Skip to main content
NCBI home page
ACS AuthorChoice logoLink to ACS AuthorChoice
. 2022 Dec 16;144(51):23358–23367. doi: 10.1021/jacs.2c09146

Taming Keteniminium Reactivity by Steering Reaction Pathways: Computational Predictions and Experimental Validations

Mark A Maskeri †, Anthony J Fernandes ‡, Giovanni Di Mauro ‡, Nuno Maulide ‡,*, K N Houk †,*
PMCID: PMC9801433  PMID: 36525680

Abstract

graphic file with name ja2c09146_0014.jpg

Keteniminium ions, the nitrogen analogues of ketenes, exhibit high reactivity toward olefins and π-systems. Previous results from the Maulide group demonstrated an unexpected propensity for an alternative intramolecular Belluš–Claisen-type rearrangement rather than an expected intramolecular (2 + 2) cycloaddition. We have conducted a cooperative density functional theory/experimental investigation of this process, seeking insights into the competition between the observed Claisen-type reaction and the historically expected (2 + 2) cyclization. Our calculations revealed a surprisingly small difference in the free energy barrier between these two intramolecular reactions. Further theoretical and experimental investigations probe the electronics of the substrate, rationalize a competing deallylation side reaction, and demonstrate the proof-of-concept for an enantioselective (2 + 2) variant.

Introduction

Ketenes hold a storied place in the organic chemist’s toolbox.1−3 Since Staudinger’s disclosure of the species in 1905 (Scheme 1, top left),4 ketenes have enabled the synthesis of cyclobutanone scaffolds through a prototypical thermal (π2s+π2a) reaction with olefins5−7 and have also served as critical rearrangement intermediates (e.g., Wolff rearrangement,8,9 among numerous others1,2,10). This prodigious reactivity is, however, often detrimental, as the desired ketene transformation may be plagued by side reactivity or dimerization.11−13 The generation of ketenes also often requires the use of sensitive intermediates (e.g., acyl chlorides) or high temperatures, which may pose challenges in complex-target synthesis or when delicate substrates are employed.

Scheme 1. Top: Early Disclosures of the Ketene and Keteniminium Ions; Bottom: Maulide’s Unexpected Belluš–Claisen-Type Rearrangement to 4 and the Expected Bicyclic Cyclobutanone-Tetrahydropyran 6.

Scheme 1

Open in a new tab

A variety of ketene equivalents and analogues have been formulated to temper errant reactivity and enable access to ketene-like behavior from convenient functional groups.14−17 The keteniminium ion, introduced by Ghosez and co-workers, represents a particularly convenient equivalent (Scheme 1, top right).18−23 Through the action of an electrophile (such as phosgene or, more typically, triflic anhydride) and a base, otherwise stalwart carboxamides can be transiently converted to keteniminium ions and subsequently engaged in polar chemistry. In addition to the ease of generation, keteniminium ions are less prone to dimerization and offer a particularly useful handle to manipulate selectivity through additional bonds to nitrogen.24

Seeking to leverage the reactivity and selectivity of keteniminium ions in a total synthesis campaign, the Maulide group in 2010 attempted the synthesis of cyclobutanone-tetrahydropyran bicycle 6 (Scheme 1, bottom).25−27 However, unlike related literature precedents for this reaction, the presence of an ethereal oxygen instead triggered a Belluš–Claisen-type rearrangement, furnishing allyl γ-lactones (4).19,28,29 None of the classic (2 + 2) product 6 was reported, and we sought to understand the factors controlling these two competing intramolecular reactions. In the process of our investigation, we engaged in a computational/experimental collaboration. The synergy between experiment and theory enabled a thorough investigation of the cycloaddition capacity of this reaction and uncovered methods by which the chemoselectivity of the reaction can be dramatically altered.

Methods

Quantum mechanical investigations of the reactions of keteniminium ion 2 and congeners were conducted with density functional theory (DFT) calculations using Gaussian 16.30 Initial geometries were prepared with Grimme’s xTB31/CREST32 and Zimmerman’s GSM33 codes. Geometry optimization was completed at the ωB97X-D/def2-SVP level with the solvation model based on density (SMD) for dichloromethane.34−36 Single-point corrections to energy were made at the ωB97X-D/def2-TZVPP level with the SMD for dichloromethane. Quasiharmonic corrections to enthalpy and entropy were made using Paton’s GoodVibes software.37−39 Temperature corrections were applied at 393.15 K (120 °C) unless otherwise specified. Visualizations were prepared with Legault’s CYLview2040 and Gilbert’s IQmol. Please see the Supporting Information for experimental details.

Results and Discussion

Initial Studies

Our computational investigations began with the mechanism proposed by Maulide et al. for the conversion of amide 1 to allyl lactone 4 (Figure 1).25 We located intermediates and transition structures (TSs) with favorable energetics proceeding from the parent amide to the keteniminium ion 2. The reversible addition of the collidine base to 2 was also established, with the resulting adduct demonstrating substantial stabilization of the keteniminium ion (see Supporting Information, section 5.3).41,42 Unlike the generation of keteniminium ions from ynamides—which is known to be endergonic43—this process is considerably exergonic (ΔG = −32.7 kcal/mol; see Supporting Information section 5.4 for ynamide model study). Keteniminium ion 2 adopts a low-energy conformation where the ethereal oxygen interacts with the π* orbital (n → π*, Figure 1 inset);44 this conformation essentially preorganizes the scaffold to proceed to the key allyloxonium ion species 9 (ΔΔG⧧ = 10.7 kcal/mol; see Supporting Information, section 5.5).43,45,46

Figure 1.

