Periodic Reporting for period 1 - ExIQ (Exponentially Improved Quantum memory)
Berichtszeitraum: 2020-06-15 bis 2022-06-14
The interaction between light and matter is a fundamental process in nature that is relevant for many quantum technologies. A prime example of such potential technologies are quantum memories. As our current IT technology relies on memories to store the information before it is processed — or send around the globe —, quantum technologies will rely heavily on quantum memories. Quantum memories will for example allow to securely communicate over global distances. Hereby, the security of the communication does not depend anymore on the limited resources of the attackers but is given by fundamental laws of physics.
During ExIQ, I worked towards the implementation of a new type of quantum memory based on a recent theoretical proposal [1]. This new type of quantum memory relies on the control of quantum interferences. In this context, the phenomenon is called selective radiance and suppresses losses, while enhancing the transfer of information into the desired channel.
The work was performed on a platform of cold atoms coupled to a nanofiber. Cold atom setups are system laser cool atoms (Caesium in our experiment) close to the absolute zero (~microkelvins). Under these conditions, the quantum state of each atom can be well controlled, and the atoms are well isolated from the environment. These properties make this technology an ideal candidate to test new theoretical proposals in a well-controlled and clean environment. In our system, quantum emitters (cold atoms) are coupled to a nanofiber via evanescent fields. Our platform allows to efficiently interface emitter in a controlled manner. Thus, it allows us to perform fundamental studies on the interaction between light send through the nanofiber and the coupled quantum emitters.
[1] A. Asenjo-Garcia, M. Moreno-Cardoner, A. Albrecht, H. J. Kimble, and D. E. Chang, ‘Exponential Improvement in Photon Storage Fidelities Using Subradiance and ``Selective Radiance’’ in Atomic Arrays’, Phys. Rev. X, vol. 7, no. 3, p. 031024, Aug. 2017
During ExIQ, I worked towards the implementation of a new type of quantum memory based on a recent theoretical proposal [1]. This new type of quantum memory relies on the control of quantum interferences. In this context, the phenomenon is called selective radiance and suppresses losses, while enhancing the transfer of information into the desired channel.
The work was performed on a platform of cold atoms coupled to a nanofiber. Cold atom setups are system laser cool atoms (Caesium in our experiment) close to the absolute zero (~microkelvins). Under these conditions, the quantum state of each atom can be well controlled, and the atoms are well isolated from the environment. These properties make this technology an ideal candidate to test new theoretical proposals in a well-controlled and clean environment. In our system, quantum emitters (cold atoms) are coupled to a nanofiber via evanescent fields. Our platform allows to efficiently interface emitter in a controlled manner. Thus, it allows us to perform fundamental studies on the interaction between light send through the nanofiber and the coupled quantum emitters.
[1] A. Asenjo-Garcia, M. Moreno-Cardoner, A. Albrecht, H. J. Kimble, and D. E. Chang, ‘Exponential Improvement in Photon Storage Fidelities Using Subradiance and ``Selective Radiance’’ in Atomic Arrays’, Phys. Rev. X, vol. 7, no. 3, p. 031024, Aug. 2017
During the project, I collaborated with a group of theoretical physicists at ICFO, Barcelona, Spain, led by Darrick Chang. During this collaboration, it was realized that the aims of the first work package could be drastically relaxed. Initially, it was planned to fill all lattice sites around the nanofiber with atoms as depicted in Figure 2.
However, based on a theoretical study carried out by Daniel Hümmer at ICFO, it was predicted that already for a filling factor in the range of 25% – 50%, the physics of selective radiance could be accessed. Our current setup already works in a regime where a filling factor up to 50% is readily obtained. In essence, those theoretical studies showed that when loading atoms at random sites, the possibility to create sub-chains with all sites filled, is reasonably high and these small sub-chains can reveal the initially predicted behaviour of selective radiance.
The second work package consisted of the implementation of a trapping scheme in which atoms are closer than half the wavelength of the emitter’s transitions. This was achieved by using two counter-propagating beams at 685nm which allows trapping atoms at a minimum distance of 447nm from each other as sketched in Figure 2. To our best knowledge, we are the first to implement a trapping scheme of atoms that are that close to each other around a nanofiber. To choose the best parameters of the trapping scheme, we performed a theoretical investigation of the trapping parameters. The trapping potentials depend on the laser powers and polarization of the laser beams at 685 nm and 935nm. As in the previous trapping scheme, the laser at 935nm attracts the atoms towards the nanofiber and the laser at 685nm prevents the atoms from falling onto nanofiber the surface, creating a trapping minimum at around 200nm from the nanofiber surface. In previous trapping schemes, the longitudinal confinement was created with two counter-propagation lasers at 935 with equal power.
Figure 3 Trapping potentials for a blue standing wave with unequal power from both sides.
In the new trapping scheme, the standing wave is created by two lasers at 685nm which complicates the situation. Due to the attractive potential of the blue detuned laser at 685nm, the atoms are trapped at a minimum of the standing wave. For equal power from both sides, the atoms would then fall onto the fibre surface. To prevent this, we calculated trapping potentials with unbalanced power from both sides. It is crucial to find a good trade-off between longitudinal and radial confinement when choosing the power in each arm of the blue laser.
Based on simulation (see Figure 3.) we choose to work with approximately 35mW and 3mW of blue power from each side. By measuring the transmission through the ensemble of trapped atoms see (Figure 4), we were able to observe atoms trapped in this laser configuration with a similar lifetime as in previous trapping schemes.
