The Shielding Effect of Phenyl Groups in the Silyl-Protecting Groups Introduced into Borneol and Isoborneol

Abstract

Protection of the hydroxy group in monoterpenoids borneol and isoborneol with various silyl-protective groups containing a different number of phenyl groups enabled complete assignments of the ¹H chemical shifts. In particular, a wider range of the protons in isoborneol and its silyl-protected derivatives were influenced than those in borneol and its silyl-protected derivatives, which is likely to be due to the anisotropic effects from the phenyl groups. Understanding these effects allows for the interpretation of the ¹H NMR spectra and provides information about how silyl-protecting groups affect these spectra.

1. Introduction

Silyl-protective groups play a particularly important role in protecting hydroxy groups in organic synthesis. The central silicon atom bears three functional groups, which can include a wide range of alkyl and aryl groups such as methyl, ethyl, propyl, tert-butyl, and phenyl groups [1,2,3,4]. These silyl-protective groups form silyl ethers by reacting with alcohols, and they can also react with amino groups [5,6,7]. Due to the availability of various sizes, they often allow selective protection of primary alcohols in the presence of secondary and/or tertiary alcohols, due to their bulkiness. This selective protection is beneficial for the synthetic studies of various complex compounds [1].

We previously demonstrated that NMR spectroscopy serves as a useful tool for monitoring subtle structural differences in the environment of organic compounds using natural products, borneol (1a) and isoborneol (2a), and their derivatives [8]. These monoterpenes have been applied to pharmaceuticals, insect repellants, and chiral ligands for many decades, and therefore protection of their hydroxy groups as silyl ethers is likely to expand the scope for synthetic utilities [9,10,11,12]. Although ¹³C NMR assignments of bornyl acetate have been reported recently [13], there has been no report studying their ¹H NMR spectroscopic behaviors after protection with silyl-protective groups. As part of our ongoing studies of monitoring behavior of various organic compounds by NMR spectroscopy, we have now monitored ¹H NMR spectroscopic behaviors for structural differences of borneol (1a) and isoborneol (2a) protected with various silyl-protective groups.

Borneol (1a) and isoborneol (2a) are monoterpenoids with a bicyclic structure, and they are epimers of each other, differing in the stereochemistry of the hydroxy group, as shown in Scheme 1. Their ¹H NMR spectra differ primarily due to the stereochemistry of this hydroxy group. In particular, the ¹H NMR chemical shifts for their three methyl groups are quite different. In CDCl₃ solution, these three methyl groups of borneol overlap, while these methyl groups are separated in isoborneol (2a). We previously observed high-frequency (hereafter referred to as downfield (i.e. deshielding)) shifts of the overlapped methyl groups after the protection of borneol (1a) or isoborneol (2a) with an acetyl group or a benzoyl group in each of their ¹H NMR spectra [8].

Here, we protected the hydroxy group of borneol (1a) and isoborneol (2a) with various silyl-protective groups containing different numbers of phenyl groups and investigated the changes in their ¹H NMR chemical shifts relative to the number of benzene rings.

In the past, the ¹³C NMR chemical shift assignments of borneol (1a) and isoborneol (2a) using chiral shift reagents were reported [14,15,16,17,18], but very little is known about the ¹H NMR chemical shift changes upon protection with a silyl-protective group. Therefore, we introduced tert-butyl dimethylsilyl (TBDMS) group, dimethylmonophenylsilyl (DMMPS) group, tert-butyldiphenylsilyl (TBDPS) group, and triphenylsilyl (TPS) group as protective groups for borneol (1a) and isoborneol (2a). These silyl-protective groups have zero (TBDMS), one (DMMPS), two (TBDPS), and three (TPS) phenyl groups, respectively. We also monitored the ¹H NMR chemical shift changes of these products in CDCl₃ and C₆D₆ solutions.

2. Results and Discussion

2.1. ¹H NMR Chemical Shift Changes of Borneol in CDCl₃ and C₆D₆ Solutions

Figure 1 shows the ¹H NMR spectra with chemical shift changes for borneol (1a) and its derivatives 1b–1e protected with each silyl-protective group in CDCl₃ solutions. From Figure 1, we observed different changes in the ¹H NMR chemical shift for each of the H-2 to H-6 protons and for the three methyl groups, H-8 to H-10, after the protection with the silyl-protective groups, depending on the number of the phenyl groups.

As shown in Figure 1, in CDCl₃ solution, the H-2exo proton (blue), which was the most deshielded, showed a slight downfield shift upon protection with the silyl groups. This downfield shift was more prominent as the number of phenyl groups increased, suggesting these changes are due to the anisotropic effects from the benzene ring. The H-3exo proton (red), which was the second most deshielded, showed low-frequency (hereafter referred to as upfield (i.e. shielding)) shifts as the number of phenyl groups increased. Conversely, the H-6endo (blue) proton exhibited downfield shifts. The H-5exo and H-4 protons remained almost unchanged. Although the H-6exo and H-5endo protons overlap with each other in borneol, the H-5endo (blue) proton exhibited a slight downfield shift, while the H-6exo (red) proton showed a slight upfield shift. The H-3endo proton remained almost unchanged. The extent of the downfield shifts observed for the H-6endo proton were the greatest among these protons that exhibited downfield shifts after the introduction of silyl protection with different numbers of phenyl groups.

Figure 2 shows the ¹H NMR chemical shift changes for the same series of compounds in C₆D₆ solution after protection with the silyl-protective groups, depending on the number of phenyl groups. Similar changes were observed in C₆D₆ solution as in CDCl₃ solution. The most deshielded proton in borneol, H-2exo (blue), exhibited larger downfield shifts in C₆D₆ solution than in CDCl₃ solution. This result is likely due to the anisotropic effects from the benzene solvent in addition to the benzene ring in the silyl groups. The H-6endo proton (blue), which was the second most deshielded in borneol (1a), showed relatively large downfield shifts, similar to those in CDCl₃ solution. Conversely, the H-3exo (red) proton showed upfield shifts, but the changes were smaller than those in CDCl₃ solution. The two protons, H-5exo and H-4, did not show changes in chemical shifts upon protection with the silyl group in CDCl₃ solution. However, H-5exo showed slight downfield shifts while H-4 remained almost unchanged in C₆D₆ solution. The H-5endo and H-6exo protons appeared overlapped in borneol (1a), but both exhibited slight downfield shifts after protection with silyl-protective groups. The H-3endo proton also showed large downfield shifts. The extent of the downfield shifts observed for the H-6endo and H-3endo protons was the greatest among the protons that exhibited downfield shifts. From these results, it appears that by introducing silyl-protective groups on the endo hydroxy group of borneol (1a), the neighboring H-3endo and H-6endo protons are affected by the anisotropic effects from the side of the benzene ring in the silyl group. Therefore, by protecting the hydroxyl group of borneol (1a) with silyl protecting groups having various numbers of phenyl groups and measuring the effect of each protecting group on the ¹H NMR spectrum, it was possible to assign the ¹H NMR chemical shifts of borneol (1a), even with overlapping signals.

Figure 3 shows the differences in the ¹H NMR chemical shifts of the methyl groups when the same series of compounds as above are protected with silyl groups containing different numbers of phenyl groups, in CDCl₃ and C₆D₆ solutions. In borneol (1a), the order of the chemical shifts for these methyl groups differed depending on the deuterated solvent. In CDCl₃ solution, the protons H-10 of the methyl group at C-10 were the most shielded, followed by the protons H-9, with the protons H-8 being the most deshielded. On the other hand, in C₆D₆ solution the methyl protons H-9 of the methyl group at C-9 were the most shielded, followed by H-8, with H-10 being the most deshielded. In CDCl₃ solution, the H-8 methyl protons of borneol (1a) and its derivatives (1b–1e) showed little change regardless of the type of the silyl group, while the H-9 and H-10 methyl groups exhibited upfield shifts as the number of phenyl groups increased. In particular, the H-9 methyl protons exhibited the largest upfield shifts. Similarly, in C₆D₆ solution, the H-9 methyl protons exhibited larger upfield shifts as the number of phenyl groups increased. The H-8 methyl protons showed little influence from the introduction of the silyl group, as observed in CDCl₃. The H-10 methyl protons showed downfield shifts after protection, but the degree of shift did not depend on the number of phenyl groups.

The chemical shift difference between the H-8 and H-9 methyl groups tended to become larger in CDCl₃ solution, while that between the H-9 and H-10 methyl groups tended to become larger in C₆D₆ solution [19]. This difference was the largest in the TBDPS-protected borneol (1d). From these observations, it was suggested that among these methyl groups, the H-9 methyl group is affected by the anisotropic effects from the benzene ring to the greatest degree. It was also suggested that among the protons attached to the bicyclic ring, the H-6endo proton is most prominently affected by the anisotropic effects.

2.2. ¹H NMR Chemical Shift Changes of Isoborneol in CDCl₃ and C₆D₆ Solutions

We also examined the ¹H NMR chemical shift changes of the protons on the bicyclic ring and three methyl groups by protecting the hydroxy group of isoborneol (2a) with various silyl-protective groups containing different numbers of benzene rings, as in the above studies on borneol (1a). Figure 4 shows the ¹H NMR spectra of this series of compounds in CDCl₃ solution, and Figure 5 shows those in C₆D₆ solution. In contrast to borneol (1a), except for the H-2 proton and three methyl groups, the ¹H NMR chemical shifts for most protons in isoborneol (2a) were observed to be significantly more overlapped with each other. Therefore, we assigned the ¹H NMR chemical shift for each proton using the HMQC spectrum, based on the cross peaks between the separated ¹³C NMR and ¹H NMR signals. In particular, as shown in Figure 4, four protons (H-3exo, H-3endo, H-4, and H-5exo) were observed to sufficiently overlap in the ¹H NMR spectrum measured in CDCl₃ solution. Among the four protons that were well-overlapped in CDCl₃ solution (H-3exo, H-3endo, H-4, and H-5exo), H-3endo and H-5exo showed upfield shifts, while the H-3exo proton exhibited downfield shifts as the number of phenyl groups in the silyl groups increased, similarly to H-2endo. The H-4 proton showed little change upon protection with the silyl groups, as observed in the previous studies for borneol (1a). Additionally, in CDCl₃ solution, H-5endo and H-6endo exhibited slight upfield shifts as the number of phenyl groups increased. In particular, the extent of the upfield shifts was prominent with the introduction of the TBDPS group and TPS group due to their phenyl groups. It is therefore conceived that an increased number of benzene rings can induce more enhanced anisotropic effects [20,21].

Next, in the C₆D₆ solution, isoborneol (2a) exhibited ¹H NMR chemical shift changes that are different from those in the CDCl₃ solution, as compared in Figure 4 and Figure 5. As shown in Figure 5, the H-2 and H-3exo protons in C₆D₆ solution showed a downfield shift as the number of phenyl groups increased due to the introduction of silyl groups, which is similar to the shift observed in the CDCl₃ solution. The degree of the downfield shift of these two protons is attributed to the solvent effect of C₆D₆, as the benzene solvent causes a larger downfield shift. On the other hand, the H-5exo, H-3endo, H-6exo, H-5endo, and H-6endo protons showed a slight upfield shift as the number of phenyl groups in the silyl group increased, which is the same as in the CDCl₃ solution. Similar to the case of CDCl₃ solution, the chemical shift for the H-4 proton remained almost unchanged.

We next focused on the chemical shifts for the three methyl groups of isoborneol (2a), and its derivatives (2b–2e), as shown in Figure 6. Each of the three methyl groups in isoborneol (2a) had distinct chemical shifts, differently from those in borneol (1a). In each solvent, the H-8 methyl group was the most shielded, and the H-9 methyl group was the most deshielded as the number of phenyl groups in the silyl group increased. In CDCl₃ solution, the chemical shifts for the H-8 methyl group of isoborneol (2a) remained almost unchanged, which is similar to borneol (1a), but the H-10 methyl group exhibited upfield shifts while the H-9 methyl group exhibited downfield shifts. Furthermore, as the number of phenyl groups in the silyl group increased, the chemical shift differences between the most shielded H-8 methyl group and the most deshielded H-9 methyl group tended to expand. In C₆D₆ solution, the H-8 methyl group was the most shielded, but its chemical shift remained almost unchanged with the increased number of phenyl groups in the silyl group upon the protection of the hydroxy group, similar to the CDCl₃ solution. On the other hand, the H-9 methyl group exhibited downfield shifts as the number of phenyl groups in the silyl group increased, which is similar to the observation in CDCl₃ solution. The H-10 methyl group showed a slight downfield shift, contrary to the observation in CDCl₃ solution. The chemical shift difference between the most shielded H-8 methyl group and the most deshielded H-9 methyl group expanded to a greater extent in C₆D₆ than in CDCl₃, which is the same tendency observed in borneol (1a) and its silyl-protected derivatives (1b–1e).

3. Materials and Methods

3.1. Chemicals

(−)-Borneol (1), purity > 95% (GC), and (±)-isobornel (2), purity > 90% (GC) were purchased from Tokyo Chemical Industry Co., Ltd. (Tokyo, Japan), and the compounds 1 and 2 were used without further purification. The deuterated solvents, CDCl₃, C₆D₆, and CD₃OD and other reagents were purchased from Fujifilm Wako Chemicals Co., Ltd. (Osaka, Japan).

3.2. NMR Measurements

All the NMR spectra were recorded using a JEOL JNM-ECZ400S NMR spectrometer (JEOL Ltd., Tokyo, Japan) at 298 K with 5 mm (o.d.) Pyrex glass tubes. ¹H and ¹³C NMR chemical shifts were measured with reference to the residual ¹H and ¹³C signals of the deuterated solvents (CDCl₃, δ_H 7.260, δ_C 77.01; C₆D₆, δ_H 7.156, δ_C 128.03) [22]. The spectra of ¹H NMR, ¹³C NMR, DEPT, and 2D correlation experiments (DQF COSY, and HMQC) were performed for these products in CDCl₃ or C₆D₆ solutions. The number of accumulations for NMR are as follows: ¹H NMR, 8 scans; ¹³C NMR, 5000 scans; DEPT, 2500 scans; DQF COSY, 8 scans; HMQC, 16 scans.

3.3. General Procedures for Protection of Borneol and Isoborneol with a Silyl-Protecting Group

Silyl-protected borneol and isoborneol were prepared with the corresponding silyl protecting reagent with imidazole and iodine as an additive according to the reported procedures [23,24]. An alcohol (borneol (1a) or isoborneol (2a)) (1.0 mmol), base (imidazole or 1-methylimidazole) (3.0 mmol), and additive (iodine or KI) (3.0 mmol) were dissolved in 5 mL of solvent (methylene chloride or acetonitrile), and a silyl reagent was then added, and the mixture was stirred for 24 h. After confirming the consumption of the starting alcohol by TLC analysis, the reaction mixture was quenched with a saturated sodium thiosulfate aqueous solution to remove iodine. The organic layer was collected using a separatory funnel, dried over anhydrous sodium sulfate, and evaporated. The silyl-protected products were purified by column chromatography on silica gel with hexane-EtOAc (95:5).

The conditions for protection with each silyl-protective group and the reaction yields are as follows. Protection with TBDMS: borneol/isoborneol (1.0 mmol), imidazole (3.0 mmol), methylene chloride (5 mL), tert-butyldimethylchlorosilane (1.5 mmol); yield: bornyl TBDMS (1b): 99%, isobornyl TBDMS (2b): 52%. Protection with DMMPS: borneol/isoborneol (1.0 mmol), imidazole (3.0 mmol), iodine (3.0 mmol), acetonitrile (3 mL), chlorodimethylphenylsilane (1.1 mmol); yield: bornyl TBDMS (1c): 56%, isobornyl TBDMS (2c): 36%. Protection with TBDPS: borneol/isoborneol (1.0 mmol), 1-methylimidazole (3.0 mmol), acetonitrile (3 mL), iodine (3.0 mmol), chlorodimethylphenylsilane (1.1 mmol); yield: bornyl TBDPS (1d): 66%, isobornyl TBDPS (2d): 21%. Protection with TPS: borneol/isoborneol (1.0 mmol), imidazole (3.0 mmol), methylene chloride (5 mL), triphenyllchlorosilane (1.5 mmol); yield: bornyl TPS (1e): 99%, isobornyl TPS (2e): 89%.

4. Conclusions

We have investigated the effects of the protection of the hydroxy group in borneol (1a) and isoborneol (2a) with various silyl-protective groups for the assignments of ¹H NMR chemical shifts. As a result, it was observed that downfield shifts and upfield shifts tended to increase as the number of phenyl groups increased, even at the positions remote from the phenyl groups, which is likely to be due to the anisotropic effects.

In the series of borneol (1a) and its silyl-protected derivatives (1b–1e), the downfield shift was particularly prominent in the H-2exo proton attached to the carbon bearing the hydroxy group and the H-6endo proton. Conversely, the H-3exo proton exhibited upfield shifts, suggesting it is affected by the anisotropic effects of the benzene ring in the silyl protecting groups. For the methyl groups of borneol (1a), the protons H-9 and H-10 of methyl groups at C-9 and C-10 exhibited upfield chemical shifts. As these methyl groups are directed toward the hydroxy group, they are likely affected by the anisotropic effects from the benzene rings.

In the series of isoborneol (2a) and its silyl-protected derivatives (2b–2e), which possesses a crowded exo hydroxy group at the C-2 position, a wide range of protons were influenced upon protection with silyl groups. All the H-2 and H-3exo protons, except for the H-3endo proton, exhibited downfield shifts, and all the H-5exo, H-5endo, H-6exo, and H-6endo protons exhibited upfield shifts, even in CDCl₃ solution, despite the distance from the hydroxy group at C-2. In particular, for the methyl groups of isoborneol (2a), the methyl protons H-8 at C-8 remained almost unchanged, but the methyl protons H-9 at C-9 exhibited larger downfield shifts as the number of phenyl groups in the silyl groups increased due to proximity to the benzene ring. In conclusion, we demonstrated that ¹H NMR chemical shift assignments are possible, even for compounds with several overlapping signals, by protecting with a silyl group containing various numbers of phenyl groups and observing the induced upfield and/or downfield shifts. We are currently studying the observed changes of the chemical shift values in more depth, in combination with quantum mechanical calculations, and the results will be reported in due course.