A co-sol-emulsion-gel synthesis of tunable and uniform hollow carbon nanospheres with interconnected mesoporous shells

Abstract

Monodispersed mesoporous hollow spheres of polymer–silica and carbon–silica nanocomposites with an “interpenetration twin” nanostructure have been successfully synthesized by a co-sol-emulsion-gel method. The obtained mesoporous hollow carbon spheres (MHCSs) exhibited an open interconnected mesoporous shell that is endowed with high specific surface area (S_BET, 2106–2225 m² g⁻¹) and large pore volume (1.95–2.53 cm³ g⁻¹). Interestingly, the diameter of the uniform MHCSs could be precisely tuned on demand, as an effective electrode material in supercapacitors, MHCSs with a diameter of 90 nm deliver the shortest time constant (τ₀ = 0.75 s), which is highly beneficial for rate capacitance (180 F g⁻¹ at 100 A g⁻¹, a full charge–discharge within 0.9 s) and cyclic retainability (3% loss after 20 000 cycles). The newly developed synthesis route leads to unique interconnected mesoporous hollow carbonaceous spheres with open-framework structures, providing a new material platform in energy storage.

---

Introduction

The unique and subtle combination of hollow architectures with mesoporous nanostructures has the advantages of large void spaces,^1,2 high surface area³ and low density,⁴ so the design and controlled fabrication of mesoporous hollow carbon spheres (MHCSs) is very significant.^5–8 MHCSs have diverse applications in the field of adsorption,⁶ drug delivery,^7,8 catalysis,^9,10 nano-devices,^11–15 and energy storage/conversion.^16–19 When MHCSs have interconnected open mesoporous channels within carefully controlled shell thicknesses and channel lengths, their performance was enhanced and even might open up various potential applications.^12–14

The template methods have been well accepted as common strategies to synthesize hollow carbon nanospheres (HCSs) over the past few decades, including hard-template and soft-template methods. In general, the hard-template method involves multiple steps in which the core formation of polymer or SiO₂ nanoparticles is a key step.^6,20,21 For instance, dispersible HCSs with high uniformity are synthesized using a strict dual core–shell-type (polystyrene@phenolic resin@SiO₂) structure in which polystyrene and SiO₂ are combined into a confined nanospace at high-temperature pyrolysis.⁶ However, their limited surface area and porosity need to be further increased through physical post-activation treatment.⁶ Furthermore, the diameter (<100 nm), pore sizes and mesostructures of the replicated HCSs are limited to their parental template.^7,22 The soft-templating method usually needs fewer steps to synthesize MHCS when compared with hard-templating techniques. Whereas the obtained MHCSs are not uniform in size (normally size-variance is greater than 150 nm and not well defined), hollow core sizes and the shell thicknesses are also restricted.^6–8,23,24 Furthermore, monodispersed nano scale MHCSs are difficult to obtain, due to their weak self-assembly with precursor components, and tendency towards cross-linking with adjacent nanospheres during the pyrolysis process.^12,13,25,26 Recently, an extension of the Stöber method was found effective for HCS synthesis by removal of silica from silica/carbon nanocomposite materials.^25–28 However, it is complicated to control core/shell size, and small nanoparticles are often found as well. Moreover, the resulting small pores do not possess accessible mesochannels, which hinder potential applications.²⁶ Nevertheless, there is a great need to prepare high-quality MHCS materials, i.e. ones with highly uniform, monodispersed tunable particle size and adjustable shell thickness.^6,13 Specifically, the generated well-defined MHCSs (<100 nm) with open interconnected mesoporous channels using a facile approach are still unavailable in the literature.^12,13 It has stimulated substantial interest for the highly effective diffusivity and ultrahigh loading capacity for active guest molecules/ions. The reason is that the guest molecules/ions can penetrate or diffuse through hollow voids easily by means of accessible open interconnected mesochannels in the shell.^5–8,13

