Chemical Engineering Journal xxx (xxxx) xxxx

Contents lists available at ScienceDirect

Chemical Engineering Journal

journal homepage: www.elsevier.com/locate/cej

A green trace K2CO3 induced catalytic activation strategy for developing

coal-converted activated carbon as advanced candidate for CO2 adsorption
and supercapacitors
⁎ ⁎
Lijie Wanga,b, Fei Suna, , Fei Haoc, Zhibin Qua, Jihui Gaoa, , Mingjun Liua, Kunfang Wanga,
Guangbo Zhaoa, Yukun Qina
a
School of Energy Science and Engineering, Harbin Institute of Technology, Harbin 150001, China
b
Datang Renewable Energy Test and Research Institute Co., Ltd., Beijing 100052, China
c
ENN Graphene Technology Co., Ltd., Langfang 065000, China

H I GH L IG H T S G R A P H I C A L A B S T R A C T

• developed
A green K CO catalytic activation is
2 3
for preparing activated
carbon.
• Potassium based catalytic component
can be regenerated in activation pro-
cess.
• sorption
Ca_AC_1 exhibits excellent CO ad-
capacity and stability.
2

• Ca_AC_1 organic supercapacitor ex-

hibits energy density up to 35.7 Wh/
kg.
• The strategy can be readily amplified
based existing physical activation
craft.

A R T I C LE I N FO A B S T R A C T

Keywords: Aiming to enhancing porosity development and reducing the dosage of activation agents in preparation of ac-
Trace K2CO3 tivated carbon, a novel green and efficient strategy of trace K2CO3 induced catalytic activation was proposed. In
Catalytic activation the design, adding small amount of K2CO3 (less than 2% weight ratio of precursor) can significantly reduce the
Coal converted activated carbons reaction barrier between coal framework and CO2 molecules, thus enhancing pore formation of as-obtained
CO2 adsorption
activated carbons. The resulting Ca_AC-1 has short range ordered microcrystalline structure and developed pore
Supercapacitors
structure, even superior to the activated carbon from chemical activation with large dosage of K2CO3 (three
times weight ratio of carbon precursor). Evaluated as CO2 adsorbent, Ca_AC-1 a high CO2 adsorption capacity
and adsorption-regeneration cycling stability. More importantly, Ca_AC-1 with optimized pore and crystalline
structure can deliver excellent supercapacitive performances in term of high energy and power densities (26–35
Wh/kg at 0.338–6.25 kW/kg in organic system) as well as 100% cycling stability. Both experiment and density
functional theory (DFT) calculations demonstrate that C-O-K structure can be regenerated in CO2 activation,
leading to continuously catalytic effects despite low-dose K2CO3 addition. Combining the low-cost sources,
simple preparation procedures and good application performances, the resulting activated carbon holds great
potentials for scalable production and applications. Such strategy can also be extented for developing high-
performance activated carbons from various kinds of solid carbon sources.

⁎
Corresponding authors.
E-mail addresses: [email protected] (F. Sun), [email protected] (J. Gao).

https://doi.org/10.1016/j.cej.2019.123205
Received 3 May 2019; Received in revised form 16 September 2019; Accepted 16 October 2019
1385-8947/ © 2019 Published by Elsevier B.V.

Please cite this article as: Lijie Wang, et al., Chemical Engineering Journal, https://doi.org/10.1016/j.cej.2019.123205
L. Wang, et al. Chemical Engineering Journal xxx (xxxx) xxxx

1. Introduction leading to the formation of activated carbon matrix with abundant pore
structures [22,23]. Although chemical activation could produce high-
Carbonaceous materials have been explored and applied in many porosity activated carbon with surface areas higher than 2000 m2g−1,
fields including adsorption/separation [1,2], energy storage [3,4], this process faces the dilemma of complicated craft and equipment re-
heterogeneous catalysis [5] and environmental management [6,7] due quirements as well as the difficulties of environmental governance due
to their advantages of controllable production and environmental to the massive water/acid consumption to remove residual impurities
friendliness. Nowadays, various functional carbon materials including [21]. Considering the weaknesses of traditional physical or chemical
graphite/graphene, carbon fibers, carbon tubes and porous carbons activation crafts, it is highly desirable to develop low-cost, facile and
have been explored in carbon materials science and engineering. green strategies for high-performance porous carbons.
Among them, porous carbons, due to their wide resources, strong In the field of coal gasification, alkali alkaline earth metals (namely
structure controllability, are the common choice for many application AAEMs) and transition metals are commonly used to increase gasifi-
fields including pollutant removal [8,9], catalyst carrier [10] and en- cation reaction rate and to decrease reaction energy barrier, so that coal
ergy storage [11]. The processes of pollutant molecule removal and structure could be transformed into small gas molecules at relatively
energy storage using porous carbons have a common scientific nature, lower reaction temperatures [26–28]. The processes of physical acti-
namely the adsorption, reaction, transport and storage of target species vation and gasification have a common reaction nature that gas com-
(such as molecules and electrolyte ions) in the pore structure of carbon ponents etch coal structure to form gas products while leaving activated
matrix [2,4]. In general, high transport rate and adsorption/storage carbon or coal char with porous structure [26]. Due to the similarity of
capacity require porous carbon with high-activity including well-de- physical activation reaction and gasification reaction, that the existence
veloped porosity and abundant active sites; long cycle life requires high of AAEM species could adjust the pore structure of resulting activated
stable carbon structure [12–14]. Thus, regulating the activity and sta- carbons. In our previous study, catalytically physical activation process
bility of porous carbon is the key to improve its application perfor- offers a new route for adjusting pore structure of coal-based activated
mances [15]. For the aim to engineering the nanostructures of porous carbons in which only a small amount of catalysts, such as intrinsic
carbons, many synthetic methods such as hydrothermal carbonization, minerals, calcium or iron-based species [29] are needed to achieve
self-assembly, soft/hard template and chemical vapor desposition have porosity regulation for the resulting activated carbons.
been adopted using various carbon resources. These methods are based This work put forward a new green synthesis concept based on trace
on so-called “bottom-up” strategies from small molecules to carbon K2CO3catalyzed CO2 activation by which high porosity activated car-
structures; the resulting novel porous carbons with uniform pores and bons are produced and more importantly the catalytic species can be
ordered structures can be synthesized under the actions of non-covalent recycled during activation process. This synthesis is achieved by in-
bonds [16], surfactants [17], template agents [18] or catalysts [19]. troducing trace K2CO3 (only 1–2% weight ratio of coal) into coal matrix
However, these methods are not well suitable for compact and complex followed by a CO2-assisted physical activation process during which the
carbon precursors, such as coal structure with macromolecular net- pore development and reorganization can be realized by enhancing the
works. In this regard, traditional “up-bottom” strategies including etching reaction between coal matrix and CO2. It is worth mentioning
physical or chemical activation are still the preferred strategies for coal that potassium-based catalytic component especially the C-O-K struc-
or biomass converted porous carbon [20–22], especially to achieve pore ture can be regenerated under the assistant role of CO2 atmosphere at
structure regulation. high temperature, leading to continuous formation and development of
Although physical and chemical activation have been well com- pore structure. The as-prepared porous carbon possesses developed
mercialized in the production of coal or biomass-based activated carbon pore structure which are even superior to the activated carbon prepared
(coke), there are still many limitations due to the intrinsic reaction by traditional chemical activation with large dosage of K2CO3 (three
characteristics between carbon matrix and activation agents [23,24]. times weight of coal). Evaluated as CO2 adsorbent and supercapacitor
More specifically, physical activation often employs H2O, CO2, flue gas electrode material, the optimized Ca_AC-1 sample exhibits high CO2
or their mixtures as activation agents to create porosity in which pro- adsorption capacity of 4.36 mmol/g at 0 °C and 1 bar and excellent
cess small volatile molecules first release from coal matrix to form the symmetrical aqueous or organic supercapacitor performances with
initial pores at the low-/medium-temperature, then gas activation maximum energy and power density up to 35.7 Wh/kg and 6 kW/kg,
agents gradually etch the coal framework from surface to inside for respectively. Such a green and low-cost strategy achieve the synergistic
creating desirable porosity along with the reactions of C + H2O → regulation of activity and stability of coal-based activated carbon,
H2 + CO(-131KJ) or C + CO2 → 2CO (-170.5KJ) at elevated tempera- holding great potentials for industrial production and applications.
tures (ca. 700–1000 °C) [20,24]. Due to the diversity in molecular size,
diffusivity and reactivity of activation agents, the optimized activation 2. Experimental section
temperature and pore texture of as-obtained activated carbons are
different. Taking the commonly used H2O and CO2 as examples, the 2.1. Materials and sample preparation
temperature of H2O activation is ca.100 °C lower than CO2 activation by
virtue of the higher diffusivity and reactivity of H2O molecules relative In this study, a kind of sub-bituminous coal of Zhundong coal
to CO2 molecules. Therefore, adjusting reaction atmosphere is the main (Proximate, ultimate analyses and ash compositions are shown in Table
means to regulate pore structure of activated carbon in physical acti- S1 and S2), was taken as carbon source. The specific reasons are as
vation process [25]. However, physical activation often leads to the follows: 1) Zhundong coal has an exploitable reserve of 164 Gt, which
resulting activated carbons with inferior porosity due to the lower re- providing abundant and cheap raw materials [30,31]; 2) the char-
activity between gaseous activation agents and coal framework. Acti- acteristics of low carbon content, high moisture and ash content of
vated carbons obtained by H2O or CO2 activation, commonly have a Zhundong coal endow it unfavorable for direct combustion, but suitable
low specific surface area (less than 1000 m2g−1) and pore volume (less for carbon source; 3) Zhundong coal has more volatile components and
than 0.5cm3g−1) [20,24], thus limiting their wide applications. In higher reactivity due to its rich aromatic structures with fewer benzene
contrast, chemical activation is an important choice in achieving high rings, and it is easy to form porous carbon under the action of activating
porosity. During chemical activation process, high-dose activation agents [32]. To investigate the effect of K2CO3 on pore configuration,
agents (such as KOH, K2CO3, ZnCl2, H3PO4 with dosage several times raw coal particles (size of less than 1 mm) were firstly underwent acid-
higher than that of carbon precursors) need to form a uniform molten assisted washing (5 M hydrochloric acid and 20 wt-% hydrofluoric acid)
thermal environment, which promotes coal structure cracking, cross- to remove minerals from the coal [33]. The X-ray diffraction (XRD)
linking and aromatization assisted with the release of gas components, patterns of raw coal sample (ZD) and mineral-free coal (ZD-HCF) shown

