Jinqiang
Kuang
a,
Xinjun
Tang
a and
Shengming
Ma
*ab
aState Key Laboratory of Organometallic Chemistry, Shanghai Institute of Organic Chemistry, Chinese Academy of Sciences, 345 Lingling Lu, Shanghai 200032, P. R. China. E-mail: masm@sioc.ac.cn
bDepartment of Chemistry, Fudan University, 220 Handan Road, Shanghai 200433, P. R. China
First published on 16th March 2015
Zinc diiodide has been identified as an effective reagent for the efficient synthesis of trisubstituted allenes from propargylic amines. Compared to CdI2 this protocol offers a green approach. Due to the easy preparation of propargylic amines through the method developed by this group, this method provides a two-step synthesis of trisubstituted allenes from 1-alkynes, ketones, and pyrrolidine. Finally, an efficient synthesis of such trisubstituted allenes from 1-alkynes, ketones, and pyrrolidine via simple filtration has been developed. Compared with the CdI2-mediated protocol, the current protocol enjoys a much wider scope for ketones and affords functionalized allenes without further cyclization in some substrates.
(1) |
To our delight, ZnI2 could also promote the transformation effectively (entry 1, Table 1), which nicely provides chemists an environmentally friendly choice to produce trisubstituted allenes. In comparison, ZnBr2, Zn(OTf)2, CuBr, and CuBr2 might also be a choice, however, much less effective (entries 2–4 and 9, Table 1). CuI, KAuCl4·2H2O,11a and AgNO311b are extremely ineffective for such transformation (entries 10–12, Table 1). Further screening of reaction temperature and loading of ZnI2 led to a set of optimal conditions: 0.6 equiv. of ZnI2 in toluene at 120 °C (entry 5, Table 1). As a comparison, CdI2 provided the product in 88% NMR yield (entry 13, Table 1).
Entry | Metallic salt (X equiv.) | Time (h) | Yield of 2ab (%) | Recovery of 1ab (%) |
---|---|---|---|---|
a The reaction was carried out on a 0.5 mmol scale in 1.5 mL of toluene. b Determined by 1H NMR analysis of the crude reaction mixture with CH2Br2 as the internal standard. c The reaction was carried out at 110 °C. d The reaction was carried out at 100 °C. | ||||
1 | ZnI2 (0.8) | 1.1 | 78 | — |
2 | ZnBr2 (0.8) | 1.1 | 74 | — |
3 | Zn(OTf)2 (0.8) | 1.1 | 69 | — |
4 | CuBr (0.8) | 14.5 | 18 | 8 |
5 | ZnI 2 (0.6) | 0.9 | 77 | — |
6 | ZnI2 (0.5) | 1.6 | 65 | — |
7c | ZnI2 (0.6) | 1.6 | 72 | — |
8d | ZnI2 (0.6) | 2.5 | 64 | — |
9 | CuBr2 (0.6) | 10 | 17 | 6 |
10 | CuI (0.6) | 3 | Trace | 75 |
11 | AgNO3 (0.6) | 3 | 2 | 45 |
12 | KAuCl4·2H2O (0.1) | 3 | 4 | 60 |
13 | CdI2 (0.8) | 2 | 88 | — |
With the optimal conditions in hand, we then investigated the scope of the transformation. Under the mediation of ZnI2, propargylic amine 1a afforded trisubstituted allene 2a in 63% isolated yield (entry 1, Table 2). Propargylic amines 1b–1f derived from phenylacetylene, pyrrolidine, and different methyl ketones were transformed to allenes 2b–2f smoothly in moderate yields (entries 2–6, Table 2). Notably, the reaction of substrate 1c bearing an isopropyl group proceeded smoothly, albeit with a longer reaction time, which may be due to the steric hindrance of i-Pr (entry 3, Table 2). Surprisingly, propargylic amines 1g–1i also require longer reaction times (entries 7–9, Table 2). Since ethyl and n-propyl groups are generally considered as not so bulky, it indicates that the reaction rate is very sensitive to the steric effect of R2 and R3. It should be noted that acyclic non-methyl ketones are not suitable for our previously reported CdI2-mediated ATA reaction to afford trisubstituted allenes.9 When propargylic amines 1j–1q, which were prepared from differently substituted arylacetylenes, were applied in the reaction, allenes bearing different aryl groups were obtained in 50–65% isolated yields (entries 10–17, Table 2). Substitutes on the phenyl group have only a slight impact on the yields; a halogen substituent survived in the reaction (entries 11, 12 and 14–16, Table 2), affording products ready for further elaborations. When propargylic amines derived from alkyl-substituted acetylenes were treated under the standard conditions, the corresponding allene products were produced in higher yields (entries 18–21, Table 2), as compared to the arylacetylene-based propargylic amines.
