Author’s Accepted Manuscript

Few Layer Nitrogen-doped Graphene with Highly

Reversible Potassium Storage

Zhicheng Ju, Peizhi Li, Guangyao Ma, Zheng Xing,

Quanchao Zhuang, Yitai Qian

www.elsevier.com/locate/ensm

PII: S2405-8297(17)30159-9
DOI: http://dx.doi.org/10.1016/j.ensm.2017.09.009
Reference: ENSM218
To appear in: Energy Storage Materials
Received date: 27 April 2017
Revised date: 12 September 2017
Accepted date: 18 September 2017
Cite this article as: Zhicheng Ju, Peizhi Li, Guangyao Ma, Zheng Xing,
Quanchao Zhuang and Yitai Qian, Few Layer Nitrogen-doped Graphene with
Highly Reversible Potassium Storage, Energy Storage Materials,
http://dx.doi.org/10.1016/j.ensm.2017.09.009
This is a PDF file of an unedited manuscript that has been accepted for
publication. As a service to our customers we are providing this early version of
the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting galley proof before it is published in its final citable form.
Please note that during the production process errors may be discovered which
could affect the content, and all legal disclaimers that apply to the journal pertain.
 Few Layer Nitrogen-doped Graphene with Highly Reversible

Potassium Storage

Zhicheng Jua,b,, Peizhi Lia Guangyao Maa, Zheng Xinga,*, Quanchao Zhuanga,

Yitai Qianb

a
School of Materials Science and Engineering, China University of Mining and Technology, Xuzhou, Jiangsu

221116, P. R. China.
b
Hefei National Laboratory for Physical Science at Microscale, Department of Chemistry,
University of Science and Technology of China, Hefei 230026, P.R. China.

Abstract

Few-layer nitrogen-doped graphene (FLNG) have been successfully prepared by a simple

bottom-up synthesis of technique using Dicyandiamide and Coal tar pitch as raw materials. The

as-synthesized FLNG with the thickness of about 2-10 nm, high surface area (479.21 m2g−1) and

high nitrogen content (14.68 at%) exhibits excellent K-ion storage performances. The FLNG

delivers competitive reversible capacity (320 mAh·g−1 at 50 mA·g−1 after 60 cycles), superior rate

capability as well as long-term cycling life at large current density (150 mAh·g−1 at 500 mA·g−1

after 500 cycles). The multiple synergistic effects of nitrogen doping, high specific surface area,

interconnected mesopores with large pore volume not only facilitate the ions transportation

throughout the electrode matrix, but also provide substantial active sites for K-ion storage.

Furthermore, Electrochemical Impedance Spectroscopy (EIS) of the FLNG electrode during the

initial K-ion intercalation process were measured to thoroughly understand the electrochemical

Correspondence and requests for materials should be addressed to Z. Ju. (email: juzc@cumt.edu.cn) or to Z.

Xing (email: xzh086@cumt.edu.cn)

reaction process. These results will be beneficial for designing novel anode materials for

alkaline-ion batteries in the future.

Keywords: N-doped graphene; synergistic effects; electrochemical impedance spectroscopy;

diffusion coefficient; potassium ion battery.

Introduction

In recent years, rechargeable lithium-ion batteries (LIBs) have been broadly applied to mobile

electronic devices and small size energy storage systems due to their noticeable advantages.

Contrary to the tremendous demand for lithium, the relative proportion in the lithosphere is

estimated to be 35 ppm,1 which would not be in enough quantities to meet the future requirement

for both automotive applications and electrical energy storage of the grid. In this case, concerns

about the shortage and uneven geographical distribution of Li resources have led to the increased

search for alternative Earth-abundant metal-ion battery systems; so, alkaline-ion batteries draw

most attentions owing to their similar physical and chemical characteristics like metal Li.2, 3

Especially, the natural abundance of sodium and potassium is much higher than lithium in the

earth's crust (relative proportion in lithosphere: Li 35 ppm, Na 28300 ppm and K 25900 ppm).1, 4

The sodium and potassium-ion battery systems are promising alternatives because of their large

reserves. So it has important strategic significance to develop the alkali metals ion battery

technology at room temperature, particularly with regard to large-scale energy storage applications.

However, the application of sodium ion battery anode materials is still restricted because only

small amount of Na ion could form reversible graphite intercalation compounds (GICs) with

graphite carbon5, 6
which is the anode material in the majority of commercial rechargeable
batteries. This phenomena is mainly associated with sodium plating on the graphite surface before

forming a stage-I GICs7 which have been explained from a thermodynamic perspective that NaC6

was the only graphite intercalation compound which was not energetically favorable.8, 9 In this

case, disordered carbons are usually applied as high capacity anode materials for NIBs instead of

graphite,10 which may have relevance to poor cyclability and rate performance and be different

from the operation mechanism of commercial LIBs.11

However, the recently results of the Xiulei Ji group12 and Liangbing Hu group13 have shown

that K ions can be reversibly intercalation/deintercalation in the graphite material and showed

better reactivity compared to NIBs. Furthermore, K ion-based electrochemical energy storage

technologies also exhibit a great promise due to the high natural abundance; and more importantly,

the redox potential of K/K+ (−2.92 V vs. standard hydrogen electrode, noted as SHE) is even

lower than that of Na/Na+ (−2.71 V vs. SHE),13 indicating a higher working voltage of K

ion-based batteries. Although potassium possesses larger ion radius than sodium, potassium

readily forms intercalated compounds (stage-I K-GICs) with graphite in which K ions are inserted

in the graphene interlayers until forming KC8 structure. These results fully demonstrate that the

potassium ion batteries (PIBs) could meet the basic requirements for stationary batteries, which is

low cost upon scaling up where the economies of scale should be applicable. However, so far only

several results about the PIBs have been reported, such as Prussian blue14, Graphite12, 15, Graphitic

Materials13, 16 and Hard Carbon Microspheres17. Therefore, the development in PIBs technology

would expand its application areas, as well as possible to put into production and life to be of great

significance.

In recent days, graphene has attracted widespread attentions18 due to its fascinating characters
and its broad applications, such as in energy storage19 optoelectronics20, environmental

remediation21 and catalysis22. Previous studies have shown that the nitrogen doping into graphene

sheets would change and enhance the characterization of graphene such as magnetic moment,23

photoluminescence (PL) property24 and band structures.25 Nitrogen doping has also been proved to

be an efficient route for promoting the electrochemical performances of graphene because the

nitrogen decorated local electronic structures could enhance the transportation of lithium ion and

raise the storage performance.26-29 Nitrogen-doped graphene synthesized at 600 °C in mixed NH3

and Ar delivers a capacity of more than 1040 mAh·g−1 at the current of 50 mA·g−1 after 30 cycles

in LIBs.30 3D nitrogen-doped graphene foams synthesized by sintering the graphene oxide in NH3

gas keeps a charge/discharge capacity of 594 mAh·g−1 at the current of 500 mA·g−1 after 150

cycles in NIBs.31 Accordingly, N-doping could significantly alter the electrical conductivity and

effectively promote the ion storage properties of graphene. Pint’s group11, 32 have thoroughly

investigated the effect of N-doping on few-layered graphene, which illustrated that N-doping

could significantly increase the K-ion storage capacity to over 350 mAh/g and applied in situ

Raman spectroscopy to research the K-ion storage mechanism in N-doped and un-doped few layer

graphene. N-doped carbon microspheres fabricated from seafood waste33 were applied as anode

electrode materials on PIBs and showed high specific capacity (about 205 mAh/g at 0.12 C for

nearly 200 cycles) and rate capability (about 180 mAh/g at 1.8 C for more than 4000 cycles). In

this case, the N-doped graphene would lead to ideal capabilities for PIBs.

Motivated by this, we utilized few-layer nitrogen-doped graphene (FLNG) fabricated by

carbonization of appropriate amounts of Dicyandiamide and Coal tar pitch mixture as the anode

material for room-temperature PIBs. The rechargeable K-ion half cell consists of K-foil anode,
glass fiber separator, KPF6 solution electrolyte, and the FLNG cathode on copper foil (Figure 1a).

In rechargeable batteries, the K anode produces K-ion during discharge corresponding the anode

reaction of K → K+ + e. The electrolyte is 0.8 M KPF6 solution in ethylene carbonate/dimethyl

carbonate (EC/DMC 1:1 v/v) solvent based on its high ionic conductivity.

The FLNG/K metal half-cells with nonaqueous electrolyte were assembled to investigate the

electrochemical intercalation process of K ions into FLNG. Extensive researches carried out on

PIBs demonstrate that the FLNG electrode exhibits high specific capacity (above 320 mAh·g−1

after 60 cycles at 50 mA·g−1), excellent rate capability (170 mAh·g−1 at 500 mA·g−1) and superior

long cycle-life at large current density (above 150 mAh·g−1 after 500 cycles at 500 mA·g−1),

which suggesting its immense promising to be an effective anode material candidate for K storage.

Figure 1 (a) Scheme of the K-ion half-cells’ structure, (b) Scheme of a proposed mechanism for the fabrication

process of the FLNG.

Experimental Section

Synthesis of nitrogen-doped graphene

All reagents were purchased from Aladdin Chemical Reagents Company with analytical purity
which were used without purification. The FLNG was synthesized using appropriate amounts of

Dicyandiamide and Coal tar pitch as reagents. Figure 1b shows the illustration of the fabrication

process of FLNG. In a typical method, Dicyandiamide (9.0 g), Coal tar pitch (2.0 g) and 20 ml

Ethylene glycol were added into a milling pot. The mixture was ball-milled for 4 h by Nanjing

University Instrument Factory planetary ball mill, and then dried by vacuum at 80 °C. The

obtained powder was sintered in a tube furnace under nitrogen atmosphere; the reaction

temperature was first increased to 580 °C by 2 °C min-1 and kept for 4 h. At 580 °C, the

dicyandiamide was aggravated into 2D g-C3N4 nanosheets, as well as the Coal tar pitch turned soft

and coated on the g-C3N4 nanosheets. Then the reaction temperature was increased to 800 °C and

kept for 2 h; the polycyclic aromatic hydrocarbon in Coal tar pitch polycondensed and patching

grown on the g-C3N4 template; the g-C3N4 was partially decomposed at the same time (Figure S1

in the Supporting Information).

If no Coal tar pitch were added, the pure g-C3N4 nanosheets would form from dicyandiamide at

580 C. The XRD pattern of the g-C3N4 nanosheets was shown in Figure S2. If the temperature

raised to 800 C, the g-C3N4 nanosheets would disappear and nothing could be collected at the

end.

Material characterization

The X-ray powder diffraction (XRD) patterns of the sample were recorded by Bruker D8

advanced X-ray diffractometer (Cu Kα radiation, λ=1.5418 Å). The morphology of the sample

were performed using a JEOL JSM-6700F field emission scanning electron microscope (SEM)

operating at 15 kV. Transmission electron microscopy (TEM) images were carried out by

