Experimental quantum simulations of many-body physics with trapped ions
TL;DR: An overview of different trapping techniques of ions as well as implementations for coherent manipulation of their quantum states and current approaches for scaling up to more ions and more-dimensional systems are given.
read more
Abstract: Direct experimental access to some of the most intriguing quantum phenomena is not granted due to the lack of precise control of the relevant parameters in their naturally intricate environment. Their simulation on conventional computers is impossible, since quantum behaviour arising with superposition states or entanglement is not efficiently translatable into the classical language. However, one could gain deeper insight into complex quantum dynamics by experimentally simulating the quantum behaviour of interest in another quantum system, where the relevant parameters and interactions can be controlled and robust effects detected sufficiently well. Systems of trapped ions provide unique control of both the internal (electronic) and external (motional) degrees of freedom. The mutual Coulomb interaction between the ions allows for large interaction strengths at comparatively large mutual ion distances enabling individual control and readout. Systems of trapped ions therefore exhibit a prominent system in several physical disciplines, for example, quantum information processing or metrology. Here, we will give an overview of different trapping techniques of ions as well as implementations for coherent manipulation of their quantum states and discuss the related theoretical basics. We then report on the experimental and theoretical progress in simulating quantum many-body physics with trapped ions and present current approaches for scaling up to more ions and more-dimensional systems.
read more
Chat with Paper
AI Agents for this Paper
Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps
Figures
Figure 4. Excerpt of the level scheme of 25Mg+ as an example of a hyperfine qubit (not to scale). 25Mg+ has a nuclear spin of I = 5/2 and thus a hyperfine-split ground state (S1/2, F = 3 and S1/2, F = 2). By applying a static magnetic field of a few Gauss, the degeneracy of the Zeeman sublevels is lifted. The Doppler cooling laser (labelled ‘BD’) is σ + polarized and detuned red by ![Figure 6. Comparison between parameters of geometric phase gates [110] with two ions using the axial motional modes and radial motional modes. (a) The parameters correspond to the gate from [123]. The axial centre-of-mass (COM) and stretch (STR) mode have a large frequency difference (2π × 1.6 MHz). The detuning of the Raman beams from the STR mode amounts to δSTR = −2π × 266 MHz. That is why the main contribution to the differential geometric phase between |↓↓〉/|↑↑〉 and |↓↑〉/|↑↓〉 is due to a (single) loop in the phase space of the STR mode. However, as already suggested in [110], the detuning from the COM mode is chosen to be an integer multiple of the detuning from the STR mode (δCOM = −5 × δSTR). Hence, there is no entanglement left between the electronic and motional modes at the gate duration Tg = |2π/δSTR| = 3.75µs. (Note that the spin-echo sequence is not included in Tg.) (b) The parameters correspond to a phase gate on two of the radial motional modes. The radial centre-of-mass (COM) and rocking (ROC) mode have a comparatively small frequency difference of only 2π × 130 kHz. The detunings from both modes are chosen to have the same absolute values resulting in (approximately) equal contributions to the acquired geometric phase from both modes. The gate duration according to the original implementation would amount to Tg = |2π/δSTR| = 15.4µs. As the displacement pulse is repeated in the second gap of the spin-echo sequence (compare figure 7) to cancel dynamic phases (compare [118] and see text), the duration increases by an additional factor of two.](/figures/figure-6-comparison-between-parameters-of-geometric-phase-1o7b7c9n.png)
Figure 6. Comparison between parameters of geometric phase gates [110] with two ions using the axial motional modes and radial motional modes. (a) The parameters correspond to the gate from [123]. The axial centre-of-mass (COM) and stretch (STR) mode have a large frequency difference (2π × 1.6 MHz). The detuning of the Raman beams from the STR mode amounts to δSTR = −2π × 266 MHz. That is why the main contribution to the differential geometric phase between |↓↓〉/|↑↑〉 and |↓↑〉/|↑↓〉 is due to a (single) loop in the phase space of the STR mode. However, as already suggested in [110], the detuning from the COM mode is chosen to be an integer multiple of the detuning from the STR mode (δCOM = −5 × δSTR). Hence, there is no entanglement left between the electronic and motional modes at the gate duration Tg = |2π/δSTR| = 3.75µs. (Note that the spin-echo sequence is not included in Tg.) (b) The parameters correspond to a phase gate on two of the radial motional modes. The radial centre-of-mass (COM) and rocking (ROC) mode have a comparatively small frequency difference of only 2π × 130 kHz. The detunings from both modes are chosen to have the same absolute values resulting in (approximately) equal contributions to the acquired geometric phase from both modes. The gate duration according to the original implementation would amount to Tg = |2π/δSTR| = 15.4µs. As the displacement pulse is repeated in the second gap of the spin-echo sequence (compare figure 7) to cancel dynamic phases (compare [118] and see text), the duration increases by an additional factor of two. 
Figure 10. Example for normal modes in a surface-electrode trap geometry similar to the one discussed in section 6. It consists of three trapping zones (red spheres) arranged in a triangle at mutual distances of 40µm. Each trapping zone has a (bare) frequency of 2π × 2 MHz corresponding to the vibration towards the centre of the triangle, 2π × 1.1 MHz for the out-of-plane vibration and 2π × 0.9 MHz for the third, perpendicular direction. The eigenvectors (green arrows) of the nine normal modes for an ion of mass m = 25 amu in each potential minimum are shown (first and last row: top view for in-plane modes; middle row: side view for out-of-plane modes; see appendix A for details of the calculation). The labels denote the frequencies corresponding to the modes. The (coupled) frequencies make up three triplets close to each bare frequency; two frequencies in each triplet are degenerate. For ![Figure 14. Schematic to illustrate the projection of the electrodes of the RF (yellow) and dc (shaded) electrodes on a surface as a way to scale towards two-dimensional arrays of ions. The black crosses indicate the positions of the minima of the pseudopotentials. (a) Cross section of the electrodes of a conventional linear RF trap with three-dimensional geometry and the electrode structures projected onto a surface. The dashed arrows point at the new location of the electrodes, the white areas represent isolating gaps. (b) Cross section (upper part) and top view (lower part) of the stripe electrodes. It has been proposed to concatenate several linear RF surface-electrode traps as depicted in (a) as a basic unit to span a two-dimensional array of ions [164] (red and blue disks representing ions in opposite spin states). For sufficiently small mutual ion distances and decoherence rates of the ions, this is an approach to scale analogue QS.](/figures/figure-14-schematic-to-illustrate-the-projection-of-the-i9ntj5w1.png)
Figure 14. Schematic to illustrate the projection of the electrodes of the RF (yellow) and dc (shaded) electrodes on a surface as a way to scale towards two-dimensional arrays of ions. The black crosses indicate the positions of the minima of the pseudopotentials. (a) Cross section of the electrodes of a conventional linear RF trap with three-dimensional geometry and the electrode structures projected onto a surface. The dashed arrows point at the new location of the electrodes, the white areas represent isolating gaps. (b) Cross section (upper part) and top view (lower part) of the stripe electrodes. It has been proposed to concatenate several linear RF surface-electrode traps as depicted in (a) as a basic unit to span a two-dimensional array of ions [164] (red and blue disks representing ions in opposite spin states). For sufficiently small mutual ion distances and decoherence rates of the ions, this is an approach to scale analogue QS. 
