Letter Cite This: Org. Lett. 2018, 20, 1228−1231
pubs.acs.org/OrgLett
Visible-Light-Enhanced Ring Opening of Cycloalkanols Enabled by Brønsted Base-Tethered Acyloxy Radical Induced Hydrogen Atom Transfer-Electron Transfer Rong Zhao,†,‡,∥ Yuan Yao,‡,∥ Dan Zhu,†,‡ Denghu Chang,†,‡ Yang Liu,‡ and Lei Shi*,†,‡,§ †
Shenzhen Graduate School, Harbin Institute of Technology, Shenzhen 518055, P. R. China MIIT Key Laboratory of Critical Materials Technology for New Energy Conversion and Storage, School of Chemistry and Chemical Engineering, Harbin Institute of Technology, Harbin 150001, P. R. China § State Key Laboratory of Elemento-Organic Chemistry, Nankai University, Tianjin 300071, P. R. China ‡
S Supporting Information *
ABSTRACT: A metal-free ring opening/halogenation of cycloalkanols, which combines both PPO/TBAX oxidant system and blue LEDs irradiation, is presented. This method produces diverse γ, δ, and even more remotely halogenated ketones in moderate to excellent yields under mild conditions. Interestingly, experimental and computational studies demonstrate the novel ring size-dependent concerted/stepwise (four-/five- to eight-membered rings) hydrogen atom transfer-electron transfer induced by Brønsted base-tethered acyloxy radical, which indicates distinct advantages brought by the cyclic structure of diacyl peroxides.
O
Scheme 1. Different Reaction Intermediates Generated from Cyclic Diacyl Peroxides
rganic peroxides are worthy of study owing to their widespread applications in industry and organic synthesis, bleaching, and disinfection and as polymerization promoters.1 Acyclic peroxides, such as tert-butyl hydroperoxide (TBHP) and 3-chloroperoxybenzoic acid (m-CPBA),2 have been extensively applied and studied; however, the research on cyclic peroxides remains largely unexplored in comparison with their similarly atom-connected counterparts.3 Despite that, the reported results about cyclic diacyl peroxides have preliminarily revealed its outstanding advantages, such as greater thermal stability,4 mild reaction conditions,5 and selectively generating oxidation products, such as phenols,6 which are more reactive than the starting commercial arenes. This convinced us there are more attractive properties about cyclic diacyl peroxides, beyond simple oxidation, which are worthy of exploration. Phthaloyl peroxides (PPO) and malonoyl peroxides (MPO) as two kinds of burgeoning cyclic diacyl peroxides exhibit the divine dichotomy mechanisms in the reported literatures. In 2013, Houk and Siegel6 described a reverse-rebound mechanism of oxidation of aromatic carbon−hydrogen bonds in the presence of PPO through diradical intermediate (Scheme 1a). In contrast, Tomkinson7 raised an zwitterion mechanism of the similar reaction oxidized by MPO supported by isotopic labeling in 2015 (Scheme 1b). Besides, PPO also achieves oxidation of sulfides/ thioketones by zwitterion mechanism in our previous work.5a Herein, we report the ring opening of the cyclic alcohols realized by unique radical/anion intermediate generated by the interaction of parent phthaloyl peroxides and halogen anions (Scheme 1c). The radical/anion intermediate effeciently enhances the ability to promote hydrogen atom transfer © 2018 American Chemical Society
(HAT)8a than a tranditional benzoyloxy radical8b (Scheme 1d), and accomplishes homolytic cleavage of pronouncedly stable oxygen−hydrogen bond (O−H BDFEs ≈ 105 kcal/mol)9 Received: January 16, 2018 Published: February 8, 2018 1228
DOI: 10.1021/acs.orglett.8b00161 Org. Lett. 2018, 20, 1228−1231
Letter
Organic Letters with the assistance of tethered hydrogen bonds (Scheme 1e), which is supported by quantum-mechanical calculations. Radical ring opening of strained cycloalkanols has been realized by different methods (Scheme S-1), and its corresponding products such as halogenated ketones (Figure S-1) are key intermediates of many bioactive molecules. In 2015, Zhu et al.10 reported transition-metal-catalyzed ring opening of cycloalkanols. In their mechanism of silver-catalyzed fluorination (Scheme 2a),10a the incorporation of cyclobutanol and a silver
Scheme 3. Scope of Radical Ring Opening of DifferentMembered Cyclic Alcohols
Scheme 2. Different Mechanisms of Radical Ring Openings of Strained Cycloalkanols
a 2 equiv of PPO-2 was used as an oxidant. bGram scale: 1a (1 g, 6.75 mmol).
