NSF Deploys $108 Million to Six New Materials Research Centers — And the Award Size Tells You How to Compete Next Cycle

Academic PIs and NIH-funded researchers watching the federal science topline contract got a rare counter-signal on July 30, when the National Science Foundation announced $108 million for six new Materials Research Science and Engineering Centers — $18 million each over six years — in an NSF news release.

The centers begin work in September 2026. In a year when nearly every federal research headline has been about subtraction, this is one of the few center-scale competitions that not only ran on schedule but funded more awards than NSF said it would.

What $108 million actually bought

NSF's announcement is unusually spare on institutional detail — it says only that the six centers are "housed at U.S. institutions of higher education in five states, including one in the NSF Established Program to Stimulate Competitive Research (NSF EPSCoR) program." Institutional announcements fill in the roster: UC San Diego, Princeton, Harvard, Columbia, MIT, and the University of Nebraska–Lincoln. Five states, two of them in Massachusetts, with Nebraska carrying the EPSCoR slot.

The science is broader than the "materials" label suggests. UC San Diego's center, funded 2026–2032 and jointly run by the Jacobs School of Engineering and the School of Physical Sciences, is organized around quantum metamaterials for optical processing (thrust leads Zhaowei Liu and Richard Averitt) and chemically tailored 2D superlattices for quantum electronics (Joshua Figueroa and Monica Allen), with Tod Pascal leading computational modeling. Columbia's Center for Advanced Electronic Materials targets semiconductors and quantum hardware. Princeton's long-running Princeton Center for Complex Materials pairs quantum materials discovery with soft matter. Harvard's center sits squarely in soft functional materials.

NSF's own summary flags three application threads that should catch the attention of anyone with an NIH portfolio: scintillator materials that make CT scanning more sensitive at reduced X-ray exposure, autonomous AI-driven experimentation labs for soft materials and drug delivery, and hybrid quantum metamaterials. The agency also lists critical mineral extraction, biotechnologies, microelectronics, and advanced manufacturing as downstream targets.

"For more than fifty years, NSF Materials Research Science and Engineering Centers have helped put America at the forefront of advanced materials discovery," said Tie Luo, head of NSF's Mathematical and Physical Sciences Directorate.

The training numbers are the part most PIs will care about operationally. Collectively the six centers will support education, mentorship, and specialized training for more than 60 early career researchers, more than 150 graduate students, and more than 250 undergraduates. That is roughly 460 funded trainee slots created by a single competition — a meaningful number in a year when lab groups are shedding people rather than adding them.

Read the arithmetic: NSF funded six small centers instead of three big ones

This is where the announcement gets genuinely informative, and it takes a look at the solicitation to see it.

The competition ran under NSF 26-500, the current MRSEC solicitation (which superseded NSF 25-532 and moved the full-proposal deadline from November 2025 to January 27, 2026). Preliminary proposals were due June 23, 2025; full proposals were by invitation only. NSF 26-500 estimated 2 to 5 awards against an anticipated $15,000,000 annual program amount.

NSF made six.

The size tells the rest of the story. MRSEC awards under NSF 26-500 run approximately $3 million per year for a two-IRG center up to a maximum of $4.5 million per year for a three-IRG center, for an initial period of up to six years. Every one of these six awards came in at $18 million over six years — exactly $3 million a year, the two-IRG floor.

So NSF did not stretch its estimate by finding extra money. It chose breadth over depth: six two-IRG centers rather than four larger three-IRG centers for the same annual outlay. For PIs planning the next cycle, that is a real signal about how the Division of Materials Research is currently weighing portfolio coverage against concentration — and it argues for building a tight, defensible two-IRG proposal rather than a sprawling three-IRG one.

Two other structural facts from NSF 26-500 are worth committing to memory now, because they govern institutional politics more than science. A MRSEC proposal must contain a minimum of two and a maximum of three Interdisciplinary Research Groups. And only one MRSEC preliminary proposal may be submitted by any one organization as the lead institution — which makes MRSEC a limited submission at essentially every research university in the country. Your internal competition, not NSF's, is the first gate you have to clear.