Entry | 1 | Time (h) | Yield of 2b (%) | |
---|---|---|---|---|
R1 | R2; R3 | |||
a The reaction was carried out on a 1.0 mmol scale in 3 mL of toluene. b Isolated yield. | ||||
1 | Ph | –(CH2)5– (1a) | 0.8 | 63 (2a) |
2 | Ph | Me; Et (1b) | 1.1 | 57 (2b) |
3 | Ph | Me; i-Pr (1c) | 5.3 | 62 (2c) |
4 | Ph | Me; n-Bu (1d) | 1.5 | 53 (2d) |
5 | Ph | Me; n-Hex (1e) | 1.0 | 55 (2e) |
6 | Ph | Me; BnCH2 (1f) | 1.0 | 56 (2f) |
7 | Ph | Et; n-Pr (1g) | 7.8 | 52 (2g) |
8 | Ph | n-Pr; n-Pr (1h) | 5.0 | 56 (2h) |
9 | Ph | Et; Et (1i) | 4.0 | 60 (2i) |
10 | 4-MeOC6H4 | –(CH2)5– (1j) | 2.0 | 65 (2j) |
11 | 4-BrC6H4 | –(CH2)5– (1k) | 2.0 | 63 (2k) |
12 | 3-BrC6H4 | –(CH2)5– (1l) | 2.0 | 60 (2l) |
13 | 4-MeOC6H4 | Me; n-Bu (1m) | 0.9 | 50 (2m) |
14 | 4-BrC6H4 | Me; n-Bu (1n) | 2.7 | 52 (2n) |
15 | 3-BrC6H4 | Me; n-Bu (1o) | 1.0 | 55 (2o) |
16 | 2-ClC6H4 | Me; n-Bu (1p) | 1.0 | 56 (2p) |
17 | 4-MeC6H4 | Me; n-Bu (1q) | 1.7 | 57 (2q) |
18 | n-C6H13 | Me; n-Bu (1r) | 2.0 | 67 (2r) |
19 | n-C8H17 | –(CH2)5– (1s) | 1.3 | 81 (2s) |
20 | n-C8H17 | Me; Me (1t) | 1.3 | 61 (2t) |
21 | n-C8H17 | –(CH2)4– (1u) | 1.6 | 58 (2u) |
The method could also tolerate substrates bearing reactive functional groups such as hydroxyl and amide functionalities (Table 3), which could easily undergo further transformations for the preparation of other functionalized allenes.1,2c,12
Entry | 1 | Time (h) | Yield of 2 (%) |
---|---|---|---|
R1; R2; R3 | |||
a The reaction was carried out on a 1.0 mmol scale in 3 mL of toluene. Ts = p-toluenesulfonyl. b Isolated yield. | |||
1 | Me2C(OH); Me; n-Bu (1v) | 2.0 | 58b (2v) |
2 | HO(CH2)2; Me; n-Bu (1w) | 1.0 | 53b(2w) |
3 | TsNHCH2; –(CH2)5– (1x) | 1.0 | 79b (2x) |
4 | TsNHCH(n-C7H15); –(CH2)5– (1y) | 1.8 | 92b (2y) |
5 | TsNHPhCH; –(CH2)5– (1z) | 1.3 | 85b (2z) |
In fact, as a comparison, when 1x was treated with CdI2, in addition to the corresponding allenyl amide 2x, the related cyclization product 4x was also obtained in 60% yield as the major product (eqn (2)).
