- Email: [email protected]
Full-gradient structured LiNi0.8Co0.1Mn0.1O2 cathode material with improved rate and cycle performance for lithium ion batteries
Accepted Manuscript Full-gradient structured LiNi0.8Co0.1Mn0.1O2 cathode material with improved rate and cycle performance for lithium ion batteries Yunpeng Jiang, Zhihao Liu, Yongzheng Zhang, Huili Hu, Xiangguo Teng, Dianlong Wang, Peng Gao, Yongming Zhu PII:
S0013-4686(19)30735-2
DOI:
https://doi.org/10.1016/j.electacta.2019.04.058
Reference:
EA 34006
To appear in:
Electrochimica Acta
Received Date: 15 January 2019 Revised Date:
9 April 2019
Accepted Date: 9 April 2019
Please cite this article as: Y. Jiang, Z. Liu, Y. Zhang, H. Hu, X. Teng, D. Wang, P. Gao, Y. Zhu, Fullgradient structured LiNi0.8Co0.1Mn0.1O2 cathode material with improved rate and cycle performance for lithium ion batteries, Electrochimica Acta (2019), doi: https://doi.org/10.1016/j.electacta.2019.04.058. 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 Full-gradient Structured LiNi0.8Co0.1Mn0.1O2 Cathode Material with Improved Rate and Cycle Performance for Lithium ion batteries Yunpeng Jiang,a,b , Zhihao Liu,a Yongzheng Zhang,a Huili Hua, Xiangguo Teng,a
RI PT
Dianlong Wang,b Peng Gaoa,* and Yongming Zhua,* a. Harbin Institute of Technology at Weihai, Department of Applied Chemistry, Weihai 264209, China. b. Harbin Institute of Technology, School of Chemical
SC
Engineering and Technology, Harbin 150001, China.
* Corresponding authors. Tel.: +86 631 5687232 (P. Gao), +86 631 5687901 (Y. Zhu).
M AN U
E-mail addresses: [email protected] (P. Gao), [email protected] (Y. Zhu).
Abstract: LiNi0.8Co0.1Mn0.1O2 cathode material with full-gradient structure is prepared by a novel co-precipitation method. Scanning electron microscope images show that the material presents regular spherical shape. X-ray diffraction pattern shows that the
TE D
material has good layered structure. Energy dispersive X-ray spectroscopy analysis for the cross section of secondary particles shows that the relative molar content of Ni
EP
gradually decreases from 84% to 76% along the center to the edge of hemisphere, while the Mn content increases gradually. Moreover, the Co content also presents a
AC C
slow gradient change. Electrochemical tests at room temperature show that the retention rates of discharge capacity for full-gradient material after 100 cycles at current density of 1 C and 5 C are 98.8% and 93.7%, respectively, which are obviously improved compared with intrinsic material. The gradient material also shows improved midpoint voltage and specific energy. Moreover, full-gradient material exhibits good cyclic stability at high temperature, and the capacity retention rate of 100 cycles is as high as 90% at 5C rate, which is obviously higher than 71.8% 1
ACCEPTED MANUSCRIPT of intrinsic material. Impedance spectroscopy studies show that the solid-state diffusion coefficient of lithium ion in gradient material has been significantly improved, thus showing good electrochemical properties. The full-gradient material shows
RI PT
excellent application prospects in the field of power batteries. Keywords: Li-ion battery, nickel-rich cathode material, LiNi0.8Co0.1Mn0.1O2, full-gradient structure, a novel co-precipitation method
SC
1. Introduction
M AN U
At present, with the development of new energy automobile technology, the energy density of lithium-ion power batteries has been put forward increasingly requirements. The actual specific capacity of commercial cathode materials is generally lower than 160 mAh/g, which has become a key factor restricting the
TE D
capacity of lithium-ion batteries. In recent years, nickel-rich ternary cathode materials have received increasing attention due to their high specific capacity and low cost [1-3]. Among them, LiNi1−x−yCoxAlyO2 and LiNi1−x−yCoxMnyO2 (x + y < 0.2) are the
EP
most promising materials [4]. However, such relatively high nickel content leads to
AC C
poor stability [5-7]. Therefore, it is necessary to modify the nickel-rich ternary materials to improve its electrochemical performance. Nevertheless, most of modified methods are limited to elemental doping (such as Mg, Al, F, Cr, Y, et al.) [8-12] and surface coating (such as Al2O3, ZrO2, AlPO4 et al.) [13-19]. In addition, Yang et al. [20] and Li et al. [21] combined doping and coating modification to improve performance, such as dual-modification strategy and lithium residues-assistant modification. These methods are further processed on the basis of intrinsic material, 2
ACCEPTED MANUSCRIPT and the structural deterioration during the charge-discharge cycles cannot be fundamentally solved [22-24]. Recent studies have shown that core-shell structure is an effective way to improve the shortcomings of high nickel, and the low nickel shell
RI PT
coated with high nickel core can achieve good modification effect, Sun et al. [25] synthesized a spherical core-shell structure with high capacity and good thermal stability, which is about the microscale spherical core-shell structure, that is, the
SC
Li[Ni0.8Co0.1Mn0.1]O2 as the core and the Li[Ni0.5Mn0.5]O2 as the shell. In addition,
M AN U
Sun and Amine et al. [26,27]. further explored the effect of core-shell structure with different thickness on electrochemical performance, which makes the thickness of core-shell structure further optimized and promotes the research of core-shell structure.
However,
the
core-shell
structure
has
the
disadvantage
of
TE D
concentration-varying, and the structure is still easy to be destroyed. Therefore, learning from the advantages of core-shell structure, and a full-gradient material was designed, which internal structure changes very smoothly and controllably. In the
EP
research of full-gradient structure, the optimization of internal structure shows better [28] reported a nickel-rich material of
AC C
modified prospects. Sun et al.
LiNi0.75Co0.1Mn0.15O2, where the atomic percentage of Co remained constant, whereas the concentration of Ni decreased and Mn increased continuously from the center towards the outer layer of the particle. It can improve the thermal stability and long life. Li et al. [29] employed a ternary cathode material of LiNi0.6Co0.2Mn0.2O2 with concentration gradient structure, which provides a nickel-rich core to deliver high capacity and a manganese-rich outer layer to provide enhanced stability and cycle life. 3
ACCEPTED MANUSCRIPT However, these previous studies basically drip the salt solution (A) with high manganese content into the salt solution (B) with high nickel content, and pump the solution (B) into the reaction kettle to synthesize the gradient material with linear
RI PT
change. It is not convenient to precisely control the synthesis of gradient structures. Moreover, because of this feeding method, only the gradient distribution of Ni and Mn can be realized, resulting in uniform distribution of Co in the synthesized material.
SC
In this study, we design a novel and convenient method to prepare
M AN U
LiNi0.8Co0.1Mn0.1O2 with full-gradient structure (LGNCMO) more accurately, that only separates the original ternary salt solution into two strands for feeding, which can accurately control the gradient synthesis. The schematic illustration of the synthesis of LGNCMO is shown in Scheme 1. Firstly, the full-gradient precursor is prepared by a
TE D
novel co-precipitation method, and then lithiation is carried out in combination with high temperature solid-phase method to synthesize the final product with full-gradient structure. The designed full-gradient material is a nickel-rich as the inner core
EP
material, and gradually transfers to the relatively low nickel and manganese-rich outer
AC C
surface, moreover, the Co content also presents a slow gradient change. The full-gradient material has improved high rate and high temperature cycle performance. In this experiment, the LiNi0.8Co0.1Mn0.1O2 intrinsic material (LNCMO) with uniform distribution of internal elements is prepared for comparison. 2. Experimental 2.1 Material synthesis Preparation of LGNCMO. The precursor of full-gradient material (GNCMO) 4
ACCEPTED MANUSCRIPT was prepared by a novel and convenient co-precipitation method, and then combined with high temperature solid-phase method for lithiation to prepare LGNCMO with full-gradient structure. The schematic diagram of the preparation process of GNCMO
RI PT
is shown in Fig. 1. In this experiment, the sulphate of nickel, cobalt and manganese was weighed according to the stoichiometric ratio of 8:1:1. NiSO4·6H2O and CoSO4·7H2O were dissolved in deionized water to prepare a salt solution of A (2
SC
mol/L), and MnSO4·H2O was dissolved in deionized water to prepare a salt solution
M AN U
of B (2 mol/L). At the same time, an aqueous solution of sodium hydroxide (4 mol/L) and an aqueous solution of aqueous ammonia (6.68 mol/L) were prepared as precipitating agent and complexing agent, respectively. The A solution, sodium hydroxide solution, and aqueous ammonia solution were simultaneously pumped into
TE D
the reaction vessel with an inert gas introduced and the B solution was pumped into the A solution. pH=11.50 was accurately controlled by adjusting the flow rate of sodium hydroxide solution, keeping the stirring speed at 800 rpm, and the whole
EP
reaction time was 36 h. In this process, the flow rate of B solution (~2.7 mL/h) was
AC C
controlled to drip out in about 35 hours, the flow rates of A solution and aqueous ammonia solution were 30 mL/h and 6 mL/h, respectively. At the same time, the flow rate of sodium hydroxide solution was adjusted with the fluctuation of pH value. As the co-precipitation proceeded, the concentration of Mn2+ in the feed increased gradually, and the relative concentrations of Ni2+ and Co2+ gradually decreased, and the precursor with gradient structure was synthesized. Added the salt solution of manganese to the salt solution of nickel and cobalt alone could regulate the gentle 5
ACCEPTED MANUSCRIPT gradient change more conveniently and accurately, and made the co-precipitation more stable. GNCMO prepared by this method had a slow and smooth change trend, which could be clearly observed by EDX analysis. Then final product (LGNCMO)
RI PT
was synthesized by mixed the precursor with lithium hydroxide (the mixing molar ratio is 1:1.05), firstly pre-calcined for 4 h at 500, then calcined at 800 for 12 h and then sintered with oxygen-assisted sintering, the flow rate of oxygen is 2.5 L/h.
SC
Preparation of the cross section of spherical secondary particles. In order to
M AN U
better detect the distribution of elements within the material, we dissected the secondary spherical particles prepared by co-precipitation method into hemispheres, and directly qualitatively and quantitatively analyzed the distribution of each element by EDX. The schematic diagram of the preparation of cross-section is shown in Fig. 2.
TE D
Mix epoxy resin A/B evenly, then took a small amount of sample powder and put it on the bottom of the mold, then filled the whole mold with the epoxy resin A/B. It was then aged in air, until the powder firmly embedded in the A/B glue. The surface of
EP
bottom sanded with the sample powder was polished with sandpaper and then placed
AC C
on a polishing machine for further processed. The cross section of spherical secondary particles was obtained which can be observed by SEM. 2.2 Material characteristics The samples were visualized by field emission scanning electron microscope
(FE-SEM, MERLIN Compact, Zeiss) to monitor the topography of surface. On this basis, Energy dispersive X-ray spectroscopy (EDX, OCTANE PLUS, EDAX) was introduced to qualitatively and quantitatively analyzed the distribution of elements. 6
ACCEPTED MANUSCRIPT X-ray diffraction (XRD, DX2700, Haoyuan) patterns was used to detect the crystal structure of materials, and the scan interval was 0.03º. Transmission electron microscopy (TEM, JEOL-2100, Japan) was carried out to observe the presence of
RI PT
lattice spacing and detecting its specific distance. The valence states of nickel were analyzed by X-ray photoelectron spectroscopy (XPS, ESCALAB 250XI, Thermo). The tap density of the sample was tested by a tap density tester (BT-301, Bettersize).
SC
2.3 Electrochemical measurements
M AN U
The final product was the active material, Super P was the conductive agent, and polyvinylidene fluoride (PVDF) was the binder, with a mass ratio of 8:1:1. First the PVDF powder was dissolved in an organic solvent, and prepared into a PVDF solution for standby. Then the active material, Super P and PVDF solution were
TE D
weighed into a bottle and stirred by magnetic force. The uniformly stirred slurry was applied to the surface of aluminum foil, then dried and cut into a positive electrode sheet of desired size. The 2025 coin cell was used, and the diaphragm was made of
EP
Celgard 2400, polypropylene micro diaphragm. The electrolyte was prepared by
AC C
dissolved 1.0 mol/L of LiPF6 in a mixture of dimethyl carbonate (DMC) + diethyl carbonate (DEC) + ethylene carbonate (EC) (volume ratio 1:1:1). In a glove box filled with inert gas (Argon, 99.999%), the coin half-cell was assembled. Battery monitoring system (Land CT2001A) was introduced to monitor the charge and discharge of coin batteries. The performance of the battery was evaluated in the voltage range of 2.75-4.3 V. Charge and discharge tests were carried at current rates of 0.1 C, 0.2 C, 0.5 C, 1.0 C, 2.0 C, 5.0 C and 10.0 C, respectively. The cyclic 7
ACCEPTED MANUSCRIPT voltammetry and impedance were tested by an electrochemical workstation (CHI 604e, Chenhua). The scanning speed of cyclic voltammetry test was 0.1 mV/s, and the potential interval was between 2.8 and 4.3 V. In the impedance test, the amplitude of
RI PT
disturbance voltage was 5 mV, and the set frequency range was 0.01 to 100 kHz. All the electrochemical impedance spectroscopy (EIS) of LNCMO and LGNCMO were conducted at open circuit voltage after complete discharge.
