Inorganic shell nanostructures to enhance performance and stability of metal nanoparticles in catalytic applications


In this article, we review the recent progress and our research activity on the synthesis of inorganic shell nanostructures to enhance the catalytic performance and stability of metal nanoparticles in catalytic applications. First, we introduce general synthetic strategies for the fabrication of inorganic nanoscale shell layers, including template-assisted sol-gel coating, hydrothermal (or solvothermal) synthesis and the self-templating process. We also discuss recent examples of metal nanoparticles (NPs) with nanoscale shell layers, namely core–shell, yolk–shell and multiple NPs-embedded nanoscale shell. We then discuss the performance and stability of metal particles in practical catalytic applications. Finally, we conclude with a summary and perspective on the further progress of inorganic nanostructure with nanoscale shell layers for catalytic applications.


It is well known that catalysts can reduce the activation energy of a chemical reaction, resulting in acceleration of the rate of chemical reaction. Recent works elucidate that reaction pathways are highly influenced by the surface properties of catalysts, resulting in the ability to precisely tune selectivity by using well-controlled catalysts [1,2,3,4]. Catalysts can be either homogeneous or heterogeneous, depending on whether they are in the same phase as reactant and product, or not. Heterogeneous catalysts, which are generally solid materials, can play with liquid and/or gas phase reactants. Since heterogeneous catalysts can be easily separated and recycled from the reaction media, a lot of practical processes, including petrochemicals, semiconductor, energy and environment, consist of heterogeneous catalytic processes.

Heterogeneous catalysts usually consist of active sites with the support materials [5, 6]. To achieve high activity of heterogeneous catalysts, active sites should be highly dispersed on the surface of support materials [7,8,9]. Thus, people usually synthesized the metal nanoparticles supported on inorganic support materials through a couple of synthetic methods. Traditionally, heterogeneous metal-supported catalysts were synthesized by impregnation, ion exchange and precipitation methods. As the synthetic method has continuously advanced, catalytic materials that have various nanostructures, including nanoparticles, nanocolloids, ordered nanoporous materials and nanocomposites, have been synthesized and studied in terms of both fundamental science and practical engineering [10,11,12,13,14,15].

Nanostructured materials have attracted much attention in a variety of application fields, due to their extraordinary characteristics and enhanced performance. Over the last few decades, as the synthetic chemistry for synthesizing the tailored nanomaterials has developed, well-defined and complicated nanostructures have been successfully synthesized [16,17,18,19,20,21,22,23]. Among various nanostructures, there has been increasing interest in nanostructures consisting of nanoscale shell layers that can have controllable physicochemical properties, such as composition, crystalline characteristics, thickness and porosity [24,25,26]. When some active materials are formed as nanoscale layers, they can provide advantageous characteristics, including relatively large surface area-to-volume ratio per unit mass, improved diffusion of molecules to active site and facile surface reactions in many chemical reactions [27,28,29,30].

Various types of nanostructures consisting of nanoscale shell layers have been studied and suggested for not only fundamental study, but also from the viewpoint of practical application [18, 24, 31]. By adapting the well-controlled sol-gel reaction followed by posttreatment, tailored mesoporous shell layers, including SiO2, TiO2 and ZrO2, have been easily synthesized [32,33,34,35,36,37,38]. Other synthetic methods, such as hydrothermal or solvothermal synthesis, allow the production of transition metal-oxide nanoshells, such as CuOx and FeOx [39, 40]. Similar to the sol-gel chemistry of the SiO2 shell, resorcinol–formaldehyde (R–F) resin was also used for the formation of microporous R–F polymer shells that can be converted to carbon shell through the following pyrolysis process [41]. It is also reported that the precursor monomer of carbonizable polymer, such as dopamine, can produce microporous shell layer on the surface of metal-oxide particles [42].

These kinds of nanostructured shell layers can play an important role in metal-supported composites when used as catalysts in chemical reactions. Nanoscale shells can provide physical barriers to prevent the sintering of metal NPs and can isolate each metal NP individually, resulting in high catalytic activity [30, 32, 33, 43]. The formation of nanoscale shell with well-controlled nanopore can allow the selective diffusion of reactant chemicals that induce selective catalytic reaction. In addition, when the core portion is empty, the nanoscale shell can produce locally homogenous void space, resulting in confined chemical reaction in nanoreactors.

In this article, we intend to summarize not only our research results, but also recent progress in the fabrication of inorganic shell nanostructures and performance enhancement in terms of the activity and stability of metal catalysts. We first introduce the general synthetic concepts for fabricating inorganic nanoscale shell layer. Then, we demonstrate several examples of nanostructured catalysts that consist of metal nanoparticles with nanoscale shell layers. We also discuss the enhancement of activity and stability of the metal–nanoscale shell catalyst in catalytic applications. Finally, we conclude with a summary and our perspective on the continuous progress with the challenges of nanostructured materials in the catalysis field.

General synthetic concepts

Template-assisted sol-gel synthesis

Since several pioneering works were reported in the early 2000s, there have been numerous synthetic technologies reported on the formation of nanostructured shell layers on the surface of substrate [44,45,46]. One of the most common methods for synthesizing the nanostructured shell layer is template-assisted sol-gel synthesis (Fig. 1). In fact, template-assisted sol-gel synthesis has been the most widely investigated method during the past decade for depositing nanostructured shell layers on the surface of colloidal particles. The principle of deposition of shell layers is straightforward and involves several sequential steps: (1) preparing a template substrate that can provide a surface for the deposition of the nanoscale shell layer; (2) depositing the nanostructured shell layer through sol-gel coating of the precursor with the controlled reaction kinetics to form the core–shell nanostructure; and (3) posttreatment, including crystallization, the introduction of porosity and/or the selective removal of substrate for target applications.

Fig. 1

Schematic showing general synthetic strategy for synthesis of inorganic shell nanostructures

Xia and coworkers reported pioneering works in which they synthesized polystyrene@TiO2 core–shell and void@TiO2 hollow nanostructure by infiltrating the hydrolyzed TiO2 precursor, followed by sequential deposition of nanoscale shell on the surface of closed packed PS (polystyrene) particles, fast evaporation of solvent and selective removal of PS particle [46]. Liz-Marzan and coworkers achieved a thin TiO2 shell layer on simultaneously formed Ag nanoparticle [44]. Yu et al. demonstrated solid core/mesoporous silica shell particles that have perpendicular mesopore channels by employing Cn-TAB (n = 12–18) chemicals as pore forming agents [47]. They regulated mesopore orientation and pore diameter by adapting different alkyl chain length of Cn-TAB and changing the synthetic parameters.

Recently, we have reported a robust sol-gel coating method for synthesizing mesoporous SiO2, TiO2, ZrO2 and R–F polymer shell nanostructures on the surface of colloidal particles [32,33,34,35,36,37, 41]. In the case of TiO2 and ZrO2 shells, we used either TBOT (titanium tetrabutoxide) or ZBOT (zirconium tetrabutoxide) as a precursor, hydroxy propyl cellulose (HPC) as a surfactant and colloidal silica sphere prepared by Stober method as a template in ethanolic solution. The shell thickness of TiO2 and ZrO2 layers can be conveniently controlled through either multiple coating or controlling the synthetic parameters, such as the concentration of surfactant and the amount of precursor (Fig. 2) [36, 37]. Figure 2a1 shows that the shell thickness of TiO2 hollow particle is conveniently tuned by repeating the cycles of the coating process. A single run of TiO2 coating produces ca. 25 nm thickness. When repeating the coating step three and five times, the shell thickness increases to ca. 50 and 75 nm, respectively (Fig. 2b1–e1). In the case of ZrO2 hollow nanostructure, the shell thickness is controlled by varying the synthetic parameters, such as the amount of ZrO2 precursor chemical (Fig. 2a2). As the amount of ZrO2 precursor (ZBOT) is increased from 0.2 to 1.4 ml, the shell thickness continuously increases from ca. 21 to 62 nm (Fig. 2b2–g2). In addition, the crystalline properties of inorganic shell nanostructures can also be tuned by using different strategies. We have developed several synthetic methods for inducing the crystallinity on the TiO2 shell layer, in which the structure integrity is well maintained at the same time. We have recently developed “silica-protected calcination,” “partial etching and recalcination” and “acid treatment followed by calcination” processes to produce the crystalline TiO2 shell nanostructure with well-defined hollow morphology [34,35,36]. The detailed procedures and chemistry are found not only in our research articles, but also in our recent review articles [18, 24].

Fig. 2

a1 Schematic showing typical template-assisted synthesis of TiO2 hollow shell nanostructures and corresponding transmission electron microscopy (TEM) images of TiO2 hollow shell nanostructures prepared using multiple sol-gel coating processes: b1 once, c1 three times, d1 five times, and e1 a plot indicating relationship between number of multiple coatings and shell thickness of hollow TiO2 samples; a2 schematic showing typical template-assisted synthesis of ZrO2 hollow shell nanostructures and corresponding TEM images of amorphous hollow ZrO2 particle, when different amounts of precursor used: b2 0.2 ml, c2 0.4 ml, d2 0.6 ml, e2 1.0 ml and f2 1.4 ml; g2 relationship between amount of ZrO2 precursor and shell thickness of hollow ZrO2 samples. Adapted with permission from Refs. [36, 37]

Template-assisted hydrothermal and solvothermal synthesis

Hydrothermal and solvothermal syntheses are not only simple and conventional, but also reliable synthetic processes for preparing powder materials by using solution-based media as a starting material. These synthetic methods are suitable for controlling the crystal growth and mass production of solid materials in mild reaction conditions. Hydrothermal and solvothermal synthetic approaches can be employed for fabricating nanoscale shell layers, as well as shell-based nanostructures. Lou et al. [48,49,50] demonstrated template-assisted hydrothermal approaches for the synthesis of SnO2 shell layer. They reported shell-by-shell templating approaches under hydrothermal conditions and successfully synthesized SnO2 shells on the surface of colloidal silica and hollow SnO2 nanostructures, followed by HF etching. Do et al. achieved the hollow multiple-shell Pt-WO3/TiO2-Au nanostructure using sucrose-derived carbon sphere as the sacrificial template under hydrothermal conditions (Fig. 3) [51]. They first synthesized carbon sphere@Pt-WO3 core–shell particle through hydrothermal synthesis by using sucrose, H2PtCl6 and Na2WO4. Then, as-synthesized carbon sphere@Pt-WO3 was coated with solvothermally pre-synthesized titanate nanodisk through an electrostatic force-induced layer-by-layer deposition method. They repeated the several cycles of electrostatic force-induced layer-by-layer deposition to obtain the carbon sphere@Pt-WO3/titanium nanodisk core–shell nanostructure and deposit Au precursor, followed by heat treatment to obtain double-shell Pt-WO3/TiO2-Au nanostructure (Fig. 3a) [51]. Figure 3b shows that hollow Pt-WO3/TiO2-Au nanostructures revealed well-defined hollow morphology of ca. 1 μm in average diameter. Au NPs with an average size of 22 nm are evenly distributed on the surface of the outer TiO2 layers (Fig. 3c, d).

Fig. 3

a Schemes of synthesis of hollow multiple-shell Pt-WO3/TiO2-Au nanostructure: (1) one-pot hydrothermal synthesis of Pt-WO3@carbon spheres, (2) coating with titanate nanodisk using a layer-by-layer strategy, followed by Au loading (3) and (4) calcination and heat treatment; b, c TEM images of hollow H:Pt-WO3/TiO2-Au; d high-resolution TEM image indicating Au nanoparticle on surface of H:Pt-WO3/TiO2-Au sample. Adapted with permission from Ref. [51]

We also demonstrated SiO2@carbon core–shell nanostructure and hollow carbon sphere through the hydrothermal synthetic route (Fig. 4). The pre-synthesized silica particles were treated with aluminum trichloride (AlCl3), followed by calcination for the introduction of acidic sites, which are catalytic sites for the acid-catalyzed polymerization of carbon precursor. The aqueous mixture of AlCl3-treated SiO2 and sucrose was charged in a stainless steel autoclave and treated under 200 °C for 10 h to obtain silica–polymer composites. SiO2@carbon core–shell particles and hollow carbon shell nanostructures were obtained by the carbonization process and followed an etching process for the removal of silica template, respectively (Fig. 4a) [52]. Figure 4b clearly shows the aggregates of the monodispersed carbon sphere. The high-magnification images show that some of the carbon particles were partially broken, and the inside space was empty, indicating that they have hollow nanostructure. The average diameter of carbon particles is estimated to be ca. 340 nm, which is similar to that of silica template (Fig. 4c).

Fig. 4

a Schematic indicating template-assisted hydrothermal synthesis of hollow carbon sphere and corresponding b scanning electron microscopy (SEM) images and c TEM image of hollow carbon sphere. Adapted with permission from Ref. [52]

Self-templated or template-free synthesis

Another synthetic strategy for synthesizing inorganic nanostructured materials having nanoscale shell layer is the self-templated or template-free method. As one of the most representative examples, Hu et al. [53] synthesized hollow TiO2 shell nanostructure by heating the amorphous TiO2 solid sphere protected with polymer in diethylene glycol (DEG) solution. DEG plays an important role as not only a solvent, but also an etchant. When amorphous TiO2 microsphere in DEG media is heated in the presence of protective polymer (poly acrylic acid, PAA), hollow particles with nanoscale shell layer are eventually produced. The functional groups of PAA can be strongly interconnected on the surface of solid TiO2 particles, resulting in it acting not only as a cross-linker, but also a protective layer to connect local surface, and allow TiO2 particles to be maintained against rapid dissolution by hot solvent. Thus, once DEG solvent penetrates the TiO2 particle and starts to etch the TiO2, preferential dissolution of the core portion can happen, resulting in hollow shell structure.

Recently, Zhang et al. [20, 29, 54] developed a simple method for converting dense solid silica particle to hollow shell counterparts by the self-templated method, which is called “surface-protected etching.” A typical surface-protected etching process involves the following two steps: (1) preparation of silica particle with a protective layer of polymer chemicals, and (2) preferential dissolution of the core portion of silica particle by using a base etchant under well-controlled conditions (Fig. 5a). Upon polyvinylpyrrolidone (PVP) protection as a protective layer on the surface of silica particles, their stability is dramatically enhanced against dissolution by base etching against chemicals such as NaOH. The PVP, which cross-links the subunit of silica surface by strong binding between surface OH groups and carbonyl groups of the PVP, allows NaOH molecules to diffuse into the silica particle and to dissolve oxide-rich area [55]. It makes the outer portion of the silica particle to retain its original particle dimension through the multiple strong interactions between the surfaces of the silica subunit and PVP molecules. Hence, unprotected core particles of silica are gradually dissolved out, resulting in nanostructured hollow shell layers. Figure 5b, c shows TEM images, which reveal that after etching for ca. 1 h, the original monodispersed solid particle becomes porous. As the etching time is elongated to 2.75 and 3.00 h, the interior of the silica particle becomes more porous, and upon continued etching, the hollow sphere can be produced (Fig. 5d, e).

Fig. 5

a Schematic showing concept of “surface-protected etching” for transforming solid SiO2 particles into permeable hollow shells; corresponding TEM images showing morphology of SiO2 particles after etching by NaOH for b 0 h, c 1.00 h, d 2.75 h and e 3.00 h. Adapted with permission from Ref. [29]

As another method, the nanostructured silica shell layer can also be synthesized by spontaneous dissolution, followed by the regrowth process. It is well known that amorphous silica colloids dispersed in an aqueous solution containing weak base chemicals, such as Na2CO3 or NaBH4, can spontaneously change from solid particle to nanostructured hollow shell [56, 57]. The dissolution of silica colloids appears due to high pH resulting from Na2CO3 or NaBH4, in which silica can be decomposed to soluble silicate species. When the concentration of silicate species is continuously increased, and the solution is eventually supersaturated at a certain condition, silicate species are preferably precipitated and re-deposited on the surface. When both spontaneous dissolution and re-deposition of silica particle occur at the same time, solid silica particle can be converted to core–shell or yolk–shell nanostructure and then finally ends with the formation of nanoscale hollow shell.

Ostwald ripening is also one of the classical phenomena in small solid dispersion or liquid sol, in which small particles dissolve and regrow on the surface of larger ones. Yang and Zeng [58] proposed the Ostwald ripening mechanism for the formation of hollow titanium dioxide shell nanostructures from TiF3 under hydrothermal conditions. They also extended the Ostwald ripening phenomena to the synthesis of hollow metal-doped TiO2 nanostructures through a similar hydrothermal process [59]. Archer et al. also carried out pioneering work in which they synthesized a nanostructured SnO2 shell by an inside–out Ostwald ripening under hydrothermal conditions [49, 60]. Spherical CeO2 hollow particles can easily be synthesized through the Ostwald ripening mechanism. Zhang et al. demonstrated porous CeO2 hollow nanostructures that were synthesized in mixed solvent conditions, including ethylene glycol, acetic acid and water under solvothermal conditions [61]. They suggested that hollow CeO2 shell nanostructure can be formed by following several sequential steps during solvothermal synthesis. At the first stage, CeO2 nanoparticles are initially formed through the hydrolysis and oxidation of CeO2 precursor under solvothermal conditions (Fig. 6b). Once the primary CeO2 nanoparticles are formed, they initially lead to aggregation of the solid particle (Fig. 6c). Then, the hollowing process can happen by Ostwald ripening and self-assembly, resulting in mesoporous CeO2 hollow shell nanostructures, when continuously elongating the reaction times under hydrothermal conditions (Fig. 6d, e). Cai et al. also synthesized the CeO2 hollow nanoshell through Ostwald ripening via a microwave-assisted aqueous hydrothermal process [62]. The ripening phenomena can be extended to the synthesis of other inorganic oxide shell nanostructure, including cuprous oxide (Cu2O), cobalt oxide (Co3O4) and many others [39, 63, 64].

Fig. 6

a Schematic indicating the formation of CeO2 hollow spheres through Ostwald ripening process; TEM images indicating morphology change of CeO2 particles obtained at different reaction time: b 2 h, c 4 h, d 6 h and e 8 h. Adapted with permission from Ref. [61]

Metal–nanostructured shell catalyst

Single nanoparticle–nanoscale shell

During the past decade, there have been a lot of studies on the synthesis and catalytic applications of single nanoparticle-supported nanoscale shell structures, which are the so-called core–shell or yolk–shell nanostructures. Most of these efforts are devoted to synthesizing metal nanoparticle (NP) core with inorganic oxide shell. In general, the single metal NP@oxide core–shell or yolk–shell nanostructure is generated through several synthetic approaches, as shown in Fig. 7. Pre-synthesized metal NP can be coated with oxide materials, such as SiO2, to construct metal@oxide core–shell nanostructures. Metal@oxide core–shell nanostructures can then be transformed to their yolk–shell counterparts through either self-hollowing of the oxide layer, or core etching of the metal nanoparticle. As another approach, metal@oxide core–shell nanostructure can be applied to the coating of other inorganic materials, such as TiO2, ZrO2 or carbon, to obtain core–shell–shell nanostructure. If intermediate oxide shell layer is rapidly dissolved out compared to the outer one, we could achieve selective etching of the intermediate layer, resulting in metal@oxide yolk–shell structure.

Fig. 7

Schematic showing synthetic strategies for preparing single metal nanoparticle-oxide shell (core–shell or yolk–shell) nanostructures by several approaches

Since we have established the reliable coating procedure of various materials on nanoscale colloidal particles, we have successfully achieved the practical synthesis of metal@oxide yolk–shell nanostructure, which can have various combinations of core and shell materials. Figure 8 demonstrates a model system of yolk–shell nanostructure that consists of Au nanoparticle as core, with either SiO2 or TiO2 as shell. Colloidal Au NPs can be easily synthesized by the citrate reduction method in hot water, and stabilized by a surfactant, such as PVP. For fabricating Au@SiO2 yolk–shell nanostructure, we chose resorcinol–formaldehyde (R–F) resin material as the intermediate sacrificial layer. Recently, it has been reported that R–F polymer has similar sol-gel chemistry compared with SiO2, resulting in either the monodispersed R–F resin microspheres being successfully synthesized or the coating of colloidal particle with nanoscale R–F layer being easily carried out, though an extension of the Stober method [41, 65]. Figure 8b shows that Au NPs can be coated with uniform R–F layer to produce Au@R–F core–shell structure. The thickness of R–F layer can be conveniently tuned by controlling synthetic parameters, such as the R–F precursor’s concentration, water-to-ethanol ratio and addition of surfactant [25, 65]. After the coating of SiO2 layer on the surface of Au@R–F core–shell particle through the modified Stober method, Au@R–F@SiO2 core–shell–shell nanostructures were obtained. When Au@R–F@SiO2 particles were calcined under air conditions, well-defined Au@SiO2 yolk–shell nanostructures were obtained, due to burning out of the intermediate R–F layer (Fig. 8c). The intermediate R–F layer can be replaced by other inorganic materials, such as SiO2 for fabricating Au@other oxide (e.g., TiO2) yolk–shell nanostructure (Fig. 8d). For synthesizing Au@TiO2 yolk–shell, PVP-stabilized Au NPs can first be coated with SiO2 to form Au@SiO2 core–shell structure. Figure 8e shows that Au NPs can be encapsulated by uniform SiO2 layer. Similar to R–F, the thickness of SiO2 layer can be tuned by varying the synthetic conditions. When Au@SiO2 particles are then coated with TiO2 layer through the sol-gel reaction of TiO2 precursor, such as titanium butoxide, under controlled conditions, Au@SiO2@TiO2 core–shell–shell nanostructures can be obtained. Since the dissolution kinetics of SiO2 on base conditions is much faster than that of TiO2, the intermediate SiO2 layer can be preferentially dissolved out in aqueous NaOH solution, resulting in Au@TiO2 yolk–shell nanostructure (Fig. 8f).

Fig. 8

a Schematic indicating synthesis of Au@SiO2 yolk–shell nanostructure and corresponding TEM images of b Au@R–F core–shell and c Au@SiO2 yolk–shell nanostructure; d schematic indicating synthesis of Au@TiO2 yolk–shell nanostructure and corresponding TEM images of e Au@SiO2 core–shell and f Au@TiO2 yolk–shell nanostructure

Consistent with other nanostructure, the physicochemical characteristics of single metal–shell nanostructure, such as core size, void size, shell thickness and crystallinity, can be controlled by varying the synthetic conditions. Recently, Lee et al. [66] have reported the relationship between geometrical parameters of Au@TiO2 yolk–shell particles and photocatalytic activity. The geometrical parameters of Au@TiO2 yolk–shell nanostructure, including metal size, void space and TiO2 shell thickness, are systemically varied by changing the synthetic parameters. Specifically, the diameter of void space inside the TiO2 shell can be controlled by the concentration of tetraethyl orthosilicate (TEOS) used during the growth of silica layer. The thickness of TiO2 shell layer can be varied by repeating the TiO2 coating step, and the size of Au nanoparticle can be controlled by seed-growing the original trapped Au nanoparticle. Figure 9 shows that the shell thickness is varied from ca. 14 to 80 nm, void diameter is varied from ca. 50 to 350 nm, and the Au NP size is varied from ca. 18 to 100 nm, respectively.

