How to store x-rays

Light is an excellent carrier of information used not only for classical communication technologies but increasingly also for quantum applications such as quantum computing. However, processing light signals is far more difficult compared to the more common electronic signals.

In quantum computing, much like in classical computing, various processes must be synchronised to each other. For instance, in a multi-thread process, each thread must wait for the others to finish before the process can continue. This renders a memory device essential, where the quantum information, so called qubits, can be stored and released at predetermined times without any loss of information.

In optical quantum computing, qubits are encoded in the various degrees of freedom of photon wave packets, such as their polarisation state, photon number, or waveform. Storing such a wave packet without losing its quantum information is a major challenge for optical quantum applications, since photons are notoriously difficult to control.

Typically, the problem is solved by transferring the quantum information to a long-lived state of a matter system. Researchers have established various “quantum memory” protocols that can force the matter system to re-emit the photon wave packet at a predetermined time allowing the qubit to be read out.

Frequency combs as quantum memory protocol

One particularly appealing quantum memory protocol uses a “frequency comb” structure. In this method, the absorption spectrum of the matter system features a series of evenly spaced atomic resonances, manifesting as absorption lines that correspond to the teeth of the comb. When a photon wave packet gets absorbed by such a comb structure, it simultaneously excites all absorption lines. The subsequent emission of the wave packet then has a remarkable property: The photons are emitted only at moments in time when all transitions radiate in phase so that constructive interference between all transitions occurs. These moments are determined by the energy spacing of the comb lines. Thus, by preparing a certain frequency comb state in the matter system before the arrival of the photon wave packet, the re-emission of the photon wave packet can be well controlled to occur at predetermined time instants without (nearly) any information loss.

This quantum memory protocol has been proved for optical light using strong laser sources to create a frequency comb in an atomic absorption spectrum. This gets increasingly more difficult for light with shorter wavelengths because light sources are much weaker at x-ray energies.

Now scientists led by DESY, together with the ESRF, have overcome this problem by using a novel approach to form a frequency comb. The team used the beamlines ID18 at the ESRF and P01 at PETRA III. Sven Velten, researcher at DESY and corresponding author of the paper, explains: "Efficiently testing our nuclear frequency comb setup required highly monochromatized, intense x-ray pulses. At ID18, we found an ideal setup for this, and the excellent support from the beamline staff was essential for the success of our experiment."  

Sasha Chumakov, scientist in charge of ID18 (now ID14) adds: “We've had a long-term collaboration with the team at DESY since years and our beamline could offer them highest flux and best beam stability they needed in their experiment".

Instead of an atomic transition, they use the nuclear transition of the isotope 57Fe at an energy of 14.4 keV corresponding to a wavelength of 86 pm, thus, at x-ray energies. Nuclear transitions feature extremely narrow energy linewidths. The transition of 57Fe, for example, has an energy linewidth of 5 neV, so nearly 13 orders of magnitude smaller than its transition energy. For these narrow transitions, Doppler shifts via mechanical motions can substantially shift the transition. Therefore, they used mechanical motions to form a frequency comb structure by using multiple moving absorber foils. In the forward emission direction, an x-ray wave packet is coherently emitted, meaning it is absorbed and re-emitted by all foils without loss of phase information. Therefore, the “nuclear frequency comb” allowed re-emission of x-ray photon wave packets at well-controlled time instants while preserving the waveform of the wave packet nearly undistorted.

The synchrotron radiation pulses used in the experiments contained at most one resonant photon at a time (roughly 0.01 resonant photons per x-ray pulse). The ability to work on a single-photon level without loss of information qualifies the nuclear frequency comb as a quantum memory – a first for x-ray energies.

It highlights the potential for applying quantum technologies at short wavelengths regimes where devices can be more compact and flexible while operating at room temperature. The nuclear frequency comb also allows for the formation of “time-bin” waveforms, a specific type of photonic qubit. Manipulating and controlling x-ray wave packets at the single-photon level opens up intriguing possibilities for technical applications such as excitation and detection of ultra-narrow nuclear transitions, as well as advancing the field quantum optics at x-ray energies.

Reference: 

Velten, S. et al, Science Advances 26 June 2024, DOI: 10.1126/sciadv.adn9825

Top image: Time-velocity histogram of the single-photon detection events resulting from the decay of a nuclear polariton spectrally prepared in a frequency comb consisting of seven teeth, equally spaced over a Doppler detuning range extending from -10 mm/s to 10 mm/s (≈ 480 neV). The color map displays the number of detected photons on a logarithmic scale.