Subscribe to RSS
DOI: 10.1055/s-0034-1380462
C(sp3)–H versus C(sp3)–C(sp) in Activation of Propargylic Amines under Transition-Metal Catalysis
Publication History
Received: 01 February 2015
Accepted after revision: 23 February 2015
Publication Date:
04 May 2015 (online)
Abstract
Transition-metal-catalyzed C–H and C–C bond activation procedures have attracted significant interest as environmentally friendly processes for organic synthesis. This account summarizes transition-metal-catalyzed transformations of propargylic amines through C(sp3)–H and C(sp3)–C(sp) activation, including hydrogen transfer, deacetylenative homocoupling, fragment exchange, and redox cross-dehydrogenative coupling (CDC). The generation of iminium intermediates is essential for the current transformations based on propargylic amines.
1 Introduction
2 Synthesis of Allenes from Propargylic Amines through Palladium-Catalyzed Hydrogen-Transfer Reactions
2.1 Propargylic Amines as Allenyl Anion Equivalents
2.2 Synthesis of Allenyl Carbinols
2.3 Synthesis of Heterocyclic Allenes
2.4 Mechanistic Study
2.5 One-Pot Synthesis of Allenes from Aryl Halides
3 Substitution Reactions of Propargylic Amines through Copper(I)-Catalyzed C(sp)–C(sp3) Bond Activation
3.1 Substitution Reactions of Propargylic Amines with Secondary Amines
3.2 Substitution Reactions of Propargylic Amines with 1-Alkynes
3.3 Deacetylenative Coupling with Various Propargylic Amines
4 Zinc(II)-Catalyzed Redox Cross-Dehydrogenative Coupling of Propargylic Amines and Terminal Alkynes
5 Conclusion
#
Key words
propargylic amines - C–H activation - C–C activation - palladium - copper - zinc - allenes - 1,6-enynesBiographical Sketch
Hiroyuki Nakamura received his Ph.D. from Tohoku University under the supervision of Professor Yoshinori Yamamoto in 1996. He became an assistant professor at Kyushu University (1995–1997) and then at Tohoku University (1997–2002). He worked as a visiting assistant professor at the University of Pittsburgh with Professor D. Curran (2000–2001). In 2002, he was appointed as an associate professor at Gakushuin University and promoted to professor in 2006. In 2013, he was appointed as a professor at Tokyo Institute of Technology. He received the Chemical Society of Japan Award for Young Chemists in 1999 and the Incitement Award of the Japanese Association for Molecular Target Therapy of Cancer in 2007. His research interests include synthetic methodology, medicinal chemistry, chemical biology, and neutron capture therapy.
Introduction
In recent years, strategies for transition-metal-catalyzed C–H bond activations have attracted significant interest as environmentally friendly processes in organic synthesis.[1] [2] [3] [4] [5] [6] In particular, the activation of C(sp3)–H bonds alpha to nitrogen has resulted in a variety of transformations. Murahashi and co-workers reported the first C(sp3)–H activation in amine-exchange reactions,[7] a process which has been systematically extended to catalytic oxidation reactions as mimics of enzymatic process.[6] Li and co-workers have developed the cross-dehydrogenative coupling (CDC) of tertiary amines with various pronucleophiles under oxidative conditions.[5] [8] Furthermore, Crabbé and co-workers developed the homologation of 1-alkynes for the synthesis of terminal allenes,[9] probably through a hydrogen-transfer pathway in the propargylic amine intermediates generated by the A3-type coupling.[10] [11] The key for these transformations is the generation of iminium intermediates assisted by the lone pair of electrons of the alpha nitrogen via a single-electron transfer (SET) mechanism (Scheme [1]).[5]
We have investigated the transition-metal-catalyzed transformations of propargylic amines and found that they undergo C(sp3)–C(sp) activation in the presence of copper(II) catalysts, or C(sp3)–H activation with palladium or zinc(II) catalysts (Scheme [2]).
Propargylic amines underwent C(sp3)–H activation in the presence of palladium catalysts to form allenes via the iminium intermediates A (path a, Scheme [3]).[12] In contrast, C(sp3)–C(sp) activation, in the presence of copper catalysts, was observed to generate the corresponding iminium intermediates B and copper acetylide (path b, Scheme [3]), which reacted with another equivalent of the starting material (R1 = H) to give the deacetylenative coupling products, 2-butyn-1,4-diamines, in the absence of a base,[13] or reacted with aldehydes, amines and/or 1-alkynes to give the corresponding fragment-exchanged propargylic amines.[14] Moreover, the propargylic amines underwent C(sp3)–H activation in the presence of zinc catalysts to afford the hydrogen-migrated vinyl zinc species C (path c, Scheme [3]), which reacted with 1-alkynes to give N-tethered 1,6-enynes.[15] In this case, coupling between the C(sp3)–H of the propargylic amines and C(sp)–H of the 1-alkynes took place and the two hydrogen atoms generated were trapped by the internal alkyne. As a result, redox CDC proceeded during the reaction, and additional oxidants were not necessary for this coupling. This account summarizes our recent results on the transformations of propargylic amines via transition-metal-catalyzed C(sp3)–H activation.
# 2
Synthesis of Allenes from Propargylic Amines through Palladium-Catalyzed Hydrogen-Transfer Reactions
2.1Propargylic Amines as Allenyl Anion Equivalents
Propargylic amines 1 undergo a hydrogen-transfer reaction in the presence of palladium catalysts to form allenes 2. In this transformation, the propargylic amines can be thought of as allenyl anion equivalents and react readily with electrophiles such as aromatic and vinyl halides, alkyl halides, and carbonyl compounds (Scheme [4]).[12]
The optimization of the reaction conditions for the synthesis of allene 2a was initially examined using N,N-diisopropyl-3-phenylprop-2-ynylamine (1a) as the starting material. The results are shown in Table [1]. The reaction proceeded in the presence of tris(dibenzylideneacetone)-dipalladium(0)–chloroform adduct/triphenylphosphine [Pd2(dba)3·CHCl3/Ph3P] as the catalyst at 80 °C in 1,4-dioxane to give phenylallene (2a) in only 9% yield (Table [1], entry 1). Although triphenyl phosphite [(PhO)3P], bis(diphenylphosphino)ethane (dppe), and bis(diphenylphosphino)ferrocene (dppf) were not effective ligands for the current transformation (Table [1], entries 2–4), 1,2-bis(diphenylphosphino)carborane (L1) and 1-phenylphosphinocarborane (L2) (Figure [1]) gave allene 2a in 58% and 20% yields, respectively (Table [1], entries 5 and 6). A carborane framework is known to be an electron-deficient cluster. Therefore, electron-deficient phosphine ligands would expected to be effective for this transformation. Finally, palladium(0) complexes, such as Pd2(dba)3·CHCl3 and Pd2(dba)3, combined with an electron-deficient phosphine ligand, such as tris(pentafluorophenyl)phosphine [(C6F5)3P], were found to be effective for the hydrogen-transfer reaction: phenylallene (2a) was obtained quantitatively after heating at 100 °C for 24 hours (Table [1], entries 8 and 9). The inclusion of (C6F5)3P is essential for the hydrogen-transfer reaction (Table [1], entry 10). It should be noted that the hydrogen-transfer reaction also proceeded in the presence of the palladium(II) catalyst, palladium(II) acetate [Pd(OAc)2] (Table [1], entries 11 and 12); in this case, a 61% yield of 2a was obtained in the absence of (C6F5)3P (Table [1], entry 10 vs 12).
a All reactions were carried out in 1,4-dioxane at 80 °C using a vial tube.
b Yields were determined by GC analysis using hexadecane as an internal standard.
c The reaction was carried out at 100 °C.
d The percentage of recovered 1a is indicated in parentheses.
The effects of amine substituents on the hydrogen-transfer reaction were also examined. Among the various 3-phenylprop-2-ynylamines employed, diisopropyl and dicyclohexyl groups proved to be the most suitable for the hydrogen-transfer reaction,[16] and phenylallene (2a) was obtained quantitatively. Using the optimized reaction conditions, various monosubstituted allenes were synthesized as shown in Table [2]. 3-Substituted N,N-diisopropylpropargylamines 1, which were readily prepared from aromatic halides and N,N-diisopropylpropargylamine via Sonogashira coupling, underwent the palladium-catalyzed hydrogen-transfer reaction to give the corresponding monosubstituted allenes in good to high yields. The reaction proceeded with propargylic amines possessing electron-donating groups, such as methoxy (MeO) and acetamide (AcNH) (Table [2], entries 3 and 4), and electron-withdrawing groups, including nitro (NO2), ethoxycarbonyl (CO2Et), and aldehyde (CHO) (Table [2], entries 5–7).
 |
||
|
Entry |
R |
Yield (%) |
|
1 |
Ph |
99 |
|
2 |
1-naphthyl |
67 |
|
3 |
4-MeOC6H4 |
99 |
|
4 |
4-AcNHC6H4 |
74 |
|
5 |
4-O2NC6H4 |
75 |
|
6 |
4-EtCO2C6H4 |
96 |
|
7 |
4-OHCC6H4 |
66 |
The palladium-catalyzed hydrogen-transfer reaction was also applicable for the synthesis of 1,3-disubstituted allenes (Table [3]).[17] In this case, a cyclohexyl group was found to be more effective than an isopropyl group for the desired transformation. The reactions proceeded in the presence of Pd(OAc)2 (5 mol%) and 1,2-bis(dipentafluorophenylphosphino)ethane [(C6F5)2PC2H4P(C6F5)2] (5 mol%) in chloroform at 100 °C, giving the corresponding 1,3-disubstituted allenes 3 in 53–73% yields (Table [3], entries 1–5). The use of Pd2(dba)3·CHCl3 and other phosphine ligands, such as (C6F5)3P, Ph3P, (PhO)3P, and dppe, was not effective for this transformation. It should be noted that aromatic substituents at R2 were not suitable for this transformation. Indeed, N,N-dicyclohexyl-1-phenyloct-1-yn-3-ylamine (Table [3], entry 6) and N,N-dicyclohexyl-1,3-diphenylprop-2-ynylamine (Table [3], entry 7) gave the corresponding 1,3-disubstituted allenes in only 26% and 10% yields, respectively.
