This is the peer reviewed version of the following article: ChemSusChem 2015, 8, 1841 – 1844, which has been published in final form http://onlinelibrary.wiley.com/doi/10.1002/cssc.201403466/abstract. This article may be used for non-commercial purposes in accordance with http://olabout.wiley.com/WileyCDA/Section/id820227.html   Visible  Light-­‐Driven  Atom  Transfer  Radical  Addition  to  Olefins  using   Bi2O3  as  Photocatalyst   Paola Riente,[a] and Miquel A. Pericàs *[a,b] [a] Institute of Chemical Research of Catalonia (ICIQ). Avda. Països Catalans, 16. E-43007 Tarragona (Spain). E-mail: mapericas@iciq.es, priente@iciq.es [b] Department de Química Orgànica. Universitat de Barcelona. Dedicated to Prof. Antonio M. Echavarren on the occasion of his 60th birthday Abstract: Bismuth oxide, an inexpensive and non-toxic semiconductor, has been successfully used as a visible light photocatalyst for the atom transfer radical addition (ATRA) reaction of organobromides to diversely functionalized terminal olefins. The reaction takes place under very mild conditions, in the absence of any additive, and with low catalyst loading (1 mol%). The corresponding ATRA products are obtained with moderate to excellent yields (up to 95%). The development of chemical transformations promoted by costless energy, such as sunlight, is at the forefront of research in catalysis.[1] Within this area, the search for greener (non-toxic) and low cost catalysts, suitable for large scale production, has driven the entire field towards the intersection of organic chemistry, nanomaterials science and photochemistry.[2] In particular, much effort has been devoted to the discovery of visible light photocatalysts allowing the replacement of the yet widely used, but economically and toxicologically inconvenient, ruthenium and iridium complexes.[3] In this context, a variety of readily available semiconductors[4] and organic dyes[5] have been evaluated as photocatalysts. Recently, we have reported the photoinduced enantioselective a-alkylation of aldehydes with a-bromocarbonyl compounds by merging the commercially available, cheap semiconducting oxide Bi2O3 as a photocatalyst and the second generation MacMillan imidazolidinone as an organocatalyst.[6] From a mechanistic perspective, these results strongly suggest that the readily available photoexcited state of Bi2O3 (this semiconductor is characterized by a low band gap of ca. 1.3 eV) is appropriate to promote the cleavage of reactive C-Br bonds leading to alkyl radicals. Bearing this in mind, we envisaged that Bi2O3 might be a convenient visible light photocatalyst for other important organic reactions initiated by electron transfer to carbon-halogen bonds. Pioneered by Kharasch and coworkers,[7] and further studied by Curran[8] and by others,[9] the Atom Transfer Radical Addition (ATRA) reaction[10] consists in the addition of haloalkanes across C-C double (or triple) bonds promoting the simultaneous formation of a C-X (X = halogen) and a C-C bond. A closely related process, the Atom Transfer Radical Polymerization (ATRP) reaction,[11,12] independently introduced by Sawamoto[12a] and Matyjaszewski,[12b] has evolved into one of the most powerful methodologies to build up well-defined polymers and copolymers. Typically, the ATRA reaction is accomplished using transition metal complexes and reducing agents under harsh reaction conditions.[13] In this regard, some valuable approaches have been reported involving the use of metal-mediated visible light photoredox catalysis as an alternative for this useful synthetic transformation (Scheme 1).