Deep Tech · Paradigm Analysis
Technology Analysis Blog
Paradigm
Shift Lab
Breaking barriers through materials & structural innovation
Skip to main content

TGV — Why Drilling Holes Through Glass Is the Hardest Problem in Packaging

Glass Substrate Series · 03

TGV — Why Drilling Holes Through Glass Is
the Hardest Problem in Packaging

Through-Glass Vias are the single biggest technical barrier to glass substrate mass production. Here's what makes them so difficult — and who's closest to solving it.

Glass Substrate TGV · Through-Glass Via Intermediate ~8 min read

Ask any expert what the hardest part of manufacturing a glass substrate is, and they'll give you the same answer: drilling the holes. What sounds almost trivially simple hides one of the most demanding engineering challenges in the history of semiconductor packaging.

By now it's industry consensus that glass substrates outperform ABF (Ajinomoto Build-up Film) on thermal expansion, planarity, and high-frequency signal loss. But translating those material advantages into an actual product requires clearing one formidable hurdle: drilling thousands of vertical holes — tens of micrometers in diameter — through glass without cracking it, then filling every one perfectly with copper. That process is the Through-Glass Via, or TGV.

What Is a Via and Why Does Glass Make It So Hard?

A chip mounted on a substrate needs to pass electrical signals down to the PCB below. The vertical conductive channels that carry those signals through the substrate are called vias. ABF substrates use laser-drilled copper-filled microvias; silicon interposers use TSVs (Through-Silicon Vias). Glass substrates need the equivalent: TGVs.

The problem is glass's fundamental nature. Glass is a brittle material — it fractures without deforming. Unlike metal or silicon, you can't simply drill into it incrementally. And the via diameters that advanced packaging demands are smaller than half a human hair (sub-30μm). Thousands of those holes, in exact positions, without cracking the substrate: that's the TGV challenge.

📐 Scale Reference

A human hair is roughly 70μm in diameter. TGV target diameters range from 20–50μm. That density also means — in theory — 4× more vias per unit area than ABF can support. The keyword is "in theory."

TGV — 4-Step Process Flow
STEP 1 Laser Drilling Glass Panel Laser forms initial holes (CO₂ / UV / ps laser) STEP 2 Wet Etching HF solution widens holes (taper angle control) STEP 3 Seed + Plating PVD seed layer, then Cu plating (void-free fill) STEP 4 CMP + Wiring Surface planarized, RDL layers formed (complete)

Three Ways to Drill Glass — None Perfect

There are three main approaches to forming holes in glass substrates. None has reached the yield and throughput required for high-volume manufacturing. Dozens of companies are competing on exactly this problem right now.

🔴
CO₂ Laser
Low equipment cost
High throughput
Inconsistent hole quality
Microcracks around holes
Limited minimum diameter
💜
UV / Picosecond Laser
High precision
Sub-20μm capable
Very high equipment cost
Slow throughput
HAZ (heat-affected zone)
🔵
LEAP (Laser + Etch)
Quality-speed balance
Most viable for scale
Complex process flow
HF etch safety requirements
Yield reproducibility
⚠ The Core Dilemma

Push for precision and throughput collapses. Push for speed and yield falls apart. Drilling tens of thousands of TGVs across a large-format glass panel (300mm+), without cracking, uniformly, at production speed — satisfying all three simultaneously is the defining challenge of TGV engineering.

Drilling Is Only Half the Problem — Filling Is Harder

Forming the holes doesn't complete a TGV. Depositing a uniform copper seed layer on the inner walls — and then filling the cavity void-free through electroplating — is an equally demanding challenge.

Unlike metal or silicon, glass is electrically inert. Before plating can begin, the surface must be activated, a seed layer deposited by PVD, and plating bath conditions precisely controlled to ensure uniform electrolyte exchange deep inside narrow vias. If any condition drifts, a void forms — an unfilled pocket in the copper. That single void, in one via among millions, causes an open circuit. In a thermal cycling reliability test (−55°C ↔ +125°C), stress concentrates around voids, and after hundreds of cycles the via fails — taking the entire package with it.

🔬 Why Voids Are Fatal

A single void in a TGV can survive initial electrical testing and still cause field failure after thermal stress. In a package housing a $10,000+ AI accelerator chip, one open circuit writes off the entire assembly. This is why void-free fill yield is treated as a make-or-break metric.

Yield Is the Real Barrier — Where the Industry Stands

TGV yield levels vary by player, but virtually no one has publicly claimed to reach the industry-standard threshold for high-volume manufacturing (typically 99.9%+). Below is a relative technology maturity estimate based on public statements and industry sources.

Philoptics (KR)
Leading
LPKF (DE)
Leading
Corning (US)
Dev.
AGC / NEG (JP)
Dev.
New entrants
Early

※ No public figures available. Relative maturity estimates based on aggregated industry sources.

The Competitive Map — Who's Where

Category Key Players Strengths Status
Equipment Philoptics (KR)
LPKF (DE)
Most TGV equipment references
LIDE process IP
Abseis supply
Glass Material Corning (US)
AGC / NEG (JP)
Large-format panel supply
Optical glass expertise
Pilot stage
Substrate Mfg. Absolics (SKC)
Samsung Electro-Mech.
First dedicated fab lines
Large-scale pkg. know-how
Pilot production
Plating Chemistry YCChem (KR)
Soulbrain (KR)
TGV-specific plating fluids
Etchant co-development
Supply chain entry
End Customers AMD · Amazon
Intel (external pivot)
AI accelerator packaging
Data center scale
2027–2028 target

Beyond TGV — What Comes Next

TGV is the first major wall in glass substrate manufacturing, but not the only one. Once vias are formed and filled, the stack still faces redistribution layer (RDL) patterning, chip-to-substrate bonding reliability, and planarity maintenance across large panels — each a nontrivial challenge in its own right.

Yet TGV commands the most R&D attention for a simple reason: whoever masters stable, high-yield TGV formation controls the critical bottleneck in the glass substrate supply chain. The technology itself is the moat. Yield is the barrier to entry, and yield is the business advantage.

📌 Key Takeaways

TGV looks deceptively simple — drill holes through glass, fill with copper — but satisfying precision, throughput, and yield simultaneously makes it one of the most demanding processes in packaging history. All three drilling approaches (CO₂, UV/ps laser, LEAP) carry their own limits, and void-free copper fill adds another layer of complexity. Philoptics and LPKF lead on equipment; Absolics and Samsung Electro-Mechanics are in pilot production. Customer adoption by AMD, Amazon, and a pivoting Intel looks most realistic in 2027–2028.

Glass Substrate TGV Through-Glass Via Semiconductor Packaging Laser Drilling Yield Advanced Packaging Deep Tech
← Previous · 02
Lead Frame → PCB → ABF → Glass: 60 Years of Substrate Evolution
How each material transition in packaging history triggered a performance leap

Comments

Figure, Tesla Optimus, 1X NEO — What's Actually Changed in 2026

Figure, Tesla Optimus, 1X NEO — What's Actually Changed in 2026 | Paradigm Shift Lab ◆ Robotics  ·  2026 Update Figure, Tesla Optimus, 1X NEO — What's Actually Changed in 2026 Not demos. Not concept renders. Real factories, real production lines, real dollar signs. Here's the honest 2026 scorecard for the humanoid robot race — and why this year is genuinely different. Figure AI Tesla Optimus 1X NEO Boston Dynamics Atlas PSL Editorial  ·  May 2026  ·  ~7 min read Let's be honest for a second. For the past three years, "humanoid robot updates" meant the same thing: a carefully choreographed demo, a funding announcement, and a CEO posting a slow-motion video of their robot picking up a box. Then everyone moved on. 2026 feels different. Not because the robots suddenly got perfect — they didn't. But...

Boston Dynamics' Atlas Is Training for the 2026 World Cup — And I'm Not Sure How to Feel

Boston Dynamics' Atlas Is Training for the 2026 World Cup — And I'm Not Sure How to Feel (Part 1) ◆ Robotics  ·  Part 1 of 2 Boston Dynamics' Atlas Is Training for the 2026 World Cup — And I'm Not Sure How to Feel I genuinely believed robots doing fine, expressive, human-like movement was decades away. Then Boston Dynamics dropped a video this week. I had to revise everything. Boston Dynamics FIFA World Cup 2026 School of Football PSL Editorial  ·  May 2026  ·  ~5 min read Hyundai × FIFA World Cup 2026 Official Partner. That figure in the background isn't a player. It's Atlas. There are things you quietly file away as "not in my lifetime." Not impossible — just comfortably distant. Fusion energy. Mars colonies. And robots doing things that require genuine physical expressiveness. Haircuts. Dance. Sp...

Thermal Management — Overcoming Glass's Achilles Heel

Glass Substrate Series · 06 Thermal Management — Overcoming Glass's Achilles Heel Glass conducts heat 100x worse than silicon. In a world where AI accelerators dissipate over 1,000 watts, that's a serious problem. Here's how the industry is solving it. Glass Substrate Thermal Management Intermediate ~9 min read ← 05. AI Accelerators & Signal Loss Glass Substrate Series 6/10 07. Flatness & Yield » Glass substrates' signal loss advantage over silicon and ABF is now well established. But every strength comes with a trade-off. Glass has roughly 1/100th the thermal conductivity of silicon and about half that of ABF. In an era where a single AI accelerator dissipates hundreds of watts, this weakness could be fatal — unless engineered around. Here's how. Thermal management is one of the last major hurdles to glass substrate commercialization. If TGV yield is the ...