www.chemsuschem.org

Accepted Article

A Journal of

Title: Microspheres self-assembled by anti-site defect-less nanocrystal-LiFePO4@C/rGO to achieve both high-rate capability and hightap density for Li-ion batteries

Authors: Hongbin Wang, Lijia Liu, Runwei Wang, Xiao Yan, Ziqi Wang,Jiangtao Hu, Haibiao Chen, Shang Jiang, Ling Ni, HailongQiu, Haitong Tang, Yingjin Wei, Zongtao Zhang, Qiu Shilun,and Feng Pan

This manuscript has been accepted after peer review and appears as anAccepted Article online prior to editing, proofing, and formal publicationof the final Version of Record (VoR). This work is currently citable byusing the Digital Object Identifier (DOI) given below. The VoR will bepublished online in Early View as soon as possible and may be differentto this Accepted Article as a result of editing. Readers should obtainthe VoR from the journal website shown below when it is publishedto ensure accuracy of information. The authors are responsible for thecontent of this Accepted Article.

To be cited as: ChemSusChem 10.1002/cssc.201800786

Link to VoR: http://dx.doi.org/10.1002/cssc.201800786

COMMUNICATION

Microspheres self-assembled by anti-site defect-less nanocrystal-

LiFePO4@C/rGO to achieve both high-rate capability and high tap

density for Li-ion batteries

Hongbin Wang, [a,b] Lijia Liu,[a] Runwei Wang,[a] Xiao Yan,[d] Ziqi Wang,[b] Jiangtao Hu,[b] Haibiao Chen,[b]

Shang Jiang,[a] Ling Ni,[a] Hailong Qiu,[c] Haitong Tang,[a] Yingjin Wei,[c] Zongtao Zhang*[a,c], Shilun Qiu[a]

and Feng Pan*[b]

Abstract: LiFePO4@C/rGO hierarchical microspheres with superior

electrochemical activity and high tap density were first synthesized

using a Fe3+-based single inorganic precursor (LiFePO4OH@RF/GO)

obtained from a template-free self-assembly synthesis, following with

direct calcination. The synthesis process requires no physical mixing

step. The phase transformation path way from tavorite LiFePO4OH to

olivine LiFePO4 upon calcination was determined by the in situ high

temperature XRD technique. Benefitted from the unique structure of

the material, these microspheres can be densely packed together,

giving a high tap density of 1.3 g cm-3, and simultaneously, the defect-

less LiFePO4 primary nanocrystals modified with highly conductive

surface carbon layer and ultrathin rGO provide good electronic and

ionic kinetics for fast electron/Li+ ion transport.

Lithium-ion batteries (LIBs), as one important technology for

electric energy storage, have revolutionarily extend the battery life

of portable electronic devices (e.g. smart phones, wearable

devices, and laptop computers), and now stand as a promising

candidate for high-power and long-life battery applications,

typically for the electric transportation off the grid and the clean

energy back-up and storage systems. [1-3] To date, olivine (e.g.

LiFePO4), layer-type (e.g. LiNi1-x-yMnxCoyO2), and spinel (e.g.

LiMn2O4) electrode materials are the three mostly used cathodes

for high power LIBs. [1, 2, 4] Among them, LiFePO4 is more stable

in thermodynamics and reaction kinetics than the other two owing

to the presence of a robust polyanionic framework. [5-7] Moreover,

LiFePO4 prevails in terms of natural abundance, cost, and

environmental friendliness. The major issue to this material is the

negative transport kinetics in both Li+ ion diffusion and electron

delivery. Size tailoring, in conjunction with surface carbon (i.e.

graphitic carbon, graphene, et al.) coatings to form LiFePO4/C

nanocomposite is acknowledged as a technology of choice to

alleviate this issue effectively. [6-9] However, when the particle size

of LiFePO4/C is decreased down to nanoscale, thermodynamic

instability and high risk of side reactions with the organic

electrolyte become into new hazards needed to be faced with. [10]

In addition, the processability and tap density (less than 1.0 g cm-

3) of the powders also require improvements when it comes to a

practical point of view. [11] For that, small LiFePO4/C nanoparticles

assembled into hierarchical architectures, ideally with a

microsphere morphology, would naturally become more easily

free-flowing and can be densely packed together, giving a higher

tap density and hence a higher volumetric energy density. [12-14]

Hydrothermal and solvothermal synthesis technologies or a

combination of them are the methods commonly used to create

LiFePO4/C microsphere with hierarchical architectures, the vast

majority of which, however, required the use of instable and costly

ferrous iron (Fe2+) salts as the raw material, and simultaneously a

reducing agent and/or inert gas to avoid the oxidation of Fe2+ ions

during reactions. [13, 15] On top of that, most attempts to develop

new routes, especially through ferric iron (Fe3+) based

approaches, focused either on the “one-pot synthesis” or “in-situ

carbon coating”, rather than considering both two concepts

together. [12-14] Thus exploring a simpler synthesis scheme to

produce LiFePO4/C microspheres with ideal structural features

will be of great interest.

In this communication, reduced GO-modified LiFePO4/C (i.e.

