- Email: [email protected]
Zinc doped H2Ti12O25 Anode and Activated Carbon Cathode for Hybrid Supercapacitor with superior performance
Accepted Manuscript Title: Zinc doped H2 Ti12 O25 Anode and Activated Carbon Cathode for Hybrid Supercapacitor with superior performance Author: Hyeong-Jong Choi Seung-Hwan Lee Jin Hyeon Kim Hong-Ki Kim Jong-Myon Kim PII: DOI: Reference:
S0013-4686(17)31735-8 http://dx.doi.org/doi:10.1016/j.electacta.2017.08.094 EA 30098
To appear in:
Electrochimica Acta
Received date: Accepted date:
12-7-2017 14-8-2017
Please cite this article as:http://dx.doi.org/10.1016/j.electacta.2017.08.094 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.
Zinc doped H2Ti12O25 Anode and Activated Carbon Cathode for Hybrid Supercapacitor with superior
ip t
performance
Hong-Ki Kima, Jong-Myon Kimc,**
cr
Hyeong-Jong Choia, 1, Seung-Hwan Leeb, 1,*, Jin Hyeon Kima,
Dept. of Electronic Materials Engineering, Kwangwoon University, Korea
b
Institute for Research in Electronics and Applied Physics, University of Maryland, College
us
a
Dept. of IT Convergence, University of Ulsan, Ulsan 44610, Korea
M
c
an
Park, Maryland 20742, USA
d
Abstract
te
Hybrid supercapacitors were fabricated with both pristine and H2Ti12-xZnxO25-x (x=0.15, 0.3, 0.45, 0.6) anode electrodes. The structural properties and morphology of particles after zinc
Ac ce p
doping were confirmed by X-ray diffraction (XRD), Rietveld refinement, X-ray photoelectron spectroscopy (XPS), and scanning electron micron microscopy (SEM). The electrochemical performances of the hybrid supercapacitors were measured. The results show that the H2Ti11.7Zn0.3O24.7 not only enhanced discharge specific capacitances of 70.7 Fg-1 at 0.5 Ag-1, but also capacitance retention of 92% after 1000 cycles. In addition, H2Ti11.7Zn0.3O24.7 has lower polarization and charge transfer resistance (Rct) of 0.142 Ω. The power densities and energy densities of H2Ti11.7Zn0.3O24.7 were 42.4-8.1 Wh kg-1 and 1825676.2 W kg-1 at 0.1 and 3 Ag-1, respectively. Consequentially, Zn doping improved structure stability and electrochemical performance of H2Ti12O25. Therefore, hybrid supercapacitors were fabricated using a H2Ti11.7Zn0.3O24.7 anode can be regarded as energy storage devices. Keywords: Hybrid Supercapacitor; electrochemical performance; H2Ti12O25 Anode; Activated Carbon cathode; Zinc Doping 1
Page 1 of 33
2*
Corresponding author’s email: [email protected] Tel: +1-301-405-5323 Fax: +1-301-314-9269
3*
Corresponding author’s email: [email protected] Tel: +82-52-259-2217, Fax: +82-52-259-
1687 These authors contributed equally to this work.
ip t
1
cr
Introduction
us
Recently, worldwide depletion of fossil fuel with climate and environmental issues has garnered serious attention. Thus, the use of renewable and environmentally friendly energy resources has been growing. The necessity for the development of a device capable of storing
an
electric energy becomes an important issue. Especially, in order to develop hybrid electric vehicles (HEV) and electric vehicles (EV), energy storage devices with high performance
M
have been attracting attention. Lithium ion batteries and supercapacitors are the most promising candidates as a power source for HEV and EVs. However, lithium ion batteries have low power density (50-200 W kg-1) and short cycle life while supercapacitors have low
d
energy density (30-90 W h kg-1) and high power density (< 3kW kg-1) [1, 2]. To overcome the
te
disadvantages of lithium ion batteries and supercapacitors, a hybrid supercapacitor was designed. This combines the mechanism of lithium ion batteries and the supercapacitors.
Ac ce p
Therefore, the hybrid supercapacitors provide high power density, high energy density, and long cycle life [3]. Various materials including Li4Ti5O12 [4], TiO2(B) [5], H2Ti3O7 [6] and H2Ti12O25 [7] have been used for the anode in hybrid supercapacitors, lithium ion batteries. The flat potential plateau of H2Ti12O25 shows the capacity of 236 mAhg-1 at 1.55 V [6]. H2Ti12O25 has a higher capacity than Li4Ti5O12 [3] and TiO2(B) [6]. Also, H2Ti12O25 was confirmed for cycling performance [7]. In order to improve the electrochemical performance of an H2Ti12O25 anode, we proposed Zn doping. In this study, a hybrid supercapacitor using H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6) as the anode and activated carbon as the cathode was investigated to see the effects of different Zn doping content on the structural properties and electrochemical performance of hybrid supercapacitors.
2
Page 2 of 33
Experiments H2Ti12-xZnxO25-x powders were prepared via a solid state method. First, ZnO powders were mixed with TiO2 at a predetermined molar ratio of 0-0.6 using a ball mill for 24 h. After
ip t
drying, the mixture was calcined at 1350ºC for 10 h in air. The precursor Na2Ti3-xZnxO7-x was prepared by a solid-state method mixing Na2CO3 (99.5%) and Ti1-xZnxO2-x in a molar ratio of 1:3. This mixture was calcined at 800oC for 20 h in air. As result, the H2Ti3-xZnxO7-x power
cr
was fabricated by Na+/H+ ion exchange reaction of Na2Ti3-xZnxO7-x in 1M HCl solution for 3 days at 60oC. Finally, the prepared H2Ti3-xZnxO7-x powder was heated at 300ºC for 5 h in air.
us
As a result, H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6) powders were fabricated. The structure of H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6) was analyzed using X-ray diffraction (XRD).
an
Furthermore, X-ray diffraction data were analyzed with Rietveld refinement program (RietanFP). Field emission scanning electron microscopy was used to observe the morphology of Zn doped H2Ti12O25. Also, the composition of H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6) was
M
analyzed via X-ray photoelectron spectroscopy (XPS). The hybrid supercapacitor was composed of the H2Ti12-xZnxO25-x anode and activated carbon (AC) cathode. An AC cathode
d
was fabricated by mixing AC (MSP-20, 90 wt%) with conductive carbon (5 wt%) and polytetrafluoroethylene (PTFE, 5 wt%). In order to fabricate the anode, H2Ti12-xZnxO25-x
te
powder, conductive carbon black binder (Super P), and polyvinylidene fluoride (PVDF) were mixed in an 83:7:10 weight ratio. N-Methyl pyrrolidinine (NMP) solvent was added to
Ac ce p
produce the slurry. This was casted on aluminum foil to a thickness of 125 µm and then dried at 100oC to remove the NMP solvent. The aluminum foil was pressed to a thickness of 70 80 µm. To carry out half-cell test, the electrodes were assembled with Li metal counter electrode and separator in an argon-gas-filled glove box. Before being impregnated with a 1.5M solution of LiBF4 solution in 1:1 ethylene carbonate (EC): dimethyl carbonate (DMC) as the electrolyte, in order to remove the moisture in the cell, the fabricated cell was dried in a vacuum oven for 48 h. The electrodes of full-cells were prepared by the above-stated method and full-cells were assembled with using Zn-doped H2Ti12O25 as anode and activated carbon as cathode in argon-gas-filled glove box. The width of the cathode, separator, and anode were 28 cm, 40 cm, and 30 cm, respectively and the heights of the cathode and anode were both 3 cm. The electrochemical measurement of the fabricated cell was carried out with various tests, such as initial capacitance and rate capability, using an Arbin BT 2042 battery test system. 3
Page 3 of 33
Results and discussion Figure 1 (a) shows the X-ray diffraction (XRD) pattern of the synthesized H2Ti12-xZnxO25-x
d
M
an
us
cr
ip t
(x=0, 0.15, 0.3, 0.45, 0.6) anode.
Figure 1 (a) X-ray diffraction pattern of the pristine and H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45,
te
0.6) powders. (b) By Zn content, shifted XRD pattern between 24 and 26 degrees.
Ac ce p
When x is less than 0.3, it can be seen that all samples were well-crystallized without impurity. This means that doped Zn content successfully entered into the Ti site. However, the impurity phases of TiO2 are observed above x=0.45 due to excessive Zn content [8, 9]. Li, Jiangang et al. we can expect that impurity phase affects a negative effect on the electrochemical performance [10]. Figure 1 (b) shows the magnified XRD patterns of the (110) planes for H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6). The diffraction peaks of the samples shifted toward higher 2θ with increasing Zn concentration. The lattice parameters of the monoclinic cell can be calculated using lattice parameter of monoclinic equation:
(1) (2) 4
Page 4 of 33
In order to obtain more precise structural parameters (lattice parameters), Rietveld refinements were performed for XRD patterns of H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6). Figure 2 (a-e) show the Rietveld refinement plots of H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45,
te
d
M
an
us
cr
ip t
0.6).
Ac ce p
Figure 2 Rietveld refinement plots of H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6) samples. This confirmed good agreement between observed XRD patterns via intensity scale of line (Yobs-Ycal). The value of the goodness of fit (GOF) can be calculated by the following formula [11]:
!"#
GOF
!$%#
(3)
where ‘Rwp’ is the residuals for the weighted pattern and ‘Rexp’ is the expected error. The calculated the GOF of the H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6) are 1.304, 1.194, 1.174, 1.294 and 1.195, respectively. These values indicate that the experimental pattern fitted well with the simulated pattern. The lattice parameter and volume of H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6) from the Rietveld refinement are depicted in Table 1. 5
Page 5 of 33
Table 1. Lattice parameter and volume of H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6) from the Rietveld refinement. b (Å)
c (Å)
β
Volume of unit (Å)
PristineH2Ti12O25
13.865
3.791
9.822
110.172
484.599
H2Ti11.85Zn0.15O24.85
13.618
3.720
9.801
111.123
H2Ti11.7Zn0.3O24.7
13.583
3.708
9.797
111.556
H2Ti11.55Zn0.45O24.55
13.510
3.699
9.741
111.724
H2Ti11.4Zn0.6O24.4
13.502
3.689
9.738
ip t
a (Å)
463.148
us
cr
458.922
111.819
452.219 450.292
an
Sample
As a result, the lattice parameters of H2Ti12-xZnxO25-x samples were slightly smaller than that
M
of pristine H2Ti12O25 [12]. This is advantageous to the lattice stability during the insertion/extraction of Li+ ion [13, 14]. The lattice contraction was generated by transition from Ti3+ (0.67 Å) to Ti4+ (0.605 Å) due to Zn2+ (0.74 Å) doping [15]. The X-ray
d
photoelectron spectroscopy (XPS) measurement confirmed the Zn doping into the H2Ti12O25
Ac ce p
(x=0.15, 0.3, 0.45, 0.6).
te
structure. Figure 3 (a) shows the XPS spectra of pristine H2Ti12O25 and H2Ti12-xZnxO25-x
Figure 3 (a) X-ray photoelectron spectroscopy (XPS) spectra with different Zn/Ti molar ratios and (b) Zn 2p peak of H2Ti11.7Zn0.3O24.7.
6
Page 6 of 33
We can show the peaks of Ti 2p and Zn 2p binding energy. The pristine H2Ti12O25 shows the Ti 2p doublet peaks at binding energy of 458.77-463.80 eV, with spin orbit splitting of 5.03 eV, which is in good agreement with the Ti4+ oxidation state [16]. Also, in the case of H2Ti12xZnxO25-x
(x=0, 0.15, 0.3, 0.45, 0.6), the peak of Zn 2p was observed at binding energy of
ip t
1021.17 eV [17]. The Zn 2p weak peak of H2Ti11.7Zn0.3O24.7 can be explained by low content of Zn doping in Figure 3 (b). When compared with pristine H2Ti12O25, the Ti 2p peak of
cr
H2Ti12-xZnxO25-x (x=0.15, 0.3, 0.45, 0.6) were moved toward lower binding energy. This can be explained by the smaller binding energy inducing expanded bond length due to the Zn
us
dopants. Thus, the mobility of Li+ will improve during charge/discharge by the relatively longer bond length of Ti-O [18]. Also, we can assume that this phenomenon is closely related with structure distortion of H2Ti12O25 by Zn doping [19]. This means that Zn was successfully
an
entered into the Ti site. When Ti is substituted for with Zn, the cation (Ti) vacancies were caused due to the different balance of cation [19]. The cation vacancies were generated
()
& ' *+, &
M
according to the following defect equations [20]: -(
∙∙ )
')×
(4)
d
The free oxygen vacancies (V O • • ) were created to balance the electric charge [20, 21]. The
te
free oxygen vacancies are proportional to Zn contents. Also, the electron hopping probability is linked closely to oxygen vacancies. Thus, the electron hopping probability can improve the
Ac ce p
electronic conductivity and is expected to increase with increasing oxygen vacancies [22]. As a result, Zn doping will improve the electrochemical performance of hybrid supercapacitors. The electrical conductivities of all samples were measured using the four-point probe measurement (RTS-8 Four-Point probe meter). The table 2 shows the measured electric conductivity.
Table 2. Electric conductivity of H2Ti12-xZnxO25-x samples. x in H2Ti12xZnxO25-x
σ/ S cm‒ 1
0
0.15
0.3
0.45
0.6
6.61 x 10-10
5.38 x 10-9
7.42 x 10-8
6.17 x 10-9
8.33 x 10-10
7
Page 7 of 33
The SEM images show pristine H2Ti12O25 and H2Ti12-xZnxO25-x (x=0.15, 0.3, 0.45, 0.6) as
us
cr
ip t
shown in Figure 4.
Figure 4 SEM images of the (a) H2Ti12O25, (b) H2Ti11.85Zn0.15O24.85, (c) H2Ti11.7Zn0.3O24.7, (d)
an
H2Ti11.55Zn0.45O24.55 and (e) H2Ti11.4Zn0.6O24.4 powders.
The initial charge-discharge curves of the H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6) and AC
Ac ce p
te
d
M
were shown in Fig. 5 (a) and (b), respectively.
Figure 5 Initial charge/discharge curves of half-cell using (a) H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6) anode and (b) AC cathode. Cyclic voltammetry curves of the half-cell using (c) Zndoped H2Ti12O25 anodes and (d) AC cathode. 8
Page 8 of 33
The pristine H2Ti12O25 shows a flat voltage plateau near 1.55 V (versus Li/Li+) with the discharge capacity of 231 mAh g-1 at current rate of 200 mA g-1, which is the lowest value than others. The discharge capacity increases with increasing Zn doping and reaches a maximum value of 270 mAh g-1 at H2Ti11.7Zn0.3O24.7. It can be inferred that the Li ion kinetics
ip t
is enhanced by the Zn doping. The H2Ti12-xZnxO25-x (x=0.45, 0.6) shows inferior to H2Ti11.7Zn0.3O24.7, which is consistent with XRD result. The initial charge-discharge curve of
cr
the activated carbon shows the linear behavior due to physical charge-storage mechanism. The activated carbon shows a capacity of 49 mAh g-1 at a current rate of 20 mA g-1.
us
To better understand individual electrodes, the CV curves of Li half-cell using AC cathode and Zn-doped H2Ti12O25 anodes were measured. CV curves of the pristine and Zn-doped H2Ti12O25 anodes show sharp peaks derived from redox reactions of Ti4+ ions. The potential
an
difference between anodic and cathodic peak decreased until the Zn doping level reaches x=0.3, as shown in Figure 5 (c). The anodic peaks (Ea) and cathodic peaks (Ec) of pristine
M
H2Ti12O25 and H2Ti12-xZnxO25-x (x=0.15, 0.3, 0.45, 0.6) anode, as seen in the CV curve (versus Li/Li+), are shown in Table 3.
Ec(V)
∆E(Ea -Ec) (V)
2.5
2.3
0.2
Ac ce p
Pristine H2Ti12O25
Ea(V)
te
Sample
d
Table 3. Polarization of the pristine and H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6).
H2Ti11.85Zn0.15O25
2.57
2.45
0.12
H2Ti11.7Zn0.3O25
2.6
2.5
0.1
H2Ti11.55Zn0.45O25
2.53
2.4
0.13
H2Ti11.4Zn0.6O25
2.54
2.38
0.16
The H2Ti12-xZnxO25-x (x=0.15, 0.3, 0.45, 0.6) anode shows higher oxidation peak current than pristine H2Ti12O25. The potential difference (∆E=Ea-Ec) between the anodic and cathodic peak is representative of the reaction kinetics [18] and is closely related to the polarization. The higher electronic conductivity can induce lower polarization, which improved the transfer rate of Li+ and the reversibility [8, 13]. Table 3 shows the polarization of pristine 9
Page 9 of 33
H2Ti12O25 and H2Ti12-xZnxO25-x (x=0.15, 0.3, 0.45, 0.6) anodes. The H2Ti12-xZnxO25-x (x=0.15, 0.3, 0.45, 0.6) anode have lower polarization and higher oxidation-reduction peak currents than a pristine H2Ti12O25 anode; the H2Ti11.7Zn0.3O24.7 anode shows the lowest polarization and largest CV area. This means that the Li+ diffusion and electronic conductivity were
ip t
improved with Zn addition. These results of the CV curves confirm that the capacitance of hybrid supercapacitors with H2Ti12-xZnxO25-x (x=0.15, 0.3, 0.45, 0.6) anode improved
cr
electrochemical performance. On the contrary, the CV of the activated carbon in the potential ranges 3.0-4.5 V at a scan rate of 2 mV s-1, as shown in Figure 5 (d). The measured curve
us
shows a typical rectangular shape, which is type of common electrical double layer capacitance (EDLC) [5]. As a result, both capacitive and redox reactions were generated in
an
the hybrid supercapacitor system.
Multiple experiments related to the electrode thickness of hybrid supercapacitors were conducted to optimize the cell balancing [7]. It is more associated with thickness of H2Ti12O25,
M
due to its slower Li+ insertion/extraction than BF4- ion adsorption/desorption at activated carbon. The discharge capacitance retentions according to the different electrode thicknesses
d
were shown in Table 4.
0.1 A
Ac ce p
HTO-AC
te
Table 4. Discharge capacitance retention for different electrode thickness
thickness
90 µm - 240 µm 90 µm - 180 µm
100 (63 Fg-1)
0.5 A
1A
2A
3A
93
85
68
48
-1
94
87
70
50
-1
100 (58 Fg )
70 µm - 240 µm
100 (60 Fg )
95
88
71
52
70 µm - 180 µm
100 (53 Fg-1)
93
85
69
49
50 µm - 240 µm
100 (56 Fg-1)
94
86
70
51
50 µm - 180 µm
100 (47 Fg-1)
89
81
64
43
40 µm - 240 µm
100 (50 Fg-1)
92
85
68
48
93
86
69
47
40 µm - 180 µm
-1
100 (43 Fg )
10
Page 10 of 33
At 0.1 A, the initial capacitances were increased proportionally with electrode thickness because both electrodes have sufficient time for electrochemical reactions. However, with increasing the current rate, the capacitance retention is decreased. The reason for this is sluggish kinetics of Li+ at H2Ti12O25. Therefore, H2Ti12O25 anode can’t use its entire thickness,
ip t
except for the upper part. On the contrary, with increasing the thickness of cathode, the capacitance retention increased regardless of current density. Based on these results, we can
cr
select the thickness composition of 70-240 µm, indicating capacitance retention of 52 % at 3.0 A as an optimum thickness.
us
The cyclic voltammetry (CV) curves of the hybrid supercapacitors with H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6) anode are shown in Figure 6(a) at a scan rate of 10 mV s-1 between
Ac ce p
te
d
M
an
0 V and 2.8 V (working voltage of activated carbon 4.3 V [23] and HTO 1.5 V [6]).
Figure 6(a) Cyclic voltammetry, (b) initial charge-discharge and (c) electrochemical impedance spectra curves of the hybrid supercapacitors using the pristine and H2Ti12-xZnxO25x
(x=0, 0.15, 0.3, 0.45, 0.6. (d) Rate capabilities and (e) cycling performance of the hybrid
supercapacitors using the pristine and H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6) from 0.1 to 3 Ag-1. 11
Page 11 of 33
It can be inferred that H2Ti11.7Zn0.3O24.7 can have a positive effect on hybrid supercapacitors. The reason for this is accelerated kinetics of Li+ by Zn doping in H2Ti12O25. Based on above results, the capacitive and redox reactions coexist in the hybrid supercapacitors.
ip t
Figure 6(b) shows the initial charge/discharge curves for the H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6) anode between 0 V and 2.8 V. The discharge specific capacitances of the hybrid supercapacitors using the H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6) anode were obtained
1
∆3×4
5 ∆6 ∆3×4
us
0
cr
using the following relationship [24]:
(5)
where 'C' is the capacitance (Fg-1), '∆V' is the voltage change, 'm' is the mass of the active
an
materials in both electrodes, 'q' is the total charge, 'i' is the current, and 't' is time. These results indicate that the capacitances of the H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6) anodes
M
are 57.3, 65.05, 70.7, 60.1, and 58.3 Fg-1, respectively, at current density of 0.5 Ag-1. As a result, the capacitance of hybrid supercapacitors with H2Ti11.7Zn0.3O24.7 anode had the highest value. In the case of the H2Ti12-xZnxO25-x (x=0.15, 0.3) anode, the capacitances were higher
d
than pristine H2Ti12O25. Also, the voltage drop (IR drop) was observed in the initial
te
charge/discharge curves of the H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6) anode during the discharge process. This IR drop shows the effect of internal resistance in the hybrid
Ac ce p
supercapacitor. In the initial charge/discharge curves, the discharge region involves an IR drop owing to the internal resistance. Also, a faradaic discharge process is related to the lithium ion insertion/extraction as well as a non-faradaic discharge process is associated with the adsorption/desorption of ions. The voltage change was caused by ion separation at the interface of the electrode and the electrolyte [25]. The IR drop can be calculated by the following equation [26]:
7
389:;<$ =3>?@89:;<$
(6)
'Vcharge' is the voltage of the cell at the end charge, 'Vdischarge' is the voltage of the cell at the starting discharge, and 'I' is the absolute value of charge and discharge current. The calculated internal resistance of the H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6) anode are 0.19, 0.161, 0.142, 0.177, and 0.183 Ω, respectively. The IR drop is significantly correlated with the 12
Page 12 of 33
charge transfer resistance (Rct) of the hybrid supercapacitor [27]. These results indicated that the enhanced Li+ diffusion and electronic conductivity by oxygen vacancies may contribute to improve of capacitance in the hybrid supercapacitor.
