Synlett 2019; 30(11): 1329-1333
DOI: 10.1055/s-0037-1611825
letter
© Georg Thieme Verlag Stuttgart · New York

Synthesis of 2,3-Dihydrotryptamines from Amide Solvents and Acyclic Materials through Metal-Free Amidoalkylarylation of Unactivated Alkenes

Yubin Ji
a   Research Center on Life Sciences and Environmental Sciences, Harbin University of Commerce, Harbin 150076, P. R. of China   Email: liuyj691@nenu.edu.cn
,
Siping Yang
b   Yunnan Engineering Technology Research Center for Plastic Films, Department of Chemistry, Kunming University, Kunming 650214, P. R. of China   Email: liangdq695@nenu.edu.cn
,
Shiyi Lin
b   Yunnan Engineering Technology Research Center for Plastic Films, Department of Chemistry, Kunming University, Kunming 650214, P. R. of China   Email: liangdq695@nenu.edu.cn
,
Yan Wang
b   Yunnan Engineering Technology Research Center for Plastic Films, Department of Chemistry, Kunming University, Kunming 650214, P. R. of China   Email: liangdq695@nenu.edu.cn
,
Chenfeng Ji
a   Research Center on Life Sciences and Environmental Sciences, Harbin University of Commerce, Harbin 150076, P. R. of China   Email: liuyj691@nenu.edu.cn
,
Yingjie Liu*
a   Research Center on Life Sciences and Environmental Sciences, Harbin University of Commerce, Harbin 150076, P. R. of China   Email: liuyj691@nenu.edu.cn
,
b   Yunnan Engineering Technology Research Center for Plastic Films, Department of Chemistry, Kunming University, Kunming 650214, P. R. of China   Email: liangdq695@nenu.edu.cn
› Author Affiliations

We gratefully acknowledge the financial support from the National Natural Science Foundation of China (21702083), the Program for Innovative Research Team (in Science and Technology) in Universities of Yunnan Province, the Yunnan Ten Thousand Talent Program for Young Top-Notch Talents, and the Science Foundation Project of Harbin University of Commerce (18XN067).
Further Information

Publication History

Received: 12 March 2019

Accepted after revision: 24 April 2019

Publication Date:
03 May 2019 (online)

 


Abstract

The first synthesis of 2,3-dihydrotryptamines from acyclic materials and an exo-selective amidoalkylation/cyclization cascade of N-allyl anilines through α-C(sp3)–H functionalization of simple amides across unactivated alkenes are presented. This reaction proceeds in mixed aqueous media and under metal-free conditions and features a broad substrate scope and a simple operation.


#

Dehydrogenative C–H functionalization is considered as an ideal transformation to construct C–C or C–heteroatom bonds due to its atom- and step-economy as well as the abundance of C–H bonds.[1] In this regard, C(sp3) –H functionalization is a significant challenge because of the high bond-dissociation energy and the lack of π-electrons which could interact with a transition-metal catalyst or an electrophile.[2] Of particular importance is the C(sp3) –H functionalization involving solvent molecules, which are inexpensive and abundant and play dual roles as both reaction media and C–H sources, thus eliminating the need for adding an additional coupling agent. For example, the α-C(sp3)–H functionalization of amide solvents could enable the direct introduction of an amidyl functionality into various systems such as alkynes,[3] isocyanides,[4] arenes,[5] heteroarenes,[6] heteroatoms,[7] [8] carbonyls,[9] or activated olefins.[10] [11] In violent contrast, there are only two papers dealing with the α-C(sp3)–H functionalization reaction of simple amides across unactivated alkenes, and the reactions are both assisted by a 1,2-aryl migration process (Scheme [1a]).[12] Hence, new chemistry for such functionalization still needs to be developed.

Zoom Image
Scheme 1 Synthesis of 2,3-dihydrotryptamines

Indolic motifs constitute privileged structures in natural products as well as in medicinal and biological chemistry because of their remarkable activities.[13] For example, 2,3-dihydrotryptamines are lead compounds for drug design and are found in skeletons of many alkaloids and clinical drugs.[14] They are also feedstocks for natural products and natural product-like structures.[15] The preparation of 2,3-dihydrotryptamines, however, is strictly restricted to the derivatization reaction of parent tryptamines (Scheme [1b]),[16] and the synthesis from acyclic materials has not been reported. Thanks to the development of N-arylacrylamide chemistry,[17] 2,3-dihydrotryptamine-2-ones could be prepared from N-arylacrylamides and amide solvents,[10] yet 2,3-dihydrotryptamine counterparts could not be accessed through oxindole reduction owing to severe functional-group intolerance.[18] In continuation of our interest in radical-based synthesis of bioactive structures,[19] we hypothesized that the α-C(sp3)–H functionalization of simple amides with the unactivated alkenic bonds of N-allyl anilines could permit the first construction of 2,3-dihydrotryptamines from acyclic materials (Scheme [1c]). Here we report the successful execution of this synthetic plan, and the title reaction proceeds in mixed aqueous media under metal-free and simple conditions and features a broad substrate scope and exo selectivity.

