This is the accepted version of the journal article: Cabré, Albert; Rafael, Sergi; Sciortino, Giuseppe; [et al.]. «Catalytic regios- elective isomerization of 2,2-disubstituted oxetanes to homoallylic alcohols». Angewandte Chemie, Vol. 59, issue 19 (May 2020), p. 7521-7527. 8 pàg. DOI 10.1002/anie.201915772 This version is available at https://ddd.uab.cat/record/279264 under the terms of the license RESEARCH ARTICLE Catalytic Regioselective Isomerization of 2,2-Disubstituted Oxetanes to Homoallylic Alcohols Albert Cabré,[a,b] Sergi Rafael,[b] Giuseppe Sciortino,[c,d] Gregori Ujaque,[c] Xavier Verdaguer[a,b] Agustí Lledós*[c] and Antoni Riera*[a,b] Dedication ((optional)) Abstract: The selective isomerization of strained heterocyclic compared to epoxides. One of the reasons is that 3-substituted compounds is an important tool in organic synthesis. An oxetanes show a significantly lower electrophilicity than unprecedented regioselective isomerization of 2,2-disubstituted epoxides so they usually have to be activated by Lewis oxetanes to homoallylic alcohols is described. The use of acids.[8a,10] tris(pentafluorophenyl)borane (B(C6F5)3), a commercially available Despite of their lower reactivity than epoxides, significant Lewis acid was key to obtain good yields and selectivities since chemistry of 3-substituted oxetanes has been developed other Lewis acids afforded mixtures of isomers and substantial focusing on ring-opening reactions[8a,11], including asymmetric polymerization. The reaction took place under exceptionally mild versions,[10] and ring-expansion reactions.[12] conditions and very low catalyst loading (0.5 mol %). DFT On the other hand, 2,2-disubstituted oxetanes are more reactive calculations disclose the mechanistic features of the isomerization than the 3-substituted ones due to their facile ring-opening. and account for the high selectivity displayed by the B(C6F5)3 catalyst. However, they are also more sensitive to hydrolysis and The synthetic applicability of the new reaction is demonstrated by polymerization. The main reactivity of 2,2-disubstituted oxetanes the preparation of -chiral alcohols using iridium catalyzed include reductive ring-opening by titanium[13] or iron[14] asymmetric hydrogenation. complexes, ring expansion[15] to five-membered cyclic ethers and perhydrolysis[16] (Figure 1). The oxetanes have also been used as a lithiation directing groups in aromatic rings. [11,17] Introduction Oxetanes are present in numerous drugs and natural products, such as Taxol (Paclitaxel), a well-known antitumoral.[1] They have found significant application in medicinal chemistry[2] since the introduction of an oxetane fragment usually has positive effects in terms of solubility and metabolic stability. [2,3] More recently, 3-substituted oxetanes have been used in polymer chemistry as monomers[4] or cross-linkers,[5] and in opto- electronic devices.[6,7] The development of new and efficient methods for the preparation of oxetanes has enhanced their role as versatile building blocks in synthetic chemistry.[8,9] Although the ring strain energy of oxetanes (25.4 kcal·mol-1) is only slightly lower than epoxides (26.8 kcal·mol-1), they have been scarcely used [a] A. Cabré, S. Rafael, Prof. X. Verdaguer and Prof. A. Riera, Institute of Research in Biomedicine (IRB Barcelona). The Figure 1. Reactivity of 2,2-disubstituted oxetanes. Barcelona Institute of Science and Technology, Baldiri Reixac 10, Barcelona 08028, Spain E-mail: antoni.riera@irbbarcelona.org [b] A. Cabré, Prof. X. Verdaguer and Prof. A. Riera Departament de Química Inorgànica i Orgànica, Secció Orgànica. Our group has recently explored the metal-catalyzed Universitat de Barcelona, Martí i Franquès 1, Barcelona 08028, isomerization of strained heterocyclic compounds (Figure 2). Spain. Particularly, we reported the selective iridium-catalyzed [c] G. Sciortino, Prof. G. Ujaque and Prof. A. Lledós, Departament de Química, Universitat Autònoma de Barcelona, Cerdanyola del isomerization of N-sulfonyl aziridines to allylic amines using Vallès, Barcelona E-08193, Spain. Crabtree’s catalyst (1) under very mild conditions and without E-mail: agusti@klingon.uab.es. external activation.