Physicists have produced the most intense hard X-ray flashes of their kind yet recorded, a result that should allow electrons inside atoms and molecules to be examined more closely than before.
The work was carried out in April at the European XFEL, the X-ray free-electron laser facility in Germany, and has been published in the journal Optica. The announcement was issued by European XFEL and reported by Phys.org on Friday.
The pulses are an estimated 300 attoseconds long. An attosecond is a billionth of a billionth of a second, and at that scale the relevant comparison is not with anything mechanical but with the motion of electrons themselves.
The achievement was in the focusing. Using an advanced Kirkpatrick-Baez mirror system supplied from Japan, with two mirrors and four reflections per pulse, one focusing the beam vertically and the other horizontally, the team squeezed the flashes into a spot measuring 180 by 330 nanometres. That is roughly 100,000 times smaller in area than the cross-section of a typical human hair.
The result was a peak intensity of more than 10 to the power of 20 watts per square centimetre. The first author, Ichiro Inoue of the University of Tokyo, described it as a world record in intensity and fluence for a hard X-ray attosecond pulse.
The two quantities are not the same thing, and the distinction matters. Fluence is the energy delivered per unit of area. Intensity also accounts for how quickly that energy arrives. A short pulse and a small spot together raise the second far faster than the first.
Verifying a number of that size is not straightforward. The team confirmed it by measuring amplified spontaneous emission from a copper foil: the signal rose steeply with pulse energy and then saturated, as theory says it should at these intensities. Anders Madsen, the group leader at the MID instrument where the experiment ran, described the moment the emission condensed into the forward direction in near-magical terms.
The corresponding author, Jiawei Yan of DESY, noted that terawatt-scale hard X-ray attosecond pulses had been demonstrated at European XFEL a few years earlier. The optics used for this experiment came from Japan; the facility's own system of the same type is expected in early 2027.
The stated purpose is to study electrons in atoms and molecules before the X-rays significantly change the sample, and in the longer term to control the behaviour of electrons during chemical reactions.
That last sentence is the one worth reading twice, because it contains the entire method, and the method is not what it first appears.
The pulse does not avoid damaging the sample. It arrives before the damage does. For 300 attoseconds the atom is in the condition it was in beforehand, because nothing it has been subjected to has yet had time to take effect. The photograph is taken during the interval in which the subject has not been told what is happening to it.
This newspaper regards the principle as sound and notes only that it is unusual to see it stated so openly. The measurement is accurate precisely because the sample has not reacted, and the sample has not reacted precisely because the measurement is faster than the consequences.
There is a corollary, which the Optica paper does not dwell on. A result obtained this way is reproducible only on material that has not previously been measured. Each sample is good for one observation, taken in the instant before it stops being the thing that was observed.
Our analysts, who have spent some time with the figures, find no fault in any of this and wish to record an unrelated observation. 10 to the power of 20 watts per square centimetre, focused into 180 by 330 nanometres, is a considerable amount of attention to direct at a copper foil that has done nothing wrong.
The Department of Unnecessary Research has asked whether the technique generalises. Its provisional finding is that it does, and that the atom is now being photographed in the only condition in which it can be relied upon: before it has had time to find out.