Nickel-Manganese Oxide Cathode Electrode
4-1 Introduction
For the lithium rich layered cathodes, the phases gradually transform from the layerd to spinel structure during the repeated cycling have been verified and reported that is caused by the interaction of electrolyte and electrode[118, 119], the lattice suffers large strain and evantually leads to the breakdown of structure is the reason for the capacity and voltage fadding. Additionally, the electrolyte decompostion at the oxide surface at high cutoff potential is the factor to form the solid electrolyte interphase (SEI) on the cathode materials. It is well known that a normal SEI is formed by reaction of the surface electrode. Hence, quantities of research have been attempted to modify the surface of materials to either relieve the phase transformation or the thick SEI formation.
The related publications with the different methods, such as sol-gel methods[120], heat treatment[105, 121, 122], and atomic layer deposition[106, 123-126] have been proposed and revealed some good results for maintaining the cycle life for lithium ion batteries. In this chapter, a novel method of the surface modification on the cathode electrode was proposed by employing the plasma enhanced chemical vapor deposition (PECVD). Polytetrafluoroethene, also known as Teflon, was selected as the deposited compound owing to its outstanding properties of low chemical reactivity, anti-corrosion capability, and thermal stability, which are expected to against the interfacial side reaction between the liquid electrolyte and the active electrode.
On the other hand, it was reported that capacity fading may be intense at elevated temperature due to SEI growth and micro-crack growth at grain boundaries [127-130].
doi:10.6342/NTU201601970
58
In order to either study the thermal stability or accelerate the side reaction of electrode to exhibit the effect of the Teflon modification on the electrochemical performance, the elevated temperature test is necessary to perform by installing the cells at high environmental temperature. However, elevating the temperature might cause the onset of thermal runaway to occur. Therefore, for the safety issue and preventing the premature cell death, the optimized temperature for the elevated temperature test was also determined in this chapter.
doi:10.6342/NTU201601970
59
4-2 Temperature Determination for Elevated Temperature Test
To know the temperature effect on the cell performance, the cycling test of Li1.5Ni0.25Mn0.75O2.5 (LrMNO) electrodes were operated at varying temperatures from 40 to 55 oC in voltages of 2.0 – 4.6 V at 67.5 mA/g (0.3 C). Specific discharge capacity as a function of cycle number is shown in Fig. 4-1. The cell operated at 40 oC is shown in Fig. 4-1 (a). After 50 cycles, the retention of the discharge capacity of the cell is 95.41 % at 40 oC that could still maintain well,. Hence, the further temperature rising was directly in series and performed within 90 cycles (Fig. 4-1 (b)). From the cycling performance of the varying temperature, the retention of the terminated cycle (ca. every 30 cycles) at 45, 50 and 55 oC are 98.86 %, 94.59 %, and 86.35 % respectively. It is obvious that the capacity fading was gotten seriously with the increasing temperature (Fig. 4-1 (b)). To further confirm the capacity fading that is major caused by the elevated temperature, instead of the advanced cycle number. The cycle test with a new cell was directly operated at 55 oC, and the performance is shown in Fig. 4-1 (c). In this case, the retention was dropped to 91.33 % after 30 cycles, which is much lower than the case of 45 oC, 98.86 %. After 50 cycles, the 86.35 % retention was also lower than the case of 40 oC, 95.41 %, which implies the capacity fading was accelerated by the elevated temperature more than the advanced cycle number. Although the fast degradation of the capacity has occurred at the elevated temperature, the coulombic efficiency in the case of 55 oC could still achieve to 99.18 % after 50 cycles, which means the cell was still functioning well.
