Yolk@shell nanoreactor for the heterogeneous Fenton reaction: a review of recent progress

Abstract

The heterogeneous Fenton reaction is a promising technology to non-selectively degrade organic pollutants, providing a feasible solution to environmental problems. However, the essential issue in the Fenton reaction is the design of highly efficient catalysts, and yolk@shell structures are usually employed as a nanoreactor to boost the activation efficiencies of H₂O₂ (or peroxymonosulfate (PMS)). Typically, a yolk@shell nanoreactor is constructed from a hollow shell and a movable core, which can transfer the Fenton reaction from the bulk solution to the internal cavity. Benefiting from the confinement effect, the chance of collision among various reactants is increased, the degradation rate is accelerated, H₂O₂ (or PMS) utilization efficiency is boosted, the environmental tolerance of catalysts is improved, and metal leaching is lowered. Thus, the yolk@shell nanoreactor provides an ideal platform for the heterogeneous Fenton reaction. In addition, many methods could be adopted to tailor the yolk@shell nanoreactor for better catalytic performance, and this inspired researchers to design nanoreactors with optimized compositions and novel structures. At present, many reviews on either the preparation and application of yolk@shell nanoreactors or the recent development of heterogeneous Fenton reactions are reported; however, strategies to improve the Fenton reaction based on yolk@shell nanoreactors have not been discussed in detail. This review illustrated the recent progress on yolk@shell nanoreactors for heterogeneous Fenton reactions and provided comprehensive information on catalyst design, catalytic performance, catalytic mechanisms, catalyst advantages, and catalyst improvement methods, aiming at developing catalysts with better performance.

---

Introduction

Persistent organic pollutants, including antibiotics, fertilizers, and synthetic dyes, produced as a result of rapid urbanization and industrialization, have posed great threats to the environment and human beings. Many techniques have been established for their disposal; however, most of them only transfer the pollutants from one phase to another, such as adsorption and flocculation.^1–4 In this case, secondary treatment must be conducted for complete decomposition. To avoid these shortcomings, advanced oxidation processes (AOPs) have drawn considerable attention in organic wastewater treatment,⁵ and the Fenton reaction is one of the most prominent AOPs. In the classical H₂O₂-based Fenton reaction, H₂O₂ acts as an oxidant and is activated by Fe(II) to generate ˙OH radicals.⁶ With an oxidation potential as high as 2.8 eV, ˙OH can degrade all organic compounds non-selectively.

After long-term research, significant progress has been achieved in the Fenton reaction. The homogeneous reaction has been gradually replaced by its heterogeneous counterparts. The peroxymonosulfate (PMS)-based Fenton reaction has emerged and shown great potential in organic pollutant removal.⁷ External energy sources were introduced to accelerate the Fenton reaction. Research attention has extended beyond catalytic performance to include other critical issues such as metal leaching, secondary pollution, reusability, and environmental adaptability.⁸

Whether homogeneous or heterogeneous, in the H₂O₂-based or PMS-based Fenton reaction, catalyst design is the essential issue that affects the final degradation performance. Benefiting from nanoscience, nanomaterials with different morphologies, structures and compositions have been rapidly developed over the past 30 years. For example, catalysts with hierarchical porous structures are beneficial for increasing the number of reactive sites,⁹ heteroatom-doped carbon materials are low-cost and environmentally friendly,¹⁰ and composites combining catalysts and co-catalysts can accelerate the rate-limiting step in the Fenton reaction.^11,12 Compared to bulk-phase materials, the catalytic activity is remarkably boosted on the surface of these nanomaterials.

At present, nanoreactors have become a hotspot in scientific research. Different from chemical reactions occurring in bulk solution, the reaction inside a nanoreactor is confined within a closed microenvironment.^13,14 Due to the limited space, many reaction parameters are altered, such as reactant adsorption, reactant concentration, catalytic mechanism, and substance exchange. This results in enhanced catalytic activity, selectivity and reaction rate. Usually, nanomaterials with hollow voids can function as nanoreactors, including nanotubes,¹⁵ hollow capsules and yolk@shell composites.^16,17 As an important extension of core@shell composites and hollow capsules, the yolk@shell composite is the most widely accepted candidate for nanoreactors. Compared to the hollow capsule, the yolk@shell nanoreactor provides an extra core, leading to multiple functions. Compared to the core@shell structure, the cavity serves as a unique reaction environment, and the loss of active sites on the core surface is avoided due to the non-seamless contact between the core and shell.

To improve the catalytic performance of the yolk@shell nanoreactor, many research groups have devoted their efforts to synthetic methods, structural design and composition optimization. The catalytic core has evolved from metal oxides to metal-organic frameworks (MOFs);^18,19 the shell has evolved from merely a protective function to also serving a catalytic role;²⁰ and the structure has evolved from a single core to multiple or functionally diverse cores.^21,22 The catalytic activity has been greatly enhanced by these nanoreactors. Additionally, several other persistent challenges in the traditional Fenton reaction have also been effectively addressed. For example, metal ion leaching has been reduced, the applicable pH range has been widened, and environmental tolerance has been improved under various impurity conditions. With continuous progress, the overwhelming advantages of yolk@shell nanoreactors in the Fenton reaction have been gradually recognized.

Several reviews have focused on yolk@shell nanoreactors, summarizing their preparation methods, structural design, catalytic advantages, and application fields.^23–27 Although these reviews offer a broad understanding of yolk@shell nanoreactors, detailed discussions on specialized catalytic reactions and recent advancements are often lacking. As far as we known, no published review has yet summarized the application of yolk@shell nanoreactors in Fenton reactions. To fill this gap, this review mainly focuses on the H₂O₂-based and PMS-based heterogeneous Fenton reactions. The fundamental principles of the Fenton reaction and the yolk@shell nanoreactor are first introduced (Fig. 1). Then, the advantages of the yolk@shell nanoreactor in Fenton reactions are illustrated from different perspectives, such as accelerating the degradation rate, suppressing metal leaching, improving environmental adaptability, and increasing oxidant utilization As the yolk@shell nanoreactor is composed of only two components (core and shell), the following section introduces the approaches to optimize their performance. Next, the advantages of yolk@shell nanoreactors in Fenton reactions coupled with external energy inputs are discussed. Finally, the challenges and future prospects of yolk@shell nanoreactors for Fenton reactions are addressed to stimulate future breakthroughs and attract the interest of potential researchers. We believe this review will help readers deepen their understanding of yolk@shell nanoreactors in Fenton reactions and inspire the development of novel and efficient yolk@shell nanoreactors.

Fig. 1 Yolk@shell nanoreactor in the Fenton reaction.

Fundamental knowledge on Fenton reaction and yolk@shell nanoreactor

H₂O₂-based Fenton reaction

In 1894, the French scholar Fenton discovered that tartaric acid could be rapidly oxidized by the Fe²⁺ + H₂O₂ system.²⁸ Nevertheless, the underlying reaction mechanism was not understood at that time. After a long period of study, the researchers uncovered that ˙OH radicals were the main reactive species responsible for tartaric acid degradation (eqn (1) and (2)). In 1964, the Canadian scientist Eisenhauer reported that phenol and alkylbenzene could be successfully degraded via the Fenton reaction for the first time.²⁹ Since then, the homogeneous H₂O₂-based Fenton reaction has been applied in wastewater treatment. Initially, the homogeneous reaction catalyzed by Fe³⁺/Fe²⁺ was popular due to its high degradation efficiency and minimal requirement for specialized equipment. Nowadays, its industrial application has already been realized. However, as research progressed, several drawbacks began to emerge. For example, Fe²⁺ cannot be recovered and reused; acidification pre-treatment is required (pH = 2.0–4.0); and a large amount of iron sludge is generated.³⁰ These disadvantages have significantly restricted further development, and researches have gradually shifted their attention to the heterogeneous Fenton reaction, in which the solid transition metals are used as catalysts to replace Fe²⁺.

Fe²⁺ + H₂O₂ → Fe³⁺ + ˙OH + OH⁻ (1)

Fe³⁺ + H₂O₂ → Fe²⁺ + ˙OOH + OH⁺ (2)

In general, transition metals with variable valences can function as catalysts for H₂O₂ activation, including Cu(II)/Cu(I), Mn(V)/Mn(III), Ni(III)/Ni(II), and so on.³¹ At present, Fe(III)/Fe(II) are still the most widely accepted catalysts in heterogeneous reactions. In addition to their low cost and high activity, they can be easily recovered using an external magnetic field, which is particularly important for heterogeneous catalysts. Some intrinsic drawbacks in other transition metals limit their application. For example, although the working pH range of Cu(II)/Cu(I) is even broader than that of Fe(III)/Fe(II), their high toxicity makes it difficult to meet discharge standards.³² The activity of Mn(V)/Mn(III) and Ni(III)/Ni(II) is lower, and their prices are also higher than those of Fe(III)/Fe(II).³³

After the development of the heterogeneous reaction, many traditional problems associated with the homogeneous reaction were effectively addressed. However, new challenges have emerged. As is commonly known, catalytic reactions primarily occur on the catalyst surface. Both the number of reactive sites and the catalytic activity are not comparable to those of ions uniformly dispersed in the reaction solution. Therefore, the key problem faced by the heterogeneous reaction is its relatively lower catalytic performance. To date, many studies have focused their efforts on improving catalytic performance, with surface area enlargement being the primary strategy. Nanomaterials with porous structure or hierarchical structures are particularly favored. Hu and co-workers synthesized ZIF-67-derived Co₃O₄–CeO₂ nanocages with a porous structure via a combination of the hard-template method and subsequent calcination, achieving a surface area of 109.9 m² g⁻¹.³⁴ Many papers have reviewed the recent developments in the H₂O₂-based Fenton reaction.^35,36 In this study, we mainly highlight the advantages of the yolk@shell nanoreactor in the Fenton reaction.

PMS-based Fenton reaction

From the H₂O₂-based Fenton reaction, scientists have recognized that radicals with powerful oxidation capacity can be used for the non-selective degradation of organic pollutants. Various radicals are generated in chemical reactions, including CO₃˙, Cl˙, SO₄˙⁻ and so on.^37,38 However, most of them were unsuitable for practical applications due to their intrinsic drawbacks. For example, the oxidation potential of CO₃˙ (1.8 eV) is lower than that of ˙OH (2.8 eV),³⁹ and Cl˙ radicals themselves are environmental pollutants.⁴⁰ Compared to ˙OH, SO₄˙⁻ exhibits a much higher oxidation potential (3.2 eV), a longer half-life (30–40 μs vs. 2.0 ns for ˙OH), and a broader applicable pH range (pH = 2.0–9.0).⁴¹ Owing to these advantages, organic pollutants can be rapidly degraded by SO₄˙⁻ over a wide pH range. Unlike the H₂O₂-based Fenton reaction, PMS can be activated both on the surface of transition metals via the radical activation path, and on heteroatom-doped carbon materials via the non-radical activation path.⁴²

Radical activation path

In general, two molecules are used for SO₄˙⁻ production: PMS and persulfate (PDS). Two SO₄˙⁻ radicals are generated from PDS activation (eqn (3)), while only one SO₄˙⁻ is produced from PMS activation (eqn (4)). At first glance, the activation efficiency of PDS appears to be higher. However, a heterolytic reaction occurs at the O–O bond in the PDS molecule, whereas a homolytic reaction takes place at the O–O bond in the PMS molecule (Fig. 2). The activation energy required for PMS is much lower than that for PDS; therefore, the PMS-based Fenton reaction is more widely used in organic wastewater treatment.

S₂O₈²⁻ → 2SO₄˙⁻ (3)

HSO₅⁻ → SO₄˙⁻ + ˙OH (4)

Fig. 2 Mechanism of PMS and PDS activation.

The activation efficiency of PMS itself was unsatisfactory in the absence of catalysts. Nevertheless, its activation could be remarkably accelerated by transition metal ions. As early as 1956, Ball and Edwards reported that PMS was efficiently decomposed into SO₄˙⁻ by Co(III) and Co(II) (eqn (5) and (6)). Similar to the development history of H₂O₂-based Fenton reaction, the PMS activation process has also undergone a transition from a homogeneous reaction to a heterogeneous reaction. Among various transition metal catalysts, Co-based catalysts displayed the most outstanding activity. Therefore, they became a focal point in PMS-based Fenton reactions for the generation of SO₄˙⁻ and ˙OH. Besides cobalt oxide, Co-based MOFs and Co single-atom catalysts were also developed.^43,44 However, given that the toxicity of cobalt ions is much higher than that of iron ions, their leaching was a problem that could not be ignored.

Co(II) + HSO₅⁻ + Co(III) + SO₄˙⁻ + OH⁻ (5)

Co(III) + HSO₅⁻ + Co(II) + SO₅˙⁻ + H⁺ (6)

Non-radical activation path

As mentioned in the above section, besides the radical activation path, PMS could also be activated on the surface of heteroatom-doped carbon materials. During the degradation process, the pollutants were successfully removed without radical production, and this was the so-called non-radical activation path. In 2013, Sun and co-workers prepared N-doped graphene oxide, and applied it for PMS activation. Nearly 100% of phenol was degraded within 45 min. For comparison, the degradation efficiency over graphene oxide was negligible. Mechanism studies indicated that the active sites in sp²-hybridized carbon within graphene possess abundant free-flowing π electrons, which could be activated via conjugation with the lone-pair electrons from the doped N atom.⁴⁵ C₃N₄, carbon nanotubes, graphene, and graphitized carbon were usually employed as catalysts.^46–48 Generally, sp²-hybrid carbon is chemically inert and hence incapable of catalytic reactions. Heteroatom doping was an effective method to tailor the electronic structure and enable PMS non-radical activation. Nevertheless, the activation paths were diverse. For example, carbon atoms could be partially replaced by the doped nonmetal atoms, leading to formation of nonmetal–C bonds. Due to differences in electronegativity, the uniform distribution of the electron cloud was disturbed, enabling PMS activation at electron-deficient carbon atoms for ¹O₂ generation.⁴⁹ In some studies, heteroatom-doped carbon acted only as an electron transfer medium, through which electrons migrated from organics to PMS, resulting in oxidative degradation of the pollutants.⁵⁰ At present, the general non-radical activation paths can be classified into ¹O₂ generation, electron transfer, and high-valence metal routes.⁵¹ The detailed mechanisms of these paths have been reviewed in other works.

These two PMS activation paths possess their own advantages and drawbacks. In the radical activation path, the oxidation potentials of reactive oxygen species (ROSs) are high enough to enable rapid degradation and mineralization of organic pollutants. Nevertheless, their performance is susceptible to pH and can be easily interfered with by impurities in aqueous solution. In an alkaline environment, the degradation rate significantly decreases, and sometimes even ceases, due to the reduced oxidation potentials of ROSs and the formation of hydroxides on the catalyst surface.⁵² Additionally, metal leaching is another serious concern. The discharge of toxic cobalt ions is strictly limited to less than 1.0 mg L⁻¹ according to the China Standard Emission (GB 25467-2010), while the limit for iron ions is 10.0 mg L⁻¹ (GT/T-31962-2015).⁵²

In the non-radical activation path, environmentally friendly carbon nanomaterials are used as catalysts and the content of doped heteroatoms is low. Therefore, the issue of metal leaching does not exist. Compared to their radical counterparts, non-radical systems exhibit much better environmental tolerance. Guan and co-workers prepared Fe, Co, N co-doped carbon capsules, where the ¹O₂ generation and electron transfer paths were responsible for 86.7% of tetracycline (TC) degradation. Importantly, the catalysts performed well across the entire pH range (pH = 0.0–14.0) and in solutions containing complex inorganic anions.⁵⁰ Unfortunately, the major limitation of the non-radical activation path is its low oxidation capacity. For example, the oxidation potential of ¹O₂ is only 2.2 eV.⁵³ From the above discussion, it is evident that the two activation paths are complementary. This provides the possibility of combining them within a yolk@shell structure to achieve enhanced catalytic performance.

Fenton reaction coupled with external energy

In the Fenton reaction, the activation of oxidants (H₂O₂ or PMS) forms the basis of pollutant degradation. Therefore, achieving high activation efficiency is key to accelerating pollutant degradation. In addition to catalyst design, introducing various external energies is a feasible strategy to enhance activation efficiency. Photo energy, electrical energy and mechanical energy are commonly involved, and the corresponding Fenton reactions are termed photo-Fenton reaction (with ultraviolet light [UV] or visible light),⁵⁴ microwave-Fenton reaction (with microwave),⁵⁵ electro-Fenton reaction (with electric current),⁵⁶ sono-Fenton reaction (with ultrasonic wave),⁵⁷ and so on.

These external energies not only accelerate ROS generation, but also bring additional benefits to the Fenton reaction. For instance, the cleavage of the O–O bond in H₂O₂ molecule becomes possible under ultraviolet light alone. Microwave irradiation can rapidly increase the solution temperature, thereby accelerating pollutant degradation under elevated temperatures. Ultrasonic waves improve the dispersity of solid catalysts in aqueous solution, enhancing Brownian motion and increasing the likelihood of collisions among pollutants, catalysts and oxidants. In electro-Fenton systems, electrons supplied at the electrode accelerate the reduction of Fe(III) to Fe(II). Consequently, coupling Fenton reactions with external energies results in pollutant degradation performance that exceeds the sum of individual processes. Feng and co-workers prepared graphene aerogel loaded with Fe_xO_y nanoparticles as catalysts for a photo-Fenton reaction. The synergetic effect was evaluated using a synergetic index (SI). In eqn (7), k, k₁, k₂ and k₃ represent the rate constants for the systems of catalysts + H₂O₂ + visible-light (photo-Fenton reaction), catalysts + H₂O₂ (Fenton reaction), catalysts + visible-light (photocatalytic reaction) and H₂O₂ + visible light (H₂O₂ self-activation reaction), respectively. In their study, the SI was 1.7, indicating that the synergistic effect of the photo-Fenton reaction provided an additional 70% efficiency in pollutant removal.⁵⁸ Using a similar method, Tong and co-workers reported an SI of 3.88 for the microwave-Fenton reaction catalyzed by FeCo/N atom-doped carbon (NC).⁵⁹

Fundamental knowledge on yolk@shell structure

At the beginning of this century, the core@shell structure attracted significant attention, as such composites could integrate the functions from both the core and the shell. In a classical core@shell structure, the shell is seamlessly coated on the surface of the core. However, in the context of the Fenton reaction, the catalytic properties of the core are difficult to exploit due to the limited contact among oxidants, pollutants and the catalytic core surface. Benefiting from the presence of hollow voids, the core and shell are spatially separated in a yolk@shell nanoreactor. As a result, chemical reaction can occur within the internal cavity, allowing the catalytic activity of the movable core to be utilized. Numerous catalytic reactions have been carried out inside yolk@shell nanoreactors, resulting in improvements in product yield, reaction rate and selectivity.⁶⁰ In Fenton reactions, oxidants diffuse into the cavity and are subsequently activated on the surface of the core to generate ROS, which then attack pollutants inside the nanoreactor. In the subsequent sections, we focus on the advantages of yolk@shell nanoreactors and the strategies for enhancing their catalytic activity.

There are many approaches for preparing yolk@shell structures, among which the hard-template-assisted etching method is the most frequently used. A sandwich-like structure is first synthesized using a hard template, followed by selective etching of the middle layer without damaging the external shell or inner core.⁶¹ However, several issues are associated with this method, such as complex preparation procedures, material wastage, and the extensive use of organic, acidic, or basic solvents. As a result, alternative approaches have been developed in recent years, including the ship-in-bottle method,⁶² surface-protected etching method,⁶³ Ostwald ripening method,⁶⁴ Kirkendall effect method,⁶⁵ galvanic replacement method,⁶⁶ chelation-competition-induced polymerization method⁶⁷ and others. To maximize the efficiency of yolk@shell nanoreactors, it is necessary to precisely control experimental parameters. For example, during the etching process, if the core is excessively etched, the number of active sites is reduced. On the other hand, if the core is insufficiently etched, the cavity size may be inadequate for efficient reaction performance. During heat treatment, the heating rate affects structural stability. Liu and co-workers reported the a two-component metal sulfide core decomposed into fragments when the heating rate exceeded 5 °C min⁻¹.⁶⁸

The traditional yolk@shell structure consists of a single core and a single shell. After 20 years of development, several non-traditional structures with improved performance have been synthesized. For instance, multiple cores can increase the number of catalytic sites,^69,70 multiple shells can offer enhanced protection⁷¹ and multiple cavities can provide diverse functionalities (Fig. 3).⁷² The detailed preparation methods, formation mechanisms, applications and advantages of yolk@shell nanoreactors have been reviewed elsewhere.⁷³ In this work, we focus exclusively on yolk@shell nanoreactors with optimized compositions, novel structures and tailored properties for enhancing catalytic activity in the Fenton reaction.

Fig. 3 Diagram of different yolk@shell nanoreactors. (a) Traditional yolk@shell nanoreactor with a single core and single shell; (b) multiple cores encapsulated within a single shell; (c) single core confined within multiple shells; (d) multiple voids.

Advantages of yolk@shell nanoreactor in Fenton reaction

Accelerating degradation rate

As mentioned above, the degradation rate of heterogeneous Fenton reaction is generally lower than that of their homogeneous counterparts, which remains a major bottleneck. Various strategies have been developed to improve catalytic performance and accelerate the degradation rate. Most of these approaches focus on catalyst design, particularly the optimization of nanostructure and chemical composition. For example, increasing the surface area to enhance the number of catalytic sites,⁷⁴ doping with different heteroatoms to tailor the electron cloudy density⁷⁵ and introducing external energy sources to activate oxidants through multiple paths.⁷⁶

When attention was focused on catalyst design, another strategy for enhancing degradation performance was largely overlooked. According to Le Chatelier's principle, increasing the concentrations of reactants is a feasible way to accelerate the degradation rate. The Fenton reaction and the subsequent degradation process can be considered a cascade reaction. Therefore, it is necessary to enhance the concentrations of all reactants, including oxidants, pollutants, ROSs, and even catalysts. In bulk solution, catalysts, oxidants and pollutants are homogeneously dispersed, resulting in low local concentrations. Consequently, both oxidant activation efficiency and pollutant degradation rate are often unsatisfactory. When the yolk@shell nanoreactor is employed as a catalyst, the primary site of the cascade reaction shifts from the bulk solution to the cavity inside the nanoreactor. The adsorption of pollutants, concentration of reactants, diffusion of products and catalytic mechanism are all altered within this confined microenvironment – an effect known as the confinement effect.¹⁶ Materials with good adsorption capacity are commonly used as shell components.^77,78 In such cases, pollutant molecules are enriched on the shell surface, increasing the local concentration. Simultaneously, oxidants diffuse into the cavity, where the catalytic core is more readily accessible. The ROSs generated are also confined within the microenvironment, leading to ultrahigh local concentrations. Based on the above analysis, it is evident that the concentrations of radicals, pollutants and oxidants are simultaneously enhanced within the yolk@shell nanoreactor, thereby accelerating the degradation rate. The degradation of pollutants results in a lower reactant concentration and a higher product concentration within the cavity. The concentration gradients between the interior and exterior of the nanoreactor facilitate the continuous influx of pollutant molecules and the efflux of products, thus sustaining the Fenton reaction.

Due to technical limitations, it is difficult directly analyze radical concentrations and degradation rates inside the cavity. Currently, degradation rates are assessed based on changes in pollutant concentration in the bulk solution, and the advantage of the confinement effect in accelerating degradation has been confirmed. It is known that the content of catalytic components in a yolk@shell nanoreactor is lower than that in the pure core, due to the presence of the shell. Nevertheless, the degradation rate observed for the nanoreactor is often higher than that of the pure core under the same total mass. For example, Deng and co-workers reported an FeS₂/MoS₂@C nanoreactor with FeS₂/MoS₂ heterojunctions encapsulated inside a carbon capsule, as shown in Fig. 4a and b. In the photo-Fenton reaction, only 19.2% of MTZ was degraded over FeS₂/MoS₂ in 25 min, whereas 90.5% of MTZ was removed in the presence of the FeS₂/MoS₂@C nanoreactor.⁷⁹

Fig. 4 (a) SEM, TEM, HRTEM and HAADF-STEM images of FeS₂/MoS₂@C.⁷⁹ (b) Schematic of the degradation mechanism of FeS₂/MoS₂@C.⁷⁹ (c) Schematic of the catalytic mechanism inside the FeS₂/MoS₂@SiO₂ nanoreactor.⁸¹ (d) TOC mineralization efficiency and H₂O₂ activation efficiency with FeS₂/MoS₂, B-FeS₂/MoS₂@SiO₂ and FeS₂/MoS₂@SiO₂.⁸¹ Reproduced from ref. 79 with permission from Elsevier, copyright 2022. Reproduced from ref. 81 with permission from Elsevier, copyright 2023.

Besides the confinement effect, another contributing factors should not be overlooked. When the Fenton reaction is carried out in bulk solution, the generated radicals must travel a relatively long distance to encounter the target pollutants, owing to their low concentrations. However, the lifetime of radials is very short, and their diffusion distance is limited. For example, the lifetime of ˙OH is less than 2.0 ns, and its diffusion distance is only 2.2 nm before quenching.⁸⁰ The shorter the distance between the radical and the pollutant molecule, the greater the probability of collision. In a yolk@shell nanoreactor, ROSs are generated on the surface of the catalytic core, while pollutant molecules are primarily adsorbed on the inner surface of the shell; their separation distance can be shortened by controlling the cavity diameter.

Boosting activation efficiency and utilization efficiency of oxidants

The prices of H₂O₂ and PMS were $18 per L and $68 per kg from Aladdin Biochemical Technology Co., Ltd, respectively, and they accounted for most of the cost in the Fenton reaction. In this case, it was particularly important to improve their utilization efficiency. Unfortunately, their utilization efficiency was usually very low in heterogeneous Fenton reactions. Hui and co-workers reported that the utilization efficiency was only 28.1% in the FeS₂/MoS₂@SiO₂ + H₂O₂ system, as shown in Fig. 4c and d.⁸¹

In H₂O₂-based Fenton reaction, H₂O₂ were first activated on the catalyst surface to generate ˙OH, accompanied by self-decomposition (eqn (8) and (9)). Herein, the activation efficiency was defined as [H₂O₂]_S/[H₂O₂]₀, where [H₂O₂]₀ was the initial H₂O₂ concentration, and [H₂O₂]_S was the H₂O₂ concentration effectively activated to ˙OH (eqn (10)). Yang and co-workers measured the generated ˙OH concentration using terephthalic acid as a fluorescence probe on a fluorescence spectrophotometer, and the corresponding [H₂O₂]_S was determined.⁸² Hui and co-workers introduced another method to evaluate [H₂O₂]_S.⁸¹ Only effectively activated H₂O₂ was responsible for pollutant mineralization. According to eqn (11), 36 mol of H₂O₂ was consumed for complete mineralization of 1.0 mol of TC. By measuring the total organic carbon value before and after the Fenton reaction, the amount of CO₂ produced was determined. Accordingly, the amount of H₂O₂ participating in pollutant mineralization was determined stoichiometrically. In PMS-based Fenton reactions, the PMS activation efficiency was measured by a similar method.⁸³

H₂O₂ + e⁻→OH⁻ + ˙OH (8)

2H₂O₂ → O₂ + 2H₂O (9)

C₂₂H₂₄N₂O₈ + 36H₂O₂ → 22CO₂ + 36H₂O + 2NH₃ (11)

In the next step, the generated ˙OH attacked pollutant molecules to form smaller molecules. Accompanying the degradation reaction, ˙OH underwent other complex reactions, such as coupling and quenching reactions (eqn (12)).⁸⁴ Obviously, the utilization efficiency of H₂O₂ (or ˙OH) was lowered by these side reactions. Additionally, the lifetime of ˙OH was short, and its transfer distance was limited. If a collision with pollutant molecules did not occur within its lifetime, ˙OH would become “dead”. The low pollutant concentration reduced the likelihood of such collisions, further decreasing the utilization efficiency. Usually, H₂O₂ utilization efficiency is defined as [H₂O₂]_A/[H₂O₂]₀ (eqn (13)), where [H₂O₂]_A is the consumed H₂O₂ concentration, which can be determined by the potassium titanyl oxalate method using a UV-Vis spectrophotometer.

˙OH + ˙OH → H₂O₂ (12)

H₂O₂ utilization efficiency was influenced by the activation process and the subsequent degradation process, with low concentrations of H₂O₂ and pollutants being major adverse factors. The advantages of yolk@shell nanoreactor were apparent. In the first step, although the H₂O₂ concentration was low in the bulk solution, it increased dramatically after diffusing into the hollow void. This accelerated the activation process, offering an advantage over the self-decomposition reaction. In the second step, the generated ˙OH was also confined within the cavity, resulting in a high local concentration. Meanwhile, pollutant molecules were enriched on the external shell via the adsorption process. According to Le Chatelier's principle, the degradation process was accelerated, and the H₂O₂ utilization efficiency was increased.

Tailoring catalytic mechanism

Usually, ˙OH and SO₄˙⁻ played the dominating role in pollutant degradation, while the contributions of ˙O₂⁻ and ¹O₂ were limited. Nevertheless, the catalytic mechanism could be altered inside the nanoreactor due to the ultra-high concentrations of ROSs in the confined void space. According to eqn (14)–(16), the transformation from ˙OH to ¹O₂ was possible under such high concentrations. Although the oxidation potential of ¹O₂ was lower than that of ˙OH and SO₄˙⁻, it was still sufficient for the degradation of most pollutants. The lifetime of ¹O₂ was much longer (2.0–3.5 μs), and its transfer distance was greater.⁸⁵ Therefore, PMS utilization efficiency was typically improved in its presence.

˙OH + ˙OH → H₂O + ¹O₂ (14)

˙OH + ˙O₂⁻ → ¹O₂ + H₂O (15)

h⁺ + ˙O₂⁻ → ¹O₂ (16)

The type of ROSs can be identified by electron spin resonance (ESR) analysis in the presence of spin-trapping agents. 5,5-Dimethyl-1-pyrroline N-oxide (DMPO) is employed as the spin-trapping agent for ˙OH and ˙O₂⁻. The characteristic intensity ratio of DMPO-˙OH was 1 : 2 : 2 : 1, while that of DMPO-˙O₂⁻ was 1 : 1 : 1 : 1. 2,2,6,6-Tetramethylpiperidine (TEMP) is used as the spin-trapping agent to detect ¹O₂, and the intermediate TEMP-¹O₂ exhibits a characteristic intensity ratio of 1 : 1 : 1 in ESR analysis.⁸⁶ The contribution of ROSs to pollutant degradation can be evaluated by radical trapping experiments, where benzoquinone served as the scavenger for ˙O₂⁻, isopropanol served as the scavenger for ˙OH only, while methanol functioned as the scavenger for both ˙OH and SO₄˙⁻. Hui and co-workers reported that the transformation from ˙OH to ¹O₂ was significantly accelerated inside the FeS₂/MoS₂@SiO₂ yolk@shell nanoreactor due to confinement effect.⁸¹ Liu and co-workers reported a nanoreactor in which SO₄²⁻-modified Co₃O₄ nanoparticles were encapsulated inside a Co₂SiO₄ shell (Fig. 5a). Under the confinement effect, the concentrations of generated ROSs were boosted inside the cavity, resulting in a rapid transformation of SO₄˙⁻ + ˙OH to ¹O₂.⁸⁷

Fig. 5 (a) Possible catalytic mechanism inside the yolk@shell SO₄-Co₃O₄@Co₂SiO₄ nanoreactor.⁸⁷ (b) Schematic of the fabrication procedure for SCYSN and MCYSN.²² Reproduced from ref. 87 with permission from Elsevier, copyright 2025. Reproduced from ref. 22 with permission from Elsevier, copyright 2023.

Suppressing metal leaching

Metal leaching is an unavoidable problem in the Fenton reaction. When a heterogeneous reaction is performed, metal ions are leached from the catalyst surface into the reaction solution, resulting in secondary pollution. The catalytic activity is gradually reduced or even completely lost after long-term use. Moreover, these leached ions can initiate a homogeneous Fenton reaction. The activation efficiency of the homogeneous reaction is much higher than that of its heterogeneous counterpart, and even a small amount of metal ions can significantly accelerate the degradation rate. This results in the illusion that the heterogeneous Fenton reaction is accelerated. Therefore, the leached metal ion concentration should be analyzed by inductively coupled plasma measurement and checked to determine whether it meets the discharge standard. Moreover, the contribution of the homogeneous reaction to pollutant degradation should be evaluated and compared with that of the heterogeneous one to determine the dominant reaction.⁸⁸

At present, the most effective method to suppress metal leaching is coating a shell on the catalytic core surface. In a yolk@shell structure, the permeable shell not only allows the free diffusion of reactants (or products) into (or out of) the hollow cavity, but also protects the core from direct erosion by the bulk solution. On the other hand, the leached metal ions from the inner core surface are confined inside the nanoreactor. Sometimes, they are adsorbed on the inner surface of the shell. Therefore, the detected metal ion concentration in the bulk solution is decreased. Compared to core@shell composites, the yolk@shell nanoreactor is an ideal candidate to lower metal leaching without sacrificing catalytic activity. Hui and co-workers showed that the degradation efficiencies of ZIF-67 and ZIF-H₂O-350-2 in the homogeneous Fenton reaction were 28.1% and 50.9%, respectively. However, due to the protective effect of the Co₂SiO₄ shell, the homogeneous reaction degradation efficiencies of ZIF@Co₂SiO₄ and ZIF@Co₂SiO₄-350-2 decreased to 29.1% and 35.1%, respectively.⁸⁹ Wang and co-workers indicated that Fe/C achieved only 41.3% of BPA degradation efficiency, and the iron leaching was as high as 0.57 mg L⁻¹. In contrast, Fe/C@mSiO₂ displayed better performance in the electro-Fenton reaction, and the leaching was only 0.11 mg L⁻¹.⁹⁰ Most yolk@shell nanoreactors can maintain their integrated structure even after several reaction cycles. The degradation efficiency of MB over a yolk@shell Co₃O₄@Fe₃O₄/C nanoreactor still reached 98.0%, and the iron leaching was only 0.18 mg L⁻¹ after 5 cycles.⁹¹ The TC degradation efficiency over Fe₃O₄/PBA@PPy remained at 84.5%, and the cobalt leaching was only 0.174 mg L⁻¹ after 4 cycles.³⁸

Improving environmental adaption

Tolerance against anions. In natural water bodies, ubiquitous anions pose a significant influence on degradation performance. Inorganic anions, including Cl⁻, NO₃⁻, SO₄²⁻, CO₃²⁻, HCO₃⁻ and PO₄³⁻, usually serve as scavengers for ROSs and interfere with pollutant degradation (Fig. 6). In addition, they can occupy the catalyst surface, resulting in unavailable active sites for H₂O₂ or pollutant molecules. In most cases, the removal efficiencies are reduced or even completely suppressed. In rare cases, the degradation rate is accelerated, as some inorganic anions function as promoters of the reaction.

Fig. 6 Inorganic anions serve as scavengers for ˙OH to generate other ROSs.

Cl⁻ can produce reactive chlorine species in aqueous solution, such as Cl˙, Cl₂˙⁻ and Cl₂. These reactive chlorine species quench ˙OH under acidic conditions, leading to a negative impact on the degradation rate (eqn (17)–(21)). Although the oxidation potentials of Cl˙ (2.4 eV) and Cl₂˙⁻ (2.0 eV) are lower than that of ˙OH (2.8 eV), they are higher than that of O₂˙⁻ (1.8 eV). Some studies have reported an accelerated degradation rate in the presence of Cl⁻.⁹²

Cl⁻ + ˙OH → HClO˙⁻ (17)

HClO˙⁻ + H⁺ → Cl˙ + H₂O (18)

HClO˙⁻ + Cl⁻ → Cl₂˙⁻ + OH⁻ (19)

Cl˙ + Cl⁻ → Cl₂˙⁻ (20)

Cl˙ + Cl˙ → Cl₂ (21)

˙OH can be consumed by CO₃²⁻ (or HCO₃⁻) to form CO₃˙⁻ (or HCO₃˙), which have low oxidation potentials (eqn (22) and (23)).⁹³ Recently, some researchers have noted the positive aspects of CO₃˙⁻ (or HCO₃˙), such as the low cost of raw materials and the environmentally friendly nature of these radicals, and CO₃˙⁻-based AOPs have been explored for pollutant removal.

HCO₃⁻ + ˙OH → CO₃˙⁻ + H₂O (22)

CO₃²⁻ + ˙OH → CO₃˙⁻ + OH⁻ (23)

NO₃⁻/NO₂⁻ could react with ˙OH to form NO₃˙/NO₂˙ (eqn (24) and (25)), leading to a retarding effect on pollutant degradation due to their low oxidative capacity. To date, NO₃⁻ has shown a negative influence on ˙OH-based AOPs in all reports, and the suppression effect becomes more pronounced with increasing concentrations.

NO₃⁻ + ˙OH → NO₃˙ + OH⁻ (24)

NO₂⁻ + ˙OH → NO₂˙ + OH⁻ (25)

PO₄³⁻ can react with ˙OH to form H₂PO₄˙⁻/H₂PO₄˙/PO₄˙²⁻ (eqn (26)–(28)). Compared to the above-mentioned inorganic anions, the retarding effect of PO₄³⁻ is the most pronounced, and the degradation process is dramatically suppressed in ˙OH-based AOPs.

˙OH + PO₄³⁻ → PO₄˙²⁻ + OH⁻ (26)

˙OH + HPO₄²⁻ → HPO₄˙⁻ + OH⁻ (27)

˙OH + H₂PO₄⁻ → H₂PO₄˙ + OH⁻ (28)

The size of inorganic anions is small enough to diffuse into the cavity through the porous channels on the yolk@shell nanoreactor. During the diffusion process, they may be adsorbed on the external shell surface and the channel surface, especially when these surfaces are positively charged. Therefore, the influence of residual anions entering the cavity is slight. Moreover, the anion concentration in the bulk solution is low, while the concentrations of both pollutants and produced radicals are increased inside the cavity. As a result, the influence of anions on the degradation process is reduced.

To study environmental adaption, the anionic solution is usually pre-prepared in the laboratory and then added to the reaction system to investigate pollutant degradation performance. Only one kind of anions is typically examined at a time. Nevertheless, this deviates from real conditions, since multiple anions co-exist in natural water bodies. Additionally, the concentrations of anions vary across different studies, making it difficult to establish a standard for assessing impact. An alternative and more acceptable experiment involves preparing the pollutant solution using local natural water sources (such as lakes, rivers, oceans),⁹⁴ in which the possible anions can be identified by ion chromatography. This method is closer to real-world applications. Zhao and co-workers reported a yolk@shell nanoreactor (MCYSN) with multiple cores, synthesized by thermolysis of ZIF-67@SiO₂ precursor (Fig. 5b). To test the anti-interference ability of MCYSN, TC was added to natural water as a reference solution. Compared to the solution prepared with deionized water, the TC removal rate in natural water was even increased by 2–3 times in the MCYSN + PMS system. Moreover, after the individual addition of H₂PO₄⁻, HA, Cl⁻ and HCO₃⁻ into the TC solution prepared with deionized water, the removal efficiency was only slightly affected, owing to the shell protection.²²

Tolerance against solution pH values. Considering that the influence of solution pH on the non-radial activation path in PMS-based Fenton reactions is slight, herein, we mainly discuss H₂O₂-based Fenton reactions and the radical activation path in PMS-based Fenton reactions.

The activation efficiency of oxidants and the degradation rate are largely influenced by solution pH. Under strongly acidic conditions (pH < 3.0), metal catalysts can be etched by H⁺, leading to severe metal leaching.⁹⁵ In contrast, under strongly alkaline conditions (pH > 12.0), metal hydroxides with inferior activity are formed and deposited on active sites, resulting in reduced activation efficiency.⁹⁶ Besides the catalysts, the radicals themselves are also ineffective under extreme pH conditions. ˙OH tends to react with H⁺ to form H₂O in strongly acidic media, rather than degrading pollutants,⁹⁷ while its oxidation potential is significantly lowered under strongly alkaline conditions.⁵² Based on the above analysis, it can be rationally inferred that catalysts usually exhibit satisfactory activity under nearly neutral to weakly acidic conditions. For ˙OH, the effective working pH range is between 2.0 and 4.0.⁹⁸ The working range of SO₄˙⁻ is slightly broader, with satisfactory activity in the pH range of 2.0 to 9.0.⁹⁹

In yolk@shell nanoreactors, the Fenton reaction primarily takes place inside the cavity. During pollutant degradation, various intermediates, including H⁺, were generated. These products can locally alter the pH inside the nanoreactor, creating a microenvironment with pH values distinct from those of the external bulk solution. Benefiting from this microenvironment, the pH resistance of the system is enhanced, and the effective operating range is widened. Ma and co-workers synthesized Co₃O₄@NC (Fig. 7a).¹⁰⁰ The Co₃O₄@NC + PMS system exhibited exceptional stability across a wide pH range, with minimal changes in the TC degradation rate observed within the pH range of 3.0 to 9.0.

Fig. 7 (a) Possible catalytic mechanisms over the Co₃O₄@NC catalyst for TC removal.¹⁰⁰ (b) Catalytic mechanism on the CoFe@NC surface.¹⁰² (c) Schematic of the catalytic mechanism toward MTZ over Co_xO_y@Co₂SiO₄/SiO₂-48.³³ Reproduced from ref. 100 with permission from Elsevier, copyright 2023. Reproduced from ref. 102 with permission from Elsevier, copyright 2024. Reproduced from ref. 33 with permission from Elsevier, copyright 2022.

Tolerance against organic matter. Compared to anions and pH values, the anti-interference property of yolk@shell nanoreactors against the organic matter molecules is much more pronounced. Humic acid (HA) is one of the most common organic substances in natural water sources, and its influence is frequently investigated in research studies (Fig. 6). Generally, HA serves as a scavenger for all major ROSs. Therefore, strategies to prevent its interference with ROSs need to be addressed. Unlike anions, the size of HA molecules is relatively large (2.0–3.0 μm).¹⁰¹ In the yolk@shell structure, the catalytic core is confined inside a shell whose pore sizes are only several nanometers. In this case, the transport of reactants (or products) is permitted, while large-sized HA molecules are excluded from diffusing inward. Consequently, in Fenton reactions catalyzed by yolk@shell nanoreactors, the influence of HA is nearly negligible. Ding and co-workers reported that degradation performance of the methylene blue (MB) over the CoFe@NC yolk@shell nanoreactor was minimally affected even in the presence of 10 mM HA solution (Fig. 7b).¹⁰²

Maximizing catalytic activity of the core in yolk@shell nanoreactors

The yolk@shell nanoreactor consists of only two components: the core and the shell. In this section, we illustrate recent progress on the core materials, and in the following section, methods to improve shell performance are discussed.

Metal oxides

Transition metal oxides are commonly used as cores in yolk@shell nanoreactors. Because of their uniform shape and good stability, the hard-template-assisted etching method is frequently employed for nanoreactor preparation. Cui and co-workers synthesized Fe₂O₃@mesoporous SiO₂ yolk@shell nanoreactors, in which a spindle-shaped Fe₂O₃ core was confined inside a mesoporous SiO₂ shell for MB degradation.¹⁰³ The void space was easily controlled by the thickness of a middle carbon layer, leading to variable degradation rates. A degradation efficiency of 90% at 7.0 h was achieved using nanoreactors with a 40 nm void space, whereas only 20% of MB was removed by pure Fe₂O₃.

In addition to monometallic oxides, bimetallic oxides and alloys can also serve as catalytic cores and exhibit enhanced activity in Fenton reaction, owing to redox reactions between different metal sites. For example, the reduction of Fe(III) to Fe(II) in NiFe₂O₄ was accelerated (eqn (29)) due to the difference in redox potentials between Fe(III)/Fe(II) (∼0.77 eV) and Ni(III)/Ni(II) (∼0.48 eV).⁷ Andikaey and co-workers reported a trimetallic CoNiSe₂@Fe-CoNiSe₂ yolk@shell nanoreactor, prepared using ZIF-67@NiCoFe Prussian blue analogue (PBA) as precursors.⁹³ Benefiting from redox reactions among Ni(III)/Ni(II), Co(III)/Co(II) and Co(IV)/Co(III), the CoNiSe₂@Fe-CoNiSe₂ nanoreactor exhibited excellent catalytic activity.

Fe(III) + Ni(II) → Fe(II) + Ni(III) (29)

Some shortcomings existed in the metal oxide cores, such as the slow rate-limiting step and lack of low valence metal sites. Nonmetal doping was a simple and effective method to further enhance the activity. The electrons could transfer from doped nonmetal atoms to the metal atoms, which was beneficial for the reduction of high valence metal sites to the low valence counterparts, and the acceleration of the rate-limiting step. Ding and co-workers introduced B atoms into the Cu_xO lattice and partially replaced O atoms (B-Cu_xO).¹⁰⁴ The electronegativity of element B was lower than that of element O, and only one electron was on their 2p orbital, making them function as an electron donor. Compared to the pure Cu_xO, the recycling of Cu(II)/Cu(I) was accelerated, due to the acceptance of electrons from doped B atoms. In the PMS-based Fenton reaction, the degradation rate was increased by 12 times, from 0.023 min⁻¹ (pure Cu_xO) to 0.267 min⁻¹ (B-Cu_xO). On the other hand, many defects were created on the metal oxide surface during the doping process. Zuo and co-workers synthesized B atom-doped NiFe₂O₄ (B-NiFe₂O_x) by the sol-gel method.¹⁰⁵ Compared to NiFe₂O_x, the formation of B–O–Fe bonds and B–O–Ni bonds in B-NiFe₂O_x destroyed the equilibrium of the original Ni–O–Fe bond. Due to the asymmetric radius, the oxygen vacancies (OVs) were generated by the local vibration of the B–O–Fe bond and B–O–Ni bond. OVs could serve as an electron transfer station for free electrons, which accelerated the charge transfer rate in catalytic reactions. In addition, the appearance of OVs led to uncoordinated metals on the catalyst surface, which served as Lewis acid sites. PMS was a well-known Lewis base, and the Fenton reaction was accelerated via a Lewis acid-base reaction. In this study, B-NiFe₂O_x could remove 85.7% of 2,4-dichlorophenoxyacetic acid within 15 min, while only 59.3% of removal efficiency was achieved over pure NiFe₂O₄.

