Lithium batteries - Process - Effects of different clamps on the cycling performance of lithium batteries

The expansion force of lithium-ion batteries during cycling has an extremely important impact on their cycling performance. 

When conducting cyclic testing on batteries, in order to eliminate the negative effects caused by gas production or electrode expansion during charge and discharge cycles, 

clamps are usually used on both sides of the battery for pressure fixation. 

Different clamping plates and fixing methods have different effects on the cycling of batteries. 

Some pressurization methods not only do not improve the cycling life of the battery cells, 

but also cause negative effects such as lithium deposition, thereby reducing the service life of the battery.

This article takes the Lishen LP2714897-50Ah battery cell as the research object,

 studies the influence of using different test clamps on the cycling performance of the battery, 

and uses SEM, ICP, XRD and other analysis methods to study the negative electrode of the cyclic EOL from the perspectives of morphology, elements, structure, etc. 

The reason for the fast cycling decay caused by the use of ordinary aluminum clamps is analyzed.

 This result has guiding significance for improving the cycling performance of individual cells, enhancing the force uniformity of individual cells in the module, 

and thus improving the lifespan of the module and even the system.

1 Experimental section

1.1 Battery preparation

The raw materials used in this experiment are the materials used for the preparation of our company's commercial lithium-ion batteries.

Mix NCM ternary positive electrode, binder PVDF, and conductive agent in a mass ratio of 95:2:3, add N-methylpyrrolidone NMP,

 and homogenize, coat, roll, and cut according to our company's production process to obtain positive electrode sheets.

Mix artificial graphite, conductive agent, carboxymethyl cellulose sodium CMC, 

and styrene butadiene rubber SBR in a mass ratio of 96:1:1:2, using water as the solvent, and perform homogenization, coating, rolling, 

and cutting according to our company's production process to obtain negative electrode sheets.

Produce a square aluminum shell battery with a rated capacity of 50Ah and model LP2714897 using our company's production process for the positive and negative electrodes and separator.

 The electrolyte used is LiPF6 based electrolyte.

1.2 Battery Testing and Analysis

The ambient temperature cycling performance of the battery is tested using the American Arbin battery tester. 

The testing process is as follows: Charge at 1C constant current to 4.2V, 

then switch to constant voltage charging until 0.05C cut-off; Sleep for 30 minutes; 1C constant current discharge to 2.8V. 

During the cycling process, measure the DC internal resistance DCIR at 50% SOC state every 200 cycles and monitor the changes in DCIR.

The surface morphology of the polarizer was observed using a scanning electron microscope. Use X-ray diffractometer for material structure analysis. 

Inductively coupled plasma emission spectrometer is used for element content analysis.

2 Experimental Results and Discussion

Figure 1 shows the cyclic performance capacity retention curves of batteries using ordinary aluminum clamps (clamp 1) and spring clamps (clamp 2) at room temperature for 1C charging/1C discharging. 

Photos of different types of test clamp devices are shown in Figure 2.

From Figure 1, it can be seen that the capacity retention rates of the two types of clamp batteries are consistent in the first 300 cycles. After 300 cycles, 

the battery cycle using clamp 1 splits. After 2500 cycles, the capacity retention rate of the battery using clamp 1 is 84.95%, while the capacity retention rate of the battery using clamp 2 is 86.71%.

 Moreover, the DC internal resistance growth of the battery using clamp 2 is significantly lower than that of the battery using clamp 1.

In terms of DCIR growth, during the entire cycle life, the DCIR growth value of the battery using clamp 2 was lower than that tested using clamp 1, 

indicating that using a spring-loaded clamp can provide space for the expansion of the battery cell during cycling, release internal stress, and improve cycling performance.

The voltage and AC internal resistance of EOL batteries using clamp 1 and clamp 2 were tested. The test results are shown in Table 1.

 From the data in the table, it can be seen that the internal resistance of the battery using clamp 1 is much higher than that of the battery using clamp 2, 

which is consistent with the increase in DC internal resistance shown in Figure 1.

Disassemble and analyze the two batteries after cycling in the drying room. The disassembled photos are shown in Figure 3. 

Figure 3 (a) shows the residual electrolyte inside the disassembled battery case. 

It can be seen that there is no electrolyte residue at the bottom of the battery case using clamp plate 1, while there is still a small amount of electrolyte residue at the bottom of the battery case using clamp plate 2. 

The color of the electrolyte is relatively clear and transparent, indicating that during the cycling process,

 clamp plate 1 is more likely to lose more capacity and produce more side reaction products under stress conditions when it cannot release internal stress.

Figure 3 (b) shows a photo of the disassembled negative electrode plate.

 It can be seen from the figure that the surface color of the negative electrode plate of the battery using clamp plate 1 is uneven, with lighter gray colors on the upper and lower edges, and darker blue colors in the middle; 

The surface color of the negative electrode of the battery using clamp plate 2 is relatively consistent, and the electrode is in good condition. 

This is because during the cycling process, the battery undergoes swelling due to side reactions such as gas production and electrode expansion.

 Using ordinary aluminum clamps, the battery is compressed, and as more gas is produced, 

the compression force increases, resulting in greater force on the middle electrode and faster insertion and extraction of lithium ions, while the edge is relatively less stressed, 

resulting in slower insertion and extraction of lithium ions. Therefore, the middle color is darker and the edge color is lighter.

 By using a spring clamp, there is a buffer space after the battery expands, and the force is relatively uniform. 

The lithium ion insertion and extraction speed at the edges and middle is the same, so the surface color of the electrode is also relatively uniform.

In order to further analyze the morphology, structure, and elemental composition of the regions with different colors of polarizer A and B in Figure 3 (b).

 Therefore, SEM, ICP, and XRD characterization were carried out.

SEM images of regions A and B. From the figure, it can be seen that the material surface in the lighter colored edge A area is relatively smooth, forming a complete and uniform sheet shape, 

while the material in the darker colored middle B area is more fragmented. 

This is due to the generation of gas during the cycling process of the battery, causing swelling of the electrode plates and the battery. 

Using ordinary aluminum clamps, the battery is compressed, and as more gas is produced during the cycling process,

 the compression force increases, and there is no buffer space, resulting in mechanical stress tension on the positive and negative electrodes, structural collapse, and ultimately particle breakage.

This also indirectly illustrates that there is no residual liquid in the battery using clamp 1 in Figure 3,

 while there is residual liquid in the battery using clamp 2 due to excessive force on the battery using clamp 1, resulting in particle breakage and SEI film rupture, 

electrolyte infiltration, continuous consumption of electrolyte, and re film formation, 

leading to a significant increase in DC internal resistance of the battery using clamp 1 during the cycling process, as shown in Figure 1.

During the charging and discharging process of lithium-ion battery positive electrode materials, metal elements will dissolve out, 

and the dissolved metal elements will enter the negative electrode with the electrolyte and deposit on the surface of the negative electrode plate.

 Quantitative analysis of the content of transition metal Ni, Co, and Mn in the negative graphite of regions A and B of the negative electrode was carried out using inductively coupled plasma emission spectrometer (ICP).

 Table 2 shows the element content analysis of regions A and B. 

The transition metal elements Ni, Co, and Mn in regions A and B were all dissolved to varying degrees and accumulated in the negative graphite, 

with Mn having a larger amount of dissolution, followed by Ni, and Co having the least.

However, the leaching of Ni, Co, and Mn metal elements in the middle B region is higher than that in the edge A region, and the Mn element content in the middle B region is much higher than that in the edge A region. 

The leaching of Mn is more severe in the middle position, which may be due to the lattice distortion of the John Teller effect caused by increased pressure, leading to an increase in the leaching of Mn ions.

 This also explains why the blue color in the middle area of the negative electrode is caused by Mn leaching.

XRD testing and analysis were conducted on regions A and B, and compared with the XRD results of fresh electrode slices. 

The results are shown in Figure 5.

 From the XRD spectrum in Figure 5, it can be seen that the characteristic peak positions and peak intensities of the sample are completely consistent with the PDF card Graphite-2H, 

and no new phase is generated after cycling. The XRD diffraction peaks in regions A and B have been widened to varying degrees and the peak intensities have decreased.

 However, the diffraction peak intensity in the edge A region is significantly stronger than that in the middle B region, indicating that the material failure in the middle region is more severe compared to the edge region.

Using JADE software to calculate the negative electrode parameters of the tested XRD spectra, the calculation results are shown in Table 4. 

From the analysis results in Table 4, it can be seen that after cycling, 

the XRD diffraction angles of the electrode plates all shifted towards lower angles, and the diffraction angle in the middle B region shifted even more towards lower angles, resulting in a larger interlayer spacing. 

This proves that there is more lithium ion embedding in the middle position, which is manifested macroscopically as the detachment of graphite layers; At the same time, due to the high stress in the middle B region,

 the structure is damaged, resulting in a decrease in diffraction peak intensity and an increase in diffraction peak width. The unit cell parameters a and c both decrease, manifested macroscopically as the detachment of graphite flakes.

In order to verify the effect of the clamp on the morphology and cycling performance of the middle and edges of the positive electrode,

 we conducted SEM tests on fresh and positive electrode sheets after 1 cycle using the clamp, as shown in Figure 6. From the comparison in the figure,

 we found that after cycling, most of the small particles in the middle and edge areas showed cracks, and even some were completely crushed to micro powder. 

On the one hand, it will increase the self discharge of the battery, 

and on the other hand, it will reduce the effective contact area between particles, greatly affecting the conductivity and lithium ion transport, thereby increasing the impedance during cycling. 

