Quantum optics

A grip on entanglement

Quantum buffers will be an essential part of quantum-information networks. A buffer that can preserve not only a 'quantum bit' but also a 'quantum image' is a major step towards creating those networks.

The long-term vision for quantum computation and communication is the use of these technologies as the basis of a quantum-information network. Quantum computers interlinked by quantum communication would form a quantum equivalent of the Internet. With the possibility of unbreakable encryption, simulations of large quantum-mechanical systems and exponentially faster computing for certain algorithms, quantum-information processing offers an important niche in information technologies, and may even be the technology of the future as classical systems reach their physical limits (for example, in transistors, the size of an atom). But, unlike their classical counterparts, all elements in the quantum network require the preservation not only of bits, but also of quantum bits.

A classical bit can have values of 0 or 1, but a quantum bit (qubit) can have a value of 0 or 1, or 0 and 1. Quantum repeaters, buffers, memories and so on all require that a quantum bit is preserved while in that element. An important advance, made by Marino et al.1 and reported on page 859, is the demonstration of an optical buffer (a device that can store light) that preserves the 'quantumness' of a light beam. A quantum buffer can be used to create quantum signals on demand, synchronize quantum bits and lengthen the distance over which a quantum-communication system can work. But Marino et al. went beyond this mark and demonstrated that their quantum buffer can preserve a precious quantum resource called entanglement — a quantum correlation that two objects can share even if physically separated. What's more, the researchers showed that entangled quantum images (images that exhibit quantum correlations in transverse dimensions) can also be preserved in the device.

In 1935, Albert Einstein, Boris Podolsky and Nathan Rosen (EPR) wrote a paper2 discussing what is now termed 'continuous-variable entanglement'. While trying to refute the idea that, in quantum mechanics, the wavefunction description of a physical system was complete, they ended up setting the stage for quantum information. They showed — correctly, even though they didn't agree with the results — that quantum mechanics predicts correlations between particles that can't be accounted for by classical mechanics.

Many researchers have contributed to our understanding of entanglement3,4,5,6,7. However, knowledge of continuous entanglement, as envisaged by EPR, was given a boost by Reid and Drummond8. Instead of looking at the correlations between the position and momentum of two particles, they considered those that can exist between 'squeezed' light beams9. Squeezed light beams obey the Heisenberg uncertainty relation (for example, the position and momentum of a particle can be known precisely, but not both at the same time) between light's electric-field components, which are referred to as quadratures and are analogues of the position and momentum.

The vacuum (minimum-uncertainty) state of light can be graphed in this quadrature representation as a circle at the origin of an xy plane — one quadrature on the x-axis and the other on the y-axis. The product of the dimensions (diameter squared) of the circle is equal to the minimum uncertainty. The simplest squeezed beam is called a squeezed vacuum, and is represented by an ellipse — rather than a circle — at the origin. It is described as 'squeezed' because, to obey Heisenberg's uncertainty principle, one axis of the circle is compressed at the expense of the elongation of the other.

In EPR squeezed beams, two light beams exhibit correlations in their quadratures that cannot be explained classically. For example, under the right conditions, the sum of their phases (which describe the positions of the light's peaks and troughs) and the difference in their intensities can be less than the possible values of two uncorrelated beams. Marino et al.1 used the inseparability criterion, which depends on the joint quadratures, as a reference for quantifying the degree of entanglement between two light beams. This criterion consists of calculating bounds (inseparability bounds) below which the beams, even if both squeezed, cannot be described independently10,11.

The quantum optical buffer used in Marino and colleagues' experiment is a gas of warm rubidium atoms. After creating two entangled, squeezed light beams (using a scheme known as four-wave mixing), one of the beams was sent to the quantum buffer. An additional pump laser beam, near Raman resonance with the squeezed light beam, amplified this beam and changed the gas's index of refraction to create a region of steep dispersion (rapid change in index as a function of wavelength of light). This steep dispersion caused the squeezed light beam to travel more slowly in the optical buffer than it would have done in free space. Marino et al. measured delays of up to 32 nanoseconds and showed that the two beams were inseparable (entangled) — not spatially but in quantum terms — both before and after one of them passed through the buffer.

But Marino and colleagues' buffer has one primary limitation. The amplification and delay of the squeezed light beam depend on the power of the pump laser: the higher the power, the more gain and delay. But crucially, this comes at the cost of a reduction in entanglement. With a cap on the power, the maximum delay for which entanglement can be maintained is fixed. However, this limitation should not be viewed as a 'no go' for squeezed-light buffers because there are ways around it, which several groups are pursuing.

Quantum information is not necessarily restricted to using qubits (two-state systems), because a photon is not limited to being in two states: it can be in more than two and in a superposition of many states at one time. Marino et al. demonstrated that their quantum buffer could also delay quantum images, which can be thought of as a superposition of many transverse states — those encoding the spatial information of the light beam. Although others have recently achieved all-optical buffering of quantum signals12,13 and classical images14,15,16,17, Marino et al. are the first to demonstrate the preservation of a quantum image (an image of the reduced Planck constant, ħ), noting the transverse correlations between their two light beams. With the hope of dramatically increased information-storage capacity, it is likely that quantum images will play an important part in the future of quantum-information processing.


  1. 1

    Marino, A. M., Pooser, R. C., Boyer, V. & Lett, P. D. Nature 457, 859–862 (2009). | Article |

  2. 2

    Einstein, A., Podolsky, B. & Rosen, N. Phys. Rev. 47, 777–780 (1935).

  3. 3

    Bohm, D. Quantum Theory (Prentice-Hall, 1951).

  4. 4

    Bell, J. S. Physics (Long Island City, 1965).

  5. 5

    Clauser, J. F., Horne, M. A., Shimony, A. & Holt, R. A. Phys. Rev. Lett. 23, 880–884 (1969).

  6. 6

    Aspect, A., Grangier, P. & Roger, G. Phys. Rev. Lett. 47, 460–463 (1981).

  7. 7

    Ekert, A. Phys. Rev. Lett. 67, 661–663 (1991).

  8. 8

    Reid, M. D. & Drummond, P. D. Phys. Rev. Lett. 60, 2731–2733 (1988).

  9. 9

    Ou, Z. Y., Pereira, S. F., Kimble, H. J. & Peng, K. C. Phys. Rev. Lett. 68, 3663–3666 (1992).

  10. 10

    Duan, L.-M., Giedke, G., Cirac, J. I. & Zoller, P. Phys. Rev. Lett. 84, 2722–2725 (2000).

  11. 11

    Simon, R. Phys. Rev. Lett. 84, 2726–2729 (2000).

  12. 12

    Hètet, G. et al. Opt. Express 16, 7369–7381 (2008).

  13. 13

    Broadbent, C. J., Camacho, R. M., Xin, R. & Howell, J. C. Phys. Rev. Lett. 100, 133602 (2008).

  14. 14

    Kasapi, A., Jain, M., Yin, G. Y. & Harris, S. E. Phys. Rev. Lett. 74, 2447–2450 (1995).

  15. 15

    Jain, M., Merriam, A. J., Kasapi, A., Yin, G. Y. & Harris, S. E. Phys. Rev. Lett. 75, 4385–4388 (1995).

  16. 16

    Camacho, R. M., Broadbent, C. J., Ali-Khan, I. & Howell, J. C. Phys. Rev. Lett. 98, 043902 (2007).

  17. 17

    Residori, S., Bortolozzo, U. & Huignard, J. P. Phys. Rev. Lett. 100, 203603 (2008).

Download references

Author information

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Howell, J. A grip on entanglement. Nature 457, 798–799 (2009) doi:10.1038/457798a

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.