Metals and alloys

Compared to metal oxides, the electron supply property of zero-valent metals is better. Therefore, they could also be confined in the hollow capsule for a nanoreactor with higher activity. Due to the high chemical activity, the composition of zero-valent metal was usually destroyed during the nanoreactor preparation process. Therefore, they could not be directly used as the hard template. Generally, the yolk@shell nanoreactor with metal oxides as the core was firstly prepared, and then the metal oxide was reduced to the corresponding zero-valent metal under an H₂ atmosphere at high temperature. Yang and co-workers first anchored Fe₃O₄ nanoparticles on the inner surface of SiO₂ nanotube, which were subsequently reduced to Fe⁰@SiO₂ nanotube under an H₂ atmosphere.¹⁰⁶

Considering the danger of explosive H₂, a pre-shell/post-core approach was established, which involved the formation of a zero-valent metal core within a preformed hollow capsule.¹⁰⁷ Liu and co-workers first prepared a hollow mesoporous SiO₂ capsule, then FeSO₄ aqueous solution was introduced into the capsule dispersed in n-hexane via a “two-solvent” impregnation method.¹⁰⁸ Finally, the encapsulated Fe₂SO₄ was reduced to Fe⁰ in the NaBH₄ solution under a N₂ atmosphere. In the H₂O₂-based Fenton reaction, 99.9% of phenol was removed over the nanoreactor within 45 min, while 180 min was required to realize the same efficiency over bare Fe⁰.

Sometimes, the carbon materials can be used to reduce metal oxides under an N₂ atmosphere at high temperature. In Wang's work, the FeCo-PBA nanocubes were firstly synthesized, and used for electrospinning of acrylonitrile.¹⁰⁹ During the calcination process under an N₂ atmosphere, polyacrylonitrile was carbonized and reduced the encapsulated PBA to FeCo alloys. As a result, a necklace-like one-dimensional structures composed of FeCo@NC yolk@shell nanoreactors were prepared. In the PMS-based Fenton reaction, nearly 100% of bisphenol A (BPA) was removed within 7.0 min over the FeCo-based nanonecklace, which was more rapid than the Fe-based nanonecklace and the Co-based nanonecklace.

MOFs

Traditional MOFs. MOF are a kind of three-dimensional inorganic-organic hybrid, which are formed by interlinking metal ions with organic linkers. The metal centers can serve as catalytic sites and display excellent performance in catalytic reactions, due to their high surface area, abundant porosity and crisscrossed channels. Numerous MOFs have been employed as catalysts in the Fenton reaction. At present, Fe-based MOFs are usually applied in the H₂O₂-based Fenton reaction, such as MIL-101(Fe) and MIL-88B(Fe). Co-based MOFs exhibit excellent activity in the PMS-based Fenton reaction, such as ZIF-67 and Co-MOF-74.¹¹⁰ Prussian blue is a famous Fe-based MOF, in which Fe(III) or Fe(II) nodes are bridged with CN⁻. After replacing partial Fe(III) (or Fe(II)) with Co(III) (or Co(II)), PBAs are obtained, and they present excellent activity in both the H₂O₂-based Fenton reaction and the PMS-based Fenton reaction. Chen and co-workers prepared [Fe(CN)₆]-based or [Co(CN)₆]-based PBAs with other metallic ions (i.e., Co, Cu, Fe, Mn and Ni).¹¹¹ Co-based or Mn-based PBAs were capable of PMS activation, while the activation performance of Cu-based, Fe-based or Ni-based PBAs was unsatisfactory.

It was difficult to obtain a yolk@shell structure with an MOF as core by the hard-template assisted etching method, as their integrated structure was easily destroyed during the preparation procedure, especially in the etching process. For some MOFs, even H₂O could serve as the etching agent. Zhang and co-workers prepared yolk@shell ZIF-67@CoSiO₄ nanoreactor in Fig. 7c.³³ At the beginning, ZIF-67 was employed as the hard template. Then, an SiO₂ shell was coated on the ZIF-67 surface by tetraethyl orthosilicate condensation. Meanwhile, ZIF-67 was etched by H₂O slowly, and the dissolved Co²⁺ diffused into the aqueous solution. Under alkaline conditions, the SiO₂ shell was etched, and the released SiO₃²⁻ reacted with Co²⁺ to form Co₂SiO₄. The ZIF-67 core was gradually etched from surface to interior, while the Co₂SiO₄ shell kept growing outward, resulting in a void between the residual ZIF-67 core and the Co₂SiO₄ shell. ZIF-67@Co₂SiO₄ yolk@shell nanoreactor was prepared and applied in the PMS-based Fenton reaction.

Some alternative methods were established to encapsulate MOFs inside a hollow capsule. Zhao and co-workers first prepared Fe₃O₄@polypyrrole (PPy) yolk@shell nanoreactor via the hard-template-assisted etching method.³⁸ Then, the Fe₃O₄ core was functioned as a reactive template in the presence of HCl and K₃[Co(CN)₆], since the released Fe^3+/2+ reacted with [Co(CN)₆]³⁻ to form a PBA nanocube inside the PPy shell. The PBA@PPy nanoreactor displayed better activity than the Fe₃O₄@PPy yolk@shell precursor in the PMS-based Fenton reaction.

MOFs with CUMSs. Although MOFs have already displayed the outstanding activity in the Fenton reaction, there is large space for further improvement. In MOFs, the metal nodes are fully coordinated by organic ligands, resulting in limited contact between oxidants and pollutants. From this viewpoint, the ligands act as barriers to enhanced catalytic activity. Therefore, engineering coordinately unsaturated metal sites (CUMSs) via partial removal of ligands has become an effective approach to increase the number of exposed metal sites.¹¹² It is easily accepted that the more exposed metal sites, the better the activity. Usually, the coordinate bond is kinetically unstable in MOFs.¹¹³ During the CUMS engineering process, the integrated MOF structure and crystal phase should not be destroyed. There are many methods to create CUMSs, including solvent exchange thermal activation,¹¹⁴ heat treatment,¹¹⁵ photoactivation,¹¹⁶ and so on. High-temperature treatment is the most commonly used technique for CUMS engineering. If the temperature is low, only a small amount of ligands is removed, and the number of CUMSs cannot be maximized. Conversely, if the temperature is too high, MOFs are oxidized to metal oxides. This puts forward a high requirement for experimental techniques.

In Zhao's further work, the as-prepared PBA@PPy yolk@shell nanoreactor was calcinated under a moderate temperature.⁸³ Following the temperature rise, the ligands were gradually lost, and more and more CUMSs were generated on the PBA surface. This was verified by Fourier transform infrared spectroscopy, X-ray diffraction patterns and digital camera photos. Meanwhile, the catalytic performance was also gradually improved, and the maximum activity was realized at 350 °C. If the temperature was higher, PBAs were oxidized to CoFe₂O₄, and the activity was lowered. According to Ostwald ripening, the MOF exterior is more compact than the interior, leading to a protective external surface. Wu's research group studied the aforementioned ZIF-67@CoSiO₄ nanoreactor and found that CUMSs were more easily produced on ZIF-67 without an external surface. Under the identical calcination temperature, 20.1% of Co–N bonds were lost in integrated ZIF-67, while 45.3% of Co–N bonds were removed in the sample without an external surface. Therefore, it was a feasible way to maximize the CUMS number by removing the MOF exterior, and 90.1% of MTZ was degraded within only 8.0 min in the presence of the optimized catalysts.⁸⁹

There was an irreconcilable contradiction between catalytic activity and metal leaching. Generally, the more ligands were lost, the more catalytic sites were exposed, the better the activity, and the easier the metal leaching. Therefore, it was difficult to improve catalytic performance and reduce metal leaching on pristine MOFs simultaneously. This issue was well addressed in yolk@shell nanoreactors, as the nanoreactor shell functioned as a barrier toward the outward diffusion of the leached ions. In Hui's work, the homogeneous degradation reaction triggered by leached cobalt ions increased from 28.1% (integrated ZIF-67) to 50.9% in CUMSs-ZIF-67 (Fig. 8a).⁸⁹ After encapsulating CUMSs-ZIF-67 into the CoSiO₄ shell, the degradation efficiency dramatically decreased to 35.1%. In Zhao's work, the concentration of leached cobalt ions from the pure PBA nanocubes was 1.73 mg L⁻¹ (Fig. 8c).⁸³ After creating the CUMSs, the concentration increased to 2.29 mg L⁻¹, due to the lack of protection from the ligands. After encapsulating CUMSs-PBA into the PPy capsule, the concentration was dramatically lowered to 0.34 mg L⁻¹. From these examples, it can be seen that enhanced catalytic performance and reduced metal leaching can be simultaneously realized by encapsulating CUMSs-MOF inside the yolk@shell nanoreactor.

Fig. 8 (a) Schematic of the catalytic mechanism inside the ZIF@Co₂SiO₄-350-2 nanoreactor.⁸⁹ (b) Catalytic mechanism on the CoS₂/NC@SiO₂-48 surface.¹²⁴ (c) Synthetic procedure of the CUMSs-PBA@PPy yolk@shell nanoreactor.⁸³ Reproduced from ref. 89 with permission from Elsevier, copyright 2024. Reproduced from ref. 124 with permission from Elsevier, copyright 2024. Reproduced from ref. 83 with permission from Elsevier, copyright 2023.

Derivatives from MOFs. At present, derivatives from MOFs have drawn much attention. They can be obtained by the simple heat treatment of MOF precursors. Excitingly, their structures and compositions can be easily tailored by changing the heat-treatment environment. For example, after calcining MOFs in air, the organic ligands are completely removed, and the metal oxides are prepared.¹¹⁷ Under an N₂ atmosphere, the ligands are carbonized, and the metal nodes are thermally reduced to the zero-valent metal.¹¹⁸ After placing MOFs in an NH₃ atmosphere, or with sulfur powder, selenium powder, or NaH₂PO₄ powder at high temperature, the corresponding metal nitride,¹¹⁹ metal sulfide,¹²⁰ metal selenide and metal phosphide are prepared.^121,122 Benefiting from their varied structures and chemical compositions, these derivatives display outstanding performance in various application fields, such as electrocatalysis, gas sensing, electromagnetic wave absorption. Zhang and co-workers mixed Co(NO₃)₂, 2-methylimidazole, and tetraethoxysilane (TEOS) together. Compared to the rapid synthesis of ZIF-67, the hydrolysis of TEOS was relatively slow. As a result, ZIF-67@SiO₂ core@shell composites were formed. After calcination in air at 600 °C, ZIF-67 was converted to Co₃O₄, and encapsulated inside the SiO₂ shell. In the PMS-based Fenton reaction, 99.1% of BPA was degraded at 23 min.³³ Tian and co-workers synthesized yolk@shell NiS₂@C spheres by one-step thermal transformation, where Ni-MOFs were used as the precursor for subsequent vulcanization.¹²³

Multiple cores

The Fenton reaction mainly took place on the core surface, and the surface area was a key factor in determining the catalytic activity. Compared with the single core in traditional yolk@shell nanoreactors, multiple cores were beneficial for the enhanced activity due to the greater number of available catalytic sites (Fig. 3b). This was also confirmed by other catalytic reactions.⁶⁹

How can a yolk@shell structure with multiple cores be prepared? It is difficult to simultaneously coat the external shell onto a surface containing multiple cores. Generally, a yolk@shell nanoreactor with a single core is first prepared, and then the single core is decomposed during the preparation process or post-treatment. Ding and co-workers first prepared a PBA@PPy yolk@shell structure. After calcination under an N₂ atmosphere, the cubic PBA core was self-decomposed into multiple FeCo alloys, and the PPy shell was carbonized to form the NC shell.¹²⁴ In the PMS-based Fenton reaction, 43.8% of MB was degraded at 30 min in the presence of PBA@PPy, while the efficiency increased to 93.8% with the FeCo@NC nanoreactor.

Besides self-decomposition, some chemical reactions can be utilized to destroy the integrated MOF into multiple fragments. During the vulcanization process, the metal nodes in the MOF react with sulfur vapor to form metal sulfide, leading to size expansion. In this case, the original shape cannot be maintained, and several metal sulfide nanoparticles are formed. According to this principle, Ding and co-workers reported that multiple CoS₂ nanoparticles were confined inside the SiO₂ shell by vulcanizing the ZIF-67@SiO₂ yolk@shell precursor (Fig. 8b).¹²⁴ Zhang and co-workers designed a yolk@shell Co@C nanoreactor with multiple cores by high-temperature carbonization.⁶⁹ Co₃O₄ was completely reduced by the carbon shell to zero-valent Co at 750 °C.

Although multiple cores can improve catalytic performance, one of the disadvantages is their aggregation inside the cavity, which leads to decreased activity. Zhang and co-workers reported the Co₂SiO₄ nanosheet as a shell (Fig. 7c).³³ Benefiting from the separation provided by the Co₂SiO₄ nanosheets on the shell's inner surface, a maze-like cavity was obtained. This not only prevented the aggregation of multiple cores, but also extended the travel path and retention time of organic pollutants trapped in the cavity, leading to improved catalytic performance.

Diverse functional cores

Sometimes, the multiple cores encapsulated in the nanoreactor were not merely intended to improve the catalytic performance, other functions were also expected to be introduced. In addition to catalytic activity, the recyclability was another important property. Among the various separation approaches, magnetic separation was commonly recognized as the simplest and most effective. Unfortunately, most Fe-based or Co-based MOFs do not exhibit superparamagnetism, such as PB, PBA, MIL-101(Fe), and ZIF-67. To solve this problem, Zhao and co-workers encapsulated both Fe₃O₄ and PBA inside PPy capsules, and a “double-yolk egg-like” Fe₃O₄/PBA@PPy nanoreactor was prepared, where the Co-based PBA displayed outstanding activity, while the Fe₃O₄ core not only contributed to the catalytic activity but also enabled magnetic separation of the catalyst.³⁸ The TC degradation rate over the Fe₃O₄/PBA@PPy nanoreactor (0.38 L min⁻¹ mol⁻¹) was 5.1 times higher than that of the reference Fe₃O₄/PBA without the PPy shell (0.074 L min⁻¹ mol⁻¹). Additionally, the “double-yolk egg-like” nanoreactor could be magnetically separated, and 84.5% of the degradation efficiency was maintained after the 4th run. For comparison, the separation efficiency of the reference PBA@PPy yolk@shell nanoreactor without Fe₃O₄ was inferior.

In 2018, Xing and co-workers reported that MoS₂ nanosheets could function as a co-catalyst for the Fe(III)/Fe(II) cycle.³⁹ Compared to organic co-catalysts, radical consumption over inorganic MoS₂ was avoided, leading to enhanced oxidant utilization. Since then, many studies have focused on accelerating the Fenton reaction in the presence of MoS₂ co-catalysts, and their positive role has been verified.^125,126 Hui and co-workers encapsulated both MoS₂ and FeS₂ inside a mesoporous SiO₂ shell to form the FeS₂/MoS₂@SiO₂ yolk@shell nanoreactor.⁸¹ Compared to the reference nanoreactor broken by ultrasonic wave (named as B-FeS₂/MoS₂@SiO₂), both the activation efficiency and utilization efficiency of H₂O₂ were improved in the integrated nanoreactor. Furthermore, 92.8% of TC was removed by FeS₂/MoS₂@SiO₂, while only 64.1% was degraded by B-FeS₂/MoS₂@SiO₂ under identical condition. This further verified the importance of the confinement effect.