[18] Similarly, the conversion of epoxides into [d] G. Sciortino Dipartimento di Chimica e Farmacia, Università di Sassari, Via Vienna 2, I-07100 Sassari, Italy aldehydes was achieved by hydrogen activation of reagent 1.[19] Inspired on these selective transformations, we wondered if 2,2- Supporting information for this article is given via a link at the end disubstituted oxetanes would be selectively isomerized when of the document.((Please delete this text if not appropriate)) using the appropriate catalyst. RESEARCH ARTICLE Gratifyingly, the reaction in dichloromethane afforded the homoallylic alcohol with excellent yield (82 %) and selectivity (3a:4a: 98:2, entry 8, Table 1). With this result in hand, we performed a solvent screening. EtOAc, THF and toluene gave excellent conversion and selectivity but the yield was not enhanced (entries 9-11, Table 1). Acetonitrile gave very low conversion (entry 12). The temperature effect was also evaluated and the reaction was performed at 0 ºC in dichloromethane (entry 13). However, no difference with the Figure 2. Selective isomerization of heterocyclic compounds catalyzed by Crabtree’s catalyst, 1. reaction performed at room temperature was observed. Finally, we reduced the catalyst loading to 0.5 mol %, and the reaction time to 2 h, thus demonstrating the outstanding catalytic activity In contrast with the regioselective isomerization of epoxides [20] of B(C6F5)3 (entry 14, Table 1). In addition, a gram scale reaction (also called Meinwald rearrangement) the regioselective was carried out, affording 3a with an excellent 84% yield (entry isomerization of oxetanes still remains a challenge and, to the 15). best of our knowledge, no synthetic protocols have been described to date. Most probably, the easy polymerization of 2- Table 1. Screening of conditions for the isomerization of 2a substituted oxetanes in acidic conditions has hampered their synthetic use. Herein, we report a new catalytic reaction that affords homoallylic alcohols from 2,2-oxetanes in high selectivity using very low loading of tris(pentafluorophenyl)borane, B(C6F5)3, as Lewis acid. The subsequent asymmetric hydrogenation allows the preparation of enantioenriched -substituted alcohols. Since Ratio Yield 3a Entry Catalyst Solvent Conv. (%)[a] 3a:4a[a] (%)[b] the preparation 2,2-disubstituted oxetanes using the Corey- Chaykovski reaction[21] is straightforward and environmentally 1 1 DCM - 35 25 friendly, we believe that the new sequence emerges as a 2[c] 1 DCM - >99 0 [d] breakthrough in the synthesis of chiral alcohols from ketones. 3 ZnCl2 DCM - >99 0 [d] 4 InCl3 DCM 4:1 50 39 5 IrCl3 DCM 4:1 >99 34 6 AlCl3 DCM 5:1 >99 51 Results and Discussion 7 BF3·Et2O DCM 1:2 >99 30 8 B(C6F5)3 DCM 98:2 >99 82 Our preliminary studies began with the synthesis of 2-phenyl-2- 9 B(C6F5)3 EtOAc 96:4 >99 70 methyloxetane 2a, which was obtained from acetophenone via 10 B(C6F5)3 THF 98:2 >99 67 double Corey-Chaykovsky reaction[21] with good yields without 11 B(C6F5)3 Toluene 98:2 >99 79 the need of further purification (see the Supporting Information). 12 B(C6F5)3 MeCN - 33 21 With 2a in hand, the isomerization reaction was performed using 13[e] B(C6F5)3 DCM 98:2 >99 80 several catalysts at 5 mol % loading at room temperature, as 14[f] B(C6F5)3 DCM 98:2 >99 82, 78[g] shown in Table 1. First, we tested Crabtree’s catalyst 1 as it 15[h] B(C6F5)3 DCM 98:2 >99 84 gave satisfactory results when using N-sulfonyl aziridines[18] and The reaction was performed in a sealed vial with 0.1 mmol of 2a, [0.1 M] and epoxides.[19] However, we observed that, without any external using 5 mol % of catalyst loading. [a] Determined by 1H NMR spectroscopy. [b] 1 H NMR yield using mesitylene as internal standard. [c] The reaction was activation, conversion was very low (entry 1, Table 1) being the performed in a pressure tube, using H2 for catalyst activation and, after 1 min, only product that could be quantified the homoallylic alcohol 3a. the vessel was fully degassed. [d] Polymerization occurred. [e] The reaction was Activation of the catalyst with H2 to form an Ir-H,H as the performed at 0 ºC. [f] 0.5 mol% of catalyst was employed, and the reaction was catalytic active species led to polymerization (entry 2, Table 1). left stirring for 2 h. [g] Isolated yield, at 0.8 mmol scale. [h] Isolated yield These disappointing results led us to study a wide range