Replies to post #225519 on Lightwave Logic Inc (LWLG)

[tedpeele]

Here you go Fruno. Chatgpt searched at least 20 industry documents/websites, to give me an answer to this question.:

My last post for the day so walter is off the hook for now..

Question: "Is it accurate to say that historically - meaning 5 years ago or more - the industry and expert opinion with regard to the likelihood that plasmonic modulator technology being used at scale in data centers was that it would take a decade or more? Is it also accurate to say that many of those same people would say today that it still is a number of years off, maybe even a full decade? Please explain with clear examples."

The answer is consistent with what I posted hear earlier in the year -- still far off, with some exceptions for 'niche' application - for the lazy - check out the summary at the end:

Short answer: Yes — with a big caveat.
Five or more years ago many roadmaps and experts treating plasmonic modulators as a distinct, non-mainstream technology expected practical, at-scale deployment in data-centers to be many years away — often a decade or more because of integration, packaging, loss/cost and reliability challenges. And today you see the same pattern: exciting lab/early-startup demos but mainstream industry and majority of review papers still treat plasmonics as promising but not yet ready for wide data-center deployment, i.e. “still several years off” (for many, plausibly a decade) — even as a few startups and research groups push toward engineering samples and niche products. Below I explain and give clear examples + citations.

Why experts were cautious (˜5+ years ago)

Roadmaps and review articles from the late-2010s / 2020s treated plasmonic modulators as an attractive high-speed option but stressed the hard engineering problems that delay real data-center use: high optical loss of plasmonic waveguides, complicated materials and processing, packaging and thermal/reliability qualification, and the need to fit into existing silicon-photonics foundry/ecosystem. That combination is precisely why many roadmaps pointed to uptake near the end of the decade rather than “next year.”
dev.photonicsmanufacturing.org
+1

Concrete examples (historical)

The IPSR-I / Integrated Photonic Systems Roadmap (international effort, 2020) lays out growth/technology waves for datacenter transceivers through 2030, and explicitly treats “alternative materials for the modulator function” (i.e., not standard plasma-dispersion silicon modulators) as part of the long-term trajectory — in other words, commercial replacement of the incumbent modulator tech was expected on a multi-year-to-decade timescale.
dev.photonicsmanufacturing.org

Review/perspective articles on integrated electro-optic modulators (e.g., AIP/JAP perspective 2021 and other reviews) catalog many promising device approaches (plasmonic, EO polymers, ferroelectrics, III–V on Si) but emphasize integration, packaging, aging and thermal issues as key obstacles that normally make vendors and datacenter operators wait years before adopting non-incumbent device types. That literature is why many expert readers thought “a decade” was a plausible adoption window.
AIP Publishing
+1

Why people still say “years off” today

The basic practical barriers haven’t magically disappeared: you still need wafer-scale manufacturing flows, packaging and test flows that meet datacenter cost/reliability targets, transceiver-level integration (modulator + driver + detector + optics), and standards/market incentives to switch. Recent road-mapping and review work still stresses those system-level gaps. So mainstream industry consensus remains cautious.
Nature

Lab breakthroughs and startup activity have accelerated (e.g., ETH Zurich groups and spinoffs such as Polariton have demonstrated very high-bandwidth plasmonic modulators and are offering engineering samples), but the reporting is explicit: these are engineering samples and record demos, not large-volume, fully-qualified datacenter transceivers. That’s the exact reason many experts who were cautious five years ago still describe broad datacenter rollout as “years away” — the devices are progressing but the system-level commercialization work remains.
optica-opn.org
+1

Concrete examples (today)

Optica / OPN and IEEE reporting in 2024–2025 highlight terahertz-bandwidth plasmonic modulators and mention that companies/universities have engineering samples or small-qty offerings, while simultaneously noting the larger engineering and ecosystem work required before replacement of incumbent modulators in datacenter transceivers at scale. Those articles are a good snapshot of “lab?startup” progress plus continued caution.
optica-opn.org
+1

Recent review papers (2023–2024) on silicon photonics roadmapping still list material / integration / packaging / reliability as the bottlenecks for next-generation modulators — the same reasons that kept timelines measured in years previously.
Nature
+1

Bottom line — concise synthesis

Accurate as a general statement: Historically (˜2018–2020) many industry roadmaps and expert reviews treated plasmonic modulators as promising but likely not to become a broadly-deployed, datacenter-scale technology within a couple of years — rather they expected the heavy lifting to happen across the 2020s (i.e., a decade-scale horizon).
dev.photonicsmanufacturing.org
+1

Accurate today, with nuance: After a wave of impressive lab results and a few startups shipping engineering samples, many of the same experts remain cautious — they see clear potential but still call full-scale datacenter adoption “several years off” (and for conservative estimates, still a decade in many scenarios). At the same time, a small but important group (academic teams and deep-tech startups) is trying to accelerate that timeline — so there is a credible path to earlier, niche commercialization even if general, large-volume adoption may be delayed.

Summary in plain terms

Five years ago, experts succinctly expected plasmonic modulators to be many years away from datacenter volume use; they viewed them as “promising future” rather than “imminent mainstream”.

Today, while performance milestones are being hit, the same experts still emphasise that the “system integration ? cost/yield ? qualification ? volume” path remains long, so mainstream datacenter deployment remains several years off (often five-ten years or more in many scenarios).

Thus your characterization is accurate: historically the view was a decade+ timeframe, and today many of those same voices would still say “a number of years off” (potentially a decade) for large-scale datacenter use of plasmonic modulators.