LiFePO4@C/rGO) hierarchical microspheres were synthesized

for the first time by a one-pot mixed-solvothermal process to form

a ferric three-component precursor of LiFePO4OH@RF/GO,

followed by direct calcination at high temperature, during which

tedious physical mixing and grinding step can be totally avoided.

The final LiFePO4@C/rGO microspheres created can be densely

packed together, giving a high tap density of 1.3 g cm-3, and

simultaneously, the small primary LiFePO4 nanoparticles (65 nm),

in conjunction with the integrated carbonous conductive network

can provide an adequate electrochemically available surface for

enhancing the high-rate capability. The phase transformation

mechanism from tavorite LiFePO4OH to the olivine LiFePO4 upon

calcination was also determined by the in situ high temperature

X-ray diffraction (XRD).

[a] Dr. H. Wang, Dr. L. Liu, Prof. R. Wang, Dr. S. Jiang, L. Ni, H. Tang,

Prof. Z. Zhang, Prof. S. Qiu

State Key Laboratory of Inorganic Synthesis and Preparative

Chemistry, International Joint Research Laboratory of Nano-Micro

Architecture Chemistry

Jilin University

Changchun 130012, P. R. China

E-mail: zzhang@jlu.edu.cn

[b] Dr. H. Wang, Dr. Z. Wang, J. Hu, Dr. H. Chen, Prof. F. Pan

School of Advanced Materials

Peking University Shenzhen Graduate School

Shenzhen 518055, P. R. China

E-mail: panfeng@pkusz.edu.cn

[c] Dr. H. Qiu, Prof. Y. Wei, Prof. Z. Zhang

Key Laboratory of Physics and Technology for Advanced Batteries

(Ministry of Education)

Jilin University

Changchun 130012, P. R. China

[d] Dr. X. Yan

School of Chemistry and Materials Science

Jiangsu Normal University

Xuzhou 221116, P. R. China

Supporting information for this article is given via a link at the end of

the document.

10.1002/cssc.201800786

Acc

epte

d M

anus

crip

t

ChemSusChem

This article is protected by copyright. All rights reserved.

COMMUNICATION

Figure 1. Schematic illustration for the synthesis process of

LiFePO4@C/rGO composite microspheres by using a three-

component precursor (LiFePO4OH@RF/GO) achieved from a

template-free self-assembly approach (Step 1), followed by direct

calcination at high temperature (Step 2).

As schematically illustrated by Figure 1, the synthesis started

with a one-step mixed-solvothermal process that proceeded at a

designated temperature of 180 °C under mechanical stirring. The

process involved the addition of a methanol solution containing

formaldehyde to an aqueous solution containing Li+, Fe3+, PO43-,

resorcinol and GO. Detail structure information and morphology

of the GO are shown in Figure S1-S4. Here, the methanol acted

as a chelating agent to disrupt the screening of charges by the

water due to its relatively lower polarity. In other words, the

methanol with a low dielectric constant can promote the

electrostatic attraction between the negative phosphate groups

and the positive Li+ and Fe3+ ions in the solution to establish stable

ionic bonds. [16] Along with the crystallization of LiFePO4OH

nanocrystals at an elevated temperature, the resorcinol added

reacted with formaldehyde simultaneously to form the resorcinol-

formaldehyde (RF) resin by a catalytic polymerization process.

The RF resin coated in situ on the surface of LiFePO4OH

nanoparticles, separating them from each other to avoid

continuous crystal growth, and linking them together via the

polymer network to assemble the hierarchical microspheres. As

thus, contact-induced growth of formed LiFePO4OH nanoparticles

was regulated by the surface RF resin coverage during reaction.

Detailed polymerization mechanism between resorcinol and

formaldehyde can be described by Figure S5. [17, 18] Moreover, the

GO nanosheets with abundant hydroxyl and/or carboxyl groups

anchored onto the LiFePO4OH@RF composite through an

electrostatic interaction and/or host-guest binding behavior,

forming a three-component composite. [19] Note that the RF resin

is more chemically stable under such a high temperature than

other organics such as sucrose, glucose, et al., which remained

highly polyfunctional during the whole reaction and was capable

of maintaining the strong chemical interactions with both

LiFePO4OH nanoparticles and GO nanosheets. Heat treatment in

Ar/H2 atmosphere led to the conversion of LiFePO4OH@RF/GO

into the final LiFePO4@C/rGO composite microspheres. During

the calcination process, the isolated nanoparticles of single

inorganic tavorite LiFePO4OH crystallized into olivine LiFePO4

nanocrystals, and meanwhile the surface RF resin and GO were

converted into a highly conductive carbon shells network and

large rGO nanosheets, respectively, and in situ restricted the

growth of primary LiFePO4 nanocrystals and the secondary

microspheres.

As suggested by the XRD pattern of LiFePO4OH@RF/GO

(Figure 2A), all the Bragg reflections can be well indexed to the

standard triclinic structure of tavorite LiFePO4OH (JCPDS No. 41-

1376) without other discernible phases. The broad diffracting

peaks of the XRD pattern presage a small particle size of the

material. On the basis of Scherrer’s formula (d=0.89 λ/β cosθ), the

mean size of primary LiFePO4OH crystals can be estimated to be

~30 nm. A lack of GO and RF polymer characteristic signals can

be explained by the lower content and the disordered state of the

materials, respectively. The successful loading of RF resin on the

LiFePO4OH crystals was synthetically verified by the Fourier

transform infrared spectroscopy (FTIR), HRTEM and TEM-energy

dispersive spectroscopy(TEM-DES) as illustrated in Figure S6-S8.