Figure 15. Illustration of the optimization results for the electrode structure for a basic triangular lattice with respect to the height of the ions above the traps at constant mutual ion distance. The white gap isolates RF and dc patches. The three red disks symbolize three ions at a constant distance of d = 40µm, hovering above the surface at a height of (a) h = 30µm, (b) h = 40µm and (c) h = 50µm. (Courtesy of Roman Schmied.) ![Figure 5. Implementations of different interaction types for hyperfine/Zeeman qubits. (a) An operation of type (a) can be implemented, for example, by two-photon stimulated-Raman transitions driven by a pair of laser beams (shown without motional dependence) or directly by a microwave field. These types of interactions can be used for single-qubit gates in QC and to simulate the effective magnetic field in the simulation of quantum spin Hamiltonians. (b) State-dependent forces (see type (c) in the text) can be created by two beams detuned by approximately the frequency of a motional mode. This interaction is used in the geometric phase gate for the displacement pulse [110, 123] or in the simulation of the quantum Ising Hamiltonian to create the effective spin–spin interaction [21, 47].](/figures/figure-5-implementations-of-different-interaction-types-for-3mxtf5gx.png)
Figure 5. Implementations of different interaction types for hyperfine/Zeeman qubits. (a) An operation of type (a) can be implemented, for example, by two-photon stimulated-Raman transitions driven by a pair of laser beams (shown without motional dependence) or directly by a microwave field. These types of interactions can be used for single-qubit gates in QC and to simulate the effective magnetic field in the simulation of quantum spin Hamiltonians. (b) State-dependent forces (see type (c) in the text) can be created by two beams detuned by approximately the frequency of a motional mode. This interaction is used in the geometric phase gate for the displacement pulse [110, 123] or in the simulation of the quantum Ising Hamiltonian to create the effective spin–spin interaction [21, 47].
![Figure 6. Comparison between parameters of geometric phase gates [110] with two ions using the axial motional modes and radial motional modes. (a) The parameters correspond to the gate from [123]. The axial centre-of-mass (COM) and stretch (STR) mode have a large frequency difference (2π × 1.6 MHz). The detuning of the Raman beams from the STR mode amounts to δSTR = −2π × 266 MHz. That is why the main contribution to the differential geometric phase between |↓↓〉/|↑↑〉 and |↓↑〉/|↑↓〉 is due to a (single) loop in the phase space of the STR mode. However, as already suggested in [110], the detuning from the COM mode is chosen to be an integer multiple of the detuning from the STR mode (δCOM = −5 × δSTR). Hence, there is no entanglement left between the electronic and motional modes at the gate duration Tg = |2π/δSTR| = 3.75µs. (Note that the spin-echo sequence is not included in Tg.) (b) The parameters correspond to a phase gate on two of the radial motional modes. The radial centre-of-mass (COM) and rocking (ROC) mode have a comparatively small frequency difference of only 2π × 130 kHz. The detunings from both modes are chosen to have the same absolute values resulting in (approximately) equal contributions to the acquired geometric phase from both modes. The gate duration according to the original implementation would amount to Tg = |2π/δSTR| = 15.4µs. As the displacement pulse is repeated in the second gap of the spin-echo sequence (compare figure 7) to cancel dynamic phases (compare [118] and see text), the duration increases by an additional factor of two.](/figures/figure-6-comparison-between-parameters-of-geometric-phase-1o7b7c9n.webp)
![Figure 14. Schematic to illustrate the projection of the electrodes of the RF (yellow) and dc (shaded) electrodes on a surface as a way to scale towards two-dimensional arrays of ions. The black crosses indicate the positions of the minima of the pseudopotentials. (a) Cross section of the electrodes of a conventional linear RF trap with three-dimensional geometry and the electrode structures projected onto a surface. The dashed arrows point at the new location of the electrodes, the white areas represent isolating gaps. (b) Cross section (upper part) and top view (lower part) of the stripe electrodes. It has been proposed to concatenate several linear RF surface-electrode traps as depicted in (a) as a basic unit to span a two-dimensional array of ions [164] (red and blue disks representing ions in opposite spin states). For sufficiently small mutual ion distances and decoherence rates of the ions, this is an approach to scale analogue QS.](/figures/figure-14-schematic-to-illustrate-the-projection-of-the-i9ntj5w1.webp)
![Figure 5. Implementations of different interaction types for hyperfine/Zeeman qubits. (a) An operation of type (a) can be implemented, for example, by two-photon stimulated-Raman transitions driven by a pair of laser beams (shown without motional dependence) or directly by a microwave field. These types of interactions can be used for single-qubit gates in QC and to simulate the effective magnetic field in the simulation of quantum spin Hamiltonians. (b) State-dependent forces (see type (c) in the text) can be created by two beams detuned by approximately the frequency of a motional mode. This interaction is used in the geometric phase gate for the displacement pulse [110, 123] or in the simulation of the quantum Ising Hamiltonian to create the effective spin–spin interaction [21, 47].](/figures/figure-5-implementations-of-different-interaction-types-for-3mxtf5gx.webp)
Citations
Generation of Non-classical Motional States of a Trapped Atom
D. M. Meekhof,Christopher Monroe,B. E. King,Wayne M. Itano,D. Leibfried,David J. Wineland +5 more
- 01 May 1996
TL;DR: In this article, the authors report the creation of thermal, Fock, coherent, and squeezed states of motion of a harmonically bound {sup 9}Be{sup +} ion.
