Muhammad Firmansyah Kasim

AWAKE, The Advanced Proton Driven Plasma Wakefield Acceleration Experiment at CERN

The Advanced Proton Driven Plasma Wakefield Acceleration Experiment (AWAKE) aims at studying plasma wakefield generation and electron acceleration driven by proton bunches. It is a proof-of-principle R&D experiment at CERN and the world׳s first proton driven plasma wakefield acceleration experiment. The AWAKE experiment will be installed in the former CNGS facility and uses the 400 GeV/c proton beam bunches from the SPS. The first experiments will focus on the self-modulation instability of the long (rms ~12 cm) proton bunch in the plasma. These experiments are planned for the end of 2016. Later, in 2017/2018, low energy (~15 MeV) electrons will be externally injected into the sample wakefields and be accelerated beyond 1 GeV. The main goals of the experiment will be summarized. A summary of the AWAKE design and construction status will be presented.

Compression of X-ray free electron laser pulses to attosecond duration

State of the art X-ray Free Electron Laser facilities currently provide the brightest X-ray pulses available, typically with mJ energy and several hundred femtosecond duration. Here we present one- and two-dimensional Particle-in-Cell simulations, utilising the process of stimulated Raman amplification, showing that these pulses are compressed to a temporally coherent, sub-femtosecond pulse at 8% efficiency. Pulses of this type may pave the way for routine time resolution of electrons in nm size potentials. Furthermore, evidence is presented that significant Landau damping and wave-breaking may be beneficial in distorting the rear of the interaction and further reducing the final pulse duration.

FPGA implementation of fast serial 64-points FFT/IFFT block without reordering block

There has been many FPGA implementation of serial Fast Fourier Transform (FFT) operation. In the most cases, output of the serial FFT block is in bit-reversed order, so it needs a reordering block to reorder the output. However, some of FFT applications do not require ordered output of FFT, such like Spectral Subtraction method. In this paper, we propose an FPGA implementation of serial FFT and IFFT architecture in one block without reordering block. By not implementing the reordering block, we can save some clock cycles latency and increase speed of the block. The architecture is implemented in Altera DE2-70 board with Cyclone II EP2C35F672C6 FPGA chip. Our 64-points FFT/IFFT block utilizes 2960 logic elements or half of logic elements utilized by Altera MegaFunctions FFT IP. The block can work in maximum frequency of 84.55MHz and perform 64-points FFT/IFFT operation in 863.4ns.

Quantitative shadowgraphy and proton radiography for large intensity modulations

Shadowgraphy is a technique widely used to diagnose objects or systems in various fields in physics and engineering. In shadowgraphy, an optical beam is deflected by the object and then the intensity modulation is captured on a screen placed some distance away. However, retrieving quantitative information from the shadowgrams themselves is a challenging task because of the nonlinear nature of the process. Here, we present a method to retrieve quantitative information from shadowgrams, based on computational geometry. This process can also be applied to proton radiography for electric and magnetic field diagnosis in high-energy-density plasmas and has been benchmarked using a toroidal magnetic field as the object, among others. It is shown that the method can accurately retrieve quantitative parameters with error bars less than 10%, even when caustics are present. The method is also shown to be robust enough to process real experimental results with simple pre- and postprocessing techniques. This adds a powerful tool for research in various fields in engineering and physics for both techniques.

Optimization of plasma amplifiers

Plasma amplifiers offer a route to side-step limitations on chirped pulse amplification and generate laser pulses at the power frontier. They compress long pulses by transferring energy to a shorter pulse via the Raman or Brillouin instabilities. We present an extensive kinetic numerical study of the three-dimensional parameter space for the Raman case. Further particle-in-cell simulations find the optimal seed pulse parameters for experimentally relevant constraints. The high-efficiency self-similar behavior is observed only for seeds shorter than the linear Raman growth time. A test case similar to an upcoming experiment at the Laboratory for Laser Energetics is found to maintain good transverse coherence and high-energy efficiency. Effective compression of a 10kJ, nanosecond-long driver pulse is also demonstrated in a 15-cm-long amplifier.

Dense plasma heating by crossing relativistic electron beams

Here we investigate, using relativistic fluid theory and Vlasov-Maxwell simulations, the local heating of a dense plasma by two crossing electron beams. Heating occurs as an instability of the electron beams drives Langmuir waves, which couple nonlinearly into damped ion-acoustic waves. Simulations show a factor 2.8 increase in electron kinetic energy with a coupling efficiency of 18%. Our results support applications to the production of warm dense matter and as a driver for inertial fusion plasmas.

Machine learning applied to proton radiography of high-energy-density plasmas

Proton radiography is a technique extensively used to resolve magnetic field structures in high-energy-density plasmas, revealing a whole variety of interesting phenomena such as magnetic reconnection and collisionless shocks found in astrophysical systems. Existing methods of analyzing proton radiographs give mostly qualitative results or specific quantitative parameters, such as magnetic field strength, and recent work showed that the line-integrated transverse magnetic field can be reconstructed in specific regimes where many simplifying assumptions were needed. Using artificial neural networks, we demonstrate for the first time 3D reconstruction of magnetic fields in the nonlinear regime, an improvement over existing methods, which reconstruct only in 2D and in the linear regime. A proof of concept is presented here, with mean reconstruction errors of less than 5% even after introducing noise. We demonstrate that over the long term, this approach is more computationally efficient compared to other techniques. We also highlight the need for proton tomography because (i) certain field structures cannot be reconstructed from a single radiograph and (ii) errors can be further reduced when reconstruction is performed on radiographs generated by proton beams fired in different directions.

Validating Continuum Lowering Models via Multi-Wavelength Measurements of Integrated X-ray Emission

X-ray emission spectroscopy is a well-established technique used to study continuum lowering in dense plasmas. It relies on accurate atomic physics models to robustly reproduce high-resolution emission spectra, and depends on our ability to identify spectroscopic signatures such as emission lines or ionization edges of individual charge states within the plasma. Here we describe a method that forgoes these requirements, enabling the validation of different continuum lowering models based solely on the total intensity of plasma emission in systems driven by narrow-bandwidth x-ray pulses across a range of wavelengths. The method is tested on published Al spectroscopy data and applied to the new case of solid-density partially-ionized Fe plasmas, where extracting ionization edges directly is precluded by the significant overlap of emission from a wide range of charge states.

Playing the game of Congklak with reinforcement learning

Reinforcement learning is a branch of machine learning that allows an agent to learn to take an action based on its observations and rewards it obtains. In this paper, reinforcement learning agents are trained to play the game of Congklak, a traditional game from Indonesia and Malaysia. Congklak is a deterministic board game played by 2 players which play in turns. However, it was found that the common rules of Congklak make it possible for the first player to win the game without giving a turn to the second player. A change in rule is suggested to make the game more fair and more challenging to train artificial intelligent agents to play the game. The agents were trained based on the suggested rules using model-free reinforcement learning method combined with artificial neural network. After being trained in 3000 games against a random-moves opponent, the agents successfully beat the opponent with a winning chance of up to 90% without any prior knowledge of the game, not even the rules of the game.