Abstract
Electron-deficient dendralenes, bearing enone substructures and possessing an unfavorable disposition of like charges at the neighboring carbons, undergo nucleophilic 1,4-addition (Michael) or 1,6-addition (anti-Michael). Diverse products are obtained, including those of simple addition as well as cyclic and ortho-fused systems arising via multistep sequences, depending on the structure of the substrate and the nature of the nucleophile. Attack of a hydride at an enone fragment triggers the formation of multisubstituted pyranones and furans; furan formation was also initiated by thiolates. A notable exception is the derivative with a five-membered cyclic enone, which prefers simple additions followed by the reshuffling of the double bonds for both H– and RS– nucleophiles. By contrast, the latter enone is the only one that can react with stabilized C-nucleophiles, yielding bicyclic compounds. Domino cyclizations can also be induced by the enolization of the enone with DBU, giving mostly polysubstituted furans. However, the dendralene with a five-membered cyclic enone and its analogue with a six-membered ring behave differently: The former gives a mixture, while the latter prefers the formation of an isocoumarin derivative, which is driven by aromatization. DFT calculations have shown that the additions of thiolates are mostly governed by the thermodynamic stability of possible products arising from complex equilibrium processes.
Introduction
Multi-bond-forming reactions are among the most efficient tools enabling organic chemists to increase molecular complexity in a single operation. A unique opportunity to develop such processes is offered by conjugated olefins, of which the double bonds react in an interconnected manner. A classic example of such a process is the venerable Diels–Alder reaction. Consequently, conjugated sp2 hydrocarbons possessing multiple reaction sites have an extraordinary potential to undergo multi-bond-forming processes, enabling a rapid construction of complex structures. In this realm, dendralenes stand out as a distinct class of polyenes, , since their cross-conjugated nature allows the Diels–Alder (DA) reaction to proceed as a series of consecutive transformations. Here, reorganization of the π-electrons in a DA process results in the formation of an intermediate diene that can undergo another DA reaction, a process known as diene-transmissive Diels–Alder reaction (DTDA). While DTDA, capable of constructing two or more cycles and numerous chiral centers, is apparently the most attractive transformation, other reactions are also conceivable, but very few have been examined to date. The first example, reported as early as in 1975, was cyclopropanation; a few other examples, such as higher cycloadditions, anionic polymerization, and selective hydroboration, appeared more recently. However, the potential of, e.g., photochemical processes and electrophilic, nucleophilic, radical, and transition-metal-promoted additions remains virtually unexplored. Hence, following up on our recently reported facile synthesis of highly polarized, electron-deficient dendralenes (see Figure for a general structure) and their reactivity in DTDA processes, investigation of nucleophilic additions (AN) to these systems appears to be particularly attractive in view of their undisputable potential to unlock new multistep complex transformations.
1.
Conceivable nucleophilic additions to electron-deficient [3]dendralenes (represented by two conformers allowing cross-conjugation). EWG stands for electron-withdrawing group.
These dendralenes, decorated at their termini with electron-withdrawing groups, possess an unfavorable disposition of like charges at the neighboring carbons (Figure ). Therefore, we expected the reversal of this apparently thermodynamically inconvenient state to become the major driving force of their reactions with nucleophiles. Notably, the arrows in Figure show that both Michael and anti-Michael (or vinylogous Michael) additions can initiate a change in the distribution of charges, with the participation of the remaining double bonds being more than conceivable. In addition, steric factors also need to be considered since a polysubstituted dendralene molecule can hardly be expected to assume a conformation that would position all three CC bonds in the same plane and thus allow an overlap of all of the p orbitals of the π-systems. In this work, we report, for the first time, that electron-deficient dendralenes are capable of structure-dependent nucleophilic additions, giving rise to a variety of structurally diverse products. We also demonstrate that the reactivity can vary from simple additions to multistep domino cyclizations, which afford ortho-fused bicyclic products. We envisaged that these AN sequences would parallel those reported for DTDA reactions and offer as yet undescribed opportunities for the synthesis of relevant natural products.
Results and Discussion
For our investigation, we have selected a set of model compounds, most of which were recently prepared by us (Figure ), namely, enones 1a–1d, connected to the dienoate moiety via the α-carbon; conjugated lactone 1e, also connected by the α-carbon; isomeric enones 1f and 1g, connected by the β-carbon; triester 1h; and sulfone 1i.
2.
Model dendralenes 1a–1i synthesized for this study.
Calculations have demonstrated that in the lowest-energy conformation of selected dendralenes 1a, 1b, and 1i the dienoate moiety is invariably planar, whereas the bulky appendage R is positioned almost perpendicular to that plane. This is illustrated by the structure of 1b in the top row of Figure . The same conformation is demonstrated for 1i by single-crystal X-ray analysis (see the Supporting Information). Therefore, the C–C bond connecting the dienoate moiety with the appendage R can be regarded, in principle, as a chiral axis. However, according to calculations, the rotation barrier here is too low (e.g., 13 kcal mol–1 for 1b) to allow isolation of the atropisomers (Note S1 of the Supporting Information).
3.

Spatial representation of the structures of dendralene 1b (top row), obtained by calculations, and dendralene 1i (bottom row), obtained by X-ray crystallography. In this and the following schemes, the bonds held in the plane as part of the conjugated π-system are portrayed as bold lines. Bonds out of that plane (approximately perpendicular) are highlighted with wedges.
Hydride as a Nucleophile
The first nucleophile that we chose to investigate was a hydride. Initial attempts with sodium borohydride as its source revealed remarkable differences between keto-diester 1a with the cyclopentenone appendage and the rest of the dendralenes. Thus, reduction of 1a with NaBH4 in aqueous methanol proceeded via a conjugate addition of H– at the β′-position of the cyclopentenone moiety, affording alcohol 5 in 71% yield (Scheme ). The presence of water in the reaction mixture proved to be beneficial since in the strictly anhydrous methanol the product was obtained in a lower yield (65%). This reactivity can be rationalized by assuming an initial borohydride attack at the enone moiety (approximately perpendicular to the dienoate segment), generating enolate 2, which can assume multiple conformations, such as 2a, 2a′, and 2a″, by rotation about the single bonds, as shown. Of these, 2a″, which would lead to enolate 3, is apparently too congested, so that the pathway via 2a′ dominates, giving intermediate 4, which undergoes protonation and reduction of the keto group (rather than any other conjugate addition), giving rise to alcohol 5 as the final product. It is pertinent to note that the alternative 6(O)-endo-trig cyclization of enolate 2a (highlighted by a blue double-crossed curved arrow) was not observed.
1. Reduction of Cyclopentenone 1a .
a In the 3D structure, the cyclopentenone ring is approximately perpendicular to the plane of the dienoate moiety.
By contrast, reduction of cyclohexenone derivative 1b followed a different pathway (Scheme ). Here, enolate 2b, generated by conjugate hydride addition at the enone unit, was found to exhibit dichotomous behavior, affording two products, arising via 1,4- and 1,2-addition (at 0 °C within 60 min). Pathway i involves an intramolecular 5(O)-exo-trig 1,4-addition (Michael) to produce, after aromatization of intermediate 6b, furan derivative 7b (41%). The competing pathway ii generated lactone 8b as a result of the 6(O)-exo-trig 1,2-attack at the ester group; the reaction was completed via a predominant 1,6-reduction (anti-Michael) of the external acrylate moiety to produce lactone 9b (27%), as evidenced by isotopic labeling, using NaBD4 (see the Supporting Information for details). Under the same conditions, but over an extended reaction time (5 h), cycloheptenone analogue 1c furnished ortho-fused cyclohepta[b]furan derivative 7c in 50% yield; formation of lactone 8c/9c was not observed in this instance.
2. Competing Pathways in the Hydride Reduction of Enones 1b and 1c .
Acyclic enone 1d followed the pattern set by cyclohexenone and cycloheptenone analogues 1b and 1c, respectively, to some extent (Scheme ). Pathway i, triggered by the 1,4-addition (Michael) of H– to the enone fragment, was found to dominate, giving rise to furan derivative 7d (41%). The competing pathway (ii) involved a direct CO reduction, followed by the formation of lactone 8d that was further reduced in the side chain to give 9d (21%) as the second major product. In analogy with the reduction of 8b, where the mechanism is evidenced by isotopic labeling, the latter reduction of 8d can be assumed to proceed in the same way.
3. Reduction of Acyclic Enone 1d .
Under Luche conditions (NaBH4/CeCl3), cyclohexenone derivative 1b was found to produce a mixture of dendralenic lactone 10 and diol 11 (Scheme ). Lactone 10 (see the Supporting Information for X-ray analysis) apparently arose via reduction of the carbonyl in enone 1b, followed by lactonization, the product of which became the substrate for further CeCl3-assisted reduction to generate the corresponding lactol that underwent another reduction to give diol 11. The latter diol became the sole product obtained in high yield (94%), when an excess of borohydride was used.
4. Reduction of 1b in the Presence of CeCl3 .
Not surprisingly, reduction of β-substituted cyclic enones 1f and 1g (Scheme ), isomeric to 1a and 1b, was found to proceed solely at the carbonyl bond since conjugate addition to the tertiary β-position is precluded by increased level of steric congestion. This reduction thus gave rise to alcohols 12f and 12g (35% and 56%, respectively). Unstable cyclopentenol derivative 12f was isolated as the only product from a rather complex mixture. In these instances, the propensity for conjugate 1,4-reduction (Michael) is obviously diminished by the enhanced steric hindrance at the β′-carbon of the enone moiety and by the weak propensity of the dienoate moiety to undergo reduction (vide infra).
5. Reduction of β-Substituted Enones 1f and 1g .
In conflict with general expectations that the acrylate moiety should readily undergo Michael addition, triester 1h was found to be inert to NaBH4, as was lactone 1e. Finally, attempted reduction of sulfonyl derivative 1i afforded an intractable mixture of products under the same conditions.
Altogether, the results of hydride additions have shown that, like a dienophile in DTDA sequences, hydride attack can initiate multiple bond-forming processes, typically proceeding in a Michael or anti-Michael fashion.
Thiolate Nucleophiles
Thiolates, recognized as strong nucleophiles, were the next reagent types to be explored. In contrast to hydride addition, treatment of cyclopentenone derivative 1a with ethanethiolate (generated by deprotonation of ethanethiol with DBU) initially afforded an unstable major product, which was assigned structure 14 based on NMR spectroscopy (Scheme ). The latter compound apparently arises via a 1,6-addition (anti-Michael), starting with an attack at the sterically less hindered terminus of the dienoate moiety followed by protonation of intermediate enolate 13. Adduct 14 then underwent isomerization to thermodynamically more stable isomer 15 upon purification by chromatography. Interestingly, the enone moiety turned out to be inert, even in the presence of an excess (3 equiv) of the thiol.
6. Anti-Michael Addition of Ethanethiolate to 1a .
On the other hand, as outlined in Scheme and Table , enones 1b and 1c followed the same pathway as that for hydride addition (Scheme ), with a preferential attack at the enone ring. Interestingly, while the combination of EtSH and DBU produced an inseparable mixture of furan 17b and its not fully aromatized counterparts (Table , entry 1), addition of I2 (10 mol %) as a catalyst , steered the reaction to completion, giving furan 17b as the only isolable product in 60% yield (entry 2). Replacing EtSH with 2-mercaptoethanol resulted in an increased yield of final product 18b to 66% (entry 3). Similarly, homologous cycloheptenone derivative 1c gave ortho-fused furans 17c and 18c in 45% and 53% yields, respectively (entries 4 and 5).
7. Addition of Thiolate to 1b and 1c .
1. Addition of Thiolates to Cyclic Dendralenes 1b and 1c .
| entry | starting material | time (h) | Nu (equiv) | product | yield (%) |
|---|---|---|---|---|---|
| 1 | 1b | 1 | EtSH (1.2) | 17b | 45 |
| 2 | 1b | 1 | EtSH (1.5) | 17b | 60 |
| 3 | 1b | 5 | HO(CH2)2SH (1.5) | 18b | 66 |
| 4 | 1c | 5 | EtSH (1.5) | 17c | 45 |
| 5 | 1c | 8 | HO(CH2)2SH (1.5) | 18c | 53 |
No I2 added.
Isolated as part of an inseparable mixture with not fully aromatized furans.
Since thiolate additions unveiled a further dramatic difference between five-membered enone 1a and the rest of the cyclic dendralenes, it was of interest to explore the reactivity of acyclic dendralenes, namely, 1d and 1i. While 1d was found to afford an inseparable mixture of products, sulfone 1i reacted in an anti-Michael fashion (1,6-addition) at the more hindered α-site of the dienoate fragment to give adduct 20 (Scheme ), which is in stark contrast to the reaction of 1a, where the addition of the same nucleophile occurs at the opposite terminus (δ-position) of the dienoate moiety (Scheme ). Whether this difference can be attributed to the stronger electron-withdrawing power of the phenylsulfonyl moiety is questionable since the vinyl sulfone group is bent away from planarity (as demonstrated by X-ray crystallography in Figure ), which would prevent the involvement of its π-system. The reaction outcome can be rationalized by assuming a nucleophilic attack at the planar dienoate segment to generate enolate 19, which would then afford 20 upon protonation. However, this mechanism would not explain the difference in the sites of the attack in 1i and 1a. An alternative scenario can thus be proposed, which would require planarization of the α,β,β′,α′-segment by rotation of the vinyl sulfone moiety together with an out-of-plane rotation of both the ester and phenyl group and also rotation of the acrylate moiety, as in 1i*, thereby minimizing conformational strain. Nucleophilic 1,6-attack at the now planarized segment (highlighted by color), followed by protonation, would then readily produce adduct 20. A detailed description of this scenario is given in the computational section (vide infra).
8. Addition of EtSH to Sulfone 1i .
Carbon Nucleophiles
Rather surprisingly, derivatives 1a and 1b were found to be inert to freshly prepared Me2CuLi and several other organometallic combinations, known to favor conjugate additions. In all of these cases, the starting material was recovered, suggesting that the organometallics could not properly coordinate to the enone moiety due to steric hindrance and served merely as bases, effecting the formation of enolates, which were converted back to the starting enones upon workup. However, interesting results were obtained with malonate-type nucleophiles. This time, only cyclopentenone-derived dendralene 1a was found to react, while its homologues, 1b and 1c, with larger rings and acyclic enone 1d turned out to be inert again. Treatment of 1a with deprotonated dimethyl malonate, as a representative stabilized carbon nucleophile, furnished bicyclic derivative 24x (Scheme and Table ). In light of the previous experiments with EtS– (Scheme ), the latter reaction can be considered to start with the anti-Michael 1,6-addition to generate enolate (21x), followed by a second deprotonation of the malonate moiety and the 6(C)-exo-trig Michael cyclization of resulting intermediate 22x toward the enone moiety. Protonation of arising bicyclic intermediate 23x would then afford 24x as the final product. A reversed scenario, i.e., one commencing with a Michael addition to the cyclopentenone moiety, followed by a 6(C)-endo-trig cyclization to the dienoate segment seems less likely in view of the reaction of 1a with EtS–, which occurs at the dienoate unit to give 15. Optimization of the base (Table , entries 1–4) identified the mixture of KOtBu and NaOMe as the optimum. Hence, while the enone moiety was inert to thiolates, the intramolecular nature of the latter reaction rendered the attack possible (as a result of the second deprotonation of the malonate moiety). A brief extension to other activated nucleophiles, namely, to CH2(CN)2 and CH2(CN)(SO2Ph), was also successful, affording bicyclic derivatives 24y and 24z, respectively (entries 5 and 6). The structure of bicyclic products 24 was corroborated by X-ray crystallography of 24x (see the Supporting Information), while NOESY measurement showed that the signal of the phenyl ring of the PhSO2 moiety in 24z correlates to that of the bridgehead hydrogen, demonstrating their cis relationship.
