Xue-Ju
Fu†
a,
Ting
Qi†
b,
Xin
Zhang†
a,
Ke
Xu†
a,
Wei
Huang
a,
Cheng
Peng
a,
Jun-Long
Li
b,
Bo
Han
*a and
Gu
Zhan
*a
aSchool of Pharmacy, Hospital of Chengdu University of Traditional Chinese Medicine, Chengdu University of Traditional Chinese Medicine, Chengdu 611137, China. E-mail: hanbo@cdutcm.edu.cn; zhangu@cdutcm.edu.cn
bAnti-Infective Agent Creation Engineering Research Centre of Sichuan Province, Sichuan Industrial Institute of Antibiotics, School of Pharmacy, Chengdu University, Chengdu 610106, China
First published on 28th January 2026
Developing efficient methods for the deconstructive nitrogenation of alkenes remains challenging because carbon–carbon double bonds are generally resistant to cleavage under mild conditions. Here we report a catalyst-free, water-promoted deconstructive amination of electron-deficient dienes that proceeds through an unprecedented single-carbon deletion pathway. The transformation operates smoothly in water without photoirradiation, oxidants, or radical precursors, providing diverse fully substituted pyrroles. Mechanistic studies indicate a sequence involving initial aziridination followed by water-assisted C–C and C–N bond cleavage of a transient aziridine intermediate. Water plays a key role in enabling single-carbon extrusion and guiding selective skeletal reorganization. This approach offers high atom- and step-economy and complements existing oxidative and radical-based strategies. The mild conditions, broad substrate scope, and straightforward access to bioactive pyrrole analogs highlight the practical utility of this method. Overall, this work provides a mechanistically distinct route to carbon–carbon double bond deconstructive nitrogenation and expands the toolkit for heterocycle synthesis.
Green foundation1. This work advances green chemistry by enabling deconstructive amination of alkenes in water under catalyst-free, oxidant-free, and organic-solvent-free conditions, providing a rare single-carbon deletion pathway to fully substituted pyrroles—valuable heterocycles with broad pharmaceutical relevance.2. The reaction exhibits excellent step-economy, with diaryl sulfide recovered as a recyclable byproduct. The pyrrole products precipitate directly from water, allowing isolation by simple filtration and reducing solvent use, waste generation, and energy input. 3. Greenness could be further enhanced by adapting the process to continuous-flow aqueous operation. Extending this water-promoted single-carbon deletion strategy to other deconstructive amination and cascade transformations may broaden its synthetic and environmental impact |
C) bonds has emerged as a frontier in modern synthetic chemistry, enabling access to complex molecular architectures that fuel advances in synthesis and drug discovery.1 Classic methods, such as ozonolysis2 and alkene metathesis,3 exemplify the versatility of C
C bond activation (Scheme 1A, left). These transformations either generate carbonyl functionalities or reorganize unsaturated frameworks, thus expanding retrosynthetic options and enhancing molecular diversity.4
Despite these advances, the direct conversion of alkenes into nitrogen-containing frameworks through concurrent C
C bond cleavage and C–N bond formation remains a formidable challenge (Scheme 1A, right). The inherent stability of the C(sp2)–C(sp2) bond often necessitates harsh conditions incompatible with sensitive nitrogen functionalities.5 Additionally, achieving precise control over both chemoselectivity and regioselectivity during concurrent bond scission and formation events poses significant difficulties. Unlike conventional alkene amination reactions—such as aziridination, aza-Michael addition, or hydroamination—that introduce nitrogen functionality onto the existing framework,6 deconstructive nitrogenation enables fundamental skeletal editing.7 This strategy unlocks distinct molecular scaffolds from alkene materials and offers a direct route to valuable nitrogen heterocycles that are inaccessible through traditional methods.
Recent advances in this area can be categorized into two principal mechanistic classes (Scheme 1B). The first involves oxidative N-atom insertion,8 particularly utilizing an in situ generated iodonitrene-like species,9 as elegantly demonstrated by the Morandi and Ball/Kürti groups.10 Ball and Kürti achieved the oxidative nitrogen insertion into silyl enol ether C
C bonds.10a Morandi reported nitrogen insertion into unactivated alkenes to form aza-allenium intermediates, which could be further transformed into nitriles or amidines.10b
The second class encompasses radical-mediated pathways.11 Jiao's copper-catalyzed aerobic oxidative cleavage of C
C bonds affords carbonyl and nitrile functionalities,12a whereas Studer achieved radical oxidative scission of C
C bonds to form oximes using N-nitrosomorpholine, followed by diversification into benzimidazoles or indazoles.12b Although these pioneering studies highlight the power of oxidative and radical-mediated deconstructive nitrogenation strategies, approaches that proceed via different mechanism—particularly those operating under mild and operationally simple conditions—still remain highly desirable.
Here, we report a water-promoted deconstructive amination of 2-allylidenemalononitriles that proceeds through an unprecedented single-carbon deletion pathway,13 employing sulfilimines as nitrogen sources (Scheme 1C).14 The reaction operates under green, catalyst-free conditions, using water as the sole solvent without photoirradiation, oxidants, or radical precursors. It features high atom- and step-economy through selective single-carbon deletion, with diaryl sulfide recovered as a recyclable byproduct, and displays a broad substrate scope encompassing diverse functionalized dienes and alkyl- or aryl-substituted sulfilimines. In line with our ongoing efforts toward the construction of pharmaceutically relevant heterocyclic scaffolds,15 this method grants direct access to fully substituted pyrroles, privileged motifs with wide pharmaceutical relevance. The pyrrole products can be readily diversified into analogs of known bioactive molecules, underscoring the practical value of this strategy. Mechanistic investigations, integrating experimental observations with DFT calculations, reveal a crucial role of water in mediating sequential C–C and C–N bond cleavage of a transient aziridine intermediate. Collectively, this work expands the mechanistic repertoire of single-carbon deletion and deconstructive amination of C
C bonds, offering a sustainable platform for the efficient construction of valuable pyrrole scaffolds.
| Entry | Solvent | Additive | Yield of 3a (%) | Yield of 4a (%) | Yield of 5a (%) |
|---|---|---|---|---|---|
| a Reaction conditions: 1a (0.12 mmol), 2a (0.10 mmol), and additive (50 μL) in solvent (1.0 mL) under Ar at 60 °C for 4 h; NMR yield. b Isolated yield. | |||||
| 1 | CDCl3 | — | 7 | 61 | 22 |
| 2 | CD2Cl2 | — | 10 | 63 | 15 |
| 3 | CD3CN | — | 12 | 47 | 23 |
| 4 | DMSO-d6 | — | 10 | 58 | 18 |
| 5 | CDCl3 | H2O | 42 | 39 | 14 |
| 6b | EtOH | — | 68 | — | 8 |
| 7b | EtOH | H2O | 78 | — | — |
| 8b | H2O | — | 81 | — | — |
Indeed, the deliberate addition of 50 μL of H2O to the reaction in CDCl3 dramatically shifted the product distribution, making 3a the major product (42% yield) together with 39% of 4a (entry 5). Encouraged by this result, we examined protic solvents and found ethanol to be particularly effective, affording 3a in 68% isolated yield with 8% of the (4 + 1) adduct 5a, while 4a was no longer detected (entry 6). Adding 50 μL of water to the ethanol medium further improved chemoselectivity, raising the yield of 3a to 78% (entry 7). Remarkably, using water alone as the solvent provided 3a in 81% yield after 4 h (entry 8). The pyrrole product precipitated directly from the aqueous phase, allowing isolation by simple filtration. Further screening showed slower conversion at room temperature and no further improvement at higher temperature (see Table S1 in SI for optimization details). Ultimately, we identified water at 60 °C as the optimal, simple, and sustainable conditions.