Figure 1

Open in a new tab

Energetics for the generation of allyl lactone 4 and (2 + 2) adduct 6 from amide 1. Inset. NBO orbitals of 2 showing n → π* interaction. NBO orbitals depicting nO (orbital 79, HOMO–7) and π*C=N (orbital 87, LUMO) are drawn at an isovalue of 0.05. Geometries optimized at the ωB97X-D/def2-SVP/SMD(CH2Cl2) level of theory. Single-point corrections and NBO orbital generation were performed at the ωB97X-D/def2-TZVPP/SMD(CH2Cl2) level of theory. Energies are given in kcal/mol. TSs for the conversion of 7 to 8 and of 2 to 9 can be found in section 5.5 of the Supporting Information.

From 9, we located Claisen-type TSs for the allyl rearrangement, the lower-energy chair conformer of which exhibits a reasonable energy barrier of 18.9 kcal/mol (TS-3).47 This keteniminium ion 2 also permitted us to identify a (2 + 2) TS (TS-4), proceeding to the cis-cyclobutane iminium ion 5 (cis-(2 + 2) adduct), that is only 2.4 kcal/mol higher in energy than the Claisen chair (see inset in Figure 1). This free energy difference is consistent with the previous report that only lactone product 4 was observed.48 However, 2.4 kcal/mol is not a particularly large free energy gap, and we hypothesized that small perturbations to the olefin electron density may promote (2 + 2) reactivity over the Claisen-type rearrangement.

Obtaining (2 + 2) Products

We next studied the methallyl congener 10, which exhibited an energetic profile similar to that of the unsubstituted parent substrate 1, until the formation of the keteniminium ion 11 (see Supporting Information, section 5.6). Following on from 11, we were gratified that the minor inductive electron donation of the methyl group was sufficient to remove the relative energy barrier between the chair sigmatropic rearrangement and the cis-(2 + 2) TSs (ΔΔG⧧ = −0.3 kcal/mol, Scheme 2).

Scheme 2. Selected Energetics for Methyl Olefin Substrate 10.

Scheme 2

Open in a new tab

Computed with ωB97X-D/def2-TZVPP/SMD(CH2Cl2)//ωB97X-D/def2-SVP/SMD(CH2Cl2), reported in kcal/mol.

While the initial report of this reaction for methyl olefin substrate 10 did not identify any of the (2 + 2) bicyclic product 13—instead reporting a 61% yield of allyl lactone 12(25)—the corresponding essentially isoenergetic transition structures (TS-5 and TS-6, Scheme 2) suggest that the (2 + 2) reaction should be a competitive pathway. This computational prediction was readily tested experimentally. Gratifyingly, conducting the experiment utilizing modern, mild amide activation procedures49 (Scheme 3A, 2-fluoropyridine at 20 °C; see Supporting Information, section 2.3) allowed us to obtain a 15% yield of the cis-(2 + 2) cycloaddition product 13. This result confirms that it is possible to steer the reaction pathway by modulating the olefin electronic traits. Notably, subjecting the unsubstituted olefin substrate 1 to these conditions provided an 85% NMR yield of the corresponding allyl lactone 4 with no evidence of the corresponding (2 + 2) cycloadduct 6. A significant byproduct exhibiting NMR spectroscopic signals indicative of a trans-(2 + 2) cycloaddition product was also identified. Both 13 and this unexpected byproduct resisted crystallization but were successfully identified by X-ray diffraction spectroscopy (XRDS) of the crystalline derivatives obtained from the addition of a Grignard reagent (Scheme 3B, 13PhCl and 14PhCl). This analysis confirmed the monomeric cis-(2 + 2) identity of 13 and identified the unknown byproduct as a dimeric species with a trans/trans tricyclic core (14).

Scheme 3. (A) Experimental Confirmation of (2 + 2) Adduct Formation and (B) XRDS Structure of 13PhCl and 14PhCl Obtained after Derivatization through Grignard Reagent Addition.

Scheme 3

Open in a new tab

(a) Pyridine base (2.2 equiv), Tf2O (1.1 equiv), 0–20 °C, 4 h, then NaHCO3(aq), 20 °C, (16 h); (b) Plot exhibits NMR yields and recovery (%) using 1,3,5-trimethoxybenzene as the internal standard (IS).

When the activation of 10 was repeated using different pyridine bases (Scheme 3A), several interesting trends emerged. While electron-rich pyridines did not lead to significant conversion of 10, highly electron-deficient pyridines led to the formation of γ-butyrolactone 15 via deallylation—as is the case in absence of a base.50 At the 20 °C temperature utilized for these studies, sterically crowded collidine produced the base-stabilized adduct 16.42 When bases of intermediate basicity and steric hindrance were employed, both (2 + 2) adducts 13 and lactone 12 were formed along with dimerized product 14. The halopyridines—particularly 2-fluoropyridine—appear appropriately tuned to the desired (2 + 2) behavior with regard to basicity and steric encumbrance.

While the observed dimer species 14 exhibited only trans cyclobutanone ring junctions, no trace of a monomeric trans-(2 + 2) adduct was detected. Significant differences manifest on a monomeric basis between the direct cis-(2 + 2) (Figure 2, TS-6) and trans-(2 + 2) (Figure 2, TS-7), whereby the trans-(2 + 2) is found to be higher in energy than its diastereomer by a barrier surpassing 20 kcal/mol. This can be rationalized by analogy to cis- and trans-cycloheptene, the geometries of which comprise the cores of these transition structures. These consequently exhibit a similar difference in energy, as can be identified by the positions of the exo- and endo-cyclic olefin protons (Figure 2, highlight).51 Additionally, the resulting trans adduct is 12.7 kcal/mol less stable than the cis, in line with the very few examples of related trans-fused small ring-containing bicycles previously reported in the literature.29,52−54

Figure 2.