Keeping all the challenges in mind, herein we report a novel synthesis of hollow-structured uniform polymer–silica and carbon–silica nanocomposite spheres with tailorable nano-size and shell thickness using a simple co-sol-emulsion-gel method. The generation process of MHCSs is illustrated in Scheme 1. It is a one-pot synthesis method which uses resorcinol-formaldehyde (RF) (polymer precursor), tetraethyl orthosilicate (TEOS, inorganic precursor) and cetyltrimethyl ammonium bromide (CTAB) as the template. For the reaction rates of RF and TEOS during the sol-gel process, the formation of silica spheres is faster and requires a shorter reaction time compared to the RF spheres under the same conditions of “the Stöber method”.^25,26 Therefore, during the co-sol-emulsion-gel process they are first formed with CTAB micelles and RF molecules as an organic phase act as a template, then TEOS is added as an inorganic precursor, which finally results in the successful synthesis of hollow nanospheres with “interpenetration twin” mesostructures. The “interpenetration twin” mesostructure of inorganic silica functions as a nanoreactor to provide a confined nanospace, which can immensely reduce structural shrinkage of the carbon precursor during high-temperature annealing. After removing silica from the “interpenetration twin” (carbon–silica) nanocomposites, open interconnected mesoporous (2–20 nm) carbon spheres with tailorable particle size (90–240 nm) and well controlled hollow voids (45–105 nm) are obtained. The adopted approach is found to be facile, and efficient for the synthesis of well-defined hollow carbon nanospheres with interconnected mesoporous shells, which may provide a potential platform for energy storage applications. This method may reveal that not only does the controllable synthesis of high quality MHCSs avoid the redundancy of the hard-template method, i.e. it avoids the first synthetic core steps, but also that it effectively prevents the monodispersed nano scale hollow nanospheres aggregating in the soft-template method. It is a promising substitute to traditional methods for the preparation of novel open interconnected mesoporous with “interpenetration twin” mesostructures of HCSs.

Scheme 1 The formation process of mesoporous hollow spheres of polymer–silica and carbon–silica nanocomposite with an “interpenetration twin” nanostructure by a co-sol-emulsion-gel method.

Experimental section

Synthesis of mesoporous hollow carbon spheres (MHCSs)

In a typical synthesis, resorcinol (0.2 g) and CTAB (0.6 g) were dissolved in a 30 ml solution containing absolute deionized water (25–22) and ethanol (5–8 ml) at 25 °C under vigorous stirring. After the resorcinol was fully dissolved, ammonia aqueous solution (NH₄OH, 0.2 mL, 25 wt%) was added and continuously stirred for 20 min, then the formaldehyde solution (0.28 mL) was added and then allowed to stand for a few minutes until it changed from transparent to an emulsion, then TEOS was immediately added to the prepared solution, stirred for 24 h at 25 °C, and subsequently heated for 24 h at 80 °C in a Teflon-lined autoclave. The solid product was obtained through centrifugation and freeze-drying. The obtained product was annealed at 850 °C for 3 h with a ramp rate of 1 °C min⁻¹ under nitrogen flow and the mesoporous hollow carbon/silica sphere (MHSs-X-C/SiO₂) composites were obtained as a result. The mesoporous hollow carbon spheres (MCHSs-X) were obtained when silica was removed by immersing in 15% HF aqueous solution at room temperature for 24 h, followed by subsequent washing with deionized water. Alternatively, the mesoporous hollow silica spheres (MHSs-X-SiO₂) were obtained by calcination of the MHSs-X-C/SiO₂ composites at 650 °C for 5 h in air. Here X represents the spherical average diameters.

Characterization

In order to see the morphology more clearly field emission scanning electron microscopy (FESEM) and transmission electron microscopy (TEM) were performed using Hitachi S-4800, JEOL-2100F, and JEM-ARM200F, respectively. Powder X-ray diffraction (XRD) patterns at low diffraction angles (10–80° in 2θ) were measured using a PANalytical-X-pert diffractometer (Cu Kα radiation at 40 kV and 40 mA). A Raman spectrum was measured at 514 nm excitation by Renishaw. Dynamic light scattering (DLS) measurements were carried out on a Malvern Zetasizer NanoZS Instrument. An Axis Ultra spectrometer was used for XPS analysis. A nitrogen adsorption–desorption isotherm was measured at 77 K with a relative pressure p/p⁰ of 3.3 × 10⁻⁷–0.989 using a Micromeritics-ASAP-2020. Prior to the measurement, the samples were degassed at 300 °C for at least 5 h. Brunauer–Emmett–Teller (BET) and density functional theory (DFT) tests were performed to measure the specific surface area and pore size distribution.