In order to know the potential fading towards the rising temperature, average potential as the function of cycle number at varying temperature was calculated and shown in Fig. 4-2. The result consistes with the capacity fading, that is, the potential
doi:10.6342/NTU201601970
60
fading was also accelerated by the elevated temperature. Additionally, the cell polarization was also investigated by differentiating the specific capacity of the voltage (dQ/dV curve), and the normalizing charge/discharge curve are shown in Fig. 4-3. From the dQ/dV plots, shown in Fig. 4-3, (a)－(c), the oxidation peak is a result of the lithium-ion extraction process while the reduction peak stands for the lithium-ion insertion process. The potential difference in reversible couples can be responded to the overpotential. Once the oxidation peak shifts to the higher potential and reduction peak shifts to the lower potential, the higher overpotential means the higher activation barrier for the electron transfering is then required to drive the electrochemical reaction. This intrinsic change in the potential between the reversible couples can be explained to the polarization of the electrode. The positions of redox potential at varying temperature were summarized in Table 4-1. The difference in the redox potentials at 45 oC are 0.064, 0.086, and 0.088 V in the 1st, 15th, and 30th cycle, respectively. However, in the case of 55 oC, the difference of the redox potential increases from 0.179 to 0.466V after 30 cycles. This apparent potential change with the rising temperature, which also can be observed from the normalized charge/discharge curves in Fig. 4-3 (d), implies a high degree of electrochemical polarization occurs because of the seriously interfacial side reaction of the electrode. And it illustrated that the potential polarization was significantly enlarged when the cell was operated at 55 oC.
Hence, from the above results, the fading and polarization effect on the cells could be enlarged with the rising temperature. However, even though the elevated temperature could accelerate the degradation of the electrode, the cells can still work functionally even at 55 oC without causing the short circuit or destructive damages.
Therefore, 55 oC would be selected as the temperature for the following elevated temperature test.
doi:10.6342/NTU201601970
61
Figure 4-1 Cycling performance of the LrMNO cells at (a) 40 oC; (b) varying temperatures from 45 to 55 oC; (c) 55 oC in 0.3 C (67.5 mA/g).
(a)
(b)
(c)
0 20 40 60 80
80 100 80 100 80 100
Cycle (number)
N o rmali zed d ischarge Capac it y (%)
55
oC
55
oC 50
oC
45
oC
40
oC
doi:10.6342/NTU201601970
62
Figure 4-2 Average potential as the function of cycle number of the LrMNO cells at varying temperatures from 45 to 55 oC.
10 20 30 40 50 60 70 80 90 100 3.2
3.4 3.6 3.8
45
oC
50
oC
55
oC
Average vo lt age vs. Li/Li
+(V)
Cycle number
doi:10.6342/NTU201601970
63
2.0 2.5 3.0 3.5 4.0 4.5
-300 0
300 45
oC
Diffe rential Capacity
E vs.(Li/Li
+)/V 1 st , 15 th , 30 th
2.0 2.5 3.0 3.5 4.0 4.5
-300 0 300
1 st , 15 th , 30 th
50
oC
Differential Capacity
E vs.(Li/Li
+)/V
(a)
(b)
doi:10.6342/NTU201601970
64
Figure 4-3 Differential capacity versus voltage (dQ/dV) plots for 1st, 15th, 30th cycles in 0.3 C (67.5 mA/g) at: (a) 45 oC; (b) 50 oC; (c) 55 oC. (d) Normalized Charge/discharge curves for every 1st, 15th, and 30th cycles at 45, 50 and 55 oC.
2.0 2.5 3.0 3.5 4.0 4.5
-300 0
300 55
oC
1 st , 15 th , 30 th
Diffe rential Capacity
E vs.(Li/Li
+)/V
(c)
0.0 0.5 1.0
2.0 2.5 3.0 3.5 4.0 4.5 5.0
45 oC 50 oC 55 oC
Vo ltage (V)
Normalized Capacity every 1
st, 15
th, 30
th(d)
doi:10.6342/NTU201601970
65
Table 4-1 Potential of the redox peaks in the 1st, 15th, and 30th cycle for the LrMNO cell operated at different temperature.
Environmental Temperature
1st cycle 15th cycle 30th cycle Oxid. Red. Delta Oxid. Red. Delta Oxid. Red. Delta 45 oC 3.788 3.724 0.064 3.787 3.701 0.086 3.788 3.700 0.088 50 oC 3.772 3.701 0.071 3.801 3.656 0.145 3.824 3.643 0.181 55 oC 3.818 3.639 0.179 3.886 3.582 0.304 3.950 3.484 0.466
doi:10.6342/NTU201601970
66
4-3 Characterization of PECVD Treated Lithium-rich Nickel-Manganese Oxide Cathode Electrode
The procedure of the PECVD process was described in section 3-3-1. In this chapter, three kinds of PECVD treated time, 3, 5, and 7minutes of the spherical Li1.5Ni0.25Mn0.75O2.5 (LrMNO) based electrodes have been studied and compared to the non-treated electrode. For these electrodes, hereafter was denoted as PTFE-3min, PTFE-5min, PTFE-7min, and Blank electrode, respectively.
The electrodes morphology in different PECVD treated time were observed by the scanning electron microscopy (SEM), as shown in Fig. 4-4 (a)－(h), and the energy dispersive X-ray spectroscopy (EDS) was used to confirm the deposition of Teflon compound on the electrodes by examining the atomic percentage of fluorine element, listed in Table 4-2. Comparing to the Blank electrode, there were no significant difference in surface morphology of the electrodes after PECVD treatment, indicating that PECVD treatment did not cause the damage to surface structure of electrodes. From the EDS analysis, for assuming the similar content of PVDF binder was added in every electrodes with the same electrode area, the increased fluorine content with the PECVD treated time can be explained to the thicker Teflon film has been deposited on the electrode.
doi:10.6342/NTU201601970
67
Figure 4-4 SEM micrographs of spherical LrMNO based (a)/(e) Blank; (b)/(f) PTFE-3min; (c)/(g) PTFE-5min; (d)/(h) PTFE-7min electrode in the magnification of 500/2000.
(a) (e)
(b) (f)
(c) (g)
(d) (h)
doi:10.6342/NTU201601970
68
Table 4-2 EDS analysis of the spherical LrMNO based electrode with different PECVD treated time.
PECVD treated time
C K Atomic %
O K Atomic %
F K Atomic %
Mn L Atomic %
Ni L Atomic %
0 min (Blank) 57.35 24.64 8.50 6.26 3.25
3 min 55.21 20.85 17.29 3.92 2.72
5 min 55.09 19.76 18.91 3.75 2.49
7 min 52.89 17.15 24.82 2.99 2.16
doi:10.6342/NTU201601970
69
4-4 Electrochemical Performance
In the electrochemical test, in the formation cycle, the first charge/discharge cycle was performed with a current density of 10 mA/g from 2.0 to 4.9 V and the following three charge/discharge cycles were performed under a current density of 20 mA/g from 2.0 to 4.6 V. Fig. 4-5 showed the comparison of the PECVD modified cells and non-modified cell in the formation cycles, and the resulting data is highlighted in the Table 4-3.
From the charge/discharge curve (Fig. 4-5 (a)), the first discharge capacity of the Blank electrode and PTFE-3, 5, 7 min electrodes are 246.09, 260.94, 240.15, and 211.49 mAh/g respectively. The lower discharge capacity in the PTFE-5, 7 min electrodes maybe because of the thicker insulated deposition of Teflon film, which slowed down the lithium-ion diffusion in the material, blocking the electron transfer. Hence, the shifted voltage plateau to the higher potential during charge and lower potential during discharge has occurred in the PTFE-7 min cell (Fig. 4-5 (a)). The polarization of the electrodes during the formation cycles might lead to non-complete electrochemical process. That is, the oxidation of nickel from Ni (II) to Ni (IV) at first inclined plateau and extraction of Li ions from Li2MnO3 to produce the electrochemical activated MnO2
above 4.5 V were encountered higher energy barrier, leading to the low capacity offering. However, the first coulombic efficiency of the PTFE-3, 5, 7 min electrodes are higher than the Blank electrode (80.48, 79.45, 78.94 %, 77.49 %) indicating there were less electrolyte decomposition occurred during the first charge process in the modified electrodes.