2
L. Wang, et al. Chemical Engineering Journal xxx (xxxx) xxxx

in Fig. S1 indicates that minerals could be removed from raw coal under a frequency range from 100 kHz to 10 MHz at an amplitude of 10 mV.
this acid treatment. In this experiment, liquid impregnation was used The specific capacitances of the symmetric supercapacitor were calcu-
for preparation of K2CO3 loaded coal samples, during which the lated by using the following Eq. (1).
weighted mineral-free coal was added to the K2CO3 (Aladdin, 99%)
C= 2IΔt/(mΔV) (1)
solution under 300r/min magnetic stirring and the mixture was then
evaporated under 80 °C. Then the pretreated coal samples underwent where I is the constant discharge current, Δt is the time for a full dis-
heating treatment at a constant heating rate of 10 °C/min to 900 °C and charge, m is the active material mass of one electrode and ΔV represents
kept for 1 h with a total flow rate of 200 mL/min using in a horizontal voltage window.
tube furnace, as shown in Fig. S2. The detailed information of the Energy density of the symmetric device was calculated by the Eq.
samples was listed in Table S3. After washing and drying process [33], (2).
various samples of Char, Ch_AC, Ph_AC, Ca_AC-1 and Ca-AC-2 were
E= CV 2/8 (2)
prepared depending on mixture ratio and reaction atmosphere.
where V is the cell operating voltage (obtained from the discharge
2.2. Structural characterizations of prepared activated carbons curve subtracted by the IR drop).
The power densities of the device (P) were calculated from the
The microstructure and surface mineral distribution of obtained following Eq. (3).
samples were characterized by transmission electron microscopy (TEM, P= E/Δt (3)
JEM-2100) and scanning electron microscopy (SEM, Helios
Nanolab600i) with energy dispersive spectrum analysis (EDS). Pore where Δt is the discharge time [35].
structure and detailed parameters including BET surface area, pore
volume and pore size distribution were measured by N2 adsorption/ 2.5. Computational methods
desorption analysis (BeiShiDe instrument co., Ltd, 3H2000PS) under
liquid nitrogen environment (−196 °C). The carbonaceous structure All the calculations were carried out with Gaussian 09 program
and composition were measured by XRD (Rigaku D/Max 2400) and packages [36]. The geometries for all stationary points investigated in
Raman spectrometer (Renishaw inVia Micro-Raman). X-ray photoelec- this study were optimized with M06-2X [37] method in conjunction
tron spectroscopy (XPS, PHI 5700 ESCA) analysis was performed on a with 6-31G (d, p) basis set. Harmonic vibrational frequencies were
System with AlKa X-ray at 14 kV and 6 mA for surface elemental ana- calculated at the same level to verify the nature of corresponding sta-
lysis. Besides, the carbon conversion rates of coal samples were re- tionary points on the potential energy surface (minima or transition
corded by the thermal gravimetric analyzer (TGA, Beijing Henven state). Also, intrinsic reaction coordinate calculations were performed
Instrument Factory, HCT-3) under simulated preparation atmosphere. to confirm that the transition states connect desired reactants and
products. Single point calculations were performed at the M06-2X/6-
2.3. CO2 adsorption test 311 + G (d, p) level to get more reliable energy barriers. The M06-2X
functional has been shown to significantly outperform widely used
In this study, CO2 was selected as probe molecule to evaluate the B3LYP in determination of saddle point geometries and barrier heights
pore structure of typical ACs. More specifically, CO2 adsorption test was [38,39]. The energy barriers of reactions investigated in our work Eb
measured using the above static volumetric analyzer at the 0 °C. Before were calculated by the Eq. (4).
adsorption, the ACs were degassed for 8 h at 180 °C under N2 atmo- Eb = ETS − EPR (4)
sphere, followed by the introduction of pure CO2 into the test system.
Meanwhile, N2 adsorption under the same condition was tested for EPR and ETS respectively represent single point energy of the com-
evaluating the selectivity of ACs. Moreover, CO2 recycling tests for the plex formed before the reaction and the transition state.
typical sample were also carried out on Nitrogen adsorption static vo-
lumetric analyzer. Specifically, the saturated sample was degassed at 3. Results and discussion
200 °C for 1.5 h to release the adsorbed CO2 molecules, then the de-
gassed sample was weighted for the next adsorption cycle. 3.1. Structural characterizations of ACs

