Abstract
Herein, we report the palladium-catalyzed decarboxylative asymmetric allylic alkylation of α-enaminones. In addition to serving as valuable synthetic building blocks, we exploit the α-enaminone scaffold and its derivatives as probes to highlight structural and electronic factors that govern enantioselectivity in this asymmetric alkylation reaction. Utilizing the (S)-t-BuPHOX ligand in a variety of nonpolar solvents, the alkylated products are obtained in up to 99% yield and 99% enantiomeric excess.
Graphical Abstract

The development of the palladium-catalyzed decarboxylative asymmetric allylic alkylation reaction has been of longstanding interest in our group.1 With applications in fields ranging from total synthesis2 to pharmaceutical development,3 the allylic alkylation reaction has proven to be an enabling method for the asymmetric construction of all-carbon quaternary centers.1,4 Our initial disclosures showcased the chemistry in the context of carbocyclic ketone substrates to be utilized in this reaction (Figure 1).5 Employing the (S)-t-BuPHOX (L1) ligand, enantioselectivities of the cyclic ketone products are generally limited to a maximum of 88–92% ee. In conjunction with the development of the electron-poor (S)-(CF3)3-t-BuPHOX (L2) ligand, subsequent expansion of the substrate scope to include N-benzoyl lactams was realized. Interestingly, under these modified conditions, the alkylated lactam products are obtained in up to 99% ee.6 This finding represents a substantial improvement in enantioselectivity with respect to the carbocyclic ketone class of substrates.
Figure 1.
(a) Previous report of allylic alkylation in the context of carbocyclic allyl β-ketoesters.5,a (b) Extension of the methodology to N-benzoyl lactams.6 (c) Hypothesis-driven substrate design presented herein.
Curious as to the origins of the enhanced reactivity and enantioselectivity of the N-benzoyl lactam substrate class, we sought to explore the underlying phenomenon through a hypothesis-driven approach to substrate design. Our approach is two-fold: (1) expand the scope of the reaction to include additional synthetically valuable substrates and (2) gain insight into the factors that dictate enantioselectivity and reactivity in the palladium-catalyzed asymmetric allylic alkylation reaction.
In comparing the previously reported carbocyclic and N-benzoyl lactam substrate classes, we note two potentially distinguishing features. First, assuming the intermediacy of a palladium-bound enolate,7 we hypothesize that the N-functionalized lactams afford a more electron-rich enolate than the corresponding carbocyclic ketone analogues. The mechanism through which electrophilic alkylation occurs in canonical palladium-catalyzed Tsuji−Trost reactions is considered to be strongly influenced by the electronics of the nucleophile.8 Experiments suggest that alkylation of stabilized “soft” nucleophiles proceed through an outer-sphere attack of the electrophilic allyl species,9 whereas basic “hard” nucleophiles may undergo allylic alkylation via an inner-sphere process.10 Thus, modulating the electronics of the enolate may in turn affect the propensity for the C−C bond-forming event to proceed through an inner-sphere or outer-sphere mechanism. Indeed, we believe that our prototypical asymmetric allylic alkylation proceeds via an inner-sphere-type mechanism.11
To probe the hypothesis that an increasingly basic palladium enolate is driving the enhanced enantioselectivity of the lactam substrates, we previously investigated β-enaminones (i.e., vinylogous amides) as electron-rich surrogates for the carbocyclic ketone substrate.12 Interestingly, these vinylogous amides maintain reactivity and selectivity comparable to that of typical ketone substrates, with enantioselectivities ranging from 52 to 90% ee but not greater than 90% ee.
In addition to perturbation of the electronics of the hypothesized palladium enolate, we considered potential chelation of the Lewis basic oxygen atom of the benzoyl lactam to the palladium center in a palladium-bound enolate. In lieu of a clear trend between enolate electronics and enantioselectivity, we sought to further explore this second hypothesis through design of a carbocyclic class of substrates bearing a Lewis basic heteroatom in the α′-position. Simultaneously considering synthetic utility of such a substrate class lead us to α-enaminones (i.e., 1).
To our delight, morpholine-containing allyl β-ketoester 1a affords the desired product 2a in quantitative yield and 95% ee under our previously reported conditions with L2 (entry 9) in THF. Subsequent screening of reaction conditions reveals that enantioselectivity is a largely invariant solvent choice, and similar results are even obtained with the standard (S)-t-BuPHOX (L1) ligand in a variety of nonpolar solvents (Table 1). Given this, we selected ethyl acetate as an inexpensive, readily available, and environmentally benign solvent for use in subsequent development.
Table 1.
Solvent and Ligand Screen for the Enantioselective Decarboxylative Allylic Alkylation of Enaminone 1aa
![]() |
Conditions: β-ketoester (0.05 mmol), Pd2(dmdba)3 (5.0 mol %), ligand (12.5 mol %), solvent (1.5 mL), 40 °C, 12 h.
Determined by analytical chiral SFC. See Supporting Information for details.
We then sought to explore the scope of the transformation across a variety of substitution patterns. Both yield and enantioselectivity are maintained across various α-substituents (Table 2). Interestingly, 2-methyl and 2-chloroallyl fragments also afford the desired product with high enantioselectivity, albeit with reduced yield in the 2-methyl case. It is worth noting that substitution at the 1 or 3 positions of the allyl group is not tolerated under these conditions. As previously mentioned, α-enaminones 2a−l represent the first carbocyclic substrate class to achieve of enantioselectivities of up to 99% ee in this asymmetric allylic alkylation.
Table 2.
Substrate Scope of the Enantioselective Decarboxylative Allylic Alkylation of Enaminonesa
![]() | |||||
|---|---|---|---|---|---|
| entry | product | R1 | R2 | yield (%)b | ee (%)c |
| 1 | 2a | Me | H | 95 (93)d | 99 (98)d |
| 2 | 2b | Et | H | 99 | 98 |
| 3 | 2c | CH2OTBS | H | 96 | 99 |
| 4 | 2d | CH2CH2OTBS | H | 93 | 99 |
| 5 | 2e | CH2CH2CO2Me | H | 98 | 97 |
| 6 | 2f | CH2CH2C(O)Me | H | 90 | 94 |
| 7 | 2g | CH2CH2CN | H | 99 | 94 |
| 8 | 2h | Bn | H | 95 | 96 |
| 9 | 2i | p-(OMe)Bn | H | 99 | 95 |
| 10 | 2j | p-(CF3)Bn | H | 87 | 92 |
| 11 | 2k | Me | Me | 92 | 99 |
| 12 | 2l | Me | Cl | 58 | 99 |
Conditions: β-ketoester (0.47 mmol), Pd2(dmdba)3 (5.0 mol %), (S)-t-BuPHOX (12.5 mol %), EtOAc (14 mL), 40 °C, 9 h.
Isolated yield.
Determined by analytical chiral SFC. See Supporting Information for details.
Conditions: β-ketoester (12.3 mmol), Pd2(dmdba)3 (0.5 mol %), (S)-t-BuPHOX (1.3 mol %), EtOAc (37 mL), 40 °C.
We then sought to probe further structural modifications to the carbocyclic backbone of the enolate fragment (Table 3). Piperidene-containing product 2m was obtained in 99% yield and 99% ee, suggesting the distal oxygen of the morpholine ring is not critical for enantioinduction. Introduction of a β-methyl group to the enaminone affords a more sluggish reaction and reduced yield, whereas enantioselectivity is only slightly diminished (2n).13 We then aimed to explore the role of the potentially coordinating α-amine through cyclohexyl substrate 1o. In accordance with our working hypothesis, product 2o was obtained in high yield (93%) but with reduced enantioselectivity (72% ee). Considering the possibility of α-nitrogen chelation, we aimed to explore the effects of perturbations in the N−Pd−O bite angle by preparing an analogous five-membered cyclic α-enaminone substrate, 1p. Under the optimized reaction conditions, substrate 1p affords the desired product 2p in 94% yield and a reduced 83% ee.
Table 3.
Further Structural Variation of Enaminonea
Conditions: β-ketoester (0.47 mmol), Pd2(dmdba)3 (5.0 mol %), (S)-t-BuPHOX (12.5 mol %), EtOAc (14 mL), 40 °C.
Isolated yield.
Determined by analytical chiral SFC. See Supporting Information for details. Values in red highlight deviations from those of model substrate 2a.
Thus far, the trends in observed product enantioenrichment with respect to substrate modification are qualitatively consistent with the proposition that enhancement in enantioselectivity of the allylic alkylation reaction may arise from secondary chelation of α-heteroatoms to the metal center in the hypothesized palladium enolate. However, we note that these are only observations, and a thorough mechanistic study will be required to delineate the potentially numerous factors that influence enantioselectivity.
