- Email: [email protected]
Metal-organic-framework derived carbon polyhedron and carbon nanotube hybrids as electrode for electrochemical supercapacitor and capacitive deionization
Accepted Manuscript Metal-organic-framework derived carbon polyhedron and carbon nanotube hybrids as electrode for electrochemical supercapacitor and capacitive deionization Tie Gao, Feng Zhou, Wei Ma, Haibo Li PII:
S0013-4686(18)30062-8
DOI:
10.1016/j.electacta.2018.01.044
Reference:
EA 31026
To appear in:
Electrochimica Acta
Received Date: 27 November 2017 Revised Date:
5 January 2018
Accepted Date: 8 January 2018
Please cite this article as: T. Gao, F. Zhou, W. Ma, H. Li, Metal-organic-framework derived carbon polyhedron and carbon nanotube hybrids as electrode for electrochemical supercapacitor and capacitive deionization, Electrochimica Acta (2018), doi: 10.1016/j.electacta.2018.01.044. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Metal-Organic-Framework Derived Carbon Polyhedron and Carbon Nanotube Hybrids as Electrode for Electrochemical Supercapacitor and Capacitive Deionization Tie Gaoa, Feng Zhouab, Wei Maa, Haibo Liab* a. Ningxia Key Laboratory of Photovoltaic Materials, Ningxia University, Yinchuan, Ningxia,
RI PT
750021, P. R. China b. State Key Laboratory of High-efficiency Utilization of Coal and Green Chemical Enginering, School of Chemistry and Chemical Engineering, Ningxia University, Yinchuan, Ningxia, 750021, P. R. China
AC C
EP
TE D
M AN U
SC
∗ Corresponding author. Fax/Tel: +86 0951 2062414; E-mail: [email protected] (Haibo Li)
1
ACCEPTED MANUSCRIPT Abstract In this work, carbon polyhedron and carbon nanotube hybrids (HCN) have been synthesized by employing ZIF-67 as precursor for electrochemical supercapacitor and capacitive deionization (CDI). Basically, uniform polyhedral nanocrystals of ZIF-67 were firstly fabricated, and then they
RI PT
were directly subjected to chemical vapor deposition for growing carbon nanotubes. It is found that the as-fabricated HCN electrode exhibits remarkable electrochemical performance, i.e. the highest capacitance of 343 F·g-1 at the scan rate of 10 mV·s-1 and excellent rate capability. This is due to that HCN can provide rich space and short ion diffusion path. Moreover, the HCN
SC
electrodes shows superior CDI performance, i.e. the electrosorption capacity reached as high as 7.08 mg·g-1 and the adsorption rate of 0.03958 mg·g-1·min-1. Metal
organic
frameworks,
Carbon
polyhedron,
M AN U
Keywords:
AC C
EP
TE D
Supercapacitor
2
Capacitive
deionization,
ACCEPTED MANUSCRIPT 1. Introduction Portable water and energy storage are absolutely essential to promote the ongoing of our society [1-3]. Fast charge-discharge supercapacitors are attractive energy storage technologies as a result of their high power capability, good operational safety and long cycling life compared to
RI PT
secondary batteries, of which the performance is greatly governed by the electrode materials [4-7]. On the other hand, capacitive deionization (CDI), as an emerging promising desalination technology, is considered to be a cost-effective, environmentally friendly and energy-efficient in purification with low-pressure pumps, non-membrane, etc [8-11]. The mechanism of CDI is
SC
similar to supercapacitor which based on electrical double layer theory. Compared with other desalination technologies, CDI is an efficient and energy-saving method profit from salty ions
M AN U
which is the minority one immobilized in the double layer between the surface of electrodes and solution. On the contrary, other methods such as distillation extract the majority phase, the water, from the seawater then much additional energy would be wasted [12-16]. Porous carbon materials have attracted intense attention due to their unique properties, including high surface area, large pore volume, and unique pore size distribution, which have
TE D
great potential for electrochemical applications [17-22]. Metal-organic frameworks (MOFs) have been regarded as a promising class of advanced materials for gas adsorption, sensing, supercapacitors, and catalysis, owing to their crystalline structures with high porosities and
EP
tunable functionalities [23-30]. In particular, due to their porous structures with high surface areas, MOFs are ideal sacrificial templates to derive into various porous carbon and other active materials, such as nanocarbons metal oxides, metal sulfides, metal carbides, metal phosphides and
AC C
their composites [31, 32]. Basically, direct pyrolysis of MOFs represents an efficient strategy to produce nanostructured carbon and related composite. Yamauchi et al. prepared nanoporous carbon though thermal pyrolysis of zeolitic immidazolate frameworks (ZIF-8) for supercapacitors, and a high specific capacitance of 214 F·g-1 at a scan rate of 5 mV·s-1 was achieved [33]. Pan carried out pioneer work on ZIF-8 derived porous carbon polyhedrons (PCPs) for CDI, exhibiting the high electrosorption capacity of 13.86 mg·g-1 when the initial NaCl concentration is 500 mg·L-1, due to their high accessible surface area and low charge transfer resistance [34]. Yang et al. designed bimetallic metal-organic frameworks (BMOFs) with different molar ratios of Zn and Co based on ZIF-8 and ZIF-67, showing the porous carbon derived from a BMOF of Zn: Co = 3:1 3
ACCEPTED MANUSCRIPT exhibits an untra-high salt removal capacity of 45.62 mg·g-1 in a 750 mg·L-1 NaCl solution at 1.4 V which is due to the large ion-accessible surface area and improved electrical conductivity of the carbon derived from BMOF [35]. However, the prospect of pure porous carbon materials derived from direct pyrolysis of MOFs in energy and water fields is still obstructed owing to the intrinsic
RI PT
properties of carbon materials, for instance, low conductivity. In this case, to explore novel electrode materials with satisfying electrochemical properties, porous carbon materials have been incorporated with other active materials. Generally, the porous carbon based composites have many special features to achieve enhanced properties. For instance, the carbon frameworks
SC
always provide facile electron transportation so that high electronic conductivity can be achieved while the active materials confined in the carbon matrix endow improved performance, which can
M AN U
be attributed to the synergetic enhancement. The porous feature of these composites leads to a facile electrolyte infiltration so that a high electrode/electrolyte interface can be obtained. Therefore, the porous carbon-based composites have many unique features and can achieve excellent performance for electrochemical applications.
Carbon nanotubes (CNTs), a 1D allotrope of carbon with extraordinary properties, have been
TE D
widely investigated in recent decades. In relation to supercapacitor and CDI, either CNTs or CNTs composite as electrode has been extensively explored due to its controllable physio-chemical properties [36, 37]. In the present work, we report the fabrication of carbon-based HCN derived
EP
from growing CNTs on ZIF-67, a well-known subfamily of MOFs, for both supercapacitor and CDI. The HCN feature with closely packed tube-box structures: a ZIF-67-derived dodecahedron of porous carbons (PCs) and chemical vapor deposited CNTs linked with each PCs as a bridge. In
AC C
this kind of novel structure, the PC with high accessible surface area is responsible for providing huge space to accommodate ions. Besides, cobalt derived from ZIF-67 is prone to be oxidized as CoxOy which is highly active to introduce the pseudo-reaction and therefore contribute to high electrochemical performance. On the other hand, CNTs act as conducting wire to facilitate the diffusion kinetics. To prove our analysis, the potential and superior performances of HCN electrode are demonstrated in liquid and solid supercapacitor as well as CDI. 2 Experimental 2.1 Synthesis of HCN 4
ACCEPTED MANUSCRIPT The synthetic strategy of HCN is schematically illustrated in Fig. 1. Cobaltous Nitrate Hexahydrate (Co(NO3)2·6H2O) (99%) was purchased from Macklin. 2-Methylimidazole (Hmim) (98%) was purchased from Aladdin. All chemicals were used as received without any special handling. Uniform polyhedral nanocrystals of ZIF-67 were first fabricated by the coordination of
RI PT
Co2+ and 2-MeIm in methanol. Specifically, methanolic solutions of cobaltous nitrate hexahydrate (6.89 g, 240 mL) and solutions of Hmim (15.57 g, 240 mL) were mixed under stirring when the methanolic solutions of Hmim was kept at 40
. After being stirred for 0.5 h, ZIF-67 nanocrystals
were collected by washing with methanol and followed by centrifuging (11000 rpm, 10 min) for
SC
three times. Finally, the product was dried overnight for ready to use.
The CNTs grown on ZIF-67 was carried in a CVD system. First of all, 0.1g ZIF-67 powder was
M AN U
sent into the chamber and the pressure in quartz tube is pumped to 1×10-2 Pa by the mechanical pump. Next, injecting Ar+4%H2 at the flow rate of 20 sccm and the pressure is kept at a balance of 2.9 Pa. Then the temperatures rose steadily from room temperature to target temperatures (500, 600, 650 and 700
) with a heating rate of 15
·min-1. Once it reaches the target temperature, the
acetylene with flow rate of 20 sccm was injected and kept for 1 hour, the pressure balance at 10
temperature below 200 2.2 Characterization
TE D
Pa during this process. Successfully, the system was cooling in Ar+4%H2 atmosphere until the . Finally, closing the Ar+4%H2 and taking out of sample.
EP
The morphology and fine structures of samples were observed using scanning electron microscope (SEM, Zeiss Supra 40) and transmission electron microscope (TEM, Hitachi
AC C
HT7700). The crystal structures information were identified using an X-ray diffractometer (DX-2700, CHN) with Cu Kα radiation at 40 kV and 30 mA over the 2θ range of 3-80°. The thermogravimetric analyses (TGA) were carried out on a thermogravimetric analyzer (SETARAM Setsys 16) from room temperature to 1000
at a heating rate of 15
·min-1 and under N2 flow
rate of 20 sccm. The Raman spectra were recorded on a spectrometer (DXR, USA) equipped with an optical microscope at room temperature. X-Ray photoelectron spectroscopy (XPS, Thermo Scientific Escalab 250Xi) spectra was collected on a monochromatized Al Kα X-ray source (1361 eV) and the X-ray spot size of 500 µm. Nitrogen adsorption and desorption isotherms were measured at 77K using accelerated surface area and porosimetry system (Tristar 5
3020 USA).
ACCEPTED MANUSCRIPT The wettability was obtained by measuring the contact angle (Powerreach JC200D2, CHN). 2.3 Fabrication of electrode The electrodes tested in electrochemical, CDI and flexible symmetric supercapacitors were fabricated by the following approach. First, the electrode material was fabricated by
RI PT
mixing the 80 wt% of HCN, 10 wt% of super conductive black and 10 wt% of polytetrafluoroethylene (PTFE) slurries. It should be noted that the mixture was vigorously grinded for several minutes until it became homogenous. Subsequently, the mixtures were
SC
daubed onto target shape graphite sheet by a medical blade, followed by drying at 70 for 10h in a vacuum oven.
M AN U
2.4 Electrochemical measurement
The electrochemical test was measured from a three-electrode system in 6M KOH solution using electrochemical workstation (CHI 660D). The working electrode was prepared according to above method. The counter electrode was a platinum wire and the reference electrode was Ag+/AgCl reference electrode. Then, the Cyclic Voltammetry (CV), Galvanostatic Charging-Discharging
TE D
(GCD) and Electrochemical Impedance Spectrum (EIS) were obtained. The specific capacitances were obtained from the CV curves by the following equation: C = ∫ idV /ν∆Vm
(1)
EP
Where C is the specific capacitance (F·g-1), i is the response current (A), v is the potential scan rate (mV·s-1), ∆V is the absolute value of the potential window (V) and m is the mass of HCN (g).
AC C
For comparison, the specific capacitance is also calculated from the GCD curves according to the following equation:
C=
I mt ∆V
(2)
Where Im is the ratio of response current and the mass of activated materials (mA·g-1), t is the discharge time (s) and ∆V is the potential change during the discharge process (V). The energy density E (Wh·Kg-1) and power density P (W·Kg-1) was calculated based on the following equations:
6
ACCEPTED MANUSCRIPT E=
1 C 1 × × V2 × 2 4 3.6
P=
E t
(3) (4)
Where C (F·g-1) is the specific capacitance and V is the voltage changed during discharge period.
