Market Definition and Technology Segmentation

This report studies the ceramic substrates used in power modules. The key ceramic substrates are DBC substrates, AMB substrates, and DBA substrates. The power modules include IGBT modules and SiC modules. DBC (Direct Bonded Copper) substrates are composed of a ceramic insulator—Al₂O₃ or AlN—onto which pure copper metal is attached by a high-temperature eutectic melting process and thus tightly and firmly joined to the ceramic. Active Metal Brazing (AMB) represents the latest development in ceramic substrates and offers the ability to produce heavy copper with AlN or SiN. The normal metallization process is not used, as AMB involves brazing pure copper on the ceramic in a high-temperature vacuum brazing process. Alongside a high-reliability substrate with unique heat dissipation, the brazing technology enables double-sided copper weights of up to 800µm on thin ceramic substrates of just 0.25mm.

The technology frontier is not a simple substitution story. DBC remains the workhorse for industrial IGBT modules where cost and mature supply chains dominate. AMB, particularly with silicon nitride (Si₃N₄) ceramic, is the enabling substrate for SiC traction inverters in electric vehicles, where thermal cycling reliability and heat dissipation determine module lifetime. NGK’s Si₃N₄-AMB substrates achieve bonding layers of several microns or less, producing almost no influence on thermal resistance, with thermal conductivity reaching 80–90 W/mK in standard grades and 100–110 W/mK in higher-conductivity variants . Toshiba Materials’ TSN-90 silicon nitride substrate reaches 90 W/mK thermal conductivity with 680 MPa three-point bending strength and 6.5 MPa·m¹/² fracture toughness . These specifications are not incremental improvements; they are the material basis for the reliability claims that automotive OEMs require from SiC traction modules.

Competitive Landscape and Recent Developments

According to QYResearch, the global key manufacturers of power module substrates include Rogers Corporation, Ferrotec, BYD, Heraeus Electronics, NGK Electronics Devices, KCC, Toshiba Materials, Denka, Shengda Tech, Mitsubishi Materials, and others. In 2023, the global top ten players held approximately 85.0% of revenue share. This concentration reflects the capital intensity of ceramic sintering and brazing, the multi-year qualification cycles required for automotive and industrial applications, and the technical service burden of co-developing substrate designs with module makers.

Recent developments illustrate the competitive dynamics at the technology frontier. Heraeus Electronics introduced Condura.ultra metal-ceramic substrates at APEC 2026—advanced silver-free AMB substrates combining high thermal conductivity with mechanical stability, positioned for demanding power density and reliability requirements . The shift to silver-free brazing systems is analytically significant: silver migration under high-voltage bias is a reliability concern in automotive applications, and eliminating silver from the brazing layer addresses a failure mechanism that conventional AMB systems must manage through design margins.

BYD’s international patent filing for a silicon nitride ceramic sheet, published in March 2026, signals the vertical integration strategy of a major automotive OEM into the substrate supply chain. The patent covers a composite plasticizer and method for preparing silicon nitride green bodies and ceramic sheets—a manufacturing process innovation rather than a materials composition claim . The strategic intent is clear: BYD’s high-voltage fast-charging vehicles depend on SiC power modules, and the silicon nitride substrate is the critical thermal and mechanical support component. By developing in-house process capability, BYD reduces dependence on external AMB suppliers and captures process know-how that is otherwise embedded in supplier relationships.

Ferrotec’s Jiangsu Fulehua Semiconductor Technology subsidiary positions itself as a full-portfolio substrate supplier spanning AMB, DCB, DPC, DBA, and TMF, with sales networks covering more than 20 countries and applications in electric vehicles, wind power, locomotive traction, and HVDC transmission . The company’s DCB ZTA (zirconia-toughened alumina) substrate offers thermal conductivity 1.1 times and flexural strength 1.7 times that of standard alumina, targeting medium-power modules requiring higher reliability . This product-tiering strategy—standard alumina for cost-sensitive industrial, ZTA for medium-power automotive, and Si₃N₄-AMB for high-power SiC—illustrates how leading suppliers capture value across the application spectrum rather than competing on a single technology.

Market Drivers

The primary demand driver is the accelerating adoption of SiC power modules in electric vehicle traction inverters. SiC MOSFETs operate at higher junction temperatures and switching frequencies than silicon IGBTs, and their thermal expansion coefficient is closer to silicon nitride than to alumina. This materials compatibility is not a secondary consideration; it is the reason Si₃N₄-AMB has become the default substrate for automotive SiC modules. As BYD and other Chinese OEMs push fast-charging platforms, the substrate’s role in enabling “充电5分钟续航400公里” class performance has moved from engineering detail to competitive specification .

A second driver is renewable energy and industrial electrification. Wind turbine converters, photovoltaic inverters, and locomotive traction systems all require power modules that survive thermal cycling and high-voltage isolation over service lives measured in decades. AMB substrates’ superior thermal cycling durability compared to DBC makes them the preferred choice in these applications, even where the cost premium is difficult to justify on first-cost grounds. The AMB substrate market is projected to reach USD 2.295 billion by 2032 at an 11.9% CAGR for DBC and AMB combined, reflecting the sustained demand pull from these segments .

A third driver is the geographic shift of production. AMB ceramic substrates are currently produced primarily in Japan, Germany, and China, which accounted for 27.3%, 24.56%, and 43.5% of market share respectively in 2023 . China’s 43.5% share is notable because it reflects both the scale of domestic demand and the rapid capacity build-out by suppliers such as Ferrotec’s Fulehua and BYD. This geographic concentration in China creates a supply-chain dynamic distinct from semiconductor photoresist or advanced logic: the substrate segment is less exposed to export-control risk because the technology is materials-process-based rather than equipment-dependent, and Chinese suppliers have accumulated meaningful production experience.

Restraints

The principal restraint is the qualification burden and its interaction with capacity timing. AMB substrate qualification for automotive SiC modules requires thermal cycling, power cycling, and high-temperature reverse bias testing that can extend over 12 to 18 months. Suppliers that build capacity ahead of qualification risk idle assets; suppliers that wait for qualification signals risk losing allocation to competitors who committed earlier. This timing mismatch is particularly acute in the current cycle, where automotive OEMs are simultaneously ramping SiC platforms and demanding supply-chain localization.

A second restraint is the cost premium of AMB relative to DBC. Silicon nitride ceramic is more expensive than alumina, and the vacuum brazing process is slower and more capital-intensive than the eutectic bonding used in DBC. For medium-power industrial modules where thermal cycling requirements are less severe, DBC remains the rational choice, and the AMB premium is difficult to justify. This cost-performance trade-off limits AMB’s addressable market to high-power, high-reliability applications where the total cost of ownership argument—including warranty costs and field failure rates—outweighs the unit price differential.

A third restraint is the technical complexity of thick copper brazing on thin ceramics. The AMB process enables double-sided copper weights up to 800µm on 0.25mm ceramic, but managing residual stress and warpage across this thickness ratio requires precise control of brazing temperature profiles and cooling rates. Warpage in the as-sintered substrate propagates through die attach and wire bonding, reducing yield and reliability. Patent activity in this area—including Mitsubishi Materials’ work on Cu₃P phase area ratios in bonding layers and low-warpage DBC designs with stress-relief grooves—indicates that warpage control remains an active engineering frontier rather than a solved problem .

Opportunity

The most significant opportunity is the localization of AMB supply in China. With 43.5% of global AMB production already located in China and domestic automotive OEMs committing to SiC platforms, the addressable market for Chinese substrate suppliers is large enough to support multiple scaled competitors . Ferrotec’s Fulehua, BYD, and Shengda Tech are positioned to capture this demand, but the qualification requirements of tier-one automotive customers mean that volume will concentrate among suppliers with demonstrated reliability data. The opportunity is not simply to replace imports; it is to set the reliability benchmark that defines the next generation of automotive SiC modules.

A second opportunity lies in substrate-level integration. As module designs become more compact and power densities increase, the boundary between substrate and thermal interface material, and between substrate and baseplate, is being renegotiated. Heraeus’s positioning of Condura.ultra as part of a “fully matched power module system” alongside die attach materials and sintering pastes reflects a recognition that substrate performance is system-dependent . Suppliers that can offer co-optimized substrate-attach-baseplate stacks will capture value that standalone substrate vendors cannot.

Analyst Observations

Two underappreciated dynamics merit attention. First, the 85.0% top-ten concentration is higher than in most semiconductor materials segments, but it is not stable. The composition of the top ten is shifting as Chinese suppliers scale. The conventional view holds that Japanese and German incumbents dominate high-reliability AMB and Chinese suppliers serve cost-sensitive domestic demand. The data complicates this: China already accounts for 43.5% of AMB production, and BYD’s patent filing shows an OEM moving into process development rather than relying on merchant suppliers . The competitive boundary is not geography but qualification: suppliers that can demonstrate automotive-grade reliability will win share regardless of origin, and the incumbents’ advantage is their accumulated reliability data, not their nationality.

Second, the 17.5% CAGR is the highest in this series of market summaries, and it is driven by a technology transition—SiC adoption—rather than by cyclical demand. This matters for strategic planning because technology-driven growth is less sensitive to macroeconomic conditions but more sensitive to qualification timelines. If SiC vehicle ramps slip by a year, substrate demand slips with them; if they accelerate, substrate supply becomes the bottleneck. The suppliers that invested in capacity ahead of qualification, and that have the process discipline to yield at automotive defect rates, will capture disproportionate value. The 17.5% CAGR is not a forecast of smooth growth; it is an average that will be realized through a lumpy, qualification-gated sequence of platform launches.

Contact Us:

If you have any queries regarding this report or if you would like further information, please contact us:

QY Research Inc.
Add: 17890 Castleton Street Suite 369 City of Industry CA 91748 United States
EN: https://www.qyresearch.com
E-mail: global@qyresearch.com
Tel: 001-626-842-1666(US)
JP: https://www.qyresearch.co.jp