However, based on a theoretical study carried out by Daniel Hümmer at ICFO, it was predicted that already for a filling factor in the range of 25% – 50%, the physics of selective radiance could be accessed. Our current setup already works in a regime where a filling factor up to 50% is readily obtained. In essence, those theoretical studies showed that when loading atoms at random sites, the possibility to create sub-chains with all sites filled, is reasonably high and these small sub-chains can reveal the initially predicted behaviour of selective radiance.
The second work package consisted of the implementation of a trapping scheme in which atoms are closer than half the wavelength of the emitter’s transitions. This was achieved by using two counter-propagating beams at 685nm which allows trapping atoms at a minimum distance of 447nm from each other as sketched in Figure 2. To our best knowledge, we are the first to implement a trapping scheme of atoms that are that close to each other around a nanofiber. To choose the best parameters of the trapping scheme, we performed a theoretical investigation of the trapping parameters. The trapping potentials depend on the laser powers and polarization of the laser beams at 685 nm and 935nm. As in the previous trapping scheme, the laser at 935nm attracts the atoms towards the nanofiber and the laser at 685nm prevents the atoms from falling onto nanofiber the surface, creating a trapping minimum at around 200nm from the nanofiber surface. In previous trapping schemes, the longitudinal confinement was created with two counter-propagation lasers at 935 with equal power.
Figure 3 Trapping potentials for a blue standing wave with unequal power from both sides.
In the new trapping scheme, the standing wave is created by two lasers at 685nm which complicates the situation. Due to the attractive potential of the blue detuned laser at 685nm, the atoms are trapped at a minimum of the standing wave. For equal power from both sides, the atoms would then fall onto the fibre surface. To prevent this, we calculated trapping potentials with unbalanced power from both sides. It is crucial to find a good trade-off between longitudinal and radial confinement when choosing the power in each arm of the blue laser.
Based on simulation (see Figure 3.) we choose to work with approximately 35mW and 3mW of blue power from each side. By measuring the transmission through the ensemble of trapped atoms see (Figure 4), we were able to observe atoms trapped in this laser configuration with a similar lifetime as in previous trapping schemes.
Inspired by the idea behind selective radiance which engineers the interferences to harness useful properties of light, we investigated the interference effects behind antibunched light. Within the low saturation regime, the effect of antibunching is due to interference that up-to-now has been largely overlooked by the community. The idea that interference between coherently and incoherently scattered light is at the origin of antibunching has already been pointed out in 1983 [2], but only recently it has been experimentally investigated [3]. With our setup, we were able to reveal the interference fringes in the second-order correlation function g^((2) )which arise when controlling the phase between the coherently and incoherently scattered light. More precisely, we used the nanofiber setup as a tool to control this phase between the coherently and incoherently scattered light. We were able to reveal interference fringes in the photon statistics g^((2) ) (τ=0) as shown in Figure 5. We are in the process of writing an article about those findings.
In a similar spirit, we predicted effects for a single atom. By using spectral filtering antibunched light from a single emitter can be transformed into bunched light. Preliminary results show that this effect observable. The experimental observations, both with the nanofiber setup and with a single atom, were made possible by a more intuitive theoretical modelling that we developed during the MSCA action. Initially, the theoretical model was indented to describe the squeezed light observation which we published at the beginning of the MSCA action [4].
Recently, I have been working in collaboration with K. Kusmierek, K. Hammer and S. Mahmoodian from Hanover University on the understanding of how large ensembles of atoms create squeezed light and how it depends on the input power. A preprint on the topic will soon be made public.
[2] J. Dalibard and S. Reynaud, ‘Correlation signals in resonance fluorescence : interpretation via photon scattering amplitudes’, J. Phys., vol. 44, no. 12, pp. 1337–1343, 1983
[3] L. Hanschke et al., ‘Origin of Antibunching in Resonance Fluorescence’, Phys. Rev. Lett., vol. 125, no. 17, p. 170402, Oct. 2020
[4] J. Hinney et al., ‘Unraveling Two-Photon Entanglement via the Squeezing Spectrum of Light Traveling through Nanofiber-Coupled Atoms’, Phys. Rev. Lett., vol. 127, no. 12, p. 123602, Sep. 2021
In a similar spirit, we predicted effects for a single atom. By using spectral filtering antibunched light from a single emitter can be transformed into bunched light. Preliminary results show that this effect observable. The experimental observations, both with the nanofiber setup and with a single atom, were made possible by a more intuitive theoretical modelling that we developed during the MSCA action. Initially, the theoretical model was indented to describe the squeezed light observation which we published at the beginning of the MSCA action [4].
Recently, I have been working in collaboration with K. Kusmierek, K. Hammer and S. Mahmoodian from Hanover University on the understanding of how large ensembles of atoms create squeezed light and how it depends on the input power. A preprint on the topic will soon be made public.
[2] J. Dalibard and S. Reynaud, ‘Correlation signals in resonance fluorescence : interpretation via photon scattering amplitudes’, J. Phys., vol. 44, no. 12, pp. 1337–1343, 1983
[3] L. Hanschke et al., ‘Origin of Antibunching in Resonance Fluorescence’, Phys. Rev. Lett., vol. 125, no. 17, p. 170402, Oct. 2020
[4] J. Hinney et al., ‘Unraveling Two-Photon Entanglement via the Squeezing Spectrum of Light Traveling through Nanofiber-Coupled Atoms’, Phys. Rev. Lett., vol. 127, no. 12, p. 123602, Sep. 2021