Condensed-matter physics

Rotating molecules as quantum magnets

The push to engineer and probe quantum many-body systems using ultracold gases has reached a milestone with the observation of controlled dynamics caused by interactions between distant molecules trapped in a lattice. See Letter p.521

Experiments with ultracold gases of atoms and molecules open up many avenues for exploring fundamental questions of many-body physics, because they offer a high level of control over the trapping potentials that confine the gases and interparticle interactions1. On page 521 of this issue, Yan and colleagues2 provide a highly anticipated demonstration of a new system in this context, with potassium–rubidium molecules trapped in separate sites of an optical lattice formed by standing waves of laser lightFootnote 1. They observe the dynamics caused by dipolar interactions between distant molecules, setting the stage for the exploration of many-body phenomena and quantum phases of matter, especially those described by quantum spin models3,4,5,6.

In Yan and colleagues' experiment, gases of potassium and rubidium at nanokelvin temperatures are associated to form K–Rb molecules by means of a magnetic-field ramp, and are then placed in the lowest-energy state of their electronic, vibrational and rotational degrees of freedom using laser-induced transitions. Ultracold gases of such heteronuclear ground-state molecules were first produced in 2008 (ref. 7), and are particularly interesting for the study of many-body physics because of the strong electric dipole moments exhibited by the molecules (0.57 debye units for K–Rb)3,6.

But the authors have now gone a step further, introducing an optical lattice that traps molecules at individual lattice sites (Fig. 1), separated by 0.5 micrometres, in which the lifetime of the molecular gas (limited by heating processes) is 25 seconds. This allows the realization of models in which the translational motion of the molecules is frozen and the dynamics are dominated by the molecules' internal motion. The molecules can each be treated as effective spin-1/2 systems by identifying two different states of their rotational motion as spin up and spin down. Applying microwave fields to couple the rotational states allows the dipole–dipole molecular interaction to exchange angular momentum between spatially separated molecules, giving rise to effective interactions between the spins.

Figure 1: Trapped in a lattice.

Yan et al.2 demonstrate a system of ultracold heteronuclear molecules (red and green) in an optical lattice (grey) that can be used to study quantum spin models. The lattice freezes the translational motion of the molecules, and a dipole–dipole interaction between molecules (dipoles shown as thick black arrows) couples different states of their rotational motion (red arrows), which can be used to represent quantum-mechanical spins. The authors' system currently has a smaller filling fraction of molecules than is shown here.

Yan et al. measured the effects of these interactions on dynamics using a Ramsey spectroscopy technique, in which the rotation of the molecules is controlled through microwave pulses. The interactions show up as oscillations of the system's collective spin state, with a dominant oscillation frequency that is characteristic of the dipole–dipole interactions between molecules in neighbouring lattice sites. The authors also demonstrate how the effects of pairwise interactions can be cancelled using multiple microwave pulses, leaving a general damping of oscillations in the collective spin state that is caused by interactions between many molecules.

These measurements highlight the two key advantages of loading molecules into an optical lattice. First, because the translational motion is frozen, the effective temperature in the rotational motion — and therefore the spin systems — can be very low, even if the initial temperature of the gas would have been too high to observe the dynamics cleanly. Second, trapping molecules in individual sites prevents collisions between molecules. In the case of K–Rb, collisions would lead to chemical reactions that produce K2 and Rb2 dimers8. This reaction releases a lot of energy, normally leading to the loss of molecules from the system.

Yan and colleagues also showed that when the molecules are allowed to move through the lattice, tunnelling weakly from site to site, their loss is further suppressed by a quantum-mechanical phenomenon known as the continuous quantum Zeno effect. In this mechanism, a dissipative process (the reactive loss) suppresses a coherent one (in this case, the tunnelling), so that the rapid loss of molecules from the same site actually decreases the likelihood that a molecule will move to a lattice site where another molecule is already present. Thus, counterintuitively, a rapid loss process slows down the loss of molecules. Such suppression has previously been observed9 for highly excited molecules of Rb2, but it has now been observed cleanly for the first time in ground-state molecules.

This experimental system has been anticipated by many theoretical proposals, which have demonstrated that many classes of spin model can be realized by tailoring the dipole–dipole interactions with external electric and microwave fields3,4,5,6. This includes spin models that describe topological quantum phases4. Such exotic phases do not fit into the standard Landau theory that characterizes phase transitions with a local-order parameter — they are instead identified by highly non-local topological properties. As shown by Yan et al., these systems are particularly suitable for the study of non-equilibrium dynamics in strongly interacting spin models. Although some spin models can be implemented with atoms in optical lattices1, the interactions in those systems are limited to contact interactions when the atoms collide, and effective spin–spin interactions involve tunnelling of particles in the lattice. Dipolar molecular systems such as those of Yan and colleagues provide a means to create spin models with stronger interactions, allowing phenomena to be observed at much higher temperatures than those required in atomic systems.

At present, the biggest challenge in the K–Rb experiment is to increase the 'filling fraction' (the proportion of lattice sites occupied by a molecule), which is currently around 10%. This should ideally be close to 100% to realize some of the most interesting many-body models, which will require the production of molecular gases with higher initial density and at a lower temperature than achieved here. In the meantime, however, Yan and colleagues' work immediately opens the door to the study of dynamics in disordered spin models, with the randomness arising from the distribution of molecules in the lattice. As technical improvements are made over the next few years, this system will provide an exciting path for the realization of exotic many-body physics.


  1. 1.

    *This article and the paper under discussion2 were published online on 18 September 2013.


  1. 1

    Bloch, I., Dalibard, J. & Nascimbène, S. Nature Phys. 8, 267–276 (2012).

    ADS  CAS  Article  Google Scholar 

  2. 2

    Yan, B. et al. Nature 501, 521–525 (2013).

    ADS  CAS  Article  Google Scholar 

  3. 3

    Carr, L. D., DeMille, D., Krems, R. V. & Ye, J. New J. Phys. 11, 055049 (2009).

    ADS  Article  Google Scholar 

  4. 4

    Micheli, A., Brennen, G. K. & Zoller, P. Nature Phys. 2, 341–347 (2006).

    ADS  CAS  Article  Google Scholar 

  5. 5

    Barnett, R., Petrov, D., Lukin, M. & Demler, E. Phys. Rev. Lett. 96, 190401 (2006).

    ADS  Article  Google Scholar 

  6. 6

    Pupillo, G., Micheli, A., Büchler, H. P. & Zoller, P. in Cold Molecules: Theory, Experiment, Applications (eds Krems, R. V., Stwalley, W. C. & Friedrich, B.) 421–470 (CRC Press, 2009).

    Google Scholar 

  7. 7

    Ni, K. K. et al. Science 322, 231–235 (2008).

    ADS  CAS  Article  Google Scholar 

  8. 8

    Ospelkaus, S. et al. Science 327, 853–857 (2010).

    ADS  CAS  Article  Google Scholar 

  9. 9

    Syassen, N. et al. Science 320, 1329–1331 (2008).

    ADS  CAS  Article  Google Scholar 

Download references

Author information



Corresponding author

Correspondence to Andrew J. Daley.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Daley, A. Rotating molecules as quantum magnets. Nature 501, 497–498 (2013).

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.


Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing