Hydrothermal synthesis and activation of graphene-incorporated nitrogen-rich carbon composite for high-performance supercapacitors

C A R B O N 7 0 ( 2 0 1 4 ) 1 3 0 – 1 4 1

.sc iencedi rect .com

ScienceDirect

journal homepage: www.elsev ier .com/ locate /carbon

Hydrothermal synthesis and activation ofgraphene-incorporated nitrogen-rich carboncomposite for high-performance supercapacitors

0008-6223/$ - see front matter � 2014 Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.carbon.2013.12.081

* Corresponding author.E-mail address: [email protected] (J. Qiu).

Xiaoming Fan, Chang Yu, Juan Yang, Zheng Ling, Jieshan Qiu *

Carbon Research Laboratory, Liaoning Key Lab for Energy Materials and Chemical Engineering, State Key Lab of Fine Chemicals,

Dalian University of Technology, Dalian 116024, PR China

A R T I C L E I N F O

Article history:

Received 3 October 2013

Accepted 25 December 2013

Available online 8 January 2014

A B S T R A C T

Graphene-incorporated nitrogen-rich carbon composite with nitrogen content of ca.

10 wt.% has been synthesized by an effective yet simple hydrothermal reaction of glucosa-

mine in the presence of graphene oxide (GO). The nitrogen content of carbon composite is

nearly twice as high as that of hydrothermal carbon without graphene. GO is favorable for

the high nitrogen doping in the carbon composite by the reaction between the glucosa-

mine-released ammonia and GO. The hydrothermal carbon composite is further activated

by KOH, and graphene in the activated carbon composite demonstrates a positive effect of

increasing specific surface area, pore volume and electrical conductivity, resulting in supe-

rior electrochemical performance. The activated carbon composite with higher specific sur-

face area and micropore volume possesses higher specific capacitance with a value of

300 F g�1 at 0.1 A g�1 in 6 M KOH aqueous solution in the two electrode cell. Larger meso-

pore volume and higher conductivity of the activated carbon composite will provide fast

ion and electron transfer, thus leading to higher rate capacity with a capacitance retention

of 76% at 8 A g�1 in comparison to the activated hydrothermal carbon without graphene.

� 2014 Elsevier Ltd. All rights reserved.

1. Introduction

Carbon electrode materials for supercapacitors have been

studied extensively in past decades due to their low cost, high

surface area, excellent electrochemical stability and electrical

conductivity [1–3]. Capacitive behavior on the carbon based

supercapacitors can be mainly classified into electrical dou-

ble-layer capacitance, which arises from the electrostatic

attraction between ions and the charged surface of carbon

electrode. However, these pure carbon-based electrical

double-layer capacitors (EDLC) cannot meet the requirements

in high energy storage field. An efficient approach to solve

this problem is to incorporate heteroatoms into carbon

materials to enhance the capacitance, in which additional

contributions derived from pseudocapacitance associated

with Faradaic charge transfer reactions between the heteroat-

oms and the electrolytes are present. It has been found that

heteroatom dopants such as oxygen, nitrogen, boron, phos-

phorus and sulfur can improve the capacitance to a certain

degree [4–12]. Of these available dopants, nitrogen doping

has attracted much attention and shows a significant

improvement for higher energy storage ability. Nitrogen spe-

cies in the carbon materials can not only act as an electron

donor due to its five valence electrons which result in a

shift of the Fermi level to the valence band but also introduce

functional groups containing nitrogen and oxygen atoms

[13–16]. Moreover, the combined and positive effects of nitro-

gen and oxygen containing groups on the electrochemical

C A R B O N 7 0 ( 2 0 1 4 ) 1 3 0 – 1 4 1 131

performance are also present in the microporous activated

carbon electrode for supercapacitors [5].

Functional carbon materials can be synthesized using vari-

ous carbon-containing sources such as hydrocarbon, polymer,

biomass and so on. From the point of view of sustainable chem-

istry, biomass and its derivatives are considered as one of highly

potential candidates for the production of functional carbon

materials. To convert biomass to nanostructured carbon, hydro-

thermal carbonization has been considered as an efficient

method operated in an environmentally friendly process

[17,18]. Nitrogen-containing hydrothermal carbons (HTCs) have

been prepared from different types of biomasses or derivatives

such as chitin, chitosan, glucosamine (GA), microalgae and so

on [19–21]. Moreover, the nitrogen-free biomass and nitrogen-

containing molecules such as ovalbumin, glycine, and ethy-

lenediamine are also employed to fabricate nitrogen-containing

HTCs [22–24]. It has also been demonstrated that HTCs are rich

in oxygen groups and the oxygen content can be tuned by dif-

ferent operating conditions. With this in mind, HTCs have been

considered as a promising candidate of electrode materials for

supercapacitors. Titirici et al. recently found that GA-derived

nitrogen-containing HTCs chemically activated by KOH exhib-

ited superior performance in supercapacitors [25]. Interestingly,

HTCs with the networks of uniformly distributed oxygen-con-

taining groups can be efficiently transformed into microporous

carbons with the optimum pore size distribution by chemically

activation. The activated carbon materials from natural poly-

saccharides-derived HTCs exhibited excellent potential for sup-

ercapacitor applications with rather high specific capacitance

and rapid charging–discharging capability [26].