We next turned our attention to investigating the synthetic utility of the chiral quaternary-center-containing α-enaminone products for access to potential synthetic building blocks. Treatment of enaminone 2a with hydrochloric acid in a methanol−water mixture revealed the latent 1,2-diketone functionality, providing compound 3 in 72% yield. Unfortunately, the analogous reaction of 2a with aniline and catalytic tosic acid in toluene affords the transaminated product 4 in only modest 35% yield. However, 2a is a competent substrate for Fisher indolization, providing indole 5 in nearly quantitative yield (over two steps). 2-Morpholine-substituted enaminones have also been utilized as precursors to pyrazole-based inhibitors of blood coagulation factor Xa by reaction with p-methoxyphenylchlorohydrazones.14 Accordingly, when treated with methyl (Z)-2-chloro-2-(2-(4-methoxyphenyl)-hydrazineylidene)acetate, enaminone 2a yields the corresponding pyrazole 6 in 46% yield (Scheme 1). Further functionalization at the β-position of enaminone 2a may be achieved through an electrophilic bromination with NBS and subsequent Suzuki−Miyaura cross-coupling with phenyl boronic acid, affording compound 7 in 58% yield over two steps.
Scheme 1.
Enaminone Product Transformations
In conclusion, we present α-enaminones as a new substrate class for the palladium-catalyzed decarboxylative allylic alkylation. Product yields of up to 99% and enantioselectivities of up to 99% ee are achieved. The quaternary-center-containing enaminone products are demonstrated to be competent precursors in a variety of postalkylation synthetic transformations. In addition to serving as useful synthetic building blocks, the α-enaminone substrate class and its derivatives offer new insight into the potential role of α-heteroatom chelation for general enhancement of enantioselectivities for the palladium-catalyzed allylic alkylation reaction. Studies to elucidate the effect of these coordinating groups are currently in progress.
Supplementary Material
ACKNOWLEDGMENTS
We thank the NIH-NIGMS (R01GM080269), and the NSF (predoctoral research fellowship to D.C.D., No. DGE-1144469) for financial support. Research reported in this publication was supported by the NIH-NIGMS under Award Number F32GM116304 (postdoctoral fellowship to J. T. Moore). We further thank Dr. Scott C. Virgil (Caltech) for assistance with chiral-SFC separation and insightful discussion.
Footnotes
Complete contact information is available at: https://pubs.acs.org/10.1021/acs.orglett.0c01441
The authors declare no competing financial interest.
ASSOCIATED CONTENT
Supporting Information
The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.orglett.0c01441.
Experimental procedures and characterization data (PDF)
Contributor Information
Douglas C. Duquette, The Warren and Katharine Schlinger Laboratory for Chemistry and Chemical Engineering, Division of Chemistry and Chemical Engineering, California Institute of Technology, Pasadena, California 91125, United States
Alexander Q. Cusumano, The Warren and Katharine Schlinger Laboratory for Chemistry and Chemical Engineering, Division of Chemistry and Chemical Engineering, California Institute of Technology, Pasadena, California 91125, United States
Louise Lefoulon, The Warren and Katharine Schlinger Laboratory for Chemistry and Chemical Engineering, Division of Chemistry and Chemical Engineering, California Institute of Technology, Pasadena, California 91125, United States.
Jared T. Moore, The Warren and Katharine Schlinger Laboratory for Chemistry and Chemical Engineering, Division of Chemistry and Chemical Engineering, California Institute of Technology, Pasadena, California 91125, United States
Brian M. Stoltz, The Warren and Katharine Schlinger Laboratory for Chemistry and Chemical Engineering, Division of Chemistry and Chemical Engineering, California Institute of Technology, Pasadena, California 91125, United States.
REFERENCES
- (1).Liu Y; Han S-J; Liu W-B; Stoltz BM Acc. Chem. Res 2015, 48 (3), 740–751. [DOI] [PMC free article] [PubMed] [Google Scholar]