Electrochemical measurements

Electrochemical performance of MHCSs was measured for the two electrode system with 6 M KOH as the electrolyte. 85 wt% of obtained carbon, 10 wt% of acetylene black and 5 wt% of PTFE binder were used for electrode preparation. Stainless steel meshes were used as current collectors for the supercapacitors, with 6 M KOH as the electrolyte. The electrodes were then dried under vacuum at 120 °C for 8 h. Then this slurry was coated on the substrate with 8–10 mg loading per cm² area (∼0.25 mm in thickness). The details are summarized in Table S4 and Fig. S7.† A polypropylene membrane was used as the separator for the supercapacitors. Cyclic voltammetry (CV), galvanostatic charge/discharge and AC impedance spectroscopy (EIS) tests were measured through the CHI660D electrochemical workstation. Impedance spectra were measured in the frequency range of 0.01 Hz–10 kHz with a voltage amplitude of 5 mV. The cutoff charge voltage for the capacitor using 6 M KOH was set at 0–1.0 V. The specific capacitance for a single electrode was calculated using eqn (1),where I(A) = discharge current, dV/dt (V s⁻¹) = slope measured through the discharge curve after the ohmic drop and (g) m = active material mass for single electrode. Specific energy density (E_cell) and specific power density (P_cell) for symmetrical supercapacitors were also calculated by using eqn (2) and (3),where ΔV(V) = cell voltage after the ohmic drop, and t(h) = discharge time.

Results and discussion

SEM images show dispersible mesoporous spheres with high uniformity and average diameter of 90 nm (Fig. 1a and b). The spherical structure is well retained, i.e., a carbon–silica composite, carbon and silica have almost the same diameter (∼90 nm) and void sizes (∼45 nm) (Fig. 1a–d). These indicate that the mesoporous hollow carbon/silica spheres have a structure based on the “interpenetration twins” structure formation of the composite materials (Fig. 1a) and that they also have a self-governing framework (Fig. 1b–d). The high angle annular dark field and bright field scanning transmission electron microscopy (BF-STEM, HAADF-STEM) images (Fig. 1e and f) and elemental mapping images (Fig. 1g–j) further confirmed that “interpenetration twin” mesostructures are crossing each other in a uniform and interlocking manner. The evidence of “interpenetration twins” is further explained through magnified and wrecked images of the nanospheres in Fig. S2 and S3.† Their equal contribution over nanospheres are confirmed through EDX analysis (Fig. 1g–j and S1†). These “interpenetration twin” mesostructures of inorganic silica can possibly function as a nanoreactor to provide a confined nanospace, which reduces structural shrinkage during annealing.⁶ By removing one template from the “interpenetration twins”, these interpenetrating interfaces are converted into interconnected mesoporous shells which are endowed with a storage space, a large pore volume and a crucial channel for active small molecules to penetrate or diffuse within hollow voids easily.

Fig. 1 Typical SEM images of (a) MHSs-90-C/SiO₂ and (b) MHCSs-90; TEM images of (c) MHCSs-90 and (d) MHSs-90-SiO₂; (e, f) BF and HAADF-STEM images of MHSs-90-C/SiO₂; (g–i) the corresponding EDX elemental mapping of carbon, silicon and oxygen, (j) along with an overlay of those three maps.

The diameters of silicon dioxide colloid spheres²⁹ and RF colloid spheres³⁰ under “Stöber method” conditions can be tailored by changing the water/ethanol volume ratio, respectively. From this insight, the diameter of MHCSs also can be facilely and precisely modulated by varying the water/ethanol volume ratio. For instance, SEM and TEM images (Fig. 2 and 3a–c) show dispersible MHCSs with high uniformity and average diameters of 125, 175, and 240 nm synthesized by changing the water/ethanol volume ratio from 25 : 5–22 : 8. Further, their corresponding hollow voids are 70, 97 and 105 nm, respectively. An open interconnected mesoporous (>5 nm) network structure on shells can be clearly observed through FESEM and HRTEM (Fig. 2 and S4†). The open interconnected mesoporous shell is important for facile mass transport channels.^8,13 Dynamic light scattering (DLS) data indicated MHCSs (MHCS-125, MHCS-175 and MHCS-240) are well dispersed with particle sizes of 130, 170 and 235 nm and their polydispersity index is less than 8% (Fig. 3d). It is important to mention that we have not used any surfactant or polymer, the uniformly dispersed system is stable with no sign of aggregated precipitation over one month in water or ethanol, electrostatic repulsion should be responsible for this stabilization. The nitrogen adsorption–desorption shows a type-IV isotherm (Fig. 3e), clearly demonstrating the unique hierarchical mesoporous characteristics of MHCSs. The pore size distribution (Fig. 3f) is calculated by using density functional theory (DFT). These multimode peaks are associated with: (a) micropores (1.1–2 nm) from the phenolic resin pyrolysis, (b) mesoporous shells (2–20 nm) by removing the surfactant and the silica from the “interpenetration twin” mesostructures of the as-made material, and (c) meso–macroporous shells (20–110 nm) which may be derived from the hollow voids and/or between the carbon spheres. The above results are in good agreement with TEM and SEM. The high S_BET (pore volume) of MHCSs is 2106–2225 m² g⁻¹ (1.95–2.53 cm³ g⁻¹), which is higher than other HCSs.^{5–7,18–28,31} The details of porosity properties are summarized in Table 1 and S2.† In addition, the shell thickness can also be tuned by simply increasing the TEOS amount (SiO₂ precursor), i.e., for 1.5 ml the shell thickness increases to 48 nm (avg) from the original (39 nm) and for 2 ml it corresponds to an increase in shell thickness to 56 nm (avg), without any obvious change of the core size of MHCS-175 (Fig. S5†). Furthermore, we found that the S_BET slightly decreases with the increase of the hollow spherical shell, i.e., from 2203 to 1965 and 1858 m² g⁻¹, however, the pore volume increases, i.e., from 1.96 to 2.46 and 2.51 cm³ g⁻¹ and the average pore diameter also increases from 5.32 to 5.93 to 6.53 (Fig. S6 and Table S3†). This effect is possibly attributed to the increase of SiO₂ in the “interpenetration twin” nanostructure which can effectively fill the space of carbon materials.

Fig. 2 SEM images of MHSs with tunable sizes (a, b) MHSs-125-C/SiO₂, (c) MHCSs-125, (d, e) MHSs-175-C/SiO₂, (f) MHCSs-175, (g, h) MHSs-240-C/SiO₂, (i) MHCSs-240.

Fig. 3 TEM images of (a) MHCSs-125, (b) MHCSs-175, (c) MHCSs-240, corresponding insets at different resolutions, (d) dynamic light scattering (DLS) curves of MHCSs-125, MHCSs-175 and MHCSs-240, inset shows photograph of the dispersed solution and the obvious Tyndall effect, (e) nitrogen adsorption–desorption isotherm, and (f) pore size distribution.

Table 1 Porosity properties and the distribution of the pore volume of typical MHCSs

Sample	S BET ^a (m^2 g^−1)	Pore vol and (pore vol%)^b (cm^3 g^−1)
		V total	V <2 nm	V 2–10 nm	V >10 nm
a Surface area is calculated with the Brunauer–Emmett–Teller (BET) method by using a relative pressure range from 0.05 to 0.28. b The volume of pores smaller than 2 nm (V<2 nm), pores between 2 and 10 nm (V2–10 nm), and pores larger than 10 nm (V>10 nm) obtained by DFT.
MHCSs-90	2225	2.53	0.23 (9.1)	0.74 (29.2)	1.56 (61.7)
MHCSs-125	2106	2.11	0.20 (9.5)	0.94 (44.5)	0.97 (46.0)
MHCSs-175	2203	1.96	0.25 (12.8)	1.07 (54.9)	0.64 (32.3)
MHCSs-240	2186	2.18	0.28 (12.9)	1.30 (59.6)	0.6 (27.5)

---

These high S_BET and total pore volumes are due to interconnected multimode mesoporous nanostructures from an “interpenetration twin” mesostructure. Hierarchical nanostructured carbons (MHSCs) within-interconnected hollow void/mesoporous shells have improved super-capacitance as compared with single-sized porous carbon materials because of improved mass transport, higher selectivity and increased S_BET.^7–11 In addition, the open accessible mesoporous channels and small size of MHSCs (especially less than 100 nm) also can be effectively used for loading, grafting, and linking functional nanoparticles/ligands for multiple therapies including drug delivery and bioimaging diagnosis.^12–14