[14] Most often, the photocatalysts employed for the photoinduced ATRA have been expensive metals based on Ir- or Ru- complexes in the presence of a Lewis acid. As an alternative to these expensive metals, Reiser and coworkers demonstrated that a homemade copper complex was also an effective visible-light photocatalyst mediating the formation of ATRA products.[15] In a different approach, Melchiorre and coworkers have very recently reported a photochemical metalfree process to mediate the ATRA reaction.[16] This work: R! + 1 R-Br Br Bi2O3 (1 mol%) R R! Visible Light DMSO, rt 2 3 Other photocatalysts in previous works: [Ir{dF(CF3)ppy}2(dtbbpy)]PF6(14a) [Ru(bpy)3]Cl2(14b) [Cu(dap)2Cl](15) p-Methoxybenzaldehyde(16) Scheme 1. Catalysts for the visible light photoinduced atom transfer radical addition (ATRA) reaction. Where dF(CF3)ppy=2-(2,4-difluorophenyl)-5-trifluoromethylpyridine; dtbbpy=4,4’-ditert-butyl-2,2’-dipyridyl; bpy=2,2’-bipyridyl, and dap=2,9-bis(para-anisyl)-1,10-phenanthroline. Herein we report the successful application of the cheap, non-toxic and commercially available Bi2O3 powder as a light-driven photocatalyst to promote the ATRA reaction of a variety of organobromides into activated and non-activated alkenes under very mild reaction conditions involving either simulated or actual sunlight. Table 1. Bi2O3 photocatalyzed ATRA reaction under [a] different conditions. Br HO + 3 EtO 2C 1a Entry Bi 2O3 (1 mol%) CO2Et 2a Br HO Conditions 23 W 3 CO2Et CO2Et 3a 2a [mmol] Solvent Time [h] Yield [%] 2 DMF 24 38 2 2 DMF 24 81 3 1.2 DMF 48 63 4 2 DMSO 20 74 1.2 DMSO 20 69 1.2 DMSO 20 90 1.2 DMSO 72 - 1 5 [c] [d] 6 7 [e] [b] 8 [f] 1.2 DMSO 72 - 9 [g] 1.2 DMSO 24 - 1.2 DMSO 6 10 [h] [h] 85 [a] Conditions: Bi2O3 (5 mg, 0.01 mmol), 1a (1 mmol), 2a (1.2 or 2 mmol), degassed solvent (2 mL) at room temperature. [b] Isolated yield after column chromatography. [c] LiBr (2 mmol) was used as an additive. [d] LiBr (0.1 mmol) was used as an additive. [e] Reaction performed without Bi2O3. [f] Reaction carried out in the dark. [g] TiO2 (P25, Degussa) was used instead of Bi2O3. [h] Reaction promoted by daylight in Tarragona (41º07'00"N) in the first week of December; daylight hours computed. To start our investigation, the reaction between 5-hexen-1-ol (1a) and diethyl bromomalonate (2a) was chosen as a model to test the performance of Bi2O3 (1 mol%) in the ATRA reaction under different conditions and using as light source a 23 W household fluorescent bulb lamp (Table 1). We initially studied the use of DMF as solvent, either in the presence or in the absence of LiBr as a Lewis acidic additive (Table 1, entries 1-3). Product 3a could be isolated in good yield using a twofold amount of the ATRA donor in the absence of additive (Table 1, entry 2) but, in contrast with previous reports,[14] the presence of LiBr had a deletereous effect on the performance of the reaction causing a significant decrease in yield (Table 1, entry 1). In a similar manner, yield also decreased when the reaction was carried out with a lower excess of diethyl bromomalonate (Table 1, entry 3). We next examined the effect on the ATRA reaction of using a slightly more polar aprotic solvent, such as DMSO. Also in this case, poorer outcomes were recorded either using 2 equiv. of 2a (entry 4) or in the presence of LiBr (entry 5). Gratifyingly, when the reaction was performed in DMSO without any additive and with a 120% molar amount of ATRA donor, a substantial increase in yield was noted (entry 6). Control experiments clearly established that the presence of both Bi2O3 and light was essential for the reaction to proceed (entries 7 and 8). For comparison purposes, titanium dioxide was also tested as a photocatalyst under optimized conditions (entry 9), but no conversion was recorded. Finally, the reaction conditions employed in entry 6 were used for an additional experiment promoted by daylight (entry 10). In the first week of December 2014 (partly cloudy weather) in Tarragona, 6 hours of exposure to daylight irradiation (see Supporting Information) sufficed to induce complete conversion of 1a. Encouraged by the results achieved in the addition of 2a to 1a, we decided to examine the scope of the ATRA reaction promoted by light and catalyzed by Bi2O3. The results of this study have been summarized in table 2. Four different ATRA donors were tested (diethyl and dimethyl bromomalonate, ethyl bromodifluoroacetate, and carbon tetrabromide). Among them, diethyl and dimethyl bromomalonate displayed the highest reactivity in the ATRA reaction promoted by light and Bi2O3. For instance, the reaction of diethyl bromomalonate with a variety of functionalized terminal olefins led the expected adducts (3a-g) in good to excellent yields (up to 95%) in relative short reaction time (< 24 hours). Interestingly, the highest yields in the preparation of some of these products are recorded when the limiting reagent is the ATRA donor (olefin/ATRA donor = 1.1), a fact not observed in the preparation of 3a. Alkyne partners could also be used in combination with this ATRA donor leading the compound 3h as a ca. 60/40 mixture of Z/E isomers.[17] On the other hand, the use of dimethyl bromomalonate as ATRA donor appears to involve slightly longer reaction times and leads to somewhat lower yields (compare 3a and 3i). It is worth mentioning that the ATRA additions of 2a performed by this procedure can be readily scaled up. Thus, 3a could be prepared at 5 mmol scale (1.52 g, 90% isolated yield) with the same experimental setup and in the same reaction time by linearly scaling the amounts of reactants, solvent and catalyst. As a further example of the robustness of this methodology, the preparation of 3d was also performed with daylight promotion (71% yield). Remarkably enough, the reactions performed in this manner involved shorter reaction times (4 vs. 24 h) than with simulated sunlight (23W lamp). Thus, the use of Bi2O3 as a photocatalyst effectively allows performing ATRA reactions promoted by costless sunlight energy. Optimal conditions for the ATRA reaction of ethyl bromodifluoroacetate (2c) involved working with a slight excess of organobromide (2c/1 = 1.2), as established in the initial screening (Table 1). Thus, while the formation of 3j took place in 69% yield under these conditions, a significant yield decrease was noted (to 50%) when 2c and 1 were used in a 1:1.1 molar ratio. A parallel behavior was observed for the formation of 3k and 3l. Table 2. Scope of ATRA reaction using Bi2O3 as the photocatalyst.[a] 1 Bi2O3 (1 mol%) + R Br EtO2C FG R DMSO, rt, 23 W 2 ATRA donors: Br Br FG 3 Br CO2Et MeO2C CO2Me CF2BrCO2Et CBr4 ATRA acceptors: R = -(CH2)4OH, -(CH2)4OBn, -(CH2)4OTBS, -CH2NHBoc, -(CH2)3Br, -(CH2)9Br, -(CH2)2Ph, -CH2Ph, -CH2CH(CO2Et)2 Br RO2C Br CO2R Br BnO CO2Et 3a, 20 h, 90% (91%)[b] yield R = Et (2a), Me (2b) Br CO2Et CO2Et Br BocHN Br Br CO2Et 3g, 24 h, 46% (91%)[b] yield Br CO2Et 3h, 72 h, 48% (56%)[b] yield CO2Et Ph CO2Et 7 3f, 24 h, 75% (91%)[b] yield CO2Et CO2Et Br CO2Et Br Br CO2Et 3e, 24 h, 64% (95%)[b] yield 3d, 24 h, (73%)[b] yield 4 h, 71% yield[c] Br CO2Et 3c, 15 h, 60% yield CO2Et CO2Et CO2Et TBSO 3b, 24 h, 76% (84%)[b] yield HO CO2Et HO HO CO2Me