9. Addition of Stabilized C-Nucleophiles to Enone 1a .
a For R1 and R2, see Table .
2. Reaction of 1a with Stabilized C-Nucleophiles.
| entry | R1 | R2 | Nu (equiv) | base (equiv) | solvent | product | yield (%) |
|---|---|---|---|---|---|---|---|
| 1 | CO2Me | CO2Me | 1 | NaOMe (0.4) | MeOH | 24x | 14 |
| 2 | CO2Me | CO2Me | 3 | NaOMe (1.0) | 4:1 CHCl3/MeOH | 24x | 42 |
| 3 | CO2Me | CO2Me | 3 | NaH (1.0) | CHCl3 | 24x | 25 |
| 4 | CO2Me | CO2Me | 3 | KOtBu (0.5), NaOMe (0.5) | 4:1 DCM/MeOH | 24x | 50 |
| 5 | CN | CN | 3 | KOtBu (0.5), NaOMe (0.5) | 4:1 DCM/MeOH | 24y | 60 |
| 6 | CN | SO2Ph | 3 | NaOMe (1.0) | 4:1 CHCl3/MeOH | 24z | 50 |
Unlike its analogues with larger rings, cyclopentenone 1a was also found to react with nitromethane in the presence of 2 equiv of NaOH (Scheme ). In this case, the only product, isolated in a moderate yield of 42%, was diester 28 with a fully aromatized six-membered ring lacking the nitro group. Here, the reaction can be assumed to commence with a 1,6-addition (anti-Michael) at the δ-site. Resulting primary adduct 25 would then undergo second deprotonation, generating 26, followed by an intramolecular 1,4-addition (Michael) to form bicyclic nitro derivative 27, from which indenone 28 is obtained via a base-catalyzed nitrous acid elimination , with ensuing air oxidation.
10. Cyclization of 1a with CH3NO2 .
Enolization
Finally, since many of these reactions encompass cyclization of enolate intermediates arising by an initial nucleophile attack, we attempted to generate enolates via simple deprotonation at the γ′-position of the enone segment. To this end, enones 1a–1d were treated with DBU at room temperature. As summarized in Scheme , the three cyclic derivatives (1a–1c) were found to differ in behavior from each other, depending on the ring size. Thus, while five-membered cyclic enone 1a afforded a complex mixture of products under these conditions, the initial steps in the case of its six- and seven-membered homologues (1b and 1c, respectively) mirrored those observed for their reduction with NaBH4 and thiolate addition (as in Schemes , , and ). Cyclohexenone derivative 1b preferred lactonization via a 1,2-addition (as in pathway ii of Scheme ), followed by an extensive reshuffling of the double bonds (via protonation/deprotonation) under the basic conditions, furnishing coumarin 31 in a high yield (87%) at room temperature in less than 1 h. Since no furan derivative (vide infra) was detected, it can be assumed that aromatization of the six-membered ring was the decisive driving force of the sequence. The reaction also proceeded when the base was used as a catalyst (0.1 equiv) but more slowly (over a period of 12 h) to give 31 in the same yield (87%) under an Ar atmosphere. In principle, enolization of 1b could also occur at the α″-position, but this scenario is less likely in view of the behavior of the other members of this group, namely, 1c and 1d, where the product structure would be incompatible with that pathway (vide infra).
11. Cyclization of Dendralenic Enones 1a–1d via Enolate Formation.
Seven-membered cyclic enone 1c, not capable of the formation of a benzene ring, afforded 7,8-dihydro-6H-cyclohepta[b]furan derivative 33 (65%), presumably via γ′-enolization, 5(O)-exo-trig Michael cyclization, and subsequent aromatization, generating the furan ring (as in pathway i of Scheme ). Acyclic dendralene 1d behaved in the same way, giving rise to tetrasubstituted furan derivative 35. In all of these instances, γ′-enolization is assumed, since the alternative α″-enolization would not lead to the formation of the products obtained.
Mechanistic Considerations
For the computational investigation of the nucleophilic reactivity of dendralenes, we chose to focus on the thiol addition reaction as it showed the most variable experimental behavior, with attacks at different positions of individual substrates. The hydride addition occurred invariably at the β′-position (Figure ), which renders this reaction mechanistically less revealing.
4.

Sites for potential nucleophilic attack at dendralenes 1a and 1b.
Since the seminal work of Thomas and Kollman, the addition of thiolates to conjugated carbon systems has been studied by theoretical methods often in the context of biological chemistry (e.g., covalent kinase inhibitors) or materials chemistry. The most common approach relied on density functional theory (DFT) methods. Many computational contributions to the field have been summarized by Roseli et al. in a recent review. The choice of DFT functional has been a matter of debate, and it has been claimed that some popular functionals cannot predict stable minima for the gas phase addition of methanethiol to methyl vinyl ketone. , However, it has also been noted that moving from vacuum to implicit solvent often cures these problems.
Since dendralenes can be regarded as cross-conjugated systems, it is of note that addition of a thiol to cross-conjugated enones has often been found to be close to thermoneutral compared to non-cross-conjugated variants.
Nucleophilic attacks on dendralenes themselves, in general (not just thiols), have previously received a limited amount of attention. An important exception is the study of Takagi et al., which was focused on polymerization of phenyl-substituted [3]dendralenes. These were generally most reactive at the central methylene group (C4 in their paper; corresponding to the α-position in compounds studied here). It is noteworthy that the reactivity at this position could not be attributed to electronic effects (LUMO orbital location). The α-position was also not particularly reactive in the dendralenes studied in this work, showing the effect of substitutions with electron-withdrawing carbonyl groups.
Our model dendralenes possess nine sp2 carbons, as shown for cyclo-enones 1a and 1b (Figure ), all of which can, a priori, undergo a nucleophilic attack, except perhaps β- and α′-positions that are sterically hindered. Thus, one can anticipate three types of attacks: 1,2-additions (black arrows), 1,4-additions (red arrows), and 1,6-additions (blue arrows). To investigate the differences in reactivity among the seven likely attack sites, we chose to computationally model the reactions of 1a, 1b, and 1i with thiolate, as this reaction exhibited the most variability. While the addition of thiolate to 1a proceeds at the δ-position (Scheme ), the product observed in the reaction of 1b requires thiolate addition at the β′-position (Scheme ). We did not expect such variation of reactivity to occur upon the extension of the ring by one methylene group. Yet another variation was observed for 1i, which was found to prefer an α-position attack, possibly reflecting the strong electron-withdrawing effect of the sulfone group.
