Internships and Thesis

Topics for internships / thesis in the Cold Atoms group (2016-2017):


Detailed subjects:


Study of a Cold Atom Random Laser

Level : from L3 to M2 (master)


Wave propagation in diffusive media is an important subject for numerous fields (medical imaging, acoustics, seismology, stellar physics, …). The experiments that we pursue at INLN make use of an original medium: a cold-atom cloud. The peculiar properties of this diffusive medium (strong resonances, quantum internal structure of the scatterers, mechanical effects of light on the atoms, quantum effects…) give rise to a very rich physics.

In this context, one of our research topic is the random laser which corresponds to a laser without cavity, in which feedback is provided by multiple scattering in the gain medium. This kind of lasers has been studied for a few years and is still a hot topic in the photonics community [1].

After several years of preliminary research, allowing us to show that a cold atomic cloud can provide enough gain to sustain lasing [2], to theoretically establish the possibility to combine gain and multiple scattering simultaneously [3], to study the best gain mechanisms [4], and to find the appropriate detection scheme, the first random laser based on a cold atomic sample has been developed [5]. An advanced study is now needed to better characterize this process, with in particular the study of its coherence properties.

The main goal of this internship is to participate to the development of the coherence properties measurements. This will also allows to understand how a cold atomic cloud is created, with all the experimental aspects, and to participate to the data acquisition and analysis.

This internship can be continued by a thesis.


[1] The physics and applications of random lasers, D. Wiersma, Nature Phys. 4, 359 (2008).
[2] Mechanisms for Lasing with Cold Atoms as the Gain Medium, W. Guerin, F. Michaud et R. Kaiser, Phys. Rev. Lett. 101, 093002 (2008).
[3] Threshold of a Random Laser with Cold Atoms, L. Froufe-Pérez, W. Guerin, R. Carminati and R. Kaiser, Phys. Rev. Lett. 102, 173903 (2009).
[4] Towards a random laser with cold atoms, W. Guerin et al. , J. Opt 12, 024002 (2010).
[5] A cold-atom random laser, Q. Beaudoin, N. Mercadier, V. Guarrera, W. Guerin et R. Kaiser, Nature Phys. 9, 357 (2013).



Subradiance with on-site disorder

Level : M2 (master)


Wave propagation in diffusive media is an important subject for numerous fields (medical imaging, acoustics, seismology, stellar physics,…). The experiments that we pursue in our team make use of an original medium: a cold-atom cloud. The peculiar properties of this diffusive medium (strong resonances, quantum internal structure of the scatterers, mechanical effects of light on the atoms, quantum effects…) give rise to a very rich physics. One of the subjects we are studying is cooperative scattering.

When a photon is sent onto an atomic ensemble, it interacts collectively with the N atoms of the cloud and not simply with one of them. This results in measurable modifications in the scattering rate, the emission diagram or the temporal dynamics. We study these cooperative effects experimentally and theoretically. Recently, we managed to observe subradiance [1], and superradiance in the linear-optics regime [2]. We are now performing complementary experiments on subradiance in order to understand the influence of multiple scattering and of the atomic thermal motion.

The next step will be to add “on-site disorder” in the system, i.e., to change randomly the atomic resonant frequencies of each atoms. One idea to do so is to use the light-shift (AC Stark shift) induced by a 3D speckle pattern illuminating the atoms. There are two reasons to perform this experiment. First, it could be a way to enhance the relative weight of the subradiance states, which would make them easier to detect and to use for quantum-optics applications. Second, as suggested recently [3], it could induce a spatial localization of the subradiant modes with phase transition induced by the disorder, sharing similitudes with the celebrated Anderson localization.

The internship can be either numerical or experimental. If numerical, the goal will be to use the coupled-dipole model to test how on-site disorder can enhance subradiance, and to determine what experimental signature to look at for detecting the localization effect. If experimental, the goal will be to design, set up and characterize the optical system to produce the 3D speckle pattern, then to set it up on the cold-atoms apparatus and, if time allows it, to participate to a first series of experiments on subradiance with on-site disorder.

This internship can be continued by a thesis.


[1] Subradiance in a large cloud of cold atoms, W. Guerin, M. O. Araújo, and R. Kaiser, Phys. Rev. Lett. 116, 083601 (2016).
[2] Superradiance in a large and dilute cloud of cold atoms in the linear-optics regime, M. O. Araújo, I. Krešić, R. Kaiser, and W. Guerin, Phys. Rev. Lett. 117, 073002 (2016).
[3] Subradiance localization in the open 3D Anderson-Dicke model, A. Biella, F. Borgonovi, R. Kaiser, and G. L. Celardo, Europhys. Lett. 103, 57009 (2013).