CO2Me 3i, 24 h, 68% (80%)[b] yield F Br Br F F F 2c CO2Et HO Br F F CO2Et BocHN CO2Et 3k, 48 h, 45% yield 3j, 48 h, 69% yield Br F F Ph CO2Et 3l, 36 h, 60% yield CBr4 2d Br Br EtO2C CBr3 EtO2C 3m, 24 h, 76% yield CBr3 BnO 3n, 48 h, 52% yield Br 3o, 48 h, 56% yield Br Ph CBr3 Ph Br CBr3 HO 3p, 24 h, 68% yield Br CBr3 3q, 48 h, 54% yield Br 7 CBr3 3r, 30 h, 73% yield [a] Conditions: Bi2O3 (5 mg, 0.01 mmol), 1 (1.0 mmol), 2 (1.2 mmol), DMSO (2 mL) at room temperature. [b] Yields in parentheses are for reactions performed with 1 (1.1 mmol) and 2a or 2b (1.0 mmol). [c] Reaction promoted by daylight in Tarragona (41º07'00"N) in Dec. 10. Daylight hours computed. Carbon tetrabromide (2d) could also be used as an ATRA donor in front of a variety of terminal olefins (3m-3r). As with 2c, the reactions proceeded better when the ATRA donor was used in slight excess with respect to the olefin partner (2d/1 = 1.2). With the aim of verifying whether the reaction catalyzed by Bi2O3 takes place through radical intermediate, the known radical scavenger TEMPO (1.2 mmol per mmol 1a) was used as an additive in the preparation of 3a under otherwise optimized conditions. As anticipated, this provoked complete inhibition of the addition reaction (Scheme 2). Br HO 3 1a (1mmol) + EtO 2C CO2Et 2a (1.2 mmol) Bi 2O3 (1 mol%) X TEMPO (1 mmol) DMSO 23 W Br HO 3 CO2Et CO2Et 3a No conversion after 24h Scheme 2. Inhibition by TEMPO of the visible-light-induced atom transfer radical addition (ATRA) reaction. This fact, along with the semiconducting properties of the photocatalyst, enables us to propose the tentative mechanism shown in Figure 1.[8b,14,18] According to it; the incident photons promote the photoexcitation of electrons on the surface of the semiconductor from the valence to the conduction band, with generation of positive holes (h+). The photoexcited electrons induce reductive cleavage of the organobromide to generate the electrophilic radical R· (I). Then, this photogenerated radical undergoes addition to the partner olefin-giving raise to the radical intermediate II. From this point, two routes are possible. In route a (a radical-polar crossover), the radical intermediate II delivers an electron to the semiconductor to neutralize a positive hole and to provide a carbocation intermediate, which ultimately reacts with bromide leading to the ATRA product 3. In route b, a radical chain propagation pathway is proposed. Radical II subtracts a bromine atom from the starting material leading directly to 3 and regenerating radical I that continues the chain. Likewise, routes a and b operate in a concomitant manner. While route b is well established in the context of the ATRA reaction,[8b,14] the nature of Bi2O3 (solid semiconductor particles) could importantly favor route a.[19] R-Br 2 + e- e- CB GAP Br - h+ VB Bi 2O3 particle - eroute a R R1 R • R1 1 R1 II Br - I R-Br 2 route b h + = Holes CB = Conduction band VB = Valence band R• I Br R R• R1 3 Figure 1. Proposed mechanism for the Bi2O3 catalyzed, visible light photoinduced atom transfer radical addition (ATRA) reaction. In conclusion, a simple catalytic system composed of a non-toxic, commercially available Bi2O3 powder operating at low loading (1%) under visible light irradiation in DMSO displays excellent performance in the ATRA reaction between a variety of olefins and organobromides. The photocatalytic reaction works specially well for dialkyl bromomalonate ATRA donors, and allows using costless daylight to promote the process. Interestingly, the present methodology does not require the use of any additive for the reaction