2 at 147.1 ppm ( 2 JC-F = 19.0 Hz), and for the N-CH3 at 32.6 ppm ( 4 JC-F = 2.5 Hz). The former is in good agreement with the literature data on 2-(fluoromethyl)-1H-benzimidazole C-2 (148.6 ppm, d, 2 JCF = 19.7 Hz) [8].
The 19 F NMR signal for 3 at −214.93 ppm is similar to the literature value of −213.92 ppm for 2-(fluoromethyl)-1H-benzimidazole [8]. The signal appeared as a triplet ( 2 JF-H = 48.0 Hz), due to 19 F-1 H coupling with the two 1 H atoms of the adjacent methylene group.
Alkoxyamine 1 is stable to visible-light and the reaction was performed at 0 °C, therefore ruling out bond homolysis as a pathway to formation of methylene fluoride 3. Assuming Selectfluor is a source of F + or F • and not fluoride, this rules out SN2 displacement of the TEMPO residue [3,9]. Incompatible polarization of the alkoxyamine C-O bond also prevents a simple SH2 mechanism. A single electron transfer (SET) pathway is now proposed, and is supported by the electrochemical oxidations of TEMPO-based alkoxyamines (TEMPO-R) with mesolytic cleavage of the alkoxyamine bond forming TEMPO + and R • [10]. In this case (Scheme 2), SET is proposed to induce mesolytic cleavage of ʹbenzylic alkoxyamineʹ 1 to produce TEMPO + and a methylene radical 4. Abstraction of F • by 4 gives reaction product 3, while reduction of the oxoammonium cation 5 by the Selectfluorderived DABCO derivative 6 gives the TEMPO free radical detected by GC-MS (Figure 2). A plausible alternative to the mesolytic cleavage is initial SN2 on the fluorine of Selectfluor by the N-3 of benzimidazole 1 to give an imidazolium fluoride [9]. The subsequent generation of 3 eliminates a TEMPO + species that would undergo reduction by 6 (as in Scheme 2) to give a TEMPO free radical. The methylene signal split into a doublet ( 1 J C-F = 165.5 Hz) in the 13 C NMR spectrum of 3 at 76.8 ppm (see Supplementary Materials for NMR spectra). 13 C-19 F NMR coupling also gave doublets for benzimidazole-C-2 at 147.1 ppm ( 2 J C-F = 19.0 Hz), and for the N-CH 3 at 32.6 ppm ( 4 J C-F = 2.5 Hz). The former is in good agreement with the literature data on 2-(fluoromethyl)-1H-benzimidazole C-2 (148.6 ppm, d, 2 J CF = 19.7 Hz) [8].
The 19 F NMR signal for 3 at −214.93 ppm is similar to the literature value of −213.92 ppm for 2-(fluoromethyl)-1H-benzimidazole [8]. The signal appeared as a triplet ( 2 J F-H = 48.0 Hz), due to 19 F-1 H coupling with the two 1 H atoms of the adjacent methylene group.
Alkoxyamine 1 is stable to visible-light and the reaction was performed at 0 • C, therefore ruling out bond homolysis as a pathway to formation of methylene fluoride 3. Assuming Selectfluor is a source of F + or F • and not fluoride, this rules out S N 2 displacement of the TEMPO residue [3,9]. Incompatible polarization of the alkoxyamine C-O bond also prevents a simple S H 2 mechanism. A single electron transfer (SET) pathway is now proposed, and is supported by the electrochemical oxidations of TEMPO-based alkoxyamines (TEMPO-R) with mesolytic cleavage of the alkoxyamine bond forming TEMPO + and R • [10]. In this case (Scheme 2), SET is proposed to induce mesolytic cleavage of 'benzylic alkoxyamine' 1 to produce TEMPO + and a methylene radical 4. Abstraction of F • by 4 gives reaction product 3, while reduction of the oxoammonium cation 5 by the Selectfluor-derived DABCO derivative 6 gives the TEMPO free radical detected by GC-MS (Figure 2). A plausible alternative to the mesolytic cleavage is initial S N 2 on the fluorine of Selectfluor by the N-3 of benzimidazole 1 to give an imidazolium fluoride [9]. The subsequent generation of 3 eliminates a TEMPO + species that would undergo reduction by 6 (as in Scheme 2) to give a TEMPO free radical. Scheme 2. Proposed mechanism for the formation of methylene fluoride 3.