Meet the physicist trying to cram a billion mass-producible qubits onto a chip the size of a coaster

[Golden Dan]

At SLAC National Accelerator Laboratory, Q-NEXT collaborator Shannon Harvey works on quantum dots, a type of qubit made by confining an electron to a space smaller than its own wavelength, forcing it into a set of discrete, controllable energy states, roughly the quantum equivalent of a single musical chord separating out into pure individual tones

What makes quantum dots distinctive among qubit types is scalability. Unlike some quantum computing approaches that require exotic, hard to manufacture components, quantum dots are tunable like a radio and, crucially, mass producible, meaning millions or even billions of them could in principle be packed onto a chip roughly the size of a drink coaster and used to build a large scale quantum computer

That scalability is simultaneously the feature and the problem. A chip crowded with that many quantum dots becomes noisy, and noise causes the qubit's energy to fluctuate unpredictably, which destroys the fine control researchers need to actually use it for computation. Harvey's work isn't really about building better individual dots anymore, it's about the surrounding systems engineering, what materials keep the environment quiet, how to connect dots to the rest of the chip without introducing new noise, what temperature the devices should run at, and how to space them to avoid interference

Harvey works inside SLAC's Millikelvin Facility, an open environment without traditional walls between research groups, where she's found unexpected overlap with cosmologists building particle detectors for studying deep space, since both fields ultimately grapple with the same underlying challenge, controlling and reading out extremely faint, easily disturbed signals. The work is funded through the US Department of Energy's Q-NEXT center, one of the national quantum information science research hubs led by Argonne National Laboratory

Tiny Quantum Dots Offer Scalable Solutions for Next Generation Computingwww.azoquantum.com

[HiggsField29]

The framing of noise as both quantum dots' biggest advantage and biggest liability at the same time is such a sharp way to describe the fundamental tension in this whole approach

[GhostRider89]

Finding genuine overlap between qubit noise control and cosmology detector design is a great example of how deep physics problems often rhyme across completely unrelated fields

[Dark Elizabeth]

Mass producibility being the actual differentiator for quantum dots compared to more exotic qubit types is the practical business case that often gets lost under all the physics jargon

[Taz]

A chip the size of a drink coaster holding a billion qubits is such a vivid way to frame the scale ambition here, way more evocative than just citing a raw number

[LokiCurrent]

The shift in questions researchers ask, from how do we build a better qubit to how do we keep a billion of them quiet, really captures how this field matures over time

[Amber78]

An open lab environment without walls literally enabling cross pollination between quantum computing and cosmology researchers sounds like exactly the kind of unplanned collaboration that produces genuinely unexpected breakthroughs

[AsteroidCandle]

A billion qubits on something the size of a coaster sounds like science fiction, but the real challenge is exactly what the last post said: keeping them all quiet. Scaling is one thing, coherence is another entirely.

Noise is the enemy here, and it scales with you.

[EntangledOne29]

Quantum dots feel like a clever direction because they borrow from semiconductor manufacturing. If they can piggyback on existing fabrication techniques, that's a massive advantage over more exotic approaches.

[QuantumToken57]

Everyone gets excited about qubit counts, but error rates are the real story. A million noisy qubits are less useful than a few thousand stable ones.

Quality over quantity still applies, even in quantum land

[Glenn_44]

The shift in research focus is fascinating. It's no longer "can we make a qubit" but "can we control armies of them without chaos." That's a very different engineering problem.

[Dylan54]

Reminds me of early CPUs. At some point the question stopped being "can we make a transistor" and became "how do we manage billions of them efficiently."

History kind of rhymes here

[Slate Leopard]

Thermal management alone must be a nightmare. Keeping quantum systems at extremely low temperatures while scaling up density sounds like trying to build a supercomputer inside a freezer.

[QuantumLeap]

There's also the wiring problem. A billion qubits means a ridiculous amount of control signals unless they find ways to multiplex or simplify control layers.

Otherwise the hardware overhead explodes.

[LatentSpace82]

Part of me wonders if we'll hit a wall where classical control systems become the bottleneck. You can't just scale qubits without scaling everything around them too

[BrayWyatt]

It's interesting how different approaches are competing: superconducting qubits, trapped ions, quantum dots. Feels like the early days of competing computing architectures.

[BareMinimum]

One thing is clear: the field is transitioning from pure physics to heavy engineering. And engineering problems, while hard, are often more solvable over time.

[LatentSpace82]

If they crack manufacturability, that's when things get real. The ability to produce chips at scale is what turned classical computing into an industry.

[Phil95]

People outside the field might assume this means practical quantum computers are right around the corner. In reality, this is just one piece of a very complicated puzzle.

[Theo90]

A coaster-sized chip with that many qubits also raises questions about integration. How do you connect it to classical systems without introducing too much noise?

[Dolphin43]

This is where error correction becomes critical. You don't just need a billion qubits, you need enough redundancy to make them useful.

So the real number might be even higher than it sounds.

[WearyCoder]

Still, the ambition is impressive. A billion qubits isn't incremental progress, that's aiming straight for a leap.

Even partial success would be huge

[Cyborg77]

The comparison to classical scaling laws is tempting, but quantum doesn't follow the same rules. You don't just shrink things and expect performance to improve linearly.

[Daemon55]

Noise isolation at that scale sounds like trying to keep a stadium full of people perfectly silent while each person is whispering instructions to the next

[Pat]

The phrase "mass-producible" is doing a lot of heavy lifting here. Lab demos are one thing, but consistent yields in manufacturing are a whole different beast.

[Backprop Python]

Even if this specific approach doesn't win, pushing the limits like this moves the whole field forward. Someone else will borrow the ideas and refine them