Optimizing performance of the shell in yolk@shell nanoreactors

After illustrating the possible approaches to improve catalytic activity by optimizing the composition and structure of the cores, herein, the methods to exploit the advantages of the shell are illustrated. Various materials can be used as shells, including polymer,¹²⁷ mesoporous SiO₂,¹²⁸ activated carbon,¹²⁹ graphene,¹³⁰ and MOF¹³¹ and others. These shells exhibit diverse functions and greatly influence the performance of the yolk@shell nanoreactor. For example, the adsorption capacity is increased in the presence of a mesoporous SiO₂ shell. The stability is improved by a cross-linked polymer shell. The reaction rate-limiting step of Fe(III)/Fe(II) is accelerated by using the carbon shell with good reducibility.

Separation of adsorption sites and catalytic sites

If transition metal nanoparticles were directly used as catalysts in the Fenton reaction, pollutant adsorption and H₂O₂ activation were performed on the metal surface at the same time. Considering the limited surface area and number of active sites, this competition was not favorable for full exploitation of both active and adsorption sites. The yolk@shell nanoreactor has provided a feasible way to address this issue. Usually, the adsorption property of shell was much better than that of the core, while it did not possess catalytic activity. In this case, pollutant adsorption mainly took place on the shell surface, and H₂O₂ activation was conducted on the inner core surface. Shells with excellent adsorption properties, large surface area and multiple functional groups were preferred, and mesoporous SiO₂ and activated carbon were frequently employed as shells.¹³²

In general, a question is raised after the spatial separation of H₂O₂ activation sites and pollutant adsorption sites. The produced ROSs, with limited lifetime, have to travel a long distance to reach the pollutant molecules. Fortunately, this problem is well resolved by the yolk@shell nanoreactor itself. As mentioned in the Accelerating Degradation Rate section, the transfer distance can be tailored by adjusting the void space during the preparation process. Mei and co-workers prepared a yolk@shell Fe₃O₄@MgSiO₃ nanoreactor (Fig. 9a).²⁹ The MgSiO₃ shell was a good adsorbent for MB, and the local MB concentration on the MgSiO₃ shell was 1.9 times higher than that in solution, according to the Boltzmann equation. For comparison, MB adsorption on the Fe₃O₄ core was limited, and the activation sites on the core surface could be fully utilized for H₂O₂ activation. In the PMS-based Fenton reaction, 96.7% of MB was degraded over the nanoreactor within 120 min. In another work, Hui and co-workers encapsulated FeS₂/MoS₂ composites into a mesoporous SiO₂ shell to form a yolk@shell nanoreactor.⁸¹ The mesoporous SiO₂ shell acted as the primary adsorption site for TC. The TC adsorption capacity on the FeS₂/MoS₂@SiO₂ surface was 13.0 mg g⁻¹, which was significantly higher than that of the FeS₂/MoS₂ composites (3.95 mg g⁻¹). Benefiting from the spatial separation of adsorption sites and activation sites, the TC removal efficiencies of these two catalysts were nearly equal in the H₂O₂-based Fenton reaction. Nevertheless, the leached iron ion concentrations were 0.213 and 2.064 mg L⁻¹ for FeS₂/MoS₂@SiO₂ and FeS₂/MoS₂, respectively, indicating that the degradation contribution from the homogeneous reaction was much higher on the latter catalyst surface.

Fig. 9 (a) Possible removal mechanism in the yolk@shell Fe₃O₄@MgSiO₃ nanoreactor toward MB.²⁹ (b) Catalytic mechanism on the CoO@Co-N-GC surface.³⁰ Reproduced from ref. 29 with permission from Elsevier, copyright 2020. Reproduced from ref. 30 with permission from Elsevier, copyright 2022.

Exploring catalytic shell in radical activation path

In the traditional yolk@shell structure, the catalytic reaction was mainly performed on the core surface. The shell was incapable of catalytic reactions and only provided protective and adsorption functions, leading to the waste of numerous catalytic sites. It was rationally inferred that the catalytic performance would be greatly enhanced if the Fenton reaction could occur on both the shell surface and core surfaces by decorating transition metal nanoparticles on the shell.

There were three different positions on the shell for loading nanoparticles: the external surface, the inner surface and embedded within the shell.³⁷ When anchored on the external surface, the nanoparticles were directly exposed to the reaction solution. In this case, although the catalytic performance was improved, detachment and leaching were unavoidable. When the nanoparticles were decorated on the shell inner surface, detachment and subsequent aggregation remained major issues after long-time use. Compared to the above two positions, embedding nanoparticles in the shell was more favorable. Although the approach did not maximize catalytic activity, it effectively prevented detachment and aggregation, and also suppressed metal leaching.

Based on the above statements, Zeng and co-workers prepared a yolk@shell nanoreactor via a solvothermal method, in which large-sized Fe₃O₄ nanoparticles (∼200 nm) served as the core, and a carbon shell embedded with small-sized Fe₃O₄ nanoparticles (several nanometers) served as the shell.¹³³ In the H₂O₂-based Fenton reaction, the 4-chlorophenol removal efficiency reached nearly 100% at 120 min, which was significantly better than that of two reference samples: pure Fe₃O₄ and SiO₂@Fe₃O₄/C yolk@shell nanoreactor with SiO₂ as the core. In another study, Yang and co-workers prepared Co₃O₄@Fe₃O₄/C yolk@shell nanoreactor, in which Fe₃O₄ nanoparticles were embedded in the carbon shell.⁹¹ In the PMS-based Fenton reaction, MB degradation efficiencies at 30 min over Co₃O₄ nanoparticles, hollow Fe₃O₄/C capsules and Co₃O₄@Fe₃O₄/C nanoreactors were 43%, 74% and 100%, respectively. The enhanced degradation performance of the nanoreactor was ascribed to the synergistic effect between the core (Co₃O₄) and the shell (Fe₃O₄/C capsule). Moreover, the leached iron ions in each cycle were less than 49 μg L⁻¹, verifying the good protective function of the carbon shell toward the embedded Fe₃O₄ nanoparticles.

MOFs could function not only as cores, but also shells, enabling the Fenton reaction to occur on both the core and shell surfaces within the nanoreactor. Niu and co-workers encapsulated core@shell Pd@Fe₃O₄ composites within a hollow Fe-based MOF capsule and applied them in H₂O₂-based Fenton reaction. Benefiting from the abundant catalytic sites on both the MOF shell and Fe₃O₄ core, the total organic carbon removal reached 85%, which was much higher than that achieved with either the hollow Fe-based MOF shell (43%) or the Pd/Fe₃O₄ core (36%).¹³⁴

Exploring catalytic shell in non-radical activation path

In the above section, the shell participated in the catalytic reaction upon the introduction of transition metal nanoparticles. In that case, both PMS-based and H₂O₂-based Fenton reactions were activated via the radical path on the shell surface. As mentioned in the Non-radical Activation Path section, PMS can be activated via both the radical path on the transition metal surface and the non-radical path on the heteroatom-doped carbon surface. Carbon materials are among the most frequently used shell materials due to their biocompatibility, large surface area, excellent stability and low cost. Therefore, yolk@shell nanoreactors with a transition metal core and heteroatom-doped carbon shell could activate PMS via both radical and non-radical activation paths simultaneously. These two activation paths could complement each other's strengths and weaknesses. (i) Leached metal ions from the heteroatom-doped carbon shell were negligible, thus resolving the issue of shell-related metal leaching. (ii) The non-radical activation path possessed high environmental tolerance, whereas the radical path was susceptive to the environmental factors. A microenvironment created by the heteroatom-doped carbon shell significantly reduced external interference. Furthermore, the metal leaching from the core was also suppressed by the shell's protective function. (iii) Benefiting from the microenvironment, the catalytic activity of the encapsulated core was enhanced via the confinement effect, compensating for the relatively low catalytic activity of the heteroatom-doped carbon shell.

When the non-radical activation path was performed on the external shell surface and the radical activation path on the inner core surface, the components of the yolk@shell nanoreactor were fully utilized, the oxidant activation efficiency was maximized, and the degradation performance was significantly enhanced. Ma and co-workers prepared a yolk@shell nanoreactor with Co, N-co-doped carbon as the shell and multiple hollow Co nanoparticles as the cores by combing the chelation-competition-induced polymerization method and the Kirkendall effect (Fig. 9b).³⁰ A total of 80.0% of TC at a concentration of 50 mg L⁻¹ was degraded within 40 min. The PMS utilization efficiency reached as high as 95.8%, while cobalt ion leaching was only 0.166 mg L⁻¹. Most importantly, the outstanding catalytic performance was maintained across the entire pH range (pH = 0.0–14.0) due to the protective function of the external shell, which greatly extended its environmental adaptability.

Application of nanoreactor in Fenton reaction coupled with external energy

Besides their application in the H₂O₂-based or PMS-based Fenton reactions, yolk@shell nanoreactors could also be employed in the Fenton reactions coupled with external energy. The photo-Fenton reaction is a prominent representative of AOPs, in which light irradiation and transition metals are combined to activate oxidants. At present, most photo-Fenton reactions are carried out in open environments. Their performance could also be improved by using yolk@shell nanoreactors. Du and co-workers prepared Fe₃O₄@TiO₂ yolk@shell nanoreactors using a hard-template-assisted etching method (Fig. 10a). In the photo-Fenton reaction, the TC removal efficiency reached nearly 100% within 6.0 min under ultraviolet light and 377 mM H₂O₂, whereas the removal efficiency over pure Fe₃O₄ nanoparticles was only 44%.¹³⁵ Deng and co-workers prepared an FeS₂/MoS₂@C yolk@shell nanoreactor using MIL-101(Fe) as a hard template via the chelation-competition-induced polymerization method. The metronidazole degradation efficiency at 25 min reached 90.5% under visible light and 20 mM H₂O₂, while the values for the FeS₂/MoS₂ core and the broken FeS₂/MoS₂@C nanoreactor were19.2% and 26.5%, respectively.⁷⁹

Fig. 10 (a) Possible mechanism proposed for photo-Fenton degradation of TC over Fe₃O₄@TiO₂ yolk@shell nanoreactors.¹³⁵ (b) Proposed mechanism for the Fe/C@mSiO₂-catalyzed heterogeneous electro-Fenton reaction for BPA degradation.⁹⁰ Reproduced from ref. 135 with permission from Elsevier, copyright 2017. Reproduced from ref. 90 with permission from Elsevier, copyright 2024.

In the electro-Fenton reaction, the input current could accelerate oxidant activation and the reduction of high-valence metal sites, leading to improved degradation performance. Catalysts with high conductivity and electroactivity are preferred in electro-Fenton reactions. Wang and co-workers prepared an Fe/C@mSiO₂ yolk@shell nanoreactor, in which the Fe/C core was derived from an MOF (Fig. 10b).⁹⁰ BPA was completely removed within 120 min under a 100 mA current, whereas the degradation efficiency over the Fe/C core alone was only 41.3%.

Conclusion and prospects

The heterogeneous Fenton reaction is a promising method for the non-selective degradation of organic pollutants. Unlike traditional catalytic reactions, the Fenton reaction and the subsequent degradation process can be regarded as a cascade reaction, in which catalyst design plays a critical role. Initially, research attention was mainly focused on catalytic activity. As studies progressed, other important characteristics were also taken into consideration, including oxidant utilization efficiency, environmental adaptability, metal leaching, and so on. This has placed higher demands on catalyst design. Fortunately, yolk@shell nanoreactors have provided a feasible solution to these challenges simultaneously. Benefiting from the confinement effect, catalytic activity was enhanced and the pollutant degradation rate was accelerated. Oxidant utilization efficiency was increased by exploiting the active sites on both the shell and the core. Environmental tolerance was improved, and metal leaching was reduced under the protection of the external shell. The advantages of yolk@shell nanoreactors in Fenton reactions have been comprehensively summarized in this review.

Although yolk@shell nanoreactors have attracted much interest, several problems still need to be resolved. As the nanoreactor is composed of a core and a shell, its performance largely depends on the development both core and shell materials, and there is still considerable room for improving catalytic activity. One the other hand, a yolk@shell nanoreactor is not a simple mixture of core and shell; the two components should complement each other's strengths and weaknesses. A representative example is a yolk@shell nanoreactor with a radical activation path occurring on the inner core and a non-radical activation path occurring on the external shell. However, such examples remain rare. Moreover, the Fenton reaction takes place within the cavity, and the hollow void is also important for the final degradation performance. Nevertheless, its influence has been largely overlooked, and no published studies have yet addressed this aspect.

The catalytic mechanism of the Fenton reaction performed in bulk solution is commonly recognized as highly complex. In many cases, the ROS generation path and transfer route are only deduced rather than direct observed. After confining the reaction within the nanoreactor, it becomes even more difficult to detect the radical types, radical concentrations and radical reactions. What actually occurs inside the nanoreactor remains a mystery. This challenge is highly dependent on advances in the development of instrument science and technology. At present, theoretical calculations should be carried out to obtain more critical information, such as catalytic sites, radical generation paths and chemical reaction mechanisms. These insights are expected to play an increasingly important role in guiding catalyst design.

Novel synthetic methods for yolk@shell nanoreactors urgently need to be explored. Currently, even the widely used hard-template-assisted etching method faces challenges in preparing nanoreactors for Fenton reactions. For example, MOFs cannot be effectively encapsulated using this method. As discussed in this review, non-typical nanoreactors often exhibit better performance than their traditional counterparts (i.e., those with a single core and single shell). Therefore, nanoreactors with novel structures and diverse functionalities should be developed. For example, nanoreactors capable of in situ H₂O₂ production within the hollow cavity represent a promising direction.

Recently, most studies on yolk@shell nanoreactors have focused on the H₂O₂-based or PMS-based Fenton reactions, while their application in Fenton reactions coupled with external energies remain scarce. It is rationally inferred that yolk@shell nanoreactors could be applied in Fenton reactions coupled with other energy sources (microwave-Fenton, sono-Fenton reactions, etc.); however, research in these areas is still limited. This suggests that more studies should be conducted in the future to fill this gap.

Although great progress has been made in the development of yolk@shell nanoreactors, their scale-up production still faces many challenges. Precise control over the preparation conditions is essential for the successful fabrication of yolk@shell nanoreactors. However, such precision is difficult to achieve in industrial-scale production. From an economic perspective, the current cost of yolk@shell nanoreactors is prohibited, as the prices of components such as MOFs, mesoporous SiO₂ and metal oxides are still high. Additionally, the number of times the catalysts can be reused in laboratory settings is usually fewer than 10, which is far from sufficient for real-world applications. To overcome these issues, it is necessary to lower production costs, update processing techniques, and improve activation efficiency. At present, research on yolk@shell structures remains at the laboratory scale, and there is still a long way to go before their practical application can be realized.

Although many achievements have been made, there is still significant room for further improvement, and numerous challenges remain to be addressed. We believe that yolk@shell nanoreactors could provide a platform for the development of new technologies and strategies in Fenton reactions. Moreover, Fenton reactions catalyzed by yolk@shell nanoreactors hold great promise as an efficient, low-cost and sustainable wastewater treatment technology.

Conflicts of interest

There are no conflicts to declare.

Data availability

Data availability is not applicable to this article as no new data were created or analyzed in this study.

Acknowledgements

This work was supported by the Natural Science Foundation of Heilongjiang Province, China (PL2024B019), and State Key Laboratory of Urban-rural Water Resources and Environment (Harbin Institute of Technology, No. 2025TS45).

