Ultra-broadband NMR and ESR
In magnetic resonance, the design of pulses with wide excitation bandwidth and reduced sensitivity to instrumental imperfections is of great interest. In NMR, quantitative analysis of nuclei with wide chemical shift ranges (19F, 31P, 13C, etc.), especially in the pharmaceutical industry for impurity analysis, requires robust excitation, with constant amplitude and phase, of signals over the full spectral width. However, the limited available radiofrequency power in NMR spectrometers distorts both signal intensities and phases and seriously degrades the accuracy of signal integration. In pulsed ESR, the problem of limited excitation bandwidth is much more serious and affects the majority of experiments.
We develop ultra-broadband pulse, especially using swept-frequency (adiabatic) pulses, purely numerical evolutionary methods like optimal control theory (OCT), or a combination of both, for a variety of applications. For example, Currently, in collaboration with AstraZeneca, we are developing new-generation ultra-broadband pulses for robust quantification and impurity analysis of fluorinated drugs.
NMR and ESR methodologies with optimal control theory
Controlling the dynamics of quantum spin systems using time-dependent Hamiltonians in the form of pulses (radiofrequency, microwave, laser) is the essence of method development in many areas of science, from magnetic resonance techniques and terahertz technologies for spectroscopy and imaging to trapped ions, and NV-centers in diamond, for quantum information processing and computing. Developing mathematical tools and numerical algorithms for robust and sophisticated manipulation of quantum spin systems and optimising experimental parameters are among the most active areas in spectroscopy and quantum measurements.
We develop software packages and numerical optimisation routines for experiment design using derivative-based optimal control and on-the-fly, sample tailored optimisation of NMR and ESR experiments using derivative-free optimisation protocols. Additionally, we explore potential applications of these techniques in other areas like optimal control of quantum metrology devices, spin tomography, quantum computation, and information processing.
Signal processing and parametric spectral estimation
Accurate modelling of magnetic resonance data, using parametric estimation techniques, significantly facilitates the process of spectral analysis and brings a major benefit to the field of NMR of biomolecules, metabolomics, complex mixtures, and pharmaceuticals. Such estimation methods give access to the spectral parameters (frequencies, amplitudes, phases, and damping factors), generally through fitting spectral data to models like a sum of damped complex exponentials in the presence of additive stationary noise.
We develop robust estimation routines, graphical user interfaces, and software packages for one- and multi-dimensional spectral estimation and analysis and explore their applications in magnetic resonance spectroscopy, including adaptive denoising of MR spectra and images, resolution enhancement and automated analysis of NMR data of small molecules, pharmaceuticals, metabolomics, and proteins for reaction monitoring and dynamics studies. Additionally, we are interested in combining the spectral estimation routines developed in our group with machine learning methods to extend the application of such techniques to the identification of unknown chemicals from magnetic resonance spectra.
NMR and ESR spectroscopy with quantum sensors
Designing sensitive spectroscopic techniques with high spatial resolution to study chemicals, living tissues, and surfaces at a cellular and molecular level is currently one of the most challenging objectives in the fields of spectroscopy and imaging. One obvious and challenging example is the study of the dynamics and the behaviour of the living cells as building blocks of living systems. In order to study the biological processes at the molecular level and in the physiologically relevant conditions, non-destructive, non-invasive spectroscopic methods with a cellular resolution are required.
One of our current research interests is the design and development of novel magnetic resonance techniques which allow us to detect magnetic resonance signals and images from samples of extremely low concentration, for example, a single cell and eventually a single molecule. Very recently, it has been shown that high-resolution NMR signals can be detected from a very small volume of samples (nanolitres) using nitrogen-vacancy defects in diamond (NV-centres) as an ultrasensitive magnetometer. These materials hold a great promise to shape our future and transform our technologies, especially in the spectroscopy and imaging of biological systems, since they can be used as biologically inert and extremely accurate measurement devices.
Associated research themes:
Advanced functional materials and interfaces
Innovative measurement and photon science
Theory and modelling in the chemical sciences