Gold-Catalyzed Diyne-Ene Annulation for the Synthesis of Polysubstituted Benzenes through Formal [3+3] Approach with Amide as the Critical Co-Catalyst

Abstract

The one-step synthesis of tetra-substituted benzenes was accomplished via gold-catalyzed diyne-ene annulation. Distinguished from prior modification methods, this novel strategy undergoes formal [3+3] cyclization, producing polysubstituted benzenes with exceptional efficiency. The critical factor enabling this transformation was the introduction of amides, which were reported for the first time in gold catalysis as covalent nucleophilic co-catalysts. This interesting protocol not only offers a new strategy to achieve functional benzenes with high efficiency, but also enlightens potential new reaction pathways within gold-catalyzed alkyne activation processes.

Benzene stands as a cornerstone in chemical industry, wielding profound significance across different domains including natural products, pharmaceuticals, and materials science.^[1] Synthesis of benzene derivatives containing various functional groups (FGs) has been a long time focus through chemistry history, evidenced by the numerous methodologies reported, from electrophilic/nucleophilic aromatic substitution to transition metal mediated coupling and C–H activation.^[2] Though with tremendous successes to access mono- and di-substituted benzenes, chemists recognized the challenge associated with the synthesis of multiple substituted benzenes with diverse functional groups due to potentially strict regiochemistry control and undesired steric impact.^[3] Nevertheless, the fast growth of bio-medicinal research and material chemistry further emphasized the urgent demand in developing new practical methods to achieve polysubstituted benzenes.^[4]

Compared with post-benzene functionalization methods (i.e. C–H activation), the formal cycloaddition strategy presents notable advantages for preparing multi-functional benzene through modular approaches (Scheme 1A).^[5] Existing methods have predominantly focused on either [4+2] or [2+2+2] cycloaddition under thermal or metal-catalyzed conditions.^[6] Although [3+3] cycloaddition could be a plausible approach to form 6-membered rings, it has seen limited use in all-carbon cyclic structure synthesis.^[7] This is due to the high reactivity associated with allyl cations, anions, and radicals, which tend to give many side reactions over the desired [3+3] cycloaddition pathway.

Scheme 1.

Synthesis of polysubstituted benzenes by gold catalysis.

Over the past decade, our group has been engaged in gold catalysis with the focus on new reactivity discovery through mechanistically guided exploration.^[8] Recent developments in gold(I) catalysis have provided new bond-forming synthetic approaches to novel materials and drug candidates.^[9] Some interesting new reactivity of conjugated diynes^[10] and enynes^[11] has been disclosed upon gold activation. One noteworthy discovery was the diyne-ene cycloaddition with amide nucleophiles, leading to the sequential formation of asymmetric dihydropyridines via enantioselective gold catalysis (Scheme 1B).^[8c] With the strong interest in expanding this methodology to the synthesis of chiral quaternary carbon centers, we extended investigations on diyne-enes to trisubstituted alkenes. Surprisingly, no dihydropyridine products were observed. Instead, amides were discovered as critical covalent nucleophilic catalysts, facilitating gold-carbene formation and formal allyl addition to yield the desired polysubstituted benzenes with high efficiency. Herein, we report this amide-assisted gold-catalyzed diyne-ene annulation, producing tetra-substituted benzenes in one-step with excellent yields (up to 95%) and good functional group tolerability (>30 substrates, Scheme 1C).

As shown in Scheme 2A, gold cation can activate 1,6-enyne to generate the cyclopropane gold carbene intermediate A. Previous reports suggest that gold carbene might initiate facile C–H insertion if the functional group is in the close proximate position.^[12] Based on this information, we postulate that gold carbene may react with the allylic C–H bond to produce conjugated diene-yne B which could subsequently cyclize to form the benzene skeleton (Scheme 2A). Notably, the substitution on alkene can significantly impact reactivity due to the formation of the [4.1.0] bicyclic gold-carbene intermediate, with the allylic C–H located at either concave or convex position.

Scheme 2.

Diyne-ene annulation for the synthesis of polysubstituted benzenes.