of performing the reaction at gram scale, using 0.5 mol% of B(C6F5)3. Lewis-acids. Again, ZnCl2 gave rise to polymerization (entry 3). With the optimal conditions in hand, we explored the substrate When moving to milder, inorganic Lewis acids such as InCl3, scope (Table 2). The aryl group was modified with a range of IrCl3 or AlCl3, 3a was the major product but the mass balance of functionalities. A fluorine atom in para- and ortho-position was the reaction was poor, probably due to polymerization (entries 4- well tolerated (entries 1 and 2, Table 2), affording 3b and 3c with 6, Table 1). At that point, we turned our attention to organic complete regioselectivity. In the latter case, the yield was Lewis-acids. First, BF3·Et2O was tested. Full conversion was moderate probably due to dimerization of the substrate. A observed, although the selectivity of the reaction was null: an substrate with stronger electron-withdrawing group such as p- equimolar mixture of 3a:(E)-4a:(Z)-4a (entry 7, Table 1) was CF3 (entry 3, Table 2) afforded the homoallylic alcohol 3d with afforded. We then tested a bulky Lewis acid namely B(C6F5)3.[22] high selectivity, albeit in moderate yield. Para-substituted RESEARCH ARTICLE halogen substituents such as p–Cl, p–Br and p-I showed from the product by column chromatography. [g] A mixture of homoallylic excellent selectivity affording the homoallylic alcohols 3e, 3g and alcohols (7:3) was also obtained. (see the Supporting Information). 3h in high yield (entries 4-7, Table 2). In the case of m–Cl (2f), the yield was high (84%, entry 5, Table 2), but up to 5% of the To prove the versatility of this reaction, we also modified the allylic alcohols 4 were also formed. We then moved on to test methyl group to other alkyl groups, such as an ethyl (2q) or the effect of the electron-donating group in the aryl substituent. benzyl group (2r). The corresponding trisubstituted homoallylic A methyl group was introduced in both para- and ortho-position alcohols 3q and 3r were formed in good yields, with high E (entries 8 and 9, Table 2). In both cases, the selectivity of the selectivity with respect to Z (Scheme 1). Also, bicyclic oxetane isomerization reaction decreased compared to EWG groups, 2s, synthesized from α-tetralone, proved highly reactive, although 3i and 3j were still obtained in good yields. exclusively affording homoallylic alcohol 3s in 84% isolated yield Alternatively, a methoxy group in the meta-position enhanced (Scheme 1). Finally, the applicability of this methodology is the selectivity ratio 3:4 up to 98%, affording 3k with an excellent exemplified by the synthesis of estrone derivative 3t, which is a 92% yield (entry 10, Table 2). Furthermore, other relevant potential novel precursor for the development of further bioactive functionalities on the aromatic core were tested, such as nitrile compounds. For this purpose, starting from estrone 3-methyl (2l) and amide (2m). For both cases, homoallylic alcohols 3l and ether, the corresponding oxetane 2t was synthesized in high 3m were obtained in excellent selectivity and in good yields yield affording a mixture of diastereomers (see the Supporting (entries 11 and 12, Table 2), albeit harsher reactions conditions Information) which was directly treated with 0.5 mol% of (2 mol% catalyst loading and leaving the reaction overnight, at B(C6F5)3. After 5 h, the homoallylic alcohol 3t was obtained with room temperature or heating) were required. In contrast, bulky a 78% yield. The synthesis of this novel compound shows a new group in para-position (p-iBu) diminished the selectivity, derivatization pathway for steroids and related compounds. although 3n was afforded in synthetically useful yield (entry 13, Table 2). The naphthyl group was also studied, affording 3o in excellent selectivity (92:8) and 83% yield (entry 14, Table 2). Interestingly, the reactivity was not limited to 2-aryl oxetanes. The dialkyl compound 2-cyclohexyl-2-methyl oxetane 2p (entry 15, Table 2) afforded a 7:3 mixture of the expected product 3p and the tetrasubstituted allyl alcohol in 56% yield (see the Supporting Information). Table 2. Study of the substrate scope. Conv. Yield 3 Ratio Entry R (%)[b] 3:4[a] (%)[a] 1 2b p-F-C6H4- 97:3 >99 77[f] 2[c] 2c o-F-C6H4- >99 >99 57 Scheme 1. Substrate scope with other alkyl substituents. 