Fig. 9

Typical TEM images of Au@Void@TiO2 nanostructures. Shown are examples for three sets made, namely, for Z@Y@X samples with varying TiO2 shell thickness (X nm), TiO2 shell inner void diameter (Y nm) and gold nanoparticle diameter (Z nm). Adapted with permission from Ref. [66]

Multiple nanoparticles–nanoscale shell

To obtain high catalytic activity, a heterogeneous catalyst must have several requirements, which are well-defined active sites with large active surface area, high dispersion, large resistance to thermal sintering and stability for long-term operation. Although single metal NP@oxide core–shell nanostructures are well-defined structure and attract a lot of attention, other shell-based catalysts, which consist of a large number of metal nanoparticles with nanoscale shell layer, are more useful for practical purposes. Yin and coworkers suggested the concept of a core–satellite catalyst that consists of a monolayer of metal nanocatalysts immobilized on the surface of SiO2 core (Fig. 10a) [54, 67]. A typical procedure for immobilizing metal nanoparticles involves the surface modification of silica particle with 3-aminopropyltriethoxysilane (3-APTES), followed by the controlled attachment of metal NPs through chemical adsorption between amine groups and metal. The SiO2 surface modified with metal nanoparticles is then overcoated with another layer of silica, as shown in Fig. 10b. When metal nanoparticles are overcoated with dense silica layer, since there is poor accessibility and slow surface reaction, the layer should become porous, allowing reactant molecules to reach the metal surface, and protecting metal nanoparticles from metal sintering during heat-treatment or catalysis reactions. Thus, the surface-protected etching is applied to convert the dense outer layer into porous shell nanostructure. Finally, well-defined Fe3O4@SiO2@multiple Au NPs@porous SiO2 nanostructures were obtained (Fig. 10c). The loading of Au nanocatalyst can be well controlled by varying the amount of Au NP added during the adsorption step, and the porosity of the outer silica layer can be tuned by elongating the etching time. As discussed in the previous section, due to the surface protection of PVP, the thickness of outer silica layer does not show any apparent change, until severe etching occurs. Final Fe3O4@SiO2@multiple Au NPs@porous SiO2 nanostructures showed advantageous characteristics, in which Au NPs are highly dispersed in porous silica layers. The porous silica layer not only allows reactant molecules to access the Au NPs for favorable surface reaction, but also provides a physical barrier to prevent Au sintering. Thus, it showed enhanced catalytic performance in terms of the stability of Au NPs and reaction recyclability under multiple runs of liquid phase 4-nitrophenol reduction, discussed in the next section.

Fig. 10

a Schematic showing synthetic procedures for fabrication of SiO2/Au/SiO2 catalysts; TEM images of SiO2/Au/SiO2 catalysts b before and c after surface-protected etching. Adapted with permission from Refs. [54]

Similar to the previous case, we also demonstrated thermally stable Au nanocatalysts encapsulated in mesoporous layer, by introducing cetyl trimethylammonium bromide (CTAB)-derived mesoporous silica on the surface of Au-decorated Fe3O4@SiO2 particles [32]. Although the surface-protected etching technique provides well-defined mesoporous outer shell layer, the porosity must be carefully controlled by monitoring the degree of etching. If the etching significantly proceeds without any monitoring in solution, the outer layer is continuously etched out, and finally disappears by complete dissolution. Sometimes the surface-protected etching process is highly dependent on the experimental conditions and individuals. To fabricate reliable mesoporous outer shell, we thus carried out overcoating Au NPs with CTAB-derived mesoporous silica, followed by calcination [32]. Figure 11a shows the synthetic procedure for preparing Fe3O4@SiO2-Au@mSiO2 sample. Hydrothermally synthesized Fe3O4 particles are coated with silica layer to produce Fe3O4@SiO2 core–shell structure. After Au nanoparticles are decorated on the surface of APTES-functionalized Fe3O4@SiO2, CTAB-derived mesoporous silica layer, which has a perpendicular pore structure, was introduced. Direct calcination under air conditions gives Fe3O4@SiO2-Au@mSiO2 core–shell–shell nanostructure. Additional water treatment makes the outer mesoporous silica layer more porous, resulting in large surface area and high pore volume. This allows reactant molecules not only to be more accessible to the encapsulated Au NPs, but also to be easily reacted on the surface of Au NPs. Figure 11b shows TEM and SEM images of our Fe3O4@SiO2-Au@mSiO2 sample, indicating well-defined core–shell nanostructure. In addition, evenly distributed and encapsulated Au nanoparticles in the porous silica layer can be observed without any aggregation. Figure 11c shows Fe3O4@SiO2-Au@mSiO2 sample which reveals continuous N2 adsorption in the range of ~ 0.5 P/P0 (relative pressure), indicating the presence of mesopore, while control sample Fe3O4@SiO2-Au showed negligible adsorption uptake. Owing to the presence of well-developed mesopore, Fe3O4@SiO2-Au@mSiO2 and water-treated Fe3O4@SiO2-Au@mSiO2-H2O showed relatively high surface area values of 508 and 398 m2·g−1, respectively, while Fe3O4@SiO2-Au sample had a low surface area value (13 m2·g−1). As discussed in the next section, Fe3O4@SiO2-Au@mSiO2 samples having well-developed mesoporosity show improved molecule diffusion and enhanced reaction kinetics.

Fig. 11

a Schematic showing synthetic procedures for preparing Fe3O4@SiO2-Au@mSiO2 samples in which outer silica layer has cylindrical pore structures by coating with CTAB-derived mesoporous silica, followed by calcination; b TEM and SEM images of water-treated Fe3O4@SiO2-Au@mSiO2-H2O; c N2 adsorption–desorption isotherms of Fe3O4@SiO2-Au, Fe3O4@SiO2-Au@mSiO2 and Fe3O4@SiO2-Au@mSiO2-H2O. Adapted with permission from Ref. [32]

Catalytic applications

Metal-nanostructured shells have been used as novel nanocatalysts in many chemical reactions, because during chemical reactions, they have several beneficial effects. Metal–nanoscale shell colloidal nanostructures can enhance the accessibility of the reactant to the metal core, because the thickness of the porous shell layer is in the range of 10–100 nm, resulting in a short diffusion pathway. In addition, a metal nanoparticle can have high stability against thermal sintering or aggregation, due to the protective function of the shell layer. Yolk–shell particle can provide a locally homogeneous reaction environment in the void space of each individual particle, resulting in the minimization of interference of the neighboring particles. Each individual particle having a single active metal, meaning that the number of active sites is limited, allows the relationship between each physiochemical property of the particles and the catalytic performance to be systemically investigated. Here, we discuss some recent results on the catalytic applications and performance enhancement of metal NPs/nanostructured shell catalysts, including single metal nanoparticle/oxide shell yolk–shell particle, mesoporous SiO2 shell-based inorganic micelle and multiple nanoparticles–nanoscale shell nanostructures.

