Synthesis of Hematite (r-Fe
2
O
3
) Nanorods: Diameter-Size and Shape Effects on Their
Applications in Magnetism, Lithium Ion Battery, and Gas Sensors
Changzheng Wu, Ping Yin, Xi Zhu, Chuanzi OuYang, and Yi Xie*
Department of Nano-materials and Nano-chemistry, Hefei National Laboratory for Physical Sciences at
Microscale, UniVersity of Science & Technology of China, Hefei, Anhui 230026, China
ReceiVed: June 1, 2006; In Final Form: July 16, 2006
We demonstrated in this paper the shape-controlled synthesis of hematite (R-Fe
2
O
3
) nanostructures with a
gradient in the diameters (from less than 20 nm to larger than 300 nm) and surface areas (from 5.9 to 52.3
m
2
/g) through an improved synthetic strategy by adopting a high concentration of inorganic salts and high
temperature in the synthesis systems to influence the final products of hematite nanostructures. The benefits
of the present work also stem from the first report on the <20-nm-diameter and porous hematite nanorods,
as well as a new facile strategy to the less-than-20-nm nanorods, because the less-than-20-nm diameter size
meets the vital size domain for magnetization properties in hematite. Note that the porous and nonporous
hematite one-dimensional nanostructures with diameter gradients give us the first opportunity to investigate
the Morin temperature evolution of nanorod diameter and porosity. Evidently, the magnetic properties for
nanorods exhibit differences compared with those for the spherical particle counterparts. Hematite nanorods
are strongly dependent on their diameter size and porosity, where the magnetization is not sensitive to the
size evolution from submicron particles to the 60-90 nm nanorods, while the magnetic properties change
significantly in the case of <20 nm. In other words, for the magnetic properties of nanorods, in a comparable
size range, the porous existence could also influence the magnetic behavior. Moreover, applications in
formaldehyde (HCHO) gas sensors and lithium batteries for the hematite nanostructures with the diameter/

transition temperature (T
M
)ofR-Fe
2
O
3
nanoparticles (0-D)
decreases with decreasing spherical particle size according to a
1/d dependence.
3
Additionally, 1-D R-Fe
2
O
3
nanostructures,
such as nanorods,
4
nanowires,
5
nanobelts,
6
and nanotubes
7
have
also been synthesized and used for investigating their peculiar
properties. For example, Woo et al. synthesized R-Fe
2
O
3
nanorods by a sol-gel mediated reaction of ubiquitous Fe

Since the Morin temperature of R-Fe
2
O
3
spherical particles
was found to be strongly dependent on particle size and tends
to disappear (<5 K) below a diameter of 8-20 nm,
11
their
counterpart nanorods/nanowires with a diameter of <20 nm are
then significantly necessary for further understanding of the
magnetic properties. Currently, because of limited studies on
R-Fe
2
O
3
nanorods/nanowires with a diameter of <20 nm and
because their <20 nm and porous nanorods have not been
obtained so far, their subsequent applications in magnetization
fields and investigations on the size-dependent properties of iron
oxides are significantly delayed. Herein, we demonstrate the
synthesis of R-Fe
2
O
3
nanorods with a gradient in the surface
* Corresponding author. Address: Department of Nano-materials and
Nano-chemistry. Hefei National Laboratory for Physical Sciences at
Microscale, University of Science and Technology of China, Hefei, Anhui
230026, P. R. China. Tel: 86-551-3603987. Fax: 86-551-3603987. E-

2
O
3
nanorods with diameters of
<20 nm. Evidently, the magnetic properties were strongly
dependent on the size of their diameter and the porosity in the
present work, while the lithium intercalation and HCHO gas
sensor properties were significantly dependent on the surface
area. Therefore, the present work provides not only the first
example of investigating the magnetic property evolution of
nanorods/nanowire diameters and porosity, but also the first
example of the fabrication of hematite nanostructure sensors
for detecting HCHO gas.
2. Experimental Section
To prepare FeOOH nanostructure precursors, 50 mL of 0.06
M iron chloride (FeCl
3
) aqueous solution, with/without the
addition of 0.300 mol of inorganic salts (NH
4
Cl, KCl, and Na
2
-
SO
4
) was put in a conical flask and stirred with a magnetic
stirrer for 30 min. The homogeneous solution was then
transferred into a 60 mL Teflon-lined stainless steel autoclave,
sealed, and then heated to 120 °C. After the autoclaves were
maintained at 120 °C for 12 h, the resulting yellow product

