Letter Cite This: Org. Lett. 2018, 20, 3049−3052
pubs.acs.org/OrgLett
Oxy-Alkylation of Allylamines with Unactivated Alkyl Bromides and CO2 via Visible-Light-Driven Palladium Catalysis Liang Sun,† Jian-Heng Ye,† Wen-Jun Zhou,*,†,§ Xin Zeng,† and Da-Gang Yu*,† †
Key Laboratory of Green Chemistry & Technology of Ministry of Education, College of Chemistry, Sichuan University, Chengdu 610064, P. R. China § College of Chemistry and Chemical Engineering, Neijiang Normal University, Neijiang 641112, P. R. China S Supporting Information *
ABSTRACT: A selective oxy-alkylation of allylamines with unactivated alkyl bromides and CO2 via visible-light-driven palladium catalysis is reported. The commercially available Pd(PPh3)4 is used as the sole catalyst in this threecomponent reaction. A variety of tertiary, secondary, and primary alkyl bromides undergo reactions to generate important 2-oxazolidinones in high yields and selectivity. The mild reaction conditions, easy scalability, and facile derivatization of products provide great potential for application in organic synthesis and pharmaceutical chemistry.
C
utilization,9 especially the light-driven process,3h,g,9c,10 we herein report a novel oxy-alkylation of allylamines with diverse unactivated alkyl bromides and CO2 via visible-light-driven palladium catalysis (Scheme 1). This method affords valuable
arbon dioxide (CO2) has been attracting much attention as a nontoxic, nonflammable, abundant, and sustainable one-carbon (C1) building block.1 Chemical fixation of CO2 into high-value-added chemicals holds significant promise for defining new paradigms in synthetic sequences.2 Among them, the carboxylative cyclization of allylamines with CO2 is a promising strategy to generate important 2-oxazolidinones,3 which widely exist in drugs, functional materials, chiral auxiliaries, and chemical intermediates.4 In 1987, Toda reported the first carboxylative cyclization of allylamines with CO2 and iodine.3a Subsequently, Muñoz3b,c and Minakata3d significantly improved the reaction with different bases or iodine reagents (e.g., t-BuOI). More recently, our3f,h and He’s3g groups have independently developed radical-initiated carboxylative cyclization of allylamines with CO2 and activated alkylation reagents, such as Togni’s reagent, CnF2n+1I or RCF2Br, to afford fluorinecontaining 2-oxazolidinones. However, the direct cyclization of allylamines with CO2 and unactivated alkyl halides remains an unresolved challenge. In the past decade, visible light photoredox catalysis brought a renaissance to transition-metal catalysis.5 The reactivity of irradiated palladium complexes beyond the ground state has also been investigated.6,7 Very recently, the groups of Gevorgyan7a and Fu7b independently reported visible-lightdriven Pd-catalyzed Heck-type couplings with a wide scope of alkyl halides. Meanwhile, we developed the visible-light-driven, Pd-catalyzed radical alkylation of C−H bonds with general unactivated alkyl bromides.7c Mechanism studies suggested that all these transformations involved alkyl radicals, which were generated through single electron transfer (SET) between alkyl halides and an irradiated Pd(0) complex. As is well-known, abundant and readily available alkyl halides are widely used as building blocks in organic synthesis. The novel visible-lightdriven Pd-catalysis would significantly expand the toolbox of modern organic synthesis through generating diverse alkyl radicals from alkyl halides.8 With continuing interest in CO2 © 2018 American Chemical Society
Scheme 1. Oxy-Alkylation of Allylamines with CO2 and Unactivated Alkyl Bromides
alkylated 2-oxazolidinones with high yield and selectivity under mild reaction conditions. It is worth mentioning that these products share the same core structure with patented 11βHSD1 inhibitors (11β-hydroxysteroid dehydrogenase type 1 inhibitors).11 We began our investigation by examining the reaction of Nbenzyl-2-phenylprop-2-en-1-amine (1a) and 1-bromoadamantane (2a) in the presence of Pd(PPh3)4 as a photocatalyst and 1 atm of CO2 (Table 1). First, various kinds of solvents were screened (entries 1−4), and the reaction proceeded smoothly in DMSO to afford the desired product 3aa in 80% isolated yield (entry 2). Subsequently, we tested various kinds of bases (entries 5−7) and found that TBD was the best choice (entry 7), providing 3aa in 86% isolated yield. The yield of 3aa decreased to 72% HPLC yield when 2.0 equiv of TBD were used (entry 8). Notably, the catalyst loading could be further reduced to 5 mol % without loss of yield by prolonging the reaction time (entry 9). Importantly, the aminative cyclization byproduct,12 i.e., aziridine 3aa′, could hardly be detected (trace, HPLC), indicating the high chemoselectivity of this reaction. Received: April 6, 2018 Published: May 2, 2018 3049
