AQuA DIP: Advanced Quantum Approaches to Double Ionisation Processes

In order to 'view' electrons in the atom, we need to be able to describe their motion on a time-scale comparable to the interactions themselves. This is akin to taking a photograph- the faster an object is moving, the shorter the exposure time required to capture it. In atomic and optical physics we effectively use ultrashort 'camera' flashes (laser pulses) to image electrons in motion. While this field of attosecond physics (1 attosecond = 1 billionth of a billionth of a second) is well established, the theoretical description and computational models of the underlying mechanisms are relatively underdeveloped.

We are particularly interested, in this project, in double-ionisation processes: mechanisms whereby two electrons escape from an atomic target. There is a two-fold technological motivation for this interest. Firstly, X-ray Free-Electron Laser systems (or XFELs for short) allow the delivery of massively energetic radiation capable of multiply ionising a target with just one or two photons. Secondly, table-top sources capable of delivering short, intense laser pulses in the visible to mid-infrared regime drive dynamics in what is known as the 'strong-field' regime, where the electric field of the laser is strong enough to distort the potential well binding the electrons to the target, allowing them to escape. Processes such as 'non-sequential double ionisation' (NSDI) or 'resonant excitation and subsequent ionisation' (RESI) are fascinating because they depend not only on the characteristics of the laser field, but also on the detailed and complex interactions taking place between the many electrons.

Because of this complexity, there is a dearth of methods capable of providing theoretical insight to these processes. Computational or analytical methods tend to specialise in one particular regime (either XFEL or Strong-field) and even then, many choose to focus on the laser-driven dynamics, ignoring the electronic-interaction part of the picture.

For this reason, in this project we will develop three complementary methods to address double-ionisation dynamics in atoms. One is the R-matrix with time-dependence, or 'RMT' approach, which will use techniques of high-performance computing to break the immensely complex calculation into smaller parts, solved in parallel by many thousands of processors. The second is the Coulomb Quantum Strong Field Approximation, or 'CQSFA', which is an analytical technique based on electron trajectories. This lends enhanced interpretational power to the CQSFA, because it provides a clear physical picture of the dynamics, which can sometimes be obscured in the full, quantum-wavefunction picture employed by RMT. The third approach we will develop is a hybrid of RMT and CQSFA: giving the best of both: a sophisticated and accurate treatment of the core dynamics (the behaviour of the atomic target in the laser field) and a clear interpretation of the continuum dynamics (the behaviour of the ionised electrons).

Having developed these three, complementary tools we will apply them to problems of double ionisation in XFEL and strong-field laser pulses. In particular we will address those processes which contain a strong influence from both the laser field and the quantum mechanical behaviour of the target, as this is the regime largely unserviced by current methodology. Hence we will perform studies of NSDI and RESI, as well as core-ionization followed by Auger decay and single photon double ionization driven by XFEL light.

Understanding these mechanisms will inform our understanding of all light-mediated electronic processes, including important biological actions such as photosynthesis and the operation of the eye. The ultimate goal of attoscience is to control these processes and realise the potential benefits for society. Tools, such as those developed in this project, may prove to be incomparably valuable in this pursuit.

FUNDAMENTAL SCIENCE

* UNDERSTAND FUNDAMENTAL PHENOMENA

Scientific problems addressed will directly inform understanding of the fundamental fields of ultrafast electron dynamics, correlation effects and atomic structure. Improved understanding in these areas impacts

- astrophysics: atomic opacities, scattering cross sections
- materials science: electron transport
- fusion energy: scattering cross sections
- laser physics: electron dynamics, ultrafast phenomena, high harmonic generation
- radiation damage: electron dynamics in molecular systems
- plasma physics: scattering cross sections

Specific insights of analytical (CQSFA & hybrid) methods will elucidate

- interpretational tools: attosecond imaging
- asymptotic expansion techniques: e.g. used in quantum chemistry
- interferometric schemes: quantum sensing

* IMPROVE/DEVELOP EXPERIMENTAL TECHNIQUES

Double ionisation phenomena in intense laser matter interaction impacts techniques e.g.

- Pump-probe spectroscopy
- Attosecond laser pulse generation
- Multidimensional spectroscopy
- Free-electron laser science

Insights will inform attosecond time-resolved measurements in biological systems, via dialogue with Angus Bain (UCL), Jason Greenwood (QUB), whose research is in time-resolved spectroscopy in biomolecules and UCL's Biological Physics group which researches quantum effects in biological systems (Alexandra Olaya-Castro) and imaging biomolecules (Bart Hoogenboom).

* UNDERSTAND, IMPROVE AND ENHANCE THEORETICAL METHODS

Computational techniques employed

- novel mixed basis-set/finite difference techniques
- mixed (shared/distributed) memory parallelism
- novel hardware for HPC (GPU accelerators)

are generally applicable in myriad HPC applications.

The analytical techniques

- boundary problems, branch cuts and complex roots
- regularisation of singularities
- complex optimization approaches

are also employed in a wide range of mathematical research.

The hybrid approach will provide general insights into combined numerical/analytical methodologies.

ECONOMY AND SOCIETY

Improved understanding of light mediated charge transfer allows understanding and control of

- vision
- photosynthesis
- nano-scale & high-speed electronics
- manufacture & efficient operation of photovoltaic cells

with clear implications for medicine, human & plant biology, emerging technologies and renewable energy respectively.

R-matrix codes are used for modelling magnetic confinement fusion with consequences for clean energy, defence and security.

HPC aspects of the work will inform the development and design of new HPC facilities, one of which will be built in Belfast in 2021. This facility will be designed to enhance research in

- Health sciences
- Agriculture
- Food Security
- Financial Technology
- Cyber Security
- Creative media

We will use the new HPC facility for research-adjacent applications of artificial intelligence (AI), such as using machine learning to allocate computer resource intelligently to the various parts of the RMT calculations, or to optimise quantum interference patterns and bound-state reconstruction in the CQSFA codes. KironTech LTD applies machine learning and AI to healthcare and is a partner of the UCL Quantum Science and Technology Institute. They will provide secondments and crash courses at UCL related to the project.

Outreach programs impact on public understanding of science, including HPC.

Both PDRAs will receive high-end technical training relevant to academia and industry, in HPC techniques, advanced analytical methodologies and fundamental science. Our record in preparing researchers for positions outside of academia is evidenced by high-level positions held by former researchers Michael Lysaght (Irish Centre for High-End Computing), Laura Moore (SAP), Jie Wu (Engine), Toni Das (AI4BD GmBH), Tom Nygren (founder and CTO KironTech), and Bradley Augstein (Finastra).