2022, why quantum computing is creating a neutral atom fever
The just-launched quantum computing startup planqc recently announced the closing of a €4.6 million funding round [1]. The funding will be used to develop a highly scalable quantum computer operating at room temperature - based on atoms trapped in an optical lattice. planqc was founded by a team of scientists from the Max Planck Institute for Quantum Optics (MPQ) and the University of Munich, and is named after the pioneer of quantum theory, Max Named after Max Planck, the pioneer of quantum theory, planqc is the first startup in the Quantum Valley of Munich, one of Europe's leading centers of quantum technology.
Figure 1 Lab of planqc
The company's founding team combines decades of international research on neutral atom quantum technology from world-leading institutions such as Harvard, Oxford, UC Berkeley, University of Colorado, University of Innsbruck, Singapore's Center for Quantum Technologies (CQT) and MPQ.
Similar to other neutral atom companies, planqc's quantum computers store information in individual atoms and arrange them in highly scalable artificial light crystals. The quantum information is then processed through quantum gates based on precisely controlled laser pulses, leveraging MPQ's groundbreaking fundamental research and the world-leading German laser and photonics industry.
Johannes Zeiher, MPQ researcher and planqc co-founder, explains: "Our atoms are more than a million times colder than deep space and more than 1,000 times colder than the superconducting quantum bits used by IBM or Google, but because of the near-perfect isolation of our quantum bits from their surroundings, we can run computers at room temperature. We've routinely captured and controlled more than 2,000 atoms in MPQ's optical lattice simulator."
In fact, as you can see from the Photonic Box story, 2022 quantum computing has already kicked off a neutral atom fever. Multiple companies have launched 100+ quantum bit systems; a University of Chicago team has successfully captured 512 atoms; programmability, versatility and scalability of neutral-atom quantum computers have been demonstrated; and Pasqal and ColdQuanta have each acquired a quantum computing software company.
Why has quantum computing created a neutral atom craze? Because of its advantages in quantum simulation.
The neutral-atom quantum computer allows the control of the device at the so-called analog level in addition to the digital mode of quantum gates describing the time evolution of quantum bits, allowing the direct manipulation of the mathematical operators (Hamiltonian quantities) describing the evolution of the atomic system. It allows not only a finer control of the pulses during the application of the gate, but also a direct use of the Hamiltonian quantities of the system as computational resources. The level of fine control allowed by this analog setup, together with the large number of possible configurations, makes it a powerful tool for quantum processing [2].
Some of the current neutral-atom quantum computing companies are working on simulation approaches (e.g. Pasqal), while others are focused on gate approaches (e.g. Atom Computing). Among them, Pasqal is developing hybrid digital/analog solutions.
Figure 2 Major neutral atom quantum computing companies
Neutral-atom quantum processors (QPUs) are capable of implementing both digital and analog quantum processing tasks. In digital computing, quantum algorithms are decomposed into a series of quantum logic gates described by quantum lines as shown in Figure 3(a). These quantum gates are implemented by shining fine-tuned laser pulses onto a selected subset of individual atoms in a register. In simulated calculations, the laser is used to realize Hamiltonian quantities. The quantum bits evolve with time according to the Schrödinger equation, as shown in Figure 3(b). The final state of the system is probed by measuring the state of each quantum bit.
Figure 3 Digital versus analog processing. (a) In digital processing, a series of gates are applied to the quantum bits to implement the quantum algorithm. Each gate is implemented by individually addressing the quantum bits with a laser beam. (b) In analog processing, the quantum bits evolve at a specific Hamiltonian H, for example by irradiating the entire register with a laser beam. The wave function|ψ〉 of the system follows the Schrödinger equation.
Neutral atom arrays are suitable for realizing quantum Hamiltonian quantities and for implementing analog quantum processing. Riedberg atoms behave as giant electric dipoles which undergo dipole-dipole interactions and map to spin Hamiltonian quantities. Each quantum bit of the register behaves as a spin with the states |↓〉=|0〉 and |↑〉=|1〉. Depending on the Riedberg states involved in the process, the spins undergo different types of interactions and are transformed into different Hamiltonian quantities. The most studied model is the Ising model, where |↓〉 is one of the ground states and |↑〉 is the Ridderberg state.
