NSF ELEGANT Puts 2D Semiconductors on an Other Transaction — and Forces You to Pick a Lane Before You Write a Word

On September 17, 2026, the National Science Foundation announced NSF ELEGANT — the Ecosystem for Low-dimensional Electronics: Growth, Assembly, Nanoelectronics, and Translation. The program's stated job is to take low-dimensional semiconductor materials out of the paper-and-demo stage and into something a U.S. fab can actually commit capital to.

The news coverage has focused on the technology. That is the less interesting half. The more consequential half is the instrument: NSF did not issue a solicitation. It issued an Other Transaction Agreement Solutions Offering (OTASO), posted to SAM.gov under notice ID fc4611184ad340358c2314151865aef8, seeking two structurally different kinds of respondent — a single Ecosystem Coordinator and a set of Expert Performer Teams across five translation themes.

And the offering contains a sentence that should reorganize your week if you work in this space:

An organization may submit a proposal as Ecosystem Coordinator or as an Expert Performer Team, but not both.

That is a one-line fork in the road, and most institutions will discover it after they have already started drafting.

What "low-dimensional" actually means here, and why translation is the hard part

Low-dimensional electronics means materials that are effectively one or two atoms thick — graphene, transition metal dichalcogenides like MoS₂ and WS₂, and their relatives. The physics case has been settled for the better part of a decade: as silicon transistors approach the point where the channel is only a few nanometers long, conventional silicon starts leaking, and an atomically thin channel does not. These materials are the leading candidate for what comes after the current scaling regime.

The physics case being settled is precisely why this is a translation program and not a research program. NSF has funded the science for years — the Penn State 2D Crystal Consortium Materials Innovation Platform, the EFRI 2-DARE program, hundreds of CAREER awards. What has not happened is the boring, expensive, unglamorous work that converts a lab result into a manufacturing decision: growing uniform films across a 300-millimeter wafer, controlling defect density at production volumes, and — critically — producing a process design kit that a circuit designer can actually use.

NSF Assistant Director for Technology, Innovation and Partnerships Erwin Gianchandani framed the goal in terms that name the real bottleneck:

NSF ELEGANT aims to de-risk advanced microelectronics manufacturing technology to the point where industry can make a timely, credible, evidence-based commitment to its adoption.

Read that carefully. The deliverable is not a demonstration. The deliverable is evidence sufficient to justify an industrial capital commitment. That is a much higher bar than a paper, and it is a materially different success metric than anything NSF's core programs reward.

The five translation themes

The OTASO seeks Expert Performer Teams in one or more of five themes:

1. Manufacturable materials platform — growth and transfer of low-dimensional films at wafer scale with production-relevant uniformity and defect control.
2. Process technology and process design kit — integration-compatible process modules, plus the PDK that lets outside designers use them.
3. Device, circuit and full-chip demonstration — moving up the stack from single devices to working circuits to a complete chip.
4. Manufacturing readiness and transition — the industrial handoff: yield, metrology, reliability, tool compatibility.
5. AI-enabled acceleration — using machine learning to compress the experimental search across the other four.

Notice the shape. Themes 1 through 4 are a sequential value chain, not four independent research topics. A materials platform with no process design kit is unusable. A full-chip demonstration built on a materials platform nobody can reproduce proves nothing. Theme 5 is a cross-cutting accelerant applied to the whole chain.

That sequencing is the reason the Ecosystem Coordinator exists at all, and it is the key to reading the program correctly.

The Coordinator role is not administrative overhead

Per NSF, the Ecosystem Coordinator "oversees the ELEGANT ecosystem, managing NSF oversight support, facilitating collaboration among Expert Performer Teams, and ensuring progress toward program objectives and translation endpoints."

In a conventional NSF center, that function is folded into the lead institution's management plan. Here it is broken out as a separately competed award — which tells you NSF expects the coordination problem to be hard enough, and the inter-team dependencies tight enough, that it wants a dedicated entity whose incentives are aligned with the whole chain rather than with any single link.

This is the same architectural instinct visible across NSF's Technology, Innovation and Partnerships directorate this year. X-Labs uses Other Transactions with Phase 0 formation funding and milestone gates before scale-up — we covered the mechanism in NSF X-Labs: How the $1.5B Quantum & Instrumentation Initiative Rewrites the Federal Funding Playbook, and the completed 2026 topic portfolio in our September 18 analysis. The $20 million deep-tech commercialization pilot announced September 14 runs milestone tranches too. TIP is converging on a house style: Other Transactions, milestone gates, explicit translation endpoints, and an orchestration layer that is competed rather than assumed.

If you have only ever written NSF proposals under the PAPPG, understand what changes. An Other Transaction is not a grant and not a procurement contract. It sits outside the Federal Acquisition Regulation and outside most of 2 CFR 200. Terms are negotiated per agreement. The upside is speed and flexibility — non-traditional performers, unconventional IP arrangements, milestone-based payment. The downside is that the familiar protections and the familiar cost principles are not automatically there. Your sponsored programs office will need to read the agreement, not a policy manual. Budget real time for that.

What NSF has not published yet — and how to handle it

Be clear-eyed about the information gap. As of this writing, NSF's public announcement and program page do not state a total program budget, per-award ceilings, a period of performance, cost-share requirements, or a proposal due date. Several outlets have covered ELEGANT without those numbers because those numbers are not in the press release.

The authoritative source is the SAM.gov notice itself, fc4611184ad340358c2314151865aef8. The NSF program page lists Michael Huff (mahuff@nsf.gov) as the contact. Do not build a schedule on a secondhand summary of this program — including this one. Pull the offering document.

What NSF has published on eligibility is usable now:

• Proposers must be U.S.-based incorporated entities.
• Eligible types: institutions of higher education, for-profits, nonprofits, or formally established consortia.
• Active SAM.gov registration is required before submission — not at award.
• Coordinator or Performer, not both.

That third bullet deserves emphasis independent of everything else. SAM registration is the single most common reason a team that did the technical work misses a federal deadline. If your entity's registration has lapsed, renewal can take weeks. Start it today, before you decide whether you are bidding.

The strategic read: three decisions, in order

Decision one: Coordinator or Performer. This is exclusive and it is the highest-leverage choice in the program. The Coordinator role rewards convening capacity, neutrality, and program management credibility — it fits consortia, industry associations, and institutions with existing multi-party microelectronics infrastructure. The Performer role rewards deep technical capability in a specific theme. An organization with a world-class 2D growth facility that bids Coordinator has traded its actual comparative advantage for an administrative seat. An organization whose real asset is a standing industry consortium that bids Performer will lose to a specialist lab.

The failure mode is bidding Coordinator as a hedge because it feels lower-risk. It is not lower-risk. It is a different competition with a different and probably smaller applicant pool, evaluated on criteria most research institutions are not optimized for.

Decision two: which theme, and how far down the chain. Because themes 1 through 4 are sequential, theme choice is also a choice about who your neighbors are. A team bidding Theme 2 — process technology and PDK — is structurally dependent on whoever wins Theme 1. If you bid the middle of the chain, your proposal needs a credible answer to "what if the upstream materials platform is not the one you assumed?" Teams bidding Theme 1 have the opposite problem: you must show that your platform is reproducible by people who did not invent it, because the entire downstream ecosystem will have to.

Decision three: what your industry evidence actually is. Gianchandani's standard is an "evidence-based commitment" by industry. Letters of support from a fab that has read nothing are worthless against that bar. What counts is a partner that has run your material on its own tools, or a metrology house that has independently characterized your films, or a designer who has used a draft version of your PDK and will say what broke. If you have that relationship today, it is your proposal. If you do not, the honest question is whether this offering is yours to win at all this cycle — and whether your better move is positioning for the next one.

Where ELEGANT sits in the 2026 landscape

Context matters here, because this announcement arrives during the worst NSF grant year in four decades — the agency is tracking toward roughly 6,100 new awards in FY2026, about 46% below its 2021–2024 average, with more than $1 billion held in a central account outside core programs. We covered that in NSF's $1 Billion Central Account.

ELEGANT is on the winning side of that reallocation. Money is not leaving NSF; it is moving from investigator-initiated core awards toward large, directed, translation-oriented instruments run out of TIP. For a materials physicist whose DMR proposal has been sitting unfunded since spring, that is cold comfort. For the same physicist with a manufacturing partner and a wafer-scale story, it is the clearest funding signal NSF has sent all year.

It also lands against a semiconductor policy backdrop that has gotten considerably more transactional — see our analysis of the CHIPS Act equity-stake renegotiation. ELEGANT is the research-side complement to that industrial policy: CHIPS money builds fabs for the nodes we know how to make, and ELEGANT is an attempt to have something to put in them afterward.

What to do this week

1. Pull the SAM.gov notice (fc4611184ad340358c2314151865aef8) and read the actual terms. Every number in circulation right now is secondhand.
2. Verify your SAM registration is active. If it is not, that is your critical path, not the technical narrative.
3. Make the Coordinator/Performer call explicitly, at the leadership level, and write down why. Do not let it get made by default by whoever starts drafting first.
4. Email Michael Huff (mahuff@nsf.gov) with specific structural questions. Program officers answer scope and eligibility questions readily, and the answers this early are worth more than they will be later.
5. Inventory your industry evidence honestly. Not relationships — evidence. Someone outside your lab who has touched your material and will document what happened.

The technology in ELEGANT has been almost ready for years. The program is a bet that the missing ingredient was never the physics. Write like you agree.

Granted tracks NSF Other Transaction offerings, SAM.gov notices, and federal microelectronics funding alongside 400,000+ opportunities. Search NSF programs or start a free account to get matched.