Acid Chlorides as Formal Carbon Dianion Linchpin Reagents in the

Letter pubs.acs.org/OrgLett

Acid Chlorides as Formal Carbon Dianion Linchpin Reagents in the Aluminum Chloride-Mediated Dieckmann Cyclization of Dicarboxylic Acids Ahlam M. Armaly,‡ Sukanta Bar,‡ and Corinna S. Schindler* Department of Chemistry, Willard Henry Dow Laboratory, University of Michigan, 930 North University Avenue, Ann Arbor, Michigan 48109, United States S Supporting Information *

ABSTRACT: The development of acid chlorides as formal dianion linchpin reagents that enable access to cyclic 2-alkyl- and 2-acyl-1,3-alkanediones from dicarboxylic acids is described herein. Mechanistic experiments relying on 13C-labeling studies confirm the role of acid chlorides as carbon dianion linchpin reagents and have led to a revised reaction mechanism for the aluminum(III)-mediated Dieckmann cyclization of dicarboxylic acids with acid chlorides.

T

he construction of ketone moieties via C−C bondforming reactions is a commonly employed strategy in retrosynthetic analysis1 of natural and pharmaceutically relevant complex molecules. The prevalence of these structural units and their utility as functional handles for further elaboration have propelled the development of potent synthetic equivalents. Specifically, strategies have been developed that use three possible synthons as linchpin reagents2,3the carbonyl dianion,4,5 dication,7 and carbene6 (Figure 1A). We recently reported an efficient protocol for the formation of complex cyclic-1,3-diones 6 in a one-step Dieckmann cyclization reaction of dicarboxylic acids in the presence of acid chlorides (Figure 1B).8 Despite these advances, highly functionalized substrates were not tolerated under the reaction conditions. As such, we initiated mechanistic studies to enhance the synthetic utility of this method. Mechanistic investigations presented herein establish acid chlorides 5 as a new class of carbon dianion linchpin reagents and represent a new synthetic strategy toward the synthesis of cyclic 1,3-diones 6 (Figure 1B). During our investigations of the aluminum(III)-mediated Dieckmann cyclization of dicarboxylic acids with acid chlorides, we observed distinct reactivity with β-branched acid chlorides, such as 3,3-dimethylbutyryl chloride 7. These substrates were poorly tolerated under the reaction conditions, resulting primarily in acid chloride self-condensation byproducts.8 While these substrates did not provide the desired Dieckmann product 6, we were able to isolate 2-acyl-1,3-dione 11 as a cross-condensation product in 12% yield (Scheme 1). Similarly, acetyl chloride 9 formed the corresponding 2-acyl-1,3-dione 12, which had previously been observed by Matoba et al.,9 albeit in low yield. However, in the case of n-alkyl acid chlorides, no 2acyl-1,3-dione products were observed. In order to elucidate the dichotomous reactivity of acetyl and β-branched acid chlorides and develop the aluminum(III)-mediated Dieckmann cycliza© 2017 American Chemical Society

Figure 1. (A) Strategies toward complex ketones relying on the synthetic equivalents of the carbonyl dianion, carbene, and dication synthons as linchpin reagents. (B) Acid chlorides as new carbon dianion linchpin reagents.

tion of dicarboxylic acids with acid chlorides into a synthetic method of general utility, we conducted mechanistic investigations into the controlling features of this transformation based on 13C-labeling experiments. Received: May 29, 2017 Published: July 19, 2017 3962

DOI: 10.1021/acs.orglett.7b01623 Org. Lett. 2017, 19, 3962−3965

Letter

Organic Letters Scheme 1. (A) n-Alkyl Acid Chlorides Provide 2-Alkyl-1,3diones 6 and (B) β-Branched Acid Chlorides and Acetyl Chloride Provide 2-Acyl-1,3-diones 11 and 12

In order to gain experimental support for this mechanism, we conducted 13C-labeling experiments (Scheme 2B). When 13Clabeled propionyl chloride 18 was subjected to the reaction conditions with glutaric acid 10, we obtained dione 19 that was devoid of any 13C incorporation. This unexpected result12 is in sharp contrast to the previously reported mechanistic hypothesis that suggests decarboxylation of one of the acid moieties to allow complete inclusion of the acid chloride fragment, which would result in 20 as the expected product. We then evaluated a fully 13C-labeled propionyl chloride 21 with glutaric acid 10 and found that the product 22 again maintained both carbonyls of glutaric acid 10 and only incorporated 13C at the C2 and methyl positions. These results establish acid chlorides as carbon dianion linchpin reagents for the formation of the desired complex ketone products in the aluminum(III)mediated Dieckmann cyclization of dicarboxylic acids. Based on these insights, we propose an alternative mechanistic hypothesis for the formation of 2-alkyl-1,3-diones in the aluminummediated Dieckmann cyclization of dicarboxylic acids (Scheme 3A). Upon acylation of the dicarboxylic acid moiety 10 with 1

Mechanistic Investigations into the Formation of 2-Alkyl-1,3-diones

A mechanistic hypothesis for the Lewis acid-mediated Dieckmann cyclization of dicarboxylic acids was first proposed by Schick et al.10 and further elaborated by Ivanov et al.11 This previous mechanistic hypothesis suggests the complete incorporation of the acid chlorides 5 into the corresponding 1,3-dione products 17 (Scheme 2A). The reaction is proposed

Scheme 3. Revised Mechanistic Hypothesis for the AlCl3· MeNO2-mediated Dieckmann Cyclization Reaction

Scheme 2. (A) Previous Mechanistic Hypothesis by Schick et al. and (B) Labeling Studies with Glutaric Acid 10 and 13CLabeled Propionyl Chloride 18 and 21a

equiv of acid chloride 5, the corresponding mixed anhydride 24 is formed. Lewis acid-mediated condensation of 24 results in the formation of cyclic anhydride 26 upon release of 25. The activated enol 25 subsequently cleaves anhydride 26,13 which results in the formation of keto acid 27. Decarboxylation of 27 and subsequent activation with another equivalent of acid chloride provides activated enol 28, which undergoes Dieckmann cyclization to yield 17. In experiments aimed at obtaining additional support for this revised mechanism, we were able to isolate intermediate cyclic anhydrides such as 26 that formed in the course of this transformation. Moreover, when glutaric anhydride 26 was subjected to the reaction conditions with propionyl chloride 29, the desired Dieckmann cyclization product was obtained, albeit in lower yields of 41% (Scheme 3B).14,15 Based on this result, we hypothesize that the formation of enol 25 is essential for the reaction to proceed. Importantly, this revised reaction mechanism accounts for both carbonyl subunits of the dicarboxylic acid being transferred to the corresponding cyclization products and confirms the role of acid chlorides as formal carbon dianion linchpin reagents. While our revised mechanistic hypothesis explains the requirement of superstoichiometric amounts of this linchpin

a

Condition: AlCl3 (3.0 equiv) and dicarboxylic acid (1.0 equiv) in MeNO2 (0.25 M) were stirred at rt for 1 h. Acid chloride (3.0 equiv) was added, and the reaction was heated at 80 °C for 3 h.

to proceed via initial acylation of the dicarboxylic acid 10 with acid chloride 5 to form β-keto acid 13. Subsequent decarboxylation of 13 leads to keto acid 14, which can undergo Lewis acid-mediated Dieckmann cyclization via intermediate enol 15 to give 17. 3963

DOI: 10.1021/acs.orglett.7b01623 Org. Lett. 2017, 19, 3962−3965

Letter

Organic Letters reagent, only 1 equiv is ultimately incorporated in the final product. Specifically, we propose that the first equivalent of acid chloride is incorporated into intermediate keto acid 32, while excess equivalents would simply serve as acylating agents to cyclized 32 to the final product 33 (Scheme 4A). In order to

Scheme 5. (A) Labeling Studies with 13C-Labeled Acetyl Chlorides 36 and 39 and (B) Revised Mechanistic Hypothesis

Scheme 4. (A) Evaluation of the Formation of Intermediate Ketone 32 Bearing a Linchpin Fragment and (B) Reaction of cis-Cyclohexanedicarboxylic Acid 34 with Propionyl Chloride as the Linchpin Reagent and Acetyl Chloride as the Sacrificial Acylating Agent

probe this hypothesis, we evaluated the ability to form intermediate 32 and induce the subsequent cyclization with a sacrificial acylating agent that is not incorporated into the final product. Specifically, when 1 equiv of propionyl chloride 29 was used in the Dieckmann cyclization of cis-1,3-cyclohexanedicarboxylic acid 34 followed by 2 equiv of acetyl chloride as a sacrificial acylating agent, the corresponding 1,3-dione 35 was obtained in 58% yield (eq 1, Scheme 4B). In comparison, when the reaction was conducted under identical conditions but in the absence of acetyl chloride, only 23% yield of 35 was obtained (eq 2). These studies provide additional support for our revised mechanistic hypothesis for the aluminum(III)mediated Dieckmann cyclization of dicarboxylic acids and justify the requirements of superstoichiometric amounts of acid chloride.

from a subsequent acylation of the resultant unsubstituted cyclic 1,3-diones 43 (Scheme 5B). In the case of β-substituted acid chlorides, the 2-alkyl-1,3-diones 45 that form are not stable under the reaction conditions and undergo Lewis acid-mediated cleavage to provide a stabilized carbocation fragment and unsubstituted dione 43. Dione 43 is subsequently acylated to the observed 2-acyl-1,3-dione product 44. Importantly, the products that arise from n-alkyl acid chlorides do not undergo this fragmentation pathway due to lack of carbocation stabilization. This is evidenced by the fact that we did not observe any 2-acyl-1,3-dione byproducts for the case of n-alkyl acid chlorides but rather the competing acid chloride selfcondensation byproducts. Altogether, these results corroborate the role of acid chlorides as a formal carbon dianion linchpin reagent for simple acid chlorides such as acetyl chloride as well as complex β-substituted acid chlorides. Lastly, we extended this one-step Dieckmann cyclization method for dicarboxylic acids to the synthesis of the related cyclic 2-acetyl-1,3alkanediones relying on our previously reported reaction protocol.8 Similar to their alkyl counterparts, the structurally related cyclic 2-acyl-1,3-alkanediones represent common motifs in many complex molecules of biological importance.17,18 Currently available literature procedures for the construction of these building blocks require harsh reaction conditions relying on reducing metals19 or the preformation of the 1,3-diketone moiety.20 Previously, Matoba et al. reported the formation of 46 and 54 from their corresponding dicarboxylic acids albeit in 10% and 15% yield, respectively.9 During the course of our studies, we found the attenuated reactivity of the AlCl3·MeNO2 complex proved superior for enabling access to a variety of 2acetyl-1,3-diones in up to 73% yield (46−54, Scheme 6).

Mechanistic Investigations into the Formation of 2-Acyl-1,3-diones

We hypothesized that acetyl chloride and β-substituted acid chlorides proceeded via a similar mechanism as n-alkyl acid chlorides (depicted in Scheme 3) to provide the 2-alkyl-1,3dione Dieckmann products. However, we suspected that the 2acyl-1,3-diones would have to arise from acylation of the unsubstituted cyclic-1,3-diones 43 (Scheme 5B).16 To reconcile these theories and obtain experimental support for this mechanistic hypothesis, we conducted 13C-labeling experiments with acetyl chloride substrates 36 and 39 (Scheme 5A). The results obtained were found to be consistent with our previous observations using 13C-labeled propionyl chloride (Scheme 2B); the 2-acyl-1,3-diones 37 and 40 maintained both carbonyl units of glutaric acid 10. Importantly, no 13C incorporation was observed when 13C-labeled MeNO2 was used as solvent. In the case of acetyl chloride, the 2-acetyl products 44 therefore arise 3964

DOI: 10.1021/acs.orglett.7b01623 Org. Lett. 2017, 19, 3962−3965

Letter

Organic Letters

(2) (a) Adachi, S.; Liew, S. K.; Lee, C. F.; Lough, A.; He, Z.; Denis, J. D.; Poda, G.; Yudin, A. K. Org. Lett. 2015, 17, 5594−5597. (b) Rolfe, A.; Lushington, G. H.; Hanson, P. R. Org. Biomol. Chem. 2010, 8, 2198−2203. (c) Amatore, M.; Beeson, T. D.; Brown, S. P.; MacMillan, D. W. C. Angew. Chem., Int. Ed. 2009, 48, 5121−5124. (d) Shi, S.; Szostak, M. Chem. - Eur. J. 2016, 22, 10420−10424. (3) For reviews, see: (a) Cornil, J.; Guerinot, A.; Cossy, J. Org. Biomol. Chem. 2015, 13, 4129−4142. (b) Denmark, S. E.; Liu, J.H.-C. Angew. Chem., Int. Ed. 2010, 49, 2978−2986. (4) For reviews, see: (a) Kirschning, A.; Kujat, C.; Schaumann, E. Eur. J. Org. Chem. 2007, 15, 2387. (b) Rentner, J.; Kljajic, M.; Offner, L.; Breinbauer, R. Tetrahedron 2014, 70, 8983−9027. (5) (a) Smith, A. B.; Boldi, A. M. J. Am. Chem. Soc. 1997, 119, 6925− 6929. (b) Smith, A. B.; Pitram, S. M.; Boldi, A. M.; Gaunt, M. J.; Sfouggatakis, C.; Moser, W. H. J. Am. Chem. Soc. 2003, 125, 14435− 14445. (c) Tietze, L. F.; Geissler, H.; Gewert, J. A.; Jakobi, U. Synlett 1994, 1994, 511−512. (6) Wu, X.-F.; Neumann, H.; Beller, M. Chem. Soc. Rev. 2011, 40, 4986. (7) Heller, S. T.; Newton, J. N.; Fu, T.; Sarpong, R. Angew. Chem., Int. Ed. 2015, 54, 9839. (8) Armaly, A. M.; Bar, S.; Schindler, C. S. Org. Lett. 2017, DOI: 10.1021/acs.orglett.7b01622. (9) Matoba, K.; Tachi, M.; Itooka, T.; Yamazaki, T. Chem. Pharm. Bull. 1986, 34, 2007. (10) Schick, H.; Lehmann, G. J. Prakt. Chem. 1968, 38, 391−396. (11) Dziomko, V. M.; Ivanov, O. V. Zh. Org. Khim. 1967, 3, 712− 717. (12) The expected products based on an initial mechanistic hypothesis by Schick and Ivanov (see refs 10 and 11). (13) Enol acetates have been reported to provide 2-alkyl-1,3-diones in the presence of anhydrides and AlCl3, see: Schick, H.; Eichhorn, I. Synthesis 1989, 1989, 477. (14) Anhydrides have previously been reported as substrates toward cyclic-2-alkyl-1,3-diones, albeit, resulting in significantly lower yields ranging from 17 to 25%. For representative examples, see: (a) Ivanov, O. V.; Tikhonova, G. I.; Dziomko, V. M. Metody Poluch. Khim. Reaktiv. Prep. 1969, 20, 16−18. (b) Dziomko, V. M.; Ivanov, O. V. Org. Khim. 1967, 3, 712−717. (15) For an example of a base-mediated conversion of anhydrides and acid chlorides to cyclic-2-alkyl-1,3-diones that proceeds through intermediate ketenes, see: Diter, P.; Magnier, E.; Blazejeqski, J.-C. J. Fluorine Chem. 2007, 128, 1235−1240. (16) Acylation of cyclic-1,3-diones gives rise to enol esters that isomerize to the 2-acyl-1,3-diones in the presence of AlCl3. See: Akhrem, A. A.; Lakhvich, F. A.; Budai, S. I.; Khlebnicova, T. S.; Petrusevich, I. I. Synthesis 1978, 1978, 925−927. (17) (a) Potapovich, M. V.; Eremin, A. N.; Rubinov, D. B.; Metelitza, D. I. Appl. Biochem. Microbiol. 2008, 44, 19−27. (b) Chabbert, T. A.; Scavizzi, M. R. Antimicrob. Agents Chemother. 1976, 9, 36−41. (c) Chu, D. T. W.; Bernstein, E.; Huckin, S. N. Can. J. Chem. 1978, 56, 1059− 1062. (d) Safak, B.; Ciftci, I. H.; Ozedemir, M.; Kiyildi, N.; Cetinkaya, Z.; Aktepe, O. C.; Altindis, M. Phytother. Res. 2009, 23, 955−957. (18) For a review, see: Rubinov, D. B.; Rubinova, I. L.; Akhrem, A. A. Chem. Rev. 1999, 99, 1047−1066. (19) (a) Lim, S.; Min, Y.; Choi, B.; Kim, D.; Lee, S. S.; Lee, I. M. Tetrahedron Lett. 2001, 42, 7645−7649. (b) Skarzewski, J. Tetrahedron 1989, 45, 4593−4598. (c) Jung, J. C.; Watkins, E. B.; Avery, M. A. Synth. Commun. 2002, 32, 3767−3777. (20) (a) Tabuchi, H.; Hamamoto, T.; Ichihara, A. Synlett 1993, 1993, 651−652. (b) Isobe, T. J. Org. Chem. 1999, 64, 6984−6988. (c) Shen, Q.; Huang, W.; Wang, J.; Zhou, X. Org. Lett. 2007, 9, 4491−4494.

Scheme 6. Cyclic 2-Acetyl-1,3-diones Accessed via the Dieckmann Dianion Linchpin Strategy

Based on mechanistic investigations relying on 13C-labeling experiments, we propose a revised reaction mechanism for the aluminum(III)-mediated Dieckmann cyclization of dicarboxylic acids with acid chlorides that establishes acid chlorides as carbon dianion linchpin reagents. Experiments based on sacrificial acylating reagents provide additional support for this revised reaction mechanism which incorporates only one equivalent of acid chloride into the corresponding products. Additionally, precious acid chlorides can be converted to the corresponding 2-alkyl-1,3-dione products with improved yields relying on acetyl chloride as a sacrificial acylating reagent which ultimately provides new opportunities to further improve this valuable synthetic procedure (vide inf ra).

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.orglett.7b01623. Experimental data as well as 1H and 13C NMR spectra for all new compounds prepared in the course of these studies (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. ORCID

Corinna S. Schindler: 0000-0003-4968-8013 Author Contributions ‡

A.M.A. and S.B. contributed equally.

Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported by the Petroleum Research Fund (PRF#54688-DNI1) and the NSF/National Science Foundation (CHE-1654223). A.M.A. thanks the National Science Foundation for Graduate Research Fellowship (Grant No. 1256260).

■

REFERENCES

(1) Hoffman, R. W. Elements of Synthetic Planning; Springer: Berlin, 2009. 3965

DOI: 10.1021/acs.orglett.7b01623 Org. Lett. 2017, 19, 3962−3965