(2) |
Furthermore, it is interesting to observe that the reaction depends highly on the structure of the amine applied: propargylic amines 1d′ and 1d′′ from phenylacetylene, 2-hexanone, and six-membered morpholine or piperidine may also afford corresponding trisubstituted allenes, however, with a very low efficiency (Scheme 1).
Scheme 1 Synthesis of trisubstituted allenes from propargylic amines derived from morpholine or piperidine. |
Since CuBr-catalyzed synthesis of propargylic amines from 1-alkynes, ketones, and pyrrolidine10 and ZnI2-promoted the synthesis of trisubstituted allenes from propargylic amines both proceeded smoothly in the same solvent, we tried to combine them in one pot, however, a poor result was obtained (eqn (3)).
(3) |
After careful study and optimizations, we found that filtration to remove Cu(I) after the first step is essential for a successful second step. Based on this, we developed a convenient two-step synthesis of trisubstituted allenes from 1-alkynes, ketones, and pyrrolidine. Propargylic amines were firstly synthesized through the CuBr-catalyzed KA2 reaction10 and then subjected to the ZnI2-mediated transformation after a simple filtration through a short pad of silica gel to remove Cu(I). The substrate scope of the reaction was evaluated for various 1-alkynes and ketones and the typical results are summarized in Table 4. Besides phenylacetylene 3a, substituted phenylacetylenes bearing 4-Cl, 4-Br, and 4-Me substituents reacted with cyclohexanone to afford the corresponding allenes in moderate yields (entries 1–4, Table 4). Alkyl-substituted 1-alkyne 3f could react with cyclohexanone to afford alkyl-substituted allene 2s in 74% isolated yield (entry 5, Table 4). Since 2,3-allenols and 2,3-allenyl amines are useful in organic synthesis,1,2c,11 we also investigated the tolerance of amino and hydroxyl groups in the reaction: encouragingly, terminal propargylic toluenesulfonylamide 3g may be applied, affording 2,3-allenyl tosylamide 2x in 30% isolated yield (entry 6, Table 4). The reaction of terminal homopropargylic alcohol 3h with cyclohexanone afforded 2,3-allenol 2w in 53% isolated yield (entry 7, Table 4). Moreover, terminal alkynes with common protecting groups such as THP and TBS were also tolerated, producing corresponding allenes 2ad and 2ae in moderate yields (entries 8 and 9, Table 4). Cyclopentanone reacted with 1-decyne 3f to afford trisubstituted allene 2u smoothly, however, with an obviously decreased yield as compared to that of cyclohexanone (entry 10 vs. 5, Table 4). When acetone was used, allene 2t bearing two methyl groups on the allene moiety could be obtained in 49% isolated yield, although two equivalents of acetone were required to ensure the yield due to the lower boiling point of acetone (entry 11, Table 4). 2-Octanone may also be used in the reaction, affording allene 2af in moderate yield (entry 12, Table 4). Meanwhile, aromatic ketones gave a disappointing result in this reaction.
