Abstract:Laser plasma acceleration (also referred to as laser acceleration) employs ultra-strong electric fields generated when an ultra-intense laser pulse interacts with the plasma to accelerate charged particles. As a promising novel acceleration method, its acceleration gradient can be 103-106 times that of conventional accelerators. Laser accelerators based on the laser acceleration process can not only generate ultra-short pulse particle/photon beams for fundamental research, but also provide high-energy protons/heavy ions for tumor radiotherapy devices, benefiting people's lives and health. Under the strong national investment around the world, basic and applied research related to laser accelerators has developed rapidly in recent years. Unlike conventional accelerators, which use macroscopic electromagnetic fields to accelerate particles in a vacuum, the laser acceleration process is realized by shooting targets with laser pulses. The scientific and technical problems related to the "target" are the core problems of the laser accelerator. In order to build an accelerator that can stably output high-energy particle beams, on the one hand, developing new target systems and deeply studying laser acceleration physics is highly necessary. On the other hand, overcoming a large number of key technical issues related to the preparation and characterization of targets and shooting is also very crucial. This paper gives an introduction to the key problems in the science and technologies for laser accelerators and highlights the progress in this field made by our team at the Institute of Heavy Ion Physics of Peking University. By exploiting nanotechnology and nanomaterials, we developed novel targets made of carbon nanotubes, which can serve as long-awaited perfect near-critical-density (NCD) targets. With such targets, we observed the laser focusing and pulse steepening in NCD plasmas for relativistic laser pulses for the first time. We also successfully achieved cascaded ion acceleration scheme by designing and building double-layer CNT targets, which solves the long-standing dilemma between the ionization and long-time acceleration of heavy ions acceleration. By collaborating with the Institute for Basic Sciences (IBS, Korea), we generated record-breaking 580 MeV carbon ions and 1.2 GeV Au ions in experiments. Recently, we numerically and experimentally studied how an ultra-intense femtosecond laser pulse interacts with a nanowire array target and eventually creates a high-energy-density plasma. The generation of soft-X-rays/EUV radiations and fusion neutrons from such a plasma were measured and studied in detail. Micro-channel targets are very interesting targets for us as well. We revealed how a laser pulse propagates through a micro-channel, peels off and accelerates electrons to GeV level, and produces corpus gamma rays with very small divergence angles. Experiments collaborating with the China Academy of Engineering Physics confirmed our prediction. Besides the above breakthroughs in acceleration sciences, we solved many key technical problems in laser ion accelerators. For example, we developed some mass-preparation methods for high-quality targets, including nanometerthin diamond-like-carbon foils and metal foils, measured the target damage thresholds of a majority of existing targets in laser plasma experiments, figured out how to precisely locate transparent targets with high-precision. For non-stop running of the laser ion accelerators, we recently developed liquid sheets, which is highly promising to realize continuous beam delivery for weeks.

Modeling of a Liquid Leaf Target TNSA Experiment using Particle-In-Cell Simulations and Deep Learning

Deep learning approaches for modeling laser-driven proton beams via phase-stable acceleration

Surrogate Models studies for laser-plasma accelerator electron source design through numerical optimisation

Generation of Heavy Ion Beams Using Short Pulse Lasers

Building robust surrogate models of laser-plasma interactions using large scale PIC simulation

Modeling the interaction of an ultra-high intensity laser pulse with nano-layered flat-top cone targets for ion acceleration

Applying Machine Learning Methods to Laser Acceleration of Protons: Lessons Learned from Synthetic Data

Deep learning with plasma plume image sequences for anomaly detection and prediction of growth kinetics during pulsed laser deposition

Efficient Laser Absorption and Enhanced Electron Yield in the Laser-Target Interaction by Using a Cone-Nanolayer Target

Physics-separating artificial neural networks for predicting sputtering and thin film deposition of AlN in Ar/N$_2$ discharges on experimental timescales

Automated and Rapid Target Position Alignment in Laser-Plasma Experiments Using Deep Learning Algorithms

Origins of plateau formation in ion energy spectra under target normal sheath acceleration

Deep learning-based predictive models for laser direct drive at the Omega Laser Facility

Three-dimensional particle-in-cell simulations of laser-driven multiradiation sources based on double-layer targets

Enhancement of Proton Acceleration Field in Laser Double-Layer Target Interaction

Toward machine-learning-assisted PW-class high-repetition-rate experiments with solid targets

Ray-tracing laser-deposition model for plasma particle-in-cell simulation

Science and Technologies in Targets for Laser Accelerators

Proton Acceleration from Laser Interaction with a Complex Double-Layer Plasma Target

Machine Learning Models Capture Plasmon Dynamics in Ag Nanoparticles