This  document  is  the  Accepted  Manuscript  version  of  a  Published  Work  that  appeared  in  final  form  in   Journal  of  the  American  Chemical  Society,  copyright  ©  American  Chemical  Society  after  peer  review  and  technical  edit-­‐ ing  by  the  publisher.  To  access  the  final  edited  and  published  work  see  http://pubs.acs.org/doi/abs/10.1021/ja5064586     Ni-catalyzed Carboxylation of Unactivated Primary Alkyl Bromides and Sulfonates with CO2 Yu Liu†, Josep Cornella† and Ruben Martin*†§ † § Institute of Chemical Research of Catalonia (ICIQ), Av. Països Catalans 16, 43007, Tarragona, Spain Catalan Institution for Research and Advanced Studies (ICREA), Passeig Lluïs Companys, 23, 08010, Barcelona, Spain Supporting Information Placeholder ABSTRACT: A Ni-catalyzed carboxylation of unactivat- ed primary alkyl bromides and sulfonates with CO2 at atmospheric pressure is described. The method is characterized by its mild conditions and a remarkable wide scope without the need for air- or moisture-sensitive reagents, thus becoming a user-friendly and operationally simple protocol en route to carboxylic acids. The utilization of CO2 as an alternative renewable feedstock has recently received a significant attention in the scientific community.1 Such interest is primarily associated to the fact that CO2 is nontoxic, abundant and nonflammable, hence constituting an opportunity for carbon sequestration and allowing the implementation of innovative, yet practical, methodologies counterintuitive at first sight.1 Beyond any doubt, the synthesis of carboxylic acids represents an ideal target in CO2 fixation since a myriad of molecules such as Atorvastatin, Beraprost, Artesunate, Pemetrezed or Pregabalin, among others, display a significant biological activity (Scheme 1).2,3 Scheme 1. Biological Significance of Carboxylic Acids. Me PhHN Me CO2H OH OH Me CO2H N O O F Atorvastatin Me H H H2N Me H CO2H Pregabalin Me OH OH Beraprost Encouraged by a seminal work of Osakada and Yamamoto,4 we5 and others6 launched a program aimed at unlocking the potential of CO2 in reductive catalytic reactions (Scheme 2, path a).7 Unlike carboxylation events based on stoichiometric, well-defined and, in some cases, air-sensitive organometallic species (path b),8,9 such reductive events offer higher flexibility and ease of execution by using simpler building blocks, thus representing an added value from a simplicity, reliability and step-economical standpoint. Unfortunately, reductive carboxylation protocols are inherently restricted to substrates that rapidly undergo oxidative addition such as aryl5,6 or benzyl halides (path a).5a,5b Ideally, this field should include the use of unactivated alkyl electrophiles possessing β-hydrogens. Indeed, these substrates are the most challenging in the cross-coupling arena due to their reluctance towards oxidative addition and the proclivity of in situ generated alkyl metal species for β-hydride elimination, homodimerization or hydrogen abstraction pathways, among others. 