To test our hypothesis, diyne-enes 1a–1 c (Scheme 2B) were prepared and applied in various gold-catalyzed conditions. Charging methyl substituted substrate 1a with gold catalyst yielded the bicyclic product 2a as the single diastereomer, suggesting the formation of gold carbene with the methyl group located at the convex face. However, the reaction of the dimethyl-substituted alkene 1b resulted in substrate decompositions under various conditions, with no clear products identified, despite complete consumption of the diyne-ene. Interestingly, the use of wet solvents gave the water double-addition product 2b (30% yield), which underwent slow decomposition over time. These results suggest that the proposed gold-carbene C–H insertion is problematic due to various undesired competing side pathways.

Considering that water nucleophiles could attack the cyclopropane gold carbene, leading to the formation of a relative stable hydroxy addition product, the reactivity of diyne-ene with various protic nucleophiles was investigated. As shown in Scheme 2B, the reaction of 1b with aniline (ArNH₂) gave complex reaction mixtures with no clear products identified. However, when methanol (MeOH) was employed, the reaction yielded the ether product 2 c. Interestingly, using benzamide as nucleophile, reaction of 1b gave a new product 3a, though in low yield (12%). The structure of 3a was determined without amide attachment. This result is exciting as it not only confirms the possibility in producing polysubstituted benzenes via the proposed diyne-ene annulation, but also highlights the critical role of the co-catalyst in promoting this transformation. Furthermore, reaction of phenyl substituted alkene 1c gave significantly improved yield (68%) with 3b structure confirmed by single-crystal X-ray diffraction. To further evaluate this critical co-catalyst, various amides were explored with 1c. The results are summarized in Figure 1.

Figure 1.

Amide impact on gold-catalyzed diyne-ene annulation.

The acetamide was identified as the optimal catalyst. With the acetamide co-catalyst, the yield of 3b could be further improved to 87% at 40 °C with complete diyne-ene conversion in 12 h (76% yield at rt). Aromatic amides gave moderate yields regardless substitutions. Less nucleophilic compounds, such as CF₃C(O)NH₂, TsNH₂, p-NO₂-aniline, and different carbamates, gave no desired products despite complete diyne-ene consumption. No reaction was observed with the more electron-rich aliphatic amine BnNH₂ (strong Nu), as gold cation was likely quenched by the amine. Formamide gave a good yield of 74%, whereas urea only produced 16% conversion of 1c. These results suggest that balanced nucleophilicity and the ability of amide toward coordinate with the gold cation is required. Interestingly, the N-substituted amides (such as DMF) could also serve as effective co-catalyst, suggesting that proton of amide was not necessary in this case. These results suggest that the amide serves as a “covalent nucleophilic catalyst” toward the cyclopropane opening followed by amide elimination. The sequential addition of alkene to gold-activated enyne leads to the formation of the tetra-substituted benzene. The reaction conditions were further optimized with variation of gold catalysts, solvents, reaction temperature, etc. (Table 1, see details in SI).

Table 1:

Optimization of reaction conditions.^[a,b]

entry	variations	convn. (%)	yield (%)
1	none	100	87
2	[Au]=PPh3AuNTf2	34	21
3	[Au]=JohnPhosAuNTf2	100	70
4	[Au]=XPhosAu(MeCN)SbF6	100	79
5	[Au]=RuPhosAuNTf2	100	82
6	[Au]=XPhosAuCl, AgNTf2	100	45
7	[Au]=IPrAuNTf2	<5	–
8	no amide	100	<5
9	0.2 equiv. amide	100	62
10	2.0 equiv. amide	100	87
11	5 % [Au]	100	85
12	2 % [Au]	50	38
13	other solvents (see SI)	0–100	<76
14	room temperature	100	76
15	no inert atmosphere	100	55

^[a]Standard conditions: under N₂ atmosphere, diyne-ene 1 c (0.1 mmol), 1 equiv. acetamide (0.1 mmol), and 10 % XPhosAuNTf₂ (0.01 mmol) were added to CHCl₃ (2 mL, stabilized with amylenes). Reaction was kept stirring at 40 °C for 12 h.

^[b]Conversion and yields were determined by ¹H NMR spectroscopy using 1,3,5-trimethoxybenzene as internal standard.

Optimized conditions were obtained with XPhosAuNTf₂ (10 mol%), acetamide (1 equiv.) in CHCl₃ at 40°C for 12 h under nitrogen atmosphere (entry 1). A series of primary phosphine ligands were investigated. PPh₃AuNTf₂ exhibited slower kinetic, resulting only 34% 1c conversion after 12 h (entry 2). Encouragingly, electron-rich phosphine ligands such as JohnPhos, XPhos, and RuPhos afforded the desired product with good yields (entries 3–5). In situ activation of the gold complex by a silver salt led to a decreased yield (entry 6). Surprisingly, IPrAuNTf₂ did not provide conversion of 1c (entry 7). It is important to notice that no desired product 3b was detected without amide co-catalysts and only diyne-ene decomposition was observed (entry 8). While reducing co-catalyst to 0.2 equiv. gave slightly dropped yield (62%, entry 9), good yield was observed with excess amount (2 equiv.) of amide (entry 10). Slightly reduced yield was observed with 5% gold loading (85%, entry 11). Further reducing the loading to 2% Au caused incomplete conversion (50%, entry 12). Impacts from alternative solvents and reaction temperature (rt) were also explored with lower yields observed (entries 13–14, see details in SI). Conducting the reaction in open-air conditions led to the reduction of yield (entry 15), likely due to the competing water addition. Application of molecular sieves quenched gold catalyst reactivity, resulting in slow conversion and gold decomposition.

With the optimized conditions in hand, the reaction scope was explored (Table 2). Various diynes worked well under this condition with different modified aryl substituted groups, giving the desired polysubstituted benzenes in good yields (3c, 3d, and 3f). Lower yields were observed with electron withdrawing groups (EWGs) modified arene substitutions (3e, 3j). The CF₃-modified diyne gave no desired product (3am), which is likely due to the competing 5-exodig cyclization, which lead to decomposition over time. Thiophene substrate showed good compatibility with gold catalyst, giving 3k in 84% yield. The quaterphenyl 3l was prepared in excellent yield (95%), suggesting the potential application in polyaromatic hydrocarbons (PAHs) synthesis. Impressively, conjugated enynes are compatible substrates, producing 3m and 3n in good yields. Aliphatic alkynes generally worked well in this transformation with excellent yields (3o–3x). Good FGs tolerability with aliphatic alkyne was observed, allowing substrates containing various FGs (cyclopropyl, –Cy, –Me, –Cl, –OTBS, ester, ether etc.). Unprotected alcohol gave lower yield (3u) due to the competing –OH addition. Ibuprofen modified diyne-ene was also compatible (3 x), highlighting the potential application in drug molecule late-stage modification. Besides NTs and NMs, ether was also suitable, generating benzyl alcohol 3 z in good yield. Interestingly, reaction of amino acid modified diyne with -NBoc₂ protection gave the desired [3+3] cyclization with partial Boc deprotection (3 aa, 55%; 3ab 29%). This result reveals a potential new approach to achieve facile peptide modification. Under the optimal conditions, improved yields of 3a were observed. The highly reactive ynamide could give desired aniline 3ac, though with low yield. Various aryl alkenes work well with this transformation, allowing good substitution pattern of the resulting arenes. Geraniol and t-Bu substituted alkene (3 aj, 3ak) gave slow conversion, likely due to the steric hinderance in the initial enyne cyclization.

Table 2:

Reaction scope of [3+3] annulation of polysubstituted benzenes.^[a,b]

^[a]Standard conditions: In a glove box with N₂ atmosphere, diyne-ene (0.5 mmol), 1 equiv. acetamide (0.5 mmol), and 10 % XPhosAuNTf₂ (0.05 mmol) were added to a CHCl₃ solvent (10 mL, stabilized with amylenes), and reaction was kept stirring at 40 °C for 12 h.

^[b]Isolated yields.

^[c]DCE solvent, 60 °C.

Overall, this gold catalyzed diyne-ene annulation works well with a large group of substrates. Considering that the formal [3+3] approach is under modular fashion, this new strategy allows the rapid synthesis of various polysubstituted benzenes via the combination of different substituted groups with high efficiency and great FG tolerability.