JEOL-2010 transmission electron microscope with accelerate voltage of 200 kV. The Raman
spectra were obtained with Horiba-Jobin Yvon LabRam HR equipped with visible laser excitation

of 514 nm. The XPS measurements were taken on Thermo Scientific ESCALAB 250 X-ray

photoelectronic spectrometer with non-monochromated Mg Kα X-ray radiation as the excitation

source. The surface areas of the products were measured by Micromeritics TriStar II 3020 surface

area analyzer and calculated by N2 adsorption-desorption isotherms. The electrode was composed

of nitrogen-doped graphene (80 wt %), conductivity agent (Super P acetylene black, 10 wt %) and

binder (polyvinylidene fluoride) (10 wt %), with cupper-foil as the current collector. The slurry

was coated on the Cu-foil by doctor-blade technique and dried at 60 C under vacuum. Electrodes

discs with diameter of 14 mm were cut for electrochemical tests. Typical electrode active material

loading was about ca. 1.0-1.2 mg on one disc. The button-type cells (2025) were assembled in an

argon-filled glove box (Mikrouna, Super 1200/750). Metal K foil was used as the counter

electrode. The separator was the Whatman glass fiber, and the electrolyte was 0.8 M KPF6

solution dissolved in ethylene carbonate/dimethyl carbonate solvent (EC/DMC, 1:1 v/v). The

galvanostatic charge/discharge performance was measured by LAND-CT2001A multichannel

tester (Wuhan Lanhe, China) in 0.01-1.5 V (vs. K+/K) at room temperature. The cyclic

voltammetry (CV) profiles were carried out in the voltage range of 0.01-3.0 V by scan rate of 0.3

mV/s on electrochemical workstation (CHI 760E, Shanghai, China). The EIS was measured in a

three-electrode system scanning from 1 MHz to 0.01 Hz with alternating current signal amplitude

of 5 mV were tested on CHI660B electrochemical workstation.

Results and discussion

Figure 2 (a) XRD pattern of FLNG prepared through the typical reaction, the inset is a schematic of the

cyameluric core inherited from the g-C3N4 template, (b) Raman spectra of FLNG, (c) Nitrogen-adsorption

isotherms of FLNG, (d) BJH desorption pore-size distribution and pore volume.

The composition and structure of the as-synthesized sample were first analyzed by XRD. It is

clearly observed that a broad diffraction peak (Figure 2a) around 13.1° corresponds to the

interlayer distance of 0.68 nm. This d-spacing is agreed with the in-plane periodicity packing

motif of the g-C3N4 as shown in Figure S2. The FLNG was synthesized by the g-C3N4 induced

templated method; so the broad peak could be attributed to the cyameluric cores (as shown the

schematic of cyameluric cores is shown in the inset of Figure 2a) caused by nitrogen doping

which is inherited from the g-C3N4 template.34 The broad diffraction peak around 25.9° with low

intensity in the XRD pattern could be indexed to the (002) diffractions of graphitic carbon (JCPDS

No. 41-1487), while the broad peak around 42.3° could be attributed to (101) diffractions of
graphitic carbon. The interlayer distance of the (002) plane could be calculated to be 3.44 nm

(which is larger than dgraphite (002) = 3.35 nm). The low intensity and larger d-spacing actually

indicate that the graphene layers in the FLNG are loosely stacked along c-direction. In addition,

the average thickness of the crystal along c-direction (designated as Lc) and the in-plane diameter

of the crystal (designated as La) could be calculated based on the Debye–Sherrer equation (Lc =

0.89λ/Bcosθ and La = 1.84λ/Bcosθ).35 The Lc of the sample is 8.9 nm and the La is 11.4 nm,

respectively.

Furthermore, the above results can be further verified by the Raman spectra. As presented in

Figure 2b, the FLNG possesses two strong peaks centered at 1348 cm-1 (D-band) and 1595 cm-1

(G-band), respectively. The G-band reflected the in-plane vibrations of the bonding between

carbon atoms involves a layer of sp2 hybridized orbitals, while the D-band could be result in the

vibration of the dangling bonding of the in-plane terminations of disordered carbon atoms,

reflecting the existence of defects and boundaries in the graphene domains. Furthermore, the weak

and broad 2D peak centered at 2750 cm-1 implies that the FLNG holds the feature structure of the

few-layered graphene.36 The value of ratio ID/IG could be generally connected with the graphitic

structures of sp2 domains in the graphene layer. The larger ID/IG value (1.08) implies the existence

of massive structural defects in the graphene layers with smaller sp2 regions.37 The average

crystalline area along the a-axis (La) of sp2 domains could be estimated by the equation: La =

(2.4×10-10)λ4(ID/IG)-1 (empirical Tuinstra-Koenig equation, λ = 514 nm).38, 39 The average size of

the ordered sp2 regions of FLNG could by calculated to be about 15.1 nm, which is larger than the

La calculated by XRD.

The surface area and porosity of the FLNG were further investigated by nitrogen
adsorption/desorption isotherms and analyzed by Brunauer-Emmett-Teller (BET) method (Figure

2c and 2d). The specific surface area of the sample is calculated up to 479.21 m2g−1. The isotherm

curve of FLNG (Figure 2c) shows a typical IV type isotherm with broad H2 type hysteresis loops

between adsorption and desorption curves at P/P0 = 0.4-0.9 which implies massive interconnected

cage-like mesopores with similar pore size.40, 41 The average pore size distribution (Figure 2d)

calculated by Barrett–Joyner–Halenda (BJH) method according to desorption curve exhibits one

main peak centered at about 5.34 nm. The pore volume calculated by BJH method is about 0.71

cm3/g. The mesopores existed in the basal planes of graphene could facilitate the K-ion

transportation, providing a fast access from graphene to the electrolyte.42 The interconnected

mesoporous structure along with the large surface area would effectively promote the intercalation

of K-ions.
 Figure 3 (a) typical SEM micrograph of wrinkled FLNG sheets, (b) high resolution TEM of an individual

graphene sheet, (c) XPS survey spectra of FLNG, (d) C 1s, (e) N 1s, (f) i: graphitic N, ii: pyridinic N, and iii:

pyrrolic N.

The morphology and structure of FLNG sample were imaged by TEM and SEM investigation.

Figure S3a and b present the representative TEM pictures of the obtained sample; it could be
clearly observed that the sample displays a vague and flexible appearance with a typical curled

and overlapped nanofilm structure, which results from thermodynamically stable bending. To

assess the thickness of the few-layered FLNG, more SEM images (which shows in Figure 3a and

Figure S4) have been taken. The average thickness of the FLNG is mainly distributed in 2-10 nm with

generally 5−30 graphene layers. The high resolution TEM image (Figure 3b) shows several typical

FLNG nanosheets; the FLNG sheet around 4.8 nm possesses approximately 14 stacked single

graphene layer. The layer-to-layer distance (d spacing) is estimated to be 0.342 nm, which is larger

than that of graphite (0.335 nm) and agreed with the interlayer spacing (d002) calculated by XRD

pattern. This kind of thin film like structure is more advantageous to the interface storage and high

rate charge and discharge.

XPS was used to investigate the chemical valence of the elements in the FLNG. As shown in

Figure 3c, there were three distinct peaks located at 284.6, 399.2, and 531.8 eV in the XPS survey

spectra, which could be assigned to C 1s of sp2 carbon, N 1s of nitrogen, and O 1s, respectively.

The O 1s spectrum could be attributed to the existence of oxygen groups including C=O, C–O,

and O=C=O, respectively.43 The C 1s spectra ranging from 280–295 eV (Figure 3d) exhibit four

peaks by curve fitting. The primary peak centered at 284.6 eV could be attributed to the sp2

carbon atom (C1) constituted graphitic regions, which demonstrates that majority of carbon atoms

are located in graphitic carbon. The peak centered at 288 eV could be identified as sp2-hybrid

carbon arranged in the aromatic nucleus which was linked with nitrogen groups. The peaks

located at 285.3 and 286 eV were originated from C–N and C–O/C=N groups, respectively.44 The

N 1s peak could be divided into three individual peaks (as presented in Figure 3e) located at 398.2

eV, 399.8 eV and 400.9 eV associated with pyridinic (N-6), pyrrolic (N-5) and graphitic nitrogen
(N-g), respectively.44, 45 The amount of nitrogen doped in the graphene is 14.68 at% calculated by

XPS elemental analysis. Furthermore, the content of the N-doping can be proper controlled by

adjusting the amount of dicyandiamide (Figure S5). The high nitrogen doping level could be

derived from the high nitrogen content of dicyandiamide. According to the XPS spectra, the

nitrogen functional groups configuration includes 49% pyridinic N, 30% pyrrolic N and 21%

graphitic N. This implies that the carbon-nitrogen double bonds and triple bonds within the

dicyandiamide are all converted into N-6, N-5 and N-g during the carbonization and graphitization

process. Depended on the above results, the possible structure illustration with nitrogen functional

groups is presented in Figure 3f. The N-5 and N-6 nitrogen atoms are generally locating on the

edge of the honeycomb-like lattice, which is of high chemical activity. Generally, the occurrence

of nitrogen atoms in the aromatic rings accompanied with the structure defects such as vacancies,

bonding disorders and non-cyclized structures would enhance the reversible capacity of graphene

with alkali metal ions.46 Meanwhile, the higher electronegativity of nitrogen (3.04) than carbon

(2.55) would also be beneficial to form stronger interaction between the graphene structure and

K-ion. Furthermore, the doping will significantly influence the electronic characteristics of the

graphene; and K would be favorable around the defects and sites in the vicinity of residual N

atoms.

The electrochemical properties of the FLNG were first tested by cyclic voltammetry (CV)

measurements in the voltage window of 0.01–3.0 V vs. K/K+ (Figure 4a). The CV record

indicates the electrode presents three cathodic peak around 0.51 V (R1), 0.31 V (R2), 0.09 V (R3)

and one rather broad anodic peak around 0.68 V (O1) during the first cycle. The first cathodic peak

(R1) in the 1st cycle could be associate with the reduction of the electrolyte and the form of solid
electrolyte interphase (SEI) film on the surfaces of FLNG. Similar peak could also be observed

during lithium intercalation process into the carbon-based materials.47 The anodic peak around

0.68 V (O1) were mainly attributed to the depotassiation progress, while the sharp cathodic peaks

centered at 0.31 V (R2) and 0.09 V (R3) are related to the stepwise intercalation of K+ into

FLNG.12 In the following anodic sweep, the broad peak between 0.30 to 0.65 V corresponds to the

deintercalation of K+ from FLNG. In the subsequent cycles, the cathodic peak R1 disappears,

which means that SEI mainly forms in the first cathodic process. Moreover, the other

cathodic/anodic peaks are highly reversible and overlap, suggesting the high reversibility of the

intercalation and deintercalation of K+ into/from FLNG.

Figure 4b presents the representative galvanostatic charge/discharge curves of selected cycles.

FLNG exhibits surprisingly high capacities of 944.5 and 352.2 mAh·g−1 in potassiation and

depotassiation, respectively. Unfortunately, the initial Columbic efficiency (CE) is very low (only

37.29%). The depotassiation capacity is larger than 279 mAh·g−1 (theoretical value) when KC8

forms; the additional capacity could be attributed to the surface storage. The initial potassiation

storage quantity is much larger than the theoretical value, which is responsible for the

decomposition of the electrolyte and formation of SEI film. This phenomenon is in agreement

with the CV results. Comparing the first and other cycle potassiation potential profiles shows that

the slope region from 1 to 0.40 V only existed in the initial cycle, and vanished in the following

cycles. This is consistent with the first-cycle behavior of graphite anode in PIBs when SEI forms

on the graphite surface.12 The following less steep slope from 0.40 to 0.01 V in the first

potassiation also contributes to the initial irreversible capacity (formation of SEI) as it is much

diminished in the second cycle. Besides, the depotassiation potential profiles almost overlap from
the first cycle.

Figure 4 Electrochemical performance of the FLNG/K metal half-cell. (a) Cyclic voltammogram, (b)

Galvanostatic charge/discharge profile at a current rate of 50 mA·g−1, (c) Discharge/charge capacity at current

density of 50 mA·g−1 and coulombic efficiency, (d) Rate capability from 50 to 500 mA·g−1, (e) Cycling

performance at a high current density of 500 mA·g-1.

