Literature DB >> 30542535

Highly diastereoselective synthesis of tricyclic fused-pyrans by sequential hydride shift mediated double C(sp3)-H bond functionalization.

Keiji Mori1,2, Nobuaki Umehara1, Takahiko Akiyama1.   

Abstract

Described herein is the Brønsted acid-catalyzed double C(sp3)-H bond functionalization of alkyl phenethyl ether derivatives. In this process, a [1,5]-[1,5]-hydride shift occurred successively to afford tricyclic fused pyran derivatives in excellent chemical yields with excellent diastereoselectivities (up to >20 : 1). The key to achieve this reaction was the introduction of two methyl groups at the benzylic position, which was effective in both hydride shift processes: (1) the Thorpe-Ingold effect for the first hydride shift and (2) conformational control in the second hydride shift.

Entities:  

Year:  2018        PMID: 30542535      PMCID: PMC6237131          DOI: 10.1039/c8sc02103a

Source DB:  PubMed          Journal:  Chem Sci        ISSN: 2041-6520            Impact factor:   9.825


Introduction

Development of an efficient synthetic method is a major topic of research interest in modern synthetic organic chemistry. Direct transformation of inert C–H bonds, in particular, has attracted much attention because of its high potential for the development of new trends in the synthesis of complex organic molecules.1 Hydride shift triggered C(sp3)–H bond functionalization has recently emerged as a powerful tool for the rapid construction of various useful organic molecules.2Fig. 1 illustrates the details of this strategy. The key feature of this transformation is the [1,5]-hydride shift of the C(sp3)–H bond α to a heteroatom. Subsequent 6-endo cyclization to a cationic species A affords fused heterocycle 2.3–6
Fig. 1

C(sp3)–H functionalization by the internal redox process.

As part of our ongoing effort to develop a new synthetic transformation method involving hydride shift triggered C(sp3)–H bond functionalization,7 we have recently successfully applied this strategy to a sequential hydride shift system (double C(sp3)–H bond functionalization).8,9 The key to achieving this reaction was the use of the reactive α,β-unsaturated trifluoromethyl ketone as the electrophilic moiety. This setting effectively promoted an unprecedented [1,4]-[1,5]-hydride shift sequence, affording linear tricyclic piperidines in excellent chemical yields and with excellent diastereoselectivities (eqn (1), Fig. 2).8 The application of this sequential system to an oxygen analogue is also highly important because it would be a promising tool for rapid construction of multisubstituted tricyclic pyrans, which are found in some biologically active compounds. The two challenges to achieve this reaction are as follows: (1) the lower electron-donating ability of an oxygen atom than that of a nitrogen atom, which is disadvantageous for the hydride shift process, and (2) the propensity to eliminate the alkoxy group.10
Fig. 2

Double C(sp3)–H bond functionalization in alkyl phenethyl ether derivatives.

We wish to report herein a solution to the issues related to the realization of the double C(sp3)–H bond functionalization in alkyl phenethyl ether derivatives (eqn (2), Fig. 2). The key to achieving this functionalization was to introduce two methyl groups at the benzylic position, thereby promoting the desired [1,5]-[1,5]-hydride shift/cyclization sequence smoothly to furnish the corresponding tricyclic pyrans in excellent chemical yields with excellent diastereoselectivities (up to >20 : 1 d.r.).

Results and discussion

An initial trial was conducted with cinnamylidene barbiturate 3 bearing a 2-benzyloxyethyl moiety at the ortho position (Scheme 1). Treatment of 3 with 10 mol% Yb(OTf)3, which exhibited excellent catalytic performance in the sequential internal redox reactions we previously developed,8,9 in refluxing ClCH2CH2Cl resulted in the recovery of the starting material 3. Gd(OTf)3 was also ineffective and the reaction did not proceed at all.
Scheme 1

Initial trial with simple cinnamylidene barbiturate 3.

We assumed that these disappointing results were ascribed to the low electron-donating ability of the oxygen atom compared with the nitrogen atom (vide supra). To improve the hydride shift ability, a substrate 5 with a PMP group (electron-donating group)7 adjacent to the oxygen atom was employed (Scheme 2). Although the starting material 5 was mostly consumed, the resulting product was not the desired adduct 6 but the tetracyclic compound 7, which was produced by BnOH elimination, E/Z isomerization, followed by the inverse electron demand Diels–Alder reaction (IEDDA).
Scheme 2

Reaction of 5 with a PMP group adjacent to the oxygen atom.

The above results suggest that there are two obstacles to realize the sequential hydride shift reaction: (1) the low reactivity of the substrate and (2) the ease of eliminating the alkoxy group. We envisioned that 8 with two alkyl groups at the benzylic position would be a suitable substrate for overcoming these issues. This structural modification would be effective in both hydride shift processes (Fig. 3): (1) the Thorpe–Ingold effect for the first hydride shift and (2) the conformational control in the second hydride shift. The cooperative effect of these two factors would lead to a dramatic improvement of the reactivity. Another advantage of using this substrate is the suppression of elimination of the alkoxy group by the absence of benzylic hydrogens.
Fig. 3

Design of substrate 8 with a malonate moiety.

As expected, the planned reaction proceeded smoothly with 8a. When 8a was treated with 10 mol% Yb(OTf)3 in refluxing ClCH2CH2Cl, the sequential [1,5]-hydride shift process proceeded to afford the tricyclic pyran derivative 9a in excellent chemical yield (90%). Only two diastereomers were observed in this reaction even though 9a has three stereogenic centers, and one diastereomer was slightly majored (d.r. = 3.3 : 1). Other metal triflates such as Sc(OTf)3, Gd(OTf)3, and Dy(OTf)3 also promoted the reaction albeit with lower diastereoselectivities (d.r. = <4.5 : 1, entries 2–4). The use of Mg(OTf)2 or TiCl4 resulted in the complete recovery of the starting material (entries 5 and 6). FeCl3 promoted the reaction effectively to furnish 9a in 87% yield with 5.6 : 1 diastereoselectivity (entry 7). TfOH exhibited high reactivity as well as satisfactory diastereoselectivity (86%, d.r. = 13.0 : 1, entry 9), although the reactivity of Tf2NH, also a strong Brønsted acid catalyst, was moderate (72%, d.r. = 5.0 : 1, entry 8). 2 mol% TfOH sufficed to complete the reaction with moderate diastereoselectivity (89%, d.r. = 6.7 : 1, entry 10). Excellent diastereoselectivity was realized even with a 2 mol% catalyst when the reaction was conducted in toluene (99%, d.r. = >20 : 1, entry 11). The structure of the major diastereomer was determined as described for 9a by X-ray crystallographic analysis, and those of others shown in Fig. 4 were surmised by analogy.11,12
Fig. 4

Substrate scope of diastereoselective synthesis of tricyclic pyrans.

With the highly diastereoselective tricyclic pyran synthesis in our hand, the substrate scope of this reaction was examined (Fig. 4).13 Various substituents such as electron-donating groups (Me and OMe) and an electron-withdrawing group (F) were tolerated in this reaction, and the corresponding tricyclic pyrans 9b–f were obtained in excellent chemical yields with excellent diastereoselectivities (>94%, d.r. = >12.5 : 1). Although the diastereoselectivity was sufficient (d.r. = 11.2 : 1), the chemical yield was modest (30%) in the case of the naphthyl substrate 9g owing to the detachment of the benzyl group.14 Both allyl ether 8h and ethyl ether 8i were also suitable substrates, and excellent diastereoselectivities were achieved for both substrates (d.r. = >20 : 1). The construction of three contiguous stereogenic centers was also attainable, that is, the reaction of the substrate 8j with a monomethyl group at the benzylic position afforded the adduct 9j in 93% with 3.1 : 1 diastereoselectivity.11 To elucidate the reaction mechanism, additional experiments were conducted (Scheme 3). No primary kinetic isotope effect was observed in both hydride shift processes: The KIE (kH/D) values were determined to be 1.63 (for the first hydride shift) and 0.96 (for the second hydride shift). These results imply the reversibility of both hydride shift processes. To check their possible reversibility, the substrates 14 and 15 having the D atom in vinylic positions were subjected to the optimized reaction conditions. No incorporation of the D atom into the position adjacent to the oxygen atom was observed, which completely ruled out the reversibility of the hydride shift steps. Furthermore, the diastereomer ratio did not change even when the adduct 9a with low diastereoselectivity (d.r. = 6.7 : 1, entry 10 in Table 1) was subjected to slightly modified optimized reaction conditions (2 mol% TfOH, toluene, reflux, 48 h). Importantly, the relative stereochemistries of the two diastereomers of 9a were determined to be cis and trans in the ring junction [JH–H = 6.8 Hz in cis-9a and JH–H = 11.6 Hz in trans-9a (major diastereomer)], which indicates that the diastereoselectivity was determined in the first hydride shift/cyclization process. The above results suggest that the excellent diastereoselectivities were achieved under kinetic control, and the rate-determining step would be the first cyclization step, and not the hydride shift processes as in the case of the sequential hydride shift system we have recently reported.8
Scheme 3

Additional experiments for clarification of the reaction mechanism.

Table 1

Screening of reaction conditions


EntryCatalystSolventYield b [%]d.r. c
1Yb(OTf)3ClCH2CH2Cl903.3 : 1
2Sc(OTf)3ClCH2CH2Cl884.2 : 1
3Gd(OTf)3ClCH2CH2Cl22 (68)2.5 : 1
4Dy(OTf)3ClCH2CH2Cl604.5 : 1
5Mg(OTf)2ClCH2CH2Cl0 (98)
6TiCl4ClCH2CH2Cl0 (90)
7FeCl3ClCH2CH2Cl875.6 : 1
8Tf2NHClCH2CH2Cl725.0 : 1
9TfOHClCH2CH2Cl8613.0 : 1
10 d TfOHClCH2CH2Cl896.7 : 1
11 d , e , f TfOHToluene99>20 : 1

Unless otherwise noted, all reactions were conducted with 0.10 mmol of 8a in the presence of an acid catalyst (10 mol%) in ClCH2CH2Cl (1.0 mL) at refluxing temperature.

Isolated yield.

Determined by the comparison of integral values of the C3 (methine) proton of each diastereomer.

MS5Å was added.

2 mol% catalyst.

The reaction was completed within 1 h.

Unless otherwise noted, all reactions were conducted with 0.10 mmol of 8a in the presence of an acid catalyst (10 mol%) in ClCH2CH2Cl (1.0 mL) at refluxing temperature. Isolated yield. Determined by the comparison of integral values of the C3 (methine) proton of each diastereomer. MS5Å was added. 2 mol% catalyst. The reaction was completed within 1 h.

Conclusions

In summary, we have developed a highly diastereoselective synthesis of fused tricyclic pyrans via double C(sp3)–H bond functionalization triggered by sequential hydride shift processes. The key to achieving this goal was the careful design of the substrate with two methyl groups at the benzylic position, and the desired sequential reaction proceeded smoothly with low catalyst loading of TfOH (only 2 mol%). Various tricyclic pyrans with electron-donating and -withdrawing substituents on the aromatic ring were obtained in excellent chemical yields with excellent diastereoselectivities (up to d.r. = >20 : 1). Additional experiments suggested that the rate-determining step was not the hydride shift step but the first cyclization step. Further investigation on the hydride shift/cyclization sequence, particularly the catalytic asymmetric version, is underway in our laboratory.

Conflicts of interest

There are no conflicts to declare. Click here for additional data file. Click here for additional data file.
  67 in total

1.  Recent Advances in C-H Functionalization.

Authors:  Huw M L Davies; Daniel Morton
Journal:  J Org Chem       Date:  2016-01-15       Impact factor: 4.354

2.  Metal-free oxidation/C(sp3)-H functionalization of unactivated alkynes using pyridine-N-oxide as the external oxidant.

Authors:  Dian-Feng Chen; Zhi-Yong Han; Yu-Ping He; Jie Yu; Liu-Zhu Gong
Journal:  Angew Chem Int Ed Engl       Date:  2012-10-26       Impact factor: 15.336

3.  Construction of Oxadiazepines via Lewis Acid-Catalyzed Tandem 1,5-Hydride Shift/Cyclization.

Authors:  Yong-Zhi Chang; Ming-Lei Li; Wen-Feng Zhao; Xiaoan Wen; Hongbin Sun; Qing-Long Xu
Journal:  J Org Chem       Date:  2015-10-02       Impact factor: 4.354

4.  Asymmetric tandem 1,5-hydride shift/ring closure for the synthesis of chiral spirooxindole tetrahydroquinolines.

Authors:  Weidi Cao; Xiaohua Liu; Jing Guo; Lili Lin; Xiaoming Feng
Journal:  Chemistry       Date:  2014-11-24       Impact factor: 5.236

5.  C-H bond functionalization in complex organic synthesis.

Authors:  Kamil Godula; Dalibor Sames
Journal:  Science       Date:  2006-04-07       Impact factor: 47.728

6.  Brønsted acid-catalyzed decarboxylative redox amination: formation of N-alkylindoles from azomethine ylides by isomerization.

Authors:  Hui Mao; Sichang Wang; Peng Yu; Huiqing Lv; Runsheng Xu; Yuanjiang Pan
Journal:  J Org Chem       Date:  2011-01-18       Impact factor: 4.354

7.  C-H bond functionalization via hydride transfer: direct coupling of unactivated alkynes and sp(3) C-H bonds catalyzed by platinum tetraiodide.

Authors:  Paul A Vadola; Dalibor Sames
Journal:  J Am Chem Soc       Date:  2009-11-18       Impact factor: 15.419

8.  Nontraditional reactions of azomethine ylides: decarboxylative three-component couplings of alpha-amino acids.

Authors:  Chen Zhang; Daniel Seidel
Journal:  J Am Chem Soc       Date:  2010-02-17       Impact factor: 15.419

9.  Chiral Magnesium Bisphosphate-Catalyzed Asymmetric Double C(sp3)-H Bond Functionalization Based on Sequential Hydride Shift/Cyclization Process.

Authors:  Keiji Mori; Ryo Isogai; Yuto Kamei; Masahiro Yamanaka; Takahiko Akiyama
Journal:  J Am Chem Soc       Date:  2018-05-11       Impact factor: 15.419

10.  Organocatalytic C-H activation reactions.

Authors:  Subhas Chandra Pan
Journal:  Beilstein J Org Chem       Date:  2012-08-27       Impact factor: 2.883

View more

北京卡尤迪生物科技股份有限公司 © 2022-2023.