Mechanism studies were substantiated by several control experiments in Scheme 3. To investigate the initial enyne cyclization, the 1,6-enyne 4a and 4b were prepared and applied into the gold catalytic conditions (Scheme 3A). Under typical gold catalytic conditions (no amide cocatalyst), diene 5a was produced with 80% yield. Addition of amide further improved yield to 97%. Methanol nucleophile provided 5b with the standard conditions, and extension reaction time to 48 hours produced 5a with 12% yield. It has been reported in literature that dimethyl substituted enyne 4b gave either diene 5 c or cyclopropane 5d under gold catalyzed conditions.^[13] Charging 4b with our optimal conditions gave skipped-diene 5e in 95% yields (78% yields without amide). Both results suggested that diene C (Scheme 3C) is likely the intermediate from the initial enyne cyclization. Notably, the choice of ligands played a critical role in gold catalyzed enyne cyclization (5-exo vs 6-endo).^[14] To the best of our knowledge, this is the first example reported that 1,6-enyne undergo 6-endo cyclization followed by elimination to give skipped-diene via gold catalysis.^[15]

Scheme 3.

Proposed mechanism and derivatization. Conditions: i) 3 b (0.5 mmol), NBS (1 mmol), TsNH₂ (0.5 mmol), DCM (10 ml), N₂, 40 °C, 20 h; ii) 3 b (0.5 mmol), NBS (1 mmol), DCM (10 ml), N₂, 40 °C, 20 h; iii) 3 b (0.5 mmol), PhB(OH)₂ (1 mmol), PPh₃AuCl (0.05 mmol), MeCN (5 ml), 60 °C, 2 h; : iv) 3 b (0.5 mmol), allyl bromide (1 mmol), K₂CO₃ (2 mmol), MeCN(10 ml), 85 °C, 12 h; v) Hoveyda- Grubbs (0.025 mmol), DCM (50 ml), N₂, rt, 2 h.

The isotope labeled Z-isomer substrate 6a was prepared to explore the function of the amide co-catalyst. (See details in SI) As shown in Scheme 3B (Eq. 1), with CD₃ modified substrate 6a, under the optimal conditions, the polysubstituted benzene 7-[D_n] was isolated with 11.4% deuterium incorporation on C-α and 17.3% D on C-β. This result clearly ruled-out the gold-carbene C–H insertion path as initially proposed in Scheme 2A. It is likely that the observed partial deuteration at C-α and C-β is from C–Au protodeauration. With non-protic amide co-catalyst DMF (Eq. 2), significantly higher percentage of deuteration were observed, providing strong evidence for the C–Au protodeauration pathway. With the CD₃ substrate available, the kinetic isotope effect (KIE) was evaluated using substrates 6a and 6a’. The calculated KIE (Scheme 3B, Eq. 3) was 0.86±0.02. The observed inverse KIE provided critical information on reaction mechanism by revealing sp² to sp³ C–H transformation involving in the rate determining step (RDS). Based on these control experiments and KIE results, a plausible mechanism involving amide covalent nucleophilic catalysis process is proposed, as shown in Scheme 3C.

Firstly, cyclopropane gold carbene intermediate A is formed through 1,6-enyne cyclization. The amide serves as effective nucleophilic catalyst for the cyclopropane ring opening followed by elimination, giving the skipped-diene C. The inverse KIE indicated the RDS likely happened at the transformation of intermediate C to D. Alkene addition to gold activated enyne followed by protodeauration and elimination/aromatization gave the resulting tetra-substituted benzene. Although the function of amide (N_cat.) in second alkyne cyclization remains uncertain at this moment, the amide-promoted gold-carbene-cyclopropane ring opening clearly plays a critical role in facilitating this transformation under the combined gold/amide dual catalysis.

Grams-scale synthesis (Scheme 3D) and function group transformations (Scheme 3E) were also performed to demonstrate the practicality of this new method. The intrinsic alkene and tosyl amide could be readily applied into the synthesis of other heterocycles through simple transformations, which, again, highlighted the power of this method in incorporating diverse functionality into benzene moiety in one step.

In summary, reported herein is the first example of gold and covalent nucleophilic dual catalysis. With the critical amide co-catalyst, the high value polysubstituted benzenes could be prepared from readily available diyne-ene starting materials through a simple formal [3+3] annulation in one step. The modular approach, mild reaction conditions, high efficiency, and good functional group tolerability make this method one attractive new approach in producing polysubstituted benzenes. The combination of gold chemistry and covalent nucleophilic catalysis opens opportunity to achieve challenging molecule synthesis. The potential applications of this new method in peptides, natural products, and drugs derivatives syntheses are expected in the future.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.