With the optimal conditions in hand, we next explored the generality of this water-promoted deconstructive amination cascade (Scheme 2). The protocol proved highly versatile toward both reaction partners, enabling straightforward access to a wide range of fully substituted pyrroles. We first examined the electronic and steric influence of N-aryl substituents on sulfilimine 2. Gratifyingly, a broad variety of N-aryl sulfilimines were well tolerated irrespective of electronic character. Electron-rich substrates (tolyl, 3b, 80%; tBu, 3c, 75%; Ph, 3d, 76%) and electron-deficient ones (NO2, 3e, 77%; COOMe, 3f, 79%) all reacted efficiently. Halogenated derivatives (F, Cl, Br; 3g–3i, 67–80%) were similarly effective, while CF3 substitution reduced reactivity (3j, 45%). Substrates bearing meta-substitution or para, meta-disubstitution displayed good reactivity (3k, 82%; 3l, 75%), whereas ortho-substitution significantly suppressed efficiency due to steric hindrance (3m, 49%). The structure of 3a was confirmed unambiguously by X-ray crystallography (CCDC 2466191; see SI for details).
![]() | ||
| Scheme 2 Substrate scope of the water-promoted deconstructive amination cascade reaction. Reaction conditions: 1 (0.12 mmol) and 2 (0.1 mmol) in 1.0 mL of H2O at 60 °C for 4 h; isolated yield. | ||
The methodology also accommodated a variety of functional handles. A vinyl-bearing sulfilimine furnished 3n in 87% yield, providing a valuable synthetic vector for further elaboration. Remarkably, 2-formyl-substituted diphenyl sulfilimines (2o–2q) delivered pharmaceutically relevant pyrrolo[1,2-a]quinazoline scaffolds 3o–3q in a single step. Beyond diaryl sulfilimines, cyclic dibenzothiophene analogs (2r–2t)14e also participated effectively, affording structurally distinct pyrroles (3r–3t). For instance, N-pyridyl 3r was obtained in 57% yield, hydroxyl-substituted 2s furnished 3s in 85% yield, and a protected amine substrate (NHCbz) provided 3t in 67% yield, underscoring excellent functional group tolerance.
We next investigated the scope of the diene component. A diverse range of aryl substituents at the R position were compatible, including para-substituted electron-donating and electron-withdrawing groups (CH3, NO2, COOMe, CN, OCH3, halogens, OCF3, SCH3), delivering products 3u–3af in good to excellent yields. Internal and terminal alkynes were also tolerated (3ag, 94%; 3ah, 84%). meta-Substituted arenes (3ai–3am) gave consistently high yields (83–90%), while sterically hindered ortho-substituted derivatives (3an–3aq) remained effective; notably, the ortho-fluoro analogue 3ao reached 92% yield. Furthermore, naphthyl (3ar, 73%) and heteroaryl dienes (3as–3au) consistently observed, with no detectable (4 + 1) or aziridination adducts (products 4–6). The breadth of substrate scope, combined with the operational simplicity of using water as the sole solvent, highlights the utility of this protocol for efficient access to diverse fully substituted pyrroles.
To evaluate the scalability of this water-promoted process, we performed the reaction on a 2.0 mmol scale (Scheme 3A). The scaled-up reaction maintained excellent practicality and efficiency, yielding 521 mg of deconstructive amination cascade product 3a (79% yield). Subsequent synthetic transformations demonstrated the versatility of this green approach in producing a diverse array of biologically active analogs of polysubstituted pyrroles and fused pyrrolic derivatives (Scheme 3B).16a Treatment of 3a with polyphosphoric acid (PPA) furnished the decarboxylated and hydrolyzed product 7 in 85% yield, while simple acetylation gave compound 8, an analogue of a metallo-β-lactamase inhibitor.16b [4 + 2] cycloaddition of 3a with isothiocyanatobenzene or 1,2-diphenylethyne smoothly provided the fused heterocycles 9 and 10, respectively. Condensation of 3b with formic acid afforded pyrrolo[2,3-d]pyrimidin-4-ol 11, which upon chlorination yielded the metallo-β-lactamase inhibitor analogue 12.16c Further diversification of 12 through thiourea incorporation introduced a mercapto group (13, 85%), while substitution with m-chloroaniline delivered EGFR tyrosine kinase inhibitor analogue 14 in 77% yield.16d In addition, the terminal alkyne-functionalized pyrrole 3ah underwent a copper-catalyzed click reaction with Zidovudine, efficiently appending a nucleoside fragment to the pyrrole scaffold and affording conjugate 15 in 91% yield.