Graphene is a two dimensional carbon material with

high surface area, electrical conductivity, mechanical

strength and chemical stability, and considered as a prom-

ising candidate of electrode materials for supercapacitors

[27,28]. Graphene has also been used as an additive in

the activated carbon, mesoporous carbon, carbon nanofiber

to improve their electrochemical performance in superca-

pacitors [29–32]. Chen et al. also reported porous graph-

ene-based bulk materials produced by KOH activation of

the sucrose-derived HTCs and graphene composite. Graph-

ene in the graphene-based bulk materials is in favor of

increasing their surface area and electrical conductivity,

resulting in superior supercapacitve performance in the io-

nic liquid electrolyte [33]. Nevertheless, few work has been

reported to investigate the role of graphene oxide (GO) as

an additive in the hydrothermal reaction of nitrogen-con-

taining carbohydrate.

Herein, we report an effective yet simple hydrothermal

method to fabricate nitrogen-rich graphene-incorporated

HTC (GHTC) using GA as a nitrogen-containing carbohy-

drate precursor (Fig. S1). GO is in favor of increasing the

nitrogen content of the GHTC (ca. 10 wt.%) to a great degree

in the hydrothermal process. The HTC and GHTC were fur-

ther activated by KOH, and the electrochemical perfor-

mance of the activated products as electrode materials

for supercapacitors was investigated. It was found that

graphene demonstrates a positive effect of increasing the

surface area, pore volume and electrical conductivity of

the activated GHTC, leading to high specific capacitance

and rate capacity.

2. Experimental

2.1. Sample preparation

2.1.1. Synthesis of the HTC and GHTCGO was prepared according a modified Hummers method

which has been described in the previous reports [34–36].

The GHTC composite was synthesized by a hydrothermal

method, the details of which are described below. A 30 mL

GO solution (3 mg mL�1) was mixed with a solution which

contained 8 g GA hydrochloride (C6H13NO5ÆHCl) dissolved in

60 mL deionized water. The pH of the reaction system was ad-

justed to 11 with NaOH aqueous solution to obtain well-

dispersed mixture [37], and then the mixture was transferred

to a Teflon autoclave and treated at 180 �C for 12 h. After cool-

ing, the product was washed with deionized water and dried

at 80 �C. For comparison, the nitrogen-doped HTC without

graphene was also prepared by directly hydrothermal treat-

ment of GA at 180 �C for 12 h.

2.1.2. Activation of the HTC and GHTC by KOHThe HTC and GHTC samples were activated by KOH at 600 �Cfor 4 h in a nitrogen atmosphere with a mass ratio of KOH to

carbonaceous materials 2:1. The as-obtained products were

then treated with diluted hydrochloric acid and deionized

water and then dried at 80 �C overnight. The activated sam-

ples were denoted as HTC-a and GHTC-a, respectively.

2.2. Characterization methods

Morphologies and microstructures of samples before and

after KOH activation were observed by field emission scan-

ning electron microscopy (FESEM) (FEI NOVA NanoSEM 450)

at 3 kV and transmission electron microscopy (TEM) (FEI Tec-

nai G220 and TF30). Specific surface area and pore size distri-

bution were calculated based on the data recorded on

physical sorption instrument (Micromeritics ASAP 2020).

The samples were degassed at 150 �C prior to nitrogen sorp-

tion measurement. Element compositions were measured

using Elementar varioEL III. The data of X-ray photoelectron

spectroscopy were collected on Thermo ESCALAB 250. The

electrical conductivities were calculated according the results

measured by the four-probe method using a S-2A four-probe

station and Keithley 2400 Sourcemeter.

2.3. Electrochemical measurements

Active materials and binder poly(tetrafluoroethylene) with

the mass ratio of 95:5 pressed onto the nickel foam were used

as the working electrode. Hg/HgO electrode and Pt foil elec-

trode were used as reference electrode and counter electrode,

respectively. Working electrodes mentioned above with the

same mass were symmetrically assembled to two electrode

cell. Cyclic voltammetry was carried out using a CHI 660D

(Shanghai Chenhua, China) electrochemical workstation in

a three electrode cell with 6 M KOH as electrolyte solution.

Galvanostatic charge–discharge profiles and electrochemical

impedance spectroscopy were also recorded in a two elec-

trode cell with 6 M KOH as electrolyte solution using a CHI

132 C A R B O N 7 0 ( 2 0 1 4 ) 1 3 0 – 1 4 1

660D electrochemical workstation. Gravimetric specific

capacitances of the cells Ccell were calculated at different cur-

rent densities according to galvanostatic charge/discharge

profiles based on the following equations:

Ccell ¼I

mtotal � ðdv=dtÞ ð1Þ

where Ccell is the specific capacitance of the two electrode

cell, I is the discharge current, dv/dt is the average slope of

the discharge curve, and mtotal is the total mass of active

materials in the two electrode cell. The specific capacitance

of single electrode could be obtained by multiplying the Ccell

by four. Energy density and power density of the cells were

calculated based on the following equations:

Ecell ¼Ccell � DV2 � 1000

2� 3600ð2Þ

Pcell ¼Ecell

Dt� 3600 ð3Þ

where Ecell and Pcell are the energy density and power density

of two electrode cell, respectively, DV is the voltage window

from the end of IR drop to the end of discharge and Dt is

the discharge time. Galvanostatic cycling was performed by

1000 cycles at the current density of 1 A g�1.

