MLCC performance competition in the era of AI data centers: why does Murata's production and solar induced electricity lead significantly? AI data centers are driving MLCC into a new round of technological upgrade cycle. In the past, servers mainly used 12V power supply, but now they are evolving towards 48V rack power supply, and may even enter the era of 800V HVDC high voltage direct current in the future. At the same time, the power consumption of AI platforms such as NVIDIA GB200 and GB300 continues to increase, and the GPU core voltage has dropped to 0.6V-0.8V, but the current of a single GPU has exceeded 1000A. For MLCC, the challenges mainly come from three directions. Firstly, there is high pressure. 48V power supply requires higher voltage resistance, higher reliability, stronger heat resistance, and mechanical stress resistance, so the demand for 100V or even higher voltage resistant MLCCs is rapidly increasing. Next is transient response. The load variation of AI GPU occurs in nanoseconds, and the power supply network must have extremely low ESL (equivalent series inductance) and extremely low impedance, otherwise voltage drop, performance degradation, and even system instability may occur. The third is spatial limitation. The PCB area around the GPU is becoming increasingly tight, and engineers hope to place more decoupling capacitors in the closest position to the GPU. Therefore, MLCC must simultaneously achieve small size, high capacitance value, and high volume efficiency. Faced with these demands, the industry has begun to develop towards high-voltage MLCC, ultra-low ESL MLCC, and ultra-high capacity MLCC. Among them, Murata Manufacturing and Solar Induction became the two most representative companies. Sun Induced Power has launched the LWDC low ESL series MLCC, which significantly reduces ESL through a reverse electrode structure, making it particularly suitable for AI GPU power supply scenarios. Simultaneously layout high-voltage MLCCs above 100V and high-capacity MLCCs, and actively promote Embedded MLCC technology. Murata continues to break industry records, maintaining a leading position in small-sized, high-capacity, and highly reliable products. What gives them a competitive advantage lies in the materials. MLCC may belong to the manufacturing industry, while high-end MLCC is closer to the materials technology industry. Its core technology chain: BaTiO ∝ dielectric powder → slurry formulation → thinning → lamination → sintering → MLCC The most difficult and challenging part is the dielectric powder. MLCC mainly uses barium titanate (BaTiO3) as the dielectric material. But the differences in BaTiO ∝ from different manufacturers are reflected in: Particle size control particle size distribution Rare earth doping system Core Shell structure Grain growth control These abilities collectively determine the ultimate performance limit. That's also why, with the same MLCC, Murata can achieve 100 μ F and solar induction can achieve 50 μ F, while most other manufacturers cannot even achieve 22 μ F? The reason is that Murata and solar induced electricity can make the dielectric layer thinner and stack more layers. For fixed size MLCC, there are only three things that can be relied upon to improve capacity: Higher dielectric constant Thinner dielectric layer More stacking layers The problem is that as the dielectric layer continues to thin, the requirements for materials will exponentially increase. If the BaTiO ∝ particles are too large, when the thickness of the dielectric layer decreases to 0.5 μ m or even lower, only two or three layers of grains may remain. At this point, leakage, breakdown, and lifespan issues will rapidly deteriorate. One of the biggest advantages of Murata and solar induced electricity is the ability to achieve extremely fine and highly uniform BaTiO3 particles, thereby continuing to promote the thinning of the dielectric layer. And particle size is only the first step. Particle size distribution is often more important. If the particle size difference is too large, it is easy to form abnormal grains, voids, and stress concentration after sintering, ultimately leading to a decrease in reliability and a deterioration in yield. High end MLCC manufacturers generally have the most advanced particle size distribution control capabilities in the industry. Further up is the Core Shell technology. High end MLCC requires a special rare earth doped layer to be coated on the outside of the BaTiO ∝ core. Core is responsible for providing high dielectric constant. Shell is responsible for controlling leakage current, improving insulation performance, and extending lifespan. This part is often one of the core technical secrets of Murata's production and solar induction. Even with the same powder, the sintering process can still result in significant differences in final performance. The temperature curve, oxygen partial pressure control, holding time, and cooling rate during the sintering process all affect grain growth. Truly leading manufacturers are not only able to produce ultrafine powders, but also maintain a small, uniform, and stable grain structure after sintering. This is also why high capacity MLCCs are extremely difficult to manufacture. The difficulty of 100 μ F is the need to stably stack hundreds or even thousands of ultra-thin dielectric layers together. Any minor defect in any layer may lead to the failure of the entire product. Therefore, high-value products are essentially a comprehensive competition of material science, process control, and yield management. From the perspective of the industrial landscape, the current high-end MLCC market roughly presents the following tiers: Murata Manufacturing - an industry leader with comprehensive leadership in materials, processes, and products. Taiyo Yuden - the closest competitor to Murata, has long maintained a leading position in the high-end MLCC field. TDK - Strong technical strength, continuously catching up with the first tier. Samsung Electro Mechanics - With outstanding manufacturing capabilities, it continues to expand in the AI server market. Yageo, Fenghua Advanced Technology and other manufacturers are continuously catching up. The MLCC that will be most needed for future AI servers is no longer a popular product in the era of consumer electronics. But at the same time, it possesses: high pressure High capacity value Ultra low ESL small size high reliability Products with five abilities. And these five abilities will ultimately point to the same source: decades of accumulated BaTiO Ⅲ powder technology, Core Shell structure design, ultra-thin dielectric manufacturing capabilities, and sintering process experience. That's also why in the era of AI data centers, what really sets the gap is not MLCC itself, but the material science behind it. Disclaimer: I hold the assets mentioned in the article, and my views are biased and not investment advice, dyor