Figure 2

Open in a new tab

Direct transition structures for the generation of monomeric (2 + 2) adducts cis-13 and trans-13. Analogous cycloheptene isomers illustrate the relative strain between the transition structures. Energies vs TS-6, ωB97X-D/def2-TZVPP/SMD(CH2Cl2)//ωB97X-D/def2-SVP/SMD(CH2Cl2) in kcal/mol.

We hypothesized that the dimer system is capable of trans/trans ring fusion as the initial intermolecular interaction does not necessitate ring strain analogous to the monomeric TS-7. This is not the whole picture, however, as we observed that different pyridine bases produce temperature-dependent product distributions (Scheme 4A).

Scheme 4. (A) Temperature and Base Identity Dependence of 13:14 Ratio; (B) Influence of Base Loading on the 13:14 Ratio and Product Distribution.

Scheme 4

Open in a new tab

(a) Pyridine base (2.2 equiv), Tf2O (1.1 equiv), 120 °C (μw), 5 min, or temperature, 4 h, then NaHCO3(aq), 20 °C, 16 h; NMR yields provided on the bars; (b) At 46% conversion.

(c) NMR yields using 1,3,5-trimethoxybenzene as an IS.

Rationalizing Base Effects

Indeed, employing 2-fluoropyridine favored 14, while collidine at elevated temperatures favored the formation of 13 (Scheme 4A). Interestingly, the same 13:14 ratio was observed when the reaction was performed using 2-fluoropyridine at 20 °C instead of 120 °C, with the dimer product being more favored at even lower temperatures (Scheme 4A). These features indicate that the pyridine base is directly involved in one or more determining steps of the mechanism.

However, the methyl olefin transition structures TS-6 and TS-7 do not require explicit involvement of base and there should be no such kinetic dependence. We, therefore, sought to elucidate potential routes by which an equivalent of a base could intercept these structures. One such possibility arose from our study of the intrinsic reaction coordinate (IRC) of TS-6. This IRC illustrates that the two new bonds of the product iminium cyclobutane 18 are formed in an energetically concerted, bonding-stepwise fashion (Figure 3; see Supporting Information, section 5.7). These bond-forming events are connected by a flat region in the IRC—a feature we have referred to in other contexts as an entropic intermediate55—that represents the intervening carbenium ion. We identified stationary points nearby the IRC for a distorted analogue of this monomeric entropic intermediate (17, Figure 3A) and the analogous dimer system (21, Figure 3D) that may be conformationally sampled over the course of the reaction.

Figure 3.

Figure 3

Open in a new tab

Computed mechanism for the formation of the trans-cyclobutane ring junction on the model monomer. (A) Distorted entropic intermediate 17 proceeding from TS-6. (B) Trapping of 17 with 2-fluoropyridine. (C) Ring expansion and expulsion of 2-fluoropyridine to produce 20. (D) Comparison with dimeric species, revealing a stationary point. Triflate counterions omitted from D for clarity. Energies vsTS-6 (2-fluoropyridine complex, see the Supporting Information), ωB97X-D/def2-TZVPP/SMD(CH2Cl2)//ωB97X-D/def2-SVP/SMD(CH2Cl2).

If such an entropic intermediate were sufficiently kinetically stable, we hypothesized that an equivalent of pyridine base may nucleophilically intercept the cationic species. The resulting complex would adopt the required geometry to produce the observed trans-cyclobutane ring junctions. While DFT geometry optimizations of the requisite dimeric TSs and intermediates did not converge, a representative analysis was successfully conducted for the monomeric system (Figure 3B,C; proposed dimer analogue shown in Scheme 5). Starting from distorted entropic intermediate 17, we located structures representative of a plausible pathway for the generation of the trans-(2 + 2) adduct (Figure 3B). Trapping 17 with an equivalent of 2-fluoropyridine produces the pyridinium oxabicyclo[5.1.0]octane species 19 via TS-8.

Scheme 5. Proposed Mechanism for the Formation of Dimer 14 Based on Monomeric trans-(2 + 2) Computational Findings.

Scheme 5

Open in a new tab

Triflate counterions omitted for clarity.

Cyclopropyloxepane 19 presents a conformation analogous to the direct trans-(2 + 2) structure TS-7, with the pyridinium ring adopting a pseudoequatorial orientation. Simultaneous cyclopropyl ring expansion and expulsion of the pyridine base proceed to the trans-cyclobutylidene iminium adduct 20 (Figure 3C).

While the free energy of the expansion TS (TS-9) is marginally higher than that of the cis-(2 + 2) TS, the small difference in energy suggests that both pathways should be competitive. That the monomeric trans-(2 + 2) analogue of 13 is not observed suggests that the homoallylic carbenium ion 17 is not sufficiently kinetically stable as hypothesized and is not intercepted by the base at a competitive rate. As the trans motif is observed in the dimer species, we propose the dimeric analogue of 17 (21) is sufficiently long-lived to interact with the pyridine base and undergo this interception/ring expansion transformation to the observed trans/trans dimer 14 (Scheme 5).

DFT studies conducted with collidine as the base similarly located TS-10, a trapping transition structure analogous to TS-9 (Figure 4). This TS exhibits considerable steric crowding and is 5.0 kcal/mol higher in energy than the corresponding cis-(2 + 2) TS. A similar energy barrier for the dimer system would suggest the trans-(2 + 2) dimer should be disfavored. Indeed, this is consistent with our collidine experiments, as the monomeric cis-(2 + 2) cycloadduct is favored 3.3:1.56 As the formation of 21 should be independent of base, the observed product distribution suggests this initial dimerization is reversible (Scheme 5). An additional consequence of this interception/expansion and reversibility hypothesis is the computational prediction that the monomer/dimer ratio should be dependent on the stoichiometry of the base; a lower concentration of base should yield a higher proportion of cis-(2 + 2) adduct 13 over dimer 14 and vice versa. Indeed, experimentally repeating the 2-fluoropyridine-mediated process with different proportions of base confirms this prediction (Scheme 4B).

Figure 4.

Figure 4

Open in a new tab

Collidine ring expansion of monomer. Computations with ωB97X-D/def2-TZVPP/SMD(CH2Cl2)//ωB97X-D/def2-SVP/SMD(CH2Cl2) in kcal/mol. Energy vs TS-6 (collidine complex; see Supporting Information, section 5.12 for coordinates).

These base concentration studies also found that substoichiometric base loadings resulted in the appearance of the deallylated γ-lactone side product 15, a side product that is also observed with the use of highly electron-deficient—and therefore less basic57—pyridines (Scheme 3A).

This had been previously observed by the Maulide group, whereby omitting the base entirely demonstrated utility for the synthesis of γ-lactones from amides.50 We sought a mechanistic understanding of this side reactivity (Scheme 6) and identified a competing pathway proceeding through the doubly cationic—though surprisingly low-energy—cyclic species 28. This structure, which readily deallylates to produce an unsubstituted γ-lactone (15, ΔΔG⧧ = +1.2 kcal/mol) after hydrolysis, is accessed by a cyclization TS-11 that is expected to be highly competitive with the base-mediated isomerization TS-14, key to the above Claisen/(2 + 2) processes (ΔΔG⧧ = +1.3 kcal/mol). Kinetically, we would expect this deallylation process to predominate when the relative concentration of the base is low, which corroborates our experimental findings (see Scheme 4).

Scheme 6. DFT Investigation of the Deallylation Side Reaction (Blue Box).

Scheme 6

Open in a new tab

Free energies versus27 computed with ωB97X-D/def2-TZVPP/SMD(CH2Cl2)//ωB97X-D/def2-SVP/SMD(CH2Cl2) and are reported in kcal/mol. col = collidine.

Steering toward (2 + 2) Products: the Reactant Olefin as a Handle

Having successfully altered the chemoselectivity of this keteniminium ion cascade, we hypothesized that additional increases in the olefin electron density may further favor the formation of the (2 + 2) adduct. We sought to test this using substituted aryl olefin derivatives (30a–f, Figure 5A). DFT calculations conducted on derivatives 30a and 30c suggest the barrier to the cis-(2 + 2) TS is even lower than for methyl derivative 10 (ΔΔG⧧ = −0.9 kcal/mol vs Claisen chair). To our delight, the aryl olefin substrates delivered high yields of the (2 + 2) adduct—in particular, the 4-methoxyphenyl derivative 30a, which produced a 65% NMR yield of the monomeric cis adduct 31a—and were in close agreement with the computationally predicted product ratio (see Supporting Information, section 5.8). The structure of the cis adduct of these aryl derivatives could be unequivocally assessed by XRDS of 31f, for which enantiomerically pure monocrystals were obtained58 (Figure 5C). Additionally, we indeed observed a correlation between olefin electron density and (2 + 2) adduct yield, corroborating our hypothesis that electron-rich olefins promote (2 + 2) activity and simultaneously suppress allyl-lactone production (Figure 5A).

Figure 5.

Figure 5

Open in a new tab

(A) Influence of the electronic nature of the arene on the product distribution. (a) 2-fluoropyridine (2.2 equiv), Tf2O (1.1 equiv), 0–20 °C, 4 h, then NaHCO3(aq), 20 °C, (16 h). NMR yields using mesitylene as an IS. (B) Influence of concentration on the product distribution of 4-fluorophenyl substrate 30d. (C) XRDS structures of 31f, 33d, and 33d′.

We sought to minimize dimer formation and performed concentration studies using 30d (Figure 5B). As anticipated, the production of monomeric cis-(2 + 2) adduct 31d was maximized at lower concentrations, while dimer 33d could be favored at higher concentrations. Production of lactone 32d was also inhibited at higher concentrations, consistent with an increasingly competitive trapping of the keteniminium ion by the allyl moiety of another activated substrate such as the aryl olefin congeners of 11 or 29.

In contrast to the methyl olefin derivative 10, the aryl olefin 30d formed the dimer compound 33d, exhibiting both trans/trans- and cis/trans-cyclobutanone configurations (33d and 33d′, Figure 5B). These diastereomers were obtained as a 1.5:1 mixture in favor of the trans/trans derivative and could be isolated pure by preparative high-performance liquid chromatography and fully characterized by NMR spectroscopy as well as XRDS (Figure 5C). Analysis of the IRC of the monomeric cis-(2 + 2)-forming TS for analogues of 30 indicates this species proceeds to a stable carbocation, analogous to the entropic intermediate of the methyl olefin substrate 17 and methyl olefin dimer 21 (see Supporting Information, section 5.9). It is reasonable to assume that the benzylic stabilization of this carbocation is the reason for the formation of the two observed diastereomers, as the stability imparted permits the molecule to sample conformers capable of proceeding to both the trans/trans and trans/cis configurations.