We began our studies by investigating the amidoalkylation/cyclization cascade of N-(2-methylallyl) acetanilide (1a1) with N,N-dimethylacetamide (DMA, Table [1]). Exposure of 1a1 to 10 mol% Cu2O and 3 equivalents of di-tert-butyl peroxide (DTBP) in DMA at 120 °C furnished 2,3-dihydrotryptamine product 2a1 in 36% yield (Table [1], entry 1). A copper salt seemed unnecessary, a similar yield was achieved in the absence of it (Table [1], entry 2). With dicumyl peroxide (DCP, Table [1], entry 3) as the oxidative initiator, dihydrotryptamine 2a1 was afforded in an improved yield of 44%, while the use of tert-butyl peroxybenzoate (TBPB, Table [1], entry 4) led to a slight depreciation in yield. Oxidants possessing a weaker O–O bond, such as benzoyl peroxide (BPO) and K2S2O8, proved ineffective for this transformation, probably due to polarity mismatching[20] during the hydrogen-atom transfer (HAT) from DMA to O radicals. A polar peroxide bond might be difficult to undergo homolysis in DMA, and tert-butyl hydroperoxide (TBHP) and Oxone are unusable initiators as well (Table [1], entry 5). The attempt to perform the reaction in other solvents using 10 equivalents of DMA met with no success (Table [1], entries 6–8). While dihydrotryptamine 2a1 was delivered in very poor yields in 1,2-dichloroethane (DCE, Table [1], entry 6) or toluene (Table [1], entry 7), the use of polar solvents such as tetrahydrofuran (THF), MeNO2, dimethyl sulfoxide (DMSO) or ethanol, proved fruitless (Table [1], entry 8). Interestingly, the yield of 2a1 was improved to 71% by changing the solvent to an DMA/H2O mixture (5:1, v/v, Table [1], entry 9). At this stage, the origin of this beneficial effect remains unclear, yet it might be associated with the hydrogen-bonding interaction. Further increasing the ratio of water to DMA led to diminished yields (Table [1], entries 10–12). Whereas no reaction occurred at 80 °C (Table [1], entry 13), the yield of dihydrotryptamine 2a1 was compromised with a reduced loading of DCP (Table [1], entry 14).

Table 1 Optimization of Reaction Conditionsa

Entry

Oxidant (equiv)

Solvent

Temp (°C)

Yield (%)

 1b

DTBP (3)

DMA

120

36

 2

DTBP (3)

DMA

120

33

 3

DCP (3)

DMA

120

44

 4

TBPB (3)

DMA

120

29

 5

BPO, TBHP,c K2S2O8, or Oxone (3)

DMA

120

trace

 6d

DCP (3)

DCE

120

15

 7d

DCP (3)

toluene

120

17

 8d

DCP (3)

THF, MeNO2, DMSO, or EtOH

120

trace

 9

DCP (3)

DMA/H2O (5:1e)

120

71

10

DCP (3)

DMA/H2O (3:1e)

120

63

11

DCP (3)

DMA/H2O (2:1e)

120

32

12

DCP (3)

DMA/H2O (1:2e)

120

trace

13

DCP (3)

DMA/H2O (5:1e)

 80

nr

14

DCP (2)

DMA/H2O (5:1e)

120

46

a Reaction conditions: 1a1 (0.2 mmol), solvent (1.2 mL), Ar, 6 h.

b Cu2O (10 mol%) was added as a catalyst.

c 5.0–6.0 mol/L in decane.

d 10 equiv of DMA were added.

e Ratios given as v/v.

Having developed optimized reaction conditions, we subsequently explored the scope of this transformation (Scheme [2]).[21] N-(2-Methylallyl) acetanilides bearing a methyl, bromo, or chloro group at the para position of the N-aryl group reacted with DMA to afford 5-substituted dihydrotryptamines 2a24 in 68–76% yields, whereas 5-phenyl product 2a5 was delivered in only a moderate yield. Propionyl, octanoyl, decanoyl, and dodecanoyl N-protecting groups (PG) were all tolerated, and the corresponding 2,3-dihydrotryptamines 2be were synthesized in moderate to high yields from N-allylated anilides bearing an electron-neutral, -rich, or -deficient N-aryl group. Stearoyl-protected products 2f1,2 were formed in modest yields probably due to steric conflict, while a high yield of indoline 2g was achieved by using the pivaloyl N-PG. tert-Butyloxy-carbonyl and ethylsulfonyl substituents on the nitrogen atom are compatible with this transformation as well, affording the corresponding 2,3-dihydrotryptamines 2h1,2 in 61% and 44% yields, respectively. An electron-withdrawing N-PG is essential, and allylated N-methylaniline reacted to deliver N-methyl indoline 2i in a very poor yield. The allylated acetanilide having a 3-chloro group on the N-aryl ring is also a competent substrate, although poor regioselectivity was observed, affording 4- and 6-chloro 2,3-dihydrotryptamines 2j and 2j′ in 43% and 25% yields, respectively. Interestingly, more encumbered product 2j is the major isomer, and the origin of such regioselectivity might be associated with intermediate stability. This amidoalkylation/cyclization cascade might be sensitive to steric effects, because the use of the allylated acetanilide bearing an ortho substituent as the substrate proved fruitless.

