Practical Evaluation of Li-Ion Batteries

Rechargeable batteries are key technology for developing many emerging applications. Thousands of academic papers have been published on this topic. It is quite often that the conclusions and claims are overstated, partially due to significant deviation of lab testing conditions from practical battery design. This paper points out common problems that could mislead the evaluation of the new materials and new devices. Rechargeable batteries are key technology for developing many emerging applications. Thousands of academic papers have been published on this topic. It is quite often that the conclusions and claims are overstated, partially due to significant deviation of lab testing conditions from practical battery design. This paper points out common problems that could mislead the evaluation of the new materials and new devices. Rechargeable batteries are a key technology for developing many emerging applications and have attracted wide attention. In 2018, 11,583 academic papers were published with the keywords “lithium” and “batteries” based on a search of the Web of Science core collection. It is quite often that the conclusions and claims in papers are overstated, due in part to unknown performances of current industry products, significant deviation of lab testing conditions from practical battery design, and boasting. The average increasing rate of energy density of Li-ion batteries is less than 3% in the last 25 years, and it is only becoming more sluggish. From a historical viewpoint, the energy density has never increased suddenly due to complicated system design and requirements on well-balanced performances for application. Creating a record in a single performance does not promise that the new battery could be commercialized in the short term. Researchers should be aware of the complicity of developing batteries. After 28 years of effort from many scientists and engineers, the energy density of 300 Wh/kg has been achieved for power batteries and 730–750 Wh/L for 3C devices from an initial 90 Wh/kg. We could read the claims frequently that the energy density of a new device could be 2–10 times higher than that of current Li-ion batteries—that means 600–3,000 Wh/kg or 1,460–7,500 Wh/L. These values are highly desired but obviously very difficult to be realized. Except for artificial exaggerating, lack of standard testing protocol for lab studying leads to overstatement. This is more significant for developing new-generation lithium batteries using metal lithium anode. Similar problems also appear frequently in Li-ion batteries. It is necessary that the researchers be aware of technological parameters of practical battery and standard testing protocol. Recently, Lin et al. have pointed out that reporting the performance based on a limited number of metrics does not give a realistic picture of the battery performance required by practical use.1Lin Z. Liu T. Ai X. Liang C. Aligning academia and industry for unified battery performance metrics.Nat. Commun. 2018; 9: 5262Crossref PubMed Scopus (189) Google Scholar In particular, many papers do not care about Coulombic efficiency (CE). CE of 99.96% is required for cycling stability up to 500 cycles for commercialization.1Lin Z. Liu T. Ai X. Liang C. Aligning academia and industry for unified battery performance metrics.Nat. Commun. 2018; 9: 5262Crossref PubMed Scopus (189) Google Scholar Recently in Joule, Chen et al.2Chen S. Niu C. Lee H. Li Q. Yu L. Xu W. Zhang J.-G. Dufek E.J. Whittingham M.S. Meng S. et al.Critical Parameters for Evaluating Coin Cells and Pouch Cells of Rechargeable Li-Metal Batteries.Joule. 2019; 3 (this issue): 1094-1105Abstract Full Text Full Text PDF Scopus (249) Google Scholar describe a set of coin cell parameters and testing conditions systems based on 300 Wh/kg pouch cell level requirements. It is helpful to expedite the discovery of new materials and their full integration into a realistic battery. As mentioned in this paper, the testing conditions in many papers for button cell testing are not suitable to evaluate new materials and new batteries.2Chen S. Niu C. Lee H. Li Q. Yu L. Xu W. Zhang J.-G. Dufek E.J. Whittingham M.S. Meng S. et al.Critical Parameters for Evaluating Coin Cells and Pouch Cells of Rechargeable Li-Metal Batteries.Joule. 2019; 3 (this issue): 1094-1105Abstract Full Text Full Text PDF Scopus (249) Google Scholar However, such knowledge is not known clearly by many authors. We fabricated a batch of pouch cells using lithium nickel cobalt aluminum layered oxide as cathode and a 50 μm Li foil as anode; the data parameters are listed in Table 1. Compared to the parameter shown in the paper by Chen et al.,2Chen S. Niu C. Lee H. Li Q. Yu L. Xu W. Zhang J.-G. Dufek E.J. Whittingham M.S. Meng S. et al.Critical Parameters for Evaluating Coin Cells and Pouch Cells of Rechargeable Li-Metal Batteries.Joule. 2019; 3 (this issue): 1094-1105Abstract Full Text Full Text PDF Scopus (249) Google Scholar it can be seen that the design parameters for most of the materials are similar, except for the quantity of injected electrolyte and selection of the electrolyte and cathode. In our case, initial injected liquid electrolyte is 1.86 g/Ah, comparable with that in Li-ion batteries. After performing an in situ solidification and use of new electrolyte design, the cell can cycle over 100 cycles with a capacity retention over 90%, as shown in Figure 1. The motivation of in situ solidification is to decrease the continuous reaction of metal lithium with liquid electrolyte, which is helpful to decrease the weight ratio of the electrolyte. The data shown here are mainly to verify and confirm the design parameter in the paper by Chen et al.2Chen S. Niu C. Lee H. Li Q. Yu L. Xu W. Zhang J.-G. Dufek E.J. Whittingham M.S. Meng S. et al.Critical Parameters for Evaluating Coin Cells and Pouch Cells of Rechargeable Li-Metal Batteries.Joule. 2019; 3 (this issue): 1094-1105Abstract Full Text Full Text PDF Scopus (249) Google Scholar Developing a practical rechargeable lithium battery needs comprehensive optimization on cycle life, rate, and safety, simultaneously. The balance of the performances could lead to the decrease of the energy density shown in Table 1 and will be discussed elsewhere.Table 1Technological Parameters of a Li/NCA Pouch Cell with an Energy Density of 385 Wh/kgComponent of CellParameterReal ValueCathodeMaterialsNCAReversible Capacity (mAh/g)205Unilateral areal density (g/m2 each side of Al)230Active material ratio (%)96.10Press density (g/cm3)3.5Unilateral areal capacity (mAh/cm2)4.5Unilateral thickness (μm)65Thickness of Al12Number of cathode sheet16AnodeMaterialsLiUnilateral areal density (mAh/cm2)10N/P ratio2Unilateral thickness (μm)50Thickness of Cu (μm, 27% porosity)8ElectrolyteInjection mass (g)8.6SeparatorThickness (μm)16Thickness of unilateral coating layer (μm)4Sealing filmThickness (μm)113TabSpecifications (mm)25*8*0.2CellVoltage (V)3.7Capacity (Ah)4.3Mass (g)41.3Volume (L)0.0198Gravimetric energy density (Wh/kg)385Volumetric energy density (Wh/L)802 Open table in a new tab As pointed out by Chen et al. in this issue of Joule and Cao et al. in Nature Nanotechnology, quantity of electrolyte, area capacity of cathode and anode, N/P ratio, and rate are very sensitive parameters on tested performances in button cell.2Chen S. Niu C. Lee H. Li Q. Yu L. Xu W. Zhang J.-G. Dufek E.J. Whittingham M.S. Meng S. et al.Critical Parameters for Evaluating Coin Cells and Pouch Cells of Rechargeable Li-Metal Batteries.Joule. 2019; 3 (this issue): 1094-1105Abstract Full Text Full Text PDF Scopus (249) Google Scholar, 3Cao Y. Li M. Lu J. Liu J. Amine K. Bridging the academic and industrial metrics for next-generation practical batteries.Nat. Nanotechnol. 2019; 14: 200-207Crossref PubMed Scopus (289) Google Scholar Besides technological parameters, we have summarized that 10 possible factors could lead to deviation, errors, or low reproducibility for button cell measurement,4Wang Q.Y. Chu G. Zhang J.N. Wang Y. Zhou G. Nie K.H. Zheng J.Y. Yu X.Q. Li H. The assembly, charge-discharge performance measurement and data analysis of lithium-ion button cell.Energy Storage Sci. Tech. 2018; 7: 327-340Google Scholar including material preparation, weighing, grinding, mixing, coating, unexpected short circuit, button cell fabrication, measuring instrument, environment control, and experimental design. Therefore, basic electrochemical performance testing using button cell should be very careful to achieve reliable, reproducible, and valuable data. Aside from the above technological parameters and experimental design, the following problems also exist and could mislead the evaluation of the new materials and devices:(1)Voltage range. When a new anode material is evaluated in the button cell using metal lithium as anode in the half cell, it is quite common that the cut-off voltage for charging is set above 2.0 V or even 3.0 V versus Li+/Li. This will lead to high initial CE and high reversible delithiation (charging) capacity. However, based on our research, only the voltage range of 0 to 0.8 V is meaningful for the full Li-ion cell design. Thus, we recommend testing in the range of 0 to 0.8 V for most anode materials. For fundamental understanding on the maximum delithiation capacity or research on high-voltage anode, such as Li4Ti5O12 or TiNb2O7, the voltage range can be extended as wide as 0–3.0 V. The researchers should be aware that high delithiation voltage for anode leads to low discharge voltage in the full cell. For the cathode evaluation, the cut-off voltage for discharging is normally above 2.9 V. Achieving high discharging capacity in a wide voltage range—in some papers, down to 1.0 V—is meaningful for the fundamental research but may not be valuable for cathode application.(2)Rate performance. High rate performance is always desirable and reported in many papers. It is highly dependent on the parameters of the material and electrode. Small particles and a thin electrode layer will be favorable to achieve a high rate performance. However, it could lead to low CE or low gravimetric energy density. Moreover, it should be known that the area capacity in commercial batteries is about 3–4 mAh/cm2 for energy-type cell. Therefore, 1C rate should be measured at a current density of 3–4 mA/cm2, while 3C rate should be measured at 9–12 mA/cm2. In literature, area capacity is much lower than this value, which will lead to apparent high rate performance. Some authors publish data using mA/g, or even A/g. It is suggested to provide the area current density mA/cm2 simultaneously since very low active material loading (mg/cm2) could lead to apparent very high mA/g, which is not very practical. It is not surprising that some authors claim C-rate performance over 100 C, corresponding to a 36 s discharging or charging. Presuming the area capacity is 3 mAh/cm2, 100 C means 300 mA/cm2, which is much higher than 10 mA/cm2 in most batteries. In the full cell, such a level of current density could cause thermal runaway. In addition, when the researchers claim high rate performance, it is also quite common to see that the capacity retention at high rate is already very low. When rate performance is claimed, the highest rate should correspond to 80% capacity retention. For example, if the capacity can maintain 80% at 3C, we can claim that this cell can discharge and charge at 3C.(3)Coulombic efficiency. Lin et al. have clearly calculated how the CE in each cycle influences the cycle life.1Lin Z. Liu T. Ai X. Liang C. Aligning academia and industry for unified battery performance metrics.Nat. Commun. 2018; 9: 5262Crossref PubMed Scopus (189) Google Scholar In the full cell of Li-ion batteries, all active lithium is provided from the cathode, and total capacity loss determines the cycle life of the full cell and the real energy density. For graphite and Li4Ti5O12 anode, the CE after the first cycle is approaching 99.9%. For many new high-capacity anode materials, CE may approach 99.5% only after 10 or even more cycles. Therefore, total irreversible capacity loss should be calculated for predicting the cyclic performance of the material in the full cell. It is recommended that total capacity loss at the first 20–50 cycles should be reported in the publication if the CE of the tested cell is lower than 99.5% in each cycle.(4)Energy density. The development of Li-ion batteries can be regarded as the history of the increase of energy density of the cell. It is always attractive and very exciting when a new material or a new cell with very high energy density is reported. In order to avoid overstatement, the following points should be known. First, when the anode and cathode materials are selected, the theoretical energy density can be calculated based on the Nernst equation from the formation energy data of the reactants and products.5Zu C.X. Li H. Thermodynamic analysis on energy densities of batteries.Energy Environ. Sci. 2011; 4: 2614-2624Crossref Scopus (677) Google Scholar It is also known that the highest ratio of the real energy density of the cell to the theoretical energy density of an electrochemical reaction is about 58%.5Zu C.X. Li H. Thermodynamic analysis on energy densities of batteries.Energy Environ. Sci. 2011; 4: 2614-2624Crossref Scopus (677) Google Scholar Possible energy density can be estimated roughly based on this ratio and the theoretical energy density. Certainly, an ideal calculation should include all materials in the cell based on reasonable data, similar to the real data provided in Table 1. It should be mentioned that some parameters in Table 1 can be modified further. The energy density of the cell with the same chemistry can be improved significantly if the reversible capacity of the cathode and thickness of the cathode layer increase, the thickness of the separator, Cu, and Al foil decrease, and the N/P ratio decreases from 2.0 to 1.2. Those improvements are possible. The thinnest Li foil of 3 μm can be prepared in industry, and many technology progresses are feasible to improve the performances. Recently, several papers have been published to discuss the energy density of new systems in practical cells.3Cao Y. Li M. Lu J. Liu J. Amine K. Bridging the academic and industrial metrics for next-generation practical batteries.Nat. Nanotechnol. 2019; 14: 200-207Crossref PubMed Scopus (289) Google Scholar, 6Betz J. Bieker G. Meister P. Placke T. Winter M. Schmuch R. Theoretical versus Practical Energy: A Plea for More Transparency in the Energy Calculation of Different Rechargeable Battery Systems.Adv. Energy Mater. 2019; 9: 1803170Crossref Scopus (252) Google Scholar, 7Liu J. Bao Z. Cui Y. Dufek E.J. Goodenough J.B. Khalifah P. Li Q. Liaw B.Y. Liu P. Manthiram A. et al.Pathways for practical high-energy long-cycling lithium metal batteries.Nat. Energy. 2019; 4: 180-186Crossref Scopus (1429) Google Scholar These calculations indicate clearly that the real energy density of the cell could be much lower than the value only obtained from a rough estimation from cathode or anode active materials. The readers should be very careful and not too optimistic regarding the reported data if the claims are only based on a few metrics or using inadequate and unsuitable cell parameters. Particularly, when a project target will be decided, the practical calculations based on full parameters are very helpful to judge whether the target can be realized. For example, for a Li/NMC cell (NMC: LiNixMnyCozO2), only when the reversible capacity of the NMC exceeds 220 mAh/g, it is possible to achieve the energy density of 500 Wh/kg.7Liu J. Bao Z. Cui Y. Dufek E.J. Goodenough J.B. Khalifah P. Li Q. Liaw B.Y. Liu P. Manthiram A. et al.Pathways for practical high-energy long-cycling lithium metal batteries.Nat. Energy. 2019; 4: 180-186Crossref Scopus (1429) Google Scholar That means that Li-rich oxide cathode could be more realistic or the only feasible option for achieving the target of 500 Wh/kg. It is not necessary to force the standardization of the electrode and button cell design for fundamental research due to diversity. However, based on discussions in Chen et al. and other recent publications,1Lin Z. Liu T. Ai X. Liang C. Aligning academia and industry for unified battery performance metrics.Nat. Commun. 2018; 9: 5262Crossref PubMed Scopus (189) Google Scholar, 2Chen S. Niu C. Lee H. Li Q. Yu L. Xu W. Zhang J.-G. Dufek E.J. Whittingham M.S. Meng S. et al.Critical Parameters for Evaluating Coin Cells and Pouch Cells of Rechargeable Li-Metal Batteries.Joule. 2019; 3 (this issue): 1094-1105Abstract Full Text Full Text PDF Scopus (249) Google Scholar, 3Cao Y. Li M. Lu J. Liu J. Amine K. Bridging the academic and industrial metrics for next-generation practical batteries.Nat. Nanotechnol. 2019; 14: 200-207Crossref PubMed Scopus (289) Google Scholar, 4Wang Q.Y. Chu G. Zhang J.N. Wang Y. Zhou G. Nie K.H. Zheng J.Y. Yu X.Q. Li H. The assembly, charge-discharge performance measurement and data analysis of lithium-ion button cell.Energy Storage Sci. Tech. 2018; 7: 327-340Google Scholar, 5Zu C.X. Li H. Thermodynamic analysis on energy densities of batteries.Energy Environ. Sci. 2011; 4: 2614-2624Crossref Scopus (677) Google Scholar, 6Betz J. Bieker G. Meister P. Placke T. Winter M. Schmuch R. Theoretical versus Practical Energy: A Plea for More Transparency in the Energy Calculation of Different Rechargeable Battery Systems.Adv. Energy Mater. 2019; 9: 1803170Crossref Scopus (252) Google Scholar, 7Liu J. Bao Z. Cui Y. Dufek E.J. Goodenough J.B. Khalifah P. Li Q. Liaw B.Y. Liu P. Manthiram A. et al.Pathways for practical high-energy long-cycling lithium metal batteries.Nat. Energy. 2019; 4: 180-186Crossref Scopus (1429) Google Scholar it is very important for researchers to understand that the experimental design, cell fabrication, and testing protocol of the button cell could have a significant influence on the results. Researchers should know the typical values in industry product and announce any exciting claim only after performing reliable experimental and suitable data analysis. It should be mentioned that in addition to basic electrochemical performance measurements, many other tests, characterizations, and analyses may also contain obvious deviations and problems. It is necessary to understand and standardize each method for achieving valuable and reliable data. This will save time, improve research efficiency, and decrease the difficulty in technology transfer. In the long history of battery development, all progress is based on systematic and effective innovations that can stand up to the scrutiny of others. This work is supported by National Key R&D Program of China (Grants No. 2016YFB0100100), Beijing Municipal Science & Technology Commission (Grants No. D181100004518003), and the National Natural Science Foundation of China (Y5JC011E21). H.L. thanks Dr. Wenjun Li and Dr. Jie Huang from Beijing Welion New Energy Ltd. for sharing cell data. Critical Parameters for Evaluating Coin Cells and Pouch Cells of Rechargeable Li-Metal BatteriesChen et al.JouleFebruary 28, 2019In BriefNumerous studies on Li anode have been published in recent years, but various experimental conditions make it extremely difficult to compare the developed materials for practical application. Based on our investigation of remarkable performance variations in rechargeable Li-metal batteries under different conditions, a set of parameters and testing conditions for Li-metal batteries were proposed to expedite the discovery of new materials and their full integration into realistic battery systems. Full-Text PDF Open Archive