3486 Chem. Mater.

2006, 18, 3486-3494

Oxides@C Core-Shell Nanostructures: One-Pot Synthesis, Rational

Conversion, and Li Storage Property
Xiaoming Sun,† Junfeng Liu,† and Yadong Li*
Department of Chemistry, Tsinghua UniVersity, Beijing 100084, People’s Republic of China
ReceiVed NoVember 30, 2005. ReVised Manuscript ReceiVed February 13, 2006

A general and facile one-pot hydrothermal method was developed to prepare oxide@C core-shell
nanostructures with carbonaceous polysaccharide shells and oxides (including hydroxides or complex
oxides) cores. “Coupled synthesis” or “post-synthesis” approaches can be applied. As-formed particles
possess hydrophilic and biocompatible shells and variable cores, thus have promising applications in the
areas such as life science or colloid science. The nanostructures can be transformed to a metal@C core/
shell structure or hollow carbon structure. Samples were characterized with X-ray diffraction, transmission
electron microscopy, high-resolution transmission electron microscopy, electron energy loss spectroscopy,
and Fourier transform infrared spectroscopy. Electrochemical characterization on calcined SnO2@C
nanostructures showed unusual reversible Li storage capacities. This implies the structural optimization
might provide new opportunities to enhance the properties of solids in nanoscale.

Introduction Fe@C) or oxides@C (e.g., V2O3@C) nanoparticles.14-21

These nanoparticles exhibited excellent properties especially
Core-shell nanostructures have stimulated great interest in battery application. It was found that carbon coated
in the previous decades because of their tunable surface CaMoO4 would give significantly enhanced reversible Li
properties, enhanced optical, electronic, and catalytic proper- storage capacities.14 The cycle performance of nanosized Sn
ties, and consequent wide potential applications.1,2 Many metal particles could be improved by encapsulation in
advantages are attributed to the core-shell structure. For spherical hollow carbon.15 Carbon hollow structures, typically
instance, when nanoparticles are used in biosystems to give in the form of capsules converted from their core/shell
optical or magnetic signals (e.g., Au, CdSe, or Fe3O4), precursors, exhibited higher current and power density when
coating a passive shell comprised of inert materials (e.g., used as a catalyst support in the direct methanol fuel cell.16
silica or polymer) on the surfaces of colloids can isolate the Many approaches have been explored to synthesize carbon-
cores from external chemical environments and prevent their shelled nanoparticles, such as microemulsion interface po-
toxicity, biodegradation, or other chemical interactions.3-6 lymerization and carbonizatoin,15,17,18 template synthesis from
On the other hand, the cores might support or cooperate with cellulose chars,19 sealed reaction at elevated temperature from
the shells, enhance their properties, or afford composite alkoxides,20 arc discharge in methane or alcohol,21 and so
properties in one system.7-13 forth.
Very recently, core-shell structured nanoparticles with a Most of the core-shell structured nanoparticles mentioned
carbon shell have stimulated great interest. A series of works above were prepared through the “post-synthesis” encapsula-
have been done on the synthesis of metal@C (e.g., Sn@C, tion method.1-13,15 Otherwise, some special equipment (e.g.,
arc discharge devices21 or elevated temperature16,19,20) was
* Corresponding author. Fax: (+86) 10-6278-8765. E-mail: ydli@ required for the formation of the core-shell structure. So
[Link]. developing a general and facile method which uses inex-
† Contributed equally to this work.