RI PT
2.5 CDI experiment and calculation
The CDI performances of HCN electrode were acquired through batch-mode experiment system, which includes a current collector, DC power, conductivity monitor and constant
SC
peristaltic pump. Electrode’s material was adhered to two graphite current collectors and separated from each other by the insulated spacer to avoid short circuit. The total volume of NaCl
M AN U
solution was 40 mL and drove by the constant peristaltic pump under a flow rate of 34 mL·min-1. The adsorption capacity of electrode’s material was evaluated by adsorption NaCl from aqueous solution at an initial conductivity of 1000 µS·cm-1 respectively. A direct voltage of 0.6-1.4 V was applied between electrodes. At the moment, the adsorption ability of the electrode was evaluated by electrosorption capacity (Γ, mg·g-1).Theoretically, the adsorption capacity was obtained from the conductivity transients curves by the following equation: (G0 − Gt )V 2m
TE D
Γ=
(5)
where G0 and Gt are the initial and final conductivity (µS·cm-1), V is the total volume of NaCl
EP
solution (L) and m is the mass of HCN (g) [38].
The simulation on CDI performance of HCN electrode is performed by employing the modified
AC C
Donnan model (mD-model) which has been proved to be valid for batch-mode experiment [9, 39-41]. In this model, the water leaving the CDI-cell is flowing to a recycle vessel and from there fed back into the CDI-cell. It also assumes that throughout the cell the salt concentration is the same everywhere, and is the same as in the recycle vessel. A simple description of the mD-model has been given in supporting information. Based on above method, we enter the following key dates to the program: a = 7.2 (m2); V = 4.0×10-5 (m3); C0 = 9.223 (mol·m-3); 7
ACCEPTED MANUSCRIPT A = 3.85×10-3 (m2); K =1.455×10-8 (m·s-1); Vcell = (0.6, 0.8, 1.0, 1.2 and 1.4V); Cst = 0.5 (F·m-2).
RI PT
2.6 Assemble of flexible supercapacitors The preparation of LiCl/polyvinyl alcohol (PVA) gel electrolyte is as follows. In a typical run, LiCl (8.4g) and PVA (4.0 g) powders were added into DI water (40 mL). The mixture was stirred till the solution became clear, followed by natural cooling to room temperature.
SC
at 90
Successfully, the electrodes for solid state flexible supercapacitor were fabricated by drop casting
drying at 70
M AN U
the HCN suspension on graphite sheet with geometrical surface area of 1×2 cm2, followed by for overnight. Finally, the two electrodes were immersed into the LiCl/PVA gel
electrolyte for 5 min. Then, the separator was sandwiched by the two electrodes to form the flexible symmetric supercapacitor [42]. 3 Results and Discussion
TE D
3.1 Morphology and structure
Fig. 2(a) and (b) show the typical SEM images of ZIF-67 and HCN obtained at 650 (HCN-650). Overall, the size and polyhedral shape of the ZIF-67 nanoparticles are retained well
EP
after CVD treatment, while the surface of the particles becomes much rougher. A large amount of long CNTs, several hundred nanometers in length, embedded in ZIF-67 derived porous carbon
AC C
can be observed, forming a network structure. The formation of CNTs is the result of the catalytic effect of Co nanoparticles on the particle surface. Further, the effect of growth temperature on HCN has been examined. Fig. 2(c) to (f) exhibited the HCN prepared at 500, 600, 650 and 700
,
respectively, and inset in each figure is the magnified image of CNTs corresponding to certain temperature. Regardless of temperature, a well polyhedral shape of ZIF-67 is maintained. Moreover, high temperature results in much rougher surface. On the other hand, it seems that a more clear and hollow structure of HCN is confirmed. More importantly, one can observed from the inset of image that the diameter of CNTs is going to increase with the rise of growth temperature. A statistic result of diameter distribution for CNTs growth at different temperature is 8
ACCEPTED MANUSCRIPT given in Fig. S1. Since the diameter of CNTs is closely related to the size of Co, we have examined the morphology of Co particle at different temperature. As shown in Fig. S2, as the temperature increase, the size of Co is became bigger and thus leads to CNTs with bigger diameter. In addition, the TEM image also reveals the presence of Co nanoparticles encapsulated
RI PT
by a few-layered carbon shell (Fig. S3), especially at the tip of the CNTs. It should be mention that these carbon-encased Co nanoparticles are inaccessible to reactants as they remain completely encapsulated within the graphitic carbon shell even after acid leaching.
The crystalline nature of HCN is further carried out by XRD in Fig. 3(a). The metallic Co is
SC
confirmed by the three typical peaks at 44.2°, 51.6°, and 75.9° respectively, consistent with (111), (200), and (220) planes of Co (PDF#89-4307) [43]. This is due to that metallic Co nanoparticles
M AN U
are quickly formed in the presence of H2 atmosphere during CVD process. Besides, the typical peak at 22°corresponding to the (002) diffraction mode of the graphic structure is revealed for all HCN samples. Moreover, the degree of graphitization is strengthened with the increase of growth temperature. Fig. 3(b) draws the TGA curve of ZIF-67. The mass loss is concentrated at 500 700
to
which is mainly caused by thermal decomposition and recombination of organic ligands
TE D
during carbonization process. Raman spectroscopy is essential for the structural analysis which is illustrated in Fig. 3(c). The vibration peaks at 188, 476, 520, and 687 cm−1 in terms of all HCN samples match with the Co3O4 phase which is in good agreement of literature [44]. Three 1360,
1580 and 2680 cm−1, which are separately unique characteristic
EP
additional bands at
D-band, G-band and G* band associated with carbon materials. The intensity ratio of D-band over G-band (ID/IG) is usually used to evaluate the degree of graphitization of carbon materials and as
AC C
the ratio decreases, the graphitization degree is higher [44, 45]. Obviously, the ID/IG is decreased from 1.05 to 0.66 as temperature increased from 600 to 700
, implying high graphitized CNTs is
facile to acquire at high growth temperature. The chemical composition is investigated by XPS spectra. The whole spectra of HCN samples are illustrated Fig. 3(d) and the corresponding atomic percent is summarized in Tab. 1. When the temperature increased from 600
to 700
, the C1s
ratio has not varied much. Meanwhile, the Co 2p ratio is very rare which is only occupy
3%.
Besides, N1s has a mild ratio which would beneficial to improve the conductivity of HCN due to the doping effect. Importantly the high-resolution C 1s spectrum can be deconvolved into several bonds (Fig. S4(a)), corresponding to Sp2 hybrid carbon (284.6 eV), C-C (285.2 eV), C-N (285.7 9
ACCEPTED MANUSCRIPT eV), C-O (286.4) and O-C=O (288.7 eV). It is noted that the presence of O signal can be attributed to residual oxygen in ZIF derived carbon and the adsorbed oxygen from environmental atmosphere. The high-resolution O 1s spectrum can be deconvolved into three bands (Fig. S4(b)), corresponding to C=O (530.8 eV), C-O-C (532 eV) and O-C=O (533.4 eV) [46]. The
at
398.5 eV and pyrrolic N at
RI PT
high-resolution N1s spectrum reveals the presence of two types of nitrogen species, pyridinic N 400.8 eV (Fig. S4(c)), together with the identification of C-N
bonds at 285.1 eV, indicating the N doping in HCN which is provided by the ZIF-67 from the pyrolysis [47]. Significantly, the N doping may beneficial to enhance the conductivity of HCN
SC
and therefore the electrochemical behavior. Further, high-resolution XPS spectra in the Co 2p region are shown in Fig S4(d). The first peak presented at 778.5 eV is assigned to be the Co 2p3/2
M AN U
while the peak located at 793.5 eV can be attributed to Co 2p1/2. Importantly, the 2p3/2 appears at 780.3 eV and 782.1 eV for Co3+ and Co2+ while 2p1/2 appears at 795.3 eV and 798 eV for Co3+ and Co2+. In addition, there are two small peaks at 787.9 eV and 803.6 eV represent shake-up peaks of Co2+ [48, 49]. The simultaneously exist of Co3+ and Co2+ indicating the nanoparticles contain Co3O4, but the area ratio between peaks of Co3+ and Co2+ below 2:1 that confirmed there
TE D
are superfluous Co2+ for Co3O4 which are exist in CoO exactly.
The N2 adsorption-desorption isotherm and pore size distribution curves are shown in Fig. 3(e) and (f). The specific values associated with pore texture are summarized in Tab. 2. Basically, the isotherm, in
EP
N2 adsorption-desorption isotherm of all HCN samples were belong to type
which the isotherms with sharp increase in volume at low relative pressure imply the presence of micropores, mainly from CNTs. The hysteresis loop at relative pressure between 0.4 and 0.9
AC C
suggests the coexistence of mesopores from porous carbon derived from ZIF-67. The pore size distribution reveals that the micropores are smaller than 1 nm in size while the mesopores are around 100 nm. Moreover, the HCN-650 have a Brunauer-Emmett-Teller (BET) surface area of 172 m2·g-1 and total volume of
0.17 cm3·g-1, both of which are highest among all samples.
In addition, the wettability of HCN-650 electrode has been tested (Fig. S5), exhibiting that the contact angle is
91° which confirms the hydrophilicity of HCN-650 electrode.
3.2 Electrochemical performance Based on the above discussions, the resulted porous carbon nanostructures can not only provide 10
ACCEPTED MANUSCRIPT sufficient reaction sites, short ion diffusion length, buffer the volume change, but also ensure direct mechanical and electrical connection with the flexible carbon support, making it a very promising electrode material for aqueous and flexible energy storage applications. To evaluate the electrochemical performance, the electrochemical activity of various HCN samples toward
RI PT
supercapacitor was first identified by CV tests conducted in saturated 6M KOH electrolyte via three electrodes method. Fig. 4(a) provide the representative CV curve of HCN derived from 500, 600, 650 and 700
at scan rate of 10 mV·s-1. Apparent redox peaks can be observed from the
CV curves, which can be assigned for the Faradic reaction between Co3+/Co4+ with OH−.
SC
From CV curves of all HCN samples (Fig. S6), the peak positions only shift a little when the scan rate increased from 10 to 100 mV·s−1, indicating that the electrode has excellent reversibility.
M AN U
The specific capacitance as a function of scan rate for all HCN electrodes is given in Fig. 4(b). Among all HCN electrodes, HCN-650 has the highest capacitance at each certain scan rate. At a low scan rate of 10 mV·s−1, HCN-650 exhibits a high capacitance of 343 F·g-1 and a high capacitance of 43 F·g-1 can be maintained even when the scan rate increased to 1000 mV·s−1, suggesting that 13% of the capacitance can be remained even when the scan rate increased 100
TE D
times. By compared with other HCN samples, the enhanced capacitance and rate capability of the HCN-650 can be attributed to the unique porous structures and high BET surface area as well as pore volume that can provide sufficient reaction cites and short ion diffusion length. The GCD curves for HCN electrode derived from 500, 600, 650 and 700
at the current density of 0.8
EP
A·g-1 are shown in Fig. 4(c), where all the curves have almost linear charge-discharge characteristics with two slight plateau located at -0.4 and -0.6 V. When the current density further
AC C
increased, a similar GCD curves can be obtained (Fig. S7). Fig. 4(d) presents the specific capacitance with respect to current density calculated from GCD curve. At each current density, HCN-650 reveals the highest capacitance. At current density of 0.5 A·g-1, a high capacitance of 190 F·g-1 can be obtained. Remarkably, when the current density increased to 2 A·g-1, the capacitance is as high as 110 F·g-1, implying 58% remaining. Besides, Fig. S8 shows Ragone plots of HCNs corresponding to the relationship between energy and power densities. The maximum energy density of HCN-650 reach 6.6 Wh·kg-1 at the current density of 0.5 A·g-1. The Nyquist plot of the EIS and corresponding equivalent circuit are drawn in Fig. 4(e) and (f), respectively. A frequency response analysis at open circuit potential over the frequency range 11
ACCEPTED MANUSCRIPT from 100 KHz to 10 mHz yielded the Nyquist plots. Commonly, the spectrum consists of a high frequency semicircle and a low frequency tail. In the equivalent circuit, the small L are used to reduces the influence of alternating current, Rs includes the intrinsic resistance of the electrode active material, the contact resistance at the interface between the electroactive material and the
RI PT
current collector and the resistance of electrolyte solution, Rct is the charge transfer resistance at the electrode/electrolyte interface, Cdl is the double layer capacitance, Cps is the pseudo capacitance [50, 51]. Tab. 3 provides the analogue value for each parameter. In the high-frequency region, the intercept of X axis for HCN-700 is smallest among all samples, showing that the bulk
SC
resistance (Rs) of HCN-700 is smallest. This is inspired us that the bulk resistance would be decreased as the growth temperature increasing which is probably due to the high crystallization
M AN U
of HCN. Moreover, the lowest charge transfer resistance corresponding to HCN-650 is acquired. In the low-frequency region, a more vertical line for the HCN-650 is observed, indicating that the HCN-650 have lowest ionic diffusion resistance among all samples.