3 2d p-CF3-C6H4- 99:1 >99 54 4 2e p-Cl-C6H4- >99 >99 95 The novelty of the transformation led us to perform a 5 2f m-Cl-C6H4- 95:5 >99 84 mechanistic study. Our previous studies of iridium-catalyzed 6 2g p-Br-C6H4- >99 >99 86 isomerization of N-sulfonyl aziridines and epoxides (Figure 2) 7[c] 2h p-I-C6H4-) >99 >99 70 disclosed a two-steps mechanism in which the initial ring- 8 2i p-Me-C6H4- 96:4 >99 79 opening step is followed by metal-assisted tautomerization.[18,19] 9 2j o-Me-C6H4- 91:9 >99 63 The selectivity of the process depends on the C-H hydrogen 10 2k m-OMe-C6H4- 98:2 >99 92 involved in the tautomerization. However, the metal-free nature 11[d] 2l m-CN-C6H4- >99 >99 55 of the oxetane isomerization raises new questions regarding the 12[e] 2m m-CONMe2-C6H4- >99 >99 68 hydrogen migration step, in particular regarding the identity of 13 2n p-iBu-C6H4- 88:12 >99 70[f] the agent assisting the H-migration and how the selectivity 14 2o 2-Naphthyl 92:8 >99 83[f] homoallylic alcohol (3)/allylic alcohol (4) is defined. Indeed, the 15 2p cyclohexyl 99:1 >99 56[g] results obtained using the two borane reagents [BF3·Et2O and B(C6F5)3] differ markedly. While the former yields an equimolar The reaction was performed in a sealed vial using 0.4 mmol of 2 and 0.5 mol % of catalyst loading. [a] Determined by 1H NMR spectroscopy from the mixture of both products, the latter is highly selective toward the crude. [b] Isolated yield. [c] 1 mol% of catalyst was used, leaving the reaction homoallylic product (Table 1). To answer these questions, here stirring overnight. [d] 2 mol% of catalyst was used, heating to 40 ºC and leaving we studied the isomerization mechanism of 2-phenyl-2- the reaction stirring overnight. [e] 2 mol% of catalyst was used, leaving the methyloxetane 2a catalyzed by both BF3·Et2O and B(C6F5)3 reaction stirring overnight. [f] Allylic alcohols 4 could not be completely removed Lewis acids. To this end, DFT calculations[23] applying the RESEARCH ARTICLE B3LYP-D3 functional and treating the DCM solvent with the Details in the Supporting Information). SMD continuum model were performed (see Computational Scheme 2. DFT computed mechanism (B3LYP-D3 in DCM) for the isomerization of oxetane 2a to homoallylic 3a and allylic (E)-4a and (Z)-4a alcohols catalyzed by a) BF3·Et2O and b) B(C6F5)3. Relative Gibbs energies (in red) in kcal·mol-1 are referred to the separated species; transition state energies are shown in purple, for the ring opening step and orange, blue or green for the proton transfer process entailing the formation of 3a, (E)-4a and (Z)-4a, respectively. As BF3·Et2O gave similar yields of both products, we started the three species have been computed leading to an energy mechanistic study using this Lewis acid reagent. The computed ordering for the Gibbs energy barrier from II that follows the catalytic cycle is shown in Scheme 2a. The corresponding Gibbs expected basicity order: O-II (8.1 kcal·mol-1 ) < O-2a (13.6 energy profile can be found at the Supporting Information kcal·mol-1 ) < O-Et2O (15.6 kcal·mol-1 ) (see the Supporting (Figure S3). The formation of the Lewis-adduct 2a·BF3 (I) Information). The favored intermolecular II-assisted TS bringing requires the previous displacement of Et2O by 2a, which takes 4a comes from a head-to-tail adduct of two unities of places by means of a SN2 substitution process demanding 12.7 intermediate II (Figure 3a). kcal·mol-1. Coordination of boron to the O-oxetane activates the C-O bond for ring opening, that only requires an activation barrier of 4.0 kcal·mol-1 (TSI-II) and leads to the zwitterionic intermediate II (-4.9 kcal·mol-1). At this point, the catalytic cycle can follow two different pathways depending on the proton that migrates to the oxygen: proton transfer from methyl group brings the homoallylic alcohol 3a, while if it happens from the C β H2 group the allylic products (E,Z)-4a are formed. The process leading to 3a has been characterized as a low barrier (8.3 kcal·mol-1) intramolecular proton transfer involving a six-member ring transition state, TSII-IV. On the contrary, the formation of the allylic products 4a via an intramolecular mechanism, involving a four-member ring transition state, can be discarded as it presents a Gibbs energy barrier of 23.8 kcal·mol-1 (see the Supporting Information), incompatible with the experimental detection of equimolar amounts of 3a and 4a products (Table 1). Therefore, we investigated the possible role of all the bases present in the reaction medium in assisting the proton migration from the CβH2 group to the oxygen. Three O-bases were tested in the intermolecular base-assisted H transfer: the ether oxygen Figure 3. Optimized geometries for: a) adduct of two unities of intermediate II; of Et2O, that of oxetane substrate 2a, and alkoxide oxygen of b) concerted intermolecular proton migration transition state, TSII-VI; c) adduct intermediate II. Base-assisted transition states involving the of two unities of intermediate IIb, and d) first step of the proton migration RESEARCH ARTICLE transition state, TSII-Vb. The most important distances are also reported in Å. products, pharmaceuticals, and fine chemicals of perfume Hydrogen atoms of Ph groups have been omitted for clarity. industry.[25] Enantioenriched -methyl alcohols are also versatile building blocks that are used as synthetic intermediates in the The proton migration has been characterized as a concerted synthesis of complex molecules.[3] For instance, (-)-citronellol is mechanism in which the Cβ-H proton, picked by the oxygen of an intermediate for the production of rose oxide. [26] one unity of intermediate II, is directly transferred to the oxygen of the second unity to form the allylic alcohol (Figure 3b). Both Focusing on catalytic processes, the asymmetric isomerization pro-(E), TSIII-V and pro-(Z), TSIII-VI, show barriers (8.1 and 9.2 and/or hydrogenation of primary allylic alcohols have emerged kcal·mol-1, respectively) very close to TSII-IV (8.3 kcal·mol-1). For as promising alternatives to synthesize γ-methyl chiral alcohols, sake of completeness, the transition state of the base-II assisted where iridium-P,N complexes have come to dominate the proton transfer leading to the major product 3a, has been also field.[25] Despite of this, serious drawbacks remain; the catalytic characterized. The intermolecular process has an energy isomerization[27] of γ-methyl allylic alcohols suffer from very low barrier of 9.9 kcal·mol-1, 1.6 kcal·mol-1 higher than the conversion and moderate enantioselectivities or the asymmetric intramolecular TSII-IV. Taking these data into account, from a hydrogenation[28] requires harsh conditions (very high H2 micro-kinetics analysis carried out with COPASI software,[24] a pressure) to avoid undesired isomerization byproducts. 1.0:1.6 mixture of 3a:4a is predicted, highlighting no selectivity of Moreover, the traditional preparation of the starting allylic is the catalyst (for further details see the Supporting Information). highly inefficient. First, their synthesis usually require two steps with poor atom economy: an olefination reaction (Horner– When using B(C6F5)3 (Scheme 2b and Gibbs energy profile at Wadsworth–Emmons -HWE-, or similar), and ester reduction the Supporting Information, Figure S2) no previous substitution (usually with reactive hydrides). Second, even more important, is required. According with the lower Lewis acidity of B(C 6F5)3 the usual poor selectivity of the HWE with ketones leads to the respect to BF3, its interaction with the O-oxetane is weaker and formation of mixtures of E and Z allylic alcohols that require a bit higher barrier (11.6 kcal·mol-1, TSI-IIb) is required in the ring- complicated chromatographic separations (Scheme 3). For this opening step that leads to the zwitterionic intermediate IIb at -4.4 reason, we believe that the strategy presented in this work is kcal·mol-1. From IIb the homoallylic alcohol 3a is obtained clearly advantageous to obtain chiral alcohols from cheap and through an intramolecular H-transfer involving a 6-member ring abundant ketones in a greener 3-step synthetic approach. To transition state TSII-IVb, requiring a Gibbs energy barrier of 12.4 demonstrate that, we moved to perform the asymmetric kcal·mol-1. As with the BF3·Et2O catalyzed process, the hydrogenation of homoallylic alcohols 3, which can be intermolecular base-II assisted mechanism displays a higher considered minimally functionalized olefins. [29] barrier of 16.0 kcal·mol-1. As for BF3, the formation of the allylic products