A yellowish-brown appearance of the final LiFePO4OH@RF/GO

composite material suggests a slight pyrolysis of the RF polymer

contained. Typical SEM images, given in Figure 2C, reveal that

the LiFePO4OH@RF/GO material shows a regular microsphere

morphology with a mean particle size of 2.2 μm in diameter

(Figure S9), on which a large number of ultrathin GO nanosheets

with lateral size up to several micrometers can be clearly

observed. At a higher magnification (inset in Figure 2C), one can

notice the microspheres are composed of densely aggregated

nanoparticles with tens of nanometers in size, in good agreement

with the above XRD analysis, confirming the hierarchical

architecture characteristic of the LiFePO4OH@RF/GO composite

microspheres. It is believed that the RF polymer formed in situ on

the surface of the LiFePO4OH nanocrystals has restricted their

grain growth effectively during the solvothermal process. Besides,

Figure 2. (A) XRD powder patterns of LiFePO4OH@RF/GO and

LiFePO4@C/rGO composite microspheres. (B) In situ high

temperature XRD analysis of LiFePO4OH@RF/GO microspheres

under an inert atmosphere at controlled temperatures (25-700 °C,

heating rate of 5 °C min-1) with 15 minutes for each collection.

Typical SEM images of (C) LiFePO4OH@RF/GO and (D)

LiFePO4@C/rGO microspheres. Insets show corresponding HR-

SEM images, and the scale bars are both 500 nm.

PO43-

Solvothermal reaction

LiFePO4OH

GO rGO

Resorcinol

Formaldehyde

RF resinLi+

Fe3+ Carbon

11 12

LiFePO4OH@RF/GO LiFePO4@C/rGO

RF resin Olivine LiFePO4

Li

FeO6

PO4

Tavorite LiFePO4OH

Fe(1)O6 Fe(2)O6

H

PO4

Li

+

GO

b

ac

b

ac

Self-assembly Calcination

(180 oC, stirring)

Inte

nsity (

a.u

.)

2 Theta (degree)

15 20 25 30 35 40 45

LiFePO4@C/rGO

LiFePO4OH@RF/GO

PDF#81-1173

PDF#41-1376

15 20 25 30 35 40 45

100

200

300

400

500

600

700

800

Tem

pera

ture

(oC

)

2 Theta (degree)

0.000100.0200.0300.0400.0500.0600.0700.0800.0900.0100011001200

C

3 μm

B

3 μm

A

D

10.1002/cssc.201800786

Acc

epte

d M

anus

crip

t

ChemSusChem

This article is protected by copyright. All rights reserved.

COMMUNICATION

GO nanosheets with highly active negative groups (hydroxyl and

carboxyl) chelated with positive Li+ and Fe3+ ions strongly, capable

of depressing the secondary particles growth and also making the

composite microspheres uniform in dimension.

Thermal heat treatment in an Ar/H2 (6 %) atmosphere led to

the conversion of LiFePO4OH@RF/GO into LiFePO4@C/rGO

composite, as suggested by the XRD pattern (Figure 2A). The

entire diffraction peaks of LiFePO4@C/rGO material match well

with the standard orthorhombic structure of LiFePO4 (JCPDS No.

81-1173). X-ray photoelectron spectroscopy (XPS) spectra were

also analyzed to check for the oxidation state of iron species in

the surface-near region (depth of <5 nm), as shown in Figure S10,

which confirmed the iron species in LiFePO4@C/rGO composite

are all at a Fe (II) oxidation state. [20] The mean particle size of

LiFePO4 nanocrystals was estimated to be 65 nm by Scherrer’s

formula. Given the fact that the LiFePO4 and its precursor

(LiFePO4OH) were comparable in particle dimension, the surface

residual carbon derived from the RF polymer in situ coated should

have effectively restricted the particle growth of LiFePO4 crystals

during heat treatment. Morphology details of LiFePO4@C/rGO

were illustrated by the SEM images (Figure 2D and Figure S11),

from which we can observe the material remained the hierarchical

microsphere morphology as LiFePO4OH@RF/GO. It is interesting

to note that part of the rGO nanosheet seems to embed into the

interior of the microsphere, while the rest wraps around the

exterior surface. Integrating large rGO nanosheets in such a

manner can help to establish a continuous and robust conductive

network, which extended from the interior to the outside of the

LiFePO4/C microspheres and even between the microspherical

particles, thus alleviating the polarization of electrode. N2

adsorption/desorption isotherm of LiFePO4@C/rGO composite

microspheres (Figure 3A) appears to be a type IV curve with a

large type H2 hysteresis loop, typical characteristic of N2

molecular adsorption in mesoporous materials. Calculated from

the adsorption isotherm, the Brunauer-Emmett-Teller (BET)

surface area is determined to be 30.5 m2 g−1. Such high internal

surface area and mesoporous structure would, we believe, not

only allow a better penetration of electrolyte to facilitate the Li+ ion

diffusion but also a better electrons transport in the interior of

LiFePO4@C/rGO microspheres, ultimately enhancing the rate

capability for lithium ion insertion/deinsertion reaction. SEM-

energy dispersive spectroscopy (SEM-EDS) mapping images of

an individual LiFePO4@C/rGO composite microsphere (shown in

Figure S12) indicate a uniform distribution of Fe, P, O and C

elements in the spherical scanning region. The sheet-like feature

in the mapping just peculiar to the C element at the edges of the

microspheres was attributable to the surface rGO nanosheets,

indicating a complete reduction of GO upon heat treatment.