SC
3. Results and discussion
M AN U
Scanning electron microscope (SEM) images of the precursors of non-gradient intrinsic material (NCMO) and full-gradient material (GNCMO) are shown in Fig. 3a and 3c. From the macroscopic morphology of spherical particles, there is no significant difference between GNCMO and NCMO, both of which are smooth and
TE D
spherical secondary particles. Further magnifying the surface of the spherical precursors finds that the surface of GNCMO is denser and smoother than NCMO. According to the tap density measurement, the tap density of NCMO is 1.85 g/ml, and
EP
the tap density of GNCMO is 1.92 g/ml.
AC C
In order to better verify the distribution of the elements inside spherical secondary particles of precursor, we dissect the secondary spherical particles prepared by co-precipitation method into hemispheres, and the distribution of elements on cross section are subjected to linear scan of EDX. The scanning results are shown in Fig. 3b and 3d. The path of linear scan begins along one edge of hemisphere and passes through the center and continues to scan linearly to the other edge of hemisphere. It can be seen that the linear scan curves of the three elements of nickel, cobalt and 8
ACCEPTED MANUSCRIPT manganese in NCMO are relatively flat, indicating that the distribution of the three elements inside the primary particles of intrinsic material is relatively uniform. The linear scan curves of nickel, cobalt and manganese in GNCMO show obvious
RI PT
fluctuating trend. The full-gradient material is a nickel-rich material with high specific capacity as the inner core, and gradually transfers to the relatively low nickel and manganese-rich outer surface. The relative molar content of nickel gradually
SC
decreases from 88% to 72% along the center to the edge of hemisphere, while the
M AN U
relative molar content of manganese increases gradually. The relative molar content of cobalt fluctuates with the contents of nickel and manganese. Moreover, the contents of the three elements changes in a smooth arc. This gentle radian change is because the degree of gradient change in the co-precipitation process can be precisely
TE D
controlled. Gradient material with radian changes are smoother than linear changes, and the primary particles deposited in precursor are more compact and denser, which proves that the gradient material synthesized by this method is relatively compact.
EP
The NCMO and GNCMO are uniformly mixed with LiOH·H2O, placed in a
AC C
tubular resistance furnace for annealing treatment to obtain the final products, namely LNCMO and LGNCMO, respectively. The SEM images of final products are shown in Fig. 3e and 3g. It can be seen that the final products prepared by calcination still maintain the same regular spherical shape as the precursors. From the magnified images of the surface of final products, it is observed that the single crystal particles on the surface are coarser than the precursors and the distribution of each crystal grain is remarkable. There is no significant difference in the macroscopic morphology 9
ACCEPTED MANUSCRIPT between the two materials, and the single crystal particles on the surface are densely packed. According to the tap density measurement, the tap density of LNCMO is 2.26 g/ml, and the tap density of LGNCMO is 2.38 g/ml. The results show that the final
RI PT
product synthesized by the compacted precursor is more compacted. 5.0 g of LNCMO and LGNCMO are weighed separately, dissolved in 50 ml of deionized water, and magnetically stirred for 10 min. After the solutions
SC
are allowed to stand for clarification, the pH of the supernatant is accurately
M AN U
measured with a pH meter. The pH values of LNCMO and LGNCMO are measured to be 11.78 and 11.57, respectively. It indicates that the content of residual alkali on the surface of the relatively low-nickel, manganese-rich gradient material is lower than that of intrinsic material. The full-gradient
TE D
structure not only can effectively solve the problem of high residual alkali content on the surface of nickel-rich materials, but also retard the side reactions of LiOH and Li2CO3 with electrolyte, prevent the material from being attacked
EP
by HF and the formation of inactive "NiO" phase [30].
AC C
EDX linear scan is performed on the elemental distribution of the cross sections of LNCMO and LGNCMO, respectively. The linear scan results are shown in Fig. 3f and 3h. The path of linear scan begins along one edge of hemisphere and passes through the center, and then continues to scan linearly to the other edge of hemisphere. It can be seen that the tendency in elemental distribution of final products prepared by lithiation are consistent with the precursors, and the three elements inside LNCMO are relatively evenly distributed. Although the elemental distribution inside LGNCMO 10
ACCEPTED MANUSCRIPT maintains a full-gradient trend, the gradient is slightly slower than that of the precursor, which transforms from the nonlinear gradient of the precursor to a gentle linear gradient. The relative molar content of the central nickel element is reduced by
RI PT
84% and the edge is increased to 76%, the relative molar contents of cobalt and manganese also showed corresponding change. This phenomenon is caused by the thermal diffusion of ions during calcination, which results in the slowing down of
SC
concentration gradient.
M AN U
The uniformity of elemental distribution is further observed by surface scan of EDX. The surface scan is performed on the cross section and the complete spherical secondary particles of precursors (Fig. 4a-d). Comparing the EDX surface scan images of the hemispheres of NCMO and GNCMO, it is found that the three and the manganese
TE D
transition metal elements in NCMO are evenly distributed,
content in GNCMO gradually is increased from the center to the edge of hemisphere, and the cobalt is relatively reduced. The nickel element appears to be relatively evenly
EP
distributed, which is due to the material itself contains relatively high content of
AC C
nickel, so the difference of EDX surface scan is not obvious. Comparing the EDX surface scan images of the complete spherical secondary particles of NCMO and GNCMO, the distribution of nickel, cobalt and manganese on the outer surface of the two materials are relatively uniform, indicating that the elemental gradient of GNCMO is longitudinally changed from the center of the spherical secondary particle to the outer surface. That is, as the reaction proceeds, the three elements are uniformly grown on the core surface and continuously stacked to form uniform and dense 11
ACCEPTED MANUSCRIPT spherical secondary particle, which again proves that our method is accurate and controllable. The surface scan is also performed on the cross section and the complete
RI PT
spherical secondary particles of final products (Fig. 4e-h). The elements in the cross section of LNCMO are evenly distributed. The nickel element in the cross section of LGNCMO is uniformly distributed due to the high content. The gradient trends of
SC
cobalt and manganese are slightly slower than that of the precursor, which is
M AN U
consistent with the results obtained by linear scan of EDX. The elements on the surface of the complete spherical secondary particles of LNCMO and LGNCMO are evenly distributed. This indicates that the single crystal particles of final products still maintain the trend of radial growth of the precursors.
TE D
X-Ray diffraction (XRD) is introduced to analyze the structures of NCMO and GNCMO (Fig. 5a). The XRD diffraction peaks of the precursors of these two materials are coincided with the peaks of β-Ni(OH)2 (PDF card No. 14-0117) and
EP
none of them showed impurity peaks. However, the (001) peak of NCMO is
AC C
significantly lower than the peaks in GNCMO and that of β-Ni(OH)2. This may be due to the introduction of cobalt and manganese in NCMO and the uniform distribution of the three elements. The synergistic effect of the three transition metal elements affects the structure of (001) crystal plane. Although the spectrum of full-gradient material is closer to that of β-Ni(OH)2, its (001) peak is also slightly lower than β-Ni(OH)2, which may be due to the uneven distribution of the three elements. From the core to the outer surface of GNCMO, the material gradually 12
ACCEPTED MANUSCRIPT transitions from a nickel-rich core of substantially Ni(OH)2 to a ternary synergistic low-nickel outer layer. The XRD diffraction peaks of LNCMO and LGNCMO are coincided with the
RI PT
peaks of the standard card of LiNi0.8Co0.1Mn0.1O2 (Fig. 5b). None of the peaks of LNCMO shows impurity peaks. However, there is a significant small split at the bottom of the (003) peak in the XRD spectrum of LGNCMO, the splitting of the peak may be
SC
caused by the uneven distribution of internal elements in the gradient material. In
M AN U
addition, the splits of (108) and (110) peak of the two materials are obvious and the ratio of I(003)/I(104) are greater than 1.2, indicating that both have good layered structure [31-33]. The XRD spectroscopy of LNCMO and LGNCMO were further subjected to structural refinement (Table 1). According to the reference [34], when the ratio of
TE D
lattice parameter c/a > 4.9, the ratio can represent that the material has formed a layered structure. It can be seen from the results of structural refinement in Table 1 that a and c of the lattice parameters of LGNCMO are slightly increased relative to
EP
LNCMO. However, the ratio of c/a is equal to LNCMO and the value is
AC C
approximately 4.94, further indicating that both materials formed good layered structure. A good layered structure facilitates the deintercalation of Li+ and has good electrochemical performance. The R value indicates the fitting error, and the smaller the R value, the better the fitting is. Table 1. XRD structural refinement of LNCMO and LGNCMO sample
I(003)/I(104)
a(Å)
c(Å)
c/a
R
LNCMO
1.53
2.864
14.163
4.945
5.66 %
LGNCMO
1.51
2.865
14.161
4.943
5.32 %
13
ACCEPTED MANUSCRIPT The microstructure is observed by transmission electron microscopy (TEM). Fig. 5c and 5d are the TEM images of NCMO and GNCMO, respectively. The layered structure of the precursors of two materials are obvious and the layer spacing is about
RI PT
0.23 nm, which corresponds to the spacing of the (101) crystal plane in the XRD spectrum. In addition, due to the gentle variation of the full-gradient, there is no significant effect on crystal growth. Fig. 5e and 5f are the TEM images of LNCMO
SC
and LGNCMO, respectively. It can be seen that the lattice fringes of the final products
M AN U
of two materials are more obvious than the precursors and the layered structure are good, which indicates that the materials after lithiation have higher degree of crystallization. The interplanar spacing of the two materials is about 0.47 nm, which corresponds to the spacing of the (003) crystal planes in the XRD spectrum.
TE D
The valence state of nickel is analyzed by X-ray photoelectron spectroscopy (XPS), and Fig. 5g shows the full-spectrum scan curves of the surface of LNCMO and LGNCMO. The diffraction peaks of both materials are the same, indicating that the
EP
binding energy of each element of gradient material is in agreement with the intrinsic
AC C
material. The curve of peak-differentiating-imitating of nickel is shown in Fig. 5h, the peaks with binding energies of 854.4 eV, 857.1 eV and 855.5 eV correspond to Ni2+ and Ni3+, respectively [35]. Firstly, we fit the original XPS data to make the curve smoother and the separation peak more convenient. Then, the positions of Ni2+ and Ni3+ peaks are determined and add to the fitted spectra. After region fitting, the spectrum after peak separation is obtained. The distribution of the valence states of nickel is obtained by peak fitting (Table S1). The total content of Ni2+ and the total content of 14
ACCEPTED MANUSCRIPT Ni3+ in LGNCMO are close to that of LNCMO, indicating that the gentle change of the gradient of each element in LGNCMO does not significantly affect the content of Ni3+. However, the distribution of Ni2+ of gradient material is significantly different
RI PT
from that of intrinsic material. This is due to the gradient distribution of the three elements of nickel, cobalt and manganese in LGNCMO. The content of Mn4+ on the surface of LGNCMO is relatively more. In order to balance the valence state of cobalt
SC
and manganese, the distribution of Ni2+ and Ni3+ fluctuates with the gradient of the
M AN U
contents of the two elements [36].