 |
|||
|
Entry |
R1 |
R2 |
Yield (%) |
|
1 |
Ph |
C16H33 |
53 |
|
2 |
Cy |
C16H33 |
45 |
|
3 |
C5H11 |
C16H33 |
73 |
|
4 |
PhCH2CH2 |
C16H33 |
72 |
|
5 |
PhCH2CH2 |
Me |
63 |
|
6 |
C5H11 |
Ph |
26 |
|
7 |
Ph |
Ph |
10 |
# 2.2
Synthesis of Allenyl Carbinols
The palladium-catalyzed hydrogen-transfer reaction also proceeded with propargylic amines possessing a hydroxy group in their structure.[12] [17] As shown, various monosubstituted allenyl carbinols (Table [4], entries 1–6) and 1,3-disubstituted allenyl carbinols (Table [4], entries 7–12) were prepared from the corresponding N,N-diisopropylaminomethyl propargylic carbinols and N,N-dicyclohexylaminomethyl propargylic carbinols, as allene precursors. The combination of Pd2(dba)3·CHCl3 (2.5 mol%) and (C6F5)3P (20 mol%) as the catalyst was suitable for the synthesis of allenyl carbinols: the hydride-transfer reaction proceeded in 1,4-dioxane at 80 °C to give the corresponding allenyl carbinols and an allenyl ether in 56–92% yields (Table [4], entries 1–6). Substituted allenyl carbinols were also synthesized from the corresponding substituted allene precursors. In the case of 1,3-disubstituted allenyl carbinols, the combination of Pd(OAc)2 (5 mol%) and (C6F5)2PC2H4P(C6F5)2 (5 mol%) as the catalyst gave better yields (36–68%) in chloroform at 100 °C (Table [4], entries 7–12).
# 2.3
Synthesis of Heterocyclic Allenes
The palladium-catalyzed hydride-transfer reaction is also useful for the synthesis of heterocyclic allenes. As shown in Table [5], various heterocyclic allenes were obtained in 51–99% yields from the corresponding 3-substituted N,N-diisopropylpropargylic amines, which were themselves prepared readily from heterocyclic bromides and N,N-diisopropylpropargylamine via Sonogashira coupling, in the presence of Pd2(dba)3·CHCl3–1,2-bis(dipentafluorophenylphosphino)ethane as the catalyst.[18]
The palladium-catalyzed hydrogen-transfer reaction was also applicable for the synthesis of allenic anilinoquinazolines. Anilinoquinazolines[19] are known as inhibitors of epidermal growth factor receptor (EGFR) tyrosine kinase, which plays an important role in cell growth signaling pathways and is often overexpressed in a wide variety of cancers. Indeed, gefitinib and erlotinib, that contain an anilinoquinazoline framework, have been approved for the treatment of non-small-cell lung cancer.[20] [21] We were interested in the utility of the allene unit as an alternative functional group for pharmaceutical drug design, and have demonstrated the introduction of an allene moiety into 4-anilinoquinazolines using the palladium-catalyzed hydrogen-transfer reaction.[22,23] As shown, in Table [6], various allenic anilinoquinazolines were synthesized in 39–99% yields from the corresponding propargylic amines in the presence of Pd2(dba)3·CHCl3–(C6F5)2PC2H4P(C6F5)2 as the catalyst in chloroform. Among the synthesized allenic quinazolines, an allenyl group substituted at the 3′ position of the aniline ring exhibited efficient inhibitory potency toward EGFR tyrosine kinase (Table [6], entries 1 and 5).
|
Entry |
Propargylic amine |
Time (h) |
Allene |
Yield (%) |
|
1 |
 |
13 |
 |
64 |
|
2 |
 |
23 |
 |
91 |
|
3 |
 |
14 |
 |
92 |
|
4 |
 |
14 |
 |
70 |
|
5 |
 |
12 |
 |
56 |
|
6 |
 |
9 |
 |
86 |
|
7 |
 |
8 |
 |
48 |
|
8 |
 |
23 |
 |
42 |
|
9 |
 |
43 |
 |
68 |
|
10 |
 |
24 |
 |
38 |
|
11 |
 |
23 |
 |
36 |
|
12 |
 |
24 |
 |
47 |
a The reactions were carried out in the presence of Pd2(dba)3·CHCl3 (2.5 mol%) and (C6F5)3P (20 mol%) in 1,4-dioxane at 80 °C (entries 1–6), or in the presence of Pd(OAc)2 (5 mol%) and (C6F5)2PC2H4P(C6F5)2 (5 mol%) in CHCl3 at 100 °C (entries 7–12).
|
Entry |
Propargylic amine |
Time (h) |
Allene |
Yield (%) |
|
1 |
 |
24 |
 |
99 |
|
2 |
 |
48 |
 |
89 |
|
3 |
 |
28 |
 |
91 |
|
4 |
 |
30 |
 |
95 |
|
5 |
 |
30 |
 |
86 |
|
6 |
 |
24 |
 |
51 |
|
7 |
 |
24 |
 |
87 |
|
8 |
 |
29 |
 |
88 |
a The reactions were carried out in the presence of Pd2(dba)3·CHCl3 (2.5 mol%) and (C6F5)2PC2H4P(C6F5)2 (10 mol%) at 100 °C in CHCl3.
a Position of the allene substituent on the aromatic ring.
b The drug concentration required to inhibit the EGFR tyrosine kinase induced phosphorylation of the poly(Glu:Tyr) substrate by 50% (IC50) was determined from semilogarithmic dose–response plots.
