Chemical oxidation, typified by Fenton and Fenton-like reactions, is a widely adopted and promising wastewater treatment technology, known for its ability to produce highly reactive oxide species that efficiently remove pollutants [1-3]. However, this process typically demands significant energy/chemical agent input to mineralize organic pollutants, converting the carbon within them into CO₂. This results in the loss of a substantial amount of chemical energy contained in the wastewater, making energy recovery quite challenging [4]. Developing approaches to regulate the oxidation pathway of pollutant from mineralization to polymerization is a crucial step in advancing water treatment from pollution control to energy recovery, ultimately reducing the carbon footprint. Direct surface oxidation of the pollutants by forming oligomers is a potential way to regulate the polymerization pathway. Elimelech and co-workers [5,6] have demonstrated that the surface of high valence metal oxides can initiate surface-dependent coupling and polymerization pathways for the removal of organic compounds via a two-electron direct oxidation process. Specifically, oxidant activation and organic adsorption occur simultaneously at adjacent sites on the catalyst surface, followed by the creation of positively charged centers on neighboring and pairs of carbon atoms in the phenolic hydroxyl group of the organics, leading to C—C coupling or spontaneous C—O polymerization of the stabilized phenoxonium ions. Three crucial functions of the nanocatalyst surface, i.e., activation, stabilization and accumulation were revealed to render the direct oxidation process thermodynamically spontaneous and kinetically favorable. Except for such direct surface oxidation process, recent studies highlighted that spatial nanoconfinement can substantially alter the thermodynamic and kinetic behaviors of chemical reactions, leading to different reaction pathways and outcomes compared to the bulk oxidation reaction [7]. Spatial nanoconfinement can trigger thermodynamically favorable oligomerization routes for the removal of organic pollutants, due to its more negative Gibbs free energy compared to the open-ring route [8,9]. Under spatial nanoconfinement, the oxidant exhibits higher adsorption energy and lower activation energy barriers on the catalyst. This facilitates the breaking of O—O or C—O bonds in the oxidants and enhances electron migration from the molecular orbitals of the metal particles, thereby activating the oxidants more easily. Additionally, spatial nanoconfinement triggers synergistic effects such as reactant enrichment, electron-metal-carrier interactions, and improved mass transfer. These effect will increase the localized reactant concentration, which is considered kinetically favorable [10,11]. The local enrichment of reactants at high concentrations promotes effective molecular collisions, shortens the mass transfer distance of ROS and facilitates efficient internal electron transfer, thus enhancing the removal of organic pollutants at higher reaction rates [12]. This phenomenon holds promise for converting organic pollutant into polymers for recovery, thereby enabling green and low-carbon sewage and wastewater treatment [8,12,13]. Nevertheless, more in-depth understanding of how nanoconfinement affects the oxidation pathway of pollutant is still lacking, and the potential mechanism underlying the selectivity enhancement of oligomers under nanoconfinement remains uncovered.

Recently, Pan's group [14] investigated how the size of spatial nanoconfinement affects the oxidation pathways of pollutants and the involved modulation mechanism, which was achieved by confining the Mn₃O₄/PMS Fenton-like system within amorphous carbon nanotubes (ACNTs) of various pore diameters. Firstly, Mn₃O₄@nACNT (n = 20, 55, and 120, denoting the pore size of ACNTs in nm) catalysts were synthesized by encapsulating Mn₃O₄ nanoparticles (NPs) within these ACNTs via a hydrothermal reaction, as illustrated in Fig. 1a. High-resolution TEM (HRTEM) images and X-ray diffraction (XRD) patterns of all catalysts revealed Mn₃O₄ NPs with octahedral rhombic shapes predominantly featuring (211) lattice facets (Figs. 1b-i). The Mn₃O₄ NPs exhibited a fairly uniform sizes distribution ranging from 15.0 nm to 24.0 nm (Fig. 1j), along with characteristic diffraction peaks indicating spinel structures (Fig. 1k). The preceding findings clearly demonstrate that, besides the nanopore diameter, these catalysts exhibit a satisfactory structural similarity, providing a rational basis for understanding how spatial nanoconfinement influences the Fenton-like reaction.

