Determination of accurate 19F chemical shift tensors with R-symmetry recoupling at high MAS frequencies (60-100 kHz)

Abstract

Fluorination is a versatile and valuable modification for numerous systems, and ¹⁹F NMR spectroscopy is the premier method for their structural characterization. ¹⁹F chemical shift anisotropy is a sensitive probe of structure and dynamics, even though ¹⁹F chemical shift tensors have been reported for only a handful of systems to date. Here, we explore γ-encoded R-symmetry based recoupling sequences for the determination of ¹⁹F chemical shift tensors in fully protonated organic solids at high, 60-100 kHz MAS frequencies. We show that the performance of ¹⁹F-RNCSA experiments improves with increasing MAS frequencies, and that ¹H decoupling is required to determine accurate chemical shift tensor parameters. In addition, these sequences are tolerant to B₁-field inhomogeneity making them suitable for a wide range of systems and experimental conditions.

Keywords: R-symmetry recoupling; chemical shift anisotropy; high-frequency (19)F MAS NMR.

Figure 1.

Schematic depiction of ¹⁹F RNCSA pulse sequences used. (a) One R14₈⁵ block. (b) Conversion of a single π-pulse to a 270°–90° composite pulse [37]. (c) One composite (270°–90°) R14₈⁵ block. (d) CP-based ¹⁹F RNCSA with a short spin echo prior to acquisition for suppression of ring-down and baseline distortion. (e) DP-based ¹⁹F RNCSA experiment with either (f) spin echo or (g) triple pulse excitation [38] prior to acquisition for suppression of ring-down and baseline distortion. The spin echo-based ¹⁹F RNCSA phase cycle is: ${\mathrm{\phi }}_0=\mathrm{y}\mathrm{y}\mathrm{y}\mathrm{y}-\mathrm{y}-\mathrm{y}-\mathrm{y}-\mathrm{y};{\mathrm{\phi }}_1=\mathrm{y}\mathrm{y}\mathrm{y}\mathrm{y}-\mathrm{y}-\mathrm{y}-\mathrm{y}-\mathrm{y};{\mathrm{\phi }}_2=\mathrm{x}-\mathrm{x}\mathrm{y}-\mathrm{y};{\mathrm{\phi }}_3=\mathrm{x}\mathrm{x}\mathrm{x}\mathrm{x}\mathrm{y}\mathrm{y}\mathrm{y}\mathrm{y}-\mathrm{x}-\mathrm{x}-\mathrm{x}-\mathrm{x}-\mathrm{y}-\mathrm{y}-\mathrm{y}-\mathrm{y};{\mathrm{\phi }}_{\mathrm{r}\mathrm{e}\mathrm{c}}=\mathrm{x}-\mathrm{y}-\mathrm{x}\mathrm{y}-\mathrm{x}\mathrm{y}\mathrm{x}-\mathrm{y}$. The triple-pulse excitation phase cycle $\left({\mathrm{\phi }}_{11},{\mathrm{\phi }}_{12},{\mathrm{\phi }}_{13},{\mathrm{\phi }}_{\mathrm{r}\mathrm{e}\mathrm{c}}\right)$ is as described in [38].

Figure 2.

Spectrum and ¹⁹F CSA line shapes of Atorvastatin calcium (a) recorded with CP-excitation at the MAS frequency of 10 kHz (b) and R12₅⁴ RNCSA sequence at the MAS frequency of 60 kHz (c, d) and 100 kHz (e, f). ¹H π-decoupling was applied to remove ¹H-¹⁹F dipolar couplings. The experimental spectra and the fits are shown in black solid lines and red dotted lines, respectively. R12₅⁴ spectra were recorded with 19 (60 kHz MAS) and 32 (100 kHz MAS) t₁ transients, respectively, resulting in approximately the same CSA evolution period.

Figure 3.

Spectrum and ¹⁹F CSA line shapes of 5F-L-Trp (a) recorded with ¹H-¹⁹F CP at the MAS frequency of 10 kHz (b) and R12₅⁴ RNCSA acquired at 60 kHz MAS (c). ¹H π-decoupling was applied to remove ¹H-¹⁹F dipolar couplings. The experimental spectra and the fits are shown in black solid and red dotted lines, respectively. The R12₅⁴ spectrum was collected with 38 t₁ transients.

Figure 4.