2.4. Electrochemical measurement Figs. 1 and S3 illustrate the morphology and element distributions
of ACs. As shown in Fig. 1a, Char derived from the pyrolysis of coal
To measure the supercapacitive performance of typical samples, features a block structure without etching and holes. Ph_AC from solely
symmetric supercapacitors were constructed by using aqueous (6 M CO2 activation exhibits microporous structure, indicating the porosity
KOH) and organic (1 M TEABF4-PC) electrolytes. To fabricate the creation function of CO2 (Fig. 1b-c). Unlike Char and Ph_AC, Ch_AC
working electrodes, the obtained porous carbons (80 wt%) were mixed prepared by K2CO3 chemical activation shows a relatively rougher
with carbon black (10 wt%) and polytetrafluoroethylene (PTFE) (10 wt surface with rich nano-scale holes distribution (Figs. 1d and S3a).
%) homogeneously. Then the mixture was rolled into a thin film and Ca_AC-1 from trace K2CO3 induced catalytic CO2 activation also ex-
dried at 100 °C for 12 h. For the device construction, the pre-dried thin hibits analogously block structure with a roughened surface and rich
film with a certain mass was pressed onto the nickel foam or conductive pores similar to Ch_AC, clearly demonstrating the pore development
carbon coated aluminum foils (current collector) to form the electrodes. can be enhanced by adding trace K2CO3. Moreover, there are differ-
The electrodes were further dried at 120 °C for 12 h under vacuum to ences between Ca_AC-1 and Ch_AC, particularly, the degree of crystal-
remove the residual moisture and solvent. Then, two identical (by linity. It can be seen from Figs. 1h–I and S3e–f, unlike the amorphous
weight and size) electrodes were assembled in a CR2032 stainless steel structure of Ph_AC and graphene-like structures of Ch_AC (Figs. 1e–f
coin cell with the glass fiber membrane (GF/D) (from Whatman) as the and S3b–c), Ca_AC-1 exhibits graphite-like microcrystalline structure
separator. The assembly of the coin cells was done in a glove box filled with a smaller size and short-range order. Besides, Fig. 1j–l show that
with Ar. The loading of the active material on each electrode the EDS analysis of Ca_AC-1 without washing indicating the uniform
was ~ 1.0 mg cm−2 [34]. distribution of the K element that is helpful to achieving uniform acti-
VMP3 Electrochemical Workstation (Bio-logic) was used for elec- vation and pore development.
trochemical tests. The electrochemical impedance spectroscopy was Fig. 2a–b and Table 1 show the N2 adsorption/desorption isotherms,
performed using a Solartron 1287/1260 electrochemical workstation in corresponding pore size distribution and detailed pore parameters of

3
L. Wang, et al. Chemical Engineering Journal xxx (xxxx) xxxx

Fig. 1. SEM and TEM images of as-prepared ACs: SEM (a) image of Char; SEM (b) and high-resolution TEM(HRTEM)(c) images of Ph_AC; SEM(d) and HRTEM (e,f)
images of Ch_AC; SEM (g) and HRTEM (h,i) images of Ca_AC-1; SEM(j) and EDS(k, l) images of Ca_AC-1(K).

prepared ACs, which exhibit significant differences due to the various the resulting Ch_AC exhibit type I isotherms with predominant ad-
preparation conditions. More specifically, Char derived from coal car- sorption below P/P0 = 0.1, which is the characteristic of micropore
bonization under N2 atmosphere presents the poorest pore structure structure [40]. As a result, the micro-porous Ch_AC has high BET sur-
with the lowest BET surface area (215 m2/g) and pore volume face area and pore volume of 1609 m2/g and 0.94 cm3/g, respectively.
(0.14 m2/g). By contrast, Ph_AC prepared by solely CO2 activation By contrast, Ca_AC-1 prepared by trace K2CO3 catalytic activation
possesses slightly higher BET specific surface area (345 m2/g) and pore presents much rich pore structure with the highest BET surface area and
volume (0.19 cm3/g) than Char, further indicating the limited effect of pore volume up to 1773 m2/g and 1.11 cm3/g, four times more than
solely CO2 atmosphere on pore development. Unlike carbonization and Ph_AC and even higher than Ch_AC, which impressively demonstrated
physical activation process, large-dosage K2CO3 chemical process gives that adding a small dose of K2CO3 into coal framework can cause a

4
L. Wang, et al. Chemical Engineering Journal xxx (xxxx) xxxx

Fig. 2. (a) N2 adsorption/desorption isotherms of prepared ACs. (b) Pore size distributions (PSDs) of prepared ACs. (c) XRD patterns of Ch_AC and Ca_AC-1. (d)
Raman spectra of Ch_AC and Ca_AC-1.

significant increase in pore structure during CO2 activation process. It’s (only 1–2% weight ratio of coal) during CO2 assisted-activation can
worth noted that such high pore structure parameters are also much cause significant enhancing for pore structure of as-obtained ACs. The
higher than that of other coal/biomass-based porous carbons by phy- pore construction and carbonaceous structure of activated carbon de-
sical or chemical activation methods, as shown in Table S4. Besides, the rived from this method are similar to that prepared by K2CO3 chemical
difference between Ca_AC-1 and Ca_AC-2 indicates that the pore para- activation. To further explore the trace K2CO3 effects, both the TGA
meters of obtained ACs can be adjusted by changing the amount of analysis of various coal samples during the simulated preparation
K2CO3 addition and the concentration of CO2 atmosphere. Except for process and yields of obtained activated carbons were evaluated. As can
pore structure analysis, the microstructure of Ch_AC and Ca_AC-1 was be seen from Fig. 3a, mineral-free coal sample in carbonization process
also evaluated by XRD and Raman, as shown in Fig. 2c–d. The two at N2 atmosphere exhibits the lowest conversion rate, and the second
samples appeared similar XRD patterns with the broad diffraction peaks one is mineral-free coal sample in CO2 activation process, indicating
at about 24 and 44° corresponding to (0 0 2) and (1 0 0) of carbon CO2 can enhance etching reaction. In contrast, trace K2CO3 loaded coal
structure [41]. Besides, the apparent peaks at ~1350 and ~1590 cm−1 sample during CO2 activation has higher carbon conversion rates,
in Raman spectra represent the D and G band of the carbon framework, which indicates that only a small amount of K2CO3 addition can sig-
respectively, which indicated both Ch_AC and Ca_AC-1 exhibit mainly nificantly enhance the reaction reactivity between coal sample and
amorphous structure characteristics [32]. activated gas. Based on the pore structure analyses, it can be safely
speculated that high porosity ascribes to high conversion rate. As ex-
3.2. Porosity development mechanism investigation pect, Fig. 3b shows that the BET specific surface areas and pore volumes
of various ACs are proportional to the conversion rates of corresponding
So far, it has been demonstrated that adding a small amount K2CO3 coal samples, demonstrating that trace K2CO3 induced catalytic

Table 1
Pore structure parameters of as-prepared ACs.
Sample SBET(m2 g−1) Vtotal(cm3 g−1) Vmic(cm3 g−1) Vmec+mac(cm3 g−1) Dave(nm)