Figure 4c shows the discharge-charge cycling property of FLNG electrode at 50 mA·g−1. After

the tenth cycle, the specific capacity becomes stable and maintains at about 320 mAh·g−1 until 60th

cycle, which was much higher than many previous reported carbonaceous materials (some
excellent papers are summarized in Table S1); the CE is increased from 37.29% to more than 97%,

indicating superior capacity retention capability. Besides the good cyclability, the rate capacity of

FLNG is also excellent (Figure 4d). Even in the higher current of 500 mA·g−1, the capacity is still

higher than 170 mAh·g−1. If the current goes back to 50 mA·g−1, the capacity returns to 305

mAh·g−1, which almost recovers the initial cycle capacity. This suggests that the special structure

of FLNG could remain steady at the large current density. The electrochemical capability at large

current density (500 mA·g−1) was also tested, which exhibits high cycling stability (Figure 4e).

Even after 500 cycles, the charge/discharge capacity still kept more than 150 mA·g−1 which is

even higher than sodium-ion batteries at the same current density.48

The K-ion storage performances of graphene without N-doping obtained through an improved

Hummers’ method and further reduction using Ar/H2 atmosphere was also investigated (Figure

S6). The data implied that the specific capacity of the electrode material decreases steadily without

N-doping. The above results sufficiently attested that the FLNG electrode possesses good

reversible capacity as well as excellent cycling capability. This is mainly originated from the

unique doping mechanism and thin film like morphology which are favorable to the insertion and

extraction of the potassium ions. Furthermore, the nitrogen functional groups could effectively

keep the graphene layers from restacking and maintain large interlayer distance,44 and

consequently improve the interfacial contact between FLNG and electrolyte, in the meantime,

ameliorate the transportation of the K+ in/on graphene layers. In order to investigated this effect,

ex-situ XRD pattern of the FLNG electrode (after 50 cycles at 500 mA·g−1) was shown in Figure

S9. It is clearly shown that the (002) Miller peak shifts to the lower angle (25.0°) after discharge

and charge progress, which suggested that significant expansion took place, and the interlayer
distance of FLNG was enlarged to about 0.36 nm. It should be pointed out that, in this research

project, the gaps or pores will be produced by the graphene thin film stack and curl, which can

offer more active sites and hold the volume changes during K-ion intercalation/de-intercalation.

EIS of the FLNG electrode during the first and fifth cycle were measured by a three-electrode

system to thoroughly understand the electrochemical reaction process. The Nyquist plots of the

EIS results at different voltages during the initial K-ion insertion procedure are summarized in

Figure 5, Figure 6 and Figure S10. At 3.0 V, the impedance spectroscopy comprised one

depressed semicircle in the high frequency area (abbreviated to HFS) and one slope line in lower

frequency area belonging to a semi-infinite solid diffusion process.49 The slightly inclined slope

was related to the blocking characteristics of the electrode material without ion insertion at the

equilibrium potential.50 The morphologies of the HFS kept unchanged when the potential was

reduced from 3.0 V to 1.0 V which were mainly because that the SEI film had not appeared on the

electrode surface before 1.0 V.51 In this case, the HFS could be only originated from the contact

resistance which mainly related to the interface between the electrolyte and FLNG. The

unchanged HFS also suggested that the SEI film generated by electrolyte reduction decomposition

has not been formed above 1.0 V, which conformed to the CV results.

Below 1.0 V, the oblique line in the lower frequency area appeared a tendency to deviate to real

axis. While the electrode polarization voltage reduced, a new depressed semicircle in

middle-frequency (MFS) was appeared. In the potential range from 0.9 to 0.6 V, the impedance

spectra constitute of two semicircles and one slope line. Comparing with the EIS of lithium

batteries,52 the HFS could be assigned to ion transportation through the SEI film overlapped the

electrode surface; the MFS could be ascribed to the charge transfer across the interfaces of
electrode and electrolyte; as well as the slope is corresponding to K-ion solid-state diffusion in the

graphene interior. When the potential decreased to 0.5 V, the MFS in the Nyquist plot below

becomes obscure and the Nyquist plots mainly consist of a HFS and a sloping line. The

morphologies of the Nyquist plots are similar even when the potential decrease to 0.1 V.

Figure 5 Nyquist plots of the FLNG electrode at different voltages from 3.0 V to 0.6 V in the initial

potassiation.

The typical impedance data measured at 0.1 V could be simulated by a suitable analog circuit

diagram in Figure 6b. According to the circuit diagram, RS means the electrolyte Ohmic resistance.

R1 represents the resistance of surface film and contact resistance which are described as
uncompensated resistance. Rct means the charge-transfer resistance in the high frequency and low

frequency regions. Qdl is the double-layer capacitance in the middle frequency regions,

respectively. The QD represents the Warburg impedance linked with the diffusion process of

K-ions into the graphene layers.

The K-ion diffusion coefficient is obtained by the EIS curve of the FLNG during the first (as

shown in Figure 6b) and fifth (as shown in Figure S10d) discharge cycle at 0.1 V, respectively.

The K-ion diffusion coefficient could be obtained by the following formula:53, 54

According to this formula, R represents the gas constant, T represents the experiment temperature,

F represents the Faraday constant. A represents the electrode surface area, n means the electrons

per molecule involving in the reaction, and C represents the K-ion concentration in graphene

electrode. The σ is the Warburg coefficient which could be calculated by the gradient of the

oblique line Z’ ∼ ω−1/2 (ω = 2πf) in the low-frequency area (shown in Figure S11). The K-ion

diffusion coefficient is about 2.36 × 10−10 cm2 s−1, which is a little smaller than the Li-ion solid

diffusion coefficient in the graphite (about 10−10 cm2 s−1) and similar as the results of N- and O-rich

carbon nanofiber and Nitrogen-rich hard carbon calculated by Galvanostatic Intermittent Titration

Technique.33, 55 After 5 discharge cycles, the diffusion coefficient (at 0.1 V) varied to 8.39 × 10−10

cm2 s−1.
Figure 6 EIS analysis of the FLNG electrode. (a) Nyquist plots at different voltages from 0.5 V to 0.1 V in the

initial K-ion insertion, (b) Typical analog circuit diagram proposed for the simulation of the FLNG electrode at 0.1

V.