(1) Caruso, F. AdV. Mater. 2001, 13, 11. pensive, “user-friendly” reagents and safe, easy-operating
(2) Schartl, W. AdV. Mater. 2000, 12, 1899. equipment to synthesize carbon-shelled core-shell nano-
(3) Lu, Y.; Yin, Y. D.; Mayers, B. T.; Xia, Y. N. Nano Lett. 2002, 2,
183. structures would open an avenue to exploring the physical
(4) Wang, Y.; Teng, X. W.; Wang, J. S.; Yang, H. Nano Lett. 2003, 3,
789. (14) Sharma, N.; Shaju, K. M.; Rao, G. V. S.; Chowdari, B. V. R.; Dong,
(5) Wang, Y. A.; Li, J. J.; Chen, H. Y.; Peng, X. G. J. Am. Chem. Soc. Z. L.; White, T. J. Chem. Mater. 2004, 16, 504.
2002, 124, 2293. (15) Lee, K. T.; Jung, Y. S.; Oh, S. M. J. Am. Chem. Soc. 2003, 125,
(6) Gohy, J. F.; Willet, N.; Varshney, S.; Zhang, J. X.; Jerome, R. Angew. 5652.
Chem., Int. Ed. 2001, 40, 3214. (16) Han, S. J.; Yun, Y. K.; Park, K. W.; Sung, Y. E.; Hyeon, T. AdV.
(7) Cho, G. J.; Fung, B. M.; Glatzhofer, D. T.; Lee, J. S.; Shul, Y. G. Mater. 2003, 15, 1922.
Langmuir 2001, 17, 456. (17) Yoon, S. B.; Sohn, K.; Kim, J. Y.; Shin, C. H.; Yu, J. S.; Hyeon, T.
(8) Oldfield, G.; Ung, T.; Mulvaney, P. AdV. Mater. 2000, 12, 1519. AdV. Mater. 2002, 14, 19.
(9) Hirakawa, T.; Kamat, P. V. J. Am. Chem. Soc. 2005, 127, 3928. (18) Jang, J.; Ha, H. Chem. Mater. 2003, 15, 2109.
(10) Zhang, Y.; Li, Y. D. J. Phys. Chem. B 2004, 108, 17805. (19) Herring, A. M.; McKinnon, J. T.; McCloskey, B. D.; Filley, J.;
(11) Teng, X. W.; Yang, H. J. Am. Chem. Soc. 2003, 125, 14559. Gneshin, K. W.; Pavelka, R. A.; Kleebe, H. J.; Aldrich, D. J. J. Am.
(12) Teng, X. W.; Black, D.; Watkins, N. J.; Gao, Y. L.; Yang, H. Nano Chem. Soc. 2003, 125, 9916.
Lett. 2003, 3, 261. (20) Pol, S. V.; Pol, V. G.; Gedanken, L. [Link]. J. 2004, 10, 4467.
(13) Yi, D. K.; Selvan, S. T.; Lee, S. S.; Papaefthymiou, G. C.; Kundaliya, (21) Dong, X. L.; Zhang, Z. D.; Jin, S. R.; Kim, B. K. J. Appl. Phys. 1999,
D.; Ying, J. Y. J. Am. Chem. Soc. 2005, 127, 4990. 86, 6701.

10.1021/cm052648m CCC: $33.50 © 2006 American Chemical Society

Published on Web 06/23/2006
Oxides@C Core-Shell Nanostructures Chem. Mater., Vol. 18, No. 15, 2006 3487

and chemical properties of the distinguished nanostructures

and extending their applications.
In previous research, we have encapsulated noble metal
nanoparticles in carbonaceous microspheres using glucose
as the carbon source22 and got shape and optical property
manipulation on inner metal cores.23,24 Herein, this strategy
is extended to develop a general and facile method for
preparation of core-shell nanostructures with carbonaceous
polysaccharide shells and oxides, including hydroxide or Figure 1. Schematic illustration of one-pot synthesis, conversion, and
oxide halogenate compound cores. decoration of oxide@C nanostructures.

Experimental Section H800, operated at 200 kV), high-resolution transmission electron