Char 215 0.14 0.11 0.02 2.74

Ph_AC 345 0.19 0.16 0.03 2.19
Ch_AC 1609 0.94 0.67 0.27 2.33
Ca_AC-1 1773 1.11 0.66 0.45 2.51
Ca_AC-2 1294 0.82 0.54 0.28 2.54

5
L. Wang, et al. Chemical Engineering Journal xxx (xxxx) xxxx

Fig. 3. (a) TGA profiles of coal samples under simulated thermal converted process. (b) The relationship between carbon conversions of pretreated coal samples and
pore structure of ACs. (c) Yields of various ACs. (d) The contribution of carbon ablation to pore structure of ACs.

activation could promote porosity of as-obtained activated carbon, mechanism of K2CO3 mentioned in previous studies [45,46]. As showed
which is consistent with our previous study [12]. Besides, yields of ACs in Fig. 4b, CO will be released from K2CO3 attached to the carbon
derived from the various process were evaluated and shown in Fig. 3c. matrix to form composite structure containing C-O-K during CO2 acti-
It is easy to understand that Char and Ph_AC with the poor porosity vation process, and then the surface carbon reduces this complex
have relatively high yields. While, it’s worth noting that although structure to form the K-C complex with CO releasing. The reduced K-C
Ch_AC and Ca_AC-1 have similar development pore structure, the yield complex will be oxidized to a new C-O-K complex when CO2 is in-
of Ch_AC is higher than Ca_AC-1, which may be due to the different troduced during high temperature environment, ensuring the sub-
pore-forming mechanisms. In general, CO2 involved activation is often sequent catalytic reaction going and pore developing. Thus, although a
limited by diffusion process, thus causing surface ablation and low yield small amount of catalytic species was added, it will be regenerated
of Ca_AC-1 [42]. For further demonstration, the contribution rates of based on the redox cycle in the assistance of CO2 atmosphere. Of
ablation carbon to pore formation were calculated (pore parameter course, it is inevitable that some K element will evaporate since the
divided by ablation rate, where ablation rate is calculated by one minus reaction temperature in this study is slightly higher than the boiling
yield), as shown in Fig. 3d. It can be seen that Ch_AC has the highest temperature of K element [47]. With the catalytic gasification reaction
ablation contribution rate, and the second is Ca_AC-1, which was much and phase evolution, pore structure of activated carbon can be further
higher than Char and Ph_AC. Comprehensively considering the lower developed. To further elucidate the existence form of K, the surface
K2CO3 dosage and high porosity creation ability, such trace K2CO3 in- elemental analyses of Ch_AC(K) and Ca_AC-1(K) without acid washing
duced catalytic activation strategy demonstrates obvious advantages in were conducted, as shown in Fig. 4c–f. As we can see from Fig. 4c and e,
preparing high-porosity activated carbons. the broad two peaks at 293.0 eV and 295.6 eV indicated the K 2p3/2 and
To further explore the role of K2CO3 in catalytic activation, the K 2p1/2 [48,49], respectively. Both activated carbons obtained from
phase of potassium-based species in typical activated carbons before catalytic activation and chemical activation show similar potassium-
washing was detected. As can be seen from Fig. 4a, Ch_AC(K) without containing functional groups fitting into K 2p3/2 peak in K–O group
washing obtained by the traditional K2CO3 chemical activation shows (denoted as KO) at 292.4 eV ( ± 0.2) and potassium with carbonate
visible potassium-containing peaks which are assigned to surrounding (denoted as KCB) at 293.2 eV ( ± 0.1) [50]. Besides, K 2p1/
K2CO3(carbonate), K2O2/KO2(oxide), K (elementary substance) and 2 peak could be deconvoluted into the same species which has a binding
K4C60/K3C60(carbide). These K-related species ascribe to the phase energy shifted from K 2p3/2 correspondingly. As shown in Fig. 4d and f,
evolution in large-dosage K2CO3 induced chemical activation. The ap- both activated carbons obtained from catalytic activation and chemical
pearance of carbides including K4C60/K3C60 further explains the for- activation show similar functional groups with C1s spectrum fitting into
mation of graphite streaks in Fig. 1e–f [43,44]. Compared with graphite at 284.5 eV (denoted as CeC), carbon singly bound to oxygen
Ch_AC(K), Ca_AC-1(K) obtained by the catalytic activation also presents (denoted as CeO) at 286.1 eV and carbon bound to two oxygen atoms
some less obvious peaks of K-based species except two broad peaks at (denoted as OeC]O) at 288.7 eV [50,51]. In general, the existence of
about 24° and 43° corresponding amorphous carbon, and it is worth KeO groups and CeO groups collectively demonstrate the existence of
mentioning that these K-related components are similar to those in CeOeK species, which is consistent with previous studies [45,46].
Ch_AC(K). The presence of these components verifies the catalytic Through experimental analyses and previous research, it can be

6
L. Wang, et al. Chemical Engineering Journal xxx (xxxx) xxxx

Fig. 4. (a) The XRD patterns of Ch_AC(K) and Ca_AC-1(K) without acid washing. (b) The schematic diagram of reaction pathway in trace K2CO3 induced catalytic CO2
activation. (c-d) XPS spectra of Ca_AC-1(K) without acid washing: (c) K 2p and (d) C 1 s. (e-f) XPS spectra of Ch_AC(K) without acid washing: (e) K 2p, (f) C 1 s.

concluded that CeOeK structure is the key to catalytic reaction. occurs with a barrier of 24.2 kJ/mol, producing fragment 1 and frag-
To further explore the catalytic role of C-O-K in CO2 induced erosion ment 2 (C3 → C4). Finally, the second CO can be released via the O (3)
reaction of coal sample, molecular models with and without C-O-K in fragment 1 attacking C (6) in fragment 2 (C4 → C5). The release of
structure (Fig. S4) were constructed and DFT calculations for these the second CO needs to overcome a barrier of 209.3 kJ/mol. Since the
models reacting with CO2 was performed. Considering Zhundong coal, transition state TS2 of H transfer from C (1) to C (4) has the highest
a kind of low-rank coal has 1–4 benzene rings in its aromatic structures, relative energy, such H transfer is the rate-determining step of the
the models used in our calculations contain three benzene rings. The erosion reaction. According to Fig. 5b, the energy barrier for the rate-
detailed pathways and potential energy profiles for the calculated re- determining step of overall process is 591.9 kJ/mol.
actions are shown in Fig. 5. As shown in Fig. 5a-b, in the case without C- When the C-O-K is involved in the coal structure, the reaction
O-K involved, CO2 firstly attacks C (1) atom via TS1, generating the first pathways and corresponding energy barriers are substantially altered.
CO molecule and C-O complex (C1 → C2). The energy barrier for this As shown in Fig. 5c–d, firstly, CO2 attacks C (1) atom, producing
step is 449.3 kJ/mol. Then the H (2) atom in the produced C-O complex complex C-CO2 (COK1 → COK2). The energy barrier for CO2 attacking
transfers to C (4) atom with an energy barrier of 326.9 kJ/mol (C2 → coal structure is 228.5 kJ/mol, nearly 200 kJ/mol lower than that in the
C3). After H transfer, the breakage of bond between C (4) and O (3) absence of C-O-K, which indicates that C-O-K can facilitate the attack