O
3
were
obtained on a Hitachi Model H-800 instrument with a tungsten
filament at an accelerating voltage of 200 kV. The selected-
area electron diffraction patterns and high-resolution transmis-
sion electron microscopy (HRTEM) images were obtained on
a JEOL-2010 TEM at an acceleration voltage of 200 kV. The
porosity and adsorption performance of R-Fe
2
O
3
were deter-
mined via a Micromeritics ASAP-2000 nitrogen adsorption
apparatus. The magnetic properties of R-Fe
2
O
3
were measured
using a vibrating sample magnetometer and superconducting
quantum interference device. The performance of the R-Fe
2
O
3
as a cathode was evaluated using a Teflon cell with a lithium
metal anode. The cathode was a mixture of β-FeOOH/acetylene
black/poly(vinylidene fluoride) with a weight ratio of 85/10/5.
The electrolyte was 1 M LiPF
6
in a 1:1 mixture of ethylene

was injected into the chamber by a microinjector. The sensitivity
could be measured when the detecting gas was mixed with air
homogeneously. Here, the response magnitude, S, is defined as
R
s(air)
/R
s(gas)
, where R
s(air)
and R
s(gas)
are the resistance of the
sensor in clean air and in detected gas, respectively.
3. Results and Discussion
3.1. Morphology, Characterization, and Formation Mech-
anism of the As-Obtained Hematite Nanostructures. The
hematite nanostructures could be originated from the well-
controlled FeOOH nanostructure precursors (see Supporting
Information), and all the synthetic conditions are summarized
in Table 1. Furthermore, the phase and morphology information
for the as-obtained products are revealed by Figure 1, where
panels c, f, i, and m are the corresponding XRD patterns for
panels a-b, d-e, g-h, and j-l, respectively. All the XRD
patterns in Figure 1 show characteristics of pure hexagonal
R-Fe
2
O
3
(JCPDS card 33-664, a ) 5.035 Å and c ) 13.74 Å).
No characteristic peak was observed for other impurities such

4
systems
S4 nanorods with porosity nanorods: 5-19
porosity: 2-16
FeCl
3
-NH
4
Cl systems
Synthesis of Hematite (R-Fe
2
O
3
) Nanorods J. Phys. Chem. B, Vol. 110, No. 36, 2006 17807
is shown in Figure 1a,b, the appearance of S1 is the
hematite submicrometer particles with the diameter range of
300-500 nm, and no hollow structures were found even from
the amplified TEM image for the direct hydrolysis system. The
calcined products (S2) from the FeCl
3
-KCl system have the
regular pores (20-50 nm) distributed along the hematite
nanorods with a diameter-size range of 60-90 nm, as shown
in Figure 1d,e. The calcining products (S3) obtained in the
FeCl
3
-Na
2
SO
4

nm (Figure 1l). Evidently, the above TEM images were
consistent with the analysis of the magnified (110) peak, as
shown in Figure 1n, where the (110) peak becomes narrower
in the sequence of S4-S1, and the sharper peaks for S1 indicate
its good crystallinity and greater grain size than the other three
products (S2-S4) whose precursors grew under the control. On
the basis of the combination analysis of TEM images and XRD
patterns, the as-obtained R-Fe
2
O
3
nanostructures possess diameter-
sized gradients in the sequence from S4 to S1 with increasing
sizes.
As described above, the systems with the addition of inorganic
salts retained the morphology of the FeOOH precursors, and
regular nanopores formed along the nanorods, while the direct
hydrolysis system produced only particles under air conditions.
The phenomenon can be explained by the difference in
thermalstability behavior based on DrTGA, as shown in Figure
2. From DrTGA curves, one can see that there is a broad
exothermal peak with a coexistence of feeble shoulder peaks
in the temperature range of 200-400 °C for each of S2-S4,
showing that their weight loss is much more lagged. The
intensity of the exothermal peak for S1 (Figure 2a) without salt
addition in the temperature range of 200-400 °C is evidently
stronger than the corresponding peaks for the other three samples
with the addition of high-concentration salt ions, indicating that
the weight loss is much quicker. Compared with the direct
hydrolysis systems (S1), the final products of FeOOH in the

