The State of the Art of Silicon-Based Negative Electrodes for Lithium-Ion Batteries

How long is the battery verification period when new materials are used?

　　The SP component analysis and BT microstructure of material batches made by different manufacturers may be different. If the company’s original adhesive, including the proportion formula, is adjusted according to a certain batch of conductive carbon black supplied by a certain factory, but the next batch of incoming materials will still be adjusted again, and this time will not be too long. About a month or two. Because adhesives and conductive agents are not the main materials, but auxiliary materials, the manufacturer may not be able to use the main or auxiliary materials for a month. If it is a new project, the manufacturer has decided to use Heimao products at the beginning. No more moving, mainly to adjust the active material, positive and negative electrodes and other main materials. Once it is fixed, it will basically not be changed, and it will be done within a month or two.

　　The current application of silicon-based negative electrodes in lithium battery fast charging.

　　Betray, the first company on the Beijing Stock Exchange, started to make silicon-based anodes. The silicon-based anodes he supplied to Panasonic were used in Tesla. The technology of silicon-based anodes has also been iterated a lot, from the earliest nano-silicon From the powder to the compounding of nano-silicon powder and graphite, to silicon-oxygen and silicon-carbon, carbon was first coated on nano-silicon powder, and later evolved into depositing silicon in porous carbon.

　　After the domestic semiconductor supply chain was established, everyone used groupfourteen’s technical route to deposit silicon in porous carbon. At present, several top companies in China, such as Betray, Shanshan ( 600884 ), and Tianmu Pioneer founded by the Institute of Physics of the Chinese Academy of Sciences, are all doing groupfourteen’s technical route, that is, meteorological deposition of silicon, and silicon oxygen is gradually being abandoned.

　　At present, most manufacturers’ battery samples on the market are composite products of nano-silicon and graphite; however, the research and development stage is all silicon-carbon materials deposited by chemical vapor deposition. The main problem encountered at present is that the silicon core reaction process leads to a large volume expansion, and it is even very likely to tear the negative electrode sheet, resulting in failure of the battery. At present, the best structural design is doping 10% silicon carbon, that is, doping 5% silicon in the negative electrode. This is a relatively mature product in the research stage. If a higher energy density is to be achieved, the proportion of silicon must reach about 40 to 50%. The silicon carbon for trial sale in the future will be half carbon and half silicon, and then compounded with graphite, which is actually equivalent to 25%. -20% silicon content, such a silicon content is the state of research by various cell factories, because there is no way to solve the problem of volume expansion.

　　The first effect of pure silicon is relatively high. Compared with silicon oxygen, the first effect is similar to that of graphite. The 1:1 silicon carbon obtained from Tianmu Leading, Betray, and Shanshan may be compounded with some graphite by the manufacturer. Or maybe some hard carbon, because different carbon materials have different effects on reducing volume expansion. The negative electrode is generally water-based coating, and it is necessary to add CMC as a suspending agent. CMC helps the slurry to disperse evenly, but adding too much CMC will cause the battery resistance to increase, and the magnification will not increase; adding less CMC will cause the slurry to be unable to disperse evenly. At present, the company that wins the era in the industry is the least added by CMC. Its mixing and mixing are relatively good, and its engineering capabilities are very strong. As a result, materials made of silicon carbide have much lower internal resistance than those made by other manufacturers.

　　At present, the ranking of domestic silicon carbon companies is Beiterui, Shanshan, Jiangxi Zichen, and Tianmu Leading. The sample supply levels of silicon carbon companies are similar, and what is more important is the ability of product engineering, including production capacity, uniformity, and batch repeatability. One of the better ones is Betray. Then there is Jiangxi Zichen, Beiterui makes silicon carbon, and Jiangxi Zichen makes silicon oxygen. Now the general price of silicon carbon and silicon oxygen is about 170,000 to 80,000 per ton. The price of nano silicon powder is about 200,000 per ton. Silicon carbon and silicon oxygen are relatively the direction of future development, mainly silicon carbon.

　　In addition, there is a relatively new technical route, which is to recycle the old solar panels and crush them. Since the solar panels themselves will be doped, they are N-type or P-type semiconductors. The electronic conductivity of the semiconductor itself is relatively high. Recycling It is also a new direction to make nano-silica powder for solar panels. But at present, I don’t know who is doing better in China, and this technology is doing better overseas.

　　The gap and characteristics of silicon-based negative electrodes from the first generation to the third generation.

　　The first to third generations are different for different people. In my opinion, the first generation is the compounding of nano-silicon, nano-silicon powder and graphite, which is a simple mixture of ball milling physics. The first generation of products has the problem of volume expansion. Through compounding, it is possible Part of it is solved, but the overall structural design is not very good.

　　The second-generation product is silicon carbon, which is to make a layer of CAD on the nano-silicon powder and wrap a layer of carbon. On the one hand, it can improve the electronic silicon conductivity of silicon, and wrapping a layer of carbon will inhibit some volume expansion and shrinkage, which will affect the formation of SEI film. There are some benefits.

　　The third-generation product is the technical route of groupfourteen, which uses silane to do chemical vapor deposition in the hole carbon to form a silicon-carbon composite, which can best reduce volume expansion.

　　Some are classified according to nano-silicon, silicon-oxygen, and silicon-carbon. Silicon-oxygen can be regarded as the second-generation product. Compared with pure nano-silicon, silicon-oxygen can reduce the size better, because the energy storage mechanism is different. Silicon-oxygen has a The phase change reaction may form some inactive oxides, which will consume part of the lithium and reduce the first effect. The inactive oxide itself has neither ion conduction nor electron conduction, resulting in poor power. Now most companies also produce silicon-oxygen products, but basically they will not use silicon-oxygen and silicon-oxygen technologies in the future. The route may end there.

　　Does the third-generation product need to add lithium supplements?

　　Yes, that’s right. Lithium supplements are generally aimed at silicon-oxygen products. Silicon carbon itself has a relatively high first effect, and graphite itself has a relatively high first effect. Silicon carbon and graphite are compounded to form a composite silicon carbon negative electrode. The first effect is not good. It will be worse than graphite, and there is no need for lithium supplementation.

　　What are the first effects of pure graphite negative electrode, silicon oxygen, and silicon carbon?

　　The first effect of silicon carbon and graphite is similar, both are above 95, and silicon oxygen is relatively low, maybe more than 80, and less than 90. The products made by different manufacturers are different, which has a lot to do with the electrode fluid used.

　　The SEI film formed on the surface of the negative electrode will consume lithium ions. Is there any difference with different negative electrodes?

　　There will be differences for different negative electrodes. Graphite itself has a layered structure, and ions are solvated in the electrolyte. When inserted into the layered structure, it must be desolvated to form an SEI film. Use EC or VC electrode solution It is better to form a film on the negative electrode. The first effect of silicon is a little higher than that of graphite, or similar to that of graphite. It is relatively better to form an SEI film, and silicon oxygen is not a problem with the SEI film. The silicon oxygen material itself will be charged during the charging process. Decomposition produces inactive substances, which will consume lithium and lead to a decrease in the first effect.

　　There is no direct difference in the SEI structure of the three materials, because whether it is silicon carbon or silicon oxygen, they are all products compounded with graphite. The actual contact with the electrode liquid on the surface is actually graphite, and silicon carbon or silicon oxygen is buried in the Under the graphite, it does not directly contact the electrodes. The process of simply forming the SEI film will consume some lithium, but not too much. If the first effect is very high, reaching more than 90%, and only 12% of lithium is consumed, there is no need to use lithium supplements.

　　Which silicon carbon negative electrode is relatively better?

　　Overseas, it is groupforteen. Their core technology is being developed by many domestic companies, such as Betray, Shanshan, Tianmu Leading, and some start-up companies. As for which company does better, it mainly depends on stability, batch repeatability, and engineering capabilities. As far as I can see, SiC basically uses Celestial Eye as the lead, and then Betray. The uniformity and stability of the products of large manufacturers are better. There is no generation difference among domestic manufacturers, and everyone is in a hurry. It is nothing more than price competition and performance comparison of a certain batch, but it does not constitute a generation difference.

　　How is silane prepared in stomatal silicon?

　　Hard carbon has many pores, which can be ventilated. Silane undergoes a chemical reaction to transform silane into precipitated silicon, which is deposited in the pores to form composite silicon. This process is called chemical vapor deposition.

　　Has anyone done the technology of depositing silicon on activated carbon? How is it done?

　　It is to deposit a layer of silicon on the ordinary carbon material, then use CAD to coat a layer of carbon, then deposit a layer of silicon, and then coat a layer of carbon. This technology is also done, such as Betray. This is also a structural design that allows silicon and carbon to mix better, and at the same time, the problem of volume expansion can be better resolved.

　　What is the current progress of large cylinders such as 4680 and 4695? When is it expected that there will be a clear landing?

　　The 4680 Tesla has already landed, and domestic Zhongchuang and Yiwei have basically already landed. Yiwei has not shipped yet. Basically, domestic R&D has been completed. It mainly depends on when the car factory will place an order. This is a purely business issue.

　　What is the expansion coefficient of silicon carbon?

　　Generally, it will not be calculated in this way, because it will not be made with pure materials, but will be compounded with pure materials and different proportions of graphite.

　　Generally, the material is compounded with graphite first. For example, after 30% silicon carbon is compounded with graphite, there will be no expansion coefficient as usual, and it will distinguish whether it is Z-direction expansion or XY-direction expansion. Because of a pole piece, the Z-direction expansion is upward expansion, and the SY-direction expansion is expansion to both sides. According to our experience, silicon carbon and graphite expand at about 37. This expansion is acceptable to everyone, because the expansion itself It is also related to the compaction of the pole piece. The expansion in the XY direction will not be too obvious, causing the pole piece to curl or break.

　　After silicon carbon and silicon oxygen are compounded with graphite, the comparison of the expansion under the same mixing ratio.

　　At present, the difference between the two should not be very big. The main difference between the third generation and the second generation is not the expansion, the main difference is that the first effect of silicon oxygen is relatively low, so it is basically not used now. Silicone has also made some microstructures such as carbon coating, which will also ease its expansion.

　　What is the main reason for the elimination of silicon oxide?

　　I think it is a cost issue, and the process will be more complicated. Because the sacrificial lithium replenishing agent pressed on the positive electrode needs to be pumped away after the gas generated is exhausted. For example, adding lithium oxalate to the positive electrode produces carbon dioxide. If a lot of lithium metal is used on the negative electrode, it will be used during slurry mixing or after coating. , It is very dangerous to get it up. In the past, some battery cell factories used lithium powder to supplement lithium, which caused accidents. Some people also add stabilizers to make lithium slurry or others. First, the cost is high, and second, the process is relatively complicated. Taking many detours to save silicon oxygen, it is actually better to use silicon carbon directly.

　　How much is the price per ton of the third generation of CVD? What is the future price trend?

　　If we buy it now, it will be less than 200,000. The cost itself is high, because it involves a lot of high-temperature domestic production, even if it is made of nano-silicon, it also involves a lot of high temperature, and the grinding cost will not be low. But how to get down to the price, I don’t know the material.

　　Did the CVD method encounter any difficult problems in the laboratory?

　　At present, many factories are doing the route of making nano-silicon by CVD, and they all follow the technology of groupforteen. CVD has little contact with traditional industries. Later, due to the support of semiconductors, many domestic equipment, including raw materials, special gas, silane, etc. , Many companies are doing mass production. Personally, I think that there is no problem in supplying and mass producing battery cells on a large scale.

　　In addition, large-scale production also needs to consider how the battery cell factory can better solve the problem of volume expansion, because the amount of silicon itself was not too large before, and its ratio of graphite to graphite is 1:9. The factory solves the problem of how to make high silicon. At present, the proportion of silicon is relatively high, and the proportion of Panasonic’s first-generation silicon is less than 1%. In addition, to solve the problem of volume expansion, many pulping technologies, adhesive technologies, and cell factories are currently doing it; the last is the problem of feeding materials, that is, the fluidized bed technology. This technical route itself is feasible, and it is only a matter of time. Question, explore some specific parameters.

　　What is the progress of the route of producing nano-silicon by PVD method?

　　Nano-silicon is used in products now, but the real research and development direction is on CVD long silicon carbon. I think it is more suitable to use ball milling method for nano-silicon, and the cost is low. Moreover, my country’s photovoltaic industry is developed, so many scrapped photovoltaic panels, directly Just use the photovoltaic board to grind it, there is no need to use PVD to make nano-silicon.

　　The scrapped photovoltaic panels themselves are doped semiconductors with high electronic conductivity. The earliest Tianmu leader used ball milling method to grind silicon powder, micro silicon or nano silicon. Nano-silicon has a larger specific surface, more side reactions, and has a larger area to grow SCI and SEI films; micro-silicon has a larger volume and smaller specific surface, so there will not be so many areas to grow SCI films, and the voltage of the interface is smaller. I think it’s better to use the ball mill method.

　　Is the main problem faced by the CVD method cost?

　　Yes, one is the cost problem, the price is a bit high, and the cost of silicon itself is not low; the other is the problem of insufficient demand.

　　Compared with the original nano-silicon silicon carbon, which theoretical cost is lower?

　　If the theoretical cost is lower, it must be the original one, but the original nano-silicon only accounts for 0.5% in the negative electrode, but if some structural design is done, like the current silicon carbon may be able to achieve 5%, silicon The content is ten times higher than that of the original product, and the problem of body expansion has also been well solved.

　　At what level can the price of silicon carbon drop to a large scale?

　　It’s not easy to judge. It has a lot to do with the market. If the demand is strong, long battery life is required, and the consumption power is strong, the price can be increased even if the price is higher. I think what really drives technological progress and large-scale application is more cost and supply chain maturity.

　　Which companies do better in silicon carbon?

　　Silicon carbon is mainly made by Japanese companies, such as Toray and Kuraray. There are also a few domestic companies, but they are incomparable with Japan. I think that the only hopeful domestic silicon carbon companies are big manufacturers such as Betray and Shanshan. Relying on their engineering capabilities, if they have a clear view of the market and want to enter the market, there is still hope.