Quantum Memory for Light – Niels Bohr Institutet - Københavns Universitet

Home
Resize Print Bookmark and Share

Quantop > News > News at NBI > Quantum Memory

 
We have published a Nature article where we report storing the quantum state of a light pulse in an atomic sample. Read more below and/or follow one of the links in the menu-box. 		 



People involved in this experiment:

  • Brian Julsgaard (Niels Bohr Institute).
  • Jacob Sherson (University of Aarhus).
  • J. Ignacio Cirac (Max Planck Institute, Garching).
  • Jaromír Fiurášek (Universite Libre de Bruxelles).
  • Eugene S. Polzik (Niels Bohr Institute).
The experiment in short:

The quantum memory for light experiment deals with the fundamental question: given an unknown input state of light, can this be stored with high precision in atoms? The answer is yes, and we demonstrate this experimentally. The results have recently been published in Nature 432, 482 (2004) together with a News and Views summary (Nature 432, 453 (2004)). Links to press coverage is given at the bottom of this page.

Below we discuss the setup of the experiment in general terms. A few technical details can be found here. For non-physicists it may be useful to read the description of quantum memory for pedestrians or to have a look at the frequently asked questions.

To understand why the experiment is important, let us discuss the limitations Nature puts on us when we wish to store some information carried by light into atoms. First of all, as Niels Bohr pointed out many years ago, there is the so-called complementarity principle in quantum mechanics (and hence in Nature). This principle states that certain properties of a physical system cannot be measured (or even defined) precisely at the same time. For a light wave the strength and the phase are complimentary (strength and phase are explained here). As a consequence, if we encode information into the strength and the phase of a light beam, this information can only be retrieved again with a certain precision. This lack of precision materializes as noise - the so-called quantum noise.

However, we might not want to retrieve the information from the light. Rather we could be interested in storing the exact information in a memory unit. If this storage procedure works with a precision better than the precision with which we could ever measure the light state directly, we have a so-called "quantum memory".


How the quantum memory works: In the picture below we see a part of the experiment. An atomic sample of cesium gas contained in a paraffin coated glass cell is placed inside a magnetic shielding (the cylinders) to prevent unwanted external fields from disturbing the experiment. The atoms can be accessed by lasers, and the outgoing laser light can be measured on photo-detectors. We have the possibility to affect the atoms via magnetic fields by sending currents through a set of coils close to the atoms.


Quantum Mapping

The quantum memory protocol runs as follows: 1. An unknown input state of light is sent to the atoms. The strength and phase of this light is not exactly defined which is symbolized by the thickness of the line in the drawing of the light wave. 2. When light passes the atomic sample there is an exchange of information between light and atoms. A part of the light is stored in atoms (e.g. the phase information). At the same time the atoms act back on the outgoing light. In this process light and atoms become entangled. 3. The outgoing light is detected (e.g. the amplitude part). The obtained result both carries information about the incoming light amplitude and information about atoms. 4. With a feedback system the atoms are rotated by an amount conditioned on the measurement result. From the beginning the atoms contained quantum noise just as the input light, but the fact that light and atoms became entangled in step 2 enables us to cancel the information about atoms in the outgoing light with the initial quantum noise of atoms. The result is a storage of the incoming light state in the atomic system.


Comparison with a classical protocol: A classical memory performs a measurement on the light directly which is shown in the picture below. 1. We are given an input state of light as before with some line thickness symbolizing the quantum noise. 2. The complementarity principle prevents us from detecting both the strength and phase of the light at the same time. We can only obtain a half-good result of the two (the measurement result is symbolized with a double thickness wave). 3. To store the measured information in atoms (or any other appropriate physical system) we must rotate atoms according to our measurement. But the atoms also contain initial noise and as a result we have tripled the noise level in the procedure.


Quantum Mapping

In conclusion, a classical memory cell will always add two units of quantum noise in addition to the initial one unit of quantum noise in the input state. A quantum memory can in principle work without adding extra noise. In our experiment we have created a memory cell which adds less noise than a classical memory cell. Hence we have demonstrated quantum memory for light.

Quantum memory has applications for quantum communication and quantum computation. These topics may some day revolutionize todays computer abilities and network structure. The ability to understand and cross the borderline between quantum and classical protocols may also lead to better classical communication performance.

Funding:

This research was funded by the Danish National Research Foundation, by EU grants QUICOV, COVAQIAL and CHIC, and by the project 'Research Center for Optics' of the Czech Ministry of Education.

Further reading:

Scientific journals:

Press coverage in English:

Press coverage in Danish:

Other languages: