Sterically controlled rhenium-catalyzed hydroxyl transposition

Easily prepared vinyliodides are transformed in a few steps into enantiomerically pure alcohol or lactone synthons. The key transformation involves the highly regioselective rhenium(VII)-catalyzed transposition of an allylic alcohol that was prepared by stereoselective addition of a vinyllithium to p -menthane-3- carboxaldehyde.


Introduction
Since the first reports on the high efficiency of trioxorhenium species as catalysts for the isomerization of allylic alcohols, [1][2][3] chemists have strived to control the position of the equilibrium between the two regioisomeric alcohols 3 and 4 in the hope of favouring a single form (Scheme 1).5][6] In absence of conjugation, a tertiary allylic alcohol usually predominates over the other regioisomer, but the selectivity is not always pronounced. 4 The ratios of regioisomers obtained when a secondary and a primary alcohol compete is particularly weak. 71][12] In the latter systems, the trapping moieties are part of the structure and are not meant to be removed, limiting the applicability of the method to this type of substrate.4][15] More ways of controlling the position of the equilibrium between allylic alcohols are still needed to increase the scope of this useful catalytic transformation.
Alkynes (1) are easy to prepare and, via their hydrometallation, can lead to alcohols 3, directly or via the corresponding vinylhalide 2 (Scheme 1).In the case where the regioisomer 4 would be needed, the Re(VII)catalyzed transposition would likely give an unusable mixture of both regioisomers 3 and 4. [4][5][6][7] We now wish to report that the chiral auxiliary p-menthane-3-carboxaldehyde may be used to supply a very high regioisomeric ratio of chiral non racemic allylic alcohols upon treatment with 2% Re(VII) catalyst under mild conditions.Cleavage of the chiral auxiliary leads directly to useful synthons and recovery of the auxiliary.Scheme 1. Alkynes as precursors to allylic alcohols as synthons.

Results and Discussion
We have developed p-methane-3-carboxaldehyde 5 (which we commonly call menthylaldehyde) to access chiral synthons like products 8-11 reliably and efficiently (Scheme 2). 16The alcohols 6 are typically prepared enantiomerically and diastereomerically pure in good yields from alkenylmetals and, after proper activation, the hydroxyl can be regio-and stereoselectively transposed to isomers 7. We have successfully transposed the hydroxyl group with carbon, nitrogen or sulfur nucleophiles using rearrangements or SN2' reactions. 16One of the main attractiveness of the method is that the auxiliary can be cleaved oxidatively to give acids or aldehydes like 9 or by ring-closing metathesis (RCM) to access such carbo-and heterocycles as shown (8, 10,  and 11).Different cleavage methods allow for the formation of acyclic or cyclic chiral constructs 8-11 that can then be used as synthons.

Scheme 2. p-Menthane-3-carboxaldehyde 5 as chiral auxiliary.
The large steric volume of the auxiliary is in part responsible for the high regioselectivity we observe when performing cuprate additions or rearrangement reactions on derivatives of 6.We were intrigued to see if this steric bulk would be enough to bias the Re-catalyzed equilibrium between two regioisomeric hydroxyl groups 6 and 12 (Scheme 3).If so, we could significantly extend the usefulness of this method to making Oheterocycles 13, cyclic alcohols 14, -hydroxy acids 15, and other useful synthons.To test the idea, we submitted allylic alcohol 16a to Osborn's catalyst 3 at 2 mol % loading under various reaction conditions.It was immediately apparent that the auxiliary was performing adequately in biasing the equilibrium between the two regioisomers 16a and 17a in favour of the latter (Table 1, entries 1 to 5).However, initially, complete epimerization of the stereocenter occurred (entry 1).It appeared that the Lewis acidity of the rhenium(VII) catalyst led to the formation of an allylic carbocation thereby destroying the stereochemical integrity of the system.Changing the solvent to toluene (entry 2) and ether (entry 3) improved this situation.We finally obtained a synthetically useful ratio of stereoisomers 17a : 18a by using ether at low temperature (entry 4).It is crucial to keep the temperature down (c.f.entry 5) and to limit the reaction time because the stereochemical integrity of the product erodes with time and side products start to appear.Dimeric ethers in particular form with time, sign that a carbocation is involved.In addition, water must be excluded from the medium because it slows the reaction considerably.Wet ether led only to traces of products in the same reaction time (entry 6).With these conditions found, we looked at the scope of the reaction.Ratios of regioisomers were excellent across the board for different alkyl groups (Table 2, entries 1 to 10).We were happy to observe that even a t-butyl group cannot compete with the menthyl moiety for supremacy over the control of the equilibrium mixture of regioisomers (Table 2, entries 3 and 4).Notably, both stereoisomeric alcohols 16 and 19 could be successfully rearranged in a highly regioselective and stereospecific manner (entries 1 to 15, and 16-17).3]17 Either starting alcohol 16 or 19 is easily obtained pure from menthylaldehyde, 16 which gives access to either stereochemistry of the desired product from the same enantiomer of the chiral auxiliary.Alternatively, the chiral auxiliary is also cheaply available in either configuration from the corresponding menthol.© AUTHOR(S) A silyl-protected alcohol gave a high yield of the desired product (entries 13 and 14).Yet, protection of a spectator alcohol is not necessary, although a higher temperature and longer reaction time are required to reach equilibrium (entries 11 and 12).This is probably due to the competing formation of the rhenate ester with the spectator alcohol, so the yields of desired products 17f and 18f were lower, and some unidentified by-products were observed.The presence of an azide was also tolerated (entry 15), but the catalyst loading had to be increased to 10% otherwise the reaction was too slow and what we believe are nitrene-derived byproducts started to appear.A trifluoromethyl gave a 9 : 1 regioisomeric ratio in favor of the desired regioisomers 17i or 18i (entries 16 and 17).This potent electron-withdrawing group renders the alcohol poorly nucleophilic and slows the reaction down.The reaction took 1 h to reach equilibrium in refluxing dichloroethane.Despite the elevated temperature, the stereospecificity remained high.The strongly electron-withdrawing trifluoromethyl group undoubtedly prevents the formation of a carbocation, keeping the stereochemical integrity of the system intact even at 84 ºC.Compound 17i was crystalline and a single crystal X-Ray diffraction analysis confirmed its structure.In addition, this X-ray confirmed the carbinol stereochemistry of 16i (and hence, 19i) because the mechanism of the rhenium(VII)-catalyzed rearrangement is known to be concerted. 3The stereochemistry of 16 is expected from the Felkin-Anh model of addition to chiral aldehydes and we had prepared 16a-b by this method. 16Rearranging both isomers 16 and 19 allows us to ascertain the stereochemical purity of the products 17 and 18.The stereochemistry of 17 and 18 was further confirmed by converting 17c and 17j to known compounds and measuring their optical rotation (vide infra).
The large trimethylsilyl group gave the starting regioisomer 16j (entry 18) mixed with the epimerized product 19j.This group is not only large but being electropositive, it prefers a sp 2 to a sp 3 carbon.Predictably, a conjugating group like phenyl (16k, entry 19) gave epimerization and no desired regioisomer 17k.
As we had with our previous systems, 16 we envisaged cleaving the auxiliary in two main ways accessing useful synthons directly in the same step.We anticipated cleavage efficiencies to be similar to previous systems.Indeed, this was the case.The first example starts with the acylation of enantiomerically pure alcohol 17c (Scheme 4).RCM cleavage of the resulting vinyl ester yielded furanone 20.This compound has been successfully converted to cognac lactone by dimethylcuprate addition. 19This is thus a 5-steps formal synthesis [20][21][22][23] of this prized natural aromatic oil 24 from 1-iodohept-1-ene.
The second example is the oxidative cleavage of alcohol 17j.Ozonolysis and in situ oxidation supplied the chiral -hydroxy acid 21. 25 Chiral -hydroxy acids are widely used synthons and are also used as depsipeptides monomers. 26We converted -hydroxyacid 21 to lactam 22 by standard means and used this lactam to make the non-racemic version of a novel glycoside hydrolase inhibitor. 27heme 4. Cleavage of the auxiliary into useful synthons.