microscopy (HRTEM; JEOL, 2010F, operated at 200 kV), and
All the reagents used were A.R. grade, purchased from Beijing Fourier transform infrared (FT-IR) spectroscopy (Nicolet 560).
Chemical Reagent Factory. Electrochemical Testing. Electrodes were prepared by coating
Coupled Synthesis of Sb8O11Cl2@C, BiOCl@C, Sb6O13@C, a copper foil substrate with the slurry of the active materials (82
SnO2@C, and MnCO3@C Nanoparticles. A total of 10 mmol wt %), carbon black (10 wt %), and poly(vinylidene fluoride) (8
SbCl3, 10 mmol BiCl3, 1 mmol SbCl5, 0.5-5 mmol SnCl2 (or wt %) dissolved in cyclopentanone. After coating, the slurry was
SnCl4), and 2 mmol KMnO4 were dissolved in 0.5 M aqueous dried at 100 °C for 12 h and then pressed between two stainless
glucose solutions, respectively. After vigorous stirring for 10 min, steel plates at 1 MPa. Prior to cell assembling, the electrodes with
the suspensions were transferred into autoclaves and kept at 180 area of 0.64 cm2 were dried at 120 °C for 4 h under vacuum. The
°C for 4 h. After reaction, the autoclaves were cooled naturally in testing cells had a typical two-electrode construction using a
air, and the suspensions were isolated by filtration and washed with polypropylene microporous sheet as the separator, and 1 M LiPF6
deionized water and alcohol three times. dissolved in ethyl carbonate and diethyl carbonate (1:1, v/v) are
Post-Synthesis of FeOOH@C, Pr(OH)3@C, and Fe3O4@C used as the electrolyte. A pure lithium foil as the negative electrode
Nanoparticles. FeOOH, Pr(OH)3, and Fe3O4 nanoparticles were and the SnO2@C nanoparticle film described above are used as
synthesized following previous reports.25-27 The first two were the working electrode. The cells were assembled in an argon-filled
vigorously stirred and ultrasoniated in aqueous glucose solutions glovebox. The cells were charged and discharged at a constant
to get even dispersions. However, Fe3O4 nanoparticles must be current of 0.10 mA and cycled between 0.00 and 2.00 V.
ultrasonicated for 10 min in 0.1 M HNO3 before being dispersed
in the aqueous glucose solution. The suspensions were transferred Results and Discussion
to autoclaves and kept at 180 °C for 4 h. The rest of the rinse steps
were the same as previously described for the coupled synthesis. The synthesis and conversion strategy of these composite
Calcination of Sb8O11Cl2@C and BiOCl@C Nanostructures. nanostructures are schematically shown in Figure 1. Encap-
The nanoparticles obtained after rinsing were accumulated in an sulation of these inorganic nanoparticles in the carbonaceous
alumina boat, which was set in the hot zone of an alumina tube shell could be accomplished by “coupled synthesis” or “post-
furnace. The content was heated at 850 °C for 2 h in a high purity synthesis”. Figure 1 shows “coupled synthesis”. The starting
argon atmosphere under ambient pressure. Then the furnace was system is a clear solution of glucose. Other solvable
naturally air cooled to room temperature under the protection of
saccharides are also applicable. As hydrolyzable metal salts
argon. Calcination of SnO2@C was performed following the same
steps except the temperatures were altered to designed ones. are introduced into the solution, they hydrolyze and dehydrate
Conversion of BiOCl in BiOCl@C to Bi Nanospheres. A total at ambient or hydrothermal conditions to form nano- or
of 0.1 g of BiOCl@C nanoparticles was dispersed in 30 mL of micro-oxide (or hydroxide, complex oxide) particles (step
water. Then 5 mL of 80% hydrazine was added. After stirring for I, as-formed particles are schematically illustrated as rods).
10 min, the suspension was sealed in autoclaves and hydrothermally As our previous works have indicated,22 the saccharides
treated at 180 °C for 10 h. Products were filtered and rinsed with would not carbonize unless the hydrothermal condition
deionized water. overcomes the threshold condition (e.g., time > 2 h at 180
Loading of Ag Nanoparticles on the Surface of Sb8O11Cl2@C °C), which could accommodate recrystallization or many
Nanoparticles. Freshly prepared Sb8O11Cl2@C nanoparticles (50 kinds of chemical reactions. The dissolved saccharides begin
mg) were ultrasonicated and dispersed in AgNO3 solution (1.0 M,
to dehydrate and carbonize as the threshold is overcome as
10 mL). The sonication continued for a total of 10 min, and then
the solution was filtrated to erase excess AgNO3. A water rinse evidenced by Raman and IR.22 As-formed carbonaceous
was repeated three times before the sample was dispersed in alcohol materials encapsulate former oxide particles to form a thin
at the aid of sonication for transmission electron microscopy (TEM) sheath (step II). The mechanism was drawn inspired by the
observation. Ag@C core/shell structure formation mechanism we stud-
Characterization. Products were characterized with X-ray ied.23 In that case, the Ag core formed first, and the
diffraction (XRD; Bruker D8 Advance X-ray diffractometer with carbonaceous shell formed consequently and covered the Ag
monochromatized Cu KR radiation, λ ) 1.5418 Å), TEM (Hitachi core. But for the case of the oxide@C core/shell structure,
it was not definitely demonstrated yet. In addition, it is true
(22) Sun, X. M.; Li, Y. D. Angew. Chem., Int. Ed. 2004, 43, 597. that formation of the carbonaceous shell could influence the
(23) Sun, X. M.; Li, Y. D. Langmuir 2005, 21, 6019.
(24) Sun, X. M.; Li, Y. D. AdV. Mater. 2005, 17, 2626. final shape of the particles as demonstrated by experiment.
(25) Chen, M.; Tang, B.; Nikles, D. E. IEEE Trans. Magn. 1998, 34, 1141. The carbonaceous sheath is penetrable for small molecules
(26) Wang, X.; Li, Y. D. Angew. Chem., Int. Ed. 2002, 41, 4790.
(27) Lian, S. Y.; Kang, Z. H.; Wang, E. B.; Jiang, M.; Hu, C. W.; Xu, L. such as amine or hydrazine.23 Thus, the oxide cores can be
Solid State Commun. 2003, 127, 605. removed by chemical or thermal means and leaving behind
3488 Chem. Mater., Vol. 18, No. 15, 2006 Sun et al.

