Control out of the Vacuum, Virtually

How virtual photons alter atomic X-ray spectra

virtual photons
Experimental setup to measure the collective Lamb shift at tantalum. Credit: DESY, Haber et al.

Certain X-ray optical properties of metal atoms can be controlled with the help of virtual photons. This has been demonstrated for the first time by a DESY research team using the highly brilliant radiation from DESY’s X-ray light source PETRA III. In the journal Physical Review Letters they report on how the X-ray spectra of metal atoms can be controlled by virtual photons. This opens up new possibilities for specifically modifying the X-ray optical properties of materials.

So-called virtual photons play an important role in the interaction of light and matter. This is quite remarkable because they do not exist in the classical sense. Virtual photons are created in the vacuum out of nothing and then disappear again after an extremely short time. If these photons interact during their short existence with the electrons of an atom, the binding energies of the electrons shift ever so slightly.

This fundamental effect was first measured on hydrogen atoms in 1947 by Willis Lamb. The eponymous Lamb shift provided theoretical physicists at the time with an crucial evidence of the virtual interactions of photons with matter. This was a breakthrough in the development of quantum electrodynamics (QED), the fundamental theory of the interaction of light with matter. Lamb received a share of the Nobel Prize in Physics in 1955 for his pioneering experiment.

The effect of virtual photon interactions changes fundamentally if many similar atoms are involved. If these atoms are very close to each other, the virtual photons can cause an interaction between the atoms. The resulting collective Lamb shift can be much larger than that of individual atoms and also depends strongly on the spatial arrangement of the atoms. Following its prediction in 1973, however, this collective shift had long evaded experimental proof, since it was not possible to prepare identical atoms in sufficient density.

The DESY group led by Ralf Röhlsberger was able to directly detect this shift for the first time in 2010 with the help of the iron isotope 57Fe. Using modern deposition processes, a large number of atoms of this isotope can be embedded in suitably structured thin-film systems, which act as so-called resonators for X-rays. In these, the interaction of the virtual photons with the atomic nuclei is further enhanced by reflecting the light back and forth between two mirrors mounted around the resonant atoms. Since the resonance line of the 57Fe is extremely narrow, smallest shifts of this line can be measured with high precision. While the collective Lamb shift could be confirmed and investigated in detail with this technique, these studies were limited to the 57Fe isotope.

Proof of the collective Lamb shift around the resonance at the L3 absorption edge of tantalum. Credit: DESY, Haber et al.

In the new work, Johann Haber’s team has now succeeded for the first time in detecting the collective Lamb shift of inner-shell atomic excited states, or resonances. Here, an electron is excited by absorption of a photon from a low-lying energy shell, i.e., an orbit around the atomic nucleus, into a higher energy shell. For this experiment, the researchers used the 9.88 keV L absorption edge of the metal tantalum, which is a particularly pronounced resonance. The atoms were prepared in a very similar fashion as for the experiments with 57Fe. As the measurements show, the collective Lamb shift of the tantalum resonance can be up to three electron volts (3 eV) depending on the excitation conditions and can even change its sign. This opens up numerous new applications since there are a number of elements with sufficiently narrow band resonances.

“These results are of great relevance for high-precision spectroscopy in the X-ray range,” says Sonia Francoual, the scientist responsible for the P09 instrument at PETRA III. In the X-ray resonators of this experiment, standing waves are formed that are similar to those occuring in certain X-ray spectroscopy investigations. These standing waves can also influence the spectroscopic signature of any matter resonance or excited state to which they couple. If standing X-ray waves are combined with spectroscopic investigations, quantum-optically induced shifts might occur which need to be taken into account, lest they affect the results

“In addition, we expect this experiment to lead to further applications of X-ray quantum optics that work with electronic resonances, such as a strong coupling between a resonator mode and an ensemble of resonant atoms,” says Haber. In fact, resonance lines such as tantalum LIII resonance have a number of advantages over 57Fe nuclear resonance. For example, the electronic resonance lines are more than ten orders of magnitude (ten billion times) wider than the nuclear resonance lines. This makes so-called multiphoton experiments much easier, since the X-rays from a synchrotron or an X-ray laser contain many more photons within the bandwidth of electronic resonances. This will open up new possibilities for resonance-amplified nonlinear optics in the X-ray range.

Collective Lamb-shift processes could also be relevant for novel techniques that exploit stimulated emission processes from electronic resonances excited by X-ray laser radiation. Finally, the availability of attosecond pulses in the hard X-ray range also opens up the investigation of quantum-optical effects with highest temporal resolution.