One Copy Won't Brew, Two Already Will ✎ express

Original: "Two-copy distillability of one-copy-undistillable negative-partial-transpose states in every dimension"
arXiv:2608.08836v1 · 2026-08-09 · arXiv non-exclusive · independent analysis · 1 min
It was thought that some quantum states couldn't be purified no matter how many copies you had, but a state has been found that needs only two.

📄 Showing the "Simple" version — "Advanced" is not ready yet. Add it to favorites to help prioritize it.

Coffee beans alone don't make a drink, but take two, and the cup fills. In the quantum world, there are similar states: from one copy, you can't extract pure entanglement, but from several, sometimes you can. For a long time, it wasn't known if such a thing existed at all.

Scientists took a family of states specifically designed as a testing ground for this question. One state in the family turned out to be a surprise: one copy (nothing, two copies) there's a result. And this holds for any system size, starting from three.

Entanglement is when particles are linked so that the state of one instantly determines the state of the other, even at a distance. Purifying entanglement means turning many weak connections into a few strong ones, suitable for precise work.

Next to this state, they found a whole region where two copies already work. A bit further, there's another region where even two copies don't help. Between them lies a band where it's still unknown how many copies will be needed.

Thus, one family of states showed that the boundary between 'hopeless' and 'useful' quantum resources is not a straight line. It bends, and within it there are islands where the second copy makes all the difference.

Entanglement cannot be copied: if you have one entangled pair of particles, you cannot make a second identical one without destroying the first. So the question 'how many copies are needed for purification' is not idle: it's about how to manage a limited resource.

Scientists
William WoottersJohn Stewart BellErwin SchrödingerAlain AspectDavid DeutschNiels Bohr
Concepts
no-cloning theorem