formed after ultrasonication reduction on the surface of the

carbonaceous shell (Figure 3E,F) were very small in size
(usually <10 nm) because of the mild reaction conditions.
Calcination in noble gas (Ar) to high temperature of 850
°C could drive off the Sb compound cores and carbonize
the shells. As formed carbon nanotubes have high purity,
uniform shell thickness (∼8 nm) and clear edges (Figure
3G,H). They had essentially the same external size of Sb8O11-
Cl2@C nanostructures, indicating a “replica” of Sb8O11Cl2
nanorod templates. The samples after calcination had local-
ized graphitization zones of ∼2.5 nm in diameter (Figure
3I), exhibiting typical characteristics of “hard carbon”, which
Figure 2. XRD patterns of (A) Sb8O11Cl2@C nanoparticles and (B) C were obtained from polyol or polyglycol precursors and were
nanotubes obtained after calcination. hard to graphitize.29 XRD patterns (Figure 2B) gave very
weak peaks corresponding to graphite (JCPDS 751621).
hollow carbonaceous shells as replicas of inner cores (step Electron energy loss spectroscopy (EELS) recorded on the
III). Otherwise, the oxide cores could react with the carbon- carbon nanotubes indicated that their profile was similar to
aceous shell or other introduced chemical reagents to convert that of amorphous carbon (Figure 3J). Two peaks at 284.5
to core-shell nanostructures, such as metal@C nanoparticles eV and 293.4 eV could be indexed to the π* and σ* K-edge
(step IV). Because carbonaceous shells stem from hydro- peaks of carbon, respectively. The former was stronger than
thermally treated saccharides, rich functional groups could the amorphous one, which might be related to the localized
be inherited from starting materials and thus could be used carbonization of samples.
for loading noble metal nanoparticles (step V). Besides
BiOCl@C Nanoparticles. On the basis of the consider-
hydrolyzation, various chemical reactions can be introduced
ation of the chemical reactivity similarity, the synthetic
under aforementioned hydrothermal conditions before car-
strategy has been extended to prepare BiOCl@C nanopar-
bonization of saccharides. This significantly enriches the
ticles. Figure 4A,B shows the typical TEM images of BiOCl
accessible core-shell structured nanoparticles.
nanorods and nanoplates encapsulated in carbonaceous shells
Sb8O11Cl2@C Nanoparticles. The Sb8O11Cl2@C nano- (<10 nm). The nanorods were tens of nanometers in diameter
particle has been studied as the first example. SbCl3 naturally and hundreds of nanometers in length. Plates were also
hydrolyzed as it was dispersed in aqueous glucose solutions. hundreds of nanometers in edge length. After calcination at
The precipitates were further hydrolyzed and crystallized 850 °C, BiOCl was reduced by a carbonaceous shell to Bi,
under hydrothermal conditions to form well-crystallized which led to shrinkage in volume, and left behind many parts
nanorods. These nanorods were encapsulated in the carbon- of a hollow carbon sheath (Figure 4C,D). When the reduction
aceous shell as glucose dehydrated under hydrothermal was carried out in aqueous hydrazine solution at 180 °C, Bi
conditions. XRD characterization (Figure 2A) indicated that nanospheres would form because of the spacial localization
as-formed nanorods were mainly phase pure Sb8O11Cl2 effects of the carbonaceous shell (Figure 4E). This phenom-
(JCPDS 771583). Some unmarked peaks might be due to enon indicated that the carbon shell formed after the
unknown Sb-O-Cl complex structure. All the Sb8O11Cl2 hydrothermal treatment is still penetrable for small molecules
nanorods, 60-100 nm in diameter and several micrometers such as hydrazine. It was in accord with our former Ag-
in length, were encapsulated in carbonaceous sheaths (6- dissolving experiment.23 In comparative experiments, when
10 nm) as revealed by TEM images (Figure 3A,B). naked BiOCl nanoparticles were reduced with hydrazine in
Because complex oxides of Sb, O, and Cl are usually autoclaves, micrometer sized irregular particles could form.
decomposable at a temperature lower than 320 °C and the XRD characterization indicated that both the samples were
boiling point of SbCl3 can be as low as 223 °C,28 the samples phase pure (Figure 4F). Following the synthetic strategy,
tended to transform into hollow structures when they were TiO2@C nanoparticles can be prepared using TiCl4 as the
irradiated by focused electron beams, which was comparable starting material (see Supporting Information). In addition,
to heating under vacuum. Figure 3A,B showed samples if SbCl3 and SnCl4 were added simultaneously, composite
before and after electron beam irradiation for 1 min. The oxides could be prepared in the carbonaceous shell (see
originally fully filled cabonaceous sheath (as marked with Supporting Information).
an ellipsoid circle in Figure 3A) turned into hollow structures
Adjustment of Sb and Sn Valence States. Because
after focused irradiation despite the inner cores being single
glucose is a mild reducing reagent, it may show the
crystals according to HRTEM and selected-area electron
oxidizability when strong reducing agents are introduced.
diffraction characterization (Figure 3C,D). Hydroxyl groups
Otherwise, its reducibility will be shown if relatively strong
distribute all over the shells, because the carbonaceous shells
oxidizing agents are introduced. Thus, the valence state of
stemmed from glucose in aqueous solution. They could be
final oxides can be “self-adjusted” by glucose. When SbCl5
used to reduce noble metal salt ions.22 Ag nanoparticles
was hydrolyzed and hydrothermally treated in glucose
(28) An, J. J.; Chen, Z. C. Inorganic Compound Synthesis Handbook;
Chemical Engineering Publisher: Beijing, China, 1986, 1st Edition, (29) Wang, Q.; Li, H.; Chen, L. Q.; Huang, X. J. Solid State Ionics 2002,
ISBN I5063-3726. 152, 43.
Oxides@C Core-Shell Nanostructures Chem. Mater., Vol. 18, No. 15, 2006 3489