ip t
To investigate the effect of the H2Ti12-xZnxO25-x anode on the electrolyte/H2Ti12-xZnxO25-x interface resistances, electrochemical impedance spectra (EIS) curves were performed in the frequency range of 10-1 to 10-6 Hz at a voltage of 2.8V, as shown in Figure 6(c). In the EIS
cr
results, a semicircle and straight line in the high and low frequency region were observed and showed the charge transfer resistance (Rct) and internal resistance (Ri) related to Li+
us
interfacial transfer at high frequency, respectively. The Rct of the H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6) anodes were approximately 0.06, 0.043, 0.034, 0.053, and 0.058 Ω,
an
respectively. As shown in Figure 6(c), the Rct values of H2Ti12-xZnxO25-x (x=0.15, 0.3) decreased. In addition, electrochemical polarization decreased with decreasing the Rct value [15]. However, Rct values of H2Ti12-xZnxO25-x (x=0.45, 0.6) increased. The straight line at the
M
low frequency is caused by Warburg impedance, which corresponds to Li+ diffusion in the H2Ti12O25 anode [28, 29]. In addition, we can calculate the diffusion coefficient (DLi) of Li+
d
by the following equation [30, 31]:
te
AB
(D
!(
E FG
)
(7)
Ac ce p
where 'R' is the gas constant, 'T'' is the absolute temperature, 'A' is the surface area of the anode, 'n' is the number of electrons, 'F' is the Faraday’s constant, 'C' is the concentration of Li+, and 'σ' is the Warburg impedance coefficient. The calculated diffusion coefficients of Li+ of the samples are 2.937 x 10-7, 5.637 x 10-7, 7.144 x 10-7, 6.307 x 10-7, and 5.259 x 10-7 cm2 s-1 for H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6) anodes, respectively. Based on these results, the H2Ti12-xZnxO25-x (x=0.15, 0.3) anode had a lower Rct value and a faster the Li+ diffusion rate than that of pristine H2Ti12O25. Therefore, we can expect that zinc doping of the anode will improve the electrochemical performance by increasing their electrical conductivity [9, 32]. Figure 6(d) shows the rate capability of the pristine H2Ti12O25 and H2Ti12-xZnxO25-x (x=0.15, 0.3, 0.45, 0.6) anodes at different charge-discharge rates of 0.1, 0.5, 1, 2, and 3 Ag-1. Regarding the rate capability, the capacitance retention of pristine H2Ti12O25 anode appears 52% 13
Page 13 of 33
at 3 Ag-1. On the other hand, the H2Ti12-xZnxO25-x (x=0.15, 0.3, 0.45, 0.6) anodes had capacitance retention of approximately 61.5%, 65%, 59% and 54%, respectively at the 3 Ag-1. The rate capability of the zinc doped H2Ti12O25 anode was improved and the properties of the samples were enhanced with the increasing capacitance retention. Figure 6(d) shows H2Ti12(x=0.15, 0.3, 0.45, 0.6) anode have a much better rate capability than pristine
ip t
xZnxO25-x
H2Ti12O25. As a result, enhanced electronic conductivity of H2Ti12-xZnxO25-x (x=0.15, 0.3,
cr
0.45, 0.6) anodes because of oxygen vacancies leads to superior capacitance retention [32]. This shows a similar trend as shown in Figure 8.
xZnxO25-x
us
Figure 6(e) shows the cycle performance of the hybrid supercapacitors with the H2Ti12(x=0, 0.15, 0.3, 0.45, 0.6) anode after 1000 cycles compared with 1 cycle at 3.0 Ag-
1
x
an
. It was found that the cycling behavior of the hybrid supercapacitor with the H2Ti12-xZnxO25(x=0.15, 0.3) anodes is better than that of pristine H2Ti12O25 during 1000 cycles. Also, the
hybrid supercapacitors with H2Ti12-xZnxO25-x (x=0.15, 0.3) anodes have specific capacitance
M
of 41.03 and 46.67 Fg-1 for the first discharge process, respectively, and maintains at 36.68 and 43.1 Fg-1 at 1000 cycles, which shows 89 and 92%, respectively, after 1000 cycles compared with 1 cycle. On the other hand, the hybrid supercapacitors with H2Ti12-xZnxO25-x
d
(x=0.45, 0.6) anodes maintain capacitances of 31.4 and 26.23 Fg-1, respectively, at 1000
te
cycles, and are 83 and 77% after 1000 cycles, respectively, compared to 1 cycle. Excessive doping of zinc (x=0.45, 0.6) induced large distortion of the lattice and increased the electrical
Ac ce p
resistance. Thus, capacitance and cycle performance is decreased [33]. As shown in Figure 10, gradually enhanced cycle performances were confirmed with Zn doping content of x=0.15, 0.3 by increasing electrical conductivity. Generally, the shrinking of the lattice or collapse of structure was caused by repetition of lithium insertion/extraction; the unchangeable radius of Zn would prevent the shrinking or collapse of structure [34, 35] and improve the structure stability. Thus, cycle performance was improved by enhanced electric conductivity and structure stability.
Figure 7 shows the Ragone plot of the hybrid supercapacitors with H2Ti11.7Zn0.3O24.7 anode with various hybrid supercapacitors.
14
Page 14 of 33
ip t cr us
an
Figure 7 Ragone plots of hybrid supercapacitors using H2Ti12-xZnxO25-x.
The energy and power densities were observed by discharging a charged device at various
I
M
current densities and were calculated using the following relationship [26]:
∆J × 4
(8)
J
I×L
(9)
d
K
te
∆J
MN:% OMN?P
(10)
Ac ce p
where ′J4 R ′ is the potential at the starting discharge, ′J4 ′ is the potential at the end discharge, ‘I’ is the charge and discharge currents, ‘m’ is the mass of active materials including the anode and cathode electrodes, and ‘t’ is the discharge time in the hybrid supercapacitor. The maximum energy and power densities of hybrid supercapacitors with H2Ti11.7Zn0.3O24.7 anode and activated carbon cathode material were 44.4 Whkg-1/182 Wkg-1 and 9.1Whkg-1/5676.2 Wkg-1 at 0.1 and 3 Ag-1 rates, respectively. We compared the power and energy densities of H2Ti11.7Zn0.3O24.7/AC with other hybrid supercapacitors composed of TNW/CNT [5], C-LTP/AC [26], C-LTO/AC [36] and C-LTP/AC [37]. As a result, the energy and power densities of the hybrid supercapacitors using H2Ti11.7Zn0.3O24.7 anode and activated carbon cathode are superior to other systems. The extraordinary electrochemical performance can be attributed to the faster kinetic of Li ion comparable to that of BF4- ion by inherent tunnel structure of H2Ti12O25 and improved conductivity by Zn doping. Consequently, this 15
Page 15 of 33
new hybrid system is expected to be sufficient for high energy storage devices in various applications.
ip t
Conclusion
In this study, hybrid supercapacitors using H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6) anode
cr
were successfully fabricated. We investigated the effect on the electrochemical performance of Zn doping. The Zn plays roles of superior reversible capacitance and lower polarization
us
and caused cation vacancies in the H2Ti12O25. In addition, the Zn doping improved structure stability to prevent lattice shrinking and structure collapse. The doped Zn content
an
significantly improved both initial specific capacitance and the capacitance retention. Also, the hybrid supercapacitor using an H2Ti11.7Zn0.3O24.7 anode showed better power density as well as energy density than the other system of lithium ion secondary batteries and
te
Acknowledgments
d
performance of hybrid supercapacitors.
M
supercapacitors. The H2Ti11.7Zn0.3O24.7 anode is important to improving the electrochemical
Ac ce p
This work was supported by the Korea Institute of Energy Technology Evaluation and Planning (KETEP) and the Ministry of Trade, Industry & Energy (MOTIE) of the Republic of Korea (No. 20162220100050, No. 20161120100350), in part by The Leading Human Resource Training Program of Regional Neo industry through the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT and future Planning (NRF-2016H1D5A1910564), and in part by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (2016R1D1A3B03931927).
16
Page 16 of 33
Reference [1] M. Yang, Y. Zhong, J. Ren, X. Zhou, J. Wei, Z. Zhou, Electrochemical Capacitors: Fabrication of High-Power Li-Ion Hybrid Supercapacitors by Enhancing the Exterior Surface
ip t
Charge Storage, Adv. Energy Mater. 5 (2015) 1500550 [2] M. Yang, Y. Zhong, J. Bao, X. Zhou, J. Wei, Z. Zhou, Achieving battery-level energy density by constructing aqueous carbonaceous supercapacitors with hierarchical porous N-
cr
rich carbon materials, J. Mater. Chem. A 3 (2015) 11387
[3] V. H. Nguyen, H. B. Gu, Synthesis and Characterization of Silver Vanadium Oxide as a
us
Cathode for Lithium Ion Batteries, Trans. Electr. Electron. Mater 17 (2016) 139
[4] S. H. Lee, S. G. Lee, J. R. Yoon, H. K. Kim, Novel performance of ultrathin AlPO4
an
coated H2Ti12O25 Exceeding Li4Ti5O12 in cylindrical hybrid supercapacitor, J. Power Sources 273 (2015) 839
[5] Q. Wang, Z. Wen, J. Li, A Hybrid Supercapacitor Fabricated with a Carbon Nanotube
M
Cathode and a TiO2–B Nanowire Anode, Adv. Funct. Mater. 16 (2006) 2141 [6] J. Akimoto, K. Chiba, N. Kijima, H. Hayakawa, S. Hayashi, Y. Gotoh, Y. Idemotob, SoftChemical Synthesis and Electrochemical Property of H2Ti12O25 as a Negative Electrode
d
Material for Rechargeable Lithium-Ion Batteries, J. Electrochem. Soc. 158 (2011) A546
te
[7] S. H. Lee, H. K. Kim, J. H. Lee, S. G. Lee, Y. H. Lee, Fabrication and electrochemical properties of cylindrical hybrid supercapacitor using H2Ti12O25 as anode material, Materials
Ac ce p
Letters 143 (2015) 101
[8] B. Zhang, H. Du, B. Li, F. Kang, Structure and Electrochemical Properties of Zn-Doped Li4Ti5O12 as Anode Materials in Li-Ion Battery, Electrochem. Solid. St. 13 (2010) A36 [9] G. Lu, S. Qiu, J. Liu, X. Wang, C. He, Y. J. Bai, Enhanced Electrochemical Performance of Zn-Doped Fe3O4 with Carbon Coating, Electrochim. Acta 117 (2014) 230 [10] J. Li, X. He, R. Zhao, C. Wan, C. Jiang, D. Xia, S. Zhang, Stannum doping of layered LiNi3/8Co2/8Mn3/8O2 cathode materials with high rate capability for Li-ion batteries, J. Power Sources 158 (2006) 524 [11] R. A. Young, D. B. Wiles, Profile shape functions in Rietveld refinements, J. Appl. Cryst. 15 (1982) 430 17
Page 17 of 33
[12] Akimoto, Junji. et al. U.S. Patent No. 8,309,253. (2012) [13] D. Wu, Y. Cheng, Enhanced high-rate performance of sub-micro Li4Ti4. 95Zn0. 05O12 as anode material for lithium-ion batteries, Ionics 19 (2013) 395
ip t
[14] C. Deng, S. Zhang, S. Y. Yang, B. L. Fu, L. Ma, Synthesis and characterization of Li2Fe0.97M0.03SiO4 (M=Zn2+, Cu2+, Ni2+) cathode materials for lithium ion batteries, J. Power
cr
Sources 196 (2011) 386
[15] T. F. Yi, H. Liu, Y. R. Zhu, L. J. Jiang, Y. Xie, R. S. Zhu, Improving the high rate
us
performance of Li4Ti5O12 through divalent zinc substitution, J. Power Sources 215 (2012) 258
an
[16] Z. Ali, S. N. Cha, J. I. Sohn, I. Shakir, C. Yan, J. M. Kim, D. J. Kang, Design and evaluation of novel Zn doped mesoporous TiO2 based anode material for advanced lithium ion batteries, J. Mater. Chem. 22 (2012) 17625
d
M
[17] M. E. Fragala, I. Cacciotti, Y. Aleeva, R. L. Nigro, A. Bianco, G. Malandrino, C. Spinella, G. Pezzottid, G. Gusmano, Core–shell Zn-doped TiO2–ZnO nanofibers fabricated via a combination of electrospinning and metal–organic chemical vapour deposition, Cryst Eng Comm. 12 (2010) 3858
te
[18] G. Yang, H. Ni, H. Liu, P. Gao, H. Ji, S. Roy, J. Pinto, X. Jiang, The doping effect on the crystal structure and electrochemical properties of LiMnxM1−x PO4 (M= Mg, V, Fe, Co, Gd), J. Power Sources 196 (2011) 4747
Ac ce p
[19] Z. H. Wang, L. X. Yuan, M. Wu, D. Sun, Y. H. Huang, Effects of Na+ and Cl− co-doping on electrochemical performance in LiFePO4/C, Electrochim. Acta 56 (2011) 8477 [20] W. Y. Wang, D. F. Zhang, T. Xu, X. F. Li, T. Zhou, X. L. Chen, Nonlinear electrical behavior and dielectric properties of (Ca, Ta)-doped TiO ceramics, J. Alloy. Comp. 335 (2002) 210 [21] H. Beltrán, M. Prades, N. Masó, E. Cordoncillo, A. R. West, Voltage-Dependent LowField Bulk Resistivity in BaTiO3:Zn Ceramics, J. Am. Ceram. Soc. 93 (2010) 500 [22] J. M. Wu, C. J. Chen, Dielectric properties of (Ba, Nb) doped TiO2 ceramics: migration mechanism and roles of (Ba, Nb), J. Mater. Sci. 23 (1988) 4157 [23] S. Dsoke, B. Fuchs, E. Gucciardi, M. W. Mehrens, The importance of the electrode mass ratio in a Li-ion capacitor based on activated carbon and Li4Ti5O12, J. Power Sources 282 18
Page 18 of 33
(2015) 385 [24] V. Aravindan, Y. L. Cheah, W. F. Mak, G. Wee, B. V. R. Chowdari, S. Madhavi, Fabrication of High Energy‐Density Hybrid Supercapacitors Using Electrospun V2O5
ip t
Nanofibers with a Self‐Supported Carbon Nanotube Network, Chempluschem 77 (2012) 570 [25] N. Yu, L. Gao, Electrodeposited PbO2 thin film on Ti electrode for application in hybrid
cr
supercapacitor, Electrochem. commun. 11 (2009) 220
[26] J. Y. Luo, Y. Y. Xia, Electrochemical profile of an asymmetric supercapacitor using
us
carbon-coated LiTi2(PO4)3 and active carbon electrodes, J. Power Sources 186 (2009) 224 [27] A. Chu, P. Braatz, Comparison of commercial supercapacitors and high-power lithium-
an
ion batteries for power-assist applications in hybrid electric vehicles: I. Initial characterization, J. Power Sources 112 (2012) 236
M
[28] S. S. Zhang, K. Xu, T. R. Jow, T.R. EIS study on the formation of solid electrolyte interface in Li-ion battery, Electrochim. Acta. 51 (2006) 1636
d
[29] S. Zhang, M. S. Ding, K. Xu, J. Allen, T. R. Jow, Understanding solid electrolyte
te
interface film formation on graphite electrodes, Electrochem Solid. St. 4 (2001) A206 [30] J. R. Yoon, E. Baek, H. K. Kim, M. Pecht, S. H. Lee, Critical dual roles of carbon
Ac ce p
coating in H2Ti12O25 for cylindrical hybrid supercapacitors, Carbon 101 (2016) 9 [31] T. F. Yi, Y. Xie, J. Shu, Z. Wang, C. B. Yue, R. S. Zhu, H. B. Qiao, Structure and Electrochemical Performance of Niobium-Substituted Spinel Lithium Titanium Oxide Synthesized by Solid-State Method, Electrochem. Soc. 158 (2011) A266 [32] J. Y. Luo, L. J. Chen, Y. J. Zhao, P. He, Y. Y. Xia, The effect of oxygen vacancies on the structure and electrochemistry of LiTi2(PO4)3 for lithium-ion batteries: A combined experimental and theoretical study, J. Power Sources 194 (2009) 1075 [33] W. Wang, J. Zhang, Z. Jia, C. Dai, Y. Hu, J. Zhou, Q. Xiao, Enhancement of the cycling performance of Li3V2(PO4)3/C by stabilizing the crystal structure through Zn2+ doping, Phys. Chem. Chem. Phys. 16 (2014) 13858 [34] Y. Hu, J. Yao, Z. Zhao, M. Zhu, Y. Li, H. Jin, H. Zhao, J. Wang, ZnO-doped LiFePO4 19
Page 19 of 33
cathode material for lithium-ion battery fabricated by hydrothermal method, Materials Chemistry and Physics. 141 (2013) 835 [35] Y. Yang, W. Xu, R. Guo, L. Liu, S. Wang, D. Xie, Y. Wan, J. Power Sources 269 (2014) 15
ip t
[36] H. G. Jung, N. Venugopal, B. Scrosati, Y. K. Sun, Synthesis and electrochemical properties of Zn-doped, carbon coated lithium vanadium phosphate cathode materials for
cr
lithium-ion batteries, J. Power Sources 221 (2013) 266
[37] V. Aravindan, W. Chuiling, M. V. Reddy, G. V. S. Rao, B. V. R. Chowdari, S. Madhavi,
us
Carbon coated nano-LiTi2(PO4)3 electrodes for non-aqueous hybrid supercapacitors, Phys.
an
Chem. Chem. Phys. 14 (2012) 5808
Figure caption,
M
Figure 1 (a) X-ray diffraction pattern of the pristine and H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6) powders. (b) By Zn content, shifted XRD pattern between 24 and 26 degrees.
d
Figure 2 Rietveld refinement plots of H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6) samples.
te
Figure 3 (a) X-ray photoelectron spectroscopy (XPS) spectra with different Zn/Ti molar ratios
Ac ce p
and (b) Zn 2p peak of H2Ti11.7Zn0.3O24.7.
Figure 4 SEM images of the (a) H2Ti12O25, (b) H2Ti11.85Zn0.15O24.85, (c) H2Ti11.7Zn0.3O24.7, (d) H2Ti11.55Zn0.45O24.55 and (e) H2Ti11.4Zn0.6O24.4 powders. Figure 5 Initial charge/discharge curves of half-cell using (a) H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6) anode and (b) AC cathode. Cyclic voltammetry curves of the half-cell using (c) Zndoped H2Ti12O25 anodes and (d) AC cathode. Figure 6(a) Cyclic voltammetry, (b) initial charge-discharge and (c) electrochemical impedance spectra curves of the hybrid supercapacitors using the pristine and H2Ti12-xZnxO25x
(x=0, 0.15, 0.3, 0.45, 0.6. (d) Rate capabilities and (e) cycling performance of the hybrid
supercapacitors using the pristine and H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6) from 0.1 to 3 Ag-1. 20
Page 20 of 33
Figure 7 Ragone plots of hybrid supercapacitors using H2Ti12-xZnxO25-x. Table caption
ip t
Table 1. Lattice parameter and volume of H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6) from the Rietveld refinement. Table 2. Electric conductivity of H2Ti12-xZnxO25-x samples.
cr
Table 3. Polarization of the pristine and H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6).
Ac ce p
te
d
M
an
us
Table 4. Discharge capacitance retention for different electrode thickness
21
Page 21 of 33
Ac ce p
te
d
M
an
us
cr
ip t
Figure 1
22
Page 22 of 33
Ac ce p
te
d
M
an
us
cr
ip t
Figure 2
23
Page 23 of 33
Ac ce p
te
d
M
an
us
cr
ip t
Figure 3
24
Page 24 of 33
Ac ce p
te
d
M
an
us
cr
ip t
Figure 4
25
Page 25 of 33
Ac ce p
te
d
M
an
us
cr
ip t
Figure 5
26
Page 26 of 33
Ac ce p
te
d
M
an
us
cr
ip t
Figure 6
27
Page 27 of 33
Ac ce p
te
d
M
an
us
cr
ip t
Figure 7
28
Page 28 of 33
Table.1
b (Å)
c (Å)
β
Volume of unit (Å)
PristineH2Ti12O25
13.865
3.791
9.822
110.172
484.599
H2Ti11.85Zn0.15O24.85
13.618
3.720
9.801
111.123
H2Ti11.7Zn0.3O24.7
13.583
3.708
9.797
111.556
H2Ti11.55Zn0.45O24.55
13.510
3.699
9.741
111.724
H2Ti11.4Zn0.6O24.4
13.502
3.689
9.738
ip t
a (Å)
463.148
us
cr
458.922
111.819
452.219 450.292
Ac ce p
te
d
M
an
Sample
29
Page 29 of 33
Table. 2
xZnxO25-x
0.15
0.3
0.45
0.6
6.61 x 10-10
5.38 x 10-9
7.42 x 10-8
6.17 x 10-9
8.33 x 10-10
Ac ce p
te
d
M
an
us
cr
σ/ S cm‒ 1
0
ip t
x in H2Ti12-
30
Page 30 of 33
Table. 3
Ec(V)
∆E(Ea -Ec) (V)
Pristine H2Ti12O25
2.5
2.3
0.2
H2Ti11.85Zn0.15O25
2.57
2.45
H2Ti11.7Zn0.3O25
2.6
2.5
H2Ti11.55Zn0.45O25
2.53
2.4
H2Ti11.4Zn0.6O25
2.54
2.38
ip t
Ea(V)
0.12
cr
0.1
0.13
0.16
Ac ce p
te
d
M
an
us
Sample
31
Page 31 of 33
Table. 4
90 µm - 180 µm 70 µm - 240 µm
100 (63 Fg-1)
0.5 A
1A
2A
3A 48
93
85
68
-1
94
87
70
-1
95
88
71
-1
69
100 (58 Fg ) 100 (60 Fg ) 100 (53 Fg )
93
85
50 µm - 240 µm
100 (56 Fg-1)
94
86
50 µm - 180 µm
100 (47 Fg-1)
89
81
40 µm - 240 µm
100 (50 Fg-1)
92
85
40 µm - 180 µm
100 (43 Fg-1)
93
86
50 52 49
70
51
64
43
68
48
69
47
Ac ce p
te
d
M
an
70 µm - 180 µm
ip t
90 µm - 240 µm
0.1 A
cr
thickness
us
HTO-AC
32
Page 32 of 33
Highlights - Hybrid supercapacitors were fabricated with H2Ti12-xZnxO25-x (x=0~0.6) anode. - Capacitance retention of H2Ti11.7Zn0.3O24.7 was 92% after 1000 cycles. - Power densities of H2Ti11.7Zn0.3O24.7 were 42.4-8.1 Wh kg-1 from 0.1 to 3 Ag-1.