10 Therefore, at the outset of our investigations it was unclear whether a metalcatalyzed carboxylation event could ever be conducted with unactivated alkyl electrophiles.11 If successful, such a process would offer an unrecognized opportunity in CO2 fixation while opening up new possibilities via unconventional bond disconnections. Herein we report a mild Ni-catalyzed carboxylation of unactivated primary alkyl bromides and sulfonates possessing β-hydrogens with CO2 (path c). The protocol represents a convenient method to rapidly access carboxylic acids from simple precursors without handling air-, moisture-sensitive reagents or cyanide sources and it is characterized by a wide scope and an excellent chemoselectivity profile. Scheme 2. Reductive Carboxylation Reactions with CO2. R1 catalyst / CO2 or R1 X path a X = I,Br,Cl,OR Ar or I R1 CO2H Ar X catalyst / CO2 R1 Restricted to aryl & activated benzyl motifs (fast oxidative addition) 1a Br CO2H Mn, CO2 (1 atm) DMA, rt 2a path c H H this work X = Br, OTs User-friendly & mild conditions No sensitive metal species Exquisite chemoselectivity Me Well-defined & sensitive metal species catalyst path b M = Li, MgX ZnX, B(OR)2 X = I, Br We initiated our investigations with 1a as the model substrate with CO2 (1 atm) at room temperature (Table 1). As expected, the conditions previously employed for the carboxylation of aryl halides5c,6 or primary benzylic halides5a,5b failed to convert 1a into 2a. Initial screening of metal complexes identified NiCl2·glyme as a competent catalyst with cheap Mn as reducing agent.12 While nitrogen donors have successfully been employed as ligands in cross-coupling reactions of unactivated alkyl halides,13 no conversion to 2a was observed with commonly employed bipyridines, tert-pyridines or oxazolines (L1-L9).12 In these cases, dimerization, βhydride elimination and recovered starting material was observed in the crude reaction mixtures. A similar reactivity pattern was found when employing simple phenanthroline-type ligands (L10-L13). We speculated that an increase in the steric bulk around the nitrogen-donor ligand could lead to more robust Ni complexes with enhanced stability and greater activity. In line with this notion, we found that L14 delivered 2a in 66% yield. Analogously, L16, a bench-stable ligand readily obtained in one-step and in bulk quantities,14 allowed for obtaining 2a in a 76% isolated yield. Dimerization and traces of β-hydride elimination byproducts account for the observed mass balance.15 Importantly, the reaction could be scaled up without any erosion in yield. Intriguingly, subtle changes in the electronic or steric environment of the 1,10-phenanthroline backbones had a deleterious effect (L15 and L17).16 Control experiments unambiguously revealed that all reaction components were necessary to promote the carboxylation of 1a.12,17 Table 1. Ligand influence on the reaction outcome.a,b R Me CO2H Me Me Me N N R R R = H (L10), 0% R = Ph (L11), 0% R M + CO2 OMe NiCl2·glyme (10 mol%) L (22 mol%) II Catalytic reductive carboxylation of unactivated alkyl bromides & tosylates R1 OMe CO2H X R1 Me N N L13, 0% N N L12, 0% R N N Me Me R = H (L14), 66% R = OMe (L15), 0% N Et N N Et L16, 76%,c 78%,c,d 35%e N nPr nPr L17, 66% a Reaction conditions: 1a (0.15 mmol), NiCl2·glyme (10 mol%), L (22 mol%), Mn (0.33 mmol), DMA (0.15 M), at rt under CO2 (1 atm) for 12 h. b Yields were determined by HPLC analysis using naphthalene as internal standard. c Isolated yield. d 1a (1.0 mmol). e NiCl2·glyme (5 mol%). Table 2. Ni-catalyzed Carboxylation of Alkyl Bromides.a,b R1 NiCl2·glyme (10 mol%) R1 L16 (22 mol%) CO2H Br Mn, CO2 (1 atm) H 2a-y H 1a-y DMA, rt Et OMe N N L16 Et Me Me CO2H 2a, 74% Me O 2j, 56% O CO2H O N Me O 2k, 66% Me 2d, 70% X CO2H 2h, X=OH; 62% 2i, X=CO2Et; 62% 2g, 59% 2f, 70% O O CO2H CO2H O CO2H 2e, 55% CO2H CN OAc O O CO2H 2c, 68% CO2H 2b, 65% S O CO2H OBn CO2H 2l, 64% R CO2H 2m, 68% CHO yltin into the