Binary computing has had a spectacular run, but we need to stop pretending that ones and zeros are the only way to process information. We stuck with them for decades because building a physical switch with two clear positions—on or off, conducting or insulating—was straightforward.
It worked, we scaled it, and then we ran face-first into the thermal and physical boundaries of silicon.
Cramming more microscopic switches onto a chip is getting absurdly expensive.
Instead of making smaller binary switches, the obvious alternative is demanding that a single switch do more work.
A research team at the Vienna University of Technology (TU Wien), led by Andrei Pimenov and first author M. Ryzhkov, took that challenge straight to the atomic lattice. In a report published in Communications Materials, the group demonstrated a nanoscale switch that settles into four separate, stable electrical polarization states.
Not two. Four. That immediately doubles the logic states you can pack into an identical physical footprint without begging lithography engineers to split another atom.
The Trouble With Two Sides
Standard ferroelectric materials are straightforward to a fault. Their crystal structure forces their electrical dipoles into one of two opposing directions: up or down, positive or negative. Push them with a voltage, and they flip. Reverse it, and they flip back. It is textbook physics, reliable, and strictly dualistic.
The Vienna group abandoned conventional ferroelectrics and picked up gadolinium manganese oxide, or GdMn2O5. This is an unconventional multiferroic compound built from rare-earth gadolinium, transition-metal manganese, and oxygen. The crystal doesn't play by ordinary structural rules.
Its intricate internal magnetic ordering breaks the spatial and crystallographic symmetries that usually force a material into a simple binary choice.
Because the lattice loses that conventional symmetry, its energy landscape changes entirely, producing four distinct, stable polarization dips where energy settles.
Breaking the Four-Stroke Trap
Finding four states in a lab isn't completely unheard of. The real headache has always been controlling them. Historically, multi-state multiferroics acted like a temperamental four-stroke lawnmower engine: intake, compression, power, exhaust. If you wanted state three, you had to pass through states one and two first. You couldn't hop directly from state four back to state two without running the entire loop.
That rigid sequence is useless for actual computing. Real computational memory requires random access. You need to jump directly to any state instantly, without waiting for the material to cycle through intermediate checkpoints. A sequential cycle is essentially an analog counter pretending to be memory.
The TU Wien team resolved this by leaning hard into the native magnetoelectric coupling of GdMn2O5. In these materials, charge and spin are tangled up together. The magnetic orientation of the manganese and gadolinium ions dictates exactly where the neighboring electrical charges shift.
Pimenov and his team mapped out how targeted electrical pulses shake up that underlying magnetic order.
Instead of nudging the material down a one-way circular track, their tailored pulses pick specific physical pathways through the material.
You send a specific pulse, and the switch steers directly to the target state.
Done. No intermediary steps, no cyclic delay.
Beyond Binary Muscle Memory
We do not need to keep shrinking transistors until quantum tunneling ruins our circuits. Moving from binary to quaternary switching yields two bits of data per cell immediately. That change cuts down on interconnect routing, reduces active circuit real estate, and drops the energy required to shuffle data back and forth across a processor.
GdMn2O5 might not replace every flash drive on the market tomorrow—multiferroic materials bring their own fabrication and temperature hurdles to commercial foundries. But the TU Wien experiment proves that multi-state physical memory is controllable. We do not have to settle for the simple binary toggles we inherited from early telephone relays. The physics allows for richer hardware; we just have to write off the habit of thinking in twos.
Have thoughts on this article?
Send your feedback. Spotted a factual error or typo? Use this form to let us know. We use your feedback to improve our reporting. Thank you!
