A Family of Mesocubes

A Family of MesocubesSai Karthik Addu,‡ Jian Zhu,‡ K. Y. Simon Ng, and Da Deng*

Department of Chemical Engineering and Materials Science, Wayne State University, 5050 Anthony Wayne Drive, Detroit, Michigan48202, United States

*S Supporting Information

ABSTRACT: It is challenging to develop a general universal procedure to fabricate mesoscale cubic structures on a large scalewith different nanoscale building units. It is always desirable to tune the chemical compositions within confined arrangementswithout damaging the mesostructures to provide the desired physiochemical properties required by various devices/applications.Herein, we report the successful design and facile preparation of a family of mesocubes with different compositions, including (a)ZnSn(OH)6, (b) evenly distributed Zn2SnO4 and SnO2 nanoparticles, (c) hollow cubes of SnO2 nanoparticles, (d) high-orderednanoparticles of Zn2SnO4&Sn@C; (e) SnO2@C core−shell subunits, (f) SnO2@C nanoparticle aggregates enclosed withoxidized carbon sheath, and (g) C nanobubbles, as building units, all, except ZnSn(OH)6, with the same confined arrangementsof nanoparticles as building units inside the same framework of cubic mesostructures. This family of mesocubes will provide arich pool of materials with different functional properties to meet demands in different applications and offer opportunities toevaluate fundamentals of structure−property−performance relationships. On the basis of the best of our knowledge, this familyof facilely prepared mesocubes with unique combination of microsize cubes and compositions was reported for the first time,especially the carbon mesocubes formed by aggregation of carbon nanobubbles as the building subunits. Additionally, wedemonstrated, for the first time, that two family members of mesocubes of Zn2SnO4&SnO2 and Zn2SnO4&Sn@C can be used asanode materials in lithium ion batteries with impressive high packing densities and superior rate performance.

1. INTRODUCTION

The design and fabrication of nanoscale functional materials toexplore fundamentals of morphology-dependent properties andnumerous advanced applications of nanomaterials have beenattracting much attention in the past 2 decades. Nanoscalematerials are promising to achieve paradigm shifts in manyfields, such as catalysis,1 drug delivery,2 energy storage,3−5 solarcells,6,7 absorption,8 photonics,9 chemical sensors,10 andreactors in confined space on the nanoscale.11 The ability torationally design and facilely fabricate nanoscale materials willenable the wide adoption of nanomaterials in many fields,achieving tremendous positive impacts. Nanomaterials aretypically prepared by template-assisted methods with multiplesteps involved, hydrothermal methods under high temperatureand high pressure conditions, top-down ball-milling with highenergy, pyrolysis under high temperature, and chemical vapordeposition. However, it is still challenging to develop a generaluniversal procedure to fabricate nanomaterials on a large scalewith the same confined arrangements on the mesoscale, but

with different chemical compositions and properties to meetdemanding requirements of various applications.Another challenging issue for nanoscale materials is the low

tapped density associated with the small particle size and largesurface area, which is not desirable for certain applications. Forexample, nanomaterials have been extensively explored foradvanced lithium ion batteries to achieve high specific energy(by mass), but the critical issue of energy density (by volume)due to the low tapped density of nanomaterials is rarelyaddressed. Low tapped density could prevent the production ofcompact batteries, which is not acceptable for mobile electronicdevices and electric vehicles with limited space. In fact, thetapped density of nanomaterials could be 1 order of magnitudelower as compared to those in the bulk state. This critical issueof low tapped density is most vividly illustrated by the following

Received: May 1, 2014Revised: July 1, 2014Published: July 8, 2014

Article

pubs.acs.org/cm

© 2014 American Chemical Society 4472 dx.doi.org/10.1021/cm501580y | Chem. Mater. 2014, 26, 4472−4485

pubs.acs.org/cm

example: the tapped density of graphite nanoparticles(commercial) of 30−40 nm in size is 0.26 g/cm3, as comparedto that of bulk graphite with density of 2.23 g/cm3, a differenceof almost 9 times. Other issues associated with electrodematerials at the nanoscale are the poor electrical properties ofthe electrode due to interparticle resistance and low Coulombicefficiency attributed to large surface area induced side reactionsbetween the electrode and electrolyte.Here we reports a facile procedure to prepare a family of

cubic mesostructures of nanoparticle aggregates confined incubes on a large scale instead of simple random nanoparticlesto simultaneously overcome all the issues discussed above, inparticular, achieving electrode materials with high packingdensity for lithium ion batteries. This family of cubicmesostructures with various chemicals compositions andbuilding substructures, including (a) ZnSn(OH)6, (b) evenlydistributed Zn2SnO4 and SnO2 nanoparticles, (c) hollow cubesof SnO2 nanoparticles, (d) high-ordered nanoparticles ofZn2SnO4&Sn@C, (e) SnO2@C core−shell subunits, (f)SnO2@C nanoparticle aggregates enclosed with oxidizedcarbon sheath, and (g) cubes with C nanobubbles as buildingunits, will provide a rich pool of materials with differentchemical and physical properties to meet demands in differentapplications. The overall idea and procedure with experimentalconditions involved in each step are illustrated in Figure 1. Tothe best of our knowledge, family members c−g above havenever been reported before. We also demonstrated, for the firsttime, that family members of b and d can be used as anodematerials in lithium ion batteries with very high packing densitywith improve performances in lithium storage.

2. EXPERIMENTAL SECTIONPreparation of Mesocubes of ZnSn(OH)6. Mesocubes of

ZnSn(OH)6 as the starting family member was synthesized througha room-temperature coprecipitation method. Typically, calculatedamounts of SnCl4 and ZnCl2 were dissolved in 50 mL of ethanol understirring, followed by the addition of 50 mL of an aqueous solution ofNaOH (0.32 M) drop-by-drop in 5 min. The mixture was stirred for 1h and kept at room temperature without stirring for another 23 h. Thewhite precipitate was collected by centrifugation and washed withethanol and water several times to remove residual ions. TheZnSn(OH)6 powder was dried in a conventional oven overnight.

Preparation of Mesocubes of Evenly Distributed Zn2SnO4and SnO2 Nanoparticle Aggregates. Mesocubes ofZn2SnO4&SnO2 nanoparticle aggregates were prepared by calcinatingmesocubes of ZnSn(OH)6. To tune the size of the building units ofevenly distributed Zn2SnO4 and SnO2 nanoparticle, mesocubes ofZnSn(OH)6 were calcinated at different temperatures of 650 and 800°C, at different ramping rates of 1 and 20 °C/min, respectively.

Preparation of Hollow Mesocubes of SnO2 NanoparticleAggregates. Typically, 50 mg of mesocubes of Zn2SnO4&SnO2

nanoparticle aggregates obtained with calcination at 650 °C describedabove were dispersed in 40 mL of 1.0 M HCl and kept for 24 h atroom temperature with stirring to etch off the Zn ions and core part ofthe cube. The white product was collected by centrifugation, washedwith deionized water several times until the solution became neutral,and then washed with ethanol and dried at 60 °C.