- (2).For examples in recently reported syntheses, see: [Google Scholar]; (a) Stoltz BM; Pritchett BP; Kikuchi J; Numajiri Y Heterocycles 2017, 95 (2), 1245. [DOI] [PMC free article] [PubMed] [Google Scholar]; (b) Kim KE; Stoltz BM Org. Lett 2016, 18 (21), 5720–5723. [DOI] [PubMed] [Google Scholar]; (c) Pritchett BP; Donckele EJ; Stoltz BM Angew. Chem., Int. Ed 2017, 56 (41), 12624–12627. [DOI] [PMC free article] [PubMed] [Google Scholar]; (d) Pritchett BP; Stoltz BM Nat. Prod. Rep 2018, 35 (6), 559–574. [DOI] [PMC free article] [PubMed] [Google Scholar]; (e) Loskot SA; Romney DK; Arnold FH; Stoltz BM J. Am. Chem. Soc 2017, 139 (30), 10196–10199. [DOI] [PMC free article] [PubMed] [Google Scholar]; (f) Defieber C; Mohr JT; Grabovyi GA; Stoltz BM Synthesis 2018, 50 (22), 4359–4368. [DOI] [PMC free article] [PubMed] [Google Scholar]; (g) Hartrampf N; Winter N; Pupo G; Stoltz BM; Trauner D J. Am. Chem. Soc 2018, 140 (28), 8675–8680. [DOI] [PMC free article] [PubMed] [Google Scholar]
- (3).Sun AW; Bulterys PL; Bartberger MD; Jorth PA; O’Boyle BM; Virgil SC; Miller JF; Stoltz BM Bioorg. Med. Chem. Lett 2019, 29 (18), 2686–2689. [DOI] [PMC free article] [PubMed] [Google Scholar]
- (4).(a) Trost BM; Jiang C Synthesis 2006, 3, 369–396. [Google Scholar]; (b) James J; Guiry PJ ACS Catal. 2017, 7, 1397–1402. [Google Scholar]; (c) Jackson M; O’Broin CQ; Müller-Bunz H; Guiry PJ Org. Biomol. Chem 2017, 15, 8166–8178. [DOI] [PubMed] [Google Scholar]; (d) Akula R; Guiry PJ Org. Lett 2016, 18, 5472–5475. [DOI] [PubMed] [Google Scholar]; (e) Akula R; Doran R; Guiry PJ Chem. - Eur. J 2016, 22, 9938–9942. [DOI] [PubMed] [Google Scholar]; (f) Tunge JA Isr. J. Chem 2020, 60, 351–359. [Google Scholar]
- (5).(a) Behenna DC; Stoltz BM J. Am. Chem. Soc 2004, 126 (46), 15044–15045. [DOI] [PubMed] [Google Scholar]; (b) Mohr JT; Behenna DC; Harned AM; Stoltz BM Angew. Chem., Int. Ed 2005, 44 (42), 6924–6927. [DOI] [PubMed] [Google Scholar]
- (6).Behenna DC; Liu Y; Yurino T; Kim J; White DE; Virgil SC; Stoltz BM Nat. Chem 2012, 4 (2), 130–133. [DOI] [PMC free article] [PubMed] [Google Scholar]
- (7).(a) Keith JA; Behenna DC; Mohr JT; Ma S; Marinescu SC; Oxgaard J; Stoltz BM; Goddard WA J. Am. Chem. Soc 2007, 129 (39), 11876–11877. [DOI] [PubMed] [Google Scholar]; (b) Keith JA; Behenna DC; Sherden N; Mohr JT; Ma S; Marinescu SC; Nielsen RJ; Oxgaard J; Stoltz BM; Goddard WA J. Am. Chem. Soc 2012, 134 (46), 19050–19060. [DOI] [PMC free article] [PubMed] [Google Scholar]
- (8).(a) Trost BM; Thaisrivongs DA J. Am. Chem. Soc 2008, 130 (43), 14092–14093. [DOI] [PubMed] [Google Scholar]; (b) Zhang K; Peng Q; Hou X-L; Wu Y-D Angew. Chem., Int. Ed 2008, 47 (9), 1741–1744. [DOI] [PubMed] [Google Scholar]; (c) Trost BM; Xu J; Schmidt T J. Am. Chem. Soc 2009, 131 (51), 18343–18357. [DOI] [PMC free article] [PubMed] [Google Scholar]
- (9).Trost BM; Verhoeven TR J. Org. Chem 1976, 41 (19), 3215–3216. [Google Scholar]
- (10).Matsushita H; Negishi E J. Chem. Soc., Chem. Commun 1982, 3, 160–161. [Google Scholar]
- (11).Sherden NH; Behenna DC; Virgil SC; Stoltz BM Angew. Chem., Int. Ed 2009, 48 (37), 6840–6843. [DOI] [PMC free article] [PubMed] [Google Scholar]
- (12).Bennett NB; Duquette DC; Kim J; Liu W-B; Marziale AN; Behenna DC; Virgil SC; Stoltz BM Chem. - Eur. J 2013, 19 (14), 4414–4418. [DOI] [PMC free article] [PubMed] [Google Scholar]
- (13). The recovered starting material (1m) ofthis reaction was isolated in 18% yield and 99% ee. When the reaction was stopped at 12 h, the recovered starting material was obtained in 75% yield and 16% ee. This corresponds to a modest s value of 3.3. These data indicate a small but notable relative preference for one enantiomer of the starting material as a substrate for the stereoablative decarboxylation beginning the catalytic cycle of the reaction, which to our knowledge is only observed in this specific substrate. In other cases, we have studied that there is indiscernible kinetic resolution of the starting allyl esters.
- (14).Pinto DJP; Orwat MJ; Koch S; Rossi KA; Alexander RS; Smallwood A; Wong PC; Rendina AR; Luettgen JM; Knabb RM; He K; Xin B; Wexler RR; Lam PYS J. Med. Chem 2007, 50 (22), 5339–5356. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.