Further investigation by TEM image, shown in Figure S13,

indicates some typical discrete nanoparticles of LiFePO4@C/rGO

microspheres obtained by grinding centre around 65 nm in size,

in good agreement with the above XRD result. High resolution

TEM (HRTEM) image (Figure S14) reveals the visible lattice

fringes correspond to a spacing of 0.25 nm (marked by a pair of

parallel red lines in the crystalline region), which matches well with

the expected d spacing of the (131) plane of olivine-type LiFePO4,

indicating the observed nanoparticles is single-phase LiFePO4

and highly crystalline. The LiFePO4 nanocrystal was coated by a

layer of amorphous carbon with a typical thickness of ~2.0 nm.

Generally speaking, the kinetic barrier for Li+ ions to diffuse

through such a thin carbon layer can be totally neglected in

contrast to other diffusion kinetic limitations upon cycling. [21, 22]

The mass fraction of carbon contained was determined by the

CHN elemental analysis to be ~6 wt. %. Detailed structure

information on the surface residual carbon was further verified by

the Raman spectroscopy (Figure S15). [12, 21, 23] Typically, an

increase in the peak intensity ratio (IG/ID) in Raman is an indicator

of a greater degree of ordering for graphitic materials. [12, 21, 23]

According to the fitting result, the IG/ID value of LiFePO4@C/rGO

was estimated to be 0.82, which is similar to or higher than those

of most LiFePO4/C modified with other types of semi-graphitized

carbon, [12, 23] implying the residual carbon pyrolyzed from the RF

polymer and GO is more graphitized, capable of providing a better

electric transfer around LiFePO4 nanoparticles. Moreover, a lack

of LiFePO4 Raman response in the range of 600-1100 cm-1

illustrates that the carbon coating on the LiFePO4 nanoparticles is

fairly complete. [12, 21] In view of the one-dimensional (1D) Li+

diffusion pathway in the LiFePO4 framework, a thorough surface

modification with a layer of highly graphitized carbon can enable

the electrons to reach all the positions where Li+ intercalation

takes place, leading to a higher rate capability. [21, 24]

A Fe·Li anti-site defect, namely Fe2+ ion occupying the Li+ site,

is commonly observed in LiFePO4 crystals, and it has been

proven unfavorable due to the blocking of the Li+ ions diffusion

pathway. [21, 25, 26] The content of Fe·Li defects in LiFePO4 can be

easily identified by FTIR spectroscopy. Theoretically, the

absorption peak concerning to the P-O symmetric stretching

vibration of PO4 tetrahedron in Fe·Li defect-free LiFePO4 should

emerge at 957 cm-1, and it shifts towards a higher wavenumber

with the number of Fe·Li anti-site defects increasing. [21, 25, 26] For

LiFePO4@C/rGO, the characteristic peak was measured to

appear at 960 cm-1 (shown in Figure 3B), that is, close to the

theoretical value of Fe·Li defect-free LiFePO4, illustrating a very

small number of Fe·Li defects in the LiFePO4@C/rGO material. To

identify the cause of the extremely low defect population, the

phase transformation mechanism of LiFePO4OH to LiFePO4 was

further studied. Figure 2B lists the in situ high temperature XRD

data showing a detailed evolution from LiFePO4OH@RF/GO to

LiFePO4@C/rGO during temperature ramping. Clearly, the

triclinic LiFePO4OH degraded starting at 400 °C and lasted until

about 500 °C with the appearance of a new crystalline phase

Figure 3. (A) Typical N2 absorption-desorption isotherms of the

LiFePO4@C/rGO microspheres with a large type H2 hysteresis

loop, typical characteristic of mesopore structure. (B) Typical FTIR

spectrum of the LiFePO4@C/rGO microspheres recorded in the

wavenumber region of 400–1200 cm−1.

400 600 800 1000 1200

1095 1

139

1070

1046

960

645

635

577

5495

00

Tra

nsm

itta

nce

Wavenumber (cm-1)

471

0.0 0.2 0.4 0.6 0.8 1.00

5

10

15

20

25

30

35

40

45

for mesopore structure

Type H2 hysteresis loop

Adsorption isotherm

Desorption isotherm

Volu

me a

dsorb

ed (

cm

3 g

-1)

Relative pressure (P/P0)

A B

10.1002/cssc.201800786

Acc

epte

d M

anus

crip

t

ChemSusChem

This article is protected by copyright. All rights reserved.