Why a materials center should be on a biomedical PI's radar

The reflex among NIH-facing investigators is to file MRSEC under "not my directorate." That reflex is expensive.

Two of the three application areas NSF chose to highlight are biomedical. Scintillators for CT imaging are a detector-physics problem with a direct clinical readout — lower dose per scan. Autonomous AI-based experimentation labs for soft materials and drug delivery are, functionally, a closed-loop formulation platform: the kind of infrastructure that a pharmacology or bioengineering lab would spend years and an R01 trying to stand up alone, sitting inside a center that is mandated to share it.

That mandate is the point. MRSECs are required to operate shared experimental facilities and seed-funding programs, and those facilities are open to users outside the host center — including outside the host institution. A center at Harvard or Columbia or Nebraska is not a closed shop; it is a piece of instrumentation infrastructure with a six-year funding guarantee and an explicit external-user mission, at a moment when almost nothing else in the federal system carries a six-year guarantee.

The practical moves for a non-materials PI are unglamorous and effective: get on the external user list for the shared facility that matches your assay, apply to the center's seed program (most MRSECs run small internal competitions specifically to pull in adjacent disciplines), and pitch yourself as a collaborator on the next renewal's IRG. Center renewals are won on demonstrated interdisciplinary reach, and directors are actively shopping for it.

Center-scale money is holding better than the topline

The contrast worth understanding is between program lines and center lines.

NSF's FY 2027 budget request to Congress, transmitted April 3, 2026, proposed a total of $3.96 billion for the agency — a 54.7 percent cut — with Research and Related Activities, the account that actually funds grants, dropping to $3.4 billion, a 52.5 percent reduction below the FY 2026 appropriated level. That is the environment in which this $108 million was obligated.

Multi-year center commitments behave differently from annual program lines under that pressure. They are already scored, already competed, and already carry six years of outyear obligation, which makes them politically and administratively harder to unwind than a rolling program that can simply stop making new awards. It is the same pattern visible at the Department of Energy, which renewed its five National Quantum Information Science Research Centers for $625 million over five years — we covered what that renewal means for researchers and the read is nearly identical: when agencies are squeezed, the center portfolio is the last thing to move.

The strategic conclusion for a PI is not "chase centers." It is that the stability of your funding now depends heavily on whether your work is attached to a multi-year center commitment or to a single-investigator line renewed annually. Those two positions are diverging fast, and 2026 is the year to move from the second to the first if you can.

What to do before the next cycle

The NSF 26-500 competition is closed. The realistic near-term actions are these.

Work the always-open Division of Materials Research programs. DMR's Topical Materials Research Programs — Biomaterials (BMAT), Condensed Matter Physics, Metals and Metallic Nanostructures, and Polymers, under NSF 20-589 — accept proposals with no deadline, as does Computational, Data-Driven, and Theoretical Materials Research under PD 12-8029. No-deadline programs are the correct place to park a proposal while center competitions are between cycles, and BMAT in particular is the natural landing spot for biomedical materials work that does not fit an NIH study section.

Start the internal limited-submission conversation now, not in 2028. One lead preliminary proposal per institution means the internal down-select happens roughly a year before NSF's prelim deadline. If you intend to lead or co-lead an IRG in the next round, your research office needs to know this quarter.

Track the adjacent mechanisms while you wait. Several are live right now with real deadlines — DARPA's Multi Modal Materials Analysis (MMoMA) program closes August 26, 2026, and the Moore Foundation's EPiQS Investigator competition is running unconstrained six-year quantum materials grants that function as a private-sector analogue to a center award.

Search live materials and quantum materials solicitations across NSF, DOE, DARPA, and private funders on Granted: grantedai.com/grants?q=materials+research. Filter for the no-deadline DMR programs first — they are the fastest path from reading this to having something submitted.