The charge and discharge test curves of LNCMO and LGNCMO are shown in Fig. 6a-c. With a voltage window of 2.75-4.3 V and a current rate of 0.1 C, the first discharge specific capacities of LNCMO and LGNCMO are 198.7 mAh/g and 198.3
TE D
mAh/g, respectively. The discharge specific capacities at the current rate of 1 C are 179.6 mAh/g and 182.5 mAh/g, respectively. Further comparison of the discharge voltage platform, it can be found that at a current rate of 0.1 C and 1 C, the voltage
EP
platforms of LGNCMO are slightly higher than that of LNCMO, corresponding to a
AC C
higher discharge specific energy. It can be seen from the rate discharge curve that when the current rate is small, the discharge specific capacities of the two materials are relatively close. When the current rate is increased to more than 1 C, the specific capacity of LGNCMO is slightly higher than that of LNCMO, and the larger the current rate, the more obvious the difference, which shows excellent application prospect of power batteries. However, when the current rate is cycled from 10 C to 0.1 C, the specific capacity of LGNCMO is still higher than that of LNCMO, indicating that 15
ACCEPTED MANUSCRIPT LGNCMO has better cycle performance. In the cycle curves, the retention rate of discharge capacity of LGNCMO after 100 cycles at current rate of 1 C and 5 C were 98.8% and 93.7%, respectively, which is higher than 96.6% and 90.3% of LNCMO.
RI PT
This is because the full-gradient material has a relatively low nickel and manganese-rich surface layer that has better cycle performance and thermal stability relative to the intrinsic material [36].
SC
The full-gradient material has a flat compositional change compared to the
M AN U
core-shell material, avoiding structural collapse due to excessive component differences during charge and discharge [26-27]. The process after first charge and discharge is further analyzed by the differential capacity vs voltage plots (dQ/dv vs V) (Fig. 6d), which is similar to the cyclic voltammogram (CV) and has the same
TE D
characteristics as the CV curve (Fig. S1). The dQ/dv vs V curves of the two materials show three sets of redox peaks, corresponding to three phase transitions during charge and discharge, that is, the hexagonal (H1) ↔ the monoclinic (M) ↔ the hexagonal (H2)
EP
↔ the hexagonal (H3) [37,38]. At the current rate of 0.1 C, the phase transition of the
AC C
dQ/dv vs V curve of LGNCMO after the first charge and discharge is obvious than that of LNCMO. Phase transformation of H2-H3 is usually irreversible, which leads to volume expansion and cracks, and ultimately leads to a rapid decline in capacity during the cycle. But from the CV curve of LGNCMO (Fig. S1), it can be seen that the oxidation peak and reduction peak shape of H2-H3 are similar, and the peak shape area is basically the same. It shows that the phase transformation of H2-H3 is reversible for LGNCMO during charging and discharging process, and almost does not lead to 16
ACCEPTED MANUSCRIPT volume expansion. This may be attributed to the structural characteristics of the full gradient. In addition, the peaks of the two materials at around 3.7 V is significantly higher than the other two peaks, indicating that more Li+ are released and embedded
RI PT
around 3.7 V, also indicating that the charge and discharge platform are around 3.7 V. Further comparison of the peak potential difference between oxidation peak and reduction peak in CV curves shows that the peak potential difference of LGNCMO is
SC
relatively small, indicating that the polarization on electrode is smaller and the
M AN U
reversibility is better than LNCMO.
At the current density of 0.1 C, the battery is placed in a battery test system for one charge and discharge cycle, and then place on an electrochemical workstation for electrochemical impedance testing (Fig. 6e). The high frequency capacitive reactance
TE D
arc in the impedance spectrum corresponds to the migration process of Li+ in the SEI film. The medium frequency capacitive reactance arc corresponds to the transfer process of Li+ at the interface between the SEI film and the electrode active material,
EP
and the low-frequency line corresponds to the diffusion process of Li+ in the solid phase
AC C
[39-41]. According to this analysis, the equivalent circuit of electrode can be established. In the equivalent circuit of the impedance test, RΩ represents the ohmic resistance of electrode, including the ohmic resistance of the solution in the diaphragm and the ohmic resistance of the electrode itself. QSEI and RSEI represent the capacitance and resistance of the SEI film, respectively. Qd represents the electric double layer capacitance, Rct represents the charge transfer resistance, and W represents the solid phase diffusion impedance [42,43]. According to the results of the equivalent circuit 17
ACCEPTED MANUSCRIPT fitting, the parameters are obtained (Table 2). Although the RΩ and RSEI of LNCMO are similar to LGNCMO, its Rct are relatively large, which is disadvantageous for Li+ transfer and diffusion. The Rct of LGNCMO are obviously smaller than LNCMO,
RI PT
which corresponds to better electron transport ability and ion diffusion capacity, and the material has good electrochemical performance.
The lithium-ion diffusion coefficient can be calculated using the following
M AN U
SC
equation (1). [44]
In this formula, R is the gas constant (8.314 J K-1 mol-1), T is 298 K, A is the surface area of the electrode, n is the number of electrons involved in reaction, F is the
TE D
Faraday constant (96 485 C mol-1), C is the concentration of lithium ion, and σ is the Warburg coefficient, which can be obtained from the line of Z′~ω-1/2. can be obtained by taking the known data into formula (1) for calculation
EP
(Table 2). From the calculated results, it can be seen that the
value of LGNCMO
AC C
is larger than that of LNCMO, indicating that the gradient structure is beneficial to the diffusion of lithium ions. The
results of impedance in the full paper are all
calculated in this way.
Table 2. Impedance fitting parameters of LNCMO and LGNCMO
sample
RΩ/Ω
RSEI/Ω
Rct/Ω
LNCMO
4.61
35.62
108.07
2.592×10-17
LGNCMO
4.75
36.16
48.31
3.073×10-17
The impedance and cyclic voltammogram of the battery after 100 cycles at current 18
ACCEPTED MANUSCRIPT densities of 1C and 5C are tested. As shown in Fig. S2, the impedance spectra shows that the total impedance of the gradient material is less than that of intrinsic material after 100 cycles at current densities of 1C and 5C. According to the result of the
RI PT
equivalent circuit fitting, the parameters are obtained (Table 3). RSEI of LGNCMO is obviously smaller than that of LNCMO. The Rct of LGNCMO after 100 cycles at 1C rate is similar to that of LNCMO, and the Rct of LGNCMO after 100 cycles at 5C rate is
values of LGNCMO are larger than
M AN U
electrochemical reactions. In addition, the
SC
smaller than that of LNCMO. This is undoubtedly conducive to the rapid progress of
that of LNCMO, indicating that the gradient structure after 100th cycles is still beneficial to the diffusion of lithium ions. Further comparison of the peak potential difference between oxidation peak and reduction peak in CV curves shows that the
TE D
peak potential difference of LGNCMO is relatively small (Fig. S3), which means the polarization on electrode is smaller and the reversibility is better than LNCMO. In addition, the differential capacity vs voltage plots for the 5th, 25th, 50th, 75th and 100th
EP
cycles at 1C rate also given in Fig. S4. It can be seen that the peak position difference of
AC C
LGNCMO curve under different cycles are smaller than that of LNCMO, which indicates that LGNCMO has better reversibility. Moreover, the curve coincidence between LGNCMO cycles is better than LNCMO, and the symmetry of redox peaks is also good, which indicates that LGNCMO has better cycle stability. In order to analyze the structure stability of LNCMO and LGNCMO after electrochemical cycles, the XRD spectra (Fig. S5) and TEM images (Fig. S6) after 100th cycles at 5 C are analyzed. As can be seen from the XRD spectra, the peaks of 19
ACCEPTED MANUSCRIPT LNCMO and LGNCMO are consistent with the standard card, and the ratio of 003/104 still maintains the trend before electrochemical cycles, but the peak splitting of 006 and 012 is slightly less obvious. The splitting of the two peaks is not obvious,
RI PT
indicating that the order of the layered structure is somewhat damaged. Generally, it can be concluded from the overall XRD spectrum that there is no significant change in the structure of the two materials after cycles. From the TEM images after
SC
electrochemical cycles, it can be seen that obvious lattice spacing can still be observed
M AN U
between the two materials. But further observation of local magnification exhibits that LNCMO has slight lattice disorder, while LGNCMO performs well, which shows that the gradient structure has better stability.
Table 3. Impedance fitting parameters of LNCMO and LGNCMO after 100 cycles at 1C and 5C RΩ/Ω
RSEI/Ω
Rct/Ω
LNCMO-1C
4.66
27.94
21.39
2.315×10-16
LGNCMO-1C
4.26
10.91
21.66
4.103×10-16
LNCMO-5C
4.32
27.81
17.9
2.847×10-16
LGNCMO-5C
4.04
14.24
15.14
4.724×10-16
EP
TE D
sample
In order to verify the stability of the material in high temperature cycle, the battery
AC C
is charged and discharged at 50. In order to facilitate large rate comparison, we directly choose the current density of 5C for charge-discharge test. As can be seen from Fig. 7a, the initial capacity of LNCMO and LGNCMO are 165.3 mAh/g and 167.1 mAh/g, respectively, and higher than the initial capacity at room temperature. However, the curve of LNCMO declines sharply after more than 65 cycles, while the curve of LGNCMO is relatively stable, and the capacity retention rate of 100 cycles is as high as 90%, which is obviously higher than 71.8% of intrinsic material. This shows that the 20
ACCEPTED MANUSCRIPT gradient material has good high temperature cyclic stability. The impedance and cyclic voltammetry after 100 cycles are measured as shown in the Fig. 7b and 7c. It can be concluded that the impedance of LGNCMO is obviously smaller than that of LNCMO,
RI PT
and the Rct of LGNCMO are obviously smaller than that of LNCMO from the parameters of impedance fitting (Table 4), which shows that LGNCMO has good high-rate performance at high temperature. Meanwhile, the
value of LGNCMO is values at room
SC
still higher than that of LNCMO and higher than that of
M AN U
temperature. It is shown that the lithium ion diffusivity can be improved at higher temperature. From the cyclic voltammetry curves, it can be seen that the peak position difference of LGNCMO is significantly smaller than that of LNCMO, and the redox current value of LGNCMO is also significantly higher than that of LNCMO, which
capacity stability.
TE D
again proves that LGNCMO has good high temperature cyclic reversibility and
Table 4. Impedance fitting parameters of LNCMO and LGNCMO after 100 cycles at 5C (at 50) RΩ/Ω
EP
sample
RSEI/Ω
Rct/Ω
3.99
11.31
50.65
9.367×10-16
LGNCMO
3.71
12.36
30.04
1.212×10-15
AC C
LNCMO
In order to further analyze the effect of the full-gradient structure on the
performance, we present the cyclic curves of midpoint voltage and energy density of cathode material in the discharge process. Fig. 8a and 8b show the midpoint voltage curves at room temperature and high temperature respectively. It can be seen that the midpoint voltage of LGNCMO is obviously higher than that of LNCMO under different current densities and temperatures. More obviously, at the current density of 21
ACCEPTED MANUSCRIPT 5C, the midpoint voltage of LNCMO decreased more seriously than that of LGNCMO. In addition, at the high temperature of 50, the cyclic stability of the midpoint voltage of LGNCMO is significantly better than that of LNCMO. Further, we analyzed the
RI PT
energy density of cathode materials as shown in Figure 8c, d, and find that the energy density of LGNCMO is higher than that of LNCMO under different conditions. This
performance and high temperature cycling stability.
M AN U
4. Conclusions
SC
further indicates that the full-gradient structure can effectively improve the high rate
Full-gradient structured LiNi0.8Co0.1Mn0.1O2 has successfully prepared by a novel and facilitate method. SEM images and XRD spectra show that gradient material has regular spherical shape and good crystal structure, respectively. EDX test results shows
TE D
that a smooth transition from inner core with 84% nickel content to the outer surface with 76% nickel content of the final product is achieved. The Electrochemical tests show that the full-gradient material has significantly improved cycle performance and
EP
high rate performance, and the higher the current multiplier, the more obvious the
AC C
improvement. What’s more, full-gradient material exhibits good cyclic stability at high temperature, and the capacity retention rate of 100 cycles is as high as 90% at 5C rate, which is obviously higher than 71.8% of intrinsic material. In addition, it can be determined from impedance analysis that the gradient material has improved diffusion coefficient of lithium ion, indicating that it has good electrochemical performance. The gradient material also shows improved midpoint voltage and specific energy, which exhibits excellent application prospect of power batteries. 22
ACCEPTED MANUSCRIPT Acknowledgements The authors acknowledge funding support from the Shandong Provincial Natural Science Foundation of China (Grant No. ZR2017MB032), and the Shandong
References
RI PT
Provincial Major Project of China (Grant No. 2015ZDZX04002).
[1] B. Dunn, H. Kamath, J.-M. Tarascon, Electrical energy storage for the grid: A
SC
battery of choices, Science 334 (2011) 928-935.
M AN U
[2] C. Liu, F. Li, L.-P. Ma, H.-M. Cheng, Advanced materials for energy storage, Adv. Mater. 22 (2010) E28-E62.
[3] B. Scrosati, J. Garche, Lithium batteries: Status, prospects and future, J. Power Sources 195 (2010) 2419-2430.
