Most mass spectrometers handle a few molecules. This new prototype handles billions
Most mass spectrometers work like trying to catch fish from Niagara Falls with a single bucket. That's Brian Chait's analogy, and he should know — he's been working with the technology since the 1990s. The century-old technique is one of biology's most powerful analytical tools, capable of identifying and quantifying molecules by their mass-to-charge ratio. But most instruments still analyze molecules one at a time or in very small groups. Rare but important signals get lost in the noise. The technology is bottlenecked.
Chait and his colleague Andrew Krutchinsky at The Rockefeller University think they've found a way through. Their prototype, called MultiQ-IT, can trap and analyze over a billion ions simultaneously — and the team reports up to ten billion ions, roughly a thousand times the capacity of conventional ion traps. The key was borrowing an idea from biology itself.
The team spent a decade studying how molecules move in and out of a cell's nucleus through structures called nuclear pore complexes. These aren't single-file channels — they're hundreds of tiny openings that handle traffic in parallel. "It was a very obvious idea," Krutchinsky says. "But how to do it with mass spectrometry wasn't obvious."
The result is a cube-shaped ion-trapping chamber lined with hundreds of electrically controlled openings. Ions enter, slow down through collisions with residual gas molecules, and distribute across the chamber rather than piling up at a single inlet. The system can then filter, hold, and redirect multiple ion populations at once. The team scaled the design from six ports to over a thousand, testing how efficiently ions could be confined and sorted across increasingly parallel configurations.
The performance numbers are striking. In tests, the 486-port version held over a billion charges at once — the team reports up to ten billion. By allowing abundant background molecules to escape while retaining rarer, information-rich ions, the system improved signal-to-noise ratios by as much as 100-fold — making proteins that were previously undetectable suddenly visible. The trick: applying a small electrical voltage barrier so that singly charged ions (often less informative) escape while multiply charged ions (frequently more biologically significant) remain trapped.
The parallelization analogy is the part that should make tech readers lean in. "What revolutionized DNA sequencing wasn't any change in the underlying chemistry," Chait says. "It was the ability to run so many chemical reactions in parallel, which took genome sequencing from a billion-dollar effort to something that costs around $100. The same thing happened in computing with GPUs. And that's what we're trying to do with mass spectrometry."
The implications extend to drug discovery and proteomics. Unlike DNA, proteins and metabolites cannot be amplified — the most abundant species in a sample may be millions of times more prevalent than the rarest. Current mass spectrometry often lacks the sensitivity to detect those low-abundance signals. If parallelization can push that sensitivity higher, researchers could potentially read the full molecular contents of a single cell and track thousands of chemical reactions simultaneously.
David Clemmer, a chemist at Indiana University who wasn't involved in the work, calls the approach "truly parallel mass analysis." The significance, he says, is that researchers "would have the chance for true discovery" without having to preselect what to study. "Nature doesn't stop and select things one at a time to look at. It does things all at once all the time."
There is a competitive context the Rockefeller team enters. In October, Waters Corporation launched a charge detection mass spectrometry product based on work by Martin Jarrold and Clemmer's Indiana University colleague. Waters acquired that technology in 2022. Clemmer suggests that pairing parallelization approaches with charge detection's ability to measure enormous molecules — including the protein complexes that are the machines of the cell — could advance the whole field. His estimate for the payoff: "an immediate 10 to 20 year horizon where we start to be able to deal with biological complexity at the next level."
For now, MultiQ-IT is a proof of concept, not a product. The team sees its role as establishing a physical blueprint — establishing that the approach works — with the development path from prototype to commercial instrument still ahead. "There was a lot of development between the discovery of a reaction for sequencing DNA and modern genomics; decades between the first transistor and putting a billion transistors on a chip," Chait says. "In both cases, someone first had to show it could be done, and then industry took over."
Whether that path leads anywhere depends partly on whether the parallelization claims hold up under broader scrutiny. But the analogy the team draws — to a pattern that has repeatedly reshaped technology when applied to information — is worth sitting with. GPUs, DNA sequencing, and now mass spectrometry. The same idea, arriving late to biology's most complicated measuring problem.
Sources: