Download PDF
Synlett 2023; 34(07): 850-854
DOI: 10.1055/a-1921-7296
cluster
Chemical Synthesis and Catalysis in India

Ring-Rearrangement Metathetic Approach to Fused 6/5/6/5/6-Oxacyclic Ring System and Bipentalene Derivatives

Sambasivarao Kotha∗
,
Kunkumita Jena

Funded by the Council of Scientific and Industrial Research, CSIR, New Delhi (02(0272)/16/EMR-II).
› Further Information
 
 PDF Download Permissions and Reprints All articles of this category


Abstract

We have developed a useful synthetic route to a 6/5/6/5/6-oxacyclic ring system and bipentalene derivatives from dimeric 7-oxonorbornene derivatives by using ring-rearrangement metathesis as a key step. This method provides access to fused oxacycles containing eight stereogenic centers in just three steps and to bipentalene derivatives in two steps only.


#

Key words

ring-rearrangement metathesis - oxacycles - pentalenes - norbornenes - Grubbs catalysts

The construction of carbocyclic frameworks has received a great deal of attention from synthetic chemists because of the structural complexity and functional diversity of such motifs. Heterocycles fused to carbocyclic rings make up the core of many marine natural products, some of which have proven to be biologically active and pharmaceutically important substances.[1] Among these, oxacycles of various ring sizes are prevalent.[2] Similarly, other fused carbocyclic frameworks consisting of five-, six-, seven-, or eight-membered rings play an important role in the design and synthesis of natural and nonnatural products. Some examples of naturally occurring compounds containing fused 5/6/5-ring skeletons are orostanal,[3] bufogargarizin A,[4] hydroxyikarugamycin A,[5] chabrolosteroid C,[6] cerrenin D,[7] cinncassiol E,[8] and hexacyclinic acid[9] (Figure [1]). Although syntheses of fused carbocyclic rings generally involve multistep sequences, there have been continual efforts to advance synthetic routes by employing olefin metathesis.[10]

Zoom Image
Figure 1 Natural products containing 5/6/5 fused units

Olefin metathesis was developed in the 1960s[11] and extended to organic synthesis in the 1990s[12] as a result of the availability of well-defined robust metathesis catalysts. Some important categories of metathesis reactions that are extensively used in organic synthesis are cross-metathesis, ring-opening metathesis, ring-opening cross-metathesis, and ring-rearrangement metathesis (RRM). Although each of these reactions has its own significance, RRM is known to be useful for creating molecular complexity from simple substrates in a single step.[13] However, the design of suitable substrates for this purpose is not a trivial task.

We have identified norbornadiene dimers 1 and 2 with C2–C2′ connections as starting materials, owing to their inherent ring strain and structural–stereochemical complexity for the synthesis of a fused 6/5/6/5/6-oxacyclic system by using RRM as a key step (Figure [2]).[14]

Zoom Image
Figure 2 Substrates for the RRM study

The diketo derivatives 1 and 2 can be synthesized by following the procedure developed by Khan and Parasuraman, which involves an inverse-electron-demand Diels–Alder[15] reaction between 2,5-dihydrothiophene 1,1-dioxide (3) and 1,2,3,4-tetrachloro-5,5-dimethoxycyclopenta-1,3-diene (4).[16] Sulfone 3, a substitute for buta-1,3-diene acts as the dienophile, and diene 4,[17] which is equivalent to the corresponding cyclopentadienone, acts as the diene. Diastereoisomers 6 and 7 both have endo–anti–endo stereochemistry. Although the C2–C2′ bond can rotate to form syn-diastereoisomers, compounds 6 and 7 prefer to adopt an anti-configuration at their 2 and 2′ positions, probably to avoid steric hindrance between their C7 and C7′ carbon atoms. The overall stereochemistry brings symmetry to these systems (Scheme [1]).

Zoom Image
Scheme 1 Synthesis of compounds 6 and 7 by a Diels–Alder reaction

With compounds 6 and 7 in hand, our next task was the realization of a dechlorination sequence. Dechlorination of norbornene derivatives can be carried out using various reagents, such as Li/ t BuOH, Na/ t BuOH, Na/liquid NH3/EtOH, or Na/EtOH, depending on the nature of the substrate.[18] Although these reactions are well established, yields in most of these cases are lower than 50%. In view of reports in the literature,[18] we used Na/ t BuOH/THF for dechlorination of compounds 6 and 7 to furnish the products 8 and 9, respectively. The yields were around 54–56% when the reaction was carried out on a one-gram scale but tended to fall considerably on scaling up the reaction. In an attempt to improve the yield, we tested a dechlorination sequence with Na/EtOH, which gave slightly improved yields of 60–65% (Scheme [2]).

Zoom Image
Scheme 2 Dechlorination of compounds 6 and 7

The dimethoxy derivatives 8 and 9 were hydrolyzed with 10% sulfuric acid to give the corresponding diketo derivatives 1 and 2 in yields of 86 and 90%, respectively. Compounds 1 and 2 were then treated with allylmagnesium bromide to give the symmetrical diallyl norbornene derivatives 10 and 11, respectively, as the major products (Scheme [3]).

Zoom Image
Scheme 3 Synthesis of the diallyl derivatives 10 and 11

In both cases, traces of the corresponding nonsymmetrical norbornene derivatives, in which both allyl groups were on the same side of the molecule, were observed as minor products (see, Supplementary Information; S5 and S20). The symmetry of compounds 10 and 11 was confirmed by NMR spectral analyses (13C and DEPT-135), and the structure and stereochemistry of compound 10 were unambiguously established by single-crystal X-ray diffraction (Figure [3]).

Zoom Image
Figure 3 Single-crystal X-ray diffraction analysis of compound 10 (front view). Ellipsoids are shown with 50% probability[21]

The dihydroxy diallyl norbornene derivative 10 was then subjected to RRM, with the aim of synthesizing the fused carbocyclic compound 12b (Scheme [4]). However, when compound 10 was treated with the G-I or G-II catalysts under various reaction conditions (Table [1]), unfortunately, only the bipentalene derivative 12a was obtained; compound 12b was not formed under any of the conditions examined. With the G-I catalyst (5 or 10 mol%) only, the starting material was recovered, even in the presence of additives such as Ti(O i Pr)4, which prevents coordination of the catalyst to the hydroxy groups.[14] [19] When compound 10 was treated with the G-II catalyst in the presence of this additive, compound 12a was formed in a moderate yield. Diastereoisomer 11 showed a similar behavior when subjected to RRM, giving only the bipentalene derivative 13a; the expected polycyclic derivative 13b was not formed under any reaction conditions (Scheme [4]).

Zoom Image
Scheme 4 RRM of compounds 10 and 11

Table 1 Optimization of the Conditions for RRM of Compound 10

Entry

Catalyst (mol%)

Additive

Time

Yielda (%)

1

G-I (5)

–

8 h

NRb

2

G-I (10)

Ti(O i Pr)4

6 h

NR

3

G-II (5)

–

8 h

NR

4

G-II (5)

Ti(O i Pr)4

8 h

40

5

G-II (10)

Ti(O i Pr)4

6 h

65

a Isolated yield.

b NR = no reaction.

Compounds 12a and 13a were further treated with G-I and G-II catalysts in the presence of Ti(O-i-Pr)4 as an additive, but no RCM products were realized. To verify the presence of interference by the free hydroxy group in the RRM, we alkylated the hydroxy groups in compounds 10 and 11 with iodomethane to give the corresponding methoxy derivatives 14 and 15. When these compounds were treated with the G-II catalyst, the corresponding ROM derivatives 16 and 17 were obtained, indicating, that the hydroxy group does not interfere with the RRM process (Scheme [5]). The structure and stereochemistry of compound 16 were confirmed by single-crystal X-ray diffraction analysis (Figure [4]).

Zoom Image
Scheme 5 RRM of compounds 14 and 15 to synthesize the bipentalene derivatives 16 and 17 respectively
Zoom Image
Figure 4 Single-crystal X-ray diffraction data for compound 16 (front view); the ellipsoids are shown with 50% probability[21]

Furthermore, treatment of compounds 16 and 17 with the G-I or G-II catalyst did not result in any ring-closure products. The inaccessibility of the fused carbocyclic compounds 12b and 13b suggests that the plausible pathway for RRM is Pathway a, and not the expected Pathway b, as shown for compound 11 in Figure [5]. RRM of compound 11 can proceed through the ROM intermediate 11a followed by RCM, either by Pathway b (ring closure of the allyl group and 5/5′ vinyl groups to form the five-membered rings and ring closure of the two vinyl groups at C6 and C6′ to form a six-membered ring) or by Pathway a (ring closure of the allyl group and the 6/6′ vinyl groups to form the five-membered rings). The two vinyl groups of compound 13a at the C5 and C5′ positions are too far apart to participate in the formation of an eight-membered ring by RCM, which explains the failure to obtain any RCM products from the bipentalene derivatives 12a, 13a, 16, and 17. At this stage, we do not understand the reason for the preference of Pathway a over Pathway b for RRM of compounds 10 and 11. However, the RCM process of ROM intermediate 11a might involve competing ring closure reactions between two vinyl groups or an allyl and a vinyl group. In this case, the allyl–vinyl ring closure to form a five-membered ring is faster than the vinyl–vinyl ring closure to form a six-membered ring. This explains the formation of 13a rather than 13b by RRM of compound 11.

Zoom Image
Figure 5 Plausible pathway for RRM of compound 11

Later, we turned our attention to the synthesis of the fused oxacyclic compounds 22 and 23 by an RRM sequence. In this regard, we treated the diketo compound 8 with NaBH4 to deliver the dihydroxy compound 18 (Scheme [6]). The structure of compound 18 was confirmed unambiguously by single-crystal X-ray diffraction. O-Allylation of compound 18 was accomplished with allyl bromide and NaH to afford compound 20 in a good yield. Treatment of compound 20 with G-I or G-II catalyst under various reaction conditions and at various concentrations resulted in complex mixtures; unfortunately, compound 22 was not obtained (Scheme [6]).

Zoom Image
Scheme 6 RRM of compound 20 and single-crystal X-ray diffraction data for compound 18 (front view). Ellipsoids are shown at a 50% probability[21]

Following a similar sequence, compound 9 was reduced with NaBH4 to obtain dihydroxy compound 19. The structure of compound 19 had been previously confirmed by X-ray diffraction studies.[14] Dihydroxy derivative 19 was further treated with allyl bromide in the presence of NaH to afford compound 21. To our delight, when compound 21 underwent an RRM sequence with the G-I or G-II catalyst, the 6/5/6/5/6-fused oxacarbocyclic compound 23 was obtained (Scheme [7]).

Zoom Image
Scheme 7 RRM of compound 21 and single-crystal X-ray diffraction data for compound 23 (front view). Ellipsoids are shown at a 50% probability[21]

In the formation of compound 23 from 21, the O-allyl double bond interacts with the vinyl groups at the 5 and 5′ carbon atoms (Pathway b, with reference to Figure [5]) of the norbornene system during the RRM process to form the six-membered oxacyclic rings, and RCM of the vinyl groups at C6 and C6′ forms the six-membered carbocyclic ring. From this observation, we conclude that the vinyl–vinyl ring closure is faster than the allyl–vinyl ring closure. The structure and the stereochemistry of compound 23 are supported by NMR and by single-crystal X-ray diffraction analyses[20] (Scheme [7]).

Overall, in this study, we have investigated the metathetic behavior of bisnorbornene derivatives and have successfully demonstrated a facile synthetic route to the fused 6/5/6/5/6-oxacyclic compound 23 and the bipentalene diols 12a and 13a through RRM as a key step. Single-crystal X-ray diffraction studies confirmed the structure and stereochemistry of compound 23 and a derivative of compound 12a (i.e., compound 16). RRM can be a useful strategy for introducing complexity in a synthetic sequence starting with a readily available simple building block by judicious selection of the required norbornene derivatives.


#

Conflict of Interest

The authors declare no conflict of interest.

Acknowledgment

The authors are grateful to the Department of Science and Technology for providing a DST-INSPIRE award and fellowship and to Indian Institute of Technology Bombay for providing the infrastructures. Also, K.J. thanks Darshan S. Mhatre and Dr. Saima Ansari for their help in solving the X-ray crystal data.

Supporting Information

    Supporting information for this article is available online at https://doi-org.accesdistant.sorbonne-universite.fr/10.1055/a-1921-7296.
  • Supporting Information

Corresponding Author

Sambasivarao Kotha
Department of Chemistry, Indian Institute of Technology Bombay
Mumbai 400076
India   

Publication History

Received: 20 June 2022

Accepted after revision: 09 August 2022

Accepted Manuscript online:
09 August 2022

Article published online:
11 October 2022

© 2022. Thieme. All rights reserved

Georg Thieme Verlag KG
Rüdigerstraße 14, 70469 Stuttgart, Germany


   Permissions and Reprints All articles of this category
Zoom Image
Figure 1 Natural products containing 5/6/5 fused units
Zoom Image
Figure 2 Substrates for the RRM study
Zoom Image
Scheme 1 Synthesis of compounds 6 and 7 by a Diels–Alder reaction
Zoom Image
Scheme 2 Dechlorination of compounds 6 and 7
Zoom Image
Scheme 3 Synthesis of the diallyl derivatives 10 and 11
Zoom Image
Figure 3 Single-crystal X-ray diffraction analysis of compound 10 (front view). Ellipsoids are shown with 50% probability[21]
Zoom Image
Scheme 4 RRM of compounds 10 and 11
Zoom Image
Scheme 5 RRM of compounds 14 and 15 to synthesize the bipentalene derivatives 16 and 17 respectively
Zoom Image
Figure 4 Single-crystal X-ray diffraction data for compound 16 (front view); the ellipsoids are shown with 50% probability[21]
Zoom Image
Figure 5 Plausible pathway for RRM of compound 11
Zoom Image
Scheme 6 RRM of compound 20 and single-crystal X-ray diffraction data for compound 18 (front view). Ellipsoids are shown at a 50% probability[21]
Zoom Image
Scheme 7 RRM of compound 21 and single-crystal X-ray diffraction data for compound 23 (front view). Ellipsoids are shown at a 50% probability[21]