Based on the inherent molecular properties of 1a and 1b (see the discussion of the Fukui function in the Supporting Information), we would not have expected any preferential site for the thiol addition. Therefore, to explain the experimental products, we performed extensive DFT calculations of the complete reaction pathways. Herein, we discuss only the relevant stable species, calculated at the ωB97M-V/def2-TZVPD//B3LYP-D3/def2-TZVP level of theory. We chose the high-level ωB97M-V functional by comparing the predicted reaction energies to the DLPNO-CCSD(T) calculation on selected systems. Despite the inability of the B3LYP functional to correctly identify local enolate minima in the gas phase, , we did not observe such behavior in the present study. For the full reaction scheme and free energies calculated at various levels of theory, see Figure S8.
Our DFT study was carried out for the reaction with MeSH instead of EtSH, as it reduced the computational cost. Calculations suggest that thiolate attacks at all relevant positions (α, δ, and β′) have no significant barriers. The only barriers are entropic, which are compensated by favorable interaction enthalpies, resulting in an effective barrier (see Figure S8) of ∼7–8 kcal mol–1, which is not affecting the reaction outcomes in any meaningful way. Therefore, the preference for the product is dictated by the overall free reaction energies (i.e., thermodynamically controlled). The exclusive isolation of product 15 from the addition of EtSH to 1a (Scheme ) is noteworthy, given the many electrophilic sites of 1a. By comparing the free energies of anionic thiolate addition intermediates 1a INT1, 1a INT2, and 1a INT3 (Scheme ), we observed an energetic preference for 1a INT2, corresponding to attack at the β′-site. However, despite the relative stability of the latter intermediate, the reaction takes a different turn, namely, the formation of less stable 6 INT1 (arising by δ-attack), whose protonation yields 1a P1 as the most stable product (corresponding to 15 in the experiment). However, the calculated free energy differences among products 1a P1– 1aP3 are very small, falling within the typical error range of DFT calculations (2–3 kcal mol–1), and the reactions are close to being thermoneutral (ca. −4 kcal mol–1). Furthermore, the most stable product varied across different DFT methods, making the results uncertain. In conflict with the experimental findings, our calculations suggest that double addition product 1a P5 should be even more stable than single addition product 1a P1, but the energy difference between them (2 kcal mol–1) is also rather small.
12. Illustration of Plausible Pathways for the Reaction of Dendralene 1a ( 1a R1 in this scheme) with Methanethiolate .
a The Gibbs free energies under the standard conditions (298.15 K and 1 M) were calculated at the ωB97M-V/def2-TZVPD//B3LYP-D3BJ/def2-TZVP level of theory. The experimentally observed pathway is highlighted by a bold arrow.
Since 1b, a homologue of 1a, undergoes preferential β′-attack, followed by cyclization (Scheme ), we attempted to find a rationale for the δ-attack in the case of 1a. Here, the β′-attack would generate enolate 1a INT2, whose 5(O)-endo-trig cyclization toward the α-carbon is precluded by a high energy barrier (>35 kcal mol–1), in consonance with the Baldwin rules. Furthermore, the alternative 5(O)-exo-trig cyclization toward the γ-carbon would yield unstable products, which cannot be optimized by DFT. The fact that the structure reverts back to reactants is presumably due to the strain in the [3.3.0] system that the latter ring formation would generate. Hence, cyclization is not preferred in either of the instances. Interestingly, after a potential aromatization by formation of the furan ring, product 1a P4 would be of practically the same energy as the product of the simple addition, 1a P1 (Scheme ). This finding indicates that it is mainly the strain in the [3.3.0] potential intermediate that disfavors this pathway. Thus, due to the lack of stabilization by subsequent cyclization, and with a view of the principal reversibility of each step in all possible conjugate additions, it is understandable that the reaction of 1a takes a path different from that of 1b. This scenario thus results in a simple δ-attack, followed by protonation, which affords 15 ( 1a P1) as a major isolable product.
By contrast, addition of thiolate to the cyclohexenone (1b) and cycloheptenone homologues (1c) occurs at the enone moiety. Subsequent Michael-type cyclization, analogous to pathway i in Scheme , then gives nonchiral furan derivatives 17b and 17c, respectively. In analogy to dendralene 1a, a computational investigation of the addition of thiolate to dendralene 1b ( 1b R1;Scheme ) indicates a thermodynamically controlled thiolate attack at the selected electrophilic sites. Furthermore, the reaction free energies for the formation of protonated products 1b P1, 1b P2, and 1b P5 were found to be consistent with those calculated for dendralene 1a (ca. −4 kcal mol–1). However, in contrast to the inability of 1a (specifically 1a INT2) to cyclize (Scheme ), the nucleophilic 5(O)-exo-trig cyclization of enolate 1b INT2 toward the γ-carbon becomes plausible. This process was found to proceed through intermediate 1b INT4, which is 4 kcal mol–1 uphill in energy (Figure S8), to ultimately reach cyclization product 1b P4, which is consistent with the experiment (17b in Scheme ). While the energy of alternative cyclization product 1b P3 is predicted to be similar to that of 1b P4, the reaction barrier of the required 5(O)-endo-trig enolate cyclization toward the α-carbon is prohibitively high (28 kcal mol–1), which is consistent with the Baldwin rules. Notably, the reaction free energy of the cyclization product from dendralene 1b (−18 kcal mol–1) is significantly lower than that of any addition product ( 1b P1 or 1b P2; Scheme ) by −14 kcal mol–1, making product 1b P4 (and its experimental counterpart 17b) a thermodynamic sink.
13. Illustration of Plausible Pathways for the Reaction of Dendralene 1b ( 1b R1 in this scheme) with Methanethiolate .
a The Gibbs free energies under the standard conditions (298.15 K and 1 M) were calculated at the ωB97M-V/def2-TZVPD//B3LYP-D3BJ/def2-TZVP level of theory. The experimentally observed pathway is highlighted by a bold arrow.
To qualitatively explain the energy difference between the α- and preferred γ-cyclization pathways (Scheme ), we employed the intrinsic bond orbital (IBO) approach. The IBO approach transforms the DFT wave function into a set of orthogonal orbitals localized at a minimal number of atoms, which has been shown to provide intuitive interpretation of chemical bonding. In our analysis, we examined the three IBO orbitals, which undergo the most changes during the enolate cyclization (Scheme ). Consistent with the Baldwin rules, the γ-attack was found to exhibit a favorable orbital overlap, due to the nearly perpendicular orientation of the interacting orbitals, which is more favorable than that for the α-attack, which would require partial breaking of the CαCβ bond.
14. Relevant Occupied Intrinsic Bond Orbitals (IBOs, exponent 2) Depicting Orbital Changes during the Enolate Attack at the Less Favorable α-Position (A) and the More Favorable γ-Position (B) of the Dienoate Unit.
Finally, we turned to sulfone 1i and investigated its reaction with thiolate (Scheme ) in a manner similar to what we used for enones 1a and 1b. The most energetically favorable anionic intermediates were predicted to correspond to an attack at the dienoate α-carbon. Depicted intermediate 1i INT1 (Figure B) represents the nucleophilic attack at the most stable conformer of 1i (Figure A), whereas 1i INT4 (in fact lower in energy and more closely resembling product 1i P1) would represent nucleophilic attack at hypothetical conformer 1i*, which was predicted to be unstable (at the B3LYP/TZVP level of theory). These intermediates are lower in energy than 1i INT2 and 1i INT3, which stands in sharp contrast with dendralenes 1a and 1b, where the most stable anionic intermediate corresponds to thiolate attack at the β′- and δ-sites, respectively (Schemes and ). This α-preference could be attributed to the electron-withdrawing effect of the phenyl sulfone group being stronger than that of the enone and acrylate moieties. Subsequent protonation of anionic intermediate 1i INT1 or 1i INT4 yields product 1i P1, an energetically most stable species among the possible products (Scheme ), which is consistent with the experimental finding (20 in Scheme ).