Conclusions
We have successfully extended our menthylaldehyde-based chiral auxiliary method to encompass secondary chiral allylic alcohols.Starting from alkynes, chiral non-racemic O-heterocycles and -hydroxy acids can be © AUTHOR(S) obtained regio-and stereospecifically in 3-5 steps.We are further expanding the method to access chiral non racemic tertiary alcohols.These are much trickier because of the easy formation of a carbocation and loss of stereochemical integrity.To the best of our knowledge, there are yet no example of a rhenium-catalyzed allylic rearrangement to make a chiral non-racemic tertiary allylic alcohol.We are designing a new chiral auxiliary to tackle this issue and will report the results in due course.

Experimental Section
General.All reactions were performed under an inert atmosphere of argon in glassware that had been flamedried, or oven dried overnight.Solvents were distilled from lithium aluminum hydride (tetrahydrofuran, ether), sodium/benzophenone (toluene), calcium hydride (DCM, triethylamine) prior to use.Proton nuclear magnetic resonance ( 1 H NMR) spectra were recorded on a 300 MHz Bruker spectrometer.NMR samples were dissolved in chloroform-d and chemical shifts are reported in ppm relative to the residual undeuterated solvent.Data are reported as follows: chemical shift, integration, multiplicity (s = singlet, d = doublet, dd = doublet of doublets, ddd = doublet of doublets of doublets, t = triplet, q = quartet, m = multiplet), coupling constant.Carbon nuclear magnetic resonance ( 13 C NMR) spectra were recorded on a 75.5 MHz Bruker spectrometer.NMR samples were dissolved in chloroform-d and chemical shifts are reported in ppm relative to the solvent.High-resolution spectrometry was performed on a Shimadzu LC-QqTOF Nexera.Reactions were monitored by thin-layer chromatography (TLC) on Silicycle 0.25 mm silica gel coated glass plate UV 254, vanillin, KMnO4, PMA, or by 1 H NMR. Silica gel (particule size: 230-400 mesh) was used for flash chromatography.Melting points are uncorrected. ]
Ester 30 (40 mg, 0.13 mmol) was added to a oven-dried round-bottom two-neck flask equipped with a stir bar and a condenser under N2 atmosphere.Dry toluene (11 mL) was added, and the remaining oxygen was removed by applying 3 freeze-thaw cycles.The solution was heated to reflux and a solution of 2 nd generation Hoveyda-Grubbs catalyst (7.8 mg, 12 µmol) in dry toluene (1.2 mL) was added over 30 min.using a syringe pump.The reaction mixture was stirred under reflux for an additional 3 h.The reaction mixture was cooled to rt and DMSO (44 µL, 0.62 mmol) was added.The reaction mixture was stirred at rt for 16 h.Solvent was removed under reduced pressure to yield the crude product as a brown mixture of oil and solids (141 mg).The crude product was purified by flash chromatography eluting with a mixture of hexanes and ethyl acetate (8:2) to yield the pure desired product 20 as a colorless oil (12 mg, 63%).

Table 1 .
Optimization of the reaction condition of the Re-catalyzed transposition of allylic alcohol 16a a. Isolated yields of 17a and 18a.