Ac ce
pt
ed
M
an
us
cr
ip t
- Energy densities of H2Ti11.7Zn0.3O24.7 were 182-5676.2 W kg-1 from 0.1 to 3 Ag-1.
Page 33 of 33
S0013-4686(17)31735-8 http://dx.doi.org/doi:10.1016/j.electacta.2017.08.094 EA 30098
To appear in:
Electrochimica Acta
Received date: Accepted date:
12-7-2017 14-8-2017
Please cite this article as:
Zinc doped H2Ti12O25 Anode and Activated Carbon Cathode for Hybrid Supercapacitor with superior
ip t
performance
Hong-Ki Kima, Jong-Myon Kimc,**
cr
Hyeong-Jong Choia, 1, Seung-Hwan Leeb, 1,*, Jin Hyeon Kima,
Dept. of Electronic Materials Engineering, Kwangwoon University, Korea
b
Institute for Research in Electronics and Applied Physics, University of Maryland, College
us
a
Dept. of IT Convergence, University of Ulsan, Ulsan 44610, Korea
M
c
an
Park, Maryland 20742, USA
d
Abstract
te
Hybrid supercapacitors were fabricated with both pristine and H2Ti12-xZnxO25-x (x=0.15, 0.3, 0.45, 0.6) anode electrodes. The structural properties and morphology of particles after zinc
Ac ce p
doping were confirmed by X-ray diffraction (XRD), Rietveld refinement, X-ray photoelectron spectroscopy (XPS), and scanning electron micron microscopy (SEM). The electrochemical performances of the hybrid supercapacitors were measured. The results show that the H2Ti11.7Zn0.3O24.7 not only enhanced discharge specific capacitances of 70.7 Fg-1 at 0.5 Ag-1, but also capacitance retention of 92% after 1000 cycles. In addition, H2Ti11.7Zn0.3O24.7 has lower polarization and charge transfer resistance (Rct) of 0.142 Ω. The power densities and energy densities of H2Ti11.7Zn0.3O24.7 were 42.4-8.1 Wh kg-1 and 1825676.2 W kg-1 at 0.1 and 3 Ag-1, respectively. Consequentially, Zn doping improved structure stability and electrochemical performance of H2Ti12O25. Therefore, hybrid supercapacitors were fabricated using a H2Ti11.7Zn0.3O24.7 anode can be regarded as energy storage devices. Keywords: Hybrid Supercapacitor; electrochemical performance; H2Ti12O25 Anode; Activated Carbon cathode; Zinc Doping 1
Page 1 of 33
2*
Corresponding author’s email: [email protected] Tel: +1-301-405-5323 Fax: +1-301-314-9269
3*
Corresponding author’s email: [email protected] Tel: +82-52-259-2217, Fax: +82-52-259-
1687 These authors contributed equally to this work.
ip t
1
cr
Introduction
us
Recently, worldwide depletion of fossil fuel with climate and environmental issues has garnered serious attention. Thus, the use of renewable and environmentally friendly energy resources has been growing. The necessity for the development of a device capable of storing
an
electric energy becomes an important issue. Especially, in order to develop hybrid electric vehicles (HEV) and electric vehicles (EV), energy storage devices with high performance
M
have been attracting attention. Lithium ion batteries and supercapacitors are the most promising candidates as a power source for HEV and EVs. However, lithium ion batteries have low power density (50-200 W kg-1) and short cycle life while supercapacitors have low
d
energy density (30-90 W h kg-1) and high power density (< 3kW kg-1) [1, 2]. To overcome the
te
disadvantages of lithium ion batteries and supercapacitors, a hybrid supercapacitor was designed. This combines the mechanism of lithium ion batteries and the supercapacitors.
Ac ce p
Therefore, the hybrid supercapacitors provide high power density, high energy density, and long cycle life [3]. Various materials including Li4Ti5O12 [4], TiO2(B) [5], H2Ti3O7 [6] and H2Ti12O25 [7] have been used for the anode in hybrid supercapacitors, lithium ion batteries. The flat potential plateau of H2Ti12O25 shows the capacity of 236 mAhg-1 at 1.55 V [6]. H2Ti12O25 has a higher capacity than Li4Ti5O12 [3] and TiO2(B) [6]. Also, H2Ti12O25 was confirmed for cycling performance [7]. In order to improve the electrochemical performance of an H2Ti12O25 anode, we proposed Zn doping. In this study, a hybrid supercapacitor using H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6) as the anode and activated carbon as the cathode was investigated to see the effects of different Zn doping content on the structural properties and electrochemical performance of hybrid supercapacitors.
2
Page 2 of 33
Experiments H2Ti12-xZnxO25-x powders were prepared via a solid state method. First, ZnO powders were mixed with TiO2 at a predetermined molar ratio of 0-0.6 using a ball mill for 24 h. After
ip t
drying, the mixture was calcined at 1350ºC for 10 h in air. The precursor Na2Ti3-xZnxO7-x was prepared by a solid-state method mixing Na2CO3 (99.5%) and Ti1-xZnxO2-x in a molar ratio of 1:3. This mixture was calcined at 800oC for 20 h in air. As result, the H2Ti3-xZnxO7-x power
cr
was fabricated by Na+/H+ ion exchange reaction of Na2Ti3-xZnxO7-x in 1M HCl solution for 3 days at 60oC. Finally, the prepared H2Ti3-xZnxO7-x powder was heated at 300ºC for 5 h in air.
us
As a result, H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6) powders were fabricated. The structure of H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6) was analyzed using X-ray diffraction (XRD).
an
Furthermore, X-ray diffraction data were analyzed with Rietveld refinement program (RietanFP). Field emission scanning electron microscopy was used to observe the morphology of Zn doped H2Ti12O25. Also, the composition of H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6) was
M
analyzed via X-ray photoelectron spectroscopy (XPS). The hybrid supercapacitor was composed of the H2Ti12-xZnxO25-x anode and activated carbon (AC) cathode. An AC cathode
d
was fabricated by mixing AC (MSP-20, 90 wt%) with conductive carbon (5 wt%) and polytetrafluoroethylene (PTFE, 5 wt%). In order to fabricate the anode, H2Ti12-xZnxO25-x
te
powder, conductive carbon black binder (Super P), and polyvinylidene fluoride (PVDF) were mixed in an 83:7:10 weight ratio. N-Methyl pyrrolidinine (NMP) solvent was added to
Ac ce p
produce the slurry. This was casted on aluminum foil to a thickness of 125 µm and then dried at 100oC to remove the NMP solvent. The aluminum foil was pressed to a thickness of 70 80 µm. To carry out half-cell test, the electrodes were assembled with Li metal counter electrode and separator in an argon-gas-filled glove box. Before being impregnated with a 1.5M solution of LiBF4 solution in 1:1 ethylene carbonate (EC): dimethyl carbonate (DMC) as the electrolyte, in order to remove the moisture in the cell, the fabricated cell was dried in a vacuum oven for 48 h. The electrodes of full-cells were prepared by the above-stated method and full-cells were assembled with using Zn-doped H2Ti12O25 as anode and activated carbon as cathode in argon-gas-filled glove box. The width of the cathode, separator, and anode were 28 cm, 40 cm, and 30 cm, respectively and the heights of the cathode and anode were both 3 cm. The electrochemical measurement of the fabricated cell was carried out with various tests, such as initial capacitance and rate capability, using an Arbin BT 2042 battery test system. 3
Page 3 of 33
Results and discussion Figure 1 (a) shows the X-ray diffraction (XRD) pattern of the synthesized H2Ti12-xZnxO25-x
d
M
an
us
cr
ip t
(x=0, 0.15, 0.3, 0.45, 0.6) anode.
Figure 1 (a) X-ray diffraction pattern of the pristine and H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45,
te
0.6) powders. (b) By Zn content, shifted XRD pattern between 24 and 26 degrees.
Ac ce p
When x is less than 0.3, it can be seen that all samples were well-crystallized without impurity. This means that doped Zn content successfully entered into the Ti site. However, the impurity phases of TiO2 are observed above x=0.45 due to excessive Zn content [8, 9]. Li, Jiangang et al. we can expect that impurity phase affects a negative effect on the electrochemical performance [10]. Figure 1 (b) shows the magnified XRD patterns of the (110) planes for H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6). The diffraction peaks of the samples shifted toward higher 2θ with increasing Zn concentration. The lattice parameters of the monoclinic cell can be calculated using lattice parameter of monoclinic equation:
(1) (2) 4
Page 4 of 33
In order to obtain more precise structural parameters (lattice parameters), Rietveld refinements were performed for XRD patterns of H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6). Figure 2 (a-e) show the Rietveld refinement plots of H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45,
te
d
M
an
us
cr
ip t
0.6).
Ac ce p
Figure 2 Rietveld refinement plots of H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6) samples. This confirmed good agreement between observed XRD patterns via intensity scale of line (Yobs-Ycal). The value of the goodness of fit (GOF) can be calculated by the following formula [11]:
!"#
GOF
!$%#
(3)
where ‘Rwp’ is the residuals for the weighted pattern and ‘Rexp’ is the expected error. The calculated the GOF of the H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6) are 1.304, 1.194, 1.174, 1.294 and 1.195, respectively. These values indicate that the experimental pattern fitted well with the simulated pattern. The lattice parameter and volume of H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6) from the Rietveld refinement are depicted in Table 1. 5
Page 5 of 33
Table 1. Lattice parameter and volume of H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6) from the Rietveld refinement. b (Å)
c (Å)
β
Volume of unit (Å)
PristineH2Ti12O25
13.865
3.791
9.822
110.172
484.599
H2Ti11.85Zn0.15O24.85
13.618
3.720
9.801
111.123
H2Ti11.7Zn0.3O24.7
13.583
3.708
9.797
111.556
H2Ti11.55Zn0.45O24.55
13.510
3.699
9.741
111.724
H2Ti11.4Zn0.6O24.4
13.502
3.689
9.738
ip t
a (Å)
463.148
us
cr
458.922
111.819
452.219 450.292
an
Sample
As a result, the lattice parameters of H2Ti12-xZnxO25-x samples were slightly smaller than that
M
of pristine H2Ti12O25 [12]. This is advantageous to the lattice stability during the insertion/extraction of Li+ ion [13, 14]. The lattice contraction was generated by transition from Ti3+ (0.67 Å) to Ti4+ (0.605 Å) due to Zn2+ (0.74 Å) doping [15]. The X-ray
d
photoelectron spectroscopy (XPS) measurement confirmed the Zn doping into the H2Ti12O25
Ac ce p
(x=0.15, 0.3, 0.45, 0.6).
te
structure. Figure 3 (a) shows the XPS spectra of pristine H2Ti12O25 and H2Ti12-xZnxO25-x
Figure 3 (a) X-ray photoelectron spectroscopy (XPS) spectra with different Zn/Ti molar ratios and (b) Zn 2p peak of H2Ti11.7Zn0.3O24.7.
