A Clear Explanation of the Essential Differences Between ABF Substrates, BT Substrates, and Glass Substrates

As AI computing chips evolve toward large sizes and multi-die Chiplet heterogeneous integration, packaging substrates, as the core base of chips, directly determine high-density interconnect capability, signal integrity, and packaging yield. BT substrates, ABF substrates, and glass-core substrates respectively represent the mature organic substrate solution, the current mainstream solution for high-end computing, and the next-generation potential upgrade direction. The three are not simply iterative replacements, but rather form a layered application landscape based on cost, performance, and process maturity.

I. BT Substrate: Mature and Reliable, but with a Performance Ceiling

BT substrates use bismaleimide triazine resin as the matrix, impregnated with glass fiber cloth and then laminated as a whole. They feature high hardness, excellent dimensional stability, and good heat resistance, making them the most mature organic packaging substrate solution currently available.

However, their technical limitations are equally prominent: BT resin is relatively rigid, making drilling difficult, and microvia processing capability is limited. Line width/line spacing typically remains in the 25–40 μm range, with layer counts concentrated at 2–4 layers, making it difficult to break through in wiring density. This characteristic determines that BT substrates cannot support high-density interconnect requirements at the tens of thousands of pins level.

Therefore, BT substrates are mainly used in memory chips (DRAM, NAND), RF modules, MEMS sensors, and other scenarios—fields that demand high reliability but do not require extreme wiring density. Memory-type substrates account for approximately two-thirds of the CSP packaging substrate market, making them the most important downstream market for BT substrates.

In short, BT substrates are the pragmatic choice of "good enough": controllable cost, mature process, but with a clear ceiling.

II. ABF Substrate: The Core Carrier for High-Density Interconnect, but Approaching Physical Limits

The core material of ABF substrates is Ajinomoto Build-up Film, which contains no glass fiber cloth. It is manufactured using multilayer film stacking combined with semi-additive process (SAP), achieving mass-production line width/line spacing of 8–15 μm, with further scaling possible in advanced processes.

Core Advantages:

· Low dielectric loss: The fiber-free structure avoids signal reflection caused by fiber weaving effects, giving significantly better high-frequency signal integrity than BT substrates.

· High layer stacking capability: Supports 4–16 layers or even higher stacking, meeting the FC-BGA packaging needs of CPUs, GPUs, AI accelerators, FPGAs, and other high-I/O-density chips.

· High surface flatness: Good via-filling and wiring consistency, suitable for fine-line processing.

Inherent Shortcomings:

The physical bottleneck of ABF substrates centers on thermal expansion mismatch. The planar coefficient of thermal expansion (CTE) of conventional ABF base material is approximately 13–17 ppm/°C, and in the Z-axis direction it reaches as high as 48 ppm/°C, while the CTE of silicon chips is only about 2.6 ppm/°C—a difference of up to 5 times. As chip packaging area continues to expand, this mismatch causes severe warpage in organic materials during lamination and thermal cycling. The larger the substrate area, the more difficult warpage becomes to control, directly affecting lithography and die-bonding precision. In addition, the measured dielectric loss of ABF at high frequencies can reach 0.018 or above in the 10 GHz range, making signal attenuation increasingly prominent in high-frequency scenarios.

On the material supply side, Ajinomoto Corporation holds a global market share of over 95% in ABF materials used for GPU and CPU packaging substrates, indicating extremely high supply concentration. As AI chip sizes increase and layer counts rise, the amount of ABF material consumed by a single high-end AI chip reaches 5 to 10 times that of a traditional CPU, and the supply-demand relationship continues to tighten.

III. Glass Substrate: A Material Upgrade Born for Next-Generation Computing

First, a common misconception must be clarified: glass substrates are not "replacing ABF substrates." The technical route of glass substrates is to replace the traditional organic core board with a glass core (not to replace the ABF film itself). ABF remains as an insulating build-up material on both the upper and lower sides of the glass core. Chip interconnect functionality is achieved jointly by the chip-side RDL, through-glass vias/copper interconnect structures, and ABF build-up layers. Glass and ABF are a complementary, coexisting combination.

Core Advantages:

· Thermal expansion matching: The CTE of glass is approximately 3.2 ppm/°C, highly matching that of silicon chips, which can fundamentally suppress warpage and solder joint fatigue caused by thermal cycling.

· Ultra-flat surface: Surface roughness of less than 0.1 μm ensures high-precision etching of fine lines, supporting line width/line spacing below 2 μm.

· Low dielectric loss: As an insulator, glass has a dissipation factor approximately 10 times lower than traditional organic substrates, supporting high-speed signal transmission exceeding 112 Gbps.

· High rigidity: Young's modulus is between 50–90 GPa, several times that of organic materials, making large-size substrates less prone to deformation.

Industrialization Challenges:

The mass production of glass substrates faces multiple process challenges. TGV processing yield is the core bottleneck. Issues such as via formation precision, defect-free copper filling, copper adhesion, multilayer RDL alignment, and thermal cycling reliability have not yet been fully resolved. The current full-process mass production yield is approximately 50%–60%, with a target of raising it above 75%. In addition, the manufacturing and cutting of large-size ultra-thin glass, panel-level alignment, and mass production cost control are also key thresholds on the path from laboratory to scaled commercial application.

Mainstream technical routes fall into two categories:

First, as a glass-core substrate, replacing the organic core board with a glass core, combined with ABF build-up, for BGA/CSP/2.5D packaging; second, as a glass interposer, replacing the silicon interposer with glass, for CoWoS and other 2.5D/3D packaging scenarios, using square glass panels to achieve larger packaging areas and lower costs.

IV. The Essential Differences Among the Three

If a chip is compared to a house, the substrate is the foundation—BT is a solid old-fashioned concrete foundation, ABF is a precision reinforced concrete structure, and glass substrate is a new type of composite foundation that combines high strength with thermal matching.

The table below intuitively presents the core differences among the three from eight dimensions:

The relationship among the three is not simply iterative replacement. BT substrates still offer cost and maturity advantages in mid-to-low-end scenarios; ABF substrates are the primary packaging base material for current high-end computing chips, but their physical limits are being continuously approached; glass substrates are the material upgrade direction to meet the next generation's larger-size, higher-density packaging needs, and industrialization still requires time. Understanding the boundaries and applicable scenarios of the three is far more valuable than debating who will "replace" whom.