15. Illustration of Plausible Pathways for the Reaction of Sulfone 1i ( 1i R1 in this Scheme) with Methanethiolate .

a The Gibbs free energies under the standard conditions (298.15 K and 1 M) were calculated at the ωB97M-V/def2-TZVPD//B3LYP-D3BJ/def2-TZVP level of theory. The experimentally observed pathway is highlighted by a bold arrow.
5.

Most stable DFT-optimized conformers of (A) sulfone 1i and (B) intermediate 1i INT1.
To conclude, reactions of thiolates with dendralenes were found to offer several pathways with negligible barriers, where the overall reactivity is governed by the thermodynamic stability of all possible products, with the exception of some cyclization reactions that are kinetically prohibited. In the case of 1a, the DFT methods are not sufficiently accurate to provide useful selectivity predictions of these vast reaction networks (Figure S8). In other instances, such as in the cyclization product originating from 1b or the addition of thiolate to 1i, we were able to rationalize the observed outcomes based on the DFT results.
Conclusions
In conclusion, this exploratory study of the chemistry of electron-deficient dendralenes with an enone fragment has, for the first time, demonstrated the variability of nucleophilic addition reactions to these intriguing compounds, proceeding in both Michael and anti-Michael fashions. The results showed that the initial nucleophilic attack triggers a variety of reaction paths, including simple additions on one hand or attractive multiple bond-forming cascades on the other. The actual reaction path depends on both the structure of the starting dendralene and the nature of the nucleophile. In general, the reactions gave rise to products, which no longer possessed like charges at the neighboring carbons or for which this “dissonant” disposition was diminished, as in 12f and 12g. Another factor with a profound influence, exercised in these reactions, seems to be the formation of a stable aromatic or bicyclic system. Specifically, hydride (from NaBH4) prefers Michael addition at the enone fragment, regardless of the substrate. However, in the case of compound 1a with the five-membered cyclic enone, a simple double bond shift in the ensuing enolate gave rise to product 5 with alternating charges at the neighboring carbons. On the other hand, dendralenes 1b and 1c with larger rings and acyclic analogue 1d afforded enolates that are capable of further cyclization to afford polysubstituted furans 7b–7d, respectively, as the major products. Switching to thiolate nucleophiles showed a remarkable difference between cyclopentenone derivative 1a and the rest of the series again. While 1a underwent a simple 1,6-addition (anti-Michael) at the less hindered δ-position of the dienoate moiety, thiolate attack at 1b and 1c produced ortho-fused furans 17b and 17c, respectively, as a result of an initial β′-attack followed by cyclization. Density functional theory calculations suggest that nucleophilic additions of thiolates to dendralenes proceed (mostly) in a reversible fashion, making them mostly thermodynamically controlled (with the exception of some kinetically controlled cyclization reactions). For 1b, furan ring formation serves as a thermodynamic sink, resulting in the formation of 17b as the stable product. With stabilized C-nucleophiles, 1a reacted to deliver bicyclo[4.3.0]nonane systems 24x–24z in a diastereoselective manner, while unexpectedly, 1b–1d turned out to be inert. Finally, generating enolates via deprotonation confirmed the trend, with dendralene 1a giving a complex mixture, while 1c and 1d afforded furan derivatives 33 and 35, respectively, in high preparative yields. By seeming contrast, six-membered enone 1b furnished coumarin derivative 31 with DBU, both as a stoichiometric base and as a catalyst. However, this cyclization mode was almost certainly dictated by the spontaneous aromatization of the six-membered carbocyclic ring even in the absence of air oxygen, an option not available with 1c and 1d. The sequences developed in this study are ready to be applied for the synthesis of natural products with the corresponding carbon skeletons or their analogues. Specifically, the facile formation of polysubstituted furans from simple, easy-to-make precursors is particularly notable, since a plethora of natural products, in which the furan core is fused to a six- or seven-membered saturated ring, such as zedoarol, crotoxides A and B, gnididione, and others, can be viewed as viable synthetic targets.
Supplementary Material
Acknowledgments
This work was supported by the Czech Science Foundation (Project 22-19209S), by New Technologies for Translational Research in Pharmaceutical Sciences/NETPHARM (Project CZ.02.01.01/00/22_008/0004607), cofunded by the European Union, and by the Ministry of Education, Youth and Sports of the Czech Republic through the e-INFRA CZ (90254). S.M.P. acknowledges financial support from Charles University (Projects GA UK 362421 and SVV 260 781).
The data underlying this study are available in the published article and its .
The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.joc.5c02397.
FAIR data, including the primary NMR FID files, for compounds 1a–1g, 1i, 5, 7b–7d, 9b, 9d, 10, 11, 12f, 12g, 14, 15, 17b, 17c, 18b, 18c, 24x–24z, 28, 31, 33, and 35 (ZIP)
Experimental procedures and spectroscopic data for all compounds, X-ray crystallographic data, computational details, and copies of NMR spectra (PDF)
The authors declare no competing financial interest.