3. Results and discussion

3.1. Enhanced nitrogen doping in GHTC composite

Great changes occur during the hydrothermal reaction of GA

in the presence of a small amount of GO, in which GA was

transformed to carbonaceous materials and GO was reduced

to graphene simultaneously [20,38]. The typical SEM images

Fig. 1 – SEM images of (a, b) HTC and (c, d) GHTC. (A

of the HTC and GHTC are shown in Fig. 1. The HTC (Fig. 1a

and b) has a compact network assembled by smaller carbon

particles and displays a rough surface. In the case of the

GHTC, a relatively smoother surface can be observed and

the graphene layers cover onto the surface of the HTC parti-

cles, of which the edge can be clearly seen in the ellipse

shown in Fig. 1d. The further TEM images in Fig. 2a and b re-

veal that the HTC from GA is consisted of agglomerated smal-

ler carbon particles with an amorphous structure, which is

consistent with the SEM results. For the GHTC, the graphene

layers curl up to some degree and are uniformly dispersed in

the resulting carbon composite, which are marked in Fig. 2c.

The compositions of the HTC and GHTC were firstly deter-

mined by the element analysis (Table S1). The two hydrother-

mal products both have relatively low carbon content and

high oxygen content, which is similar to the results reported

in the literatures [20,39]. It is interesting that the weight con-

tent of nitrogen in the GHTC nearly reaches 10.0 wt.%, being

an increase of twice in comparison to that in the HTC. The

nitrogen content in the GHTC remains a high value up to

10.1 wt.% even if the sample is further treated at 600 �C for

4 h under the nitrogen atmosphere. It is also noted that the

carbon content also increases while the oxygen content de-

creases a lot after heat treatment. This indicates that carbo-

naceous material with high and stable nitrogen content can

be achieved through a ‘‘GO assisted’’ hydrothermal reaction

of nitrogen-containing carbohydrates.

The content of the element C, O, N on the surface of the

HTC and GHTC was also evaluated using X-ray photoelectron

spectroscopy (XPS) and the detailed information is summa-

rized in Table 1. It is very surprising that the atom content

of N in the GHTC is 3.5 times as much as that in the HTC from

the XPS results. The N/C ratio of the GHTC also increases by

color version of this figure can be viewed online.)

Table 1 – Surface compositions of the samples before and after KOH activation determined by XPS.

Sample C/at.% N/at.% O/at.% N/C O/C

HTC 79.1 1.8 19.1 0.023 0.24GHTC 77.1 6.3 16.7 0.082 0.22HTC-a 85.9 2.3 11.8 0.027 0.14GHTC-a 86.0 2.2 11.8 0.026 0.14

Fig. 2 – TEM images of (a, b) HTC and (c, d) GHTC. (A color version of this figure can be viewed online.)

C A R B O N 7 0 ( 2 0 1 4 ) 1 3 0 – 1 4 1 133

3.5 times but the O/C ratio decreases slightly. This indicates

that the reaction pathway corresponding to nitrogen-contain-

ing functional groups may change to some degree in the pres-

ence of GO during the hydrothermal process, which will be

further discussed in the following section. Further informa-

tion about the changes of surface chemistry has been ob-

tained from the deconvoluted high-resolution XPS spectra of

three interesting regions (C, O, and N) (Fig. 3 and Fig. S2).

The C1s spectra of the HTC and GHTC in Fig. 3a and b can

be deconvoluted into five components, and the carbon spe-

cies correspond to carbide carbon (283.9 eV); graphite carbon

(284.6 eV); carbon in C–O–C or R–OH (285.3 eV); carbon in

C@O (286.2 eV); and carbon in COOH or C(O)–O–C (288.5 eV)

(Fig. 3a and b) [40,41]. No obvious differences are observed

in the carbon components of the two samples. However, there

exist the differences between the O1s spectra of the HTC and

GHTC to some extent. The oxygen species of the HTC repre-

sent oxygen double bonded to carbon (C@O) (531.8 eV),

oxygen single bonded to carbon in C–O (532.8 eV) and oxygen

in hydroxyl OH (533.7 eV), while the O1s region of the GHTC

contains three components, corresponding to quinone

(530.7 eV), C@O (531.8 eV) and C–O (532.9 eV) (Fig. 3c and d)

[5,41]. Great changes of nitrogen components in the HTC

and GHTC are very obvious, which is in agreement with ele-

ment analysis results. The nitrogen content of the HTC on

the surface is 1.8 at.%, and the N1s peak shows four compo-

nents at 398.4, 399.5, 400.3 and 401.3 eV, which can be as-

cribed to pyridinic (N-6, 0.7 at.%), amine (NH2, 0.5 at.%),

pyrrolic/pyridone (N-5, 0.3 at.%) and quaternary (N-Q,

0.3 at.%) nitrogen, respectively (Fig. 3c and d) [5,13]. In the

case of the GHTC, the surface nitrogen content increases up

to 6.3 at.%. The nitrogen species include N-6 (2.0 at.%), NH2

(2.5 at.%), and N-5 (1.7 at.%) (Fig. 3e and f), nevertheless, the

characteristic peak of the N-Q nitrogen disappears. Based

Fig. 3 – Deconvoluted high-resolution (a, b) C1s, (c, d) O1s and (e, f) N1s spectra of the HTC and GHTC. (A color version of this

figure can be viewed online.)

134 C A R B O N 7 0 ( 2 0 1 4 ) 1 3 0 – 1 4 1

on the above results, we conclude that GO will provide a new

possibility to increase the nitrogen doping in the HTC com-

posite. GO is favorable for the nitrogen fixation and the nitro-

gen-doping enhancement in the HTC composite with

increasing the content of pyridinic, amine and pyrrolic nitro-

gen (Fig. 3e and f).