To evaluate the differences in pollutant transformation pathways between the spatially nanoconfined Mn₃O₄@nACNT/PMS system and the open Mn₃O₄/PMS system, Pan and co-workers selected the aromatic pollutant PhOH as a model compound, quantifying its oxidation products under various oxidation systems. In bulk Mn₃O₄/PMS system, PhOH oxidation primarily yield degradation products such as quinone, organic acids, and carbonyl compounds. However, in the confined Mn₃O₄@nACNT/PMS system, the proportion of degradation products decreased while oligomeric products increased notably as the ACNTs tube size decreased from 120 nm to 20 nm. In particular, the Mn₃O₄@20ACNT/PMS system exhibited a significant oligomer yield (36.7%) at 50.0% PhOH conversion, which was 20 times higher than that of the bulk Mn₃O₄/PMS system (2.9%). Moreover, the yield of organic acids decreased from 17.9% to 2.7%, and the lower yields of ring-opening products indicated inhibited degradation pathway in narrower nanopore (Fig. 2a). Additionly, the distribution of dimers, trimers, and tetramers at different PhOH conversions further supported the favored oligomerization pathway is in narrower pores (Fig. 2b).

To elucidate the mechanism behind the enhanced selectivity of oligomerization under nanoconfinement, Pan and co-workers conducted a series of spin trapping and chemical scavenging experiments. They first identified the principal dissolved active oxidants, ruling out HO^•, SO₄^•– and ¹O₂ in all the systems. Next, they investigated surface-bound active species, excluding Mn–PMS complexes and Mn(Ⅴ) in PhOH conversion, and clarifying Mn(Ⅳ) as the principal active oxidant in all the systems. Finally, they proposed that PhOH conversion proceeds via electron transfer. Based above, the elementary steps of PhOH conversion were portrayed. The reaction between Mn(Ⅳ) and PhOH predominantly proceeded through inner-sphere electron transfer rather than outer-sphere electron transfer with surface-active Mn(Ⅳ) generated by Mn₃O₄ and PMS in both confined and bulk oxidation systems. First, PhOH forms a complex with surface-bound Mn(Ⅳ) and then undergoes one-electron transfer to produce PhO^∙, where the C—O and C—C coupling of PhO^∙ leads to polymerization. Additionly, PhO^∙ may be oxidized by surface-bound Mn(Ⅳ) to form a phenoxenium cation (PhO⁺). The electrophilic substitution of PhO⁺ on PhOH preferentially leads to C—C coupled dimers, or it undergoes one-electron oxidation/reduction to produce ring-opening products (Fig. 2c).

The above results indicate that PhO^∙ is the key precursor of the polymerization reaction. Importantly, the higher proportion of dimers produced by PhO^∙ coupling suggests that the increasing concentration of PhO^∙ facilitates self-coupling and/or cross-coupling with other intermediates in the confined systems, thereby modulating the catalytic oxidation process. With lower pH in the narrower pores, the higher proton concentration favors the protonation of Mn₃O₄ nanoparticles, which increases the reduction potential and/or promotes the generation of active Mn(Ⅳ) sites and/or the subsequent electron transfer process for the conversion of PhOH to PhO^∙ (Fig. 2d). Furthermore, the enrichment of PhO^∙ within the catalysts further enhances the selectivity of polymerization (Fig. 2e). These results demonstrate that nano-confinement effectively promotes the generation and mutual collision of phenoxy radicals, driving the polymerization reaction and changing the product distribution.

In conclusion, this study by Pan's group systematically elucidates that the carbon evolution pathway can be readily tuned by varying the spatial nanoconfinement sizes. It also reveals that these sizes drive the selectivity and yield of the oxidation of pollutant products, from the fragmented molecule to polymer, through affecting the production of PhO^∙. These findings offer a valuable theoretical foundation and methodological references for the innovative development of green, low-carbon treatment technology. Despite the above exciting progress, a gap remains between the fundamental understanding of nanoconfinement oxidation systems and their application in water treatment. Factors such as the impact of multiple contaminants, the reusability and long-term stability of catalysts need to be considered under realistic conditions. Meanwhile, developing assisted separation processes for polymer collection present significant challenges. Moreover, studying the mechanism of nanoconfinement effect on the reaction transition state is crucial. This will provide guidance for modulating reaction species and pathways, optimizing kinetics to control polymer formation, and obtaining high-quality polymer products. Additionally, the rational design of catalyst materials with reaction-oriented properties holds great promise for precisely manipulating the reaction pathways of pollutant molecules. These in-depth studies will further advance the application of nanoconfinement oxidation for water treatment from both fundamental and applied perspectives.

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Yanhua Peng: Writing – original draft, Conceptualization. Xin Yu: Writing – review & editing. Ting Wang: Writing – review & editing, Conceptualization.