TE D
[4] S.-T. Myung, F. Maglia, K.-J. Park, C. S. Yoon, P. Lamp, S.-J. Kim, Y. Sun, Nickel-rich layered cathode materials for automotive lithium-ion batteries: achievements and perspectives, Acs Energy Lett. 2 (2017) 196-223.
EP
[5] J. B. Goodenough, Y. Kim, Challenges for rechargeable Li batteries, Chem.
AC C
Mater. 22 (2010) 587-603. [6] V. Etacheri, R. Marom, R. Elazari, G. Salitra, D. Aurbach, Challenges in the development of advanced Li-ion batteries: A review, Energ. Environ. Sci. 4 (2011) 3243-3262.
[7] H.-J. Noh, S. Youn, C. S. Yoon, Y.-K. Sun, Comparison of the structural and electrochemical properties of layered Li[NixCoyMnz]O2 (x=1/3, 0,5, 0,6, 0,7,
23
ACCEPTED MANUSCRIPT 0,8 and 0,85) cathode material for lithium- ion batteries, J. Power Sources, 233 (2013) 121-130. [8] L.-J. Li, Z.-X. Wang, Q.-C. Liu, C. Ye, Z.-Y. Chen, L. Gong, Effect of
RI PT
chromium on the structural, surface chemistry and electrochemical of layered LiNi0.8-xCo0.1Mn0.1CrxO2, Electrochim. Acta 77 (2012) 89-96.
[9] P. Yue, Z. Wang, X. Li, X. Xiong, J. Wang, X. Wu, H. Guo, The enhanced
SC
electrochemical performance of LiNi0.6Co0.2Mn0.2O2 cathode materials by
M AN U
low temperature fluorine substitution, Electrochim. Acta 95 (2013) 112-118. [10]W. Luo, F. Zhou, X. Zhao, Z. Lu, X. Li, J. R. Dahn, Synthesis, characterization, and thermal stability of Li[Ni1/3Mn1/3Co1/3-z(MnMg)(z/2)]O2, Chem. Mater. 22 (2010) 5065-5073.
TE D
[11]S. W. Woo, S. T. Myung, H. Bang, D. W. Kim, Y. K. Sun, Improvement of electrochemical and thermal properties of Li[Ni0.8Co0.1Mn0.1]O2 positive electrode materials by multiple metal (Al, Mg) substitution, Electrochim.
EP
Acta 54 (2009) 3851-3856.
AC C
[12]M. Wang, Y. Chen, F. Wu, Y. Su, L. Chen, D. Wang, Characterization of yttrium substituted LiNi0.33Mn0.33Co0.33O2 cathode material for lithium secondary cells, Electrochim. Acta 55 (2010) 8815-8820.
[13]Y. Cho, P. Oh, J. Cho, A new type of protective surface layer for high-capacity Ni-based cathode materials: nanoscaled surface pillaring layer, Nano Lett. 13 (2013) 1145-1152.
24
ACCEPTED MANUSCRIPT [14]C.-H. Jo, D.-H. Cho, H.-J. Noh, H. Yashiro, Y.-K. Sun, S. T. Myung, An effective method to reduce residual lithium compounds on Ni-rich Li[Ni0.6Co0.2Mn0.2]O2 active material using a phosphoric acid derived Li3PO4
RI PT
nanolayer, Nano Res. 8 (2015) 1464-1479. [15]D. Mohanty, K. Dahlberg, D. M. King, L. A. David, A. S. Sefat, D. L. Wood, C. Daniel, S. Dhar, V. Mahajan, M. Lee, F. Albano, Modification of Ni-rich
SC
FCG NMC and NCA cathodes by atomic layer deposition: preventing surface
M AN U
phase transitions for high-voltage lithium-ion batteries, Sci. Rep. 6 (2016) 26532.
[16]J. Cho, T.-J. Kim, J. Kim, M. Noh, B. Park, Synthesis, thermal, and electrochemical properties of AlPO4-Coated LiNi0.8Co0.1Mn0.1O2 cathode
TE D
materials for a Li-ion cell, J. Electrochem. Soc. 151 (2004) A1899-A1904. [17]M. G. Kim, J. J. Cho, Air stable Al2O3-coated Li2NiO2 cathode additive as a surplus current consumer in a Li-ion cell, Mater. Chem. 18 (2008) 5880-5887.
EP
[18]H. Lee, Y. Kim, Y. S. Hong, Y. Kim, M. G. Kim; N. S. Shin, J. Cho,
AC C
Structural Characterization of the Surface-Modified LixNi0.9Co0.1O2 Cathode Materials by MPO4 Coating (M = Al , Ce, SrH, and Fe) for Li-Ion Cells, J. Electrochem. Soc. 153 (2006) A781-A786.
[19]X. Meng, X.-Q. Yang, X. Sun, Meng X , Yang X Q , Sun X . Emerging Applications of Atomic Layer Deposition for Lithium-Ion Battery Studies, Adv. Mater. 24 (2012) 3589-3615.
25
ACCEPTED MANUSCRIPT [20]H. Yang, H. Wu, M. Ge, L. Li, Y. Yuan, Q. Yao, J. Chen, L. Xia, J. Zheng, Z. Chen, J. Duan, K. Kisslinger, Q. Zhang, J. Lu, Simultaneously Dual Modification of Ni-rich Layered Oxide Cathode for High-energy Lithium-ion
RI PT
Batteries, Adv. Funct. Mater. (2019) 1808825. [21]L. Li, Z. Chen, Q. Zhang, M. Xu, X. Zhou, H. Zhu and K. Zhang, A hydrolysis-hydrothermal route for the synthesis of ultrathin LiAlO2-inlaid
M AN U
batteries, J. Mater. Chem. A 3 (2015) 894-904.
SC
LiNi0.5Co0.2Mn0.3O2 as a high-performance cathode material for lithium ion
[22]W. Liu, P. Oh, X. Liu, M. J. Lee, W. Cho, S. Chae, Y. Kim, J. Cho, Nickel-rich layered lithium transition-metal oxide for high-energy lithium-ion batteries, Angew. Chem. Int. Ed. Engl. 54 (2015) 4440-4457.
TE D
[23]K. Meng, Z. Wang, H. Guo, X. Li, D. Wang, Improving the cycling performance of LiNi0.8Co0.1Mn0.1O2 by surface coating with Li2TiO3, Electrochim. Acta 211 (2016) 822-831.
EP
[24]J. M. Lim, T. Hwang, D. Kim, M. S. Park, K. Cho, M. Cho, Intrinsic origins
AC C
of crack generation in ni-rich LiNi0.8Co0.1Mn0.1O2 layered oxide cathode material, Sci. Rep. 7 (2017) 39669.
[25]Y.-K. Sun, S.-T. Myung, M.-H. Kim, J. Prakash, K. Amine, Synthesis and characterization
of
Li[(Ni0.8Co0.1Mn0.1)0.8(Ni0.5Mn0.5)0.2]O2
with
the
microscale core-shell structure as the positive electrode material for lithium batteries, J. Am. Chem. Soc. 127 (2005) 13411-13418.
26
ACCEPTED MANUSCRIPT [26]Y.-K. Sun, S.-T. Myung, B.-C. Park, K. Amine, Synthesis of spherical nano-to microscale core-shell particles Li[(Ni0.8Co0.1Mn0.1)1-x(Ni0.5Mn0.5)x]O2 and their applications to lithium batteries, Chem. Mater. 18 (2006)
RI PT
5159-5163. [27]Y.-K. Sun, S.-T. Myung, H.-S. Shin, Y.-C. Bae, C. S. Yoon, Novel core-shell structured Li[(Ni0.8Co0.1Mn0.1)0.8(Ni0.5Mn0.5)0.2]O2 via coprecipitation as
SC
positive electrode material for lithium secondary batteries, J. Phys. Chem. B
M AN U
110 (2006) 6810-6815.
[28]Y.-K. Sun, Z. Chen, H.-J. Noh, D.-J. Lee, H.-G. Ren, Y. Jung, S. Wang, C. S. Yoon, S.-T. Myung, K. Amine, Nanostructured high-energy cathode materials for advanced lithium batteries, Nat. Mater. 11 (2012) 942-947.
TE D
[29]Y. Li, R. Xu, Y. Ren, J. Lu, H. Wu, L. Wang, D. J. Miler, Y. Sun, K. Amine, Z. Chen, Synthesis of full concentration gradient cathode studied by high energy X-ray diffraction, Nano Energy 19 (2016) 522-531.
EP
[30]K. Matsumoto, R. Kuzuo, K. Takeya, A. Yamanaka, Effects of CO2 in air on
AC C
Li deintercalation from LiNi1-x-yCoxAlyO2, J. Power Sources 81-82 (1999) 558-561.
[31]A. Rougier, P. Gravereau, C. Delmas, Optimization of the composition of the Li1-zNi1+zO2 electrode materials: structural, magnetic, and electrochemical
studies, J. Eletrochem. Soc. 143 (1996) 1168-1175.
27
ACCEPTED MANUSCRIPT [32]J. R. Dahn, U. V. Sacken, C. A. Michal, Structure and electrochemistry of Li1±yNiO2 and a new Li2NiO2 phase with the Ni(OH)2 structure, Solid State Ionics 44 (1990) 87-97.
RI PT
[33]J. Morales, C. Perez-Vicente, J. L. Tirado, Cation distribution and chemical deintercalation of Li1-xNi1+xO2, Mater. Res. Bull. 25 (1990) 623-630.
[34]J.-H. Kim, C. W. Park, Y.-K. Sun, Synthesis and electrochemical behavior of
SC
Li[Li0.1Ni0.35-x/2CoxMn0.55-x/2]O2 cathode materials, Solid State Ionics 163
M AN U
(2003) 43-49.
[35]J. Duan, G. Hu, Y. Cao, C. Tan, C. Wu, K. Du, Z. Peng, Enhanced electrochemical performance and storage property of LiNi0.815Co0.15Al0.035O2 via Al gradient doping, J. Power Sources 326 (2016) 322-332.
TE D
[36]Y. Wei, J. Zheng, S. Cui, X. Song, Y. Su, W. Deng, Z. Wu, X. Wang, W. Wang, M. Rao, Y. Lin, C. Wang, K. Amine, F. Pan, Kinetics tuning of Li-ion diffusion in layered Li(NixMnyCoz)O2, J. Am. Chem. Soc. 137 (2015)
EP
8364-8367.
AC C
[37]W. Li, J. N. Reimers, J. R. Dahn, In situ X-ray diffraction and electrochemical studies of Lil-xNiO2, Solid State Ionics 67 (1993) 123-130.
[38]J. Barker, R. Koksbang, M. Y. Saidi, An electrochemical investigation into the lithium insertion properties of LixNiO2 (0≤x≤1), Solid State Ionics 89
(1996) 25-35. [39]D. Aurbach, L. Mikhael, E. Levi, H. Teller, B. Markovsky, G. Salitra, U. Heider, L. Heider, Common electroanalytical behavior of Li intercalation 28
ACCEPTED MANUSCRIPT processes into graphite and transition metal oxides, J. Electrochem. Soc. 145 (1998) 3024-3034. [40]D. Aurbach, K. Gamolsky, B. Markovsky, G. Salitra, Y. Gofer, U. Heider, R.
RI PT
Oesten, M. Schmidt, The Study of surface phenomena related to electrochemical lithium intercalation into LixMOy host materials (M = Ni, Mn), J. Electrochem. Soc. 147 (2000) 1322-1331.
SC
[41]D. Aurbach, M. D. Levi, K. Gamulski, B. Markovsky, G. Salitra, E. Levi, U.
M AN U
Heider, L. Heider, R. Oesten, Capacity fading of LixMn2O4 spinel electrodes studied by XRD and electroanalytical techniques, J. Power Sources 81-82 (1999) 472-479.
[42]Q.-C. Zhuang, T. Wei, L.-L. Du, Y.-L. Cui, L. Fang, S.-G. Sun, An
TE D
electrochemical impedance spectroscopic study of the electronic and ionic transport properties of spinel LiMn2O4, J. Phys. Chem. C 114 (2010) 8614-8621.
EP
[43]G. Gi, Y. Ko, T.-H. Kim, H.-K. Song, S. Bin, S.-M. Park, Fourier transform
AC C
electrochemical impedance spectroscopic studies on LiFePO4 nanoparticles of hollow sphere secondary structures, J. Electrochem. Soc. (2013) 158 A1267-A1274.