6
Page 6 of 33
We can show the peaks of Ti 2p and Zn 2p binding energy. The pristine H2Ti12O25 shows the Ti 2p doublet peaks at binding energy of 458.77-463.80 eV, with spin orbit splitting of 5.03 eV, which is in good agreement with the Ti4+ oxidation state [16]. Also, in the case of H2Ti12xZnxO25-x
(x=0, 0.15, 0.3, 0.45, 0.6), the peak of Zn 2p was observed at binding energy of
ip t
1021.17 eV [17]. The Zn 2p weak peak of H2Ti11.7Zn0.3O24.7 can be explained by low content of Zn doping in Figure 3 (b). When compared with pristine H2Ti12O25, the Ti 2p peak of
cr
H2Ti12-xZnxO25-x (x=0.15, 0.3, 0.45, 0.6) were moved toward lower binding energy. This can be explained by the smaller binding energy inducing expanded bond length due to the Zn
us
dopants. Thus, the mobility of Li+ will improve during charge/discharge by the relatively longer bond length of Ti-O [18]. Also, we can assume that this phenomenon is closely related with structure distortion of H2Ti12O25 by Zn doping [19]. This means that Zn was successfully
an
entered into the Ti site. When Ti is substituted for with Zn, the cation (Ti) vacancies were caused due to the different balance of cation [19]. The cation vacancies were generated
()
& ' *+, &
M
according to the following defect equations [20]: -(
∙∙ )
')×
(4)
d
The free oxygen vacancies (V O • • ) were created to balance the electric charge [20, 21]. The
te
free oxygen vacancies are proportional to Zn contents. Also, the electron hopping probability is linked closely to oxygen vacancies. Thus, the electron hopping probability can improve the
Ac ce p
electronic conductivity and is expected to increase with increasing oxygen vacancies [22]. As a result, Zn doping will improve the electrochemical performance of hybrid supercapacitors. The electrical conductivities of all samples were measured using the four-point probe measurement (RTS-8 Four-Point probe meter). The table 2 shows the measured electric conductivity.
Table 2. Electric conductivity of H2Ti12-xZnxO25-x samples. x in H2Ti12xZnxO25-x
σ/ S cm‒ 1
0
0.15
0.3
0.45
0.6
6.61 x 10-10
5.38 x 10-9
7.42 x 10-8
6.17 x 10-9
8.33 x 10-10
7
Page 7 of 33
The SEM images show pristine H2Ti12O25 and H2Ti12-xZnxO25-x (x=0.15, 0.3, 0.45, 0.6) as
us
cr
ip t
shown in Figure 4.
Figure 4 SEM images of the (a) H2Ti12O25, (b) H2Ti11.85Zn0.15O24.85, (c) H2Ti11.7Zn0.3O24.7, (d)
an
H2Ti11.55Zn0.45O24.55 and (e) H2Ti11.4Zn0.6O24.4 powders.
The initial charge-discharge curves of the H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6) and AC
Ac ce p
te
d
M
were shown in Fig. 5 (a) and (b), respectively.
Figure 5 Initial charge/discharge curves of half-cell using (a) H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6) anode and (b) AC cathode. Cyclic voltammetry curves of the half-cell using (c) Zndoped H2Ti12O25 anodes and (d) AC cathode. 8
Page 8 of 33
The pristine H2Ti12O25 shows a flat voltage plateau near 1.55 V (versus Li/Li+) with the discharge capacity of 231 mAh g-1 at current rate of 200 mA g-1, which is the lowest value than others. The discharge capacity increases with increasing Zn doping and reaches a maximum value of 270 mAh g-1 at H2Ti11.7Zn0.3O24.7. It can be inferred that the Li ion kinetics
ip t
is enhanced by the Zn doping. The H2Ti12-xZnxO25-x (x=0.45, 0.6) shows inferior to H2Ti11.7Zn0.3O24.7, which is consistent with XRD result. The initial charge-discharge curve of
cr
the activated carbon shows the linear behavior due to physical charge-storage mechanism. The activated carbon shows a capacity of 49 mAh g-1 at a current rate of 20 mA g-1.
us
To better understand individual electrodes, the CV curves of Li half-cell using AC cathode and Zn-doped H2Ti12O25 anodes were measured. CV curves of the pristine and Zn-doped H2Ti12O25 anodes show sharp peaks derived from redox reactions of Ti4+ ions. The potential
an
difference between anodic and cathodic peak decreased until the Zn doping level reaches x=0.3, as shown in Figure 5 (c). The anodic peaks (Ea) and cathodic peaks (Ec) of pristine
M
H2Ti12O25 and H2Ti12-xZnxO25-x (x=0.15, 0.3, 0.45, 0.6) anode, as seen in the CV curve (versus Li/Li+), are shown in Table 3.
Ec(V)
∆E(Ea -Ec) (V)
2.5
2.3
0.2
Ac ce p
Pristine H2Ti12O25
Ea(V)
te
Sample
d
Table 3. Polarization of the pristine and H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6).
H2Ti11.85Zn0.15O25
2.57
2.45
0.12
H2Ti11.7Zn0.3O25
2.6
2.5
0.1
H2Ti11.55Zn0.45O25
2.53
2.4
0.13
H2Ti11.4Zn0.6O25
2.54
2.38
0.16
The H2Ti12-xZnxO25-x (x=0.15, 0.3, 0.45, 0.6) anode shows higher oxidation peak current than pristine H2Ti12O25. The potential difference (∆E=Ea-Ec) between the anodic and cathodic peak is representative of the reaction kinetics [18] and is closely related to the polarization. The higher electronic conductivity can induce lower polarization, which improved the transfer rate of Li+ and the reversibility [8, 13]. Table 3 shows the polarization of pristine 9
Page 9 of 33
H2Ti12O25 and H2Ti12-xZnxO25-x (x=0.15, 0.3, 0.45, 0.6) anodes. The H2Ti12-xZnxO25-x (x=0.15, 0.3, 0.45, 0.6) anode have lower polarization and higher oxidation-reduction peak currents than a pristine H2Ti12O25 anode; the H2Ti11.7Zn0.3O24.7 anode shows the lowest polarization and largest CV area. This means that the Li+ diffusion and electronic conductivity were
ip t
improved with Zn addition. These results of the CV curves confirm that the capacitance of hybrid supercapacitors with H2Ti12-xZnxO25-x (x=0.15, 0.3, 0.45, 0.6) anode improved
cr
electrochemical performance. On the contrary, the CV of the activated carbon in the potential ranges 3.0-4.5 V at a scan rate of 2 mV s-1, as shown in Figure 5 (d). The measured curve
us
shows a typical rectangular shape, which is type of common electrical double layer capacitance (EDLC) [5]. As a result, both capacitive and redox reactions were generated in
an
the hybrid supercapacitor system.
Multiple experiments related to the electrode thickness of hybrid supercapacitors were conducted to optimize the cell balancing [7]. It is more associated with thickness of H2Ti12O25,
M
due to its slower Li+ insertion/extraction than BF4- ion adsorption/desorption at activated carbon. The discharge capacitance retentions according to the different electrode thicknesses
d
were shown in Table 4.
0.1 A
Ac ce p
HTO-AC
te
Table 4. Discharge capacitance retention for different electrode thickness
thickness
90 µm - 240 µm 90 µm - 180 µm
100 (63 Fg-1)
0.5 A
1A
2A
3A
93
85
68
48
-1
94
87
70
50
-1
100 (58 Fg )
70 µm - 240 µm
100 (60 Fg )
95
88
71
52
70 µm - 180 µm
100 (53 Fg-1)
93
85
69
49
50 µm - 240 µm
100 (56 Fg-1)
94
86
70
51
50 µm - 180 µm
100 (47 Fg-1)
89
81
64
43
40 µm - 240 µm
100 (50 Fg-1)
92
85
68
48
93
86
69
47
40 µm - 180 µm
-1
100 (43 Fg )
10
Page 10 of 33
At 0.1 A, the initial capacitances were increased proportionally with electrode thickness because both electrodes have sufficient time for electrochemical reactions. However, with increasing the current rate, the capacitance retention is decreased. The reason for this is sluggish kinetics of Li+ at H2Ti12O25. Therefore, H2Ti12O25 anode can’t use its entire thickness,
ip t
except for the upper part. On the contrary, with increasing the thickness of cathode, the capacitance retention increased regardless of current density. Based on these results, we can
cr
select the thickness composition of 70-240 µm, indicating capacitance retention of 52 % at 3.0 A as an optimum thickness.
us
The cyclic voltammetry (CV) curves of the hybrid supercapacitors with H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6) anode are shown in Figure 6(a) at a scan rate of 10 mV s-1 between
Ac ce p
te
d
M
an
0 V and 2.8 V (working voltage of activated carbon 4.3 V [23] and HTO 1.5 V [6]).
Figure 6(a) Cyclic voltammetry, (b) initial charge-discharge and (c) electrochemical impedance spectra curves of the hybrid supercapacitors using the pristine and H2Ti12-xZnxO25x
(x=0, 0.15, 0.3, 0.45, 0.6. (d) Rate capabilities and (e) cycling performance of the hybrid
supercapacitors using the pristine and H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6) from 0.1 to 3 Ag-1. 11
Page 11 of 33
It can be inferred that H2Ti11.7Zn0.3O24.7 can have a positive effect on hybrid supercapacitors. The reason for this is accelerated kinetics of Li+ by Zn doping in H2Ti12O25. Based on above results, the capacitive and redox reactions coexist in the hybrid supercapacitors.
ip t
Figure 6(b) shows the initial charge/discharge curves for the H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6) anode between 0 V and 2.8 V. The discharge specific capacitances of the hybrid supercapacitors using the H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6) anode were obtained
1
∆3×4
5 ∆6 ∆3×4
us
0
cr
using the following relationship [24]:
(5)
where 'C' is the capacitance (Fg-1), '∆V' is the voltage change, 'm' is the mass of the active
an
materials in both electrodes, 'q' is the total charge, 'i' is the current, and 't' is time. These results indicate that the capacitances of the H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6) anodes
M
are 57.3, 65.05, 70.7, 60.1, and 58.3 Fg-1, respectively, at current density of 0.5 Ag-1. As a result, the capacitance of hybrid supercapacitors with H2Ti11.7Zn0.3O24.7 anode had the highest value. In the case of the H2Ti12-xZnxO25-x (x=0.15, 0.3) anode, the capacitances were higher
d
than pristine H2Ti12O25. Also, the voltage drop (IR drop) was observed in the initial
te
charge/discharge curves of the H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6) anode during the discharge process. This IR drop shows the effect of internal resistance in the hybrid
Ac ce p
supercapacitor. In the initial charge/discharge curves, the discharge region involves an IR drop owing to the internal resistance. Also, a faradaic discharge process is related to the lithium ion insertion/extraction as well as a non-faradaic discharge process is associated with the adsorption/desorption of ions. The voltage change was caused by ion separation at the interface of the electrode and the electrolyte [25]. The IR drop can be calculated by the following equation [26]:
7
389:;<$ =3>?@89:;<$
(6)
'Vcharge' is the voltage of the cell at the end charge, 'Vdischarge' is the voltage of the cell at the starting discharge, and 'I' is the absolute value of charge and discharge current. The calculated internal resistance of the H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6) anode are 0.19, 0.161, 0.142, 0.177, and 0.183 Ω, respectively. The IR drop is significantly correlated with the 12
Page 12 of 33
charge transfer resistance (Rct) of the hybrid supercapacitor [27]. These results indicated that the enhanced Li+ diffusion and electronic conductivity by oxygen vacancies may contribute to improve of capacitance in the hybrid supercapacitor.
