Experiments in quantum mechanics are a little like conversations: the answer you get depends on how you ask the question. That’s made clear in a study of Auger electron emission by an international team of scientists led by Markus Sch?ffler of the Goethe University of Frankfurt in Germany [1]. In this process, x-rays kick out electrons from deep within the core of the electron cloud binding atoms into molecules.

Previous investigations of how the phenomenon works have given apparently conflicting results, but Sch?ffler’s team shows that the discrepancy can be understood as a natural consequence of quantum behaviour, where the act of measurement can determine the result.

A high-energy x-ray photon hits a molecule, in this case nitrogen (N₂), and knocks an electron (called the photoelectron) out of the inner shell, where the electrons of lowest energy reside. This creates a hole which is filled when an outer-shell electron ’falls’ down into it, shedding energy that is carried away by a second electron (the Auger electron) ejected from the outer ’valence’ shell.

The researchers set out to investigate the nature of the short-lived hole, residing in a molecular orbital in the inner shell. Is it localized around one of the two nitrogen atoms, or delocalized around both? Valence electrons are delocalized - that’s how they bind atoms together. But for core electrons, the quantum theory of chemical bonding can accommodate either a delocalized or a localized picture.

A delocalized hole creates two different molecular-orbital states separated by a small energy gap, and in 2006 a Japanese team found experimental evidence for such a gap [2]. But studies last year by another Japanese group suggested on the contrary that the hole is localized [3]. Sch?ffler’s team now concludes that both groups were right - because they measured the effect in different ways.

The researchers looked at how the emission intensity of both the photoelectron and the Auger electron differs in different directions relative to the nitrogen molecule’s N-N axis. They found that these two electrons are in a quantum entangled state, meaning that that the behaviour of one is affected by measurements on the other. As a result, the direction-dependence of the photoelectron can look consistent either with a localized or a delocalized hole, depending on the direction in which the Auger electron is emitted. In other words, entanglement means that the measurements themselves determine what the hole’s molecular orbitals look like.

Because core orbitals play little part in bonding, Sch?ffler says the findings don’t directly affect current ideas about chemical bonds. But they are relevant at a deep level, says Kiyoshi Ueda of Tohoku University’s Institute of multidisciplinary research for advanced materials in Japan. ’Molecular orbitals of core electrons are like the foundations of a house,’ he explains. ’People who live in the house don’t normally notice them, but the house can’t be built without them.’

Philip Ball