Polarization of light scattered by atoms

Level : from M1 to M2 (master)


Light scattering by atoms gives rise to polarization when coherent superposition of quantum states is taking place, and the Hanle effect results from the partial decoherence caused by an external magnetic field. It is measured in the solar spectrum where it has allowed to show that weak magnetic fields are ubiquitous outside sunspots, in the so-called « quiet Sun » [1,2]. However, some puzzling anomalies have been recorded in the scattering polarization observed in some solar spectral lines,  as for example the D1 line of sodium, which shows a polarization peak at its resonance frequency whereas, according to the standard scattering theory it should not be polarizable.  Recently a new theoretical framework has been proposed to explain this « D1-enigma » through a quantum interference effect specific to multi-level atoms [3].

We propose to carry out a laboratory experiment to test this extended theory.  A Rubidium cell will be illuminated by a monochromatic laser beam which is tuned to scan the absorption line frequencies in the D2 range. The experimental setup will be aimed at measuring with the required accuracy the scattered polarization for various configurations of the exciting laser polarization, while compensating for the Earth magnetic field.

This internship proposal results from a common interest of astrophysicists and physicists in the exploration of matter-radiation interactions in light scattering processes [4]. It is the first step into a wide field exploration, where the next step will be to extend the experiment to the D1 range (the expected polarization rates are much smaller in the D1 range).  A similar experiment has been carried out by a Swiss group [3] but with potassium atoms and in slightly different conditions where the polarization may be destroyed by collisions in the atomic cell.  The experiment will be carried out at INLN in collaboration with M. Faurobert from the Lagrange Laboratory (University of Nice-Sophia Antipolis and Observatory of the Côte d’Azur) together with the group of Jan Stenflo  (Locarno observatory), and the group of  Dr. Nagendra (Bangalore Institute for Astrophysics).


[1] Investigation of weak solar magnetic fields. New observational results for the SrI 460.7 nm linear polarization and radiative transfer modeling, M. Faurobert et al., A&A 378, 627 (2001).  
[2] Solar magnetic fields as revealed by Stokes polarimetry, J.O. Stenflo, Astron. Astrophys. Rev. 21, 66 (2013).
[3] Physics of polarized scattering at multi-level atomic systems, J.O. Stenflo, ApJ. 801, 70 (2015).
[4] Cold and hot atomic vapors: a testbed for astrophysics?, Q. Baudouin, W. Guerin, R. Kaiser, in Annual Review of Cold Atoms and Molecules, vol. 2 (World Scientific, 2014).



Photon condensation

Level : from L3 to M2 (master)


Pour des bosons, une conséquence spectaculaire des collisions est la condensation des particules dans un état quantique occupé de façon macroscopique (condensation de Bose-Einstein). Ce phénomène de condensation peut être considéré comme un effet purement quantique. Cependant, différents travaux théoriques récents ont montré qu'une onde classique peut aussi exhiber un phénomène de condensation, dont les propriétés thermodynamiques sont analogues à celles de la condensation de Bose-Einstein, en dépit du caractère complètement classique du système d'ondes considéré [1]. De façon inattendue, cet effet de thermalisation peut être caractérisé par un processus d’auto-organisation de l’onde : il est thermodynamiquement avantageux pour l’onde de générer une structure cohérente à grande échelle afin d’atteindre l’état d’équilibre le plus désordonné. La condensation d’ondes classiques illustre ce phénomène de façon remarquable.

La mise en évidence expérimentale de ce phénomène nécessite un milieu non linéaire fortement défocalisant. Des résultats utilisant un cristal photoréfractif ont été publiés récemment [2]. De nombreuses questions restent cependant ouvertes et nous proposons de mettre en évidence ce phénomène de condensation dans un système conceptuellement plus simple afin d'en approfondir son étude expérimentale. Le milieu non linéaire dans lequel se propage l'onde optique est constitué par une vapeur chaude de rubidium, permettant de contrôler la non-linéarité à travers la puissance et la fréquence du laser incident. Le champ optique classique est produit par un diffuseur holographique. Ensuite nous étudierons les caractéristiques de l'intensité sortant de la cellule de rubidium chaud en champ proche et en champ lointain, en mettant comme source un champ de speckle correspondant à une distribution ‘thermique’ initiale. L’évolution d’une telle distribution de champ lors de la propagation non linéaire permet d’étudier la croissance d’une population ‘condensée’ ainsi que de la cohérence à grande portée. La mise en évidence d’un effet de thermalisation sera étudiée, ainsi que d’autres signatures de condensation d’ondes classiques.

Ce stage est de nature expérimentale, mais il est aussi possible d’effectuer une étude numérique sur les effets attendus lors de la propagation non linéaires des ondes.

Ce stage peut se prolonger en thèse.


[1] Condensation of Classical Nonlinear Waves, C. Connaughton, C. Josserand, A. Picozzi, Y. Pomeau and S. Rica, Phys. Rev. Lett. 95, 263901 (2005).
[2] Observation of the kinetic condensation of classical waves, C. Sun, S. Jia, C. Barsi, S. Rica, A. Picozzi and J. W. Fleischer, Nature Phys. 8, 471 (2012).