Ashok Kumar Raigar,
Manju
,
Kamlesh Saini
and
Anjali Guleria
*
Department of Chemistry, University of Rajasthan, Jaipur 302004, India. E-mail: dranjaliguleria@uniraj.ac.in
First published on 5th September 2025
A one-pot strategy was developed for the synthesis of α-substituted 2-benzofuranmethamines from salicylaldehydes, phenylacetylenes, and cyclic secondary amines using CuFe2O4 as a bifunctional catalyst. The reaction proceeds at 80 °C in 1,4-dioxane using Cs2CO3 as a base, enabling sequential A3-coupling, 5-exo-dig cyclization, and 1,3-allylic rearrangement in a single operation. Unlike previous methods, this protocol employs non-precious metal catalysts and mild reagents, operates under moderate conditions, and provides direct access to α-substituted 2-benzofuranmethamines in good yields (80–96%) with broad substrate compatibility. Furthermore, the catalyst is magnetically recoverable and exhibits excellent reusability over five consecutive cycles without significant loss of activity.
![]() | ||
Fig. 1 Multistep synthesis of α-aryl-2-benzofuranmethamines (5) via carbinol intermediates (c) and corresponding chlorides (d), as reported by Pestellini et al. (US Pat. 4485112, 1984). |
In subsequent years, McLean et al.5 and Nun et al.6 independently reported the synthesis of a single α-aryl-2-benzofuranmethamine, 4-(2-benzofuranylphenylmethyl)morpholine, via the Petasis boronic acid Mannich reaction under microwave irradiation, with yields ranging from 23%5 to 95%.6 While their studies broadly focused on microwave-assisted Petasis chemistry and produced various 2-hydroxyaryl derivatives, only one compound bearing this scaffold was reported. Nonetheless, the protocols required high temperatures and energy input, and in the case of McLean et al., delivered modest yields (Fig. 2).
![]() | ||
Fig. 2 Comparison of more recent catalytic strategies and the present CuFe2O4-catalyzed one-pot protocol for synthesizing α-substituted 2-benzofuranmethamines. |
A significant advancement was later reported by Wongsa et al.,7 who developed a two-step protocol for synthesizing 5 using salicylaldehyde, phenylacetylene, and cyclic/acyclic secondary amines as starting materials. The method involved an initial AgNO3-catalyzed 5-exo-dig cycloisomerization of the A3-coupling product 4 to afford 2-arylmethylidenebenzofurans (4′), followed by a second step, an AgNO3-promoted 1,3-allylic rearrangement under Lewis acidic conditions, conducted at 100 °C over 18 hours to deliver the target compound 5 (Fig. 2). While the protocol successfully provided access to the desired scaffold in moderate to good yields, it was limited by its multistep nature, high temperature, extended reaction times, formation of by-products, and reliance on a costly silver-based catalyst.
Encouraged by the findings of Wongsa et al.,7 we sought to develop a more sustainable catalytic system that addresses the key operational and environmental limitations of previous methods. Our attention turned to several parallel studies that described the synthesis of either 2-aryl(alkyl)methylidenebenzofurans (4′) or 2-aryl(alkyl)-3-aminobenzofurans (6) through copper-catalyzed processes (Fig. 3).
![]() | ||
Fig. 3 Previous studies demonstrating the role of copper catalysis in the synthesis of benzofuran derivatives (4′ and 6) through tandem cyclization and isomerization pathways. |
Ukhin et al.8 reported the synthesis of 4′ from the propargylamine intermediate A, prepared via condensation of salicylaldehyde with secondary amines and propargyl alcohols. Further cyclization using either AgNO3 or CuI (10 mol%) in MeCN under reflux (15–40 min) gave products 4′ in 39–74% yield. These findings were supported by Nguyen et al.,9 who demonstrated that microwave-assisted treatment of salicylaldehyde with cyclic amines and propargyl alcohols or alkynes bearing remote hydroxyl groups, in the presence of CuI (5 mol%) at 130 °C for 30 min, afforded 4′ in 44–88% yield. Conditions employing AgCl (5 mol%) in MeCN at 80 °C for 16 h were also effective, though they gave slightly lower yields. Interestingly, 3,3-dimethylbut-1-yne failed to cyclize, suggesting that the hydroxyl group in the alkyne plays a key role in metal-assisted coordination during the cyclization (intermediate A).