via an intramolecular mechanism, entailing a four- member ring transition state, implies high activation energy (21.3 kcal·mol-1, see the Supporting Information). Regarding the base- assisted mechanism, the same basicity trend than for BF 3 reaction has been observed: the II-assisted process is favored over the 2a-assisted proton transfer, with barriers of 16.9 and 18.5 kcal·mol-1, respectively. In contrast to the process with the BF3 catalyst, the proton migration needs two different steps: first, with an activation barrier of 16.9 kcal·mol -1, the Cβ-H proton is transferred to the oxygen of the second unity of intermediate IIb (TSIII-Vb); then, in a practically barrierless process, involving a relative reorientation of the two subunits, the proton migrates to Scheme 3. Comparison of this work with the traditional approach. form the allylic product 4a (see the Supporting Information). Contrary to what happen for the case of BF3 catalyzed During the last decade our group has developed several isomerization, TSIII-Vb has an energy barrier higher enough with modular P stereogenic chiral ligands [30] that have proved to be respect the transition state TSII-IVb (4.5 kcal·mol-1) to assure a excellent precursors of chiral catalysts. Although our iridium-P,N practically complete (99.5 %) selectivity in favor of the MaxPHOX[31] family of catalysts (6) have been successfully homoallylic alcohol product 3a, in line with the experimental applied to the asymmetric hydrogenation of cyclic enamides, [31a] results. The reasons behind the different performances of the aryl and alkyl imines[32] and minimally functionalized olefins, [33] two catalysts can be found on the steric hindrance introduced by including 2-aryl N-allyl phthalimides,[34] we found that in this the aryl rings of B(C6F5)3 that destabilizes the proton migration particular case, using 3a as model substrate, [((4S,5S)-Cy2- step by placing the oxygen atom that will accept the proton Ubaphox)Ir(COD)]BArF4 7[35] gave the best results. further away (Figure 3c). Table 3. Iridium-catalyzed asymmetric hydrogenations of homoallylic alcohols. Last, to showcase the applicability of the novel synthetic protocol, we performed the enantioselective hydrogenation of the resulting homoallylic alcohols to give enantioenriched alcohols. Alcohols with an stereogenic center in γ-position constitute an important chemical motif, present in a wide range of natural RESEARCH ARTICLE Acknowledgements This work was supported by the Spanish Ministry of Economy, Industry and Competitiveness (MINECO) through grants CTQ2017-87840-P and CTQ2017-87889-P. We also acknowledge institutional funding from MINECO through the Centres of Excellence Severo Ochoa award given to IRB Barcelona and from the Catalan Government through the CERCA Program. A.C. thanks MINECO for a FPU PhD Entry Catalyst R Product Conv. (%) [a] ee (%) [b] fellowships. G.S. tanks Regione Autonoma della Sardegna 1 3a 6 H 5a >99 71 (grant RASSR79857) for a post-doctoral fellowship. 2 3a 7 H 5a >99 90 3 3c 7 o-F 5c >99 94 4 3e 7 p-Cl 5e >99 92 Keywords: catalysis • isomerization • oxetanes • allyl alcohols • 5 3i 7 p-Me 5i >99 90 asymmetric hydrogenation 6 3k 7 m-OMe 5k >99 94 The reaction was performed in a pressure reactor using 1 mol% of catalyst 6 and left [1] (a) M. C. Wani, H. L. Taylor, M. E. Wall, P. Coggon, A. T. Mcphail, J. [a] 1 [b] Am. Chem. Soc. 1971, 93, 2325–2327. (b) M. Wang, B. Cornett, J. stirring overnight. Determined by H NMR spectroscopy from the crude. Measured by Nettles, D. C. Liotta, J. P. Snyder, J. Org. Chem. 2000, 65, 1059–1068. chiral HPLC. [2] (a) J. A. Burkhard, G. Wuitschik, M. Rogers-Evans, K. Müller, E. M. Carreira, Angew. Chem. Int. Ed. 2010, 49, 9052–9067. (b) G. Wuitschik, Using 1 mol % of this commercially available catalyst 7 E. M. Carreira, B. Wagner, H. Fischer, I. Parrilla, F. Schuler, M. Rogers- developed by Pfaltz and co-workers in dichloromethane under 1 Evans, K. Müller, J. Med. Chem. 2010, 53, 3227–3246. (c) G. Wuitschik, bar of H2 pressure, 5a was obtained in 90% ee (entry 2, Table 3). M. Rogers-Evans, K. Müller, H. Fischer, B. Wagner, F. Schuler, L. The scope of this reaction was expanded to other homoallylic Polonchuk, E. M. Carreira, Angew. Chem. Int. Ed. 2006, 45, 7736–7739. alcohols prepared from the