7
L. Wang, et al. Chemical Engineering Journal xxx (xxxx) xxxx

Fig. 5. (a) Reactants for C model. (b) Potential energy profiles for reactions of C model. (c) Reactants for COK model. (d) Potential energy profiles for reactions of
COK model.

between CO2 and coal structure, thus increasing the possibility of abundant low-rank coal and quite a few of inexpensive K2CO3 (only
modifying carbon structures. Then the dissociation of the bond between 1–2% weight ratio of coal) as raw materials and requires a small
O (3) and C (6) occur, producing the first CO molecule and C-O complex amount of water or low concentration acid to remove residual po-
(COK2 → COK3). After the release of CO, H (2) bonded to C (1) in C-O tassium base components removal, reducing the difficulty of subsequent
complex transfers to C (5) via two H transfer processes. The energy environmental governance.
barriers for H transfer from C (1) to C (4) and from C (4) to C (5) are
50.2 kJ/mol and 54.1 kJ/mol respectively (COK3 → COK4 → COK5). 3.3. CO2 adsorption property of typical ACs
After H transfer, the breakage of the bond between C (1) and C (4) as
well as C (4) attacking C (7) co-occur (COK5 → COK6). The relative To further evaluate the adsorption activity and structural stability of
energy of TS9 in this step is highest in the overall erosion reaction. As as-obtained ACs as gaseous adsorbents, CO2 adsorption performance
shown in Fig. 5d, the energy barrier for the rate-determining step is was tested on the two typical ACs (Ch_AC and Ca_AC-1). As shown in
413.1 kJ/mol, 178.8 kJ/mol lower than that without C-O-K involved. Fig. 7a, CO2 and N2 adsorption capacity of Ca_AC-1 at 0 °C are
Thus, it can be concluded that C-O-K structure plays a catalytic role in 4.36 mmol/g and 0.70 mmol/g, respectively. Despite Ch_AC possesses
the overall erosion reaction. the higher CO2 adsorption capacity of 4.47 mmol/g and lower N2 ad-
Based on the above analysis, this study proposes the pore formation sorption capacity of 0.58 mmol/g at 0 °C than Ca_AC-1, low cost and
mechanism of trace K2CO3 induced catalytic activation and conven- simple process of Ca_AC-1 can make up for the slight weakness in ad-
tional large-dosage K2CO3 chemical activation, as shown in Fig. 6. In sorption capacity and separation performance. Furthermore, such ca-
our novel synthetic strategy, adding little K2CO3 in CO2 assisted acti- pacity of Ca_AC-1 is higher than many reported coal/biomass-based
vation can significantly promote etching reactions between coal sample activated carbons derived from physical or chemical activation, as
and CO2 atmosphere, thus enhancing pore formation and creation. summarized in Table 2. To explore the stability of Ca_AC-1 in the
Compared with solely CO2 activation, K2CO3-catalyzed CO2 activation multiple recycling processes, “adsorption-regeneration” experiments of
has much higher porosity-contribution rate, although it may cause re- Ca_AC-1 were conducted, as shown in Fig. 6b. It can be seen that Ca_AC-
latively obvious surface ablation and lower yield. Although frequently- 1 can maintain more than 97% CO2 adsorption capacity after ten times,
used K2CO3 chemical activation has both high yield and porosity-con- demonstrating excellent cyclic stability and good potential in CO2
tribution rate, it requires a large amount of activation agent dose (mass capture and storage.
ratio of activation agent to carbon precursor up to 3) and complex post-
treatment process, thus significantly increasing the cost of raw mate-
3.4. Electrochemical performance of constructed symmetric supercapacitors
rials and operation. From the perspective of industrial production, this
study proposes a simple, green and economic method for preparation
The optimized Ca_AC-1 was further explored as EDLC electrode
activated carbon with developed pore structure. This craft only needs
materials to construct symmetric device in both aqueous and organic

8
L. Wang, et al. Chemical Engineering Journal xxx (xxxx) xxxx

Fig. 6. The pore formation mechanism of trace K2CO3 induced catalytic activation and conventional large-dosage K2CO3 chemical activation.

electrolytes. Fig. 8 shows the electrochemical performance of the voltage (IR) drops than those of Ch_AC, indicating a better rate per-
Ca_AC-1 and Ch_AC constructed symmetric supercapacitor in 6 M KOH formance relative to Ch_AC. Calculated form the charge-discharge
electrolyte. Fig. 8a and b show the CV curves of Ca_AC-1 and Ch_AC curves in Fig. S5c–d, the gravimetric capacitance of Ca_AC-1 at a low
electrode at the scan rates of 20 mV/s and 200 mV/s. Compared with current density of 0.5 A/g is ~146F/g; even at an ultra-high current
Ch_AC, although Ca_AC-1 electrode presents the lower capacitance at density of 20 A/g, the Ca_AC-1 still provides a gravimetric capacitance
the scan rate of 20 mV/s, Ca_AC-1 electrode keeps a better rectangular- of 91F/g with limited IR drop of around 0.2 V. However, although
like CV profile at the scan rate of 200 mV/s, suggesting the high-rate Ch_AC presents the higher gravimetric capacitance of 178F/g at 0.5A/g,
capability, which can be clearly illustrated by the detailed CV curves at the gravimetric capacitance of Ch_AC at 20 A/g is only ~49F/g with
various scan rates from 10 to 200 mV/s of Ca_AC-1 and Ch_AC shown in obvious IR drop up to 0.45 V. Fig. 8e further compare the electro-
Fig. S5a and b. Fig. 8c and d compare the galvanostatic charge-dis- chemical impedance spectroscopy (EIS) of Ca_AC-1 and Ch_AC elec-
charge curves of Ca_AC-1 and Ch_AC at 0.5 A/g and 5 A/g, clearly trode. Both Ca_AC-1 and Ch_AC electrode exhibit straight line in the
demonstrating similar conclusion. At the lower current density of 0.5A/ low-frequency region and a semicircle in the high-frequency region,
g, Ca_AC-1 has shorter discharge time, but at the higher current density indicating typical capacitive behaviors. Compared with Ch_AC, Ca_AC-1
of 5A/g, Ca_AC-1 presents the both higher capacitance and smaller exhibits the relatively lower intrinsic ohmic resistances (first intercept

Fig. 7. (a)CO2 and N2 adsorption isotherms of Ch_AC and Ca_AC-1 at 0 °C under atmospheric pressure. (b) CO2 cycling performance of Ca_AC-1 at 0 °C.