[44]Z. Li, F. Du, X. Bie, D. Zhang, Y. Cai, X. Cui, C. Wang, G. Chen, Y. Wei, Electrochemical Kinetics of the Li[Li0.23Co0.3Mn0.47]O2 Cathode Material Studied by GITT and EIS, J. Phys. Chem. C 114 (2010) 22751-22757.
29
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
S0013-4686(19)30735-2
DOI:
https://doi.org/10.1016/j.electacta.2019.04.058
Reference:
EA 34006
To appear in:
Electrochimica Acta
Received Date: 15 January 2019 Revised Date:
9 April 2019
Accepted Date: 9 April 2019
Please cite this article as: Y. Jiang, Z. Liu, Y. Zhang, H. Hu, X. Teng, D. Wang, P. Gao, Y. Zhu, Fullgradient structured LiNi0.8Co0.1Mn0.1O2 cathode material with improved rate and cycle performance for lithium ion batteries, Electrochimica Acta (2019), doi: https://doi.org/10.1016/j.electacta.2019.04.058. 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 Full-gradient Structured LiNi0.8Co0.1Mn0.1O2 Cathode Material with Improved Rate and Cycle Performance for Lithium ion batteries Yunpeng Jiang,a,b , Zhihao Liu,a Yongzheng Zhang,a Huili Hua, Xiangguo Teng,a
RI PT
Dianlong Wang,b Peng Gaoa,* and Yongming Zhua,* a. Harbin Institute of Technology at Weihai, Department of Applied Chemistry, Weihai 264209, China. b. Harbin Institute of Technology, School of Chemical
SC
Engineering and Technology, Harbin 150001, China.
* Corresponding authors. Tel.: +86 631 5687232 (P. Gao), +86 631 5687901 (Y. Zhu).
M AN U
E-mail addresses: [email protected] (P. Gao), [email protected] (Y. Zhu).
Abstract: LiNi0.8Co0.1Mn0.1O2 cathode material with full-gradient structure is prepared by a novel co-precipitation method. Scanning electron microscope images show that the material presents regular spherical shape. X-ray diffraction pattern shows that the
TE D
material has good layered structure. Energy dispersive X-ray spectroscopy analysis for the cross section of secondary particles shows that the relative molar content of Ni
EP
gradually decreases from 84% to 76% along the center to the edge of hemisphere, while the Mn content increases gradually. Moreover, the Co content also presents a
AC C
slow gradient change. Electrochemical tests at room temperature show that the retention rates of discharge capacity for full-gradient material after 100 cycles at current density of 1 C and 5 C are 98.8% and 93.7%, respectively, which are obviously improved compared with intrinsic material. The gradient material also shows improved midpoint voltage and specific energy. Moreover, full-gradient material exhibits good cyclic stability at high temperature, and the capacity retention rate of 100 cycles is as high as 90% at 5C rate, which is obviously higher than 71.8% 1
ACCEPTED MANUSCRIPT of intrinsic material. Impedance spectroscopy studies show that the solid-state diffusion coefficient of lithium ion in gradient material has been significantly improved, thus showing good electrochemical properties. The full-gradient material shows
RI PT
excellent application prospects in the field of power batteries. Keywords: Li-ion battery, nickel-rich cathode material, LiNi0.8Co0.1Mn0.1O2, full-gradient structure, a novel co-precipitation method
SC
1. Introduction
M AN U
At present, with the development of new energy automobile technology, the energy density of lithium-ion power batteries has been put forward increasingly requirements. The actual specific capacity of commercial cathode materials is generally lower than 160 mAh/g, which has become a key factor restricting the
TE D
capacity of lithium-ion batteries. In recent years, nickel-rich ternary cathode materials have received increasing attention due to their high specific capacity and low cost [1-3]. Among them, LiNi1−x−yCoxAlyO2 and LiNi1−x−yCoxMnyO2 (x + y < 0.2) are the
EP
most promising materials [4]. However, such relatively high nickel content leads to
AC C
poor stability [5-7]. Therefore, it is necessary to modify the nickel-rich ternary materials to improve its electrochemical performance. Nevertheless, most of modified methods are limited to elemental doping (such as Mg, Al, F, Cr, Y, et al.) [8-12] and surface coating (such as Al2O3, ZrO2, AlPO4 et al.) [13-19]. In addition, Yang et al. [20] and Li et al. [21] combined doping and coating modification to improve performance, such as dual-modification strategy and lithium residues-assistant modification. These methods are further processed on the basis of intrinsic material, 2
ACCEPTED MANUSCRIPT and the structural deterioration during the charge-discharge cycles cannot be fundamentally solved [22-24]. Recent studies have shown that core-shell structure is an effective way to improve the shortcomings of high nickel, and the low nickel shell
RI PT
coated with high nickel core can achieve good modification effect, Sun et al. [25] synthesized a spherical core-shell structure with high capacity and good thermal stability, which is about the microscale spherical core-shell structure, that is, the
SC
Li[Ni0.8Co0.1Mn0.1]O2 as the core and the Li[Ni0.5Mn0.5]O2 as the shell. In addition,
M AN U
Sun and Amine et al. [26,27]. further explored the effect of core-shell structure with different thickness on electrochemical performance, which makes the thickness of core-shell structure further optimized and promotes the research of core-shell structure.
However,
the
core-shell
structure
has
the
disadvantage
of
TE D
concentration-varying, and the structure is still easy to be destroyed. Therefore, learning from the advantages of core-shell structure, and a full-gradient material was designed, which internal structure changes very smoothly and controllably. In the
EP
research of full-gradient structure, the optimization of internal structure shows better [28] reported a nickel-rich material of
AC C
modified prospects. Sun et al.
LiNi0.75Co0.1Mn0.15O2, where the atomic percentage of Co remained constant, whereas the concentration of Ni decreased and Mn increased continuously from the center towards the outer layer of the particle. It can improve the thermal stability and long life. Li et al. [29] employed a ternary cathode material of LiNi0.6Co0.2Mn0.2O2 with concentration gradient structure, which provides a nickel-rich core to deliver high capacity and a manganese-rich outer layer to provide enhanced stability and cycle life. 3
ACCEPTED MANUSCRIPT However, these previous studies basically drip the salt solution (A) with high manganese content into the salt solution (B) with high nickel content, and pump the solution (B) into the reaction kettle to synthesize the gradient material with linear
RI PT
change. It is not convenient to precisely control the synthesis of gradient structures. Moreover, because of this feeding method, only the gradient distribution of Ni and Mn can be realized, resulting in uniform distribution of Co in the synthesized material.
SC
In this study, we design a novel and convenient method to prepare
M AN U
LiNi0.8Co0.1Mn0.1O2 with full-gradient structure (LGNCMO) more accurately, that only separates the original ternary salt solution into two strands for feeding, which can accurately control the gradient synthesis. The schematic illustration of the synthesis of LGNCMO is shown in Scheme 1. Firstly, the full-gradient precursor is prepared by a
TE D
novel co-precipitation method, and then lithiation is carried out in combination with high temperature solid-phase method to synthesize the final product with full-gradient structure. The designed full-gradient material is a nickel-rich as the inner core
EP
material, and gradually transfers to the relatively low nickel and manganese-rich outer
AC C
surface, moreover, the Co content also presents a slow gradient change. The full-gradient material has improved high rate and high temperature cycle performance. In this experiment, the LiNi0.8Co0.1Mn0.1O2 intrinsic material (LNCMO) with uniform distribution of internal elements is prepared for comparison. 2. Experimental 2.1 Material synthesis Preparation of LGNCMO. The precursor of full-gradient material (GNCMO) 4
ACCEPTED MANUSCRIPT was prepared by a novel and convenient co-precipitation method, and then combined with high temperature solid-phase method for lithiation to prepare LGNCMO with full-gradient structure. The schematic diagram of the preparation process of GNCMO
RI PT
is shown in Fig. 1. In this experiment, the sulphate of nickel, cobalt and manganese was weighed according to the stoichiometric ratio of 8:1:1. NiSO4·6H2O and CoSO4·7H2O were dissolved in deionized water to prepare a salt solution of A (2
SC
mol/L), and MnSO4·H2O was dissolved in deionized water to prepare a salt solution
M AN U
of B (2 mol/L). At the same time, an aqueous solution of sodium hydroxide (4 mol/L) and an aqueous solution of aqueous ammonia (6.68 mol/L) were prepared as precipitating agent and complexing agent, respectively. The A solution, sodium hydroxide solution, and aqueous ammonia solution were simultaneously pumped into
TE D
the reaction vessel with an inert gas introduced and the B solution was pumped into the A solution. pH=11.50 was accurately controlled by adjusting the flow rate of sodium hydroxide solution, keeping the stirring speed at 800 rpm, and the whole
EP
reaction time was 36 h. In this process, the flow rate of B solution (~2.7 mL/h) was
AC C
controlled to drip out in about 35 hours, the flow rates of A solution and aqueous ammonia solution were 30 mL/h and 6 mL/h, respectively. At the same time, the flow rate of sodium hydroxide solution was adjusted with the fluctuation of pH value. As the co-precipitation proceeded, the concentration of Mn2+ in the feed increased gradually, and the relative concentrations of Ni2+ and Co2+ gradually decreased, and the precursor with gradient structure was synthesized. Added the salt solution of manganese to the salt solution of nickel and cobalt alone could regulate the gentle 5
ACCEPTED MANUSCRIPT gradient change more conveniently and accurately, and made the co-precipitation more stable. GNCMO prepared by this method had a slow and smooth change trend, which could be clearly observed by EDX analysis. Then final product (LGNCMO)
RI PT
was synthesized by mixed the precursor with lithium hydroxide (the mixing molar ratio is 1:1.05), firstly pre-calcined for 4 h at 500, then calcined at 800 for 12 h and then sintered with oxygen-assisted sintering, the flow rate of oxygen is 2.5 L/h.
SC
Preparation of the cross section of spherical secondary particles. In order to
M AN U
better detect the distribution of elements within the material, we dissected the secondary spherical particles prepared by co-precipitation method into hemispheres, and directly qualitatively and quantitatively analyzed the distribution of each element by EDX. The schematic diagram of the preparation of cross-section is shown in Fig. 2.
TE D
Mix epoxy resin A/B evenly, then took a small amount of sample powder and put it on the bottom of the mold, then filled the whole mold with the epoxy resin A/B. It was then aged in air, until the powder firmly embedded in the A/B glue. The surface of
EP
bottom sanded with the sample powder was polished with sandpaper and then placed
AC C
on a polishing machine for further processed. The cross section of spherical secondary particles was obtained which can be observed by SEM. 2.2 Material characteristics The samples were visualized by field emission scanning electron microscope
(FE-SEM, MERLIN Compact, Zeiss) to monitor the topography of surface. On this basis, Energy dispersive X-ray spectroscopy (EDX, OCTANE PLUS, EDAX) was introduced to qualitatively and quantitatively analyzed the distribution of elements. 6
ACCEPTED MANUSCRIPT X-ray diffraction (XRD, DX2700, Haoyuan) patterns was used to detect the crystal structure of materials, and the scan interval was 0.03º. Transmission electron microscopy (TEM, JEOL-2100, Japan) was carried out to observe the presence of
RI PT
lattice spacing and detecting its specific distance. The valence states of nickel were analyzed by X-ray photoelectron spectroscopy (XPS, ESCALAB 250XI, Thermo). The tap density of the sample was tested by a tap density tester (BT-301, Bettersize).
SC
2.3 Electrochemical measurements
M AN U
The final product was the active material, Super P was the conductive agent, and polyvinylidene fluoride (PVDF) was the binder, with a mass ratio of 8:1:1. First the PVDF powder was dissolved in an organic solvent, and prepared into a PVDF solution for standby. Then the active material, Super P and PVDF solution were
TE D
weighed into a bottle and stirred by magnetic force. The uniformly stirred slurry was applied to the surface of aluminum foil, then dried and cut into a positive electrode sheet of desired size. The 2025 coin cell was used, and the diaphragm was made of
EP
Celgard 2400, polypropylene micro diaphragm. The electrolyte was prepared by
AC C
dissolved 1.0 mol/L of LiPF6 in a mixture of dimethyl carbonate (DMC) + diethyl carbonate (DEC) + ethylene carbonate (EC) (volume ratio 1:1:1). In a glove box filled with inert gas (Argon, 99.999%), the coin half-cell was assembled. Battery monitoring system (Land CT2001A) was introduced to monitor the charge and discharge of coin batteries. The performance of the battery was evaluated in the voltage range of 2.75-4.3 V. Charge and discharge tests were carried at current rates of 0.1 C, 0.2 C, 0.5 C, 1.0 C, 2.0 C, 5.0 C and 10.0 C, respectively. The cyclic 7
ACCEPTED MANUSCRIPT voltammetry and impedance were tested by an electrochemical workstation (CHI 604e, Chenhua). The scanning speed of cyclic voltammetry test was 0.1 mV/s, and the potential interval was between 2.8 and 4.3 V. In the impedance test, the amplitude of
RI PT
disturbance voltage was 5 mV, and the set frequency range was 0.01 to 100 kHz. All the electrochemical impedance spectroscopy (EIS) of LNCMO and LGNCMO were conducted at open circuit voltage after complete discharge.