With the aid of the synthetic strategies discussed in the previous section, Au NPs@TiO2 yolk–shell nanostructures have been successfully prepared [30]. Figure 12a shows TEM image, in which Au@TiO2 yolk–shell nanostructure consists of an Au NP particle individually encased in a TiO2 shell with ~ 200 nm in diameter and ~ 20 nm in thickness. Although the Au@TiO2 yolk–shell catalyst is calcined at high temperature up to 775 K, it shows similar dimension of Au NP core, and no change of the structural integrity of the TiO2 shell (Fig. 12 b). In practical Au NPs catalysis, one of the most severe problems is thermal sintering during either high temperature calcination, or catalytic reaction. Au NPs tend to sinter and grow into big particles, resulting in losing the unique catalytic activity that is observed in the original particles. Since small Au nanoparticle is encapsulated and protected by TiO2 shell layer in Au@TiO2 yolk–shell particles, there is no change of dimension and shape of the Au NPs. However, the Au/TiO2-P25 catalyst, in which pre-synthesized Au NPs are supported on commercial P25-TiO2, displays significant metal sintering and growth to large Au NPs under heat treatment at 775 K (Fig. 12c, d). The diffusion and surface reaction in Au@TiO2 yolk–shell nanocatalyst are evaluated by conducting gas phase CO adsorption and CO oxidation. CO molecules are clearly observed to be adsorbed on the surface of Au NPs. This indicates that CO molecules can diffuse inwards through the TiO2 shell. The catalytic activity toward CO oxidation by using the Au@TiO2 yolk–shell catalyst was also evaluated. It is considerably active in promoting the oxidation of CO and demonstrates a higher turn of frequency (TOF) value than the conventional Au/TiO2-P25 catalyst (Fig. 12e). In addition to the above example, we have also studied photocatalytic hydrogen production by using Au@TiO2 yolk–shell nanostructures that have different physicochemical characteristics, such as core size, void size, shell thickness and TiO2 crystallinity [66, 68]. It is generally known that the photocatalysis activity of TiO2-based catalyst is significantly dependent on the crystalline property. Indeed, the hydrogen production performance of Au@TiO2 yolk–shell is highly influenced by the crystallinity of the TiO2 shell. We have been systemically studying and finding out the relationship between photoluminescence lifetime decay, hydrogen production rate and crystalline properties [68].

Fig. 12

TEM images of a, b Au@TiO2 catalyst and c, d 1 wt% Au/TiO2-P25 reference sample, all shown as prepared a, c and after calcination at 775 K b, d, where thermal sintering of the nanoparticles in Au/TiO2-P25 catalysts is indicated by red circles; e time dependence of CO coverage on Au (Θ) and CO2 partial pressure (P) during oxidation of CO with O2. The first and third panels were obtained by first introducing 26.66 kPa of CO into the cell and then adding 26.66 kPa of O2; while in the second and fourth panels, the sequence was reversed. Adapted with permission from Ref. [30]

Recently, Zhang et al. [69] suggested the interesting concept of nanoscale shell-based “inorganic micelle” catalyst, which has different hydrophilic and hydrophobic functional groups on the inner void space and outer surface of particle, respectively. To achieve selective surface functionalization, CTAB micelles were first explored as a pore-blocking agent during the surface functionalization step using a hydrophobic chemical and then removed, to open the mesopore channel, enabling the remaining inner void surface to be hydrophilic. Figure 13a shows the concept of hollow shell or yolk–shell type “inorganic micelle,” in which the SiO2 shells are modified to hydrophobic–hydrophilic interfaces. When Au@SiO2 inorganic micelle was used as a catalyst in 4-nitrophenol reduction in aqueous phase, it showed slower reaction kinetics than its pristine hydrophilic Au@SiO2 counterpart. This indicates that the hydrophobic outer surface of micelle structure influences either the dispersion of Au@SiO2 micelle particle in aqueous reaction media or the diffusion of reactant molecule. They also demonstrated that the inorganic micelle particle can be used as highly efficient catalysts in the liquid phase catalyst in organic solvent. The bromination of alcohols was tested in organic solvents, such as dichloromethane (CH2Cl2), because alkyl halides are valuable chemicals in organic chemistry. With the help of hollow SiO2 and Au@SiO2 yolk–shell inorganic micelles, not only the rate of bromination of benzyl alcohol and α-methylbenzyl alcohol is accelerated, but also over 90% yields are obtained (Fig. 13b, c).

Fig. 13

a Schematics of hollow SiO2 micelle and Au@SiO2 micelles; catalytic activities for bromination of b benzyl alcohol and c α-methylbenzyl alcohol over various catalysts. Adapted with permission from Ref. [69]

Another example of the stabilization effect and catalytic efficiency enhancement of using metal/nanostructured oxide shell is multiple metal nanoparticles encapsulated in porous silica layer, which is prepared by surface-protected etching process. As mentioned before, mesoporous SiO2 framework can effectively stabilize the embedded Au NPs and prevent the efficiency reduction in catalysis caused by aggregation or detachment of nanoparticles. The structural stability and catalytic activity of unprotected Fe3O4/SiO2/Au nanocomposites are severely decreased by agglomerating the Au NPs after six successive cycles of 4-nitrophenol reduction. In the first cycle of 4-nitrophenol reduction, the unprotected Fe3O4/SiO2/Au catalyst showed high activity in terms of kinetics and conversion, as all the Au NPs on the core surface contribute to the catalysis. The catalytic activity of unprotected Fe3O4/SiO2/Au catalyst is continuously decreased during the six successive runs of catalytic reaction, since Au NPs are gradually detached out, and the active surface area of Au NPs is dramatically decreased (Fig. 14a). However, Au catalysts-protected porous SiO2 shell (Fe3O4/SiO2/Au/porous SiO2) showed high metal dispersion, without any loss of Au NPs and deformation of structural integrity. Fe3O4/SiO2/Au/porous SiO2 catalysts maintained their activity well for successive cycles of chemical reaction, with a slight drop of 4-nitrophenol conversion (Fig. 14b).

Fig. 14

Catalytic conversion results of 4-nitrophenol over a Fe3O4/SiO2/Au catalyst and b Fe3O4/SiO2/Au/porous SiO2 catalyst as a function of reaction time in six successive cycles. Adapted with permission from Ref. [67]

Our group also demonstrated the stabilization effect of mesoporous oxide shell on metal nanoparticle catalysts. As previously discussed, we have demonstrated Au NPs encapsulated in mesoporous oxide layer, which is prepared by either SiO2 overcoating followed by base etching or the formation of CTAB-derived mesoporous SiO2 layer followed by calcination. As expected, the Fe3O4@SiO2-Au catalyst that has exposed Au NP showed the highest reaction kinetics (k is ca. 0.223 min−1) in the first run of catalytic reaction. Fe3O4@SiO2-Au@nSiO2, which is prepared by overcoating with nonporous silica layer, showed the lowest rate constant (k is ca. 0.00034 min−1). Fe3O4@SiO2-Au@mSiO2 and water-etched Fe3O4@SiO2-Au@mSiO2-H2O samples showed reasonable k value, such as 0.0771 and 0.104 min−1, respectively. We also evaluated the stability and recyclability of each catalyst by conducting multiple runs of catalytic reaction. While Fe3O4@SiO2-Au catalyst showed the highest conversion in the first cycle, the value rapidly decreased and finally reached < 10% in the successive five reaction runs (Fig. 15a). Since the reactant could not diffuse inward to the active metals due to the dense silica layer, the Fe3O4@SiO2-Au@nSiO2 displayed negligible activity during five reaction runs (Fig. 15b). Fe3O4@SiO2-Au@mSiO2 and Fe3O4@SiO2-Au@mSiO2-H2O samples showed stable conversion values during five successive runs (Fig. 15c, d). The Fe3O4@SiO2-Au@mSiO2-H2O sample showed the best performance in the recycling test, in terms of reaction kinetics and stability. It should be noted that introducing a mesoporous SiO2 followed by calcination and water treatment provides a large amount of active Au NPs encapsulated porous layer without metal sintering. The porous layer allows facile molecule diffusion and retains Au NPs without detachment during multiple reaction runs, resulting in the most stable recyclability with enhanced activity, while exposed Au NPs in Fe3O4@SiO2-Au sample is easily detached out by destabilization of Au-APTES interaction during the 4-NP reduction, resulting in the continuous decrease in the number of active Au, leading to low conversion.

Fig. 15

Cycling experimental results of different catalysts: a Fe3O4@SiO2-Au, b Fe3O4@SiO2-Au@nSiO2, c Fe3O4@SiO2-Au@mSiO2 and d water-treated Fe3O4@SiO2-Au@mSiO2-H2O (insets being TEM images of each catalyst). Adapted with permission from Ref. [32]

Summary and outlook

During the past decades, extensive progress has been made on the synthesis, property control and practical applications of nanostructured materials, as synthetic chemistry and characterization tools have been continuously developed. In the past few years, we have studied the synthesis and catalytic applications of novel metal-oxide shell nanostructures. In this mini-review, we briefly review our research results and the recent progress in metal-inorganic shell nanostructures for catalytic applications. First, we introduce the general synthetic methodology for preparing nanostructured inorganic shell layer, including template-assisted sol-gel chemistry, template-assisted hydrothermal synthesis and the self-templating process. We also discuss two representative concepts of metal NPs/inorganic shell nanostructures, namely core (or yolk)–shell and multiple NPs encapsulated in nanoscale shell. Finally, we discuss the stability and performance enhancement of metal-inorganic shell nanostructures in practical catalysis.

The examples discussed in this article represent not only our efforts, but also recent progress in the engineering of inorganic shell nanostructures for designing highly active and sustainably stable catalysts. Although there have been pioneering works and advanced studies in this research field, extensive research works are still necessary to address many important challenges in practical catalysis industries. The first challenge is that chemical reaction over metal NPs/nanoscale shell nanostructures is limited within simple model reactions. Even though we discussed simple catalytic applications, such as CO oxidation, bromination of benzyl alcohol in organic solvent and 4-nitrophenol reduction, the use of inorganic shell nanostructure catalysts should be extended to other important catalytic reactions, such as hydrogenation, partial oxidation and condensation. Recently, there have been several pioneering works, in which more-complicated nanostructured catalysts have been fabricated and used as a catalyst for other important chemical reactions, such as the Suzuki coupling reaction, epoxidation and olefin hydrogenation [37, 70,71,72]. We believe that these efforts will have positive effects on the use and extension of nanostructured catalysts in the practical catalysis field. The second challenge is that inorganic shell nanostructure catalysts should have not only the improved activity and sustainable stability, but also highly tunable selectivity on a targeted chemical reaction. One intuitive approach for improving selectivity is that metal nanoparticles can be fabricated with well-controlled shape and introduced to shell nanostructures. There have been well-known examples, including shape-controlled Pt, Pd and other nanocrystals, which have demonstrated significantly improved selectivity during chemical reactions [2, 3, 10]. As another synthetic approach, some other functional groups, such as acidic site or base site, can be introduced to the surface of nanostructured shell [37], resulting in improved selectivity toward targeted reactions, such as dehydration, isomerization and condensation. One drawback to be overcome is the need to develop suitable synthetic methods either to encapsulate shape-controlled nanoparticles with exposed active surface or to selectively introduce other active sites on the targeted surface. In addition, there are also the important issues of how to maintain the shape of metal nanoparticle and stability of active site during posttreatment and catalysis. The final challenge is the mass production of inorganic shell nanostructure catalysts. Although inorganic nanostructured catalysts showed unique catalytic properties at laboratory scale, the production amount of the catalyst is limited to the scale of several milligrams to grams. Mass production is still one of the most challengeable works. Once suitable synthetic processes for preparing large amounts of nanostructured catalysts are developed, and are tested in either bench, pilot or commercial scale reactor, it is believed that nanostructured catalysts will contribute to solving many drawbacks in practical catalysis industries.


  1. [1]

    Lee H, Habas SE, Kweskin S, Butcher D, Somorjai GA, Yang P. Morphological control of catalytically active platinum nanocrystals. Angew Chem Int Ed. 2006;45(46):7824.

    CAS  Google Scholar 

  2. [2]

    Bratlie KM, Lee H, Komvopoulos K, Yang P, Somorjai GA. Platinum nanoparticle shape effects on benzene hydrogenation selectivity. Nano Lett. 2007;7(10):3097.

    CAS  Google Scholar 

  3. [3]

    Lee I, Delbecq F, Morales R, Albiter MA, Zaera F. Tuning selectivity in catalysis by controlling particle shape. Nat Mater. 2009;8:132.

    CAS  Google Scholar 

  4. [4]

    Lee I, Zaera F. Catalytic conversion of olefins on supported cubic platinum nanoparticles: selectivity of (100) versus (111) surfaces. J Catal. 2010;269(2):359.

    CAS  Google Scholar 

  5. [5]

    Kim P, Joo JB, Kim H, Kim W, Kim Y, Song IK, Yi J. Preparation of mesoporous Ni–alumina catalyst by one-step sol–gel method: control of textural properties and catalytic application to the hydrodechlorination of o-dichlorobenzene. Catal Lett. 2005;104(3):181.

    CAS  Google Scholar 

  6. [6]

    Kim P, Kim H, Joo JB, Kim W, Song IK, Yi J. Effect of nickel precursor on the catalytic performance of Ni/Al2O3 catalysts in the hydrodechlorination of 1,1,2-trichloroethane. J Mol Catal A Chem. 2006;256(1–2):178.

    CAS  Google Scholar 

  7. [7]

    Kim P, Joo JB, Kim W, Kim J, Song IK, Yi J. Preparation of highly dispersed Pt catalyst using sodium alkoxide as a reducing agent and its application to the methanol electro-oxidation. J Mol Catal A Chem. 2007;263(1–2):15.

    CAS  Google Scholar 

  8. [8]

    Joo JB, Kim P, Kim W, Yi J. Preparation and application of mesocellular carbon foams to catalyst support in methanol electro-oxidation. Catal Today. 2008;131(1–4):219.

    CAS  Google Scholar 

  9. [9]

    Kim P, Joo JB, Kim W, Kim J, Song IK, Yi J. NaBH4-assisted ethylene glycol reduction for preparation of carbon-supported Pt catalyst for methanol electro-oxidation. J Power Sources. 2006;160(2):987.

    CAS  Google Scholar 

  10. [10]

    Kim C, Lee H. Shape effect of Pt nanocrystals on electrocatalytic hydrogenation. Catal Commun. 2009;11(1):7.

    CAS  Google Scholar 

  11. [11]

    Jia CJ, Schüth F. Colloidal metal nanoparticles as a component of designed catalyst. Phys Chem Chem Phys. 2011;13:2457.

    CAS  Google Scholar 

  12. [12]

    Mahmoud MA, Tabor CE, El-Sayed MA, Ding Y, Wang ZL. A new catalytically active colloidal platinum nanocatalyst: the multiarmed nanostar single crystal. J Am Chem Soc. 2008;130(14):4590.

    CAS  Google Scholar 

  13. [13]

    Taguchi A, Schüth F. Ordered mesoporous materials in catalysis. Micropor Mesopor Mat. 2005;77(1):1.

    CAS  Google Scholar 

  14. [14]

    Li W, Wu Z, Wang J, Elzatahry AA, Zhao D. A perspective on mesoporous TiO2 materials. Chem Mater. 2014;26(1):287.

    CAS  Google Scholar 

  15. [15]

    Davidson M, Ji Y, Leong GJ, Kovach NC, Trewyn BG, Richards RM. Hybrid mesoporous silica/noble-metal nanoparticle materials—synthesis and catalytic applications. ACS Appl Nano Mater. 2018;1(9):4386.

    CAS  Google Scholar 

  16. [16]

    Zhang Q, Joo JB, Lu Z, Dahl M, Oliveira D, Ye M, Yin Y. Self-assembly and photocatalysis of mesoporous TiO2 nanocrystal clusters. Nano Res. 2011;4(1):103.

    CAS  Google Scholar 

  17. [17]

    Park J, Song H. Metal@Silica yolk–shell nanostructures as versatile bifunctional nanocatalysts. Nano Res. 2011;4(1):33.