# 2.4
Mechanistic Study
To clarify the mechanism of the palladium-catalyzed hydrogen-transfer reaction, a deuterium labeling study was performed (Scheme [5]).[18] The reaction of d-1a proceeded in the presence of Pd2(dba)3·CHCl3 (2.5 mol%) and (C6F5)3P (20 mol%) at 100 °C in 1,4-dioxane to afford allene d-2a. The kinetic isotope effect (k H/k D) was 4.0, suggesting that the activation of the C–H bond adjacent to the nitrogen atom was a kinetically relevant process in the present hydrogen-transfer reaction.
Based on the result of the deuterium labeling study, the proposed mechanisms for the palladium(0)- and palladium(II)-catalyzed hydrogen-transfer reactions are shown in Schemes 6 and 7, respectively. In the case of the palladium(0)-catalyzed process, π-coordination of palladium(0) with alkyne 1 at the carbon–carbon triple bond would form the complex 3.[24] [25] Next, hydrogen transfer from the isopropyl carbon assisted by the nitrogen lone pair of electrons would generate the palladium anionic species 4. Migration of the hydrogen on palladium to the alkyne moiety of 4 followed by rearrangement of the π-bond would give the allene 2 and the corresponding imine, while palladium(0) is regenerated. Indeed, the generation of cyclohexanone was observed in the reactions of propargylic dicyclohexylamines. It is considered that stabilization of the anionic palladium intermediate 4 in the equilibrium between 3 and 4 would accelerate the generation of allenes 2 in the catalytic cycle. Thus electron-deficient phosphine ligands, such as 1,2-bis(diphenylphosphino)carborane (L1) and (C6F5)3P, are essential for the hydrogen-transfer reaction.
In the case of the palladium(II)-catalyzed reaction, palladium(II) would be a reactive intermediate through the catalytic cycle, as shown in Scheme [7]. The coordination of Pd(OAc)2 with 1 would form the complex 5, which enters into rapid equilibrium with the iminium ion complex 6 assisted by the nitrogen lone pair of electrons. The hydride transfer from H–Pd–OAc to the alkyne moiety of 6 followed by rearrangement of π-bond would give the allene 2 and the corresponding imine, and Pd(OAc)2 is regenerated. It has been reported that the iminium ion can be generated by the insertion of palladium coordinated to the nitrogen lone pair into the carbon–hydrogen bond adjacent to nitrogen.[7] [26] Since isopropyl and cyclohexyl groups (R in Table [2]) substituted on the nitrogen were effective for the reaction, it can be assumed that these substituents stabilize iminium ion 4 (or 6) in the equilibrium thereby accelerating the catalytic cycle.
# 2.5
One-Pot Synthesis of Allenes from Aryl Halides
The synthesis of allenes from organic halides (described above) needed two steps including the Sonogashira coupling (step a) and the hydrogen-transfer reaction (step b) as shown in Scheme [8]. Therefore, we next demonstrated the palladium-catalyzed one-pot synthesis of allenes from organic halides.[27] The correct choice of ligands for the transition-metal-catalyzed processes was found to be crucial for expanding the scope of the reactions in the current transformation.
We first examined the one-pot allene synthesis from N-(4-iodophenyl)acetamide and N,N-dicyclohexylpropargylamine using various phosphine ligands (Table [7]). The reaction proceeded in the presence of Pd2(dba)3·CHCl3 (2.5 mol%), Ph3P (20 mol%), copper(I) iodide (CuI) (15 mol%) and triethylamine (Et3N) (150 mol%) to give the allene 2b in 23% yield, along with the Sonogashira coupling product 1b in 70% yield (Table [7], entry 1). The use of (PhO)3P as the ligand increased the yield of 2b to 40% (Table [7], entry 2). Although the reaction did not proceed in the presence of (C6F5)3P as the ligand, which is potent for the palladium-catalyzed hydride-transfer step (Table [7], entry 3), electron-deficient bidentate phosphine ligands, such as (C6F5)2PC2H4P(C6F5)2 and L1, were found to be effective for both the Sonogashira coupling and hydride-transfer steps (Table [7], entries 4 and 5). The use of 1,2-bis(diethoxyphosphino)-ortho-carborane (L3) (Figure [1]) led to a reduced yield of 2b (Table [7], entry 6). Other solvents, such as 1,4-dioxane (Table [7], entry 7), chloroform (Table [7], entry 8), dimethoxyethane, THF, toluene, and dimethylformamide, were not suitable for the one-pot allene synthesis.
a All reactions were carried out in the presence of Pd2(dba)3·CHCl3 (2.5 mol%), ligand [20 mol% of monodentate ligands (entries 1–3) or 10 mol% of bidentate ligands (entries 4–8)], CuI (15 mol%) and Et3N (150 mol%) at 80 °C for 3 h, after which the temperature was increased to 100 °C for 20 h.
b Recovered amount of N-(4-iodophenyl)acetamide is indicated in parentheses.
Various allenes were synthesized from aromatic halides using the optimized one-pot procedure. The results are summarized in Table [8]. Aryl iodides with electron-donating groups, such as acetamide, methoxy, and dimethyl, were converted into the corresponding allenes in 86–99% yields (Table [8], entries 1–4), in one-pot. Although allenylbenzene was produced from iodobenzene, quantitatively (Table [8], entry 5), 2-allenylnaphthalene was obtained from 2-iodonaphthalene in 60% yield (Table [8], entry 6). In the case of 4-bromoiodobenzene, the reaction took place at the iodo-substituted position, selectively (Table [8], entry 7), whereas, 1,4-diiodobenezene gave 1,4-diallenylbenzene in 65% yield by treatment with two equivalents of the propargylic amine (Table [8], entry 8). The aryl iodides with electron-withdrawing groups, such as acetyl and ester, gave the corresponding allenes in 74–99% yields (Table [8], entries 9–12). It should be noted that (C6F5)2PC2H4P(C6F5)2 was a suitable ligand in the reactions shown in entries 9 and 10 (Table [8]). We also demonstrated the one-pot synthesis of allenes from iodoheterocycles: the corresponding allenes were obtained in 67–99% yields (Table [8], entries 13–15). The combination of (C6F5)2PC2H4P(C6F5)2 with chloroform as the solvent was effective for the synthesis of heterocyclic allenes.
a All reactions were carried out in the presence of Pd2(dba)3·CHCl3 (2.5 mol%), L1 (10 mol%), CuI (15 mol%) and Et3N (150 mol%) at 80 °C for 3 h, after which the temperature was increased to 100 °C.
b The diallene product was obtained exclusively.
c (C6F5)2PC2H4P(C6F5)2 was used instead of L1.
d The reaction was carried out in CHCl3.
To clarify the reaction rate of this one-pot allene synthesis, we monitored the concentration of N-(3-iodophenyl)acetamide, the Sonogashira adduct 1c, and N-(3-allenylphenyl)acetamide (2c) during the reaction. As shown in Scheme [9,] N-(3-iodophenyl)acetamide was consumed within 20 minutes after the addition of N,N-dicyclohexylpropargylamine at 80 °C, and the Sonogashira coupling product 1c was produced in 25% yield along with the allene 2c in 3% yield. The consumption of N-(3-iodophenyl)acetamide and the formation of 1c and 2c plateaued after three hours. The yield of 2c increased when the temperature was raised. It was essential to heat the reaction up to 100 °C in order to generate the allenes in high yields.
#
# 3
Substitution Reactions of Propargylic Amines through Copper(I)-Catalyzed C(sp)–C(sp3) Bond Activation
3.1Substitution Reactions of Propargylic Amines with Secondary Amines
Propargylic amines undergo substitution reactions with various secondary amines in the presence of copper(I) bromide (CuBr) (Table [9]).[14] The substitution reactions of N,N-dicyclohexylpropargylamines with three equivalents of dibenzylamine proceeded smoothly to afford the corresponding amine-exchanged products in 55–73% yields (Table [9], entries 1–3). The reaction also proceeded with 1-substituted propargylic amines (Table [9], entries 4–6). Not only dibenzylamine, but also various secondary amines reacted with N,N-dicyclohexyl-1-phenylpropargylamine to give the corresponding amine-exchanged products in good to high yields. The use of toluene (Table [9], entries 8–11 and 15) or 1,4-dioxane (Table [9], entry 12) as the solvent instead of THF proved more effective in certain cases. The substitution reaction using N-methylaniline resulted in a relatively low 30% yield of the expected product (Table [9], entry 15). N,N-Diethyl-1-phenyl-propargylamine and N,N-diisopropyl-1-phenylpropargylamine also reacted with dibenzylamine (Table [9], entries 16 and 17).
a The reaction was carried out with three equivalents of HNR4R5 in the presence of CuBr (20 mol%) as the catalyst at 100 °C in a sealed vial.
Furthermore, interesting results were observed as shown in Scheme [10]. When propargylamine 1d was treated with dibenzylamine in the presence of 4-methylbenzaldehyde, N,N-dibenzyl-1-phenylpropargylamine (7) was obtained as the major product (63%) along with N,N-dibenzyl-1-(4-tolyl)propargylamine (8) as a minor product (11%). In contrast, when propargylamine 1e was treated with dibenzylamine in the presence of benzaldehyde, compound 7 was again obtained as the major product (40%), and alkyne 8 was formed as the minor product (18%). Based on these observations, the fragment-exchange reactions would involve C(sp)–C(sp3) bond cleavage at the propargylic position to generate the iminium cationic species and copper acetylide as shown in Scheme [11]. It occurred to us that the alkyne substitution reaction of propargylic amines would be possible in the presence of additional alkynes.
# 3.2
Substitution Reactions of Propargylic Amines with 1-Alkynes
The alkyne substitution reaction of propargylic amines with additional 1-alkynes was examined in the presence of copper(I) chloride (CuCl) as the catalyst (Scheme [12]).[14] N,N-Dicyclohexylpropargylamine was treated with three equivalents of phenylacetylene in the presence of CuCl (20 mol%) and different bases to produce the corresponding propargylic amines 9. Tributylamine (9a–c,e), trioctylamine (9f–i) and disodium hydrogen phosphate (Na2HPO4) (9d,j–l) were effective as bases for the alkyne substitution reaction. Various propargylic amines and 1-alkynes could be applied in this reaction, including examples with labile functional groups such as hydroxy (9e) and ester (9l).
A plausible mechanism is shown in Scheme [13]. Propargylic amine 1 would undergo the C(sp)–C(sp3) bond cleavage initiated by copper(I) to generate the iminium ion 10 and the copper alkynylide 11. The amino group (–NR1 2) of the iminium ion 10 would be substituted with another amino group (–NR3R4) by addition of another secondary amine (HNR3R4) to form the corresponding iminium ion 10′, which reacts with the copper alkynylide 11 to afford the amine-substituted product 9. According to the results shown in Scheme [10], it is also possible to exchange the R2 group of iminium ion 10′ by adding an extra aldehyde. Furthermore, the alkyne-substitution reaction also proceeded in the presence of an additional alkyne.
# 3.3
Deacetylenative Coupling with Various Propargylic Amines
The copper-catalyzed homocoupling reaction of terminal alkynes, known as the Glaser–Hay coupling reaction, is a well-established process for the synthesis of symmetric 1,3-diynes. In general, the Glaser–Hay reaction and the related deprotonative dimerizations proceed in the presence of amines and oxygen as the base and the terminal oxidant, respectively. Surprisingly, we found that the deacetylenative coupling reaction proceeded in the absence of an amine.[13] The effect of 4-(N,N-dimethylamino)pyridine (DMAP) as the base on the coupling reactions is shown in Table [10]. The reaction proceeded in the presence of CuCl (10 mol%) as the catalyst in THF at 100 °C for 24 hours to give the deacetylenated coupling product 13a in 59% yield along with a 27% recovery of substrate 1f (Table [10], entry 1). The addition of DMAP increased the quantity of the Glaser–Hay-type homocoupling product 12 (Table [10], entries 2–5).
a Isolated total yield of products 12 and 13a after silica gel chromatography. Recovered 1f is indicated in parentheses.
The deacetylenative coupling reaction proceeded with various propargylic amines and the results are summarized in Scheme [14]. Propargylic amines having an aliphatic group at R1 were converted into the corresponding 1,4-diamino-2-butynes 13b–i in the presence of CuCl (20 mol%) as the catalyst in 45–84% yields. However, propargylic amines with an aromatic group at R1 gave the corresponding products 13j–l in lower yields (39–44%), whilst a propargylic amine with no substituent at R1 gave product 13m in 37% yield. These results indicate that electron-donating substituents at the propargylic position lead to efficient deacetylenative coupling reactions. In all cases, the Glaser–Hay-type homocoupling products were not detected.
A plausible mechanism is shown in Scheme [15]. Propargylic amine 1 would undergo the C(sp)–C(sp3) bond cleavage in the presence of CuCl to generate iminium intermediate 10 and a copper acetylide, which reacts with another molecule of propargylic amine 1 to afford the corresponding propargylic acetylide 14 and acetylene. Addition of 14 to the iminium intermediate 10 would produce the deacetylenated coupling product 13 and CuCl is regenerated. The C(sp)–C(sp3) bond cleavage assisted by the lone pair of electrons of the nitrogen is essential for the deacetylenative coupling prior to the Glaser–Hay homocoupling reaction.
#
# 4
Zinc(II)-Catalyzed Redox Cross-Dehydrogenative Coupling of Propargylic Amines and Terminal Alkynes
Propargylic amines undergo cross-dehydrogenative coupling (CDC) with terminal alkynes in the presence of zinc(II) catalysts to afford N-tethered 1,6-enynes.[15] In this reaction, C(sp3)–C(sp) bond formation occurs between the carbon adjacent to the nitrogen atom in the propargylic amine and the terminal carbon of the alkyne. In general CDC reactions, stoichiometric oxidants such as tert-butyl hydroperoxide are necessary. However, in the current redox CDC (alternatively called cross-dehydrogenative hydrogenative coupling), the C–C triple bond of the propargylic amines acts as an internal oxidant and reacts with the generated hydrogen atoms resulting in its transformation into a C–C double bond.
Scheme [16] shows the redox CDC of ethynyl benzene with various propargylic amines. The reactions proceeded in the presence of zinc(II) bromide (20 mol%) in toluene at 100 °C for 24 hours to give the corresponding N-tethered 1,6-enynes 15. In the cases of 1,2,3,4-tetrahydroisoquinoline and N-methylbutylamine derivatives, regioselective CDCs were observed: the N-tethered 1,6-enynes 15f and 15g were obtained predominantly. These results indicate that the alkynylation takes place preferentially at the benzylic position adjacent to the nitrogen and that the secondary C–H bond undergoes the alkynylation rather than the primary C–H bond. With propargylic amines having a substituent at R4 (Me or Bu), the corresponding 1,6-enynes 15h and 15i were obtained in moderate yields with diastereomeric ratios of 1.7:1 and 1.4:1, respectively.
The redox CDCs of N,N-diisobutylpropargylamine with various 1-alkynes were also examined (Scheme [17]). 1-Alkynes having aromatic substituents afforded the corresponding N-tethered 1,6-enynes in better yields compared to those having aliphatic, trimethylsilyl, and ester substituents (compare 15a–d vs 15e–n). Diynes also underwent the redox CDC to give products 15o and 15p in 53% and 30% yields, respectively.