SC
3. Results and discussion
M AN U
Scanning electron microscope (SEM) images of the precursors of non-gradient intrinsic material (NCMO) and full-gradient material (GNCMO) are shown in Fig. 3a and 3c. From the macroscopic morphology of spherical particles, there is no significant difference between GNCMO and NCMO, both of which are smooth and
TE D
spherical secondary particles. Further magnifying the surface of the spherical precursors finds that the surface of GNCMO is denser and smoother than NCMO. According to the tap density measurement, the tap density of NCMO is 1.85 g/ml, and
EP
the tap density of GNCMO is 1.92 g/ml.
AC C
In order to better verify the distribution of the elements inside spherical secondary particles of precursor, we dissect the secondary spherical particles prepared by co-precipitation method into hemispheres, and the distribution of elements on cross section are subjected to linear scan of EDX. The scanning results are shown in Fig. 3b and 3d. The path of linear scan begins along one edge of hemisphere and passes through the center and continues to scan linearly to the other edge of hemisphere. It can be seen that the linear scan curves of the three elements of nickel, cobalt and 8
ACCEPTED MANUSCRIPT manganese in NCMO are relatively flat, indicating that the distribution of the three elements inside the primary particles of intrinsic material is relatively uniform. The linear scan curves of nickel, cobalt and manganese in GNCMO show obvious
RI PT
fluctuating trend. The full-gradient material is a nickel-rich material with high specific capacity as the inner core, and gradually transfers to the relatively low nickel and manganese-rich outer surface. The relative molar content of nickel gradually
SC
decreases from 88% to 72% along the center to the edge of hemisphere, while the
M AN U
relative molar content of manganese increases gradually. The relative molar content of cobalt fluctuates with the contents of nickel and manganese. Moreover, the contents of the three elements changes in a smooth arc. This gentle radian change is because the degree of gradient change in the co-precipitation process can be precisely
TE D
controlled. Gradient material with radian changes are smoother than linear changes, and the primary particles deposited in precursor are more compact and denser, which proves that the gradient material synthesized by this method is relatively compact.
EP
The NCMO and GNCMO are uniformly mixed with LiOH·H2O, placed in a
AC C
tubular resistance furnace for annealing treatment to obtain the final products, namely LNCMO and LGNCMO, respectively. The SEM images of final products are shown in Fig. 3e and 3g. It can be seen that the final products prepared by calcination still maintain the same regular spherical shape as the precursors. From the magnified images of the surface of final products, it is observed that the single crystal particles on the surface are coarser than the precursors and the distribution of each crystal grain is remarkable. There is no significant difference in the macroscopic morphology 9
ACCEPTED MANUSCRIPT between the two materials, and the single crystal particles on the surface are densely packed. According to the tap density measurement, the tap density of LNCMO is 2.26 g/ml, and the tap density of LGNCMO is 2.38 g/ml. The results show that the final
RI PT
product synthesized by the compacted precursor is more compacted. 5.0 g of LNCMO and LGNCMO are weighed separately, dissolved in 50 ml of deionized water, and magnetically stirred for 10 min. After the solutions
SC
are allowed to stand for clarification, the pH of the supernatant is accurately
M AN U
measured with a pH meter. The pH values of LNCMO and LGNCMO are measured to be 11.78 and 11.57, respectively. It indicates that the content of residual alkali on the surface of the relatively low-nickel, manganese-rich gradient material is lower than that of intrinsic material. The full-gradient
TE D
structure not only can effectively solve the problem of high residual alkali content on the surface of nickel-rich materials, but also retard the side reactions of LiOH and Li2CO3 with electrolyte, prevent the material from being attacked
EP
by HF and the formation of inactive "NiO" phase [30].
AC C
EDX linear scan is performed on the elemental distribution of the cross sections of LNCMO and LGNCMO, respectively. The linear scan results are shown in Fig. 3f and 3h. The path of linear scan begins along one edge of hemisphere and passes through the center, and then continues to scan linearly to the other edge of hemisphere. It can be seen that the tendency in elemental distribution of final products prepared by lithiation are consistent with the precursors, and the three elements inside LNCMO are relatively evenly distributed. Although the elemental distribution inside LGNCMO 10
ACCEPTED MANUSCRIPT maintains a full-gradient trend, the gradient is slightly slower than that of the precursor, which transforms from the nonlinear gradient of the precursor to a gentle linear gradient. The relative molar content of the central nickel element is reduced by
RI PT
84% and the edge is increased to 76%, the relative molar contents of cobalt and manganese also showed corresponding change. This phenomenon is caused by the thermal diffusion of ions during calcination, which results in the slowing down of
SC
concentration gradient.
M AN U
The uniformity of elemental distribution is further observed by surface scan of EDX. The surface scan is performed on the cross section and the complete spherical secondary particles of precursors (Fig. 4a-d). Comparing the EDX surface scan images of the hemispheres of NCMO and GNCMO, it is found that the three and the manganese
TE D
transition metal elements in NCMO are evenly distributed,
content in GNCMO gradually is increased from the center to the edge of hemisphere, and the cobalt is relatively reduced. The nickel element appears to be relatively evenly
EP
distributed, which is due to the material itself contains relatively high content of
AC C
nickel, so the difference of EDX surface scan is not obvious. Comparing the EDX surface scan images of the complete spherical secondary particles of NCMO and GNCMO, the distribution of nickel, cobalt and manganese on the outer surface of the two materials are relatively uniform, indicating that the elemental gradient of GNCMO is longitudinally changed from the center of the spherical secondary particle to the outer surface. That is, as the reaction proceeds, the three elements are uniformly grown on the core surface and continuously stacked to form uniform and dense 11
ACCEPTED MANUSCRIPT spherical secondary particle, which again proves that our method is accurate and controllable. The surface scan is also performed on the cross section and the complete
RI PT
spherical secondary particles of final products (Fig. 4e-h). The elements in the cross section of LNCMO are evenly distributed. The nickel element in the cross section of LGNCMO is uniformly distributed due to the high content. The gradient trends of
SC
cobalt and manganese are slightly slower than that of the precursor, which is
M AN U
consistent with the results obtained by linear scan of EDX. The elements on the surface of the complete spherical secondary particles of LNCMO and LGNCMO are evenly distributed. This indicates that the single crystal particles of final products still maintain the trend of radial growth of the precursors.
TE D
X-Ray diffraction (XRD) is introduced to analyze the structures of NCMO and GNCMO (Fig. 5a). The XRD diffraction peaks of the precursors of these two materials are coincided with the peaks of β-Ni(OH)2 (PDF card No. 14-0117) and
EP
none of them showed impurity peaks. However, the (001) peak of NCMO is
AC C
significantly lower than the peaks in GNCMO and that of β-Ni(OH)2. This may be due to the introduction of cobalt and manganese in NCMO and the uniform distribution of the three elements. The synergistic effect of the three transition metal elements affects the structure of (001) crystal plane. Although the spectrum of full-gradient material is closer to that of β-Ni(OH)2, its (001) peak is also slightly lower than β-Ni(OH)2, which may be due to the uneven distribution of the three elements. From the core to the outer surface of GNCMO, the material gradually 12
ACCEPTED MANUSCRIPT transitions from a nickel-rich core of substantially Ni(OH)2 to a ternary synergistic low-nickel outer layer. The XRD diffraction peaks of LNCMO and LGNCMO are coincided with the
RI PT
peaks of the standard card of LiNi0.8Co0.1Mn0.1O2 (Fig. 5b). None of the peaks of LNCMO shows impurity peaks. However, there is a significant small split at the bottom of the (003) peak in the XRD spectrum of LGNCMO, the splitting of the peak may be
SC
caused by the uneven distribution of internal elements in the gradient material. In
M AN U
addition, the splits of (108) and (110) peak of the two materials are obvious and the ratio of I(003)/I(104) are greater than 1.2, indicating that both have good layered structure [31-33]. The XRD spectroscopy of LNCMO and LGNCMO were further subjected to structural refinement (Table 1). According to the reference [34], when the ratio of
TE D
lattice parameter c/a > 4.9, the ratio can represent that the material has formed a layered structure. It can be seen from the results of structural refinement in Table 1 that a and c of the lattice parameters of LGNCMO are slightly increased relative to
EP
LNCMO. However, the ratio of c/a is equal to LNCMO and the value is
AC C
approximately 4.94, further indicating that both materials formed good layered structure. A good layered structure facilitates the deintercalation of Li+ and has good electrochemical performance. The R value indicates the fitting error, and the smaller the R value, the better the fitting is. Table 1. XRD structural refinement of LNCMO and LGNCMO sample
I(003)/I(104)
a(Å)
c(Å)
c/a
R
LNCMO
1.53
2.864
14.163
4.945
5.66 %
LGNCMO
1.51
2.865
14.161
4.943
5.32 %
13
ACCEPTED MANUSCRIPT The microstructure is observed by transmission electron microscopy (TEM). Fig. 5c and 5d are the TEM images of NCMO and GNCMO, respectively. The layered structure of the precursors of two materials are obvious and the layer spacing is about
RI PT
0.23 nm, which corresponds to the spacing of the (101) crystal plane in the XRD spectrum. In addition, due to the gentle variation of the full-gradient, there is no significant effect on crystal growth. Fig. 5e and 5f are the TEM images of LNCMO
SC
and LGNCMO, respectively. It can be seen that the lattice fringes of the final products
M AN U
of two materials are more obvious than the precursors and the layered structure are good, which indicates that the materials after lithiation have higher degree of crystallization. The interplanar spacing of the two materials is about 0.47 nm, which corresponds to the spacing of the (003) crystal planes in the XRD spectrum.
TE D
The valence state of nickel is analyzed by X-ray photoelectron spectroscopy (XPS), and Fig. 5g shows the full-spectrum scan curves of the surface of LNCMO and LGNCMO. The diffraction peaks of both materials are the same, indicating that the
EP
binding energy of each element of gradient material is in agreement with the intrinsic
AC C
material. The curve of peak-differentiating-imitating of nickel is shown in Fig. 5h, the peaks with binding energies of 854.4 eV, 857.1 eV and 855.5 eV correspond to Ni2+ and Ni3+, respectively [35]. Firstly, we fit the original XPS data to make the curve smoother and the separation peak more convenient. Then, the positions of Ni2+ and Ni3+ peaks are determined and add to the fitted spectra. After region fitting, the spectrum after peak separation is obtained. The distribution of the valence states of nickel is obtained by peak fitting (Table S1). The total content of Ni2+ and the total content of 14
ACCEPTED MANUSCRIPT Ni3+ in LGNCMO are close to that of LNCMO, indicating that the gentle change of the gradient of each element in LGNCMO does not significantly affect the content of Ni3+. However, the distribution of Ni2+ of gradient material is significantly different
RI PT
from that of intrinsic material. This is due to the gradient distribution of the three elements of nickel, cobalt and manganese in LGNCMO. The content of Mn4+ on the surface of LGNCMO is relatively more. In order to balance the valence state of cobalt
SC
and manganese, the distribution of Ni2+ and Ni3+ fluctuates with the gradient of the
M AN U
contents of the two elements [36].
The charge and discharge test curves of LNCMO and LGNCMO are shown in Fig. 6a-c. With a voltage window of 2.75-4.3 V and a current rate of 0.1 C, the first discharge specific capacities of LNCMO and LGNCMO are 198.7 mAh/g and 198.3
TE D
mAh/g, respectively. The discharge specific capacities at the current rate of 1 C are 179.6 mAh/g and 182.5 mAh/g, respectively. Further comparison of the discharge voltage platform, it can be found that at a current rate of 0.1 C and 1 C, the voltage
EP
platforms of LGNCMO are slightly higher than that of LNCMO, corresponding to a
AC C
higher discharge specific energy. It can be seen from the rate discharge curve that when the current rate is small, the discharge specific capacities of the two materials are relatively close. When the current rate is increased to more than 1 C, the specific capacity of LGNCMO is slightly higher than that of LNCMO, and the larger the current rate, the more obvious the difference, which shows excellent application prospect of power batteries. However, when the current rate is cycled from 10 C to 0.1 C, the specific capacity of LGNCMO is still higher than that of LNCMO, indicating that 15
ACCEPTED MANUSCRIPT LGNCMO has better cycle performance. In the cycle curves, the retention rate of discharge capacity of LGNCMO after 100 cycles at current rate of 1 C and 5 C were 98.8% and 93.7%, respectively, which is higher than 96.6% and 90.3% of LNCMO.