bromination by basic byproduct carboxylate F, and generates the α,β-unsaturated ketone 5 (Scheme S-2). Subsequently, four- to eight-membered cyclic alcohols were used to test the influences of ring strain. Gratifyingly, all ring sizes evaluated in this study underwent β-scission readily. With the increase of the ring’s members, yields of halogenated linear ketones declined from excellent to low yields in general, and reaction times gradually lengthened to several hours in bromination and iodination reactions, which coincide with reduced reactivities of less strained cyclic alcohols. On account of the unsuccessful combination of PPO and TBACl, the Cl source was changed to alkali salt CsCl, and a low yield was obtained from the same blue LEDs irradiation conditions (Scheme S-2). Efforts to reduce these constraints are ongoing. Subsequently, we turned our attention to study the substitution pattern on cyclic alcohols (Scheme 4). Most surprising was all of the reactions achieved within 3 h under standard conditions, which are in general the shortest reaction times in reported literatures as far as we know. Different steric substituted γ-halogenated ketones were obtained with good-
salt exchanges F with SelectFluor by proton transfer (PT), and then alkoxy radical is generated by subsequent electron transfer (ET). Knowles et al.11 employed the intermolecular cooperation of a Brønsted base and an oxidant to generate alkoxy radical through proton-coupled electron transfer (PCET) in 2016 (Scheme 2b). In comparison, our multisite radical/anion intermediate in this work comprises an oxidizing acyloxy radical outfitted with a tethered carboxylate base (Scheme 2c). The oxidative carboxylic radical can, in conjuction with the Brønsted base, engender the HAT by simultaneous hydrogen-bond interaction rather than a PCET process (Table S-2). In all of the above-mentioned reactions, there are the generations of alkoxy radicals, and more stable C-centered radicals are subsequently obtained by ring-strain-promoted rearrangements. However, in this ring size-dependent concerted/stepwise mechanism, no alkoxy radical is found in the quantummechanical calculations of concerted hydrogen atom-electron transfer mechanism about the cyclobutanols, supported by intrinsic reaction coordinate (IRC) calculations (Figure S-4), while a stepwise mechanism is proposed for five- to eightmembered cycloalkanols. Currently, most ring openings of strained cycloalkanols rely on metal catalysts (Scheme S-1), and just a few cases can produce ring openings by oxidants,12a,b photocatalysts,12c or Brønsted bases.12d Intrigued by the electrophilic halogenation in nature,13 we combined various peroxides and bromide sources and chose cyclobutanol 1a as the model substrate to optimize the ring opening/bromination reaction (Table S-1). Despite being intuitively more favorable on entropic grounds than cyclic peroxides, acyclic benzoyl peroxide (BPO) led to much lower yield. After condition screening, PPO was mainly used with tetrabutylammonium bromide (TBAB) in the ring opening of cyclic alcohols, and PPO-2 served as an oxidant replacement. Having established the optimized reaction conditions, we first attempted to explore the influence of different ring sizes (Scheme 3). Unfortunately, cyclopropanol 4 proceeds elimination after
Scheme 4. Substrate Scope of Radical Ring Opening of Different Substituted Cyclic Alcohols
a
1229
2 equiv of PPO-2 was used as an oxidant. DOI: 10.1021/acs.orglett.8b00161 Org. Lett. 2018, 20, 1228−1231
Letter
Organic Letters yields (2f−h). Compared with the moderate yields of cyclobutanols with electron-donating groups on phenyl (1n, 1p), substrates with electron-withdrawing groups usually gave the corresponding γ-brominated ketones in better yields (2i−m). Additionally, low yields of aromatic heterocyclic substituted substrates (1t, 1u) were likely due to their instability to be easily oxidized to other byproducts. Importantly, we observed that different substituents on six-membered ring (1x−z) could be accommodated and furnish excellent yields (77−93%), suggesting that steric influence on the ring is not a prerequisite for efficient bond scission. Besides aromatic groups, cyclic and chain aliphatic substituted cycloalkanols also proceed smoothly with good yields (2v, 2w). Furthermore, the ring-opening/bromination reaction of cycloalkanols on different sites of the adamantane skeleton (2aa, 2ab) and late-stage functionalization of the derivative of the natural product were performed successfully as well (2ac). Notably, PPO-2 can be an alternative oxidant to significantly increase yields of most inactive substrates (2m, 2q, 2r, and 2s). To gain insights into the possible mechanistic pathways, several control experiments (Scheme 5) and UV/vis detection