Figure 3. TEM and HRTEM images of Sb8O11Cl2@C nanoparticles: (A) A typical initial sample; (B) the same sample with the inner core volatilized under
electron beam irradiation; (C and D) HRTEM images, (inset) corresponding electron diffractions; (E) Ag nanoparticles-doped sample; (F) sample before
doping; (G) pure amorphous carbon nanotubes obtained by thermally driving out the inner cores; (H) HRTEM image of a bundle of nanotubes; (I) partly
magnified image to show the local crystallinity; and (J) EELS spectrum of C nanotubes.

solutions, they would convert to Sb6O13 (upper curve in temperature (400 and 500 °C), the grain size of SnO2
Figure 5, JCPDS 330111). When SnCl2 was introduced into nanoparticles increased but no new phase formed (Figure
the aqueous solution, SnO2 would form (bottom curve in 7B,C). TEM images indicated that samples after calcination
Figure 5, JCPDS 770448). remained core-shell structures approximately the same as
TEM and HRTEM characterizations gave structural in- the precursive products despite the SnO2 proportion getting
formation for SnO2@C nanoparticles prepared from SnCl2 smaller (Figure 6E) or bigger (Figure 6F). HRTEM images
(Figure 6). General images of the nanostructures indicated gave further insight into the SnO2@C nanostructure after
submicrometer branch-like frameworks (Figure 6A,C). Partly calcination: the central parts of the nanostructure comprised
magnified images (Figure 6B,D) revealed actual core-shell a large number of ellipsoidal nanoparticles (Figure 6G); at
structures for each “branch”. The samples were tilted by the edge, very short nanorods could be observed (Figure 6H).
(25° to ensure that the nanoparticles were encapsulated in These images indicated some anisotropic growth tendency
the carbon shell rather than attached on the surface. The of SnO2 nanoparticles, which might be attributed to the
shape of SnO2 nanoparticles encapsulated can be tuned by anisotropic growth and dynamic diffusion process during
adjusting the SnCl2 concentration. For instance, ∼4 nm formation of SnO2 nanocrystals.
roughly round particles formed when 0.1 g of SnCl2 was The carbon thermal reduction of SnO2 could take place
added (Figure 6B), whereas nanorods ∼10 nm in length when the calcination temperature reached 600 °C. The Sn
formed when 1.0 g SnCl2 was added (Figure 6D). The element emerged as a new phase in XRD patterns (Figure
nanoparticles and nanorods aggregated and were encapsu- 7D). At the same time, SnO2 peaks in the XRD pattern were
lated by the surrounding carbonaceous bulk solids. significantly sharpened. This implied that the existence of
Both the samples exhibited essentially the same XRD elemental Sn might accelerate the SnO2 grain size growth.
patterns (Figure 7A). When calcined at relatively low After calcination at 700 °C for 2 h, the conversion from SnO2
3490 Chem. Mater., Vol. 18, No. 15, 2006 Sun et al.

