Compared with the traditional lithium battery wet process, the biggest feature of dry electrode technology is that it does not use any solvent, and has outstanding advantages such as environmental friendliness, high energy density, good electrical conductivity, and more suitable for pre-lithiation technology and solid-state battery technology.
In this paper, graphite dry electrode pieces were successfully prepared, the mechanism of electrode forming was elaborated, the influence of different binder dosages on the performance of the electrode pieces was investigated, and the electrochemical performance of dry electrode pieces and traditional wet electrode pieces was compared, which confirmed that the "binder network structure" of dry electrode pieces is conducive to reducing the polarization during the charging and discharging process of graphite anode, which is expected to improve the rate performance of graphite anode.
1 Experiment
All the graphite main materials used in the test are produced by BTR New Materials Group Co., Ltd. The formula of wet electrode tablets is graphite: conductive agent: CMC: SBR = 95.5: 1.5: 1.2: 1.8. The formula of dry electrode tablets changes with the change of binder dosage (2%~10%) (the amount of conductive agent is fixed at 1.5%).
Characterization tests include:
(1) Scanning electron microscopy test. (2) Buckle charge and discharge test: the dry or wet electrode piece is cut into a round electrode with a diameter of 8mm, and the lithium plate is used as the counter electrode and the reference electrode to assemble the CR-2016 button battery for testing. The blue electric test system is adopted, the setting current is 0.1 C, and the voltage range is 0.001~1.5V. (3) Cyclic voltammetry test: using strong transmission electrochemical workstation, scanning range 0~0.5V, scanning speed 0.05mV/s. (4) AC impedance test: using a strong electrochemical workstation, first adjust the state of charge of the electrode piece to 50% and then test, the test frequency is 0.01~105Hz, and the amplitude is 5mV.
2 Results and discussion
2.1 Preparation and forming mechanism of dry electrode piece
According to literature and patent reports, the preparation process of dry electrode pieces can be divided into electrostatic spraying method, hot compression molding method, calendering method, etc. In this work, the "calendering method" is used as the preparation method, the core material of the method is polyperfluoroethylene binder (PTFE) with "fibrosis" ability, in the actual preparation, the active material, conductive agent, PTFE are mixed, in the mixing process of various materials are uniformly dispersed, and at the same time under the action of shear force the binder will be physically stretched from the original spherical shape into fine filaments (that is, the so-called "fibrosis"), the formed network structure can connect the active particles to each other. So as to achieve better bonding. Then the mixed powder is formed by a calender, that is, a dry diaphragm is obtained, and then the diaphragm is pressed onto the copper foil to obtain a dry electrode piece.
Figure 1 (a) shows the SEM photo of PTFE binder, you can see that the particles are spherical, the diameter is 200~400nm, and a small number of uneven thickness filaments are also seen in the electron microscope field of view, indicating that a small number of PTFE particles have been fibrosized and stretched into fine filament, which may be caused by external forces such as collision and extrusion of PTFE powder during transportation, so be careful in the process of transferring and weighing PTFE to avoid fibrosis in advance due to shearing of the binder.
Figure 1(b) shows the SEM photo of graphite, conductive agent and PTFE after uniform mixing, it can be seen that PTFE particles are distributed around the graphite large particles relatively uniformly, and a certain number of filaments are found in the field of vision, connecting the graphite large particles to each other, indicating that PTFE has undergone a certain degree of fibrosis after mixing.
Figure 2 shows the appearance and related characteristics of graphite dry diaphragm and pole piece. The powder was calendered by calender, and a graphite dry diaphragm with a thickness of less than 100μm was obtained, as shown in Figure 2(a), it can be seen that the appearance of the diaphragm is relatively smooth and shows good flexibility, indicating that the degree of dispersion and fibrosis of the binder is better. Next, the diaphragm is compounded with copper foil to obtain a graphite dry electrode piece, as shown in Figure 2(b), it can be seen that the electrode piece also shows good flexibility. The SEM photo of the test pole piece, as shown in Figure 2(c), shows that the fibrotic filamentous PTFE is densely distributed in the field of view, forming a network structure that tightly connects the graphite particles together, and the amount of filament is much higher than the mixed powder in Figure 1(b), which indicates that the PTFE has undergone a higher degree of fibrosis after the powder is calendered.
It is worth mentioning that although the degree of fibrosis of PTFE after calendering is higher, the step that really determines the quality of the film (strength, flexibility) lies in the powder mixing step, and the uniform dispersion degree and fibrosis degree of PTFE after mixing are the key. Generally speaking, the mixed powder should have a certain viscosity, its texture and the powder before mixing has a clear difference, the higher the viscosity, the stronger its tableting ability, but too high viscosity will lead to powder agglomeration (plasticine-like), which is not conducive to the powder feeding and the uniformity of diaphragm strength during the calendering process. In general, the selection of mixing equipment and the optimization of process parameters are the key to dry electrode technology.