corresponding oxetanes 2. Both (d) M. McLaughlin, R. Yazaki, T. C. Fessard, E. M. Carreira, Org. Lett. 2014, 16, 4070–4073. electron-withdrawing substituents such as o-F, (3c, entry 3) and [3] (a) J. A. Burkhard, G. Wuitschik, J. M. Plancher, M. Rogers-Evans, E. p-Cl (3e, entry 4) and electron-donating substituents such as p- M. Carreira, Org. Lett. 2013, 15, 4312–4315. (b) A. F. Stepan, K. Karki, Me (3i, entry 5) and m-OMe (3k, entry 6) gave full conversions W. S. McDonald, P. H. Dorff, J. K. Dutra, K. J. Dirico, A. Won, C. and excellent enantioselectivities (up to 94% ee). Subramanyam, I. V. Efremov, C. J. O’Donnell, et al., J. Med. Chem. 2011, 54, 7772–7783. [4] J. V. Crivello, J. Polym. Sci. Part A Polym. Chem. 2014, 52, 2934–2946. [5] B. Ghosh, M. W. Urban, Science 2009, 323, 1458-1460. Conclusion [6] H. Kudo, T. Nishikubo, J. Polym. Sci. Part A: Polym. Chem. 2007, 45, 709–726 In conclusion, we have reported a highly regioselective [7] T. Soga, H. Kudo, T. Nishikubo, S. Sato, Polym. J. 2008, 40, 832–836. isomerization of 2,2-disubstituted oxetanes using B(C6F5)3 as [8] (a) J. A. Bull, R. A. Croft, O. A. Davis, R. Doran, K. F. Morgan, Chem. catalyst, using extremely mild conditions (0.5 mol% of catalyst Rev. 2016, 116, 12150–12233. (b) O. A. Davis, J. A. Bull, Angew. loading and very short reaction times). DFT calculations shed Chem. Int. Ed. 2014, 53, 14230–14234. (c) O. A. Davis, R. A. Croft, J. light on the reaction mechanism and account for the high A. Bull, Chem. Commun. 2015, 51, 15446–15449. For enantioselective selectivity toward the homoallylic alcohol product displayed by synthesis of oxetanes, see: (d) T. Sone, G. Lu, S. Matsunaga, M. Shibasaki, Angew. Chem. Int. Ed. 2009, 48, 1677–1680. the B(C6F5)3 catalyst. This novel selective isomerization of [9] S. Ahmad, M. Yousaf, A. Mansha, N. Rasool, A. F. Zahoor, F. Hafeez, oxetanes can be used as key step in multiple synthetic organic S. M. A. Rizvi, Synth. Commun. 2016, 46, 1397–1416. transformations. As a leading example, we have disclosed a [10] (a) Z. Wang, Z. Chen, J. Sun, Org. Biomol. Chem. 2014, 12, 6028– highly efficient 3-step protocol towards the enantioselective 6032. (b) Z. Wang, Z. Chen, J. Sun, Angew. Chem. Int. Ed. 2013, 52, synthesis of γ-aryl butanols starting from abundant inexpensive 6685–6688. (c) Z. Chen, Z. Wang, J. Sun, Chem. Eur. J. 2013, 19, acetophenone derivatives. First, a double Corey-Chaykovksy 8426–8430. (d) R. N. Loy, E. N. Jacobsen, J. Am. Chem. Soc. 2009, reaction, which only generates dimethyl sulfoxide as by-product, 131, 2786-2787. is carried out. Then, and without further purification, the selective [11] R. A. Croft, J. J. Mousseau, C. Choi, J. A. Bull, Chem. Eur. J. 2016, 22, 16271–16276. isomerization of the resulting oxetanes takes place followed by [12] For the ring expansion of oxetanes to carbonates, see: (a) C. J. iridium-catalyzed asymmetric hydrogenation (Figure 3). These Whiteoak, E. Martin, M. M. Belmonte, J. Benet-Buchholz, A. W. Kleij, last two steps, which are catalytic, accomplishes a greener, Adv. Synth. Catal. 2012, 354, 469–476. (b) J. Rintjema, A. W. Kleij, more sustainable and efficient approach to the enantioselective Synth. 2016, 48, 3863–3878. (c) W. Guo, V. Laserna, J. Rintjema, A. W. synthesis of valuable alcohols. Moreover, the applicability of the Kleij, Adv. Synth. Catal. 2016, 358, 1602–1607. (d) J. Rintjema, W. Guo, isomerization reaction was further demonstrated by the E. Martin, E. C. Escudero-Adán, A. W. Kleij, Chem. Eur. J. 2015, 21, isomerization of spiranic oxetanes and the preparation of a new 10754–10762. For another recent ring-expansion example of 3- estrone-derivative. substituted oxetanes, see: S. Jana, Z. Yang, C. Pei, X. Xu, R. M. Koenigs, Chem. Sci. 2019,10, 10129-10134. [13] (a) A. Gansäuer, N. Ndene, T. Lauterbach, J. Justicia, I. Winkler, C. Mück-Lichtenfeld, S. Grimme, Tetrahedron 2008, 64, 11839–11845. (b) RESEARCH ARTICLE N. Takekoshi, K. Miyashita, N. Shoji, S. Okamoto, Adv. Synth. Catal. [33] M. Biosca, E. Salomó, P. De La Cruz-Sánchez, A. Riera, X. Verdaguer, 2013, 355, 2151–2157. O. Pàmies, M. Diéguez, Org. Lett. 2019, 21, 807–811. [14] Y. K. Sugiyama, S. Heigozono, S. Okamoto, Org. Lett. 2014, 16, 6278– [34] A. Cabré, E. Romagnoli, P. Martínez-Balart, X. Verdaguer, A. Riera, 6281. Org. Lett. 2019, 21, 9709-9713. [15] (a) E. D. Butova, A. V. Barabash, A. A. Petrova, C. M. Kleiner, P. R. [35] F. Menges, A. Pfaltz, Adv. Synth. Catal. 