Sakai et al.10 advanced the method by eliminating the need for a propargylic hydroxyl group. They employed salicylaldehyde (1.5 equiv.), a secondary amine (1 equiv.), and 1-TMS-alkynes (1.5 equiv.), with Cu(OTf)2 (5 mol%), CuCl (5 mol%), and DMAP (1 equiv.) in MeCN at reflux for 6 h. Their results indicated that 4′, though not isolated, underwent base-induced isomerization to 6 in variable yields (22–99%). Later, a related study by Li et al.11 further supported this transformation using salicylaldehyde (2 equiv.), a secondary amine (1 equiv.), and a terminal alkyne (1.5 equiv.) in the presence of CuI (20 mol%), K2CO3 (1 equiv.), and Bu4NBr (1 equiv.) in toluene at 110 °C for 2–3 h. Notably, in the absence of a base, the propargylamine intermediate 4 was isolated in 84% yield. Moreover, use of 1-octyne gave a separable mixture of 4′ and 6 (∼1:
1), suggesting that under less basic or milder conditions, compounds with the general structure 4′ may be stabilized or isolated more selectively. These findings influenced our decision to conduct the reaction under basic conditions to suppress the formation of 4′ as the predominant product.
Zhang et al.12 later reported that related cyclizations to afford 6 could be performed using salicylaldehyde, a secondary amine (1.2 equiv.), and a terminal alkyne (1.5 equiv.) with CuI (10 mol%) and [bmim]OAc (20 mol%) in [bmim]PF6 at 80 °C for 6–9 h. Additionally, several other cyclization reactions of propargyl alcohol analogues of 4 (R6H or aryl) were reported by Harkat et al. and several other research groups13 to yield 3-hydroxy-2,3-dihydro-2-arylmethylidenebenzofurans (analogues of 4′), which could undergo rearrangement to 2-hydroxymethylbenzofurans or 2-alkoxybenzofurans under acidic or alcoholic conditions.
Together, these studies highlighted the versatility of copper-based systems in promoting both cyclization and rearrangement processes relevant to our scaffold. Thus, building on these mechanistic insights and the established role of Lewis acids in promoting 1,3-allylic rearrangements, we hypothesized that CuFe2O4, a mixed-metal oxide containing both Cu2+ and Fe3+ centers, could serve as an effective dual-function catalyst.14 These metal centers are known to exhibit Lewis acidity, which could promote not only the initial A3-coupling and cyclization but also the subsequent 1,3-allylic rearrangement. Furthermore, CuFe2O4 offers practical advantages such as low cost, magnetic recoverability, excellent reusability for up to five cycles without significant loss of activity, and overall environmental compatibility. The operational and environmental advantages of our protocol compared to previous strategies are illustrated in Fig. 2
As a Brønsted base, Cs2CO3 was selected due to its reliable solubility in polar aprotic solvents and its common use in multicomponent and rearrangement reactions. Its use allowed us to avoid strongly basic or nucleophilic additives that might interfere with Lewis acidic centers or promote undesired side reactions. Finally, 1,4-dioxane was chosen as the reaction solvent owing to its high boiling point, chemical stability, and ability to solubilize both organic and inorganic components, enabling efficient progression of the reaction under mild thermal conditions.
