Clark group

• Professor Stephen Clark

  Theme Lead for Theoretical Physics and Professor of Physics

Biography

My academic journey began with a MSci Physics degree at here in Bristol where I first encountered quantum theory and its curious features like entanglement. My final year project supervised by Prof Sandu Popescu cemented my interest in all things “quantum”. After completing a Masters in Advanced Studies (Part III Maths Tripos) at Cambridge in 2003 I began a PhD (or more accurately DPhil) at Oxford under the supervision of Prof Dieter Jaksch. There my work exploited quantum information inspired “tensor network” methods to accurately calculate the equilibrium and dynamical behaviour of synthetic solids built from ultra-cold atoms trapped in optical lattices.

Over the 12 years I remained in Oxford as a post-doctoral researcher I continued to develop tensor network methods and broaden my research interests. This was made possible by holding two research fellowships at Keble College and joint positions with other institutions. One position was with the Max Planck Institute for the Structure and Dynamics of Matter in Hamburg where I began a highly fruitful collaboration with Prof Andrea Cavalleri’s group modelling various novel effects in quantum materials subjected to ultra-fast optical excitation. I also held a joint position at the Centre for Quantum Technologies in Singapore where I studied the behaviour of quantum technologies like optically driven coupled resonator arrays and ion-traps.

In 2015 I took up a Lectureship at Bath and began assembling my own research group, before returning in 2018 to Bristol where I am now a Professor of Theoretical Physics.

Research Interests & Activities

I am a theoretical physicist with a life-long (so far) passion for trying to tackle the quantum many-body problem in its numerous guises, be it in condensed matter physics, ultra-cold atoms, or inside quantum technologies. The main methodology I have focused on over the past 20 years is tensor network theory. This approach makes a deep connection between the classical simulability of many-body quantum systems and the structure of entanglement, correlations and quantum mutual information in their ground and thermal states. Exploiting this tensor networks can accurately (lossless as possible) decompose and compress many-body states into smaller pieces (tensors) connected together in a network. It has been enormously successful for one-dimensional systems and in principle can also handle two-dimensional systems, albeit at an increased computational cost. Much of my work has involved applying tensor networks to understand equilibrium properties and time evolution of archetypal many-body systems, e.g. spin chains and Hubbard models of bosons or fermions. Using this I have explored a wide range of effects including rapidly quenching bosons between Mott insulating and superfluid phases, dynamically stabilising superconductivity, and engineering high-fidelity transport of spins states across chains. This work has applications in ultra-cold atoms, where experiments can implement model systems with unprecedented precision and test fundamental many-body physics. It also applies to quantum materials subjected to intense ultra-fast optical excitation, where the dynamical behaviour of electrons can be probed on a timescale commensurate with their motion. Furthermore many-body calculations are central for verifying quantum technologies by simulating underlying physical platforms and the quantum circuits under realistic noisy conditions. I have also worked on connecting and combining tensor networks with other successful numerical techniques such as variational Monte Carlo and artificial neural networks to help crack the many-body problem.

In recent years my interests have focused on non-Markovian open quantum systems. This is where a small quantum system is coupled to one or more much larger thermal baths. Such a setting can describe the non-equilibrium steady state of a nanoscale autonomous quantum machine, like an engine or a clock, and there are many open questions about how interactions influence their efficiency or precision. Tensor networks are very well suited to this problem since any thermal baths can be described as a one-dimensional chain. This success indicates how tensor networks can be combined with another powerful method for many-body systems, Dynamical Mean-Field Theory, which is based on solving an archetypal open quantum system called a quantum impurity problem. My current work is in this direction with the applications to strongly correlated materials and the diagnosis of entanglement in their novel phases.

Current researchers and PhD students