To further demonstrate the potential application of the HCN for flexible energy storage devices, a flexible supercapacitor was assembled with a PVA-LiCl solid-state electrolyte. Fig. S9 exhibits
TE D
a digital image of brighten LED lighted by flexible supercapacitors based on two HCN-650 electrode connected in series. The CV curves of the flexible supercapacitor are shown in Fig. S10(a), in which quasi-rectangular curves without sharp redox peaks can be observed, showing
EP
that full cell has good capacitive properties. The GCD curves of the flexible supercapacitor are shown in Fig. S10(b), where all the curves have almost linear GCD characteristics. Basically, the specific capacitance of HCN-650 flexible supercapacitor calculated based on CV curves reaches
AC C
0.08 F·cm-2 at the scan rate of 10 mV·s-1. Importantly, it remained about 10% (0.008 F·cm-2) when the scan rate increase to 1000 mV·s-1. On the other hand, the specific capacitance calculated based on GCD curves even up to 0.05 F·cm-2 at the current density of 0.1 A·g-1. When the current density increased to 0.5 A·g-1, the capacitance is as high as 0.025 F·cm-2, implying 50% remaining. Apparently, the excellent rate capability is in good agreement with that in KOH electrolyte via three electrodes’ method. 3.3 CDI performance In order to evaluate the desalination performance, batch-mode experiment using symmetric CDI in the feed water with fixed initial concentrations at electrical potentials of 12
ACCEPTED MANUSCRIPT 1.2 V were carried out. Fig. 5(a) shows the curve of conductivity change with time under 1.2 V at the initial conductivity of 1000 µS·cm-1. The simultaneous change of voltage and current in adsorption and desorption process was observed in Fig. 5(b). In the initial stage of adsorption, a voltage of 1.2 V was applied, the conductivity decreases rapidly which is
RI PT
due to the fact that the open porous surface was quickly occupied by salty ions. After 100 minutes adsorption, the decrease on conductivity gets slowly which is tends to reach a steady value, suggesting the saturation. During this process, the current decreases rapidly as the adsorption commences until it arrived at an ultra-small value due to the presence of
SC
leakage current. Once the circuit was shorted, adsorbed ions were released from the electrode and the conductivity recovered to the initial level gradually. Fig. 5(c) shows the
M AN U
concentration varies with time under voltage ranging from 0.8 to 1.4 V, together with theoretical curve simulated according to mD-model. The concentration decreased gradually as the potential was increased, which is due to the increase of the electrostatic force. High cell voltage resulted in high electrosorption capacity. Significantly, the experimental dates are in good agreement with the theoretical value at each cell voltage. In
which is
7.2 m2
TE D
details, the effective electrode area derived from the simulation is approximately
32% of the BET value (172m2·g-1 × 0.13g) suggesting that quite a
significantportion of the measured area is valid for electrosorption. Moreover, the Stern
EP
layer capacity Cst is as high as 0.5 F· m-2 indicating that the adsorption process is dominantly governed by electrical double layer. Additionally It should be noticed that the adsorption capacity reached as high as 7.08 mg·g-1 under 1.4 V, which show high potential
AC C
in application to CDI. Even if the cell potential decreased to 0.6 V it still remain a remarkable adsorption capacity of 1.66 mg·g-1. On the other hand, adsorption kinetics is also investigated using Lagergren’s pseudo-first-order adsorption kinetics as shown below:
log(qe − q ) = log(qe ) −
K1t 2.303
(6)
Where qe (mg·g-1) and q (mg·g-1) are the amount of NaCl absorbed at equilibrium and time t, respectively. K1 is the adsorption rate constant of pseudo-first-order equation [13]. The fitting curves are obtained with high regression coefficients as shown in Fig. 5(d) and the relevant coefficients are listed in Tab. 4. The high value of R2 (above 0.986) indicating the 13
ACCEPTED MANUSCRIPT experiment data corresponds to Lagergren’s pseudo-first-order adsorption kinetics regardless of cell voltage. Fig. S11 depicts the rate as a function of capacitance with the voltage between 0.6 and 1.4 V in a 500 mg/L NaCl solution. It is clear that at low voltage of 0.6 V, the HCN-650 have the lowest capacitance. When the voltage increased from 0.6
RI PT
to 1.4 V, the capacitance increases as well. Meanwhile, the Ragone Kim-Yoon-Plot turn righter and upper, suggesting the increase of both the capacitance and adsorption rate. 4 Conclusions
SC
In summary, we synthesized a novel carbon polyhedron and carbon nanotube hybrids (HCN) by chemical vapor deposition approach using ZIF-67 as the precursor directly. Basically, HCN
M AN U
exhibited typical network structure, within which the polyhedron porous carbon were linked tightly by ultra-long carbon nanotubes. By examining the effect of growth temperature on HCN, it is found that the HCN synthesized at 650 area of
(HCN-650) obtaining the highest specific surface
172 m2·g-1 among all samples. Further, when evaluate the HCN-650 as electrode for
both liquid and flexible electrochemical double layer capacitor, the high specific capacitance of
TE D
343 F·g-1 in 6M KOH electrolyte and 0.08 F·cm-2 in PVA-LiCl solid-state electrolyte with excellent rate capability is prone to be demonstrated, respectively. On the other hand, the electrosorption capacity of HCN-650 exhibit as high as 7.08 mg·g-1 according to adsorption
Donnan model.
EP
kinetics. Besides, the electrosorption behavior of HCN-650 can be explained by a modified
AC C
Acknowledgements
This research is supported by National Science Foundation of China (No.21403120) References
[1] M. A. Shannon, P. W. Bohn, M. Elimelech, J. G. Georgiadis, B. J. Marinas, A. M. Mayes, Science and technology for water purification in the coming decades, Nature 452 (2008) 301-310. [2] J. Schewe, J. Heinke, D. Gerten, I. Haddeland, N. W. Arnell, D. B. Clark, R. Dankers, S. Eisner, B. M. Fekete, F. J. Colón-González, S.N. Gosling, H. Kim, X. Liu, Y. Masaki, F. T. Portmann, Y. Satoh, T. Stacke, Q. Tang, Y. Wada, D. Wisser, T. Albrecht, K. Frieler, F. 14
ACCEPTED MANUSCRIPT Piontek, L. Warszawski, P. Kabat, Multimodel assessment of water scarcity under climate change, Proc. Natl. Acad. Sci. U. S. A., 111 (2014) 3245-3250. [3] P. Simon, Y. GoGotSi, Materials for electrochemical capacitors, Nature Mater. 7 (2008) 845-854.
RI PT
[4] G. Wang, L. Zhang, J. Zhang, A review of electrode materials for electrochemical supercapacitors, Chem. Soc. Rev. 41 (2012) 797-828.
[5] L. Wang, Y. Han, X. Feng, J. Zhou, P. Qi, B. Wang, Metal-organic frameworks for energy storage: Batteries and supercapacitors, Coord. Chem. Rev. 307 (2016) 361-381.
SC
[6] P. Suktha, P. Chiochan, P. Iamprasertkun, J. Wutthiprom, N. Phattharasupakun, M. Suksomboon, T. Kaewsongpol, P. Sirisinudomkit, T. Pettong, M. Sawangphruk,
M AN U
High-Performance Supercapacitor of Functionalized Carbon Fiber Paper with High Surface Ionic and Bulk Electronic Conductivity: Effect of Organic Functional Groups, Electrochim. Acta 176 (2015) 504-513.
[7] R. R. Salunkhe, Y. Kamachi, N. L. Torad, S. M. Hwang, Z. Sun, S. X. Dou, J. H. Kimc, Y. Yamauchi,
Fabrication
of
symmetric
supercapacitors
based
on
MOF-derived
TE D
nanoporouscarbons, J. Mater. Chem. A 2 (2014) 19848-19854.
[8] M. E. Suss, S. Porada, X. Sun, P. M. Biesheuvel, J. Yoon, V. Presser, Water desalination via capacitive deionization: what is it and what can we expect from it? Energy Environ. Sci. 8
EP
(2015) 2296-2319.
[9] S. Porada, R. Zhao, A. van der Wal, V. Presser, P. M. Biesheuvel, Review on the science and technology of water desalination by capacitive deionization, Prog. Mater. Sci. 58 (2013)
AC C
1388-1442.
[10] A. Thamilselvan, A. S. Nesaraj, M. Noel, Review on carbon-based electrode materials for application in capacitive deionization process, Int. J. Environ. Sci. Technol. 13 (2016) 2961-1976.
[11] L. Chang, J. Li, X. Duan, W. Liu, Porous carbon derived from Metal-organic framework (MOF) for capacitive deionization electrode, Electrochim. Acta 176 (2015) 956-964. [12] H. Li, S. Liang, M. Gao, G. Li, J. Li, L. He, The study of capacitive deionization behavior of a carbon nanotube electrode from the perspective of charge efficiency, Water Sci. Technol. 71 (2015) 131-136. 15
ACCEPTED MANUSCRIPT [13] H. Li, S. Liang, J. Li, L. He, The capacitive deionization behaviour of a carbon nanotube and reduced graphene oxide composite, J. Mater. Chem. A 1 (2013) 6335-6341. [14] H. Li, F. Zaviska, S. Liang, J. Li, L. Hea, H. Y. Yang, A high charge efficiency electrode by self-assembling sulphonated reduced graphene oxide onto carbon fibre: towards enhanced
RI PT
capacitive deionization, J. Mater. Chem. A 2 (2014) 3484-3491. [15] Z. Wang, T. Yan, G. Chen, L. Shi, D. Zhang, High Salt Removal Capacity of Metal-Organic Gel Derived Porous Carbon for Capacitive Deionization, ACS Sustainable Chem. Eng. 5 (2017) 11637-11644.
SC
[16] P. Liu, T. Yan, L. Shi, H. S. Park, X. Chen, Z. Zhao, D. Zhang, Graphene-based materials for capacitive deionization, J. Mater. Chem. A 5 (2017) 13907-13943.
M AN U
[17] Y. Liu, L. Pan, T. Chen, X. Xu, T. Lu, Z. Sun, D. H.C. Chua, Porous carbon spheres via microwave-assisted synthesis for capacitive deionization, Electrochim. Acta 151 (2015) 489-496.
[18] R. Jin, Y. Jiang, G. Li, Y. Meng, Amorphous carbon coated multiwalled carbon nanotubes@transition metal sulfides composites as high performance anode materials for
TE D
lithium ion batteries, Electrochim. Acta 257 (2017) 20-30.
[19] M. Sevilla, R. Mokaya, Energy storage applications of activated carbons: supercapacitors and hydrogen storage, Energy Environ. Sci. 7 (2014) 1250-1280.