RI PT
This is because the full-gradient material has a relatively low nickel and manganese-rich surface layer that has better cycle performance and thermal stability relative to the intrinsic material [36].
SC
The full-gradient material has a flat compositional change compared to the
M AN U
core-shell material, avoiding structural collapse due to excessive component differences during charge and discharge [26-27]. The process after first charge and discharge is further analyzed by the differential capacity vs voltage plots (dQ/dv vs V) (Fig. 6d), which is similar to the cyclic voltammogram (CV) and has the same
TE D
characteristics as the CV curve (Fig. S1). The dQ/dv vs V curves of the two materials show three sets of redox peaks, corresponding to three phase transitions during charge and discharge, that is, the hexagonal (H1) ↔ the monoclinic (M) ↔ the hexagonal (H2)
EP
↔ the hexagonal (H3) [37,38]. At the current rate of 0.1 C, the phase transition of the
AC C
dQ/dv vs V curve of LGNCMO after the first charge and discharge is obvious than that of LNCMO. Phase transformation of H2-H3 is usually irreversible, which leads to volume expansion and cracks, and ultimately leads to a rapid decline in capacity during the cycle. But from the CV curve of LGNCMO (Fig. S1), it can be seen that the oxidation peak and reduction peak shape of H2-H3 are similar, and the peak shape area is basically the same. It shows that the phase transformation of H2-H3 is reversible for LGNCMO during charging and discharging process, and almost does not lead to 16
ACCEPTED MANUSCRIPT volume expansion. This may be attributed to the structural characteristics of the full gradient. In addition, the peaks of the two materials at around 3.7 V is significantly higher than the other two peaks, indicating that more Li+ are released and embedded
RI PT
around 3.7 V, also indicating that the charge and discharge platform are around 3.7 V. Further comparison of the peak potential difference between oxidation peak and reduction peak in CV curves shows that the peak potential difference of LGNCMO is
SC
relatively small, indicating that the polarization on electrode is smaller and the
M AN U
reversibility is better than LNCMO.
At the current density of 0.1 C, the battery is placed in a battery test system for one charge and discharge cycle, and then place on an electrochemical workstation for electrochemical impedance testing (Fig. 6e). The high frequency capacitive reactance
TE D
arc in the impedance spectrum corresponds to the migration process of Li+ in the SEI film. The medium frequency capacitive reactance arc corresponds to the transfer process of Li+ at the interface between the SEI film and the electrode active material,
EP
and the low-frequency line corresponds to the diffusion process of Li+ in the solid phase
AC C
[39-41]. According to this analysis, the equivalent circuit of electrode can be established. In the equivalent circuit of the impedance test, RΩ represents the ohmic resistance of electrode, including the ohmic resistance of the solution in the diaphragm and the ohmic resistance of the electrode itself. QSEI and RSEI represent the capacitance and resistance of the SEI film, respectively. Qd represents the electric double layer capacitance, Rct represents the charge transfer resistance, and W represents the solid phase diffusion impedance [42,43]. According to the results of the equivalent circuit 17
ACCEPTED MANUSCRIPT fitting, the parameters are obtained (Table 2). Although the RΩ and RSEI of LNCMO are similar to LGNCMO, its Rct are relatively large, which is disadvantageous for Li+ transfer and diffusion. The Rct of LGNCMO are obviously smaller than LNCMO,
RI PT
which corresponds to better electron transport ability and ion diffusion capacity, and the material has good electrochemical performance.
The lithium-ion diffusion coefficient can be calculated using the following
M AN U
SC
equation (1). [44]
In this formula, R is the gas constant (8.314 J K-1 mol-1), T is 298 K, A is the surface area of the electrode, n is the number of electrons involved in reaction, F is the
TE D
Faraday constant (96 485 C mol-1), C is the concentration of lithium ion, and σ is the Warburg coefficient, which can be obtained from the line of Z′~ω-1/2. can be obtained by taking the known data into formula (1) for calculation
EP
(Table 2). From the calculated results, it can be seen that the
value of LGNCMO
AC C
is larger than that of LNCMO, indicating that the gradient structure is beneficial to the diffusion of lithium ions. The
results of impedance in the full paper are all
calculated in this way.
Table 2. Impedance fitting parameters of LNCMO and LGNCMO
sample
RΩ/Ω
RSEI/Ω
Rct/Ω
LNCMO
4.61
35.62
108.07
2.592×10-17
LGNCMO
4.75
36.16
48.31
3.073×10-17
The impedance and cyclic voltammogram of the battery after 100 cycles at current 18
ACCEPTED MANUSCRIPT densities of 1C and 5C are tested. As shown in Fig. S2, the impedance spectra shows that the total impedance of the gradient material is less than that of intrinsic material after 100 cycles at current densities of 1C and 5C. According to the result of the
RI PT
equivalent circuit fitting, the parameters are obtained (Table 3). RSEI of LGNCMO is obviously smaller than that of LNCMO. The Rct of LGNCMO after 100 cycles at 1C rate is similar to that of LNCMO, and the Rct of LGNCMO after 100 cycles at 5C rate is
values of LGNCMO are larger than
M AN U
electrochemical reactions. In addition, the
SC
smaller than that of LNCMO. This is undoubtedly conducive to the rapid progress of
that of LNCMO, indicating that the gradient structure after 100th cycles is still beneficial to the diffusion of lithium ions. Further comparison of the peak potential difference between oxidation peak and reduction peak in CV curves shows that the
TE D
peak potential difference of LGNCMO is relatively small (Fig. S3), which means the polarization on electrode is smaller and the reversibility is better than LNCMO. In addition, the differential capacity vs voltage plots for the 5th, 25th, 50th, 75th and 100th
EP
cycles at 1C rate also given in Fig. S4. It can be seen that the peak position difference of
AC C
LGNCMO curve under different cycles are smaller than that of LNCMO, which indicates that LGNCMO has better reversibility. Moreover, the curve coincidence between LGNCMO cycles is better than LNCMO, and the symmetry of redox peaks is also good, which indicates that LGNCMO has better cycle stability. In order to analyze the structure stability of LNCMO and LGNCMO after electrochemical cycles, the XRD spectra (Fig. S5) and TEM images (Fig. S6) after 100th cycles at 5 C are analyzed. As can be seen from the XRD spectra, the peaks of 19
ACCEPTED MANUSCRIPT LNCMO and LGNCMO are consistent with the standard card, and the ratio of 003/104 still maintains the trend before electrochemical cycles, but the peak splitting of 006 and 012 is slightly less obvious. The splitting of the two peaks is not obvious,
RI PT
indicating that the order of the layered structure is somewhat damaged. Generally, it can be concluded from the overall XRD spectrum that there is no significant change in the structure of the two materials after cycles. From the TEM images after
SC
electrochemical cycles, it can be seen that obvious lattice spacing can still be observed
M AN U
between the two materials. But further observation of local magnification exhibits that LNCMO has slight lattice disorder, while LGNCMO performs well, which shows that the gradient structure has better stability.
Table 3. Impedance fitting parameters of LNCMO and LGNCMO after 100 cycles at 1C and 5C RΩ/Ω
RSEI/Ω
Rct/Ω
LNCMO-1C
4.66
27.94
21.39
2.315×10-16
LGNCMO-1C
4.26
10.91
21.66
4.103×10-16
LNCMO-5C
4.32
27.81
17.9
2.847×10-16
LGNCMO-5C
4.04
14.24
15.14
4.724×10-16
EP
TE D
sample
In order to verify the stability of the material in high temperature cycle, the battery
AC C
is charged and discharged at 50. In order to facilitate large rate comparison, we directly choose the current density of 5C for charge-discharge test. As can be seen from Fig. 7a, the initial capacity of LNCMO and LGNCMO are 165.3 mAh/g and 167.1 mAh/g, respectively, and higher than the initial capacity at room temperature. However, the curve of LNCMO declines sharply after more than 65 cycles, while the curve of LGNCMO is relatively stable, and the capacity retention rate of 100 cycles is as high as 90%, which is obviously higher than 71.8% of intrinsic material. This shows that the 20
ACCEPTED MANUSCRIPT gradient material has good high temperature cyclic stability. The impedance and cyclic voltammetry after 100 cycles are measured as shown in the Fig. 7b and 7c. It can be concluded that the impedance of LGNCMO is obviously smaller than that of LNCMO,
RI PT
and the Rct of LGNCMO are obviously smaller than that of LNCMO from the parameters of impedance fitting (Table 4), which shows that LGNCMO has good high-rate performance at high temperature. Meanwhile, the
value of LGNCMO is values at room
SC
still higher than that of LNCMO and higher than that of
M AN U
temperature. It is shown that the lithium ion diffusivity can be improved at higher temperature. From the cyclic voltammetry curves, it can be seen that the peak position difference of LGNCMO is significantly smaller than that of LNCMO, and the redox current value of LGNCMO is also significantly higher than that of LNCMO, which
capacity stability.
TE D
again proves that LGNCMO has good high temperature cyclic reversibility and
Table 4. Impedance fitting parameters of LNCMO and LGNCMO after 100 cycles at 5C (at 50) RΩ/Ω
EP
sample
RSEI/Ω
Rct/Ω
3.99
11.31
50.65
9.367×10-16
LGNCMO
3.71
12.36
30.04
1.212×10-15
AC C
LNCMO
In order to further analyze the effect of the full-gradient structure on the
performance, we present the cyclic curves of midpoint voltage and energy density of cathode material in the discharge process. Fig. 8a and 8b show the midpoint voltage curves at room temperature and high temperature respectively. It can be seen that the midpoint voltage of LGNCMO is obviously higher than that of LNCMO under different current densities and temperatures. More obviously, at the current density of 21
ACCEPTED MANUSCRIPT 5C, the midpoint voltage of LNCMO decreased more seriously than that of LGNCMO. In addition, at the high temperature of 50, the cyclic stability of the midpoint voltage of LGNCMO is significantly better than that of LNCMO. Further, we analyzed the
RI PT
energy density of cathode materials as shown in Figure 8c, d, and find that the energy density of LGNCMO is higher than that of LNCMO under different conditions. This
performance and high temperature cycling stability.
M AN U
4. Conclusions
SC
further indicates that the full-gradient structure can effectively improve the high rate
Full-gradient structured LiNi0.8Co0.1Mn0.1O2 has successfully prepared by a novel and facilitate method. SEM images and XRD spectra show that gradient material has regular spherical shape and good crystal structure, respectively. EDX test results shows
TE D
that a smooth transition from inner core with 84% nickel content to the outer surface with 76% nickel content of the final product is achieved. The Electrochemical tests show that the full-gradient material has significantly improved cycle performance and
EP
high rate performance, and the higher the current multiplier, the more obvious the
AC C
improvement. What’s more, full-gradient material exhibits good cyclic stability at high temperature, and the capacity retention rate of 100 cycles is as high as 90% at 5C rate, which is obviously higher than 71.8% of intrinsic material. In addition, it can be determined from impedance analysis that the gradient material has improved diffusion coefficient of lithium ion, indicating that it has good electrochemical performance. The gradient material also shows improved midpoint voltage and specific energy, which exhibits excellent application prospect of power batteries. 22
ACCEPTED MANUSCRIPT Acknowledgements The authors acknowledge funding support from the Shandong Provincial Natural Science Foundation of China (Grant No. ZR2017MB032), and the Shandong
References
RI PT
Provincial Major Project of China (Grant No. 2015ZDZX04002).
[1] B. Dunn, H. Kamath, J.-M. Tarascon, Electrical energy storage for the grid: A
SC
battery of choices, Science 334 (2011) 928-935.
M AN U
[2] C. Liu, F. Li, L.-P. Ma, H.-M. Cheng, Advanced materials for energy storage, Adv. Mater. 22 (2010) E28-E62.
[3] B. Scrosati, J. Garche, Lithium batteries: Status, prospects and future, J. Power Sources 195 (2010) 2419-2430.
TE D
[4] S.-T. Myung, F. Maglia, K.-J. Park, C. S. Yoon, P. Lamp, S.-J. Kim, Y. Sun, Nickel-rich layered cathode materials for automotive lithium-ion batteries: achievements and perspectives, Acs Energy Lett. 2 (2017) 196-223.
