1. Introduction

Grob fragmentations are the direct breakage of heterolytic sigma bonds that have been widely studied since their discovery in the 1950s [1], which are the key steps in numerous syntheses of natural products [2] involving the very important and intriguing diterpene Vinigrol [2a]. Among Grob fragmentations, 1, 3-diheterofunctionalized compounds are typically substrates featuring a nucelophilic atom with a negative charge or lone electron pair (electrofuge) and a leaving group (nucleofuge) in a 1, 3-relationship, undergoing heterolytic fragmentation to form three fragments: a positive fragment Ⅰ (the "electrofuge"), an unsaturated neutral fragment Ⅱ, and a negative ion Ⅲ (the "nucleofuge") (Fig. 1). Usually, the positive ion could be a carbonium ion; the neutral fragment could be an alkene and the negative fragment could be a tosyl ion. Bases or nucleophiles induced fragmentations are most common in the reactions. However, Brønsted and Lewis acids mediated or catalyzed Grob fragmentations are relatively rare due to the side reactions triggered by the carbonium ion intermediate in the substrates [3].

2. Results and discussion

In the course of our study towards the synthesis of sesquiterpenes containing a decalin skeleton, an unexpected and unprecedented Grob fragmentation was uncovered (Scheme 1). In order to remove the silyl group, the vinylogous β-silyloxycyclobutanone 1 [4] was treated with 40% aqueous HF in CH₃CN at 0 ℃ for 24 h [5], expecting to obtain the desired target alcohol product 3. However, this reaction led to an unexpected product 2 in 90% yield, in which the cyclobutanone was converted to the γ-lactone and the silyloxy group was removed. The structure of 2 was confirmed by single-crystal X-ray analysis (Fig. 2). Actually, the ketone carbonyl group and the silyloxy group are in a 1, 5-relationship. To the best of our knowledge, this is the first example of Grob fragmentation of a 1, 5-difunctionalized substrate under acidic conditions and we envisioned that the double bond plays key role in this long-distance Grob fragmentation [6].

In order to identify the role of the double bond and the reaction mechanism in this fragmentation, therefore, substrate 4 without double bond was synthesized and treated with the same conditions as described in Scheme 1, the desired target alcohol product was obtained smoothly. Then, in the presence of 4-nitrobenzoyl chloride and pyridine, the ester 5 was obtained in 82% yield over two steps (Scheme 2). As the result illustrated above we can see that the double bond is undoubtedly key and important in this fragmentation. The structure of 5 was confirmed by singlecrystal X-ray analysis (Fig. 2).

According to the result above, we proposed a plausible mechanism of this fragmentation (Scheme 3). Initially, the TBS (tert-butyldimethylsilyl) group was removed by fluoride ion to afford the alcohol 6. Under aqueous acidic conditions, the ketone carbonyl group hydrolyzed to the hydrate intermediate 7. Then, after protonation of the hydroxyl group, water acted as a leaving group and the four-membered ring of 8 opened to generate the 1, 3-butadiene intermediate 9 involving a carboxylic acid group. From the transition state of 8 we can see that the lone electron pair on O (H), the C-OH₂⁺, and carbon—carbon bond of CH—C(OH)OH are anti-periplanar for maximal orbital overlap and the double bond can transmit electrons in the fragmentation. Finally, by proton transfer one double bond of the diene 10 could be protonated and was attacked by the internal carboxylic acid group to form the cislactone 2. In this fragmentation the double bond behaves like a shuttle bus which can transmit the electrons from one side to another [7]. Unsaturated decalinic lactones obtained are important building blocks and key synthetic intermediates in many organic syntheses [8].

3. Conclusion

In conclusion, we have disclosed an unprecedented formal oxy transposition of vinylogous β-silyloxycyclobutanone (a 1, 5-difunctionalized substrate) into unsaturated decalinic γ-lactone under acidic conditions by Grob fragmentation and we propose a plausible mechanism of this reaction. We envision that this unique fragmentation will provide new opportunities for organic synthesis.

4. Experimental

All reactions were monitored by thin-layer chromatography (TLC). All organic extracts were washed with brine before being dried over Na₂SO₄. For column chromatography, silica gel (200-300 mesh) and light petroleum ether (PE, b.p. 60-90 ℃) were used. Melting points (mp) were measured on a Kofler hot stage and are uncorrected. FTIR spectrum was recorded on a Nicolet MAGNA-560 FTIR spectrometer. ¹H NMR, ¹³C NMR and DEPT135 NMR spectra were recorded on a Bruker AV400 instrument; Chemical shifts are reported as δ using residual solvent as an internal standard. High resolution mass spectral analyses (HRMS) were measured on a Varian 7.0 TFTICR MS spectrometer by means of ESI.

4.1. Preparation of 5-methyl-6, 6a, 7, 8, 9, 10-hexahydro-1H-naphtho [1, 8a-b]furan-2(3aH)-one (2)

To a solution of 1 (20 mg, 0.094 mmol) in acetonitrile (5 mL) was added the aqueous HF solution (0.2 mL). The mixture was stirred for 24 h at 0 ℃. The reaction mixture was concentrated in vacuo and the residue was purified through flash column chromatography (petroleum ether/ethyl acetate = 4/1) to afford the final product (17 mg) with 90% yield after concentration. White solid. Mp: 132-134 ℃; ¹H NMR (400 MHz, CDCl₃): δ 5.53 (d, 1H, J = 1.8 Hz), 4.52 (s, 1H), 2.75 (d, 1H, J = 17.3 Hz), 2.21 (d, 1H, J = 17.2 Hz), 2.04 (qd, 2H, J = 18.2, 6.2 Hz), 1.79 (m, 1H), 1.76 (s, 3H), 1.72-1.63 (m, 1H), 1.61-1.55 (m, 3H), 1.48-1.37 (m, 3H), 1.28-1.22 (m, 1H); ¹³C NMR (100 MHz, CDCl₃): δ 176.3, 140.2, 116.8, 81.1, 40.8, 40.7, 33.5, 32.3, 32.1, 27.2, 23.6, 22.9, 22.5; IR (neat): 1774 cm^-1; HRMS (ESI) exact mass calcd. for: C₁₃H₁₈O₂Na (m+Na)⁺ 229.1199, found 229.1196.

4.2. Preparation of 4-methyl-2-oxodecahydro-1H-cyclobuta[d] naphthalen-5-yl 4-nitrobenzoate (5)

To a solution of 4 (12 mg, 0.037 mmol) in acetonitrile (3 mL) was added the aqueous HF solution (0.2 mL). The mixture was stirred for 24 h at 0 ℃. The reaction mixture was concentrated in vacuo. To the residue was added ethyl acetate (10 mL). The organic solution was washed with aqueous saturated aqueous NaHCO₃ (8 mL), brine (8 mL) and dried over Na₂SO₄ and concentrated. To the crude product was added dicholomethane (2 mL), pyridine (8.0 mL, 0.1 mmol) and 4-nitrobenzoyl chloride (13 mg, 0.07 mmol). The mixture was stirred for 4 h at room temperature. The reaction mixture was concentrated in vacuo and the residue was purified through flash column chromatography (petroleum ether/ethyl acetate = 20/1) to afford the final product (11 mg) with 82% yield after concentration. White solid. Mp: 140-142 ℃; ¹H NMR (400 MHz, CDCl₃): δ 8.34-8.29 (m, 2H), 8.24-8.19 (m, 2H), 5.23 (dd, 1H, J = 7.1, 3.7 Hz), 3.17 (dd, 1H, J = 16.9, 2.1 Hz), 2.94 (m, 1H), 2.70 (dd, 1H, J = 16.9, 3.8 Hz), 2.11 (m, 1H), 2.06-1.95 (m, 3H), 1.88-1.78 (m, 1H), 1.72-1.63 (m, 2H), 1.55-1.46 (m, 3H), 1.37 (m, 1H), 1.28-1.22 (m, 1H), 1.02 (d, 3H, J = 6.8 Hz); ¹³C NMR (100 MHz, CDCl₃): δ 209.5, 164.3, 150.6, 135.8, 130.7, 123.7, 80.6, 60.2, 57.2, 38.0, 37.7, 32.8, 31.0, 26.2, 25.1, 23.6, 23.5, 19.4; IR(neat): 1776, 1718 cm^-1; HRMS (ESI) exact mass calcd. for:C₂₀H₂₃NO₅Na (m + Na)⁺ 380.1468 found 380.1464.

Acknowledgment

We are grateful for financial support from National Natural Science Foundation of China (NSFC, Nos. 21302177, 21672049 and 21672030).

Appendix A. Supplementary data

Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.cclet.2016.12.017.