EP
[20] H. Duan, T. Yan, G. Chen, J. Zhang, L. Shi, D. Zhang, A facile strategy for the fast construction of porous graphene frameworks and their enhanced electrosorption performance, Chem. Commun. 53 (2017) 7465-7468.
AC C
[21] K. Kailasam, J. D. Epping, A. Thomas, S. Losse, H. Junge, Mesoporous carbon nitride-silica composites by a combined sol-gel/thermalcondensation approach and their application as photocatalysts, Energy Environ. Sci. 4 (2011) 4668-4674. [22] X. Liu, S. Zou, S. Chen, Ordered mesoporous carbons codoped with nitrogen and iron as effective catalysts for oxygen reduction reaction, Nanoscale 46 (2016) 19249-19255. [23] L. Chang, J. Li, X. Duan, W. Liu, Porous carbon derived from Metal-organic framework (MOF) for capacitive deionization electrode, Electrochim. Acta 176 (2015) 956-964. [24] B. Liu, H. Shioyama, H. Jiang, X. Zhang, Q. Xu, Metal-organic framework (MOF) as a template for syntheses of nanoporous carbons as electrode materials for supercapacitor, 16
ACCEPTED MANUSCRIPT Carbon 48 (2010) 456-463. [25] Z. Wang, T. Yan, L. Shi, D. Zhang, In situ expanding pores of dodecahedron-like carbon frameworks derived from MOFs for enhanced capacitive deionization, ACS Appl. Mater. Inter. 9 (2017) 15068-15078.
RI PT
[26] X. Li, C. Hao B. Tang, Y. Wang, M. Liu, Y. Wang, Y. Zhu, C. Lu, Z. Tang, Supercapacitor electrode materials with hierarchically structured pores from carbonization of MWCNTs and ZIF-8 composites, Nanoscale 9 (2017) 2178-2187.
[27] B. Liu, H. Shioyama, T. Akita, Q. Xu, Metal-Organic Framework as a Template for Porous
SC
Carbon Synthesis, J. Am. Chem. Soc. 130 (2008) 5390-5391.
[28] P. Pachfule, D. Shinde, M. Majumder, Q. Xu, Fabrication of carbon nanorods and graphene
M AN U
nanoribbons from a metal-organic framework, Nature Chem. 8 (2016) 718-724. [29] H. Wang, Q.-L. Zhu, R. Zou, Q. Xu, Metal-Organic Frameworks for Energy Applications, Chem. 2 (2017) 52-80.
[30] A. Mahmood, R. Zou, Q. Wang, W. Xia, H. Tabassum, B. Qiu, R. Zhao, Nanostructured Electrode Materials Derived from Metal-Organic Framework Xerogels for High Energy
TE D
Density Asymmetric Supercapacitor, ACS Appl. Mater. Interfaces 8 (2016) 2148-2157. [31] Y. V. Kaneti, J. Tang, R. R. Salunkhe, X. Jiang, A. Yu, K. C.-W. Wu, Y. Yamauchi, Nanoarchitectured Design of Porous Materials and Nanocomposites from Metal-Organic
EP
Frameworks, Adv. Mater. 29 (2016) 1604898. [32] S. Dang, Q.-L. Zhu, Q. Xu, Nanomaterials derived from metal-organic frameworks,Nat. Rev. Mater. 3 (2017) 17075.
AC C
[33] W. Chaikittisilp, M. Hu, H. Wang, H.-S. Huang, T. Fujita, K. C.-W. Wu, L.-C. Chen, Y. Yamauchi, K. Ariga, Nanoporous carbons through direct carbonization of a zeolitic imidazolate framework for supercapacitor electrodes Chem. Commun. 48 (2012) 7259-7261. [34] Y. Liu, X. Xu, W. Miao, T. Lu, Z. Sun, L. Pan, Metal-organic frameworks-derived porous carbon polyhedrons for highly efficient capacitive deionization, Chem. Commun. 51 (2015) 12020-12023. [35] M. Ding, W. Shi, L. Guo, Z. Y. Leong, A. Baji, H. Y. Yang, Bimetallic Metal-Organic Frameworks Derived Porous Carbon Nanostructure for High Performance Membrane Capacitive Desalination, J. Mater. Chem. A 5 (2017) 6113-6121. 17
ACCEPTED MANUSCRIPT [36] R. H. Baughman, A. A. Zakhidov, W. A. de Heer, Carbon Nanotubes-the Route Toward Applications, Science 297 (2002) 787-792. [37] X. Zhang, L. Chen, T. Yuan, H. Huang, Z. Sui, R. Du, X. Li, Y. Lu, Q. Li, Dendrimer-linked, renewable and magnetic carbon nanotube aerogels, Mater. Horiz. 1 (2014) 232-236.
RI PT
[38] H. Li, L. Pan, Y Zhang, L. Zou, C. Sun, Y. Zhan, Z. Sun, Kinetics and thermodynamics study for electrosorption of NaCl onto carbon nanotubes and carbon nanofibers electrodes. Chem. Phys. Lett. 485 (2010) 161-166.
[39] H. Li, S. Liang, M. Gao, J. Li, Uniform carbon hollow sphere for highly efficient
SC
electrosorption, J. Porous Mater. 23 (2016) 1575-1580.
[40] R. Zhao, P. M. Biesheuvel, H. Miedema, H. Bruningand A. van der Wal, Charge Efficiency:
M AN U
A Functional Tool to Probe the Double-Layer Structure Inside of Porous Electrodes and Application in the Modeling of Capacitive Deionization Phy. Chem. Lett. 1 (2010) 205-210. [41] P. M. Biesheuvel, B. van Limpt, A. van der Wal, Dynamic Adsorption/Desorption Process Model for Capacitive Deionization, J. Phys. Chem. C, 113 (2009) 5630-5640. [42] C.-C. Wang, J. Liang, Y.-H. Liao, S.-Y. Lu, 3D porous graphene nanostructure from a simple,
TE D
fast, scalable process for high performance flexible gel-type supercapacitors, ACS Sustain. Chem. Eng. 5 (2017) 4457-4467.
[43] Z. Zhang, L.-L. Kong, S. L., G.-R. Li, X.-P. Gao, A High-Efficiency Sulfur/Carbon
EP
Composite Based on 3D Graphene Nanosheet@Carbon Nanotube Matrix as Cathode for Lithium-Sulfur Battery, Adv. Energy Mater. 7 (2017) 1602543. [44] J. Zhou, Y. Dou, A. Zhou, R.-M. Guo, M.-J. Zhao, J.-R. Li, MOF Template-Directed
AC C
Fabrication of Hierarchically Structured Electrocatalysts for Efficient Oxygen Evolution Reaction, Adv. Energy Mater. 7 (2017) 1602643. [45] Z. Wang, T. Yan, J. Fang, L. Shi, D. Zhang, Nitrogen-doped porous carbon derived from bimetallic metal-organic framework as highly efficient electrodes for flow-through deionization capacitors, J. Mater. Chem. A 4 (2016) 10858-10868. [46] Y. Lei, M. Gan, L. Ma, M. Jin, X. Zhang, G. Fu, P. Yang, M. Yan, Synthesis of nitrogen-doped porous carbon from zeolitic imidazolate framework-67 and phenolic resin for high performance supercapacitors, Ceram. Int. 43 (2017) 6502-6510. [47] J. Zhou, N. Lin, W. Cai, C. Guo, K. Zhang, J. Zhou, Y. Zhu, Y. Qian, Synthesis of S/CoS 2 18
ACCEPTED MANUSCRIPT Nanoparticles-Embedded N-doped Carbon Polyhedrons from Polyhedrons ZIF-67 and their Properties in Lithium-Sulfur Batteries, Electrochim. Acta, 218 (2016) 243-251. [48] Z. Wang, Y. Lu, Y. Yan, T. Y. P. Larissa, X. Zhang, D. Wuu, H. Zhang, Y. Yang, X. Wang, Core-shell carbon materials derived from metal-organic frameworks as an efficient oxygen
RI PT
bifunctional electrocatalyst, Nano Energy 30 (2016) 368-378. [49] B. L. Fu, Z. Liu, Y. Liu, B. Han, P. Hu, L. Cao, D. Zhu, Bead Cobalt Oxide Nanoparticlaes along Carbon Nanotubes: Towards Moer Highly Integrated Electronic Devices, Adv. Mater. 17 (2005) 217-221.
SC
[50] A. S. Claye, J. E. Fischer, C. B. Huffman, A. G. Rinzler, and R. E. Smalley, Solid-State Electrochemistry of the Li Single Wall Carbon Nanotube System, J. Electrochem. Soc. 147
M AN U
(2000) 2845-2852.
[51] C. Zheng, C. Cao, Z. Ali, J. Hou, Enhanced electrochemical performance of ball milled CoO
AC C
EP
TE D
for supercapacitor applications, J. Mater. Chem. A 2 (2014) 16467-16473.
19
ACCEPTED MANUSCRIPT Table 1. Atomic percent of C, N, O and Co. C1s (%)
N1s (%)
O1s (%)
Co2p (%)
HCN-500
72.68
10.65
10.65
6.03
HCN-600
83.56
6.29
6.5
3.64
HCN-650
83.93
7.05
5.87
3.16
HCN-700
85.46
6.74
4.91
2.89
Table 2. Textural characteristics for HCN.
RI PT
Sample
Sample
Surface area (m2· g-1)
Pore volume (cm3· g-1)
HCN-500
104.45
0.1264
HCN-600
125.11
0.1558
HCN-650
171.60
0.1748
4.0752
HCN-700
116.28
0.1427
4.9102
Pore width (nm) 4.8424
M AN U
SC
4.9795
Table 3. Simulated values of L, Rs, Cdl, Cps and Rct according to the equivalent circuit. L(H)
Rs (Ω)
HCN-500
1.063E-6
1.905
HCN-600
1.98E-20
1.265
HCN-650
2.766E-7
1.083
HCN-700
5.425E-7
0.881
TE D
Sample
Cdl (F)
Cps (F)
Rct (Ω )
0.4136
0.4619
123.2
0.3438
0.2102
231.9
0.3402
0.3417
293.2
0.2402
0.5272
431.3
Table 4. Coefficients of pseudo-first-order adsorption kinetics. First order equation
K1t 2.303
qe K1 R2
Applied potential 0.6V
0.8V
1.0V
1.2V
1.4V
1.66 0.03591 0.99622
3.2 0.03654 0.99283
4.52 0.02516 0.98603
5.48 0.04204 0.99675
7.08 0.03958 0.9892
AC C
EP
log(qe − q ) = log(qe ) −
Parameter
20
ACCEPTED MANUSCRIPT Figure caption Fig. 1 Schematics of the synthesis of HCN. Fig. 2 SEM images of (a) ZIF-67 nanocrystals and (b) HCN-650; TEM images of (c) HCN-500, (d) HCN-600, (e) HCN-650 and (f) HCN-700 (inset in each figure is the magnified TEM image of
RI PT
CNTs from the selected area). Fig. 3 (a) XRD patterns of ZIF-67 and HCNs, (b) TGA curve of ZIF-67, (c) Raman spectra of HCNs, (d) XPS spectra of HCNs, (e) N2 adsorption-desorption isotherm of HCNs, (f) pore size distribution.
SC
Fig. 4 (a) CV curve of HCNs at the scan rate of 10 mV·s-1, (b) specific capacitance with respect to scan rate, (c) GCD curve of HCNs at the current density of 0.8 A·g-1, (d) specific capacitance as
M AN U
function of current density, (e) Impedance spectra of HCNs, (f) equivalent circuit. Fig. 5 (a) Electrosorption behaviours of HCN-650 at 1.2 V with an initial conductivity of 1000 µS·cm-1, (b) corresponding current and voltage variation, (c) salt concentration vs. time at different voltages (Lines: theory, dots: experimental data), (d) electrosorption capacitance with
AC C
EP
TE D
respect to time according to pseudo-first-order adsorption kinetic.
21
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Figure 1
22
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Figure 2
23
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Figure 3
24
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Figure 4
25
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Figure 5
26
S0013-4686(18)30062-8
DOI:
10.1016/j.electacta.2018.01.044
Reference:
EA 31026
To appear in:
Electrochimica Acta
Received Date: 27 November 2017 Revised Date:
5 January 2018
Accepted Date: 8 January 2018
Please cite this article as: T. Gao, F. Zhou, W. Ma, H. Li, Metal-organic-framework derived carbon polyhedron and carbon nanotube hybrids as electrode for electrochemical supercapacitor and capacitive deionization, Electrochimica Acta (2018), doi: 10.1016/j.electacta.2018.01.044. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Metal-Organic-Framework Derived Carbon Polyhedron and Carbon Nanotube Hybrids as Electrode for Electrochemical Supercapacitor and Capacitive Deionization Tie Gaoa, Feng Zhouab, Wei Maa, Haibo Liab* a. Ningxia Key Laboratory of Photovoltaic Materials, Ningxia University, Yinchuan, Ningxia,
RI PT
750021, P. R. China b. State Key Laboratory of High-efficiency Utilization of Coal and Green Chemical Enginering, School of Chemistry and Chemical Engineering, Ningxia University, Yinchuan, Ningxia, 750021, P. R. China
AC C
EP
TE D
M AN U
SC
∗ Corresponding author. Fax/Tel: +86 0951 2062414; E-mail: [email protected] (Haibo Li)
1
ACCEPTED MANUSCRIPT Abstract In this work, carbon polyhedron and carbon nanotube hybrids (HCN) have been synthesized by employing ZIF-67 as precursor for electrochemical supercapacitor and capacitive deionization (CDI). Basically, uniform polyhedral nanocrystals of ZIF-67 were firstly fabricated, and then they
RI PT
were directly subjected to chemical vapor deposition for growing carbon nanotubes. It is found that the as-fabricated HCN electrode exhibits remarkable electrochemical performance, i.e. the highest capacitance of 343 F·g-1 at the scan rate of 10 mV·s-1 and excellent rate capability. This is due to that HCN can provide rich space and short ion diffusion path. Moreover, the HCN
SC
electrodes shows superior CDI performance, i.e. the electrosorption capacity reached as high as 7.08 mg·g-1 and the adsorption rate of 0.03958 mg·g-1·min-1. Metal
organic
frameworks,
Carbon
polyhedron,
M AN U
Keywords:
AC C
EP
TE D
Supercapacitor
2
Capacitive
deionization,
ACCEPTED MANUSCRIPT 1. Introduction Portable water and energy storage are absolutely essential to promote the ongoing of our society [1-3]. Fast charge-discharge supercapacitors are attractive energy storage technologies as a result of their high power capability, good operational safety and long cycling life compared to
RI PT
secondary batteries, of which the performance is greatly governed by the electrode materials [4-7]. On the other hand, capacitive deionization (CDI), as an emerging promising desalination technology, is considered to be a cost-effective, environmentally friendly and energy-efficient in purification with low-pressure pumps, non-membrane, etc [8-11]. The mechanism of CDI is
SC
similar to supercapacitor which based on electrical double layer theory. Compared with other desalination technologies, CDI is an efficient and energy-saving method profit from salty ions
M AN U
which is the minority one immobilized in the double layer between the surface of electrodes and solution. On the contrary, other methods such as distillation extract the majority phase, the water, from the seawater then much additional energy would be wasted [12-16]. Porous carbon materials have attracted intense attention due to their unique properties, including high surface area, large pore volume, and unique pore size distribution, which have
TE D
great potential for electrochemical applications [17-22]. Metal-organic frameworks (MOFs) have been regarded as a promising class of advanced materials for gas adsorption, sensing, supercapacitors, and catalysis, owing to their crystalline structures with high porosities and
EP
tunable functionalities [23-30]. In particular, due to their porous structures with high surface areas, MOFs are ideal sacrificial templates to derive into various porous carbon and other active materials, such as nanocarbons metal oxides, metal sulfides, metal carbides, metal phosphides and
AC C
their composites [31, 32]. Basically, direct pyrolysis of MOFs represents an efficient strategy to produce nanostructured carbon and related composite. Yamauchi et al. prepared nanoporous carbon though thermal pyrolysis of zeolitic immidazolate frameworks (ZIF-8) for supercapacitors, and a high specific capacitance of 214 F·g-1 at a scan rate of 5 mV·s-1 was achieved [33]. Pan carried out pioneer work on ZIF-8 derived porous carbon polyhedrons (PCPs) for CDI, exhibiting the high electrosorption capacity of 13.86 mg·g-1 when the initial NaCl concentration is 500 mg·L-1, due to their high accessible surface area and low charge transfer resistance [34]. Yang et al. designed bimetallic metal-organic frameworks (BMOFs) with different molar ratios of Zn and Co based on ZIF-8 and ZIF-67, showing the porous carbon derived from a BMOF of Zn: Co = 3:1 3
ACCEPTED MANUSCRIPT exhibits an untra-high salt removal capacity of 45.62 mg·g-1 in a 750 mg·L-1 NaCl solution at 1.4 V which is due to the large ion-accessible surface area and improved electrical conductivity of the carbon derived from BMOF [35]. However, the prospect of pure porous carbon materials derived from direct pyrolysis of MOFs in energy and water fields is still obstructed owing to the intrinsic
RI PT
properties of carbon materials, for instance, low conductivity. In this case, to explore novel electrode materials with satisfying electrochemical properties, porous carbon materials have been incorporated with other active materials. Generally, the porous carbon based composites have many special features to achieve enhanced properties. For instance, the carbon frameworks
SC
always provide facile electron transportation so that high electronic conductivity can be achieved while the active materials confined in the carbon matrix endow improved performance, which can
M AN U
be attributed to the synergetic enhancement. The porous feature of these composites leads to a facile electrolyte infiltration so that a high electrode/electrolyte interface can be obtained. Therefore, the porous carbon-based composites have many unique features and can achieve excellent performance for electrochemical applications.
Carbon nanotubes (CNTs), a 1D allotrope of carbon with extraordinary properties, have been
TE D
widely investigated in recent decades. In relation to supercapacitor and CDI, either CNTs or CNTs composite as electrode has been extensively explored due to its controllable physio-chemical properties [36, 37]. In the present work, we report the fabrication of carbon-based HCN derived
EP
from growing CNTs on ZIF-67, a well-known subfamily of MOFs, for both supercapacitor and CDI. The HCN feature with closely packed tube-box structures: a ZIF-67-derived dodecahedron of porous carbons (PCs) and chemical vapor deposited CNTs linked with each PCs as a bridge. In
AC C
this kind of novel structure, the PC with high accessible surface area is responsible for providing huge space to accommodate ions. Besides, cobalt derived from ZIF-67 is prone to be oxidized as CoxOy which is highly active to introduce the pseudo-reaction and therefore contribute to high electrochemical performance. On the other hand, CNTs act as conducting wire to facilitate the diffusion kinetics. To prove our analysis, the potential and superior performances of HCN electrode are demonstrated in liquid and solid supercapacitor as well as CDI. 2 Experimental 2.1 Synthesis of HCN 4
ACCEPTED MANUSCRIPT The synthetic strategy of HCN is schematically illustrated in Fig. 1. Cobaltous Nitrate Hexahydrate (Co(NO3)2·6H2O) (99%) was purchased from Macklin. 2-Methylimidazole (Hmim) (98%) was purchased from Aladdin. All chemicals were used as received without any special handling. Uniform polyhedral nanocrystals of ZIF-67 were first fabricated by the coordination of
RI PT
Co2+ and 2-MeIm in methanol. Specifically, methanolic solutions of cobaltous nitrate hexahydrate (6.89 g, 240 mL) and solutions of Hmim (15.57 g, 240 mL) were mixed under stirring when the methanolic solutions of Hmim was kept at 40
. After being stirred for 0.5 h, ZIF-67 nanocrystals
were collected by washing with methanol and followed by centrifuging (11000 rpm, 10 min) for
SC
three times. Finally, the product was dried overnight for ready to use.
The CNTs grown on ZIF-67 was carried in a CVD system. First of all, 0.1g ZIF-67 powder was
M AN U
sent into the chamber and the pressure in quartz tube is pumped to 1×10-2 Pa by the mechanical pump. Next, injecting Ar+4%H2 at the flow rate of 20 sccm and the pressure is kept at a balance of 2.9 Pa. Then the temperatures rose steadily from room temperature to target temperatures (500, 600, 650 and 700
) with a heating rate of 15
·min-1. Once it reaches the target temperature, the
acetylene with flow rate of 20 sccm was injected and kept for 1 hour, the pressure balance at 10
temperature below 200 2.2 Characterization
TE D
Pa during this process. Successfully, the system was cooling in Ar+4%H2 atmosphere until the . Finally, closing the Ar+4%H2 and taking out of sample.
EP
The morphology and fine structures of samples were observed using scanning electron microscope (SEM, Zeiss Supra 40) and transmission electron microscope (TEM, Hitachi
AC C
HT7700). The crystal structures information were identified using an X-ray diffractometer (DX-2700, CHN) with Cu Kα radiation at 40 kV and 30 mA over the 2θ range of 3-80°. The thermogravimetric analyses (TGA) were carried out on a thermogravimetric analyzer (SETARAM Setsys 16) from room temperature to 1000
at a heating rate of 15
·min-1 and under N2 flow
rate of 20 sccm. The Raman spectra were recorded on a spectrometer (DXR, USA) equipped with an optical microscope at room temperature. X-Ray photoelectron spectroscopy (XPS, Thermo Scientific Escalab 250Xi) spectra was collected on a monochromatized Al Kα X-ray source (1361 eV) and the X-ray spot size of 500 µm. Nitrogen adsorption and desorption isotherms were measured at 77K using accelerated surface area and porosimetry system (Tristar 5
3020 USA).
ACCEPTED MANUSCRIPT The wettability was obtained by measuring the contact angle (Powerreach JC200D2, CHN). 2.3 Fabrication of electrode The electrodes tested in electrochemical, CDI and flexible symmetric supercapacitors were fabricated by the following approach. First, the electrode material was fabricated by
RI PT
mixing the 80 wt% of HCN, 10 wt% of super conductive black and 10 wt% of polytetrafluoroethylene (PTFE) slurries. It should be noted that the mixture was vigorously grinded for several minutes until it became homogenous. Subsequently, the mixtures were
SC
daubed onto target shape graphite sheet by a medical blade, followed by drying at 70 for 10h in a vacuum oven.
M AN U
2.4 Electrochemical measurement
The electrochemical test was measured from a three-electrode system in 6M KOH solution using electrochemical workstation (CHI 660D). The working electrode was prepared according to above method. The counter electrode was a platinum wire and the reference electrode was Ag+/AgCl reference electrode. Then, the Cyclic Voltammetry (CV), Galvanostatic Charging-Discharging
TE D
(GCD) and Electrochemical Impedance Spectrum (EIS) were obtained. The specific capacitances were obtained from the CV curves by the following equation: C = ∫ idV /ν∆Vm
(1)
EP
Where C is the specific capacitance (F·g-1), i is the response current (A), v is the potential scan rate (mV·s-1), ∆V is the absolute value of the potential window (V) and m is the mass of HCN (g).
AC C
For comparison, the specific capacitance is also calculated from the GCD curves according to the following equation:
C=
I mt ∆V
(2)
Where Im is the ratio of response current and the mass of activated materials (mA·g-1), t is the discharge time (s) and ∆V is the potential change during the discharge process (V). The energy density E (Wh·Kg-1) and power density P (W·Kg-1) was calculated based on the following equations:
6
ACCEPTED MANUSCRIPT E=
1 C 1 × × V2 × 2 4 3.6
P=
E t
(3) (4)
Where C (F·g-1) is the specific capacitance and V is the voltage changed during discharge period.