COMMUNICATION

associated with single orthorhombic LiFePO4. Thereafter, the

degree of crystallization of the LiFePO4 increased gradually as the

temperature increased. It thus follows that the phase

transformation pathway upon heat treatment from LiFePO4OH to

LiFePO4 obeyed a local crystal to crystal phase transformation

model, which is believed better for depressing the formation of

Fe·Li anti-site defects in LiFePO4 by avoiding the complex

crystallographic phase transition and long-range solid-solid mass

transfer process upon crystallization.

Remarkable rate capability and cycle stability are desirable

features for high power LIBs. Galvanostatic charge/discharge

performance of LiFePO4@C/rGO microspheres are shown in

Figure 4A. Here, 1 C rate corresponds to a current density of 170

mA g-1. Notably, at a low rate of 0.2 C, the material discharged to

a capacity of 157.2 mAh g-1, that is, close to its estimated

theoretical value (159.8 mAh g-1) calculated by considering a well-

defined mixture of active LiFePO4 and the surface residual carbon

(~6 wt. %). The material still delivered reversible capacities of

146.7, 139.8, 127.5, and 112.4 mAh g-1 at intermediate rates of 1

C, 2 C, 5 C, and 10 C, respectively, and remained relatively flat

voltage plateaux at all rates. A flatter potential plateau during the

charge/discharge process for LiFePO4@C/rGO indicates a less

pseudocapacitive contribution from surface/interfacial lithium

storage. [26] Even at high rates of 30 C and 60 C tested,

LiFePO4@C/rGO can still release stable discharge capacities of

86.5 and 60.8 mAh g-1. After the incremental rate test up to 60 C

rate, the reversible capacities almost restored to the initial values

of 109.3 and 157 mAh g-1 successively when the discharging rate

was changed back to 10 C and 0.2 C, as shown in Figure 4B,

illustrating the capacity decay at high rates is majorly due to the

electrode polarization associated with diffusion-limited transfer of

Li+ ion across the two-phase interface, rather than the structure

instability of LiFePO4@C/rGO material. [21] From a topological

point of view, the superior electrochemical performance of

Figure 4. (A) Charge and discharge curves of LiFePO4@C/rGO

measured by a galvanostatic testing model with rates from 0.2 C

up to 60 C. (B) Rate-varying cycle stability of LiFePO4@C/rGO

evaluated by raising the rate continuously from 0.2 C up to 60 C,

and then back to 10 C and 0.2 C with ten cycles for each rate. (C)

Long-term cycling stability of LiFePO4@C/rGO measured by

repeatedly charging and discharging the cell at 2 C rate for 300

cycles, and then at 10 C rate for 700 cycles.

LiFePO4@C/rGO can be derived from the unique hierarchical

composite structure of LiFePO4OH@RF/GO precursor, which

was formed from fully sealing RF resin polymer “shells” with a

spherical 3D-connected network and single-source inorganic

LiFePO4OH “cores” where the ideal elements of Li, Fe, P, and O

are evenly distributed at the atomic level, thus capable of

minimizing the formation of undesired impurities and Fe·Li lattice

defects for the LiFePO4 obtained owing to the absence of complex

crystallographic phase transition and long-range solid-solid mass

transfer process upon crystallization. [21, 27] In addition, the

ultrathin rGO nanosheets anchored on the surface can establish

a continuous and robust conductive network for inter-particle

electrical connection to lower the soft contact impedance. [19, 28]

To prove this, we synthesized LiFePO4@C in an identical process

except the addition of GO during the hydrothermal synthesis for

comparison (Figure S16), whose rate performance was inferior to

that of LiFePO4@C/rGO (Figure S17). The good electronic and

ionic kinetics of LiFePO4@C/rGO were further supported by cyclic

voltammetry (CV) and electrochemical impedance spectroscopy

(EIS) analyses, as shown in Figure S18 and S19. The material

was found to exhibit higher rate capability than those of the

LiFePO4/C microspherical samples reported in the references

(shown in Figure S20). [12-15, 29-31] Although the performance of

LiFePO4@C/rGO at higher rates (>10 C) was slightly lower than

that of the nanosized LiFePO4/C reported by the previous studies, [21, 24] the LiFePO4@C/rGO material achieved a tap density of 1.3

g cm−3, which is almost 30 % higher than that currently reported

for carbon-coated irregular LiFePO4 nanomaterials (≤1.0 g cm−3),

thus better for volumetric energy density. Some microsized

LiFePO4/C [32, 33] had been reported to show a higher tap density

(1.6 or even 1.9 g cm−3), but their rate capabilities are markedly

lower than LiFePO4@C/rGO due to the slow Li+ ion transport in

the bulk (imposed by the larger primary particles). Therefore, it

can be regarded the LiFePO4@C/rGO composite microspheres

provide a good compromise between tap density and rate

capacity, giving a relatively high tap density and simultaneously a

high rate capability.

The long-term cycling stability of obtained LiFePO4@C/rGO

microspheres was evaluated by continuous charge/discharge at

2 C rate for 300 cycles, and then at a high rate of 10 C for 700

cycles, as shown in Figure 4C. The capacity retention of the

material is superior. Almost no capacity fading can be observed

after cycling at 2 C for 300 cycles. When cycled at 10 C for another

700 cycles, the final capacity retention was higher than 95 %,

equivalent to less than 0.007 % capacity loss per cycle.

Corresponding coulombic efficiency as shown in the inset of

Figure 4C stayed at about 100 %. In addition to the contributions

of the stable LiFePO4 nanocrystals, the robust spherical 3D

conductive carbon network along with large rGO nanosheets

formed in situ on the surface of LiFePO4 particles are also

supposed to play a critical role in the good cycling reversibility by

relaxing the strain energy of the entire electrode upon Li+ ion

insertion/extraction.

Conclusions

In summary, we have developed a unique method to create

hierarchical LiFePO4@C/rGO composite microspheres through a

template-free self-assembly approach to form a Fe3+-based

single-source inorganic precursor-LiFePO4OH@RF/GO, followed

0 200 400 600 800 10000

30

60

90

120

150

180

210

240

20

40

60

80

100

120

Charge capacity Discharge capacityCa

pa

city (m

Ah

g-1

)

Coulombic efficiency

10 C

2 C

Cycle number (n)

Co

ulo

mb

ic e

ffic

ien

cy (

%)

0 10 20 30 40 50 60 70 80 900

30

60

90

120

150

180

210

0.2 C

10 C

60 C

30 C

10 C5 C

2 C1 C

0.2 C

Capacity (

mA

h g

-1)

Cycle number (n)

0 20 40 60 80 100 120 140 1602.1

2.4

2.7

3.0

3.3

3.6

3.9

4.2

5 C, 1 C,60 C,30 C,10 C, 2 C, 0.2 C

Voltage (

V v

s L

i/Li+

)

Capacity (mAh g-1

)

B

C

A

10.1002/cssc.201800786

Acc

epte

d M

anus

crip

t

ChemSusChem

This article is protected by copyright. All rights reserved.

COMMUNICATION

by direct calcination. Benefitted from the unique hierarchical

architecture of the material, these microspheres can be densely

packed together, giving a high tap density of 1.3 g cm-3, and

simultaneously, the defect-less LiFePO4 primary nanocrystals

modified with highly conductive surface carbon layer and ultrathin

rGO provide good electronic and ionic kinetics, thus achieving

superior electrochemical performance including good rate

capability and cycling stability. The synthesis route we reported

also has the advantages in simplicity, low cost, and environmental

friendliness for scale-up and commercialization. We hope this

work would enlighten a brand-new way to the mass production of

microspherical LiFePO4/C cathodes for high power LIBs.

Experimental Section

Materials synthesis. The graphene oxide (GO) was produced from

natural graphite flakes (Sigma-Aldrich, <20 μm) by a modified Hummers

method (Y. Xu, H. Bai, G. Lu, et al., J. Am. Chem. Soc., 2008, 130, 5856–

5857; X. Yan, Y. Li, F. Du, et al., Nanoscale, 2014, 6, 4108-4116). GO-

modified LiFePO4OH/RF (named as LiFePO4OH@RF/GO) hierarchical

microspheres were synthesized by a one-pot mixed solvothermal method.

Typically, 0.01 mol H3PO4 (85 wt. %), 0.02 mol lithium hydroxide

monohydrate, 0.01 mol Fe(NO3)3·9H2O and 0.005 mol resorcinol were

dissolved in 10 g of graphene oxide suspension (~2 mg mL-1). Afterwards,

10 g of methanol solution containing 0.01 mol formaldehyde was added.

The mixture solution was then transferred into an autoclave and heated at

180 °C for 6 h under agitating conditions (v=600 r min-1). After being cooled

down to the room temperature by water quench, the brown precipitate was

collected by filtration, washed several times with deionized water, and

dried for 6 h at 70 °C. rGO-modified LiFePO4/C (LiFePO4@C/rGO)

hierarchical composite microspheres were obtained by a subsequent

calcination of the LiFePO4OH@RF/GO precursor at 700 °C for 6 h under

an Ar/H2 (6 %) atmosphere. The synthesis of LiFePO4OH@RF was similar

to that of LiFePO4OH@RF/GO except the addition of GO. LiFePO4@C was

obtained by calcining LiFePO4OH@RF under the condition identical to the

preparation of LiFePO4@C/rGO.

Materials characterizations. X-ray diffraction (XRD) patterns were

recorded on a Rigaku D/MAX-2550 diffractometer at 50 kV and 200 mA

with Cu Ka radiation (λ=1.5418 Å). Fourier transform infrared spectrometer

(FTIR) spectra were tested by a Bruker IFS-66V/S spectrometer. X-ray

photoelectron spectroscope (XPS) tests were performed on an ESCA LAB

MARK Ⅱ spectrometer. Morphology details of the materials were

examined by JEOL JSM-6700F scanning electron microscope (SEM) and

JEOL JSM-3010 transmission electron microscopy (TEM). Nitrogen

absorption-desorption isotherms were evaluated on a Micromeritics ASAP

2020M instrument. The specific surface areas were calculated by

employing the Brunauer-Emmett-Teller (BET) equation. CHN analysis was

carried out by an Elementer Vario MICRO cube analyzer. Raman spectrum

was collected by a Renishaw inVia Raman microscope with 514.5 nm laser

radiation and fitted using a Gaussian-Lorentzian model. Atomic force

microscopy (AFM) images were obtained using a Bruker Dimension Icon

scanning probe microscope. The tap density was measured from the

method reported by Qian et. al. (J. Phys. Chem. C 2010, 114, 3477–3482):

a certain mass of the product was placed in a small measuring cylinder

and tapped on for at least 5 min by hand. The measured volume of the

tapped powder and its mass were used to calculate the tap density.

Electrochemistry tests. For the electrochemical test, 80 wt. % of active

material, 10 wt. % of acetylene black (AB) and 10 wt. % of polyvinylidene

fluoride (PVDF) were firstly mixed evenly with N-methyl-2-pyrrolidinone

(NMP). The blended slurry formed was then cast on an aluminum foil

current collector. After being dried thoroughly in vacuum at 110 oC, the

electrode sheet was tailored into disks of Φ10 mm and assembled in

CR2032 coin-type cells with lithium metal as counter electrode under an

Ar-filled glovebox. The loading mass of active material in the electrode is

~2 mg (area of 0.785 cm2). A mixed solution of ethylene carbonate (EC),

ethylmethyl carbonate (EMC) and dimethyl carbonate (DMC) (1:1:1 by

volume) containing 1 M LiPF6 was used as electrolyte. Celgard 2400

microporous membrane served as the separator. Galvanostatic

charge/discharge measurements were carried out in the voltage window

of 2.5-4.1 V (vs. Li/Li+) using a battery test system (CT2001A, Land). Cyclic

voltammetry (CV) and electrochemical impedance spectroscopy (EIS)

were measured on a Bio-Logic VSP electrochemical workstation. CV

curves were recorded with increasing scanning rates from 0.2 to 5 mV s-1

(2.5-4.1 V). EIS was collected at open circuit voltage (OCV) after three

cycles of activation in the frequency range of 200 KHz-10 mHz (amplitude

of 5 mV). All the electrochemical tests were performed at ambient

temperature (~25 °C).

Acknowledgements

The research was financially supported by the National Natural

Science Foundation of China (Nos. 21261130584, 91022030,

21771081), the Chang Bai Mountain Scholars Program (No.

440020031182), Provincial major project (No. 21390394),

National Materials Genome Project (No. 2016YFB0700600),

National Project for EV Batteries (No. 20121110, OptimumNano,

Shenzhen), Shenzhen Science and Technology Research Grant

(Nos. ZDSY20130331145131323, JCYJ20140903101633318,

JCYJ20140903101617271), Guangdong Innovation Team

Project (No. 2013N080) and the China Postdoctoral Science

Foundation (Nos. 2017M620497, 2017M620520).

Keywords: LiFePO4@C/rGO • hierarchical microspheres • anti-

site defects • LiFePO4OH • lithium ion batteries

[1] J. Lu, Z. Chen, Z. Ma, F. Pan, L. A. Curtiss and K. Amine, Nat. Nanotech.

2016, 11, 1031-1038.

[2] A. Manthiram, B. Song and W. Li, Energy Storage Mater. 2017, 6, 125-

139.

[3] D. Larcher and J. M. Tarascon, Nat. Chem. 2015, 7, 19-29.

[4] Z. Wang, R. Tan, H. Wang, L. Yang, J. Hu, H. Chen and F. Pan, Adv.

Mater. 2017, 1704436.

[5] J. Hu, Y. Jiang, S. Cui, Y. Duan, T. Liu, H. Guo, L. Lin, Y. Lin, J. Zheng,

K. Amine and F. Pan, Adv. Energy Mater. 2016, 6, 1600856.

[6] A. K. Padhi, J. B. Goodenough and K. S. Nanjundaswamy, J.

Electrochem. Soc. 1997, 144, 1188-1194.

[7] N. Nitta, F. Wu, J. T. Lee and G. Yushin, Mater. Today 2015, 18, 252-

264.

[8] J. Hu, Y. Xiao, H. Tang, H. Wang, Z. Wang, C. Liu, H. Zeng, Q. Huang,

Y. Ren, C. Wang, W. Zhang and F. Pan, Nano Lett. 2017, 17, 4934-4940.

[9] P. S. Herle, B. Ellis, N. Coombs and L. F. Nazar, Nat. Mater. 2004, 3,

147-152.

[10] A. S. ARICÒ, P. BRUCE, B. SCROSATI, J. TARASCON and W. V.

SCHALKWIJK, Nat. Mater. 2005, 4, 366-377.

[11] Y.-G. Guo, J.-S. Hu and L.-J. Wan, Adv. Mater. 2008, 20, 2878-2887.

[12] J. Qian, M. Zhou, Y. Cao, X. Ai and H. Yang, J. Phys. Chem. C 2010,

114, 3477-3482.

[13] C. Sun, S. Rajasekhara, J. B. Goodenough and F. Zhou, J. Am. Chem.

Soc. 2011, 133, 2132-2135.

[14] S. W. Oh, S. T. Myung, S. M. Oh, K. H. Oh, K. Amine, B. Scrosati and Y.

K. Sun, Adv. Mater. 2010, 22, 4842-4845.

10.1002/cssc.201800786

Acc

epte

d M

anus

crip

t

ChemSusChem

This article is protected by copyright. All rights reserved.

COMMUNICATION

[15] J. Su, X.-L. Wu, C.-P. Yang, J.-S. Lee, J. Kim and Y.-G. Guo, J. Phys.

Chem. C 2012, 116, 5019-5024.

[16] M.-Y. Cho, Y. S. Lim, S.-M. Park, K.-B. Kim and K. C. Roh,

CrystEngComm 2015, 17, 6149-6154.

[17] C. Liang, K. Hong, G. A. Guiochon, J. W. Mays and S. Dai, Angew. Chem.

Int. Ed. 2004, 43, 5785-5789.

[18] R. W. Pekala, J. Mater. Sci. 1989, 24, 3221-3227.

[19] X. Yan, Y. Li, F. Du, K. Zhu, Y. Zhang, A. Su, G. Chen and Y. Wei,

Nanoscale 2014, 6, 4108-4116.

[20] R. Dedryvère, M. Maccario, L. Croguennec, F. L. Cras, C. Delmas and

D. Gonbeau, Chem. Mater. 2008, 20, 7164-7170.

[21] H. Wang, R. Wang, L. Liu, S. Jiang, L. Ni, X. Bie, X. Yang, J. Hu, Z. Wang,

H. Chen, L. Zhu, D. Zhang, Y. Wei, Z. Zhang, S. Qiu and F. Pan, Nano

Energy 2017, 39, 346-354.

[22] R. Dominko, M. Bele, M. Gaberscek, M. Remskar, D. Hanzel, S. Pejovnik

and J. Jamnik, J. Electrochem. Soc. 2005, 152, A607-A610.

[23] J. Wang and X. Sun, Energy Environ. Sci. 2012, 5, 5163-5185.

[24] Y. Wang, Y. Wang, E. Hosono, K. Wang and H. Zhou, Angew. Chem. Int.

Ed. 2008, 47, 7461-7465.

[25] X. Qin, J. Wang, J. Xie, F. Li, L. Wen and X. Wang, Phys. Chem. Chem.

Phys. 2012, 14, 2669-2677.

[26] Y. Zhao, L. Peng, B. Liu and G. Yu, Nano Lett. 2014, 14, 2849-2853.

[27] Y. Jiang, H. Zhuang, D. Pan, Z. Jiao, X. Que, X. Ling, M. Zhong, Y. Chu

and B. Zhao, J. Appl. Electrochem. 2013, 43, 611-617.

[28] J. Zheng, J. Lu, K. Amine and F. Pan, Nano Energy 2017, 33, 497-507.

[29] Y. Zhou, J. Lu, C. Deng, H. Zhu, G. Z. Chen, S. Zhang and X. Tian, J.

Mater. Chem. A 2016, 4, 12065–12072.

[30] D. Xu, Y.-B. He, X. Chu, Z. Ding, B. Li, J. He, H. Du, X. Qin, F. Kang,

ChemSusChem 2015, 8, 1009–1016.

[31] W. Li, J. Hwanga, W. Chang, H. Setiadic, K. Y. Chung, J. Kim, J.

Supercrit. Fluid. 2016, 116, 164–171.

[32] H. M. Xie, R. S. Wang, J. R. Ying, L. Y. Zhang, A. F. Jalbout, H. Y. Yu,

G. L. Yang, X. M. Pan and Z. M. Su, Adv. Mater. 2006, 18, 2609-2613.

[33] R. Dominko, M. Bele, J.-M. Goupil, M. Gaberscek, D. Hanzel, I. Arcon, J.

Jamnik, Chem. Mater. 2007, 19, 2960-2969.

10.1002/cssc.201800786

Acc

epte

d M

anus

crip

t

ChemSusChem

This article is protected by copyright. All rights reserved.

COMMUNICATION

Entry for the Table of Contents

COMMUNICATION

Hongbin Wang, Lijia Liu, Runwei Wang,

Xiao Yan, Ziqi Wang, Jiangtao Hu,

Haibiao Chen, Shang Jiang, Ling Ni,

Hailong Qiu, Haitong Tang, Yingjin Wei,

Zongtao Zhang,* Shilun Qiu, and Feng

Pan*

Page No. – Page No.

Microspheres self-assembled by anti-

site defect-less nanocrystal-

LiFePO4@C/rGO to achieve both high-

rate capability and high tap density

for Li-ion batteries

The major merit of this work is to determine a unique method to synthesize hierarchical LiFePO4@C/rGO composite microspheres with

enhanced electrochemical activity and high tap density by using ferric iron-based single inorganic precursor (LiFePO4OH@RF/GO)

achieved by a template-free self-assemble synthesis, followed by direct calcination. The preparation process requires no physical

mixing step and is proven effective in depressing the Fe·Li anti-site defects in LiFePO4 crystals.

PO43-

Solvothermal reaction

LiFePO4OH

GO rGO

Resorcinol

Formaldehyde

RF resinLi+

Fe3+ Carbon

11 12

LiFePO4OH@RF/GO LiFePO4@C/rGO

RF resin Olivine LiFePO4

Li

FeO6

PO4

Tavorite LiFePO4OH

Fe(1)O6 Fe(2)O6

H

PO4

Li

+

GO

b

ac

b

ac

Self-assembly Calcination

(180 oC, stirring)

10.1002/cssc.201800786

Acc

epte

d M

anus

crip

t

ChemSusChem

This article is protected by copyright. All rights reserved.