ip t
To investigate the effect of the H2Ti12-xZnxO25-x anode on the electrolyte/H2Ti12-xZnxO25-x interface resistances, electrochemical impedance spectra (EIS) curves were performed in the frequency range of 10-1 to 10-6 Hz at a voltage of 2.8V, as shown in Figure 6(c). In the EIS
cr
results, a semicircle and straight line in the high and low frequency region were observed and showed the charge transfer resistance (Rct) and internal resistance (Ri) related to Li+
us
interfacial transfer at high frequency, respectively. The Rct of the H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6) anodes were approximately 0.06, 0.043, 0.034, 0.053, and 0.058 Ω,
an
respectively. As shown in Figure 6(c), the Rct values of H2Ti12-xZnxO25-x (x=0.15, 0.3) decreased. In addition, electrochemical polarization decreased with decreasing the Rct value [15]. However, Rct values of H2Ti12-xZnxO25-x (x=0.45, 0.6) increased. The straight line at the
M
low frequency is caused by Warburg impedance, which corresponds to Li+ diffusion in the H2Ti12O25 anode [28, 29]. In addition, we can calculate the diffusion coefficient (DLi) of Li+
d
by the following equation [30, 31]:
te
AB
(D
!(
E FG
)
(7)
Ac ce p
where 'R' is the gas constant, 'T'' is the absolute temperature, 'A' is the surface area of the anode, 'n' is the number of electrons, 'F' is the Faraday’s constant, 'C' is the concentration of Li+, and 'σ' is the Warburg impedance coefficient. The calculated diffusion coefficients of Li+ of the samples are 2.937 x 10-7, 5.637 x 10-7, 7.144 x 10-7, 6.307 x 10-7, and 5.259 x 10-7 cm2 s-1 for H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6) anodes, respectively. Based on these results, the H2Ti12-xZnxO25-x (x=0.15, 0.3) anode had a lower Rct value and a faster the Li+ diffusion rate than that of pristine H2Ti12O25. Therefore, we can expect that zinc doping of the anode will improve the electrochemical performance by increasing their electrical conductivity [9, 32]. Figure 6(d) shows the rate capability of the pristine H2Ti12O25 and H2Ti12-xZnxO25-x (x=0.15, 0.3, 0.45, 0.6) anodes at different charge-discharge rates of 0.1, 0.5, 1, 2, and 3 Ag-1. Regarding the rate capability, the capacitance retention of pristine H2Ti12O25 anode appears 52% 13
Page 13 of 33
at 3 Ag-1. On the other hand, the H2Ti12-xZnxO25-x (x=0.15, 0.3, 0.45, 0.6) anodes had capacitance retention of approximately 61.5%, 65%, 59% and 54%, respectively at the 3 Ag-1. The rate capability of the zinc doped H2Ti12O25 anode was improved and the properties of the samples were enhanced with the increasing capacitance retention. Figure 6(d) shows H2Ti12(x=0.15, 0.3, 0.45, 0.6) anode have a much better rate capability than pristine
ip t
xZnxO25-x
H2Ti12O25. As a result, enhanced electronic conductivity of H2Ti12-xZnxO25-x (x=0.15, 0.3,
cr
0.45, 0.6) anodes because of oxygen vacancies leads to superior capacitance retention [32]. This shows a similar trend as shown in Figure 8.
xZnxO25-x
us
Figure 6(e) shows the cycle performance of the hybrid supercapacitors with the H2Ti12(x=0, 0.15, 0.3, 0.45, 0.6) anode after 1000 cycles compared with 1 cycle at 3.0 Ag-
1
x
an
. It was found that the cycling behavior of the hybrid supercapacitor with the H2Ti12-xZnxO25(x=0.15, 0.3) anodes is better than that of pristine H2Ti12O25 during 1000 cycles. Also, the
hybrid supercapacitors with H2Ti12-xZnxO25-x (x=0.15, 0.3) anodes have specific capacitance
M
of 41.03 and 46.67 Fg-1 for the first discharge process, respectively, and maintains at 36.68 and 43.1 Fg-1 at 1000 cycles, which shows 89 and 92%, respectively, after 1000 cycles compared with 1 cycle. On the other hand, the hybrid supercapacitors with H2Ti12-xZnxO25-x
d
(x=0.45, 0.6) anodes maintain capacitances of 31.4 and 26.23 Fg-1, respectively, at 1000
te
cycles, and are 83 and 77% after 1000 cycles, respectively, compared to 1 cycle. Excessive doping of zinc (x=0.45, 0.6) induced large distortion of the lattice and increased the electrical
Ac ce p
resistance. Thus, capacitance and cycle performance is decreased [33]. As shown in Figure 10, gradually enhanced cycle performances were confirmed with Zn doping content of x=0.15, 0.3 by increasing electrical conductivity. Generally, the shrinking of the lattice or collapse of structure was caused by repetition of lithium insertion/extraction; the unchangeable radius of Zn would prevent the shrinking or collapse of structure [34, 35] and improve the structure stability. Thus, cycle performance was improved by enhanced electric conductivity and structure stability.
Figure 7 shows the Ragone plot of the hybrid supercapacitors with H2Ti11.7Zn0.3O24.7 anode with various hybrid supercapacitors.
14
Page 14 of 33
ip t cr us
an
Figure 7 Ragone plots of hybrid supercapacitors using H2Ti12-xZnxO25-x.
The energy and power densities were observed by discharging a charged device at various
I
M
current densities and were calculated using the following relationship [26]:
∆J × 4
(8)
J
I×L
(9)
d
K
te
∆J
MN:% OMN?P
(10)
Ac ce p
where ′J4 R ′ is the potential at the starting discharge, ′J4 ′ is the potential at the end discharge, ‘I’ is the charge and discharge currents, ‘m’ is the mass of active materials including the anode and cathode electrodes, and ‘t’ is the discharge time in the hybrid supercapacitor. The maximum energy and power densities of hybrid supercapacitors with H2Ti11.7Zn0.3O24.7 anode and activated carbon cathode material were 44.4 Whkg-1/182 Wkg-1 and 9.1Whkg-1/5676.2 Wkg-1 at 0.1 and 3 Ag-1 rates, respectively. We compared the power and energy densities of H2Ti11.7Zn0.3O24.7/AC with other hybrid supercapacitors composed of TNW/CNT [5], C-LTP/AC [26], C-LTO/AC [36] and C-LTP/AC [37]. As a result, the energy and power densities of the hybrid supercapacitors using H2Ti11.7Zn0.3O24.7 anode and activated carbon cathode are superior to other systems. The extraordinary electrochemical performance can be attributed to the faster kinetic of Li ion comparable to that of BF4- ion by inherent tunnel structure of H2Ti12O25 and improved conductivity by Zn doping. Consequently, this 15
Page 15 of 33
new hybrid system is expected to be sufficient for high energy storage devices in various applications.
ip t
Conclusion
In this study, hybrid supercapacitors using H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6) anode
cr
were successfully fabricated. We investigated the effect on the electrochemical performance of Zn doping. The Zn plays roles of superior reversible capacitance and lower polarization
us
and caused cation vacancies in the H2Ti12O25. In addition, the Zn doping improved structure stability to prevent lattice shrinking and structure collapse. The doped Zn content
an
significantly improved both initial specific capacitance and the capacitance retention. Also, the hybrid supercapacitor using an H2Ti11.7Zn0.3O24.7 anode showed better power density as well as energy density than the other system of lithium ion secondary batteries and
te
Acknowledgments
d
performance of hybrid supercapacitors.
M
supercapacitors. The H2Ti11.7Zn0.3O24.7 anode is important to improving the electrochemical
Ac ce p
This work was supported by the Korea Institute of Energy Technology Evaluation and Planning (KETEP) and the Ministry of Trade, Industry & Energy (MOTIE) of the Republic of Korea (No. 20162220100050, No. 20161120100350), in part by The Leading Human Resource Training Program of Regional Neo industry through the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT and future Planning (NRF-2016H1D5A1910564), and in part by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (2016R1D1A3B03931927).
16
Page 16 of 33
Reference [1] M. Yang, Y. Zhong, J. Ren, X. Zhou, J. Wei, Z. Zhou, Electrochemical Capacitors: Fabrication of High-Power Li-Ion Hybrid Supercapacitors by Enhancing the Exterior Surface
ip t
Charge Storage, Adv. Energy Mater. 5 (2015) 1500550 [2] M. Yang, Y. Zhong, J. Bao, X. Zhou, J. Wei, Z. Zhou, Achieving battery-level energy density by constructing aqueous carbonaceous supercapacitors with hierarchical porous N-
cr
rich carbon materials, J. Mater. Chem. A 3 (2015) 11387
[3] V. H. Nguyen, H. B. Gu, Synthesis and Characterization of Silver Vanadium Oxide as a
us
Cathode for Lithium Ion Batteries, Trans. Electr. Electron. Mater 17 (2016) 139
[4] S. H. Lee, S. G. Lee, J. R. Yoon, H. K. Kim, Novel performance of ultrathin AlPO4
an
coated H2Ti12O25 Exceeding Li4Ti5O12 in cylindrical hybrid supercapacitor, J. Power Sources 273 (2015) 839
[5] Q. Wang, Z. Wen, J. Li, A Hybrid Supercapacitor Fabricated with a Carbon Nanotube
M
Cathode and a TiO2–B Nanowire Anode, Adv. Funct. Mater. 16 (2006) 2141 [6] J. Akimoto, K. Chiba, N. Kijima, H. Hayakawa, S. Hayashi, Y. Gotoh, Y. Idemotob, SoftChemical Synthesis and Electrochemical Property of H2Ti12O25 as a Negative Electrode
d
Material for Rechargeable Lithium-Ion Batteries, J. Electrochem. Soc. 158 (2011) A546
te
[7] S. H. Lee, H. K. Kim, J. H. Lee, S. G. Lee, Y. H. Lee, Fabrication and electrochemical properties of cylindrical hybrid supercapacitor using H2Ti12O25 as anode material, Materials
Ac ce p
Letters 143 (2015) 101