EP
[5] J. B. Goodenough, Y. Kim, Challenges for rechargeable Li batteries, Chem.
AC C
Mater. 22 (2010) 587-603. [6] V. Etacheri, R. Marom, R. Elazari, G. Salitra, D. Aurbach, Challenges in the development of advanced Li-ion batteries: A review, Energ. Environ. Sci. 4 (2011) 3243-3262.
[7] H.-J. Noh, S. Youn, C. S. Yoon, Y.-K. Sun, Comparison of the structural and electrochemical properties of layered Li[NixCoyMnz]O2 (x=1/3, 0,5, 0,6, 0,7,
23
ACCEPTED MANUSCRIPT 0,8 and 0,85) cathode material for lithium- ion batteries, J. Power Sources, 233 (2013) 121-130. [8] L.-J. Li, Z.-X. Wang, Q.-C. Liu, C. Ye, Z.-Y. Chen, L. Gong, Effect of
RI PT
chromium on the structural, surface chemistry and electrochemical of layered LiNi0.8-xCo0.1Mn0.1CrxO2, Electrochim. Acta 77 (2012) 89-96.
[9] P. Yue, Z. Wang, X. Li, X. Xiong, J. Wang, X. Wu, H. Guo, The enhanced
SC
electrochemical performance of LiNi0.6Co0.2Mn0.2O2 cathode materials by
M AN U
low temperature fluorine substitution, Electrochim. Acta 95 (2013) 112-118. [10]W. Luo, F. Zhou, X. Zhao, Z. Lu, X. Li, J. R. Dahn, Synthesis, characterization, and thermal stability of Li[Ni1/3Mn1/3Co1/3-z(MnMg)(z/2)]O2, Chem. Mater. 22 (2010) 5065-5073.
TE D
[11]S. W. Woo, S. T. Myung, H. Bang, D. W. Kim, Y. K. Sun, Improvement of electrochemical and thermal properties of Li[Ni0.8Co0.1Mn0.1]O2 positive electrode materials by multiple metal (Al, Mg) substitution, Electrochim.
EP
Acta 54 (2009) 3851-3856.
AC C
[12]M. Wang, Y. Chen, F. Wu, Y. Su, L. Chen, D. Wang, Characterization of yttrium substituted LiNi0.33Mn0.33Co0.33O2 cathode material for lithium secondary cells, Electrochim. Acta 55 (2010) 8815-8820.
[13]Y. Cho, P. Oh, J. Cho, A new type of protective surface layer for high-capacity Ni-based cathode materials: nanoscaled surface pillaring layer, Nano Lett. 13 (2013) 1145-1152.
24
ACCEPTED MANUSCRIPT [14]C.-H. Jo, D.-H. Cho, H.-J. Noh, H. Yashiro, Y.-K. Sun, S. T. Myung, An effective method to reduce residual lithium compounds on Ni-rich Li[Ni0.6Co0.2Mn0.2]O2 active material using a phosphoric acid derived Li3PO4
RI PT
nanolayer, Nano Res. 8 (2015) 1464-1479. [15]D. Mohanty, K. Dahlberg, D. M. King, L. A. David, A. S. Sefat, D. L. Wood, C. Daniel, S. Dhar, V. Mahajan, M. Lee, F. Albano, Modification of Ni-rich
SC
FCG NMC and NCA cathodes by atomic layer deposition: preventing surface
M AN U
phase transitions for high-voltage lithium-ion batteries, Sci. Rep. 6 (2016) 26532.
[16]J. Cho, T.-J. Kim, J. Kim, M. Noh, B. Park, Synthesis, thermal, and electrochemical properties of AlPO4-Coated LiNi0.8Co0.1Mn0.1O2 cathode
TE D
materials for a Li-ion cell, J. Electrochem. Soc. 151 (2004) A1899-A1904. [17]M. G. Kim, J. J. Cho, Air stable Al2O3-coated Li2NiO2 cathode additive as a surplus current consumer in a Li-ion cell, Mater. Chem. 18 (2008) 5880-5887.
EP
[18]H. Lee, Y. Kim, Y. S. Hong, Y. Kim, M. G. Kim; N. S. Shin, J. Cho,
AC C
Structural Characterization of the Surface-Modified LixNi0.9Co0.1O2 Cathode Materials by MPO4 Coating (M = Al , Ce, SrH, and Fe) for Li-Ion Cells, J. Electrochem. Soc. 153 (2006) A781-A786.
[19]X. Meng, X.-Q. Yang, X. Sun, Meng X , Yang X Q , Sun X . Emerging Applications of Atomic Layer Deposition for Lithium-Ion Battery Studies, Adv. Mater. 24 (2012) 3589-3615.
25
ACCEPTED MANUSCRIPT [20]H. Yang, H. Wu, M. Ge, L. Li, Y. Yuan, Q. Yao, J. Chen, L. Xia, J. Zheng, Z. Chen, J. Duan, K. Kisslinger, Q. Zhang, J. Lu, Simultaneously Dual Modification of Ni-rich Layered Oxide Cathode for High-energy Lithium-ion
RI PT
Batteries, Adv. Funct. Mater. (2019) 1808825. [21]L. Li, Z. Chen, Q. Zhang, M. Xu, X. Zhou, H. Zhu and K. Zhang, A hydrolysis-hydrothermal route for the synthesis of ultrathin LiAlO2-inlaid
M AN U
batteries, J. Mater. Chem. A 3 (2015) 894-904.
SC
LiNi0.5Co0.2Mn0.3O2 as a high-performance cathode material for lithium ion
[22]W. Liu, P. Oh, X. Liu, M. J. Lee, W. Cho, S. Chae, Y. Kim, J. Cho, Nickel-rich layered lithium transition-metal oxide for high-energy lithium-ion batteries, Angew. Chem. Int. Ed. Engl. 54 (2015) 4440-4457.
TE D
[23]K. Meng, Z. Wang, H. Guo, X. Li, D. Wang, Improving the cycling performance of LiNi0.8Co0.1Mn0.1O2 by surface coating with Li2TiO3, Electrochim. Acta 211 (2016) 822-831.
EP
[24]J. M. Lim, T. Hwang, D. Kim, M. S. Park, K. Cho, M. Cho, Intrinsic origins
AC C
of crack generation in ni-rich LiNi0.8Co0.1Mn0.1O2 layered oxide cathode material, Sci. Rep. 7 (2017) 39669.
[25]Y.-K. Sun, S.-T. Myung, M.-H. Kim, J. Prakash, K. Amine, Synthesis and characterization
of
Li[(Ni0.8Co0.1Mn0.1)0.8(Ni0.5Mn0.5)0.2]O2
with
the
microscale core-shell structure as the positive electrode material for lithium batteries, J. Am. Chem. Soc. 127 (2005) 13411-13418.
26
ACCEPTED MANUSCRIPT [26]Y.-K. Sun, S.-T. Myung, B.-C. Park, K. Amine, Synthesis of spherical nano-to microscale core-shell particles Li[(Ni0.8Co0.1Mn0.1)1-x(Ni0.5Mn0.5)x]O2 and their applications to lithium batteries, Chem. Mater. 18 (2006)
RI PT
5159-5163. [27]Y.-K. Sun, S.-T. Myung, H.-S. Shin, Y.-C. Bae, C. S. Yoon, Novel core-shell structured Li[(Ni0.8Co0.1Mn0.1)0.8(Ni0.5Mn0.5)0.2]O2 via coprecipitation as
SC
positive electrode material for lithium secondary batteries, J. Phys. Chem. B
M AN U
110 (2006) 6810-6815.
[28]Y.-K. Sun, Z. Chen, H.-J. Noh, D.-J. Lee, H.-G. Ren, Y. Jung, S. Wang, C. S. Yoon, S.-T. Myung, K. Amine, Nanostructured high-energy cathode materials for advanced lithium batteries, Nat. Mater. 11 (2012) 942-947.
TE D
[29]Y. Li, R. Xu, Y. Ren, J. Lu, H. Wu, L. Wang, D. J. Miler, Y. Sun, K. Amine, Z. Chen, Synthesis of full concentration gradient cathode studied by high energy X-ray diffraction, Nano Energy 19 (2016) 522-531.
EP
[30]K. Matsumoto, R. Kuzuo, K. Takeya, A. Yamanaka, Effects of CO2 in air on
AC C
Li deintercalation from LiNi1-x-yCoxAlyO2, J. Power Sources 81-82 (1999) 558-561.
[31]A. Rougier, P. Gravereau, C. Delmas, Optimization of the composition of the Li1-zNi1+zO2 electrode materials: structural, magnetic, and electrochemical
studies, J. Eletrochem. Soc. 143 (1996) 1168-1175.
27
ACCEPTED MANUSCRIPT [32]J. R. Dahn, U. V. Sacken, C. A. Michal, Structure and electrochemistry of Li1±yNiO2 and a new Li2NiO2 phase with the Ni(OH)2 structure, Solid State Ionics 44 (1990) 87-97.
RI PT
[33]J. Morales, C. Perez-Vicente, J. L. Tirado, Cation distribution and chemical deintercalation of Li1-xNi1+xO2, Mater. Res. Bull. 25 (1990) 623-630.
[34]J.-H. Kim, C. W. Park, Y.-K. Sun, Synthesis and electrochemical behavior of
SC
Li[Li0.1Ni0.35-x/2CoxMn0.55-x/2]O2 cathode materials, Solid State Ionics 163
M AN U
(2003) 43-49.
[35]J. Duan, G. Hu, Y. Cao, C. Tan, C. Wu, K. Du, Z. Peng, Enhanced electrochemical performance and storage property of LiNi0.815Co0.15Al0.035O2 via Al gradient doping, J. Power Sources 326 (2016) 322-332.
TE D
[36]Y. Wei, J. Zheng, S. Cui, X. Song, Y. Su, W. Deng, Z. Wu, X. Wang, W. Wang, M. Rao, Y. Lin, C. Wang, K. Amine, F. Pan, Kinetics tuning of Li-ion diffusion in layered Li(NixMnyCoz)O2, J. Am. Chem. Soc. 137 (2015)
EP
8364-8367.
AC C
[37]W. Li, J. N. Reimers, J. R. Dahn, In situ X-ray diffraction and electrochemical studies of Lil-xNiO2, Solid State Ionics 67 (1993) 123-130.
[38]J. Barker, R. Koksbang, M. Y. Saidi, An electrochemical investigation into the lithium insertion properties of LixNiO2 (0≤x≤1), Solid State Ionics 89
(1996) 25-35. [39]D. Aurbach, L. Mikhael, E. Levi, H. Teller, B. Markovsky, G. Salitra, U. Heider, L. Heider, Common electroanalytical behavior of Li intercalation 28
ACCEPTED MANUSCRIPT processes into graphite and transition metal oxides, J. Electrochem. Soc. 145 (1998) 3024-3034. [40]D. Aurbach, K. Gamolsky, B. Markovsky, G. Salitra, Y. Gofer, U. Heider, R.
RI PT
Oesten, M. Schmidt, The Study of surface phenomena related to electrochemical lithium intercalation into LixMOy host materials (M = Ni, Mn), J. Electrochem. Soc. 147 (2000) 1322-1331.
SC
[41]D. Aurbach, M. D. Levi, K. Gamulski, B. Markovsky, G. Salitra, E. Levi, U.
M AN U
Heider, L. Heider, R. Oesten, Capacity fading of LixMn2O4 spinel electrodes studied by XRD and electroanalytical techniques, J. Power Sources 81-82 (1999) 472-479.
[42]Q.-C. Zhuang, T. Wei, L.-L. Du, Y.-L. Cui, L. Fang, S.-G. Sun, An
TE D
electrochemical impedance spectroscopic study of the electronic and ionic transport properties of spinel LiMn2O4, J. Phys. Chem. C 114 (2010) 8614-8621.
EP
[43]G. Gi, Y. Ko, T.-H. Kim, H.-K. Song, S. Bin, S.-M. Park, Fourier transform
AC C
electrochemical impedance spectroscopic studies on LiFePO4 nanoparticles of hollow sphere secondary structures, J. Electrochem. Soc. (2013) 158 A1267-A1274.
[44]Z. Li, F. Du, X. Bie, D. Zhang, Y. Cai, X. Cui, C. Wang, G. Chen, Y. Wei, Electrochemical Kinetics of the Li[Li0.23Co0.3Mn0.47]O2 Cathode Material Studied by GITT and EIS, J. Phys. Chem. C 114 (2010) 22751-22757.
29
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT