Multi-Component Syntheses of Spiro[furan-2,3′-indoline]-3-carboxylate Derivatives Using Ionic Liquid Catalysts

Two previously described Brønsted acidic ionic liquids, 3,3′-(1,6-hexanediyl)bis(1-methyl)-1H-imidazolium hydrogen sulfate (Cat1) and 1,1′-(1,6-hexanediyl)bis(pyridinium) hydrogen sulfate (Cat2), were used as catalysts for the preparation of spiro[furan-2,3′-indoline]-3-carboxylate derivatives via a three-component reaction of anilines, isatins (N-alkyl-indoline-2,3-diones), and diethyl acetylenedicarboxylate, in high yields. The use of ultrasonic (US) irradiation led to the targeted products (1a–15a) in high yields ranging from 80% to 98%. Under the same conditions, the use of sulfuric acid and acetic acid as a Brønstedt catalyst did not yield the desired benchmark product 1a.

In continuation of our recent work using a heterogeneous sulfonic acid polyvinyl pyridinium IL catalyst for multi-component synthesis of spiroindole-3,5-pyrano [2,3-d]pyrimidines and pyrazines [41], we report herein the synthesis of spiro[furan-2,3′-indoline]-3-carboxylate derivatives using the two easy-to-make and versatile homogeneous Brønsted-acidic IL catalysts Cat1 and Cat2 in three-component reactions using isatin (Nalkyl-indoline-2,3-dione) derivatives, anilines, and diethyl acetylenedicarboxylate (Scheme 2).The use of ultrasonic irradiation was motivated by our own experience with such multi-component reactions suffering from limited miscibility of the components due In continuation of our recent work using a heterogeneous sulfonic acid polyvinyl pyridinium IL catalyst for multi-component synthesis of spiroindole-3,5-pyrano [2,3-d]pyrimidines and pyrazines [41], we report herein the synthesis of spiro[furan-2,3 ′ -indoline]-3-carboxylate derivatives using the two easy-to-make and versatile homogeneous Brønsted-acidic IL catalysts Cat1 and Cat2 in three-component reactions using isatin (N-alkyl-indoline-2,3dione) derivatives, anilines, and diethyl acetylenedicarboxylate (Scheme 2).The use of ultrasonic irradiation was motivated by our own experience with such multi-component reactions suffering from limited miscibility of the components due to their different hydrophilic/hydrophobic properties and the wish to use only one solvent.

Optimization of the Reaction Conditions
First, the synthesis of 1a from the reaction of 1-ethylindoline-2,3-dione (1 mmol), aniline (1 mmol), and dimethylacetylenedicarboxylate (1 mmol) was studied under ultrasonic (US) irradiation, varying solvent, catalyst amounts, and temperature (Scheme 3).Without the catalyst and when the reaction was stirred at ambient temperature (r.t.) for 24 h, no product was obtained (Table 1).Similarly, when the reaction was heated at 80 • C under the solvent-free condition the substrates remain unreacted even after prolonged heating (24 h).Next the reaction was investigated under reflux conditions, and we found no product formation below 60 • C; among the solvents, EtOH was found to be suitable.Finally, we found that 30 mol% of the ionic liquid catalyst shows optimum performance with yields of 96% or 88% using Cat1 or Cat2, respectively (Table 1).
(US) irradiation, varying solvent, catalyst amounts, and temperature (Scheme 3).Without the catalyst and when the reaction was stirred at ambient temperature (r.t.) for 24 h, no product was obtained (Table 1).Similarly, when the reaction was heated at 80 °C under the solvent-free condition the substrates remain unreacted even after prolonged heating (24 h).Next the reaction was investigated under reflux conditions, and we found no product formation below 60 °C; among the solvents, EtOH was found to be suitable.Finally, we found that 30 mol% of the ionic liquid catalyst shows optimum performance with yields of 96% or 88% using Cat1 or Cat2, respectively (Table 1).
Under the same conditions (EtOH, US, Reflux for 3 h), we also attempted the synthesis of 1a using 30% acetic acid (HOAc) or sulfuric acid (H2SO4), but failed to detect the product in the reaction mixtures or after workup, although full conversion of the starting materials was achieved, as 1 H NMR spectra unequivocally showed.In particular, the Nethyl, O-ethyl, and the NH functions of the starting materials and product were indicative of this.Instead of the desired product, we obtained mixtures of unidentifiable products (Figures S31-S34, Supplementary Materials).
Table 1.Optimization of the reaction conditions for the preparation of 1a a .