Figure 7 Scheme of the mixed mechanisms for potassium ions storage: a. K-ion reversibly trapping in the surface;

K-ion reversibly trapping in the defect sites.

The high potassium storage performance of FLNG results from its unique structures especially

the nitrogen doping defects, large specific surface area, and few layer structure of graphene ,

which provide multiple potassium storage positions as shown in Figure 7. (a) Amorphous

crystallites with a large amount of interlayer structure defects including nitride pores and holes

offer active sites for K-ion storage which could give contribution to the capacity. (b) The high

proportion of nitrogen-doping would improve the conductivity of the FLNG, which is propitious

to achieve good cycle stability and rate performance. The conductivity of the FLNG (10 Ω·mm2/m)
was measured by a Suzhou Tongchuang SZT-2A four probe resistivity tester, which very closed to

conductive carbon black (1.2-1.8 Ω·mm2/m). (c) The large surface area with interconnected

nanopores and defects could shorten the K-ion diffusion pathway and guarantee sufficient contact

in the electrode/electrolyte interface for facile charge transfer. (d) The N-doping could

simultaneously activate distributed reversible storage sites in the graphene lattice while

maintaining the energetic pathway for the composing of KC8. All these structures with their

multiple synergistic effects lead to the excellent potassium storage of the FLNG.

Conclusions

In summary, the FLNG have been synthesized by an improved method using Dicyandiamide

and Coal tar pitch by a solid state reaction. The FLNG possesses unique structures such as high

specific surface area (479.21 m2g−1), and high proportion of nitrogen-doping (14.68 at%) defects

which could offer sufficient active sites for K-ion storage and promote ions and electrons transfer

with nanosized diffusion pathways. Therefore, the FLNG exhibits excellent electrochemical

properties for PIBs such as large specific capacity (320 mAh·g−1 until 60th cycle at 50 mA·g−1),

superior rate capability and long-term cycling life (more than 150 mAh·g−1 after 500 cycles at 500

mA·g−1), which is a very competitive performance. As a result, the FLNG would be expected to

become available in the future for potential applications in the areas of PIBs.

Acknowledgments

This work was supported by the Fundamental Research Funds for the Central University

2014QNA15.