reaction (Scheme 5b), which suggested that a radical-mediated mechanism might be involved. To further evaluate the cleavage of the O−H bond, the kinetic isotope effect was roughly estimated as 1.4 from parallel experiments of protiated (1a) or deuterated substrate (1a′) with 1n as a reference substance (Scheme 5c), due to the indistinguishability of γ-brominated ketones from direct KIE experiments. The influence of hydrogen−deuterium exchange between substrates is not excluded, but may be ignored due to short reaction time. This approximate result of 1.4 matches previously reported KIE values (1.41,14a 1.4,14b and 1.4 ± 0.114c) for HAT process of active C−H bonds realized by oxygen radical, lending support to the proposed HAT process. Similarly, secondary kinetic isotope effects have been measured under standard conditions using protiated and deuterated substrates in the same reaction flask. The apparent distinctions between cyclobutanol (KIE ≈ 1.31) and five- and six-membered cycloalkanols (KIE ≈ 1.15, 1.12) possibly indicated different mechanisms on turnover-limiting steps between four- and other-membered rings (Scheme 5c). To validate our hypothesis, we then performed DFT calculations to uncover the fundamental mechanism. PPO directly yields diradical molecule (INT1A) through O−O bond homolysis, followed by single-electron transfer (SET) with bromide anion to give radical/anion intermediate INT4 and bromide radical (Figure S-3). Compound 1a undergoes a concerted HAT and C−C bond cleavage of the ring to directly provide carbon radical INT6 (Figure 1). The spin density
Scheme 5. Control and Competitive Experiments and Mechanistic Studies
Figure 1. Potential energy profiles for the radical ring opening of cyclic alcohols.
analysis confirmed that HAT process, not PCET, is involved in the reaction, in agreement with KIE results mentioned above. In sharp contrast, 1b−e undergo a HAT to generate stable alkoxy radicals INT2D, followed by the subsequential C−C bond cleavage to give carbon radicals INT6. The activation free energies for this stage gradually increased from 1a to 1e, consistent with reaction times shown in Scheme 3. Based on mechanistic observations and DFT calculations, a proposed mechanism was outlined in Scheme 6. Intermediate B, in situ produced by PPO and TBAB, led to the homolytic cleavage of O−Br bond under illumination condition to generate radical/anion intermediate C and bromide radical. When cyclobutanol served as a substrate, a kinetic coupling of hydrogen atom abstraction and electron transfer enabled the homolyzing of strong O−H bond and C−C bond of the strained ring in concert to provide carbon radical G and byproduct F directly. In contrast, five- to eight-membered cyclic alcohols first undergo HAT driven by active radical/anion intermediate C. The generated alkoxy
(Figure S-2) were performed. First, to investigate substituent effects on phenyl, the product ratio obtained from the pairwise comparison experiments among 1a, 1n, and 1l demonstrated that the relative reactivities increase obviously with the increasing electron-withdrawing abilities of the para substituents of the aromatic ring of 1-arylcyclobutan-1-ol (Scheme 5a), which is consistent with yields shown in Scheme 4. Subsequently, the additions of known effective radical inhibitors as TEMPO and BHT under identical conditions significantly suppressed the 1230
DOI: 10.1021/acs.orglett.8b00161 Org. Lett. 2018, 20, 1228−1231
Letter
Organic Letters Author Contributions
Scheme 6. Proposed Mechanism of the Radical Ring-Opening of Cyclic Alcohols
∥
R.Z. and Y.Y. contributed equally.
Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS The work was financially supported by the “Fundamental Research Funds for the Central University” (HIT.BRETIV.201502). The authors thank Dr. Xiao Zhang in the School of Chemistry and Chemical Engineering, Harbin Institute of Technology, for assistance with the single-crystal X-ray analysis of byproduct F.
■
radicals E were tautomerized to similar carbon radical G subsequently. At last, carbon radical G reacts with bromide radical forming the brominated ketones 2. This mechanism is well supported by our experimental and computational results. In conclusion, a visible-light-enhanced method for the preparation of diverse halogenated ketones has been developed by the ring opening/halogenation reaction of cyclic alcohols using a combination of phthaloyl peroxide and tetrabutylammonium halide. The reaction was conducted under mild conditions with blue LED irradiation, which is flexible enough to extend to different substrates, even the late-stage functionalization of the derivative of natural products in moderate to excellent yields. The PPO/TBAX oxidant system is one of the rare examples of metal-free ring opening of cyclic alcohols with the shortest reaction times. After mechanistic experiments and quantummechanical calculations, a ring-size-dependent concerted/ stepwise (four-/five- to eight-membered rings) hydrogen atom transfer−electron transfer mechanism induced by Brønsted base tethered acyloxy radical was proposed. Further investigations on the protocol for Brønsted base assisted interactions are currently ongoing in our laboratories.
■
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.orglett.8b00161. Experimental procedures, compound characterization data, mechanistic studies, crystallographic data, and calculation details for products (PDF) Accession Codes
CCDC 1580138 contains the supplementary crystallographic data for this paper. These data can be obtained free of charge via www.ccdc.cam.ac.uk/data_request/cif, or by emailing data_
[email protected], or by contacting The Cambridge Crystallographic Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK; fax: +44 1223 336033.
■
REFERENCES
(1) (a) Campos-Martin, J. M.; Blanco-Brieva, G.; Fierro, J. L. G. Angew. Chem., Int. Ed. 2006, 45, 6962−6984. (b) Shanley, E. S. J. Chem. Educ. 1951, 28, 260−266. (c) Her, B.; Jones, A.; Wollack, J. W. J. Chem. Educ. 2014, 91, 1491−1494. (d) Klement, I.; Knochel, P. Synlett 1995, 1995, 1113−1114. (e) Held, P. A.; Gao, H.