References
- For reviews, see:; a Hopf H.. The Dendralenes - A Neglected Group of Highly Unsaturated-Hydrocarbons. Angew. Chem., Int. Ed. 1984;23(12):948–960. doi: 10.1002/anie.198409481. [DOI] [Google Scholar]; b Hopf H.. Dendralenes: The Breakthrough. Angew. Chem., Int. Ed. 2001;40(4):705–707. doi: 10.1002/1521-3773(20010216)40:4<705::AID-ANIE7050>3.0.CO;2-8. [DOI] [PubMed] [Google Scholar]; c Hopf H., Sherburn M. S.. Dendralenes Branch Out: Cross-Conjugated Oligoenes Allow the Rapid Generation of Molecular Complexity. Angew. Chem., Int. Ed. 2012;51(10):2298–2338. doi: 10.1002/anie.201102987. [DOI] [PubMed] [Google Scholar]
- Hopf, H. ; Sherburn, M. S. . Cross Conjugation: Modern Dendralene, Radialene and Fulvene Chemistry; John Wiley & Sons: Hoboken, NJ, 2016. [Google Scholar]
- For representative examples of DTDA reactions, see, e.g., the following:; a Payne A. D., Willis A. C., Sherburn M. S.. Practical Synthesis and Diels-Alder Chemistry of [4]Dendralene. J. Am. Chem. Soc. 2005;127(35):12188–12189. doi: 10.1021/ja053772+. [DOI] [PubMed] [Google Scholar]; b Chung S.-I., Seo J., Cho C.-G.. Tandem Diels-Alder Cycloadditions of 2-Pyrone-5- acrylates for the Efficient Synthesis of Novel Tetracyclolactones. J. Org. Chem. 2006;71(17):6701–6704. doi: 10.1021/jo061119e. [DOI] [PubMed] [Google Scholar]; c M. Brummond K., Mitasev B., Yan B.. Cycloaddition Reactions of Amino-Acid Derived Cross-Conjugated Trienes: Stereoselective Synthesis of Novel Heterocyclic Scaffolds. Heterocycles. 2006;70:367–388. doi: 10.3987/COM-06-S(W)36. [DOI] [Google Scholar]; d Cergol K. M., Newton C. G., Lawrence A. L., Willis A. C., Paddon-Row M. N., Sherburn M. S.. 1,1-Divinylallene. Angew. Chem., Int. Ed. 2011;50(44):10425–10428. doi: 10.1002/anie.201105541. [DOI] [PubMed] [Google Scholar]; e Lindeboom E. J., Willis A. C., Paddon-Row M. N., Sherburn M. S.. Computational and Synthetic Studies with Tetravinylethylenes. J. Org. Chem. 2014;79(23):11496–11507. doi: 10.1021/jo5021294. [DOI] [PubMed] [Google Scholar]; f George J., Sherburn M. S.. Diene-Transmissive Enantioselective Diels–Alder Reactions and Sequences Involving Substituted Dendralenes. J. Org. Chem. 2019;84(22):14712–14723. doi: 10.1021/acs.joc.9b02296. [DOI] [PubMed] [Google Scholar]; g Fan Y.-M., Yu L.-J., Gardiner M. G., Coote M. L., Sherburn M. S.. Enantioselective oxa-Diels–Alder Sequences of Dendralenes. Angew. Chem., Int. Ed. 2022;61(39):e202204872. doi: 10.1002/anie.202204872. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kostikov R. R., Molchanov A. P.. Reaction of Dichlorocarbene with 2-Vinylbuta-1,3-diene. J. Org. Chem. USSR (English Transl.) 1975;11(2):438–449. [Google Scholar]
- Bojase G., Nguyen T. V., Payne A. D., Willis A. C., Sherburn M. S.. Synthesis and properties of the ivyanes: the parent 1,1-oligocyclopropanes. Chem. Sci. 2011;2:229–232. doi: 10.1039/C0SC00500B. [DOI] [Google Scholar]
- Zhang C., Tian J., Ren J., Wang Z.. Intramolecular Parallel [4+3] Cycloadditions of Cyclopropane 1,1-Diesters with [3]Dendralenes: Efficient Construction of [5.3.0]Decane and Corresponding Polycyclic Skeletons. Chem. - Eur. J. 2017;23(6):1231–1236. doi: 10.1002/chem.201605190. [DOI] [PubMed] [Google Scholar]
- For anionic polymerization, see: [Google Scholar]; a Takamura Y., Takenaka K., Toda T., Takeshita H., Miya M., Shiomi T.. Anionic Polymerization of 2-Hexyl[3]dendralene. Macromol. Chem. Phys. 2018;219(1):1700046. doi: 10.1002/macp.201700046. [DOI] [Google Scholar]; b Takagi T., Toda T., Miya M., Takenaka K.. Stable and Highly Regioselective Anionic Polymerization of (Z)-1-Phenyl[3]dendralene. Macromolecules. 2021;54(9):4326–4332. doi: 10.1021/acs.macromol.1c00260. [DOI] [Google Scholar]
- Desfeux C., Besnard C., Mazet C.. [n]Dendralenes as a Platform for Selective Catalysis: Ligand-Controlled Cu-Catalyzed Chemo-, Regio-, and Enantioselective Borylations. Org. Lett. 2020;22(21):8181–8187. doi: 10.1021/acs.orglett.0c01892. [DOI] [PubMed] [Google Scholar]
- Antal R., Staś M., Perdomo S. M., Štemberová M., Brůža Z., Matouš P., Kratochvíl J., Růžička A., Rulíšek L., Kuneš J., Kočovský P., Andris E., Pour M.. Synthesis of highly polarized [3]dendralenes and their Diels–Alder reactions. Org. Chem. Front. 2023;10:5568–5578. doi: 10.1039/D3QO01221B. [DOI] [Google Scholar]
- The term “dissonant compounds” was suggested by D. A. Evans for the description of such species in the 1970s. For a thorough discussion of this topic, see: Hudlický, T. ; Reed, J. W. . The Way of Synthesis; Wiley-VCH Verlag GmbH & KGaA: Wienheim, Germany, 2007; pp 186–187. [Google Scholar]
- The additions of a nucleophile to an α-carbon (as in Figure , in blue) can, in principle, be termed anti-Michael additions, reflecting opposite regiochemistry to Michael reactions. However, since the α-carbons are also in the δ-positions to another oxo group, the reactions can equally well be classified as (vinylogous) 1,6-conjugate Michael additions.
- The configuration at the tetrasubstituted CC bond in 5 was established by 2D NMR spectroscopy.
- The order of events in the formation of 9d (pathway ii) could a priori be reversed; i.e., the dienoate unit might first undergo conjugate reduction, and the intermediate would then by lactonized. However, since triester 1a was found to be inert toward treatment with NaBH4, it is unlikely that the dienoate moiety would be reduced in this instance.