References
1. Teng, F. Z.; McDonough, W. F.; Rudnick, R. L.; Dalpé, C.; Tomascak, P. B.; Chappell, B. W.; Gao, S.,
Lithium isotopic composition and concentration of the upper continental crust. Geochimica et
Cosmochimica Acta 2004, 68, 4167-4178.
2. Slater, M. D.; Kim, D.; Lee, E.; Johnson, C. S., Sodium-Ion Batteries. Advanced Functional Materials
2013, 23, 947-958.
3. Eftekhari, A.; Jian, Z.; Ji, X., Potassium Secondary Batteries. Acs Applied Materials & Interfaces
2017, 9, 4404-4419.
4. Seyfried, W. E.; Janecky, D. R.; Mottl, M. J., Alteration of the oceanic crust: Implications for
geochemical cycles of lithium and boron. Geochimica et Cosmochimica Acta 1984, 48, 557-569.
5. Wen, Y.; He, K.; Zhu, Y.; Han, F.; Xu, Y.; Matsuda, I.; Ishii, Y.; Cumings, J.; Wang, C., Expanded
graphite as superior anode for sodium-ion batteries. Nature Communication 2014, 5, 4033.
6. Sun, J.; Lee, H.-W.; Pasta, M.; Yuan, H.; Zheng, G.; Sun, Y.; Li, Y.; Cui, Y., A phosphorene–graphene
hybrid material as a high-capacity anode for sodium-ion batteries. Nature Nanotechnology 2015, 10,
980-985.
7. Ge, P.; Fouletier, M., Electrochemical intercalation of sodium in graphite. Solid State Ionics 1988,
28, 1172-1175.
8. DiVincenzo, D. P.; Mele, E. J., Cohesion and structure in stage-1 graphite intercalation compounds.
Physical Review B 1985, 32, 2538-2553.
9. Wang, Z.; Ratvik, A. P.; Grande, T.; Selbach, S. M., Diffusion of alkali metals in the first stage
graphite intercalation compounds by vdW-DFT calculations. RSC Advances 2015, 5, 15985-15992.
10. Tang, K.; Fu, L.; White, R. J.; Yu, L.; Titirici, M.-M.; Antonietti, M.; Maier, J., Hollow Carbon
Nanospheres with Superior Rate Capability for Sodium-Based Batteries. Advanced Energy Materials
2012, 2, 873-877.
11. Share, K.; Cohn, A. P.; Carter, R.; Rogers, B.; Pint, C. L., Role of Nitrogen-Doped Graphene for
Improved High-Capacity Potassium Ion Battery Anodes. Acs Nano 2016, 10, 9738-9744.
12. Jian, Z.; Luo, W.; Ji, X., Carbon Electrodes for K-Ion Batteries. Journal of the American Chemical
Society 2015, 137, 11566-11569.
13. Luo, W.; Wan, J.; Ozdemir, B.; Bao, W.; Chen, Y.; Dai, J.; Lin, H.; Xu, Y.; Gu, F.; Barone, V.; Hu, L.,
Potassium Ion Batteries with Graphitic Materials. Nano Letters 2015, 15, 7671-7677.
14. Wessells, C. D.; Peddada, S. V.; Huggins, R. A.; Cui, Y., Nickel Hexacyanoferrate Nanoparticle
Electrodes For Aqueous Sodium and Potassium Ion Batteries. Nano Letters 2011, 11, 5421-5425.
15. Xing, Z.; Qi, Y.; Jian, Z.; Ji, X., Polynanocrystalline Graphite: A New Carbon Anode with Superior
Cycling Performance for K-Ion Batteries. Acs Applied Materials & Interfaces 2017, 9, 4343-4351.
16. Ju, Z.; Zhang, S.; Xing, Z.; Zhuang, Q.; Qiang, Y.; Qian, Y., Direct Synthesis of Few-Layer F-Doped
Graphene Foam and Its Lithium/Potassium Storage Properties. ACS Applied Materials & Interfaces
2016, 8, 20682-20690.
17. Jian, Z.; Xing, Z.; Bommier, C.; Li, Z.; Ji, X., Hard Carbon Microspheres: Potassium‐Ion Anode
Versus Sodium‐Ion Anode. Advanced Energy Materials 2016, 6, 1501874.
18. Cui, H.; Zhou, Z.; Jia, D., Heteroatom-doped graphene as electrocatalysts for air cathodes.
Materials Horizons 2017, 4, 7-19.
19. Xu, C.; Xu, B.; Gu, Y.; Xiong, Z.; Sun, J.; Zhao, X. S., Graphene-based electrodes for electrochemical
energy storage. Energy & Environmental Science 2013, 6, 1388-1414.
20. Chang, H.; Wu, H., Graphene-Based Nanomaterials: Synthesis, Properties, and Optical and
Optoelectronic Applications. Advanced Functional Materials 2013, 23, 1984-1997.
21. Sun, H.; Liu, S.; Zhou, G.; Ang, H. M.; Tadé, M. O.; Wang, S., Reduced graphene oxide for catalytic
oxidation of aqueous organic pollutants. Acs Applied Materials & Interfaces 2012, 4, 5466-5471.
22. Xia, B.; Yan, Y.; Wang, X.; Lou, X. W., Recent progress on graphene-based hybrid electrocatalysts.
Materials Horizons 2014, 1, 379-399.
23. Li, Y.; Zhou, Z.; Shen, P.; Chen, Z., Spin Gapless Semiconductor−Metal−Half-Metal Properties in
Nitrogen-Doped Zigzag Graphene Nanoribbons. ACS Nano 2009, 3, 1952-1958.
24. Chiou, J. W.; Ray, S. C.; Peng, S. I.; Chuang, C. H.; Wang, B. Y.; Tsai, H. M.; Pao, C. W.; Lin, H. J.;
Shao, Y. C.; Wang, Y. F.; Chen, S. C.; Pong, W. F.; Yeh, Y. C.; Chen, C. W.; Chen, L. C.; Chen, K. H.; Tsai, M.
H.; Kumar, A.; Ganguly, A.; Papakonstantinou, P.; Yamane, H.; Kosugi, N.; Regier, T.; Liu, L.; Sham, T. K.,
Nitrogen-Functionalized Graphene Nanoflakes (GNFs:N): Tunable Photoluminescence and Electronic
Structures. The Journal of Physical Chemistry C 2012, 116, 16251-16258.
25. Cervantes-Sodi, F.; Csányi, G.; Piscanec, S.; Ferrari, A. C., Edge-functionalized and substitutionally
doped graphene nanoribbons: Electronic and spin properties. Physical Review B 2008, 77, 165427.
26. Xing, Z.; Ju, Z.; Zhao, Y.; Wan, J.; Zhu, Y.; Qiang, Y.; Qian, Y., One-pot hydrothermal synthesis of
Nitrogen-doped graphene as high-performance anode materials for lithium ion batteries. Scientific
Reports 2016, 6, 26146.
27. Ma, C.; Shao, X.; Cao, D., Nitrogen-doped graphene nanosheets as anode materials for lithium
ion batteries: a first-principles study. Journal of Materials Chemistry 2012, 22, 8911-8915.
28. Gu, X.; Tong, C.-J.; Rehman, S.; Liu, L.-M.; Hou, Y.; Zhang, S., Multifunctional Nitrogen-Doped
Loofah Sponge Carbon Blocking Layer for High-Performance Rechargeable Lithium Batteries. Acs
Applied Materials & Interfaces 2016, 8, 15991-16001.
29. Shan, H.; Li, X.; Cui, Y.; Xiong, D.; Yan, B.; Li, D.; Lushington, A.; Sun, X., Sulfur/Nitrogen
Dual-doped Porous Graphene Aerogels Enhancing Anode Performance of Lithium Ion Batteries.
Electrochimica Acta 2016, 205, 188-197.
30. Wu, Z.-S.; Ren, W.; Xu, L.; Li, F.; Cheng, H.-M., Doped Graphene Sheets As Anode Materials with
Superhigh Rate and Large Capacity for Lithium Ion Batteries. Acs Nano 2011, 5, 5463-5471.
31. Xu, J.; Wang, M.; Wickramaratne, N. P.; Jaroniec, M.; Dou, S.; Dai, L., High-Performance Sodium
Ion Batteries Based on a 3D Anode from Nitrogen-Doped Graphene Foams. Advanced Materials 2015,
27, 2042-2048.
32. Share, K.; Cohn, A. P.; Carter, R. E.; Pint, C. L., Mechanism of potassium ion intercalation staging in
few layered graphene from in situ Raman spectroscopy. Nanoscale 2016, 8, 16435-16439.
33. Chen, C.; Wang, Z.; Zhang, B.; Miao, L.; Cai, J.; Peng, L.; Huang, Y.; Jiang, J.; Huang, Y.; Zhang, L.;
Xie, J., Nitrogen-rich hard carbon as a highly durable anode for high-power potassium-ion batteries.
Energy Storage Materials 2017, 8, 161-168.
34. Goettmann, F.; Fischer, A.; Antonietti, M.; Thomas, A., Chemical Synthesis of Mesoporous Carbon
Nitrides Using Hard Templates and Their Use as a Metal‐Free Catalyst for Friedel–Crafts Reaction of
Benzene. Angewandte Chemie International Edition 2006, 45, 4467-4471.
35. Warren, B. E.; Bodenstein, P., The diffraction pattern of fine particle carbon blacks. Acta
Crystallographica 1965, 18, 282-286.
36. Ferrari, A. C.; Meyer, J. C.; Scardaci, V.; Casiraghi, C.; Lazzeri, M.; Mauri, F.; Piscanec, S.; Jiang, D.;
Novoselov, K. S.; Roth, S.; Geim, A. K., Raman Spectrum of Graphene and Graphene Layers. Physical
Review Letters 2006, 97, 187401.
37. Li, X.-H.; Kurasch, S.; Kaiser, U.; Antonietti, M., Synthesis of Monolayer-Patched Graphene from
Glucose. Angewandte Chemie International Edition 2012, 51, 9689-9692.
38. Zhang, C.; Fu, L.; Liu, N.; Liu, M.; Wang, Y.; Liu, Z., Synthesis of Nitrogen-Doped Graphene Using
Embedded Carbon and Nitrogen Sources. Advanced Materials 2011, 23, 1020-1024.
39. Tuinstra, F.; Koenig, J. L., Raman Spectrum of Graphite. The Journal of Chemical Physics 1970, 53,
1126-1130.
40. Fan, Z.; Liu, Y.; Yan, J.; Ning, G.; Wang, Q.; Wei, T.; Zhi, L.; Wei, F., Template-Directed Synthesis of
Pillared-Porous Carbon Nanosheet Architectures: High-Performance Electrode Materials for
Supercapacitors. Advanced Energy Materials 2012, 2, 419-424.
41. Matsuoka, K.; Yamagishi, Y.; Yamazaki, T.; Setoyama, N.; Tomita, A.; Kyotani, T., Extremely high
microporosity and sharp pore size distribution of a large surface area carbon prepared in the
nanochannels of zeolite Y. Carbon 2005, 43, 876-879.
42. Yao, F.; Güneş, F.; Ta, H. Q.; Lee, S. M.; Chae, S. J.; Sheem, K. Y.; Cojocaru, C. S.; Xie, S. S.; Lee, Y. H.,
Diffusion Mechanism of Lithium Ion through Basal Plane of Layered Graphene. Journal of the
American Chemical Society 2012, 134, 8646-8654.
43. Ou, J.; Zhang, Y.; Chen, L.; Zhao, Q.; Meng, Y.; Guo, Y.; Xiao, D., Nitrogen-rich porous carbon
derived from biomass as a high performance anode material for lithium ion batteries. Journal of
Materials Chemistry A 2015, 3, 6534-6541.
44. Tian, L.-L.; Wei, X.-Y.; Zhuang, Q.-C.; Jiang, C.-H.; Wu, C.; Ma, G.-Y.; Zhao, X.; Zong, Z.-M.; Sun,
S.-G., Bottom-up synthesis of nitrogen-doped graphene sheets for ultrafast lithium storage. Nanoscale
2014, 6, 6075-6083.
45. Wang, H.; Zhang, C.; Liu, Z.; Wang, L.; Han, P.; Xu, H.; Zhang, K.; Dong, S.; Yao, J.; Cui, G.,
Nitrogen-doped graphene nanosheets with excellent lithium storage properties. Journal of Materials
Chemistry 2011, 21, 5430-5434.
46. Hou, Z.; Wang, X.; Ikeda, T.; Terakura, K.; Oshima, M.; Kakimoto, M.-a., Electronic structure of
N-doped graphene with native point defects. Physical Review B 2013, 87, 165401.
47. Tang, K.; Fu, L.; White, R. J.; Yu, L.; Titirici, M. M.; Antonietti, M.; Maier, J., Hollow Carbon
Nanospheres with Superior Rate Capability for Sodium‐Based Batteries. Advanced Energy Materials
2012, 2, 873-877.
48. Ma, G.; Huang, K.; Zhuang, Q.; Ju, Z., Superior cycle stability of nitrogen-doped graphene
nanosheets for Na-ion batteries. Materials Letters 2016, 174, 221-225.
49. Li, G.-C.; Li, G.-R.; Ye, S.-H.; Gao, X.-P., A Polyaniline-Coated Sulfur/Carbon Composite with an
Enhanced High-Rate Capability as a Cathode Material for Lithium/Sulfur Batteries. Advanced Energy
Materials 2012, 2, 1238-1245.
50. Chang, Y. C.; Sohn, H. J., Electrochemical Impedance Analysis for Lithium Ion Intercalation into
Graphitized Carbons. Journal of The Electrochemical Society 2000, 147, 50-58.
51. Holzapfel, M.; Martinent, A.; Alloin, F.; Le Gorrec, B.; Yazami, R.; Montella, C., First lithiation and
charge/discharge cycles of graphite materials, investigated by electrochemical impedance
spectroscopy. Journal of Electroanalytical Chemistry 2003, 546, 41-50.
52. Xu, S.-D.; Zhuang, Q.-C.; Tian, L.-L.; Qin, Y.-P.; Fang, L.; Sun, S.-G., Impedance Spectra of
Nonhomogeneous, Multilayered Porous Composite Graphite Electrodes for Li-Ion Batteries:
Experimental and Theoretical Studies. The Journal of Physical Chemistry C 2011, 115, 9210-9219.
53. Mertens, A.; Vinke, I. C.; Tempel, H.; Kungl, H.; de Haart, L. G. J.; Eichel, R.-A.; Granwehr, J.,
Quantitative Analysis of Time-Domain Supported Electrochemical Impedance Spectroscopy Data of
Li-Ion Batteries: Reliable Activation Energy Determination at Low Frequencies. Journal of The
Electrochemical Society 2016, 163, H521-H527.
54. Wang, X.; Hao, H.; Liu, J.; Huang, T.; Yu, A., A novel method for preparation of macroposous
lithium nickel manganese oxygen as cathode material for lithium ion batteries. Electrochimica Acta
2011, 56, 4065-4069.
55. Adams, R. A.; Syu, J.-M.; Zhao, Y.; Lo, C.-T.; Varma, A.; Pol, V. G., Binder-Free N- and O-Rich Carbon
Nanofiber Anodes for Long Cycle Life K-Ion Batteries. Acs Applied Materials & Interfaces 2017, 9,
17872-17881.
 Graphical abstract

Few-layer nitrogen-doped graphene derived from Dicyandiamide and Coal tar pitch

demonstrate ultrahigh rate capability and ultralong cycle life as anode for

potassium-ion batteries.