9
L. Wang, et al. Chemical Engineering Journal xxx (xxxx) xxxx

Table 2 along the real axis) and a similar equivalent series resistance (ESR, by
A comparison of reported biomass/coal based active carbons with our samples extrapolating the vertical portion to the real axis), demonstrating the
on CO2 adsorption properties at 273 K. lower contact resistance. Furthermore, the straight line in the low-fre-
Raw material Activation method Pressure Capacity Ref. quency region of Ca_AC-1 electrode has a higher slope compared to that
(mmol/g) of the Ch_AC electrode, indicating increased ion/electron transporta-
tion. The EIS results further explain a rather inferior rate performance
Starch KOH (1:1) 101 kPa 3.02 [52]
of Ca_AC-1 relative to Ch_AC. Fig. 8f shows the Ragone plot of the
Gelatin KOH (1:1) 101 kPa 4.25 [52]
Coffee residue CO2 101 kPa 3.2 [53] Ca_AC-1 and Ch_AC based supercapacitor. The Ca_AC-1 based super-
Almond shells O2(3%) 101 kPa 3.11 [53] capacitor can deliver a energy density of 5.1 Wh/kg at power density of
Olive stones O2(3%) 101 kPa 3.10 [54] 0.125 kW/kg and still 2.0 Wh/kg at a high power density of 4 kW/kg,
Zhundong coal NH3 (75%) 800 mmHg 3.71 [32]
much higher than Ch_AC based aqueous supercapacitor with rapidly
Palm stone H3PO4 (2:1) 101 kPa 3.1 [55]
Palm stone ZnCl2 (2:1) 101 kPa 2.4 [55]
decreased energy densities.
Zhundong coal CO2 (30%) with trace 1 bar 4.36 This Furthermore, organic supercapacitors were constructed to increase
K2CO3 (2%) work the energy density of EDLC devices. Fig. 9 shows the electrochemical
performance of the Ca_AC-1 and Ch_AC constructed symmetric

Fig. 8. Electrochemical performances of Ca_AC-1 and Ch_AC constructed symmetric supercapacitor in 6 M KOH electrolyte. (a) CV curves of Ca_AC-1 and Ch_AC at
20 mV/s. (b) CV curves of Ca_AC-1 and Ch_AC at 200 mV/s. (c) Galvanostatic charge–discharge curves of Ca_AC-1 and Ch_AC at a charge–discharge current density of
0.5A/g. (d) Galvanostatic charge–discharge curves of Ca_AC-1 and Ch_AC at a charge–discharge current density of 5A/g. (e) Nyquist plots of Ca_AC-1 and Ch_AC,
insert shows the enlarged view. (f) Ragone plots of the Ca_AC-1 and Ch_AC capacitor.

10
L. Wang, et al. Chemical Engineering Journal xxx (xxxx) xxxx

Fig. 9. Electrochemical performance of the Ca_AC-1 and Ch_AC constructed symmetric supercapacitor in 1 M tetraethylammonium tetrafluoroborate in propylene
carbonate (1 M TEA BF4-PC) electrolyte. (a) CV curves of Ca_AC-1 and Ch_AC at 20 mV/s. (b) CV curves of Ca_AC-1 and Ch_AC at 200 mV/s. (c) Galvanostatic
charge–discharge curves of Ca_AC-1 and Ch_AC at a charge–discharge current density of 0.5A/g. (d) Galvanostatic charge–discharge curves of Ca_AC-1 and Ch_AC at
a charge–discharge current density of 5A/g. (e) Nyquist plot of Ca_AC-1 and Ch_AC, insert shows the enlarged view. (f) Ragone plots of the Ca_AC-1 and Ch_AC
capacitor.

supercapacitor in organic electrolytes. Fig. 9a–d compare the CV curves Ch_AC at 10 A/g is only ~101F/g. Fig. 9e further compare the elec-
and galvanostatic charge-discharge curves of Ca_AC-1 and Ch_AC trochemical impedance spectroscopy (EIS) of Ca_AC-1 and Ch_AC
electrodes from which Ca_AC-1 based supercapacitors has a slightly electrode in organic electrolyte. Compared with Ch_AC, Ca_AC-1 ex-
lower capacitances in the low scan rates and current densities, but hibits the relatively lower intrinsic ohmic resistances and equivalent
provides an obviously higher capacitances at high scan rates and cur- series resistance in the high-frequency region and a higher slope in the
rent densities, indicating the high-rate capability in organic system. The low-frequency region, demonstrating the lower contact resistance,
detailed CV curves at various scan rates and galvanostatic charge–- better electrolyte accessibility and good ion/electron transportation.
discharge curves at various current densities of Ca_AC-1 and Ch_AC The Ragone plots of Ca_AC-1 and Ch_AC based supercapacitor in or-
electrode shown in Fig. S6 Calculated form the charge-discharge curves ganic electrolyte shown in Fig. 9f reveal that the Ca_AC-1 super-
of constructed symmetrical organic supercapacitor in Fig. S6c–d, the capacitor can deliver high energy densities of 26–35 Wh/kg at power
gravimetric capacitance of Ca_AC-1 at a low current density of 0.5 A/g densities of 0.3375–6.25 kW/kg, indicating superior rate performance
is 141F/g; even at a high current density of 10 A/g, the Ca_AC-1 still relative to Ch_AC.
provides a gravimetric capacitance of 121F/g with limited IR drop of Fig. 10a further shows the Ragone plot of the Ca_AC-1 capacitor as
around 0.2 V. However, although Ch_AC presents the higher gravi- well as its comparison with the representative porous carbon-based
metric capacitance of 148F/g at 0.5A/g, the gravimetric capacitance of supercapacitors in a similar electrolyte [56–58], which demonstrates

11
L. Wang, et al. Chemical Engineering Journal xxx (xxxx) xxxx

Fig. 10. (a) Ragone plots of Ca_AC-1-based capacitor and a comparison with reported symmetric carbon capacitors. (b) cycling stability of Ca_AC-1 at a current
density of 5 A/g, inset shows the galvanostatic charge–discharge curves of Ca_AC-1 at the first three circles and the last three circles.