-Y.; Liu, L.; Mück-Lichtenfeld, C.; Timmer, A.; Mönig, H.; Barton, D.; Neugebauer, J.; Fuchs, H.; Studer, A. Angew. Chem., Int. Ed. 2016, 55, 9777−9782. (2) Mayer, R. J.; Tokuyasu, T.; Mayer, P.; Gomar, J.; Sabelle, S.; Mennucci, B.; Mayr, H.; Ofial, A. R. Angew. Chem., Int. Ed. 2017, 56, 13279−13282. (3) Zhao, R.; Chang, D.; Shi, L. Synthesis 2017, 49, 3357−3365. (4) Fujimori, K.; Oshibe, Y.; Hirose, Y.; Oae, S. J. Chem. Soc., Perkin Trans. 2 1996, 0, 413−417. (5) (a) Gan, S.; Yin, J.; Yao, Y.; Liu, Y.; Chang, D.; Zhu, D.; Shi, L. Org. Biomol. Chem. 2017, 15, 2647−2654. (b) Chang, D.; Zhu, D.; Shi, L. J. Org. Chem. 2015, 80, 5928−5933. (6) Yuan, C.; Liang, Y.; Hernandez, T.; Berriochoa, A.; Houk, K. N.; Siegel, D. Nature 2013, 499, 192−196. (7) Dragan, A.; Kubczyk, T. M.; Rowley, J. H.; Sproules, S.; Tomkinson, N. C. O. Org. Lett. 2015, 17, 2618−2621. (8) (a) Pike, S. J.; Hutchinson, J. J.; Hunter, C. A. J. Am. Chem. Soc. 2017, 139, 6700−6706. (b) Mukherjee, S.; Maji, B.; Tlahuext-Aca, A.; Glorius, F. J. Am. Chem. Soc. 2016, 138, 16200−16203. (9) Dean, J. A.; Lange, N. A. Lange’s Handbook of Chemistry, 15th ed.; McGraw Hill, Inc.: New York, 1999. (10) (a) Zhao, H.; Fan, X.; Yu, J.; Zhu, C. J. Am. Chem. Soc. 2015, 137, 3490−3493. (b) Ren, R.; Zhao, H.; Huan, L.; Zhu, C. Angew. Chem., Int. Ed. 2015, 54, 12692−12696. (c) Wang, D.; Ren, R.; Zhu, C. J. Org. Chem. 2016, 81, 8043−8049. (d) Ren, R.; Wu, Z.; Zhu, C. Chem. Commun. 2016, 52, 8160−8163. (e) Fan, X.; Zhao, H.; Yu, J.; Bao, X.; Zhu, C. Org. Chem. Front. 2016, 3, 227−232. (f) Huan, L.; Zhu, C. Org. Chem. Front. 2016, 3, 1467−1471. (g) Ren, R.; Wu, Z.; Xu, Y.; Zhu, C. Angew. Chem., Int. Ed. 2016, 55, 2866−2869. (h) Wang, M.; Wu, Z.; Zhu, C. Org. Chem. Front. 2017, 4, 427−430. (i) Ren, R.; Zhu, C. Synlett 2016, 27, 1139−1144. (11) Yayla, H. G.; Wang, H.; Tarantino, K. T.; Orbe, H. S.; Knowles, R. R. J. Am. Chem. Soc. 2016, 138, 10794−10797. (12) (a) Wang, S.; Guo, L.-N.; Wang, H.; Duan, X.-H. Org. Lett. 2015, 17, 4798−4801. (b) Aureliano Antunes, C. S. A.; Bietti, M.; Lanzalunga, O.; Salamone, M. J. Org. Chem. 2004, 69, 5281−5289. (c) Bloom, S.; Bume, D. D.; Pitts, C. R.; Lectka, T. Chem. - Eur. J. 2015, 21, 8060−8063. (d) Zhang, W.-C.; Li, C.-J. J. Org. Chem. 2000, 65, 5831−5833. (13) Podgoršek, A.; Zupan, M.; Iskra, J. Angew. Chem., Int. Ed. 2009, 48, 8424−8450. (14) (a) Tanko, J. M.; Friedline, R.; Suleman, N. K.; Castagnoli, N., Jr. J. Am. Chem. Soc. 2001, 123, 5808−5809. (b) Griller, D.; Howard, J. A.; Marriott, P. R.; Scaiano, J. C. J. Am. Chem. Soc. 1981, 103, 619−623. (c) Yayla, H. G.; Peng, F.; Mangion, I. K.; McLaughlin, M.; Campeau, L.C.; Davies, I. W.; DiRocco, D. A.; Knowles, R. R. Chem. Sci. 2016, 7, 2066−2073.
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. ORCID
Yuan Yao: 0000-0003-0033-971X Lei Shi: 0000-0002-6865-1104 1231
DOI: 10.1021/acs.orglett.8b00161 Org. Lett. 2018, 20, 1228−1231