![]() | ||
| Scheme 3 (A) Scale-up reactions with diphenyl sulfide recovery. (B) Derivatization of fully substituted pyrrole products. | ||
To elucidate the transformation mechanism, we combined experimental and computational investigations. HRMS of the standard reaction mixture detected ions consistent with the 1a–2a adduct (Scheme 4A, top left). To probe the fate of the extruded carbon, a diene 1x (from dimethyl acetylenedicarboxylate) was subjected to the standard water-mediated reaction with sulfilimine 2a (Scheme 4A, top right), smoothly delivering pyrrole 16. HRMS further revealed methyl glyoxylate, indicative of C
C bond scission. Collectively, these results suggest that, in water, the excised single-carbon unit is released as an aldehyde species (or its hydrate).
![]() | ||
| Scheme 4 (A) Experimental investigations on the reaction mechanism. (B) DFT calculation of the water-promoted pyrrole synthesis through deconstructive amination. | ||
Given the transient nature of the reaction intermediates, we turned to in situ NMR monitoring of the model reaction in EtOH between 1a and 2a to gain further insight (Scheme 4A, bottom). These results revealed the rapid accumulation of aziridine intermediate 4a, whose concentration peaked quickly before being gradually consumed to form the final pyrrole product 3a. This observation provides direct evidence for a kinetically competent, semi-stable intermediate and confirms that 4a is a key on-pathway precursor for the deconstructive amination.
With these experimental insights, we performed DFT calculations (M06-2X functional)17 to map the detailed energy profile (Scheme 4B, see Fig. S1–S3 in SI for more calculation details). The computational studies revealed that the reaction initiates with addition of sulfilimine 2a to diene 1avia transition state TS1 (ΔG‡ = 16.2 kcal mol−1) to afford adduct IM1. From this point, two competing pathways emerge: (i) intramolecular cyclopropanation leads to aziridine 4a with concomitant release of diphenyl sulfide (Ph2S), placing 4a in a deep energy well (ΔG = −28.9 kcal mol−1) consistent with its observed accumulation; (ii) alternatively, IM1 can undergo intramolecular cyclization to form the more stable (4 + 1) cycloadduct 6a. However, this latter pathway features a significantly higher activation barrier (ΔΔG‡ = 8.2 kcal mol−1), explaining why the reaction selectively proceeds through aziridination to form 3a rather than 6a under aqueous conditions.
The subsequent transformations account for the single-carbon deletion process. Water-promoted ring opening of aziridine 4avia C–C bond cleavage completes the scission of the alkene unit and furnishes hemiaminal IM2. This intermediate then undergoes C–N bond cleavage with elimination of formaldehyde to complete the deconstructive amination process, yielding intermediate IM3. Water-assisted intramolecular cyclization of IM4 delivers IM5, which undergoes stepwise proton transfers and aromatization to afford the fully substituted pyrrole 3a as the thermodynamically favored product. These experimental and computational findings define a unique water-promoted single-carbon deletion mechanism for the deconstructive amination and pyrrole synthesis.
C bonds, providing a sustainable and general strategy for constructing valuable nitrogen-containing frameworks.
CCDC 2466191 (3a) contains the supplementary crystallographic data for this paper.18
Footnote |
| † These authors contributed equally to this work. |
| This journal is © The Royal Society of Chemistry 2026 |