to proceed and offers advantages in cost and atom economy over previously reported methods involving the use of expensive metals or large amounts of organic materials. Experimental Section General procedure for ATRA reaction: To a sealed vial filled with argon, containing Bi2O3 powder (5 mg, 0.01 mmol), the corresponding organobromide (1.2 mmol, 1.2 equiv) and degassed dimethylsulfoxide (2 mL) was added through a septum. To this suspension, the alkene (1 mmol, 1.0 equiv) was added via syringe, and the mixture was degassed for 10 min by bubbling argon through the reaction medium. Inlet and outlet needles were removed; the vial was sealed (parafilm) and placed at a distance of 10 cm from a household bulb lamp (26 W). When the reaction was complete according to TLC or 1H NMR, the crude was poured into a funnel containing ethyl acetate (5 mL) and H2O (5 mL). The layers were separated; the organic phase was extracted with ethyl acetate (3 x 5 mL), washed with brine, dried over MgSO4 and concentrated. The resulting residue was purified by column chromatography on silicagel (cyclohexane/ethyl acetate) to afford the corresponding product. Acknowledgements Financial support from the Institute of Chemical Research of Catalonia (ICIQ) Foundation, MINECO (grant CTQ2012–38594-C02–01), EU-ITN network Mag(net)icFun (PITN-GA-2012-290248) and DEC Generalitat de Catalunya (Grant 2014SGR827) is gratefully acknowledged. We also thank MINECO for support through Severo Ochoa Excellence Accreditation 2014–2018 (SEV-2013-0319). Keywords: ATRA reactions • bismuth oxide • photocatalysis • radical reactions • visible light [1] a) A. Albini, M. Fagnoni, Green Chem. 2004, 6, 1-6; b) Y. Xi, H. Yi, A. Lei, Org. Biomol. Chem. 2013, 11, 2387–2403; c) D. M. Schultz, T. P. Yoon, Science 2014, 343, 1239176. [2] a) Y. Han, H. Huang, H. Zhang, Y. Liu, X. Han, R. Liu, H. Li, Z. Kang, ACS Catal. 2014, 4, 781-787; b) D. W. Manley, R. T. McBurney, P. Miller, J. C. Walton, A. Mills, C. O'Rourke, J. Org. Chem. 2014, 79, 1386–1398; For reviews, see: c) X. Lang, X. Chen, J. Zhao, Chem. Soc. Rev. 2014, 43, 473–486; d) Y. Wang, X. Wang, M. Antonietti, Angew. Chem. 2012, 124, 70–92; Angew. Chem. Int. Ed. 2012, 51, 68-89. [3] For reviews, see: a) T. P. Yoon, M. A. Ischay, J. Du, Nature Chem. 2010, 2, 527-532; b) C. K. Prier, D. A. Rankic, D. W. C. MacMillan, Chem. Rev. 2013, 113, 5322-5363; c) T. P. Yoon, ACS Catal. 2013, 3, 895-902; d) T. Koike, M. Akita, Inorg. Chem. Front. 2014, 1, 562-576. [4] For reviews, see: a) D. Ravelli, D. Dondi, M. Fagnoni, A. Albini, Chem. Soc. Rev. 2009, 38, 1999-2011; b) X. Lang, X. Chen, J. Zhao, Chem. Soc. Rev. 2014, 43, 473-486. [5] a) K. Fidaly, C. Ceballos, A. Falguières, M. S.-I. Veitia, A. Guy, C. Ferroud, Green Chem. 2012, 14, 1293-1297; For reviews, see: b) D. A. Nicewicz, T. M. Nguyen, ACS Catal. 2014, 4, 355-360; c) S. Fukuzumi, K.Ohkubo, Org. Biomol. Chem. 2014, 12, 6059-6071. [6] P. Riente, A. M. Adams, J. Albero, E. Palomares, M. A. Pericàs, Angew. Chem. 2014, 126, 9767– 9770; Angew. Chem. Int. Ed. 2014, 53, 9613–9616. [7] a) M. S. Kharasch, E. V. Jensen, W. H. Urry, Science 1945, 102, 128; b) M. S. Kharasch, P. S. Skell, P. Fisher, J. Am. Chem. Soc. 1948, 70, 1055-1059. [8] a) D. P. Curran, Synthesis 1988, 417-439; b) D. P. Curran, Synthesis 1988, 489-513; c) D. P. Curran, E. Bosch, J. Kaplan, M. Newcomb, J. Org. Chem. 1989, 54, 1826-1831; d) D. P. Curran, M.-H. Chen, E. Spletzer, C. M. Seong, C.-T. Chang, J. Am. Chem. Soc. 1989, 111, 8872-8878; e) D. P. Curran, C. M. Seong, J. Am. Chem. Soc. 1990, 112, 9401-9403; f) D. P. Curran, J. Tamine, J. Org. Chem. 1991, 56, 2746-2750; g) D. P. Curran, S.