2002, 344, 40–41. Schreiner, A. A. Fokin, J. Org. Chem. 2010, 75, 6229–6235. (b) C. Gronnier, S. Kramer, Y. Odabachian, F. Gagosz, J. Am. Chem. Soc. 2012, 134, 828–831. (c) D. J. Mack, L. A. Batory, J. T. Njardarson, Org. Lett. 2012, 14, 378–381. (d) C. Larksarp, H. Alper, J. Org. Chem. 1999, 64, 4152–4158. (e) S. K. Pawar, D. Vasu, R.-S. Liu, Adv. Synth. Catal. 2014, 356, 2411-2416. [16] (a) P. H. Dussault, T. K. Trullinger, F. Noor-e-Ain, Org. Lett. 2002, 4, 4591–4593. (b) W. B. Han, Y. Wu, Org. Lett. 2014, 16, 5706–5709. [17] (a) D. I. Coppi, A. Salomone, F. M. Perna, V. Capriati, Angew. Chem. Int. Ed. 2012, 51, 7532–7536. (b) D. I. Coppi, A. Salomone, F. M. Perna, V. Capriati, Chem. Commun. 2011, 47, 9918–9920. [18] A. Cabré, G. Sciortino, G. Ujaque, X. Verdaguer, A. Lledós, A. Riera, Org. Lett. 2018, 20, 5747–5751. [19] A. Cabré, J. Cabezas-Giménez, G. Sciortino, G. Ujaque, X. Verdaguer, A. Lledós, A. Riera, Adv. Synth. Catal. 2019, 3624–3631. [20] J. Meinwald, S. S. Labana, M. S. Chadha, J. Am. Chem. Soc. 1963, 85, 582–585. [21] K. Okuma, Y. Tanaka, S. Kaji, H. Ohta, J. Org. Chem. 1983, 48, 5133– 5134. [22] For the use of B(C6F5)3 in isomeization or ring-opening of epoxides, see: (a) K. Ishihara, N. Hanaki, H. Yamamoto, Synlett 1995, 1995, 721- 722. (b) M. R. Tiddens, R. J. M. Klein Gebbink, M. Otte, Org. Lett. 2016, 18, 3714–3717. (c) M. N. Bhagat, C. K. Bennett, G. F. Chang, Y. Zhu, A. Raghuraman, M. E. Belowich, S. T. Nguyen, L. J. Broadbelt, J. M. Notestein, ACS Catal. 2019, 9, 9663–9670. [23] Gaussian 09 (Revision D.01): Frisch, M. J. et al. Gaussian, Inc., Wallingford CT, 2011. [24] Hoops, S.; Sahle, S.; Gauges, R.; Lee, C.; Pahle, J.; Simus, N.; Singhal, M.; Xu, L.; Mendes, P.; Kummer, U. Bioinformatics 2006, 22, 3067- 3074 [25] L. A. Saudan, Acc. Chem. Res. 2007, 40, 1309–1319. [26] P. L. Alsters, W. Jary, V. Nardello-Rataj, J. M. Aubry, Org. Process Res. Dev. 2010, 14, 259–262. [27] (a) H. Li, C. Mazet, Acc. Chem. Res. 2016, 49, 1232–1241. (b) J. Q. Li, B. Peters, P. G. Andersson, Chem. Eur. J. 2011, 17, 11143–11145. [28] (a) M. Bernasconi, V. Ramella, P. Tosatti, A. Pfaltz, Chem. Eur. J. 2014, 20, 2440–2444. (b) J. Q. Li, J. Liu, S. Krajangsri, N. Chumnanvej, T. Singh, P. G. Andersson, ACS Catal. 2016, 6, 8342-8349. [29] (a) J. J. Verendel, O. Pàmies, M. Diéguez, P. G. Andersson, Chem. Rev. 2014, 114, 2130–2169. (b) S. J. Roseblade, A. Pfaltz, Acc. Chem. Res. 2007, 40, 1402–1411. (c) D. H. Woodmansee, A. Pfaltz, Chem. Commun. 2011, 47, 7912–7916. (d) C. Margarita, P. G. Andersson, J. Am. Chem. Soc. 2017, 139, 1346–1356. (e) A. Cabré, X. Verdaguer, A. Riera, Adv. Synth. Catal. 2019, 361, 4196–4200. [30] (a) M. Revés, C. Ferrer, T. León, S. Doran, P. Etayo, A. Vidal-Ferran, A. Riera, X. Verdaguer, Angew. Chem. Int. Ed. 2010, 49, 9452–9455. (b) T. León, A. Riera, X. Verdaguer, J. Am. Chem. Soc. 2011, 133, 5740– 5743. (c) E. Cristõbal-Lecina, P. Etayo, S. Doran, M. Revés, P. Martín- Gago, A. Grabulosa, A. R. Costantino, A. Vidal-Ferran, A. Riera, X. Verdaguer, Adv. Synth. Catal. 2014, 356, 795–804. (d) T. Leon, M. Parera, A. Roglans, A. Riera, X. Verdaguer, Angew. Chem. Int. Ed. 2012, 51, 6951–6955. (e) A. Prades, S. Núñez-Pertíñez, A. Riera, X. Verdaguer, Chem. Commun. 2017, 53, 4605–4608. [31] (a) E. Salomó, S. Orgué, A. Riera, X. Verdaguer, Angew. Chem. Int. Ed. 2016, 55, 7988–7992. (b) S. Orgué, A. Flores-Gaspar, M. Biosca, O. Pàmies, M. Diéguez, A. Riera, X. Verdaguer, Chem. Commun. 2015, 51, 17548–17551. [32] (a) E. Salomó, P. Rojo, P. Hernández-Lladó, A. Riera, X. Verdaguer, J. Org. Chem. 2018, 83, 4618–4627. (b) E. Salomó, A. Gallen, G. Sciortino, G. Ujaque, A. Grabulosa, A. Lledós, A. Riera, X. Verdaguer, J. Am. Chem. Soc. 2018, 140, 16967–16970. RESEARCH ARTICLE Entry for the Table of Contents (Please choose one layout) RESEARCH ARTICLE Albert Cabré,[a,b] Sergi Rafael,[b] Giuseppe Sciortino,[c,d] Gregori Ujaque,[c] Xavier Verdaguer[a,b] Agustí Lledós*[c] and Antoni Riera*[a,b] Page No. – Page No. Catalytic Regioselective Isomerization of 2,2-Disubstituted Oxetanes to Homoallylic Alcohols min. font size 6-7 pt.)) Text for Table of Contents, 450 characters with spaces.