Deuterium labeling studies were carried out to obtain further insight into the reaction mechanism. As shown in Scheme [18], the deuterium at the carbon adjacent to the nitrogen atom of d-16 shifted to a position cis to the amino methyl group in 1,6-enyne d-17. The kinetic isotope effect (k H/k D) was 3.0, suggesting that the activation of the C–H bond adjacent to the nitrogen atom was a kinetically relevant process in this reaction. Furthermore, H/D scrambling between d-17 and 18 was not observed in the presence of ZnBr2 as the catalyst, indicating that the redox CDC proceeded with an intramolecular hydride shift of the propargylic amine. In the case of the reaction of 16 with ethynyl benzene (93% d), the deuterium atom incorporated at the alkene position of d-17 was cis orientated and was located preferentially on the terminal carbon of the alkene position rather than on the substituted carbon.
Based on the observations from the deuterium labeling study, the proposed mechanism for the redox CDC is shown in Scheme [19]. ZnBr2 would be expected to react with the 1-alkyne to generate a zinc alkynylide species, and an equilibrium might exist between 19 and 20 in the case of propargylic amine 1 bearing a terminal alkyne (R4 = H). Next, hydride migration (H a ) to the triple bond of propargylic amine 19 would take place with the assistance of the lone pair of electrons of the alpha nitrogen and coordination of zinc(II) to the give complex 21 which, in turn, would afford iminium intermediate 22 via a 1,5-hydride shift. Attack of the zinc alkynylide on the iminium ion followed by protonation of the σ-vinyl zinc complex 23 furnishes 1,6-enyne 15. Finally, C(sp3)–C(sp) bond formation between C(sp)–H b of the 1-alkyne and C(sp3)–H a adjacent to the nitrogen atom of the propargylic amine 1 takes place with the generation of an allyl group in the molecule.
# 5
Conclusion
This account has described C(sp3)–H activation versus C(sp3)–C(sp) activation in propargylic amines. Palladium catalysts activate C(sp3)–H adjacent to the nitrogen atom in propargylic amines to generate iminium intermediates that underwent hydrogen transfer to give the corresponding allenes. In this transformation, the generated iminium moiety acted as a good leaving group. Copper catalysts activated C(sp3)–C(sp) at propargylic positions to generate iminium species and copper acetylides that underwent deacetylenative homocoupling or fragment-exchange reactions depending on the presence of additional amines, aldehydes, or 1-alkynes in the reaction. Furthermore, zinc catalysts activated C(sp3)–H similar to palladium catalysts. In these cases, the vinyl zinc species generated by migration of the hydrogen from the carbon atom adjacent to the nitrogen atom in the propargylic amine is stable and the iminium moiety in the vinyl zinc intermediates can undergo nucleophilic attack by the alkynyl zinc species to give N-tethered 1,6-enynes. This cross-coupling attracts attention from an atom economic point-of-view because no waste is generated at the end of the reaction. Since iminium intermediates are key for the transformations described in this account, C–H and C–C activations assisted by adjacent nitrogen atoms are attractive strategies for the construction of new molecular skeletons in organic synthesis.
#
#
Acknowledgment
I gratefully acknowledge my colleagues, Professor J. F. Biellmann, Mr. T. Kamakura, Mr. M. Ishikura, Mr. S. Onagi, Mr. S. Tashiro, Mr. A. Kimura, Miss. Y. Tanaka, Miss. M. Hatori, Dr. H. S. Ban, Dr. Y. Kim, and Dr. T. Sugiishi, for their devotion to this research.
-
References
- 1 Dyker G. Angew. Chem. Int. Ed. 1999; 38: 1698
- 2 Jia CG, Kitamura T, Fujiwara Y. Acc. Chem. Res. 2001; 34: 633
- 3 Ritleng V, Sirlin C, Pfeffer M. Chem. Rev. 2002; 102: 1731
- 4 Chen X, Engle KM, Wang DH, Yu JQ. Angew. Chem. Int. Ed. 2009; 48: 5094
- 5 Li CJ. Acc. Chem. Res. 2009; 42: 335