[8] B. Zhang, H. Du, B. Li, F. Kang, Structure and Electrochemical Properties of Zn-Doped Li4Ti5O12 as Anode Materials in Li-Ion Battery, Electrochem. Solid. St. 13 (2010) A36 [9] G. Lu, S. Qiu, J. Liu, X. Wang, C. He, Y. J. Bai, Enhanced Electrochemical Performance of Zn-Doped Fe3O4 with Carbon Coating, Electrochim. Acta 117 (2014) 230 [10] J. Li, X. He, R. Zhao, C. Wan, C. Jiang, D. Xia, S. Zhang, Stannum doping of layered LiNi3/8Co2/8Mn3/8O2 cathode materials with high rate capability for Li-ion batteries, J. Power Sources 158 (2006) 524 [11] R. A. Young, D. B. Wiles, Profile shape functions in Rietveld refinements, J. Appl. Cryst. 15 (1982) 430 17
Page 17 of 33
[12] Akimoto, Junji. et al. U.S. Patent No. 8,309,253. (2012) [13] D. Wu, Y. Cheng, Enhanced high-rate performance of sub-micro Li4Ti4. 95Zn0. 05O12 as anode material for lithium-ion batteries, Ionics 19 (2013) 395
ip t
[14] C. Deng, S. Zhang, S. Y. Yang, B. L. Fu, L. Ma, Synthesis and characterization of Li2Fe0.97M0.03SiO4 (M=Zn2+, Cu2+, Ni2+) cathode materials for lithium ion batteries, J. Power
cr
Sources 196 (2011) 386
[15] T. F. Yi, H. Liu, Y. R. Zhu, L. J. Jiang, Y. Xie, R. S. Zhu, Improving the high rate
us
performance of Li4Ti5O12 through divalent zinc substitution, J. Power Sources 215 (2012) 258
an
[16] Z. Ali, S. N. Cha, J. I. Sohn, I. Shakir, C. Yan, J. M. Kim, D. J. Kang, Design and evaluation of novel Zn doped mesoporous TiO2 based anode material for advanced lithium ion batteries, J. Mater. Chem. 22 (2012) 17625
d
M
[17] M. E. Fragala, I. Cacciotti, Y. Aleeva, R. L. Nigro, A. Bianco, G. Malandrino, C. Spinella, G. Pezzottid, G. Gusmano, Core–shell Zn-doped TiO2–ZnO nanofibers fabricated via a combination of electrospinning and metal–organic chemical vapour deposition, Cryst Eng Comm. 12 (2010) 3858
te
[18] G. Yang, H. Ni, H. Liu, P. Gao, H. Ji, S. Roy, J. Pinto, X. Jiang, The doping effect on the crystal structure and electrochemical properties of LiMnxM1−x PO4 (M= Mg, V, Fe, Co, Gd), J. Power Sources 196 (2011) 4747
Ac ce p
[19] Z. H. Wang, L. X. Yuan, M. Wu, D. Sun, Y. H. Huang, Effects of Na+ and Cl− co-doping on electrochemical performance in LiFePO4/C, Electrochim. Acta 56 (2011) 8477 [20] W. Y. Wang, D. F. Zhang, T. Xu, X. F. Li, T. Zhou, X. L. Chen, Nonlinear electrical behavior and dielectric properties of (Ca, Ta)-doped TiO ceramics, J. Alloy. Comp. 335 (2002) 210 [21] H. Beltrán, M. Prades, N. Masó, E. Cordoncillo, A. R. West, Voltage-Dependent LowField Bulk Resistivity in BaTiO3:Zn Ceramics, J. Am. Ceram. Soc. 93 (2010) 500 [22] J. M. Wu, C. J. Chen, Dielectric properties of (Ba, Nb) doped TiO2 ceramics: migration mechanism and roles of (Ba, Nb), J. Mater. Sci. 23 (1988) 4157 [23] S. Dsoke, B. Fuchs, E. Gucciardi, M. W. Mehrens, The importance of the electrode mass ratio in a Li-ion capacitor based on activated carbon and Li4Ti5O12, J. Power Sources 282 18
Page 18 of 33
(2015) 385 [24] V. Aravindan, Y. L. Cheah, W. F. Mak, G. Wee, B. V. R. Chowdari, S. Madhavi, Fabrication of High Energy‐Density Hybrid Supercapacitors Using Electrospun V2O5
ip t
Nanofibers with a Self‐Supported Carbon Nanotube Network, Chempluschem 77 (2012) 570 [25] N. Yu, L. Gao, Electrodeposited PbO2 thin film on Ti electrode for application in hybrid
cr
supercapacitor, Electrochem. commun. 11 (2009) 220
[26] J. Y. Luo, Y. Y. Xia, Electrochemical profile of an asymmetric supercapacitor using
us
carbon-coated LiTi2(PO4)3 and active carbon electrodes, J. Power Sources 186 (2009) 224 [27] A. Chu, P. Braatz, Comparison of commercial supercapacitors and high-power lithium-
an
ion batteries for power-assist applications in hybrid electric vehicles: I. Initial characterization, J. Power Sources 112 (2012) 236
M
[28] S. S. Zhang, K. Xu, T. R. Jow, T.R. EIS study on the formation of solid electrolyte interface in Li-ion battery, Electrochim. Acta. 51 (2006) 1636
d
[29] S. Zhang, M. S. Ding, K. Xu, J. Allen, T. R. Jow, Understanding solid electrolyte
te
interface film formation on graphite electrodes, Electrochem Solid. St. 4 (2001) A206 [30] J. R. Yoon, E. Baek, H. K. Kim, M. Pecht, S. H. Lee, Critical dual roles of carbon
Ac ce p
coating in H2Ti12O25 for cylindrical hybrid supercapacitors, Carbon 101 (2016) 9 [31] T. F. Yi, Y. Xie, J. Shu, Z. Wang, C. B. Yue, R. S. Zhu, H. B. Qiao, Structure and Electrochemical Performance of Niobium-Substituted Spinel Lithium Titanium Oxide Synthesized by Solid-State Method, Electrochem. Soc. 158 (2011) A266 [32] J. Y. Luo, L. J. Chen, Y. J. Zhao, P. He, Y. Y. Xia, The effect of oxygen vacancies on the structure and electrochemistry of LiTi2(PO4)3 for lithium-ion batteries: A combined experimental and theoretical study, J. Power Sources 194 (2009) 1075 [33] W. Wang, J. Zhang, Z. Jia, C. Dai, Y. Hu, J. Zhou, Q. Xiao, Enhancement of the cycling performance of Li3V2(PO4)3/C by stabilizing the crystal structure through Zn2+ doping, Phys. Chem. Chem. Phys. 16 (2014) 13858 [34] Y. Hu, J. Yao, Z. Zhao, M. Zhu, Y. Li, H. Jin, H. Zhao, J. Wang, ZnO-doped LiFePO4 19
Page 19 of 33
cathode material for lithium-ion battery fabricated by hydrothermal method, Materials Chemistry and Physics. 141 (2013) 835 [35] Y. Yang, W. Xu, R. Guo, L. Liu, S. Wang, D. Xie, Y. Wan, J. Power Sources 269 (2014) 15
ip t
[36] H. G. Jung, N. Venugopal, B. Scrosati, Y. K. Sun, Synthesis and electrochemical properties of Zn-doped, carbon coated lithium vanadium phosphate cathode materials for
cr
lithium-ion batteries, J. Power Sources 221 (2013) 266
[37] V. Aravindan, W. Chuiling, M. V. Reddy, G. V. S. Rao, B. V. R. Chowdari, S. Madhavi,
us
Carbon coated nano-LiTi2(PO4)3 electrodes for non-aqueous hybrid supercapacitors, Phys.
an
Chem. Chem. Phys. 14 (2012) 5808
Figure caption,
M
Figure 1 (a) X-ray diffraction pattern of the pristine and H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6) powders. (b) By Zn content, shifted XRD pattern between 24 and 26 degrees.
d
Figure 2 Rietveld refinement plots of H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6) samples.
te
Figure 3 (a) X-ray photoelectron spectroscopy (XPS) spectra with different Zn/Ti molar ratios
Ac ce p
and (b) Zn 2p peak of H2Ti11.7Zn0.3O24.7.
Figure 4 SEM images of the (a) H2Ti12O25, (b) H2Ti11.85Zn0.15O24.85, (c) H2Ti11.7Zn0.3O24.7, (d) H2Ti11.55Zn0.45O24.55 and (e) H2Ti11.4Zn0.6O24.4 powders. Figure 5 Initial charge/discharge curves of half-cell using (a) H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6) anode and (b) AC cathode. Cyclic voltammetry curves of the half-cell using (c) Zndoped H2Ti12O25 anodes and (d) AC cathode. Figure 6(a) Cyclic voltammetry, (b) initial charge-discharge and (c) electrochemical impedance spectra curves of the hybrid supercapacitors using the pristine and H2Ti12-xZnxO25x
(x=0, 0.15, 0.3, 0.45, 0.6. (d) Rate capabilities and (e) cycling performance of the hybrid
supercapacitors using the pristine and H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6) from 0.1 to 3 Ag-1. 20
Page 20 of 33
Figure 7 Ragone plots of hybrid supercapacitors using H2Ti12-xZnxO25-x. Table caption
ip t
Table 1. Lattice parameter and volume of H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6) from the Rietveld refinement. Table 2. Electric conductivity of H2Ti12-xZnxO25-x samples.
cr
Table 3. Polarization of the pristine and H2Ti12-xZnxO25-x (x=0, 0.15, 0.3, 0.45, 0.6).
Ac ce p
te
d
M
an
us
Table 4. Discharge capacitance retention for different electrode thickness
21
Page 21 of 33
Ac ce p
te
d
M
an
us
cr
ip t
Figure 1
22
Page 22 of 33
Ac ce p
te
d
M
an
us
cr
ip t
Figure 2
23
Page 23 of 33
Ac ce p
te
d
M
an
us
cr
ip t
Figure 3
24
Page 24 of 33
Ac ce p
te
d
M
an
us
cr
ip t
Figure 4
25
Page 25 of 33
Ac ce p
te
d
M
an
us
cr
ip t
Figure 5
26
Page 26 of 33
Ac ce p
te
d
M
an
us
cr
ip t
Figure 6
27
Page 27 of 33
Ac ce p
te
d
M
an
us
cr
ip t
Figure 7
28
Page 28 of 33
Table.1
b (Å)
c (Å)
β
Volume of unit (Å)
PristineH2Ti12O25
13.865
3.791
9.822
110.172
484.599
H2Ti11.85Zn0.15O24.85
13.618
3.720
9.801
111.123
H2Ti11.7Zn0.3O24.7
13.583
3.708
9.797
111.556
H2Ti11.55Zn0.45O24.55
13.510
3.699
9.741
111.724
H2Ti11.4Zn0.6O24.4
13.502
3.689
9.738
ip t
a (Å)
463.148
us
cr
458.922
111.819
452.219 450.292
Ac ce p
te
d
M
an
Sample
29
Page 29 of 33
Table. 2
xZnxO25-x
0.15
0.3
0.45
0.6
6.61 x 10-10
5.38 x 10-9
7.42 x 10-8
6.17 x 10-9
8.33 x 10-10
Ac ce p
te
d
M
an
us
cr
σ/ S cm‒ 1
0
ip t
x in H2Ti12-
30
Page 30 of 33
Table. 3
Ec(V)
∆E(Ea -Ec) (V)
Pristine H2Ti12O25
2.5
2.3
0.2
H2Ti11.85Zn0.15O25
2.57
2.45
H2Ti11.7Zn0.3O25
2.6
2.5
H2Ti11.55Zn0.45O25
2.53
2.4
H2Ti11.4Zn0.6O25
2.54
2.38
ip t
Ea(V)
0.12
cr
0.1
0.13
0.16
Ac ce p
te
d
M
an
us
Sample
31
Page 31 of 33
Table. 4
90 µm - 180 µm 70 µm - 240 µm
100 (63 Fg-1)
0.5 A
1A
2A
3A 48
93
85
68
-1
94
87
70
-1
95
88
71
-1
69
100 (58 Fg ) 100 (60 Fg ) 100 (53 Fg )
93
85
50 µm - 240 µm
100 (56 Fg-1)
94
86
50 µm - 180 µm
100 (47 Fg-1)
89
81
40 µm - 240 µm
100 (50 Fg-1)
92
85
40 µm - 180 µm
100 (43 Fg-1)
93
86
50 52 49
70
51
64
43
68
48
69
47
Ac ce p
te
d
M
an
70 µm - 180 µm
ip t
90 µm - 240 µm
0.1 A
cr
thickness
us
HTO-AC
32
Page 32 of 33
Highlights - Hybrid supercapacitors were fabricated with H2Ti12-xZnxO25-x (x=0~0.6) anode. - Capacitance retention of H2Ti11.7Zn0.3O24.7 was 92% after 1000 cycles. - Power densities of H2Ti11.7Zn0.3O24.7 were 42.4-8.1 Wh kg-1 from 0.1 to 3 Ag-1.
Ac ce
pt
ed
M
an
us
cr
ip t
- Energy densities of H2Ti11.7Zn0.3O24.7 were 182-5676.2 W kg-1 from 0.1 to 3 Ag-1.
Page 33 of 33