- For catalysis by I2, see:; a Breugst M., Detmar E., von der Heiden D.. Origin of the Catalytic Effects of Molecular Iodine: A Computational Analysis. ACS Catal. 2016;6(5):3203–3212. doi: 10.1021/acscatal.6b00447. [DOI] [Google Scholar]; b von der Heiden D., Bozkus S., Klussmann M., Breugst M.. Reaction Mechanism of Iodine-Catalyzed Michael Additions. J. Org. Chem. 2017;82(8):4037–4043. doi: 10.1021/acs.joc.7b00445. [DOI] [PubMed] [Google Scholar]; c Čebular K., Stavber S.. Molecular iodine as a mild catalyst for cross-coupling of alkenes and alcohols. Pure Appl. Chem. 2018;90(2):377–386. doi: 10.1515/pac-2017-0414. [DOI] [Google Scholar]; d von der Heiden D., Detmar E., Kuchta R., Breugst M.. Activation of Michael Acceptors by Halogen-Bond Donors. Synlett. 2018;29(8):1307–1313. doi: 10.1055/s-0036-1591841. [DOI] [Google Scholar]; e Marsili L. A., Pergomet J. L., Gandon V., Riveira M.. Iodine-Catalyzed Iso-Nazarov Cyclization of Conjugated Dienals for the Synthesis of 2-Cyclopentenones. Org. Lett. 2018;20(22):7298–7303. doi: 10.1021/acs.orglett.8b03229. [DOI] [PubMed] [Google Scholar]; f Breugst M., von der Heiden D.. Mechanisms in Iodine Catalysis. Chem. - Eur. J. 2018;24(37):9187–9199. doi: 10.1002/chem.201706136. [DOI] [PubMed] [Google Scholar]; g Tran U. P. N., Oss G., Breugst M., Detmar E., Pace D. P., Liyanto K., Nguyen T. V.. Carbonyl–Olefin Metathesis Catalyzed by Molecular Iodine. ACS Catal. 2019;9(2):912–919. doi: 10.1021/acscatal.8b03769. [DOI] [Google Scholar]; h Oss J., Nguyen T. V.. Iodonium-Catalyzed Carbonyl–Olefin Metathesis Reactions. Synlett. 2019;30(17):1966–1970. doi: 10.1055/s-0039-1690297. [DOI] [Google Scholar]; i Bandi V., Kavala V., Konala A., Hsu C.-H., Villuri B. K., Reddy S. R., Lin L.-C., Kuo C. W., Yao C.-F.. Synthesis of Polysubstituted Cyclopentene and Cyclopenta[b]carbazole Analogues from Unsymmetrical 4-Arylidene-3,6-diarylhex-2-en-5-ynal and Indole Derivatives via an Iodine Mediated Electrocyclization Reaction. J. Org. Chem. 2019;84(6):3036–3044. doi: 10.1021/acs.joc.8b02168. [DOI] [PubMed] [Google Scholar]; j Koenig J. J., Arndt T., Gildemeister N., Neudörfl J.-M., Breugst M.. Iodine-Catalyzed Nazarov Cyclizations. J. Org. Chem. 2019;84(12):7587–7605. doi: 10.1021/acs.joc.9b01083. [DOI] [PubMed] [Google Scholar]; k Hamlin T. A., Fernández I., Bickelhaupt F. M.. How Dihalogens Catalyze Michael Addition Reactions. Angew. Chem., Int. Ed. 2019;58(26):8922–8926. doi: 10.1002/anie.201903196. [DOI] [PMC free article] [PubMed] [Google Scholar]; l Hamlin T. A., Bickelhaupt F. M., Fernández I.. The Pauli Repulsion-Lowering Concept in Catalysis. Acc. Chem. Res. 2021;54(8):1972–1981. doi: 10.1021/acs.accounts.1c00016. [DOI] [PubMed] [Google Scholar]; m Arndt T., Wagner P. K., Koenig J. J., Breugst M.. Iodine-Catalyzed Diels-Alder Reactions. ChemCatChem. 2021;13(12):2922–2930. doi: 10.1002/cctc.202100342. [DOI] [Google Scholar]
- For halogen bonding, see:; a Bulfield D., Huber S. M.. Halogen Bonding in Organic Synthesis and Organocatalysis. Chem. - Eur. J. 2016;22(41):14434–14450. doi: 10.1002/chem.201601844. [DOI] [PubMed] [Google Scholar]; b Kolář M. H., Hobza P.. Computer modelling of halogen bonds and other σ-hole interactions. Chem. Rev. 2016;116(9):5155–5187. doi: 10.1021/acs.chemrev.5b00560. [DOI] [PubMed] [Google Scholar]; c Costa P. J.. The halogen bond: Nature and applications. Phys. Sci. Rev. 2017;2(11):20170136. doi: 10.1515/psr-2017-0136. [DOI] [Google Scholar]; d Breugst M., von der Heiden D., Schmauck J.. Novel Noncovalent Interactions in Catalysis: A Focus on Halogen, Chalcogen, and Anion-π Bonding. Synthesis. 2017;49(15):3224–3236. doi: 10.1055/s-0036-1588838. [DOI] [Google Scholar]; e Kesharwani M. K., Manna D., Sylvetsky N., Martin J. M. L.. The X40 × 10 Halogen Bonding Benchmark Revisited: Surprising Importance of (n–1)d Subvalence Correlation. J. Phys. Chem. A. 2018;122(8):2184–2197. doi: 10.1021/acs.jpca.7b10958. [DOI] [PubMed] [Google Scholar]; f Breugst M., Koenig J. J.. σ-Hole Interactions in Catalysis. Eur. J. Org. Chem. 2020;2020(34):5473–5487. doi: 10.1002/ejoc.202000660. [DOI] [Google Scholar]; g Mallada B., Gallardo A., Lamanec M., de la Torre B., Špirko V., Hobza P., Jelínek P.. Real-space imaging of anisotropic charge of σ-hole by means of Kelvin probe force microscopy. Science. 2021;374:863–877. doi: 10.1126/science.abk1479. [DOI] [PubMed] [Google Scholar]
- Ballini R., Palmieri A.. Formation of Carbon-Carbon Double Bonds: Recent Developments via Nitrous Acid Elimination (NAE) from Aliphatic Nitro Compounds. Adv. Synth. Catal. 2019;361(22):5070–5097. doi: 10.1002/adsc.201900563. [DOI] [Google Scholar]
- For recent examples of nitrous acid elimination, see, e.g., the following:; a Roy S.. An Unusual Route to Synthesize Indolizines through a Domino SN2/Michael Addition Reaction Between 2-Mercaptopyridine and Nitroallylic Acetates. Eur. J. Org. Chem. 2019;2019(4):765–769. doi: 10.1002/ejoc.201801426. [DOI] [Google Scholar]; b Chen W. T., Luo F., Wu Y., Cen J., Shao J. A., Yu Y. P.. Efficient access to fluorescent benzofuro[3,2-b]carbazoles via TFA-promoted cascade annulations of sulfur ylides, 2-hydroxy-β-nitrostyrenes and indoles. Org. Chem. Front. 2020;7:873–878. doi: 10.1039/D0QO00137F. [DOI] [Google Scholar]
- Thomas B. E., Kollman P. A.. An ab initio molecular orbital study of the first step of the catalytic mechanism of thymidylate synthase: the Michael addition of sulfur and oxygen nucleophiles. J. Org. Chem. 1995;60:8375–8381. doi: 10.1021/jo00131a012. [DOI] [Google Scholar]
- a Raycroft M. A. R., Racine K. É., Rowley C. N., Keillor J. W.. Mechanisms of Alkyl and Aryl Thiol Addition to N-Methylmaleimide. J. Org. Chem. 2018;83:11674–11685. doi: 10.1021/acs.joc.8b01638. [DOI] [PubMed] [Google Scholar]; b Awoonor-Williams E., Isley W. C., Dale S. G., Johnson E. R., Yu H., Becke A. D., Roux B., Rowley C. N.. Quantum Chemical Methods for Modeling Covalent Modification of Biological Thiols. J. Comput. Chem. 2020;41:427–438. doi: 10.1002/jcc.26064. [DOI] [PubMed] [Google Scholar]; c Costa A. M., Bosch L., Petit E., Vilarrasa J.. Computational Study of the Addition of Methanethiol to 40+ Michael Acceptors as a Model for the Bioconjugation of Cysteines. J. Org. Chem. 2021;86:7107–7118. doi: 10.1021/acs.joc.1c00349. [DOI] [PMC free article] [PubMed] [Google Scholar]; d Watt S. K. I., Charlebois J. G., Rowley C. N., Keillor J. W.. A mechanistic study of thiol addition to N-acryloylpiperidine. Org. Biomol. Chem. 2023;21:2204–2212. doi: 10.1039/D2OB02223K. [DOI] [PubMed] [Google Scholar]; e Liu R., Vázquez-Montelongo E. A., Ma S., Shen J.. Quantum Descriptors for Predicting and Understanding the Structure–Activity Relationships of Michael Acceptor Warheads. J. Chem. Inf. Model. 2023;63:4912–4923. doi: 10.1021/acs.jcim.3c00720. [DOI] [PMC free article] [PubMed] [Google Scholar]; f Ma S., Patel H., Peeples C. A., Shen J.. QM/MM Simulations of Afatinib-EGFR Addition: The Role of β-Dimethylaminomethyl Substitution. J. Chem. Theory Comput. 2024;20:5528–5538. doi: 10.1021/acs.jctc.4c00290. [DOI] [PMC free article] [PubMed] [Google Scholar]