The Ising Hamiltonian is a typical model for solving many problems in condensed matter. An example is a model describing how quantum magnets in materials science evolve at very low temperatures. In a neutral atom device, such a model could be implemented in a 1D, 2D or 3D array containing hundreds of atoms and would theoretically far exceed the computational power of a classical computer.
Another example of a spin model that could be implemented is the XY Hamiltonian. It naturally appears in the spin state |↓〉 and |↑〉 as two dipole-coupled Rydberg states. Coherent exchange between the spin states converts the pair of states |↓↑〉 to |↑↓〉. The exchange interaction is well suited for studying frustrated quantum magnets or excitation transport, especially in the context of photosynthesis, to understand how light energy is carried to the reaction center in the light-trapping complex. Correlating with the controlled geometry of the quantum bits in the register, it can also address the conductivity of topological materials such as organic polymers.
Combining various states and exploiting the geometry of the spins in the register, the neutral atom QPU allows the realization of a wide variety of spin Hamiltonian quantities. Among all candidates for analog quantum processing, Riedberg atoms in optical arrays are particularly suitable because they provide a very favorable quality factor Q~102.
The first natural application of the fully programmable neutral atom QPU is to explore and solve complex quantum phenomena in many scientific fields, from the behavior of solid-state materials to the kinetics of chemical and biochemical reactions. By guiding quantum entanglement and superposition, one can reproduce in the device the key elements considered sufficient to explain such physical phenomena.
In this sense, quantum devices can serve as simulators of fundamental natural processes that can be used to facilitate scientific discovery while significantly reducing computational costs. Due to their quantum nature, the scientific problems explored in this quantum simulation framework are not easily solved on classical devices. The source of the difficulty lies in the exponential growth of the size of Hilbert space with the number of interacting particles.
Quantum Simulation can be further divided into Analog and Digital quantum simulations, with the following differences.
Table 1 Analog and digital quantum simulations
Applications of quantum simulations of neutral atom arrays include all the fields of many-body physics, i.e., the study of the behavior of interacting quantum particle populations. This is a very broad field that includes almost all of condensed matter physics and quantum chemistry, in addition to nuclear physics and high energy physics.
1. Condensed matter physics
By allowing the simulation of quantum spin systems, neutral atom devices will open up a variety of new opportunities for condensed matter physics. In the last 60 years, spin models have been extensively studied in a variety of contexts, such as magnetism and excited transport. However, many important open questions remain the subject of active research, such as the nature of phase diagrams when spins are arranged with geometric frustration as a feature, the dynamics of systems after a sudden change of one parameter of the Hamiltonian quantity, the role of disorder in coupling, or their combination with cases where topology plays a role. In addition to quantum spin systems, atomic arrays can bring new insights into other solid-state systems of interest, such as electron systems.
Future research along these lines will help investigate new materials that may provide unprecedented functionality for energy transport and storage, or exhibit transformative properties such as high-temperature superconductivity.
2. Quantum chemistry
The ability to simulate electronic systems extends to quantum chemistry and biochemistry problems. While classical computational mechanics is sufficient to describe most of the properties of these systems (e.g., molecular dynamics), the introduction of quantum effects helps to understand some physical processes at the microscopic level.
The introduction of many-body quantum effects allows us to improve models and better understand the reactivity of some molecules by providing a more complete model of the electronic degrees of freedom of the molecular active site. Such studies usually amount to describing the low eigenstate of a very large electronic Hamiltonian. Quantum methods for finding eigenvalues rely on quantum phase estimation (QPE) algorithms, which can provide exponential speedups beyond classical methods, but remain impractical for NISQ quantum devices without error correction. Therefore, there is a need to exploit the capabilities of quantum hardware through variational procedures.
3. High energy physics, nuclear physics and cosmology
[2]https://arxiv.org/pdf/2006.12326