C(sp3)–Br bond or a carboxylation event on the C–Sn bond were detected in the crude material.19 While conformational restrictions might account for the former, the latter is particularly interesting since organotin reagents have been reported to efficiently undergo carboxylation events.20 Similarly, L14 provided better results for 2s. Site-selectivity could be accomplished in the presence of electrophilic sites amenable for Nicatalyzed cross-coupling reactions such as aryl pivalates (2t),21 acetates (2y),21 carbamates (2v)21 or aryl fluorides (2r).22 While aryl chlorides,6a tosylates6a or pivalates5a have been used in reductive carboxylation reactions, we found exclusive CO2 insertion into the C(sp3)–Br bond (2t, 2u-2x). The synthetic value of this transformation is illustrated by a concise synthesis of compounds that exhibit potent biological activities such as MCPB (2x) and 12 α-CEHC (2y) from available precursors. Table 3. Ni-catalyzed Carboxylation of Alkyl Sulfonates.a,b O O O CO2H 2n, R=n-hexyl; 73%c 2o, R=NEt2; 73% X Me F O SnBu3 CO2H CO2H 2p, 14% (56%c ) 2q, 45% 2 CO2H R1 OR H 3a-f 2r, 70% Cl N N L14 Me OMe O 2s, X=OH; 53% (70%c ) 2t, X=OPiv; 62% 2u, X=OTs; 58% 2v, X=OCONMe2; 59% CO2H Cl NiBr2·glyme (10 mol%) R1 L14 (26 mol%) CO2H Mn, CO2 (1 atm) H 4 DMF, 50 ºC Me O Me OPiv 2w, 68% Me OTs 3a, 76% CO2H AcO OTs OR 3e, 69%c, 5%e 3b (R=Ts), 74% 3c (R=Ms), 61%c 3d (R=COCF3), 31%d OTs 3f, 62%f O Me CO2H 2x (MCPB), 73% (Phenoxybutyric herbicide) CO2H O Me Me 2y (!-CEHC), 85% (Antioxidant & antitumoral) a Reaction conditions: 1a-u (0.30 mmol), NiCl2·glyme (10 mol%), L16 (22 mol%), Mn (0.66 mmol), DMA (0.15 M), at rt under CO2 (1 atm) for 12 h. b Isolated yields, average of at least two independent runs. c Using L14 (22 mol%). Encouraged by these findings, we set out to explore the preparative scope of our reaction. As shown in Table 2, a host of unactivated primary alkyl bromides possessing β-hydrogens could be equally accommodated in good yields.18,19 Particularly illustrative is the chemoselectivity profile of our protocol as esters (2f, 2i-l, 2r, 2t, 2w and 2y), nitriles (2g), heterocycles (2k and 2l), acetals (2e), amides (2o), ketones (2j and 2n) and even aldehydes (2q) were tolerated. Notably, unprotected aliphatic alcohols (2h), phenols (2s) or carbonyl compounds containing relatively acidic α-protons (2g, 2i, 2j, 2n, 2o and 2w) did not compete with the efficacy of the carboxylation event. At the current level of development, unactivated secondary alkyl bromides cannot be employed as coupling partners.19 Surprisingly, the reaction could also be conducted in the presence of aryltin reagents (2p) with L14, thus providing ample opportunities for subsequent manipulation. Interestingly, no macrocycle resulting from an intramolecular addition of the ar- a 3a-f (0.30 mmol), NiBr2·glyme (10 mol%), L14 (26 mol%), Mn (2.4 equiv), DMF (0.15M) at 50 ºC for 12 h. b Isolated yields, average of at least two independent runs. c 60 ºC. d 100 ºC. e 7.5 mol% NiBr2·glyme. f 70 ºC. In light of these results we wondered whether we could extend our Ni-catalyzed reductive carboxylation event to unactivated alkyl sulfonates. While the reaction of 3b under the optimized conditions for alkyl bromides (Table 2) resulted in lower conversions to products, the combination of NiBr2·glyme, L14 and DMF as the solvent at 50 ºC under 1 atm CO2 was optimal, furnishing the corresponding carboxylic acid in 76% yield.12,14 Interestingly, alkyl mesylates (3c) or trifluoroacetates (3d) could also be employed, albeit in lower yields. Notably, the presence of other C-O electrophiles such as alkyl pivalates did not interfere, resulting in the selective carboxylation of the alkyl sulfonate backbone (3f). Overall, we believe the results in Tables 2 and 3 shows the robustness and the prospective impact of our Ni-catalyzed carboxylative protocol when employing unactivated alkyl bromides or alkyl sulfonates.23 Although an in depth mechanistic study should await further investigations, we wondered whether the reaction was initiated by β-hydride elimination followed by a hydrocarboxylation event.24 To such end, we subjected 5-phenyl pentene (5) under our optimized conditions. Under the limits of detection, we did not detect any carboxylation reaction.12 A similar result was obtained when exposing n-butylMnBr (6) to our Ni/L16 system in the presence or absence of Mn, thus leaving some doubt about the intermediacy of organomanganese species.12 In order to shed light into the mechanism, we decided to study the carboxylation reaction of 7a and 7b (Scheme 3).12 A diastereomerically pure 8 was anticipated for a mechanism consisting of a “classical” oxidative addition;25 on the contrary, a statistical mixture of diastereoisomers in 8 would account for a free-radical mechanism via single electron transfer (SET). As shown in Scheme 3, 1H-NMR spectroscopical analysis of the crude mixture revealed the loss of stereochemical integrity at C1.26 A similar behavior was found for 7c and 7d, an observation that might indicate a scenario consisting of SET processes via Ni(I) species. 27-31 In line with this notion, we observed that radical clocks such as (bromomethyl)cyclopropane and 1-bromo-5-hexene resulted in ring-opened dimerization products. Scheme 3. Mechanistic experiments. As Tables 2 and 3 Ph D 45-65% Ph D 8 (1:1) CO2 DH 2 1 X I D + L2Ni(I)X L2Ni(0) (1) (2) (3) (4) (5) (6) Ph DH 1 H REFERENCES 65-69% 2 D D H L2Ni(0) 7a (X=Br; J1,2 = 5.1 Hz) 7b (X=OTs; J1,2 = 5.4 Hz) As Tables 2 and 3 CO2H Johnson Matthey, Umicore and Nippon Chemical Industrial are acknowledged for a gift of metal & ligand sources. Y.L. and J.C thank COFUND and European Union (FP7PEOPLE-2012-IEF-328381) for a fellowship. This paper is dedicated to the memory of Prof. Gregory L. Hillhouse. X (7) HD 7c (X=Br; J1,2 = 12.4 Hz) 7d (X=OTs (J1,2 = 9.4 Hz) In summary, we have reported a new catalytic carboxylation of unactivated primary alkyl bromides and sulfonates possessing β−hydrogens with CO2 that gives access to valuable carboxylic acids. This method is characterized by its exquisite functional group compatibility, mild conditions, readily availability of the starting materials and ease of execution without the need for airor moisture-sensitive materials. Further investigations into the mechanism and the extension to more challenging substrate combinations are currently underway. (8) (9) (10) (11) ASSOCIATED CONTENT Supporting Information. Experimental procedures and spectral data. This material is available free of charge via the Internet at http://pubs.acs.org. (12) (13) AUTHOR INFORMATION Corresponding Author * rmartinromo@iciq.es Funding Sources No competing financial interests have been declared. ACKNOWLEDGMENT We thank ICIQ, the European Research Council (ERC277883) and MINECO (CTQ2012-34054) for support. (14) (15) For selected reviews, see: (a) Zhang, L.; Hou, Z. Chem. Sci. 2013, 4, 3395. (b) Tsuji, Y.; Fujihara, T. Chem. Commun. 