The possible mechanism of the enhanced nitrogen doping

in the GHTC composite has been proposed here. The degrada-

tion of GA in water at high temperature has been investigated

in many literatures, in which GA is transformed into hydroxy-

methyl furfural (HMF) and releases ammonia in solution

simultaneously (Fig. 4a); subsequently, ammonia reacts with

HMF to form different type of nitrogen-containing molecules

(Fig. 4b (1)) [20,42,43]. When GO was added in the hydrother-

mal reaction of GA, new reaction between GO and the re-

leased ammonia from GA would occur simultaneously in

the system. Gao et al. reported that the urea decomposes

and releases ammonia, and the released ammonia reacts

with the oxygen functional groups (carboxylic acid and epox-

ide) on the GO surface to form the intermediates (amide and

amine) under the hydrothermal conditions. These nitrogen-

containing intermediates can further pass through the reac-

tion of dehydration (pyridine or pyridone) or decarbonylation

(pyrrole) to form more stable structures (Fig. 4b (2)) [44]. That

is to say, the competitive reaction for ammonia would occur

between GO and the GA-released HMF, which may also

change the elemental composition of the final product in

the system. Furthermore, it is worth pointing out that HMF

is the main reactive compound in the hydrothermal process

resulting in the formation of carbonaceous material with

the furane rings, which is abundant in the final nitrogen-

doped product (Fig. 4b (3)) [23,45]. On the basis of the discus-

sion mentioned above, the released ammonia from GA under-

goes two reactions simultaneously in which ammonia reacts

with HMF and GO, and the possible mechanism is proposed

and illustrated in Fig. 4c.

3.2. KOH activation of HTC and GHTC

KOH activation is a general method applied to increase the spe-

cific area and pore volume in carbon materials, which contrib-

utes to obtaining electrode materials with high capacitance in

Fig. 4 – (a) Degradation reaction of GA to HMF and ammonia. (b) (1, 2) Possible reactions of HMF and GO for ammonia; (3) Main

reaction of HMF transformed into carbonaceous material with furanic structure. (c) Schematic representation of the

hydrothermal progress of HMF, GO and ammonia to fabricate graphene-incorporated nitrogen-doped carbon composite. (A

color version of this figure can be viewed online.)

C A R B O N 7 0 ( 2 0 1 4 ) 1 3 0 – 1 4 1 135

supercapacitor application. The HTC and GHTC were acti-

vated by KOH with a mass ratio of KOH to the carbonaceous

materials 2:1 at 600 �C for 4 h to obtain the products HTC-a

and GHTC-a. The activation mechanism follows several

simultaneous/consecutive reactions of carbon and KOH be-

low 700 �C [46]. Fig. 5 shows the morphologies of the HTC-a

and GHTC-a, in which the two samples both exhibit the typ-

ical morphologies of activated carbon composed of the bro-

ken carbon particles. The differences that can be observed

in the GHTC-a are that the graphene layers are incorporated

in the carbon matrix and cover onto the carbon particles

(marked by the ellipse in Fig. 5d). This is also why the

GHTC-a has a high conductivity, nearly being three orders of

magnitude in comparison to that of the HTC-a (Table 2). Fur-

ther TEM examination of the GHTC-a reveals that the microp-

ores with a very small diameter are present in the carbon

particle in Fig. 6b, which will provide the possibility for

enhancing the capacitance of carbon composite, while the

graphene layers in the carbon composite are wrinkled and

stacked with multilayers (marked by the ellipse in Fig. 6d).

The porous structures of the HTC-a and GHTC-a were ana-

lyzed by nitrogen sorption technique and their nitrogen

adsorption/desorption isotherms are shown in Fig. 7a. The

two samples both exhibit a combined characteristics of type

I/IV isotherms, indicating that the samples contain microp-

ores and mesopores. Pore size distributions of the two sam-

ples are obtained using the density functional theory (DFT)

method, which are shown in Fig. 7b. It can be seen that most

of the pore volumes of the two samples arise from microp-

ores. Compared to the HTC-a, the GHTC-a contains a large

volume of micropores with pore size below 0.7 nm that is con-

sidered as very active for double layer capacitance [47]. More-

over, the micropore size of the HTC-a concentrates in the pore

size region below 1 nm, whereas the GHTC-a contains more

pores with the size larger than 1 nm including mesopores.

This will be a positive factor for rate capability of the GHTC-

a in the supercapacitor application [48]. Pore volume and

Brunauer–Emmett–Teller (BET) specific surface area are calcu-

lated from the isotherms data and the results are shown in

Table 2. It is noted that specific surface area, total pore

Fig. 5 – SEM images of (a, b) HTC-a and (c, d) GHTC-a. (A color version of this figure can be viewed online.)

Fig. 6 – TEM images of the GHTC-a. (A color version of this figure can be viewed online.)

Table 2 – Porous properties and electrical conductivity values of the HTC-a and GHTC-a.

Sample SBET/m2 g�1 Vta/cm3 g�1 Vm

b/cm3 g�1 Conductivity/S m�1

HTC-a 1147 0.60 0.37 0.013GHTC-a 1646 0.82 0.54 15.1a Single point total pore volume (Vt) from adsorption isotherms at P/P0 � 0.99.b Micropore volume (Vm) calculated using the DFT method.

136 C A R B O N 7 0 ( 2 0 1 4 ) 1 3 0 – 1 4 1

Fig. 7 – (a) Nitrogen adsorption/desorption isotherms and (b)

pore size distributions of the HTC-a and GHTC-a. (A color

version of this figure can be viewed online.)

C A R B O N 7 0 ( 2 0 1 4 ) 1 3 0 – 1 4 1 137

volume and micropore volume of the GHTC-a are higher than

that of the HTC-a, demonstrating that graphene has a positive

effect on the enhancement of surface area and porosity of the

activated carbon composite. Nevertheless, the graphene-in-

duced enhancement of specific surface area and porosity de-

pends on the content and existence form of graphene in

carbon composite [30,33].