Conditions Cat1 Cat2
Time (h)/Yield (%) Time (h)/Yield (%) EtOH; US; Cat.: none; T ( Under the same conditions (EtOH, US, Reflux for 3 h), we also attempted the synthesis of 1a using 30% acetic acid (HOAc) or sulfuric acid (H 2 SO 4 ), but failed to detect the product in the reaction mixtures or after workup, although full conversion of the starting materials was achieved, as 1 H NMR spectra unequivocally showed.In particular, the N-ethyl, O-ethyl, and the NH functions of the starting materials and product were indicative of this.Instead of the desired product, we obtained mixtures of unidentifiable products (Figures S31-S34, Supplementary Materials).
Other comparable [2+2+1] three-component reactions reported the use of aniline derivatives, dialkyl acetylenedicarboxylates and aromatic aldehydes to produce 3,4,5substituted furan-2(5H)-one derivatives.Using (Bu 4 N)(HSO 4 ) as homogeneous catalyst allowed reactions in EtOH at ambient T with up to 93% yields [35], which stands in contrast to our attempts to produce 1a at ambient T, resulting in no product.However, the amount of catalyst used was not provided in this report.Elevated T of 45 • C in CH 2 Cl 2 were necessary when using 2-pyrrolidonium hydrogen sulfate ((HNMP)(HSO 4 )) as catalyst (40 mol%) for similar products, which were obtained in up to 98% yield [36].The best conditions for similar reactions using Al(HSO 4 ) 3 (31 mol%) as catalyst were EtOAc as solvent, 8 h at ambient T with 78% yields, whereas in EtOH the yield dropped to 51% [34].A number of heterogeneous Lewis acidic catalysts have shown similar performance with maximum yields ranging from 75% to 97% and reaction times from 1 to 12 h, at elevated temperatures of up to 100 • C [26][27][28][29][30][31][32][33].Within this frame, our catalyst systems perform on a good-to-excellent level.

Mechanistic Considerations
Mechanistically, the reaction is supposed to start via the sulfonation of the triple bond of the acetylenic ester and the subsequent replacement of the OSO 3 group by the amine.Thus, diethyl 2-(phenylamino)fumarate is obtained as the intermediate (A).At the same time, 1-ethylindoline-2,3-dione is activated by ionic liquid through protonation at the 3-oxo group (B).The nucleophilic reaction of diethyl 2-(phenylamino)fumarate (A) with the protonated 1-ethylindoline-2,3-dione (B) gives the intermediate (C) which undergoes a cyclization reaction and elimination of EtOH (Scheme 4), while the catalyst is recycled.
For the three-component [2+2+1] cycloaddition reactions using diethyl acetylenedicarboxylate, N-substituted isatins, and 2,2-dimethyl-4,4-dimethyl-butane isonitrile, the attack of the isonitrile C atom to the triple bond in a Michael-type addition was proposed [42,43] and for the reaction of PPh 3 with diethyl acetylenedicarboxylate and N-alkylisatins a dipolar intermediate resulting from the addition of the nucleophilic PPh 3 to the triple bond was suggested [44].In the following steps, a nucleophilic addition of the carbonyl (isatin) leads to intermediates being able to close the butyrolactone ring.
Importantly, both mechanisms are proposed for non-catalyzed reactions and their reaction mixtures both contain strong nucleophiles (isonitrile or PPh 3 ).We have considered, alternatively to Scheme 4, that the isatin attacks the triple bond.However, in the abovementioned mechanistic studies, this was not an option.Further, our failure to use the simple Brønstedt acids HOAc or H 2 SO 4 as catalysts suggests that the IL-character of our catalysts, especially the ionic forces between the cationic pyridinium and imidazolium moieties and the anionic HSO 4 − , under the reaction conditions, make hydrogen sulfate a suitable nucleophile to attack the triple bond.However, the mechanistic details remain to be supported by studying the proposed sulfonated acetylenic esters and their reactivity towards amines, as well as by quantum chemical calculations.This will be the topic of a future study.
moieties and the anionic HSO4 − , under the reaction conditions, make hydrogen sulfate a suitable nucleophile to attack the triple bond.However, the mechanistic details remain to be supported by studying the proposed sulfonated acetylenic esters and their reactivity towards amines, as well as by quantum chemical calculations.This will be the topic of a future study.

Studying the Substrate Scope
Using the optimized conditions (EtOH, reflux, US, 3 h), the catalytic potential of ionic liquid catalysts on the preparation of further furan-2(5H)-ones was examined applying various anilines (Scheme 5).For products 1a-8a, starting from 1-ethylindoline-2,3-dione, the yields ranged from 85% to 98% (Table 2).Relatively low yields and long reaction times were observed for 4-chloroaniline, which might be due to the electron-withdrawing nature of the chloride substituent.However, the reactions using 1-benzylindoline-2,3-dione (products 9a to 11a) or 5-chloroisatine (1-benzyl-5-chloro-indole-2,3-dione) (products 12a to 15a) did not show such a dependence.The same substrates were also studied using Cat2, and high yields were obtained (Table 2).Generally, the yields for Cat2 lie slightly lower compared with the results for Cat1, which is in line with the results obtained already for the optimization of the synthesis of 1a (Table 1).We ascribe this to the higher acidity of the HSO4 − moieties in Cat1 compared with Cat2, which is in line with the protonation step in our proposed mechanism (Scheme 4).

Studying the Substrate Scope
Using the optimized conditions (EtOH, reflux, US, 3 h), the catalytic potential of ionic liquid catalysts on the preparation of further furan-2(5H)-ones was examined applying various anilines (Scheme 5).For products 1a-8a, starting from 1-ethylindoline-2,3-dione, the yields ranged from 85% to 98% (Table 2).Relatively low yields and long reaction times were observed for 4-chloroaniline, which might be due to the electron-withdrawing nature of the chloride substituent.However, the reactions using 1-benzylindoline-2,3-dione (products 9a to 11a) or 5-chloroisatine (1-benzyl-5-chloro-indole-2,3-dione) (products 12a to 15a) did not show such a dependence.The same substrates were also studied using Cat2, and high yields were obtained (Table 2).Generally, the yields for Cat2 lie slightly lower compared with the results for Cat1, which is in line with the results obtained already for the optimization of the synthesis of 1a (Table 1).We ascribe this to the higher acidity of the HSO 4 − moieties in Cat1 compared with Cat2, which is in line with the protonation step in our proposed mechanism (Scheme 4).
be supported by studying the proposed sulfonated acetylenic esters and their reactivity towards amines, as well as by quantum chemical calculations.This will be the topic of a future study.

Studying the Substrate Scope
Using the optimized conditions (EtOH, reflux, US, 3 h), the catalytic potential of ionic liquid catalysts on the preparation of further furan-2(5H)-ones was examined applying various anilines (Scheme 5).For products 1a-8a, starting from 1-ethylindoline-2,3-dione, the yields ranged from 85% to 98% (Table 2).Relatively low yields and long reaction times were observed for 4-chloroaniline, which might be due to the electron-withdrawing nature of the chloride substituent.However, the reactions using 1-benzylindoline-2,3-dione (products 9a to 11a) or 5-chloroisatine (1-benzyl-5-chloro-indole-2,3-dione) (products 12a to 15a) did not show such a dependence.The same substrates were also studied using Cat2, and high yields were obtained (Table 2).Generally, the yields for Cat2 lie slightly lower compared with the results for Cat1, which is in line with the results obtained already for the optimization of the synthesis of 1a (Table 1).We ascribe this to the higher acidity of the HSO4 − moieties in Cat1 compared with Cat2, which is in line with the protonation step in our proposed mechanism (Scheme 4).In summary, the procedures proved to be effective in achieving a broad scope of 2 ′ ,5dioxo-5H-spiro[furan-2,3 ′ -indoline]-3-carboxylate derivatives in good-to-excellent yields.A simple filtration procedure allowed for the isolation of the pure products.In summary, the procedures proved to be effective in achieving a broad scope of 2′,5dioxo-5H-spiro[furan-2,3′-indoline]-3-carboxylate derivatives in good-to-excellent yields.A simple filtration procedure allowed for the isolation of the pure products.
In summary, the procedures proved to be effective in achieving a broad scope of 2′,5dioxo-5H-spiro[furan-2,3′-indoline]-3-carboxylate derivatives in good-to-excellent yields.A simple filtration procedure allowed for the isolation of the pure products.

Conclusions
A simple synthetic method for the preparation of 2′,5-dioxo-5H-spiro[furan-2,3′-indoline]-3-carboxylate derivatives was described, which provides short reaction times and good yields using the ionic liquid catalysts 3,3′-(1,6-hexanediyl)bis(1-methyl)-1H-imidazolium hydrogen sulfate and 1,1′-(1,6-hexanediyl)bis(pyridinium) hydrogen sulfate.Importantly, the use of acidic acid (HOAc) or sulfuric acid (H2SO4) as catalysts under the same conditions failed.The IL catalysts are easy to prepare, cheap, can be easily separated, and showed high levels of catalytic activity for the three-component reaction of anilines, isatins (N-alkyl-indoline-2,3-diones), and diethyl acetylene dicarboxylate in EtOH under reflux, using ultrasonic (US) irradiation.The method is superior to non-catalyzed procedures, having far shorter reaction times and higher yields.Due to the complete conversion, only minute amounts of polymeric by-products were obtained, which can be  In summary, the procedures proved to be effective in achieving a broad scope of 2′,5dioxo-5H-spiro[furan-2,3′-indoline]-3-carboxylate derivatives in good-to-excellent yields.A simple filtration procedure allowed for the isolation of the pure products.