2012, 48, 9956. (c) Cokoja, M.; Bruckmeier, C.; Rieger, B.; Herrmann, W.A.; Kühn, F. E. Angew. Chem., Int. Ed. 2011, 50, 8510. (d) Martin, R.; Kleij, A. W. ChemSusChem 2011, 4, 1259. (e) Huang, K.; Sun, C.–L.; Shi, Z.–J. Chem. Soc. Rev. 2011, 40, 2435. (g) Sakakura, T.; Choi, J.-C.; Yasuda, H. Chem. Rev. 2007, 107, 2365 (a) Patai, S. The chemistry of acid derivatives. Germany: Wiley, 1992. (b) Goossen, L. J.; Rodríguez, N.; Goossen, K. Angew. Chem., Int. Ed. 2008, 47, 3100. Maag, H. Prodrugs of carboxylic acids. New York: Springer, 2007. Osakada, K.; Sato, R.; Yamamoto, T. Organometallics 1994, 13, 4645 (a) Correa, A.; León, T.; Martin, R. J. Am. Chem. Soc. 2014, 136, 1062. (b) León, T.; Correa, A.; Martin, R. J. Am. Chem. Soc. 2013, 135, 1221. (c) Correa, A.; Martin, R. J. Am. Chem. Soc. 2009, 131, 15974 (a) Fujihara, T.; Nogi, K.; Xu, T.; Terao, J.; Tsuji, Y. J. Am. Chem. Soc. 2012, 134, 9106. (b) Tran-Vu, H.; Daugulis, O. ACS Catal. 2013, 3, 2417. For electrochemical carboxylative processes of organic halides, see for example: (a) Ohkoshi, M.; Michinishi, J.– Y.; Hara, S.; Senboku, H. Tetrahedron 2010, 66, 7732. (b) Amatore, C.; Jutand, A.; Khalil, F.; Nielsen, M. F. J. Am. Chem. Soc. 1992, 114, 7076 For a review, see: Correa, A.; Martin, R. Angew. Chem., Int. Ed. 2009, 48, 6201 For selected carboxylation of alkyl organoboranes: (a) Ohishi, T.; Zhang, L.; Nishiura, M.; Hou, Z. Angew. Chem., Int. Ed. 2011, 50, 8114. (b) Ohmiya, H.; Tanabe, M.; Sawamura, M. Org. Lett. 2011, 13, 1086. For the carboxylation of alkyl organozincs: (c) Ochiai, H.; Jang, M.; Hirano, K.; Yorimitsu, H.; Oshima, K. Org. Lett. 2008, 10, 2681. For selected reviews using unactivated alkyl halides: (a) Kambe, N.; Iwasaki, T.; Terao, J. Chem. Soc. Rev. 2011, 40, 4937. (b) Hu, X. Chem. Sci. 2011, 2, 1867. (c) Jana, R.; Pathak, T. P.; Sigman, M. S. Chem. Rev. 2011, 111, 1417. For photoinduced events with excess of Sm/SmI2 using tungsten lamps under hv, see: Nomoto, A.; Kojo, Y.; Shiino, G.; Tomisaka, Y.; Mitani, I.; Tatsumi, M.; Ogawa, A. Tetrahedron Lett. 2010, 51, 6580. See Supporting Information for details Selected references: (a) Everson, D. A.; Jones, B. A.; Weix, D. J. J. Am. Chem. Soc. 2012, 134, 6146. (b) Choi, J.; Fu, G. C. J. Am. Chem. Soc. 2012, 134, 9102. (c) Binder, J. T.; Cordier, C. J.; Fu, G. C. J. Am. Chem. Soc. 2012, 134, 17003. (d) Phapale, V. B.; Guisán-Ceinos, M.; Buñuel, E.; Cárdenas, D. J. Chem. Eur. J. 2009, 15, 12681. (e) Phapale, V. B.; Buñuel, E.; García-Iglesias, M.; Cárdenas, D. J. Angew. Chem. Int. Ed. 2007, 46, 8790. (f) Powell, D. A.; Maki, T.; Fu, G. C. J. Am. Chem. Soc. 2005, 127, 510. (g) Powell, D. A.; Fu, G. C. J. Am. Chem. Soc. 2004, 126, 7788. Pijper, P. J.; Van der Goot, H.; Timmerman, H.; Nauta, W. T. Eur. J. Med. Chem. 1984, 19, 399. Lowering down the loading of the reaction components for the carboxylation of 1a was not successful. A lower Ni:L ratio resulted in higher amounts of dimerization. At present (16) (17) (18) (19) (20) (21) (22) (23) (24) (25) (26) (27) (28) (29) (30) (31) we believe that a L:Ni ≥ 2 is needed to stabilize the resting state of the active Ni(0) catalyst while avoiding undesired pathways. A similar Ni:L effect has been observed in other reductive events: (a) Wu, F.; Lu, W.; Qian, Q.; Ren, Q.; Gong, H. Org. Lett. 2012, 14, 3044; (b) see also ref. 5 and 6. At present, we believe that electron-rich ligands such as L15 might prevent the binding of CO2 to the Ni center While the use of unactivated alkyl chlorides resulted in no conversion, the coupling of alkyl iodides delivered 14-19% yield with significant amounts of dimerization events. All attempts to improve these results were not successful. The coupling of phenethyl electrophiles primarily resulted in dimerization with traces of styrene derivatives Dimerization and β-hydride elimination account for the mass balance. Although a screening was conducted for substrates with low yields, the results were not satisfactory. For selected examples: (a) Wu, J.; Hazari, N. Chem. Commun. 