2.2 Effect of different binder dosage on electrochemical performance of dry electrode piece
In the early experiments, the influence of different binder content (2%~10%) on the performance of graphite electrodes was explored, and the first and most intuitive result was that the more binder used, the more the number of binder filaments in the electrode pieces, so the strength and flexibility of the electrode pieces were better. We also tested the charge and discharge performance of the electrode and lithium plates into a button half-cell, and Figure 2(d) shows the first cycle charge-discharge curves (0.1C) for a 2% PTFE and 10% PTFE electrode.
It can be seen that the discharge curve (lithium insertion curve) of the two electrodes is quite different, specifically, in the voltage range of 0.5~0.1V, the reduction potential of the 10% curve is significantly higher than the 2% curve, that is, there is an additional discharge platform, and the general graphite lithium embedded platform is less than 0.1V, that is, the additional discharge platform is not the lithium insertion platform of graphite, the above analysis shows that the 10% pole piece has obvious side reactions in the lithium insertion process. In fact, both electrodes have different degrees of side reactions, but the more binder is used, the greater the degree of side reactions, the more obvious the side reaction platform, these experimental results show that the binder participates in the occurrence of side reactions, and eventually leads to the decline of coulomb efficiency in the first week of graphite anode.
The buckling results showed that the first effect of the 2% diaphragm was 90.2%, while the first effect of the 10% diaphragm was 79.1%, the difference between the two was obvious, while the first effect of the graphite material in the wet electrode piece was about 92%, that is, the first effect of the two dry electrode pieces was lower than that of the wet electrode piece. For the phenomenon of side reactions of PTFE in the process of graphite lithium insertion, it has been discussed in the work of Chen Liquan's research group of the Institute of Physics of the Chinese Academy of Sciences in the 90s of last century, and they believe that the reaction mechanism is as follows:
According to the above formula, PTFE will irreversibly react with active lithium to generate amorphous carbon and lithium fluoride during the process of graphite lithium insertion, resulting in the loss of active lithium, obviously with the increase of PTFE consumption, the more electricity consumed by the reaction, the more obvious the side reaction platform, and our experimental results are also consistent with this conclusion. Therefore, in the preparation of graphite dry electrode pieces, the amount of PTFE should be reduced as much as possible under the premise of ensuring the strength of the pole piece, so as to reduce the impact of PTFE side reactions on battery performance, but the reduction of the amount of binder will put forward higher requirements for the mixing process and calendering process, and the preparation of the electrode piece is more difficult, and there is a balance point between the two.
2.3 Comparison of electrochemical performance of dry electrode pieces and traditional wet electrode pieces
Figure 3 compares the first cycle charge-discharge curves of dry and conventional wet electrode pieces in a button half-cell. Observing the lithium insertion curve of the two, it can be found that the potential of the dry electrode piece has always been higher than that of the wet pole piece in the voltage range of 0.15~1.5V, which once again confirms the side reaction phenomenon of PTFE; When the voltage < 0.15V, the lithium insertion curves of the two tend to be similar but do not completely coincide, and the lithium insertion capacity of the dry electrode piece is larger, indicating that the graphite lithium intercalation reaction mainly occurs in this voltage range, but at the same time, the dry pole piece is also accompanied by a small amount of PTFE side reaction. Looking at the charging curve of the two, it can be found that the two almost completely coincide, which indicates that the specific capacity of the two electrodes is almost equal, because when the greatly excess lithium sheet is used as the opposite electrode, the greatly excessive lithium source can make up for the loss of active lithium caused by the PTFE side reaction in the dry electrode piece and make the graphite in a complete lithium insertion state, so it can also get rid of the same number of lithium ions as the wet electrode sheet when the delition reaction occurs.
The reasons for the coulomb efficiency of 88.7% and 92.0% for dry and wet electrode pieces have been elaborated in Section 2.2 and will not be repeated here. In addition, in the actual full battery, the counter electrode (positive electrode) used is not a large excess of lithium sheets, but a variety of cathode materials with limited lithium sources, such as ternary materials, lithium cobalt oxide, lithium iron phosphate, etc., so the PTFE side reaction in the first charge reaction will not only lead to a decrease in the efficiency of Coulombs in the first week, but also lead to the loss of reversible charge and discharge capacity in the whole battery, so that the energy density of the battery decreases.
In order to further explore the difference between dry and wet electrode sheets, other electrochemical characterization of the two was performed. Figure 4(a) compares the cyclic voltammetry curves of the two polar pieces in the second week, and the reason for the comparison is that the PTFE side reaction has almost been completed in the first cycle, so the second week curve basically only reflects the electrochemical behavior of graphite intercalation and lithium removal. It can be seen from Figure 4(a) that three pairs of obvious redox peaks appear in both polar pieces, reflecting the different stages of graphite intercalation and delithium, and a significant feature can also be observed from this figure, that is, the peak spacing of the three pairs of redox peaks of the dry electrode piece is smaller than that of the wet electrode piece, which indicates that the electrochemical polarization of the graphite intercalation and delithiumization reaction in the dry electrode piece is smaller, that is, the reaction is more reversible. To further prove this conclusion, the two electrodes were assembled with the lithium plates to form a button half-cell, and the half-cell AC impedance curve (EIS) of the two electrodes at 50% SOC was tested.