After reprogramming this process toward (2 + 2) adduct formation, we attempted to formulate an enantioselective variant by installing stereogenic elements on the keteniminium ion. Pleasingly, reactions conducted with α-methoxymethylpyrrolidine substrate 34 demonstrated asymmetric induction with a small but detectable enantiomeric ratio (e.r.) of 58:42 (Scheme 7), a selectivity in-line with previous keteniminium ion processes.24,25,59 Our DFT analysis of the four diastereomeric TSs finds that, while one pair of diastereomers—producing a single enantiomer on hydrolysis of the resulting iminium ion—is indeed favored, there is considerable conformational flexibility to the scaffold that renders the energy gap minute (predicted: 71:29 e.r. at 0 °C; see Supporting Information, section 5.10). While minor, this result does provide an interesting proof-of-concept for further experimental and computational development. Other chiral amine scaffolds, such as MacMillan’s imidazolidinones,60−63 were found to exhibit larger energy gaps between diastereomeric transition structures as determined by DFT (see Supporting Information, section 5.11). Unfortunately, substrate 35 did not produce observable (2 + 2) or Claisen products, presumably either due to side reactivity between the keteniminium ion and imidazolidinone pendent aryl ring or competition for Tf2O between both amide groups.64

Scheme 7. Enantioselective Variant Using Chiral Auxiliaries: α-Methoxymethylpyrrolidine 34 or MacMillan Imidazolidinone 35.

Scheme 7

Open in a new tab

(a) 2-Fluoropyridine (2.2 equiv), Tf2O (1.1 equiv), 0–20 °C, 4 h, then NaHCO3(aq), 20 °C, 16 h; (b) Isolated yields; (c) NMR yields using mesitylene as an IS.

Conclusions

We have conducted a thorough computational analysis of Maulide’s 2010 Belluš–Claisen-type rearrangement of keteniminium ion scaffolds, identifying a significant opportunity to alter the observed chemoselectivity of the reaction toward an intramolecular (2 + 2) reaction. Computational predictions of substrates with greater electron density on the olefin moiety indicated that critical cyclization TSs should become highly competitive with the Claisen-type TSs, a prediction that was experimentally confirmed. A remarkable unexpected product forming in these reactions was identified as a dimeric species resulting from twofold (2 + 2) cycloadditions. We provide potential mechanisms—with experimental backing—for the production of monomeric cis-(2 + 2) adduct and (2 + 2) dimers, as well as DFT rationalization of a competing deallylation side reaction. We demonstrated experimentally the response of the (2 + 2) cyclization to olefin electronics as well as a proof-of-concept enantioselective (2 + 2) variant. These data demonstrate the mutability of useful intramolecular processes, illustrate the impact of small changes to scaffolds, and highlight the beneficial synergy of computation and experimentation in modern synthetic organic chemistry. Our study elucidates subtle differences between reaction pathways and demonstrates the control over chemoselectivity necessary for future applications of keteniminium ion chemistry to total synthesis and the further advancement of this versatile functional group.

Acknowledgments

Computations for this work were performed on the Hoffman2 cluster available through the UCLA Institute for Digital Research and Education (IDRE). We are grateful to Dr. T. Grüne (UVienna) for XRDS measurements and the University of Vienna for generous support of our research programs. We are grateful to Dr. Saad Shaaban and Dr. Daniel Kaiser for their helpful advice and discussions.

Glossary

Abbreviations

DFT

density functional theory

SMD

solvation model based on density

TS

transition structure

d.r.

diastereomeric ratio

DTBMP

2,6-(di-tert-butyl)-4-methylpyridine

IRC

intrinsic reaction coordinate

IS

internal standard

e.r.

enantiomeric ratio

XRDS

X-ray diffraction spectroscopy

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/jacs.2c09146.

  • Computational and experimental methods and procedures, energies, Cartesian coordinates of optimized structures, and experimental characterization data and spectra (PDF)

  • XYZ files for optimized structures (ZIP)

Author Contributions

@ M.A.M. and A.J.F. contributed equally. The manuscript was written through the contributions of all authors.

Open Access is funded by the Austrian Science Fund (FWF). We acknowledge support from the Austrian Research Fund (FWF, P32607 and MolTag DK, W1232), the European Research Council (ERC CoG VINCAT 682002), and the National Science Foundation (CHE 1764328 to KNH).

The authors declare no competing financial interest.

Supplementary Material

ja2c09146_si_001.pdf (10.5MB, pdf)
ja2c09146_si_002.zip (57.5KB, zip)