Preparation of Mesocubes of High-Order Zn2SnO4&[email protected], the porous mesocubes of evenly distributed Zn2SnO4 andSnO2 nanoparticle aggregates prepared by calcination at 800 °Cdescribed above were placed into a ceramic crucible and heated to 650°C in a quartz tube furnace with ramping rate of 20 °C/min under Arflow. The chemical vapor deposition (CVD) process was carried out at650 °C for 1 h with a flow of 100 sccm of mixture gas (10% acetylenewith argon as the balance). The tube furnace was purged with argonfor at least 1 h to remove oxygen before CVD and cooled downnaturally under argon after CVD. SnO2 was reduced to metallic Sn byacetylene under CVD conditions but not Zn2SnO4, and the black colorindicated a carbon coating.

Preparation of Mesocubes with SnO2@C Core−ShellSubunits. Typically, 25 mg of mesocubes of Zn2SnO4&Sn@Cobtained above was dispersed in 20 mL of 2 M HCl for 2 days toselectively etch off Zn ions and metallic Sn. The as-treated sample stillblack in color was collected by centrifugation, washed with deionizedwater several times until the filtrate became neutral, and then washedwith ethanol.

Preparation of Solid Mesocubes of SnO2@Oxidized CSheath Nanoparticle Aggregates. Typically, concentrated nitricacid (65 wt %, 15 M) was used to etch off Zn ions and oxidize metallicSn particles into SnO2 and oxidize the C sheath from the precursormesocubes of Zn2SnO4&Sn@C.

Preparation of Mesocubes of C Nanobubbles as BuildingUnits. The mesocubes of carbon were obtained by completelyremoving Zn and Sn enclosed in the mesocubes of Zn2SnO4&Sn@Cby washing Zn2SnO4&Sn@C with concentrated HCl (12 M) etchingfor 2 days. The black, acid-etched product was washed thoroughly andcollected.

Figure 1. Schematic of the idea and procedure to prepare the family of mesocubes with different compositions and building substructures: (a)mesocube of ZnSn(OH)6 as the starting family member; (b) cube of evenly distributed Zn2SnO4 and SnO2 nanoparticles obtained by annealing a;(c) hollow cubes of SnO2 nanoparticles aggregates obtained by selective etching b in 1 M HCl; (d) cube of high-ordered nanoparticles ofZn2SnO4&Sn@C aggregates obtained by CVD treatment of b under C2H2; (e) cube of SnO2@C core−shell subunits obtained by etching d in 2 MHCl; (f) cube of SnO2@C nanoparticle aggregates enclosed with an oxidized carbon sheath obtained by oxidizing and etching d under concentratedHNO3; and (g) cube with C nanobubbles as building units obtained by completely removing Zn and Sn elements in d with concentrated HCl.

Chemistry of Materials Article

dx.doi.org/10.1021/cm501580y | Chem. Mater. 2014, 26, 4472−44854473

Materials Characterization. Powder X-ray diffraction (XRD) wascarried out with a Rigaku Smartlab X-ray diffractometer using Cu Kαradiation (λ = 0.154 18 nm). The morphologies of the products wascharacterized by field emission scanning electron microscopy (JSM-7600 FE SEM, equipped with Pegasus Apex 2 integrated EDS, withaccelerating voltage of 15 kV) and by transmission electronmicroscopy (JEOL 2010 TEM instrument, with accelerating voltageof 200 kV).Electrochemical Measurements. A homogeneous slurry was

prepared by mixing 80 wt % of the as-prepared active materials, 10 wt% of conductivity enhancer (Super-P carbon black, Timcal), and 10 wt% of polyvinylidene fluoride (PVDF) binder in N-methylpyrrolidone(NMP). The slurry was then applied to copper disks as currentcollectors and dried in a vacuum oven at 80 °C for 24 h. Coin-typecells were assembled in an argon-filled glovebox using the coatedcopper disk as the working electrode, metallic lithium foil as thecounter electrode, 1 M solution of LiPF6 in a mixture of ethylenecarbonate (EC) and diethyl carbonate (DEC) (1:1, v/v) as theelectrolyte, and PP/PE/PP trilayer membrane (Celgard 2320) as theseparator. The electrochemical cells were charged and dischargedgalvanostatically at room temperature in the voltage window of 0.005−3 V on a MTI BST8-WA battery tester.

3. RESULTS AND DISCUSSIONSFigure 2 shows the XRD patterns of three important familymembers: (a) mesocubes of ZnSn(OH)6 as the starting

materials, (b) mesocubes of evenly distributed Zn2SnO4&SnO2nanoparticles obtained by annealing ZnSn(OH)6 mesocubes,and (c) mesocubes of high-ordered nanoparticles ofZn2SnO4&Sn@C aggregates obtained by CVD treatment ofZn2SnO4&SnO2 mesocubes. The XRD pattern of the startingfamily member obtained by coprecipitation can be assigned toprimitive cubic ZnSn(OH)6 (JCPDS card no. 20-1455). Noother peak was observed, indicating the purity of as-preparedZnSn(OH)6 (Figure 2a). The XRD pattern of sample preparedby calcinating ZnSn(OH)6 precursor at 800 °C can be assignedto Zn2SnO4 with cubic crystal structure (JCPDS card no. 24-1470) and tetragonal rutile SnO2 (JCPDS card no. 41-1445)and no other peaks observed (Figure 2b). This XRD patternsuggests the thorough conversion of ZnSn(OH)6 into Zn2SnO4and SnO2 under heat treatment. After the CVD treatmentunder acetylene, the composition of the products changed to amixture of Zn2SnO4 and tetragonal tin (JCPDS card no. 04-0673) (Figure 2c). The distinguishable peak at around 2θ = 26°for SnO2(110) in Figure 2b disappeared in Figure 2c, and acharacteristic couple of peaks at around 2θ = 30° and 32°

associated with metallic Sn appeared in Figure 2c after CVDtreatment, indicating the successful reduction of SnO2 to Sn byacetylene under CVD conditions. The reduction of SnO2 tometallic Sn under such a CVD condition is well-docu-mented.12−15

Mesocubes of ZnSn(OH)6. We identified zinc hydrox-ystannate [ZnSn(OH)6] as the starting family member for therich chemistry itself as well as its derivatives offered. Zinchydroxystannate can be easily prepared as cubic nanostructureson a large scale.16−19 Preparation of cubes about 100 nm in sizeat room temperature has been reported by Li et al. fromNa2SnO3·3H2O and ZnCl2

20 and by Cao et al. by grinding.21

Hollow cubes around 500 nm were prepared by Wang et al.through a room temperature alkali-assisted dissolutionprocess.19 Polyhedral microcrystals with core−shell structurewith size around 1 μm were prepared by a room temperatureNH3 bubble templating method.

22 The thermal decompositionof ZnSn(OH)6 into Zn2SnO4 and SnO2 has been well-documented.23−26 Those derivatives from decomposed ZnSn-(OH)6 not only offer different chemical and physical propertiesbut also can find important applications in gas sensor andlithium ion batteries. A size of around 1 μm was the largestamong all the zinc hydroxystannate prepared at roomtemperature, to the best of our knowledge. Here, we prepareduniform ZnSn(OH)6 cubes with a size of ∼2 μm without anysurfactants in the mixture of ethanol and water system at roomtemperature for the first time. We developed a simplecoprecipitation method with a water and ethanol mixture asthe solvent to provide the right conditions to generate largeamounts of mesocubes of ZnSn(OH)6 at room temperature.The chemical composition was confirmed by XRD (Figure 2a).The morphology of the as-prepred mesocubes of ZnSn(OH)6is revealed by FESEM at different magnifications in Figure 3a−c. All mesocubes of ZnSn(OH)6 are similar in size, as shown inthe low-magnification FESEM image (Figure 3a). The high-magnification FESEM image shows the perfect cubic structureof ZnSn(OH)6, with flat surfaces and sharp edges (Figure 3b,c),and some nanoparticles adsorbed on the surface were observed.The cubic structure and the solid nature of the mesocubes wereconfirmed by TEM (inset of Figure 3a). The XRD (Figure 2a)with sharp diffraction peaks and the nearly perfect cubicstructure from TEM and SEM characterization (Figure S2 inSupporting Information) all suggest that the meoscubes ofZnSn(OH)6 are highly crystalline. The EDS of ZnSn(OH)6 isshown in Figure 3d, and the atomic ratio of Zn:Sn is ∼1:1, asexpected.