Conclusions
A simple synthetic method for the preparation of 2′,5-dioxo-5H-spiro[furan-2,3′-indoline]-3-carboxylate derivatives was described, which provides short reaction times and good yields using the ionic liquid catalysts 3,3′-(1,6-hexanediyl)bis(1-methyl)-1H-imidazolium hydrogen sulfate and 1,1′-(1,6-hexanediyl)bis(pyridinium) hydrogen sulfate.Importantly, the use of acidic acid (HOAc) or sulfuric acid (H2SO4) as catalysts under the same conditions failed.The IL catalysts are easy to prepare, cheap, can be easily separated, and showed high levels of catalytic activity for the three-component reaction of anilines, isatins (N-alkyl-indoline-2,3-diones), and diethyl acetylene dicarboxylate in EtOH under reflux, using ultrasonic (US) irradiation.The method is superior to non-catalyzed procedures, having far shorter reaction times and higher yields.Due to the complete conversion, only minute amounts of polymeric by-products were obtained, which can be

Conclusions
A simple synthetic method for the preparation of 2 ′ ,5-dioxo-5H-spiro[furan-2,3 ′indoline]-3-carboxylate derivatives was described, which provides short reaction times and good yields using the ionic liquid catalysts 3,3 ′ -(1,6-hexanediyl)bis(1-methyl)-1Himidazolium hydrogen sulfate and 1,1 ′ -(1,6-hexanediyl)bis(pyridinium) hydrogen sulfate.Importantly, the use of acidic acid (HOAc) or sulfuric acid (H 2 SO 4 ) as catalysts under the same conditions failed.The IL catalysts are easy to prepare, cheap, can be easily separated, and showed high levels of catalytic activity for the three-component reaction of anilines, isatins (N-alkyl-indoline-2,3-diones), and diethyl acetylene dicarboxylate in EtOH under reflux, using ultrasonic (US) irradiation.The method is superior to non-catalyzed procedures, having far shorter reaction times and higher yields.Due to the complete conversion, only minute amounts of polymeric by-products were obtained, which can be separated together with the catalyst through centrifugation and filtration.Amongst other so far reported homogeneous acid catalysts, the separation of our systems is more convenient and does not require chromatographic separation of targeted products.Compared with heterogeneous acid catalysts, our systems are more reactive, and the simple handling makes our IL catalyst systems very interesting for further reactions catalyzed by Brønsted acids.

Instrumentation
1 H and 13 C NMR spectra were recorded on a Bruker Avance DPX 500 MHz instrument (Bruker, Rheinhausen, Germany).Chemical shifts δ (in ppm) were given vs. the tetramethylsilane standard.A Heraeus CHN-O-Rapid analyzer (Heraeus, Darmstadt, Germany) was used for elemental analysis.Melting points were recorded on an Electrothermal IA9100 melting point apparatus (Electrothermal Ltd., London, UK) and are uncorrected.Thin-layer chromatography (TLC) was carried out using silica plates (Merck, Darmstadt, Germany) as the solid support and n-hexane/ethyl acetate (9/1) or c-hexane/ethyl acetate (2/1) as the eluting solvent.

General Procedure for the Syntheses
In a 50 mL flask equipped with a condenser, the ionic liquid catalyst (30 mol%) was added to a mixture of amine (1 mmol), diethyl acetylene dicarboxylate (1 mmol), and 1-ethylindoline-2,3-dione (1 mmol) in 20 mL EtOH under stirring.The mixture was irradiated using an ultrasonic bath at 80 • C. The progress of the reaction was monitored by thin-layer chromatography (TLC).After the reaction was completed, the solvent was evaporated to dryness and the residual solid was suspended in 5 mL EtOH and cooled.After centrifugation, the suspension was carefully poured through a 2 cm thick silica sludge deposited on a glass filter which collected the precipitated catalyst and polymeric by-products (yellow to brown sticky oil), while the pure products were obtained from the filtrate.

Table 1 .
Optimization of the reaction conditions for the preparation of 1a a .
b Isolated yields.