the state-of-the-art performances. Fig. 10b shows the cycling stability of References
the Ca_AC-1 supercapacitor, which maintains ~100% of the initial ca-
pacitance after 5000 cycles at a current density of 5A/g, indicating a [1] X.C. Ma, L.Q. Li, R. Chen, C.H. Wang, Porous carbon materials based on biomass for
superior long cycling life. Combining the low-cost and simple pre- acetone adsorption: Effect of surface chemistry and porous structure, Appl. Surf. Sci.
459 (2018) 657–664.
paration craft, Ca_AC-1 holds great potential in commercial super- [2] J. Wang, P.X. Zhang, L. Liu, Y. Zhang, J.F. Yang, Z.L. Zeng, S.G. Deng, Controllable
capacitor applications. synthesis of bifunctional porous carbon for efficient gas-mixture separation and
high-performance supercapacitor, Chem. Eng. J. 348 (2018) 57–66.
[3] G.F. Ma, Q. Yang, K.J. Sun, H. Peng, F.T. Ran, X.L. Zhao, Z.Q. Lei, Nitrogen-doped
4. Conclusions porous carbon derived from biomass waste for high-performance supercapacitor,
Bioresource Technol. 197 (2015) 137–142.
[4] G. Chen, T. Chen, K. Hou, Robust, hydrophilic graphene/cellulose nanocrystal fiber-
In conclusion, this study puts forward a facile, economical and ef-
based electrode with high capacitive performance and conductivity, Carbon 127
ficient strategy for pore development and stability of coal converted (2018) 218–227.
activated carbon. The synthesis is achieved by adding very small amunt [5] J.F. Wang, L.J. Jin, Y. Zhou, Effect of Ca(NO3)2 addition in coal on properties of
of K2CO3 (less than 2 wt% weight ratio of coal simple) followed by CO2 activated carbon for methane decomposition to hydrogen, Fuel Process. Technol.
176 (2018) 85–95.
activation. The potassium-base component embedded in the coal matrix [6] G. Wang, S. Chen, X. Quan, Enhanced activation of peroxymonosulfate by nitrogen
acts as a catalyst for intense erosion reaction between coal sample and doped porous carbon for effective removal of organic pollutants, Carbon 115 (2017)
CO2 molecules, thus enhancing pore formation and creation of as-ob- 730–739.
[7] M. Teng, J. Qiao, F. Li, Electrospun mesoporous carbon nanofibers produced from
tained activated carbon. Through experimental analysis and DFT cal- phenolic resin and their use in the adsorption of large dye molecules, Carbon 50
culations, it is verified that the potassium-based catalytic component (2012) 2877–2886.
(mainly C-O-K structure) can be regenerated during CO2 atmosphere at [8] Y.Q. Chen, X. Zhang, W. Chen, H.P. Yang, H.P. Chen, The structure evolution of
biochar from biomass pyrolysis and its correlation with gas pollutant adsorption
high temperature, so that adding very small amount of K2CO3 can cause performance, Bioresource Technol. 246 (2017) 101–109.
a continuous catalytic effect. The optimized Ca_AC-1 possesses a short- [9] S. Deng, H. Wei, T. Chen, Superior CO2 adsorption on pine nut shell-derived acti-
range ordered microcrystalline structure and a developed pore structure vated carbons and the effective micropores at different temperatures, Chem. Eng. J.
253 (2014) 46–54.
with the highest specific surface area of 1773 m2/g and pore volume of [10] L.Y. Wang, X.X. Cheng, Z.Q. Wang, C.Y. Ma, Investigation on Fe-Co binary metal
1.11 cm3/g, three times more than Ph_AC from solely CO2 activation oxides supported on activated semi-coke for NO reduction by CO, Appl. Catal. B
and even more than Ch_AC prepared by chemical activation with large Environ. 201 (2016) 636–651.
[11] X. Xu, J. Gao, Q. Tian, Walnut shell derived porous carbon for a symmetric all-solid-
dosage of K2CO3. Evaluated as adsorbents for small CO2 molecule,
state supercapacitor, Appl. Surf. Sci. 411 (2017) 170–176.
Ca_AC-1 exhibits high CO2 adsorption capacity. Evaluated as super- [12] L.J. Wang, F. Sun, J.H. Gao, Y.W. Zhu, Pore reorganization of porous carbon during
capacitor electrode materials, Ca_AC-1 exhibits superior performances trace calcium-catalyzed coal activation for adsorption applications, Energy Fuel 32
in both aqueous and organic systems, particularly superior rate cap- (2018) 9191–9201.
[13] H.S. Chavan, B. Hou, A.T.A. Ahmed, Y.J. Jo, S. Cho, J. Kim, S.M. Pawar, S.N. Cha,
ability and cycling stability. Due to the low-cost carbon source, low- A.I. Inamdar, H. Im, H. Kim, Nanoflake NiMoO4 based smart supercapacitor for
dose catalyst and simple preparation process, it offers a simple, green intelligent power balance monitoring, Sol. Energy Mat. Sol. Cell 185 (2018)
and scalable preparation strategy for high-performance coal-based ac- 166–173.
[14] A.I. Inamdar, J. Kim, Y. Jo, H. Woo, S. Cho, S.M. Pawar, S. Lee, J.L. Gunjakar,
tivated carbon, and such strategy can be easily applied for other carbon Y. Cho, B. Hou, S.N. Cha, J. Kwak, Y. Park, H. Kim, H. Im, Highly efficient electro-
sources. optically tunable smart-supercapacitors using an oxygen-excess nanograin tungsten
oxide thin film, Sol. Energy Mat. Sol. Cell 166 (2017) 78–85.
[15] S.Y. Tao, Y.C. Wang, D. Shi, Facile synthesis of highly graphitized porous carbon
Acknowledgments monoliths with a balance on crystallization and pore-structure, J. Mater. Chem. A 2
(2014) 12785–12791.
[16] B. Hu, K. Wang, L. Wu, Engineering carbon materials from the hydrothermal car-
This work was financially supported by National Key Research and
bonization process of biomass, Adv. Mater. 2 (2010) 813–828.
Development Project of China (grant numbers 2017YFB0602900). [17] K. Chen, C. Li, L. Shi, Growing three-dimensional biomorphic graphene powders
Moreover, the calculation was carried out at National Supercomputer using naturally abundant diatomite templates towards high solution processability,
Nature Commun. 7 (2016) 13440.
Center in Tianjin on TianHe-1(A) instrument.
[18] L. Fang, Y.P. Xie, Y.Y. Wang, Z.W. Zhang, P.F. Liu, N. Cheng, Facile synthesis of
hierarchical porous carbon nanorods for supercapacitors application, Appl. Surf.
Appendix A. Supplementary data Sci. 464 (2019) 479–487.
[19] X. Xu, Y. Wang, C. Bing, Phosphorus-doped porous graphene nanosheet as metal-
free electrocatalyst for triiodide reduction reaction in dye-sensitized solar cell, Appl.
Supplementary data to this article can be found online at https:// Surf. Sci. 405 (2017) 308–315.
doi.org/10.1016/j.cej.2019.123205. [20] P.J.M. Carrott, M.M.L.R. Suhas, C.I. Carrott, L.A. Guerrero, Delgado, Reactivity and