Figure 4(b) is an analog equivalent circuit diagram, where the series resistance Rrs represents the ohmic internal resistance of the entire buckle, the two RC parallel circuits correspond to the SEI film impedance (Rf) and charge transfer resistance (Rct) of the graphite anode, and Zw represents the Warburg impedance related to mass transfer. Observing the EIS curves of the two, it can be found that two semicircles appear in both curves, and it is generally believed that the mid-high frequency band semicircle corresponds to Rf, and the middle and low frequency band semicircle corresponds to Rct.
First of all, it can be seen that the diameter of the low and medium frequency semicircle of the dry pole piece is significantly smaller than that of the wet pole piece, that is to say, the charge transfer resistance Rct of the dry pole piece is smaller than that of the wet pole piece, and by fitting the two curves, it can be seen that the Rct of the dry pole piece is 25.35Ω, and the wet pole piece is 31.70Ω, the former is only 80% of the latter, The difference is as high as 6.25 Ω, while the Rf of the dry and wet methods is 4.058 and 2.948 Ω, respectively, the difference is only 1.1 Ω, and the Rs are 1.581 and 1.183 Ω, respectively. From the comparison of these resistance values, it can be seen that the difference between the two electrodes mainly lies in the charge transfer resistance Rct, and the size of Rct means the speed of the electrochemical reaction kinetics, the smaller the Rct, the faster the reaction kinetics, so from the EIS results, it can be seen that the graphite intercalation and delithium reaction kinetics of the dry electrode sheet are faster than the wet electrode piece, which proves the conclusion of Figure 4(a).
However, the essential cause of the difference in electrochemical polarization between the two could not be found through electrochemical characterization, so SEM characterization was carried out on the two polar pieces. Figure 4 (c) and (d) are SEM photos of dry electrode piece and wet electrode piece, respectively, in which the morphology of the dry electrode piece is similar to Figure 2 (c), which also presents a network morphology of fibrotic PTFE connecting graphite particles together, while the morphology of the wet electrode piece is very different from the former, because the binder used in the wet electrode piece is CMC and SBR without fibrosis, so the wet slurry is coated on the current collector and dried, the binder will be coated on the surface of the graphite particles, as shown in Figure 4 (d), The graphite particles exhibit an uneven and irregular interface, which is the result of the binder coating the graphite particles.
Based on the SEM results, the following conjectures are made: for wet electrode sheets, because the graphite particles are generally coated with binder, the conductivity of the material surface is worse [the "shadow" phenomenon in the SEM photo in Figure 4(d) also indicates that the conductivity of the electrode piece is not good], which makes the lithium ion transport at the graphite/electrolyte interface more difficult, and the final result is that the electrochemical polarization of the graphite intercalation of lithium in the wet electrode sheet is greater; For dry pole pieces, because the binder is distributed in a network between graphite particles, the surface of most graphite particles is exposed to electrolyte, lithium ions can be unhindered on the surface of graphite for intercalation reaction, and the electronic conductivity between graphite particles is also better, so the electrochemical polarization of dry pole pieces is smaller. Based on the above experimental results and conjectures, we have reason to believe that the rate performance of the dry electrode piece will be better than that of the wet electrode piece, and in the next experiment, it is planned to assemble the dry electrode piece and the wet electrode piece into a full battery for verification.
3 Conclusion
In this paper, graphite dry electrode pieces were successfully prepared, and it was pointed out that the "fibrosis" behavior of the binder was the key to the forming of the electrode sheets. The results show that the PTFE binder will have a side reaction during the process of lithium insertion, resulting in the decrease of efficiency and irreversible capacity loss in the first week of the battery, so it is necessary to reduce the amount of PTFE as much as possible, but at the same time, the strength and flexibility of the pole piece should be considered to find the balance point of PTFE dosage. At present, graphite electrodes with a PTFE dosage of 2% can be stably prepared and assembled into lithium-ion batteries with good performance, and the amount of binder can be further reduced in the future. The results of cyclic voltammetry and AC impedance test showed that the dry electrode piece had faster reaction kinetics than the wet electrode piece, and the SEM results showed that the network structure of the binder in the dry electrode piece may be the key to reducing polarization, so it was speculated that the dry electrode piece had more advantages in rate performance.
References: OUYANG Jiaxing, LI Shoutao, WU Zixia, et al. Preparation and Properties of Graphite Dry Electrode Sheets for Lithium-ion Batteries[J]. Chinese Journal of Power Sources, 2022(009):046.