Consequently, the possible synthesis mechanism of polymer–silica and carbon–silica nanocomposites hollow spheres with “interpenetration twin” mesostructures is proposed here. Initially emulsion droplets are formed due to hydrogen bonding of CTAB micelles and numerous hydroxymethyl-substituted species (formed by the quick reaction of resorcinol and formaldehyde in water/alcohol solution with ammonia molecules as a catalyst).^32–34 The spherical diameter is limited by the formation of emulsion droplets.³⁰ A strong electrostatic interface interaction occurs between silicate anions and surfactant molecule cations (CTAB).^35–37 Oligomer silicate species also react with the hydroxyl group of hydroxymethyl-substituted species or resorcinol-formaldehyde (RF).³⁴ During the co-sol–gel process they get assembled, cross-linked and condensed such as to form an “interpenetration twin” framework.²⁶ While RF precursors exhibit structural similarities to silanes, i.e. similar coordination sites, tetrahedral geometry, three-dimensional (3D) network nanostructures and formed a spherical morphology which also undergoes hydrolysis and the subsequent condensation process under “the classical Stöber method” conditions.^25–31 The formation of silica spheres occurs quickly and requires shorter reaction time compared to the RF spheres under the same conditions of the sol–gel process.^25,26 Therefore, emulsion droplets are first present with CTAB micelles and hydroxymethyl-substituted species or RF through electrostatic interactions to generate CTAB-coated hydroxymethyl-substituted species hybrid aggregates as a soft template.²⁶ Then TEOS is added as an inorganic precursor, which can bind to the surface of emulsion droplets via the strong electrostatic interactions with CTAB cations (CTA⁺),^26,31 avoiding the excessive cross-linking between RF micelles. It also hinders RF which could diffuse into the emulsion droplets and polymerize with the silicates, therefore allowing hollow RF-silica nanospheres to be obtained. These act as a template for each other in a homogeneous manner and finally successful synthesis of monodisperse hollow nanospheres with “interpenetration twin” mesostructures takes place.^26,31

By increasing the water/ethanol volume ratio, the hydroxymethyl-substituted species or RF molecules and silicate oligomers have lower bending energy and surface tension, which results in smaller emulsion droplets and eventually, smaller polymer–silica colloid spheres are obtained.^{25,26,29–31} At the same time, with the higher water concentration, the hydrolysis rate of TEOS was increased, thus resulting in a fast nucleation and a decrease of their final diameter.³⁸

The XRD pattern shows two weak and broad characteristic peaks corresponding to the (002) and (101) planes (Fig. S7a†). Fig. S7b† presents the Raman spectrum of MHCS-90 which indicates the presence of a G-band at 1593 cm⁻¹ (related to graphitic carbon) higher than its D-band at 1363 cm⁻¹ (attributed to defects). The peak intensity ratio of the G-band to the D-band is around 1.15 which reflects that MHCS-90 have relatively high graphitized content.³⁹ X-ray photoelectron spectroscopy (XPS) (Fig. S7c†) shows C, O, and N contents in MHCS-90 are 94.1, 4.0, and 1.28 wt%, respectively.

The MHCSs provide a hollow core with an open interconnected mesoporous structure, a thin shell for ion-buffering reservoirs and facile ion diffusion channels. Therefore, they decrease the ion transport length to enhance the charge transfer rate, which are basic requirements of extensive application as high-rate energy storage devices.^40–44 For example, in supercapacitors the MHCSs have high S_BET, sufficient active sites and facile carriage channels for rapid mass transport, which should be beneficial for electrochemical energy storage in high-performance electrode materials.

Electrochemical characterization of symmetric MHCS supercapacitors

The supercapacitor performances of the MHCSs were measured by a two-electrode symmetrical system in 6 M KOH. The cyclic voltammetry (CV) curves (Fig. S8a†) of all samples showed symmetrical rectangular shape at 0.1 V s⁻¹, which indicate an ideal double-layer capacitance behavior.⁴⁰ With MHCSs diameter increasing, the area of CV curves gradually reduces with a highly distorted shape at relatively higher scan rate of 1 V s⁻¹ (Fig. 4a). This indicates that the thin shell of MHCSs delivers the advantage of short ion-transport distance. Galvanostatic charge–discharge tests were performed at various current densities (Fig. 4b). By the increase of MHCS diameter (90–240 nm), the capacitances of all MHCSs are almost equal at low current density, i.e., ∼230 F g⁻¹ at 0.5 A g⁻¹ and maintained a relatively high value ∼200 F g⁻¹ under a high current density of 10 A g⁻¹. The high electrochemical performances obtained are most likely due to large S_BET and interconnected hierarchical porosity for enhanced active sites, facilitating fast ion diffusion kinetics.⁴⁰ However, at a high current density of 100 A g⁻¹, Φ = 90 nm of MHCSs the specific capacitance is still maintained at 180 F g⁻¹ (∼78% of the capacitance retention), much greater than the other, i.e., Φ = 125, 175 and 240 nm which reveal different specific capacitances of 107, 45 and 26 F g⁻¹ (∼47%, ∼21% and ∼11% of the capacitance retention), respectively (Fig. 4b). The thin shell thickness of MHCSs with open interconnected mesoporous shells can shorten the ion-transport distance and ease the ion-transfer pathway, ultimately leading to high-rate supercapacitance. Therefore, electrochemical performances of the MHCS-90 could be better than previously reported phenolic resin-based porous carbon,^31,41 and even superior to some advanced carbon materials for high power supercapacitors (120–160 F g⁻¹ at 50 A g⁻¹ (ref. 42–45)). The Nyquist plots show an almost vertical curve in the low frequency region, indicating good capacitor behavior for MHCSs (Fig. 4c).^40–45 While in the high frequency region (Fig. 4c inset) the 90 nm MHCSs not only have relatively lower equivalent series resistance than others but they also show a much shorter Warburg region portion than that of 120, 175 and 240 nm MHCSs, indicating better ion diffusion efficiency.^42–46 The corresponding time constant τ₀ (ref. 47) also responded in a similar fashion (Fig. 4d), i.e., with 2.7 times reduction of MHCSs diameter corresponding to a τ₀ decrease by a factor of 4.7 (3.55–0.75 s). MHCS-90's superior performances can primarily be attributed to its thin shell thicknesses and pore walls compared with others.³¹ In addition, MHCSs with a diameter of 90 nm exhibited a small IR drop with smooth, symmetric and well-defined charge–discharge lines at a high current density (Fig. S8b†). They also showed high stability of 97% after 20 000 cycles at 5 A g⁻¹ compared to the initial cycle (Fig. S8c†). Power and energy density comparison of the MHCS-90-based supercapacitors with other advanced porous carbon materials can be seen in Fig. S10† which confirm its excellent performance especially in terms of power density (>10 kW kg⁻¹).^31,42–45

Fig. 4 Electrochemical characterization of MHCSs (Φ = 90, 125, 175 and 240 nm, respectively) in the 6 M KOH electrolyte in a two-electrode system. (a) Cyclic voltammograms at 1 V s⁻¹; (b) specific capacitances at different current densities; (c) Nyquist plots and (d) Bode plots of phase angle versus frequency.

The superior electrochemical performance is achieved through MHCS electrodes due to the multiple synergistic effects⁴⁶ of the unique material structure: (i) high S_BET (2106–2225 m² g⁻¹) can provide adequate active sites for adsorbing ions, and as a result large capacitance;^41–46 (ii) hollow cores are able to form ion buffering reservoirs and offer a short diffusion path between electrodes and ions to facilitate electron transfer;³¹ (iii) interconnected multimode mesoporous shells can afford fast ion channels to accelerate ion transport which minimizes polarization effects and maximizes the rate;^43–46 (iv) nanoscale spherical morphology provides a high microstructure and electrochemical stability; and (v) the thinner shell thickness of MHCSs has a shorter ion-transport distance and easier ion-transfer pathway, ultimately leading to higher rate supercapacitance.

Conclusion

In summary, we have designed a novel strategy for an “interpenetration twin” mesostructure to synthesize dispersible and uniform mesoporous hollow spheres using a co-sol-emulsion-gel method. Remarkably, open interconnected mesoporous (2–20 nm) hollow carbon spheres with tunable particle size (90–240 nm), corresponding hollow voids (45–105 nm), high surface area (∼2200 m² g⁻¹) and pore volume (1.95–2.53 cm³ g⁻¹) are obtained through this technique. Furthermore, this method is favorable for its potential application in high-rate energy storage devices. This “interpenetration twin” mesostructure design strategy can also be applied for the preparation of well-defined interconnected porous hollow sphere materials, e.g., silica, polymers and a wide range of nanocomposites. Our synthesized polymer, silicon dioxide and carbon spheres are important for various applications including electrode materials, photonics, adsorbents, catalyst loading and multi-drug delivery applications.

Acknowledgements

This work was supported by the National Natural Science Foundation (NNSF), China (grant no. 21371023) and the National Key Basic Research Program of China (2015CB251100). The authors have not declared any competing financial interests.

Notes and references

1. F. Caruso, R. A. Caruso and H. Möhwald, Science, 1998, 6, 1111–1114.
2. D. Walsh and S. Mann, Nature, 1995, 377, 320–323.
3. A. Imhof and D. J. Pine, Nature, 1997, 389, 948–951.
4. S. A. Johnson, P. J. Ollivier and T. E. Mallouk, Science, 1999, 12, 963–965.
5. A. H. Lu, W. C. Li, G. P. Hao, B. Spliethoff, H. J. Bongard, B. B. Schaack and F. Schüth, Angew. Chem., Int. Ed., 2010, 122, 1659–1662.
6. A. H. Lu, T. Sun, W. C. Li, Q. Sun, F. Han, D. H. Liu and Y. Guo, Angew. Chem., Int. Ed., 2011, 50, 11765–11768.
7. Y. Fang, D. Gu, Y. Zou, Z. Wu, F. Li, R. Che, Y. Deng, B. Tu and D. Zhao, Angew. Chem., Int. Ed., 2010, 49, 7987–7991.
8. Y. Chen, P. Xu, M. Wu, Q. Meng, H. Chen, Z. Shu, J. Wang, L. Zhang, Y. Li and J. Shi, Adv. Mater., 2014, 26, 4294–4301.
9. X. Sun and Y. Li, Angew. Chem., Int. Ed., 2004, 43, 597–601.
10. S. H. Joo, S. J. Choi, I. Oh, J. Kwak, Z. Liu, O. Terasaki and R. Ryoo, Nature, 2001, 412, 169–172.
11. G. Zheng, S. W. Lee, Z. Liang, H. W. Lee, K. Yan, H. Yao, H. Wang, W. Li, S. Chu and Y. Cui, Nat. Nanotechnol., 2014, 9, 618–623.
12. B. Z. Fang, J. H. Kim, M. S. Kim and J. S. Yu, Acc. Chem. Res., 2013, 46, 1397–1406.
13. Y. Li and J. Shi, Adv. Mater., 2014, 26, 3176–3205.
14. J. Liu, N. P. Wickramaratne, S. Qiao and M. Jaroniec, Nat. Mater., 2015, 14, 763–774.
15. R. Liu, S. M. Mahurin, C. Li, R. R. Unocic, J. C. Idrobo, H. Gao, S. J. Pennycook and S. Dai, Angew. Chem., Int. Ed., 2011, 50, 6799–6802.
16. G. H. Wang, J. Hilgert, F. H. Richter, F. Wang, H. J. Bongard, B. Spliethoff, C. Weidenthaler and F. Schüth, Nat. Mater., 2014, 13, 293–300.
17. B. You, J. Yang, Y. Suna and Q. Sua, Chem. Commun., 2011, 47, 12364–12366.
18. J. Liu, T. Yang, D. Wang, G. Lu, D. Zhao and S. Qiao, Nat. Commun., 2014, 4, 2798–2804.
19. A. D. Roberts, X. Li and H. Zhang, Chem. Soc. Rev., 2014, 43, 4341–4356.
20. R. J. White, K. Tauer, M. Antonietti and M. M. Titirici, J. Am. Chem. Soc., 2010, 132, 17360–17363.
21. F. Böttger-Hiller, P. Kempe, G. Cox, A. Panchenko, N. Janssen, A. Petzold, T. Hurn-Albrecht, T. L. Borchardt, M. Rose, S. Kaskel, C. Georgi, H. Lang and S. Spange, Angew. Chem., Int. Ed., 2013, 52, 6088–6091.
22. Y. Wan, H. Yang and D. Zhao, Acc. Chem. Res., 2006, 39, 423–432.
23. J. Fu, Q. Xu, J. Chen, Z. Chen, X. Huang and X. Tang, Chem. Commun., 2010, 46, 6563–6565.
24. Z. Yang, Y. Zhang, J. H. Kong, S. Y. Wong, X. Li and J. Wang, Chem. Mater., 2013, 25, 704–710.
25. A. B. Fuertes, V. V. Patricia and S. Marta, Chem. Commun., 2012, 48, 6124–6126.
26. X. Zhang, Y. Li and C. Cao, J. Mater. Chem., 2012, 22, 13918–13921.
27. X. Fang, S. Liu, J. Zang, C. Xu, M. Zheng, Q. Dong, D. Sun and N. Zheng, Nanoscale, 2013, 5, 6908–6916.
28. B. Guan, X. Wang, Y. Xiao, Y. Liu and Q. Huo, Nanoscale, 2013, 5, 2469–2475.
29. W. Stöber, A. Fink and E. Bohn, J. Colloid Interface Sci., 1968, 26, 62–69.
30. J. Liu, S. Qiao, H. Liu, J. Chen, A. Orpe, D. Zhao and G. Lu, Angew. Chem., Int. Ed., 2011, 50, 5947–5951.
31. Z. A. Qiao, B. Guo, A. J. Binder, J. Chen, G. M. Veith and S. Dai, Nano Lett., 2013, 13, 207–212.
32. D. Gu, H. Bongard, Y. Deng, D. Feng, Z. Wu, Y. Fang, J. Mao, B. Tu, F. Schüth and D. Zhao, Adv. Mater., 2010, 22, 833–837.
33. R. Liu, Y. Shi, Y. Wan, Y. Meng, F. Zhang, D. Gu, Z. Chen, B. Tu and D. Zhao, J. Am. Chem. Soc., 2006, 128, 11652–11662.
34. A. H. Lu, G. P. Hao and Q. Sun, Angew. Chem., Int. Ed., 2011, 50, 9023–9025.
35. Q. Huo, D. I. Margolese, U. Ciesla, P. Feng, T. E. Gier, P. Sieger, R. Leon, P. M. Petroff, F. Schüth and G. D. Stucky, Nature, 1994, 368, 317–321.
36. S. H. Tolbert, A. Firouzi, G. D. Stucky and B. F. Chmelka, Science, 1997, 278, 264–268.
37. Y. Han and J. Y. Ying, Angew. Chem., Int. Ed., 2005, 44, 288–292.
38. Z. Teng, S. Wang, X. Su, G. Chen, Y. Liu, Z. Luo, W. Luo, Y. Tang, H. Ju, D. Zhao and G. Lu, Adv. Mater., 2014, 26, 3741–3747.
39. C. Hu, L. Wang, Y. Zhao, M. Ye, Q. Chen, Z. Feng and L. Qu, Nanoscale, 2014, 6, 8002–8009.
40. J. Hou, C. Cao, X. Ma, F. Idrees, B. Xu, X. Hao and W. Lin, Sci. Rep., 2014, 4, 7260–7265.
41. X. Fang, J. Zhang, X. Wang, M. S. Zheng and N. Zheng, J. Mater. Chem. A, 2014, 2, 6191–6197.
42. Y. S. Yun, S. Y. Cho, J. Shim, B. H. Kim, S. J. Chang, S. J. Baek, Y. S. Huh, Y. Tak, Y. W. Park, S. Park and H. J. Jin, Adv. Mater., 2013, 25, 1993–1998.
43. K. Xie, X. Qin, X. Wang, Y. Wang, H. Tao, Q. Wu, L. Yang and Z. Hu, Adv. Mater., 2012, 24, 347–352.
44. W. Qian, F. Sun, Y. Xu, L. Qiu, C. Liu, S. Wang and F. Yan, Energy Environ. Sci., 2014, 7, 379–386.
45. B. You, J. Yang, Y. Sun and Q. Su, Chem. Commun., 2011, 47, 12364–12366.
46. J. Hou, C. Cao, F. Idrees and X. Ma, ACS Nano, 2015, 9, 2556–2564.
47. X. Yang, C. Cheng, Y. Wang, L. Qiu and D. Li, Science, 2013, 341, 534–537.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5nr06279a

---