Feedback-enhanced algorithm for aberration correction of holographic atom traps

Abstract

We show that a phase-only spatial light modulator can be used to generate non-trivial light distributions suitable for trapping ultracold atoms, when the hologram calculation is included within a simple and robust feedback loop that corrects for imperfect device response and optical aberrations. This correction reduces the discrepancy between target and experimental light distribution to the level of a few percent (RMS error). We prove the generality of this algorithm by applying it to a variety of target light distributions of relevance for cold atomic physics. arXiv:1409.3151v1 [cond-mat.quant-gas]

Quantum reflection of bright solitary matter waves from a narrow attractive potential

“…Modeling of the system suggests that the exact spatial form of the potential well is crucial in determining the amount of reflection observed, with the presence of multiple optical diffraction maxima, rather than a single Gaussian maximum, playing an essential role. These results indicate that carefully engineered attractive multiwell potentials, readily generated using spatial light modulators [49], could be developed as robust beam splitters for use in solitary wave interferometry. Here the narrow, self-trapped nature of the solitary waves makes them ideal for measuring the transmitted and reflected fractions of a wave packet incident on a beam splitter [19,20,23,24].…”

An atomtronic flux qubit: a ring lattice of Bose–Einstein condensates interrupted by three weak links

“…To increase the accuracy of the output potential we use the computationally generated kinoform and produce an image of the optical potential in the monitoring arm of our system, and use this as a further source of feedback to the MRAF algorithm. Our method is broadly similar to Bruce et al [43], however it is specialised for producing ring-lattices. Figure 8 shows a flow chart of our improved algorithm.…”

Comparative simulations of Fresnel holography methods for atomic waveguides

“…where N is the number of pixels (in the simulation), I˜is the final intensity, and T is the target intensity distribution, both intensity distributions are normalised by the mean of the pixels in T that are brighter than 50% of the maximum value [31].…”

“…Despite the success of these simulations, they are limited to wavelength resolution due to the Helmholtz propagation method used and to a size of 4.38×4.38mm 2 by the memory requirements of the simulation. The calculation process explicitly does not include an algorithm for iterative error correction of the kinoform meaning that both FZP and SLM rms errors may find improvements with the use of algorithms similar to those used in [14,31].…”