-B. Ko, Tetrahedron Lett. 1998, 39, 6629- 6632. [9] a) H. Yorimitsu, T. Nakamura, H. Shinokubo, K. Oshima, J. Org. Chem. 1998, 63, 8604-8605; b) H. Yorimitsu, T. Nakamura, H. Shinokubo, K. Oshima, K. Omoto, H. Fujimoto, J. Am. Chem. Soc. 2000, 122, 11041-11047; c) K. Weidner, A. Giroult, P. Panchaud, P. Renaud, J. Am. Chem. Soc. 2010, 132, 17511-17515; d) J. M. Muñoz-Molina, A. Caballero, M. M. Díaz-Requejo, S. Trofimenko, T. R. Belderraín, P. J. Pérez, Inorg. Chem. 2007, 46, 7725-7730; e) L. Cao, K. Weidner, P. Renaud, Adv. Synth. Catal. 2011, 353, 3467-3472. [10] a) T. Pintauer, K. Matyjaszewski, Encyclopedia of Radicals in Chemistry, Biology and Materials, Vol. 4, John Wiley & Sons, 2012, pp. 1851-1894; b) J. M. Muñoz-Molina, T. R. Belderrain, P. J. Pérez, Eur. J. Inorg. Chem. 2011, 3155-3164. [11] For review, see: K. Matyjaszewski, N. V. Tsarevsky, J. Am. Chem. Soc. 2014, 136, 6513-6533. [12] a) M. Kato, M. Kamigaito, M. Sawamoto, T. Higashimura, Macromolecules 1995, 28, 1721-1723; b) J.-S. Wang, K. Matyjazewski, J. Am. Chem. Soc. 1995, 117, 56145615; For selected examples, see: c) B. Wenn, M. Conradi, A. D. Carreiras, D. M. Haddleton, T. Junkers, Polym. Chem. 2014, 5, 3053-3060; d) V. A. Williams, T. G. Ribelli, P. Chmielarz, S. Park, K. Matyjazewski, J. Am. Chem. Soc. 2015, 135, 14281431; Metal-free ATRP: e) N. J. Treat, H. Sprafke, J. W. Kramer, P. G. Clark, B. E. Barton, J. Read de Alaniz, B. P. Fors, C. J. Hawker, J. Am. Chem. Soc. 2014, 136, 16096-16101. [13] a) R. A. Gossage, L. A. van de Kuil, G. van Koten, Acc. Chem. Res. 1998, 31, 423-431; b) C. L. Mero, N. A. Porter, J. Am. Chem. Soc. 1999, 121, 5155-5160; c) L. Quebatte, K. Thommes, K. Severin, J. Am. Chem. Soc. 2006, 128, 7440-7441; d) W. T. Eckenhoff, T. Pintauer, Inorg. Chem. 2007, 46, 5844-5846; e) T. Pintauer, W. T. Eckenhoff, C. Ricardo, M. N. C. Balili, A. B. Biernesser, S. J. Noonan, M. J. W. Taylor, Chem. Eur. J. 2008, 15, 38-41; f) W. T. Eckenhoff, T. Pintauer, Dalton Trans. 2011, 40, 4909-4917; g) K. Thommes, M. A. Fernández-Zúmel, C. Buron, A. Godinat, R. Scopelliti, K. Severin, Eur. J. Org. Chem. 2011, 249-255. [14] a) J. D. Nguyen, J. W. Tucker, M. D. Konieczynska, C. R. J. Stephenson, J. Am. Chem. Soc. 2011, 133, 4160-4163; b) C.-J. Wallentin, J. D. Nguyen, P. Finkbiener, C. R. J. Stephenson, J. Am. Chem. Soc. 2012, 134, 8875-8884. [15] a) M. Pirtsch, S. Paria, T. Matsuno, H. Isobe, O. Reiser, Chem. Eur. J. 2012, 18, 7336-7340; b) S. Paria, M. Pirtsch, V. Kais, O. Reiser, Synthesis 2013, 45, 26892698. [16] E. Arceo, E. Montroni, P. Melchiorre, Angew. Chem. 2014, 126, 12260– 12264; Angew. Chem. Int. Ed. 2014, 53, 12064-1268. [17] Diastereomeric ratios were determined by 1H NMR analysis. The E and Z configurations were unambiguously assigned by 2D 1H-1H NOESY experiments. [18] F. Minisci, Acc. Chem. Res. 1975, 8, 165–171. [19] The participation in the process of solid semiconductor particles could allow that all the events involved in the generation of radical intermediate II and its conversion into 3 through route a could take place in solvent cages containing 1 and 2 on the surface of Bi2O3 particles. As a consequence, route a would benefit from a very favorable entropy factor. Let the sunshine in: Bismuth oxide, a cheap and non-toxic semiconductor, efficiently converts sunlight into chemical energy. The ATRA addition of organobromides to terminal olefins is efficiently promoted by Bi2O3 (1 mol%) under the influence of actual or simulated sunlight. e- CB GAP h+ VB Bi 2O3 particle DMSO, rt R-Br + R' Br R 18 examples, up to 95% yield R'