RI PT
2.5 CDI experiment and calculation
The CDI performances of HCN electrode were acquired through batch-mode experiment system, which includes a current collector, DC power, conductivity monitor and constant
SC
peristaltic pump. Electrode’s material was adhered to two graphite current collectors and separated from each other by the insulated spacer to avoid short circuit. The total volume of NaCl
M AN U
solution was 40 mL and drove by the constant peristaltic pump under a flow rate of 34 mL·min-1. The adsorption capacity of electrode’s material was evaluated by adsorption NaCl from aqueous solution at an initial conductivity of 1000 µS·cm-1 respectively. A direct voltage of 0.6-1.4 V was applied between electrodes. At the moment, the adsorption ability of the electrode was evaluated by electrosorption capacity (Γ, mg·g-1).Theoretically, the adsorption capacity was obtained from the conductivity transients curves by the following equation: (G0 − Gt )V 2m
TE D
Γ=
(5)
where G0 and Gt are the initial and final conductivity (µS·cm-1), V is the total volume of NaCl
EP
solution (L) and m is the mass of HCN (g) [38].
The simulation on CDI performance of HCN electrode is performed by employing the modified
AC C
Donnan model (mD-model) which has been proved to be valid for batch-mode experiment [9, 39-41]. In this model, the water leaving the CDI-cell is flowing to a recycle vessel and from there fed back into the CDI-cell. It also assumes that throughout the cell the salt concentration is the same everywhere, and is the same as in the recycle vessel. A simple description of the mD-model has been given in supporting information. Based on above method, we enter the following key dates to the program: a = 7.2 (m2); V = 4.0×10-5 (m3); C0 = 9.223 (mol·m-3); 7
ACCEPTED MANUSCRIPT A = 3.85×10-3 (m2); K =1.455×10-8 (m·s-1); Vcell = (0.6, 0.8, 1.0, 1.2 and 1.4V); Cst = 0.5 (F·m-2).
RI PT
2.6 Assemble of flexible supercapacitors The preparation of LiCl/polyvinyl alcohol (PVA) gel electrolyte is as follows. In a typical run, LiCl (8.4g) and PVA (4.0 g) powders were added into DI water (40 mL). The mixture was stirred till the solution became clear, followed by natural cooling to room temperature.
SC
at 90
Successfully, the electrodes for solid state flexible supercapacitor were fabricated by drop casting
drying at 70
M AN U
the HCN suspension on graphite sheet with geometrical surface area of 1×2 cm2, followed by for overnight. Finally, the two electrodes were immersed into the LiCl/PVA gel
electrolyte for 5 min. Then, the separator was sandwiched by the two electrodes to form the flexible symmetric supercapacitor [42]. 3 Results and Discussion
TE D
3.1 Morphology and structure
Fig. 2(a) and (b) show the typical SEM images of ZIF-67 and HCN obtained at 650 (HCN-650). Overall, the size and polyhedral shape of the ZIF-67 nanoparticles are retained well
EP
after CVD treatment, while the surface of the particles becomes much rougher. A large amount of long CNTs, several hundred nanometers in length, embedded in ZIF-67 derived porous carbon
AC C
can be observed, forming a network structure. The formation of CNTs is the result of the catalytic effect of Co nanoparticles on the particle surface. Further, the effect of growth temperature on HCN has been examined. Fig. 2(c) to (f) exhibited the HCN prepared at 500, 600, 650 and 700
,
respectively, and inset in each figure is the magnified image of CNTs corresponding to certain temperature. Regardless of temperature, a well polyhedral shape of ZIF-67 is maintained. Moreover, high temperature results in much rougher surface. On the other hand, it seems that a more clear and hollow structure of HCN is confirmed. More importantly, one can observed from the inset of image that the diameter of CNTs is going to increase with the rise of growth temperature. A statistic result of diameter distribution for CNTs growth at different temperature is 8
ACCEPTED MANUSCRIPT given in Fig. S1. Since the diameter of CNTs is closely related to the size of Co, we have examined the morphology of Co particle at different temperature. As shown in Fig. S2, as the temperature increase, the size of Co is became bigger and thus leads to CNTs with bigger diameter. In addition, the TEM image also reveals the presence of Co nanoparticles encapsulated
RI PT
by a few-layered carbon shell (Fig. S3), especially at the tip of the CNTs. It should be mention that these carbon-encased Co nanoparticles are inaccessible to reactants as they remain completely encapsulated within the graphitic carbon shell even after acid leaching.
The crystalline nature of HCN is further carried out by XRD in Fig. 3(a). The metallic Co is
SC
confirmed by the three typical peaks at 44.2°, 51.6°, and 75.9° respectively, consistent with (111), (200), and (220) planes of Co (PDF#89-4307) [43]. This is due to that metallic Co nanoparticles
M AN U
are quickly formed in the presence of H2 atmosphere during CVD process. Besides, the typical peak at 22°corresponding to the (002) diffraction mode of the graphic structure is revealed for all HCN samples. Moreover, the degree of graphitization is strengthened with the increase of growth temperature. Fig. 3(b) draws the TGA curve of ZIF-67. The mass loss is concentrated at 500 700
to
which is mainly caused by thermal decomposition and recombination of organic ligands
TE D
during carbonization process. Raman spectroscopy is essential for the structural analysis which is illustrated in Fig. 3(c). The vibration peaks at 188, 476, 520, and 687 cm−1 in terms of all HCN samples match with the Co3O4 phase which is in good agreement of literature [44]. Three 1360,
1580 and 2680 cm−1, which are separately unique characteristic
EP
additional bands at
D-band, G-band and G* band associated with carbon materials. The intensity ratio of D-band over G-band (ID/IG) is usually used to evaluate the degree of graphitization of carbon materials and as
AC C
the ratio decreases, the graphitization degree is higher [44, 45]. Obviously, the ID/IG is decreased from 1.05 to 0.66 as temperature increased from 600 to 700
, implying high graphitized CNTs is
facile to acquire at high growth temperature. The chemical composition is investigated by XPS spectra. The whole spectra of HCN samples are illustrated Fig. 3(d) and the corresponding atomic percent is summarized in Tab. 1. When the temperature increased from 600
to 700
, the C1s
ratio has not varied much. Meanwhile, the Co 2p ratio is very rare which is only occupy
3%.
Besides, N1s has a mild ratio which would beneficial to improve the conductivity of HCN due to the doping effect. Importantly the high-resolution C 1s spectrum can be deconvolved into several bonds (Fig. S4(a)), corresponding to Sp2 hybrid carbon (284.6 eV), C-C (285.2 eV), C-N (285.7 9
ACCEPTED MANUSCRIPT eV), C-O (286.4) and O-C=O (288.7 eV). It is noted that the presence of O signal can be attributed to residual oxygen in ZIF derived carbon and the adsorbed oxygen from environmental atmosphere. The high-resolution O 1s spectrum can be deconvolved into three bands (Fig. S4(b)), corresponding to C=O (530.8 eV), C-O-C (532 eV) and O-C=O (533.4 eV) [46]. The
at
398.5 eV and pyrrolic N at
RI PT
high-resolution N1s spectrum reveals the presence of two types of nitrogen species, pyridinic N 400.8 eV (Fig. S4(c)), together with the identification of C-N
bonds at 285.1 eV, indicating the N doping in HCN which is provided by the ZIF-67 from the pyrolysis [47]. Significantly, the N doping may beneficial to enhance the conductivity of HCN
SC
and therefore the electrochemical behavior. Further, high-resolution XPS spectra in the Co 2p region are shown in Fig S4(d). The first peak presented at 778.5 eV is assigned to be the Co 2p3/2
M AN U
while the peak located at 793.5 eV can be attributed to Co 2p1/2. Importantly, the 2p3/2 appears at 780.3 eV and 782.1 eV for Co3+ and Co2+ while 2p1/2 appears at 795.3 eV and 798 eV for Co3+ and Co2+. In addition, there are two small peaks at 787.9 eV and 803.6 eV represent shake-up peaks of Co2+ [48, 49]. The simultaneously exist of Co3+ and Co2+ indicating the nanoparticles contain Co3O4, but the area ratio between peaks of Co3+ and Co2+ below 2:1 that confirmed there
TE D
are superfluous Co2+ for Co3O4 which are exist in CoO exactly.
The N2 adsorption-desorption isotherm and pore size distribution curves are shown in Fig. 3(e) and (f). The specific values associated with pore texture are summarized in Tab. 2. Basically, the isotherm, in
EP
N2 adsorption-desorption isotherm of all HCN samples were belong to type
which the isotherms with sharp increase in volume at low relative pressure imply the presence of micropores, mainly from CNTs. The hysteresis loop at relative pressure between 0.4 and 0.9
AC C
suggests the coexistence of mesopores from porous carbon derived from ZIF-67. The pore size distribution reveals that the micropores are smaller than 1 nm in size while the mesopores are around 100 nm. Moreover, the HCN-650 have a Brunauer-Emmett-Teller (BET) surface area of 172 m2·g-1 and total volume of
0.17 cm3·g-1, both of which are highest among all samples.
In addition, the wettability of HCN-650 electrode has been tested (Fig. S5), exhibiting that the contact angle is
91° which confirms the hydrophilicity of HCN-650 electrode.
3.2 Electrochemical performance Based on the above discussions, the resulted porous carbon nanostructures can not only provide 10
ACCEPTED MANUSCRIPT sufficient reaction sites, short ion diffusion length, buffer the volume change, but also ensure direct mechanical and electrical connection with the flexible carbon support, making it a very promising electrode material for aqueous and flexible energy storage applications. To evaluate the electrochemical performance, the electrochemical activity of various HCN samples toward
RI PT
supercapacitor was first identified by CV tests conducted in saturated 6M KOH electrolyte via three electrodes method. Fig. 4(a) provide the representative CV curve of HCN derived from 500, 600, 650 and 700
at scan rate of 10 mV·s-1. Apparent redox peaks can be observed from the
CV curves, which can be assigned for the Faradic reaction between Co3+/Co4+ with OH−.
SC
From CV curves of all HCN samples (Fig. S6), the peak positions only shift a little when the scan rate increased from 10 to 100 mV·s−1, indicating that the electrode has excellent reversibility.
M AN U
The specific capacitance as a function of scan rate for all HCN electrodes is given in Fig. 4(b). Among all HCN electrodes, HCN-650 has the highest capacitance at each certain scan rate. At a low scan rate of 10 mV·s−1, HCN-650 exhibits a high capacitance of 343 F·g-1 and a high capacitance of 43 F·g-1 can be maintained even when the scan rate increased to 1000 mV·s−1, suggesting that 13% of the capacitance can be remained even when the scan rate increased 100
TE D
times. By compared with other HCN samples, the enhanced capacitance and rate capability of the HCN-650 can be attributed to the unique porous structures and high BET surface area as well as pore volume that can provide sufficient reaction cites and short ion diffusion length. The GCD curves for HCN electrode derived from 500, 600, 650 and 700
at the current density of 0.8
EP
A·g-1 are shown in Fig. 4(c), where all the curves have almost linear charge-discharge characteristics with two slight plateau located at -0.4 and -0.6 V. When the current density further
AC C
increased, a similar GCD curves can be obtained (Fig. S7). Fig. 4(d) presents the specific capacitance with respect to current density calculated from GCD curve. At each current density, HCN-650 reveals the highest capacitance. At current density of 0.5 A·g-1, a high capacitance of 190 F·g-1 can be obtained. Remarkably, when the current density increased to 2 A·g-1, the capacitance is as high as 110 F·g-1, implying 58% remaining. Besides, Fig. S8 shows Ragone plots of HCNs corresponding to the relationship between energy and power densities. The maximum energy density of HCN-650 reach 6.6 Wh·kg-1 at the current density of 0.5 A·g-1. The Nyquist plot of the EIS and corresponding equivalent circuit are drawn in Fig. 4(e) and (f), respectively. A frequency response analysis at open circuit potential over the frequency range 11
ACCEPTED MANUSCRIPT from 100 KHz to 10 mHz yielded the Nyquist plots. Commonly, the spectrum consists of a high frequency semicircle and a low frequency tail. In the equivalent circuit, the small L are used to reduces the influence of alternating current, Rs includes the intrinsic resistance of the electrode active material, the contact resistance at the interface between the electroactive material and the
RI PT
current collector and the resistance of electrolyte solution, Rct is the charge transfer resistance at the electrode/electrolyte interface, Cdl is the double layer capacitance, Cps is the pseudo capacitance [50, 51]. Tab. 3 provides the analogue value for each parameter. In the high-frequency region, the intercept of X axis for HCN-700 is smallest among all samples, showing that the bulk
SC
resistance (Rs) of HCN-700 is smallest. This is inspired us that the bulk resistance would be decreased as the growth temperature increasing which is probably due to the high crystallization
M AN U
of HCN. Moreover, the lowest charge transfer resistance corresponding to HCN-650 is acquired. In the low-frequency region, a more vertical line for the HCN-650 is observed, indicating that the HCN-650 have lowest ionic diffusion resistance among all samples.
