Bottom line first: a glass core substrate is not "the next PCB" — it is an attempt at a generational material change for advanced packaging substrates, and what it replaces is the core layer, not the build-up, the fine lines or the back-end assembly. For PCB and substrate makers the opportunity is therefore larger than the threat. The practical conclusion for procurement is that there is no need to switch routes within the next 12–24 months, but three verifiable signals have to be watched closely.
Why the industry is looking at glass
- CTE matching: glass has a CTE of about 3–4 ppm/°C, clearly below the 12–17 ppm/°C of organic substrates and much closer to silicon, so warpage in large packages is smaller.
- Rigidity and flatness: high stiffness, very low moisture absorption and dimensional stability support finer RDL line widths and thinner dielectrics.
- Panel-level potential: glass suits large panel-level substrates; compared with round wafers it improves area utilisation and dilutes cost per unit.
Three engineering gates that have not been cleared
The material advantages are real, but they only matter once the process chain behind them holds. Three gates are still open.
- TGV formation: glass vias are commonly formed by laser-induced modification followed by etching, with hole diameters in the 20–100 μm range; aspect ratio, wall roughness and chipping control decide usability directly.
- Metallisation adhesion: glass and copper adhere poorly by nature, so an extra seed or adhesion layer is required, and interface reliability stays a long-term failure risk.
- Brittleness and yield: edge micro-cracks, handling breakage and local defects on large panels raise both the defect rate and the inspection cost — the key variable in today's cost curve.
What it pulls through the high-end PCB and substrate supply chain
It is important to separate "changing the core" from "changing the whole process". Glass replaces the carrying and rigidity role; the build-up above it, the RDL and fine lines, copper thickness control, surface finish and back-end assembly still have to be done by someone. Three judgements follow from that.
- Plants that already have mSAP fine-line and substrate processing capability are better placed to take a position on the glass route.
- Demand for thick copper and high-current structures does not disappear with glass; power plane design and heat paths become more, not less, dependent on engineering capability.
- Inspection has to be rebuilt: the transparency and brittleness of glass mean that conventional AOI and cross-section interpretation both need re-calibration.
A 12–24 month response cadence
- No switch in the short term: make dual-source qualification of existing organic substrates and high-multilayer boards solid, and do not hand mainstream capacity to a route that has not yet settled.
- Watch three signals: whether TGV via yield stabilises, whether panel-level size yield is acceptable, and whether back-end assembly and thermal cycling compatibility close the loop.
- Certify capability early: low-loss materials, fine lines and thick copper structures are the same capability base that the glass era will need.
- At contract level: require glass suppliers to provide TGV cross-sections, thermal cycling and warpage data, so that a single passing sample is not mistaken for readiness for mass production; assess sole-source risk at the same time.
Conclusion
The direction will most likely hold; the timing is not yet fixed. For procurement and engineering the best answer is not to bet on one route but to make the general capability set — high layer count, fine lines, thick copper — solid. Whether the core layer ends up organic or glass, that capability will be re-validated once more, and it happens to be the part that can be accumulated starting now.
AIPCB.cc (Shenzhen Lingxin Technology Co., Ltd.) serves AI computing and high-end applications with 1–44 layers, a maximum panel size of 610×1100 mm and finished copper up to 13oz (outer-layer base copper up to 18oz), a first-pass rate of 99.8% and a mass-production yield of 99.5%. To evaluate a high-end stackup and process route, send Gerber data and stackup requirements to info@ipcba.com.cn or call +86 138 2310 9230.
Send your Gerber files and stackup requirements; a 24-hour expedite channel is available.