Pulse engineering plays an important role in high-resolution NMR spectroscopy because the performance of existing pulses depends on experimental parameters like bandwidth or magnetic field inhomonegeities. The GRAPE optimization algorithm can be used to find the best pulse for a given set of parameters. This method has been used to design band-selective pulses, robust broadband excitation and inversion pulses and various universal rotation pulses.
The first part of this work is an extension of the systematic studies on broadband pulses. This time the GRAPE algorithm is used to design broadband 30° and 60° excitation pulses as well as universal rotation pulses with the same flip angles. Correlations between the best achievable quality factor and pulse duration have been measured for different bandwidths and degrees of rf-inhomogeneity tolerance. Minimum pulse durations for a given quality factor have been evaluated and compared to studies of 90° and 180° pulses. The obtained pulse shapes are similar to previously published point-to-point and universal rotation pulses optimized with this method.
The second part of this work is concerned with the design of ultra-broadband 19F-CMPG and 19F-PROJECT pulse sequences that could be used for ligand-based binding studies. The best CPMG sequence was a combination of a BURBOP-90 pulse with a BURBOP-180 pulse. For PROJECT, the best results were achieved using the same 90° pulse and a pair of BIBOP pulses instead of a universal rotation pulse. Simulations showed that the PROJECT sequence performs significantly better than the CPMG sequence in the presence of fluorine-fluorine couplings.
Table of Contents
1 Introduction
1.1 NMR spectroscopy
1.2 Pulse optimization
1.3 Exploring the physical limits of broadband 30° and 60° pulses
1.4 Designing broadband universal rotation pulses for 19F CPMG and 19F PROJECT
2 Theoretical background
2.1 NMR spectroscopy
2.1.1 Nuclear spins in a static magnetic field
2.1.2 The effect of a short radiofrequency pulse
2.1.3 Relaxation and signal detection
2.2 The CPMG experiment
2.3 Quantum mechanical description of nuclear spins
2.3.1 Basic concepts
2.3.2 The density matrix
2.3.3 The Hamiltonian
2.3.4 The Liouville-von-Neumann equation
2.4 Gradient ascent pulse engineering
2.4.1 Classes of NMR pulses
2.4.2 Optimal control theory
2.4.3 Optimization of point-to-point pulses
2.4.4 Optimization of universal rotation pulses
2.5 Broadband pulses in NMR spectroscopy
2.5.1 Adiabatic pulses
2.5.2 BIR-4 pulses
2.5.3 CHIRP pulses
2.5.4 Broadband excitation and inversion pulses (BEBOP and BIBOP)
2.5.5 Broadband universal rotation pulses (BURBOP)
2.5.6 ICEBERG pulses
3 Materials and Methods
3.1 Optimization of 30° and 60° excitation pulses
3.2 Optimization of 30° and 60° universal rotation pulses
3.3 Optimization of broadband 90° universal rotation pulses with different starting shapes
3.4 Simulations
4 Results and Discussion
4.1 Physical limits of broadband 30° and 60° excitation pulses
4.1.1 Nomenclature
4.1.2 Global quality factor as a function of pulse duration
4.1.3 Minimum pulse duration
4.1.4 Frequency offset profiles and local quality factors
4.1.5 Peak rf-amplitudes and pulse shapes
4.2 Broadband 30° and 60° universal rotation pulses
4.2.1 Nomenclature
4.2.2 Global quality factor and pulse duration
4.2.3 Frequency offset profiles and local quality factors
4.2.4 Peak rf-amplitudes and pulse shapes
4.3 Design of broadband CPMG and PROJECT sequences
4.3.1 Optimization of broadband 90° UR pulses
4.3.2 Optimization of 180° PP pulses
4.3.3 Simulation of different CPMG and PROJECT sequences
5 Conclusions and outlook
5.1 Broadband 30° and 60° pulses
5.2 Design of broadband CPMG and PROJECT sequences
A Pulses used in the CPMG and PROJECT sequences
Research Objectives and Core Topics
The primary objective of this thesis is to utilize the GRadient Ascent Pulse Engineering (GRAPE) algorithm to optimize specialized pulses and pulse sequences for high-resolution NMR spectroscopy, specifically targeting improvements in performance across broadband frequency ranges to facilitate ligand-based binding studies. The research addresses the limitations of standard pulses concerning bandwidth, rf-inhomogeneity, and signal distortion caused by fluorine-fluorine couplings.
- Optimization of broadband 30° and 60° excitation and universal rotation pulses.
- Exploration of the physical limits of small flip angle pulses under various constraints.
- Design of ultra-broadband 19F-CPMG and 19F-PROJECT pulse sequences.
- Refinement of refocusing pulses to minimize signal loss in multi-spin systems with scalar coupling.
- Comparative analysis of sequence performance in the presence and absence of fluorine-fluorine coupling.
Excerpt from the Thesis
4.1.2 Global quality factor as a function of pulse duration
The parameters for the optimizations are listed in Table 1. In Fig. 7, the maximum global quality factors Φ of BEBOP-30 and power-BEBOP-30 pulses are plotted as a function of pulse duration.
The global quality factors of power-BEBOP-30 pulses that were optimized for a specified bandwidth Δν and no rf-amplitude variation show an exponential time-dependence for short pulse durations. Similar plots were reported for 90° power-BEBOP pulses [8]. For smaller flip angles, the quality factors for a given pulse length and bandwidth are much higher than for 90° pulses.
An additional observation is that the exponential growth is only a small part of the plots (see Fig. 7.B). The other part of the plots probably wasn’t observed in the previous work because only pulses with Φ ≲ 0.995 were optimized. In contrast, this work allowed convergence to higher quality factors, as long as tp ≤ 200 µs. The second part of the plots shows the following behaviour: Once a quality factor of approx. 0.9999 is reached, there is almost no improvement in performance, but there are strong fluctuations (see Fig. 7.B). In the case of point-to-point pulses, it should always be possible to find a longer pulse with at least equal performance because the additional time could be used as a delay before the shorter pulse. Therefore, if some pulses have lower quality factors than shorter pulses with the same bandwidth, this means that the algorithm did not always converge to a global minimum. This can happen when using first order or second order approximations of gradients in numerical optimizations. In order to obtain the best possible quality factors in such situations, it is neccesary to perform more optimizations with different starting shapes.
Summary of Chapters
1 Introduction: Provides an overview of NMR spectroscopy, the challenges of pulse optimization, and the motivation for designing broadband pulses for 19F applications.
2 Theoretical background: Covers the physics of NMR, spin dynamics, the principles of optimal control theory, and established broadband pulse methodologies.
3 Materials and Methods: Details the algorithmic approach, including software (OCTOPUS/GRAPE), constraints, and simulation parameters used for pulse and sequence optimization.
4 Results and Discussion: Presents the findings regarding pulse physical limits, quality factor behavior, and the performance evaluation of CPMG and PROJECT sequences in different systems.
5 Conclusions and outlook: Summarizes the key results and suggests potential future research directions, such as experimental validation and the testing of different phase cycles.
A Pulses used in the CPMG and PROJECT sequences: Provides graphical representations and specifications of the optimized pulse shapes used in the computational studies.
Keywords
NMR spectroscopy, Pulse engineering, GRAPE algorithm, 19F-NMR, Pulse optimization, CPMG sequence, PROJECT sequence, Broadband pulses, Optimal control theory, Scalar coupling, Fluorine-fluorine coupling, Universal rotation pulses, Point-to-point pulses, Global quality factor, Signal sensitivity
Frequently Asked Questions
What is the core focus of this master thesis?
The thesis focuses on the optimization of radiofrequency (rf) pulses and pulse sequences for NMR spectroscopy using the GRAPE algorithm, particularly to achieve robust performance over broad frequency ranges in 19F-NMR.
What are the primary theoretical pillars of this research?
The work utilizes quantum mechanical descriptions of nuclear spins, the density matrix formalism for ensemble statistics, and optimal control theory to maximize pulse performance metrics.
What is the primary goal of the pulse optimization described?
The goal is to design excitation and universal rotation pulses that are robust against rf-inhomogeneity and bandwidth limitations, facilitating the design of more sensitive 19F-CPMG and PROJECT sequences for ligand-binding studies.
Which scientific methods are employed throughout the study?
The study employs the GRadient Ascent Pulse Engineering (GRAPE) algorithm, numerical simulations using Matlab and Python, and systematic comparative analysis of pulse quality factors against experimental constraints.
What does the main body of the work address?
The main part of the thesis discusses the systematic optimization of 30° and 60° pulses, the calculation of their physical performance limits, and the structural design of broadband 19F sequence variants.
Which specific terminology characterizes this work?
The research frequently uses terms such as "global quality factor," "broadband pulses," "offset profiles," "point-to-point (PP) pulses," and "universal rotation (UR) pulses" to describe pulse efficacy.
How does the PROJECT sequence differ from the standard CPMG experiment according to the author?
The PROJECT (Periodic Refocusing of J Evolution by Coherence Transfer) sequence refocused homonuclear J couplings, avoiding the J-modulation distortion that often affects conventional CPMG spectra when dealing with polyfluorinated compounds.
What impact does the choice of the phase factor have on the optimization of UR pulses?
The results indicate that the choice of the phase factor (eiφ) is critical for algorithm convergence in universal rotation (UR) pulse optimizations, as it prevents ambiguity in the cost function.
Why are "power-restricted" pulses considered advantageous?
Power-restricted pulses allow for higher peak rf-amplitudes compared to standard amplitude-restricted pulses, which often results in higher quality factors and shorter pulse durations for a given bandwidth.
What strategy did the author implement for the 180° PP pulse optimization to avoid local minima?
The author implemented a scheme involving the addition of noise to the best pulse and performing subsequent optimization rounds with tighter convergence parameters to escape local minima and improve the pulse quality factor.
- Citar trabajo
- Stella Slad (Autor), 2019, Optimization of Pulses and Pulse Sequences for NMR Spectroscopy, Múnich, GRIN Verlag, https://www.grin.com/document/1269029