2011, 47, 1069. (b) Shi, M.; Nicholas, K. M. J. Am. Chem. Soc. 1997, 119, 5057, and citations therein For selected reviews dealing with C–O electrophiles, see: (a) Mesganaw, T.; Garg, N. K. Org. Process Res. Dev. 2013, 17, 29. (b) Yamaguchi, J.; Muto, K.; Itami, K. Eur. J. Org. Chem. 2013, 19. (c) Rosen, B. M; Quasdorf, K. W.; Wilson, D. A.; Zhang, N.; Resmerita, A.-M.; Garg, N. K.; Percec, V. Chem. Rev. 2011, 111, 1346. The coupling of 1-bromo-4-(6-bromohexyl)benzene resulted in recovered starting material. The carboxylative event could be conducted in the dark with no erosion in yields, thus suggesting that light is not necessary for the reaction to occur. For recent metal-catalyzed hydrocarboxylation reactions: (a) Greenhalgh, M. D.; Thomas, S. P. J. Am. Chem. Soc. 2012, 134, 11900. (b) Williams, C. M.; Johnson, J. B.; Rovis, T. J. Am. Chem. Soc. 2008, 130, 14936. For related Pd-catalyzed events using isotopically-labelled intermediates: (a) Monks, B. M.; Cook, S. P. J. Am. Chem. Soc. 2012, 134, 15297. (b) Netherton, M. R.; Fu, G. C. Angew. Chem. Int. Ed. 2002, 41, 3910. (c) Stokes, B. J.; Opra, S. M.; Sigman, M. S. J. Am. Chem. Soc. 2012, 134, 11408. Care must bet taken when selecting the appropriate model substrate for isotope labelling. Whereas the J1,2 in 7a and 7c are significantly different (Scheme 3), other related γ,γunsubstituted alkyl bromide such as (5-bromopentyl)benzene had similar J1,2 in both erythro and threo-isomers. In line with such hypothesis, we found that radical scavengers such as TEMPO or BHT inhibited the reaction. Furthermore, aliphatic alcohols are obtained as byproducts when employing alkyl tosylates, an observation that is consistent with a radical pathway. See for example: Madabhushi, S.; Kumar, B. A.; Narender, R. Tetrahedron Lett. 1998, 39, 2847; (b) Closson, W. D.; Wriede, P.; Bank, S. J. Am. Chem. Soc. 1966, 88, 1581, and citations therein. In the absence of CO2 7a-d resulted predominantly in dimerization events and traces of β-hydride elimination. We cannot rule out that L14 and L16 act as a “redox-non innocent ligands”. See for example: G. D.; Martin, J. L.: McFarland, C.; Allen, O. R.; Hall, R. E.; Haley, A. D.; Brandon, R. J.: Konovalova, T.; Desrochers, P. J.; Pulay, P.; Vicic, D. A. J. Am. Chem. Soc. 2006, 128, 13175. For the intermediacy of Ni species in odd oxidation states: (a) Refs. 5a, 5b, 6a. (b) Laskowski, C. A.; Bungum, D. J.; Baldwin, S. M.; Del Ciello, S. A.; Iluc, V. M.; Hillhouse, G. L. J. Am. Chem. Soc. 2013, 135, 18272. (c) Breitenfeld, J.; Ruiz, J.; Wodrich, M. D.; Hu, X. J. Am. Chem. Soc. 2013, 135, 12004. (d) Biswas, S.; Weix, D. J. J. Am. Chem. Soc. 2013, 135, 16192. For a mechanistic hypothesis, see Supporting information. Catalytic carboxylation of unactivated alkyl bromides & sulfonates Ni catalyst / L14 or L16 R1 X H X = Br, OTs R1 CO2H H Operationally-simple & mild conditions 31 examples up to 85% yield Exquisite chemoselectivity profile No sensitive metal species CO2 (1 atm), rt N R N R=Me, L14 R=Et, L16 R