To further demonstrate the potential application of the HCN for flexible energy storage devices, a flexible supercapacitor was assembled with a PVA-LiCl solid-state electrolyte. Fig. S9 exhibits
TE D
a digital image of brighten LED lighted by flexible supercapacitors based on two HCN-650 electrode connected in series. The CV curves of the flexible supercapacitor are shown in Fig. S10(a), in which quasi-rectangular curves without sharp redox peaks can be observed, showing
EP
that full cell has good capacitive properties. The GCD curves of the flexible supercapacitor are shown in Fig. S10(b), where all the curves have almost linear GCD characteristics. Basically, the specific capacitance of HCN-650 flexible supercapacitor calculated based on CV curves reaches
AC C
0.08 F·cm-2 at the scan rate of 10 mV·s-1. Importantly, it remained about 10% (0.008 F·cm-2) when the scan rate increase to 1000 mV·s-1. On the other hand, the specific capacitance calculated based on GCD curves even up to 0.05 F·cm-2 at the current density of 0.1 A·g-1. When the current density increased to 0.5 A·g-1, the capacitance is as high as 0.025 F·cm-2, implying 50% remaining. Apparently, the excellent rate capability is in good agreement with that in KOH electrolyte via three electrodes’ method. 3.3 CDI performance In order to evaluate the desalination performance, batch-mode experiment using symmetric CDI in the feed water with fixed initial concentrations at electrical potentials of 12
ACCEPTED MANUSCRIPT 1.2 V were carried out. Fig. 5(a) shows the curve of conductivity change with time under 1.2 V at the initial conductivity of 1000 µS·cm-1. The simultaneous change of voltage and current in adsorption and desorption process was observed in Fig. 5(b). In the initial stage of adsorption, a voltage of 1.2 V was applied, the conductivity decreases rapidly which is
RI PT
due to the fact that the open porous surface was quickly occupied by salty ions. After 100 minutes adsorption, the decrease on conductivity gets slowly which is tends to reach a steady value, suggesting the saturation. During this process, the current decreases rapidly as the adsorption commences until it arrived at an ultra-small value due to the presence of
SC
leakage current. Once the circuit was shorted, adsorbed ions were released from the electrode and the conductivity recovered to the initial level gradually. Fig. 5(c) shows the
M AN U
concentration varies with time under voltage ranging from 0.8 to 1.4 V, together with theoretical curve simulated according to mD-model. The concentration decreased gradually as the potential was increased, which is due to the increase of the electrostatic force. High cell voltage resulted in high electrosorption capacity. Significantly, the experimental dates are in good agreement with the theoretical value at each cell voltage. In
which is
7.2 m2
TE D
details, the effective electrode area derived from the simulation is approximately
32% of the BET value (172m2·g-1 × 0.13g) suggesting that quite a
significantportion of the measured area is valid for electrosorption. Moreover, the Stern
EP
layer capacity Cst is as high as 0.5 F· m-2 indicating that the adsorption process is dominantly governed by electrical double layer. Additionally It should be noticed that the adsorption capacity reached as high as 7.08 mg·g-1 under 1.4 V, which show high potential
AC C
in application to CDI. Even if the cell potential decreased to 0.6 V it still remain a remarkable adsorption capacity of 1.66 mg·g-1. On the other hand, adsorption kinetics is also investigated using Lagergren’s pseudo-first-order adsorption kinetics as shown below:
log(qe − q ) = log(qe ) −
K1t 2.303
(6)
Where qe (mg·g-1) and q (mg·g-1) are the amount of NaCl absorbed at equilibrium and time t, respectively. K1 is the adsorption rate constant of pseudo-first-order equation [13]. The fitting curves are obtained with high regression coefficients as shown in Fig. 5(d) and the relevant coefficients are listed in Tab. 4. The high value of R2 (above 0.986) indicating the 13
ACCEPTED MANUSCRIPT experiment data corresponds to Lagergren’s pseudo-first-order adsorption kinetics regardless of cell voltage. Fig. S11 depicts the rate as a function of capacitance with the voltage between 0.6 and 1.4 V in a 500 mg/L NaCl solution. It is clear that at low voltage of 0.6 V, the HCN-650 have the lowest capacitance. When the voltage increased from 0.6
RI PT
to 1.4 V, the capacitance increases as well. Meanwhile, the Ragone Kim-Yoon-Plot turn righter and upper, suggesting the increase of both the capacitance and adsorption rate. 4 Conclusions
SC
In summary, we synthesized a novel carbon polyhedron and carbon nanotube hybrids (HCN) by chemical vapor deposition approach using ZIF-67 as the precursor directly. Basically, HCN
M AN U
exhibited typical network structure, within which the polyhedron porous carbon were linked tightly by ultra-long carbon nanotubes. By examining the effect of growth temperature on HCN, it is found that the HCN synthesized at 650 area of
(HCN-650) obtaining the highest specific surface
172 m2·g-1 among all samples. Further, when evaluate the HCN-650 as electrode for
both liquid and flexible electrochemical double layer capacitor, the high specific capacitance of
TE D
343 F·g-1 in 6M KOH electrolyte and 0.08 F·cm-2 in PVA-LiCl solid-state electrolyte with excellent rate capability is prone to be demonstrated, respectively. On the other hand, the electrosorption capacity of HCN-650 exhibit as high as 7.08 mg·g-1 according to adsorption
Donnan model.
EP
kinetics. Besides, the electrosorption behavior of HCN-650 can be explained by a modified
AC C
Acknowledgements
This research is supported by National Science Foundation of China (No.21403120) References
[1] M. A. Shannon, P. W. Bohn, M. Elimelech, J. G. Georgiadis, B. J. Marinas, A. M. Mayes, Science and technology for water purification in the coming decades, Nature 452 (2008) 301-310. [2] J. Schewe, J. Heinke, D. Gerten, I. Haddeland, N. W. Arnell, D. B. Clark, R. Dankers, S. Eisner, B. M. Fekete, F. J. Colón-González, S.N. Gosling, H. Kim, X. Liu, Y. Masaki, F. T. Portmann, Y. Satoh, T. Stacke, Q. Tang, Y. Wada, D. Wisser, T. Albrecht, K. Frieler, F. 14
ACCEPTED MANUSCRIPT Piontek, L. Warszawski, P. Kabat, Multimodel assessment of water scarcity under climate change, Proc. Natl. Acad. Sci. U. S. A., 111 (2014) 3245-3250. [3] P. Simon, Y. GoGotSi, Materials for electrochemical capacitors, Nature Mater. 7 (2008) 845-854.
RI PT
[4] G. Wang, L. Zhang, J. Zhang, A review of electrode materials for electrochemical supercapacitors, Chem. Soc. Rev. 41 (2012) 797-828.
[5] L. Wang, Y. Han, X. Feng, J. Zhou, P. Qi, B. Wang, Metal-organic frameworks for energy storage: Batteries and supercapacitors, Coord. Chem. Rev. 307 (2016) 361-381.
SC
[6] P. Suktha, P. Chiochan, P. Iamprasertkun, J. Wutthiprom, N. Phattharasupakun, M. Suksomboon, T. Kaewsongpol, P. Sirisinudomkit, T. Pettong, M. Sawangphruk,
M AN U
High-Performance Supercapacitor of Functionalized Carbon Fiber Paper with High Surface Ionic and Bulk Electronic Conductivity: Effect of Organic Functional Groups, Electrochim. Acta 176 (2015) 504-513.
[7] R. R. Salunkhe, Y. Kamachi, N. L. Torad, S. M. Hwang, Z. Sun, S. X. Dou, J. H. Kimc, Y. Yamauchi,
Fabrication
of
symmetric
supercapacitors
based
on
MOF-derived
TE D
nanoporouscarbons, J. Mater. Chem. A 2 (2014) 19848-19854.
[8] M. E. Suss, S. Porada, X. Sun, P. M. Biesheuvel, J. Yoon, V. Presser, Water desalination via capacitive deionization: what is it and what can we expect from it? Energy Environ. Sci. 8
EP
(2015) 2296-2319.
[9] S. Porada, R. Zhao, A. van der Wal, V. Presser, P. M. Biesheuvel, Review on the science and technology of water desalination by capacitive deionization, Prog. Mater. Sci. 58 (2013)
AC C
1388-1442.
[10] A. Thamilselvan, A. S. Nesaraj, M. Noel, Review on carbon-based electrode materials for application in capacitive deionization process, Int. J. Environ. Sci. Technol. 13 (2016) 2961-1976.
[11] L. Chang, J. Li, X. Duan, W. Liu, Porous carbon derived from Metal-organic framework (MOF) for capacitive deionization electrode, Electrochim. Acta 176 (2015) 956-964. [12] H. Li, S. Liang, M. Gao, G. Li, J. Li, L. He, The study of capacitive deionization behavior of a carbon nanotube electrode from the perspective of charge efficiency, Water Sci. Technol. 71 (2015) 131-136. 15
ACCEPTED MANUSCRIPT [13] H. Li, S. Liang, J. Li, L. He, The capacitive deionization behaviour of a carbon nanotube and reduced graphene oxide composite, J. Mater. Chem. A 1 (2013) 6335-6341. [14] H. Li, F. Zaviska, S. Liang, J. Li, L. Hea, H. Y. Yang, A high charge efficiency electrode by self-assembling sulphonated reduced graphene oxide onto carbon fibre: towards enhanced
RI PT
capacitive deionization, J. Mater. Chem. A 2 (2014) 3484-3491. [15] Z. Wang, T. Yan, G. Chen, L. Shi, D. Zhang, High Salt Removal Capacity of Metal-Organic Gel Derived Porous Carbon for Capacitive Deionization, ACS Sustainable Chem. Eng. 5 (2017) 11637-11644.
SC
[16] P. Liu, T. Yan, L. Shi, H. S. Park, X. Chen, Z. Zhao, D. Zhang, Graphene-based materials for capacitive deionization, J. Mater. Chem. A 5 (2017) 13907-13943.
M AN U
[17] Y. Liu, L. Pan, T. Chen, X. Xu, T. Lu, Z. Sun, D. H.C. Chua, Porous carbon spheres via microwave-assisted synthesis for capacitive deionization, Electrochim. Acta 151 (2015) 489-496.
[18] R. Jin, Y. Jiang, G. Li, Y. Meng, Amorphous carbon coated multiwalled carbon nanotubes@transition metal sulfides composites as high performance anode materials for
TE D
lithium ion batteries, Electrochim. Acta 257 (2017) 20-30.
[19] M. Sevilla, R. Mokaya, Energy storage applications of activated carbons: supercapacitors and hydrogen storage, Energy Environ. Sci. 7 (2014) 1250-1280.