DOI: 10.1021/acs.orglett.8b01079 Org. Lett. 2018, 20, 3049−3052
Letter
Organic Letters Table 1. Optimization of the Reaction Conditionsa
entry 1 2 3 4 5 6 7 8c 9d 10e 11f 12
base
solvent
yield (%)b
DBU DBU DBU DBU DABCO DIPEA TBD TBD TBD TBD TBD none
DMF DMSO MeCN EtOH DMSO DMSO DMSO DMSO DMSO DMSO DMSO DMSO
80 84 (80) 67 N.D. 59 20 89 (86) 72 89 (86) N.D. N.D. 23
undergo this reaction with an acceptable yield. Although steric hindrance slightly affected the yields (3ba, 3ea, and 3na), both electron-donating and weak electron-withdrawing groups on the arenes were tolerated (3fa−3na). A variety of functional groups, such as fluoro (3ha), amide (3ka), ether (3la), and ester (3ma), showed good compatibility in this reaction. The arenes with ortho-, meta-, or para-substituents were amenable to this reaction, providing the desired products in moderate to good yields. 2-Naphthyl substrate 1i also reacted smoothly. However, allylamines bearing no aryl substituent on the alkenes showed no reactivity, which may result from low reactivity of the electron-rich alkenes toward tertiary alkyl radicals and difficulty in generation of a less stable secondary radical. Next, the scope of alkyl halides was examined with slightly modified reaction conditions (Scheme 3). Gratifyingly, alkyl Scheme 3. Scope of Unactivated Alkyl Bromidesa
a
1a (0.2 mmol), 2a (0.3 mmol), Pd(PPh3)4 (10 mol %), base (0.6 mmol), solvent (0.1 M), CO2 (1 atm), rt, 10 W blue LED, 8 h. bYields were determined by HPLC using benzophenone as internal standard. Isolated yields are given in parentheses. cTBD (2 equiv). dPd(PPh3)4 (5 mol %), 24 h. eIn dark. fWithout Pd(PPh3)4. N.D. = Not detected. Ad = adamantan-1-yl. DBU = 1,8-Diazabicyclo[5.4.0]undec-7-ene. TBD = (1,5,7-Triazabicyclo[4.4.0]dec-5-ene).
Moreover, control experiments showed that both irradiation (entry 10) and a palladium catalyst (entry 11) were crucial for this process, while a base (entry 12) is important for the high conversion. With the optimized conditions (Table 1, entry 9) in hand, the scope of allylamines was investigated. As shown in Scheme 2, a variety of allylamines were examined to afford the Scheme 2. Scope of Allylamines
a
The standard reaction conditions, 2 (0.4 mmol, 2 equiv). Isolated yields are shown. b2 (0.3 mmol). c5 mol % of Pd(PPh3)4. d24 h. e60 h. f 72 h.
a
bromides (2a−2s), including tertiary, secondary, and primary ones,13 reacted smoothly to provide the desired products in moderate to good yields. This reaction tolerated many kinds of functional groups, including N-Boc (2f), oxetanyl (2l), ether (2m), and acetal (2r) moieties, all of which may be beneficial for subsequent transformations. Alkyl iodide 4 was also a suitable substrate in this reaction. To gain more insight into the reaction mechanism, a series of experiments were conducted (Scheme 4). Aziridine 3aa′ was Scheme 4. Mechanistic Investigation
a The standard reaction conditions: 1 (0.2 mmol), 2a (0.3 mmol), Pd(PPh3)4 (5 mol %), TBD (0.6 mmol), DMSO (2 mL), CO2 (1 atm), rt, 10 W blue LED, 24 h. Isolated yields are shown. b48 h. c2a (0.4 mmol).
corresponding 2-oxazolidinones. In addition to 1a, the substrate 1c with an n-butyl group on the amine nitrogen could also give the product 3ca in high yield. However, those with electronwithdrawing groups, such as Bz, Boc, and Ts, failed to afford the desired products, probably due to the poor nucleophilicity. It is worth mentioning that the primary amine substrate 1d, which was unreactive in our previous work,3f,h could also 3050
DOI: 10.1021/acs.orglett.8b01079 Org. Lett. 2018, 20, 3049−3052
Letter
Organic Letters formed in very low yield (