- a Brown J. S., Ruttinger A. W., Vaidya A. J., Alabi C. A., Clancy P.. Decomplexation as a rate limitation in the thiol-Michael addition of N-acrylamides. Org. Biomol. Chem. 2020;18:6364–6377. doi: 10.1039/D0OB00726A. [DOI] [PubMed] [Google Scholar]; b Berne D., Lemouzy S., Guiffrey P., Caillol S., Ladmiral V., Manoury E., Poli R., Leclerc E.. Catalyst-Free Thia-Michael Addition to α-Trifluoromethylacrylates for 3D Network Synthesis. Chem. - Eur. J. 2023;29:e202203712. doi: 10.1002/chem.202203712. [DOI] [PubMed] [Google Scholar]
- Roseli R. B., Keto A. B., Krenske E. H.. Mechanistic aspects of thiol additions to Michael acceptors: Insights from computations. WIREs Comput. Mol. Sci. 2023;13:e1636. doi: 10.1002/wcms.1636. [DOI] [Google Scholar]
- Smith J. M., Jami Alahmadi Y., Rowley C. N.. Range-Separated DFT Functionals are Necessary to Model Thio-Michael Additions. J. Chem. Theory Comput. 2013;9:4860–4865. doi: 10.1021/ct400773k. [DOI] [PubMed] [Google Scholar]
- Ting J. Y. C., Roseli R. B., Krenske E. H.. How does cross-conjugation influence thiol additions to enones? A computational study of thiol trapping by the naturally occurring divinyl ketones zerumbone and α-santonin. Org. Biomol. Chem. 2020;18:1426–1435. doi: 10.1039/C9OB02709B. [DOI] [PubMed] [Google Scholar]
- Takagi T., Toda T., Miya M., Takenaka K.. DFT study on the anionic polymerization of phenyl-substituted [3]dendralene derivatives: reactivities of monomer and chain end carbanion. Polym. J. 2022;54:643–652. doi: 10.1038/s41428-022-00615-1. [DOI] [Google Scholar]
- Mardirossian N., Head-Gordon M.. ωB97M-V: A combinatorially optimized, range-separated hybrid, meta-GGA density functional with VV10 nonlocal correlation. J. Chem. Phys. 2016;144(21):214110. doi: 10.1063/1.4952647. [DOI] [PubMed] [Google Scholar]
- a 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. doi: 10.1039/b508541a. [DOI] [PubMed] [Google Scholar]; b Rappoport D., Furche F.. Property-optimized Gaussian basis sets for molecular response calculations. J. Chem. Phys. 2010;133(13):134105. doi: 10.1063/1.3484283. [DOI] [PubMed] [Google Scholar]
- a Vosko S. H., Wilk L., Nusair M.. Accurate spin-dependent electron liquid correlation energies for local spin density calculations: a critical analysis. Can. J. Phys. 1980;58(8):1200–1211. doi: 10.1139/p80-159. [DOI] [Google Scholar]; b Lee C., Yang W., Parr R. G.. Development of the Colle-Salvetti correlation-energy formula into a functional of the electron density. Phys. Rev. B. 1988;37:785–789. doi: 10.1103/PhysRevB.37.785. [DOI] [PubMed] [Google Scholar]; c Becke A. D.. Density-functional thermochemistry. III. The role of exact exchange. J. Chem. Phys. 1993;98(7):5648–5652. doi: 10.1063/1.464913. [DOI] [Google Scholar]; d Stephens P. J., Devlin F. J., Chabalowski C. F., Frisch M. J.. Ab Initio Calculation of Vibrational Absorption and Circular Dichroism Spectra Using Density Functional Force Fields. J. Phys. Chem. 1994;98(45):11623–11627. doi: 10.1021/j100096a001. [DOI] [Google Scholar]; e Grimme S., Antony J., Ehrlich S., Krieg H.. A consistent and accurate ab initio parametrization of density functional dispersion correction (DFT-D) for the 94 elements H-Pu. J. Chem. Phys. 2010;132(15):154104. doi: 10.1063/1.3382344. [DOI] [PubMed] [Google Scholar]; f Grimme S., Ehrlich S., Goerigk L.. Effect of the damping function in dispersion corrected density functional theory. J. Comput. Chem. 2011;32(7):1456–1465. doi: 10.1002/jcc.21759. [DOI] [PubMed] [Google Scholar]
- a Knizia G.. Intrinsic Atomic Orbitals: An Unbiased Bridge between Quantum Theory and Chemical Concepts. J. Chem. Theory Comput. 2013;9(11):4834–4843. doi: 10.1021/ct400687b. [DOI] [PubMed] [Google Scholar]; b Knizia G., Klein J. E. M. N.. Electron Flow in Reaction Mechanisms Revealed from First Principles. Angew. Chem., Int. Ed. 2015;54(18):5518–5522. doi: 10.1002/anie.201410637. [DOI] [PubMed] [Google Scholar]
- Chatani S., Nair D. P., Bowman C. N.. Relative reactivity and selectivity of vinyl sulfones and acrylates towards the thiol–Michael addition reaction and polymerization. Polym. Chem. 2013;4:1048–1055. doi: 10.1039/C2PY20826A. [DOI] [Google Scholar]
- Shiobara Y., Asakawa Y., Kodama M., Takemoto T.. Zedoarol, 13-Hydroxygermacrone and Curzeone, Three Sesquiterpenoids from Curcuma zedoaria . Phytochemistry. 1986;25(6):1351–1353. doi: 10.1016/S0031-9422(00)81288-1. [DOI] [Google Scholar]
- Jogia M. K., Andersen R. J., Parkanyi L., Clardy J., Dublin H. T., Sinclair A. R. E.. Crotofolane Diterpenoids from the African Shrub Croton dichogamus Pax. J. Org. Chem. 1989;54(7):1654–1657. doi: 10.1021/jo00268a029. [DOI] [Google Scholar]
- Kupchan S. M., Shizuri Y., Baxter R. L., Haynes H. R.. Gnididione, a New Furanosesquiterpene from Gnidia latifolia . J. Org. Chem. 1977;42(2):348–350. doi: 10.1021/jo00422a040. [DOI] [PubMed] [Google Scholar]
- For cyclohex[b]furan natural products, see:; a Nakamura H., Kobayashi J., Kobayashi M., Ohizumi Y., Hirata Y.. Xestoquinone. A Novel Cardiotonic Marine Natural Product Isolated From The Okinawan Sea Sponge Xestospongia sapra . Chem. Lett. 1985;14(6):713–716. doi: 10.1246/cl.1985.713. [DOI] [Google Scholar]; b Hikino H., Agatsuma K., Takemoto T.. Structure of Curzerenone, Epicurzerenone, and Isofuranogermacrene (Curzerene) Tetrahedron Lett. 1968;9(24):2855–2858. doi: 10.1016/S0040-4039(00)75646-2. [DOI] [Google Scholar]; c McPherson D. D., Che Ch.-T., Cordell G. A., Doel Soejarto D., Pezzuto J. M., Fong H. H. S.. Diterpenoids from Caesalpinia pulcherrima . Phytochemistry. 1985;25(1):167–170. doi: 10.1016/S0031-9422(00)94522-9. [DOI] [Google Scholar]; d Macias F. A., Fronczek F. R., Fischer N. H.. Menthofurans from Calamintha ashei and the Absolute Configuration of Desacetylcalaminthone. Phytochemistry. 1989;28(1):79–82. doi: 10.1016/0031-9422(89)85013-7. [DOI] [Google Scholar]; e Ryu S. Y., Lee Ch. O., Choi S. U.. In Vitro Cytotoxicity of Tanshinones from Salvia miltiorrhiza . Planta Med. 1997;63(4):339–342. doi: 10.1055/s-2006-957696. [DOI] [PubMed] [Google Scholar]; f Syu W.-J., Shen Ch.-Ch., Don M.-J., Ou J.-Ch., Lee G.-H., Sun Ch.-M.. Cytotoxicity of Curcuminoids and Some Novel Compounds from Curcuma zedoaria . J. Nat. Prod. 1998;61(12):1531–1534. doi: 10.1021/np980269k. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data underlying this study are available in the published article and its .


