EP
[20] H. Duan, T. Yan, G. Chen, J. Zhang, L. Shi, D. Zhang, A facile strategy for the fast construction of porous graphene frameworks and their enhanced electrosorption performance, Chem. Commun. 53 (2017) 7465-7468.
AC C
[21] K. Kailasam, J. D. Epping, A. Thomas, S. Losse, H. Junge, Mesoporous carbon nitride-silica composites by a combined sol-gel/thermalcondensation approach and their application as photocatalysts, Energy Environ. Sci. 4 (2011) 4668-4674. [22] X. Liu, S. Zou, S. Chen, Ordered mesoporous carbons codoped with nitrogen and iron as effective catalysts for oxygen reduction reaction, Nanoscale 46 (2016) 19249-19255. [23] L. Chang, J. Li, X. Duan, W. Liu, Porous carbon derived from Metal-organic framework (MOF) for capacitive deionization electrode, Electrochim. Acta 176 (2015) 956-964. [24] B. Liu, H. Shioyama, H. Jiang, X. Zhang, Q. Xu, Metal-organic framework (MOF) as a template for syntheses of nanoporous carbons as electrode materials for supercapacitor, 16
ACCEPTED MANUSCRIPT Carbon 48 (2010) 456-463. [25] Z. Wang, T. Yan, L. Shi, D. Zhang, In situ expanding pores of dodecahedron-like carbon frameworks derived from MOFs for enhanced capacitive deionization, ACS Appl. Mater. Inter. 9 (2017) 15068-15078.
RI PT
[26] X. Li, C. Hao B. Tang, Y. Wang, M. Liu, Y. Wang, Y. Zhu, C. Lu, Z. Tang, Supercapacitor electrode materials with hierarchically structured pores from carbonization of MWCNTs and ZIF-8 composites, Nanoscale 9 (2017) 2178-2187.
[27] B. Liu, H. Shioyama, T. Akita, Q. Xu, Metal-Organic Framework as a Template for Porous
SC
Carbon Synthesis, J. Am. Chem. Soc. 130 (2008) 5390-5391.
[28] P. Pachfule, D. Shinde, M. Majumder, Q. Xu, Fabrication of carbon nanorods and graphene
M AN U
nanoribbons from a metal-organic framework, Nature Chem. 8 (2016) 718-724. [29] H. Wang, Q.-L. Zhu, R. Zou, Q. Xu, Metal-Organic Frameworks for Energy Applications, Chem. 2 (2017) 52-80.
[30] A. Mahmood, R. Zou, Q. Wang, W. Xia, H. Tabassum, B. Qiu, R. Zhao, Nanostructured Electrode Materials Derived from Metal-Organic Framework Xerogels for High Energy
TE D
Density Asymmetric Supercapacitor, ACS Appl. Mater. Interfaces 8 (2016) 2148-2157. [31] Y. V. Kaneti, J. Tang, R. R. Salunkhe, X. Jiang, A. Yu, K. C.-W. Wu, Y. Yamauchi, Nanoarchitectured Design of Porous Materials and Nanocomposites from Metal-Organic
EP
Frameworks, Adv. Mater. 29 (2016) 1604898. [32] S. Dang, Q.-L. Zhu, Q. Xu, Nanomaterials derived from metal-organic frameworks,Nat. Rev. Mater. 3 (2017) 17075.
AC C
[33] W. Chaikittisilp, M. Hu, H. Wang, H.-S. Huang, T. Fujita, K. C.-W. Wu, L.-C. Chen, Y. Yamauchi, K. Ariga, Nanoporous carbons through direct carbonization of a zeolitic imidazolate framework for supercapacitor electrodes Chem. Commun. 48 (2012) 7259-7261. [34] Y. Liu, X. Xu, W. Miao, T. Lu, Z. Sun, L. Pan, Metal-organic frameworks-derived porous carbon polyhedrons for highly efficient capacitive deionization, Chem. Commun. 51 (2015) 12020-12023. [35] M. Ding, W. Shi, L. Guo, Z. Y. Leong, A. Baji, H. Y. Yang, Bimetallic Metal-Organic Frameworks Derived Porous Carbon Nanostructure for High Performance Membrane Capacitive Desalination, J. Mater. Chem. A 5 (2017) 6113-6121. 17
ACCEPTED MANUSCRIPT [36] R. H. Baughman, A. A. Zakhidov, W. A. de Heer, Carbon Nanotubes-the Route Toward Applications, Science 297 (2002) 787-792. [37] X. Zhang, L. Chen, T. Yuan, H. Huang, Z. Sui, R. Du, X. Li, Y. Lu, Q. Li, Dendrimer-linked, renewable and magnetic carbon nanotube aerogels, Mater. Horiz. 1 (2014) 232-236.
RI PT
[38] H. Li, L. Pan, Y Zhang, L. Zou, C. Sun, Y. Zhan, Z. Sun, Kinetics and thermodynamics study for electrosorption of NaCl onto carbon nanotubes and carbon nanofibers electrodes. Chem. Phys. Lett. 485 (2010) 161-166.
[39] H. Li, S. Liang, M. Gao, J. Li, Uniform carbon hollow sphere for highly efficient
SC
electrosorption, J. Porous Mater. 23 (2016) 1575-1580.
[40] R. Zhao, P. M. Biesheuvel, H. Miedema, H. Bruningand A. van der Wal, Charge Efficiency:
M AN U
A Functional Tool to Probe the Double-Layer Structure Inside of Porous Electrodes and Application in the Modeling of Capacitive Deionization Phy. Chem. Lett. 1 (2010) 205-210. [41] P. M. Biesheuvel, B. van Limpt, A. van der Wal, Dynamic Adsorption/Desorption Process Model for Capacitive Deionization, J. Phys. Chem. C, 113 (2009) 5630-5640. [42] C.-C. Wang, J. Liang, Y.-H. Liao, S.-Y. Lu, 3D porous graphene nanostructure from a simple,
TE D
fast, scalable process for high performance flexible gel-type supercapacitors, ACS Sustain. Chem. Eng. 5 (2017) 4457-4467.
[43] Z. Zhang, L.-L. Kong, S. L., G.-R. Li, X.-P. Gao, A High-Efficiency Sulfur/Carbon
EP
Composite Based on 3D Graphene Nanosheet@Carbon Nanotube Matrix as Cathode for Lithium-Sulfur Battery, Adv. Energy Mater. 7 (2017) 1602543. [44] J. Zhou, Y. Dou, A. Zhou, R.-M. Guo, M.-J. Zhao, J.-R. Li, MOF Template-Directed
AC C
Fabrication of Hierarchically Structured Electrocatalysts for Efficient Oxygen Evolution Reaction, Adv. Energy Mater. 7 (2017) 1602643. [45] Z. Wang, T. Yan, J. Fang, L. Shi, D. Zhang, Nitrogen-doped porous carbon derived from bimetallic metal-organic framework as highly efficient electrodes for flow-through deionization capacitors, J. Mater. Chem. A 4 (2016) 10858-10868. [46] Y. Lei, M. Gan, L. Ma, M. Jin, X. Zhang, G. Fu, P. Yang, M. Yan, Synthesis of nitrogen-doped porous carbon from zeolitic imidazolate framework-67 and phenolic resin for high performance supercapacitors, Ceram. Int. 43 (2017) 6502-6510. [47] J. Zhou, N. Lin, W. Cai, C. Guo, K. Zhang, J. Zhou, Y. Zhu, Y. Qian, Synthesis of S/CoS 2 18
ACCEPTED MANUSCRIPT Nanoparticles-Embedded N-doped Carbon Polyhedrons from Polyhedrons ZIF-67 and their Properties in Lithium-Sulfur Batteries, Electrochim. Acta, 218 (2016) 243-251. [48] Z. Wang, Y. Lu, Y. Yan, T. Y. P. Larissa, X. Zhang, D. Wuu, H. Zhang, Y. Yang, X. Wang, Core-shell carbon materials derived from metal-organic frameworks as an efficient oxygen
RI PT
bifunctional electrocatalyst, Nano Energy 30 (2016) 368-378. [49] B. L. Fu, Z. Liu, Y. Liu, B. Han, P. Hu, L. Cao, D. Zhu, Bead Cobalt Oxide Nanoparticlaes along Carbon Nanotubes: Towards Moer Highly Integrated Electronic Devices, Adv. Mater. 17 (2005) 217-221.
SC
[50] A. S. Claye, J. E. Fischer, C. B. Huffman, A. G. Rinzler, and R. E. Smalley, Solid-State Electrochemistry of the Li Single Wall Carbon Nanotube System, J. Electrochem. Soc. 147
M AN U
(2000) 2845-2852.
[51] C. Zheng, C. Cao, Z. Ali, J. Hou, Enhanced electrochemical performance of ball milled CoO
AC C
EP
TE D
for supercapacitor applications, J. Mater. Chem. A 2 (2014) 16467-16473.
19
ACCEPTED MANUSCRIPT Table 1. Atomic percent of C, N, O and Co. C1s (%)
N1s (%)
O1s (%)
Co2p (%)
HCN-500
72.68
10.65
10.65
6.03
HCN-600
83.56
6.29
6.5
3.64
HCN-650
83.93
7.05
5.87
3.16
HCN-700
85.46
6.74
4.91
2.89
Table 2. Textural characteristics for HCN.
RI PT
Sample
Sample
Surface area (m2· g-1)
Pore volume (cm3· g-1)
HCN-500
104.45
0.1264
HCN-600
125.11
0.1558
HCN-650
171.60
0.1748
4.0752
HCN-700
116.28
0.1427
4.9102
Pore width (nm) 4.8424
M AN U
SC
4.9795
Table 3. Simulated values of L, Rs, Cdl, Cps and Rct according to the equivalent circuit. L(H)
Rs (Ω)
HCN-500
1.063E-6
1.905
HCN-600
1.98E-20
1.265
HCN-650
2.766E-7
1.083
HCN-700
5.425E-7
0.881
TE D
Sample
Cdl (F)
Cps (F)
Rct (Ω )
0.4136
0.4619
123.2
0.3438
0.2102
231.9
0.3402
0.3417
293.2
0.2402
0.5272
431.3
Table 4. Coefficients of pseudo-first-order adsorption kinetics. First order equation
K1t 2.303
qe K1 R2
Applied potential 0.6V
0.8V
1.0V
1.2V
1.4V
1.66 0.03591 0.99622
3.2 0.03654 0.99283
4.52 0.02516 0.98603
5.48 0.04204 0.99675
7.08 0.03958 0.9892
AC C
EP
log(qe − q ) = log(qe ) −
Parameter
20
ACCEPTED MANUSCRIPT Figure caption Fig. 1 Schematics of the synthesis of HCN. Fig. 2 SEM images of (a) ZIF-67 nanocrystals and (b) HCN-650; TEM images of (c) HCN-500, (d) HCN-600, (e) HCN-650 and (f) HCN-700 (inset in each figure is the magnified TEM image of
RI PT
CNTs from the selected area). Fig. 3 (a) XRD patterns of ZIF-67 and HCNs, (b) TGA curve of ZIF-67, (c) Raman spectra of HCNs, (d) XPS spectra of HCNs, (e) N2 adsorption-desorption isotherm of HCNs, (f) pore size distribution.
SC
Fig. 4 (a) CV curve of HCNs at the scan rate of 10 mV·s-1, (b) specific capacitance with respect to scan rate, (c) GCD curve of HCNs at the current density of 0.8 A·g-1, (d) specific capacitance as
M AN U
function of current density, (e) Impedance spectra of HCNs, (f) equivalent circuit. Fig. 5 (a) Electrosorption behaviours of HCN-650 at 1.2 V with an initial conductivity of 1000 µS·cm-1, (b) corresponding current and voltage variation, (c) salt concentration vs. time at different voltages (Lines: theory, dots: experimental data), (d) electrosorption capacitance with
AC C
EP
TE D
respect to time according to pseudo-first-order adsorption kinetic.
21
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Figure 1
22
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Figure 2
23
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Figure 3
24
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Figure 4
25
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Figure 5
26