2020 Volume 68 Issue 4 Pages 332-335
Herein, we describe a novel synthetic method for 2,5-disubstituted tetrazoles from 5-substituted tetrazoles using cobalt-catalyzed intermolecular hydroamination reaction of nonactivated olefins. Owing to its mild conditions, the method enabled the use of substrates having acid-labile functional groups, such as silyloxy and methoxymethyloxy groups. By using optically active cobalt complexes, asymmetric intermolecular hydroamination of nonactivated olefins, a longstanding challenge in synthetic organic chemistry, was developed to produce optically active disubstituted tetrazoles.
Tetrazoles are one of the most important heterocycles found in pharmaceutical sciences, material sciences, and synthetic chemistry.1) Structurally, the 1,5- and 2,5-substitutions are the two possible regioisomeric forms in disubstituted tetrazoles. 1,5-Disubstituted tetrazoles are difficult to synthesize via common SN2 alkylation of 5-substituted tetrazoles. However, a variety of multi-component coupling reactions and 1,5-electrocyclization reactions have allowed easy access to diverse 1,5-isomers.1,2) On the other hand, the synthesis of 2,5-disubstituted tetrazoles are mostly limited to traditional SN2 and SN1 reactions as well as Mitsunobu chemistries, in which (pseudo)alkyl halides and alcohols were used as alkylating reagents, respectively.2–4) Furthermore, these reactions typically afford small amounts of 1,5-isomers as side products and require separation of each isomer.3) Olefins are attractive alternatives as alkylating agents because of their high abundance, low cost, and the high atom economy of their transformation. However, the scope of the olefins for this purpose is limited to activated olefins, such as allenes5) and olefins conjugated with electron-withdrawing groups,6–11) or hydroamination with simple olefins requires harsh reaction conditions.12) We recently developed cobalt-catalyzed hydroamination (HA) reactions of nonactivated olefins with benzotriazoles, which proceeded under mild conditions with high chemo- and regio-selectivities.13) In this study, we have applied our procedure to HA reactions using tetrazoles as amine moieties.12,14–16)
We began the optimization of the HA reaction between 4-phenyl-1-butene (1a) and 5-phenyltetrazole (2a) (Table 1) and found that the HA reaction proceeded at a higher concentration in toluene (0.2 M) than that with benzotriazoles, which required a higher dilution of 0.01 M for obtaining optimal yields (for details of the optimization of reaction conditions, see: Supplementary Materials). The choice of the silane and the oxidant was crucial, and the combination of PhSiH3 and 1-fluoro-2,4,6-trimethylpyridinium tetrafluoroborate (TMFP-BF4) alone gave 3a in satisfactory yield (55% GC yield) (entries 1–5). Cobalt complex Co2 comprising a tetramethylethylenediamine backbone gave the same result as that with Co1 (entry 6). Co3, which has the (±)-1,2-diphenylethylenediamine backbone, was found to be the best catalyst for this reaction (entry 7). Next, we explored the optimization of the PhSiH3 equivalents and found that they could be reduced to 1 eq (entry 8). Although PhSiH3 bears three hydrogen atoms on the silicon atom, only one of them seems to participate in the reaction, because reducing the amount of silane to 0.6 and 0.4 eq resulted in lowered yields of 64 and 44%, respectively (entries 9 and 10). Finally, the amount of TMFP-BF4 could be reduced to 1.5 eq without loss of yield (entry 11). It is worth noting that neither 1,5-disubstituted tetrazole17,18) nor the anti-Markovnikov isomer were formed in these experiments.
 | |||||
|---|---|---|---|---|---|
| Entry | Silane | X | Oxidant | Co | Yieldb) |
| 1 | Et3SiH | 4 | TMFP-BF4 | Co1 | 0% |
| 2 | Ph2SiH2 | 4 | TMFP-BF4 | Co1 | 0% |
| 3 | PhSiH3 | 4 | TMFP-BF4 | Co1 | 55% |
| 4 | PhSiH3 | 4 | TMFP-OTf | Co1 | 18% |
| 5 | PhSiH3 | 4 | FP-BF4 | Co1 | 6% |
| 6 | PhSiH3 | 4 | TMFP-BF4 | Co2 | 55% |
| 7 | PhSiH3 | 4 | TMFP-BF4 | Co3 | 82% |
| 8 | PhSiH3 | 1 | TMFP-BF4 | Co3 | 84% |
| 9 | PhSiH3 | 0.6 | TMFP-BF4 | Co3 | 64% |
| 10 | PhSiH3 | 0.4 | TMFP-BF4 | Co3 | 44% |
| 11c) | PhSiH3 | 1 | TMFP-BF4 | Co3 | 83% |
a) Reaction conditions: 2a (1 equiv), 1a (2 equiv), Co cat. (20 mol%), silane (X equiv), oxidant (2 equiv), toluene (0.2 M), room temperature (r.t.), 24 h. b) GC yields. c) TMFP-BF4 (1.5 equiv) was used.
With the optimized conditions in hand, we next explored the substrate generality of the developed method (Fig. 1). The reaction employing various nonactivated olefins proceeded with outstanding Markovnikov selectivity. Bromo, benzyloxy, and ester groups were tolerated under the reaction conditions (3b–3d). Substrates carrying acid-labile functional groups, such as methoxymethyloxy (MOMO) and t-butyldimethylsilyl (TBS)-oxy groups, which are not suitable for conventional acid-mediated HA reaction conditions,12) gave desired products (3e and 3f) in good yields. Allyltrimethylsilane was also a suitable substrate for our method, although the yield of the product 3h was moderate. Internal olefins were also applicable in the HA reaction (3i–3k). In the case of indene, 2a was installed selectively at the benzylic position (3l).
a) All yields are of isolated products. Unless otherwise noted, all reactions were carried out with 2 (0.3 mmol, 1 equiv), 1 (2 equiv), Co3 (20 mol%), PhSiH3 (1.2 equiv), and TMFP-BF4 (1.5 equiv) in toluene (0.2 M) at r.t. for 24 h. b)1 (3 equiv), PhSiH3 (2 equiv), and TMFP-BF4 (2 equiv) were used.
Next, we explored the substrate scope of the tetrazoles beyond 2a. Tetrazoles substituted with aryl groups, such as tolyl and halophenyl groups, gave the desired HA products (3m–3p) in good yields. The developed reaction was also applicable to 5-benzyltetrazole, although the yield of 3q was lower than those of the abovementioned aryl tetrazoles. It is worth noting that the developed reaction could be applied to the N2-alkylation of irbesartan, an angiotensin II receptor antagonist, to prepare 3r.
The developed reaction was also conducted on a gram-scale (Chart 1). Using 1.00 g (6.84 mmol) of 1a, we obtained 1.54 g of 3a in 81% isolated yield, which is almost the same as the 82% obtained on the 0.3 mmol scale (see, Fig. 1).
We then explored the possibility of asymmetric hydroamination by using optically active cobalt complexes (Table 2). In particular, the asymmetric intermolecular hydroamination of nonactivated olefins still remains one of the unsolved challenges in modern synthetic organic chemistry,19) with reported reactions resulting in low enantioselectivity20) or requiring activated olefin substrates, such as styrenes,20–28) allenes,5) dienes,29,30) allylamines,31–33)etc.6–11,34–42) We chose 1a, 1g, and 1l as test substrates for the asymmetric reaction. Although the asymmetric induction on 1a did not improve beyond 20% enantiomeric excess (ee) with any of the tested catalysts, 1g with (S,S)-Co5 and 1l with (S,S)-Co6 gave (−)-3g and (+)-3l with a moderate enantioselectivity 46% ee and 39% ee, respectively (entries 4 and 5). Although lower reaction temperature did not improve the enantioselectivity in the case of 1a and 1g (entries 6 and 7), hydroamination of 1l at 0°C gave the best selectivity, 66% ee (entry 8)
 | ||||
|---|---|---|---|---|
| Entry | Co | Productb) | ||
| (−)-3a: Yield/ee | (−)-3g: Yield/ee | (+)-3l: Yield/ee | ||
| 1 | (R,R)-Co1 | 41% / 8% ee | 53%c) / 17% ee | 58% / 3% ee |
| 2 | (R,R)-Co3 | 82% / 12% ee | 60%c) / 19% ee | 60% / −8% ee j) |
| 3 | (S,S)-Co4 | 81% / −2% ee h) | 83%c) / −3% ee i) | 81% / 18% ee |
| 4 | (S,S)-Co5 | 52%f) / 17% ee | 46%c,f) / 46% ee | 31% / 7% ee |
| 5 | (S,S)-Co6 | 26% / 7% ee | 58%c) / −11% ee i) | 47%f) / 39% ee |
| 6d) | (S,S)-Co5 | 23%f) / 14% ee | 76%c,f) / 42% ee | —g) |
| 7e) | (S,S)-Co5 | 3%f) / 33% ee | 22%c,f) / 33% ee | —g) |
| 8d) | (S,S)-Co6 | —g) | —g) | 24%f) / 66% ee |
| 9e) | (S,S)-Co6 | —g) | —g) | N.D.k) |
a) Reaction conditions: 2a (1 equiv), 1 (2 equiv), Co cat. (20 mol%), silane (1.2 equiv), oxidant (1.5 equiv), toluene (0.2 M), r.t., 24 h. b) Unless otherwise noted, all yields are GC yields. Ee values are determined with HPLC using chiral columns. c) 1g (3 equiv), PhSiH3 (2 equiv), and TMFP-BF4 (2 equiv) were used. d) at 0°C. e) at −20°C. f) Isolated yield. g) not tested under this conditions. h) (+)-3a was obtained. i) (+)-3g was obtained. j) (−)-3l was obtained. k) Desired product was not detected.
In conclusion, we have utilized nonactivated olefins for alkylating 5-substituted tetrazoles via cobalt-catalyzed hydroamination under mild conditions for preparing 2,5-disubstituted tetrazoles exclusively, with the absence of the 1,5-disubstituted isomer or anti-Markovnikov product formation. The reaction could be applied to substrates having acid-labile functional groups, which are not suitable for traditional acid-mediated conditions. Furthermore, we have also developed an asymmetric intermolecular hydroamination reaction using optically active cobalt catalysts.
This work was supported in part by a JSPS Grant-in-Aid for Young Scientists (Grant Number: JP18K14868), the Teijin Pharma Award in Synthetic Organic Chemistry Japan, a Grant-in-Aid from the Tokyo Biochemical Research Foundation, the Keihanshin Consortium for Fostering the Next Generation of Global Leaders in Research (K-CONNEX), established by Building of Consortia for the Development of Human Re-sources in Science and Technology, Ministry of Education, Culture, Sports, Science and Technology (MEXT), and the Platform Project for Supporting Drug Discovery and Life Science Research from AMED (Grant Number: JP19am0101084).
The authors declare no conflict of interest.
The online version of this article contains supplementary materials.