¹⁹F spectrum and CSA line shapes of Mefloquine (a) recorded with single-pulse excitation at the MAS frequency of 10 kHz (b) and R12₅⁴ RNCSA acquired at 60 kHz (c, d) and 100 kHz (e, f). ¹H π-decoupling was applied to remove ¹H-¹⁹F dipolar couplings. The experimental spectra and the fits are shown in black solid and red dotted lines, respectively. R12₅⁴ spectra were collected with 38 (MAS frequency of 60 kHz) and 64 (MAS frequency of 100 kHz) t₁ transients, respectively, resulting in approximately the same CSA evolution period.

Figure 5.

Effect of ¹H decoupling on Mefloquine ¹⁹F RNCSA line shapes recorded at a MAS frequency of 100 kHz. Left and right are line shapes recorded with R12₅⁴ and R14₅³ sequences, respectively. Top and bottom are line shapes extracted for peaks with isotropic chemical shifts of 16.2 and 8.8 ppm, respectively. Line shapes recorded without and with π- and CW-decoupling are depicted in black, purple, and blue, respectively. R12₅⁴ and R14₅³ spectra were collected with 64 and 32 t₁ transients, respectively.

Figure 6.

¹H-decoupled Mefloquine ¹⁹F RNCSA line shapes recorded at MAS frequencies of 100 kHz (purple), 80 kHz (blue), and 60 kHz (black). Left and right are R12₅⁴ and R14₅³ line shapes, respectively, and top and bottom are peaks with isotropic chemical shifts of 16.2 and 8.8 ppm, respectively. π-pulse ¹H decoupling was applied in all experiments. R12₅⁴ spectra were collected with 38 (MAS frequency of 60 kHz), 51 (MAS frequency of 80 kHz), and 64 (MAS frequency of 100 kHz) t₁ transients, respectively, resulting in approximately the same CSA evolution period. R14₅³ spectra were collected with 19 (MAS frequency of 60 kHz), 26 (MAS frequency of 80 kHz), and 32 (MAS frequency of 100 kHz) t₁ transients, respectively, resulting in approximately the same CSA evolution period.

Figure 7.

¹⁹F RNCSA line shapes of Mefloquine and 5F-L-Trp at a MAS frequency of 60 kHz MAS using different R-symmetry sequences. The line shapes for two Mefloquine peaks at isotropic shifts 16.2 ppm (left) and 8.8 ppm (middle) as well as the line shapes for 5F-L-Trp (right) are shown. The R-symmetry numbers are indicated on the left and the corresponding RF power is indicated on the right. All spectra were collected with 38 t₁ transients, except for cR10₉³ and cR14₈⁵, which were collected with 48 and 24 t₁ transients, respectively. ¹H π-decoupling was used during t₁ evolution period of all sequences, but R14₅³ and cR14₈⁵, for which by ¹H CW-decoupling was employed.

Figure 8.

Effect of RF field mismatch on the ¹⁹F reduced anisotropy in RNCSA experiments of Mefloquine (peaks at 16.2 (a) and 8.8 (b) ppm) and microcrystalline 5F-L-Trp (c). All experiments were performed at a MAS frequency of 60 kHz. ${\delta }_{\sigma }$ values extracted from R12₇⁴, R14₈⁵, R14₆⁵, R12₅⁴, and R14₅³ experiments are shown as squares, stars, circles, triangles, and diamonds, respectively. The number of t₁ transients and ¹H decoupling schemes are as described in the legend of Fig. 7. RF field mismatch effect was measured by recollecting the ¹⁹F-RNCSA experiment with an adjusted ¹⁹F RF field during the CSA recoupling period (t₁), followed by fitting of the obtained line shapes.

Figure 9.

Off-resonance effects on the recoupled line shape in ¹⁹F RNCSA experiments of Mefloquine at a MAS frequency of 60 kHz. R12₅⁴ (a-f) and cR14₈⁵ (g-l) line shapes of Mefloquine peaks with isotropic shifts 16.2 ppm (a-c, g-i) and 8.8 ppm (d-f, j-l) with on-resonance irradiation at 13 ppm (a, d, g, j), 40 ppm off-resonance irradiation (b, e, h, k), and 80 ppm off-resonance irradiation (c, f, i, l).

Figure 10.

¹H-decoupled ¹⁹F cR14₈⁵ line shapes of Mefloquine at MAS frequencies of 60 kHz (black) and 100 kHz (purple) for peaks with isotropic chemical shifts of 16.2 (left) and 8.8 ppm (right), respectively. cR14₈⁵ spectra were recorded with 24 (MAS frequency of 60 kHz and CW ¹H decoupling) and 40 (MAS frequency of 100 kHz and π-pulse ¹H decoupling) t₁ transients, respectively, resulting in the same CSA evolution period.