References

1. M. Hubert, T. Meyn, M. C. Hansen, S. E. Hale and H. P. H. Arp, Per- and polyfluoroalkyl substance (PFAS) removal from soil washing water by coagulation and flocculation, Water Res., 2024, 249, 120888,  DOI:10.1016/j.watres.2023.120888.
2. K. Chen, X. Huang, G. Zhu, H. Pang, J. Lu and Z. Zhang, Unraveling the triple mechanisms of advanced coagulation for removal of emerging and conventional contaminants: Oxidation, hydrolytic coagulation and surface hydroxylation adsorption, Chem. Eng. J., 2024, 484, 149473,  DOI:10.1016/j.cej.2024.149473.
3. S.-Y. Woo, J.-S. Kim and Y.-D. Kim, Assessment of the suitability of adsorption water treatment as a concentration process for resource recovery and wastewater treatment, Sep. Purif. Technol., 2024, 335, 126177,  DOI:10.1016/j.seppur.2023.126177.
4. C. Xia, Y. Liang, X. Li, H. A. Garalleh, M. Garaleh, J. M. Hill and A. Pugazhendhi, Remediation competence of nanoparticles amalgamated biochar (nanobiochar/nanocomposite) on pollutants: A review, Environ. Res., 2023, 218, 114947,  DOI:10.1016/j.envres.2022.114947.
5. L. Liu, Z. Chen, J. Zhang, D. Shan, Y. Wu, L. Bai and B. Wang, Treatment of industrial dye wastewater and pharmaceutical residue wastewater by advanced oxidation processes and its combination with nanocatalysts: A review, J. Water Process Eng., 2021, 42, 102122,  DOI:10.1016/j.jwpe.2021.102122.
6. S. Xiao, M. Cheng, H. Zhong, Z. Liu, Y. Liu, X. Yang and Q. Liang, Iron-mediated activation of persulfate and peroxymonosulfate in both homogeneous and heterogeneous ways: A review, Chem. Eng. J., 2020, 384, 123265,  DOI:10.1016/j.cej.2019.123265.
7. J. Zuo, B. Wang, J. Kang, P. Yan, J. Shen, S. Wang, D. Fu, X. Zhu, T. She, S. Zhao and Z. Chen, Activation of peroxymonosulfate by nanoscaled NiFe₂O₄ magnetic particles for the degradation of 2,4-dichlorophenoxyacetic acid in water: Efficiency, mechanism and degradation pathways, Sep. Purif. Technol., 2022, 297, 121459,  DOI:10.1016/j.seppur.2022.121459.
8. X. Hu, Y. Ye, Y. Chen, M. Liu, W. Zhang and M. Zhu, The synergistic interactions of reaction parameters in heterogeneous peroxymonosulfate oxidation: Reaction kinetic and catalytic mechanism, J. Hazard. Mater., 2022, 421, 126841,  DOI:10.1016/j.jhazmat.2021.126841.
9. S. Huo, Q. Zhao, W. Fu, X. Wu, Y. Wang, H. Xie and M. Gao, Porous Mn_xFe_3-xO₄ spinel oxides trigger H₂O₂ activation for boosted degradation of tetracycline: The formation and dominant role of surface-bound radicals, Sep. Purif. Technol., 2024, 347, 127555,  DOI:10.1016/j.seppur.2024.127555.
10. W. Xiao, A. Chen, M. Cheng, W. Xiong, Y. Liu, J. Wang, G. Wang, G. Zhang, L. Li, H. Liu and Q. Shi, Mechanism insights into metal-organic framework-derived carbon materials activating periodate for p-chlorophenol removal: The role of S and Fe co-doping, Water Res., 2025, 268, 122735,  DOI:10.1016/j.watres.2024.122735.
11. D. Huang and J. Zhao, Speeding Up Fenton Reactions with a Heterogeneous Inorganic Co-catalyst, Chem, 2020, 6, 1512–1514,  DOI:10.1016/j.chempr.2020.06.008.
12. Q. Tian, J. Chang, B. Yu, Y. Jiang, B. Gao, J. Yang, Q. Li, Y. Gao and X. Xu, Co-catalysis strategy for low-oxidant-consumption Fenton-like chemistry: From theoretical understandings to practical applications and future guiding strategies, Water Res., 2024, 267, 122488,  DOI:10.1016/j.watres.2024.122488.
13. X. Ren, M. Guo, H. Li, C. Li, L. Yu, J. Liu and Q. Yang, Microenvironment engineering of ruthenium nanoparticles incorporated into silica nanoreactors for enhanced hydrogenations, Angew. Chem., Int. Ed., 2019, 58, 14483–14488,  DOI:10.1002/anie.201908602.
14. J. Zhang, N. Jin, N. Ji, X. Chen, Y. Shen, T. Pan, L. Li, S. Li, W. Zhang and F. Huo, The encounter of biomolecules in metal–organic framework micro/nano reactors, ACS Appl. Mater. Interfaces, 2021, 13, 52215–52233,  DOI:10.1021/acsami.1c09660.
15. L. Wu, T. Wu, Z. Liu, W. Tang, S. Xiao, B. Shao, Q. Liang, Q. He, Y. Pan, C. Zhao, Y. Liu and S. Tong, Carbon nanotube-based materials for persulfate activation to degrade organic contaminants: Properties, mechanisms and modification insights, J. Hazard. Mater., 2022, 431, 128536,  DOI:10.1016/j.jhazmat.2022.128536.
16. C. Wu, Z. Xing, B. Fang, Y. Cui, Z. Li and W. Zhou, Polyoxometalate-based yolk@shell dual Z-scheme superstructure tandem heterojunction nanoreactors: encapsulation and confinement effects, J. Mater. Chem. A, 2022, 10, 180–191,  10.1039/d1ta07800c.
17. J. Ying, J. Li, G. Jiang, Z. P. Cano, Z. Ma, C. Zhong, D. Su and Z. Chen, Metal-organic frameworks derived platinum-cobalt bimetallic nanoparticles in nitrogen-doped hollow porous carbon capsules as a highly active and durable catalyst for oxygen reduction reaction, Appl. Catal., B, 2018, 225, 496–503,  DOI:10.1016/j.apcatb.2017.11.077.
18. L. Chen and Q. Xu, Metal-organic framework composites for catalysis, Matter, 2019, 1, 57–89,  DOI:10.1016/j.matt.2019.05.018.
19. T. Zeng, X. Zhang, S. Wang, H. Niu and Y. Cai, Spatial confinement of a Co₃O₄ catalyst in hollow metal–organic frameworks as a nanoreactor for improved degradation of organic pollutants, Environ. Sci. Technol., 2015, 49, 2350–2357,  DOI:10.1021/es505014z.
20. Y.-H. Chiu, S. B. Naghadeh, S. A. Lindley, T.-H. Lai, M.-Y. Kuo, K.-D. Chang, J. Z. Zhang and Y.-J. Hsu, Yolk-shell nanostructures as an emerging photocatalyst paradigm for solar hydrogen generation, Nano Energy, 2019, 62, 289–298,  DOI:10.1016/j.nanoen.2019.05.008.
21. X. Wang, B. Guan, Y. He, D. An, Y. Zhang, Y. Cao, X. Li, Y. Liu and Q. Huo, Megranate-like nanoreactor with multiple cores and an acidic mesoporous shell for a cascade reaction, Nanoscale, 2015, 7, 3719–3725,  10.1039/c4nr06341d.
22. Z. Zhao, M. Zhang, J. Ruan, L. Wang, J. Wang, W. Zhang and W. Qiao, An ideal confined catalytic model via MOFs derived yolk-shell nanoreactors: The formation mechanism and catalytic performance for single-core and multi-core, Appl. Surf. Sci., 2023, 623, 156958,  DOI:10.1016/j.apsusc.2023.156958.
23. H. Jiu, Y. Sun, L. Zhang, C. Zhang, J. Zhang and J. Liu, Generalized synthesis of yolk–shell metal oxide spheres, Ceram. Int., 2014, 40, 3149–3154,  DOI:10.1016/j.ceramint.2013.09.129.
24. M. Wang, Y. Boyjoo, J. Pan, S. Wang and J. Liu, Advanced yolk-shell nanoparticles as nanoreactors for energy conversion, Chin. J. Catal., 2017, 38, 970–990,  DOI:10.1016/s1872-2067(17)62818-3.
25. J.-Y. Wu, Y.-C. Wei, T. Torimoto, Y.-A. Chien, C.-Y. Chen, T.-F. M. Chang, M. Sone, P.-Y. Hsieh and Y.-J. Hsu, Yolk@Shell Nanostructures for Water Splitting: Current Development and Future Prospects, ACS Mater. Lett., 2024, 6, 4066–4089,  DOI:10.1021/acsmaterialslett.4c00790.
26. G. Arora, M. Yadav, R. Gaur, R. Gupta, P. Yadav, R. Dixit and R. K. Sharma, Fabrication, functionalization and advanced applications of magnetic hollow materials in confined catalysis and environmental remediation, Nanoscale, 2021, 13, 10967–11003,  10.1039/d1nr01010g.
27. Y. Wang, M. Zhong, F. Ma, C. Wang and X. Lu, Shell-induced enhancement of Fenton-like catalytic performance towards advanced oxidation processes: Concept, mechanism, and properties, Water Res., 2025, 268, 122655,  DOI:10.1016/j.watres.2024.122655.
28. J. Xiao, S. Guo, D. Wang and Q. An, Fenton–Like Reaction: Recent Advances and New Trends, Chem. – Eur. J., 2024, 30, e202304337,  DOI:10.1002/chem.202304337.
29. Y. Mei, Y. Qi, J. Li, X. Deng, S. Ma, T. Yao and J. Wu, Construction of yolk/shell Fe₃O₄@MgSiO₃ nanoreactor for enhanced Fenton-like reaction via spatial separation of adsorption sites and activation sites, J. Taiwan Inst. Chem. Eng., 2020, 113, 363–371,  DOI:10.1016/j.jtice.2020.08.007.
30. S. Ma, D. Yang, Y. Guan, Y. Yang, Y. Zhu, Y. Zhang, J. Wu, L. Sheng, L. Liu and T. Yao, Maximally exploiting active sites on yolk@shell nanoreactor: Nearly 100% PMS activation efficiency and outstanding performance over full pH range in Fenton-like reaction, Appl. Catal., B, 2022, 316, 121594,  DOI:10.1016/j.apcatb.2022.121594.
31. M. Sui, L. She, L. Sheng, J. Wei, L. Zhang and S. Huang, Ordered mesoporous manganese oxide as catalyst for hydrogen peroxide oxidation of norfloxacin in water, Chin. J. Catal., 2013, 34, 536–541,  DOI:10.1016/s1872-2067(11)60492-0.
32. S. Hajimohammadi, S. Gharibi, V. Pourbarkhordar, S. R. Mousavi and H. S. Izadi, Acute poisoning of copper sulfate: a case report and review literature, Eur. J. Intern. Med., 2022, 34, 2–5,  DOI:10.1186/s43162-022-00168-y.
33. Y. Zhang, Y. Mei, S. Ma, Y. Yang, X. Deng, Y. Guan, T. Zhao, B. Jiang, T. Yao, Q. Yang and J. Wu, A simple and green method to prepare non-typical yolk/shell nanoreactor with dual-shells and multiple-cores: Enhanced catalytic activity and stability in Fenton-like reaction, J. Hazard. Mater., 2022, 436, 129234,  DOI:10.1016/j.jhazmat.2022.129234.
34. K. Hu, J. Li, Y. Han, D. H. L. Ng, N. Xing and Y. Lyu, A colorimetric detection strategy and micromotor-assisted photo-Fenton like degradation for hydroquinone based on the peroxidase-like activity of Co₃O₄-CeO₂ nanocages, Catal. Sci. Technol., 2022, 12, 7161–7170,  10.1039/d2cy01192a.
35. E. M. Cuerda-Correa, M. F. Alexandre-Franco and C. Fernández-González, Advanced Oxidation Processes for the Removal of Antibiotics from Water. An Overview, Water, 2019, 12, 102,  DOI:10.3390/w12010102.
36. Y. Liu, Y. Zhao and J. Wang, Fenton/Fenton-like processes with in situ production of hydrogen peroxide/hydroxyl radical for degradation of emerging contaminants: Advances and prospects, J. Hazard. Mater., 2021, 404, 124191,  DOI:10.1016/j.jhazmat.2020.124191.
37. T. Yao, T. Cui, H. Wang, L. Xu, F. Cui and J. Wu, A simple way to prepare Au@polypyrrole/Fe₃O₄ hollow capsules with high stability and their application in catalytic reduction of methylene blue dye, Nanoscale, 2014, 6, 7666–7674,  10.1039/c4nr00023d.
38. T. Zhao, Y. Yang, X. Deng, S. Ma, M. Wu, Y. Zhang, Y. Guan, Y. Zhu, T. Yao, Q. Yang and J. Wu, Preparation of double-yolk egg-like nanoreactor: Enhanced catalytic activity in Fenton-like reaction and insight on confinement effect, J. Colloid Interface Sci., 2022, 625, 774–784,  DOI:10.1016/j.jcis.2022.06.065.
39. M. Xing, W. Xu, C. Dong, Y. Bai, J. Zeng, Y. Zhou, J. Zhang and Y. Yin, Metal Sulfides as Excellent Co-catalysts for H₂O₂ Decomposition in Advanced Oxidation Processes, Chem, 2018, 4, 1359–1372,  DOI:10.1016/j.chempr.2018.03.002.
40. Q. Zhang, M. H. Hadizadeh, Y. Hu, X. Zhang, Z. Su, Z. Wu, X. Wang, F. Xu, Y. Sun, Q. Zhang and W. Wang, The effects of the gas–liquid interface and gas phase on Cl/ClO radical interaction with water molecules, Phys. Chem. Chem. Phys., 2023, 25, 23296–23305,  10.1039/d3cp02796a.
41. M.-M. Chen, H.-Y. Niu, C.-G. Niu, H. Guo, S. Liang and Y.-Y. Yang, Metal-organic framework-derived CuCo/carbon as an efficient magnetic heterogeneous catalyst for persulfate activation and ciprofloxacin degradation, J. Hazard. Mater., 2022, 424, 127196,  DOI:10.1016/j.jhazmat.2021.127196.
42. L. Wu, B. Li, Y. Li, X. Fan, F. Zhang, G. Zhang, Q. Xia and W. Peng, Preferential Growth of the Cobalt (200) Facet in Co@N–C for Enhanced Performance in a Fenton-like Reaction, ACS Catal., 2021, 11, 5532–5543,  DOI:10.1021/acscatal.1c00701.
43. Y. Pei, X. Liu, M. Cao, Z. Wang and H. Yang, Enhanced Fenton-like process on Co₉S₈ catalyst with Mixed-Valence Cobalt(0)/Cobalt(II), Sep. Purif. Technol., 2025, 355, 129649,  DOI:10.1016/j.seppur.2024.129649.
44. W. Xie, Y. Yuan, W. Jiang, S.-R. Zhang, G.-J. Xu, Y.-H. Xu and Z.-M. Su, Heterogeneous activation of peroxymonosulfate by stable Co-MOF for the efficient degradation of organic dye pollutants, CrystEngComm, 2022, 24, 6786–6792,  10.1039/d2ce00932c.
45. H. Sun, Y. Wang, S. Liu, L. Ge, L. Wang, Z. Zhu and S. Wang, Facile synthesis of nitrogen doped reduced graphene oxide as a superior metal-free catalyst for oxidation, Chem. Commun., 2013, 49, 9914,  10.1039/c3cc43401j.