References

  1. Tidwell T. T. The first century of ketenes (1905-2005): the birth of a versatile family of reactive intermediates. Angew. Chem., Int. Ed. 2005, 44, 5778–5785. 10.1002/anie.200500098. [DOI] [PubMed] [Google Scholar]
  2. Tidwell T. T. Ketene Chemistry after 100 Years: Ready for a New Century. Eur. J. Org. Chem. 2006, 2006, 563–576. 10.1002/ejoc.200500452. [DOI] [Google Scholar]
  3. Allen A. D.; Tidwell T. T. New Directions in Ketene Chemistry: The Land of Opportunity. Eur. J. Org. Chem. 2011, 2012, 1081–1096. 10.1002/ejoc.201101230. [DOI] [Google Scholar]
  4. Staudinger H. Ketene, eine neue Körperklasse. Ber. Dtsch. Chem. Ges. 1905, 38, 1735–1739. 10.1002/cber.19050380283. [DOI] [Google Scholar]
  5. Woodward R. B.; Hoffmann R. The Conservation of Orbital Symmetry. Angew. Chem., Int. Ed. 1969, 8, 781–853. 10.1002/anie.196907811. [DOI] [Google Scholar]
  6. Montaigne R.; Ghosez L. Stereospezifischecis-Addition von Dichlorketen ancis- undtrans-Cycloocten. Angew. Chem. 1968, 80, 194. 10.1002/ange.19680800511. [DOI] [Google Scholar]
  7. Huisgen R. Cycloaddition mechanism and the solvent dependence of rate. Pure Appl. Chem. 1980, 52, 2283–2302. 10.1351/pac198052102283. [DOI] [Google Scholar]
  8. Wolff L. Ueber Diazoanhydride. Justus Liebigs Ann. Chem. 1902, 325, 129–195. 10.1002/jlac.19023250202. [DOI] [Google Scholar]
  9. Kirmse W. 100 Years of the Wolff Rearrangement. Eur. J. Org. Chem. 2002, 2002, 2193–2256. . [DOI] [Google Scholar]
  10. Machiguchi T.; Hasegawa T.; Ishiwata A.; Terashima S.; Yamabe S.; Minato T. Ketene Recognizes 1,3-Dienes in Their s-Cis Forms through [4 + 2] (Diels–Alder) and [2 + 2] (Staudinger) Reactions. An Innovation of Ketene Chemistry. J. Am. Chem. Soc. 1999, 121, 4771–4786. 10.1021/ja990072u. [DOI] [Google Scholar]
  11. Chick F.; Wilsmore N. T. M. LXXXIX.-Acetylketen: a polymeride of keten. J. Chem. Soc., Trans. 1908, 93, 946–950. 10.1039/ct9089300946. [DOI] [Google Scholar]
  12. Staudinger H.; Bereza S. Über Ketene. 14. Mitteilung. äthylketen-carbonsäureester. Ber. Dtsch. Chem. Ges. 1909, 42, 4908–4918. 10.1002/cber.190904204113. [DOI] [Google Scholar]
  13. Rice F. O.; Roberts R. The Structure of Diketene1. J. Am. Chem. Soc. 1943, 65, 1677–1681. 10.1021/ja01249a006. [DOI] [Google Scholar]
  14. Bartlett P. D.; Tate B. E. Dehydronorcamphor from α-Acetoxyacrylonitrile. J. Am. Chem. Soc. 1956, 78, 2473–2474. 10.1021/ja01592a035. [DOI] [Google Scholar]
  15. Corey E. J.; Weinshenker N. M.; Schaaf T. K.; Huber W. Stereo-controlled synthesis of dl-prostaglandins F2.alpha. and E2. J. Am. Chem. Soc. 1969, 91, 5675–5677. 10.1021/ja01048a062. [DOI] [PubMed] [Google Scholar]
  16. Ranganathan S.; Ranganathan D.; Mehrotra A. K. Ketene Equivalents*. Synthesis 1977, 1977, 289–296. 10.1055/s-1977-24362. [DOI] [Google Scholar]
  17. Aggarwal V. K.; Ali A.; Coogan M. P. The development and use of ketene equivalents in [4+2] cycloadditions for organic synthesis. Tetrahedron 1999, 55, 293–312. 10.1016/s0040-4020(98)00924-7. [DOI] [Google Scholar]
  18. Ghosez L.; Haveaux B.; Viehe H. G. Alkyl and Arylα-Chloro Enamines. Angew. Chem., Int. Ed. 1969, 8, 454–455. 10.1002/anie.196904541. [DOI] [Google Scholar]
  19. Marchand-Brynaert J.; Ghosez L. Cycloadditions of keteneimmonium cations to olefins and dienes. New synthesis of four-membered rings. J. Am. Chem. Soc. 1972, 94, 2870–2872. 10.1021/ja00763a062. [DOI] [Google Scholar]
  20. De Poortere M.; Marchand-Brynaert J.; Ghosez L. A General Method for the Synthesis of 2-Azetidinones and 2-Azetidinylideneammonium Salts. Angew. Chem., Int. Ed. 1974, 13, 267–268. 10.1002/anie.197402672. [DOI] [Google Scholar]
  21. Madelaine C.; Valerio V.; Maulide N. Revisiting keteniminium salts: more than the nitrogen analogs of ketenes. Chem.–Asian J. 2011, 6, 2224–2239. 10.1002/asia.201100108. [DOI] [PubMed] [Google Scholar]
  22. Evano G.; Lecomte M.; Thilmany P.; Theunissen C. Keteniminium Ions: Unique and Versatile Reactive Intermediates for Chemical Synthesis. Synthesis 2017, 49, 3183–3214. 10.1055/s-0036-1588452. [DOI] [Google Scholar]
  23. Ramirez M.; Li W.; Lam Y. H.; Ghosez L.; Houk K. N. Mechanisms and Conformational Control of (4 + 2) and (2 + 2) Cycloadditions of Dienes to Keteniminium Cations. J. Org. Chem. 2020, 85, 2597–2606. 10.1021/acs.joc.9b03340. [DOI] [PubMed] [Google Scholar]
  24. Chen L.-y.; Ghosez L. Study of chiral auxiliaries for the intramolecular [2+2] cycloaddition of a keteniminium salt to an olefinic double bond. A new asymmetric synthesis of cyclobutanones. Tetrahedron Lett. 1990, 31, 4467–4470. 10.1016/s0040-4039(00)97650-0. [DOI] [Google Scholar]
  25. Madelaine C.; Valerio V.; Maulide N. Unexpected electrophilic rearrangements of amides: a stereoselective entry to challenging substituted lactones. Angew. Chem., Int. Ed. 2010, 49, 1583–1586. 10.1002/anie.200906416. [DOI] [PubMed] [Google Scholar]