12
L. Wang, et al. Chemical Engineering Journal xxx (xxxx) xxxx

porosity development during pyrolysis and physical activation in CO2 or steam of [39] Y. Zhao, D.G. Truhlar, Exploring the limit of accuracy of the global hybrid meta
kraft and hydrolytic lignins, J. Analy. Appl. Pyroly. 82 (2008) 264–271. density functional for main-group thermochemistry, kinetics, and noncovalent in-
[21] X. He, R. Li, J.S. Qiu, Synthesis of mesoporous carbons for supercapacitors from coal teractions, J. Chem. Theory. Comput. 4 (2008) 1849–1868.
tar pitch by coupling microwave-assisted KOH activation with a MgO template, [40] Q. Liang, L. Ye, Z.H. Huang, A honeycomb-like porous carbon derived from pomelo
Carbon 50 (2012) 4911–4921. peel for use in high-performance supercapacitors, Nanoscale 6 (2014)
[22] K.Y. Foo, B.H. Hameed, Preparation and characterization of activated carbon from 13831–13837.
sunflower seed oil residue via microwave assisted K2CO3 activation, Bioresource [41] Z.J. Li, W.Y. Zhang, Y.C. Li, Activated pyrene decorated graphene with enhanced
Technol. 102 (2011) 9794-7799. performance for electrochemical energy storage, Chem. Eng. J. 334 (2017)
[23] D. Changchao, W. Jiafeng, J. Yang, H3PO4, solution hydrothermal carbonization 845–854.
combined with KOH activation to prepare Argy wormwood-based porous carbon for [42] D.D. Liu, J.H. Gao, Q.X. Cao, S.H. Wu, Y.K. Qin, Improvement of activated carbon
high-performance supercapacitors, Appl. Surf. Sci. 444 105 (2018) 117. from Jixi bituminous coal by air Preoxidation, Energy Fuels 31 (2017) 1406–1415.
[24] M. João, N. Valente, P. Nunes, P.J.M. Carrott, Production of Activated Carbons from [43] J. Dı́azTerán, D.M. Nevskaia, J.L.G. Fierro, Study of chemical activation process of a
Coffee Endocarp by CO2 and Steam Activation, Fuel Process. Technol. 89 (2008) lignocellulosic material with KOH by XPS and XRD, Micropor. Mesopor. Mat. 60
262–268. (2003) 173–181.
[25] Y.H. Bai, P. Lv, X.H. Yang, M.Q. Gao, S.H. Zhu, L.J. Yan, F. Li, Gasification of coal [44] Y.B. Xi, D.J. Yang, X.Q. Qiu, Renewable lignin-based carbon with a remarkable
char in H2O/CO2 atmospheres: Evolution of surface morphology and pore structure, electrochemical performance from potassium compound activation, Ind. Crop.
Fuel 218 (2018) 236–246. Prod. 124 (2018) 747–754.
[26] Y.L. Wang, S.H. Zhu, M.Q. Gao, A study of char gasification in H2O and CO2 mix- [45] J. Kopyscinski, R. Habibi, C.A. Mims, J.M. Hill, K2CO3-catalyzed CO2 gasification of
tures: role of inherent minerals in the coal, Fuel Process. Technol. 141 (2016) 9–15. ash-free coal: Kinetic study, Energy Fuels 27 (2013) 4875–4883.
[27] J.T. Wei, Y. Gong, L. Ding, J.Q. Yu, G.S. Yu, Influence of biomass ash additive on [46] D.M. Quyn, J.I. Hayashi, C.Z. Li, Volatilisation of alkali and alkaline earth metallic
reactivity characteristics and structure evolution of coal char-CO2 gasification, species during the gasification of a Victorian brown coal in CO2, Fuel Process.
Energy Fuels 32 (2018) 10428–10436. Technol. 86 (2005) 1241–1251.
[28] J.T. Wei, Q.H. Guo, L. Ding, Y. Gong, J.Q. Yu, G.S. Yu, Understanding the effect of [47] J. Kopyscinski, M. Rahman, R. Gupta, J.M. Hill, K2CO3 catalyzed CO2 gasification of
different biomass ash additions on pyrolysis product distribution, char physico- ash-free coal: Interactions of the catalyst with carbon in N2 and CO2 atmosphere,
chemical characteristics, and char gasification reactivity of bituminous coal, Energy Fuel 117 (2014) 1181–1189.
Fuels 33 (2019) 3068–3076. [48] A. Miyakoshi, A. Ueno, M. Ichikawa, XPS and TPD characterization of manganese-
[29] L. Peng, Y. Bai, X. Yang, F. Li, Impacts of char structure evolution and inherent substituted iron–potassium oxide catalysts which are selective for dehydrogenation
alkali and alkaline earth metallic species catalysis on reactivity during the coal char of ethylbenzene into styrene, Appl. Catal. A Gen. 219 (2001) 249–258.
gasification with CO2/H2O, Int. J. Energy Res. 42 (2018) 3633–3642. [49] H. Kim, A. Watthanaphanit, N. Saito, Synthesis of colloidal MnO2 with sheet-like
[30] J.B. Zhou, X.G. Zhuang, A. Alastuey, X. Querol, J.H. Li, Geochemistry and miner- structure by one-pot plasma discharge in permanganate aqueous solution, RSC Adv.
alogy of coal in the recently explored Zhundong large coal field in the Junggar 6 (2015) 2826–2834.
basin, Int. J. Coal Geol. 82 (2010) 51–67. [50] S. Selvakumar, N. Nuns, M. Trentesaux, V.S. Batrab, J.M. Giraudon, Reaction of
[31] X. Su, H. Jin, L. Guo, S. Guo, Z. Ge, Experimental study on zhundong coal gasifi- formaldehyde over birnessite catalyst: a combined XPS and ToF-SIMS study, Appl.
cation in supercritical water with a quartz reactor: reaction kinetics and pathway, Catal. B: Environ. 223 (2018) 192–200.
Int. J. Hydrogen Energy 40 (2015) 7424–7432. [51] X.X. Pi, F. Sun, J.H. Gao, Z.B. Qu, A new insight into the SO2 adsorption behavior of
[32] F. Sun, J.H. Gao, Y.Q. Yang, One-step ammonia activation of Zhundong coal gen- oxidized carbon materials using model adsorbents and DFT calculations, Phys.
erating nitrogen-doped microporous carbon for gas adsorption and energy storage, Chem. Chem. Phys. 21 (2019) 9181–9188.
Carbon 109 (2016) 747–754. [52] A. Alabadi, S. Razzaque, Y. Yang, Highly porous activated carbon materials from
[33] L.J. Wang, F. Sun, J.H. Gao, X.X. Pi, Z.B. Qu, G.B. Zhao, Adjusting the porosity of carbonized biomass with high CO2 capturing capacity, Chem. Eng. J. 281 (2015)
coal-based activated carbons based on a catalytic physical activation process for gas 606–612.
and liquid adsorption, Energy Fuels 32 (2018) 1255–1264. [53] M.G. Plaza, A.S. González, C. Pevida, J.J. Pis, F. Rubiera, Valorisation of spent
[34] T. Pei, F. Sun, J.H. Gao, L.J. Wang, Introducing catalytic gasification into chemical coffee grounds as CO2 adsorbents for postcombustion capture applications, Appl.
activation for the conversion of natural coal into hierarchically porous carbons with Energy 99 (2012) 272–279.
broadened pore size for enhanced supercapacitive utilization, RSC Advances 8 [54] M.G. Plaza, A.S. González, J.J. Pis, F. Rubiera, C. Pevida, Production of microporous
(2018) 37880–37889. biochars by single-step oxidation: Effect of activation conditions on CO2 capture,
[35] F. Sun, H.B. Wu, X. Liu, Nitrogen-rich carbon spheres made by a continuous Appl. Energy 114 (2014) 551–562.
spraying process for high-performance supercapacitors, Nano Research 9 (2016) [55] D.P. Vargas, L. Giraldo, A. Erto, Chemical modification of activated carbon mono-
3209–3221. liths for CO2 adsorption, J. Therm. Anal. Calorim. 114 (2013) 1039–1047.
[36] Gaussian 09, Revision D 01, M.J. Frisch, G.W. Trucks, H.B. Schlegel, Gaussian, Inc., [56] N. Guo, M. Li, Y. Wang, Soybean root-derived hierarchical porous carbon as elec-
Wallingford CT, 2013. trode material for high-performance supercapacitors in ionic liquids, ACS Appl.
[37] Y. Zhao, D.G. Truhlar, The M06 suite of density functionals for main group ther- Mater. Inter. 8 (2016) 33626–33634.
mochemistry, thermochemical kinetics, noncovalent interactions, excited states, [57] P. Cheng, S. Gao, P. Zang, Hierarchically porous carbon by activation of shiitake
and transition elements: two new functionals and systematic testing of four M06- mushroom for capacitive energy storage, Carbon 93 (2015) 315–324.
class functionals and 12 other functionals, Theor. Chem. Account. 120 (2008) [58] G.P. Hao, A.H. Lu, W. Dong, Z.Y. Jin, X.Q. Zhang, J.T. Zhang, W.C. Li, Sandwich-
215–241. type microporous carbon nanosheets for enhanced supercapacitor performance,
[38] Y. Zhao, D.G. Truhlar, Density functionals with broad applicability in chemistry, Adv. Energy Mater. 3 (2013) 1421–1427.
Acc Chem. Res. 41 (2008) 157–167.

13