Thus, we developed a one-pot transformation that integrates the key steps of A3-coupling, cycloisomerization, and 1,3-allylic rearrangement within a single catalytic system. CuFe2O4 was employed as a bifunctional catalyst, with surface Cu2+ and Fe3+ centers capable of activating alkynes, stabilizing intermediates, and promoting the Lewis acid-driven 1,3-allylic rearrangement. Cs2CO3 was chosen as a mild, soluble base compatible with the catalytic system, and the reaction was performed in 1,4-dioxane, a thermally robust and chemically suitable solvent. This approach addresses the synthetic and environmental limitations of previous methods by streamlining the process, minimizing manipulations, and eliminating the need for precious metal catalysts, harsh reagents, and elevated temperatures. In addition to operational simplicity, the protocol delivers good yields across a broad substrate scope and employs a magnetically separable catalyst that retains high activity over five consecutive cycles without significant loss of efficiency. Hence, the work presented herein demonstrates the utility of this strategy, delineates its scope, and evaluates its potential as a more sustainable and broadly applicable method for the synthesis of α-substituted 2-benzofuranmethamines.
Entry | Catalyst (%) | Solvent | Temp. (°C) | Time (min) | Yieldb % |
---|---|---|---|---|---|
a Reaction conditions: salicylaldehyde (1.0 mmol, 0.12 mL), phenylacetylene (1.1 mmol, 0.14 mL), 3-methylpiperidine (1.3 mmol, 0.12 mL), cesium carbonate (0.31 mmol, 0.10 g), 1,4-dioxane (5 mL), and the required amount of the catalyst were treated at the indicated temperatures.b Isolated yields after column chromatography.c Reactions did not reach completion within the indicated times, and extended durations up to 600 minutes yielded no further conversion, as confirmed by TLC using ethyl acetate/hexane (2![]() ![]() |
|||||
1 | — | — | RT, reflux | 600 | No reaction |
2 | — | H2O, EtOH, H2O/EtOH(1![]() ![]() |
RT, reflux | 600 | No reaction |
3 | 2 | H2O, EtOH, H2O/EtOH(1![]() ![]() |
RT, reflux | 180 | Traces |
4 | 2 | Toluene | RT, reflux | 180 | Traces, 22c |
5 | 2 | DMSO | RT, reflux | 180 | Traces, 26c |
6 | 2 | 1,4-Dioxane | RT | 180 | 12c |
7 | 2 | 1,4-Dioxane | 50, 70 | 180 | 42c, 51c |
8 | 4.4 | 1,4-Dioxane | RT, 50 | 180 | 36c, 66c |
9 | 4.4 | 1,4-Dioxane | 80 | 60 | 68c |
10 | 4.4 | 1,4-Dioxane | 80 | 120 | 78c |
11 | 4.4 | 1,4-Dioxane | 80 | 180 | 97, 94d |
12 | 4.4 | 1,4-Dioxane | Reflux | 180 | 97 |
13 | 4.4 | 1,4-Dioxane | 80 | 240 | 97 |
14 | 8.8 | 1,4-Dioxane | 80 | 180 | 97 |
15 | 15 | 1,4-Dioxane | 80 | 180 | 97 |
In the absence of both catalyst and solvent, no reaction occurred, even after prolonged heating at room temperature (RT) or reflux (entry 1). Similarly, conducting the reaction in water, ethanol, or their mixtures (1:
1 v/v, reflux temperature = 78.15 °C) without a catalyst resulted in no product formation (entry 2), indicating that neither polar protic media nor thermal activation alone were sufficient to promote the transformation.
When 2 mol% CuFe2O4 was introduced under the same solvent systems (water, ethanol, or EtOH/H2O mixtures), only trace product formation was observed after 180 minutes, regardless of temperature (entry 3). This demonstrated that these protic environments are unsuitable for catalysis, likely due to poor solubility or deactivation of the catalytic surface.
The solvent effect was further examined using organic media. In toluene and DMSO, only trace formation was observed at RT; however, upon heating to reflux, yields improved modestly to 22% and 26%, respectively (entries 4 and 5). These results suggested that while nonpolar (toluene) and polar aprotic (DMSO) solvents could partially enable the reaction under forcing conditions, the conversion remained incomplete, likely due to limited solubility of the base or suboptimal stabilization of reactive intermediates.