Figure 4. (A and B) Typical TEM images of BiOCl nanorods and nanoplates encapsulated in amorphous carbonaceous shells; (C and D) typical TEM
images of the hollow carbon sheath formed after calcination at 850 °C; (E) Bi nanospheres formed when BiOCl@C was reduced by hydrazine; and (F) XRD
patterns of the samples.

to Sn was almost complete (Figure 7E). TEM characteriza-

tion indicated formation of Sn@C nanostructures (Figure 6I).
MnCO3@C Nanoparticles. The redox reaction induced
precipitation and encapsulation could be applied not only
on the main-group metal but also on the transition metal
compounds. The synthesis of MnCO3@C nanoparticles gave
a good example. As soon as KMnO4 and glucose were mixed
at room temperature, the following reaction took place:

24MnO4- + 5C6H12O6 + 6H2O f

24Mn(OH)2 +24OH- + 30CO2

As-formed Mn(OH)2 could convert to phase pure MnCO3

under hydrothermal conditions (Figure 8A). Carbonization
of residual glucose could lead to encapsulation of the MnCO3
particles to form MnCO3@C nanoparticles (as shown in
Figure 8B). Recrystallization of inner cores under hydro- Figure 5. XRD patterns of Sb6O13 (upper) and SnO2 (bottom) obtained
after hydrothermal treatment in an aqueous glucose solution.
thermal conditions could densify the cores and produce a
separating air layer between the MnCO3 cores and the
carbonaceous shell (Figure 8C). After ultrasonication in 0.1 From Coupled Synthesis to Post-Synthesis Encapsula-
M HCl for 10 min and aging overnight, the core-shell tion. To further enrich the accessible types of compounds@C
structures turned hollow because of the removal of inner core-shell structures, the synthesis strategy was extended
cores (Figure 8D,E). from “coupled synthesis” to “post-synthesis” encapsulation.
Oxides@C Core-Shell Nanostructures Chem. Mater., Vol. 18, No. 15, 2006 3491

Figure 6. (A and B) TEM images of SnO2@C nanoparticles with 0.1 g of SnCl2 used for hydrolyzation; (C and D) TEM images of SnO2@C nanoparticles
with 1.0 g of SnCl2 used for hydrolyzation; (E) typical TEM image of the same sample as in part B after calcination at 500 °C for 2 h; (F) typical TEM
image of the same sample as in part D after calcination at 500 °C for 2 h; (G and H) HRTEM images of the same sample as in part F; and (I) typical TEM
image of the same sample as in part D after calcination at 700 °C for 2 h.

Figure 7. XRD patterns of SnO2@C nanoparticles: (A) hydrothermally

treated sample; (B) sample calcined at 400 °C for 2 h; (C) sample calcined
at 500 °C for 2 h; (D) sample calcined at 600 °C for 2 h; and (E) sample Figure 8. (A) XRD pattern; (B and C) TEM images of MnCO3@C core-
calcinated at 700 °C for 2 h. shell structured particles; and (D and E) TEM images of the carbonaceous
shell left behind after core removal.
Some examples of oxide@C or hydroxide@C are shown
below. the core entirely (inset in Figure 9B). Acid washing could
Encapsulation of Pr(OH)3 nanorods was investigated as remove the cores and leave behind carbonaceous sheathes,
the first example. Figure 9A showed the TEM image of the that is, the nanotubes (Figure 9C). Rice-like FeOOH el-
nanorods synthesized in concentrated KOH solution.26 After lipsoidal nanoparticles could be encapsulated in the carbon-
encapsulation in glucose solution under the hydrothermal aceous shell through a similar procedure (Figure 9D).
condition, sheathes (∼8 nm) formed on the hydroxide In contrast to the encapsulation of hydroxides or hydroxyl
nanorods cores (Figure 9B). The partially magnified image oxides that possess rich hydroxyl groups on the surface,
indicated that the sheath was smooth, uniform, and covering encapsulation of oxides required a process of activation. If
3492 Chem. Mater., Vol. 18, No. 15, 2006 Sun et al.

Figure 9. TEM images of (A) raw Pr(OH)3 nanorods; (B) Pr(OH)3@C nanorods; (C) carbonaceous nanotubes left behind after removal of inner Pr(OH)3
nanorods; (D) FeOOH@C nanoparticles; and (E) Fe3O4@C nanorods.

