News Focus
News Focus
icon url

tedpeele

03/10/26 6:11 AM

#230866 RE: tedpeele #230865

More q&a: isn’t it true than the light waves per ketamine material never even has to touch the foundry? If that’s true, then couldn’t the same be said for thin lithium bait and other materials? I’m not sure why the comment that those materials are not allowed to enter the boundaries was made if the reality is that per canine isn’t really entering the foundries either.

This is the "million-dollar question" in the photonics industry, and you’ve touched on a nuance that most investors and even some engineers overlook.

The short answer is: You are correct that the polymer itself never touches the foundry. However, the reason Thin-Film Lithium Niobate (TFLN) is "banned" while polymers are "welcomed" isn't about the material itself—it’s about how the material is physically integrated into the chip.

Here is the breakdown of why this distinction matters so much for mass production:

1. The "Etch" vs. "Slot" Difference


The biggest hurdle isn't just "presence" in the building; it’s the process steps required to make the material work.

TFLN (The Subtractive Process): To make a TFLN modulator, you usually have to etch the Lithium Niobate. Lithium Niobate is notoriously difficult to etch; it requires specialized "dry etching" chemistry that can contaminate a standard silicon line. Even if you do it in the "Back-End," the tools needed to shape it are not standard.

LWLG (The Additive/Passive Process): Lightwave Logic uses a "Slot" design. The foundry builds a tiny trench (the slot) using standard silicon. The foundry does zero work on the polymer. They simply ship a standard silicon wafer with empty holes in it. The polymer is then "spun-on" (like a liquid coating) in a separate facility (like LWLG’s Denver lab or a specialized packaging house).

2. The "Cross-Contamination" Risk

Foundries like TSMC or GlobalFoundries have "Cleanroom Grades."

Lithium Niobate contains Lithium, which is a mobile ion. In a high-end fab, Lithium is considered "poison." If even a few atoms of Lithium migrate into the silicon transistors of the Front-End, the entire $200 million batch of wafers can be ruined.

Perkinamine® Polymers are organic (carbon-based). While foundries still don't want them in the "Front-End" (where transistors are made), organic materials are already common in the "Back-End" (as photoresists or packaging insulators). Foundries are much more comfortable with organic "spin-on" materials because they don't migrate through the silicon and "kill" the transistors.

3. Comparison of Integration

Feature Thin-Film Lithium Niobate (TFLN) LWLG Perkinamine® Polymer
Form Factor Solid crystal "wafer" or "die" Liquid/Spin-on material
Integration

Heterogeneous Bonding: Must be "glued" or bonded onto the silicon, then etched.

Simple Coating: Fills existing holes (slots) already made by the foundry.
Foundry Effort Requires new, non-standard tools and "clean" vs "dirty" line separation. Requires zero new tools. The foundry just makes standard "holey" silicon.
Thermal Match TFLN expands/contracts differently than silicon (causing cracks). Polymers are flexible and "buffer" the stress of temperature changes.


Why the "Allowed" Comment from Yves Matters

When Yves (the Luceda CEO) says their material is "allowed," he means the workflow is allowed.

READ THIS:

If a customer goes to a top-tier foundry and says, "I want to bond Lithium Niobate dies onto your wafers," the foundry often says, "No, our robots and etch tools aren't set up for that material."

But if they say, "I want to use the SilTerra/Luceda PDK to etch standard silicon slots, and I'll handle the coating later," the foundry says, "No problem—that’s just standard silicon processing to us."

The Reality Check

You are right to be skeptical: TFLN is moving toward "Back-End" integration (called Heterogeneous Integration). However, TFLN still requires a "bond-and-etch" or "micro-transfer" step, which is mechanically complex and expensive.

LWLG's "Spin-on" approach is essentially the "spray paint" of the chip world—it’s faster, cheaper, and uses the same machines used to apply basic coatings on every chip in your phone.
icon url

prototype_101

03/10/26 7:32 AM

#230869 RE: tedpeele #230865

that's right teddyturner, I am glad to see you made the shift to Gemini, it is the better AI tool for this type of analysis, I guess you finally recognized that from my use of Gemini, also your post is great confirmation of my post from yesterday, here

Understanding the Production run being done at one of the Foundries already, first here is what investors learned in Ghent

Cohibas - One customer is producing transceivers for 1.6 terra bit per second. First devices are expected back from the foundry by end of Q2!!!! Then they coat and protect it in Denver and we have a first product ready for sale. Still happening this year.
https://investorshub.advfn.com/boards/read_msg.aspx?message_id=177371626

KCC - did he say they initiated a production order for the first Stage 3 partner targeting 1.6T? A production order is more than just an R&D trial run.
https://investorshub.advfn.com/boards/read_msg.aspx?message_id=177371680

Ok, now to understand what is really happening read the following >>

The wafers being sent to Denver for Back-End-of-Line (BEOL) processing are far more advanced than just loose modulators; they represent the transition from a "material science project" to a functional Optical Engine.

In the context of the Fortune Global 500 partnership (specifically the Asian Tier-1 partner), the Denver facility serves as the critical "integration hub" where Lightwave Logic's (LWLG) Perkinamine® polymer is added to pre-fabricated silicon wafers.

What is on the Wafers?
When the wafers arrive in Denver, they are "Base Silicon" Photonic Integrated Circuits (PICs). These wafers have already been processed at the primary foundry (such as GlobalFoundries or AMF) and contain the static optical components:

Passive Waveguides: The "pipes" that guide light.

Grating Couplers: The ports that allow light to enter and exit the chip.

Germanium Detectors: For receiving optical signals.

"Empty" Modulator Slots: Mach-Zehnder structures that are missing their active material.

The Denver BEOL Process: Creating the "Optical Engine"
In Denver, the "magic" happens. The wafers undergo Spin-Coating and Poling, which transforms the passive silicon into an active Optical Engine:

Polymer Deposition: The Perkinamine® polymer is spin-coated across the wafer, filling the "slots" in the modulators.

Poling: A high-voltage field is applied to align the polymer molecules, enabling the Pockels Effect (ultra-fast switching at low voltage).

Encapsulation: The polymer is sealed to protect it from the environment.

Modulator vs. Optical Engine vs. Transceiver
To answer your specific question on the stage of integration:

It is NOT just modulators: The wafer contains thousands of individual modulator devices already etched into the silicon.

It IS an Optical Engine: After the Denver process, the chip is a complete "engine"—it can take an electrical signal and convert it into a 200G/lane optical signal.

It is NOT a full Transceiver yet: A full 1.6T transceiver requires additional components that are added after the wafer is diced into individual chips. These include:

The Laser: Usually a separate Indium Phosphide (InP) chip.

The ASIC/DSP: The "brain" that processes the data.

The Housing: The metal pluggable or CPO module.

https://investorshub.advfn.com/boards/read_msg.aspx?message_id=177373680
Bullish
Bullish