The surface chemical properties of the activated carbon

materials are of importance to their supercapacitive behav-

iors. The changes of the surface compositions in the as-pre-

pared KOH-activated carbon materials obtained from XPS

measurement are also shown in Table 1. It can be seen that

the values of N/C and O/C ratios in the two activated carbon

materials are nearly equal, demonstrating the similar surface

chemical compositions, which can be further confirmed by

the detailed high-resolution XPS spectra of each element.

Fig. 8 shows the deconvoluted high-resolution C1s, O1s and

N1s XPS spectra of the HTC-a and GHTC-a, and the details

of each identified component of the two samples are shown

in Fig. S3. Compared to the HTC-a, the GHTC-a possesses

more than 18% of sp2 carbon–carbon bonds corresponding

to the graphite carbon at 284.6 eV, which can be ascribed to

the presence of graphene layers. This is also why the GHTC-

a has an enhanced electrical conductivity. Moreover, the oxy-

gen species of the two samples only have a slight difference

in the percentage of each component. An obvious change of

surface chemical compositions for the samples before and

after the process of KOH activation is that the two activated

samples have almost the same nitrogen species, although

the nitrogen content of the GHTC is 3.5 times higher than that

of the HTC. The reason for this could be that the nitrogen

components in the GHTC bonded to the edge carbon atoms

of graphene layers may be etched more easily in the KOH acti-

vation process. How to immobilize nitrogen during the KOH

activation process needs to be further investigated in the fu-

ture. Nevertheless, the similar surface chemistry leads one

to believe that the supercapacitve performances between

the activated HTC-a and GHTC-a should be governed by the

improved physical properties such as pore structure and elec-

trical conductivity.

3.3. Electrochemical behaviors of HTC-a and GHTC-a

The electrochemical performances of the HTC-a and GHTC-a

were analyzed using the technologies of cyclic voltammetry,

galvanostatic charge–discharge and electrochemical imped-

ance spectroscopy in the 6 M KOH aqueous solution. Fig. 9

illustrates cyclic voltammograms (CVs) of the HTC-a and

GHTC-a measured in the three-electrode cell at different scan

rates. Both the two samples exhibit a typical capacitive

behavior with quasi-rectangular shaped voltammetry charac-

teristics as well as broad humps in the CVs at a low scan rate

of 5 mV s�1, indicating that the capacitance results from the

combination of electrical double-layer capacitance and

pseudocapacitance related to redox reactions of the heteroat-

oms in the electrode materials. The redox peaks observed

more obviously in acidic electrolyte in the CVs (Fig. S4) further

reveal the presence of the pseudocapacitance. Moreover, for

the GHTC-a containing graphene sheets in the carbon matrix,

a faster electrochemical response can be observed, revealing

that graphene in the GHTC-a reduces the resistance of the

electrode materials to achieve faster current response, and in-

duces the GHTC-a to possess a better EDLC behavior. It is also

noted that the CVs of the GHTC-a maintain more rectangular

shape than those of the HTC-a with increasing the scan rate;

meanwhile, the specific capacitances of the GHTC-a remain a

higher value according to the areas of the CVs.

It is helpful for determining the intrinsic electrochemical

characteristics of an electrode material in the three-electrode

system. However, to reproduce the physical configuration,

internal voltages, and charge transfer occurred in a packaged

supercapacitor, a two-electrode cell test is preferred to pro-

vide the best indication of the electrochemical performance

[49]. In this respect, galvanostatic charge–discharge technique

was performed in the two-electrode cell symmetrically

assembled by the HTC-a or GHTC-a electrodes to further

investigate their capacitive behavior and cycling stability,

the results of which are shown in Fig. 10. The charge–dis-

charge profiles at different current loads from 0.2 to 2 A g�1

are shown in Fig. 10a and b. Unlike linear characteristics,

the deviations from the linear curves can be observed in the

charge–discharge tests of the HTC-a and GHTC-a, indicating

the presence of redox reactions during the charge–discharge

process [50]. This is consistent with the results of cyclic vol-

tammetry. It can be also found that the Ohmic drop of the

GHTC-a electrode is smaller than that of the HTC-a electrode,

proving the higher electrical conductivity of the GHTC-a,

which is in agreement with the results shown in Table 2.

The specific capacitances of the two samples calculated from

the results of charge–discharge profiles are shown in Fig. 10c.

The GHTC-a possesses a specific capacitance of 300 F g�1 at

Fig. 8 – Deconvoluted high-resolution (a, b) C1s, (c, d) O1s and (e, f) N1s spectra of the HTC-a and GHTC-a. (A color version of

this figure can be viewed online.)

Fig. 9 – Cyclic voltammograms of the (a) HTC-a and (b) GHTC-a at different scan rates of 5, 10, 20, 50 and 100 mV s�1 in the

three-electrode cell. (A color version of this figure can be viewed online.)

138 C A R B O N 7 0 ( 2 0 1 4 ) 1 3 0 – 1 4 1

0.1 A g�1 in the 6 M KOH aqueous solution, the value of which

is 20% higher than that of the HTC-a. This can be attributed to

the fact that the GHTC-a has larger specific area and micro-

pore volume than HTC-a since the GHTC-a and HTC-a have

the similar surface chemical properties to eliminate the im-

pact of surface functional groups. Moreover, the GHTC-a re-

tains higher capacitance retention of 76% (228 F g�1 at

8 A g�1) in comparison to the HTC-a (88 F g�1 at 8 A g�1). The

higher mesopore volume and electrical conductivity of the

GHTC-a will provide a faster ions and electron migration in

Fig. 10 – Galvanostatic charge–discharge curves of the (a) HTC-a and (b) GHTC-a in the two-electrode cell; (c) Specific

capacitances of the HTC-a and GHTC-a at various current densities measured in the two-electrode cell (the results were

calculated from the specific capacitance of the cell multiplied by four); (d) Capacitance retention of the GHTC-a measured at

1 A g�1 in the two-electrode cell. (A color version of this figure can be viewed online.)

C A R B O N 7 0 ( 2 0 1 4 ) 1 3 0 – 1 4 1 139

the charge–discharge process resulting in a higher capaci-

tance retention. The GHTC-a also shows good cycling

stability, exhibiting a 90% capacitance retention when cycling

1000 times at 1 A g-1 (Fig. 10d).

Fig. 11a shows the electrochemical impedance spectros-

copy measured in the range from 0.01 to 100 kHz in the

two-electrode cell. The equivalent series resistance of the

GHTC-a is slightly smaller than that of the HTC-a. The semi-

circle of the Nyquist plot in the high-frequency range is asso-

ciated with the surface properties of the porous electrode and

corresponds to the charge transfer resistance. The smaller

diameter of semicircle corresponds to the lower charge trans-

fer resistance. The diameter of semicircle in the Nyquist plot

of GHTC-a is much smaller than that of HTC-a, indicating that

the GHTC-a electrode containing graphene sheets can offer

Fig. 11 – (a) Nyquist plots of the HTC-a and GHTC-a measured in

of specific capacitances with frequency for the HTC-a and GHTC

figure can be viewed online.)

faster charge transportation, which can be contributed to

the improved pore structure and electrical conductivity as

discussed above. The line of the Nyquist plot in the low-fre-

quency range is related to the capacitive behavior of the elec-

trode. Compared to the HTC-a, a nearly vertical line can be

observed on the Nyquist plot of the GHTC-a electrode, which

reflects a better supercapacitive behavior. The relationship

between the imaginary part of gravimetric capacitance and

frequency is also shown in Fig. 11b, which is calculated from

the results of electrochemical impedance spectroscopy

according to the method reported by Gogotsi et al. [51]. The

imaginary part of capacitance reaches the maximum at the

frequency f0, the reciprocal of which yields a time constant,

s, that is a quantitative measure of how fast the device can

be charged and discharged reversibly. The GHTC-a has a time

the two-electrode cell; (b) The correlation of imaginary parts

-a resulted from the Nyquist plots. (A color version of this

Fig. 12 – Ragone plots of the HTC-a and GHTC-a measured in

the two-electrode cell. (A color version of this figure can be

viewed online.)

140 C A R B O N 7 0 ( 2 0 1 4 ) 1 3 0 – 1 4 1

constant s of 15 s, which is much smaller than that of the

HTC-a (48 s). This implies that the GHTC-a exhibit a faster fre-

quency response and a better charge–discharge behavior at a

high current load. Fig. 12 demonstrates the relationship be-

tween the power density and energy density measured in

the two-electrode cell symmetrically assembled by the HTC-

a or GHTC-a electrode. The energy density of the GHTC-a

reaches 8.4 W h kg�1 at a power density of 22.5 W kg�1, being

higher than that of the HTC-a, and the GHTC-a maintains a

much higher energy density than the HTC-a when the power

density increases to the relatively high value.

4. Conclusions

Graphene-incorporated nitrogen-rich carbon composite has

been successfully fabricated from the mixture of GA and GO

by an effective yet simple hydrothermal method. The mor-

phologies, microstructures and surface chemistry of the as-

made products are governed by the existing graphene. The

presence of GO in the hydrothermal reaction of GA provides

an additional reaction pathway between the ammonia and

GO, resulting in the enhancement of nitrogen doping in the

GHTC. After KOH activation, the GHTC-a exhibits higher spe-

cific capacitance with a value of 300 F g�1 at 0.1 A g�1 and rate

capacity with a capacitance retention of 76% at 8 A g�1 in

comparison to the HTC-a because of the larger surface area,

enhanced mesopore volume and electrical conductivity of

the GHTC-a. In general, graphene-incorporated carbon com-

posite from low-cost biomass by an effective yet simple

hydrothermal method coupled with KOH activation will pro-

vide a promising candidate to gain high-performance acti-

vated carbon material for the applications in supercapacitor

and other fields.

Acknowledgements

This work was partly supported by the National Natural Sci-

ence Foundation of China (Nos. 20923006, 21336001 and

U1203292) and Dalian Municipal Science & Technology Project

of China (No. 2011A15GX023).

Appendix A. Supplementary data

Supplementary data associated with this article can be found,

in the online version, at http://dx.doi.org/10.1016/j.carbon.

2013.12.081.

R E F E R E N C E S

[1] Ghosh A, Lee YH. Carbon-based electrochemical capacitors.ChemSusChem 2012;5(3):480–99.

[2] Simon P, Gogotsi Y. Capacitive energy storage innanostructured carbon-electrolyte systems. Acc Chem Res2013;46(5):1094–103.

[3] Zhai Y, Dou Y, Zhao D, Fulvio PF, Mayes RT, Dai S. Carbonmaterials for chemical capacitive energy storage. Adv Mater2011;23(42):4828–50.

[4] Lota G, Grzyb B, Machnikowska H, Machnikowski J,Frackowiak E. Effect of nitrogen in carbon electrode on thesupercapacitor performance. Chem Phys Lett 2005;404(1–3):53–8.

[5] Hulicova-Jurcakova D, Seredych M, Lu GQ, Bandosz TJ.Combined effect of nitrogen- and oxygen-containingfunctional groups of microporous activated carbon on itselectrochemical performance in supercapacitors. Adv FunctMater 2009;19(3):438–47.

[6] Wang D-W, Li F, Chen Z-G, Lu GQ, Cheng H-M. Synthesis andelectrochemical property of boron-doped mesoporous carbonin supercapacitor. Chem Mater 2008;20(22):7195–200.

[7] Guo H, Gao Q. Boron and nitrogen co-doped porous carbonand its enhanced properties as supercapacitor. J PowerSources 2009;186(2):551–6.

[8] Iyyamperumal Eswaramoorthi, Wang S, Dai L. Verticallyaligned BCN nanotubes with high capacitance. ACS Nano2012;6(6):5259–65.

[9] Wu ZS, Winter A, Chen L, Sun Y, Turchanin A, Feng X, et al.Three-dimensional nitrogen and boron co-doped graphenefor high-performance all-solid-state supercapacitors. AdvMater 2012;24(37):5130–5.

[10] Zhao X, Wang A, Yan J, Sun G, Sun L, Zhang T. Synthesis andelectrochemical performance of heteroatom-incorporatedordered mesoporous carbons. Chem Mater2010;22(19):5463–73.

[11] Tsubota T, Takenaka K, Murakami N, Ohno T. Performance ofnitrogen- and sulfur-containing carbon material derivedfrom thiourea and formaldehyde as electrochemicalcapacitor. J Power Sources 2011;196(23):10455–60.

[12] Hulicova-Jurcakova D, Kodama M, Shiraishi S, Hatori H, ZhuZH, Lu GQ. Nitrogen-enriched nonporous carbon electrodeswith extraordinary supercapacitance. Adv Funct Mater2009;19(11):1800–9.

[13] Wang D-W, Li F, Yin L-C, Lu X, Chen Z-G, Gentle IR, et al.Nitrogen-doped carbon monolith for alkalinesupercapacitors and understanding nitrogen-induced redoxtransitions. Chem Eur J 2012;18(17):5345–51.

[14] Seredych M, Hulicova-Jurcakova D, Lu GQ, Bandosz TJ.Surface functional groups of carbons and the effects of theirchemical character, density and accessibility to ions onelectrochemical performance. Carbon 2008;46(11):1475–88.

[15] Hulicova-Jurcakova D, Seredych M, Jin Y, Lu GQ, Bandosz TJ.Specific anion and cation capacitance in porous carbonblacks. Carbon 2010;48(6):1767–78.

[16] Hulicova-Jurcakova D, Fiset E, Lu GQ, Bandosz TJ. Changes insurface chemistry of carbon materials upon electrochemical

C A R B O N 7 0 ( 2 0 1 4 ) 1 3 0 – 1 4 1 141

measurements and their effects on capacitance in acidic andneutral electrolytes. ChemSusChem 2012;5(11):2188–99.

[17] Hu B, Wang K, Wu L, Yu SH, Antonietti M, Titirici M-M.Engineering carbon materials from the hydrothermalcarbonization process of biomass. Adv Mater2010;22(7):813–28.

[18] Titirici M-M, Antonietti M. Chemistry and materials optionsof sustainable carbon materials made by hydrothermalcarbonization. Chem Soc Rev 2010;39(1):103–16.

[19] White RJ, Antonietti M, Titirici M-M. Naturally inspirednitrogen doped porous carbon. J Mater Chem2009;19(45):8645–50.

[20] Zhao L, Baccile N, Gross S, Zhang Y, Wei W, Sun Y, et al.Sustainable nitrogen-doped carbonaceous materials frombiomass derivatives. Carbon 2010;48(13):3778–87.

[21] Falco C, Sevilla M, White RJ, Rothe R, Titirici M-M. Renewablenitrogen-doped hydrothermal carbons derived frommicroalgae. ChemSusChem 2012;5(9):1834–40.

[22] Baccile N, Antonietti M, Titirici M-M. One-step hydrothermalsynthesis of nitrogen-doped nanocarbons: albuminedirecting the carbonization of glucose. ChemSusChem2010;3(2):246–53.

[23] Baccile N, Laurent G, Coelho C, Babonneau F, Zhao L, TitiriciM-M. Structural insights on nitrogen-containinghydrothermal carbon using solid-state magic angle spinning13C and 15N nuclear magnetic resonance. J Phys Chem C2011;115(18):8976–82.

[24] Liu Z, Zhang C, Luo L, Chang Z, Sun X. One-pot synthesis andcatalyst support application of mesoporous N-dopedcarbonaceous materials. J Mater Chem 2012;22(24):12149–54.

[25] Zhao L, Fan LZ, Zhou MQ, Guan H, Qiao S, Antonietti M, et al.Nitrogen-containing hydrothermal carbons with superiorperformance in supercapacitors. Adv Mater2010;22(45):5202–6.

[26] Wei L, Sevilla M, Fuertes AB, Mokaya R, Yushin G.Hydrothermal carbonization of abundant renewable naturalorganic chemicals for high-performance supercapacitorelectrodes. Adv Energy Mater 2011;1(3):356–61.

[27] Zhu Y, Murali S, Stoller MD, Ganesh KJ, Cai W, Ferreira PJ,et al. Carbon-based supercapacitors produced by activationof graphene. Science 2011;332(6037):1537–41.