1,4-Dioxane proved superior in promoting the reaction. At RT with 2 mol% catalyst, a 12% yield was observed (entry 6), which increased significantly to 42% at 50 °C and 51% at 70 °C under identical conditions (entry 7). This established both solvent identity and temperature as key variables. Increasing the catalyst loading to 4.4 mol% further improved the performance, affording 36% yield at RT and 66% at 50 °C (entry 8).
To refine the protocol further, the reaction was examined at 80 °C. At this temperature and 4.4 mol% catalyst loading, a 68% yield was obtained after 60 minutes (entry 9), increasing to 78% at 120 minutes (entry 10), and reaching full conversion (97%) at 180 minutes (entry 11). Interestingly, conducting the same reaction under an inert nitrogen atmosphere using dried 1,4-dioxane gave a slightly lower yield of 94%, suggesting that ambient conditions may subtly influence catalyst surface activity or solvation effects (entry 11, footnote d).
Neither prolonging the reaction time to 240 minutes nor raising the temperature to reflux further improved the yield, which remained at 97% (entries 12 and 13). These observations confirmed that 80 °C for 180 minutes is the most efficient condition. Increasing the catalyst loading to 8.8 and 15 mol% (entries 14 and 15) also showed no additional benefit, demonstrating that 4.4 mol% CuFe2O4 is sufficient to achieve maximum conversion.
A control reaction performed under the optimized conditions in the absence of Cs2CO3 revealed that the primary product was the A3-coupling intermediate, 1-(phenyl(2-hydroxybenzyl))-3-methylpiperidine (4a), with negligible formation of the target product. This finding highlights the crucial role of the base in enabling the 5-exo-dig cyclization that leads to subsequent 1,3-allylic rearrangement.
In conclusion, the optimized conditions for this transformation were established as 4.4 mol% CuFe2O4 and 0.31 mmol Cs2CO3 in 1,4-dioxane at 80 °C for 180 minutes, affording the desired product 5a in excellent isolated yield (97%).
![]() | ||
Scheme 2 A plausible mechanism for the CuFe2O4 nanoparticles catalyzed synthesis of α-aryl-2-benzofuranmethamines (5a–h). |
The resulting intermediate (4) then undergoes a 5-exo-dig cyclization, aided by Cs2CO3-mediated deprotonation of the phenolic hydroxyl. At the same time, CuFe2O4 may assist the process through a metal coordination interaction with the alkyne, helping to align the reacting centers for smooth ring closure.
This gives rise to intermediate (4′), presumed to be a benzofuranylmethylidene species, which subsequently undergoes a 1,3-allylic rearrangement to yield the final product (5). As expected in a one-pot protocol, intermediate 4′ was not isolated, which aligns with previous reports where similar species were formed transiently and not separated.10–12 In this rearrangement step, CuFe2O4 might act as a Lewis acid, owing to its Cu2+ and Fe3+ surface sites, facilitating conversion to the final product through transition state II. The basic environment provided by Cs2CO3 likely plays a supporting role in enabling charge reorganization. Altogether, this process integrates multicomponent coupling, cyclization, and rearrangement into a single operational sequence under mild conditions using a recyclable and magnetically recoverable catalyst.
The catalyst consistently delivered high yields with minimal change in reaction time across all cycles, indicating excellent retention of catalytic activity (Fig. 4). After the fifth run, 9.2 mg of catalyst was recovered from the initial 10 mg, corresponding to 92% mass recovery. The modest loss (∼8%) is likely attributable to mechanical handling during separation and washing rather than structural degradation.
SEM images (Fig. 5a and b) of both the fresh and recycled CuFe2O4 catalyst show the presence of nanosized spherical domains distributed across the surface of larger particulate structures. The larger underlying features likely reflect agglomerated or sintered CuFe2O4 material formed during synthesis or subsequent thermal treatments.