46. Y. Gao, Q. Wang, G. Ji and A. Li, Degradation of antibiotic pollutants by persulfate activated with various carbon materials, Chem. Eng. J., 2022, 429, 132387,  DOI:10.1016/j.cej.2021.132387.
47. J. Lim, H. Kim, J. Park, G.-H. Moon, J. J. M. Vequizo, A. Yamakata, J. Lee and W. Choi, How g-C₃N₄ Works and Is Different from TiO₂ as an Environmental Photocatalyst: Mechanistic View, Environ. Sci. Technol., 2019, 54, 497–506,  DOI:10.1021/acs.est.9b05044.
48. H. Luo, H. Fu, H. Yin and Q. Lin, Carbon materials in persulfate-based advanced oxidation processes: The roles and construction of active sites, J. Hazard. Mater., 2022, 426, 128044,  DOI:10.1016/j.jhazmat.2021.128044.
49. G. Wang, Y. Liu, X. Dong and X. Zhang, Transforming radical to non-radical pathway in peroxymonosulfate activation on nitrogen doped carbon sphere for enhanced removal of organic pollutants: Combined effect of nitrogen species and carbon structure, J. Hazard. Mater., 2022, 437, 129357,  DOI:10.1016/j.jhazmat.2022.129357.
50. Y. Guan, G. Fu, Q. Wang, S. Ma, Y. Yang, B. Xin, J. Zhang, J. Wu and T. Yao, Fe, Co, N co-doped hollow carbon capsules as a full pH range catalyst for pollutant degradation via a non-radical path in Fenton-like reaction, Sep. Purif. Technol., 2022, 299, 121699,  DOI:10.1016/j.seppur.2022.121699.
51. J. Yu, L. Tang, Y. Pang, X. Liang, Y. Lu, H. Feng, J. Wang, L. Deng, J. Zou, X. Zhu and J. Tang, Non-radical oxidation in environmental catalysis: Recognition, identification, and perspectives, Chem. Eng. J., 2022, 433, 134385,  DOI:10.1016/j.cej.2021.134385.
52. M. Wang, Q. Gao, M. Zhang, Y. He, Y. Zhang, R. Shen, J. Hu and G. Wu, Boosting peroxymonosulfate activation via highly active and durable cobalt catalysts, J. Mater. Chem. A, 2021, 9, 2308–2318,  10.1039/d0ta10658e.
53. L. Xu, L. Li, L. Yu and J. C. Yu, Efficient generation of singlet oxygen on modified g-C₃N₄ photocatalyst for preferential oxidation of targeted organic pollutants, Chem. Eng. J., 2022, 431, 134241,  DOI:10.1016/j.cej.2021.134241.
54. S. Sun, H. Yao, W. Fu, S. Xue and W. Zhang, Enhanced degradation of antibiotics by photo-fenton reactive membrane filtration, J. Hazard. Mater., 2020, 386, 121955,  DOI:10.1016/j.jhazmat.2019.121955.
55. H. Wang, Z. Zhao, X. Zhang, W. Dong, Z. Cao, L. He and X. Wang, Rapid decomplexation of Ni-EDTA by microwave-assisted Fenton reaction, Chem. Eng. J., 2020, 381, 122703,  DOI:10.1016/j.cej.2019.122703.
56. J. Anotai, M.-C. Lu and P. Chewpreecha, Kinetics of aniline degradation by Fenton and electro-Fenton processes, Water Res., 2006, 40, 1841–1847,  DOI:10.1016/j.watres.2006.02.033.
57. A. Kawee-ai, A. Srisuwun, N. Tantiwa, W. Nontaman, P. Boonchuay, A. Kuntiya, T. Chaiyaso and P. Seesuriyachan, Eco-friendly processing in enzymatic xylooligosaccharides production from corncob: Influence of pretreatment with sonocatalytic–synergistic Fenton reaction and its antioxidant potentials, Ultrason. Sonochem., 2016, 31, 184–192,  DOI:10.1016/j.ultsonch.2015.12.018.
58. T. Yao, W. Jia, Y. Feng, J. Zhang, Y. Lian, J. Wu and X. Zhang, Preparation of reduced graphene oxide nanosheet/Fe_xO_y/nitrogen-doped carbon layer aerogel as photo-Fenton catalyst with enhanced degradation activity and reusability, J. Hazard. Mater., 2019, 362, 62–71,  DOI:10.1016/j.jhazmat.2018.08.084.
59. X. Tong, S. Ma, Y. Qi, J. Li, T. Yao and J. Wu, Synthesis of FeCo alloy encapsulated nitrogen-doped graphitized carbon: High catalytic activation and low metal ion leaching in microwave assisted Fenton reaction, J. Taiwan Inst. Chem. Eng., 2020, 108, 64–70,  DOI:10.1016/j.jtice.2020.01.009.
60. L. Chen, X. Zuo, S. Yang, T. Cai and D. Ding, Rational design and synthesis of hollow Co₃O₄@Fe₂O₃ core-shell nanostructure for the catalytic degradation of norfloxacin by coupling with peroxymonosulfate, Chem. Eng. J., 2019, 359, 373–384,  DOI:10.1016/j.cej.2018.11.120.
61. Y. Li, L. Li, Y. Feng, H. Wang, J. Liao, J. Ren, W. Zhou, M. He and H. Li, Rattle-structured CuO/Co₃O₄@C microspheres, a potent bifunctional catalyst for hydrogen production from ammonia borane hydrolysis and methanolysis, Appl. Surf. Sci., 2023, 636, 157840,  DOI:10.1016/j.apsusc.2023.157840.
62. L. Zhang, H. Yu, S. Gao, H. Wang, Z. He and K. Huang, Synthesis of yolk-shell magnetic porous organic nanospheres supported Pd catalyst for oxidation of alcohols and Heck reactions, Chem. Eng. J., 2021, 423, 130237,  DOI:10.1016/j.cej.2021.130237.
63. S. Liang, Z. Ziyu, J. Han and D. Xiaoyan, Facile synthesis of magnetic mesoporous silica spheres for efficient removal of methylene blue via catalytic persulfate activation, Sep. Purif. Technol., 2021, 256, 117801,  DOI:10.1016/j.seppur.2020.117801.
64. J. Zhong, C. Cao, H. Liu, Y. Ding and J. Yang, Fabrication of hollow and yolk–shell structured η-Fe₂O₃ nanoparticles with versatile configurations, Ind. Eng. Chem. Res., 2013, 52, 1303–1308,  DOI:10.1021/ie302652b.
65. G. D. Moon, Yolk–shell nanostructures: Syntheses and applications for lithium-Ion battery anodes, Nanomaterials, 2020, 10, 675,  DOI:10.3390/nano10040675.
66. Z. Guo, W. Liu, Z. He, Z. Wang, W. Li and M. Zhang, A carbon-promoted galvanic replacement method to synthesize efficient PdNi nanoalloy catalyst, J. Colloid Interface Sci., 2024, 663, 369–378,  DOI:10.1016/j.jcis.2024.02.158.
67. S. Xiang, D. Wang, K. Zhang, W. Liu, C. Wu, Q. Meng, H. Sun and B. Yang, Chelation competition induced polymerization (CCIP): construction of integrated hollow polydopamine nanocontainers with tailorable functionalities, Chem. Commun., 2016, 52, 10155–10158,  10.1039/c6cc05489g.
68. Y. Liu, L. Gong, Y. Zhang, P. Wang, G. Wang, F. Bai, Z. Zhao, F. Gong and J. Liu, Metal Sulfides Yolk–Shell Nanoreactors with Dual Component for Enhanced Acidic Electrochemical Hydrogen Production, Small Struct., 2022, 4, 2200247,  DOI:10.1002/sstr.202200247.
69. M. Zhang, J. Qiu, Z. Xin and X. Sun, In situ reduced multi-core yolk–shell Co@C nanospheres for broadband microwave absorption, Materials, 2021, 14, 4610,  DOI:10.3390/ma14164610.
70. H.-H. He, J.-P. Yuan, P.-Y. Cai, K.-Y. Wang, L. Feng, A. Kirchon, J. Li, L.-L. Zhang, H.-C. Zhou and Y. Fang, Yolk–shell and hollow Zr/Ce-UiO-66 for manipulating selectivity in tandem reactions and photoreactions, J. Am. Chem. Soc., 2023, 145, 17164–17175,  DOI:10.1021/jacs.3c03883.
71. Y.-F. Wang, T.-T. Yang, W.-L. Liu, D. Zhao, M.-M. Ren, F.-G. Kong, S.-J. Wang, X.-Q. Wang and X.-L. Duan, Design of double-shelled and dual-cavity structures in Fe₃O₄@Void@PMAA@Void@TiO₂ nanocomposite particles for comprehensive photocatalyst and adsorbent applications, Colloid Polym. Sci, 2018, 296, 1719–1728,  DOI:10.1007/s00396-018-4390-z.
72. L. W. Ye, Y. F. Yuan, L. N. Wang, M. Zhu, S. M. Yin, Y. B. Chen and S. Y. Guo, NiCo₂S₄/Co₃S₄ heterogeneous double-shelled nanocages for high-performance electrochemical energy storage, Mater. Lett., 2018, 229, 152–155,  DOI:10.1016/j.matlet.2018.07.005.
73. Z. Li, M. Li, Z. Bian, Y. Kathiraser and S. Kawi, Design of highly stable and selective core/yolk–shell nanocatalysts—A review, Appl. Catal., B, 2016, 188, 324–341,  DOI:10.1016/j.apcatb.2016.01.067.
74. S.-X. Liang, Q. Zhang, Z. Jia, W. Zhang, W. Wang and L.-C. Zhang, Tailoring surface morphology of heterostructured iron-based Fenton catalyst for highly improved catalytic activity, J. Colloid Interface Sci., 2021, 581, 860–873,  DOI:10.1016/j.jcis.2020.07.138.
75. J. Wang, Q. Wang, W. She, C. Xie, X. Zhang, M. Sun, J. Xiao and S. Wang, Tuning the electron density distribution of the Co-N-C catalysts through guest molecules and heteroatom doping to boost oxygen reduction activity, J. Power Sources, 2019, 418, 50–60,  DOI:10.1016/j.jpowsour.2019.02.027.
76. A. Babuponnusami and K. Muthukumar, Advanced oxidation of phenol: A comparison between Fenton, electro-Fenton, sono-electro-Fenton and photo-electro-Fenton processes, Chem. Eng. J., 2012, 183, 1–9,  DOI:10.1016/j.cej.2011.12.010.
77. Y. Chang, Y. Li, C. Zhang, T. Zhao, X. Tuo, J. Guo and Y. Gong, Formaldehyde-controlled synthesis of multishelled hollow mesoporous SiO₂ microspheres, Langmuir, 2019, 35, 14517–14521,  DOI:10.1021/acs.langmuir.9b00880.
78. X. Chen, M. Zhang, S. Li, L. Li, L. Zhang, T. Wang, M. Yu, Z. Mou and C. Wang, Facile synthesis of polypyrrole@metal–organic framework core–shell nanocomposites for dual-mode imaging and synergistic chemo-photothermal therapy of cancer cells, J. Mater. Chem. B, 2017, 5, 1772–1778,  10.1039/c6tb03218d.
79. X. Deng, W. Hui, Y. Guan, Y. Zhang, T. Zhao, C. Guo, B. Xin, Y. Yang, T. Yao and J. Wu, A nanoreactor with Z-scheme FeS₂/MoS₂ heterojunctions encapsulated inside the carbon capsule: Insight on preparation method and enhanced performance in photo-Fenton reaction, Chem. Eng. J., 2022, 450, 138221,  DOI:10.1016/j.cej.2022.138221.
80. T. Olmez-Hanci and I. Arslan-Alaton, Comparison of sulfate and hydroxyl radical based advanced oxidation of phenol, Chem. Eng. J., 2013, 224, 10–16,  DOI:10.1016/j.cej.2012.11.007.
81. W. Hui, X. Deng, Y. Zhu, T. Zhao, Y. Zhang, Y. Guan, W. Ding, Y. Cao, T. Yao and J. Wu, Insight for FeS₂/MoS₂ @SiO₂ nanoreactor with spatial separation of H₂O₂ activation sites and pollutant adsorption sites: Enhanced H₂O₂ activation efficiency and pollutant degradation performance in Fenton reaction, Colloids Surf., A, 2023, 678, 132496,  DOI:10.1016/j.colsurfa.2023.132496.
82. Y. Yang, W. Zhen, T. Zhao, M. Wu, S. Ma, L. Zhao, J. Wu, L. Liu, J. Zhang and T. Yao, Engineering low-valence Mo^δ+ (0<δ<4) sites on MoS₂ surface: Accelerating Fe³⁺/Fe²⁺ cycle, maximizing H₂O₂ activation efficiency, and extending applicable pH range in photo-Fenton reaction, J. Cleaner Prod., 2023, 404, 136918,  DOI:10.1016/j.jclepro.2023.136918.
83. T. Zhao, W. Hui, H. Liu, Y. Yang, M. Wu, J. Wu and T. Yao, Yolk@shell nanoreactor with coordinately unsaturated metal sites on MOF surface inside polypyrrole capsule: Enhanced catalytic activity and lowered metal leaching in fenton-like reaction, Chem. Eng. J., 2023, 474, 145599,  DOI:10.1016/j.cej.2023.145599.
84. Ç. Çalık and D. İ. Çifçi, Comparison of kinetics and costs of Fenton and photo-Fenton processes used for the treatment of a textile industry wastewater, J. Environ. Manage., 2022, 304, 114234–114240,  DOI:10.1016/j.jenvman.2021.114234.
85. Y. Jiang, Z. Wang, J. Huang, F. Yan, Y. Du, C. He, Y. Liu, G. Yao and B. Lai, A singlet oxygen dominated process through photocatalysis of CuS-modified MIL-101(Fe) assisted by peroxymonosulfate for efficient water disinfection, Chem. Eng. J., 2022, 439, 135788,  DOI:10.1016/j.cej.2022.135788.
86. X. Chen, Y. Wang, X. Fan, G. Zhu, Y. Liu and X. Quan, Efficient electro-Fenton degradation of organic pollutants via the synergistic effect of ¹O₂ and ˙OH generated on single Fe-N₄ sites, Sci. Total Environ., 2024, 932, 173042,  DOI:10.1016/j.scitotenv.2024.173042.
87. H. Liu, W. Ding, W. Hui, T. Zhao, Y. Zhang, W. Lv, X. Guo, S. Ma, M. Wu, T. Yao and J. Wu, Fenton-like reaction via combining confinement effect and Lewis acid-base reaction inside a nanoreactor, J. Water Process Eng., 2025, 69, 106582,  DOI:10.1016/j.jwpe.2024.106582.
88. B. Jain, A. K. Singh, H. Kim, E. Lichtfouse and V. K. Sharma, Treatment of organic pollutants by homogeneous and heterogeneous Fenton reaction processes, Environ. Chem. Lett., 2018, 16, 947–967,  DOI:10.1007/s10311-018-0738-3.
89. W. Hui, Z. Li, S. Ma, H. Liu, Y. Zhang, W. Ding, Y. Cao, T. Yao and J. Wu, Maximize coordinately unsaturated metal sites via removal of compact MOF external surface: Enhanced Fenton-like performance inside yolk@shell nanoreactor, Chem. Eng. J., 2024, 497, 154667,  DOI:10.1016/j.cej.2024.154667.
90. J. Wang, H. Li, P. Xia, H. Liu, X. Chen, Z. Ye, Q. He and I. Sirés, Confined MOF pyrolysis within mesoporous SiO₂ core–shell nanoreactors for superior activity and stability of electro-Fenton catalysts, Chem. Eng. J., 2024, 483, 149230,  DOI:10.1016/j.cej.2024.149230.
91. R. Yang, Q. Peng, A. Ahmed, F. Gao, B. Yu, Y. Shen and H. Cong, Yolk–shell Co₃O₄@Fe₃O₄/C Nanocomposites as a Heterogeneous Fenton–like Catalyst for Organic Dye Removal, Chem. – Eur. J., 2023, 29, e202203097,  DOI:10.1002/chem.202203097.
92. Y. Zhu, B. Chang, X. Sun, H. Luo, W. Wang and C. Li, Chloride-mediated electrochemical degradation of the venlafaxine antidepressant, Environ. Technol. Innovation, 2022, 25, 102189,  DOI:10.1016/j.eti.2021.102189.
93. Z. Andikaey, A. A. Ensafi, B. Rezaei and J.-S. Hu, CoNiSe₂/Fe-CoNiSe₂ yolk–shell nanoboxes from metal-organic frameworks for high-performance supercapacitor, Electrochim. Acta, 2022, 417, 140338,  DOI:10.1016/j.electacta.2022.140338.
94. M. Y. Soomro, A. Balouch, E. Alveroglu, R. Larik, K. Shah and S. A. Chang, Fe/Ni bimetallic magnetic nano-alloy (INBMNA): An efficient heterogeneous catalyst for photo-Fenton-like degradation of phenol in aqueous environment, Environ. Pollut., 2024, 360, 124635,  DOI:10.1016/j.envpol.2024.124635.
95. X. Wang, W. Zhou, S. Zhai, X. Chen, Z. Peng, Z. Liu, W. Q. Deng and H. Wu, Metal–Organic Frameworks: Direct Synthesis by Organic Acid–Etching and Reconstruction Disclosure as Oxygen Evolution Electrocatalysts, Angew. Chem., Int. Ed., 2024, 63, 1–8,  DOI:10.1002/anie.202400323.
96. R. Wang, Z. He, W. Wang, J. Bu, D. Wang, G. Zeng, C. Zhou, W. Xiong and Y. Yang, Rational design of cobalt sulfide anchored on nitrogen-doped carbon derived from cyanobacteria waste enables efficient activation of peroxymonosulfate for organic pollutants oxidation, Chemosphere, 2023, 314, 137733,  DOI:10.1016/j.chemosphere.2022.137733.
97. Y. Wang, D. Cao and X. Zhao, Heterogeneous degradation of refractory pollutants by peroxymonosulfate activated by CoO_x-doped ordered mesoporous carbon, Chem. Eng. J., 2017, 328, 1112–1121,  DOI:10.1016/j.cej.2017.07.042.
98. H. Liu, T. A. Bruton, F. M. Doyle and D. L. Sedlak, In Situ Chemical Oxidation of Contaminated Groundwater by Persulfate: Decomposition by Fe(III)- and Mn(IV)-Containing Oxides and Aquifer Materials, Environ. Sci. Technol., 2014, 48, 10330–10336,  DOI:10.1021/es502056d.
99. D. Yue, X. Qian, M. Ren, M. Fang, J. Jia and Y. Zhao, Secondary battery inspired α-nickel hydroxide as an efficient Ni-based heterogeneous catalyst for sulfate radical activation, Sci. Bull., 2018, 63, 278–281,  DOI:10.1016/j.scib.2018.02.002.
100. R. Ma, G. Zhou, M. Gu, X. Tang, W. Ding, Y. Guan, Y. Jiang, J. Yin, L. Zhang and E. H. Huixiang Ang, Cobalt leaching inhibition: Transforming coordination polymers into spherical Co₃O₄@NC catalysts for accelerated tetracycline degradation via enhanced PMS activation, Appl. Surf. Sci., 2024, 648, 158980,  DOI:10.1016/j.apsusc.2023.158980.
101. M. Zhang, C. Xiao, X. Yan, S. Chen, C. Wang, R. Luo, J. Qi, X. Sun, L. Wang and J. Li, Efficient Removal of Organic Pollutants by Metal–organic Framework Derived Co/C Yolk–Shell Nanoreactors: Size-Exclusion and Confinement Effect, Environ. Sci. Technol., 2020, 54, 10289–10300,  DOI:10.1021/acs.est.0c00914.
102. W. Ding, T. Zhao, B. Li, H. Liu, Y. Cao, W. Hui, B. Xin, P. Chen, Y. Guo and J. Wu, A feasible method to prepare yolk@shell nanoreactor with multiple cores and enhanced Fenton-like performance by confinement effect, J. Alloys Compd., 2024, 1003, 175637,  DOI:10.1016/j.jallcom.2024.175637.
103. Z.-M. Cui, Z. Chen, C.-Y. Cao, L. Jiang and W.-G. Song, A yolk–shell structured Fe₂O₃@mesoporous SiO₂ nanoreactor for enhanced activity as a Fenton catalyst in total oxidation of dyes, Chem. Commun., 2013, 49, 2332,  10.1039/c3cc38649j.
104. Y. Ding, D. Li, S. Zuo, Z. Guan and S. Ding, Boron-doping accelerated Cu(II)/Cu(I) cycle for enhancing peroxymonosulfate activation, Sep. Purif. Technol., 2022, 282, 120086,  DOI:10.1016/j.seppur.2021.120086.
105. J. Zuo, J. Shen, J. Kang, P. Yan, B. Wang, S. Wang, D. Fu, W. Wang, T. She, S. Zhao and Z. Chen, B-doped NiFe₂O_x based on the activation of peroxymonosulfate for degrading 2,4-dichlorophenoxyacetic acid in water, Chem. Eng. J., 2023, 459, 141565,  DOI:10.1016/j.cej.2023.141565.
106. S. Yang, L. Peng, C. Cao, F. Wei, J. Liu, Y. N. Zhu, C. Liu, X. Wang and W. Song, Preparation of Magnetic Tubular Nanoreactors for Highly Efficient Catalysis, Chem. – Asian J., 2016, 11, 2797–2801,  DOI:10.1002/asia.201600454.
107. H. J. Hah, J. I. Um, S. H. Han and S. M. Koo, New synthetic route for preparing rattle-type silica particles with metal cores, Chem. Commun., 2004, 1012,  10.1039/b401387e.
108. C. Liu, J. Li, J. Qi, J. Wang, R. Luo, J. Shen, X. Sun, W. Han and L. Wang, Yolk–Shell Fe⁰@SiO₂ Nanoparticles as Nanoreactors for Fenton-like Catalytic Reaction, ACS Appl. Mater. Interfaces, 2014, 6, 13167–13173,  DOI:10.1021/am503063m.
109. C. Wang, H. Wang, J. Na, Y. Yao, A. Azhar, X. Yan, J. Qi, Y. Yamauchi and J. Li, 0D–1D hybrid nanoarchitectonics: tailored design of FeCo@N–C yolk–shell nanoreactors with dual sites for excellent Fenton-like catalysis, Chem. Sci., 2021, 12, 15418–15422,  10.1039/d1sc05000a.
110. F. Zhang, J. Zhang, B. Zhang, L. Zheng, X. Cheng, Q. Wan, B. Han and J. Zhang, Improved catalytic performance of Co-MOF-74 by nanostructure construction, Green Chem., 2020, 22, 5995–6000,  10.1039/d0gc02048f.
111. K.-Y. A. Lin, B.-J. Chen and C.-K. Chen, Evaluating Prussian blue analogues MII₃[MIII(CN)₆]₂ (MII = Co, Cu, Fe, Mn, Ni; MIII = Co, Fe) as activators for peroxymonosulfate in water, RSC Adv., 2016, 6, 92923–92933,  10.1039/c6ra16011e.
112. L. Zhang, Y. Guo, C. Guo, T. Chen, C. Feng, S. Qiao and J. Wang, Construction of defective Zeolitic Imidazolate Frameworks with improved photocatalytic performance via Vanillin as modulator, Chem. Eng. J., 2021, 421, 127839,  DOI:10.1016/j.cej.2020.127839.
113. Ü. Kökçam-Demir, A. Goldman, L. Esrafili, M. Gharib, A. Morsali, O. Weingart and C. Janiak, Coordinatively unsaturated metal sites (open metal sites) in metal–organic frameworks: design and applications, Chem. Soc. Rev., 2020, 49, 2751–2798,  10.1039/c9cs00609e.
114. J. Bae, J. S. Choi, S. Hwang, W. S. Yun, D. Song, J. Lee and N. C. Jeong, Multiple Coordination Exchanges for Room-Temperature Activation of Open-Metal Sites in Metal–Organic Frameworks, ACS Appl. Mater. Interfaces, 2017, 9, 24743–24752,  DOI:10.1021/acsami.7b07299.
115. J. Guo, H. Jia, A. Zhang, Z. Pei, M. Luo, J. Xue, Q. Shen, X. Liu and B. Xu, MIL-100 (Fe) with mix-valence coordinatively unsaturated metal site as Fenton-like catalyst for efficiently removing tetracycline hydrochloride: Boosting Fe(III)/Fe(II) cycle by photoreduction, Sep. Purif. Technol., 2021, 262, 118334,  DOI:10.1016/j.seppur.2021.118334.
116. J. Espín, L. Garzón-Tovar, A. Carné-Sánchez, I. Imaz and D. Maspoch, Photothermal Activation of Metal–Organic Frameworks Using a UV–Vis Light Source, ACS Appl. Mater. Interfaces, 2018, 10, 9555–9562,  DOI:10.1021/acsami.8b00557.
117. S. Khalil, A. Ganguly, D. Mariotti and S. Chakrabarti, Transformation study and characterization of Cu-BTC MOF-derived nanoporous copper oxide, Mater. Horiz., 2025, 12, 862–876,  10.1039/d4mh01085j.
118. W. S. Lamme, J. Zečević and K. P. de Jong, Influence of Metal Deposition and Activation Method on the Structure and Performance of Carbon Nanotube Supported Palladium Catalysts, ChemCatChem, 2018, 10, 1552–1555,  DOI:10.1002/cctc.201701991.
119. Y. Guo, Q. Huang, J. Ding, L. Zhong, T.-T. Li, J. Pan, Y. Hu, J. Qian and S. Huang, CoMo carbide/nitride from bimetallic MOF precursors for enhanced OER performance, Int. J. Hydrogen Energy, 2021, 46, 22268–22276,  DOI:10.1016/j.ijhydene.2021.04.084.
120. Y. Shi, B. Zhu, X. Guo, W. Li, W. Ma, X. Wu and H. Pang, MOF-derived metal sulfides for electrochemical energy applications, Energy Storage Mater., 2022, 51, 840–872,  DOI:10.1016/j.ensm.2022.07.027.
121. M. Aydin, T. Devic, A. Şems Ahsen, N. Gautier and R. Demir-Cakan, Transition Metal (Co, Ni, Fe) Selenides by Selenization of Gallic Acid based MOFs used as Na–Ion Battery Anodes, ChemElectroChem, 2024, 11, e202400385,  DOI:10.1002/celc.202400385.
122. R. Dang, M. Xie, M. Li and X. Xu, Hollow metal composite phosphides derived from MOFs as highly efficient and durable bifunctional electrocatalysts for water splitting, New J. Chem., 2023, 47, 1887–1893,  10.1039/d2nj05475b.
123. Y. Tian, H. Huang, G. Liu, R. Bi and L. Zhang, Metal–organic framework derived yolk–shell NiS₂/carbon spheres for lithium–sulfur batteries with enhanced polysulfide redox kinetics, Chem. Commun., 2019, 55, 3243–3246,  10.1039/c9cc00486f.
124. W. Ding, Y. Zhang, W. Hui, Y. Cao, S. Ma, M. Wu, T. Yao, B. Xin and J. Wu, Enhanced peroxymonosulfate-based Fenton-like degradation performance by confined radical activation path and non-radical activation path inside yolk@shell nanoreactor, J. Alloys Compd., 2024, 985, 173992,  DOI:10.1016/j.jallcom.2024.173992.
125. Y. Li, B. Yu, B. Liu, X. Yu, G. Qin, M. Fan, Y. Zhang and L. Wang, Superior Fenton-like and photo-Fenton-like activity of MoS₂@TiO₂/N-doped carbon nanofibers with phase-regulated and vertically grown MoS₂ nanosheets, Chem. Eng. J., 2023, 452, 139542,  DOI:10.1016/j.cej.2022.139542.
126. J. Liu, Y. Hu, X. Li, C. Xiao, Y. Shi, Y. Chen, J. Cheng, X. Zhu, G. Wang and J. Xie, High-efficient degradation of chloroquine phosphate by oxygen doping MoS₂ co-catalytic Fenton reaction, J. Hazard. Mater., 2023, 458, 131894,  DOI:10.1016/j.jhazmat.2023.131894.
127. L. Zhao, H. Liu, F. Wang and L. Zeng, Design of yolk–shell Fe₃O₄@PMAA composite microspheres for adsorption of metal ions and pH-controlled drug delivery, J. Mater. Chem. A, 2014, 2, 7065–7074,  10.1039/c4ta00976b.
128. L. Zhang, T. Liu and Y. Chen, Magnetic conducting polymer/mesoporous SiO₂ yolk/shell nanomaterials: multifunctional nanocarriers for controlled release of doxorubicin, RSC Adv., 2016, 6, 8572–8579,  10.1039/c5ra23580d.
129. C. Wei, Z. Yu, Z. Guo, N. Guan, W. Xing, W. Du, L. Liu, Y. Zhang and C. Cheng, Formation of carbon coated yolk-shelled Fe₃O₄-CeO₂ hollow spheres toward remarkable performance supercapacitors, J. Energy Storage, 2022, 54, 105269,  DOI:10.1016/j.est.2022.105269.
130. Y. Zhang, K. Sun, Z. Liang, Y. Wang and L. Ling, N-doped yolk-shell hollow carbon sphere wrapped with graphene as sulfur host for high-performance lithium-sulfur batteries, Appl. Surf. Sci., 2018, 427, 823–829,  DOI:10.1016/j.apsusc.2017.06.288.
131. J. Li, T. Bao, C. Zhang, H. Song, Y. Zou, L. Yuan, Y. Xi, C. Yu and C. Liu, A general strategy for direct growth of yolk-shell MOF-on-MOF hybrids, Chem. Eng. J., 2023, 472, 144926,  DOI:10.1016/j.cej.2023.144926.
132. X. Zhao, H. Li, J. Zhang, L. Shi and D. Zhang, Design and synthesis of NiCe@m-SiO₂ yolk-shell framework catalysts with improved coke- and sintering-resistance in dry reforming of methane, Int. J. Hydrogen Energy, 2016, 41, 2447–2456,  DOI:10.1016/j.ijhydene.2015.10.111.
133. T. Zeng, X. Zhang, S. Wang, Y. Ma, H. Niu and Y. Cai, Assembly of a Nanoreactor System with Confined Magnetite Core and Shell for Enhanced Fenton–Like Catalysis, Chem. – Eur. J., 2014, 20, 6474–6481,  DOI:10.1002/chem.201304221.
134. H. Niu, Y. Zheng, S. Wang, L. Zhao, S. Yang and Y. Cai, Continuous generation of hydroxyl radicals for highly efficient elimination of chlorophenols and phenols catalyzed by heterogeneous Fenton-like catalysts yolk/shell Pd@Fe₃O₄@metal organic frameworks, J. Hazard. Mater., 2018, 346, 174–183,  DOI:10.1016/j.jhazmat.2017.12.027.
135. D. Du, W. Shi, L. Wang and J. Zhang, Yolk-shell structured Fe₃O₄@void@TiO₂ as a photo-Fenton-like catalyst for the extremely efficient elimination of tetracycline, Appl. Catal., B, 2017, 200, 484–492,  DOI:10.1016/j.apcatb.2016.07.043.

---