The states handled by quantum computers are vulnerable to external "noise." An international research team has mathematically defined the conditions under which a "quantum catalyst," an auxiliary resource repeatedly used as a catalyst in quantum computation, can maintain its effectiveness even when faced with realistic errors.

UNIST announced on March 4 that a study led by Professor Seok-Hyung Lee of the Department of Physics found that even a very small error in the process of preparing the input state could gradually transform the catalyst in many previously proposed quantum catalyst scenarios, shaking the premise of “reuse.”

The solution the researchers focused on was a "catalytic channel." This approach structurally prevents the catalyst from changing as it wears out during computation by designing computations so that the catalyst always returns to its original state (restoration) regardless of the input. The research paper demonstrated that only this framework can guarantee a generally noise-resistant catalytic process.

However, it was also pointed out that even if a catalytic channel is used, not all quantum resources generate 'additional gains'. The research team discussed the impossibility theorem, which states that it is difficult to obtain fundamental benefits even with catalytic channels in representative resource theories such as entanglement and coherence, while also discussing that the effect of catalytic channels can be established even in the presence of noise under certain thermodynamic conditions.

The results of this study are confirmed in a paper published in Physical Review Letters (Volume 136, Paper No. 050202) and in an open access version available on arXiv. The PRL journal page includes the paper title, author, and publication information.