Short answer: Sympathetic cooling removes heat from a trapped-ion qubit by cooling a neighbouring ion instead of the qubit itself, so that the qubit's internal state is not disturbed. Two papers, from 2003 and 2021, report cooling a two-species crystal and cooling a long ion chain by laser-cooling a small fraction of its ions. These are laboratory experiments.
Why cool one ion to cool another?
The 2021 Physical Review Letters abstract notes that trapped ions are one of the leading platforms in quantum information science. For quantum computing with large circuit depth and quantum simulation with long evolution time, the authors say it is crucial to cool large ion crystals at runtime without affecting the internal states of the computational qubits. That is the need for sympathetic cooling: the cooling light acts on helper ions, and the motion is shared across the whole crystal through the Coulomb interaction.
What did the 2003 experiment show?
The 2003 Physical Review A abstract reports cooling of a two-species ion crystal containing one ion of each of two species. The species names did not render in the page text we read, so they are not named here. The authors say the cooling transitions of the two species are separated by more than 30 nm, so laser manipulation of one ion has negligible effect on the other even when the ions are not individually addressed. They describe this as a useful system for reinitialising the motional state in an ion trap quantum computer without affecting the qubit information. They also report that the mass difference between the ions enables a method to detect and then eliminate the effects of radio-frequency micromotion.
What did the 2021 long-chain experiment show?
The authors report multi-ion sympathetic cooling on a long ion chain. They use a narrow cooling beam focused on two adjacent ions, and optimise the choice of the cooling ions according to the collective oscillation modes of the chain. They show that by cooling a small fraction of the ions, cooling effects close to the global Doppler cooling limit can be achieved. They describe this as an important enabling step for quantum information processing with large ion crystals, which is their assessment and not a reported computation.
What about the Penning-trap article?
A third source, a Journal of Modern Optics article on sympathetic cooling in two-species ion crystals in a Penning trap, was on the list. Only its title and journal were readable in the page text, so nothing in this post is taken from it.
How to read this
The 2003 work cools between different species, while the 2021 work cools a single-species chain by choosing which ions receive the cooling light, so the two approaches are different. This post is based on abstracts, and full texts were not checked. For a related topic, see our post on ground-state cooling with sideband methods. It makes no claim about products.
FAQ
What is sympathetic cooling?
Cooling an ion indirectly through its Coulomb coupling to another ion that is laser cooled.
Why use two species?
Their cooling transitions are far apart, so the cooling light does not disturb the qubit ion, per the 2003 abstract.
How many ions were cooled directly in 2021?
A small fraction of the chain, using a beam on two adjacent ions.
How close to the Doppler limit did it get?
The authors say close to the global Doppler cooling limit, with no figure in the abstract.
All field notes