[28] El-Kady MF, Strong V, Dubin S, Kaner RB. Laser scribing ofhigh-performance and flexible graphene-basedelectrochemical capacitors. Science 2012;335(6074):1326–30.

[29] Zhang K, Ang BT, Zhang LL, Zhao XS, Wu J. Pyrolyzedgraphene oxide/resorcinol-formaldehyde resin compositesas high-performance supercapacitor electrodes. J MaterChem 2011;21(8):2663–70.

[30] Seredych M, Chen R, Bandosz TJ. Effects of the addition ofgraphite oxide to the precursor of a nanoporous carbon onthe electrochemical performance of the resultingcarbonaceous composites. Carbon 2012;50(11):4144–4154.

[31] Tai Z, Yan X, Lang J, Xue Q. Enhancement of capacitanceperformance of flexible carbon nanofiber paper by addinggraphene nanosheets. J Power Sources 2012;199:373–8.

[32] Lei Z, Zhang J, Zhao XS. Ultrathin MnO2 nanofibers grown ongraphitic carbon spheres as high-performance asymmetricsupercapacitor electrodes. J Mater Chem 2012;22(1):153–60.

[33] Zhang L, Zhang F, Yang X, Long G, Wu Y, Zhang T, et al.Porous 3D graphene-based bulk materials with exceptionalhigh surface area and excellent conductivity forsupercapacitors. Sci Rep 2013;3:1408.

[34] Hummers Jr WS, Offeman RE. Preparation of graphitic oxide. JAm Chem Soc 1958;80(6):1339.

[35] Stankovich S, Dikin DA, Piner RD, Kohlhaas KA,Kleinhammes A, Jia Y, et al. Synthesis of graphene-basednanosheets via chemical reduction of exfoliated graphiteoxide. Carbon 2007;45(7):1558–65.

[36] Li D, Muller MB, Gilje S, Kaner RB, Wallace GG. Processableaqueous dispersions of graphene nanosheets. NatNanotechnol 2008;3(2):101–5.

[37] Bai Y, Rakhi RB, Chen W, Alshareef HN. Effect of pH-inducedchemical modification of hydrothermally reduced grapheneoxide on supercapacitor performance. J Power Sources2013;233:313–9.

[38] Zhou Y, Bao Q, Tang LAL, Zhong Y, Loh KP. Hydrothermaldehydration for the ‘‘green’’ reduction of exfoliated grapheneoxide to graphene and demonstration of tunable opticallimiting properties. Chem Mater 2009;21(13):2950–6.

[39] Yu L, Falco C, Weber J, White RJ, Howe JY, Titirici M-M.Carbohydrate-derived hydrothermal carbons: a thoroughcharacterization study. Langmuir 2012;28(33):12373–83.

[40] Puziy AM, Poddubnaya OI, Socha RP, Gurgul J, Wisniewski M.XPS and NMR studies of phosphoric acid activated carbons.Carbon 2008;46(15):2113–23.

[41] Arrigo R, Vecker MH, Wrabetz S, Blume R, Lerch M, McGregorJ, et al. Tuning the acid-base properties of nanocarbons byfunctionalization via amination. J Am Chem Soc2010;132(28):9616–30.

[42] Shu C-K. Degradation products formed from glucosamine inwater. J Agric Food Chem 1998;46(3):1129–31.

[43] Jun M, Shao Y, Ho C-T, Koetter U, Lech S. Structuralidentification of nonvolatile dimerization products ofglucosamine by gas chromatography–mass spectrometry,liquid chromatography–mass spectrometry, and nuclearmagnetic resonance analysis. J Agric Food Chem2003;51(21):6340–6.

[44] Sun L, Wang L, Tian C, Tan T, Xie Y, Shi K, et al. Nitrogen-doped graphene with high nitrogen level via a one-stephydrothermal reaction of graphene oxide with urea forsuperior capacitive energy storage. RSC Adv2012;2(10):4498–506.

[45] Falco C, Perez Caballero F, Babonneau F, Gervais C, Laurent G,Titirici M-M, et al. Hydrothermal carbon from biomass:structural differences between hydrothermal and pyrolyzedcarbons via 13C solid state NMR. Langmuir2011;27(23):14460–71.

[46] Otowa T, Tanibata R, Itoh M. Production and adsorptioncharacteristics of MAXSORB: high-surface-area activecarbon. Gas Sep Purif 1993;7(4):241–5.

[47] Seredych M, Koscinski M, Sliwinska-Bartkowiak M, BandoszTJ. Active pore space utilization in nanoporous carbon-basedsupercapacitors: effects of conductivity and poreaccessibility. J Power Sources 2012;220:243–52.

[48] Wang D-W, Li F, Liu M, Lu GQ, Cheng H-M. 3D Aperiodichierarchical porous graphitic carbon material for high-rateelectrochemical capacitive energy storage. Angew Chem IntEd 2007;47(2):373–6.

[49] Stoller MD, Ruoff RS. Best practice methods for determiningan electrode material’s performance for ultracapacitors.Energy Environ Sci 2010;3(9):1294–301.

[50] Gu W, Sevilla M, Magasinski A, Fuertes AB, Yushin G. Sulfur-containing activated carbons with greatly reduced content ofbottle neck pores for double-layer capacitors: a case study forpseudocapacitance detection. Energy Environ Sci2013;6(8):2465–76.

[51] Chmiola J, Yushin G, Dash R, Gogotsi Y. Effect of pore size andsurface area of carbide derived carbons on specificcapacitance. J Power Sources 2006;158(1):765–72.