  26. Valerio V.; Madelaine C.; Maulide N. Steering reaction pathways: from benzyl Claisen rearrangements to powerful ionic shifts. Chem.--Eur. J. 2011, 17, 4742–4745. 10.1002/chem.201003591. [DOI] [PubMed] [Google Scholar]
  27. For a recent example highlighting the importance of related transformations in total synthesis, please see:Liu R.; Xia M.; Ling C.; Fu S.; Liu B. Construction of the Tetracyclic Core Structure of Dysiherbols A-C. Org. Lett. 2022, 24, 1642–1646. 10.1021/acs.orglett.2c00159. [DOI] [PubMed] [Google Scholar]
  28. Gonda J. The Belluš-Claisen Rearrangement. Angew. Chem., Int. Ed. 2004, 43, 3516–3524. 10.1002/anie.200301718. [DOI] [PubMed] [Google Scholar]
  29. Marko I.; Ronsmans B.; Hesbain-Frisque A. M.; Dumas S.; Ghosez L.; Ernst B.; Greuter H. Intramolecular [2+2] cycloadditions of ketenes and keteniminium salts to olefins. J. Am. Chem. Soc. 1985, 107, 2192–2194. 10.1021/ja00293a073. [DOI] [Google Scholar]
  30. Frisch M. J.et al. Gaussian 16; Gaussian, Inc.: Wallingford, CT, 2016 [Google Scholar]; , see Supporting Information section 5.2 for full citation.
  31. Bannwarth C.; Ehlert S.; Grimme S. GFN2-xTB-An Accurate and Broadly Parametrized Self-Consistent Tight-Binding Quantum Chemical Method with Multipole Electrostatics and Density-Dependent Dispersion Contributions. J. Chem. Theory Comput. 2019, 15, 1652–1671. 10.1021/acs.jctc.8b01176. [DOI] [PubMed] [Google Scholar]
  32. Pracht P.; Bohle F.; Grimme S. Automated exploration of the low-energy chemical space with fast quantum chemical methods. Phys. Chem. Chem. Phys. 2020, 22, 7169–7192. 10.1039/c9cp06869d. [DOI] [PubMed] [Google Scholar]
  33. Zimmerman P. M. Growing string method with interpolation and optimization in internal coordinates: method and examples. J. Chem. Phys. 2013, 138, 184102. 10.1063/1.4804162. [DOI] [PubMed] [Google Scholar]
  34. Chai J. D.; Head-Gordon M. Long-range corrected hybrid density functionals with damped atom-atom dispersion corrections. Phys. Chem. Chem. Phys. 2008, 10, 6615–6620. 10.1039/b810189b. [DOI] [PubMed] [Google Scholar]
  35. Weigend F.; Ahlrichs R. Balanced basis sets of split valence, triple zeta valence and quadruple zeta valence quality for H to Rn: Design and assessment of accuracy. Phys. Chem. Chem. Phys. 2005, 7, 3297–3305. 10.1039/b508541a. [DOI] [PubMed] [Google Scholar]
  36. Marenich A. V.; Cramer C. J.; Truhlar D. G. Universal solvation model based on solute electron density and on a continuum model of the solvent defined by the bulk dielectric constant and atomic surface tensions. J. Phys. Chem. B 2009, 113, 6378–6396. 10.1021/jp810292n. [DOI] [PubMed] [Google Scholar]
  37. Luchini G.; Alegre-Requena J. V.; Funes-Ardoiz I.; Paton R. S. GoodVibes: automated thermochemistry for heterogeneous computational chemistry data. F1000Research 2020, 9, 291. 10.12688/f1000research.22758.1. [DOI] [Google Scholar]
  38. Li Y.-P.; Gomes J.; Mallikarjun Sharada S.; Bell A. T.; Head-Gordon M. Improved Force-Field Parameters for QM/MM Simulations of the Energies of Adsorption for Molecules in Zeolites and a Free Rotor Correction to the Rigid Rotor Harmonic Oscillator Model for Adsorption Enthalpies. J. Phys. Chem. C 2015, 119, 1840–1850. 10.1021/jp509921r. [DOI] [Google Scholar]
  39. Grimme S. Supramolecular binding thermodynamics by dispersion-corrected density functional theory. Chem.—Eur. J. 2012, 18, 9955–9964. 10.1002/chem.201200497. [DOI] [PubMed] [Google Scholar]
  40. CYLview20 . Legault C. Y.Université de Sherbrooke. 2020, http://www.cylview.org/ (accessed Oct 31, 2022).
  41. Charette A. B.; Grenon M. Spectroscopic studies of the electrophilic activation of amides with triflic anhydride and pyridine. Can. J. Chem. 2001, 79, 1694–1703. 10.1139/v01-150. [DOI] [Google Scholar]
  42. Peng B.; Geerdink D.; Farès C.; Maulide N. Chemoselective Intermolecular α-Arylation of Amides. Angew. Chem., Int. Ed. 2014, 53, 5462–5466. 10.1002/anie.201402229. [DOI] [PubMed] [Google Scholar]
  43. Maryasin B.; Kaldre D.; Galaverna R.; Klose I.; Ruider S.; Drescher M.; Kählig H.; González L.; Eberlin M. N.; Jurberg I. D.; Maulide N. Unusual mechanisms in Claisen rearrangements: an ionic fragmentation leading to a meta-selective rearrangement. Chem. Sci. 2018, 9, 4124–4131. 10.1039/c7sc04736c. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Newberry R. W.; Raines R. T. The n→π* Interaction. Acc. Chem. Res. 2017, 50, 1838–1846. 10.1021/acs.accounts.7b00121. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Potthast A.; Rosenau T.; Sartori J.; Sixta H.; Kosma P. Hydrolytic processes and condensation reactions in the cellulose solvent system N,N-dimethylacetamide/lithium chloride. Part 2: degradation of cellulose. Polymer 2003, 44, 7–17. 10.1016/s0032-3861(02)00751-6. [DOI] [Google Scholar]