Mesocubes of Distributed Zn2SnO4 and SnO2 Nano-particle Aggregates.We successfully confined both Zn2SnO4and SnO2 nanoparticles in mesocubes by simply calcinatingmesocubes of ZnSn(OH)6. Both Zn2SnO4 and SnO2 nano-particles are highly functional, attracting much attentionrecently. Zn2SnO4 nanoparticles can be used as transparentconducting oxide,27 in gas sensors,28 in conductive inks in inkjetprinting,29 in dye-sensitized solar cells (DSSCs),26 and as anodematerials in lithium ion batteries.30−32 Zn2SnO4 nanostructureswere typically prepared by complex and energy intensivemethods of hydrothermal, microwave-assisted hydrothermaland vapor transport approaches.18,25,29,33 On the other hand,SnO2 is a well-known wide band gap n-type semiconductor (3.6eV). SnO2 nanostructures have been intensively explored toenhance its performances in many applications, in particular,gas sensors and lithium ion batteries.34−37 The cubic structureof the ZnSn(OH)6 precursor provides the template to generate

Figure 2. Selected representative XRD patterns of mesocubes toconfirm the compositions of (a) ZnSn(OH)6, (b) Zn2SnO4&SnO2nanoparticle aggregates obtained by calcinating ZnSn(OH)6, and (c)Zn2SnO4&Sn@C obtained by CVD treatment of Zn2SnO4&SnO2,with all peaks assigned.



Figure 3. Mesocubes of ZnSn(OH)6: FESEM images at (a) low-magnification overall view showing a similar size distribution, and the inset shows atypical mesocube viewed under TEM; (b) high-magnification zoomed-in view of a few mesocubes showing a rough surface; and (c) zoomed-in viewof a typical mesocube with the surface clearly revealed. (d) EDS of mesocubes of ZnSn(OH)6, and the atomic ratio of Zn:Sn is ∼1:1.

Figure 4. Mesocubes of distributed Zn2SnO4 and SnO2 nanoparticle aggregates. FESEM images of (a) low-magnification overall view showing theperfect preservation of cubic structure after calcination treatment of its precursor of ZnSn(OH)6 and (b) high-magnifcaition zoomed-in view moreclearly showing the surface roughness and the aggregation of Zn2SnO4 and SnO2 nanoparticles and the porous nature. TEM images of (c) a singlemesocube with light contrast around the edges, suggesting porous structure, where dashed lines outline the orentiation of the mesocube and (d) thezoomed-in view more clearly shows the aggregation of building nanounits. The samples were obtained by calcinating mesocubes of ZnSn(OH)6 at800 °C.



nanoparticles of Zn2SnO4 and SnO2 confined locally anddistributed evenly. The thermal decomposition reaction shouldbe26

→ + +2ZnSn(OH) Zn SnO SnO 6H O6 2 4 2 2 (1)

The morphology of the mesocubes of Zn2SnO4&SnO2nanoparticle aggregates was revealed by FESEM and TEM(Figure 4). The low-magnification overall view shows that thecubic structures were well-preserved after thermal decom-position (Figure 4a). More details of the structures wererevealed in the high-magnification FESEM image (Figure 4b).The surface of the mesocubes of Zn2SnO4 and SnO2nanoparticle aggregates is porous, indicating successfulconversion and formation of particle aggregates as comparedto its precursor of solid ZnSn(OH)6 mesocubes. Themesocubes of mixed oxides were further characterized byTEM (Figure 4c). The light contrast and course edges clearlyreveal those subunits of evenly distributed Zn2SnO4 and SnO2nanoparticles. The high-magnification TEM image (Figure 4d)more clearly shows the subunits and their aggregation. Thereare evenly distributed nanoparticles in the size ranges of about35 nm and 12 nm, which could be assigned to Zn2SnO4 andSnO2 nanoparticles, respectively. The sizes measured fromhigh-magnification TEM agree with those estimated from XRD(Figure 2b). The nanoparticle building units are randomlydistributed inside the mainframe of cubes, forming mesocubesof nanoparticle aggregates. The void spaces between nano-particle aggregations are observed. Those porous structurescould facilitate the diffusion of acetylene during CVD throughthe cubes on the nanoscale. The porosity through the

mesocubes can also be indirectly proved by the uniformcoating of carbon, the complete reduction of distributed SnO2nanoparticles in mesocubes into Sn, and the successfulpreparation of mesocubes of carbon bubbles, as will bediscussed. The EDS results of as-prepared Zn2SnO4&SnO2are shown in Figure S1a (Supporting Information), with anatomic ratio of Zn:Sn of ∼1:1. The atomic ratio of O (46.55atom %) was decreased compared to the atomic ratio of O inZnSn(OH)6 (77.26 atom %) due to removal of the waterduring the heating process and the conversion of ZnSn(OH)6to Zn2SnO4 and SnO2.

23,24,26,38

Hollow Mesocubes of SnO2 Nanoparticle Aggregates.We successfully synthesized hollow mesocubes of SnO2nanoparticle aggregates by selectively removing all Zn2+ ionsfrom the mesocubes of Zn2SnO4&SnO2 nanoparticle aggre-gates discussed above. The rational design was based on ourunderstanding that that Zn2+ ions can be easly etched off bydilute hydrochloride acid.39 At the same time, SnO4

4− can beprotonated and dehydrated to SnO2, and SnO2 is relativelystable, preserving the Sn. The possible reactions involved are

+ → ++ +Zn SnO 4H 2Zn H SnO2 42

4 4 (2)

→ +H SnO SnO 2H O4 4 2 2 (3)

Therefore, upon the complete removal of Zn2+ ions, thereshould still be SnO2 nanoparticles remaining in the framework.The morphology of as-prepared hollow SnO2 mesocubes isrevealed by SEM and TEM (Figure 5). The cubic structure waswell-maintained, as revealed by low-magnification SEM (Figure5a). It is interesting to note that the as obtained mesocubes of

Figure 5. Hollow mesocubes of SnO2 nanoparticle aggregates: SEM images of (a) low-magnification overall view showing the well-preservedmesocubes and (b) high-magnification of a few typical mesocubes and the broken shell highlighted by arrows that reveals their hollow nature; TEMimages of (c) a few typical hollow mesocubes of SnO2 nanoparticle aggregates with clear contrast between the core and shell parts in each cube,which further confirmed that they are hollow and (d) high-magnification zoomed-in view of a corner of the mesocube that clearly shows theaggregation of SnO2 nanoparticles and shell thickness.



SnO2 are hollow. This is clearly revealed by the high-magnification view of a few typical mesocubes with brokensurfaces, and the holes highlighted by white arrows (Figure 5b).Addtionally, we observed that the surface of mesocubes ofSnO2 was slightly deflated (Figure 5b), which could be possiblyattributed to attack by acid etching, and that the void coregenerated could not support the shell as compared to the solidcore before etching. The hollow structure was confirmed byTEM (Figure 5c), with a clear contrast between the core andthe shell parts in each mesocube. The corner of a typical hollowmesocube is shown in a zoomed-in TEM image, and thethickness of the shell was about 280 nm (Figure 5d). The high-magnification image confirms the hollow nature of themesocubes and that the building units of the shell are SnO2nanoparticle aggregates. The hollow mesocubes were entirelyformed by aggregated nanoparticles closely compacted, andhigh-order random packing of nanoparticles creates nanoporesin the shell as well (Figure S3, Supporting Information). Theformation of unique hollow cubes, instead of porous solidstructures duplicated from its precursor of solid cubes, might beascribed as the following: During the preparation of mesocubesof Zn2SnO4&SnO2 nanoparticle aggregates by heating ZnSn-(OH)6, the heat was transferred from the shell to the core partof the cubes and the diffusion of water molecues in reversed indirection. Thus, the Zn2SnO4&SnO2 nanoparticle formed onthe shell should have a larger grain size as compared to thosenanoparticle aggregates formed in the core part. In other words,the smaller nanoparticles in the core part with larger surfaceenergy could be more easily attacked by acid etching ascompared to those bigger particles on the shell part.40

Otherwise stated, the shell will be more densely packed ascompared to the core part, forming hollow structures aftertreatement. The complete removal of Zn2+ ions is confiremd byEDS analysis (Figure S1b, Supporting Information), where theSn peaks are dominant without any distinguishable Zn peaks, incontrast to the EDS pattern of its precuror, Zn2SnO4&SnO2.

Mesocubes of High-Order Zn2SnO4&Sn@C Nano-particle Aggregates. Another family member of mesocubesof high-order Zn2SnO4&Sn@C were successfully prepared byCVD treatment of the mesocubes of distributed Zn2SnO4 andSnO2 nanoparticle aggregates as precursors. Here acetylene wasselected to play dual roles: (1) as reducing agent to selectivelyreduce SnO2 to metallic Sn and (2) as carbon source to coatZn2SnO4&Sn with carbon sheaths. The reduction of SnO2 tometallic Sn by acetylene is well-documented.14,15,41 The porous3-D structures of precursor of mesocubes of distributedZn2SnO4 and SnO2 nanoparticle aggregates provide 3-Dchannels for acetylene to diffuse through the inside of eachcube and allows CVD to occur locally. In other words, thecarbon sheaths can encapsulate Zn2SnO4&Sn nanoparticlesthrough the cubes. The successful reduction of SnO2 to metallicSn by acetylene was confirmed by XRD (Figure 2c). Theresults also suggest that Zn2SnO4 is highly stable under theCVD conditions. The morphology of the Zn2SnO4&Sn@Cmesocubes was revealed by FESEM and TEM (Figure 6). Theuniform cubic structure was well-preserved after the CVDprocess, as revealed by the low-magnification FESEM overallview (Figure 6a). The zoomed-in view of two typicalmesocubes of Zn2SnO4&Sn@C (Figure 6b) reveals that thereare structures like bubbles/broken bubbles formed on the cube,

Figure 6. Mesocubes of high-order Zn2SnO4&Sn@C nanoparticle aggregates: FESEM images of (a) low-magnification overall view and (b) high-magnificaton top-view of two typical mesocubes; TEM images of (c) a typical mesocubes with light contrast around the edges, indicating preservedporosity, and (d) high-magnificaton zoomed-in view more clearly showing the carbon sheath wrapping the Zn2SnO4&Sn nanoparticles. Thethickness of the carbon shell is 5 nm.



which confirmed the formation of carbon sheath. Additionally,the porous structures remained with the surface still highlyrough, which indicates that the carbon coating occurred locallyon the nanoscale. Otherwise stated, mesoscale cubic structureswere well-preserved with high-ordered Zn2SnO4&Sn@C nano-particles as building subunits. This is further confirmed byTEM (Figure 6c). The light contrast around the edges andrough edge lines all suggest its high porosity and theaggregation of Zn2SnO4&Sn@C nanoparticles. This is furtherconfirmed by high-magnification TEM clearly showing thedetails of the subunits (Figure 6d). The building subunits ofZn2SnO4&Sn@C nanoparticles are wrapped by carbon sheathsof about 5 nm in thickness or about 14 layers of graphenesforming the carbon shell. The arrangement and order of theaggregates were random, but there are visible layers of carbonshell wrapping those Zn2SnO4&Sn nanoparticles. The voidbetween particles is filled with carbon. In other words, theaggregation is more packed as compared to its precursor(Figure S4, Supporting Information). Elemental mapping andEDS analysis further confirmed the presence and evendistribution of elements (C, O, Sn, and Zn) in each mesocube(Figure 7). The uniform distribution of carbon indicates thatthe carbon formed throughout the whole cubes, not just on thesurface of cubes, as expected. Other family members of

mesocubes of SnO2@C aggregates and carbon bubbleaggregates derived from this mesocubes of high-orderZn2SnO4&Sn@C nanoparticle aggregates as discussed nextalso confirmed that there are carbon sheaths throughout thecube. Additionally, the EDS analysis shows that the atomicration of Zn:Sn remains 1:1, like its precursor (Figure 7f),indicating there was no Sn loss during the CVD process,although its melting point (less than 232 °C) is significantlylower than the CVD temperature (650 °C). This again suggestthat the carbon can completely encapsulate them. The carboncontent is 30.52 atom % from EDS analysis (Figure 7f).

Mesocubes with SnO2@C Core−Shell Subunits. Mes-ocubes with SnO2@C rattlelike core−shell as building subunitscould be derived from mesocubes of Zn2SnO4&Sn@C byetching with dilute HCl solution. This rational design was basedon our understanding that metallic Sn and Zn2+ ions can beeasily etched off in HCl solution, which is well-documented,but carbon is stable. Meanwhile, it is understood that SnO4

4−

can be protonated and dehydrated to SnO2 inside the carbonunder dilute HCl, forming SnO2@C core−shell particles. Theuniform cubic structure can be well-preserved after the acidetching, as shown in the low-magnification FESEM image(Figure 8a). The high-magnification FESEM in Figure 8bshows more details about the nanosphere subunits at the corner

Figure 7. Mesocubes of high-order Zn2SnO4&Sn@C nanoparticle aggregates analyzed by elemental mapping and EDS: (a) FESEM image of thetwo mesocubes selected for elemental mapping, (b−e) the corresponding elemental mapping of C, O, Sn, and Zn, and (f) EDS analysis revealingthat the atomic ration Zn:Sn remains 1:1 as before CVD.



of the mesocube (in the area marked by a red dash in Figure8a). The edges are light in contrast, indicating the partialremoval of core and that carbon bubbles remain at the edge andon the surface. This is further confirmed by TEM (Figure 8c).As expected, the overall aggregation of nanoparticle building

units was the same as its precursors due to the preservation ofcarbon shells. As compared to its precursor of mesocube ofZn2SnO4&Sn@C, the mesocube of SnO2@C is much lighter incontrast under TEM through the cube, which suggests theremoval of Zn2+ ions and metallic Sn. The high-magnification

Figure 8. Mesocubes with SnO2@C core−shell subunits: FESEM images of (a) low-magnification overall view and (b) high-magnification corner-view of one typical mesocube; TEM images of (c) a typical mesocube with clear contrast between dark SnO2 nanoparticle and light carbon preservedand (d) high-magnification zoomed-in view that clearly shows the SnO2@C core−shell nanoparticles as the building subunits and the space betweenthe dark SnO2 core and carbon shell; (e) FESEM image of the selected two mesocubes used for elemental mapping analysis and (f, g, and h)corresponding elemental mapping results for C, O and Sn, respectively.



TEM (Figure 8d) shows details about the SnO2@C core−shellnanosphere subunits, and there are spaces between the SnO2core and carbon bubbles, forming a rattlelike structure. Thereare also carbon bubbles with the whole core part removed,forming hollow carbon bubbles, which could be attributed tothe extract amount of HCl that can also remove SnO4

4− due tothe formation of H2[SnCl6]. The EDS mapping of SnO2@Cmesocubes in Figure 8e−h demonstrates the uniformity ofcarbon, oxygen, and tin on the mesocubes by the colors red,green, and yellow, respectively. The EDS results (Figure S1c,Supporting Information) show that carbon is dominant in thecomposition, with atomic ratio of 90.26%, while a small amount(3.72 atom %) of Sn element and O (6.03 atom %) exist in themesocubes.Mesocubes of SnO2 Nanoparticle Aggregates En-

closed by Oxidized C Sheaths. Mesocubes of SnO2nanoparticle aggregates enclosed by oxidized C sheaths withholes were derived from mesocubes of Zn2SnO4&Sn@C. Therational design was based on our understanding thatconcentrated HNO3, as a strong oxidant, can oxidize Sn backto SnO2. The redox reaction between Sn and nitric acid can beascribed as42

+ → + ↑ +Sn 4HNO (concd) SnO 4NO 2H O3 2 2 2 (4)

Note that metastannic acid (H2SnO3) may be generated as theintermediate under highly concentrated HNO3 but could easilydehydrate to form SnO2 during the drying process.43

Meanwhile, concentrated HNO3 has been widely used tooxidize and functionalize carbon and open carbon nanotubes.Under concentrated HNO3, amorphous carbon could form andgraphene layers could be cut open, providing additionaldefective sites and increasing the carbon reactivity.44 In this

case, highly concentrated HNO3 was used instead of diluteHCl, to obtain more SnO2 and to oxidize and open the carbonsheath. The EDS analysis of the mesocube of SnO2@oxidizedcarbon nanoparticle aggregates showed a much higher contentof Sn (51.3 atom %) and lower content of carbon (40.1 atom%) (Figure S1d,Supporting Information), as compared to theEDS of dilute HCl treated SnO2@C sample (Figure S1c,Supporting Information). Also, no peak for zinc can beobserved, indicating the full removal of Zn2+ ions. The uniformcubic structure can be preserved after the treatment underconcentrated HNO3, as revealed by low-magnification FESEM(Figure 9a). The high-magnification FESEM image of a fewtypical treated mesocubes shows more details and surfacetexture (Figure 9b). As compared to dilute HCl treatedmesocubes (Figure 8b), the mesocubes of SnO2@oxidizedcarbon have coarse surfaces, which could be attributed tooxidation of Sn and oxidization and opening of carbonsheaths.40,42,44 The porous structure and aggregation ofnanoparticles of the mesocube were further characterized byTEM (Figure 9c). As compared to the dilute HCl treatedsample of SnO2@C mesocubes (Figure 8c), the concentratedHNO3 treated cube has more solid nanosubunits due to moreSnO2 being encapsulated, which was added through theoxidization of metallic Sn into SnO2 by HNO3. A high-magnification TEM (Figure 9d) shows that the aggregatednanoparticles encapsulated within severely damaged carbonsheaths and holes were observed on the carbon sheaths.Additionally, the thickness of the carbon sheaths wassignificantly reduced to about 3 nm as compared to that ofits precursor (Figure 6d). The mechanism of breaking andthinning of carbon sheaths should be similar to shortening andthinning of multiwalled carbon nanotubes by concentrated

Figure 9.Mesocubes of SnO2 nanoparticle aggregates enclosed by oxidized C sheaths: (a) low-magnification overall view and (b) high-magnificationview of a few typical mesocubes with highly rough surface; TEM images of (c) typical mesocubes with light contrast around the edges and a darkbody and (d) high-magnification zoomed-in view that clearly shows the SnO2 nanoparticle and oxidized or damaged C sheaths with holes.



HNO3.44 The opening of carbon sheaths could provide

additional reactivity sites for the SnO2 nanoparticles encapsu-lated, and this should be beneficial to certain applications, e.g.,sensors and photocatalysis.Mesocubes of Carbon Nanobubbles. We also developed

the first general procedure to produce mesocubes of carbonnanobubbles using the mesocubes of Zn2SnO4&Sn@C asprecursor. This method, in principle, could be applied toprepare carbon bubble aggregates in different geometriesstarting from given ZnSn(OH)6 template with various shapesinstead of a cube, such as spheres or polyhedrons. Here, insteadof dilute HCl, concentrated HCl was used to completelyremove both Zn2SnO4 and Sn inside the precursor. Weunderstand that, under concentrated HCl, the H4SnO4generated according to eq 2 will undergo the followingreaction:45,46

+ → +H SnO 6HCl H [SnCl ] 4H O4 4 2 6 2 (5)

Therefore, both Zn and Sn can be completely removed frommesocubes of Zn2SnO4&Sn@C, which left behind carbon onlyunder concentrated HCl. The as-prepared mesocubes of carbonbubbles were characterized by FESEM and TEM (Figure 10).The low-magnification FESEM shows that the all-cubicstructures were well-preserved after the acid etching inconcentrated HCl (Figure 10a). The high-magnificationFESEM image of a few typical carbon mesocubes indicatesthat there was no collapse observed, with the cubic outlineswell-maintained. The good structure stability suggests thatthere are carbon bubbles packed inside the cube to support thestructure. The complete removal of both Zn2SnO4 and Sn ismore clearly revealed by TEM (Figure 10c). Compared to themesocube of Zn2SnO4&Sn@C precursor (Figure 6c), the acid-etched mesocube is much lighter in contrast under TEM, withthe size and shape well-preserved. The TEM image under highmagnification (Figure 10d) reveals some bubblelike hollow

Figure 10. Mesocubes of C nanobubbles: (a) low-magnification overall view and (b) high-magnification view of a few typical mesocubes; TEMimages of (c) a typical mesocubes of carbon nanobubbles as building subunits and (d) high-magnification zoom-in view clearly shows only carbonbubbles as the building subunits and the complete removal of both Zn2SnO4 and Sn; (e) FESEM image of one selected mesocube used for elementalmapping analysis and (f) the corresponding elemental carbon map.



structures, which were the hollow carbon structures obtainedby removing the core of Zn2SnO4&Sn. The overall surfacedistribution of the carbon bubbles is almost the same as itsprecursor of mesocubes of high-order Zn2SnO4&Sn@C, andthe light contrast of the TEM image also suggests that thewrapped cores of Zn2SnO4&Sn were almost completelyremoved (Figure S5, Supporting Information). This observa-tion again demonstrates that the carbon was uniformly coatedon all the subunits after CVD treatment, not just coated on theoutside surfaces of the mesocubes. EDS of the carbonmesocubes (Figure S1e, Supporting Information) shows thatall the peaks associated with the elements of Sn and Zn are notdistinguishable. It is more evidence of the feasibility of thismethod, and almost all of the Zn2SnO4 and Sn inside thecarbon was removed. The elemental mapping of the carbonmesocube in Figure 10f reveals the uniform distribution ofcarbon on the whole mesocube. The uniform coating of carbonon all nanosubunits, even at the core part of the mesocubes, ismore evidence to demonstrate that the Zn2SnO4&SnO2

mesocubes were porous so that the acetylene vapor can

transport into the mesocubes and deposited evenly through themesocubes.

Application of Mesocubes in LIBs with High PackingDensity. To demonstrate the tremendous potentials of thisfamily of mesocubes, we selected two members, mesocubes ofZn2SnO4&SnO2 and Zn2SnO4&Sn@C, to evaluate in thispreliminary investigation, although all the members areelectrochemically active in reversible lithium storage andother applications. Both nanoscale Zn2SnO4 and SnO2 havebeen extensively explored as carbon-alternative anode materialsfor lithium ion batteries (LIBs).18,39,47−53 However, thesynergistic effect of evenly distributed Zn2SnO4 and SnO2

confined in a cube on reversible lithium storage has beenrarely studied. In our case, although the mesocubes are inmicroscale, they are still electrochemically active, due to the factthat the nanoparticles of Zn2SnO4&SnO2 as the building unitsand the porous structure could facilitate the diffusion of Li ions.In other words, the salient advantages of nanoparticles, such asshort Li+ diffusion paths and high rate performance, are not losteven they are packed into mesocubes. Similarly, metallic Sn-

Figure 11. Electrochemical performances of mesocubes of Zn2SnO4&SnO2 and Zn2SnO4&Sn@C: (a) charge−discharge profiles of the first twocycles of mesocubes of Zn2SnO4&SnO2; (b) corresponding differential capacity profiles (dQ/dV vs V) of part a; (c) charge−discharge profiles of thefirst two cycles of mesocubes of Zn2SnO4&Sn@C; (d) corresponding differential capacity profiles (dQ/dV vs V) of part c; (e) cycling performancesof both mesocubes Zn2SnO4&SnO2 and Zn2SnO4&Sn@C at different currents of 50 and 100 mA/g.



based materials are also considered as an attractive lithium-storage material, with a theoretical capacity of 990 mA h g−1 (orLi4.4Sn).

54,55 However, Sn-based materials suffer from poorcyclability issues caused by volume changes during charging anddischarging. There are two main strategies to address the poorcyclability issue: (1) to prepare materials on the nanoscale and(2) to prepare Sn-based material/carbon composites. Forexample, Zn2SnO4@C,18,31 SnO2@C, and Sn@C55−59 havedemonstrated improved performance in LIBs. The encapsulat-ing of Zn2SnO4&Sn in thin carbon sheaths and packing into acube on the mesoscale for lithium storage have not beenreported. Another issue associated with nanoscale materialswidely reported in literature is low tapped density, which makesit difficult to improve the packing density of electrodes. Ourpreliminary results show that both mesocubes ofZn2SnO4&SnO2 and Zn2SnO4&Sn@C can be highly useful asboth high-energy and high-packing-density anode materials.The results of preliminary investigation of the electro-

chemical performances of both mesocubes of Zn2SnO4&SnO2and Zn2SnO4&Sn@C are summarized in Figure 11. In the firstcycle discharge profiles of both mesocubes of Zn2SnO4&SnO2(Figure 11a) and Zn2SnO4&Sn@C (Figure 11c), plateausaround 0.5 V (vs Li/Li+) are observed, which can be attributedto the lithium insertion into Zn2SnO4 and subsequentformation of alloy with Sn or Zn.32,60,61 This provide additionalelectrochemical evidence of the presence of Zn2SnO4 in bothmesocubes. The first cycle irreversible capacity losses (ICLs)for mecosubes of Zn2SnO4&SnO2 and Zn2SnO4&Sn@C are39.6% and 24.9%, respectively. This first cycle ICLs can beattributed to irreversible reduction of SnO2 and Zn2SnO4 andformation of solid electrolyte interphase (SEI) in the former,while only irreversible reduction of Zn2SnO4 and formation ofSEI in the later as SnO2 has been chemically reduced to metallicSn under CVD. The difference in electrochemical reactionsinvolved in the two mesocubes could explain the difference inthe first cycle ICL observed. From the second cycle onward, thecharge−discharge profiles are highly overlapped, indicating thesame electrochemical reactions. Given all the active materialsinvolved in the two mesocubes, the possible electrochemicalreactions are32,62

+ + → + ++ −Zn SnO 8Li 8e 2Zn Sn 4Li O2 4 2 (6)

+ + → ++ −SnO 4Li 4e Sn 2Li O2 2 (7)

+ + ↔ + ≤ ≤+ −x x xSn Li e Sn Li Sn (0 4.4)x (8)

+ + ↔ ≤+ −y y yZn Li e Li Zn ( 1)y (9)

+ + ↔+ −6C Li e LiC6 (10)

To better interpret the reactions involved in the twomesocubes during the charge/discharge cycles, differentialcapacity profiles (dQ/dV vs V) were plotted (Figure 11b,d).For both mesocubes with the presence of the commoncomponent Zn2SnO4, the cathodic peaks at about 0.45 and 0.14V for the first discharge processes are observed, which can beattributed to eq 6 and forward eqs 8 9, respectively. The anodicpeaks at ∼0.6 and 1.34 V for the first charge process can beattributed to the backward eqs 8 and 9 and a partially reversiblereaction in eq 6. For the second cycle, the cathodic peak at 0.45V disappeared and another dominant peak at ∼1 V wasobserved, indicating different lithium insertion reactions.32,60,61

Keeping in mind the different compositions of the twomesocubes, unique peaks associated with SnO2 were observedin mesocubes of Zn2SnO4&SnO2 (Figure 11b), and uniquepeaks associated with metallic Sn and carbon sheaths wereobserved in mesocubes of Zn2SnO4&Sn@C (Figure 11d). ForZn2SnO4&SnO2, a stronger cathodic peak for SnO2 reductionat about 0.85 V can be attributed to the reaction of SnO2 withlithium ions and the formation of Sn and Li2O (Figure 11b),according to eq 7, which was not observed in Zn2SnO4&Sn@C(Figure 11d), indicating the presence and absence of SnO2 inthe former and latter, respectivlely, as expected. ForZn2SnO4&Sn@C, the unique cathodic peak at about 0.6 Vcould be assigned to the alloy of Li and Sn and the formation ofLixSn (Figure 11d), according to eq 8, while the oxidationpeaks between 0.4 and 0.8 V in the charging cycle could beassigned to dealloying reactions of LixSn.