Importantly, EDX analysis (Fig. 5c–f) confirms homogeneous elemental distribution (Cu, Fe, O) across both small and large domains, suggesting the absence of any extraneous support or carrier material. This indicates that both morphological features belong to the same CuFe2O4 phase, differing only in the degree of aggregation or surface roughness. The preservation of spherical domains and consistent elemental composition supports the physical robustness and chemical homogeneity of the catalyst under the applied reaction conditions.
PXRD patterns (Fig. 5g–i) confirm the crystalline spinel structure of CuFe2O4 in both fresh and recycled catalysts. The fresh catalyst shows characteristic diffraction peaks at 2θ ≈ 18.9° (111), 30.2° (220), 35.4° (311), 37.6° (222), 43.1° (400), 53.6° (422), 57.6° (511), 62.5° (440), and 74.7° (533), in good agreement with the standard pattern for cubic CuFe2O4 (JCPDS no. 34-0425). After five catalytic cycles, all major peaks are preserved in the recycled sample, indicating that the spinel structure remains intact without phase decomposition. A slight broadening and reduction in peak intensities are observed in the recycled sample (Fig. 5h), which may result from surface-level modifications, minor agglomeration, or decreased long-range ordering associated with repeated heating and magnetic separation. In the recycled sample, two additional weak reflections appear at 2θ ≈ 23.10° and 27.08°, which do not correspond to CuFe2O4. These peaks likely originate from carbonaceous residues formed during reaction cycles involving organic substrates, and their presence does not impact the integrity of the ferrite phase. Upon calcination of the recycled catalyst at 400 °C for 2 h, these low-angle peaks disappear, confirming the thermal removal of residual organic deposits. However, new weak reflections emerge at 2θ ≈ 48.09°, 66.02°, and 68.21°, which are not present in the fresh catalyst and are attributed to minor surface oxidation or phase restructuring during calcination, possibly due to the formation of low-percentage CuO or spinel-related secondary domains. The peak at 48.09° may correspond to the (200) reflection of monoclinic CuO, while those at 66.02° and 68.21° can be linked to higher-order reflections of surface-generated copper oxide phases. These changes are limited to surface regions and do not indicate any structural decomposition of the core CuFe2O4 phase. However, these differences are minimal and do not imply any significant compromise in crystallinity or performance.
The crystallite size of the fresh and recycled catalyst was calculated using the Debye–Scherrer equation:
D = (k × λ)/(β![]() ![]() |
β = FWHM (°) × π/180 |
For the recycled catalyst, the corresponding reflections appeared at 2θ ≈ 35.66° and 61.76°, giving crystallite sizes of 10.59 nm and 9.25 nm, respectively, with an average value of 9.92 ± 0.67 nm. The increase in crystallite size after calcination and reuse suggests partial growth or sintering of crystallites during thermal and catalytic cycling. The crystallite size data for both samples is presented in Table 2.
Diffraction plane | 2θ (°) | θ (°) | FWHM (radians) | Crystallite size (nm) | ||||
---|---|---|---|---|---|---|---|---|
Fresh | Recycled | Fresh | Recycled | Fresh | Recycled | Fresh | Recycled | |
(311) | 35.46 | 35.66 | 17.73429 | 17.83 | 0.03501 | 0.01436 | 4.34 | 10.59 |
(440) | 62.57 | 61.76 | 31.28565 | 30.88 | 0.03461 | 0.01825 | 4.90 | 9.25 |
Average | 4.62 ± 0.28 | 9.92 ± 0.67 |
Together, the SEM and PXRD analyses demonstrate that CuFe2O4 retains its structural integrity and morphology after prolonged use, reinforcing its suitability as a robust and recyclable heterogeneous catalyst for this transformation.
The reaction mixture was then concentrated under reduced pressure, and the crude residue was purified by column chromatography using ethyl acetate/hexane (1:
9) as the eluent to afford the corresponding α-aryl-2-benzofuranmethamine derivatives (5a–h) as pure products.
This journal is © The Royal Society of Chemistry 2025 |