  46. This low-energy, “folded” conformation of keteniminium ion 2 is 3.1 kcal/mol lower in energy than the corresponding linear conformation. Coordinates for the linear conformation of 2 are provided in the Supporting Information as 2—linear conformer.
  47. The corresponding boat conformer is essentially isoenergetic to the chair (ΔΔG = +0.1 kcal/mol). See Supporting Information, section 5.12 for coordinates.
  48. For reference, an energy gap of 2.4 kcal/mol at 120 °C would translate to a 96:4 product ratio, thus consistent with the minor component of such a mixture potentially evading detection by NMR.
  49. For the use of 2-halopyridines in amide activation, see:; a Movassaghi M.; Hill M. Synthesis of Substituted Pyridine Derivatives via the Ruthenium-Catalyzed Cycloisomerization of 3-Azadienynes. J. Am. Chem. Soc. 2006, 128, 4592–4593. 10.1021/ja060626a. [DOI] [PubMed] [Google Scholar]; b Movassaghi M.; Hill M. Single-Step Synthesis of Pyrimidine Derivatives. J. Am. Chem. Soc. 2006, 128, 14254–14255. 10.1021/ja066405m. [DOI] [PubMed] [Google Scholar]; c Peng B.; Geerdink D.; Maulide N. Electrophilic Rearrangements of Chiral Amides: A Traceless Asymmetric α-Allylation. J. Am. Chem. Soc. 2013, 135, 14968–14971. 10.1021/ja408856p. [DOI] [PubMed] [Google Scholar]
  50. Valerio V.; Petkova D.; Madelaine C.; Maulide N. Direct room-temperature lactonisation of alcohols and ethers onto amides: an ″amide strategy″ for synthesis. Chem.—Eur. J. 2013, 19, 2606–2610. 10.1002/chem.201203906. [DOI] [PubMed] [Google Scholar]
  51. Squillacote M. E.; DeFellipis J.; Shu Q. How stable is trans-cycloheptene?. J. Am. Chem. Soc. 2005, 127, 15983–15988. 10.1021/ja055388i. [DOI] [PubMed] [Google Scholar]
  52. Lenz G. R. The photocycloaddition of an 5α-androst-1-en-3-one to olefins. J. Chem. Soc., Perkin Trans. 1 1984, 2397–2401. 10.1039/p19840002397. [DOI] [Google Scholar]
  53. Smith A. B.; Wood J. L.; Keenan T. P.; Liverton N.; Visnick M. General Photoisomerization Approach to trans-Benzobicyclo[5.1.0]octenes: Synthetic and Mechanistic Studies. J. Org. Chem. 1994, 59, 6652–6666. 10.1021/jo00101a025. [DOI] [Google Scholar]
  54. García-Expósito E.; Álvarez-Larena A.; Branchadell V.; Ortuño R. M. [2 + 2]-Photocycloaddition of 1,1-Diethoxyethylene to Chiral Polyfunctional 2-Cyclohexenones. Regioselectivity and π-Facial Discrimination. J. Org. Chem. 2004, 69, 1120–1125. 10.1021/jo035532n. [DOI] [PubMed] [Google Scholar]
  55. Yang Z.; Jamieson C. S.; Xue X.-S.; Garcia-Borràs M.; Benton T.; Dong X.; Liu F.; Houk K. N. Mechanisms and Dynamics of Reactions Involving Entropic Intermediates. Trends Chem. 2019, 1, 22–34. 10.1016/j.trechm.2019.01.009. [DOI] [Google Scholar]
  56. The steric congestion of TS-10 suggests that bulkier pyridine bases, such as 2,6-(di-tertbutyl)-4-methylpyridine (DTBMP), may further disfavor the dimeric trans-(2 + 2) pathway. However, when DTBMP was used as a base, no trace of (2 + 2) cycloadduct—cis or trans—could be detected by NMR, presumably due to excessive steric hindrance.
  57. Grandberg I. I.; Faizova G. K.; Kost A. N. Comparative basicities of substituted pyridines and electronegativity series for substituents in the pyridine series. Chem. Heterocycl. Compd. 1967, 2, 421–425. 10.1007/bf00509435. [DOI] [Google Scholar]
  58. Mechanical separation of enantiomeric crystals was possible in this system. See Supporting Information, section 4 for XRDS details.
  59. Peng B.; Huang X.; Xie L. G.; Maulide N. A Brønsted Acid Catalyzed Redox Arylation. Angew. Chem., Int. Ed. 2014, 53, 8718–8721. 10.1002/anie.201310865. [DOI] [PubMed] [Google Scholar]
  60. Ahrendt K. A.; Borths C. J.; MacMillan D. W. C. New Strategies for Organic Catalysis: The First Highly Enantioselective Organocatalytic Diels–Alder Reaction. J. Am. Chem. Soc. 2000, 122, 4243–4244. 10.1021/ja000092s. [DOI] [Google Scholar]
  61. Jen W. S.; Wiener J. J. M.; MacMillan D. W. C. New Strategies for Organic Catalysis: The First Enantioselective Organocatalytic 1,3-Dipolar Cycloaddition. J. Am. Chem. Soc. 2000, 122, 9874–9875. 10.1021/ja005517p. [DOI] [Google Scholar]
  62. Gordillo R.; Carter J.; Houk K. N. Theoretical Explorations of Enantioselective Alkylation Reactions of Pyrroles and Indoles Organocatalyzed by Chiral Imidazolidinones. Adv. Synth. Catal. 2004, 346, 1175–1185. 10.1002/adsc.200404107. [DOI] [Google Scholar]
  63. Gordillo R.; Houk K. N. Origins of Stereoselectivity in Diels–Alder Cycloadditions Catalyzed by Chiral Imidazolidinones. J. Am. Chem. Soc. 2006, 128, 3543–3553. 10.1021/ja0525859. [DOI] [PMC free article] [PubMed] [Google Scholar]
  64. Marchand-Brynaert J.; Ghosez L. Electrophilic aminoalkenylation of aromatics with α-chloroenamines. J. Am. Chem. Soc. 1972, 94, 2869–2870. 10.1021/ja00763a061. [DOI] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

ja2c09146_si_001.pdf (10.5MB, pdf)
ja2c09146_si_002.zip (57.5KB, zip)

Articles from Journal of the American Chemical Society are provided here courtesy of American Chemical Society

RESOURCES