After formation cycles, AC electrochemical impedance of the cells was measured during charge cycle at 4.0 V. The Nyquist plots of the electrodes is shown in Fig. 4-6 (a). In generally, the impedance profile includes two semi-circles over the
high-to-doi:10.6342/NTU201601970
70
medium frequency range and an inclined straight line over the low-frequency range.
The first intercept on the real-part axis Zre may be attributed to the bulk electrolyte resistance, while the widths of the two semi-circles on the Zre axis are considered to be the surface-film (solid-electrolyte interphase, also known as SEI) resistance, Rs, and the overall charge-transfer resistance, Rct, respectively. The inclined line corresponds to the Warburg impedance, which accounts for Li-ion diffusion within the oxide particle. The equivalent circuit is shown in the inset in Fig. 4-6 (a), and the circuit parameters determined by simulation are summarized in Table 4-4.
From the fitting results for impedance profiles, Re value between the electrodes are approximate to each other, which means the bulk electrolyte offers the similar resistance to the cells. However, for the Rs resistance, the PECVD treated electrodes showed lower value than that of the Blank electrode. It might be attributed to the less SEI formation on the surface of the electrode after formation cycles. After then, the cells were installed into the oven set at 55 oC for the following elevated temperature test. Before the electrochemical test, the AC impedance of the sixth charge state was executed at 55 oC, and the profiles are shown in Fig. 4-6 (b). Compared to the impedance profiles of the fifth charge state measured at room temperature, the similar value of Rs was kept while the noticeable decreasing in Rct was observed at 55 oC. It was due to the kinetics for charge transfer was promoted by the elevated temperature, while the SEI still there provides the Rs resistance.
The cycling test was conducted at 55 oC under the current density of 67.5 mA/g (0.3 C) for 100 cycles. The discharge capacity and normalized capacity as the function of cycle number is shown in Fig. 4-7.The certain cycles of specific capacity and capacity retention are listed in Table 4-5. For the cycling test, except for the specific capacity, the cycle performance in all LrMNO electrodes have similar tendency before
doi:10.6342/NTU201601970
71
50 cycles. The capacity retention of the Blank electrode and PTFE-3, 5, 7 min electrode are 82.65, 86.52, 86.55, and 84.70 %, respectively. And the higher specific capacity in the PTFE electrodes may due to the thermal stability of the electrode interphases by the Teflon membrane. The divergent capacity retention between the Blank electrode and PTFE electrodes was started in the remaining 50 cycles. The final capacity retention after 100 cycles in the Blank electrode and PTFE-3, 5, 7 min electrode are 50.00, 62.90, 62.72, and 64.85 % respectively. The seriously capacity decay in the Blank cell was due to the polarization of the electrode. From the Fig. 4-8, the dQ/dV plots of the electrodes in every 50 cycles was demonstrated, and the locations of redox peaks are recorded in the Table 4-6. It is obvious that the shifted redox peaks with the advanced cycles is more intensive in the Blank electrode, indicating the more serious voltage polarization has occurred. For the LrMNO electrodes, It is well know that the SEI formation, phase transformation, and dissolution of transition metals from the active cathode material are the factors leading to the voltage polarization and capacity fading of the cells durring cycling. The comparison of the charge/discharge curves of electrodes are shown in Fig.
4-9 (a)－(c). It is clear to see that with the advanced cycles, the electrochemical plateau was significant decline in the Blank electrode more than that of in the PTFE electrodes during the charge/discharge process as shown in Fig. 4-9 (c). The impedance analysis was also conducted after cycling at 55 oC as shown in Fig. 4-9 (d). The PTFE-7 min electrode exhibits the smallest overall resistance (Re+Rs+Rct), which is responed with the best cycling performance in this electrode. In the other hand, the serious cell polarization of the Blank electrodes was because of the high cell resistance caused by the dramatic interfacial side reaction between the electrode and electrolyte, leading to the thick insulated SEI formation on the electrode and deteriorating the active material.