14 Additionally, therewere no irreversible metallic Sn surface reaction peaks between1.05 and 1.55 V observed in the first discharge cycle, suggestingthat no metallic Sn was exposed to electrolyte, which providesmore evidence to show that all the metallic Sn was encapsulatedby carbon phase. In fact, the broad satellite peak around 0.9 Vcould be assigned to the formation of SEI on the carbon surfacedue to decomposition of electrolyte.The capacity vs cycle number plots for the two mesocubes of

Zn2SnO4&SnO2 and Zn2SnO4&Sn@C are shown in Figure11e. The specific capacities of 448 and 542 mAh/g wereobta ined for mesocubes of Zn2SnO4&SnO2 andZn2SnO4&Sn@C, respectively, after 20 cycles tested at acurrent of 50 mA/g. When the currents were doubled to 100mA/g, there was no noticeable fading in capacities observed(392 and 512 mA h/g for Zn2SnO4&SnO2 and Zn2SnO4&Sn@C, respectively), which indicates that the materials may havegood rate performance. Zn2SnO4&Sn@C even shows animproved rate performance, for it only had very small fadingof 30 mA h/g, smaller than the fading of 56 mA h/g forZn2SnO4&SnO2 when charge/discharge currents were doubled.After 35 cycles, capacities of 250 and 370 mA h g−1 were stillmaintained for mesocubes of Zn2SnO4&SnO2 andZn2SnO4&Sn@C, respectively. Although the initial capacity ofthe former is higher, the carbon coating could significantlyimprove the cycling performance of the latter. Carbon coatingcould improve electrode conductivity and buffer volumevariation, which are beneficial to cycling performance.It is particularly interesting to highlight that the mesocubes

of both Zn2SnO4&SnO2 and Zn2SnO4&Sn@C nanoparticleaggregates have high tapped densities. Figure 12 illustrates thevolume occupied by the same weight of mesocubes ascompared to commercial TiO2 (AEROXIDE TiO2 P25). Thetapped densities for Zn2SnO4&SnO2 and Zn2SnO4&Sn@Cwere estimated to be 1.14 and 0.98 g/cm3, respectively, whichare much higher than the tapped density of commercial TiO2 at0.13 g/cm3. The significant high tapped densities of bothmesocubes of Zn2SnO4&SnO2 and Zn2SnO4&Sn@C can beattributed to close-compaction of cubic structures on themesoscale. High packing density is highly desirable to achieveuseful batteries and dramatically reduce the volume taken up bybattery systems for various applications. To illustrate thesignificance of high packing density, the capacity densities (byvolume) were estimated on the basis of specific capacities (bymass). The specific capacities of mesocubes of Zn2SnO4&SnO2and Zn2SnO4&Sn@C at the 35th cycle are 250 and 370 mA hg−1, and we assume that commercial TiO2 has a theoreticalspecific capacity of 168 mA h g−1. We find the capacity densities



for mesocubes of Zn2SnO4&SnO2 and Zn2SnO4&Sn@C andP25 TiO2 to be 285, 363, and 22 mA h cm−3, respectively.Otherwise stated, capacity densities could be about 13 and 17times higher than that of commercial TiO2, even based on non-optimized mesocubes.

4. CONCLUSIONSIn summary, we reports a facile procedure to prepare a family ofcubic mesostructures of nanoparticle aggregates confined incubes on a large scale to simultaneously overcome issues ofdifficulty in synthesis, tuning compositions, and properties, inparticular, achieving electrode materials with high packingdensity for LIBs. This family of cubic mesostructures withvarious chemical compositions and building substructures willprovide a rich pool of materials with different chemical andphysical properties to meet demands in different applications.The chemical and physical properties of materials were tunedby altering the compositions of the mesocubes while the samecubic structures were preserved, ranging from semiconductors,including Zn2SnO4 and SnO2, to conductive materials,including Sn and carbon. The carbon mesocubes formed byaggregation of carbon nanobubbles as the building subunitswere reported. We also demonstrated, for the first time, thattwo family members of mesocubes, Zn2SnO4&SnO2 andZn2SnO4&Sn@C, can be used as anode materials in lithiumion batteries with very high packing densities. It is our ongoingeffort to explore applications for all members of this family ofmesocubes, and the results will be updated once available.

■ ASSOCIATED CONTENT*S Supporting InformationEDS results of Zn2SnO4&SnO2 mesocubes, hollow SnO2mesocubes, and mesocubes prepared from Zn2SnO4&Sn@C,including SnO2@carbon, SnO2@ oxidized carbon and carbonbubble aggregates, obtained by treated with dilute HCl,

concentrate HNO3 and concentrate HCl, respectively, andEM images. This material is available free of charge via theInternet at http://pubs.acs.org.

■ AUTHOR INFORMATIONCorresponding Author*E-mail: [email protected] Contributions‡S.A. and J.Z. contributed equally.NotesThe authors declare no competing financial interest.

■ ACKNOWLEDGMENTSWe thank the Lumigen Instrument Center, Wayne StateUniversity, Detroit, MI.

■ REFERENCES(1) Zaera, F. ChemSusChem 2013, 6, 1797.(2) Biju, V. Chem. Soc. Rev. 2014, 43, 744.(3) Zhang, Q. F.; Uchaker, E.; Candelaria, S. L.; Cao, G. Z. Chem.Soc. Rev. 2013, 42, 3127.(4) Wang, H. L.; Dai, H. J. Chem. Soc. Rev. 2013, 42, 3088.(5) Wagemaker, M.; Mulder, F. M. Acc. Chem. Res. 2013, 46, 1206.(6) Chen, G. Y.; Seo, J.; Yang, C. H.; Prasad, P. N. Chem. Soc. Rev.2013, 42, 8304.(7) Jariwala, D.; Sangwan, V. K.; Lauhon, L. J.; Marks, T. J.; Hersam,M. C. Chem. Soc. Rev. 2013, 42, 2824.(8) Ma, J.; Zhu, Z. L.; Chen, B.; Yang, M. X.; Zhou, H. M.; Li, C.; Yu,F.; Chen, J. H. J. Mater. Chem. A 2013, 1, 4662.(9) Ding, C. Q.; Zhu, A. W.; Tian, Y. Acc. Chem. Res. 2014, 47, 20.(10) Potyrailo, R.; Naik, R. R. Annu. Rev. Mater. Res. 2013, 43, 307.(11) Lin, C. H.; Liu, X. Y.; Wu, S. H.; Liu, K. H.; Mou, C. Y. J. Phys.Chem. Lett. 2011, 2, 2984.(12) Lee, S. H.; Mathews, M.; Toghiani, H.; Wipf, D. O.; Pittman, J.C. U. Chem. Mater. 2009, 21, 2306.(13) Deng, D.; Kim, M. G.; Lee, J. Y.; Cho, J. Energy Environ. Sci.2009, 2, 818.(14) Deng, D.; Lee, J. Y. J. Mater. Chem. 2010, 20, 8045.(15) Deng, D.; Lee, J. Y. ACS Nano 2009, 3, 1723.(16) Yin, J.; Gao, F.; Wei, C.; Lu, Q. Inorg. Chem. 2012, 51, 10990.(17) Cai, P.; Ma, D.-K.; Liu, Q.-C.; Zhou, S.-M.; Chen, W.; Huang,S.-M. J. Mater. Chem. A 2013, 1, 5217.(18) Zhao, Y.; Huang, Y.; Wang, Q.; Wang, K.; Zong, M.; Wang, L.;Sun, X. Ceram. Int. 2014, 40, 2275.(19) Wang, L.; Tang, K.; Liu, Z.; Wang, D.; Sheng, J.; Cheng, W. J.Mater. Chem. 2011, 21, 4352.(20) Wei, W.; Yang, S.; Zhou, H.; Lieberwirth, I.; Feng, X.; Mullen,K. Adv. Mater. 2013, 25, 2909.(21) Cao, Y.; Jia, D.; Zhou, J.; Sun, Y. Eur. J. Inorg. Chem. 2009, 2009,4105.(22) Fan, H.; Ai, S.; Ju, P. CrystEngComm 2011, 13, 113.(23) Jena, H.; Kutty, K. V. G.; Kutty, T. R. N. Mater. Chem. Phys.2004, 88, 167.(24) Stambolova, I.; Toneva, A.; Blaskov, V.; Radev, D.; Tsvetanova,Y.; Vassilev, S.; Peshev, P. J. Alloys Compd. 2005, 391, L1.(25) Wrobel, G.; Piech, M.; Dardona, S.; Ding, Y.; Gao, P.-X. Cryst.Growth Des. 2009, 9, 4456.(26) Liu, R.; Du, W.; Chen, Q.; Gao, F.; Wei, C.; Sun, J.; Lu, Q. RSCAdv. 2013, 3, 2893.(27) Young, D. L.; Moutinho, H.; Yan, Y.; Coutts, T. J. J. Appl. Phys.2002, 92, 310.(28) Chen, Z.; Cao, M.; Hu, C. J. Phys. Chem. C 2011, 115, 5522.(29) Lehnen, T.; Zopes, D.; Mathur, S. J. Mater. Chem. 2012, 22,17732.(30) Cherian, C. T.; Zheng, M.; Reddy, M. V.; Chowdari, B. V. R.;Sow, C. H. ACS Appl. Mater. Interfaces 2013, 5, 6054.

Figure 12. Optical image to compare the same weight of (a)mesocubes of Zn2SnO4&SnO2 nanoparticle aggregates, (b) mesocubesof Zn2SnO4&Sn@C nanoparticle aggregates, and (c) commercial TiO2nanoparticles (AEROXIDE, P25). The tapped densities are estimatedto be 1.14, 0.98, and 0.13 g/cm3 for a, b, and c, respectively.



http://pubs.acs.org

mailto:[email protected]

(31) Zhao, Y.; Huang, Y.; Zhang, W.; Wang, Q.; Wang, K.; Zong, M.;Sun, X. RSC Adv. 2013, 3, 23489.(32) Rong, A.; Gao, X. P.; Li, G. R.; Yan, T. Y.; Zhu, H. Y.; Qu, J. Q.;Song, D. Y. J. Phys. Chem. B 2006, 110, 14754.(33) Li, B.; Luo, L.; Xiao, T.; Hu, X.; Lu, L.; Wang, J.; Tang, Y. J.Alloys Compd. 2011, 509, 2186.(34) Gyger, F.; Hubner, M.; Feldmann, C.; Barsan, N.; Weimar, U.Chem. Mater. 2010, 22, 4821.(35) Zhang, J.; Wang, S.; Xu, M.; Wang, Y.; Xia, H.; Zhang, S.; Guo,X.; Wu, S. J. Phys. Chem. C 2009, 113, 1662.(36) Lou, X. W.; Wang, Y.; Yuan, C.; Lee, J. Y.; Archer, L. A. Adv.Mater. 2006, 18, 2325.(37) Lou, X. W.; Deng, D.; Lee, J. Y.; Archer, L. A. Chem. Mater.2008, 20, 6562.(38) Wang, C.; Wang, X. M.; Zhao, J. C.; Mai, B. X.; Sheng, G. Y.;Peng, P. A.; Fu, J. M. J. Mater. Sci. 2002, 37, 2989.(39) Zhang, L.; Wu, H. B.; Liu, B.; Lou, X. W. Energy Environ. Sci.2014, 7, 1013.(40) Ma, G.; Zou, R.; Jiang, L.; Zhang, Z.; Xue, Y.; Yu, L.; Song, G.;Li, W.; Hu, J. CrystEngComm 2012, 14, 2172.(41) Deng, D.; Lee, J. Y. Angew. Chem., Int. Ed. 2009, 48, 1660.(42) Mecucci, A.; Scott, K. J. Chem. Technol. Biotechnol. 2002, 77,449.(43) Khallaf, H.; Chen, C.-T.; Chang, L.-B.; Lupan, O.; Dutta, A.;Heinrich, H.; Haque, F.; del Barco, E.; Chow, L. Appl. Surf. Sci. 2012,258, 6069.(44) Rosca, I. D.; Watari, F.; Uo, M.; Akasaka, T. Carbon 2005, 43,3124.(45) Xu, J.; Zhang, C.; Qu, H.; Tian, C. J. Appl. Polym. Sci. 2005, 98,1469.(46) Al-Hinai, A. T.; Al-Hinai, M. H.; Dutta, J. Mater. Res. Bull. 2014,49, 645.(47) Poizot, P.; Laruelle, S.; Grugeon, S.; Dupont, L.; Tarascon, J.Nature 2000, 407, 496.(48) Bruce, P. G.; Scrosati, B.; Tarascon, J. M. Angew. Chem., Int. Ed.2008, 47, 2930.(49) Ji, L.; Lin, Z.; Alcoutlabi, M.; Zhang, X. Energy Environ. Sci.2011, 4, 2682.(50) Liu, F.; Song, S.; Xue, D.; Zhang, H. Nanoscale Res. Lett. 2012, 7,149.(51) Zhu, J.; Ng, K. Y.; Deng, D. ACS Appl. Mater. Interfaces 2014, 6,2996.(52) Zhang, H.-X.; Feng, C.; Zhai, Y.-C.; Jiang, K.-L.; Li, Q.-Q.; Fan,S.-S. Adv. Mater. 2009, 21, 2299.(53) Zhang, W.-M.; Hu, J.-S.; Guo, Y.-G.; Zheng, S.-F.; Zhong, L.-S.;Song, W.-G.; Wan, L.-J. Adv. Mater. 2008, 20, 1160.(54) Du, W.; Wang, Q.; Saxner, D.; Deskins, N. A.; Su, D.;Krzanowski, J. E.; Frenkel, A. I.; Teng, X. J. Am. Chem. Soc. 2011, 133,15172.(55) Derrien, G.; Hassoun, J.; Panero, S.; Scrosati, B. Adv. Mater.2007, 19, 2336.(56) Wang, X.; Cao, X.; Bourgeois, L.; Guan, H.; Chen, S.; Zhong, Y.;Tang, D.-M.; Li, H.; Zhai, T.; Li, L.; Bando, Y.; Golberg, D. Adv. Funct.Mater. 2012, 22, 2682.(57) Wang, G.; Wang, B.; Wang, X.; Park, J.; Dou, S.; Ahn, H.; Kim,K. J. Mater. Chem. 2009, 19, 8378.(58) Wang, Y.; Wu, M.; Jiao, Z.; Lee, J. Y. Chem. Mater. 2009, 21,3210.(59) Yu, Y.; Gu, L.; Wang, C.; Dhanabalan, A.; van Aken, P. A.;Maier, J. Angew. Chem., Int. Ed. 2009, 48, 6485.(60) Zhu, X. J.; Geng, L. M.; Zhang, F. Q.; Liu, Y. X.; Cheng, L. B. J.Power Sources 2009, 189, 828.(61) Kim, K.; Annamalai, A.; Park, S. H.; Kwon, T. H.; Pyeon, M. W.;Lee, M.-J. Electrochim. Acta 2012, 76, 192.(62) Deng, D.; Lee, J. Y. Chem. Mater. 2008, 20, 1841.



A Family of Mesocubes

Documents