- 6 Murahashi SI, Zhang D. Chem. Soc. Rev. 2008; 37: 1490
- 7 Murahashi SI, Hirano T, Yano T. J. Am. Chem. Soc. 1978; 100: 348
- 8 Li ZP, Bohle DS, Li CJ. Proc. Natl. Acad. Sci. U.S.A. 2006; 103: 8928
- 9 Crabbé P, Fillion H, André D, Luche J.-L. J. Chem. Soc., Chem. Commun. 1979; 859
- 10 Gommermann N, Koradin X, Polborn K, Knochel P. Angew Chem. Int. Ed. 2003; 42: 5763
- 11 Peshkov VA, Pereshivko OP, Van der Eycken EV. Chem. Soc. Rev. 2012; 41: 3790
- 12 Nakamura H, Kamakura T, Ishikura M, Biellmann JF. J. Am. Chem. Soc. 2004; 126: 5958
- 13 Kim Y, Nakamura H. Chem. Eur. J. 2011; 17: 12561
- 14 Sugiishi T, Kimura A, Nakamura H. J. Am. Chem. Soc. 2010; 132: 5332
- 15 Sugiishi T, Nakamura H. J. Am. Chem. Soc. 2012; 134: 2504
- 16 Nakamura H, Tashiro S, Kamakura T. Tetrahedron Lett. 2005; 46: 8333
- 17 Nakamura H, Ishikura M, Sugiishi T, Kamakura T, Biellmann JF. Org. Biomol. Chem. 2008; 6: 1471
- 18 Nakamura H, Onagi S, Kamakura T. J. Org. Chem. 2005; 70: 2357
- 19 Fry D, Kraker A, McMichael A, Ambroso L, Nelson J, Leopold W, Connors R, Bridges A. Science 1994; 265: 1093
- 20 Linardou H, Dahabreh IJ, Bafaloukos D, Kosmidis P, Murray S. Nat. Rev. Clin. Oncol. 2009; 6: 352
- 21 Kobayashi S, Boggon TJ, Dayaram T, Janne PA, Kocher O, Meyerson M, Johnson BE, Eck MJ, Tenen DG, Halmos B. N. Engl. J. Med. 2005; 352: 786
- 22 Ban HS, Onagi S, Uno M, Nabeyama W, Nakamura H. ChemMedChem 2008; 3: 1094
- 23 Ban HS, Tanaka Y, Nabeyama W, Hatori M, Nakamura H. Bioorg. Med. Chem. 2010; 18: 870
- 24 Zeni G, Larock RC. Chem. Rev. 2004; 104: 2285
- 25 Carvalho MF. N. N, Galvao AM, Pombeiro AJ. L, Cermak J, Sabata S, Vojtisek P, Podlaha J. J. Organomet. Chem. 2000; 598: 318
- 26 Murahashi S, Watanabe T. J. Am. Chem. Soc. 1979; 101: 7429
- 27 Nakamura H, Kamakura T, Onagi S. Org. Lett. 2006; 8: 2095
-
References
- 1 Dyker G. Angew. Chem. Int. Ed. 1999; 38: 1698
- 2 Jia CG, Kitamura T, Fujiwara Y. Acc. Chem. Res. 2001; 34: 633
- 3 Ritleng V, Sirlin C, Pfeffer M. Chem. Rev. 2002; 102: 1731
- 4 Chen X, Engle KM, Wang DH, Yu JQ. Angew. Chem. Int. Ed. 2009; 48: 5094
- 5 Li CJ. Acc. Chem. Res. 2009; 42: 335
- 6 Murahashi SI, Zhang D. Chem. Soc. Rev. 2008; 37: 1490
- 7 Murahashi SI, Hirano T, Yano T. J. Am. Chem. Soc. 1978; 100: 348
- 8 Li ZP, Bohle DS, Li CJ. Proc. Natl. Acad. Sci. U.S.A. 2006; 103: 8928
- 9 Crabbé P, Fillion H, André D, Luche J.-L. J. Chem. Soc., Chem. Commun. 1979; 859
- 10 Gommermann N, Koradin X, Polborn K, Knochel P. Angew Chem. Int. Ed. 2003; 42: 5763
- 11 Peshkov VA, Pereshivko OP, Van der Eycken EV. Chem. Soc. Rev. 2012; 41: 3790
- 12 Nakamura H, Kamakura T, Ishikura M, Biellmann JF. J. Am. Chem. Soc. 2004; 126: 5958
- 13 Kim Y, Nakamura H. Chem. Eur. J. 2011; 17: 12561
- 14 Sugiishi T, Kimura A, Nakamura H. J. Am. Chem. Soc. 2010; 132: 5332
- 15 Sugiishi T, Nakamura H. J. Am. Chem. Soc. 2012; 134: 2504
- 16 Nakamura H, Tashiro S, Kamakura T. Tetrahedron Lett. 2005; 46: 8333
- 17 Nakamura H, Ishikura M, Sugiishi T, Kamakura T, Biellmann JF. Org. Biomol. Chem. 2008; 6: 1471
- 18 Nakamura H, Onagi S, Kamakura T. J. Org. Chem. 2005; 70: 2357
- 19 Fry D, Kraker A, McMichael A, Ambroso L, Nelson J, Leopold W, Connors R, Bridges A. Science 1994; 265: 1093
- 20 Linardou H, Dahabreh IJ, Bafaloukos D, Kosmidis P, Murray S. Nat. Rev. Clin. Oncol. 2009; 6: 352
- 21 Kobayashi S, Boggon TJ, Dayaram T, Janne PA, Kocher O, Meyerson M, Johnson BE, Eck MJ, Tenen DG, Halmos B. N. Engl. J. Med. 2005; 352: 786
- 22 Ban HS, Onagi S, Uno M, Nabeyama W, Nakamura H. ChemMedChem 2008; 3: 1094
- 23 Ban HS, Tanaka Y, Nabeyama W, Hatori M, Nakamura H. Bioorg. Med. Chem. 2010; 18: 870
- 24 Zeni G, Larock RC. Chem. Rev. 2004; 104: 2285
- 25 Carvalho MF. N. N, Galvao AM, Pombeiro AJ. L, Cermak J, Sabata S, Vojtisek P, Podlaha J. J. Organomet. Chem. 2000; 598: 318
- 26 Murahashi S, Watanabe T. J. Am. Chem. Soc. 1979; 101: 7429
- 27 Nakamura H, Kamakura T, Onagi S. Org. Lett. 2006; 8: 2095