In the Teflon treated electrode, the Teflon membrane could act as the physical barrier,
doi:10.6342/NTU201601970
72
giving the capability to suppress the side reaction and enhance the thermal stability, consequently prolong the cycle life of the cells.
doi:10.6342/NTU201601970
doi:10.6342/NTU201601970
74
Figure 4-5 Comparison of the charge/discharge voltage profiles of the (a) first cycle;
(b) second cycle; (c) third cycle; (d) forth cycle in formation test at room temperature
0 50 100 150 200 250 300 350
doi:10.6342/NTU201601970
75
Table 4-3 Electrochemical performance of LrMNOelectrodes in formation cycles.
cell 1st Chargea (mAh/g)
1st Dischargea (mAh/g)
1st Coulombic efficiency (%)
2nd Dischargeb (mAh/g)
Blank electrode 317.58 246.09 77.49 208.20
PTFE-3min 324.22 260.94 80.48 206.64
PTFE-5min 302.27 240.15 79.45 198.86
PTFE-7min 267.91 211.49 78.94 181.08
a. between 2.0 and 4.9 V.
b. between 2.0 and 4.6 V.
doi:10.6342/NTU201601970
76
Figure 4-6 Nyquist plots performed at 4.0 V in the (a) fifth charge cycle at room temperature; (b) sixth cycle at 55 oC of the electrodes.
0 50 100 150 200
0 20 40
60 Blank electroed cell
PTFE-3min cell
doi:10.6342/NTU201601970
77
Table 4-4. Equivalent-circuit parameters of the LrMNO based cells in fifth charge state at room temperature.
cell Re (Ω) Rs (Ω) Rct (Ω) Rtotal (Ω)
Blank electrode 2.1 42.6 87.8 132.5
PTFE-3min 2.5 35.0 77.8 115.3
PTFE-5min 2.7 34.5 72.5 109.7
PTFE-7min 2.3 30.3 72.2 104.8
doi:10.6342/NTU201601970
78
Figure 4-7 Cycling performance in the (a) Discharge capacity ; (b) Normalized discharge capacity as the function of cycle number of LrNMO based cells charge/discharge in 0.3C (67.5mAh/g) at 55oC
10 20 30 40 50 60 70 80 90 100
50 75 100
Blank electroed cell PTFE-3min cell PTFE-5min cell PTFE-7min cell
Normalized Discharge Capacity (%)
Cycle number
10 20 30 40 50 60 70 80 90 100
100 150 200 250
Blank electroed cell PTFE-3min cell PTFE-5min cell PTFE-7min cell
Disc harg e Capacity (mAh/ g)
Cycle number
(a)
(b)
doi:10.6342/NTU201601970
79
Table 4-5 Electrochemical performance of LrMNOelectrodes in cycling test.
Cell
1st cycle 50th cycle 100th cycle Discharge
capacity (mAh/g)
Discharge capacity (mAh/g)
Capacity Retention
(%)
Discharge capacity (mAh/g)
Capacity Retention
(%) Blank electrode 229.69 189.84 82.65 114.84 50.00
PTFE-3min 234.766 203.13 86.52 147.66 62.90 PTFE-5min 233.71 202.27 86.55 146.59 62.72 PTFE-7min 222.97 188.85 84.70 144.59 64.85
doi:10.6342/NTU201601970
80
Figure 4-8 dQ/dV plots in 1st, 50th, and 100th cycle for the (a) Blank electrode, (b) PTFE-3min (c) PTFE-5min, and (d) PTFE-7min at 55 oC under the current density of 67.5 mA/g
doi:10.6342/NTU201601970
doi:10.6342/NTU201601970
82
Figure 4-9 Charge/discharge curves of LrMNO electrodes in (a) 1st cycle, (b) 50th cycle, and (c) 100th cycle.; (d) Nyquist plots performed at 4.0 V after 100 cycling at 55 oC of the electrodes.
doi:10.6342/NTU201601970
83