    CAS  Google Scholar 

  18. [18]

    Zhang Q, Lee I, Joo JB, Zaera F, Yin Y. Core–shell nanostructured catalysts. Acc Chem Res. 2012;46(8):1816.

    Google Scholar 

  19. [19]

    Chen D, Cao L, Huang F, Imperia P, Cheng YB, Caruso RA. Synthesis of monodisperse mesoporous titania beads with controllable diameter, high surface areas, and variable pore diameters (14–23 nm). J Am Chem Soc. 2010;132(12):4438.

    CAS  Google Scholar 

  20. [20]

    Zhang Q, Ge J, Goebl J, Hu Y, Lu Z, Yin Y. Rattle-type silica colloidal particles prepared by a surface-protected etching process. Nano Res. 2009;2(7):583.

    CAS  Google Scholar 

  21. [21]

    Yun HJ, Lee H, Joo JB, Kim W, Yi J. Influence of aspect ratio of TiO2 nanorods on the photocatalytic decomposition of formic acid. J Phys Chem C. 2009;113(8):3050.

    CAS  Google Scholar 

  22. [22]

    Yin Y, Rioux RM, Erdonmez CK, Hughes S, Somorjai GA, Alivisatos AP. Formation of hollow nanocrystals through the nanoscale kirkendall effect. Science. 2004;304(5671):711.

    CAS  Google Scholar 

  23. [23]

    Yang Z, Niu Z, Lu Y, Hu Z, Han CC. Templated synthesis of inorganic hollow spheres with a tunable cavity size onto core–shell gel particles. Angew Chem Int Ed. 2003;42(17):1943.

    CAS  Google Scholar 

  24. [24]

    Joo JB, Dahl M, Li N, Zaera F, Yin Y. Tailored synthesis of mesoporous TiO2 hollow nanostructures for catalytic applications. Energ Environ Sci. 2013;6:2082.

    CAS  Google Scholar 

  25. [25]

    Joo JB, Liu H, Lee YJ, Dahl M, Yu H, Zaera F, Yin Y. Tailored synthesis of C@TiO2 yolk–shell nanostructures for highly efficient photocatalysis. Catal Today. 2016;264(15):261.

    CAS  Google Scholar 

  26. [26]

    Moon GD, Joo JB, Dahl M, Jung H, Yin Y. Nitridation and layered assembly of hollow TiO2 shells for electrochemical energy storage. Adv Funct Mater. 2014;24(6):848.

    CAS  Google Scholar 

  27. [27]

    Liang X, Li J, Joo JB, Gutiérrez A, Tillekaratne A, Lee I, Yin Y, Zaera F. Diffusion through the shells of yolk–shell and core–shell nanostructures in the liquid phase. Angew Chem Int Ed. 2012;51(32):8034.

    CAS  Google Scholar 

  28. [28]

    Li J, Liang X, Joo JB, Lee I, Yin Y, Zaera F. Mass transport across the porous oxide shells of core–shell and yolk–shell nanostructures in liquid phase. J Phys Chem C. 2013;117(39):20043.

    CAS  Google Scholar 

  29. [29]

    Zhang Q, Zhang T, Ge J, Yin Y. Permeable silica shell through surface-protected etching. Nano Lett. 2008;8(9):2867.

    CAS  Google Scholar 

  30. [30]

    Lee I, Joo JB, Yin Y, Zaera F. A Yolk@Shell nanoarchitecture for Au/TiO2 catalysts. Angew Chem Int Ed. 2011;50(43):10208.

    CAS  Google Scholar 

  31. [31]

    Joo JB, Zhang Q, Dahl M, Zaera F, Yin Y. Synthesis, crystallinity control, and photocatalysis of nanostructured titanium dioxide shells. J Mater Res. 2013;28(3):362.

    CAS  Google Scholar 

  32. [32]

    Lee H, Jeong U, Kim Y, Joo JB. Magnetically-separable and thermally-stable Au nanoparticles encapsulated in mesoporous silica for catalytic applications. Top Catal. 2017;60(9–11):763.

    CAS  Google Scholar 

  33. [33]

    Jeong U, Joo JB, Kim Y. Au nanoparticle-embedded SiO2–Au@SiO2 catalysts with improved catalytic activity, enhanced stability to metal sintering and excellent recyclability. RSC Adv. 2015;5:55608.

    CAS  Google Scholar 

  34. [34]

    Joo JB, Lee I, Dahl M, Moon GD, Zaera F, Yin Y. Controllable synthesis of mesoporous TiO2 hollow shells: toward an efficient photocatalyst. Adv Funct Mater. 2013;23(34):4246.

    CAS  Google Scholar 

  35. [35]

    Joo JB, Zhang Q, Dahl M, Lee I, Goebl J, Zaera F, Yin Y. Control of the nanoscale crystallinity in mesoporous TiO2 shells for enhanced photocatalytic activity. Energ Environ Sci. 2012;5:6321.

    CAS  Google Scholar 

  36. [36]

    Joo JB, Zhang Q, Lee I, Dahl M, Zaera F, Yin Y. Mesoporous anatase titania hollow nanostructures though silica-protected calcination. Adv Funct Mater. 2012;22(1):166.

    CAS  Google Scholar 

  37. [37]

    Joo JB, Vu A, Zhang Q, Dahl M, Gu M, Zaera F, Yin Y. A sulfated ZrO2 hollow nanostructure as an acid catalyst in the dehydration of fructose to 5-hydroxymethylfurfural. ChemSusChem. 2013;6(10):2001.

    CAS  Google Scholar 

  38. [38]

    Lou XW, Yuan C, Zhang Q, Archer LA. Platinum-functionalized octahedral silica nanocages: synthesis and characterization. Angew Chem Int Ed. 2006;45(23):3825.

    CAS  Google Scholar 

  39. [39]

    Teo JJ, Chang Y, Zeng HC. Fabrications of hollow nanocubes of Cu2O and Cu via reductive self-assembly of CuO nanocrystals. Langmuir. 2006;22(17):7369.

    CAS  Google Scholar 

  40. [40]

    Cao SW, Zhu YJ, Cheng GF, Huang YH. Preparation and photocatalytic property of α-Fe2O3 hollow core/shell hierarchical nanostructures. J Phys Chem Solids. 2010;71(12):1680.

    CAS  Google Scholar 

  41. [41]

    Li N, Zhang Q, Liu J, Joo JB, Lee A, Gan Y, Yin Y. Sol–gel coating of inorganic nanostructures with resorcinol–formaldehyde resin. Chem Commun. 2013;49:5135.

    CAS  Google Scholar 

  42. [42]

    Liu R, Mahurin SM, Li C, Unocic RR, Idrobo JC, Gao H, Pennycook SJ, Dai S. Dopamine as a carbon source: the controlled synthesis of hollow carbon spheres and yolk-structured carbon nanocomposites. Angew Chem Int Ed. 2011;50(30):6799.

    CAS  Google Scholar 

  43. [43]

    Lee I, Joo JB, Yin Y, Zaera F. Au@Void@TiO2 yolk–shell nanostructures as catalysts for the promotion of oxidation reactions at cryogenic temperatures. Surf Sci. 2016;648:150.

    CAS  Google Scholar 

  44. [44]

    Pastoriza-Santos I, Koktysh DS, Mamedov AA, Giersig M, Kotov NA, Liz-Marzán LM. One-pot synthesis of Ag@TiO2 core–shell nanoparticles and their layer-by-layer assembly. Langmuir. 2000;16(6):2731.

    CAS  Google Scholar 

  45. [45]

    Mayya KS, Gittins DI, Dibaj AM, Caruso F. Nanotubes prepared by templating sacrificial nickel nanorods. Nano Lett. 2001;1(12):727.

    CAS  Google Scholar 

  46. [46]

    Zhong Z, Yin Y, Gates B, Xia Y. Preparation of mesoscale hollow spheres of TiO2 and SnO2 by templating against crystalline arrays of polystyrene beads. Adv Mater. 2000;12(3):206.

    CAS  Google Scholar 

  47. [47]

    Yoon SB, Kim JY, Kim JH, Park YJ, Yoon KR, Park SK, Yu JS. Synthesis of monodisperse spherical silica particles with solid core and mesoporous shell: mesopore channels perpendicular to the surface. J Mater Chem. 2007;17:1758.

    CAS  Google Scholar 

  48. [48]

    Lou XW, Yuan C, Archer LA. Shell-by-shell synthesis of tin oxide hollow colloids with nanoarchitectured walls: cavity size tuning and functionalization. Small. 2007;3(2):261.

    CAS  Google Scholar 

  49. [49]

    Chen JS, Archer LA, Lou XW. SnO2 hollow structures and TiO2 nanosheets for lithium-ion batteries. J Mater Chem. 2011;21:9912.

    CAS  Google Scholar 

  50. [50]

    Lou XW, Yuan C, Archer LA. Double-walled SnO2 nano-cocoons with movable magnetic cores. Adv Mater. 2007;19(20):3328.

    CAS  Google Scholar 

  51. [51]

    Nguyen CC, Vu NN, Do TO. Efficient hollow double-shell photocatalysts for the degradation of organic pollutants under visible light and in darkness. J Mater Chem A. 2016;4:4413.

    CAS  Google Scholar 

  52. [52]

    Joo JB, Kim P, Kim W, Kim J, Kim ND, Yi J. Simple preparation of hollow carbon sphere via templating method. Curr Appl Phys. 2008;8(6):814.

    Google Scholar 

  53. [53]

    Hu Y, Ge J, Sun Y, Zhang T, Yin Y. A self-templated approach to TiO2 microcapsules. Nano Lett. 2007;7(6):1832.

    CAS  Google Scholar 

  54. [54]

    Zhang Q, Lee I, Ge J, Zaera F, Yin Y. Surface-protected etching of mesoporous oxide shells for the stabilization of metal nanocatalysts. Adv Funct Mater. 2010;20(14):2201.

    CAS  Google Scholar 

  55. [55]

    Pirzada T, Arvidson SA, Saquing CD, Shah SS, Khan SA. Hybrid carbon silica nanofibers through sol–gel electrospinning. Langmuir. 2014;30(51):15504.

    CAS  Google Scholar 

  56. [56]

    Zhang T, Ge J, Hu Y, Zhang Q, Aloni S, Yin Y. Formation of hollow silica colloids through a spontaneous dissolution-regrowth process. Angew Chem Int Ed. 2008;47(31):5806.

    CAS  Google Scholar 

  57. [57]

    Fang X, Chen C, Liu Z, Liu P, Zheng N. A cationic surfactant assisted selective etching strategy to hollow mesoporous silica spheres. Nanoscale. 2011;3:1632.

    CAS  Google Scholar 

  58. [58]

    Yang HG, Zeng HC. Preparation of hollow anatase TiO2 nanospheres via ostwald ripening. J Phys Chem B. 2004;108(11):3492.

    CAS  Google Scholar 

  59. [59]

    Li J, Zeng HC. Hollowing Sn-doped TiO2 nanospheres via ostwald ripening. J Am Chem Soc. 2007;129(51):15839.

    CAS  Google Scholar 

  60. [60]

    Lou XW, Wang Y, Yuan C, Lee JY, Archer LA. Template-free synthesis of SnO2 hollow nanostructures with high lithium storage capacity. Adv Mater. 2006;18(17):2325.

    CAS  Google Scholar 

  61. [61]

    Xu Y, Zhang Y, Zhou Y, Xiang S, Wang Q, Zhang C, Sheng X. CeO2 hollow nanospheres synthesized by a one pot template-free hydrothermal method and their application as catalyst support. RSC Adv. 2015;5:58237.

    CAS  Google Scholar 

  62. [62]

    Cao CY, Cui ZM, Chen CQ, Song WG, Cai W. Ceria hollow nanospheres produced by a template-free microwave-assisted hydrothermal method for heavy metal ion removal and catalysis. J Phys Chem C. 2010;114(21):9865.

    CAS  Google Scholar 

  63. [63]

    Chang Y, Teo JJ, Zeng HC. Formation of colloidal CuO nanocrystallites and their spherical aggregation and reductive transformation to hollow Cu2O nanospheres. Langmuir. 2004;21(3):1074.

    Google Scholar 

  64. [64]

    Liu B, Zeng HC. Symmetric and asymmetric ostwald ripening in the fabrication of homogeneous core–shell semiconductors. Small. 2005;1(5):566.

    CAS  Google Scholar 

  65. [65]

    Liu J, Qiao SZ, Liu H, Chen J, Orpe A, Zhao D, Lu GQ. Extension of the Stöber method to the preparation of monodisperse resorcinol-formaldehyde resin polymer and carbon spheres. Angew Chem Int Ed. 2011;50(26):5947.

    CAS  Google Scholar 

  66. [66]

    Lee YJ, Joo JB, Yin Y, Zaera F. Evaluation of the effective photoexcitation distances in the photocatalytic production of H2 from water using Au@Void@TiO2 yolk–shell nanostructures. ACS Energy Lett. 2016;1(1):52.

    CAS  Google Scholar 

  67. [67]

    Ge J, Zhang Q, Zhang T, Yin Y. Core-satellite nanocomposite catalysts protected by a porous silica shell: controllable reactivity, high stability, and magnetic recyclability. Angew Chem Int Ed. 2008;47(46):8924.

    CAS  Google Scholar 

  68. [68]

    Dillon RJ, Joo JB, Zaera F, Yin Y, Bardeen CJ. Correlating the excited state relaxation dynamics as measured by photoluminescence and transient absorption with the photocatalytic activity of Au@TiO2 core–shell nanostructures. Phys Chem Chem Phys. 2013;15:1488.

    CAS  Google Scholar 

  69. [69]

    Zhang Q, Shu XZ, Lucas JM, Toste FD, Somorjai GA, Alivisatos AP. Inorganic micelles as efficient and recyclable micellar catalysts. Nano Lett. 2014;14(1):379.

    CAS  Google Scholar 

  70. [70]

    Kim M, Park JC, Kim A, Park KH, Song H. Porosity control of Pd@SiO2 yolk–shell nanocatalysts by the formation of nickel phyllosilicate and its influence on Suzuki coupling reactions. Langmuir. 2012;28(15):6441.

    CAS  Google Scholar 

  71. [71]

    Nabid MR, Bide Y, Abuali M. Copper core silver shell nanoparticle–yolk/shell Fe3O4@chitosan-derived carbon nanoparticle composite as an efficient catalyst for catalytic epoxidation in water. RSC Adv. 2014;4:35844.

    CAS  Google Scholar 

  72. [72]

    Li X, Zhang W, Zhang L, Yang H. Pd nanoparticles confined in fluoro-functionalized yolk–shell-structured silica for olefin hydrogenation in water. Chin J Catal. 2013;34(6):1192.

    CAS  Google Scholar 

Download references


This work is supported by the Korea Institute of Energy Technology Evaluation and Planning (KETEP) and the Ministry of Trade, Industry and Energy (MOTIE, No. 20174010201490). This work is also financially supported by the Korea Environment Industry & Technology Institute (KEITI) through “The Chemical Accident Prevention Technology Development Project” granted by the Korea Ministry of Environment (MOE, No. 2017001960004).

Author information



Corresponding author

Correspondence to Ji Bong Joo.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Choi, I., Lee, H.K., Lee, G.W. et al. Inorganic shell nanostructures to enhance performance and stability of metal nanoparticles in catalytic applications. Rare Met. 39, 767–783 (2020).

Download citation


  • Nanostructures
  • Inorganic shell
  • Metal nanoparticle
  • Stabilization
  • Performance enhancement
  • Catalyst