Entry | 3 | R2; R3 | t 1/t2 (h) | Yield of 2b (%) |
---|---|---|---|---|
R1 | ||||
a Ts = p-toluenesulfonyl; THP = 2-tetrahydropyranyl; TBS = t-butyldimethylsilyl. b Isolated yield. c 2.0 equiv. of acetone were used. d 1.3 equiv. of 2-octanone were used. | ||||
1 | Ph (3a) | –(CH2)5– | 1.1/1.5 | 61 (2a) |
2 | 4-ClC6H4 (3c) | –(CH2)5– | 1.5/1.5 | 57 (2aa) |
3 | 4-BrC6H4 (3d) | –(CH2)5– | 1.5/0.5 | 51 (2k) |
4 | 4-MeC6H4 (3e) | –(CH2)5– | 1.1/0.8 | 51 (2ab) |
5 | n-C8H17 (3f) | –(CH2)5– | 1.0/0.7 | 74 (2s) |
6 | TsNHCH2 (3g) | –(CH2)5– | 1.9/3.5 | 30 (2x) |
7 | HO(CH2)2 (3h) | –(CH2)5– | 0.9/1.0 | 53 (2ac) |
8 | TBSOCH2 (3i) | –(CH2)5– | 1.2/0.8 | 52 (2ad) |
9 | THPOCH2 (3j) | –(CH2)5– | 1.1/0.9 | 43 (2ae) |
10 | n-C8H17 (3f) | –(CH2)4– | 1.3/1.3 | 50 (2u) |
11c | n-C8H17 (3f) | Me; Me | 3.1/0.9 | 49 (2t) |
12d | n-C8H17 (3f) | Me; n-C6H13 | 1.9/0.6 | 45 (2af) |
When the reaction was carried out on a 50 mmol-scale, only 1.5 mol% of CuBr was needed10 and allene 2a was afforded in 67% isolated yield (eqn (4)).
(4) |
Based on recent reports of the ATA reactions,9 we proposed a plausible mechanism for this reaction. In the first place, ketoniminium 7, formed in situ from ketone and pyrrolidine, would react with alkynyl copper species 6 to give propargylic amine 1. Then the triple in propargylic amine 1 coordinates with ZnI2 which was followed by 1,5-hydride transfer and β-elimination to afford the corresponding trisubstituted allene 2 (Scheme 2). We reasoned that the coordination of Zn2+ with the in situ generated imine makes the Zn2+ much less active after the reaction.
To another Schlenk tube was added anhydrous ZnI2 (191.4 mg, 0.6 mmol). The Schlenk tube was then dried under vacuum with a heating gun. The above crude product was then dissolved in toluene (3 mL) and transferred to the Schlenk tube via a syringe under an Ar atmosphere. The Schlenk tube was then equipped with a condenser and placed in a pre-heated oil bath of 120 °C with stirring. After 1.5 h, the reaction was complete as monitored by TLC, the crude reaction mixture was cooled to room temperature and filtrered through a short pad of silica gel with a sand-core funnel eluted with ethyl ether (20 mL). After evaporation, the residue was purified by chromatography on silica gel to afford 2a13 (112.1 mg, 61%) as a liquid (eluent: petroleum ether). 1H NMR (300 MHz, CDCl3) δ 7.33–7.22 (m, 4 H, ArH), 7.20–7.10 (m, 1 H, ArH), 6.02–5.96 (m, 1 H, CH), 2.34–2.12 (m, 4 H, 2 × CH2), 1.80–1.46 (m, 6 H, 3 × CH2).
To another three-necked flask equipped with a condenser was added anhydrous ZnI2 (9.78 g, 30 mmol). The flask was then dried under vacuum with a heating gun. The above crude product was then dissolved in toluene (150 mL) and transferred to the flask via a syringe under an Ar atmosphere. The flask was then placed in a pre-heated oil bath of 120 °C with stirring. After 3.0 h, the reaction was complete as monitored by TLC, the crude reaction mixture was cooled to room temperature and filtered through a short pad of silica gel with a sand-core funnel eluted with ethyl ether (100 mL). After evaporation, the residue was purified by chromatography on silica gel to afford 2a13 (6.13 g, 67%) as a liquid (eluent: petroleum ether). 1H NMR (400 MHz, CDCl3) δ 7.32–7.25 (m, 4 H, ArH), 7.20–7.12 (m, 1 H, ArH), 6.02–5.97 (m, 1 H, CH), 2.32–2.12 (m, 4 H, 2 × CH2), 1.77–1.47 (m, 6 H, 3 × CH2).
Footnote |
† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5qo00047e |
This journal is © the Partner Organisations 2015 |