O
3
and reminiscent
of the orientation-ordered nanostructures for R-Fe
2
O
3
in air
conditions based on the combined analysis of DrTGA and TEM
results.
Additionally, it is interesting that when the precursors are
prepared in high-concentration Cl
-
ions, the calcined products
have the appearance of porosity (S2 and S4), while in high-
concentration SO
4
2-
, S3 only has the solid appearance. These
results indicate that the existence of large amounts of Cl
-
ions
held in the tunnels in β-FeOOH might favor the appearance of
porosity in the calcined products such as S2 and S4. However,
the high-concentration large anions such as SO
4
2-
, which existed
in the surface sites,
14

lithium batteries, and gas sensors.
3.3. Magnetic Properties for r-Fe
2
O
3
Nanostructures.
Owing to their gradient in the BET surface area and the diameter
size, the magnetic behavior of as-obtained R-Fe
2
O
3
nanostruc-
tures, which is of importance for practical applications, was
investigated for samples S1-S4. Figure 3 shows the curves for
the temperature dependence of zero-field-cooled (ZFC) and
field-cooled (FC) magnetizations from 4 to 300 K, under an
applied field of 500 Oe. The insets are the corresponding
differential ZFC curves.
As for the submicron solid particle sample S1, the FC and
ZFC magnetization curves overlap in the entire concerned
temperature range, as shown in Figure 3a, displaying the
characteristic behavior for R-Fe
2
O
3
with a Morin transition
temperature (T
M
) of 255 K, which is determined by the sharp
peak in the differential ZFC curve (inset in Figure 3a). Normally,

4
system (c), and the FeCl-NH
4
Cl system (d) at
heating rate of 10 °C min
-1
, from which the calcining influence
parameters for the formation of S1, S2, S3, and S4 could be discussed,
respectively. The relation between these system and the final products
of S1, S2, S3, and S4 mentioned in this work can be seen in Table 1.
Synthesis of Hematite (R-Fe
2
O
3
) Nanorods J. Phys. Chem. B, Vol. 110, No. 36, 2006 17809
Figure 3c, the FC and ZFC magnetization curves split signifi-
cantly; the FC magnetization rises significantly, while the ZFC
curve decreases slowly. The split between the FC and ZFC
curves reflects the existence of a large size distribution of
magnetic units resulting from the decrease in effective size,
whose moments block progressively with decreasing tempera-
ture.
16
Additionally, the Morin transition temperature for S3 (235
K) is lower than that for S1 and S2, which may be related to
the decrease in diameters for 1-D nanohematite, agreeing with
the theory that T
M
decreases with decreasing particle size. Since
the temperature of the overlap point is much higher than 100

Nanostructures in a Lithium-Ion Battery. It
is found that the lithium intercalation performance is related to
the intrinsic crystal structure, where the lithium ions can
intercalate into the interlayer, the tunnels, and the holes in the
crystal structure.
18
As for the hematite crystal structure, each
Fe atom is surrounded by six O atoms, whereas each O atom is
bound to four Fe atoms in a typical hematite crystal unit. A
hematite crystal has a rhombohedrally centered hexagonal
structure of the corundum type with a closed-packed lattice in
which two-thirds of the octahedral sites are occupied by Fe
3+
ions (see Supporting Information). As seen from the hematite
structure along [001], [100], and [110], there are no interlayer
spacings and tunnels through the crystal structure (See Sup-
porting Information). Upon careful observation of the hematite
surface structure, the holes could be observed in the first
octahedral layer projected along [001] and [100]; however, the
tunnels could not be seen as the layer number was increased,
as shown in Figure 4. That is to say, holes existed in the surface
hematite crystal, which allowed foreign atoms or molecules to
be introduced, for example, Li
+
ions. When the introduction of
lithium ions to the holes in the hematite surface is concerned,
it gives us the impression that the lithium intercalation perfor-
mance will improve by increasing the surface area or the
porosity of the hematite crystals. Therefore, the synthesis of
hematite nanocrystals with higher surface area or porosity

R-Fe
2
O
3
particles.
20
The S4 electrode exhibited a high discharge
capacity of 1151 mAh/g, corresponding to 6.8 Li per R-Fe
2
O
3
,
while the S3, S2, and S1 electrodes exhibited 1088, 981, and
894 mAh/g, corresponding to 6.5, 5.8, and 5.3 Li per R-Fe
2
O
3
,
respectively. According to the results presented above, it is
evident that the electrochemical properties of the first discharge
capacity possess the sequence of S4 > S3 > S2 > S1, which
is consistent with that of the surface areas for the as-obtained
R-Fe
2
O
3
nanostructures in this case.
3.5. The r-Fe
2
O

concentration. The gas sensitivity, S
g
, is defined as R
air
/R
gas
,
where R
air
and R
gas
are the electrical resistances for sensors in
air and in gas.
22
Although the sensitivity of all the Fe
2
O
3
Figure 4. Schematic hematite structure projected along either [001]
(a) or [100] (b), where holes can be observed in the first octahedral
layer. No tunnels can be found as the layer number is increased.
Figure 5. First charge-discharge curves of hematite (R-Fe
2
O
3
) samples
(S1-S4) at a current density of 0.2 mA cm
-2
.
Figure 6. Room-temperature sensor sensitivity to formaldehyde

example, the superior sensing properties for S4 could be due to
its porous structure associated with the small grain size, which
enables HCHO gas to access more surfaces of the porous-
nanorod structures contained in the sensing unit. Therefore, the
higher surface area for the R-Fe
2
O
3
nanostructure provides more
chances to adsorb and desorb HCHO gas molecules, thus leading
to higher sensitivity at room temperature. This will give us a
guideline to devise the R-Fe
2
O
3
sensors for detecting the
concentration of HCHO gas, which is certainly scientifically
and technically interesting.
4. Conclusions
In summary, we have described in this paper the shape-
controlled synthesis of hematite (R-Fe
2
O
3
) nanostructures with
a gradient in the diameters (from less than 20 nm to larger than
300 nm) and surface areas (from 5.9 to 52.3 m
2
/g) through an
improved synthetic strategy. The benefits of the present work

Supporting Information Available: Crystal structural analy-
sis, synthesis, characterization, and discussion about the forma-
tion mechanism for R- and β-FeOOH, as well as the crystal
structural analysis for R-Fe
2
O
3
. This material is available free
of charge via the Internet at .
References and Notes
(1) Yu, D. B.; Yam, V. W. J. Am. Chem. Soc. 2004, 126, 13200.
(2) (a) Huynh, W.; Peng, X. G.; Alivisatos, A. P. AdV. Mater. 1999,
11, 923. (b) Mattoussi, H.; Radzilowski, L. H.; Dabbousi, B. O.; Thomas,
E. L.; Bawendi, M. G.; Rubner, M. F. J. Appl. Phys. 1998, 83, 7965.
(3) Zysler, R. D. Phys. ReV.B2003, 68, 212408.
(4) Vayssieres, L.; Beermann, N.; Lindquist, S E.; Hagfeldt, A. Chem.
Mater. 2001, 13, 233.
(5) Wang, R. M.; Chen, Y. F.; Fu, Y. Y.; Zhang, H.; Kisielowski, C.
J. Phys. Chem. B 2005, 109, 12245.
(6) Wen, X. G.; Wang, S. H.; Ding, Y.; Wang, Z. L.; Yang, S. H. J.
Phys. Chem. B 2005, 109, 215.
(7) Jia, C. J.; Sun, L. D.; Yan, Z. G.; You, L. P.; Luo, F.; Han, X. D.;
Pang, Y. C.; Zhang, Z.; Yan, C. H. Angew. Chem., Int. Ed. 2005, 44, 4328.
(8) Woo, K.; Lee, H. J.; Ahn, J. P.; Park, Y. S. AdV. Mater. 2003, 15,
1761.
(9) Fu, Y. Y.; Wang, R. M.; Xu, J.; Chen, J.; Yan, Y.; Narlikar, A. V.;
Zhang, H. Chem. Phys. Lett. 2003, 379, 373.
(10) Xiong, Y. J.; Li, Z. Q.; Li, X. X.; Hu, B.; Xie, Y. Inorg. Chem.
2004, 43, 6540.
(11) Amin, N.; Arajs, S. Phys. ReV.B1987, 35, 4810. (b) Nininger, C.,