Zoom Image
Scheme 2 Synthesis of 2,3-dihydrotryptamines from N-allyl anilines

Our initial studies focused on the amidoalkylative cyclization of N-(2-methylallyl) substrates, in order to steer clear of formal 6-endo-trig mode of ring closure, which is a potential competing reaction according to Baldwin’s rule.[22] After achieving excellent exo selectivity in all the above reactions, we evaluated a non-methyl-branched N-allyl aniline as the substrate. To our delight, corresponding 5-exo-trig product 2k was furnished selectively in 60% yield, and still no endo product was observed.

The scope of the amide solvent was also investigated. When N,N-dimethylformamide (DMF)/H2O mixture was used as the solvent, the desired dihydrotryptamine product 2l was produced in 36% yield, along with amidated indoline 2l′ furnished in 15% yield. Pure products were obtained in the reactions carried out in aqueous N,N-diethylacetamide or N,N-dimethylpropionamide, yet their NMR spectra are too complex and defy analysis due to rotamers and/or diastereoisomers. The use of N-methylpyrrolidin-2-one (NMP) furnished thermodynamically controlled product 2m in a high yield. Interestingly, a reversed selectivity was observed in the cases of N-butyl-N-methylacetamide or N-isopropyl-N-methylacetamide probably because of the increased steric bulks, and the α-C(sp3)–H functionalization occurred on the less hindered N-alkyl groups, affording 2,3-dihydrotryptamines 2n,o in poor yields. It is worthy of notice that the present amidoalkylation/cyclization cascade could be extended to N-methylacetamide which bears a free amidyl N–H moiety and is unusable in other α-C(sp3)–H functionalizations of amides, providing acetyl-protected primary amine 2p albeit in a moderate yield. Thus, we have presented a general protocol for the assembly of both secondary and primary 2,3-dihydrotryptamines protected by an acyl group.

Some control experiments were conducted to probe the reaction mechanism (Scheme [3a]). As might be expected, the amidoalkylative cyclization of the model substrate 1a1 under standard conditions was almost completed suppressed upon addition of 2 equivalents of 2,2,6,6-tetramethylpiperidine-1-oxyl (TEMPO, Scheme [3a1]), while 2,3-dihydrotryptamine product 2a1 was produced in a modest yield with butylated hydroxytoluene (BHT, Scheme [3a2]) as the radical scavenger. Furthermore, amidoalkyl-BHT adduct 3 was formed in 14% yield in the BHT experiment, and the yield was improved to 67% by subjecting BHT to our standard conditions (Scheme [3a3]). These results suggest that amidoalkyl radicals might be involved in the title reaction.

Zoom Image
Scheme 3 Mechanistic investigations

On the basis of the above observations and previous reports,[3] [4] [5] [6] [7] [8] [9] [10] [11] [12] a plausible mechanism is proposed (Scheme [3b]). In the beginning, thermal decomposition of DCP occurs to afford cumyloxyl radical A, which subsequently abstracts a hydrogen atom from DMA to form amidoalkyl radical B. Addition of B to the N-tethered alkenic bond of substrate 1a1 leads to alkyl radical intermediate C with a newly formed C–C bond. Intramolecular radical tapping by the phenyl group ensues, affording ring closure intermediate D. Finally, HAT from D to cumyloxyl radical A occurs to release the 2,3-dihydrotryptamine product 2a1 as well as a molecule of 2-phenylpropan-2-ol.

In conclusion, a metal-free and exo-selective amidoalkylation/cyclization cascade of N-allyl anilines through α-C(sp3)–H functionalization of simple amides across unactivated alkenes was developed, allowing the first synthesis of 2,3-dihydrotryptamines from acyclic materials. Unactivated double bonds act as radical acceptors, while simple amides play dual roles as both solvents and radical precursors. This reaction proceeds in aqueous media and features a broad substrate scope and simple conditions.


#

Supporting Information



Zoom Image
Scheme 1 Synthesis of 2,3-dihydrotryptamines
Zoom Image
Scheme 2 Synthesis of 2,3-dihydrotryptamines from N-allyl anilines
Zoom Image
Scheme 3 Mechanistic investigations