702
Nobel Lecture: Controlling photons in a box and exploring the quantum to classical boundary
TL;DR: The experiments performed with this ''photon box'' at Ecole Normale Superieure (ENS) belong to the domain of quantum optics called ''cavity quantum electrodynamics'' as discussed by the authors, and have led to the demonstration of basic steps in quantum information processing, including the deterministic entanglement of atoms and the realization of quantum gates using atoms and photons as quantum bits.
622
Nobel Lecture: Superposition, entanglement, and raising Schrödinger’s cat
TL;DR: Experimental control of quantum systems has been pursued widely since the invention of quantum mechanics as mentioned in this paper, with many of these works focusing on the control of internal and external states of trapped atomic ions, such as Bose-Einstein condensation.
541
•Journal Article
An Open-System Quantum Simulator with Trapped Ions
Julio T. Barreiro,Markus Mueller,Philipp Schindler,Daniel Nigg,Thomas Monz,M. Chwalla,Markus Hennrich,Christian F. Roos,Peter Zoller,Rainer Blatt +9 more
TL;DR: In this paper, the authors realized an experimental toolbox for simulating an open quantum system with up to five quantum bits (qubits) using a quantum computing architecture with trapped ions, combining multi-qubit gates with optical pumping to implement coherent operations and dissipative processes.
Spread of Correlations in Long-Range Interacting Quantum Systems
TL;DR: This work analyzes a local quantum quench in the long-range Ising model in a transverse field to identify three dynamical regimes: short-range-like with an emerging light cone for α>2, weakly long range for 1<α<2 without a clear light cone but with a finite propagation speed of almost all excitations, and fully nonlocal for α<1 with instantaneous transmission of correlations.
385
References
Quantum computation and quantum information
TL;DR: This book inchides well-known classics such as "On the Criteria to be used in Decomposing Systems into Modules," "On a 'Buzzword': Hierarchical Structure," and "Software Aging."
20.4K
Cramming More Components Onto Integrated Circuits
Gordon E. Moore
- 01 Jan 1998
TL;DR: Integrated circuits will lead to such wonders as home computers or at least terminals connected to a central computer, automatic controls for automobiles, and personal portable communications equipment as mentioned in this paper. But the biggest potential lies in the production of large systems.
Quantal phase factors accompanying adiabatic changes
TL;DR: In this article, it was shown that the Aharonov-Bohm effect can be interpreted as a geometrical phase factor and a general formula for γ(C) was derived in terms of the spectrum and eigen states of the Hamiltonian over a surface spanning C.
9.9K
Many-Body Physics with Ultracold Gases
TL;DR: In this article, a review of recent experimental and theoretical progress concerning many-body phenomena in dilute, ultracold gases is presented, focusing on effects beyond standard weakcoupling descriptions, such as the Mott-Hubbard transition in optical lattices, strongly interacting gases in one and two dimensions, or lowest-Landau-level physics in quasi-two-dimensional gases in fast rotation.
Simulating physics with computers
TL;DR: In this paper, the authors describe the possibility of simulating physics in the classical approximation, a thing which is usually described by local differential equations, and the possibility that there is to be an exact simulation, that the computer will do exactly the same as nature.