not, carbonaceous microspheres rather than core-shell fading.15,31 Amorphous composite tin oxide materials seem
structured nanoparticles could form accompanying naked able to overcome the disadvantages30 where Li was stored
oxide nanoparticles. According to our preliminary experi- by a two-step reaction mechanism:32-34 lithium first reacts
mental results, the carbonaceous shell seemed to have weak with tin oxide to produce amorphous Li2O and Sn irreversibly
acidity and tended to adhere on basic surfaces with rich (formula 1) and then Sn alloys with Li reversibly (formula
hydroxyl groups (e.g., FeOOH). The surface redox reaction 2). The Li2O matrix acts as a buffer and dispersive medium
could induce the in situ precipitation of carbonaceous units for nanosized Sn grains, which is critical for improving the
to form sheaths profitably. The Fe3O4@C nanoparticle is circling performance. Inactive glass-forming elements such
another example. The Fe3O4 nanorods prepared following as B or P act as a stabilizer to improve the cyclic performance
the coprecipitation method in basic solutions were immersed further.30,35,36
in 0.1 M HNO3, H2SO4, and HCl solutions, respectively, to
activate the particle surface. Only those particles activated 2Li+ + SnO (or SnO2) + 2e- f Sn +Li2O (1)
in HNO3 were coated with uniform carbonaceous sheaths
(Figure 9E). This implied that the redox reaction between xLi+ + Sn + xe- f LixSn (x e 4.4) (2)
adsorbed nitrate ions and glucose on the oxide surface might
facilitate the carbonaceous coating process. Generally speak- yLi+ + C6 + ye- T LiyC6 (y e 4.4) (3)
ing, the chemical interactions of the surface functional groups
and the coating agents were critical for encapsulation. It is obvious that if the buffer medium itself has Li storage
Li Storage Property of SnO2@C Nanoparticles. En- capacities and high electronic conductivity, the capacity of
capsulation of compound nanoparticles in a nontoxic car- the composite tin oxide anodes could be further enhanced.
bonaceous shell could effectively endow rich functional The carbonaceous matrix from glucose was demonstrated
groups on the surface and consequent promising applications to be an excellent candidate.29,37
of the core/shell particles in the areas such as biology or Figure 10A showed the charge and discharge curves of
diagnostics. In addition to that, the encapsulation of oxide SnO2@C core-shell nanostructures after calcination at 400
particles in carbonaceous sheaths also provides an op- °C for 200 min. The anode in the first discharge process
portunity to investigate the distinct properties of the com- had a long potential plateau and a large capacity loss (∼55%).
posite nanostructures. An example for SnO2@C core-shell It could be attributed to the irreversible reaction 1 and
nanostructure is shown below.
It is well-known that Sn, SnO2, and hard carbon from (31) Yang, J.; Winter, M.; Besenhard, J. O. Solid State Ionics 1996, 90,
saccharides are all Li storage materials.15,29,30 Metallic Sn 281.
(32) Courtney, I. A.; Dahn, J. R. J. Electrochem. Soc. 1997, 144, 2045.
has a theoretical specific capacity of 992 mA‚h/g, which is (33) Liu, W. F.; Huang, X. J.; Wang, Z. X.; Li, H.; Chen, L. Q. J.
much higher than that of already-commercialized graphite Electrochem. Soc. 1998, 145, 59.
(34) Li, H.; Huang, X. J.; Chen, L. Q. Electrochem. Solid State Lett. 1998,
(∼350 mA‚h/g).15 However, its (and its alloy’s) commercial 1, 241.
use are hindered by their drastic volume variation during Li (35) Hayashi, A.; Nakai, M.; Tatsumisago, M.; Minami, T.; Katadab, M.
insertion/extraction cycles, which leads to fast capacity J. Electrochem. Soc. 2003, 150, A582.
(36) Wan, K. B.; Li, S. F. Y.; Gao, Z. Q.; Siow, K. S. J. Power Sources
1998, 75, 9.
(30) Idota, Y.; Kubota, T.; Matsufuji, A.; Maekawa, Y.; Miyasaka, T. (37) Xing, W. B.; Dunlap, R. A.; Dahn, J. R. J. Electrochem. Soc. 1998,
Science 1997, 276, 1395. 145, 62.
Oxides@C Core-Shell Nanostructures Chem. Mater., Vol. 18, No. 15, 2006 3493

Figure 10. (A) Electrochemical performance of the SnO2@C nanoparticles electrode in a test Li battery. (B) Relationship between the specific discharge
capacity (black) and Coulombic efficiency (blue) and the cycling number.

irreversible trapping of lithium by the carbonaceous matrix. SnO2 is a good semiconductor. The decomposed product
The latter might relate to the surface ([H], [O], etc.) reaction from tin oxides is a nanocomposite as evidenced before.34,43,44
and solid electrolyte interface formation.38 For the SnO2@C core-shell nanostructures, the SnO2
However, the reversible capacity of the second discharge nanoparticles are embedded in the carbonaceous shell, which
process (857 mA‚h/g) was far beyond our expectation. SnO2 ensures that formed Li2O and Sn contact well without other
has a theoretical specific reversible capacity of 786 mA‚h/ separation. In addition, internal Li2O/Sn products are elec-
g. According to the thermogravimetric analysis (TGA) tronically contacted directly with the carbonaceous shell
results, SnO2 was ∼46% in weight content. That means the during cycling. This provides a kinetic advantage for the
specific capacity from SnO2 should not exceed 361 mA‚h/ reverse reaction 1. We believe that reaction 1 is at least
g. The rest of the carbon should contribute 496 mA‚h, partially reversible in our case, which contributes to the
corresponding to 918 mA‚h/g carbon. It is known that the unusually high reversible capacity.
reversible Li storage capacity of hard carbon is hard to The cyclic performance of the as-prepared SnO2@C
beyond 600 mA‚h/g.39 According to experimental results, material is not satisfying, which may need further optimiza-
the second discharge capacity of nano-SnO2 prepared fol- tion. Despite that, this result illustrates a good prospect that
lowing the same hydrothermal (180 °C) and thermal (400 the capacity of oxide anodes may be significantly enhanced
°C, 200 min) procedure was only 681 mA‚h/g, and that of by optimized microstructure. Moreover, because many
hard carbon following the same procedure was only 156.6 transition-metal compounds have been demonstrated as
mA‚h/g. Both of the samples faded quickly (see Supporting promising high-capacity negative-electrode materials for
Information, Figure s8). Where do the excessive capacities lithium-ion batteries,40,42 there still exists plenty of room for
come from? oxide@C core-shell composite structures in Li storage
Some precursive work on SnO2 anodes provided some properties. Further research is in process.
hints to give elucidation. It has been reported that transition
metal oxides show reversible Li storage behaviors.40 In those Conclusion
cases, a transition metal can react with Li2O upon Li
Several types of oxide@C nanostructrues were synthesized
extraction. That means reaction 1 is reversible, which
in one pot, following “coupled synthesis” or “post-synthesis”
contributes very high reversible capacity. Up to now, no approaches. The methods can be extended to the synthesis
experimental evidence shows whether reaction 1 for Sn-based of core-shell structures of other types of compounds such
oxides is completely irreversible or not. However, one paper as composite oxides, sulfides, nitrides, carbides, or metal. It
has mentioned that Li-O bonds are not stable when the also provided an opportunity to synthesize carbide or carbide-
charging voltage is above 1.3 V.41 A recent work indicated shelled nanostructures. Because the particles possess hydro-
that reaction 1 for most of metal compounds is reversible in philic and biocompatible shells and variable cores, their
view of the thermodynamics, and the theoretic voltage for applications in the areas such as life science or colloid science
SnO2 is 1.5 V. The reversal of reaction 1 from LiX (e.g., are promising. SnO2@C nanostructures showed unusual
Li2O) and metal (e.g., Sn) back to metal compounds is mainly reversible Li storage capacities. This implies that the
influenced by the intrinsic conductivity of MX compound, structural optimization might provide new opportunities to
grain size of LiX and metal, separated distance between LiX enhance the properties of solids in nanoscale.
and metal, and electronic contact with conductive additive.42
Acknowledgment. The authors truly thank Dr. Hong Li and
(38) Winter, M.; Besenhard, J. O.; Spahr, M. E.; Novak, P. AdV. Mater. Prof. Joseph F. Chiang for helpful discussions and Mr. Jie Shu
1998, 10, 725.
(39) Dahn, J. R.; Zheng, T.; Liu, Y. H.; Xue, J. S. Science 1995, 270, 590.
(40) Poizot, P.; Laruelle, S.; Grugeon, S.; Dupont, L.; Tarascon, J. M. (43) Brousse, T.; Retoux, R.; Herterich, U.; Schleich, D. M. J. Electrochem.
Nature 2000, 407, 496. Soc. 1998, 145, 1.
(41) Courtney, I. A.; Dahn, J. R. J. Electrochem. Soc. 1997, 144, 2943. (44) Retoux, R.; Brousse, T.; Schleich, D. M. J. Electrochem. Soc. 1999,
(42) Li, H.; Balaya, P.; Maier, J. J. Electrochem. Soc. 2004, 151, A1878. 146, 2472.
3494 Chem. Mater., Vol. 18, No. 15, 2006 Sun et al.

for electrochemical measurements. This work was supported Supporting Information Available: Three TEM images, three
by NSFC (90406003, 20401010, 50372030, 20025102, XRD patterns, three specific capacity vs circling time curves, a
20131030), the Foundation for the Author of National Excellent FT-IR spectrum, and TGA curves. This material is available free
Doctoral Dissertation of P.R. China, and the state key project of charge via the Internet at [Link]
of fundamental research for nanomaterials and nanostructures
(2003CB716901). CM052648M