The surface morphology of large particles is basically similar to that before cycling, because the internal tightness of secondary particles is not easy to break. 

To observe the internal morphology of large particles, Ar ion polishing is needed to observe whether there are cracks in the internal morphology.

Figure 7 shows the cross-sectional view of the positive electrode plate after Ar ion polishing, both fresh and after cycling with clamp plate 1. 

Observation shows that some small particles are broken before the cycle, but the secondary large particle structure is intact with a small amount of pores inside, which may be related to the synthesis process of the material;

 From the cross-sectional view after cycling, it can be observed that a large number of cracks have formed along the grain boundaries within the particles. 

The cracking degree in the middle area is significantly higher than that in the edge area, which indicates that the middle of the material is heavily stressed and lithium ions are de embedded quickly. 

The material is shrinking and expanding with the insertion and removal of lithium ions in the charge discharge process.

 Due to the anisotropy of the particles, the stress generated makes the internal grain boundaries of the secondary particles gradually obvious and produces tiny cracks. 

The electrolyte enters these cracks and acts on the surface of the cathode material to form a new interface facial mask. 

The interface reaction continuously intensifies the proliferation of the internal gaps, and these effects accumulate each other, eventually leading to the fragmentation of the material particles. 

The breakage of material particles can lead to the peeling of active materials or a decrease in electronic contact, resulting in an increase in battery polarization and a decrease in the content of effective active substances, 

thereby reducing the reversible capacity of the battery. This result corresponds to the inconsistent anatomical color of the negative electrode.

XRD analysis was performed on fresh positive electrode plates and positive electrode plates after cycling with clamp plate 1, as shown in Figure 8. 

From the figure, 

it can be seen that the characteristic peaks of the ternary layered material are basically consistent with the characteristic diffraction peaks of the material before and after cycling, indicating that

the structure is stable before and after cycling, and there is no formation of rock salt phase and spinel phase. 

Comparing before and after cycling, it was found that the diffraction peak (003) slightly shifted towards lower angles after cycling, indicating an increase in interlayer spacing; 

After cycling, the splitting degree of diffraction peaks (006)/(102) and (108)/(110) increased compared to before cycling, 

indicating that the positive electrode material has a good layered structure as the charge discharge cycle continues.

Table 5 shows the data calculated by JADE software from XRD characteristic diffraction peaks, including lattice parameters, unit cell volume, crystal size, and cation anion mixing degree.

 It was found that the lattice parameter a remained unchanged and c increased, indicating that the material underwent volume expansion along the c-axis; C (middle region)>c (edge region) indicates more severe expansion in the middle region,

 and a value of c/a (>4.899) represents a good layered structure; 

The intensity ratio I (003)/I (104) of the characteristic diffraction peaks of (003) and (104) crystal planes represents the degree of lithium nickel atom mixing in NCM811 material. 

The larger the ratio, the smaller the mixing degree, and the stronger the material's ability to remove lithium. 

A ratio below 1.2 indicates a high degree of cation mixing. After cycling, the I (003)/I (104) ratio decreases compared to before cycling, but both are greater than 1.2, indicating that lithium nickel mixing has almost not occurred;

 The size of (I (006)+I (102))/I (101) represents the quality of the hexagonal structure, and the smaller the ratio, the better the hexagonal structure. 

The grain size decreased from 64.4nm in fresh batteries to 50nm and 57nm, respectively, indicating a reduction in grain size. 

The size of the grain affects the length of the diffusion path of lithium ions. Small grains have fast diffusion, strong lithium ion transport ability, and high discharge capacity. 

However, small grains also mean that there are many interfaces in the material reaction process, and there are more side reactions after contact with the electrolyte, resulting in poorer cycling stability.

 As the number of cycles increases, the grain size decreases and the specific surface area increases, making it highly susceptible to side reactions with the electrolyte, resulting in a decrease in capacity.

3 Conclusion

This article takes LP2714897-50Ah NCM ternary/graphite square aluminum shell battery as the research object,

 and studies the influence of different test clamps on the battery's cycling performance. 

The capacity retention rates of the two types of clamp batteries were consistent in the first 300 cycles. 

After 300 cycles, the battery with ordinary aluminum clamp experienced splitting and accelerated cycle decay. 

After 2500 cycles, the capacity retention rate of the battery was 84.95%, 

while the capacity retention rate of the battery with spring clamp was 86.71%, indicating a significant improvement in cycle performance.

And from the perspectives of morphology, elements, and structure, the reason for the rapid cyclic decay caused by the use of ordinary aluminum clamps was analyzed.

 This is because the use of ordinary aluminum clamps cannot buffer the expansion force caused by cyclic gas production in the battery, resulting in uneven force distribution, large force in the middle, 

particle breakage, and dissolution of transition metal elements, ultimately leading to accelerated cyclic decay. 

By using spring clamps, the force on the battery during cycling is uniform, and the cycling performance is significantly improved.