State of matter with a second time dimension makes quantum computers more robust


Adds study co-author Justin Bohnet, a quantum physicist at Quantinuum in Broomfield, Colorado, where the experiments took place. “It’s unlike traditional error correction protocols, where you’re constantly taking measurements on small parts of the system to see if there’s any errors, and then going and correcting them.”

Quantinuum’s H1 quantum processor consists of a string of ten ytterbium ions in a vacuum chamber. Lasers precisely control the positions and states of these ten qubits. Experts speak of an “ion trap” here – a standard technique that physicists use to manipulate ions. In their first attempt to create an error-stable topological phase, Potter, Dumitrescu, and their colleagues attempted to impart simple time symmetry to the processor. To do this, they gave the ions, which are all lined up along a straight line in one dimension, periodic collisions with regularly recurring laser pulses. “Our calculations showed that this [den Quantenprozessor] protect against mistakes,” says Potter. This is similar to a steady drumbeat keeping several dancers in rhythm.

At first nothing worked: The disturbances got worse

To check if they were right, the researchers ran the program several times on the Quantinuum processor, each time checking if the resulting quantum state of all qubits matched their theoretical predictions. “It didn’t work at all,” says Potter, and laughs. “It turned out completely incomprehensible stuff.” Whenever errors accumulated in the system, its performance deteriorated within 1.5 seconds. The team soon realized that just adding time symmetry wasn’t enough. Instead of preventing the qubits from being affected by outside shock and noise, the periodic laser pulses amplified the tiny hiccups in the system and made small perturbations worse, Potter explains.

So he and his colleagues went back to the drawing board, until they finally had an insight: if there were a pattern of pulses that was not random but somehow ordered, but did not repeat regularly, they could create a more resilient topological phase. They calculated that such a “quasi-periodic” pattern could potentially induce multiple symmetries in the processor’s ytterbium qubits while avoiding the unwanted gains. As a pattern, they chose the mathematically well-researched Fibonacci sequence, in which the next number in the sequence is always the sum of the two previous ones. So while a regular periodic laser pulse train might alternate between two frequencies from two lasers as “A, B, A, B,…”, a pulsating Fibonacci sequence would run as “A, AB, ABA, ABAAB, ABAABABA,…”.

Actually, this pattern arises from a rather complex arrangement of two collectives of different laser pulses. Still, according to Potter, the system can be thought of simply as “two lasers, pulsing at two different frequencies,” ensuring that the pulses never overlap in time. For the calculations, the team theoretically imagined that these two independent pulse collectives run along two separate time lines; each collective effectively pulses in its own dimension of time. The two time dimensions can be traced on the surface of a torus, i.e. a donut-shaped structure. The quasi-periodic nature of the two timelines is evident from the way they wrap around the torus over and over again, “at an odd angle that never repeats itself,” says Potter.

When the team implemented the new quasi-periodic sequence program, Quantinuum’s processor was actually protected for the entire duration of the test: 5.5 seconds. “That doesn’t sound like much, but it makes a clear difference,” says Bohnet. »This is clear proof that the demonstration works.«



Source link -69